Citation
Integrative Use of Perennial and Annual Cover Crops for Weed Management in Organic Citrus

Material Information

Title:
Integrative Use of Perennial and Annual Cover Crops for Weed Management in Organic Citrus
Creator:
LINARES B., JOSE CLEMENTE
Copyright Date:
2008

Subjects

Subjects / Keywords:
Citrus trees ( jstor )
Clover ( jstor )
Cover crops ( jstor )
Crops ( jstor )
Nitrogen ( jstor )
Peanuts ( jstor )
Perennials ( jstor )
Seasons ( jstor )
Soil science ( jstor )
Weeds ( jstor )
City of Gainesville ( local )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Jose Clemente Linares B. Permission granted to University of Florida to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
8/31/2006
Resource Identifier:
649814560 ( OCLC )

Downloads

This item has the following downloads:

linares_j ( .pdf )

linares_j_Page_120.txt

linares_j_Page_206.txt

linares_j_Page_032.txt

linares_j_Page_063.txt

linares_j_Page_069.txt

linares_j_Page_059.txt

linares_j_Page_125.txt

linares_j_Page_198.txt

linares_j_Page_023.txt

linares_j_Page_033.txt

linares_j_pdf.txt

linares_j_Page_017.txt

linares_j_Page_038.txt

linares_j_Page_149.txt

linares_j_Page_018.txt

linares_j_Page_175.txt

linares_j_Page_041.txt

linares_j_Page_106.txt

linares_j_Page_043.txt

linares_j_Page_001.txt

linares_j_Page_064.txt

linares_j_Page_195.txt

linares_j_Page_142.txt

linares_j_Page_202.txt

linares_j_Page_165.txt

linares_j_Page_035.txt

linares_j_Page_046.txt

linares_j_Page_124.txt

linares_j_Page_097.txt

linares_j_Page_050.txt

linares_j_Page_189.txt

linares_j_Page_055.txt

linares_j_Page_039.txt

linares_j_Page_065.txt

linares_j_Page_168.txt

linares_j_Page_076.txt

linares_j_Page_136.txt

linares_j_Page_132.txt

linares_j_Page_091.txt

linares_j_Page_187.txt

linares_j_Page_011.txt

linares_j_Page_134.txt

linares_j_Page_156.txt

linares_j_Page_167.txt

linares_j_Page_146.txt

linares_j_Page_099.txt

linares_j_Page_111.txt

linares_j_Page_025.txt

linares_j_Page_172.txt

linares_j_Page_048.txt

linares_j_Page_006.txt

linares_j_Page_119.txt

linares_j_Page_186.txt

linares_j_Page_024.txt

linares_j_Page_190.txt

linares_j_Page_185.txt

linares_j_Page_082.txt

linares_j_Page_072.txt

linares_j_Page_093.txt

linares_j_Page_193.txt

linares_j_Page_105.txt

linares_j_Page_127.txt

linares_j_Page_147.txt

linares_j_Page_135.txt

linares_j_Page_199.txt

linares_j_Page_148.txt

linares_j_Page_022.txt

linares_j_Page_158.txt

linares_j_Page_129.txt

linares_j_Page_128.txt

linares_j_Page_192.txt

linares_j_Page_203.txt

linares_j_Page_057.txt

linares_j_Page_184.txt

linares_j_Page_114.txt

linares_j_Page_013.txt

linares_j_Page_143.txt

linares_j_Page_183.txt

linares_j_Page_052.txt

linares_j_Page_169.txt

linares_j_Page_062.txt

linares_j_Page_153.txt

linares_j_Page_191.txt

linares_j_Page_176.txt

linares_j_Page_140.txt

linares_j_Page_012.txt

linares_j_Page_181.txt

linares_j_Page_100.txt

linares_j_Page_030.txt

linares_j_Page_180.txt

linares_j_Page_164.txt

linares_j_Page_074.txt

linares_j_Page_107.txt

linares_j_Page_094.txt

linares_j_Page_078.txt

linares_j_Page_071.txt

linares_j_Page_066.txt

linares_j_Page_045.txt

linares_j_Page_087.txt

linares_j_Page_079.txt

linares_j_Page_179.txt

linares_j_Page_101.txt

linares_j_Page_086.txt

linares_j_Page_090.txt

linares_j_Page_014.txt

linares_j_Page_123.txt

linares_j_Page_196.txt

linares_j_Page_067.txt

linares_j_Page_104.txt

linares_j_Page_154.txt

linares_j_Page_007.txt

linares_j_Page_021.txt

linares_j_Page_042.txt

linares_j_Page_083.txt

linares_j_Page_197.txt

linares_j_Page_178.txt

linares_j_Page_118.txt

linares_j_Page_058.txt

linares_j_Page_157.txt

linares_j_Page_056.txt

linares_j_Page_110.txt

linares_j_Page_207.txt

linares_j_Page_188.txt

linares_j_Page_108.txt

linares_j_Page_053.txt

linares_j_Page_019.txt

linares_j_Page_073.txt

linares_j_Page_031.txt

linares_j_Page_139.txt

linares_j_Page_200.txt

linares_j_Page_037.txt

linares_j_Page_182.txt

linares_j_Page_028.txt

linares_j_Page_075.txt

linares_j_Page_085.txt

linares_j_Page_051.txt

linares_j_Page_174.txt

linares_j_Page_004.txt

linares_j_Page_159.txt

linares_j_Page_144.txt

linares_j_Page_201.txt

linares_j_Page_171.txt

linares_j_Page_016.txt

linares_j_Page_161.txt

linares_j_Page_034.txt

linares_j_Page_170.txt

linares_j_Page_047.txt

linares_j_Page_166.txt

linares_j_Page_133.txt

linares_j_Page_205.txt

linares_j_Page_005.txt

linares_j_Page_081.txt

linares_j_Page_109.txt

linares_j_Page_070.txt

linares_j_Page_026.txt

linares_j_Page_103.txt

linares_j_Page_068.txt

linares_j_Page_116.txt

linares_j_Page_036.txt

linares_j_Page_054.txt

linares_j_Page_117.txt

linares_j_Page_049.txt

linares_j_Page_173.txt

linares_j_Page_027.txt

linares_j_Page_131.txt

linares_j_Page_160.txt

linares_j_Page_060.txt

linares_j_Page_061.txt

linares_j_Page_204.txt

linares_j_Page_088.txt

linares_j_Page_040.txt

linares_j_Page_113.txt

linares_j_Page_092.txt

linares_j_Page_095.txt

linares_j_Page_121.txt

linares_j_Page_115.txt

linares_j_Page_096.txt

linares_j_Page_010.txt

linares_j_Page_141.txt

linares_j_Page_080.txt

linares_j_Page_089.txt

linares_j_Page_084.txt

linares_j_Page_009.txt

linares_j_Page_155.txt

linares_j_Page_002.txt

linares_j_Page_194.txt

linares_j_Page_137.txt

linares_j_Page_020.txt

linares_j_Page_015.txt

linares_j_Page_102.txt

linares_j_Page_177.txt

linares_j_Page_008.txt

linares_j_Page_151.txt

linares_j_Page_003.txt

linares_j_Page_077.txt

linares_j_Page_145.txt

linares_j_Page_122.txt

linares_j_Page_098.txt

linares_j_Page_150.txt

linares_j_Page_163.txt

linares_j_Page_029.txt

linares_j_Page_126.txt

linares_j_Page_162.txt

linares_j_Page_138.txt

linares_j_Page_130.txt

linares_j_Page_112.txt

linares_j_Page_044.txt

linares_j_Page_152.txt


Full Text











INTEGRATIVE USE OF PERENNIAL AND ANNUAL COVER CROPS FOR WEED
MANAGEMENT IN ORGANIC CITRUS















By

JOSE CLEMENTE LINARES B.


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


2006

































Copyright 2006

by

Jose Clemente Linares B.



































A Dios, mis Maestros, mis padres y familiar que me impulsan a mejorar como ser human















ACKNOWLEDGMENTS

I would like to acknowledge my major professor Johan M. Scholberg for his

support, guidance, his effort and patience throughout my graduate training program. I

would also like to thank the other members of my supervisory committee, Dr. K. Boote,

Dr. C. Chase, Dr. D. Graetz, and Dr. R. McSorley, for their support, excellent advice, and

assistance during my program, and for their contributions to my dissertation. Special

thanks go to Andy Schreffler, Jim Boyer, and staff of the UF-IFAS Plant Science

Research and Education Unit in Citra for their help with field studies. I want to further

thank Corey Cherr, Robert Wavestraut, Alicia Lusiardo, Hannah Snyder, Huazhi Liu,

Scott Prospect, Scott Tubbs, Kari Reno, Susan Sorell, Dipen Patel, Jorge Gomez,

Jonathan Bracho, Amy Van Scoick, John McQueen, and Laura Avila, among others, for

their assistance and friendship. I want to express my appreciation for the support and

technical assistance of Juan Carlos Rodriguez and Marty Mesh from Florida Organic

Growers. I also want to acknowledge Dr. Ramon Littell, Dr. Ken Portier, Salvador

Gezan, Enrique Darghan, and Meghan Brennan for their assistance with statistical

analysis.

I further want to express my gratitude to the "Universidad del Tachira", Venezuela,

for providing me with an opportunity to come to the University of Florida and to the

USDA/ CSREES for the financial support of my program.

I further want to thank Drs. Paul. Pfahler and his family, Heartwell Allen, and

Maria Luisa Izaguirre for their friendship and encouragement during the past years. I also









sincerely value the friendship and help of Belkys Bracho, Marco, Nicary, and Ver6nica

Emhart, during my stay in Gainesville and Sonia, Betty, Chavela, Maria de los Angeles,

Padr6n, and Alexis for supporting me in difficult moments.

I give thanks to God my Lord for assisting me to embrace the challenges I faced

during the past years and for all the blessings I have enjoyed as well. I would like to

express my gratitude to my parents, sisters, and brother for being so supportive and for

their continuous encouragement during my studies and my stay in Florida.
















TABLE OF CONTENTS

page

A C K N O W L E D G M E N T S ................................................................................................. iv

LIST OF TABLES ....................................................... .... .... .... ... ..... ..... ix

LIST OF FIGURES ......... ....... .................... .. ....... ........... xiv

A BSTR A C T ......... .............. ................................................ ... ... ....... .. xv

CHAPTER

1 IN T R O D U C T IO N ............................................................................... .............. ..

2 GROWTH AND EFFECTS OF ANNUAL COVER CROPS ON WEED
SUPPRESSION ......................................................................... ............. 9

Intro du action ....................................................................................................... ... .. 9
M materials and M methods ........................................................................ .................. 14
Set-up and Experim ental D esign....................................................................... 14
Data Collection, Measurements, and Analysis..........................................16
Summer Cover Crops ...............................................................................................18
Sum m er C over C rops ............................................................18
S u m m e r 2 0 0 2 ............................................................................................... 1 9
S u m m er 2 0 0 3 ............................................................................................... 2 0
S u m m e r 2 0 0 4 ............................................................................................... 2 1
Summer 2005 .............. ... ................21
W inter C ov er C rop s........................................................................................ 2 2
W in ter 2 0 0 2 /2 0 0 3 .................................................................................... 2 2
W in ter 2 0 0 3/2 0 0 4 .................................................................................... 2 3
W in ter 2 0 0 4 /2 0 0 5 .................................................................................... 2 4
D isc u ssio n .............................................................................................................. 2 5
Sum m er Cover Crops ............................................... ............... 25
W inter C ov er C rop s........................................................................................ 2 9
System Dynam ics ............................................................................ 31
C o n c lu sio n s............................................................................................................ 3 8









3 ESTABLISHMENT OF PERENNIAL PEANUT AND ITS EFFECTIVENESS
IN SUPPRESSING WEEDS IN CITRUS ROW MIDDLES .................................... 52

In tro du ctio n ...................................... ................................................ 52
M materials and M methods ....................................................................... ..................55
Set-up and Experim ental D esign............... ....................................................... 55
Data Collection, Measurements, and Analysis........................................58
R e su lts ......................................................... ................. ................ 6 0
P erennial P eanut 2 002 .............................................................. .....................60
Perennial Peanut 2003 ................... .... .......... ............. ....... .....61
P erennial P eanut 2 004 .............................................................. .....................62
Perennial Peanut 2005 ............... ................................. ............................. 62
Perennial Peanut Productivity (2005).................................. ..... ............... 63
Annual Cover Crops (2002-2005) ........... ................................ ...............64
System D ynam ics ........................ ................ ................... ..... .... 65
D isc u ssio n ............................................................................................................. 6 6
C o n c lu sio n s........................................................................................................... 7 7

4 EFFECTS OF PERENNIAL PEANUT (Arachis glabrata Benth.) AND
COMMON BERMUDAGRASS (Cynodon dactylon L.) ON NITROGEN AND
W A TER U PTA K E O F CITRU S ...............................................................................91

In tro d u ctio n .......................................... ... ......................... ................ 9 1
M materials and M methods ....................................................................... ..................94
Set-up and Experim ental Design ................................................. .............. 94
Irrigation and ET Calculations ........................................ ........................ 97
N itrogen A application ................. ............................ .... ...... .. .. ........ .... 98
N itrogen E xtraction .............................. ........................ .. ........ .... ............98
N itrogen U ptake Calculation........................................ ........................... 99
Final Plant Sam pling ......................... ..... ... .. .. ... ...............100
Statistical A n aly sis .............................................. ... .. ...... .. .... ............100
R esu lts ....................... .... ... .. ............ ........... ..................... 102
Groundcover Biomass Production and N Accumulation ...............................104
Final Citrus and Groundcover Growth and N Accumulation..........................104
D iscu ssio n ................ ..... ..1.. ...... ........ ...... ................................ 10 5
Groundcover Biomass Production and N Accumulation ...............................109
Final Citrus and Groundcover Growth and N Accumulation..........................110
C o n c lu sio n s................................................... .................. 1 1 1

5 EFFECTS OF ANNUAL AND PERENNIAL COVER CROPS ON SOIL AND
CITRUS TREE CHARACTERISTICS, CITRUS TREE ROW GROUND
COVER, AND CITRUS YIELD AND QUALITY ...............................................126

In tro du ctio n ................................................................................................ ..... 12 6
M materials and M methods ........................................... ....................................... 129
Set-up and Experimental Design .............................................. ...............129
Data Collection and Measurements ............... ...........................................132









S o il ....................................................... 1 3 2
Nematodes ...................................................................... ......... .................133
W eed Growth Dynamics ............. .................................. ................... 133
Citrus Tree Perform ance ........................................ ........................ 134
Data Analysis ......................................................................... ........ ........ .... ........ 134
R e su lts ......... ............................................................. ........................... 13 5
S o il p H ..................................................... 13 5
Soil C, N and C:N ratio ................................................................... 135
Soil Nematodes ............. .. .... ... .. ... ................. 136
Tree Row Ground Cover in Perennial Cover Crop Study..............................137
Citrus Tree Growth Characteristics, Citrus Leaves N, and Fruit Quality .........137
Discussion ........................ ..........................138
S o il p H ..................................................... 13 8
Soil C, N, and C:N Ratio ......... ......... ......... ...................... 139
N em atode Counts ........... .......... ......................... ......... .. .............. 141
Tree Row Ground Cover in Perennial Cover Crop Study..............................141
Citrus Tree Perform ance......................................................... ............... 142
C o n clu sio n s.................................................... ................ 14 3

6 SUM M ARY AND CONCLUSIONS ............................. .................................... 155

A annual C over C rop Study ............................................... ............................. 156
Perennial Cover Crop Study ........................................ .....................................157
Citrus, Perennial Peanut, and Bermudagrass Competition for Nitrogen and Water 158
Effect of Cover Crops on Soil Characteristics, Tree Row Cover and Citrus
G row th and Y field ............. .............................................. ............ .. .... .. ... .. 16 1
Im plications of the R research ......................................................... ............... 162
Future Research Recom m endations ........................................ ...... ............... 163

APPENDIX

A ANALYSES OF VARIANCE FOR PERENNIAL PEANUT STUDY................. 165

B ANALYSES OF VARIANCE FOR EFFECTS OF PERENNIAL PEANUT
(Arachis glabrata Benth.) AND COMMON BERMUDA GRASS (Cynodon
dactylon L.) ON NITROGEN AND WATER UPTAKE OF CITRUS.................... 166

C INITIAL SOIL CONDITIONS OF EXPERIMENTAL SITE, DECEMBER 2001
(SOIL AN ALY SES RESULTS)......................................... .......................... 168

D ANALYSES OF VARIANCE FOR ANNUAL AND PERENNIAL COVER
CROPS ON SOIL AND CITRUS TREE CHARACTERISTICS, CITRUS TREE
ROW GROUND COVER, AND CITRUS YIELD AND QUALITY ..................... 169

L IST O F R E FE R E N C E S ........................................................................ ................... 173

BIOGRAPHICAL SKETCH ............................................................. ............... 191















LIST OF TABLES


Table p


2.1 Overview of annual summer and winter cover crops used during the 2002 and
2003 grow ing seasons. ....................................... ......................... 39

2.2 Overview of annual summer and winter cover crops used during the 2004 and
2005 grow ing seasons. ........................ .................... ............................ 40

2.3 Overview of seeding rates, space between rows and cultivars used as annual
summer and winter cover crops used from 2002 to 2005. .....................................41

2.4 Outline of planting and harvest dates and duration of summer and winter cover
crop s. ............................................................................... 42

2.5 Outline of cover crop weed index (CCWI) categories.........................................42

2.6 Rainfall measured at Plant Research and Education Unit (Citra) Florida
Automated Weather station Net work (FAWN)1 during the 2002-2005 summer
C C g row in g season n ......................................................................... ................ .. 4 3

2.7 Shoot dry weight accumulation (DW); shoot N concentration (Nconc); shoot N
accumulation (Naccum), maximum observed leaf area index (LAI max) for
summer cover crops grown during the 2002 growing season................................44

2.8 Shoot dry weight accumulation (DW); shoot N concentration (Nconc); shoot N
accumulation (Naccum), maximum observed leaf area index (LAI max) for
summer cover crops grown during the 2003 growing season..............................45

2.9 Shoot dry weight accumulation (DW); shoot N concentration (Nconc); shoot N
accumulation (Naccum); weed dry weight accumulation (DW) for summer
cover crops during the 2004 growing season. ................. ............................... 46

2.10 Shoot dry weight accumulation (DW); shoot N concentration (Nconc); shoot N
accumulation (Naccum); weed dry weight accumulation (DW) for summer
cover crops during the 2005 growing season. ................. ............................... 47

2.11 Rainfall measured at Plant Research and Education Unit (Citra) Florida
Automated Weather station Net work (FAWN)1 during the 2002-2005 winter
C C grow ing season ............ ... .......................................................... ........ .. ...... .. 4 8









2.12 Shoot dry weight accumulation (DW); shoot N concentration (Nconc); shoot N
accumulation (Naccum); maximum observed leaf area index (LAI max) for
winter cover crops during the 2002-2003 growing season. .................................48

2.13 Shoot dry weight accumulation (DW); shoot N concentration (Nconc); shoot N
accumulation (Naccum); weed dry weight accumulation (DW) for winter cover
crops during the 2003/2004 growing season ........... .............................................49

2.14 Shoot dry weight accumulation (DW); shoot N concentration (Nconc); shoot N
accumulation (Naccum); weed dry weight accumulation (DW) for winter cover
crops during the 2004/2005 growing season ........... .............................................50

3.1 Overview of experimental treatments during 2002-2005. ....................................78

3.2 Overview of seeding rates and row spacing for annual summer and winter cover
crops used betw een 2002 and 2005 ....................................................................... 79

3.3 Outline of planting and harvest dates and duration for summer and winter cover
crop s. ............................................................................... 79

3.4 Outline of cover crop weed index (CCWI) categories.........................................80

3.5 Rainfall measured in the Plant Science Research and Education Unit (Citra)1
during 2002-2005. ........................... ........... ........ .......... .... .... 80

3.6 Effect of planting time and over-seeding of perennial peanut (PP) on shoot
number m-2 (Shoot#), leaf area index (LAIpp), shoot dry weight (DWpp), and N
accumulation (Nacc-pp); weed dry weight (DWWD), N accumulation in weeds
(Nacc-wD), and cover crop weed index (CCWI) in 2002......................................81

3.7 Effect of planting time and over-seeding of perennial peanut (PP) on shoot
number m-2 (Shoot#), leaf area index (LAIpp), shoot dry weight (DWpp), and N
accumulation (Nacc-pp); weed dry weight (DWWD), N accumulation in weeds
(Nacc-wD), and cover crop weed index (CCWI) in 2003......................................82

3.8 Effect of planting time and over-seeding of perennial peanut (PP) on shoot
number m-2 (Shoot#), leaf area index (LAIpp), shoot dry weight (DWpp), and N
accumulation (Nacc-pp); weed dry weight (DWWD), N accumulation in weeds
(Nacc-wD), and cover crop weed index (CCWI) in 2004......................................83

3.9 Effect of planting time and over-seeding of perennial peanut (PP) on shoot
number m-2 (Shoot#), leaf area index (LAIpp), shoot dry weight (DWpp), and N
accumulation (Nacc-pp); weed dry weight (DWWD), N accumulation in weeds
(Nacc-wD), and cover crop weed index (CCWI) in 2005......................................84

3.10 Effect of planting season date and over seeding on perennial peanut (PP), weeds,
and system (PP+weed) dry weight, N accumulation in PP, weeds and in the
system in 2005 .................................... ................................ .........85









3.11 Total dry weight (DW) in the system (CC+weeds), corresponding percentage of
total dry weight in CC (% DW CC), total N accumulation (Total Nacc) in the
system (CC and weeds) and corresponding percentage of N in CC (%N in CC)
in 2005 ................................................................. ........ ........... 86

4.1 Effect of cropping system on citrus, bermudagrass, and perennial peanut (PP) N
and water uptake for three different seasons during 2005.................................... 114

4.2 Effect of cropping system on citrus, bermudagrass, and perennial peanut (PP) N
and water uptake for two different growth cycles during 2005 .............................115

4.3 Comparison of effect of cropping system on citrus, bermudagrass (BG), and
perennial peanut (PP) N uptake at the final harvest (end of the growing period)
using 15N and SUM techniques. ................. ......................................116

4.4 Nitrogen accumulation by citrus and ground covers based on 15N results ...........116

4.5 Overview of parameters for N uptake regression model............... .................. 117

4.6 Effect of cropping system on bermudagrass and perennial peanut (PP) shoot dry
weights (DW), nitrogen concentration (Nconc) and nitrogen accumulation
(Naccum) for different growing seasons during 2004 and 2005 ............................118

4.7 Effect of cropping system on citrus root dry weight (DW), root length, stem dry
weight, diameter (Diam), leaf dry weight, leaf area (LA), total dry weight, root
nitrogen accumulation (Naccum), stem N accumulation, leafN accumulation,
and total N accumulation at the end of the growing season................................ 119

4.8 Effect of cropping system on bermudagrass (BG) and perennial peanut (PP) root
dry weight (DW), root length, shoot dry weight (DW), leaf area (LA), root
nitrogen concentration (Nconc), shoot nitrogen concentration, root nitrogen
accumulation (Naccum), shoot N accumulation, and total N accumulation at the
end of the grow ing season. ............................................................................. ..... 119

4.9 Percentage of N distribution in different tissues for the diverse cropping
sy stem s. ......................................................................... 12 0

5.1 Effect of year, season, location, and treatments on soil pH for the perennial
cover crop study during 2003-2005. ........................................... ............... 145

5.2 Effect of year, season, location, and treatments on soil pH for the annual cover
crop study during 2003-2005. ..........................................................................146

5.3 Effect of year, location, and treatment on soil C, N, and C:N for the perennial
cover crop study during 2003-2005. ........................................... ............... 147

5.4 Effect of year, location, and treatment on soil C, N, and C:N ratio for the annual
cover crop study during 2003-2005. ........................................... ............... 148









5.5 Number of plant-parasitic nematode for the perennial cover crop study during
2004 and 2005 .........................................................................149

5.6 Number of plant-parasitic nematode for the annual cover crop study during 2004
an d 2 0 0 5. .......................................................................... 150

5.7 Percentages of ground cover in the tree row for most commonly observed weed
species as affected by year and season in tree rows for the perennial cover crop
study during 2003-2005. ............................................... ............................. 151

5.8 Effect of year, season, and treatments on tree height and trunk diameters for
'Hamlin' oranges (perennial cover crop study) during 2002-2005.....................152

5.9 Effect of year, season, and treatments on tree height and trunk diameters for
'Navel' oranges (annual cover crop study) during 2003-2005.............................. 153

5.10 Effect of cover crop treatment on citrus yield and fruit quality (degree Brix and
acidity) for the perennial cover crop study during 2005. .......................................154

A. 1 Analyses of variance for perennial peanut (PP) shoot dry weight (DWpp), N
accumulation in PP shoots (Nacc-pp), PP leaf area index (LAIpp), number of PP
shoots per square meter (shoot#), Weed dry weight (DWWD), N accumulation in
weeds (Nacc-wD), and Cover crop weed index (CCWI). .......................................165

B. 1 Analyses of variance for the effect of ground covers on N and water uptake. ......166

B.2 Analyses of variance for the effect of cropping system on bermuda grass and
perennial peanut shoot dry weight (DW), nitrogen concentration (Nconc) and
nitrogen accumulation (Naccum). ............................................... ............... 167

D. 1 Analyses of variance for the effect of perennial cover crops (PCC) and annual
cover crops (A CC) on soil pH ....................................................... .............. 169

D.2 Analyses of variance for the effect of perennial cover crops on soil carbon (C),
nitrogen (N ) and C :N ratio. ............................................ ............................ 170

D.3 Analyses of variance for the effect of annual cover crops on soil carbon (C),
nitrogen (N ) and C :N ratio. ............................................ ............................ 170

D.4 Analyses of variance for the effect of perennial cover crops (PCC) and annual
cover crops (ACC) on soil nematode populations. .............................................. 171

D.5 Analyses of variance for the effect of perennial cover crops on tree-row cover.... 171

D.6 Analyses of variance for the effect of perennial cover crops on citrus tree height
(H eight) and diam eter (D iam ). ......................................................................... 172









D.7 Analyses of variance for the effect of annual cover crops on citrus tree height
(Height) and diameter (Diam). ............................................................................ 172

D.8 Analyses of variance for the effect of perennial cover crops (PCC) and annual
cover crops (ACC) on nitrogen citrus leaf concentration.............. ............... 172















LIST OF FIGURES


Figure page

2.1 Leaf area index values for summer cover crops 2002 (CP= cowpea; VB= velvet
bean; SH= sunnhemp; AC= Alyceclover; HI= hairy indigo)...............................51

2.2 Leaf area development for winter cover crops during 2002/2003...........................51

3.1 Dry matter of perennial peanut (PP) over time. ....................................... .......... 87

3.2 Dry m atter of w eeds across the years .................................... .......... ............... 88

3.3 Cover Crop Weed Index (CCWI) for perennial peanut across the years ...............89

3.4 Regression between PP dry weight (DWpp) and weed dry weight during spring
(SPDww), summer (SUDww), and fall (FALLDww) for all PP treatments.................90

3.5 Regression between PP dry weight (DW) and N accumulation in weeds during
spring (SPN-w), summer (SUN-w), and fall (FALLN-w) for all PP treatments..........90

4.1 Overview of soil-N uptake monitoring (SUM) system.......................................121

4.2 Minima, maxima, and soil average temperature during the experimental period..122

4.3 Solar radiation in the greenhouse during the experimental period.........................122

4.4 Regression between SUM-based N uptake and 15N based N uptake ...................123

4.5 Nitrogen uptake dynamics for different cropping systems across time. ..............124

4.6 Nitrogen uptake as a function of cumulative uptake temperature during 14-day
pre-clipped vs. post-clipped uptake period for bermudagrass mono-crop ...........125

4.7 Nitrogen uptake as a function of cumulative radiation during the 14-day
preclipped vs. post-clipped uptake period for bermudagrass mono-crop. .............125

C.1 Initial soil conditions at the experimental site in December 2001 (soil analyses
results from Analytical Research Lab. IFAS, Gainesville, FL.) ..........................168















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

INTEGRATIVE USE OF PERENNIAL AND ANNUAL COVER CROPS FOR WEED
MANAGEMENT IN ORGANIC CITRUS

By

Jose Clemente Linares B.

August, 2006

Chair: Johannes M. Scholberg
Major Department: Agronomy

Citrus is one of the most important crops in Florida. During the past decade

increased international competition and urban developments, diseases, and more stringent

environmental regulations have greatly affected the citrus industry. Citrus growers

transitioning to organic production may benefit from premium prices, but they also face

many challenges, including development of efficient weed management strategies. Cover

crops (CC) may constitute an environmentally sound alternative for improved weed

management in organic systems. Two field experiments were conducted at Citra in North

Central Florida, to test performance and the effectiveness of annual and perennial CC to

suppress weeds in organic groves. A greenhouse trial was also implemented to evaluate

potential competition between citrus and groundcovers. For annual CC, summer CC had

the greatest biomass and N accumulation in comparison with winter CC. Sunnhemp,

hairy indigo, cowpea, and alyceclover provided excellent weed suppression, which was

superior to tillage fallow. Mono-cropped winter CC did not always perform consistently









well. Use of winter CC mixtures resulted in more consistent overall CC performance,

greater dry matter production, N accumulation, and more effective weed suppression. In

both annual and perennial systems, weeds played a complementary role in nutrient

retention and recycling. Perennial peanut (PP) showed slow initial growth and summer

planting of PP was the most successful compared with spring planting. Over-seeding PP

planted in summer with crimson clover reduced PP growth and its effectiveness in

suppressing weeds. Initial weed suppression by PP was very poor to poor; however,

effectiveness of PP to reduce weed growth improved gradually over time. Annual CC

provided much better weed control than PP. For both PP and annual CC, weed biomass

typically was inversely related to CC DW accumulation due to competition for resources.

In a greenhouse experiment, citrus and bermudagrass appeared to compete for N during

summer, while citrus and PP did not compete. Citrus, bermudagrass, and PP competed

for water uptake during the spring and summer seasons. In general, perennial and annual

CC treatments did not affect soil pH, C, N, and C:N ratio during the initial 3 years of

field studies. Nematode populations did not reach high levels. Cover crop treatments in

row middles did not affect weed growth dynamics in the tree row. However, planting

cowpea in the tree row did decrease bahiagrass and crabgrass populations in tree rows.

Planting tall cover crops such as sunnhemp near young citrus trees reduced initial tree

growth. Cover crop treatments did not affect citrus leaf N, fruit yield, and quality during

initial growth. Additional research is needed to assess long-term effects of cover crops on

soil quality and fruit yield.














CHAPTER 1
INTRODUCTION

Florida is the largest citrus producing state in the U.S. and accounted for 74% of

the U.S citrus production during the 2004-2005 season, with 302,929 ha bearing fruit

(Florida Agricultural Statistics Service, 2006). Although citrus accounts for 49% of the

total certified organic farm area in Florida, less than 1% of Florida citrus is currently

grown organically (Athearn, 2004).

Total value of Florida citrus during 2004-2005 was US $742 million, which was

the lowest since 1985-86 (Florida Agricultural Statistics Service, 2006). This decline was

related to a reduction in citrus consumption (especially of orange juice) in the U.S.A. due

to the popularity of low carbohydrate diets, increased international competition, and

relatively low on-tree citrus prices. Recent pest and disease outbreaks, competition with

residential development for land and water resources, along with more stringent

environmental regulations pose tremendous challenges for conventional citrus growers

(Athearn, 2004).

In contrast, organic agriculture is the fastest increasing segment of US agriculture.

Organic sales have increased by 20% annually since 1990 (Dimitri and Greene, 2002),

and retail sales in the U.S were estimated to be on the order of $17 billion during 2005.

The leading revenue source for the US organic food market is the fruit and vegetable

sector, which generated total revenues of $7 billion in 2005, which correspond to 41% of

the overall organic market. Although it is projected that organic sales will increase to $35









billion by the end of 2010, organic production still accounts for less than 5% of US

agricultural area for most commodities (Datamonitor, 2005).

In Florida, 85% of the wells that exceeded the maximum contaminant level (MCL)

for NO3-N were located in conventional citrus production areas (McNeal et al., 1995).

Similarly, some citrus-applied herbicides have been discovered in both ground and

surface water (Troiano and Garretson, 1998; Liu and O'Connell, 2002). Since excessive

use of non-renewable resources and/or potentially harmful agrochemicals may impact

biodiversity, environmental quality, food safety, and health of farmers, there is increased

interest in more sustainable production systems including organic farming (Reganold et

al., 2001). Organic production may not only protect natural resources and the

environment but also meets current consumer's health concerns and food safety

requirements (Igual et al., 2002). Conversion to organic production systems may also

allow growers to benefit from marketing niches and grower-friendly price mechanisms

associated with organic production (Athearn, 2004). By reducing regional pesticide and

fertilizer use, organic production can preserve both groundwater resources and fragile

ecosystems for future generations (Mader et al., 2002).

Organic agriculture relies on ecologically based principles and practices, such as

nutrient recycling, increased biodiversity, and biological pest management. It aims to

achieve more autonomous management of local agroecosystems and to enhance the

resilience of such systems by increasing reliance on local resources, biodiversity, and

synergistic biological interactions. In this manner local production capacity of the soil

can be sustained (Altieri, 1995; Gliessman, 1998). However, this requires the adoption of

alternative techniques to enhance both crop and soil health, including improved use of









cover crops to suppress weeds, prevent erosion, and restore soil organic matter (Nelson,

2004).

Results from a 21-year study of agronomic and ecological performance of organic,

biodynamic, and conventional farming systems in central Europe were reported by Mader

et al. (2002). They found that although crop yields were 20% lower in the organic

systems, input of fertilizer and energy was reduced by 34 to 53% and pesticide input by

97%, while soil fertility and biodiversity were enhanced and the use of external inputs

was being reduced. The authors concluded that organically-managed legume-based crop

rotations utilizing organic fertilizers from the farm itself provided a viable alternative to

conventional farming systems.

Organic citrus production emerged as a commercial sector in Florida during the

early 1990. A 1993 survey identified 16 organic citrus growers covering 230 ha (Swisher

et al., 1994). This acreage was increased to 2,400 ha in 2004-2005, while the number of

organic citrus growers increased to 39 (Athearn, 2004).

However, despite the rapid growth of organic agriculture, information pertaining to

organic production in general and organic citrus in particular is scarce. As a result, the

organic grower community requested that the United States Department of Agriculture

(USDA) to create special programs and providing grant funding for this research was one

of the first initiatives. Research priority areas ranged from development of weed

management practices during the transition from conventional to organic farming to

economic research on markets and profitability of organic farming systems (SCOAR,

2003).









In response to the growing interest in organic agriculture in the U.S. and the

implementation of the National Organic Program in 2002, USDA created the Sustainable

Agriculture Research and Education Program (SARE), which looks at both conventional

and organic systems. Previously, the USDA had initiated a sustainable agriculture

research and education program in 1988 and this program was originally referred to as

the Low-Input Sustainable Agriculture (LISA) program (SARE, 2006). Other non-

profitable organizations focusing on organic includes the Organic Farming Research

Foundation which was founded in 1990 by certified organic farmers and provides funding

to a limited number of research programs to address critical issues in organic agriculture

(Organic Farming Research Foundation, 2006).

Conversion from conventional to organic production will result in appreciable

modification in agroecosystem management (Ngouajio and McGiffen, 2002). Weed

suppression is one of the most important components to be considered during this

conversion process since important changes in weed population dynamics occur, which

will require implementation of alternative weed management strategies (Lanini et al.,

1994; Liebman and Davis, 2000).

Weed control in conventionally grown citrus accounts for 27% of annual

production costs. In organic citrus groves, weed management accounts for over 30% of

annual production costs and the majority of the labor costs (Muraro et al., 2003; Athearn,

2004). A national survey showed that the number one research priority for organic

growers was improved weed management (Sooby, 2003). Florida growers were no

exception to this finding, and both Florida organic citrus growers and grower









organizations emphasized that weed control was the most critical factor for growers to be

successful during the transition to organic production (SCOAR, 2003).

One ecological approach for weed management includes improved soil coverage

through use of cover crops (CC) and/or mulches. However, in the absence of appropriate

recommendations, lack of effective weed management practices pertinent to organic

systems may hamper successful transition from conventional to organic citrus production

(Sooby, 2003; SCOAR, 2003).

In interviews with Florida citrus growers, the majority expressed a strong interest in

the use of cover crops including perennial peanut (Arachis glabrata Benth.) to prevent

soil degradation and suppress weed growth (Scholberg, unpublished). Perennial peanut

(PP) may provide an environmentally sound and ecologically important component of

sustainable citrus production in Florida, since it does not require pesticides or N fertilizer

(French et al., 1994; Mullahey et al., 1994). Due to its low water and nutrient

requirements, perennial peanut fits the model of sustainable production. In contrast to

annual cover crops, it needs to be planted only once and it may reduce labor costs

associated with weed control in citrus. It may also provide 67 to 112 kg of N ha-1 yr-1 to

citrus trees among many other benefits (French et al., 1994; Woodward et al., 2002).

During the past decade, perennial peanut has been evaluated as a groundcover for

vegetable crops (Roe et al., 1994) and citrus (Coleman, 1995). Several citrus growers

have also successfully integrated this cover crop into their production system. However,

current practices for establishment of perennial peanut were typically developed for

conventional forage production and are not appropriate for organic citrus. Similarly,

although perennial peanut use as a cover crop in conventional systems in South Florida









has been studied extensively (Roe et al., 1994; Mullahey et al., 1994), no detailed

information is available regarding its establishment in organic citrus production systems.

Support for the research outlined in this dissertation was provided by the USDA

Organic Transition Program via a grant entitled "Integrative Use of Perennial Peanut for

Cost-Effective Weed Control in Organic Citrus". Originally, the main focus of this

project was on the use of perennial peanut in organic citrus groves. However, based on

comments of panel reviewers the scope of this program was extended to also include a

study focusing on both annual winter and summer cover crops. These studies were

intended to evaluate the effect of cover crops on weed suppression, soil quality, and citrus

tree growth for a newly-planted certified organic citrus production system.

The general objectives of this work were to 1) determine growth characteristics of

annual and perennial cover crops in organic citrus groves; 2) evaluate changes in weed

growth as affected by annual and perennial cover crop treatments; 3) quantify the effect

of perennial peanut and/or common bermudagrass on citrus N and water uptake under

controlled conditions; and 4) assess how cover crop treatments affect soil quality, tree

height and diameter, leaf N, fruit yield, and fruit quality.

The corresponding hypotheses were as follows: 1) annual CC will suppress weeds

effectively and summer CC will accumulate more biomass and consequently will

suppress weeds more effectively compared to winter CC; 2) in organic citrus systems,

planting PP during the summer will increase the competitiveness of PP systems via

enhanced initial growth compared to spring plantings and overseeding PP with crimson

clover in fall will help to increase the PP effectiveness in suppressing weeds; 3) weed

suppression with annual CC will be more effective than perennial peanut; 4) citrus,









perennial peanut, and common bermudagrass will differentially compete for nitrogen and

water uptake and competition for N and water uptake will be most evident during periods

of high demand; 5) annual and perennial CC will increase soil quality over time and

reduce pest nematode populations in organic citrus fields; and 6) cover crop treatments in

the row middles will also affect weed growth in the tree rows; and cover crops will not

affect significantly citrus growth characteristics.

This dissertation includes six chapters. Although each chapter forms a separate

entity they are also intrinsically linked. The current (first) chapter provides a conceptual

framework for this dissertation and includes a brief historic perspective of this work. It

outlines program objectives and hypotheses while in the following part an outline of

consecutive chapters is provided to emphasize the inner-connectivity among subsequent

chapters.

The second chapter outlines the performance of different winter and summer cover

crops in a recently established organic citrus orchard under Florida conditions and their

effectiveness in suppressing weed growth. The use of the cover crop weed index (CCWI),

which is the ratio of cover crop dry weight (CCDw) to weed dry weight (WeedDw)

associated with a specific cover crop (CCWI=CCDw/WeedDw), will be discussed along

with the use of this index for improved assessment of the effectiveness of different cover

crops to suppress weeds.

The third chapter evaluates initial establishment, growth dynamics, weed

suppression capacity, and productivity of perennial peanut in a recently established

certified organic citrus grove. Treatments included date of planting, association with

winter annual cover crops (over-seeding perennial peanut with crimson clover in fall),









and a system with annual cover crops only (which provided a linkage with the previous

chapter).

After successful establishment of perennial peanut as a groundcover for organic

citrus groves, perennial peanut plants may also expand into the tree rows and potentially

compete with citrus trees for water and nutrients. Therefore, competition between citrus

and perennial peanut for water and nutrients was studied under controlled conditions as

outlined in the fourth chapter. In this chapter, the water and N uptake dynamics for pure

and mixed systems of perennial peanut, weeds (bermudagrass) and citrus, and potential

competition for water and nitrogen uptake are presented.

Besides the effects of cover crops on weed suppression in organic citrus groves,

cover crops also have potential effects on soil chemical, physical, and biological

properties, including soil organic matter, soil nitrogen, pH, and nematode populations.

Some of these and/or a combination of these parameters may be used as an indicator of

"soil health", system sustainability, and potential "suppressiveness capacity" of these

soils. In the fifth chapter we summarize how cover crops affect some of these parameters

and also overall tree growth and initial production. In the last chapter we synthesize and

summarize previous chapters and also outline future research priorities and practical

implementation for growers.














CHAPTER 2
GROWTH AND EFFECTS OF ANNUAL COVER CROPS ON WEED SUPPRESSION

Introduction

Cover crops (CC) are herbaceous plants, annuals or perennials (usually grasses or

legumes) grown in pure or mixed stands to cover the soil during part of one or more

cropping cycles. The CC may be incorporated into the soil by tillage in seasonal CC

systems, or retained as live or dead plants on the soil surface for several seasons

(Gliessman, 1998).

Cover crops may suppress weeds by either removal of resources (Ngouajio and

Mennan, 2005; Ross et al., 2001) or by increased soil microbial diversity (Kremer and Li,

2003; Kennedy, 1998), or may inhibit weeds via allelopathy (Reberg-Horton et al., 2005;

Fennimore and Jackson, 2003). Also, CC can suppress soil pathogens by allelochemicals

(Bailey and Lazarovits, 2003) or increased presence of beneficial organisms that suppress

pest organisms such as nematodes (Reddy et al., 1986b; McSorley, 2001; Macchia et al.,

2003). In addition, use of CC enhances soil biological and chemical properties by

promoting the creation of cooler and moister soil surface and subsurface habitats (Kremer

and Li, 2003).

Cover crops enhance soil fertility via improved nutrient cycling and nitrogen

fixation by legumes CC (Ramos et al., 2001; Perin et al., 2004), carbon sequestration

(Sainju et al., 2003), and increased nutrient retention by roots (Vos and van der Putten,

2001; Kristensen and Thorup-Kristensen, 2004).









Cover crops shield the soil surface from sunlight, wind, and the physical impact of

raindrops thereby reducing soil erosion and soil organic matter losses (Sainju et al.,

2003). Cover crops also increase biological activity in the root zone and thereby enhance

the formation of more stable soil aggregates and macropores while reducing soil

compaction and soil bulk density. As a result, they can result in improved soil structure

(Kremer and Li, 2003), water infiltration, and root penetration (Justes et al., 1999), while

reducing soil crust formation, runoff, and soil erosion (Gliessmann, 1998).

Although CC may provide many environmental and agronomic benefits, there also

may be negative effects of using CC in a cropping system. Establishment costs may be

cost-prohibitive, thus hampering their use in resource-limited production systems.

Residues or breakdown products of incorporated CC may produce growth-suppressing

(allelopathic) substances that impact the growth of commercial crops. Damaging

herbivores or disease organisms may find CC to be suitable alternate hosts before moving

on to the subsequent main crops. The CC residue may also interfere with cultivation,

weeding, harvesting, and/or other farming activities. Some CC when used as an intercrop

or live mulch may be excessively tall and shade the commercial crops (Reeves, 1994;

Gliessman, 1998).

Cover crops can be classed as annually seeded winter-growing grasses and

legumes, reseeding winter annual grasses and legumes, summer annuals, perennial

grasses and legumes, and other cover crop plants (Altieri, 1995). Advantages of annual

summer CC include rapid initial growth and effective competition with weeds thus cover

crops should perform well within organic systems. On the other hand, annual CC

typically need to be replanted. During the first yearss, poor inoculation may also hamper









the growth of leguminous CC (Slattery et al. 2001; Carsky et al., 2001; Tian and Kang,

1998). Many leguminous CC, especially winter CC, are more demanding and require a

suitable pH and supplemental nutrients (mainly P, S, K, and Mo) to ensure adequate

nodulation, N fixation, and biomass accumulation (Slattery et al., 2001). Therefore, they

may be poorly adapted to the sandy, shallow, and low fertility soils prevailing in most

citrus production areas in Florida.

A number of studies have shown that some of the annual CC have recalcitrant

seeds with hard coats which can become part of the soil seed bank and thus may re-

establish themselves naturally (Benech-Arnold et al., 2000). When used in vegetable

crops, use of "hard-seeded" CC may not be desirable since they promote weedinesss",

which can complicate CC management (Bond and Grundy, 2001). However, in mature

citrus orchards, a natural seed-bank of selected leguminous CC may be desirable. In this

case, different combinations of species may proliferate each year, which may result in a

natural selection of plants best-suited for prevailing soil and/or current climatic

conditions, while also eliminating the annual cost for CC re-establishment.

Weeds in agroecosystems are known to compete with crops for water, nutrients,

and light. They are potential hosts for pests and diseases and can also interfere with soil

tillage, irrigation, and harvest operations (Liebman and Davis, 2000). As a result, they

increase labor requirements and production costs. On the other hand, certain weeds may

form important components of the agroecosystem because they provide alternative food

sources such as pollen, nectar, foliage, or prey for beneficial insects (Altieri and

Letourneau, 1982; Chacon and Gliessman, 1982).









Weed control in conventionally grown citrus accounts for 27% of annual

production costs (Muraro et al., 2003). Weed control programs include the application of

pre-emergence and post-emergence herbicides within tree rows especially for young

trees, chemical control with herbicides like Roundup in the drive middles between rows,

mowing and disk harrowing (Futch, 2005). For organic citrus groves, weed control

accounts for over 30% of annual production costs and the majority of the labor costs

(Muraro et al., 2003). It includes disking, mowing, and hand labor to remove vines

growing into tree canopies and/or weeds near tree trunks of young trees. There are some

other strategies for weed management like improved soil coverage through use of

mulches and/or appropriate use of CC.

Conversion from conventional to organic production will result in appreciable

modifications in agroecosystem management. Appropriate weed management is one of

the most challenging components during this conversion because of the important

changes in weed population dynamics, which will require implementation of alternative

weed management strategies (Bond and Grundy, 2001; Lanini et al., 1994). A national

survey showed that the number one research priority for organic growers was improved

weed management (Sooby, 2003). Based on interviews with Florida citrus growers

(Scholberg, unpublished), the majority expressed a strong interest in the use of cover

crops to prevent soil degradation and suppress weed growth.

Successful weed suppression using summer CC has been reported for annual crops

such as rice (Oryza sativa L.) preceded by pigeon pea (Cajanus cajan L.) as a CC (Roder

et al., 1998), lettuce (Lactuca sativa L.) planted after cowpea (Vigna unguiculata L.

Walp.) or sorghum-sudan grass (Sorghum bicolor L. Moench) (Ngouajio and Mennan,









2003), and corn (Zea mays L.) by using velvet bean (Mucuna atropurpureum L.)

(Buckles and Triomphe, 1999; Caamal-Maldonado et al., 2001). Using annual winter CC

such as rye (Secale cereale L.) (Fennimore and Jackson, 2003) or crimson clover

(Trifolium incarnatum L.) and subterranean clover (Trifolium subterraneum L.) (Barberi

and Mazzoncini, 2001) may provide adequate aposteriori weed control in corn. Equally

suitable weed suppression has been reported in strawberries (Fragaria x aananssa

Duchesne.) employing winter rye and wheat (Triticum aestivum L.) (Whitworth, 1995).

Similarly in field pea (Pisum sativum L. spp. arvense (L.) Poir) weeds were suppressed

when it was preceded by sweet clover (Melilotus officinalis L.) (Blackshaw et al., 2001).

Cover crops have been useful in suppressing weeds in perennial production systems

as well. Bradshaw and Lanini (1995) obtained acceptable weed control in coffee by using

Desmodium ovalifolium, Commelina difusa, and Arachis pintoi. Adequate weed

suppression was obtained through use of winter rye as a CC around several forest and

ornamental tree species, but tree growth reduction occurred with sod CC treatments

(Calkins and Swanson, 1995).

Although the use of perennial and annual CC in conventional vegetables and

perennial CC in citrus systems in south Florida has been studied extensively (Rouse and

Mullahey, 1997; Coleman 1995; Roe et al., 1994; Mullahey et al., 1994), no information

is available regarding the effectiveness of annual CC in suppressing weeds in organic

citrus production systems.

The overall objectives of this study were to 1) determine growth characteristics of

annual cover crops in organic citrus groves; 2) determine changes in weed growth as

affected by annual cover crop treatments; 3) identify suitable cover crop species and









evaluate their effectiveness in suppressing weeds in organic citrus groves; and 4) develop

optimal cover crop associations suited for organic citrus production.

The following hypotheses were tested in organic citrus systems: 1) annual CC will

suppress weeds effectively; 2) summer CC will accumulate more biomass and

consequently will suppress weeds better than winter CC; and 3) weed suppression by CC

will be related to their adaptation to environmental conditions.

Materials and Methods

Set-up and Experimental Design

A one-hectare block was planted with 'navel' orange, a fresh market orange variety

[Citrus sinensis (L.) Osb.cv. Navel] grafted on Swingle Citrumelo (C. paradisi Macf. x

P. trjfoliata (L.) Raf.) during the spring of 2003 at the Plant Science Research and

Education Unit in Citra, Florida (29.68 N, 82.35 W). Tree spacing was 4.6 m in the row

and 6.1 m between rows. The main emphasis of this study was the evaluation of annual

CC used as ground cover to suppress weeds in the strips between tree rows (row

middles).

Prevailing soil types at the experimental site were a Candler fine sand (Typic

Quarzipsamments, hyperthermic, uncoated, 98% sand in the upper 15 cm) and a Tavares

fine sand (Typic Quarzipsamments, hyperthermic,uncoated, 97% sand in the upper 15

cm). The initial soil pH ranged from 4.8 to 5.1 and soil organic matter content was 9.3 g

kg-1. At the beginning of the experiment (Fall of 2001), soil was prepared by disking

followed by repeated rototilling. During the spring of 2002, both lime (2.5 Mg ha-1) and

chicken manure litter (2.5 Mg ha-1) were applied to the entire production block. During

subsequent years, chicken manure was applied exclusively to a 1.8-m wide strip

straddling the tree rows. Manure was applied during early spring at 4-11 Mg ha1.









Manure application rates were based on IFAS (Institute of Food and Agricultural

Sciences) N-recommendations for newly-planted trees and based on estimated N

mineralization and N concentration following N recommendations for newly-planted

trees (Tucker et al., 1995). To enhance growth of winter cover crops, a non-synthetic

(mined) K2SO4 (SQM North America Corp., Atlanta, GA) approved by the Organic

Materials Review Institute (OMRI) was broadcast over the entire area at a rate of 45 kg

K20 ha-1 prior to planting of winter CC. Due to a buildup of residual soil P, use of

chicken manure was discontinued after 2004, and starting in 2005, an OMRI-approved

natural fertilizer derived from feather-meal and potassium sulfate (Nature Safe, Griffin

Industries, Cold Spring, KY) with 9-0-9 (N, P205, K20) was applied to tree rows using

standard recommendations (Tucker et al., 1995).

Trees were irrigated with microjet sprinklers with a 1.8-m spray diameter and a

1800 spray pattern placed 0.4 m NW of the trees. During the winter the irrigation

sprinklers were placed inside 0.6 m high PVC pipes and also used for frost protection if

temperatures dropped below -2 C. Row middles and cover crops were not irrigated in

order to evaluate the suitability of different species for typical citrus orchard conditions.

Prior to the planting of the orange trees, one cropping cycle of both summer and winter

cover crops was completed for initial screening of suitable cover crops.

Cover crop treatments are outlined in Tables 2.1 (2002 and 2003) and 2.2 (2004

and 2005). Each CC treatment plot consisted of a total area of 6.1 m x 27 m straddling a

row of 5 citrus trees. However, Treatment 1 (a mixture of cowpea and sunnhemp)

required larger plots, so in this case, a total of three rows of 6 trees were used. Cover crop

treatments were arranged using a randomized complete block design with four replicated









blocks, each containing all the different CC treatments. Different cover crops (summer

vs. winter CC) were planted twice a year. During 2002, only a grass fallow was used as

the control, while starting in 2003, a tillage fallow was also included as an experimental

treatment. After initial establishment, grass fallows and perennial peanut plots were

mowed at 3-4 wk intervals throughout the spring, summer, and fall, while tillage fallows

were tilled twice a year before CC planting. Annual CC were planted with a "zero-till"

planter (Sukup 2100, Sukup Manufacturing Company, Shefield, IA) using a suitable row

spacing and recommended planting rates as outlined in Table 2.3. Planting dates for both

summer and winter CC are presented in Table 2.4. Except for winter CC planted in 2003,

when a zero-tillage system was used, previous CC and/or weeds were soil-incorporated

with two to three passes of a rototiller using a tillage depth of 10 cm prior to planting

cover crops.

All leguminous CC were inoculated before planting with the appropriate strain of

rhizobium (Nitragin brand, Milwaukee WI). Inoculant and untreated seeds were obtained

from local seed companies since varieties from certified organic supply companies are

typically poorly adapted to Florida conditions. Cover crops were not irrigated during the

growing season, except after planting, if soil moisture was inadequate to ensure uniform

germination. In this case, 25 mm irrigation was applied uniformly to the entire block to

mimic a typical rainfall event. Insect pests in citrus were controlled when needed with

allowed products in organic production systems following the national organic program

standards (USDA, 2000).

Data Collection, Measurements, and Analysis

Representative sections of row middles of CC were sampled at monthly intervals to

evaluate above-ground CC and weed biomass using rectangular sampling frames with an









internal surface area of 0.22 m2. Sample areas were selected in such way that the selected

area closely matched the ground cover of both CC and weeds of the entire plot. Sampled

areas invariably included a mix of weeds and annual CC. Weed biomass was not

segregated into different weed species. Weeds were harvested at ground level while

for the CC, the corresponding root system was also excavated. During 2002 and 2003, a

more detailed growth analysis of annual CC was performed and above-ground biomass

was separated into stems and leaves. Roots were washed and cleaned to remove soil and

organic debris. Leaf area was determined using a Licor leaf area meter (LI-3000, Li-cor;

Lincoln, NE).

Groundcover of CC was determined using leaf area index (LAI) values. Above-

ground biomass of weeds and CC were determined by dry weight. In order to quantify the

effectiveness of annual CC to suppress weed growth, we developed a cover crop/weed

index (CCWI). This index consists of ratios of CC and weed biomass

(CCWI=CCDw/WeedDw) calculated in each repetition. The qualitative interpretation of

this index is defined in Table 2.5. During 2002, above-ground weed biomass was

determined only at the end of the annual cover crops cycle using representative 0.5-m2

plot areas. During the summer of 2003, weed above-ground biomass was determined at

monthly intervals. During 2004 and 2005, weed above-ground biomass was determined

at bimonthly intervals using representative 0.22-m2 plot areas. Roots, stems, and leaves of

annual CC and shoots of weeds were oven dried at 65 C for 72 hours until constant

weight and dry weights were recorded. Afterwards, shoots were ground in a Wiley mill

through a 1-mm screen, and a thoroughly mixed portion (ca. 4 g) was subsequently stored

in scintillation vial. Ground tissue was digested using a wet-acid Kjeldahl digestion









(Gallaher, et al., 1975). After digestion, samples were diluted, filtered, and analyzed for

total Kjeldahl N at the UF-IFAS Analytical Research Lab (University of Florida,

Gainesville, FL) using EPA method 351.2 (Jones and Case, 1991). Analysis of variance

was performed on all data using Proc GLM of SAS (SAS Inst. Inc., 2002). Means were

compared using the Duncan test (DMRT) with a p-value of 0.05.

Results

Summer Cover Crops

Total precipitation throughout the growing period of summer cover crops (from

June to October) varied between 426 mm in 2002 (drier year) to 1060 mm in 2004 (Table

2.6).

Summer 2002

Rainfall was relatively low during 2002, but since rainfall was evenly distributed

during the growing season, no obvious water stress occurred. During the summer of

2002, sunnhemp had the highest dry matter production, followed by hairy indigo and

cowpea, whereas velvet bean performed rather poorly (Table 2.7).

Tissue N concentrations were highest for velvet bean followed by hairy indigo and

cowpea. Sunnhemp and alyceclover had a relatively low N concentration, probably due to

their relatively high end-of-season stem fraction. Overall N accumulation was greatest for

sunnhemp followed by hairy indigo and cowpea whereas both velvet bean and

alyceclover accumulated relatively little N.

In terms of maximum canopy density as LAI, cowpea and hairy indigo had the

highest LAI at 6 and 10 weeks, respectively. Due to its rapid canopy closure, cowpea was

very effective in early weed suppression, yet its canopy started to thin within 6-10 weeks

(Fig. 4.1). Hairy indigo, on the other hand, had a slower initial canopy development, but









it retained its canopy longer compared to cowpea. Sunnhemp had intermediate canopy

development rates, LAI values, and relatively long persistence (Table 2.7). Both velvet

bean and alyceclover had sparse canopies.

Although velvet bean had larger leaves and more dense canopies, overall soil

coverage was relatively poor due to low plant populations. Alyceclover, on the other

hand, had high planting densities, but plants were short, leaves were very small, and

canopies were rather sparse. Sunnhemp grew up to 2.4 m high causing some shading of

the young citrus trees. Approximate heights for cowpea, hairy indigo, alyceclover, and

velvet bean were 0.3, 1.2, 0.3, and 0.3 m, respectively.

Weed suppression typically followed CC biomass production trends and sunnhemp

also had the highest CCWI value, translating to outstanding weed control, followed by

hairy indigo and cowpea, whereas velvet bean provided poor weed control (CCWI < 1),

which may be related to the use of a bushy genotype. Weed growth was reduced 86% by

both sunnhemp and cowpea and 83% by hairy indigo compared with the grass fallow. In

these cases, only between 10 and 20% of the soil area was covered with weeds, in

contrast to velvet bean which reduced weeds only by 18%.

Summer 2003

Rainfall distribution was relatively favorable in 2003 (Table 2.6), and sunnhemp,

cowpea, hairy indigo, lablab, and alyceclover grew well. Conversely, velvet bean

performed relatively poor due to uneven germination, resulting in low plant populations

and nodulation was also relatively poor. Annual peanut performed poorly which was

related to incidence of diseases. Perennial peanut had a very poor performance because of

its slow initial establishment and competition from bermudagrass (Cynodon dactylon L.),

which inhibited its initial establishment and growth.









Total dry matter production was again greatest for sunnhemp followed by hairy

indigo, and cowpea (Table 2.8). Overall shoot N concentrations were relatively high

during 2003 compared to 2002, which may be related to a build up of soil rhizobial

inoculant and more favorable rainfall conditions. Overall N accumulation was highest for

sunnhemp and hairy indigo.

In terms of canopy development, lablab, velvet bean and cowpea had the earliest

canopy closure, which contributed to effective early season weed suppression. Similar to

2002, growth of hairy indigo was initially slow and sunnhemp and hairy indigo

developed their maximum LAI values four weeks later than the other CC (Table 2.8).

Similar to 2002, a significant lower weed biomass occurred with alyceclover, 'Iron

Clay' cowpea, sunnhemp, and lablab, resulting in weed suppression of about 94% 85%,

83%, and 65%, compared to the grass fallow (control).These CC provided similar or

better weed control than rototilling, which in turn reduced weeds by 59%

compared to the grass fallow treatment. Nitrogen accumulation by weeds was greatest

in the mowed fallow and ranged from 3 to 47 kg N ha-1 for the other treatments, which is

much lower compared to N accumulation by superior leguminous CC such as sunnhemp,

hairy indigo, and cowpea.

Sunnhemp had the highest CCWI, followed by alyceclover and cowpea. Although

hairy indigo was a proficient biomass producer, its CCWI value was low due to the

proliferation of alyceclover that volunteered in the hairy indigo plots.

Summer 2004

Precipitation during 2004 was relatively high in comparison with the other years

(Table 2.6), which was related to four major hurricanes passing through Florida, two of

which resulted in rainfall intensities in excess of 100 mm day-1. Rainfall during the rest of









the growing season was relatively evenly distributed. During the summer of 2004, pigeon

pea and hairy indigo accumulated the greatest biomass (Table 2.9), and 'Iron Clay'

cowpea clearly outperformed 'Cream-40' a commercial cowpea variety. In 2004,

sunnhemp only accumulated 5.3 Mg ha-1 which was 46% less than the average dry matter

accumulated in 2002 and 2003 (11.1 Mg ha-1) due to infection by Verticillium sp., a soil-

borne pathogenic fungus.

Shoot N concentration was highest for velvet bean and 'Iron Clay' cowpea

followed by lablab, pigeon pea, and perennial peanut. Low N values were obtained for

sunnhemp because verticillium wilt resulted in premature leaf drop.

Overall N accumulation was greatest for pigeon pea followed by hairy indigo and

cowpea. As in previous years, a lower (P<0.05) weed biomass resulted with alyceclover

and 'Iron Clay' cowpea. Use of alyceclover, 'Iron Clay' cowpea, hairy indigo, and

pigeon pea resulted in weed suppressions of about 97%, 92%, 65%, and 53%,

respectively, compared to the grass fallow (control). Rototilling reduced weeds by 69% in

comparison with grass fallow, which was low compared to alyceclover and 'Iron Clay'

cowpea. Nitrogen accumulation in weeds was greatest in the plots with 'Cream 40'

cowpea and least for alyceclover.

High CCWI values for alyceclover and 'Iron Clay'cowpea indicated outstanding

weed control (Table 2.9). On the other hand, velvet bean, lablab, perennial peanut,

'Cream 40' cowpea, and mung bean provided relatively poor weed control.

Summer 2005

Shoot dry weight was greatest for sunnhemp, followed by hairy indigo and pigeon

pea (Table 2.10). The lower biomass obtained for sorghum-sudangrass resulted from the









low seeding rate used, because it was intended to only provide support for lablab and

velvet bean without competing with the leguminous CC.

Shoot N concentration was highest for velvet bean and perennial peanut followed

by lablab. Overall N accumulation was greatest for sunnhemp followed by hairy indigo

and pigeon pea. Several crops had relatively low N accumulation (< 20 kg N ha1).

Significant reductions in weed biomass observed for 'Iron Clay' cowpea,

sunnhemp, and hairy indigo, translated to weed suppressions of about 92%, 90%, and

78%, respectively, compared with the grass fallow (control). Rototilling reduced weed

biomass by 51% and was thus less effective compared to sunnhemp and 'Iron Clay'

cowpea. Nitrogen accumulation in weeds was greatest in the mowed fallow (Table 2.10).

'Iron Clay' cowpea had the highest CCWI followed by sunnhemp and hairy indigo,

which is indicative of outstanding weed control. Mixing of lablab with velvet bean and

sorghum-sudangrass provided an excellent weed control as well, but intercropping of

velvet bean with sorghum-sudangrass provided only moderate weed control.

Winter Cover Crops

Monthly total precipitation amounts during the growing season of winter CC are

shown in Table 2.11. Total precipitation throughout the growing period varied between

334 in 2003/04 to 472 mm in 2002/03 (Table 2.11).

Winter 2002/2003

Rainfall was relatively high during the 2002/03 winter growing season in

comparison with other years, but since rainfall was unevenly distributed during the

growing season, obvious water stress did occur. During the winter of 2002/03 lupin had

the highest dry matter production, whereas red, berseem, and sweet clover performed

rather poorly (Table 2.12).









Tissue N concentrations were highest for red clover followed by the other

leguminous CC, while winter rye had the lowest value. Overall N accumulation was

greatest for lupin. In terms of maximum canopy density, planting of rye resulted in a

more rapid increase in LAI compared to other CC (Fig. 2.2). Leguminous CC had a

slower initial canopy development and in comparison with summer CC, canopy densities

were also lower. Rye and crimson clover had somewhat higher canopy densities, sweet

clover intermediate values, and all other crops had very sparse canopies (Table 2.12).

Canopy persistence was best for crimson clover but poor for most other tested species

(Fig. 2.2).

Red clover had the greatest weed biomass in comparison with the other CC. Use of

rye reduced weed growth by 92% compared to a 64% reduction for sweet clover. Use of

other leguminous crops like lupin, berseem and crimson clover, and cahaba vetch reduced

weed growth by about 50%. On the other hand, red clover decreased weed biomass by

only 32%. Nitrogen accumulation in weeds was greatest in the grass fallow and red

clover, while weeds in rye had the lowest value, but it was not significantly different

from other CC except red clover.

Rye had the greatest CCWI (excellent weed control) followed by crimson clover

and lupin which provided moderate weed control. Use of other legumes did not greatly

affect weed growth and thus provided poor weed control due to low plant populations

associated with low germination, ineffective nodulation, and poor adaptation to Florida

soils and environmental conditions.

Winter 2003/2004

An uneven distribution and lower precipitation along with lack of soil tillage

hampered growth of some CC during the 2003-04 growing season. Despite unfavorable









growth conditions, radish had the highest dry matter production (3.2 Mg ha-1). Biomass

of rye, crimson clover, black oat, lupin, sweet clover, and subterranean clover were

intermediate; while hairy vetch and lupin accumulated the lowest biomass (Table 2.13).

Nitrogen in shoots was highest for hairy vetch followed by lupin, crimson clover,

and subterranean clover. Non-leguminous CC had the lowest N concentrations in shoots

(<13 g N kg-1). Overall N accumulation was greatest for crimson clover, rye, and radish,

whereas, lupin, hairy vetch, subterranean clover while black oat accumulated less than 15

kg N ha1.

In terms of actual weed suppression, planting radish resulted in the lowest weed

biomass (Table 2.13), with only between 5 to 10% of the soil area covered with weeds.

Use of radish reduced weed growth by 88%, followed by crimson clover, rye, and hairy

vetch which decreased weed growth by 68-71%. The reduction in weed biomass for other

leguminous crops like lupin and subterranean clover was low, only about 16%.

Rototilling reduced weed biomass by 56% in comparison with grass fallow.

Nitrogen accumulation in weeds was greatest for grass fallow followed by

subterranean clover and lupin. The other treatments including the tillage fallow had

relatively low N accumulation in weeds (<20 kg N ha-1). Overall CCWI values were

greatest for radish which provided outstanding weed control whereas values for rye and

crimson clover were intermediate, indicative of moderate weed control. Use of other

leguminous CC did not greatly reduce weed growth and resulted in poor weed control.

Winter 2004/2005

Less precipitation occurred in winter 2004-05 compared to winter 2002, but overall

distribution was relatively even throughout the entire growing season. During the winter

of 2004/05, radish intercropped with rye and crimson clover had the highest dry matter









production followed by rye, rye intercropped with crimson clover, crimson clover, black

oat intercropped with crimson clover radish and black oat mixed with crimson clover and

radish all produced > 4.3 Mg ha-1, whereas other CC produced between 2.8 and 3.6 Mg

ha-1 (Table 2.14).

Shoot N concentration was significantly greater for crimson clover and black

oat+crimson clover. Non-leguminous CC had the lowest N concentrations in shoots.

Overall N accumulation was greatest for crimson clover and rye+crimson clover+ radish

due to the higher biomass accumulation, whereas black oat accumulated relatively little

N.

In terms of weed suppression, all CC treatments were statistically similar and

resulted in a weed biomass less than 0.73 Mg ha-1 which translated to a reduction of weed

growth by 80% for radish up to 98% for either the double or triple mix of cover crops.

Rototilling reduced weeds by 53% in comparison with grass fallow. Weed N

accumulation was statistically similar and less than 12 Kg N ha-1 for all CC treatments,

whereas grass fallow greatly exceeded this level (Table 2.14). The CCWI values varied

from excellent for radish to outstanding for the other monocrops and intercropping

systems.

Discussion

Summer Cover Crops

In general, sunnhemp was the most prolific summer CC. However, during 2004

continuous cultivation of sunnhemp along with wet and windy conditions increased the

dispersal and incidence of verticillium infection, thereby reducing biomass accumulation,

N fixation, and weed suppression (CCWI= 2 in 2004 vs. 64 in 2005 when the crop was

properly rotated). Based on this, it is obvious that despite the fact that sunnhemp









performance was superior to most other summer CC, repetitive use in the same site may

be undesirable and use of sound crop rotation with CC such as cowpea should be

considered. Overall sunnhemp dry weight (DW) accumulation was 10 Mg ha-1, which

was similar to the findings of Ramos et al. (2001), Steinmaier and Ngoliya (2001), and

Perin et al. (2004) under tropical conditions, but superior to the results from Jeranyama et

al. (2000) and Balkcom and Reeves (2005).

Hairy indigo performed consistently and average DW production was 8.0 Mg ha-1

similar to values reported by Reddy et al. (1986a). Overall DW accumulation for pigeon

pea was 5.7 Mg ha-1. Reported values ranged from 4.5 Mg ha-1 under tropical conditions

(Mafongoya and Dzowela, 1999) to 9.5 Mg ha-1 in Florida (Reddy et al., 1986a). The

steep decline in DW accumulation in 2005 was related to hairy indigo volunteering in

plots during 2005. 'Iron Clay' cowpea averaged 3.5 Mg ha-1, similar to values reported

by Jerenyama et al., (2000) and Muir (2002). Lablab had inconsistent biomass production

and overall DW accumulation was only 2.2 Mg ha-1. Similar values were reported by

Muir (2002) but other authors reported values ranging from 3.8 to 8.0 Mg ha-1 (Fischler

and Wortmann, 1999; Wortmann et al., 2000; Carsky et al., 2001; and Steinmaier and

Ngoliya, 2001).

Alyceclover accumulated 2.6 Mg ha-1 while velvet bean produced only 1.7 Mg ha-1

which was similar to the findings of Creamer and Baldwin (2000) but lower than values

ranging between 3.6 and 9.1 Mg ha-1 reported by Wortmann et al. (2000), Steinmaier and

Ngoliya (2001), and Carsky et al. (2001). This may be due to the low germination rate

and sparse CC population. Dry matter production of peanut and perennial peanut in 2003,

'Cream 40' cowpea, mung bean, and perennial peanut in 2004 was low, due to their poor









adaptation to the sandy soils and their ineffectiveness to successfully compete with

weeds.

Overall N accumulation by sunnhemp was 148 kg N ha-1. Similar values were

reported by Balkcom and Reeves (2005) while other studies showed a range from 195 to

305 kg N ha-1 (Ramos et al., 2001; Steinmaier and Ngoliya, 2001; Perin et al., 2004).

Overall N accumulation for hairy indigo was 132 kg N ha-1 which was similar to values

reported by Reddy et al. (1986a). Cowpea produced 61.4 kg N ha-l which was greater

than the findings by Jerenyama et al. (2000) and by Muir (2002). Pigeon pea accumulated

an average of 120 kg N ha-l which was lower than the value (170 kg N ha-l) reported by

Tian et al. (2000), which may be related with the decrease in pigeon pea DW in 2005

above discussed and the use of a more compact variety in our study.

Lablab accumulated N at 60 kg N ha-1, which was much less than the results of 113,

137 and 177 kg N ha-l reported by Steinmaier and Ngoliya (2001), Carsky et al.(2001),

and McDonald et al. (2001), respectively. This may be related to inconsistent

performance of lablab across years. Velvet bean produced only 50 kg N ha-l, much lower

than values of 163 and 281 kg N ha-l reported by Steinmaier and Ngoliya (2001) and

Carsky et al. (2001), respectively. This may be related to the use of a bushy type of

velvetbean that did seem to be less vigorous than the more commonly used vining types.

Overall weed reduction was highest for 'Iron Clay' cowpea (90% of control)

followed by sunnhemp (77%), alyceclover (74%), and hairy indigo (64%). Fallow tillage

was less effective in reducing weeds compared to the best CC with an average reduction

of 60%. Weed reduction for alyceclover and hairy indigo probably could have been

higher due to in 2003 alyceclover volunteered in hairy indigo plots and in 2005 hairy









indigo volunteered in alyceclover plots. Use of velvet bean, on the other hand, reduced

weed growth by only 42%, which contrasted with the 68% reduction in a study with

maize by Caamal-Maldonado et al. (2001). This may be related to the bushy variety used

in the current study, along with the low germination in field, wider row spacing, and

stronger weed competition.

The presence of alyceclover, hairy indigo, and cowpea as volunteer crops shows

that some of these crops may have good potential for reseeding during subsequent years,

due to their capacity to produce large numbers of dormant and/or hard coated seeds.

These species would be able to become part of the soil seed bank and germinate over

many years as suggested by Benech-Amold et al. (2000), which could provide a cost-

effective self sustaining practice in a mature citrus system.

Provided that via use of sound crop rotation the build up of disease in sunnhemp

plots was prevented, sunnhemp and cowpea provided outstanding weed control and

CCWI values were 33 and 31, respectively. Corresponding values for alyceclover and

hairy indigo were 14 and 11 (excellent weed control). Except for alyceclover, this

suppression was closely related with CC biomass production which appears to be the

main mechanism for weed suppression due to direct competition for resources (light,

nutrients and water).

Velvet bean was moderately effective in suppressing weeds (CCWI=1.2), which

contrasted with the results from Caamal-Maldonado et al. (2001) and Buckles and

Triomphe (1999) in central America and from the findings of Carsky et al. (2001) and

Fisher and Wortmann (1999) in Africa, who obtained excellent weed suppression in









maize with a viny cultivar. These differences may be related to the poor germination,

nodulation, vigor and competitiveness of the bushy type that was used for our studies.

Winter Cover Crops

Radish and rye were the most prolific biomass producers among the mono-cropped

winter CC. In our study radish accumulated 4.6 Mg ha-1 compared to 1.6 Mg ha-1

reported by Vyn et al. (2000). However, Justes et al. (1999) reported values up to 6.4 Mg

ha-1 with no N added. Rye generated 3.5 Mg ha-1 which was similar to results of Ngouajio

and Mennan (2005) and Akemo et al. (2000), but lower compared to the 5-10 Mg ha-1

reported by Bauer and Reeves (1999) and Reberg-Horton et al. (2005). Crimson clover

yielded 2.9 Mg ha-1 which was similar to values reported by Daniel et al. (1999) and

Schomberg and Endale (2004) but lower than the 4.9 Mg ha-1 reported by Dyck et al.

(1995) and Odhiambo and Bomke (2001).

Use of a triple CC mix (rye+crimson clover+radish ) resulted in the greatest

biomass (8.8 Mg ha-1), which was about two times greater than the 4.6 Mg ha-1 for radish

planted alone. Karpenstein-Machan and Stuelpnagel (2000) reported similar findings for

a mixed CC system in Germany consisting of rye and crimson clover, and for rye with

winterpea (Pisum sativum L. ssp. arvense (L.) Poir). Similar results were reported by

Juskiw et al. (2000) for small grain cereals in Canada.

During the 2004-2005 winter season, DW accumulation of triple CC mixes was

comparable to the 8-9 Mg ha-1 produced by summer CC systems in 2004 but lower than

the 10-14 Mg ha-1 obtained in 2005. This may be related to the synergistic combination of

the complementary traits of the constituents of the mix, with rye providing vigorous and

rapid growth along with allelopathic activity, radish breaking through compaction layers

and enhancing biodiversity and soil structure, and while crimson clover providing









additional N via N fixation. This synergistic interaction of complementarities in root and

canopy structure may facilitate improved adaptation to different ecological niches, soil

types and weather conditions, providing multiple benefits and improved nutrient

retention, cycling, and N-fixation as suggested by Gliessman (1998), Altieri (1999), and

Karpenstein-Machan and Stuelpnagel (2000). Mixed CC systems thus mimic natural

systems and crop components may therefore compete more effectively with weeds, which

explains the superior performance of these systems.

Dry weight accumulation by red, berseem, sweet and subterranean clover, cahaba

white vetch, and lupin was relatively low. This may be related to the uneven rainfall

distribution during 2002/03 and 2003/2004; to low soil organic matter, pH, and K values;

to poor initial nodulation and growth by these crops, hampering their ability to effectively

compete with weeds; and to an overall poor adaptation of these crops to coarse sandy

soils. Row middles were not fertilized and chicken manure was applied only to the tree

rows in order to reduce weed vigor and to provide leguminous CC with a competitive

edge. Besides this, lower temperatures and light intensities during winter and the uneven

rainfall distribution in comparison with other years could have hampered CC growth.

Overall crop N accumulation was greater for leguminous CC probably due to N

fixation. Crimson clover accumulated 70 kg N hal-. Similar values were reported by

Daniel et al. (1999) while others recorded values were between 120-125 kg N ha-1 (Dyck

et al. 1995; Odhiambo and Bomke, 2001). Lupin and radish both accumulated around 47

kg N ha-1. Rye and black oat averaged 32 and 19 kg N ha-1, respectively, which was

lower than values for other studies (Bauer and Reeves, 1999; Odhiambo and Bomke,

2001).









Intercropping of rye+crimson clover+radish resulted in the highest N accumulation

(110 kg N ha-1) whereas black oat+crimson+radish accumulated 74 kg N ha-1. The high N

accumulation by these CC systems may be related to N retention (by rye, radish, and

black oat) and/or additional N-fixation (crimson clover) as suggested by Justes et al.

(1999), Vos and van der Putten (2001), and Kristensen and Thorup-Kristensen (2004) for

rye and radish.

Similar to the positive effects of mixed CC systems on N accumulation,

intercropping two and three-way-mixtures reduced weed growth by 98% compared to the

mowed fallow. Corresponding values for crimson clover, rye, and radish monocrops were

78-84%. For rye, weed suppression may be related to allelopathy which is often reported

in the literature (Weston, 1996; Fennimore and Jackson, 2003; Reberg-Horton et al.,

2005).

Compared to the best CC, fallow tillage was less effective in reducing weeds, with

reductions in weed biomass averaging only 55%. Similarly, two- and three-component-

mixtures resulted in the highest CCWI values. The outstanding weed control of mixed

systems may be related to competition and synergistic allelopathic activities of radish,

possibly due to the glucosinolate content reported for radish and other members of the

Brassicae family (Norsworthy et al., 2005, Morra and Kirkegaard, 2002). Similar

allelopathic action has been also reported for black oat (Bauer and Reeves, 1999).

System Dynamics

Summer CC had greater DW production capacity compared to winter CC. The four

highest biomass producers were in descending order: sunnhemp>hairy indigo>pigeon

pea>cowpea with respective values of 10, 8, 5.7, and 3.5 Mg ha-1 which translates to 83,

67, 48, and 29 kg ha-1 d-1, respectively, which was similar to the values reported by









Reddy et al. (1986a), Mafongoya and Dzowela (1999), Jerenyama et al. (2000), and Perin

et al. (2004). When weeds were included the sequence was as follows: sunnhemp>hairy

indigo>pigeon pea>'Cream-40' cowpea, with corresponding daily DW accumulation

values of 93, 81, 69, 49 kg ha-1 d-.

The above results contrasted with the biomass produced by the system during the

winter season, in which monocrops of radish, rye, crimson clover, and black oat produced

3.8, 3.4, 2.9, and 2.4 Mg ha-l, with an average of 3.1 Mg ha-1per season and 25, 23, 19,

and 16 kg ha-1 day-', respectively, which were lower than the values reported by Justes et

al. (1999), Odhiambo and Bomke (2001), and Reberg-Horton et al. (2005). When weeds

were included in the balance, the total biomass production by the system was greater for

radish, rye, crimson clover, and black oat with values between 4.4 and 3.6 Mg ha-1, with

an average of 4.0 Mg ha-1 per season and 27 kg ha-1 d-1, which was much less than that


-22
obtained during summer season. This due may be to the lower radiation (138 w m-2 in

winter vs. 188 w m-2 in summer)and temperatures in winter (average temperatures from

2002 to 2005 in winter were 14.2 C, min temp= -2.2 C max temp= 29.2 C vs. average

temperatures in summer 26.3 C, min temp= 18.8 C and max temp= 36.6 C). In

addition, rainfall was higher and relatively more evenly distributed during the summer.

Finally, most winter CC may not be well-adapted to growth environments in Florida, and

nodulation of many leguminous crops tends to be erratic on sandy soils during the first

few years of their cultivation.

Use of intercropping allowed for increases in production capacity during the winter

months due probably to the synergistic interaction among crops as suggested by Kabir

and Koide (2002) and by Karpenstein-Machan and Stuelpnagel (2000). For winter









intercropping CC, the biomass production was between 8.0 to 2.8 Mg ha-1, with an

average of 5.3 Mg ha-1 per season or 35 kg ha-1 d1.

In general, N accumulation was greater for leguminous species probably due to N

fixation and was greatest during the summer season. The four highest N accumulators

were in descending order: sunnhemp>hairy indigo>pigeon pea>cowpea with values

between 61 to 148 kg N ha-1, with an average of 129 kg N ha-1 per season or 1.1 kg N ha-1

kg ha-1 d-1, similar to the values reported by others (Reddy et al., 1986a; Mafongoya and

Dzowela, 1999; Jerenyama et al., 2000; Ramos et al., 2001; Muir et al., 2002).

When weeds were included in the balance, the total N content in the system was

greater for pigeon pea, hairy indigo, sunnhemp, and velvet bean, with values between 123

and 222 kg N ha-1, with an average of 174 kg N ha-1 per season or 1.5 kg N ha-1 d-1, which

was significantly greater than the N accumulated only by the mowed fallow (66 kg N ha-

1). These results underline the ecological role of weeds in the system in capturing C and

N within the system, because without this component, an important fraction of N could

be lost from the system through leaching or runoff as proposed by Vos and van der

Putten, (2001) and by Kristensen and Thorup-Kristensen, (2004).

The four highest N accumulators in winter were in descending order: crimson

clover>radish >lupin>rye with values ranging between 32 to 69 kg N ha-1. The N

accumulation average was 49 kg N ha-1 per season (38% of N accumulation summer

average) or 0.3 kg N ha-1 d-1, which were lower than the values reported by Justes et al.

(1999), Odhiambo and Bomke (2001), and Reberg-Horton et al. (2005). When weeds

were included in the balance, N in the system was greatest for crimson clover followed

by lupin, red clover, and radish, with values ranging between 84 to 58 kg N ha-1, with an









average of 69 N kg ha-1 per season or about 0.5 kg N ha-ld-, which was 40% of that

obtained with summer cover crops, due to similar reasons explained above for biomass

production.

Use of mixed crop systems increased crop performance due to synergetic

interactions of crop components, and N accumulation ranged from 55 to 110 kg N ha-1,

with an average of 0.5 kg N ha-l d1 Although this is 67 % higher compared to monocrop

systems, it is still only 45 % of N accumulated by summer CC systems.

Nitrogen accumulation values for intercropping system were similar to the values

reported by Justes et al. (1999); Odhiambo and Bomke (2001); Reberg-Horton et al.

(2005) and superior to N accumulated in the control (mowed fallow) which amounted to

53 kg N ha-1 season'.

In terms of weed suppression, summer monocrops (sunnhemp, cowpea, hairy

indigo and alyceclover) were more effective in outcompeting weeds compared to mono-

cropped winter CC systems (rye, radish, and crimson clover). The CCWI values for the

winter CC were below 13, which was mainly associated with the lower biomass

production by winter CC discussed above. Actual weed reduction was similar in both

seasons, with average weed reductions of 76 and 80% for superior summer and winter

CC, respectively. Volunteering of mainly alyceclover and hairy indigo in the other

treatments at times created potential but may be arbitrary "weed" issues in subsequent

crops. Winter intercropped CC had CCWI greater than 39 due to higher biomass and

probably the allelopathic suppression of weeds by rye and black oat as discussed by

Weston (1996) and Putnam (1988).









Observed weeds species did not follow any special pattern associated with CC

treatments. The main species observed in field during late spring, summer, and early fall

(warm-season weeds) were: bermudagrass (Cynodon dactylon (L.) Pers.), large crabgrass

(Digitaria sanguinalis (L.), bahiagrass (Paspalum notatum Fluegge), goosegrass

(Eleusine indica (L.) Gaertn), Scop), crowfootgrass (Dactyloctenium aegyptium (L.)

Willd.), globe sedge (Cyperus globulosus Aubl.), cylindric sedge (Cyperus retrorsus

Chapm.), Florida pusley (Richardia scabra L.), and carpetweed (Mollugo verticillata L.)

among the species more prevalent and frequently observed in field; whereas purple

nutsedge (Cyperus rotundus L.), spreading dayflower (Commelina diffusa Burm. F.),

common pigweed (Amaranthus hybridus L.), common ragweed (Ambrosia artemisiifolia

L.), southern sida (Sida acuta Burm. F.), common purslane (Portulaca oleracea L.),

poorjoe (Diodia teres Walt.) were observed only in localized spots; and in some plots,

alyceclover (Alysicarpus vaginalis (L.) DC.) and hairy indigo (Indigofera hirsuta L.)

volunteered during subsequent summer seasons.

In winter and early spring, the dominant weeds found were red sorrel (Rumex

acetosella L.), oldfield toadflax (Linaria canadensis (L) Dumont), common

venuslookingglass (Triodanis perfoliata (L.) Nieuwl.), wandering cudweed (Gnaphalium

pensylvanicum Willd.); whereas virginia pepperweed (Lepidium virginicum L.), Carolina

geranium (Geranium carolinianum L.), and cutleaf evening primrose (Oenothera

laciniata Hill) were only observed in certain spots.

There appears to be excellent prospective for the use of sunnhemp, 'Iron Clay'

cowpea, alyceclover, and hairy indigo as summer CC for weed suppression in organic

Florida citrus systems. However, hairy indigo and sunnhemp appeared to have some









drawbacks relative to cowpea and alyceclover. Hairy indigo grew relatively tall and

bushy and since it is a hard-seeded crop, seeds could persist for a long time in the soil

seed bank (Benech-Arnold et al., 2000) with a weedy potential around young citrus trees,

so hairy indigo would have to be mowed before it goes to seed when used in newly-

planted orchards. Sunnhemp, although it did not branch profusely, grew very tall so that

it might create problems with shading if it were planted too close to young trees. Also

sunnhemp appears to need rotation to avoid problems with soil fungi such as Verticillium

spp.

Alyceclover and 'Iron Clay' cowpea had more compact low-growing canopies,

which would facilitate their integration into citrus production systems. Under our

experimental conditions, 'Iron Clay' cowpea reseeded itself, but seed viability in the field

was less than one year, so when mowed in time it could be managed more easily than

hairy indigo. However, some selections can be rather "viny" and grow around young

trees, which may limit its use as a cover crop in the near vicinity of small trees (<2 m),

although mechanical weeding can easily address this potential problem. Even though

alyceclover reseeded readily, it did not seem to interfere with citrus trees or citrus

irrigation and therefore may be the most suitable species to be planted in the vicinity of

trees. Based on this, we propose for tree rows a "sandwich" system consisting of a tree

strip of 1.6-1.8 m planted with alyceclover as a summer CC. Bordering this strip would

be 'Iron Clay' cowpea.

If it would be desirable to apply mulch to the tree row during the winter time, a

strip of sunnhemp could be planted in the row middle for this purpose. However,









alternatively use of 'Iron Clay' cowpea in row middles may be preferable as discussed

above.

During the winter season we propose the use of a system consisting of a tree strip

planted with a mixture of crimson clover and black oat as winter CC due to their compact

canopy and low probability of competition with citrus trees for light. Intercropping rye

with crimson clover and radish would be desirable for row middles.

However, seed availability of suitable winter CC including the use of seeds

produced in different continents may pose some problems. Limited availability of non

treated/certified organic seed sources of varieties and/or cultivars adapted to the southeast

of United States appears to be one of the key issues that may hamper effective integration

of CC in organic citrus production systems.

Increased N accumulation in CC-based systems during the summer season may

provide benefits to subsequent CC crops and/or citrus trees via mineralization. Use of

continuous CC sequence may also reduce potential nutrient losses due to leaching (Vos

and van der Putten, 2001; Kristensen and Thorup-Kristensen, 2004).

Continuous growth of CC combined with reduced tillage may also enhance C

sequestration and N cycling and retention in the soil (see Chapter 5 for more detailed

discussion about CC effect on soil quality). Augmented soil organic matter is considered

a desirable characteristic of sustainable systems. In organic systems, this approach may

also foster the development of soils that can enhance natural suppression of: weeds

(Gallandt et al., 1999; Jordan et al., 2000), soil borne diseases (Bailey and Lazarovits,

2003), and insect populations (Altieri and Nichols, 2003). All these processes are related

through the mechanism of increased soil organic matter and soil microbial diversity









(Kennedy, 1998; Kremer and Li, 2003). As a result, such an approach may be cost-

effective due to reduced requirements of external inputs.

Conclusions

Overall dry matter, N accumulation, and weed suppression by annual CC varied

depending on plant species and season. In general, summer CC had the highest biomass

and N accumulation, in which the more consistent performers in terms of biomass

production, N accumulation, and weed suppression were sunnhemp, hairy indigo,

cowpea, and alyceclover. Although pigeon pea was consistent for biomass and N

accumulation, its weed suppression capacity was not always consistent. The most

consistent and best performing winter CC were radish, rye, and crimson clover.

The best summer and winter CC, DW production averaged 6.8 and 3.1 Mg ha-1,

respectively while corresponding total biomass (CC + weeds) were 9.7 and 4.0 Mg ha-1.

Cover crop N accumulation averaged 129 and 49 kg ha-1 during summer and winter

seasons, respectively and total N accumulation (CC + weeds) was 174 and 69 kg N ha-1,

which underlines the complementary role of weeds in nutrient retention and recycling.

Throughout the course of the study, use of selected CC provided excellent weed control,

which was superior to other methods including tillage. Use of two- or three-component

winter CC mixes resulted in higher DW and N accumulation and more effective weed

suppression, due probably to the synergistic interaction among system components.











Table 2.1. Overview of annual summer and winter cover crops used during the 2002 and
2003 growing seasons.


Treatment Summer


Cropping Season
Winter/Spring


2002


'Iron Clay' cowpea (Vigna
unguiculata L. Walp.)
Velvet bean (Mucuna atropurpureum
L.) DC)
Sunnhemp (Crotalariajuncea L.)

Alyceclover (Alysicarpus vaginalis
L.)
Hairy indigo (Indigofera hirsuta L.)


Grass fallow
20

Sunnhemp/'Iron Clay' cowpea

'Iron Clay' cowpea
Velvet bean
Hairy indigo

Lablab (Lablab purpureus L.)

Peanut (Arachis hypogea L.)
Perennial peanut (Arachis glabrata
Benth.)
Grass fallow
Tillage fallow
Alyceclover


Crimson Clover (Trifolium
incarnatum L. )
Red clover (Trifolium pratense L.)

Lupin/ Cahaba vetch (Lupinus
angustifolius L./Vicia sativa L.)
Rye (Secale cereale L.)

Berseem Clover/Sweet clover
(Trifolium alexandrinum L./
Melilotus officinalis L.)
Grass fallow
03

Rye/Hairy vetch (S. cereale/ Vicia
villosa Roth)
Crimson clover
Radish (Raphanus sativus cv. rufus)
Black oat/Lupin (Avena strigosa
Schreb/ Lupinus angustifolius L.)
Subterranean clover (Trifolium
subterraneum L.)
Lupin
Perennial peanut

Grass fallow
Tillage fallow










Table 2.2. Overview of annual summer and winter cover crops used during the 2004 and
2005 growing seasons.


Treatment Summer


Sunnhemp/ 'Iron Cla
'Iron Clay' cowpea
Hairy indigo
Velvet bean
Pigeon pea (Cajanus
Lablab
Perennial Peanut
Grass fallow
Tillage fallow
Cream-40 cowpea
Mung bean
Alyceclover


Cropping season
Winter/Spring
2004
y' cowpea Winter rye (WR)
Crimson clover (CR)
Radish (R)
Black oat (BO)
cajan L.) WR + CR
BO + CR
Perennial peanut
Grass fallow
Tillage fallow
WR + CR + R
BO + CR + R


Velvet bean/Sudan grass
Sunnhemp/'Iron Clay' cowpea
Hairy indigo
Alyceclover
Lablab/velvet bean
Lablab/sorghum sudan grass
(Sorghum bicolor L.)
Perennial Peanut
Grass fallow
Tillage fallow
Pigeon pea


2005
Winter rye (WR)
Crimson clover (CR)
Radish (R)
Hairy vetch (HV)
WR + CR
WR + CR+R

Perennial peanut
Grass fallow
Tillage fallow
WR/HV (67-33%)
WR/HV (33-67%)









Table 2.3. Overview of seeding rates, space between rows and cultivars used as annual
summer and winter cover crops used from 2002 to 2005.
Species Seeding rate Space between Cultivar
(kg ha-1) plants (cm)
Pigeon pea 67 36
Velvet bean 56 36 Bushy type
Cowpea 56 36 'Cream-40' ('04);


'Iron Clay' ('02-
'03&'05)


Sunnhemp
Mung bean
Lablab
Alyceclover
Sudan grass
Hairy indigo


Winter rye



Black oat
Lupin
Crimson clover
Subterranean clover
Red Clover
Cahaba white vetch
Hairy vetch
Berseem clover
Radish


112 (55 in '03-'04)



112
112
28
28
28
22
22 (11 in '03-'04)
22
22


Abruzzi ('02-
'03&'05) Florida
401(2004)
Soil saver
Tifblue
AB Dixie
Mt Barker


Rufus









Table 2.4. Outline of planting and harvest dates and duration of summer and winter
cover crops.
Year Summer Winter

Planting Mowing Duration Planting Mowing Duration

(days) (days)

2002 30 July 11 Oct 102 1 Dec 15 May 165

2003 10 June 16 Oct 127 28 Oct 31 March 154

2004 11 June 10 Oct 121 1 Nov 5 April 156

2005 21 June 25 Oct 125 5 Dec 9 May 155




Table 2.5. Outline of cover crop weed index (CCWI) categories.
CCWI value Cover crop Weed Weed control
< 0.5 CC not competitive Weeds dominate Very poor (>70%
weeds)

0.5-1 CC coexist Weeds coexist Poor

1-3 CC start prevailing Weeds prevail in Moderate
niches

3-5 CC prevail Weeds fail to Adequate
dominate

5-15 CC predominate (70-90%) < 10-30% weeds Excellent

>15 CC dominate completely <5% weeds Outstanding
It is assumed if CCWI >15 then weed control is considered outstanding since weeds only
cover account for less than 5% of the total biomass. It should be noted that in the absence
of weeds the CCWI will approach infinity, and the upper boundary is thus not defined.












Table 2.6. Rainfall measured at Plant Research and Education Unit (Citra) Florida
Automated Weather station Net work (FAWN)1 during the 2002-2005
summer CC growing season.

Year


Month


2002


2003


2004


2005


-------------------------------Rainfall (mm)----- -----------


June

July


August 150

Sept 120

October 15

November 2

Total 426


145

100


124

269

157

418

93


1060


1 Data obtained from the website http//fawn.ifas.ufl.edu on 12/10/2005. Blank spaces
mean period of time when CC were not grown.


636











Table 2.7. Shoot dry weight accumulation (DW); shoot N concentration (Nconc); shoot
N accumulation (Naccum), maximum observed leaf area index (LAI max)
with time of occurrence in brackets; weed dry weight accumulation (DW);and
cover crop weed index (CCWI) for summer cover crops grown during the
2002 growing season.


Cover Crops


Weeds


DW
Treatment Mg ha-1


Cowpea


Velvet bean



Sunnhemp



Alyceclover



Hairy indigo


3.89 bct


0.98 c


Nconc
g kg-1


Naccum
kg N ha-1


16.5 bc 61.2 c


24.1 a 23.5 c


LAI
max
2 -2
m2m-2
(wks)

5.3 a
(6)

2.4 b


12.06 a 13.2 c 158.3 a 3.4 b
(10)


2.49 c


5.87b


14.6 c 36.9 c


2.3 b


18.7 b 106.5 b 5.9 a


DW
Mg ha-1


0.61 b


3.62 a



0.64 b



1.21 b


0.74 b


(10)


Grass
Fallow


4.43 a


CCWI


11.8b


0.8 c



22.0 a



3.6c


14.9 a


t Means within the same column followed by the same letter do not differ statistically
based on the Duncan's Multiple Range test (P<0.05).









Table 2.8. Shoot dry weight accumulation (DW); shoot N concentration (Nconc); shoot N
accumulation (Naccum), maximum observed leaf area index (LAI max) with
time of occurrence in brackets; weed dry weight accumulation (DW); weed N
accumulation (Naccum); and cover crop weed index (CCWI) for summer
cover crops grown during the 2003 growing season
Cover Crops Weeds
DW Nconc Naccum LAI DW Naccum
Treatment Mg ha-1 g kg-1 kg N max Mg ha- kg N CCWI
ha- m2m-2 ha-
(wks)


Cowpea (I)1


Sunnhemp


Cowpea


1.38 cbt


24.9 b


10.24 a 21.8 b


34.6 c 2.04 bc 0.95 bc


(12)
223.1 a 3.99 a
(16)


2.37 bc 24.9 b 59.2 c 3.40 a
(12)


0.68 c

0.60c


18.3 cd 4.9 b


16.5 cd 44.2 a


10.8 d 16.8 ab


Velvet Bean


Hairy indigo


2.78 bc 33.5 a


9.17 a


93.0 bc 3.71 a
(12)


17.9 c 162.8 ab 3.22 a
(16)


2.04 ab 37.1 bc 1.6 b


2.72 a


39.1 abc 4.1 b


3.71 b 27.8 b 100.4 bc 3.86 a 1.36 bc 27.6 bcd 3.4 b
(12)


Peanut

Perennial P

Alyceclover


Grass Fallow


1.07 bc 27.7 b 29.1 c 1.02 bc
(12)
0.02 c 20.7 bc 0.41 d 0.01 d
(16)


3.91 b


18.0 c 70.4 bc 2.5 ab
(16)


2.89 a

2.56 a

0.25 c


3.90 a


46.7 ab 0.4 b

43.5 ab 0.01 b


3.2 d


60.1 a


22.4 ab


1.58 bc 33.8 bcd


t Means within the same column followed by the same
based on the Duncan's Multiple Range test (P<0.05).
t 'Iron Clay' cowpea intercropped with sunnhemp


letter do not differ statistically


Lablab


Tillage
Fallow









Table 2.9. Shoot dry weight accumulation (DW); shoot N concentration (Nconc); shoot N
accumulation (Naccum); weed dry weight accumulation (DW); weed N
accumulation (Naccum); and cover crop weed index (CCWI) for summer
cover crops during the 2004 growing season.

Cover Crops Weeds

DW Nconc Naccum DW Naccum CCWI
Treatment Mg ha-1 g kg-1 kg N ha- Mg ha- kg N ha-

'Cream 40' 0.72 det 19.1 cd 14.2 bc 5.22 a 97.2 a 0.2 c


cowpea

Mung bean

Alyceclover

Sunnhemp

'Iron Clay'
Cowpea

Hairy indigo

Velvet Bean

Pigeon Pea

Lablab

Perennial P.

Grass Fallow

Tillage
fallow


0.24 e

2.89 c

5.31 b

5.08 b


7.59 a

1.28 cde

7.60 a

0.76 de

0.16 e


16.0 de

17.0 d

12.0 e

25.8 b


16.7 d

34.0 a

22.2 bc

24.8 b

23.5 bc


4.0 c

49.5 bc

63.45 b

63.9 b


127.0 a

43.7 bc

174.7 a

20.2 bc

3.6c


4.87 ab

0.43 d

2.86 cb

0.16 d


1.90 cd

3.84 abc

2.56 c

3.53 abc

3.64 abc

5.49 a

1.70 abc


85.4 ab

4.5 e

42.9 bcde

22.6 de


28.6 cde

72.5 abc

47.1 bcde

87.6 ab

36.3 cde

72.0 abc

60.2 abcd


0.1 c

28.0 a

2.1 c

15.5 ab


6.5 bc

0.6 c

3.8 c

0.3 c

0.04 c


tMeans within the same column


followed by the same letter do not differ statistically


based on the Duncan's Multiple Range test (P<0.05).










Table 2.10. Shoot dry weight accumulation (DW); shoot N concentration (Nconc); shoot N
accumulation (Naccum); weed dry weight accumulation (DW); weed N accumulation
(Naccum); and cover crop weed index (CCWI) for summer cover crops during the
2005 growing season.


Cover Crops


Weeds


Treatment

Cowpea (I)j


DW
Mg ha-


Nconc
g kg-1


Naccum
kg N ha1


2.75 cdt 16.7 c 46.0 c


Sunnhemp 12.61 a 19.0 c 242.6 a 0.55 e


Alyceclover

Hairy indigo

Velvet bean
+Sudangrass


1.00 cd 17.8 c 19.3 c


9.46 b


16.9 c 161.7 b 1.22 de


1.93 cd 24.7 a 48.3 c


Sudangrass 1.42 cd 6.3 d
+ velvet bean


Lablab| +
Velvet bean

Velvet bean +
lablab

Lablab +
sudangrass

Sudangrass +
lablab


9.0 c


1.44 cd 16.7 c 24.6 c


1.53 cd


24.2 ab 50.7 c


2.18 cd 19.9 bc 42.1 c


1.62 cd 8.8 d


12.9 c


Pigeon Pea 3.71 c 17.8 c 64.1 be 2.68 bcd


47.1 abc 1.6c


Perennial P

Grass Fallow

Tillage
Fallow


0.24 d 21.0 ab 2.1 c


DW
Mg ha'


0.42 e


3.56 bc


Naccum
kg N ha1


2.1 e

2.3 e


20.5 de

3.9 e

5.1 e


5.1 e


3.0 e


3.0 e


32.2 cd


32.2 cd


CCWI



64.0 ab

80.0 a

0.5 c

17.2 bc

1.8c


1.1 c


13.5 bc


12.6 bc


12.0 be


6.5 bc


1.99 cde



1.99 cde



1.32 de


1.32 de


1.54 de


1.54 de


4.22 ab


0.1 c


5.56 a


2.71 bcd


57.5 ab

70.8 a

36.1 bcd


tMeans within the same column followed by the same letter do not differ statistically based on
the Duncan's Multiple Range test (P<0.05).j 'Iron Clay' cowpea intercropped with sunnhemp










Table 2.11. Rainfall measured at Plant Research and Education Unit (Citra) Florida
Automated Weather station Net work (FAWN)1 during the 2002-2005 winter
CC growing season.
Year


Month


2002/2003


2003/2004
Rainfall (mm)


2004/2005


October 28
November 46 34
December 147 21 38
January 3 43 21
February 128 141 64
March 181 54 119
April 13 59
May 1
Total 472 334 335
1 Data obtained form the website http//fawn.ifas.ufl.edu on 12/10/2005. Blank spaces
mean period of time when CC were not grown.

Table 2.12. Shoot dry weight accumulation (DW); shoot N concentration (Nconc); shoot
N accumulation (Naccum); maximum observed leaf area index (LAI max)
with time of occurrence in brackets; weed dry weight accumulation (DW);
weed N accumulation (Naccum); and cover crop weed index (CCWI) for
winter cover crops during the 2002-2003 growing season.
Cover Crops Weeds
DW Nconc Naccum LAI max DW Naccum CCWI
Treatment Mg ha1 g kg- kgN ha- m2m-2 Mg ha) kgN ha)
(wks)
Crimson 1.9 ab 23.4 44.5 b 1.06 a 1.71 bc 26.0 ab 1.6 b
clover ab (12)
Red clover 0.8 ab 30.2 a 22.0 b 0.26 c 2.53 ab 36.5 a 0.5 b
(12)
Berseem 0.2 b 22.4 b 4.3 b 0.45 cb 1.72 bc 25.8 ab 0.1 b
clover (12)
Sweet 0.5 ab 26.4 14.3 b 0.66 b 1.32 bc 22.3 ab 0.5 b
clover ab (12)
Lupin 3.1 a 28.2 83.4 a 1.12 a 1.63 bc 24.1 ab 1.7 b
ab (12)
Cahaba 1.3 ab 25.6 34.2 b 0.44 cb 1.31 bc 27.1 ab 1.1 b
vetch ab (12)
Winter rye 3.2 a 10.0 c 31.3 b 0.94 a 0.30 c 5.9 b 7.8 a
(12)
Grass -- -- -- -- 3.70 a 38.6 a --
Fallow
tMeans within the same column followed by the same letter do not differ statistically
based on the Duncan's Multiple Range test (P<0.05).










Table 2.13. Shoot dry weight accumulation (DW); shoot N concentration (Nconc); shoot
N accumulation (Naccum); weed dry weight accumulation (DW); weed N
accumulation (Naccum); and cover crop weed index (CCWI) for winter cover
crops during the 2003/2004 growing season.
Cover Crops Weeds

DW Nconc Naccum DW Naccum CCWI
Treatment Mg ha-1 g kg-1 kg N ha- Mg ha-1 kg N ha-

Hairy vetch 0.20 ct 38.7 a 7.8 bc 1.23 bc 19.4 bc 0.5 b

Winter rye 0.89 cb 13.1 d 15.4 abc 1.23 bc 19.4 bc 2.5 b

Crimson 1.69 b 23.6 bc 38.0 a 1.10 bc 13.9 bc 1.8 b
Clover

Radish 3.24 a 10.1 d 33.5 ab 0.48 c 7.0 c 15.0 a

Black oat 1.25 cb 8.0 d 9.8 bc 1.82 abc 21.7 bc 0.8 b

Lupin (I)J 0.37 cb 28.7 b 10.3 bc 1.82 abc 21.7 bc 0.2 b


Subterranean 0.281 cb 21.8 c 5.9 c 2.25 ab 32.3 b 0.2 b
Clover

Lupin 0.20 c 24.1 bc 5.3 c 2.53 ab 30.8 b 0.1 b

Grass Fallow -- -- -- 3.84 a 52.1 a

Tillage -- -- 1.70 abc 20.1 bc --
fallow

tMeans within the same column followed by the same letter do not differ statistically
based on the Duncan's Multiple Range test (P<0.05).
t Lupin intercropped with black oat









Table 2.14. Shoot dry weight accumulation (DW); shoot N concentration (Nconc); shoot
N accumulation (Naccum); weed dry weight accumulation (DW); weed N
accumulation (Naccum); and cover crop weed index (CCWI) for winter cover
crops during the 2004/2005 growing season.


Cover Crops


DW


Treatment


Nconc Naccum DW


Weeds

Naccum


Mg ha1 g kg1 kg N ha1 Mg ha1 kg N ha1 CCWI


Winter rye (WR)

Crimson clov.
(CR)

Radish (R)

Black oat (0)


WR + CR


O+CR


WR + CR + R

BO + CR + R

Grass fallow

Tillage fallow


5.96 bt


8.3 c


49.1 cd


5.02 bc 24.6 a 125.2 a


4.31bcd 14.3 bc 62.9 cd


3.56 cd 8.4 c


27.6 d


5.34 bc 10.9 c 58.7 cd

2.76 de 19.2 ab 53.4 cd


0.28 b 4.8 b

0.19b 3.3b


0.73 b 11.6b

0.39 b 4.0 b

0.07 b 0.7 b

0.07 b 1.1 b


7.99 a 14.2 bc 110.0 ab 0.08 b ND

4.95 cb 14.7 bc 73.72 cb 0.08 b ND


3.58 a 66.5 a

1.70 ab 24.7 b


25.2 cd

35.3 c


10.6 d

16.2 d

76.2 b

39.4 c

99.9 a

61.9b


tMeans within the same column followed by the same letter do not differ statistically
based on the Duncan's Multiple Range test (P<0.05).











7
-e-CP
S6 VB
E SH
E 5 AC

4-

.; 3


1


0 14 28 42 56 70 84 98 112
Days after planting

Figure 2.1. Leaf area index values for summer cover crops 2002 (CP= cowpea; VB
velvet bean; SH= sunnhemp; AC= Alyceclover; HI= hairy indigo).


28 56 84 112 140


Days after planting


Figure 2.2. Leaf area development for winter cover crops during 2002/2003 (CCL=
Crimson clover; C-CL= Red clover; B-CL= Berseem clover; S-CL= Sweet
clover; LP= Lupin; C-V= Cahaba Vetch; Rye= Winter rye).














CHAPTER 3
ESTABLISHMENT OF PERENNIAL PEANUT AND ITS EFFECTIVENESS IN
SUPPRESSING WEEDS IN CITRUS ROW MIDDLES

Introduction

Perennial or rhizoma peanut (Arachis glabrata, Benth.) is a rhizomatous warm-

season perennial legume native to South America with a wide area of adaptation ranging

from 31 N to 350 S latitude (Prine et al., 1981). It was introduced into Florida in 1936,

and it is used as a living mulch in association with other crops for soil conservation; as a

pasture crop for grazing and/or hay production; and as an ornamental to replace turf

(French et al., 2001). Due to its high crude protein content (13 to 20%) combined with a

digestibility between 55 to 67% and high palatability, it produces a high quality forage

similar to alfalfa (Saldivar et al., 1990; Terril et al., 1996; French et al., 2001).

Perennial peanut (PP) is adapted to well-drained soils in the southern and Gulf

Coast areas of USA (French and Prine, 1991). After initial establishment, it is drought

tolerant, has excellent persistency under grazing because of the rhizomatous habit, and it

is not prone to insects, nematode or disease damage (Prine, 1981, Baltensperger et al.,

1986, French et al., 2001).

The most commonly grown cultivars are 'Florigraze' (Prine et al., 1981; Prine et

al., 1986) and 'Arbrock' (Prine et al., 1986). 'Arbrock' is considered to be more drought

tolerant, but it is less cold tolerant and may also decline if mowed frequently (Canudas et

al., 1989; French and Prine, 1991). Under non-irrigated conditions and/or dry conditions,









attaining full ground cover with the 'Florigraze' cultivar requires at least two to three

years (Prine et al., 1986; Williams, 1993; Johnson et al., 1994).

Despite being pest, disease, and drought tolerance, its requirements for vegetative

propagation combined with very slow initial growth hampers the use of PP as a cover

crop (CC) and/or forage crop (Coleman, 1995; Rice et al., 1996; Williams et al., 1997).

Perennial peanut typically produces very few or no sexual seeds. As a result, it is

exclusively propagated vegetatively by rhizomes that provide carbon reserves for shoot

growth during initial establishment and for regrowth in the spring (Saldivar, 1992a; Rice

et al. 1996). The first criterion for determining optimal planting time of PP is the need for

adequate soil moisture from rainfall and/or irrigation during the initial 2 to 3 months after

planting. Secondly, any frost period shortly after planting should be avoided (Williams,

1994a; Williams et al., 1997).

Due to changes in weed population dynamics during the conversion from

conventional to organic production, alternative weed management strategies may be

required (Lanini et al., 1994; Bond and Grundy, 2001). A national survey showed that

improved weed management was the number one research priority for organic growers

(Sooby, 2003). Lack of effective weed management practices pertinent to organic citrus

production systems thus may hamper successful transition from conventional to organic

citrus production (Mesh, personal communication).

Successful weed suppression via use of a perennial CC has been reported by

Bradshaw and Lanini (1995); Aguilar (2001); and Perez-Nieto et al. (2005) in coffee by

using Arachispintoi L. (non-rhizomal perennial peanut). It also provided suitable weed

control for heart-of-palm (Clement and DeFrank, 1998) and coconut (Mullen et al., 1997)









while perennial strawberry clover has been successfully used to suppress weeds in

vineyards (King and Berry, 2005).

Other perennial species such as bahiagrass (Paspalum notatum Flugge) and

bermudagrass (Cynodon dactylon (L.) Pers) are either planted and/or volunteer in grove

row middles and may reduce growth of weeds and prevent soil erosion (Rouse and

Mullahey, 1997). However, these systems require frequent mowing and/or chemical

control; they may harbor nematodes and citrus arthropod pests; and in many cases grasses

compete with citrus for water and nutrients (Rouse and Mullahey, 1997).

More recently use of PP as a groundcover for both vegetable crops (Roe et al.,

1994) and citrus (Coleman, 1995) has also gained attention. Due to its low water and

nutrient requirements, PP may provide an environmentally sound and ecologically

important component of sustainable citrus production in Florida (Mullahey et al., 1994).

Its use can minimize soil erosion and nutrition losses due to leaching and runoff, and

therefore, it can also enhance water quality (Woodard et al., 2002). When mowed 2 to 3

times a year, it may provide 60 to 112 kg N ha-1 yr1 to citrus trees. Alternatively, it can

provide a high quality and valuable forage crop and additional source of income for citrus

farmers during initial establishment of citrus groves (Coleman, 1995). Some citrus

growers may thus opt to use PP to also enhance the profitability of their citrus production

systems.

Although general production practices for PP are well-established, there is

relatively little information on the effective use of PP for weed suppression in citrus

(Mullahey et al., 1994). The majority of previous studies with PP have focused on

development of optimal planting strategies of pure stands for hay production in North and









Central Florida (Prine, et al. 1986; Williams et al., 1997; Freire et al., 2000). Moreover,

current practices for establishment of PP were typically developed for conventional

forage production (Williams, 1993; Rice et al., 1996; Valencia et al., 1997; Ruiz et al.,

2000; Williams et al., 2002). Although the use of PP as a CC in conventional citrus and

vegetable systems in south Florida has been studied extensively (Mullahey et al., 1994;

Roe et al., 1994; Coleman, 1995; and Rouse et al., 2001), there is no information on its

use to suppress weeds in organic production systems.

The overall objectives of this study were to 1) evaluate the effect of planting time

(spring vs. summer) on initial PP establishment, growth, and dry matter production; 2)

evaluate the effectiveness of over-seeding PP with crimson clover in the fall on system

performance and weed control; and 3) contrast the performance and weed growth

dynamics of PP-based systems with that of an annual CC system.

The corresponding hypotheses were 1) in organic citrus systems, planting PP

during the summer will increase the competitiveness of PP systems via enhanced initial

growth compared to spring plantings; 2) overseeding PP with crimson clover in fall will

help to increase the PP effectiveness in suppressing weeds; and 3) weed suppression with

annual CC is more effective than perennial peanut.

Materials and Methods

Set-up and Experimental Design

A one hectare field was planted with 'Hamlin', a processing orange cultivar [Citrus

sinensis (L.) Osb.] grafted on Swingle citrumelo (C. paradisi Macf. x P. trifoliata (L.)

Raf) during the summer of 2002 at the Plant Science Research and Education Unit in

Citra, Florida (29.68 N, 82.35 W). Tree spacing was 4.6 m in the row and 6.3 m between

rows. The main emphasis of this study was to evaluate initial growth of perennial peanut









(Arachis glabrata Benth. cultivar Florigraze) and its effectiveness in suppressing weed

growth in row middles compared to annual CC systems.

Prevailing soil types at the experimental site were a Candler fine sand (Typic

Quarzipsamments, hyperthermic, uncoated, 98% sand in the upper 15 cm) and a Tavares

fine sand (Typic Quarzipsamments, hyperthermic, uncoated, 97% sand in the upper 15

cm). The initial soil pH ranged from 4.8 to 5.1 and soil organic matter content was 9.3 g

kg-1. At the beginning of the experiment (fall of 2001), the soil was prepared by disking

before applying lime (2.5 Mg ha-) to the entire production block. Trees were irrigated

with microjet sprinklers with a 1.8-m spray diameter and a 1800 spray pattern positioned

0.4 m NW of the trees. During the winter the irrigation sprinklers were placed inside 0.6

m high PVC pipes and also used for frost protection. Row middles were non-irrigated in

order to evaluate the adaptation of different species under typical citrus orchard

conditions.

Chicken manure was applied exclusively to a 1.8 m wide strip straddling the tree

rows. Manure was applied during early spring at 4-11 Mg ha1 and application rates were

based on the estimated N mineralization rate, N content, and citrus tree age following

IFAS (Institute of Food and Agricultural Sciences) N-recommendations for newly

planted trees (Tucker et al., 1995). To enhance growth of winter cover crops, a non-

synthetic (mined) K2SO4 fertilizer approved by the Organic Materials Review Institute

(OMRI) was broadcast over the entire production block at a rate of 45 kg K20 ha- prior

to planting of winter CC. Due to a build up of residual soil P, use of chicken manure was

discontinued after 2004. Starting in 2005, an OMRI-approved 9-0-9 material derived

from feather-meal and potassium sulfate (Nature Safe, Griffin Industries, Cold Spring,









KY) was applied to tree rows following IFAS N-recommendations for newly planted

trees (Tucker et al., 1995).

A randomized complete block design was used with four replications and included

the following treatments: 1) Annual cover crop (ACC) included sunnhemp (Crotalaria

juncea L.) and/or cowpea (Vigna unguiculata L. Walp) planted in summer followed by

crimson clover (Trifolium incarnatum L.) and/or rye (Secale cereale L.) or a triple mix of

rye+crimson clover+radish (Raphanus sativus cv. Rufus) planted during fall (non-

perennial cover crop); 2) Perennial peanut (PP) planted in spring (PPsp); 3) Crimson

clover was planted in the fall of 2001 and was followed by PP planted in the summer of

2002, with plots being over-seeded with crimson clover during the fall (PPsu-os);and 4)

Fallow in spring (2002) and PP planted in summer 2002(PPsu). An outline of these

treatments is presented in Table 3.1.

Plot size was 18.9 m x 27.0 m and plots contained three row middles and two tree

rows of five trees each. For annual CC treatment (ACC) and PP-plots overseeded with

crimson clover in fall (PPsu-os), ground covers were planted with a "zero-till" planter

(Model Sukup 2100, Sukup Manufacturing Company, Shefield, IA) using appropriate

agronomic practices (Table 3.2). Mowing dates for both summer and winter CC are

outlined in Table 3.3.

Except for winter CC planted in 2003, when a zero-tillage system was used,

previous CC and/or weeds were soil-incorporated with 2 to 3 passes of a rototiller using a

tillage depth of 10 cm prior to planting a subsequent CC. 'Florigraze' perennial peanut

was planted in March and July 2002 for spring and summer treatments, respectively, with

a rhizome planting rate of 10 m3 ha-1, using a bermudagrass planter (Bermuda King









model #79, Kingfisher, OK) with a row spacing of 0.5 m. All the plots were

"cultipacked" after planting.

All annual leguminous CC were inoculated before planting with the appropriate

strain of rhizobium (Nitragin brand, Milwaukee, WI). Inoculants and untreated seeds

were obtained from local seed companies since most varieties from certified organic

commercial supply companies are typically poorly adapted to Florida conditions. Both PP

and annual CC were not irrigated during the growing season except after planting if soil

moisture was inadequate to ensure uniform germination. In this case, 25 mm of irrigation

was applied uniformly to the entire field to mimic a typical rainfall. Insect pests were

controlled when needed with allowed products in organic production systems following

the National Organic Program Standards (USDA, 2000). After initial establishment, plots

for PP-based treatments were mowed at 4-wk intervals during spring, summer, and fall.

After each sampling, PP-based systems were mowed.

Data Collection, Measurements, and Analysis

Representative sections of row middles were sampled at eight week intervals to

evaluate above-ground dry weights (DW) of both PP and weeds using rectangular

sampling frame with an internal surface area of 0.5 m2 for one representative sample for

each row-middle thus resulting in 3 samples per plot. Each sample area was selected in

such way that it closely matched the ground cover of both PP and weeds of the entire row

middle. Sampled areas included invariably a mix of weeds and PP. Weed biomass was

not segregated into different weed species.

Both PP and weeds were harvested at ground level. This approach was used since

the main focus was on weed growth dynamics. Weeds were separated from PP shoots but

individual weed species were not segregated. Perennial peanut shoots were separated into









stems and leaves and plant material was oven-dried at 65 C for 72 hours until constant

weight and dry weights were recorded. Perennial peanut leaf area was determined using a

Licor leaf area meter (LI-3000, Li-cor, Lincoln, NE) and used to calculate leaf area index

(LAI) values. Plant tissue material was ground in a Wiley mill through a 1-mm screen,

and a thoroughly mixed portion (ca. 4 g) was subsequently stored in scintillation vials.

Ground tissue was digested using wet-acid Kjeldahl digestion (Gallaher et al., 1975).

After digestion, samples were diluted, filtered, and analyzed for total Kjeldahl N at the

UF-IFAS Analytical Research Lab (University of Florida, Gainesville, FL) using EPA

method 351.2; (Jones and Case 1991).

In order to quantify the effectiveness of PP to suppress weed growth, we also

calculated cover crop weed indices (CCWI). This index expresses PP growth and

biomass production relative to weed growth. An overview of the qualitative interpretation

of this index is presented in Table 3.4.

Since the weed and PP biomass sampling approach differed from standard forage

sampling procedures, complementary forage productivity sampling was included during

2005. Overall forage productivity of PP was determined by taking representative row

middle samples from the PPsp, PPsu-os, and PPsu treatments. Perennial peanut and weeds

were cut at 6 cm of height above ground level for a 10-m-long and 0.53 m wide (5.3 m2)

sampling strip using a manual mower. Total fresh weight of the total harvested biomass

was recorded. A representative sub-sample of perennial peanut-weeds of 0.5 m2 was

harvested in a similar fashion. This subsample was segregated into perennial peanut and

weed shoots, and plant materials were oven dried at 65 C for 72 hours until constant

weight before recording dry weight.









Analysis of variance was performed on all data using Proc Mixed of the Statistical

Analysis Systems (SAS) software (SAS Inst. Inc., 2002). Since growth characteristics of

PP were different from those of ACC, two separate statistical analyses were conducted.

During the first analysis, perennial peanut treatments were compared among each other,

during a subsequent analysis the PPsu, which turned out to be the superior PP treatment,

was contrasted with the ACC treatment. Shoot dry weight for annual cover crops (DWacc)

and perennial peanut (DWpp); corresponding shoot N accumulation (Nacc and Npp) and

leaf area index (LAIacc and LAIpp), number of PP shoots (Shoot#) along with weed dry

weight (DWwd), N accumulation in weeds (Nwd), and cover crop weed index (CCWI)

were evaluated. The abbreviations outlined here will be used throughout the remainder of

this chapter. If significant interaction (P<0.05) occurred between year, treatment, season,

and/or sampling time, specific effects were tested and shown separately. The LSMEANS

procedure adjusted by Tukey test (P<0.05) was used to compare treatment means.

Results

Monthly rainfall during the PP growing season ranged from 901 mm in 2002 (drier

year) to 1522 mm in 2004 (Table 3.5). Since year by treatment interaction terms were

significant (Appendix A-1), results are presented separately for each year.

Perennial Peanut 2002

Rainfall was very low and unevenly distributed during the 2002 spring growing

season (56 mm total), so obvious water stress occurred during the establishment of PP in

spring 2002 (PPsp).

Although the summer PP planting occurred 3 months after the spring planting,

initial growth for summer plantings was better and overall crop growth was similar

within 2 months after the summer planting (Table 3.6). Due to continuous growth, fall









shoot#, LAIp, DWpp, and Nacc-pp were greater for summer plantings, while PPsp

treatments did not show a significant increase in growth after initial establishment. Due to

the rototilling at planting, weed biomass and Nacc-wd were lower for summer plantings.

As a result, the CCWI were greatest for summer plantings. However, values decreased

over time and during the fall overall weed growth parameters were similar for all

treatments.

Perennial Peanut 2003

Although rainfall was relatively high in 2003 in comparison with 2002, it was not

evenly distributed (Table 3.5). Overall shoot#, LAI, DW, and Naccpp values were lowest

during the spring and similar for summer and fall samplings (Table 3.7). Although no

difference in shoot# and LAIpp occurred between summer plantings during 2002 and the

spring of 2003, leaf area expansion and dry weight accumulation for PPsu-os during the

summer and fall of 2003 were lower compared to the PPsu treatment. Toward the fall of

2003, the reductions in shoot# and LAIpp for the PPsp compared to PPsu were 89% and

91%, respectively. Dry matter (DM) accumulation for the PPsu-os and PPsp treatments were

32% and 88% less than the PPsu treatment, and the reductions in Naccpp for the PPsu-os and

PPsp compared to the PPsu treatment were 31% and 90%, respectively.

Due to the predominance of perennial grasses and use of frequent mowing, overall

DWwd and Nacc-wd values were relatively constant throughout the year. During the spring

and summer of 2003, overall DWwd and Nacc-wd values were greatest for the PPsp

treatment and not affected by overseeding. However, during the course of 2003 the

summer-planted treatments started to diverge (as indicated by differences in DW). By the

fall of 2003, CCWI values for the PPsu-os and PPsp treatments were reduced by 43 and

93%, respectively compared to the PPsu treatment.









Perennial Peanut 2004

During 2004, total rainfall was the highest of all four years (Table 3.5) and rainfall

was also relatively evenly distributed and no obvious water stress occurred. In general,

shoot# was relatively constant across seasons, while LAI and Nacc-pp had the lowest

values in spring. The PPsu treatment performed best and had the highest shoot#, LAI,

DW, and Nacc-pp, PPsu-os had intermediate values, while PPsp performed poorly (Table

3.8). The reductions in LAIpp, DWpp, and Nacc-pp, for the PPsu-os and PPsp compared to

PPsu were 58 and 97%, respectively, by the fall of 2004.

During the fall of 2004, overall DWwd and Nacc-wd values were greatest for the PPsp

treatment and values were not affected by overseeding during the late fall/winter for PPsu-

os treatment. However during the course of summer and fall of 2004 the summer

treatments started to diverge similar as was the case during the fall of 2003 (Fig. 3.2).

Compared to PPsp treatment, PPsu and PPsu-os had 42 and 23% less weeds, respectively

(Table 3.8). For PPsu, CCWI was greatest, while PPsp and PPsu-os had similar and

relatively low CCWI values. Overall CCWI values for PPsu were highest during the

summer and fall. By the fall of 2004, CCWI values for the PPsu-os and PPsp treatments

were 68 and 98% lower compared to the PPsu treatment.

Perennial Peanut 2005

Rainfall in 2005 was relatively high (1370 mm) and evenly distributed (Table 3.5).

Shoot# was similar across the season for PPsp and PPsu-os, while PPsu showed an increase

in shoot# during the summer (Table 3.9). Overall LAIpp and DWpp values were lowest in

spring for summer plantings while spring crops had overall low growth. The Nacc-pp

value for PPsu was lowest in spring. During the summer and fall, PPsu had the best growth

and highest DW and N content, PPsu-os had intermediate values, while PPsp again









performed poorly across the entire season (Table 3.9). Toward the fall of 2005, the

reductions in shoot# for the PPsu-os and PPsp compared to PPsu were 65 and 96%,

respectively. Overall LAI, DW, and Nacc-pp values were reduced by about 62 and 97%,

for the PPsu-os and PPsp, respectively.

During the fall of 2005, overall DWwd and Nacc-wd values were greatest for the PPsp

treatment in fall while weed growth with summer planted PP was not affected by

overseeding. However, during the course of summer and fall of 2005 the results with

summer treatments started to diverge again similarly as was the case during the later part

of 2004 (Fig. 3.2). For PPsu and PPsu-os treatments, weed dry weight was 39% and 24%

lower than PPsp, respectively (Table 3.9). The cover crop weed index (CCWI) was

greatest for PPsu while values were similar for PPsp and PPsu-os. Overall CCWI values did

not differ across different seasons. By the fall of 2005, CCWI values for the PPsu-os and

PPsp treatments were reduced by 64 and 98%, respectively compared to the PPsu

treatment.

Perennial Peanut Productivity (2005)

Perennial peanut productivity was greatest for PPsu treatment, intermediate for PPsu-

os, whereas PPsp had the lowest productivity. Compared to the PPsu treatment, cumulative

biomass values for PPsu-os and PPsp were reduced by 76 and 97%, respectively (Table

3.10). It should be noted that yields presented in Table 3.10 are based on solid PP stands.

However, PP only covered 64% of the orchard area and PP production on a total land

area basis would be about 2.1, 0.5, and 0.06 Mg ha-1 for PPsu, PPsu-os, and PPsp,

respectively.

The PPsu treatment had the lowest weed biomass, followed by PPsu-os and PP sp,

however weeds made up 98% of the DW content for the PPsp treatment. Corresponding









values of contribution from PP to the total DW were 50, 13, and 2% while corresponding

PP treatments accounted for 67, 23, and 3% of overall N accumulation (Table 3.11).

These results followed a similar trend to those from the growth analyses (Tables 3.6 to

3.9).

Annual Cover Crops (2002-2005)

Annual CC outperformed PP treatments and had greater N and DW content and

were also more effective in suppressing weeds across all the seasons and years, except in

2004 when 'Cream-40' cowpea was used as CC (Tables 3.6 to 3.9). Overall LAIacc,

DWacc, and Nacc values were greater for summer CC (sunnhemp and cowpea) than for

crimson clover, except in 2004. The values in LAIacc, DWcc, and Nacc had the following

order: sunnhemp>crimson clover>'Cream-40'. The three-way winter CC-mix (winter

rye, crimson clover, and radish) generated 7.4 Mg ha-1 and outperformed mono-cropped

systems. Overall Nacc values for winter CC systems ranged from 38-40 kg N ha-1

(crimson clover) to 121 kg N ha-1 (3-way mix). Corresponding values for summer CC

systems were: 25 kg N ha-1 cowpeaa) to 103-201 kg N ha-1 (sunnhemp).

Winter CC triple mix, sunnhemp, crimson clover, 'Cream-40' cowpea, PPsu, and

PPsu-os reduced overall weed growth by 96, 92, 67, 67, 41, and 24% respectively in

comparison with PPsp. Overall DWwd and Nwd values were consistently lower in annual

CC when compared with PPsu, which was the best performing PP system. Overall DWwd

and Nwd followed the order: 'Cream-40' cowpea> crimson clover> sunnhemp> winter CC

triple mix. The CCWI score varied from outstanding weed control (CCWI>20) for triple

mix and sunnhemp, to moderate weed control (1
weed control for 'Cream-40' cowpea. Across 2002, 2003, and 2005 and compared to the

PPsu system, annual CC systems had 1200, 2090, and 1800% greater LAI values and DW









and N content, respectively. In comparison with annual CC systems, weed DW and N

content was 1067 and 772% greater for the PPsu systems.

System Dynamics

During 2005, total (PP+weed) biomass and N accumulation in PP-based systems

as affected by season were as follows: summer>fall>spring, whereas total cumulative

biomass and N were similar for all the treatments. In comparison with PPsu system, PPs0-os

and PPsp systems had 76 and 97% lower DWpp and Nacpp values, respectively. Overall

reductions in weed growth were 15 and 49% for the PPsu-os and PPsu systems, respectively

(Table 3.11). Summer CC system accumulated a similar amount of biomass and N in

comparison with winter CC (triple mix) system. Percentages of contribution of CC to the

total system DW and N content were greater for annual CC (98% and 99%, respectively).

Corresponding values for PP-based systems were very low for PPsp (2%), low for PPsu-os

(14-19%), and intermediate for PPsu (50-55%). When comparing annual CC-based

systems with the best PP-based system (PPsu), annual CC system had 114% greater DW

production capacity and also accumulated 50% more N.

Across years (2002-2005), summer CC produced 6.2 Mg ha-1 per season or 53 kg

ha-1 d-1. Winter annual CC averaged 3.2 Mg ha-1 per season or 20 kg ha-1 d-1. These

results contrasted with PP DW, which averaged 0.1, 0.7, and 2.3 Mg ha-1 for PPsp, PPsu-

os, and PPsu, respectively. When weeds were included in the balance, total biomass

production was 6.8 Mg ha-1 or 57 kg ha-1 d-1 for warm-season annual CC. Cold-season

annual CC and PP-based systems had average values of 3.5 and 7.5 Mg ha-1, or 23 and 31

kg ha-1 day-', respectively.

If we consider that almost all the PP stand was lost in PPsp, we could regard this

treatment as a grass fallow or control, then the average across all the years in percentage









of weed suppression in PPsu and PPsu-os systems were 40 and 26%, respectively,

compared to PPsp. However this level of weed suppression is low in comparison with the

93% weed suppression achieved in summer and winter annual CC. There was an inverse

correlation between PP DW and weed DW (Fig 3.4) and between PP DW and N

accumulation in weeds (Fig 3.5) in summer and fall.

Discussion

Low initial LAI and DW production for the current PP study may be related to

slow initial growth (Prine, 1986). In addition, lack of supplemental irrigation combined

with low organic matter content, and poor water retention capacity of the sandy soil may

have further hampered initial growth and leaf area expansion.

Because of the cost of plant material, sprigs are typically planted in strips with a

row spacing of 0.5 m as a standard practice (French et al., 2001). Under our conditions,

complete row closing only occurred during the third year of growth, probably due to

pronounced weed competition during the first years. These findings agree with those

reported by Williams (1993); however Ruiz et al. (2000) reported row closing within one

year. However, in that case, mechanical and/or chemical weed control along with

supplemental fertilizers and irrigation applications were used. The latter are not feasible

for citrus row middles and synthetic herbicides are not permitted in organic production.

The relatively high shoot number during the two first years after planting for the

PPsu and PPsu-os systems was probably related to more favorable initial soil moisture

conditions for summer plantings and warmer soil temperatures. Although the PPsu

treatment had the highest LAI and DW across the years, it still did not perform as well as

the annual cover crop system.









The consistently poor performance of the PPsp system may have been related to

erratic rainfall and prolonged dry period during its initial establishment. Similar results

were reported by Williams et al. (1997) and Saldivar et al. (1992b), who concluded that

adequate soil moisture during the initial 2 or 3 months after planting is the most critical

factor for PP survival. It should be noted that citrus rows middles are typically not

irrigated, and that citrus soils typically also have lower water holding capacities

compared to soils commonly used for pasture systems.

The increased divergence between different planting systems over time may be

related to the PPsp treatment failing to develop a critical density required to effectively

compete with weeds. In the absence of overhead irrigation, initial growth for the spring

planted system was very poor and the few sprouts that grew often senesced within the

first few weeks, similar to the findings reported by Williams (1993, 1994a). As a result,

overall stands were very erratic and in many cases perennial grasses and weeds prevented

effective PPsp establishment, and maximum observed DW for this treatment did not

increase over time. Delaying planting until the onset of the summer rains improved initial

establishment and increased initial growth and appeared to be a more viable strategy for

citrus systems. Similar recommendations were made for forage systems by Williams et

al. (1997), whereas French and Prine (1991) proposed that January to March was the best

time for PP planting.

For summer plantings, overall maximum DWpp, LAIpp, and Npp occurred during

early fall, which is in agreement with findings by Ocumpaugh (1990). Although

overseeding did not affect initial growth of summer plantings, it appeared to hamper plant

growth during subsequent years. The zero till planter used for cover crops may have









caused some damage to the rhizomes. Alternatively, it could be argued that planting

crimson clover during the fall may have reduced initial regrowth during early spring

since crimson clover may persist up to April/May. As a result, regrowth of PP may be

slower and maximum productivity did not occur until fall. This in turn may have affected

assimilate storage in rhizomes and subsequent regrowth as suggested by Saldivar et al.

(1992a) and by Rice et al. (1996).

The rationale for over-seeding PP in fall was to maintain a ground cover during the

winter when PP is dormant and to also add additional N to the cropping system.

However, since DWp for the PPsu-os treatment was less than PPsu and weed pressure was

not reduced, there is no justification for the extra cost associated with overseeding with

crimson clover during the fall for this system. Dunavin (1990, 1992), on the other hand,

reported that over-seeding PP with rye or a rye grass mixture and crimson clover for the

cool season provided a superior cropping system that had no negative impact on

subsequent growth of PP. However, in these studies, the companion CC were broadcast

instead of planted in rows while in our case, the knives of the zero-till planter may have

caused some injury to the PP rhizomes. Additional research may be needed to assess

whether broadcasting without mechanical soil incorporation would enhance system

performance without hampering PP growth.

The continuous increase in maximum observed DWpp values for the PPsu treatment

appears to be associated with higher shoot densities resulting from more favorable

conditions for initial growth and rhizome formation. Calculated cumulative productivity

was low compared to reported potential PP hay yields of about 8 Mg ha-1 in conventional

pure PP stands (Ocumpaugh, 1990; Johnson et al., 1994; Terril et al., 1996). Relatively









low yields for our studies could be explained by the competition between PP and grassy

weeds (Dunavin, 1992). Canudas et al. (1989) concluded that grass weeds hampered PP

establishment and reduced PP yield by 50% due to competition. As a result, PP storage

reservoirs may become depleted, resulting in reduced re-growth and poor performance in

following years (Saldivar et al. 1992a, Williams, 1994b). Low inherent soil K levels may

also have hampered PP-rhizobium activity as suggested by Slattery et al. (2001). The

limited soil water storage capacity combined with the lack of supplemental irrigation may

also have resulted in additional reductions in growth and productivity as discussed above.

Dunavin (1992) and Valencia et al. (1999) reported similar or lower yields for PP mixed

with grasses.

The shift in maximum N accumulation from summer to fall for the PPsu-os treatment

could have been related to shading of perennial peanut by crimson clover. Although N

concentration in PP tissues were within expected range of 21 and 29 g N kg-1 (Saldivar et

al., 1990; Terill et al., 1996; Venuto et al., 2000), the overall N content was much lower

than the 192 and 162 kg N ha-1 calculated from the data provided by Ocumpaugh (1990)

and Terril et al.(1996) for pure PP pure stands. This discrepancy may be related to weeds

competing for light and nutrients in mixed stands thereby reducing productivity of PP as

suggested by Dunavin (1992). The low inherent initial soil fertility of the field site may

also have resulted in poor performance of rhizobium bacteria symbiont, as was proposed

by O'Hara (2001).

The reduction in DWwd for the PPsu provides evidence that this treatment was

relatively more successful in competing with perennial weeds. The overall effectiveness

of PP in suppressing weeds was inversely related to DWpp (Fig 3.4) with correlation









coefficients r=-0.78 and -0.72 for PPs and PPsu-os, respectively, which may be related to

resource competition between PP and weeds (Dunavin, 1992; and Valencia et al., 1999).

Lower DWwd during the summer 2002 was probably due to the effect of soil tillage

on weed biomass. For all treatments DWwd was greatest during the fall, except for 2005

when values peaked during the summer. Although the PPsu systems had a relatively high

initial weed biomass compared to the PPs0-os system, within a year this trend was

reversed. This underlies the observation that over-seeding did not enhance weed

suppression. It may be possible that weed species that are effectively being suppressed by

crimson clover did not prevail during the summer, whereas weed species competing with

PP during the summer may also be dormant during the winter and thus were not greatly

affected by over seeding. Alternatively, the rhizomes of perennial weedy grasses may

have been more tolerant to potential injury of the zero-till planter.

Reduction in weed suppression by annual CC in the early spring 2004 may be

related to the use of the zero-till system initiated with the planting of winter CC in 2003.

The spike in weed growth observed in the fall of 2004 (Fig. 3.2) was related to the use of

'Cream-40' cowpea, a precocious variety with poor canopy persistence, that was

relatively ineffective in suppressing weeds.

Observed weed biomass values for PP-based systems during the first year were

similar to those for conventional plots treated with pre-emergence herbicides only

(Canudas et al., 1989). The PPsp system did not show appreciable decrease in weed

growth over time, which was consistent with DWpp not increasing over time, suggesting

that PP requires a critical initial density to effectively suppress weeds.









It should be noted that a key weed management strategy for this trial was to

withhold N-based nutrient sources from row middles and thereby provide PP with a

competitive edge. This strategy appeared to work for the PPsu system, which showed a

gradual decline in weed growth over time. Based on field observations, frequent mowing

during the first year greatly reduced the incidence of broadleaf weeds. However, it also

promoted the growth of perennial grasses such as bahia and bermudagrass, similar to the

results reported by Wright et al. (2003). In areas with higher inherent soil fertility,

bermudagrass and broadleaf weeds grew more vigorously and out-competed PP, and this

effect appeared to be most prominent for spring plantings.

The proportion of weeds (grasses) to PP on a DW basis was about 70-30 in PPsu in

2004 and 2005, which was similar to the values reported by Dunavin (1992) in the fourth

year of a mix of 'Pensacola' bahiagrass-PP. The proportions were 99-1 and 90-10 for

PPsp and PPsu-os systems, respectively. The low PP components in these systems reflect

the negative effect of the management practices associated with these systems on the

overall competitiveness of PP as explained above. In 2005, for the productivity trial, this

ratio (weeds:PP) was lower (50:50) in PPsu which may be related to the cutting height

used (Ocumpaugh, 1990).

Similar to DWwd, Nwd was greatest for the PPsp treatment. The lower weed DW and

N content for the PPsu treatment may be related to PP start attaining dominance in this

system due to more favorable rainfall distribution during initial establishment as was

proposed by Williams (1994a).

High CCWI during the first sampling, was related to the short term effects of

mechanical tillage on weeds. However over time, weeds began to dominate the system









again because they were more competitive than PP. Repeated mowing favored prevalence

of C-4 grass species including bahia and bermudagrass, which have higher growth rates

compared to PP which is a C-3 plant (Paterson et al., 1996, Newman et al., 2005).

Although PPsu initially did not compete well with grass weeds, over time PPsu

gradually became more competitive possibly due to the fact that it is more drought

tolerant and can prevail in low nutrient environments (French et al., 2001). As a result, its

CCWI thus gradually increased over time as PP gained a competitive edge over the

grasses. On the other hand, since PPsp shoots often senesced before they reached full size,

they could not contribute to restoring carbohydrate reserves of the rhizomes during the

fall and this was reflected by a gradual decline in DW and CCWI over time. As a result,

PP storage reservoirs may have become depleted (Saldivar et al., 1992a) resulting in poor

re-growth and performance in following years, giving weeds a competitive edge. It could

be argued that a critical mass of initial growth is required for PP to invest in rhizome

storage reserves expansion in order to develop dominance over time, as discussed above.

The results of the PP productivity trial were low (3.3 Mg ha-1) compared to the 8

Mg ha-1 reported by Ocumpaugh (1990), Johnson et al. (1994), and Terril et al. (1996).

However, these studies featured pure PP stands (compared to 50% PP in our study) and

their plots were treated with chemical herbicides and fertilizers. Relatively low PP yields

could be explained by the competition between PP and grass weeds, as suggested by

Canudas et al. (1989) who reported that competition from grass weeds reduced PP

establishment and PP yield by half. Dry weight of the PPsu treatment nearly matched the

results obtained by Dunavin (1992) and Valencia et al. (1999) for PP mixed with grasses.









The low productivity for the PPsu-os was probably due to over-seeding PP with

crimson clover with the zero-till drill damaging rhizomes and/or depleting carbohydrates

reserves which are critical for vigorous regrowth during a subsequent spring season

(Saldivar et al., 1992a; Rice et al., 1996). Moreover, it may be possible that crimson

clover reduced light availability to newly emerged PP sprouts during the spring. Another

factor that could account for this relatively low productivity is the low initial soil K

content (see chapter 5 for more details about soil fertility dynamics), which could have

hampered PP-rhizobium activity as proposed by Slattery et al. (2001). According to these

results, contrary to those by Prine et al. (1981) and Williams (1993), planting PP in spring

in the absence of supplemental irrigation, increases the risk of poor stands on poor soils

and/or during dry springs as was the case for the 2002 planting.

For annual CC, average sunnhemp biomass accumulation was similar to the 9-11

Mg ha-1 reported by Steinmaier and Ngoliya (2001); Ramos et al. (2001); Perin et al.

(2004) under tropical conditions, and superior to the results from Balkcom and Reeves

(2005) and Jeranyama et al. (2000) for subtropical regions.

The dry matter accumulation by 'Cream-40' was low compared to other cowpea

varieties (Jerenyama et al., 2000; Muir, 2002). This was related to the short season and

precocious reproductive cycle of this cultivar, because after 6 weeks almost all the pods

were formed and at 8 weeks all the foliage had senesced, allowing light to penetrate to

the soil surface and decreasing weed suppression (Fig 3.2-3.3 and Table 3.8).

The average DW accumulation of crimson clover (1.7 Mg ha-1) was lower than the

2.5-4.9 Mg ha-1 reported previously (Dyck et al., 1995; Daniel et al., 1999; Odhiambo

and Bomke, 2001; and Schomberg and Endale, 2004).









Superior performance of mixed winter CC systems is in agreement with findings of

Karpenstein-Machan and Stuelpnagel (2000) for mixed rye-crimson clover and rye-

winter pea (Lathyrus hirsutus L.) systems in Germany and reports by Juskiw et al. (2000)

for mixed small grain cereals systems in Canada. The excellent performance of the mixed

system may be related to the synergistic combination of complementary characteristics of

the constituents of the mix (Kabir and Koide, 2002). Winter CC mixes also had a higher

N accumulation which may be related to the combination of enhanced N retention of

deep rooting and fast growing species (rye and radish) with additional N-fixation by

crimson clover (Justes et al., 1999; Vos and van der Putten, 2001; Kristensen and

Thorup-Kristensen, 2004).

The overall greatest N accumulation for sunnhemp is in agreement with reports by

Balkcom and Reeves (2005) while the poor performance of 'Cream-40' is related to its

short growth cycle which makes it more suitable for short-term fallows. The relatively

low (40 kg N ha-N ) N content of crimson clover compared to other studies (Daniel et

al., 1999; Odhiambo and Bomke, 2001), was probably due to the low pH and poor initial

soil fertility hampering rhizobium colonization (Slattery et al., 2001).

The effectiveness of annual CC in suppressing weeds compared to PP may be

related to both higher growth rates of annual CC, the use of mechanical tillage disrupting

weed growth cycles, and allelopathic action of winter rye (Reberg-Horton et al., 2005).

The superiority of weed suppression for annual CC, especially by the triple mix

(rye+crimson clover+ radish) and sunnhemp was probably related to superior resource

pre-emption by these annual CC systems (Craine et al., 2005).









In terms of system dynamics, higher total biomass found in annual CC systems was

related to the partial dormancy observed in C4 grasses and PP during the winter season.

In contrast, winter CC are well adapted to low temperatures (Qi et al., 1999; Teasdale et

al., 2004), and the higher overall N accumulation in annual CC systems was thus partly

due to the contribution of winter CC. Therefore, use of winter CC may enhance nutrient

retention and soil C sequestration, thereby outperforming PP-based systems while also

providing superior weed suppression compared to PP-based systems.

As discussed in Chapter 2, increased N accumulation in the system during the

summer or winter season could be mineralized later and benefit either citrus trees or

subsequent CC. However, strategies should be developed to avoid potential N leaching in

sandy soils during the winter by including NO3 trap crops such as rye and/or radish as

suggested by Justes et al. (1999), Vos and van der Putten (2001), and Kristensen and

Thorup-Kristensen (2004).

Weeds contributed significantly to DW and N accumulation in PP-based systems,

underlining their important role of capturing N in the system and reducing environmental

risks associated with potential N leaching (Vos and van der Putten, 2001; Woodward et

al., 2002). However, potential benefits of specific management practices of PP-based

systems in reducing environmental impacts should be evaluated in more detail in future

research.

Although PPsu had low CCWI values and did not effectively suppress weeds during

the first years, it is able to persist under adverse conditions (French and Prine, 1991). As

was expected, over time PP gradually became more competitive, although its overall

performance was still inferior to annual CC. However, in addition to weed suppression









other potential advantages of PP include its potential for provide additional income as

hay and its use within an integrated systems with animals such as sheep or goats (French

et al., 2001). However, PP also has some clear drawbacks such as a relatively high

establishment cost ($400-900 ha-1), very slow initial growth (it takes 2-3 years to

complete establishment), poor initial weed suppression and the requirement for frequent

mowing (Coleman, 1995; Rice et al., 1996; Williams et al., 1997). Since it is very

important to ensure a clean and weed-free seed bed for PP planting (Williams, 1993;

Williams et al., 1997) use of repeated tillage followed by CC crops such as sunnhemp

and winter rye may be beneficial to reduce weed populations in organic systems for a

minimum of one year prior to planting perennial peanut. In terms of PP systems, we also

propose the use of an integrated management system with PP being planted in early

summer in row middles following repeated rototilling of a winter rye CC crop. Annual

compact, self-reseeding CC can be planted near young trees, complemented with manure

or natural fertilizer amendment applied to the tree rows only. Once the trees are 5-6 years

old and the perennial peanut is established, sheep can be introduced in the system to

graze the row middles.

Introducing sheep may reduce labor and energy requirements for maintaining a

short canopy and we aim to test this system as soon as the trees reach a tree height of 3-

3.6 m. In the mean time, we aim to establish self-reseeding cover crops such as crimson

clover (winter) and alyceclover or cowpea (summer) in the three rows to reduce labor

requirements. Additional research is needed to assess the fate of N (immobilization and

release) in the different N-soil pools derived from either CC or added manures or

fertilizers in these sandy soils, supported by some lysimeter and/or resin trap studies, in









order to evaluate potential leaching and environmental risks from these management

practices.

Conclusions

In general PP had slow establishment, and spring plantings were severely hampered

by lack of adequate soil moisture, while competition with weeds and grasses resulted in

erratic initial growth and poor stands. Under our production settings, planting PP after the

onset of the rainy season resulted in better initial stands and more effective weed control.

Initial weed suppression by PP was very poor to poor, which was due to its slow initial

growth and high weed pressure. Overseeding PP with crimson clover in fall reduced PP

vigor and its effectiveness in suppressing weeds.

Compared to PP, annual CC provided much better weed control, especially when

species were used that have allelopathic properties (rye) and/or retain adequately dense

canopies for prolonged periods of time (rye and sunnhemp). For both PP and ACC, weed

biomass was typically inversely related to DW content of either PP or ACC probably due

to competition for light, water, and nutrients. Presence of leguminous CC increased

overall N accumulation, but weeds also contributed to enhanced N retention and nutrient

cycling.









Table 3.1. Overview of experimental treatments during 2002-2005.
Season
Year Treatments
Spring Summer Fall
Annual CCO Fallow1 Sunnhemp (SH) Crimson Clover
(CrC1)
PPspI PP Fallow1 Fallow1
2002
PPsu-osU PP PP PP/ CrCl

PPsu PP PP PP

Annual CC Fallow SH + Cowpea* Crimson Clover

PPsp Fallow1 Fallow1 Fallow1
2003
PPsu-os PP PP PP/ CrCl

PPsu PP PP PP

Annual CC Fallow Cowpea Rye + Radish + CrCl

PPsp Fallow1 Fallow1 Fallow1
2004
PPsu-os PP PP PP/ CrCl

PPsu PP PP PP

Annual CC Fallow SH + Cowpea* Rye + Radish + CrCl+
hairy vetch
2005 PPsp Fallow1 Fallow1 Fallow1

PPsu-os PP PP PP/ CrCl

PPsu PP PP PP

' Spring plantings of perennial peanut (PP) were not successful in most of the plots and served as
a partial control instead. These fallows were maintained by frequent mowing (every 3-4 weeks).
Annual cover crops: crimson clover, sunnhemp and cowpea. t Perennial peanut (PP) planted in
spring. I PP planted in summer, the following years PP was over-seeded with crimson clover in
fall. PP planted in summer. Sunnhemp planted in the center of row middles surrounded by
cowpea.









Table 3.2. Overview of seeding rates and row spacing for annual summer and winter
cover crops used between 2002 and 2005

Cover crop Row spacing Seeding rate Inocula Variety
(m) kg ha-1

Sunnhemp 0.36 40 Cowpea strain --

Cowpea 0.36 56 Cowpea strain Iron clay

Crimson clover 0.18 28 Trifolium Dixie


strain


0.36


0.18


112



22


Abruzzi ('02-
'03&'05)
Florida 401
(2004)

Rufus


Table 3.3. Outline of planting and harvest dates and duration for summer and winter
cover crops.
Summer Winter

Year Planting Mowing Duration Planting Mowing Duration
(days) (days)

2002 30 Jul 11 Oct 102 1 Dec 15 May 165

2003 10 Jun 16 Oct 127 28 Oct 31 March 154

2004 11 Jun 10 Oct 121 1 Nov 5 April 156

2005 21 Jun 25 Oct 125 5 Dec 9 May 155


Winter rye


Radish









Table 3.4. Outline of cover crop weed index (CCWI) categories.
CCWI value Cover crop Weed Weed control

< 0.5 CC not competitive Weeds dominate Very poor (>70%
weeds)

0.5-1 CC coexist Weeds coexist Poor

1-3 CC start prevailing Weeds prevail in niches Moderate

3-5 CC prevail Weeds fail to dominate Adequate

5-15 CC predominate (70-90%) < 10-30% weeds Excellent

>15 CC dominate completely <5% weeds Outstanding

It is assumed if CCWI >15 then weed control is considered outstanding since weeds only
cover account for less than 5% of the total biomass. It should be noted that in the absence
of weeds the CCWI will approach infinity, and the upper boundary is thus not defined.

Table 3.5. Rainfall measured in the Plant Science Research and Education Unit (Citra)1
during 2002-2005.
Year

Month 2002 2003 2004 2005

----------------------------- Rainfall (mm)--------------------------
January 61 4 44 23
February 26 129 143 65
March 35 182 55 121
April 21 14 25 148
May 0 33 70 163
June 135 238 142 197
July 105 130 272 102
August 153 148 160 196
September 122 101 420 102
October 15 114 117 121
November 67 46 35 58
December 160 22 39 75
Total 901 1162 1522 1370
1Data obtained from the website http//fawn.ifas.ufl.edu on 1/25/2006










Table 3.6. Effect of planting time and over-seeding of perennial peanut (PP) on shoot number m2 (Shoot#), leaf area index (LAIpp), shoot dry
weight (DWpp), and N accumulation (Nacc-pp); weed dry weight (DWWD), N accumulation in weeds (Nacc-WD), and cover crop weed
index (CCWI) in 2002. LAI, DW, Nacc, DWx,and Nacc-xw for Annual Cover Crops (ACC) are included for purpose of comparison.
Perennial Peanut


Treatment


Shoot#


Naccpp


Spring Summer Fall
------------ # m ------
12 a A b 15 b A
33 aB 70 aA
29 aB 57aA


Spring Summer
------------- m2 m-2
0.03aA
0.02 a B
0.02 a B


Fall
-----------
0.03 b A
0.21 aA
0.16 aA


Spring Summer Fall
------------ Mg ha -------------
0.04 aA 0.05 bA
0.02 aB 0.29 aA
0.02 aB 0.28 aA


Spring Summer
------------ kg N ha1
0.8 aA
0.7 aB
0.8 aB


1.6 B ND


Spring Summer
----------- Mg ha'1
2.3 aA
0.1bB
0.1bB


Fall
-------------
3.6aA
3.7aA
4.8 aA


Spring Summer Fall
------------ kg N ha -------------
32.2 aA 48.1aA
1.7 bB 39.0 aA
1.2 bB 47.6 aA


Spring Summer Fall
----------- Mg Mg--------
0.02 bA 0.03 aA
0.31 aA 0.11 aB
0.35 aA 0.12aB


ACC ND 0.3 ND 5.3 ND 25.1
ACC vs PPsu ** ** **
t Perennial peanut (PP) planted in spring. I PP planted in summer, the following years perennial peanut was over-seeded with crimson clover in
fall. PP planted in summer. Means within the same column followed by the same lower case letter and means within the same row followed by
the same upper case letter do not differ statistically by the LSMEANS adjusted by Tukey test (P<0.05). **Contrast between ACC and PPsu,
significant at P=0.01.


PPsp,
PPsu-os


ACC
ACC vs PPsu


DWwd


Fall
-----------
1.5 bA
6.6aA
6.6aA


pp t
PPsp,
PP .. -os
PPsu
PPSU


2.74
**
Weeds
Nacc-wd


7.72 A
**


CCWI













Table 3.7. Effect of planting time and over-seeding of perennial peanut (PP) on shoot number m (Shoot#), leaf area index (LAIpp), shoot dry
weight (DWpp), and N accumulation (Nacc-pp); weed dry weight (DWwD), N accumulation in weeds (Nacc-WD), and cover crop weed
index (CCWI) in 2003. LAI, DW, Nacc, DWx,and Nacc-xw for Annual Cover Crops (ACC) are included for purpose of comparison.
Perennial Peanut


Shoot#

Spring Summer Fall
------------ # m2 -----
4 bA 12 bA 1 bA
35aB 88aA 82aA
60aA 107aA 127 aA


Nacc-pp


Spring Summer
-------------2 m2
0.00aA 0.03 c A
0.03aB 0.13bA
0.05 aB 0.19 aA


0.95 B
**


DWwd


Spring Summer
----------- Mg ha-1
3.5aA 3.4aA
2.5 bA 2.5 b A
2.8 bA 2.6 b A


Fall
-------------
3.2aA
2.9aA
2.5 b A


Fall
-----------
0.02 c A
0.13 b A
0.19aA


Spring Summer Fall
------------ Mg ha -------------
0.00 aA 0.04cA 0.04cA
0.02 aB 0.21 bA 0.23 b A
0.05 aB 0.31aA 0.34aA


3.12 A 1.8 B
** **
Weeds
Nacc-wd


Spring Summer Fall
------------ kg N ha -------------
44.8 aA 46.3 aA 46.5 aA
34.6aA 35.0aA 32.4bA
37.8aA 32.6aA 25.6bA


Spring Summer
------------ kg N ha1
0.2aA 1.1cA
1.0 aB 5.0bA
2.0aB 7.2aA


9.6 A 40.3 B
** **


Spring
-----------
0.00 a B
0.01 aB
0.02 a B


CCWI

Summer Fall
Mg Mg -------------
0.01lcA 0.01lcA
0.08 bA 0.08 b A
0.12aA 0.14aA


ND 0.4 A
**


11.9A ND
**


t Perennial peanut (PP) planted in spring. I PP planted in summer, the following years perennial peanut was over-seeded with crimson clover in
fall. PP planted in summer. Means within the same column followed by the same lower case letter and means within the same row followed by
the same upper case letter do not differ statistically by the LSMEANS adjusted by Tukey test (P<0.05). **Contrast between ACC and PPsu,
significant at P=0.01.


Treatment



PPspt
P P .. .
PPsu-os*
PPsu,


ACC
ACC vs PPsu


PP T
PPsT
PP sos
PPsu


Fall
-----------
0.7 bA
5.0 aA
7.2 aA


201 A
**


ACC
ACC vs PPsu


0.6 A
**


6.8 A
**


2.87 B
**


23.7 A
**










Table 3.8. Effect of planting time and over-seeding of perennial peanut (PP) on shoot number m2 (Shoot#), leaf area index (LAIpp), shoot dry
weight (DWpp), and N accumulation (Nacc-pp); weed dry weight (DWWD), N accumulation in weeds (Nacc-WD), and cover crop weed
index (CCWI) in 2004. LAI, DW, Nacc, DWx,and Nacc-xw for Annual Cover Crops (ACC) are included for purpose of comparison.
Perennial Peanut


Shoot#

Spring Summer Fall
------------ # m ------
5 bAT 8 bA 16 bA
81abA 107bA 153 bA
158aA 303 aA 346 aA


Nacc-pp


Spring Summer
------------- 2 m2
0.00aA 0.02 cA
0.03 aB 0.18 bA
0.06 aB 0.51aA


1.1A
**


DWwd


Spring Summer
----------- Mg ha-1
1.8 aB 2.7 aB
1.2 aA 2.1abA
1.1 aA 1.8 bA


Fall
-------------
4.3 aA
3.3 abA
2.5 bA


Spring
------------
29.4 aB
17.9 aA
15.2 aA


Fall
-----------
0.02 cA
0.25 bA
0.58 aA


0.65 B
NS
Weeds
Nacc-wd


Summer
kg N ha'
29.6 aB
26.5 aA
22.2 aA


Spring Summer Fall
------------ Mg ha -------------
0.00 aA 0.02 cA 0.03 cA
0.04 aB 0.22 bB 0.42 bA
0.08 aB 0.62 aB 1.00 aA


1.83 A
**


Fall
-------------
55.8 aA
35.8 abA
28.6 bA


Spring Summer
------------ kg N ha1
0.0 aA 0.5 bA
1.5 aA 5.6 bA
3.3 aB 16.8 aA


1.21 A 38.1 A ND
NS **

CCWI

Spring Summer Fall
----------- Mg Mg ---------
0.00 aA 0.01 bA 0.01 bA
0.04 aA 0.11 bA 0.13bA
0.08 aB 0.36 aA 0.41 aA


Fall
------------
0.7 cA
9.6 bA
23.1 aA


24.8 A
NS


0.1 B 1.4 A
** *


14.5 B 1.2 B 38.3 A
NS ** NS


1.92 A ND


Treatment


PPsp
PPs ...
PPs,


ACC
ACC vs PPsu


PP P
PPsp
PP u-os
PPsu
ppSU


ACC
ACC vs PPsu


0.9 B
NS


t Perennial peanut (PP) planted in spring. I PP planted in summer, the following years perennial peanut was over-seeded with crimson clover in
fall. PP planted in summer. Means within the same column followed by the same lower case letter and means within the same row followed by
the same upper case letter do not differ statistically by the LSMEANS adjusted by Tukey test (P<0.05). **Contrast between ACC and PPsu,
significant at P=0.01.


0.86 A










Table 3.9. Effect of planting time and over-seeding of perennial peanut (PP) on shoot number m2 (Shoot#), leaf area index (LAIpp), shoot dry
weight (DWpp), and N accumulation (Nacc-pp); weed dry weight (DWwD), N accumulation in weeds (Nacc-WD), and cover crop weed
index (CCWI) in 2005. LAI, DW, Nacc, DWx,and Nacc-xw for Annual Cover Crops (ACC) are included for purpose of comparison.
Perennial Peanut


Shoot#

Spring Summer Fall
------------ # m ------
2 b A 13bA 20bA
76 bA 150 bA 196 b A
225aB 479 aA 558 aA


Naccpp


Spring Summer
------------- 2 m2
0.OOaA 0.03 b A
0.03 aB 0.22 b A
0.09 aB 0.87aA


Fall
-----------
0.02 c A
0.24 b A
0.60 a A


Spring Summer Fall
------------ Mg ha -------------
0.00 aA 0.04 cA 0.03 cA
0.03 aC 0.29 bB 0.37 b A
0.10 aB 1.23aA 0.98aA


Spring Summer
------------ kg N ha1
0.1 aA 1.4bA
1.0 aA 7.3bA
4.1 aB 29.3 aA


DWwd


Spring Summer
----------- Mg ha-1
2.3aB 3.8aA
1.0 bB 3.4aA
1.2 bB 2.4 b A


1.91
**
Weeds
Nacc-wd


Fall
-------------
3.3 aA
2.5 b B
2.0 b A


Spring
------------
29.8 aA
11.4aB
14.7 aA


7.41 A
**


Summer Fall
kg N ha -------------
51.5aA 42.4aA
45.9 abA 29.7 aA
29.6 bA 23.9 a A


6.42 A 121 A
** **


Spring
-----------
0.00 a A
0.02 a A
0.10aA


CCWI

Summer Fall
Mg Mg -------------
0.01bA 0.01lbA
O.lObA 0.18bA
0.52 aA 0.50 aA


0.0 A 0.2 A
** **


1.1 A 0.2 A 2.3 A
** ** **


Treatment


PPsp
PPs ...
PPs,


ACC
ACC vs PPsu


Fall
------------
0.9 bA
8.0 bA
21.7 aA


PP P
PPsp
PP u-os
PPsu
ppSU


103 A
**


ACC
ACC vs PPsu


0.1 A
**


75.2 A
**


t Perennial peanut (PP) planted in spring. I PP planted in summer, the following years perennial peanut was over-seeded with crimson clover in
fall. PP planted in summer. Means within the same column followed by the same lower case letter and means within the same row followed by
the same upper case letter do not differ statistically by the LSMEANS adjusted by Tukey test (P<0.05). **Contrast between ACC and PPsu,
significant at P=0.01.


32.1 A
**




Full Text

PAGE 1

INTEGRATIVE USE OF PERENNIAL AN D ANNUAL COVER CROPS FOR WEED MANAGEMENT IN ORGANIC CITRUS By JOSE CLEMENTE LINARES B. A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2006

PAGE 2

Copyright 2006 by Jose Clemente Linares B.

PAGE 3

A Dios, mis Maestros, mis padres y familia que me impulsan a mejorar como ser humano

PAGE 4

iv ACKNOWLEDGMENTS I would like to acknowledge my major professor Johan M. Scholberg for his support, guidance, his effort and patience th roughout my graduate training program. I would also like to thank the other members of my supervisory committee, Dr. K. Boote, Dr. C. Chase, Dr. D. Graetz, and Dr. R. McSo rley, for their support, excellent advice, and assistance during my program, and for their contributions to my dissertation. Special thanks go to Andy Schreffler, Jim Boyer, and staff of the UF-IFAS Plant Science Research and Education Unit in Citra for their help with field studies I want to further thank Corey Cherr, Robert Wavestraut, A licia Lusiardo, Hannah Snyder, Huazhi Liu, Scott Prospect, Scott Tubbs, Kari Reno, Su san Sorell, Dipen Patel, Jorge Gomez, Jonathan Bracho, Amy Van Scoick, John Mc Queen, and Laura Avila, among others, for their assistance and friendship. I want to express my appreciation for the support and technical assistance of Juan Carlos Rodri guez and Marty Mesh from Florida Organic Growers. I also want to acknowledge Dr. Ramon Littell, Dr. Ken Portier, Salvador Gezan, Enrique Darghan, and Meghan Brenna n for their assistan ce with statistical analysis. I further want to express my gratitude to the “Universidad del Tachira”, Venezuela, for providing me with an opportunity to come to the University of Florida and to the USDA/ CSREES for the financ ial support of my program. I further want to thank Drs. Paul. Pfah ler and his family, Heartwell Allen, and Maria Luisa Izaguirre for their friendship and encouragement dur ing the past years. I also

PAGE 5

v sincerely value the friendship and help of Belkys Bracho, Marco, Nicary, and Vernica Emhart, during my stay in Gainesville and S onia, Betty, Chavela, Ma ra de los Angeles, Padrn, and Alexis for supporting me in difficult moments. I give thanks to God my Lord for assi sting me to embrace the challenges I faced during the past years and for all the blessings I have enj oyed as well. I would like to express my gratitude to my pa rents, sisters, and brother fo r being so supportive and for their continuous encouragement during my studies and my stay in Florida.

PAGE 6

vi TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES.............................................................................................................ix LIST OF FIGURES.........................................................................................................xiv ABSTRACT....................................................................................................................... xv CHAPTER 1 INTRODUCTION........................................................................................................1 2 GROWTH AND EFFECTS OF AN NUAL COVER CROPS ON WEED SUPPRESSION............................................................................................................9 Introduction................................................................................................................... 9 Materials and Methods...............................................................................................14 Set-up and Experimental Design.........................................................................14 Data Collection, Measurements, and Analysis....................................................16 Results........................................................................................................................ .18 Summer Cover Crops..........................................................................................18 Summer 2002...............................................................................................18 Summer 2003...............................................................................................19 Summer 2004...............................................................................................20 Summer 2005...............................................................................................21 Winter Cover Crops.............................................................................................22 Winter 2002/2003.........................................................................................22 Winter 2003/2004.........................................................................................23 Winter 2004/2005.........................................................................................24 Discussion...................................................................................................................25 Summer Cover Crops..........................................................................................25 Winter Cover Crops.............................................................................................29 System Dynamics................................................................................................31 Conclusions.................................................................................................................38

PAGE 7

vii 3 ESTABLISHMENT OF PERENNIAL PEANUT AND ITS EFFECTIVENESS IN SUPPRESSING WEEDS IN CITRUS ROW MIDDLES.....................................52 Introduction.................................................................................................................52 Materials and Methods...............................................................................................55 Set-up and Experimental Design.........................................................................55 Data Collection, Measurements, and Analysis....................................................58 Results........................................................................................................................ .60 Perennial Peanut 2002.........................................................................................60 Perennial Peanut 2003.........................................................................................61 Perennial Peanut 2004.........................................................................................62 Perennial Peanut 2005.........................................................................................62 Perennial Peanut Productivity (2005)..................................................................63 Annual Cover Crops (2002-2005).......................................................................64 System Dynamics................................................................................................65 Discussion...................................................................................................................66 Conclusions.................................................................................................................77 4 EFFECTS OF PERENNIAL PEANUT ( Arachis glabrata Benth ) AND COMMON BERMUDAGRASS ( Cynodon dactylon L.) ON NITROGEN AND WATER UPTAKE OF CITRUS................................................................................91 Introduction.................................................................................................................91 Materials and Methods...............................................................................................94 Set-up and Experimental Design.........................................................................94 Irrigation and ET Calculations............................................................................97 Nitrogen Application...........................................................................................98 Nitrogen Extraction.............................................................................................98 Nitrogen Uptake Calculation...............................................................................99 Final Plant Sampling.........................................................................................100 Statistical Analysis............................................................................................100 Results.......................................................................................................................1 02 Groundcover Biomass Production and N Accumulation..................................104 Final Citrus and Groundcover Gr owth and N Accumulation............................104 Discussion.................................................................................................................105 Groundcover Biomass Production and N Accumulation..................................109 Final Citrus and Groundcover Gr owth and N Accumulation............................110 Conclusions...............................................................................................................111 5 EFFECTS OF ANNUAL AND PERE NNIAL COVER CROPS ON SOIL AND CITRUS TREE CHARACTERISTICS CITRUS TREE ROW GROUND COVER, AND CITRUS YIELD AND QUALITY.................................................126 Introduction...............................................................................................................126 Materials and Methods.............................................................................................129 Set-up and Experimental Design.......................................................................129 Data Collection and Measurements...................................................................132

PAGE 8

viii Soil.............................................................................................................132 Nematodes..................................................................................................133 Weed Growth Dynamics............................................................................133 Citrus Tree Performance............................................................................134 Data Analysis.....................................................................................................134 Results.......................................................................................................................1 35 Soil pH...............................................................................................................135 Soil C, N, and C:N ratio....................................................................................135 Soil Nematodes..................................................................................................136 Tree Row Ground Cover in Perennial Cover Crop Study.................................137 Citrus Tree Growth Characteristics, C itrus Leaves N, and Fruit Quality.........137 Discussion.................................................................................................................138 Soil pH...............................................................................................................138 Soil C, N, and C:N Ratio...................................................................................139 Nematode Counts..............................................................................................141 Tree Row Ground Cover in Perennial Cover Crop Study.................................141 Citrus Tree Performance....................................................................................142 Conclusions...............................................................................................................143 6 SUMMARY AND CONCLUSIONS.......................................................................155 Annual Cover Crop Study........................................................................................156 Perennial Cover Crop Study.....................................................................................157 Citrus, Perennial Peanut, and Bermudagrass Competition for Nitrogen and Water158 Effect of Cover Crops on Soil Charac teristics, Tree Row Cover and Citrus Growth and Yield.................................................................................................161 Implications of the Research....................................................................................162 Future Research Recommendations.........................................................................163 APPENDIX A ANALYSES OF VARIANCE FOR PERENNIAL PEANUT STUDY...................165 B ANALYSES OF VARIANCE FOR EFFECTS OF PERENNIAL PEANUT ( Arachis glabrata Benth.) AND COMMON BERMUDA GRASS ( Cynodon dactylon L.) ON NITROGEN AND WA TER UPTAKE OF CITRUS....................166 C INITIAL SOIL CONDITIONS OF EXPERIMENTAL SITE, DECEMBER 2001 (SOIL ANALYSES RESULTS)...............................................................................168 D ANALYSES OF VARIANCE FOR ANNUAL AND PERENNIAL COVER CROPS ON SOIL AND CITRUS TR EE CHARACTERISTICS, CITRUS TREE ROW GROUND COVER, AND CITR US YIELD AND QUALITY.....................169 LIST OF REFERENCES.................................................................................................173 BIOGRAPHICAL SKETCH...........................................................................................191

PAGE 9

ix LIST OF TABLES Table page 2.1 Overview of annual summer and winter cover crops used during the 2002 and 2003 growing seasons..............................................................................................39 2.2 Overview of annual summer and winter cover crops used during the 2004 and 2005 growing seasons.............................................................................................40 2.3 Overview of seeding rates, space between rows and cultivars used as annual summer and winter cover crops used from 2002 to 2005........................................41 2.4 Outline of planting and harvest dates and duration of summer and winter cover crops.........................................................................................................................4 2 2.5 Outline of cover crop weed index (CCWI) categories.............................................42 2.6 Rainfall measured at Plant Research and Education Unit (Citra) Florida Automated Weather station Net work (FAWN)1 during the 2002-2005 summer CC growing season...................................................................................................43 2.7 Shoot dry weight accumulation (DW); s hoot N concentration (Nconc); shoot N accumulation (Naccum), maximum observed leaf area index (LAI max) for summer cover crops grown during the 2002 growing season..................................44 2.8 Shoot dry weight accumulation (DW); s hoot N concentration (Nconc); shoot N accumulation (Naccum), maximum observed leaf area index (LAI max) for summer cover crops grown during the 2003 growing season..................................45 2.9 Shoot dry weight accumulation (DW); s hoot N concentration (Nconc); shoot N accumulation (Naccum); weed dry weight accumulation (DW) for summer cover crops during the 2004 growing season...........................................................46 2.10 Shoot dry weight accumulation (DW); s hoot N concentration (Nconc); shoot N accumulation (Naccum); weed dry weight accumulation (DW) for summer cover crops during the 2005 growing season...........................................................47 2.11 Rainfall measured at Plant Research and Education Unit (Citra) Florida Automated Weather station Net work (FAWN)1 during the 2002-2005 winter CC growing season...................................................................................................48

PAGE 10

x 2.12 Shoot dry weight accumulation (DW); s hoot N concentration (Nconc); shoot N accumulation (Naccum); maximum observed leaf area index (LAI max) for winter cover crops during the 2002-2003 growing season......................................48 2.13 Shoot dry weight accumulation (DW); s hoot N concentration (Nconc); shoot N accumulation (Naccum); weed dry weight accumulation (DW) for winter cover crops during the 2003/2004 growing season............................................................49 2.14 Shoot dry weight accumulation (DW); s hoot N concentration (Nconc); shoot N accumulation (Naccum); weed dry weight accumulation (DW) for winter cover crops during the 2004/2005 growing season............................................................50 3.1 Overview of experimental treatments during 2002-2005........................................78 3.2 Overview of seeding rates and row sp acing for annual summer and winter cover crops used between 2002 and 2005..........................................................................79 3.3 Outline of planting and harvest dates a nd duration for summer and winter cover crops.........................................................................................................................7 9 3.4 Outline of cover crop weed index (CCWI) categories.............................................80 3.5 Rainfall measured in the Plant Scien ce Research and Education Unit (Citra)1 during 2002-2005.....................................................................................................80 3.6 Effect of planting time and over-seedi ng of perennial peanut (PP) on shoot number m-2 (Shoot#), leaf area index (LAIPP), shoot dry weight (DWPP), and N accumulation (Nacc-PP); weed dry weight (DWWD), N accumulation in weeds (Nacc-WD), and cover crop weed index (CCWI) in 2002.........................................81 3.7 Effect of planting time and over-seedi ng of perennial peanut (PP) on shoot number m-2 (Shoot#), leaf area index (LAIPP), shoot dry weight (DWPP), and N accumulation (Nacc-PP); weed dry weight (DWWD), N accumulation in weeds (Nacc-WD), and cover crop weed index (CCWI) in 2003.........................................82 3.8 Effect of planting time and over-seedi ng of perennial peanut (PP) on shoot number m-2 (Shoot#), leaf area index (LAIPP), shoot dry weight (DWPP), and N accumulation (Nacc-PP); weed dry weight (DWWD), N accumulation in weeds (Nacc-WD), and cover crop weed index (CCWI) in 2004.........................................83 3.9 Effect of planting time and over-seedi ng of perennial peanut (PP) on shoot number m-2 (Shoot#), leaf area index (LAIPP), shoot dry weight (DWPP), and N accumulation (Nacc-PP); weed dry weight (DWWD), N accumulation in weeds (Nacc-WD), and cover crop weed index (CCWI) in 2005.........................................84 3.10 Effect of planting season date and over seeding on pe rennial peanut (PP), weeds, and system (PP+weed) dry weight, N accumulation in PP, weeds and in the system in 2005..........................................................................................................85

PAGE 11

xi 3.11 Total dry weight (DW) in the system (CC+weeds), corresponding percentage of total dry weight in CC (% DW CC), to tal N accumulation (Total Nacc) in the system (CC and weeds) and correspondi ng percentage of N in CC (%N in CC) in 2005......................................................................................................................86 4.1 Effect of cropping system on citrus, berm udagrass, and perennial peanut (PP) N and water uptake for three different seasons during 2005......................................114 4.2 Effect of cropping system on citrus, berm udagrass, and perennial peanut (PP) N and water uptake for two different growth cycles during 2005.............................115 4.3 Comparison of effect of cropping syst em on citrus, bermudagrass (BG), and perennial peanut (PP) N upt ake at the final harvest (end of the growing period) using 15N and SUM techniques..............................................................................116 4.4 Nitrogen accumulation by citrus and ground covers based on 15N results.............116 4.5 Overview of parameters fo r N uptake regression model........................................117 4.6 Effect of cropping system on bermudagra ss and perennial peanut (PP) shoot dry weights (DW), nitrogen concentrati on (Nconc) and nitrogen accumulation (Naccum) for different growing seasons during 2004 and 2005............................118 4.7 Effect of cropping system on citrus root dry weight (DW), root length, stem dry weight, diameter (Diam), leaf dry weight leaf area (LA), total dry weight, root nitrogen accumulation (Naccum), stem N accumulation, leaf N accumulation, and total N accumulation at the end of the growing season...................................119 4.8 Effect of cropping system on bermudagrass (BG) and perennial peanut (PP) root dry weight (DW), root le ngth, shoot dry weight (DW) leaf area (LA), root nitrogen concentration (Nconc), shoot nitrogen concentration, root nitrogen accumulation (Naccum), shoot N accumulation, and total N accumulation at the end of the growing season......................................................................................119 4.9 Percentage of N distri bution in different tissues for the diverse cropping systems...................................................................................................................120 5.1 Effect of year, season, location, and tr eatments on soil pH for the perennial cover crop study during 2003-2005.......................................................................145 5.2 Effect of year, season, location, and trea tments on soil pH for the annual cover crop study during 2003-2005.................................................................................146 5.3 Effect of year, location, and treatment on soil C, N, and C:N for the perennial cover crop study during 2003-2005.......................................................................147 5.4 Effect of year, location, and treatment on soil C, N, and C:N ratio for the annual cover crop study during 2003-2005.......................................................................148

PAGE 12

xii 5.5 Number of plant-parasitic nematode for the perennial cover crop study during 2004 and 2005........................................................................................................149 5.6 Number of plant-parasitic nematode for the annual cover crop study during 2004 and 2005.................................................................................................................150 5.7 Percentages of ground cover in the tree row for most commonly observed weed species as affected by year and season in tree rows for the perennial cover crop study during 2003-2005.........................................................................................151 5.8 Effect of year, season, and treatments on tree height and trunk diameters for ‘Hamlin’ oranges (perennial co ver crop study) during 2002-2005........................152 5.9 Effect of year, season, and treatments on tree height and trunk diameters for ‘Navel’ oranges (annual c over crop study) during 2003-2005...............................153 5.10 Effect of cover crop treatme nt on citrus yield and fru it quality (degree Brix and acidity) for the perennial cover crop study during 2005........................................154 A.1 Analyses of variance for perennial peanut (PP) shoot dry weight (DWPP), N accumulation in PP shoots (Nacc-PP), PP leaf area index (LAIPP), number of PP shoots per square meter (shoot#), Weed dry weight (DWWD), N accumulation in weeds (Nacc-WD), and Cover crop weed index (CCWI)........................................165 B.1 Analyses of variance for the effect of ground covers on N and water uptake.......166 B.2 Analyses of variance for the effect of cropping system on bermuda grass and perennial peanut shoot dry weight (D W), nitrogen concentration (Nconc) and nitrogen accumulation (Naccum)...........................................................................167 D.1 Analyses of variance for the effect of perennial cover crops (PCC) and annual cover crops (ACC) on soil pH................................................................................169 D.2 Analyses of variance for the effect of perennial cover crops on soil carbon (C), nitrogen (N) and C:N ratio.....................................................................................170 D.3 Analyses of variance for the effect of annual cover crops on soil carbon (C), nitrogen (N) and C:N ratio.....................................................................................170 D.4 Analyses of variance for the effect of perennial cover crops (PCC) and annual cover crops (ACC) on soil nematode populations.................................................171 D.5 Analyses of variance for the effect of perennial cover crops on tree-row cover....171 D.6 Analyses of variance for the effect of perennial cover crops on citrus tree height (Height) and diameter (Diam)................................................................................172

PAGE 13

xiii D.7 Analyses of variance for the effect of annual cover crops on citrus tree height (Height) and diameter (Diam)................................................................................172 D.8 Analyses of variance for the effect of perennial cover crops (PCC) and annual cover crops (ACC) on nitrogen ci trus leaf concentration.......................................172

PAGE 14

xiv LIST OF FIGURES Figure page 2.1 Leaf area index values for summer c over crops 2002 (CP= cowpea; VB= velvet bean; SH= sunnhemp; AC= Alyceclover; HI= hairy indigo)...................................51 2.2 Leaf area development for wint er cover crops during 2002/2003...........................51 3.1 Dry matter of perennial p eanut (PP) over time........................................................87 3.2 Dry matter of weeds across the years.......................................................................88 3.3 Cover Crop Weed Index (CCWI) for pe rennial peanut across the years.................89 3.4 Regression between PP dry weight (DW pp) and weed dry weight during spring (SPDWW), summer (SUDWW), and fall (FALLDWW) for all PP treatments.................90 3.5 Regression between PP dry weight (D W) and N accumulation in weeds during spring (SPN-W), summer (SUN-W), and fall (FALLN-W) for all PP treatments...........90 4.1 Overview of soil-N uptake monitoring (SUM) system..........................................121 4.2 Minima, maxima, and soil average temper ature during the experimental period..122 4.3 Solar radiation in the greenhouse during the experimental period.........................122 4.4 Regression between SUM-based N uptake and 15N based N uptake.....................123 4.5 Nitrogen uptake dynamics for diffe rent cropping systems across time.................124 4.6 Nitrogen uptake as a function of cumu lative uptake temperature during 14-day pre-clipped vs. post-clipped uptake period for bermudagrass mono-crop.............125 4.7 Nitrogen uptake as a function of cu mulative radiation during the 14-day preclipped vs. post-clipped uptake period for bermudagrass mono-crop..............125 C.1 Initial soil conditions at the experimental site in December 2001 (soil analyses results from Analytical Research Lab. IFAS, Gainesville, FL.)............................168

PAGE 15

xv Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy INTEGRATIVE USE OF PERENNIAL AN D ANNUAL COVER CROPS FOR WEED MANAGEMENT IN ORGANIC CITRUS By Jose Clemente Linares B. August, 2006 Chair: Johannes M. Scholberg Major Department: Agronomy Citrus is one of the most important cr ops in Florida. Du ring the past decade increased international competition and urban developments, diseases, and more stringent environmental regulations have greatly aff ected the citrus indu stry. Citrus growers transitioning to organic produc tion may benefit from premium pr ices, but they also face many challenges, including development of effi cient weed management strategies. Cover crops (CC) may constitute an environmen tally sound alternative for improved weed management in organic systems. Two field e xperiments were conducted at Citra in North Central Florida, to test performance and th e effectiveness of annual and perennial CC to suppress weeds in organic groves. A greenhouse trial was also implemented to evaluate potential competition between citrus and groundcovers. For annual CC, summer CC had the greatest biomass and N accumulation in comparison with winter CC. Sunnhemp, hairy indigo, cowpea, and al yceclover provided excellent weed suppression, which was superior to tillage fallow. Mono-cropped winter CC did not always perform consistently

PAGE 16

xvi well. Use of winter CC mixtures resulted in more consistent overall CC performance, greater dry matter production, N accumulation, and more effective weed suppression. In both annual and perennial systems, weeds pl ayed a complementary role in nutrient retention and recycling. Perenni al peanut (PP) showed slow initial growth and summer planting of PP was the most successful comp ared with spring planting. Over-seeding PP planted in summer with crimson clover redu ced PP growth and its effectiveness in suppressing weeds. Initial w eed suppression by PP was very poor to poor; however, effectiveness of PP to reduce weed grow th improved gradually over time. Annual CC provided much better weed control than PP. For both PP and annual CC, weed biomass typically was inversely related to CC DW accumulation due to competition for resources. In a greenhouse experiment, citrus and berm udagrass appeared to compete for N during summer, while citrus and PP did not compet e. Citrus, bermudagrass, and PP competed for water uptake during the spring and summer seasons. In general, perennial and annual CC treatments did not affect so il pH, C, N, and C:N ratio dur ing the initial 3 years of field studies. Nematode populations did not reach high levels. Cover crop treatments in row middles did not affect weed growth dyna mics in the tree row. However, planting cowpea in the tree row did decrease bahiag rass and crabgrass populat ions in tree rows. Planting tall cover crops such as sunnhemp near young citrus trees reduced initial tree growth. Cover crop treatments di d not affect citrus leaf N, fruit yield, and quality during initial growth. Additional resear ch is needed to assess long-t erm effects of cover crops on soil quality and fruit yield.

PAGE 17

1 CHAPTER 1 INTRODUCTION Florida is the largest citrus producing st ate in the U.S. and accounted for 74% of the U.S citrus production during the 2 004-2005 season, with 302,929 ha bearing fruit (Florida Agricultural Statistics Service, 2006). Although citrus accounts for 49% of the total certified organic farm area in Florida, le ss than 1% of Florida citrus is currently grown organically (Athearn, 2004). Total value of Florida citrus duri ng 2004-2005 was US $742 million, which was the lowest since 1985-86 (Florida Agricultura l Statistics Service, 2006). This decline was related to a reduction in citrus consumption (especially of ora nge juice) in the U.S.A. due to the popularity of low carbohydrate diets, increased international competition, and relatively low on-tree citrus prices. Recent pe st and disease outbreaks, competition with residential development for land and wate r resources, along with more stringent environmental regulations pose tremendous ch allenges for conventional citrus growers (Athearn, 2004). In contrast, organic agricultu re is the fastest increasing segment of US agriculture. Organic sales have increased by 20% a nnually since 1990 (Dimitri and Greene, 2002), and retail sales in the U.S were estimated to be on the order of $17 billion during 2005. The leading revenue source for the US organi c food market is the fruit and vegetable sector, which generated total revenues of $7 billion in 2005, which correspond to 41% of the overall organic market. Although it is project ed that organic sale s will increase to $35

PAGE 18

2 billion by the end of 2010, organic production still accounts for less than 5% of US agricultural area for most commodities (Datamonitor, 2005). In Florida, 85% of the wells that exc eeded the maximum contaminant level (MCL) for NO3-N were located in conventional citrus production areas (McNeal et al., 1995). Similarly, some citrus-applied herbicides have been discovered in both ground and surface water (Troiano and Garretson, 1998; Li u and OÂ’Connell, 2002). Since excessive use of non-renewable resources and/or poten tially harmful agrochemicals may impact biodiversity, environmental quality, food safety, and health of farmers, there is increased interest in more sustainabl e production systems including organic farming (Reganold et al., 2001). Organic production may not onl y protect natural resources and the environment but also meets current consum erÂ’s health concerns and food safety requirements (Igual et al., 2002). Conversion to organic producti on systems may also allow growers to benefit from marketing ni ches and grower-friendly price mechanisms associated with organic production (Athearn, 2004). By reducing re gional pesticide and fertilizer use, organic pro duction can preserve both groundw ater resources and fragile ecosystems for future generations (Mader et al., 2002). Organic agriculture relies on ecologically ba sed principles and practices, such as nutrient recycling, increased biodiversity, and biological pest management. It aims to achieve more autonomous management of local agroecosystems and to enhance the resilience of such systems by increasing re liance on local resources biodiversity, and synergistic biological interac tions. In this manner local pr oduction capacity of the soil can be sustained (Altieri, 1995; Gliessman, 1998) However, this requires the adoption of alternative techniques to e nhance both crop and soil healt h, including improved use of

PAGE 19

3 cover crops to suppress weeds, prevent erosion, and restor e soil organic matter (Nelson, 2004). Results from a 21-year study of agronomic and ecological performance of organic, biodynamic, and conventional farming systems in central Europe were reported by Mader et al. (2002). They found th at although crop yields were 20% lower in the organic systems, input of fertilizer and energy wa s reduced by 34 to 53% and pesticide input by 97%, while soil fertility and biodiversity were enhanced and the use of external inputs was being reduced. The authors concluded th at organically-managed legume-based crop rotations utilizing organic fertilizers from the farm itself provided a viable alternative to conventional farming systems. Organic citrus production emerged as a co mmercial sector in Florida during the early 1990. A 1993 survey identified 16 organic citrus growers covering 230 ha (Swisher et al., 1994). This acreage wa s increased to 2,400 ha in 2004-2005, while the number of organic citrus growers incr eased to 39 (Athearn, 2004). However, despite the rapid growth of orga nic agriculture, inform ation pertaining to organic production in general a nd organic citrus in particular is scarce. As a result, the organic grower community reque sted that the United States Department of Agriculture (USDA) to create special programs and providi ng grant funding for this research was one of the first initiatives. Research priority areas ranged from development of weed management practices during the transition from conventi onal to organic farming to economic research on markets and profitabili ty of organic farming systems (SCOAR, 2003).

PAGE 20

4 In response to the growing interest in organic agricu lture in the U.S. and the implementation of the National Organic Pr ogram in 2002, USDA created the Sustainable Agriculture Research and Education Program (SARE), which looks at both conventional and organic systems. Previously, the USDA had initiated a sustainable agriculture research and education program in 1988 and th is program was origin ally referred to as the Low-Input Sustainable Agriculture (LISA) program (SARE, 2006). Other nonprofitable organizations focusing on organics includes the Organic Farming Research Foundation which was founded in 1990 by certifi ed organic farmers and provides funding to a limited number of research programs to address critical issues in organic agriculture (Organic Farming Research Foundation, 2006). Conversion from conventional to organi c production will result in appreciable modification in agroecosystem manageme nt (Ngouajio and McGiffen, 2002). Weed suppression is one of the most important co mponents to be considered during this conversion process since important changes in weed population dynamics occur, which will require implementation of alternative weed management strategies (Lanini et al., 1994; Liebman and Davis, 2000). Weed control in conventionally grown citrus accounts for 27% of annual production costs. In organic citrus groves, weed management accounts for over 30% of annual production costs and the majority of the labor costs (Muraro et al., 2003; Athearn, 2004). A national survey showed that the nu mber one research priority for organic growers was improved weed management (Sooby, 2003). Florida growers were no exception to this finding, and both Florid a organic citrus growers and grower

PAGE 21

5 organizations emphasized that weed control was the most critical fact or for growers to be successful during the transition to organic production (SCOAR, 2003). One ecological approach for weed manage ment includes improved soil coverage through use of cover crops (CC) and/or mulche s. However, in the absence of appropriate recommendations, lack of effective weed ma nagement practices pertinent to organic systems may hamper successful transition from conventional to organic citrus production (Sooby, 2003; SCOAR, 2003). In interviews with Florida citrus growers, the majority expressed a strong interest in the use of cover crops in cluding perennial peanut ( Arachis glabrata Benth.) to prevent soil degradation and suppress weed growth (Scholberg, unpublished). Perennial peanut (PP) may provide an environmentally sound and ecologically impor tant component of sustainable citrus production in Florida, since it does not requ ire pesticides or N fertilizer (French et al., 1994; Mullahey et al., 1994) Due to its low water and nutrient requirements, perennial peanut fits the mode l of sustainable produc tion. In contrast to annual cover crops, it needs to be plante d only once and it may reduce labor costs associated with weed control in citrus It may also provide 67 to 112 kg of N ha-1 yr-1 to citrus trees among many other benefits (French et al., 1994; Woodward et al., 2002). During the past decade, perennial peanut has been evaluated as a groundcover for vegetable crops (Roe et al., 1994) and citr us (Coleman, 1995). Several citrus growers have also successfully integrat ed this cover crop into thei r production system. However, current practices for establishment of pere nnial peanut were typically developed for conventional forage production and are not appropriate for organic citrus. Similarly, although perennial peanut use as a cover crop in conventional systems in South Florida

PAGE 22

6 has been studied extensively (Roe et al ., 1994; Mullahey et al ., 1994), no detailed information is available regarding its establis hment in organic citrus production systems. Support for the research outlined in this dissertation was provided by the USDA Organic Transition Program via a grant entitled “Integrative Use of Perennial Peanut for Cost-Effective Weed Control in Organic Citr us”. Originally, the main focus of this project was on the use of perennial peanut in organic citrus groves. However, based on comments of panel reviewers the scope of th is program was extended to also include a study focusing on both annual winter and summ er cover crops. These studies were intended to evaluate the effect of cover crops on weed suppre ssion, soil quality, and citrus tree growth for a newly-planted certified organic citrus production system. The general objectives of this work were to 1) determine growth characteristics of annual and perennial cover crops in organic citrus groves; 2) evaluate changes in weed growth as affected by annual and perennial c over crop treatments; 3) quantify the effect of perennial peanut and/or common berm udagrass on citrus N and water uptake under controlled conditions; and 4) a ssess how cover crop treatments affect soil quality, tree height and diameter, leaf N, fruit yield, and fruit quality. The corresponding hypotheses were as follo ws: 1) annual CC will suppress weeds effectively and summer CC will accumulate more biomass and consequently will suppress weeds more effectively compared to winter CC; 2) in orga nic citrus systems, planting PP during the summer will increase the competitiveness of PP systems via enhanced initial growth comp ared to spring plantings a nd overseeding PP with crimson clover in fall will help to increase the PP effectiveness in suppressing weeds; 3) weed suppression with annual CC will be more eff ective than perennial peanut; 4) citrus,

PAGE 23

7 perennial peanut, and common bermudagrass w ill differentially compete for nitrogen and water uptake and competition for N and water up take will be most evident during periods of high demand; 5) annual and perennial CC will increase soil quality over time and reduce pest nematode populations in organic citrus fields; a nd 6) cover crop treatments in the row middles will also affect weed growth in the tree rows; and cover crops will not affect significantly citrus growth characteristics. This dissertation includes six chapters Although each chapter forms a separate entity they are also intrinsically linked. The current (first) chapter provides a conceptual framework for this dissertation and includes a br ief historic perspective of this work. It outlines program objectives and hypotheses while in the following part an outline of consecutive chapters is provided to empha size the inner-connectiv ity among subsequent chapters. The second chapter outlines the performan ce of different winter and summer cover crops in a recently established organic citrus orchard under Florida conditions and their effectiveness in suppressing weed growth. The use of the cover crop weed index (CCWI), which is the ratio of c over crop dry weight (CCDW) to weed dry weight (WeedDW) associated with a specific cover crop (CCWI=CCDW/WeedDW), will be discussed along with the use of this index for improved assessm ent of the effectiveness of different cover crops to suppress weeds. The third chapter evaluate s initial establishment, growth dynamics, weed suppression capacity, and productivity of pere nnial peanut in a recently established certified organic citrus grove Treatments included date of planting, association with winter annual cover crops (over-seeding pere nnial peanut with crimson clover in fall),

PAGE 24

8 and a system with annual cover crops only (w hich provided a linkage with the previous chapter). After successful establishment of perennial peanut as a groundcover for organic citrus groves, perennial peanut plants may al so expand into the tree rows and potentially compete with citrus trees for water and nutri ents. Therefore, competition between citrus and perennial peanut for water and nutrients was studied under cont rolled conditions as outlined in the fourth chapter. In this chap ter, the water and N uptake dynamics for pure and mixed systems of perennial peanut, weed s (bermudagrass) and citrus, and potential competition for water and nitrogen uptake are presented. Besides the effects of cove r crops on weed suppression in organic citrus groves, cover crops also have potential effect s on soil chemical, physical, and biological properties, including soil organic matter, soil nitrogen, pH, and nematode populations. Some of these and/or a combin ation of these parameters may be used as an indicator of “soil health”, system sustai nability, and potential “suppres siveness capacity” of these soils. In the fifth chapter we summarize how co ver crops affect some of these parameters and also overall tree growth and initial produc tion. In the last chap ter we synthesize and summarize previous chapters and also outline future research priorities and practical implementation for growers.

PAGE 25

9 CHAPTER 2 GROWTH AND EFFECTS OF ANNUAL CO VER CROPS ON WEED SUPPRESSION Introduction Cover crops (CC) are herbaceous plants, annua ls or perennials (usually grasses or legumes) grown in pure or mixed stands to c over the soil during part of one or more cropping cycles. The CC may be incorporated into the soil by tillage in seasonal CC systems, or retained as live or dead pl ants on the soil surface for several seasons (Gliessman, 1998). Cover crops may suppress weeds by either removal of resources (Ngouajio and Mennan, 2005; Ross et al., 2001) or by increase d soil microbial diversity (Kremer and Li, 2003; Kennedy, 1998), or may inhibit weeds via allelopathy (Reberg-Horton et al., 2005; Fennimore and Jackson, 2003). Also, CC can s uppress soil pathogens by allelochemicals (Bailey and Lazarovits, 2003) or increased pres ence of beneficial organisms that suppress pest organisms such as nematodes (Reddy et al., 1986b; McSorley, 2001; Macchia et al., 2003). In addition, use of CC enhances soil biological and chemical properties by promoting the creation of cooler and moister soil surface and subsurf ace habitats (Kremer and Li, 2003). Cover crops enhance soil fertility via improved nutrient cycling and nitrogen fixation by legumes CC (Ramos et al., 2001; Perin et al., 2004), carbon sequestration (Sainju et al., 2003), and increased nutrient retention by roots (Vos and van der Putten, 2001; Kristensen and Thorup-Kristensen, 2004).

PAGE 26

10 Cover crops shield the soil surface from sunlight, wind, and the physical impact of raindrops thereby reducing soil erosion and so il organic matter losses (Sainju et al., 2003). Cover crops also increase biological activity in the ro ot zone and thereby enhance the formation of more stable soil aggr egates and macropores while reducing soil compaction and soil bulk density. As a result, they can result in improved soil structure (Kremer and Li, 2003), water infiltration, and ro ot penetration (Justes et al., 1999), while reducing soil crust formation, runoff, and soil erosion (Gliessmann, 1998). Although CC may provide many environmental and agronomic benefits, there also may be negative effects of using CC in a cr opping system. Establishment costs may be cost-prohibitive, thus hampering their use in resource-limited production systems. Residues or breakdown products of incorpor ated CC may produce growth-suppressing (allelopathic) substances that impact the growth of commercial crops. Damaging herbivores or disease organism s may find CC to be suitable alternate hosts before moving on to the subsequent main crops. The CC resi due may also interfere with cultivation, weeding, harvesting, and/or other farming activi ties. Some CC when used as an intercrop or live mulch may be excessively tall and shade the commercial crops (Reeves, 1994; Gliessman, 1998). Cover crops can be classed as annuall y seeded winter-growing grasses and legumes, reseeding winter annual grasses and legumes, summer annuals, perennial grasses and legumes, and other cover crop pl ants (Altieri, 1995). Advantages of annual summer CC include rapid initial growth and e ffective competition with weeds thus cover crops should perform well within organic systems. On the other hand, annual CC typically need to be replanted. During the fi rst year(s), poor inocul ation may also hamper

PAGE 27

11 the growth of leguminous CC (Slattery et al. 2001; Carsky et al ., 2001; Tian and Kang, 1998). Many leguminous CC, especially wint er CC, are more demanding and require a suitable pH and supplemental nutrients (mainl y P, S, K, and Mo) to ensure adequate nodulation, N fixation, and biomass accumulation (S lattery et al., 2001). Therefore, they may be poorly adapted to the sandy, shallow, and low fertility soils prevailing in most citrus production areas in Florida. A number of studies have shown that so me of the annual CC have recalcitrant seeds with hard coats which can become pa rt of the soil seed bank and thus may reestablish themselves naturally (Benech-Ar nold et al., 2000). When used in vegetable crops, use of “hard-seeded” CC may not be desirable since they promote “weediness”, which can complicate CC management (B ond and Grundy, 2001). However, in mature citrus orchards, a natural seed-bank of select ed leguminous CC may be desirable. In this case, different combinations of species may proliferate each year, which may result in a natural selection of plants best-suited fo r prevailing soil and/or current climatic conditions, while also eliminating the a nnual cost for CC re-establishment. Weeds in agroecosystems are known to compete with crops for water, nutrients, and light. They are potential hos ts for pests and diseases and can also interfere with soil tillage, irrigation, and harvest operations (L iebman and Davis, 2000). As a result, they increase labor requirements and production co sts. On the other hand, certain weeds may form important components of the agroecosy stem because they provide alternative food sources such as pollen, nectar, foliage, or prey for beneficial insects (Altieri and Letourneau, 1982; Chacon and Gliessman, 1982).

PAGE 28

12 Weed control in conventionally grown citrus accounts for 27% of annual production costs (Muraro et al., 2003). Weed co ntrol programs include the application of pre-emergence and post-emergence herbicid es within tree rows especially for young trees, chemical control with herbicides lik e Roundup in the drive middles between rows, mowing and disk harrowing (Futch, 2005). Fo r organic citrus groves, weed control accounts for over 30% of annual production cost s and the majority of the labor costs (Muraro et al., 2003). It in cludes disking, mowing, and hand labor to remove vines growing into tree canopies and/or weeds near tree trunks of young trees. There are some other strategies for weed management lik e improved soil coverage through use of mulches and/or appropriate use of CC. Conversion from conventional to organi c production will result in appreciable modifications in agroecosystem management. Appropriate weed management is one of the most challenging components during th is conversion because of the important changes in weed population dynamics, which wi ll require implementation of alternative weed management strategies (Bond and Gr undy, 2001; Lanini et al., 1994). A national survey showed that the number one research priority for organic growers was improved weed management (Sooby, 2003). Based on inte rviews with Florida citrus growers (Scholberg, unpublished), the majority expresse d a strong interest in the use of cover crops to prevent soil degradat ion and suppress weed growth. Successful weed suppression using summer CC has been reported for annual crops such as rice ( Oryza sativa L.) preceded by pigeon pea ( Cajanus cajan L.) as a CC (Roder et al., 1998), lettuce ( Lactuca sativa L.) planted after cowpea ( Vigna unguiculata L. Walp.) or sorghum-sudan grass ( Sorghum bicolor L. Moench) (Ngouajio and Mennan,

PAGE 29

13 2003), and corn ( Zea mays L.) by using velvet bean ( Mucuna atropurpureum L.) (Buckles and Triomphe, 1999; Caamal-Maldona do et al., 2001). Us ing annual winter CC such as rye ( Secale cereale L.) (Fennimore and Jackson, 2003) or crimson clover ( Trifolium incarnatum L.) and subterranean clover ( Trifolium subterraneum L.) (Barberi and Mazzoncini, 2001) may provide adequate a posteriori weed control in corn. Equally suitable weed suppression has b een reported in strawberries ( Fragaria x ananassa Duchesne .) employing winter rye and wheat ( Triticum aestivum L.) (Whitworth, 1995). Similarly in field pea ( Pisum sativum L. spp. arvense (L.) Poir) weeds were suppressed when it was preceded by sweet clover ( Melilotus officinalis L.) (Blackshaw et al., 2001). Cover crops have been useful in suppre ssing weeds in perennial production systems as well. Bradshaw and Lanini (1995) obtaine d acceptable weed cont rol in coffee by using Desmodium ovalifolium, Commelina difusa, and Arachis pintoi Adequate weed suppression was obtained through use of wint er rye as a CC around several forest and ornamental tree species, but tree growth reduction occurred with sod CC treatments (Calkins and Swanson, 1995). Although the use of perenni al and annual CC in conve ntional vegetables and perennial CC in citrus systems in south Fl orida has been studied extensively (Rouse and Mullahey, 1997; Coleman 1995; Roe et al., 1994; Mullahey et al., 1994), no information is available regarding the effectiveness of annua l CC in suppressing weeds in organic citrus production systems. The overall objectives of this study were to 1) determine growth characteristics of annual cover crops in organic citrus groves; 2) determine changes in weed growth as affected by annual cover crop treatments; 3) identify suitable cove r crop species and

PAGE 30

14 evaluate their effectiveness in suppressing weeds in organic ci trus groves; and 4) develop optimal cover crop associations suit ed for organic citrus production. The following hypotheses were tested in or ganic citrus systems: 1) annual CC will suppress weeds effectively; 2) summer CC will accumulate more biomass and consequently will suppress weeds better than winter CC; and 3) weed suppression by CC will be related to their adapta tion to environmental conditions. Materials and Methods Set-up and Experimental Design A one-hectare block was planted with ‘navel ’ orange, a fresh market orange variety [ Citrus sinensis (L.) Osb.cv. Navel] graf ted on Swingle Citrumelo ( C. paradisi Macf. x P. trjfoliata (L.) Raf.) during the spring of 2003 at the Plant Science Research and Education Unit in Citra, Florida (29.68 N, 82.35 W). Tree spacing wa s 4.6 m in the row and 6.1 m between rows. The main emphasis of this study was the evaluation of annual CC used as ground cover to suppress weeds in the strips between tree rows (row middles). Prevailing soil types at the experimental site were a Candler fine sand (Typic Quarzipsamments, hyperthermic, uncoated, 98% sand in the upper 15 cm) and a Tavares fine sand (Typic Quarzipsamments, hypert hermic,uncoated, 97% sand in the upper 15 cm). The initial soil pH ranged from 4.8 to 5.1 and soil organic matter content was 9.3 g kg-1. At the beginning of the experiment (F all of 2001), soil was prepared by disking followed by repeated rototilling. During the spring of 2002, both lime (2.5 Mg ha-1) and chicken manure litter (2.5 Mg ha-1) were applied to the enti re production block. During subsequent years, chicken manure was a pplied exclusively to a 1.8-m wide strip straddling the tree rows. Manure was a pplied during early sp ring at 4-11 Mg ha–1.

PAGE 31

15 Manure application rates were based on IF AS (Institute of Food and Agricultural Sciences) N-recommendations for newlyplanted trees and based on estimated N mineralization and N concentration followi ng N recommendations for newly-planted trees (Tucker et al., 1995). To enhance grow th of winter cover crops, a non-synthetic (mined) K2SO4 (SQM North America Corp., Atla nta, GA) approved by the Organic Materials Review Institute (OMRI) was broadc ast over the entire area at a rate of 45 kg K2O ha-1 prior to planting of winter CC. Due to a buildup of resi dual soil P, use of chicken manure was discontinued after 2004, and starting in 2005, an OMRI-approved natural fertilizer derived from feather-meal and potassium sulfate (Nature Safe, Griffin Industries, Cold Spring, KY) with 9-0-9 (N, P2O5, K2O) was applied to tree rows using standard recommendations (Tucker et al., 1995). Trees were irrigated with microjet spri nklers with a 1.8-m spray diameter and a 180 spray pattern placed 0.4 m NW of the trees. During the wi nter the irrigation sprinklers were placed inside 0.6 m high PVC pi pes and also used for frost protection if temperatures dropped below -2 C. Row middles and cover crops were not irrigated in order to evaluate the suitability of different species for typical ci trus orchard conditions. Prior to the planting of the orange trees, one cropping cycle of bot h summer and winter cover crops was completed for initia l screening of suitable cover crops. Cover crop treatments are outlined in Tables 2.1 (2002 and 2003) and 2.2 (2004 and 2005). Each CC treatment plot consisted of a total area of 6.1 m x 27 m straddling a row of 5 citrus trees. However, Treatm ent 1 (a mixture of cowpea and sunnhemp) required larger plots, so in this case, a total of three rows of 6 trees were used. Cover crop treatments were arranged using a randomized complete block design with four replicated

PAGE 32

16 blocks, each containing all th e different CC treatments. Different cover crops (summer vs. winter CC) were planted twice a year. During 2002, only a grass fallow was used as the control, while starting in 2003, a tillage fa llow was also included as an experimental treatment. After initial establishment, grass fallows and perennial peanut plots were mowed at 3-4 wk intervals throughout the spri ng, summer, and fall, while tillage fallows were tilled twice a year befo re CC planting. Annual CC were planted with a “zero-till” planter (Sukup 2100, Sukup Manufacturing Compa ny, Shefield, IA) using a suitable row spacing and recommended planting rates as out lined in Table 2.3. Planting dates for both summer and winter CC are presented in Tabl e 2.4. Except for winter CC planted in 2003, when a zero-tillage system was used, previous CC and/or weeds were soil-incorporated with two to three passes of a rototiller usi ng a tillage depth of 10 cm prior to planting cover crops. All leguminous CC were inoculated before planting with the appropriate strain of rhizobium (Nitragin brand, Milw aukee WI). Inoculant and un treated seeds were obtained from local seed companies since varieties from certified organic supply companies are typically poorly adapted to Florida conditions. Cover crops were not irrigated during the growing season, except after plan ting, if soil moisture was inad equate to ensure uniform germination. In this case, 25 mm irrigation was applied uniformly to the entire block to mimic a typical rainfall event. Insect pests in citrus were controlled when needed with allowed products in organic production sy stems following the national organic program standards (USDA, 2000). Data Collection, Measurements, and Analysis Representative sections of row middles of CC were sampled at monthly intervals to evaluate above-ground CC and weed biomass using rectangular sampling frames with an

PAGE 33

17 internal surface area of 0.22 m2. Sample areas were selected in such way that the selected area closely matched the ground cover of both CC and weeds of the entire plot. Sampled areas invariably included a mix of weed s and annual CC. Weed biomass was not segregated into different weed species. Weeds were harvested at ground level while for the CC, the corresponding root system was also excavated. During 2002 and 2003, a more detailed growth analysis of annu al CC was performed and above-ground biomass was separated into stems and leaves. Roots we re washed and cleaned to remove soil and organic debris. Leaf area was determined using a Licor leaf area meter (LI-3000, Li-cor; Lincoln, NE). Groundcover of CC was determined using leaf area index (L AI) values. Aboveground biomass of weeds and CC were determined by dry weight. In order to quantify the effectiveness of annual CC to suppress weed growth, we developed a cover crop/weed index (CCWI). This index consists of ratios of CC and weed biomass (CCWI=CCDW/WeedDW) calculated in each repetition. Th e qualitative interpretation of this index is defined in Table 2.5. Du ring 2002, above-ground weed biomass was determined only at the end of the annual c over crops cycle using representative 0.5-m2 plot areas. During the summer of 2003, weed above-ground biomass was determined at monthly intervals. During 2004 and 2005, weed above-ground biomass was determined at bimonthly intervals using representative 0.22-m2 plot areas. Roots, st ems, and leaves of annual CC and shoots of weeds were oven dr ied at 65 C for 72 hours until constant weight and dry weights were recorded. Afterwards, shoots were ground in a Wiley mill through a 1-mm screen, and a thoroughly mixed por tion (ca. 4 g) was subsequently stored in scintillation vial. Ground tissue was digested using a wet-acid Kjeldahl digestion

PAGE 34

18 (Gallaher, et al., 1975). After digestion, sample s were diluted, filtered, and analyzed for total Kjeldahl N at the UF-IF AS Analytical Research La b (University of Florida, Gainesville, FL) using EPA method 351.2 (Jones and Case, 1991). Analysis of variance was performed on all data using Proc GLM of SAS (SAS Inst. Inc., 2002). Means were compared using the Duncan test (DMRT) with a p-value of 0.05. Results Summer Cover Crops Total precipitation th roughout the growing period of summer cover crops (from June to October) varied between 426 mm in 2002 (drier year) to 1060 mm in 2004 (Table 2.6). Summer 2002 Rainfall was relatively low during 2002, but si nce rainfall was evenly distributed during the growing season, no obvious water stress occurred. During the summer of 2002, sunnhemp had the highest dry matter production, followed by hairy indigo and cowpea, whereas velvet bean perf ormed rather poorly (Table 2.7). Tissue N concentrations were highest for velvet bean followed by hairy indigo and cowpea. Sunnhemp and alyceclover had a relati vely low N concentration, probably due to their relatively high end-of-s eason stem fraction. Overall N accumulation was greatest for sunnhemp followed by hairy indigo and co wpea whereas both velvet bean and alyceclover accumulated relatively little N. In terms of maximum canopy density as LAI, cowpea and hairy indigo had the highest LAI at 6 and 10 weeks, respectively. Due to its rapid canopy closure, cowpea was very effective in early weed suppression, yet its canopy started to thin within 6-10 weeks (Fig. 4.1). Hairy indigo, on the other hand, had a slower init ial canopy development, but

PAGE 35

19 it retained its canopy longer compared to cowpea. Sunnhemp had intermediate canopy development rates, LAI values, and relatively long persistence (Table 2.7). Both velvet bean and alyceclover had sparse canopies. Although velvet bean had la rger leaves and more dense canopies, overall soil coverage was relatively poor due to low pl ant populations. Alycec lover, on the other hand, had high planting densities, but plants were short, l eaves were very small, and canopies were rather sparse. Sunnhemp grew up to 2.4 m high causing some shading of the young citrus trees. Approximate heights for cowpea, hairy indigo, alyceclover, and velvet bean were 0.3, 1.2, 0.3, and 0.3 m, respectively. Weed suppression typically followed CC bi omass production trends and sunnhemp also had the highest CCWI value, translati ng to outstanding weed control, followed by hairy indigo and cowpea, whereas velvet bean provided poor weed control (CCWI < 1), which may be related to the use of a bus hy genotype. Weed growth was reduced 86% by both sunnhemp and cowpea and 83% by hairy indi go compared with the grass fallow. In these cases, only between 10 and 20% of th e soil area was covered with weeds, in contrast to velvet bean which reduced weeds only by 18%. Summer 2003 Rainfall distribution was relatively favorab le in 2003 (Table 2.6), and sunnhemp, cowpea, hairy indigo, lablab, and alyceclove r grew well. Conversely, velvet bean performed relatively poor due to uneven ge rmination, resulting in low plant populations and nodulation was also relatively poor. A nnual peanut performed poorly which was related to incidence of diseases. Perennial pe anut had a very poor performance because of its slow initial establishment and competition from bermudagrass ( Cynodon dactylon L.), which inhibited its initial establishment and growth.

PAGE 36

20 Total dry matter production was again gr eatest for sunnhemp followed by hairy indigo, and cowpea (Table 2.8). Overall shoot N concentrations were relatively high during 2003 compared to 2002, which may be re lated to a build up of soil rhizobial inoculant and more favorable rainfall condi tions. Overall N accumulation was highest for sunnhemp and hairy indigo. In terms of canopy development, lablab, ve lvet bean and cowpea had the earliest canopy closure, which contributed to effectiv e early season weed suppression. Similar to 2002, growth of hairy indigo was initially slow and sunnhemp and hairy indigo developed their maximum LAI values four weeks later than the other CC (Table 2.8). Similar to 2002, a significant lower weed bi omass occurred with alyceclover, ‘Iron Clay’ cowpea, sunnhemp, and lablab, resulti ng in weed suppression of about 94% 85%, 83%, and 65%, compared to the grass fallow (control).These CC provided similar or better weed control than rototilling, wh ich in turn reduc ed weeds by 59% compared to the grass fallow treatment. Nitrogen accumulation by weeds was greatest in the mowed fallow and ranged from 3 to 47 kg N ha-1 for the other treatments, which is much lower compared to N accumulation by superior leguminous CC such as sunnhemp, hairy indigo, and cowpea. Sunnhemp had the highest CCWI, followed by alyceclover and cowpea. Although hairy indigo was a proficient biomass producer, its CCWI value was low due to the proliferation of alyceclover that volun teered in the hairy indigo plots. Summer 2004 Precipitation during 2004 was relatively hi gh in comparison with the other years (Table 2.6), which was related to four majo r hurricanes passing th rough Florida, two of which resulted in rainfall inte nsities in excess of 100 mm day-1. Rainfall during the rest of

PAGE 37

21 the growing season was relatively evenly distributed. During the summer of 2004, pigeon pea and hairy indigo accumulated the greate st biomass (Table 2.9), and ‘Iron Clay’ cowpea clearly outperformed ‘Cream-40’ a commercial cowpea variety. In 2004, sunnhemp only accumulated 5.3 Mg ha-1 which was 46% less than the average dry matter accumulated in 2002 and 2003 (11.1 Mg ha-1) due to infection by Verticillium sp., a soilborne pathogenic fungus. Shoot N concentration was highest fo r velvet bean and ‘Iron Clay’ cowpea followed by lablab, pigeon pea, and perennial peanut. Low N values were obtained for sunnhemp because verticillium wilt resulted in premature leaf drop. Overall N accumulation was greatest for pigeon pea followed by hairy indigo and cowpea. As in previous years, a lower (P< 0.05) weed biomass resu lted with alyceclover and ‘Iron Clay’ cowpea. Use of alyceclove r, ‘Iron Clay’ cowpea, hairy indigo, and pigeon pea resulted in weed suppressi ons of about 97%, 92%, 65%, and 53%, respectively, compared to the grass fallow (c ontrol). Rototilling reduced weeds by 69% in comparison with grass fallow, which was low compared to alyceclover and ‘Iron Clay’ cowpea. Nitrogen accumulation in weeds was greatest in the plots with ‘Cream 40’ cowpea and least for alyceclover. High CCWI values for alyceclover and ‘Iron Clay’cowpea indicated outstanding weed control (Table 2.9). On the other ha nd, velvet bean, lablab, perennial peanut, ‘Cream 40’ cowpea, and mung bean pr ovided relatively poor weed control. Summer 2005 Shoot dry weight was greatest for sunnhe mp, followed by hairy indigo and pigeon pea (Table 2.10). The lower biomass obtaine d for sorghum-sudangrass resulted from the

PAGE 38

22 low seeding rate used, because it was inte nded to only provide s upport for lablab and velvet bean without competing with the leguminous CC. Shoot N concentration was highest for velv et bean and perennial peanut followed by lablab. Overall N accumulation was greate st for sunnhemp followed by hairy indigo and pigeon pea. Several crops had rela tively low N accumulation (< 20 kg N ha-1). Significant reductions in weed biom ass observed for ‘Iron Clay’ cowpea, sunnhemp, and hairy indigo, translated to weed suppressions of about 92%, 90%, and 78%, respectively, compared with the grass fallow (control). Rototilling reduced weed biomass by 51% and was thus less effective compared to sunnhemp and ‘Iron Clay’ cowpea. Nitrogen accumulation in weeds was gr eatest in the mowed fallow (Table 2.10). ‘Iron Clay’ cowpea had the highest CCWI followed by sunnhemp and hairy indigo, which is indicative of outstanding weed contro l. Mixing of lablab w ith velvet bean and sorghum-sudangrass provided an excellent w eed control as well, but intercropping of velvet bean with sorghum-sudangrass pr ovided only moderate weed control. Winter Cover Crops Monthly total precipitation amounts duri ng the growing season of winter CC are shown in Table 2.11. Total precipitation thr oughout the growing pe riod varied between 334 in 2003/04 to 472 mm in 2002/03 (Table 2.11). Winter 2002/2003 Rainfall was relatively high during th e 2002/03 winter growing season in comparison with other years, but since rainfall was uneve nly distributed during the growing season, obvious water stress did occu r. During the winter of 2002/03 lupin had the highest dry matter production, whereas re d, berseem, and sweet clover performed rather poorly (Table 2.12).

PAGE 39

23 Tissue N concentrations were highest for red clover followed by the other leguminous CC, while winter rye had the lowest value. Overall N accumulation was greatest for lupin. In terms of maximum canop y density, planting of rye resulted in a more rapid increase in LAI compared to other CC (Fig. 2.2). Leguminous CC had a slower initial canopy development and in comparison with summer CC, canopy densities were also lower. Rye and crimson clover had somewhat higher canopy densities, sweet clover intermediate values, and all other cr ops had very sparse canopies (Table 2.12). Canopy persistence was best fo r crimson clover but poor for most other tested species (Fig. 2.2). Red clover had the greatest weed biomass in comparison with the other CC. Use of rye reduced weed growth by 92% compared to a 64% reducti on for sweet clover. Use of other leguminous crops like lupin, berseem and crimson clover, and cahaba vetch reduced weed growth by about 50%. On the other hand, red clover decreased weed biomass by only 32%. Nitrogen accumulation in weeds wa s greatest in the grass fallow and red clover, while weeds in rye had the lowest value, but it was not significantly different from other CC except red clover. Rye had the greatest CCWI (excellent weed control) followed by crimson clover and lupin which provided moderate weed cont rol. Use of other legumes did not greatly affect weed growth and thus provided poor weed control due to low plant populations associated with low germination, ineffectiv e nodulation, and poor ad aptation to Florida soils and environmental conditions. Winter 2003/2004 An uneven distribution and lower precipit ation along with lack of soil tillage hampered growth of some CC during the 2003-04 growing season. Despite unfavorable

PAGE 40

24 growth conditions, radish had the hi ghest dry matter production (3.2 Mg ha-1). Biomass of rye, crimson clover, black oat, lupin, sweet clover, and subterranean clover were intermediate; while hairy vetc h and lupin accumulated the lowest biomass (Table 2.13). Nitrogen in shoots was highest for hairy vetch followed by lupin, crimson clover, and subterranean clover. Non-leguminous CC ha d the lowest N concentrations in shoots (<13 g N kg-1). Overall N accumulation was greatest for crimson clover, rye, and radish, whereas, lupin, hairy vetch, subterranean cl over while black oat accumulated less than 15 kg N ha-1. In terms of actual weed suppression, planti ng radish resulted in the lowest weed biomass (Table 2.13), with only between 5 to 10% of the soil area covered with weeds. Use of radish reduced weed growth by 88%, followed by crimson clover, rye, and hairy vetch which decreased weed growth by 68-71%. The reduction in weed biomass for other leguminous crops like lupin and subterra nean clover was low, only about 16%. Rototilling reduced weed biomass by 56% in comparison with grass fallow. Nitrogen accumulation in weeds was greatest for grass fallow followed by subterranean clover and lupin. The other treatments including the tillage fallow had relatively low N accumulation in weeds (<20 kg N ha-1). Overall CCWI values were greatest for radish which provided outstanding weed control whereas values for rye and crimson clover were intermediate, indicative of moderate weed control. Use of other leguminous CC did not greatly reduce weed grow th and resulted in poor weed control. Winter 2004/2005 Less precipitation occurred in winter 2004-05 compared to winter 2002, but overall distribution was relatively even throughout the entire growing season. During the winter of 2004/05, radish intercropped with rye and crimson clover had the highest dry matter

PAGE 41

25 production followed by rye, rye intercropped wi th crimson clover, crimson clover, black oat intercropped with crimson clover radish and black oat mixed with crimson clover and radish all produced 4.3 Mg ha-1, whereas other CC produced between 2.8 and 3.6 Mg ha-1 (Table 2.14). Shoot N concentration was significantly greater for crimson clover and black oat+crimson clover. Non-leguminous CC had th e lowest N concentrations in shoots. Overall N accumulation was greatest for crimson clover and rye+crimson clover+ radish due to the higher biomass accumulation, whereas black oat accumulated relatively little N. In terms of weed suppression, all CC tr eatments were statistically similar and resulted in a weed biomass less than 0.73 Mg ha-1 which translated to a reduction of weed growth by 80% for radish up to 98% for eith er the double or triple mix of cover crops. Rototilling reduced weeds by 53% in comparison with grass fallow. Weed N accumulation was statistically similar and less than 12 Kg N ha-1 for all CC treatments, whereas grass fallow greatly exceeded this le vel (Table 2.14). The CCWI values varied from excellent for radish to outstanding for the other monocrops and intercropping systems. Discussion Summer Cover Crops In general, sunnhemp was the most pr olific summer CC. However, during 2004 continiuous cultivation of sunnhemp along w ith wet and windy conditions increased the dispersal and incidence of verticillium in fection, thereby reducing biomass accumulation, N fixation, and weed suppression (CCWI= 2 in 2004 vs. 64 in 2005 when the crop was properly rotated). Based on this, it is obvi ous that despite the fact that sunnhemp

PAGE 42

26 performance was superior to most other summ er CC, repetitive use in the same site may be undesirable and use of sound crop rotati on with CC such as cowpea should be considered. Overall sunnhemp dry weight (DW) accumulation was 10 Mg ha-1, which was similar to the findings of Ramos et al (2001), Steinmaier a nd Ngoliya (2001), and Perin et al. (2004) under tropical conditions, but superior to the results from Jeranyama et al. (2000) and Balkcom and Reeves (2005). Hairy indigo performed consistently and average DW production was 8.0 Mg ha-1 similar to values reported by Reddy et al (1986a). Overall DW accumulation for pigeon pea was 5.7 Mg ha-1. Reported values ranged from 4.5 Mg ha-1 under tropical conditions (Mafongoya and Dzowela, 1999) to 9.5 Mg ha-1 in Florida (Reddy et al., 1986a). The steep decline in DW accumulation in 2005 wa s related to hairy indigo volunteering in plots during 2005. ‘Iron Clay’ cowpea averaged 3.5 Mg ha-1, similar to values reported by Jerenyama et al., (2000) a nd Muir (2002). Lablab had in consistent biomass production and overall DW accumulation was only 2.2 Mg ha-1. Similar values were reported by Muir (2002) but other authors reported values ranging from 3.8 to 8.0 Mg ha-1 (Fischler and Wortmann, 1999; Wortmann et al., 2000; Ca rsky et al., 2001; and Steinmaier and Ngoliya, 2001). Alyceclover accumulated 2.6 Mg ha-1 while velvet bean produced only 1.7 Mg ha-1 which was similar to the findings of Creamer and Baldwin (2000) but lower than values ranging between 3.6 and 9.1 Mg ha-1 reported by Wortmann et al. (2000), Steinmaier and Ngoliya (2001), and Carsky et al. (2001). This may be due to the low germination rate and sparse CC population. Dry matter production of peanut and perennial peanut in 2003, ‘Cream 40’ cowpea, mung bean, and perennial p eanut in 2004 was low, due to their poor

PAGE 43

27 adaptation to the sandy soils and their ineff ectiveness to successf ully compete with weeds. Overall N accumulation by sunnhemp was 148 kg N ha-1. Similar values were reported by Balkcom and Reeves (2005) while ot her studies showed a range from 195 to 305 kg N ha-1 (Ramos et al., 2001; Steinmaier and Ngoliya, 2001; Perin et al., 2004). Overall N accumulation for hairy indigo was 132 kg N ha-1 which was similar to values reported by Reddy et al. (1986a ). Cowpea produced 61.4 kg N ha-1 which was greater than the findings by Jerenyama et al. (2000) and by Muir (2002). Pigeon pea accumulated an average of 120 kg N ha-1 which was lower than the value (170 kg N ha-1) reported by Tian et al. (2000), which may be related w ith the decrease in pigeon pea DW in 2005 above discussed and the use of a mo re compact variety in our study. Lablab accumulated N at 60 kg N ha-1, which was much less than the results of 113, 137 and 177 kg N ha-1 reported by Steinmaier and Ngoliy a (2001), Carsky et al.(2001), and McDonald et al. (2001), respectively. This may be related to inconsistent performance of lablab across years. Velvet bean produced only 50 kg N ha-1, much lower than values of 163 and 281 kg N ha-1 reported by Steinmaier and Ngoliya (2001) and Carsky et al. (2001), respectively. This may be related to the use of a bushy type of velvetbean that did seem to be less vigorous than the more commonly used vining types. Overall weed reduction was highest for ‘Iron Clay’ cowpea (90% of control) followed by sunnhemp (77%), alyceclover (74% ), and hairy indigo (64%). Fallow tillage was less effective in reducing weeds compared to the best CC with an average reduction of 60%. Weed reduction for alyceclover and hairy indigo probably could have been higher due to in 2003 alyceclover volunteered in hairy indigo plots and in 2005 hairy

PAGE 44

28 indigo volunteered in alyceclover plots. Use of velvet bean, on the other hand, reduced weed growth by only 42%, which contrasted with the 68% reduction in a study with maize by Caamal-Maldonado et al. (2001). This may be related to the bushy variety used in the current study, along w ith the low germination in field, wider row spacing, and stronger weed competition. The presence of alyceclover, hairy in digo, and cowpea as volunteer crops shows that some of these crops may have good poten tial for reseeding duri ng subsequent years, due to their capacity to produce large number s of dormant and/or hard coated seeds. These species would be able to become pa rt of the soil seed bank and germinate over many years as suggested by Benech-Arnold et al. (2000), which could provide a costeffective self sustaining practi ce in a mature citrus system. Provided that via use of sound crop rotati on the build up of disease in sunnhemp plots was prevented, sunnhemp and cowp ea provided outstanding weed control and CCWI values were 33 and 31, respectivel y. Corresponding values for alyceclover and hairy indigo were 14 and 11 (excellent weed control). Except for alyceclover, this suppression was closely relate d with CC biomass production which appears to be the main mechanism for weed suppression due to direct competition for resources (light, nutrients and water). Velvet bean was moderately effectiv e in suppressing weeds (CCWI=1.2), which contrasted with the results from Caam al-Maldonado et al. (2001) and Buckles and Triomphe (1999) in central America and from the findings of Carsky et al. (2001) and Fishler and Wortmann (1999) in Africa, w ho obtained excellent weed suppression in

PAGE 45

29 maize with a viny cultivar. These differences may be related to the poor germination, nodulation, vigor and competitiveness of the bus hy type that was used for our studies. Winter Cover Crops Radish and rye were the most prolific biomass producers among the mono-cropped winter CC. In our study radish accumulated 4.6 Mg ha-1 compared to 1.6 Mg ha-1 reported by Vyn et al. (2000). However, Justes et al. (1999) reporte d values up to 6.4 Mg ha-1 with no N added. Rye generated 3.5 Mg ha-1 which was similar to results of Ngouajio and Mennan (2005) and Akemo et al. (2000), but lower compared to the 5-10 Mg ha-1 reported by Bauer and Reeves (1999) and Re berg-Horton et al. (2005). Crimson clover yielded 2.9 Mg ha-1 which was similar to values repor ted by Daniel et al. (1999) and Schomberg and Endale (2004) but lower than the 4.9 Mg ha-1 reported by Dyck et al. (1995) and Odhiambo and Bomke (2001). Use of a triple CC mix (rye+crimson clove r+radish ) resulted in the greatest biomass (8.8 Mg ha-1), which was about two times greater than the 4.6 Mg ha-1 for radish planted alone. Karpenstein-M achan and Stuelpnagel (2000) reported similar findings for a mixed CC system in Germany consisting of rye and crimson clover, and for rye with winterpea ( Pisum sativum L. ssp. arvense (L.) Poir). Similar results were reported by Juskiw et al. (2000) for sma ll grain cereals in Canada. During the 2004-2005 winter season, DW accumulation of triple CC mixes was comparable to the 8-9 Mg ha-1 produced by summer CC syst ems in 2004 but lower than the 10-14 Mg ha-1 obtained in 2005. This may be related to the synergistic combination of the complementary traits of the constituents of the mix, with rye providing vigorous and rapid growth along with allel opathic activity, radish break ing through compaction layers and enhancing biodiversity and soil struct ure, and while crimson clover providing

PAGE 46

30 additional N via N fixation. This synergistic in teraction of compleme ntarities in root and canopy structure may facilitate improved adap tation to different ecological niches, soil types and weather conditions, providing multiple benefits and improved nutrient retention, cycling, and N-fixati on as suggested by Gliessman (1998), Altieri (1999), and Karpenstein-Machan and St uelpnagel (2000). Mixed CC systems thus mimic natural systems and crop components may therefore compete more effectively with weeds, which explains the superior performance of these systems. Dry weight accumulation by red, berseem, sw eet and subterranean clover, cahaba white vetch, and lupin was relatively low. This may be related to the uneven rainfall distribution during 2002/03 and 2003/2004; to low soil organic matter, pH, and K values; to poor initial nodulation and growth by these cr ops, hampering their ability to effectively compete with weeds; and to an overall poor adaptation of these crops to coarse sandy soils. Row middles were not fertilized and ch icken manure was applied only to the tree rows in order to reduce weed vigor and to provide leguminous CC with a competitive edge. Besides this, lower temperatures and lig ht intensities during winter and the uneven rainfall distribution in comp arison with other years could have hampered CC growth. Overall crop N accumulation was greater for leguminous CC probably due to N fixation. Crimson clover accumulated 70 kg N ha-1. Similar values were reported by Daniel et al. (1999) whil e others recorded values were between 120-125 kg N ha-1 (Dyck et al. 1995; Odhiambo and Bomke, 2001). L upin and radish both accumulated around 47 kg N ha-1. Rye and black oat averaged 32 and 19 kg N ha-1, respectively, which was lower than values for other studies (Bau er and Reeves, 1999; Odhiambo and Bomke, 2001).

PAGE 47

31 Intercropping of rye+crimson clover+radish resulted in the hi ghest N accumulation (110 kg N ha-1) whereas black oat+crimson+radish accumulated 74 kg N ha-1. The high N accumulation by these CC systems may be related to N retention (by rye, radish, and black oat) and/or additional N-fixation (cri mson clover) as suggested by Justes et al. (1999), Vos and van der Putten (2001), and Kr istensen and Thorup-Kristensen (2004) for rye and radish. Similar to the positive effects of mixed CC systems on N accumulation, intercropping two and three-way-mixtures reduced weed growth by 98% compared to the mowed fallow. Corresponding values for crimson clover, rye, and radish monocrops were 78-84%. For rye, weed suppression may be rela ted to allelopathy which is often reported in the literature (Weston, 1996; Fennimore and Jackson, 2003; Reberg-Horton et al., 2005). Compared to the best CC, fallow tillage was less effective in reducing weeds, with reductions in weed biomass averaging only 55 %. Similarly, twoand three-componentmixtures resulted in the highest CCWI valu es. The outstanding weed control of mixed systems may be related to competition and syne rgistic allelopathic activities of radish, possibly due to the glucosinolate content repo rted for radish and other members of the Brassicae family (Norsworthy et al ., 2005, Morra and Kirkegaard, 2002). Similar allelopathic action has been also reported for black oat (Bauer and Reeves, 1999). System Dynamics Summer CC had greater DW production capac ity compared to winter CC. The four highest biomass producers were in descending order: s unnhemp>hairy indigo>pigeon pea>cowpea with respective values of 10, 8, 5.7, and 3.5 Mg ha-1 which translates to 83, 67, 48, and 29 kg ha-1 d-1, respectively, which was similar to the values reported by

PAGE 48

32 Reddy et al. (1986a), Mafongoya and Dzowela (1999), Jerenyama et al. (2000), and Perin et al. (2004). When weeds were included th e sequence was as follows: sunnhemp>hairy indigo>pigeon pea>‘Cream-40’ cowpea, w ith corresponding daily DW accumulation values of 93, 81, 69, 49 kg ha-1 d-1. The above results contrasted with the biomass produced by the system during the winter season, in which monocrops of radish, rye, crimson clover, and black oat produced 3.8, 3.4, 2.9, and 2.4 Mg ha-1, with an average of 3.1 Mg ha-1per season and 25, 23, 19, and 16 kg ha-1 day-1, respectively, which were lower than the values reported by Justes et al. (1999), Odhiambo and Bomke (2001), and Re berg-Horton et al. (2005). When weeds were included in the balance, the total bi omass production by the system was greater for radish, rye, crimson clover, and black oa t with values between 4.4 and 3.6 Mg ha-1, with an average of 4.0 Mg ha-1 per season and 27 kg ha-1 d-1, which was much less than that obtained during summer season. This due ma y be to the lower radiation (138 w m-2 in winter vs. 188 w m-2 in summer)and temperatures in wint er (average temp eratures from 2002 to 2005 in winter were 14.2 C, min temp= -2.2 C max temp= 29.2 C vs. average temperatures in summer 26.3 C, min temp = 18.8 C and max temp= 36.6 C). In addition, rainfall was higher and relatively more evenly distributed during the summer. Finally, most winter CC may not be well-adapte d to growth environments in Florida, and nodulation of many leguminous crops tends to be erratic on sandy soils during the first few years of their cultivation. Use of intercropping allowed for increases in production capacity during the winter months due probably to the synergistic in teraction among crops as suggested by Kabir and Koide (2002) and by Karpenstein-Mach an and Stuelpnagel (2000). For winter

PAGE 49

33 intercropping CC, the biomass pr oduction was between 8.0 to 2.8 Mg ha-1, with an average of 5.3 Mg ha-1 per season or 35 kg ha-1 d-1. In general, N accumulation was greater for leguminous species probably due to N fixation and was greatest duri ng the summer season. The four highest N accumulators were in descending order: sunnhemp>hairy indigo>pigeon pea>cowpea with values between 61 to 148 kg N ha-1, with an average of 129 kg N ha-1 per season or 1.1 kg N ha-1 kg ha-1 d-1, similar to the values reported by others (Reddy et al., 1986a; Mafongoya and Dzowela, 1999; Jerenyama et al., 2000; Ra mos et al., 2001; Muir et al., 2002). When weeds were included in the balance, the total N content in the system was greater for pigeon pea, hairy indigo, sunnhemp, and velvet bean, with values between 123 and 222 kg N ha-1, with an average of 174 kg N ha-1 per season or 1.5 kg N ha-1 d-1, which was significantly greater than the N accu mulated only by the mowed fallow (66 kg N ha1). These results underline the ecological role of weeds in the system in capturing C and N within the system, because without this component, an important fraction of N could be lost from the system through leaching or runoff as proposed by Vos and van der Putten, (2001) and by Kristensen and Thorup-Kristensen, (2004). The four highest N accumulators in winter were in descending order: crimson clover>radish >lupin>rye with valu es ranging between 32 to 69 kg N ha-1. The N accumulation average was 49 kg N ha-1 per season (38% of N accumulation summer average) or 0.3 kg N ha-1 d-1, which were lower than the va lues reported by Justes et al. (1999), Odhiambo and Bomke (2001), and Rebe rg-Horton et al. (2005). When weeds were included in the balance, N in the system was greatest for crimson clover followed by lupin, red clover, and ra dish, with values ranging between 84 to 58 kg N ha-1, with an

PAGE 50

34 average of 69 N kg ha-1 per season or about 0.5 kg N ha-1d-1, which was 40% of that obtained with summer cover crops, due to si milar reasons explained above for biomass production. Use of mixed crop systems increased crop performance due to synergetic interactions of crop components, and N accumulation ranged from 55 to 110 kg N ha-1, with an average of 0.5 kg N ha-1 d-1 Although this is 67 % higher compared to monocrop systems, it is still only 45 % of N accumulated by summer CC systems. Nitrogen accumulation values for intercroppi ng system were similar to the values reported by Justes et al. (1999); Odhiambo and Bomke ( 2001); Reberg-Horton et al. (2005) and superior to N accumulated in the control (mowed fallow) which amounted to 53 kg N ha-1 season-1. In terms of weed suppression, summe r monocrops (sunnhemp, cowpea, hairy indigo and alyceclover) were more effective in outcompeting weeds compared to monocropped winter CC systems (rye, radish, and crimson clover). The CCWI values for the winter CC were below 13, which was mainly associated with the lower biomass production by winter CC discu ssed above. Actual weed reduction was similar in both seasons, with average weed reductions of 76 and 80% for superior summer and winter CC, respectively. Volunteering of mainly al yceclover and hairy indigo in the other treatments at times created potential but ma y be arbitrary “weed” issues in subsequent crops. Winter intercropped CC had CCWI gr eater than 39 due to higher biomass and probably the allelopathic s uppression of weeds by rye and black oat as discussed by Weston (1996) and Putnam (1988).

PAGE 51

35 Observed weeds species did not follow any special pattern associated with CC treatments. The main species observed in fi eld during late spring, summer, and early fall (warm-season weeds) were: bermudagrass ( Cynodon dactylon (L.) Pers.), large crabgrass ( Digitaria sanguinalis (L.), bahiagrass ( Paspalum notatum Fluegge), goosegrass ( Eleusine indica (L.) Gaertn), Scop ), crowfootgrass ( Dactyloctenium aegyptium (L.) Willd.), globe sedge ( Cyperus globulosus Aubl.), cylindric sedge ( Cyperus retrorsus Chapm.), Florida pusley ( Richardia scabra L.), and carpetweed ( Mollugo verticillata L.) among the species more prevalent and freque ntly observed in fi eld; whereas purple nutsedge ( Cyperus rotundus L.), spreading dayflower ( Commelina diffusa Burm. F.), common pigweed ( Amaranthus hybridus L.), common ragweed ( Ambrosia artemisiifolia L.), southern sida ( Sida acuta Burm. F.), common purslane ( Portulaca oleracea L.), poorjoe ( Diodia teres Walt.) were observed only in local ized spots; and in some plots, alyceclover ( Alysicarpus vaginalis (L.) DC.) and hairy indigo ( Indigofera hirsuta L.) volunteered during subsequent summer seasons. In winter and early spring, the dom inant weeds found were red sorrel ( Rumex acetosella L.), oldfield toadflax ( Linaria canadensis (L) Dumont), common venuslookingglass ( Triodanis perfoliata (L.) Nieuwl.), wandering cudweed ( Gnaphalium pensylvanicum Willd.); whereas virg inia pepperweed ( Lepidium virginicum L.), Carolina geranium ( Geranium carolinianum L.), and cutleaf evening primrose ( Oenothera laciniata Hill) were only obser ved in certain spots. There appears to be excellent prospectives for the use of sunnhemp, ‘Iron Clay’ cowpea, alyceclover, and hairy indigo as su mmer CC for weed suppression in organic Florida citrus systems. However, hairy i ndigo and sunnhemp appeared to have some

PAGE 52

36 drawbacks relative to cowpea and alyceclove r. Hairy indigo grew relatively tall and bushy and since it is a hard-seed ed crop, seeds could persist for a long time in the soil seed bank (Benech-Arnold et al., 2000) with a weedy potential around young citrus trees, so hairy indigo would have to be mowed befo re it goes to seed when used in newlyplanted orchards. Sunnhemp, although it did not branch profusely, grew very tall so that it might create problems with shading if it were planted too close to young trees. Also sunnhemp appears to need rotation to a void problems with soil fungi such as Verticillium spp. Alyceclover and ‘Iron Clay’ cowpea had more compact low-growing canopies, which would facilitate their integration in to citrus production systems. Under our experimental conditions, ‘Iron Clay’ cowpea reseed ed itself, but seed vi ability in the field was less than one year, so when mowed in time it could be managed more easily than hairy indigo. However, some selections can be rather “viny” and grow around young trees, which may limit its use as a cover crop in the near vicinity of small trees (<2 m), although mechanical weeding can easily addr ess this potential problem. Even though alyceclover reseeded readily, it did not seem to interfere with citrus trees or citrus irrigation and therefore may be the most suitabl e species to be planted in the vicinity of trees. Based on this, we propose for tree rows a “sandwich” system consisting of a tree strip of 1.6-1.8 m planted with alyceclover as a summer CC. Bordering this strip would be ‘Iron Clay’ cowpea. If it would be desirable to apply mulch to the tree row during the winter time, a strip of sunnhemp could be planted in the row middle for this purpose. However,

PAGE 53

37 alternatively use of ‘Iron Clay’ cowpea in ro w middles may be preferable as discussed above. During the winter season we propose the us e of a system consisting of a tree strip planted with a mixture of crimson clover and black oat as winter CC due to their compact canopy and low probability of competition with citrus trees for light. Intercropping rye with crimson clover and radish w ould be desirable for row middles. However, seed availability of suitabl e winter CC including the use of seeds produced in different continents may pose so me problems. Limited availability of non treated/certified organic seed s ources of varieties and/or cult ivars adapted to the southeast of United States appears to be one of the key issues that may hamper effective integration of CC in organic citrus production systems. Increased N accumulation in CC-based systems during the summer season may provide benefits to subsequent CC crops and/ or citrus trees via mineralization. Use of continuous CC sequence may also reduce potenti al nutrient losses due to leaching (Vos and van der Putten, 2001; Kristensen and Thorup-Kristensen, 2004). Continuous growth of CC combined with reduced tillage may also enhance C sequestration and N cycling and retention in the soil (see Chapter 5 for more detailed discussion about CC effect on soil quality). Au gmented soil organic matter is considered a desirable characteristic of sustainable systems. In organic systems, this approach may also foster the development of soils that can enhance natural suppression of: weeds (Gallandt et al., 1999; Jordan et al., 2000), soil borne diseases (Bailey and Lazarovits, 2003), and insect populations (Altieri and Nic hols, 2003). All these processes are related through the mechanism of incr eased soil organic matter a nd soil microbial diversity

PAGE 54

38 (Kennedy, 1998; Kremer and Li, 2003). As a re sult, such an approach may be costeffective due to reduced requi rements of external inputs. Conclusions Overall dry matter, N accumulation, and weed suppression by annual CC varied depending on plant species and season. In general, summer CC had the highest biomass and N accumulation, in which the more consistent performers in terms of biomass production, N accumulation, and weed suppres sion were sunnhemp, hairy indigo, cowpea, and alyceclover. Although pigeon pea was consistent for biomass and N accumulation, its weed suppression capacity was not always consistent. The most consistent and best performing winter CC were radish, rye, and crimson clover. The best summer and winter CC, DW production averaged 6.8 and 3.1 Mg ha-1, respectively while corresponding total biom ass (CC + weeds) were 9.7 and 4.0 Mg ha-1. Cover crop N accumulation averaged 129 and 49 kg ha-1 during summer and winter seasons, respectively and total N accumul ation (CC + weeds) was 174 and 69 kg N ha-1, which underlines the complementary role of weeds in nutrient rete ntion and recycling. Throughout the course of the study, use of sele cted CC provided excellent weed control, which was superior to other methods includi ng tillage. Use of twoor three-component winter CC mixes resulted in higher DW a nd N accumulation and more effective weed suppression, due probably to the synergis tic interaction among system components.

PAGE 55

39 Table 2.1. Overview of annual summer and wi nter cover crops used during the 2002 and 2003 growing seasons. Cropping Season Treatment Summer Winter/Spring 2002 1 ‘Iron Clay’ cowpea ( Vigna unguiculata L. Walp.) Crimson Clover ( Trifolium incarnatum L. ) 2 Velvet bean ( Mucuna atropurpureum L.) DC) Red clover ( Trifolium pratense L.) 3 Sunnhemp ( Crotalaria juncea L.) Lupin/ Cahaba vetch ( Lupinus angustifolius L ./Vicia sativa L.) 4 Alyceclover ( Alysicarpus vaginalis L.) Rye ( Secale cereale L.) 5 Hairy indigo ( Indigofera hirsuta L.) Berseem Clover/Sweet clover ( Trifolium alexandrinum L./ Melilotus officinalis L.) 6 Grass fallow Grass fallow 2003 1 Sunnhemp/ ‘Iron Clay’ cowpea Rye/Hairy vetch ( S. cereale / Vicia villosa Roth) 2 ‘Iron Clay’ cowpea Crimson clover 3 Velvet bean Radish ( Raphanus sativus cv. rufus) 4 Hairy indigo Black oat/Lupin ( Avena strigosa Schreb/ Lupinus angustifolius L.) 5 Lablab ( Lablab purpureus L.) Subterranean clover ( Trifolium subterraneum L.) 6 Peanut ( Arachis hypogea L.) Lupin 7 Perennial peanut ( Arachis glabrata Benth.) Perennial peanut 8 Grass fallow Grass fallow 9 Tillage fallow Tillage fallow 10 Alyceclover

PAGE 56

40 Table 2.2. Overview of annual summer and wi nter cover crops us ed during the 2004 and 2005 growing seasons. Cropping season Treatment Summer Winter/Spring 2004 1 Sunnhemp/ ‘Iron Clay’ cowpea Winter rye (WR) 2 ‘Iron Clay’ cowpea Crimson clover (CR) 3 Hairy indigo Radish (R) 4 Velvet bean Black oat (BO) 5 Pigeon pea ( Cajanus cajan L.) WR + CR 6 Lablab BO + CR 7 Perennial Peanut Perennial peanut 8 Grass fallow Grass fallow 9 Tillage fallow Tillage fallow 10 Cream-40 cowpea WR + CR + R 11 Mung bean BO + CR + R 12 Alyceclover 2005 1 Velvet bean/Sudan grass Winter rye (WR) 2 Sunnhemp/ ‘Iron Clay’ cowpea Crimson clover (CR) 3 Hairy indigo Radish (R ) 4 Alyceclover Hairy vetch (HV) 5 Lablab/velvet bean WR + CR 6 Lablab/sorghum sudan grass ( Sorghum bicolor L.) WR + CR+R 7 Perennial Peanut Perennial peanut 8 Grass fallow Grass fallow 9 Tillage fallow Tillage fallow 10 Pigeon pea WR/HV (67-33%) 11 WR/HV (33-67%)

PAGE 57

41 Table 2.3. Overview of seeding rates, space be tween rows and cultivars used as annual summer and winter cover crops used from 2002 to 2005. Species Seeding rate (kg ha-1) Space between plants (cm) Cultivar Pigeon pea 67 36 Velvet bean 56 36 Bushy type Cowpea 56 36 ‘Cream-40’ (’04); ‘Iron Clay’ (’02’03&’05) Sunnhemp 39 36 Mung bean 28 18 Lablab 22 36 Alyceclover 22 18 Sudan grass 22 36 Hairy indigo 11 36 Winter rye 112 (55 in ’03-’04) 36 Abruzzi (’02’03&’05) Florida 401 (2004) Black oat 112 36 Soil saver Lupin 112 36 Tifblue Crimson clover 28 18 AB Dixie Subterranean clover 28 18 Mt Barker Red Clover 28 18 Cahaba white vetch 22 18 Hairy vetch 22 (11 in ’03-’04) 18 Berseem clover 22 18 Radish 22 36 Rufus

PAGE 58

42 Table 2.4. Outline of planting and harvest date s and duration of summ er and winter cover crops. Summer Winter Year Planting Mowing Duration (days) Planting Mowing Duration (days) 2002 30 July 11 Oct 102 1 Dec 15 May 165 2003 10 June 16 Oct 127 28 Oct 31 March 154 2004 11 June 10 Oct 121 1 Nov 5 April 156 2005 21 June 25 Oct 125 5 Dec 9 May 155 Table 2.5. Outline of cover crop weed index (CCWI) categories. CCWI value Cover crop Weed Weed control < 0.5 CC not competitive Weeds dominate Very poor (>70% weeds) 0.5-1 CC coexist Weeds coexist Poor 1-3 CC start prevailing Weeds prevail in niches Moderate 3-5 CC prevail Weeds fail to dominate Adequate 5-15 CC predominate (70-90%) < 10-30% weeds Excellent >15 CC dominate completely <5% weeds Outstanding It is assumed if CCWI >15 then weed contro l is considered outstanding since weeds only cover account for less than 5% of the total biom ass. It should be noted that in the absence of weeds the CCWI will approach infinity, and the upper boundary is thus not defined.

PAGE 59

43 Table 2.6. Rainfall measured at Plant Res earch and Education Unit (Citra) Florida Automated Weather station Net work (FAWN)1 during the 2002-2005 summer CC growing season. Year Month 2002 2003 2004 2005 -----------------------------Ra infall (mm) --------------------------June 155 124 123 July 139 127 269 99 August 150 145 157 193 Sept 120 100 418 100 October 15 81 93 120 November 2 Total 426 608 1060 636 1 Data obtained from the website http //fawn.ifas.ufl.edu on 12/10/2005. Blank spaces mean period of time when CC were not grown.

PAGE 60

44 Table 2.7. Shoot dry weight accumulation (DW) ; shoot N concentration (Nconc); shoot N accumulation (Naccum), maximum obser ved leaf area index (LAI max) with time of occurrence in brackets; weed dry weight accumulation (DW);and cover crop weed index (CCWI) for su mmer cover crops grown during the 2002 growing season. Cover Crops Weeds Treatment DW Mg ha-1 Nconc g kg-1 Naccum kg N ha-1 LAI max m2m-2 (wks) DW Mg ha-1 CCWI Cowpea 3.89 bc† 16.5 bc 61.2 c 5.3 a (6) 0.61 b 11.8 b Velvet bean 0.98 c 24.1 a 23.5 c 2.4 b (6) 3.62 a 0.8 c Sunnhemp 12.06 a 13.2 c 158.3 a 3.4 b (10) 0.64 b 22.0 a Alyceclover 2.49 c 14.6 c 36.9 c 2.3 b (6) 1.21 b 3.6 c Hairy indigo 5.87 b 18.7 b 106.5 b 5.9 a (10) 0.74 b 14.9 a Grass Fallow ----4.43 a -† Means within the same column followed by the same letter do not differ statistically based on the Duncan’s Multiple Range test (P<0.05).

PAGE 61

45 Table 2.8. Shoot dry weight accumulation (DW); shoot N concentration (Nconc); shoot N accumulation (Naccum), maximum observed leaf area index (LAI max) with time of occurrence in brackets; weed dry weight accumulation (DW); weed N accumulation (Naccum); and cover crop weed index (CCWI) for summer cover crops grown during the 2003 growing season Cover Crops Weeds Treatment DW Mg ha-1 Nconc g kg-1 Naccum kg N ha-1 LAI max m2m-2 (wks) DW Mg ha-1 Naccum kg N ha-1 CCWI Cowpea (I)‡ 1.38 cb† 24.9 b 34.6 c 2.04 bc (12) 0.95 bc 18.3 cd 4.9 b Sunnhemp 10.24 a 21.8 b 223.1 a 3.99 a (16) 0.68 c 16.5 cd 44.2 a Cowpea 2.37 bc 24.9 b 59.2 c 3.40 a (12) 0.60c 10.8 d 16.8 ab Velvet Bean 2.78 bc 33.5 a 93.0 bc 3.71 a (12) 2.04 ab 37.1 bc 1.6 b Hairy indigo 9.17 a 17.9 c 162.8 ab 3.22 a (16) 2.72 a 39.1 abc 4.1 b Lablab 3.71 b 27.8 b 100.4 bc 3.86 a (12) 1.36 bc 27.6 bcd 3.4 b Peanut 1.07 bc 27.7 b 29.1 c 1.02 bc (12) 2.89 a 46.7 ab 0.4 b Perennial P 0.02 c 20.7 bc 0.41 d 0.01 d (16) 2.56 a 43.5 ab 0.01 b Alyceclover 3.91 b 18.0 c 70.4 bc 2.5 ab (16) 0.25 c 3.2 d 22.4 ab Grass Fallow ----3.90 a 60.1 a -Tillage Fallow ----1.58 bc 33.8 bcd -† Means within the same column followed by the same letter do not differ statistically based on the Duncan’s Multiple Range test (P<0.05). ‡ ‘Iron Clay’ cowpea intercropped with sunnhemp

PAGE 62

46 Table 2.9. Shoot dry weight accumulation (DW); shoot N concentration (Nconc); shoot N accumulation (Naccum); weed dry weight accumulation (DW); weed N accumulation (Naccum); and cover crop weed index (CCWI) for summer cover crops during the 2004 growing season. Cover Crops Weeds Treatment DW Mg ha-1 Nconc g kg-1 Naccum kg N ha-1 DW Mg ha-1 Naccum kg N ha-1 CCWI ‘Cream 40’ cowpea 0.72 de† 19.1 cd 14.2 bc 5.22 a 97.2 a 0.2 c Mung bean 0.24 e 16.0 de 4.0 c 4.87 ab 85.4 ab 0.1 c Alyceclover 2.89 c 17.0 d 49.5 bc 0.43 d 4.5 e 28.0 a Sunnhemp 5.31 b 12.0 e 63.45 b 2.86 cb 42.9 bcde 2.1 c ‘Iron Clay’ Cowpea 5.08 b 25.8 b 63.9 b 0.16 d 22.6 de 15.5 ab Hairy indigo 7.59 a 16.7 d 127.0 a 1.90 cd 28.6 cde 6.5 bc Velvet Bean 1.28 cde 34.0 a 43.7 bc 3.84 abc 72.5 abc 0.6 c Pigeon Pea 7.60 a 22.2 bc 174.7 a 2.56 c 47.1 bcde 3.8 c Lablab 0.76 de 24.8 b 20.2 bc 3.53 abc 87.6 ab 0.3 c Perennial P. 0.16 e 23.5 bc 3.6 c 3.64 abc 36.3 cde 0.04 c Grass Fallow ---5.49 a 72.0 abc -Tillage fallow ---1.70 abc 60.2 abcd -†Means within the same column followed by the same letter do not differ statistically based on the Duncan’s Multiple Range test (P<0.05).

PAGE 63

47 Table 2.10. Shoot dry weight accumulation (DW) ; shoot N concentration (Nconc); shoot N accumulation (Naccum); weed dry weight accumulation (DW); weed N accumulation (Naccum); and cover crop weed index (CCWI) for summer cover crops during the 2005 growing season. Cover Crops Weeds Treatment DW Mg ha-1 Nconc g kg-1 Naccum kg N ha-1 DW Mg ha-1 Naccum kg N ha-1 CCWI Cowpea (I)‡ 2.75 cd† 16.7 c 46.0 c 0.42 e 2.1 e 64.0 ab Sunnhemp 12.61 a 19.0 c 242.6 a 0.55 e 2.3 e 80.0 a Alyceclover 1.00 cd 17.8 c 19.3 c 3.56 bc 20.5 de 0.5 c Hairy indigo 9.46 b 16.9 c 161.7 b 1.22 de 3.9 e 17.2 bc Velvet bean +Sudangrass 1.93 cd 24.7 a 48.3 c 1.99 cde 5.1 e 1.8 c Sudangrass + velvet bean 1.42 cd 6.3 d 9.0 c 1.99 cde 5.1 e 1.1 c Lablab¦ + Velvet bean 1.44 cd 16.7 c 24.6 c 1.32 de 3.0 e 13.5 bc Velvet bean + lablab 1.53 cd 24.2 ab 50.7 c 1.32 de 3.0 e 12.6 bc Lablab + sudangrass 2.18 cd 19.9 bc 42.1 c 1.54 de 32.2 cd 12.0 bc Sudangrass + lablab 1.62 cd 8.8 d 12.9 c 1.54 de 32.2 cd 6.5 bc Pigeon Pea 3.71 c 17.8 c 64.1 bc 2.68 bcd 47.1 abc 1.6 c Perennial P 0.24 d 21.0 ab 2.1 c 4.22 ab 57.5 ab 0.1 c Grass Fallow ---5.56 a 70.8 a -Tillage Fallow ---2.71 bcd 36.1 bcd -†Means within the same column followed by the same letter do not differ statistically based on the Duncan’s Multiple Range test (P<0.05).‡ ‘Iron Clay’ cowpea intercropped with sunnhemp

PAGE 64

48 Table 2.11. Rainfall measured at Plant Res earch and Education Unit (Citra) Florida Automated Weather station Net work (FAWN)1 during the 2002-2005 winter CC growing season. Year Month 2002/2003 2003/2004 2004/2005 Rainfall (mm) October 28 November 46 34 December 147 21 38 January 3 43 21 February 128 141 64 March 181 54 119 April 13 59 May 1 Total 472 334 335 1 Data obtained form the website http//f awn.ifas.ufl.edu on 12/10/2005. Blank spaces mean period of time when CC were not grown. Table 2.12. Shoot dry weight a ccumulation (DW); shoot N concentration (Nconc); shoot N accumulation (Naccum); maximum observed leaf area index (LAI max) with time of occurrence in brackets; weed dry weight accumulation (DW); weed N accumulation (Naccum); and cover crop weed index (CCWI) for winter cover crops during the 2002-2003 growing season. Cover Crops Weeds Treatment DW Mg ha-1 Nconc g kg-1 Naccum kgN ha-1 LAI max m2m-2 (wks) DW Mg ha-1 Naccum kgN ha-1 CCWI Crimson clover 1.9 ab† 23.4 ab 44.5 b 1.06 a (12) 1.71 bc 26.0 ab 1.6 b Red clover 0.8 ab 30.2 a 22.0 b 0.26 c (12) 2.53 ab 36.5 a 0.5 b Berseem clover 0.2 b 22.4 b 4.3 b 0.45 cb (12) 1.72 bc 25.8 ab 0.1 b Sweet clover 0.5 ab 26.4 ab 14.3 b 0.66 b (12) 1.32 bc 22.3 ab 0.5 b Lupin 3.1 a 28.2 ab 83.4 a 1.12 a (12) 1.63 bc 24.1 ab 1.7 b Cahaba vetch 1.3 ab 25.6 ab 34.2 b 0.44 cb (12) 1.31 bc 27.1 ab 1.1 b Winter rye 3.2 a 10.0 c 31.3 b 0.94 a (12) 0.30 c 5.9 b 7.8 a Grass Fallow ----3.70 a 38.6 a -†Means within the same column followed by the same letter do not differ statistically based on the Duncan’s Multiple Range test (P<0.05).

PAGE 65

49 Table 2.13. Shoot dry weight a ccumulation (DW); shoot N concentration (Nconc); shoot N accumulation (Naccum); weed dry weight accumulation (DW); weed N accumulation (Naccum); and cover crop weed index (CCWI) for winter cover crops during the 2003/2004 growing season. Cover Crops Weeds Treatment DW Mg ha-1 Nconc g kg-1 Naccum kg N ha-1 DW Mg ha-1 Naccum kg N ha-1 CCWI Hairy vetch 0.20 c† 38.7 a 7.8 bc 1.23 bc 19.4 bc 0.5 b Winter rye 0.89 cb 13.1 d 15.4 abc 1.23 bc 19.4 bc 2.5 b Crimson Clover 1.69 b 23.6 bc 38.0 a 1.10 bc 13.9 bc 1.8 b Radish 3.24 a 10.1 d 33.5 ab 0.48 c 7.0 c 15.0 a Black oat 1.25 cb 8.0 d 9.8 bc 1.82 abc 21.7 bc 0.8 b Lupin (I)‡ 0.37 cb 28.7 b 10.3 bc 1.82 abc 21.7 bc 0.2 b Subterranean Clover 0.28† cb 21.8 c 5.9 c 2.25 ab 32.3 b 0.2 b Lupin 0.20 c 24.1 bc 5.3 c 2.53 ab 30.8 b 0.1 b Grass Fallow ---3.84 a 52.1 a -Tillage fallow ---1.70 abc 20.1 bc -†Means within the same column followed by the same letter do not differ statistically based on the Duncan’s Multiple Range test (P<0.05). ‡ Lupin intercropped with black oat

PAGE 66

50 Table 2.14. Shoot dry weight a ccumulation (DW); shoot N concentration (Nconc); shoot N accumulation (Naccum); weed dry weight accumulation (DW); weed N accumulation (Naccum); and cover crop weed index (CCWI) for winter cover crops during the 2004/2005 growing season. Cover Crops Weeds Treatment DW Mg ha-1 Nconc g kg-1 Naccum kg N ha-1 DW Mg ha-1 Naccum kg N ha-1 CCWI Winter rye (WR) 5.96 b† 8.3 c 49.1 cd 0.28 b 4.8 b 25.2 cd Crimson clov. (CR) 5.02 bc 24.6 a 125.2 a 0.19 b 3.3 b 35.3 c Radish (R) 4.31bcd14.3 bc 62.9 cd 0.73 b 11.6 b 10.6 d Black oat (O) 3.56 cd 8.4 c 27.6 d 0.39 b 4.0 b 16.2 d WR + CR 5.34 bc 10.9 c 58.7 cd 0.07 b 0.7 b 76.2 b O + CR 2.76 de 19.2 ab 53.4 cd 0.07 b 1.1 b 39.4 c WR + CR + R 7.99 a 14.2 bc 110.0 ab 0.08 b ND 99.9 a BO + CR + R 4.95 cb 14.7 bc 73.72 cb 0.08 b ND 61.9 b Grass fallow ---3.58 a 66.5 a -Tillage fallow ---1.70 ab 24.7 b -†Means within the same column followed by the same letter do not differ statistically based on the Duncan’s Multiple Range test (P<0.05).

PAGE 67

51 0 1 2 3 4 5 6 7 014284256708498112 Days after plantingLeaf area index (m2m-2) CP VB SH AC HI Figure 2.1. Leaf area index values for summ er cover crops 2002 (CP= cowpea; VB= velvet bean; SH= sunnhemp; AC= Alyceclover; HI= hairy indigo). 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 0285684112140 Days after plantingLeaf area index (m2 m-2) C-CL R-CL B-CL S-CL LP C-V RYE Figure 2.2. Leaf area development for wi nter cover crops during 2002/2003 (CCL= Crimson clover; C-CL= Red clover; B-CL= Berseem clover; S-CL= Sweet clover; LP= Lupin; C-V= Cahaba Vetch; Rye= Winter rye).

PAGE 68

52 CHAPTER 3 ESTABLISHMENT OF PERENNIAL PE ANUT AND ITS EFFECTIVENESS IN SUPPRESSING WEEDS IN CITRUS ROW MIDDLES Introduction Perennial or rhizoma peanut ( Arachis glabrata, Benth.) is a rhizomatous warmseason perennial legume native to South Americ a with a wide area of adaptation ranging from 31 N to 35 S latitude (Prine et al., 1981). It was introduced into Florida in 1936, and it is used as a living mulch in associati on with other crops for soil conservation; as a pasture crop for grazing and/or hay production; and as an or namental to replace turf (French et al., 2001). Due to its high crude pr otein content (13 to 20%) combined with a digestibility between 55 to 67% and high pala tability, it produces a high quality forage similar to alfalfa (Saldivar et al., 1990; Te rril et al., 1996; French et al., 2001). Perennial peanut (PP) is ad apted to well-drained soils in the southern and Gulf Coast areas of USA (French and Prine, 1991). After initial establishment, it is drought tolerant, has excellent persistency under graz ing because of the rhizomatous habit, and it is not prone to insects, nematode or diseas e damage (Prine, 1981, Ba ltensperger et al., 1986, French et al., 2001). The most commonly grown cultivars are ‘Flo rigraze’ (Prine et al., 1981; Prine et al., 1986) and ‘Arbrock’ (Prine et al., 1986). ‘A rbrock’ is considered to be more drought tolerant, but it is less cold to lerant and may also decline if mowed frequently (Canudas et al., 1989; French and Prine, 1991). Under non-ir rigated conditions a nd/or dry conditions,

PAGE 69

53 attaining full ground cover with the ‘Florigraze ’ cultivar requires at least two to three years (Prine et al., 1986; Williams 1993; Johnson et al., 1994). Despite being pest, disease, and drought to lerance, its requirements for vegetative propagation combined with very slow initial growth hampers the use of PP as a cover crop (CC) and/or forage crop (Coleman, 1995; Rice et al., 1996; Williams et al., 1997). Perennial peanut typically produces very fe w or no sexual seeds. As a result, it is exclusively propagated vegetatively by rhizom es that provide carbon reserves for shoot growth during initial establishment and for regrowth in the spring (Saldivar, 1992a; Rice et al. 1996). The first criterion for determining optimal planting time of PP is the need for adequate soil moisture from rainfall and/or ir rigation during the initia l 2 to 3 months after planting. Secondly, any frost period shortly af ter planting should be avoided (Williams, 1994a; Williams et al., 1997). Due to changes in weed population d ynamics during the conversion from conventional to organic production, alternativ e weed management strategies may be required (Lanini et al., 1994; Bond and Gr undy, 2001). A national survey showed that improved weed management was the number on e research priority for organic growers (Sooby, 2003). Lack of effective weed management practices pertinent to organic citrus production systems thus may hamper successful transition from conventional to organic citrus production (Mesh, personal communication). Successful weed suppression via use of a perennial CC has been reported by Bradshaw and Lanini (1995); Aguilar (2001); an d Perez-Nieto et al. (2005) in coffee by using Arachis pintoi L. (non-rhizomal perennial peanut ). It also provided suitable weed control for heart-of-palm (Clement and DeFr ank, 1998) and coconut (M ullen et al., 1997)

PAGE 70

54 while perennial strawberry clover has been successfully used to suppress weeds in vineyards (King and Berry, 2005). Other perennial species such as bahiagrass ( Paspalum notatum Flugge) and bermudagrass ( Cynodon dactylon (L.) Pers) are either plante d and/or volunteer in grove row middles and may reduce growth of w eeds and prevent soil erosion (Rouse and Mullahey, 1997). However, these systems re quire frequent mowing and/or chemical control; they may harbor nematodes and citrus arthropod pe sts; and in many cases grasses compete with citrus for water and nutrients (Rouse and Mullahey, 1997). More recently use of PP as a groundcover for both vegetable crops (Roe et al., 1994) and citrus (Coleman, 1995) has also ga ined attention. Due to its low water and nutrient requirements, PP may provide an environmentally s ound and ecologically important component of sustaina ble citrus production in Flor ida (Mullahey et al., 1994). Its use can minimize soil erosion and nutriti on losses due to leaching and runoff, and therefore, it can also enhance water quality (Woodard et al., 2002). When mowed 2 to 3 times a year, it may provide 60 to 112 kg N ha-1 yr-1 to citrus trees. Alternatively, it can provide a high quality and valuable forage crop and additional source of income for citrus farmers during initial establishment of c itrus groves (Coleman, 1995). Some citrus growers may thus opt to use PP to also enhan ce the profitability of their citrus production systems. Although general production practices fo r PP are well-established, there is relatively little information on the effective use of PP for weed suppression in citrus (Mullahey et al., 1994). The majority of previous studies with PP have focused on development of optimal planting strategies of pure stands for hay production in North and

PAGE 71

55 Central Florida (Prine, et al. 1986; Williams et al., 1997; Freire et al., 2000). Moreover, current practices for establishment of PP were typically developed for conventional forage production (Williams, 1993; Rice et al., 1 996; Valencia et al., 1997; Ruiz et al., 2000; Williams et al., 2002). Although the use of PP as a CC in conventional citrus and vegetable systems in south Florida has been studied extensively (M ullahey et al., 1994; Roe et al., 1994; Coleman, 1995; and Rouse et al., 2001), there is no information on its use to suppress weeds in organic production systems. The overall objectives of this study were to 1) evaluate the effe ct of planting time (spring vs. summer) on initial PP establishm ent, growth, and dry matter production; 2) evaluate the effectiveness of over-seeding PP with crimson clover in the fall on system performance and weed control; and 3) c ontrast the performan ce and weed growth dynamics of PP-based systems with that of an annual CC system. The corresponding hypotheses were 1) in or ganic citrus systems, planting PP during the summer will increase the competitiveness of PP systems via enhanced initial growth compared to spring plantings; 2) ove rseeding PP with crimson clover in fall will help to increase the PP effectiveness in s uppressing weeds; and 3) weed suppression with annual CC is more effective than perennial peanut. Materials and Methods Set-up and Experimental Design A one hectare field was planted with ‘H amlin’, a processing orange cultivar [ Citrus sinensis (L.) Osb.] grafted on Swingle citrumelo ( C. paradisi Macf. x P. trifoliata (L.) Raf) during the summer of 2002 at the Plan t Science Research a nd Education Unit in Citra, Florida (29.68 N, 82.35 W). Tree spaci ng was 4.6 m in the row and 6.3 m between rows. The main emphasis of this study was to evaluate initial growth of perennial peanut

PAGE 72

56 ( Arachis glabrata Benth. cultivar Florigraze) and its effectiveness in suppressing weed growth in row middles compared to annual CC systems. Prevailing soil types at the experimental site were a Candler fine sand (Typic Quarzipsamments, hyperthermic, uncoated, 98% sand in the upper 15 cm) and a Tavares fine sand (Typic Quarzipsamments, hyperthe rmic, uncoated, 97% sand in the upper 15 cm). The initial soil pH ranged from 4.8 to 5.1 and soil organic matter content was 9.3 g kg-1. At the beginning of the experiment (fall of 2001), the soil was prepared by disking before applying lime (2.5 Mg ha-1) to the entire production bl ock. Trees were irrigated with microjet sprinklers with a 1.8-m spray diameter and a 180 spray pattern positioned 0.4 m NW of the trees. During the winter the irrigation spri nklers were placed inside 0.6 m high PVC pipes and also used for frost pr otection. Row middles we re non-irrigated in order to evaluate the adaptation of diffe rent species under typical citrus orchard conditions. Chicken manure was applied exclusively to a 1.8 m wide strip straddling the tree rows. Manure was applied during early spring at 4-11 Mg ha–1 and application rates were based on the estimated N mineralization rate, N content, and citrus tree age following IFAS (Institute of Food and Agricultural Sciences) N-recommendations for newly planted trees (Tucker et al ., 1995). To enhance growth of winter cover crops, a nonsynthetic (mined) K2SO4 fertilizer approved by the Orga nic Materials Review Institute (OMRI) was broadcast over the entire production block at a rate of 45 kg K2O ha-1 prior to planting of winter CC. Due to a build up of residual soil P, use of chicken manure was discontinued after 2004. Starting in 2005, an OMRI-approved 9-0-9 material derived from feather-meal and potassium sulfate (Nat ure Safe, Griffin Indus tries, Cold Spring,

PAGE 73

57 KY) was applied to tree rows following IF AS N-recommendations for newly planted trees (Tucker et al., 1995). A randomized complete block design was us ed with four replications and included the following treatments: 1) Annual cover crop (ACC) included sunnhemp ( Crotalaria juncea L.) and/or cowpea ( Vigna unguiculata L. Walp) planted in summer followed by crimson clover ( Trifolium incarnatum L.) and/or rye ( Secale cereale L.) or a triple mix of rye+crimson clover+radish ( Raphanus sativus cv. Rufus) planted during fall (nonperennial cover crop); 2) Perennial peanut (PP) planted in spring (PPsp); 3) Crimson clover was planted in the fall of 2001 and wa s followed by PP planted in the summer of 2002, with plots being over-seeded with crimson clover during the fall (PPsu-os);and 4) Fallow in spring (2002) and PP planted in summer 2002(PPsu). An outline of these treatments is presented in Table 3.1. Plot size was 18.9 m x 27.0 m and plots contained three row middles and two tree rows of five trees each. For annual CC tr eatment (ACC) and PP-plots overseeded with crimson clover in fall (PPsu-os), ground covers were plante d with a “zero-till” planter (Model Sukup 2100, Sukup Manufacturing Company, Shefield, IA) using appropriate agronomic practices (Table 3.2). Mowing da tes for both summer and winter CC are outlined in Table 3.3. Except for winter CC planted in 2003, wh en a zero-tillage system was used, previous CC and/or weeds were soil-incorporated with 2 to 3 passes of a rototiller using a tillage depth of 10 cm prior to planting a s ubsequent CC. ‘Florigraze’ perennial peanut was planted in March and July 2002 for spri ng and summer treatments, respectively, with a rhizome planting rate of 10 m3 ha-1, using a bermudagrass planter (Bermuda King

PAGE 74

58 model #79, Kingfisher, OK) with a row spacing of 0.5 m. All the plots were “cultipacked” after planting. All annual leguminous CC were inoculated before planting with the appropriate strain of rhizobium (Nitragin brand, Milwaukee, WI). Inoc ulants and untreated seeds were obtained from local seed companies si nce most varieties from certified organic commercial supply companies are typically poorly adapted to Florida conditions. Both PP and annual CC were not irrigated during the gr owing season except after planting if soil moisture was inadequate to ensure uniform germination. In this case, 25 mm of irrigation was applied uniformly to the entire field to mimic a typical rainfall. Insect pests were controlled when needed with allowed produc ts in organic production systems following the National Organic Program Standards (USDA, 2000). After initial es tablishment, plots for PP-based treatments were mowed at 4-wk intervals during spring, summer, and fall. After each sampling, PP-based systems were mowed. Data Collection, Measurements, and Analysis Representative sections of row middles were sampled at eight week intervals to evaluate above-ground dry weights (DW) of both PP and weeds using rectangular sampling frame with an inte rnal surface area of 0.5 m2 for one representative sample for each row-middle thus resulting in 3 samples pe r plot. Each sample area was selected in such way that it closely matched the ground c over of both PP and weeds of the entire row middle. Sampled areas included invariably a mix of weeds and PP. Weed biomass was not segregated into different weed species. Both PP and weeds were harvested at ground level. This approach was used since the main focus was on weed growth dynamics Weeds were separated from PP shoots but individual weed species were not segregated. Perennial peanut shoots were separated into

PAGE 75

59 stems and leaves and plant material was ove n-dried at 65 C for 72 hours until constant weight and dry weights were r ecorded. Perennial peanut leaf area was determined using a Licor leaf area meter (LI-3000, Li-cor, Lincoln, NE ) and used to calculate leaf area index (LAI) values. Plant tissue material was ground in a Wiley mill through a 1-mm screen, and a thoroughly mixed portion (ca. 4 g) was subsequently stor ed in scintillation vials. Ground tissue was digested using wet-acid Kj eldahl digestion (Ga llaher et al., 1975). After digestion, samples were diluted, filtered, and analyzed for total Kjeldahl N at the UF-IFAS Analytical Research Lab (Universit y of Florida, Gainesville, FL) using EPA method 351.2; (Jones and Case 1991). In order to quantify the e ffectiveness of PP to suppress weed growth, we also calculated cover crop weed indices (CCWI). This index expresses PP growth and biomass production relative to weed growth. An overview of the qualitative interpretation of this index is presented in Table 3.4. Since the weed and PP biomass sampling ap proach differed from standard forage sampling procedures, complementary forage productivity sampling was included during 2005. Overall forage productivity of PP was determined by taking representative row middle samples from the PPsp, PPsu-os, and PPsu treatments. Perennial peanut and weeds were cut at 6 cm of height above gro und level for a 10-m-long and 0.53 m wide (5.3 m2) sampling strip using a manual mower. Total fr esh weight of the total harvested biomass was recorded. A representative sub-samp le of perennial peanut-weeds of 0.5 m2 was harvested in a similar fashion. This subsampl e was segregated into perennial peanut and weed shoots, and plant materials were ove n dried at 65 C for 72 hours until constant weight before reco rding dry weight.

PAGE 76

60 Analysis of variance was perf ormed on all data using Proc Mixed of the Statistical Analysis Systems (SAS) software (SAS Inst. Inc., 2002). Since growth characteristics of PP were different from those of ACC, two se parate statistical analyses were conducted. During the first analysis, perennial peanut tr eatments were compared among each other, during a subsequent analysis the PPsu, which turned out to be the superior PP treatment, was contrasted with the ACC treatment. S hoot dry weight for a nnual cover crops (DWacc) and perennial peanut (DWpp); corresponding shoot N accumulation (Nacc and Npp) and leaf area index (LAIacc and LAIpp), number of PP shoots (S hoot#) along with weed dry weight (DWwd), N accumulation in weeds (Nwd), and cover crop weed index (CCWI) were evaluated. The abbreviations outlined he re will be used throughout the remainder of this chapter. If signi ficant interaction (P<0.05) occurred between year, treatment, season, and/or sampling time, specific effects were tested and shown separately. The LSMEANS procedure adjusted by Tukey test (P<0.05) was used to compare treatment means. Results Monthly rainfall during the PP growing s eason ranged from 901 mm in 2002 (drier year) to 1522 mm in 2004 (Table 3.5). Sin ce year by treatment interaction terms were significant (Appendix A-1), results are pr esented separately for each year. Perennial Peanut 2002 Rainfall was very low and unevenly dist ributed during the 2002 spring growing season (56 mm total), so obvious water stress o ccurred during the esta blishment of PP in spring 2002 (PPsp). Although the summer PP planting occurred 3 months after the spring planting, initial growth for summer plantings was better and overall crop growth was similar within 2 months after the summer planting (Table 3.6). Due to continuous growth, fall

PAGE 77

61 shoot#, LAIpp, DWpp, and Nacc-pp were greater for summer plantings, while PPsp treatments did not show a signifi cant increase in growth after initial establishment. Due to the rototilling at plantin g, weed biomass and Nacc-wd were lower for summer plantings. As a result, the CCWI were greatest for su mmer plantings. However, values decreased over time and during the fall overall weed growth parameters were similar for all treatments. Perennial Peanut 2003 Although rainfall was relatively high in 2003 in comparison with 2002, it was not evenly distributed (Table 3.5). Overall shoot#, LAI, DW, and Nacc-pp values were lowest during the spring and similar for summer a nd fall samplings (Table 3.7). Although no difference in shoot# and LAIPP occurred between summer plantings during 2002 and the spring of 2003, leaf area expansion and dry weight accumulation for PPsu-os during the summer and fall of 2003 were lower compared to the PPsu treatment. Toward the fall of 2003, the reductions in shoot# and LAIPP for the PPsp compared to PPsu were 89% and 91%, respectively. Dry matter (DM) accumulation for the PPsu-os and PPsp treatments were 32% and 88% less than the PPsu treatment, and the reductions in Nacc-pp for the PPsu-os and PPsp compared to the PPsu treatment were 31% and 90%, respectively. Due to the predominance of perennial gras ses and use of frequent mowing, overall DWwd and Nacc-wd values were relatively constant throughout the year. During the spring and summer of 2003, overall DWwd and Nacc-wd values were greatest for the PPsp treatment and not affected by overseeding. However, during the course of 2003 the summer-planted treatments started to diverge (a s indicated by differen ces in DW). By the fall of 2003, CCWI values for the PPsu-os and PPsp treatments were reduced by 43 and 93%, respectively compared to the PPsu treatment.

PAGE 78

62 Perennial Peanut 2004 During 2004, total rainfall was the highest of all four years (Table 3.5) and rainfall was also relatively evenly di stributed and no obvious water st ress occurred. In general, shoot# was relatively constant across seasons, while LAI and Nacc-PP had the lowest values in spring. The PPsu treatment performed best a nd had the highest shoot#, LAI, DW, and Nacc-PP, PPsu-os had intermediate values, while PPsp performed poorly (Table 3.8). The reductions in LAIPP, DWPP, and Nacc-PP, for the PPsu-os and PPsp compared to PPsu were 58 and 97%, respectively, by the fall of 2004. During the fall of 2004, overall DWwd and Nacc-wd values were greatest for the PPsp treatment and values were not affected by overseeding during the late fall/winter for PPsuos treatment. However during the course of summer and fall of 2004 the summer treatments started to diverge similar as was the case during the fall of 2003 (Fig. 3.2). Compared to PPsp treatment, PPsu and PPsu-os had 42 and 23% less weeds, respectively (Table 3.8). For PPsu, CCWI was greatest, while PPsp and PPsu-os had similar and relatively low CCWI values. Overall CCWI values for PPsu were highest during the summer and fall. By the fall of 2004, CCWI values for the PPsu-os and PPsp treatments were 68 and 98% lower compared to the PPsu treatment. Perennial Peanut 2005 Rainfall in 2005 was relatively high (1370 mm ) and evenly distributed (Table 3.5). Shoot# was similar across the season for PPsp and PPsu-os, while PPsu showed an increase in shoot# during the summer (Table 3.9). Overall LAIPP and DWPP values were lowest in spring for summer plantings while spring crops had overall low growth. The Nacc-PP value for PPsu was lowest in spring. Duri ng the summer and fall, PPsu had the best growth and highest DW and N content, PPsu-os had intermediate values, while PPsp again

PAGE 79

63 performed poorly across the entire seas on (Table 3.9). Toward the fall of 2005, the reductions in shoot# for the PPsu-os and PPsp compared to PPsu were 65 and 96%, respectively. Overall LAI, DW, and Nacc-PP values were reduced by about 62 and 97%, for the PPsu-os and PPsp, respectively. During the fall of 2005, overall DWwd and Nacc-wd values were greatest for the PPsp treatment in fall while weed growth with summer planted PP was not affected by overseeding. However, during the course of summer and fall of 2005 the results with summer treatments started to diverge again sim ilarly as was the case during the later part of 2004 (Fig. 3.2). For PPsu and PPsu-os treatments, weed dry weight was 39% and 24% lower than PPsp, respectively (Table 3.9). The cover crop weed index (CCWI) was greatest for PPsu while values were similar for PPsp and PPsu-os. Overall CCWI values did not differ across different seasons. By the fall of 2005, CCWI values for the PPsu-os and PPsp treatments were reduced by 64 and 98%, respectively compared to the PPsu treatment. Perennial Peanut Productivity (2005) Perennial peanut productivity was greatest for PPsu treatment, intermediate for PPsuos, whereas PPsp had the lowest productivity. Compared to the PPsu treatment, cumulative biomass values for PPsu-os and PPsp were reduced by 76 and 97%, respectively (Table 3.10). It should be noted that yields presen ted in Table 3.10 are based on solid PP stands. However, PP only covered 64% of the orch ard area and PP production on a total land area basis would be about 2.1, 0.5, and 0.06 Mg ha-1 for PPsu, PPsu-os, and PPsp, respectively. The PPsu treatment had the lowest weed biomass, followed by PPsu-os and PP sp, however weeds made up 98% of the DW content for the PPsp treatment. Corresponding

PAGE 80

64 values of contribution from PP to the tota l DW were 50, 13, and 2% while corresponding PP treatments accounted for 67, 23, and 3% of overall N accumulation (Table 3.11). These results followed a similar trend to t hose from the growth analyses (Tables 3.6 to 3.9). Annual Cover Crops (2002-2005) Annual CC outperformed PP treatments and had greater N and DW content and were also more effective in suppressing weeds across all the seasons a nd years, except in 2004 when ‘Cream-40’ cowpea was used as CC (Tables 3.6 to 3.9). Overall LAIacc, DWacc, and Nacc values were greater for summer CC (sunnhemp and cowpea) than for crimson clover, except in 2004. The values in LAIacc, DWCC, and Nacc had the following order: sunnhemp>crimson clover>‘Cream-40’ The three-way winter CC-mix (winter rye, crimson clover, and radish) generated 7.4 Mg ha-1 and outperformed mono-cropped systems. Overall Nacc values for winter CC systems ranged from 38-40 kg N ha-1 (crimson clover) to 121 kg N ha-1 (3-way mix). Corresponding values for summer CC systems were: 25 kg N ha-1 (cowpea) to 103-201 kg N ha-1 (sunnhemp). Winter CC triple mix, sunnhemp, crimson clover, ‘Cream-40’ cowpea, PPsu, and PPsu-os reduced overall weed growth by 96, 92, 67, 67, 41, and 24% respectively in comparison with PPsp. Overall DWwd and Nwd values were consiste ntly lower in annual CC when compared with PPsu, which was the best performing PP system. Overall DWwd and Nwd followed the order: ‘Cream-40’ cowpea > crimson clover> sunnhemp> winter CC triple mix. The CCWI score varied from outst anding weed control (C CWI>20) for triple mix and sunnhemp, to moderate weed contro l (1
PAGE 81

65 and N content, respectively. In comparis on with annual CC systems, weed DW and N content was 1067 and 772% greater for the PPsu systems. System Dynamics During 2005, total (PP+weed) biomass and N accumulation in PP-based systems as affected by season were as follows: summer>fall>spring, whereas total cumulative biomass and N were similar for all the treatments. In comparison with PPsu system, PPsu-os and PPsp systems had 76 and 97% lower DWPP and NacPP values, respectively. Overall reductions in weed growth were 15 and 49% for the PPsu-os and PPsu systems, respectively (Table 3.11). Summer CC system accumulated a similar amount of biomass and N in comparison with winter CC (tri ple mix) system. Percentages of contribution of CC to the total system DW and N content were greater for annual CC (98% and 99%, respectively). Corresponding values for PP-based systems were very low for PPsp (2%), low for PPsu-os (14-19%), and intermediate for PPsu (50-55%). When comp aring annual CC-based systems with the best PP-based system (PPsu), annual CC system had 114% greater DW production capacity and also accumulated 50% more N. Across years (2002-2005), summer CC produced 6.2 Mg ha-1 per season or 53 kg ha-1 d-1. Winter annual CC averaged 3.2 Mg ha-1 per season or 20 kg ha-1 d-1. These results contrasted with PP DW, which averaged 0.1, 0.7, and 2.3 Mg ha-1 for PPsp, PPsuos, and PPsu, respectively. When weeds were included in the balance, total biomass production was 6.8 Mg ha-1 or 57 kg ha-1 d-1 for warm-season annual CC. Cold-season annual CC and PP-based systems had average values of 3.5 and 7.5 Mg ha-1, or 23 and 31 kg ha-1 day-1, respectively. If we consider that almost all the PP sta nd was lost in PPsp, we could regard this treatment as a grass fallow or control, then the average across all th e years in percentage

PAGE 82

66 of weed suppression in PPsu and PPsu-os systems were 40 and 26%, respectively, compared to PPsp. However this level of weed suppression is low in comparison with the 93% weed suppression achieved in summer and winter annual CC. There was an inverse correlation between PP DW and weed DW (Fig 3.4) and between PP DW and N accumulation in weeds (Fig 3.5) in summer and fall. Discussion Low initial LAI and DW production for the current PP study may be related to slow initial growth (Prine, 1986). In additi on, lack of supplemental irrigation combined with low organic matter content, and poor wa ter retention capacity of the sandy soil may have further hampered initial gr owth and leaf area expansion. Because of the cost of plant material, spri gs are typically planted in strips with a row spacing of 0.5 m as a standard practice (French et al., 2001). Under our conditions, complete row closing only occurred during the third year of growth, probably due to pronounced weed competition during the first years. These findings agree with those reported by Williams (1993); however Ruiz et al (2000) reported row closing within one year. However, in that case, mechanical and/or chemical weed control along with supplemental fertilizers and irri gation applications were use d. The latter are not feasible for citrus row middles and synthetic herbic ides are not permitted in organic production. The relatively high shoot number during th e two first years af ter planting for the PPsu and PPsu-os systems was probably related to more favorable initial soil moisture conditions for summer plantings and wa rmer soil temperatures. Although the PPsu treatment had the highest LAI a nd DW across the years, it still did not perform as well as the annual cover crop system.

PAGE 83

67 The consistently poor performance of the PPsp system may have been related to erratic rainfall and prolonged dry period during its initial establishment. Similar results were reported by Williams et al. (1997) and Sa ldivar et al. (1992b), who concluded that adequate soil moisture during the initial 2 or 3 months after planting is the most critical factor for PP survival. It should be noted th at citrus rows middles are typically not irrigated, and that citrus soils typically also have lower water holding capacities compared to soils commonly used for pasture systems. The increased divergence between differe nt planting systems over time may be related to the PPsp treatment failing to develop a critic al density required to effectively compete with weeds. In the absence of overh ead irrigation, initial gr owth for the spring planted system was very poor and the few spr outs that grew often senesced within the first few weeks, similar to the findings reported by Williams (1993, 1994a). As a result, overall stands were very erra tic and in many cases perennial grasses and weeds prevented effective PPsp establishment, and maximum observe d DW for this treatment did not increase over time. Delaying planting until the onset of the summer rains improved initial establishment and increased initial growth and appeared to be a more viable strategy for citrus systems. Similar recommendations we re made for forage systems by Williams et al. (1997), whereas French and Prine (1991) prop osed that January to March was the best time for PP planting. For summer plantings, overall maximum DWpp, LAIpp, and Npp occurred during early fall, which is in agreement w ith findings by Ocumpaugh (1990). Although overseeding did not affect initia l growth of summer plantings, it appeared to hamper plant growth during subsequent year s. The zero till planter used for cover crops may have

PAGE 84

68 caused some damage to the rhizomes. Alte rnatively, it could be argued that planting crimson clover during the fall may have redu ced initial regrowth during early spring since crimson clover may persist up to April/ May. As a result, regrowth of PP may be slower and maximum productivity did not occur unt il fall. This in turn may have affected assimilate storage in rhizomes and subsequent regrowth as suggested by Saldivar et al. (1992a) and by Rice et al. (1996). The rationale for over-seeding PP in fall wa s to maintain a ground cover during the winter when PP is dormant and to also add additional N to the cropping system. However, since DWpp for the PPsu-os treatment was less than PPsu and weed pressure was not reduced, there is no justification for the extra cost associated with overseeding with crimson clover during the fall for this sy stem. Dunavin (1990, 1992), on the other hand, reported that over-seeding PP with rye or a rye grass mixture and crimson clover for the cool season provided a supe rior cropping system that had no negative impact on subsequent growth of PP. However, in thes e studies, the companion CC were broadcast instead of planted in rows while in our case, the knives of the zerotill planter may have caused some injury to the PP rhizomes. Add itional research may be needed to assess whether broadcasting without mechanical soil incorporation would enhance system performance without hampering PP growth. The continuous increase in maximum observed DWpp values for the PPsu treatment appears to be associated with higher shoot densities resulting from more favorable conditions for initial growth and rhizome fo rmation. Calculated cumulative productivity was low compared to reported poten tial PP hay yields of about 8 Mg ha-1 in conventional pure PP stands (Ocumpaugh, 1990; Johnson et al., 1994; Terril et al., 1996). Relatively

PAGE 85

69 low yields for our studies could be explaine d by the competition between PP and grassy weeds (Dunavin, 1992). Canudas et al. (1989) co ncluded that grass weeds hampered PP establishment and reduced PP yield by 50% due to competition. As a result, PP storage reservoirs may become depleted, resulting in reduced re-growth and poor performance in following years (Saldivar et al. 1992a, Willia ms, 1994b). Low inherent soil K levels may also have hampered PP-rhizobium activity as suggested by Slattery et al. (2001). The limited soil water storage capacity combined wi th the lack of supplemental irrigation may also have resulted in additional reductions in growth and productivity as discussed above. Dunavin (1992) and Valencia et al. (1999) re ported similar or lower yields for PP mixed with grasses. The shift in maximum N accumulation from summer to fall for the PPsu-os treatment could have been related to shading of pe rennial peanut by crimson clover. Although N concentration in PP tissues were with in expected range of 21 and 29 g N kg-1 (Saldivar et al., 1990; Terill et al., 1996; Venuto et al., 2000), the overall N content was much lower than the 192 and 162 kg N ha-1 calculated from the data provided by Ocumpaugh (1990) and Terril et al.(1996) for pure PP pure stands. This discrepancy may be related to weeds competing for light and nutrients in mixed st ands thereby reducing productivity of PP as suggested by Dunavin (1992). The low inherent initial soil fertility of the field site may also have resulted in poor pe rformance of rhizobium bacteria symbiont, as was proposed by OÂ’Hara (2001). The reduction in DWwd for the PPsu provides evidence that this treatment was relatively more successful in competing with perennial weeds. Th e overall effectiveness of PP in suppressing weeds was inversely related to DWpp (Fig 3.4) with correlation

PAGE 86

70 coefficients r=-0.78 and -0.72 for PPsu and PPsu-os, respectively, which may be related to resource competition between PP and weeds (D unavin, 1992; and Valencia et al., 1999). Lower DWwd during the summer 2002 was probably due to the effect of soil tillage on weed biomass. For all treatments DWwd was greatest during th e fall, except for 2005 when values peaked during the summer. Although the PPsu systems had a relatively high initial weed biomass compared to the PPsu-os system, within a year this trend was reversed. This underlies th e observation that over-seed ing did not enhance weed suppression. It may be possible that weed species that are effectively being suppressed by crimson clover did not prevail during the summ er, whereas weed species competing with PP during the summer may also be dormant dur ing the winter and thus were not greatly affected by over seeding. Alternatively, the rhizomes of perennial weedy grasses may have been more tolerant to potenti al injury of the zero-till planter. Reduction in weed suppression by annua l CC in the early spring 2004 may be related to the use of the zerotill system initiated with the planting of winter CC in 2003. The spike in weed growth observed in the fall of 2004 (Fig. 3.2) was related to the use of ‘Cream-40’ cowpea, a precocious variety with poor canopy persistence, that was relatively ineffective in suppressing weeds. Observed weed biomass values for PP-based systems during the first year were similar to those for conventional plots tr eated with pre-emergence herbicides only (Canudas et al., 1989). The PPsp system did not show appreciable decrease in weed growth over time, which was consistent with DWpp not increasing over time, suggesting that PP requires a critica l initial density to eff ectively suppress weeds.

PAGE 87

71 It should be noted that a key weed mana gement strategy for this trial was to withhold N-based nutrient sources from ro w middles and thereby provide PP with a competitive edge. This strategy appeared to work for the PPsu system, which showed a gradual decline in weed gr owth over time. Based on field observations, frequent mowing during the first year greatly reduced the inci dence of broadleaf weeds. However, it also promoted the growth of perennial grasses such as bahia and bermudagrass, similar to the results reported by Wright et al. (2003). In areas with higher i nherent soil fertility, bermudagrass and broadleaf weeds grew more vigorously and out-competed PP, and this effect appeared to be most prominent for spring plantings. The proportion of weeds (grasses) to PP on a DW basis was about 70-30 in PPsu in 2004 and 2005, which was similar to the values reported by Dunavin (19 92) in the fourth year of a mix of ‘Pensacola’ bahiagrassPP. The proportions we re 99-1 and 90-10 for PPsp and PPsu-os systems, respectively. The low PP components in these systems reflect the negative effect of the management pract ices associated with these systems on the overall competitiveness of PP as explained a bove. In 2005, for the produ ctivity trial, this ratio (weeds:PP) was lower (50:50) in PPsu which may be related to the cutting height used (Ocumpaugh, 1990). Similar to DWwd, Nwd was greatest for the PPsp treatment. The lower weed DW and N content for the PPsu treatment may be related to PP start attaining dominance in this system due to more favorable rainfall distribution during initial establishment as was proposed by Williams (1994a). High CCWI during the first sampling, was related to the short term effects of mechanical tillage on weeds. However over time, weeds began to dominate the system

PAGE 88

72 again because they were more competitive than PP. Repeated mowing favored prevalence of C-4 grass species including bahia and bermudagrass, whic h have higher growth rates compared to PP which is a C-3 plant (Pat erson et al., 1996, Newman et al., 2005). Although PPsu initially did not compete well with grass weeds, over time PPsu gradually became more competitive possibly due to the fact that it is more drought tolerant and can prevail in lo w nutrient environments (French et al., 2001). As a result, its CCWI thus gradually increased over time as PP gained a competitive edge over the grasses. On the other hand, since PPsp shoots often senesced before they reached full size, they could not contribute to restoring carbohydr ate reserves of the rhizomes during the fall and this was reflected by a gradual dec line in DW and CCWI over time. As a result, PP storage reservoirs may have become deplet ed (Saldivar et al., 1992a ) resulting in poor re-growth and performance in following years, giving weeds a competitive edge. It could be argued that a critical mass of initial growth is required for PP to invest in rhizome storage reserves expansion in order to devel op dominance over time, as discussed above. The results of the PP productivity trial were low (3.3 Mg ha-1) compared to the 8 Mg ha-1 reported by Ocumpaugh (1990), Johnson et al. (1994), and Te rril et al. (1996). However, these studies featured pure PP sta nds (compared to 50% PP in our study) and their plots were treated with chemical herbicides and fertilizers. Relatively low PP yields could be explained by the competition between PP and grass weeds, as suggested by Canudas et al. (1989) who reported that competition from grass weeds reduced PP establishment and PP yield by half. Dry weight of the PPsu treatment nearly matched the results obtained by Dunavin (1992) and Valencia et al. (1999) for PP mixed with grasses.

PAGE 89

73 The low productivity for the PPsu-os was probably due to over-seeding PP with crimson clover with the zerotill drill damaging rhizomes and/or depleting carbohydrates reserves which are critical for vigorous regrowth during a subsequent spring season (Saldivar et al., 1992a; Rice et al., 1996). Mo reover, it may be possible that crimson clover reduced light availabi lity to newly emerged PP sprouts during the spring. Another factor that could account for this relativel y low productivity is the low initial soil K content (see chapter 5 for more details about soil fertility dynamics), which could have hampered PP-rhizobium activity as proposed by Slattery et al. (2001). According to these results, contrary to those by Prine et al. (1981) and Williams (1993), planting PP in spring in the absence of supplemental irrigation, incr eases the risk of poor stands on poor soils and/or during dry springs as wa s the case for the 2002 planting. For annual CC, average sunnhemp bioma ss accumulation was similar to the 9-11 Mg ha-1 reported by Steinmaier and Ngoliya (2 001); Ramos et al. (2001); Perin et al. (2004) under tropical conditions, and superior to the results from Balkcom and Reeves (2005) and Jeranyama et al. ( 2000) for subtropical regions. The dry matter accumulation by ‘Cream-40’ was low compared to other cowpea varieties (Jerenyama et al., 2000; Muir, 2002). This was related to the short season and precocious reproductive cycle of this cultivar because after 6 weeks almost all the pods were formed and at 8 weeks all the foliage had senesced, allowing li ght to penetrate to the soil surface and decreasing weed suppression (Fig 3.2-3.3 and Table 3.8). The average DW accumulation of crimson clover (1.7 Mg ha-1) was lower than the 2.5-4.9 Mg ha-1 reported previously (Dyck et al ., 1995; Daniel et al., 1999; Odhiambo and Bomke, 2001; and Schomberg and Endale, 2004).

PAGE 90

74 Superior performance of mixed winter CC sy stems is in agreement with findings of Karpenstein-Machan and St uelpnagel (2000) for mixed rye-crimson clover and ryewinter pea ( Lathyrus hirsutus L.) systems in Germany and reports by Juskiw et al. (2000) for mixed small grain cereals systems in Cana da. The excellent performance of the mixed system may be related to the synergistic combination of comp lementary characteristics of the constituents of the mix (Kabir and Koide, 2002). Winter CC mixes also had a higher N accumulation which may be related to the combination of enhanced N retention of deep rooting and fast growing species (ry e and radish) with additional N-fixation by crimson clover (Justes et al., 1999; Vos a nd van der Putten, 2001; Kristensen and Thorup-Kristensen, 2004). The overall greatest N accumulation for s unnhemp is in agreement with reports by Balkcom and Reeves (2005) while the poor perf ormance of ‘Cream-40’ is related to its short growth cycle which makes it more suitable for short-term fallows. The relatively low (40 kg N ha-1N ) N content of crimson clover comp ared to other studies (Daniel et al., 1999; Odhiambo and Bomke, 2001), was proba bly due to the low pH and poor initial soil fertility hampering rhizobium co lonization (Slattery et al., 2001). The effectiveness of annual CC in suppressing weeds compared to PP may be related to both higher growth rates of annual CC, the use of mechanical tillage disrupting weed growth cycles, and alle lopathic action of winter ry e (Reberg-Horton et al., 2005). The superiority of weed suppression for annual CC, especially by the triple mix (rye+crimson clover+ radish) and sunnhemp wa s probably related to superior resource pre-emption by these annual CC sy stems (Craine et al., 2005).

PAGE 91

75 In terms of system dynamics, higher total biomass found in annual CC systems was related to the partial dormancy observed in C4 grasses and PP during the winter season. In contrast, winter CC are well adapted to low temperatures (Q i et al., 1999; Teasdale et al., 2004), and the higher overall N accumulation in annual CC systems was thus partly due to the contribution of wi nter CC. Therefore, use of winter CC may enhance nutrient retention and soil C sequestration, thereby outperforming PP-based systems while also providing superior weed suppression compared to PP-based systems. As discussed in Chapter 2, increased N accumulation in the system during the summer or winter season could be mineralized later and benefit either citrus trees or subsequent CC. However, strate gies should be developed to a void potential N leaching in sandy soils during the winter by including NO3 trap crops such as rye and/or radish as suggested by Justes et al. (1999), Vos and van der Putten (2001), and Kristensen and Thorup-Kristensen (2004). Weeds contributed significantly to DW and N accumulation in PP-based systems, underlining their important role of capturing N in the system and reducing environmental risks associated with pote ntial N leaching (Vos and va n der Putten, 2001; Woodward et al., 2002). However, potential benefits of sp ecific management practices of PP-based systems in reducing environmental impacts should be evaluated in more detail in future research. Although PPsu had low CCWI values and did not effectively suppress weeds during the first years, it is able to persist under adverse conditions (French and Prine, 1991). As was expected, over time PP gradually became more competitive, although its overall performance was still inferior to annual CC. However, in addition to weed suppression

PAGE 92

76 other potential advantages of PP include its potential for provide additional income as hay and its use within an integrated systems with animals such as sheep or goats (French et al., 2001). However, PP also has some cl ear drawbacks such as a relatively high establishment cost ($400-900 ha-1), very slow initial grow th (it takes 2-3 years to complete establishment), poor initial weed suppression and the requirement for frequent mowing (Coleman, 1995; Rice et al., 1996; W illiams et al., 1997). Since it is very important to ensure a clean and weed-free seed bed for PP planting (Williams, 1993; Williams et al., 1997) use of repeated tillage followed by CC crops such as sunnhemp and winter rye may be beneficial to redu ce weed populations in organic systems for a minimum of one year prior to planting perennial peanut. In terms of PP systems, we also propose the use of an integrated management system with PP being planted in early summer in row middles following repeated rototilling of a winter rye CC crop. Annual compact, self-reseeding CC can be planted near young trees, complemented with manure or natural fertilizer amendment applied to th e tree rows only. Once the trees are 5-6 years old and the perennial peanut is established, sheep can be introduced in the system to graze the row middles. Introducing sheep may reduce labor and energy requirements for maintaining a short canopy and we aim to test this system as soon as the trees reach a tree height of 33.6 m. In the mean time, we aim to establis h self-reseeding cover cr ops such as crimson clover (winter) and alyceclover or cowpea (sum mer) in the three rows to reduce labor requirements. Additional research is needed to assess the fate of N (immobilization and release) in the different N-soil pools deri ved from either CC or added manures or fertilizers in these sandy soils, supported by so me lysimeter and/or resin trap studies, in

PAGE 93

77 order to evaluate potential leaching and environmental ri sks from these management practices. Conclusions In general PP had slow establishment, a nd spring plantings were severely hampered by lack of adequate soil moisture, while comp etition with weeds a nd grasses resulted in erratic initial growth and poor stands. Under our production settings, planting PP after the onset of the rainy season result ed in better initial stands and more effective weed control. Initial weed suppression by PP was very poor to poor, which was due to its slow initial growth and high weed pressure. Overseedi ng PP with crimson clover in fall reduced PP vigor and its effectiveness in suppressing weeds. Compared to PP, annual CC provided much be tter weed control, especially when species were used that have allelopathic pr operties (rye) and/or retain adequately dense canopies for prolonged periods of time (rye and sunnhemp). For both PP and ACC, weed biomass was typically inversel y related to DW content of either PP or ACC probably due to competition for light, water, and nutrie nts. Presence of leguminous CC increased overall N accumulation, but weeds also contri buted to enhanced N retention and nutrient cycling.

PAGE 94

78 Table 3.1. Overview of experime ntal treatments during 2002-2005. Season Year Treatments Spring Summer Fall Annual CC§ Fallow1 Sunnhemp (SH) Crimson Clover (CrCl) PPsp † PP Fallow1 Fallow1 PPsu-os ‡ PP PP PP/ CrCl 2002 PPsu PP PP PP Annual CC Fallow SH + Cowpea* Crimson Clover PPsp Fallow1 Fallow1 Fallow1 PPsu-os PP PP PP/ CrCl 2003 PPsu PP PP PP Annual CC Fallow Cowpea Rye + Radish + CrCl PPsp Fallow1 Fallow1 Fallow1 PPsu-os PP PP PP/ CrCl 2004 PPsu PP PP PP Annual CC Fallow SH + Cowpea* Rye + Radish + CrCl+ hairy vetch PPsp Fallow1 Fallow1 Fallow1 PPsu-os PP PP PP/ CrCl 2005 PPsu PP PP PP 1 Spring plantings of pere nnial peanut (PP) were not successful in most of the plots and served as a partial control instead. These fallows were main tained by frequent mowing (every 3-4 weeks).§ Annual cover crops: crimson clover, sunnhemp and cowpea. † Perennial peanut (PP) planted in spring. ‡ PP planted in summer, the following y ears PP was over-seeded with crimson clover in fall. PP planted in summer. Sunnhemp planted in the center of row middles surrounded by cowpea.

PAGE 95

79 Table 3.2. Overview of seeding rates and ro w spacing for annual summer and winter cover crops used between 2002 and 2005 Cover crop Row spacing (m) Seeding rate kg ha-1 Inocula Variety Sunnhemp 0.36 40 Cowpea strain -Cowpea 0.36 56 Cowpea strain Iron clay Crimson clover 0.18 28 Trifolium strain Dixie Winter rye 0.36 112 -Abruzzi (Â’02Â’03&Â’05) Florida 401 (2004) Radish 0.18 22 -Rufus Table 3.3. Outline of planting and harvest da tes and duration for summer and winter cover crops. Summer Winter Year Planting Mowing Duration (days) Planting Mowing Duration (days) 2002 30 Jul 11 Oct 102 1 Dec 15 May 165 2003 10 Jun 16 Oct 127 28 Oct 31 March 154 2004 11 Jun 10 Oct 121 1 Nov 5 April 156 2005 21 Jun 25 Oct 125 5 Dec 9 May 155

PAGE 96

80 Table 3.4. Outline of cover crop weed index (CCWI) categories. CCWI value Cover crop Weed Weed control < 0.5 CC not competitive Weeds dominate Very poor (>70% weeds) 0.5-1 CC coexist Weeds coexist Poor 1-3 CC start prevailing Weeds pr evail in niches Moderate 3-5 CC prevail Weeds fail to dominate Adequate 5-15 CC predominate (70-90%) < 10-30% weeds Excellent >15 CC dominate completely <5% weeds Outstanding It is assumed if CCWI >15 then weed contro l is considered outstanding since weeds only cover account for less than 5% of the total biom ass. It should be noted that in the absence of weeds the CCWI will approach infinity, and the upper boundary is thus not defined. Table 3.5. Rainfall measured in the Plant Sc ience Research and E ducation Unit (Citra)1 during 2002-2005. Year Month 2002 2003 2004 2005 -------------------------------Ra infall (mm) ----------------------------January 61 4 44 23 February 26 129 143 65 March 35 182 55 121 April 21 14 25 148 May 0 33 70 163 June 135 238 142 197 July 105 130 272 102 August 153 148 160 196 September 122 101 420 102 October 15 114 117 121 November 67 46 35 58 December 160 22 39 75 Total 901 1162 1522 1370 1 Data obtained from the website http//fawn.ifas.ufl.edu on 1/25/2006

PAGE 97

81Table 3.6. Effect of planting time and over-seedi ng of perennial peanut (PP) on shoot number m-2 (Shoot#), leaf area index (LAIPP), shoot dry weight (DWPP), and N accumulation (Nacc-PP); weed dry weight (DWWD), N accumulation in weeds (Nacc-WD), and cover crop weed index (CCWI) in 2002. LAI, DW, Nacc, DWWD,and Nacc-WD for Annual Cover Crops (ACC) are included for purpose of comparison. Perennial Peanut Treatment Shoot# LAIPP DWPP Nacc-PP Spring Summer Fall Spring Summer Fall Spring Summer Fall Spring Summer Fall -----------# m-2 ------------------------m2 m-2 ---------------------Mg ha-1 -----------------------kg N ha-1 ----------PPsp † 12 a A 15 b A 0.03aA 0.03 b A 0.04 a A 0.05 b A 0.8 a A 1.5 b A PPsu-os ‡ 33 a B 70 a A 0.02 a B 0.21 a A 0.02 a B 0.29 a A 0.7 a B 6.6 a A PPsu § 29 a B 57 a A 0.02 a B 0.16 a A 0.02 a B 0.28 a A 0.8 a B 6.6 a A ACC ND 2.74 1.6 B ND 7.72 A ND 146 ACC vs PPsu ** ** ** Weeds DWwd Nacc-wd CCWI Spring Summer Fall Spring Summer Fall Spring Summer Fall ----------Mg ha-1 -----------------------kg N ha-1 ----------------------Mg Mg-1 ------------PPsp † 2.3 a A 3.6 a A 32.2 a A 48.1 a A 0.02 bA 0.03 a A PPsu-os ‡ 0.1 b B 3.7 a A 1.7 b B 39.0 a A 0.31 a A 0.11 a B PPsu § 0.1 b B 4.8 a A 1.2 b B 47.6 a A 0.35 a A 0.12 a B ACC ND 0.3 ND 5.3 ND 25.1 ACC vs PPsu ** ** ** † Perennial peanut (PP) planted in spring. ‡ PP planted in su mmer, the following years perennial peanut was over-seeded with cr imson clover in fall. § PP planted in summer. Means within the same column followed by the same lower case le tter and means within the same r ow followed by the same upper case letter do not differ statistically by the LSME ANS adjusted by Tukey test (P <0.05). **Contrast between ACC a nd PPsu, significant at P=0.01.

PAGE 98

82 Table 3.7. Effect of planting time and over-seedi ng of perennial peanut (PP) on shoot number m-2 (Shoot#), leaf area index (LAIPP), shoot dry weight (DWPP), and N accumulation (Nacc-PP); weed dry weight (DWWD), N accumulation in weeds (Nacc-WD), and cover crop weed index (CCWI) in 2003. LAI, DW, Nacc, DWWD,and Nacc-WD for Annual Cover Crops (ACC) are included for purpose of comparison. Perennial Peanut Shoot# LAIPP DWPP Nacc-PP Treatment Spring Summer Fall Spring Summer Fall Spring Summer Fall Spring Summer Fall -----------# m-2 ------------------------m2 m-2 ---------------------Mg ha-1 -----------------------kg N ha-1 ----------PPsp † 4 bA 12 b A 11 b A 0.00aA 0.03 c A 0.02 c A 0.00 a A 0.04 c A 0.04 c A 0.2 a A 1.1 c A 0.7 b A PPsu-os ‡ 35 a B 88 a A 82 a A 0.03 a B 0.13 b A 0.13 b A 0.02 a B 0.21 b A 0.23 b A 1.0 a B 5.0 b A 5.0 a A PPsu § 60 a A 107 a A 127 a A 0.05 a B 0.19 a A 0.19 a A 0.05 a B 0.31 a A 0.34 a A 2.0 a B 7.2 a A 7.2 a A ACC 0.95 B ND 3.12 A 1.8 B ND 9.6 A 40.3 B ND 201 A ACC vs PPsu ** ** ** ** ** ** Weeds DWwd Nacc-wd CCWI Spring Summer Fall Spring Summer Fall Spring Summer Fall ----------Mg ha-1 -----------------------kg N ha-1 ----------------------Mg Mg-1 ------------PPsp † 3.5 a A 3.4 a A 3.2 a A 44.8 a A 46.3 a A 46.5 a A 0.00 a B 0.01 c A 0.01 c A PPsu-os ‡ 2.5 b A 2.5 b A 2.9 a A 34.6 a A 35.0 a A 32.4 b A 0.01 a B 0.08 bA 0.08 b A PPsu § 2.8 b A 2.6 b A 2.5 b A 37.8 a A 32.6 a A 25.6 b A 0.02 a B 0.12 a A 0.14 a A ACC 0.6 A ND 0.4 A 11.9 A ND 6.8 A 2.87 B ND 23.7 A ACC vs PPsu ** ** ** ** ** ** † Perennial peanut (PP) planted in spring. ‡ PP planted in su mmer, the following years perennial peanut was over-seeded with cr imson clover in fall. § PP planted in summer. Means within the same column followed by the same lower case le tter and means within the same r ow followed by the same upper case letter do not differ statistically by the LSME ANS adjusted by Tukey test (P <0.05). **Contrast between ACC a nd PPsu, significant at P=0.01.

PAGE 99

83Table 3.8. Effect of planting time and over-seedi ng of perennial peanut (PP) on shoot number m-2 (Shoot#), leaf area index (LAIPP), shoot dry weight (DWPP), and N accumulation (Nacc-PP); weed dry weight (DWWD), N accumulation in weeds (Nacc-WD), and cover crop weed index (CCWI) in 2004. LAI, DW, Nacc, DWWD,and Nacc-WD for Annual Cover Crops (ACC) are included for purpose of comparison. Perennial Peanut Shoot# LAIPP DWPP Nacc-PP Treatment Spring Summer Fall Spring Summer Fall Spring Summer Fall Spring Summer Fall -----------# m-2 ------------------------m2 m-2 ---------------------Mg ha-1 -----------------------kg N ha-1 -----------PPsp † 5 bA 8 bA 16 bA 0.00aA 0.02 cA 0.02 cA 0.00 aA 0.02 cA 0.03 cA 0.0 aA 0.5 bA 0.7 cA PPsu-os ‡ 81abA 107 bA 153 bA 0.03 aB 0.18 bA 0.25 bA 0.04 aB 0.22 bB 0.42 bA 1.5 aA 5.6 bA 9.6 bA PPsu § 158aA 303 aA 346 aA 0.06 aB 0.51 aA 0.58 aA 0.08 aB 0.62 aB 1.00 aA 3.3 aB 16.8 aA 23.1 aA ACC 1.1 A ND 0.65 B 1.83 A ND 1.21 A 38.1 A ND 24.8 A ACC vs PPsu ** NS ** NS ** NS Weeds DWwd Nacc-wd CCWI Spring Summer Fall Spring Summer Fall Spring Summer Fall ----------Mg ha-1 -----------------------kg N ha-1 ----------------------Mg Mg-1 ------------PPsp † 1.8 aB 2.7 aB 4.3 aA 29.4 aB 29.6 aB 55.8 aA 0.00 aA 0.01 bA 0.01 bA PPsu-os ‡ 1.2 aA 2.1abA 3.3 abA 17.9 aA 26.5 aA 35.8 abA 0.04 aA 0.11 bA 0.13 bA PPsu § 1.1 aA 1.8 bA 2.5 bA 15.2 aA 22.2 aA 28.6 bA 0.08 aB 0.36 aA 0.41 aA ACC 0.9 B 0.1 B 1.4 A 14.5 B 1.2 B 38.3 A 1.92 A ND 0.86 A ACC vs PPsu NS ** NS ** NS ** † Perennial peanut (PP) planted in spring. ‡ PP planted in su mmer, the following years perennial peanut was over-seeded with cr imson clover in fall. § PP planted in summer. Means within the same column followed by the same lower case le tter and means within the same r ow followed by the same upper case letter do not differ statistically by the LSME ANS adjusted by Tukey test (P <0.05). **Contrast between ACC a nd PPsu, significant at P=0.01.

PAGE 100

84Table 3.9. Effect of planting time and over-seedi ng of perennial peanut (PP) on shoot number m-2 (Shoot#), leaf area index (LAIPP), shoot dry weight (DWPP), and N accumulation (Nacc-PP); weed dry weight (DWWD), N accumulation in weeds (Nacc-WD), and cover crop weed index (CCWI) in 2005. LAI, DW, Nacc, DWWD,and Nacc-WD for Annual Cover Crops (ACC) are included for purpose of comparison. Perennial Peanut Shoot# LAIPP DWPP Nacc-PP Treatment Spring Summer Fall Spring Summer Fall Spring Summer Fall Spring Summer Fall -----------# m-2 ------------------------m2 m-2 ---------------------Mg ha-1 -----------------------kg N ha-1 -----------PPsp † 2 b A 13 b A 20 b A 0.00aA 0.03 b A 0.02 c A 0.00 a A 0.04 c A 0.03 c A 0.1 a A 1.4 b A 0.9 b A PPsu-os ‡ 76 bA 150 b A 196 b A 0.03 a B 0.22 b A 0.24 b A 0.03 a C 0.29 b B 0.37 b A 1.0 a A 7.3 b A 8.0 b A PPsu § 225aB 479 a A 558 a A 0.09 a B 0.87 a A 0.60 a A 0.10 a B 1.23 a A 0.98 a A 4.1 a B 29.3 a A 21.7 aA ACC ND ND 1.91 7.41 A ND 6.42 A 121 A ND 103 A ACC vs PPsu ** ** ** ** ** Weeds DWwd Nacc-wd CCWI Spring Summer Fall Spring Summer Fall Spring Summer Fall ----------Mg ha-1 -----------------------kg N ha-1 ----------------------Mg Mg-1 ------------PPsp † 2.3 a B 3.8 a A 3.3 a A 29.8 a A 51.5 a A 42.4 a A 0.00 a A 0.01 bA 0.01 b A PPsu-os ‡ 1.0 b B 3.4 a A 2.5 b B 11.4 a B 45.9 abA 29.7 a A 0.02 a A 0.10 bA 0.18 b A PPsu § 1.2 b B 2.4 b A 2.0 b A 14.7 a A 29.6 b A 23.9 a A 0.10 a A 0.52 a A 0.50 a A ACC 0.1 A 0.0 A 0.2 A 1.1 A 0.2 A 2.3 A 75.2 A ND 32.1 A ACC vs PPsu ** ** ** ** ** ** ** ** † Perennial peanut (PP) planted in spring. ‡ PP planted in su mmer, the following years perennial peanut was over-seeded with cr imson clover in fall. § PP planted in summer. Means within the same column followed by the same lower case le tter and means within the same r ow followed by the same upper case letter do not differ statistically by the LSME ANS adjusted by Tukey test (P <0.05). **Contrast between ACC a nd PPsu, significant at P=0.01. .

PAGE 101

85Table 3.10. Effect of planting season date and over seeding on perennial peanut (P P), weeds, and system (PP+weed) dry weight, N accumulation in PP, weeds and in the system in 2005. Season Treatment Spring Summer Fall Total PP Weeds Total PP Weeds Total PP Weeds Total PP Weeds Total Dry weight ---------------------------------------------------------Mg ha-1 -----------------------------------------------------------------PPsp † 0.02 b 1.9 a 1.9 a 0.07 b 2.7 a 2.8 a 0.03 b 1.9 a 1.9 a 0.1 b 6.5 a 6.6 a PPsu-os ‡ 0.2 b 1.2 a 1.4 a 0.4 b 2.4 a 2.8 a 0.2 b 1.8 a 2.0 a 0.8 b 5.5 a 6.3 a PPsu§ 0.9 a 0.9 b 1.8 a 1.4 a 1.4 b 2.8 a 1.0 a 1.1 b 2.1 a 3.3 a 3.3 b 6.6 a N accumulation -----------------------------------------------------------kg ha-1 --------------------------------------------------------------PPsp † 0.5 b 24.6 a 25.0 a 1.6 b 36.6 a 38.2 a 0.7 b 24.5 a 25.2 a 3 b 85.7 a 89 a PPsu-os ‡ 5.0 b 13.7 b 18.7 a 10.0 b 32.4a 42.4 a 5.1 b 21.4 a 26.5 a 20 b 67.5 a 87 a PPsu § 22.6 a 11.0 b 33.6 a 35.7 a 17.3 b 53.0 a 24.6 a 13.2 b 37.8 a 83 a 41.4 b 124 a † Perennial peanut (PP) plante d in spring. ‡ PP planted in summer, the following years perennial peanut was over-seeded with cr imson clover in fall. § PP planted in summer. Means within the same column followed by the same letter do not differ statistically by the LSMEANS adjusted by Tukey test (P<0.05).

PAGE 102

86 Table 3.11. Total dry weight (DW) in the sy stem (CC+weeds), corresponding percentage of total dry weight in CC (% DW CC), total N accumulation (Total Nacc) in the system (CC and weeds) and corresponding percentage of N in CC (%N in CC) in 2005. Total DW % DW CC Total Nacc % N in CC Mg ha-1 kg N ha-1 PPsp † 6.6 2 89 3 PPsu-os ‡ 6.3 13 87 23 PPsu § 6.6 50 124 67 Annual CC (total) 14.1 98 227 99 Summer CC 6.6 97 105 98 Winter CC¦ 7.5 99 121 100 † Perennial peanut (PP) plan ted in spring. ‡ PP planted in summer, the following years perennial peanut was over-seede d with crimson clover in fall. § PP planted in summer. Sunnhemp+ ‘Iron clay cowpea’.¦ Triple mi x: Winter rye+crimson clover +radish

PAGE 103

87 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.43/026/029/0212/023/036/039/0312/033/046/049/0412/043/056/059/05DateDry matter perenial peanut (Mg ha-1) PPspr-p Ppsum-p+ W-CC PPsum-p Figure 3.1. Dry matter of perennial peanut (PP) over time (PPspr-p= PP planted in spring; PPsum-p+W-CC= PP planted in summer and overseeded with crimson clover in fall; PPsum-p= PP planted in summer).

PAGE 104

88 0 1 2 3 4 5 63/026/029/0212/023/036/039/0312/033/046/049/0412/043/056/059/05DateAbove-ground weed biomass (Mg ha-1) Annual CC PPspr-p Ppsum-p+ W-CC PPsum-p Figure 3.2. Dry matter of weeds across the years (Annual CC= Annual cover crop; PPsprp= Perennial peanut (PP) planted in spring; PPsum-p+W-CC= PP planted in summer and overseeded with crimson clover in fall; PPsum-p= PP planted in summer).

PAGE 105

89 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.73/026/029/0212/023/036/039/0312/033/046/049/0412/043/056/059/05DateCover-crop weed index PPspr-p Ppsum-p+ W-CC PPsum-p Figure 3.3. Cover Crop Weed Index (CCWI) for perennial peanut across the years (PPsprp= Perennial peanut (PP) planted in spring; PPsum-p+W-CC= PP planted in summer and overseeded with crimson clover in fall; PPsum-p= PP planted in summer).

PAGE 106

90 y = 3.8 -1.155x r2 = 0.608** y = 3.2 -1.397x r2 = 0.515**0 1 2 3 4 5 6 0.00.51.01.52.0 Perennial peanut DW (Mg ha-1)Weed DW (Mg ha-1) Sp DW W Su DW W Fall DW W Linear (Su DW W) Linear (Fall DW W) Figure 3.4. Regression between PP dry weight (DWpp) and weed dry weight during spring (SPDWW), summer (SUDWW), and fall (FALLDWW) for all PP treatments; ** coefficient of determination significant at P=0.01. y = 51.8 -18.478x r2 = 0.414* y = 40.4 -18.226x r2 = 0.471*0 10 20 30 40 50 60 70 80 900.00.51.01.52.0PP DW Mg ha-1N accum in weeds kg ha-1 Sp N-Wds Su N-Wds Fall N-Wds Linear (Su N-Wds) Linear (Fall N-Wds) Figure 3.5. Regression between PP dry wei ght (DW) and N accumulation in weeds during spring (SPN-W), summer (SUN-W), and fall (FALLN-W) for all PP treatments; coefficient of determination significant at P=0.05.

PAGE 107

91 CHAPTER 4 EFFECTS OF PERENNIAL PEANUT ( Arachis glabrata Benth ) AND COMMON BERMUDAGRASS ( Cynodon dactylon L.) ON NITROGEN AND WATER UPTAKE OF CITRUS Introduction Citrus is one of the most important cr ops in Florida and currently 302,929 hectares are under citrus production (Florida Agricultural Statistics Serv ice, 2006). In most citrus production systems, row middles are manage d to facilitate harvesting and grove maintenance operations while reducing eff ects of both weeds a nd soil erosion on tree production (Wright et al., 2003; Matheis and Victoria, 2005). Annual or perennial species planted in field areas that are not occupied by commercial crops are referred to as cover crops (Gliessman, 1998). C over crops (CC) can suppress weeds in the row middl es of conventional citrus groves (Coleman, 1995; Rouse and Mullahey, 1997; Matheis a nd Victoria, 2005) and in other agronomic and forest systems (Clement and DeFrank, 1998; Aguila r, 2001; King and Berry, 2005; Perez-Nieto et al., 2005). However, no detailed information is available on the effe cts of cover crops, on either nutrient or water uptak e of citrus (Yao et al., 2005). Bermudagrass ( Cynodon dactylon L. Pers.) is a C4 perennial grass (Paterson et al., 1996) and selected/ improved cultivars have been successfully used as a forage crop (Johnson et al., 2001). Sometimes repeated mowing of sods of grasses such as bermudagrass may allow their use as groundc over in row middles. Bermudagrass is a relatively short grass that spreads via aerial st olons and subterranean rhizomes that over time form dense fibrous mats. Its dispersa l can be increased by inversion tillage

PAGE 108

92 (Guglielmini and Satorre, 2004) and its gr owth may also be enhanced by mowing (Wright et al., 2003). As a result, it may volunt eer as a weed and can compete with citrus roots for water and nutrients (Wright et al., 2003) and become a persistent weed (Fernandez et al., 2002). Perennial peanut ( Arachis glabrata, Benth.) is a leguminous CC with low water and nutrient requirements. Since it does not re quire pesticides or N fertilizer, it may fit well within the model of sustainable production systems (Mullahey et al., 1994). Perennial peanut has been reported as a su itable groundcover for co conut (Mullen et al., 1997), coffee (Bradshaw and Lanini, 1995; A guilar, 2001; PerezNieto et al., 2005), pejiyabe (Clement and DeFrank, 1998), and citrus (Coleman, 1995; Mullahey et al., 1994). Its use may reduce labor costs associated with weed control, minimize soil erosion and nutrient losses due to leaching and runoff, and provide supplemental farm income via hay production revenues during early grove es tablishment. Florida citrus growers expressed interest in the use of perennia l peanut as a ground cover to suppress weed growth (Scholberg, unpublished). However, they also expressed conc erns about potential competition with (young) citrus trees for water and nutrients. Competition can be defined as the “negative interaction between organisms that place simultaneous demands on limited resources”, thereby potentially reducing growth of either one or both organisms (Booth et al., 2003; Craine et al ., 2005). Plants compete for resources such as light, water, and nutrients. Among all nutrients, crop N demand tends to be greatest, and plants thus compete strongly for this limited resource (Radosevich, 1995; Craine et al., 2005). Altern atively, cover crops can also act as catch

PAGE 109

93 crops and thus reduce nutrient leaching in pe rennial orchard systems (Wiedenfeld et al., 1999; Stork and Jerie, 2003). Competition for water, which is another cr itical crop production factor, has also been reported to occur in dr y years and in this case pres ence of cover crops may reduce tree growth (Pool et al., 1990) Due to a dense root system and high demand for N, bermudagrass has been reported as a very efficient competitor for both water and nutrients (Cohn et al., 1989; Fernandez et al ., 2002). However, Alsaadawi and Alrubeaa (1985), on the other hand, reported that berm udagrass seedling growth may be reduced through allelopathic interac tion with sour orange ( Citrus aurantium L ). Nitrogen is an essential co mponent of key molecules su ch as proteins, nucleic acids, and chlorophyll (Marschne r, 2003). Citrus requires ade quate N supply to ensure optimal growth, canopy density, yield, and fruit quality (Zekri and Obreza, 2003). Nitrogen deficiency is a widespread probl em in many citrus soils, and suboptimal N availability may greatly hamper tree gr owth and production (Maust and Williamson, 1994). Current N recommendations for c itrus trees are 0.17-0.34, 0.34-0.67, and 0.67 to 1.0 kg N tree-1 during the first, second, and third year after initial tree establishment, respectively (Tucker et al., 1995). Citrus N upt ake has been studied extensively for young citrus trees (Legaz and Primo-Millo, 1988; Maust and Williamson, 1994; Guazzelli et al., 1995; Syvertsen and Smith, 1996; Lea-Cox and Syvertsen, 1996, Scholberg et al., 2002). However, there is no information on the eff ect of cover crops on water and N uptake of young citrus trees. Stable N isotopes (15N) have been used to study N uptake (Mooney and Richardson, 1992; Lea-Cox and Syvertsen, 1996), N allocation and redistribution within

PAGE 110

94 the tree (Feigenbaum et al., 1987; Legaz and Primo-Millo, 1988; Maust and Williamson, 1994), N uptake efficiency (Weinbaum, 1978; Qu inones et al., 2003), and to evaluate N competition between species (Staples et al., 1999). However, 15N can interfere with subsequent 15N applications. The price of labeled fertilizer and subsequent sample analysis may also be cost-prohibitive, and only a limited number of laboratories are equipped for 15N determination (Scholberg et al ., 2001). Alternatively, nitrogen uptake can be determined via less expensive methods such as the Soil N Uptake Monitoring (SUM) system methodology developed by Schol berg et al. (2001). This system was shown to be suitable to monitor short-term N uptake dynamics by citrus (Scholberg et al., 2001; Scholberg et al., 2002). The objectives of this study were to 1) qua ntify the effect of perennial peanut and bermudagrass on citrus nitrogen and water uptake under controlled conditions; 2) assess the potential for competition for water and nitrogen between citrus and ground cover species such as perennial peanut and be rmudagrass; and 3) assess the effect of competition for water and nitrogen on ci trus tree growth. We also tested the corresponding hypotheses 1) plant species will differentially compete for nitrogen and water uptake; and 2) competition for uptake for water and nitrogen will be most articulated during periods of high demand. Materials and Methods Set-up and Experimental Design A greenhouse experiment was conducted at the Agronomy and Physiology research facility at the Univ ersity of Florida in Gainesvi lle (29.68 N, 82.35 W) between April 2004 and September 2005. A Soil-N Uptake Monitoring (SUM) system was used to determine N and water uptake dynamics of citrus (C), perennial peanut (PP), and

PAGE 111

95 common bermudagrass (BG) and to assess the potential for competition for water and N uptake between citrus and species used fo r groundcover. The SUM system consisted of four main components: i) uptak e columns, ii) nitrogen extract ion (leaching) system, iii) collection reservoirs, and iv) vacuum system (Scholberg et al., 2001). The uptake columns consisted of 0.45 m high PVC cylinders with an inner diameter of 0.30 m (Freedom Plastics Inc., Ft. Pierce, FL) that were filled with 25 kg of a Candler fine sand soil (Typic Quarzipsamment s, hyperthermic, uncoated, 98% sand in the upper 15 cm). The soil pH was 6.1 while the soil organic matter content and total (Kjehldal-extractabl e) soil N contents were 13.0 and 0.53 g kg-1, respectively. Individual columns were placed in a woode n case to facilitate their handling and these cases were placed on reinforced w ooden tables (Fig. 4.1). The bottom of each column was inserted into a 30.5-cm Schedule 40 PVC end cap (Freedom Plastics Inc. Ft. Pierce, FL) after applying waterproof caulki ng along the edge of the PVC pipe to prevent leaks. A circular 15-cm triple folded piece of nylon screen was positioned at the bottom of the columns to prevent soil from spilling out. A center hole was drilled in the end-cap that was threaded to fit a 12.7-mm o.d. adapter piece (Part No 62016, Thogus Products, Avon Lake, OH) connected to a flexible vinyl drainage tube with a 9.5-mm i.d. (Termoplastic Processes, Stirli ng, NJ). Inserted in the middle of this drainage tube was a PVC ball valve (Part No 22250, Thogus Products, Avon Lake, OH) to prevent NO3-N leakage from the columns prior to sampling. The drainage tube wa s connected to a high vacuum 24-L Nalgene heavy-duty vacuum carboy (Nalgene Nunc International, Rochester, NY) using a polyethylene quick -disconnect coupling (Part No 64027, Thogus

PAGE 112

96 Products, Avon Lake, OH). These carboys, in turn were connected to a partial (0.083 MPa) vacuum using a commercial 45-L compression tank as a vacuum chamber. Drip emitters (Chapin tube-weights, Chapin Watermatics, Inc. Watertown, NY) were inserted into a 12.7 mm i.d. irrigation lin e and placed on top of the column and used for irrigation and N-extraction. A pressu re regulator of 0.069 MPa (10 PSI, Senninger Irrigation, Inc. Orlando, FL) was used to ensu re even output from the irrigation/leaching system. Each column had four emitters, whic h were calibrated to give in a total water application rate of 0.5 L min-1. Treatments included were 1) Citrus (CIT ); 2) Bermudagrass (BG); 3) Citrus + Bermudagrass (CIT + BG); 4) Perennial peanut (PP); a nd 5) Citrus + Perennial peanut (CIT + PP). Nitrogen-amended bare soil (refer ence) columns were also included to determine crop N uptake. Additional bare soil (control) treatments that did not receive any supplemental fertilizer were used to es timate soil mineralization rates. Treatments were replicated 4 times and arranged in a randomized complete block design. Columns were either planted with 8month-old ‘Hamlins’ orange trees [ Citrus sinensis (L.) Osb.] grafted on Swingle citrumelo ( C. paradisi Macf. x Poncirus trjfoliata (L.) Raf.) and/or with groundcovers (PP or BG) on April 15, 2004, or alternatively left non-planted (reference and control) columns. Citrus trees were obtained from a comme rcial nursery. Trees were selected for uniformity and average initial tree fresh weight was 130 g tree-1. The potting mix was carefully rinsed from the root system prio r to replanting. Sod strips of ‘Florigraze’ perennial peanut ( Arachis glabrata Benth.) and common bermudagrass ( Cynodon dactylon L.) were excavated from the Plant Science Research and Education Unit

PAGE 113

97 (PSREU) in Citra, Florida. Soil and organic de bris were carefully rinsed prior to planting using a rate of 40 g fresh weight of rhizom es per column. Citrus tree heights and stem diameters were recorded every 6 months. Ground cover were clipped after leaching, clipping process divided the gr ound cover growth cycle in pr e-clipped (before leaching) and post-clipped. Bermudagrass was clipped every 4 weeks and perennial peanut at variable intervals as affect ed by its regrowth rate. Air temperature, soil temperatures, and so lar radiation were re corded at one hour intervals using a Watchdog datalogger (Spect rum Technologies, Plainfield, IL). Soil temperature degree-days (DD) for a specific uptake periods were calculated based on the approach outlined by Scholberg et al. (2002) assuming a base temperature (Tbase) of 10 oC as outlined in Equation 1. t=14d DD = t=0 (Tmean – Tbase) with Tbase = 10 C (Equation 1) Where Tmean is daily mean temperature and Tbase is the minimum temperature for development. Irrigation and ET Calculations A heavy duty waterproof bench scale (O haus CW-11, Champ Bench Scale, Ohaus Co. Florham Park, NJ) was used to weigh columns before and after each leaching event. Columns were irrigated at 3-7 d interval s using graduated cylinders ( 5 mL L-1) to maintain optimal and relatively constant so il moisture levels. The volumes added to specific columns were recorded and included in the water balance. Crop evapotranspiration was calculated from wei ght losses during uptake periods corrected for the amount of water added between 2 consecutive N extractions.

PAGE 114

98 Nitrogen Application Nitrogen was applied to a ll uptake columns as KNO3 from April 2004 to January 2005 and as Ca(NO3)2*4H2O (Fisher Scientific Inc., Ha mpton, NH) from February 2005 to September 2005 at biweekly interv als at a rate of 1000 mg N column-1. Since continuous use of potassium nitrate resulted in hyper-accumulation of K in leaf tissues and induced nutrient imbalances, only N uptake with calcium nitrate will be presented. A N-depleted modified Hoagland-Arnon so lution (Maust and Williamson, 1994) was prepared from stock solutions of laboratory-grade chemi cals (Fisher Scientific Inc., Hampton, NH) and applied at biweekly intervals at 7321mL column-1. Nitrogen Extraction Residual soil N was leached after an uptake period of 14 days was leached using three pore volumes of water combined w ith a partial vacuum (0.012 MP) since this facilitated rapid and complete N extraction (Scholberg et al., 2001). Upon completion of the leaching cycle (< 1 hr), the partial vacuum was main tained for an additional 30 minutes to remove excessive soil moisture and to bring to soil back to field capacity. Four control columns receiving no additional fertilizer were extracted biweekly to estimate N mineralization rates. After completion of the drying cycle (v acuum only), the leachate collection reservoirs (24 L) were weighed to determ ine volume gravimetrically using a heavy duty water proof bench balance (Ohaus CW-11, Champ Bench Scale, Ohaus Co., Florham Park, NJ). After weighing, containers were shaken and representative subsamples were collected, filtered (#42; Whatman, Maidst one, UK) and stored in labeled 20-mL scintillation vials which were stored up at –18 C until further analysis. Samples were analyzed using an air-segmented automate d spectrophotometer (Flow solution IV, OI

PAGE 115

99 Analytical, College Station, TX) coupled w ith a Cd-reduction appr oach (modified US EPA Method 353.2). Nitrogen Uptake Calculation Nitrogen uptake (Nupt) for a specific treatment (trtm) was determined by: Nupt= Nref x Vref – Ntrtm*Vtrtm; where N= N c oncentration in leachate, V= the leachate volume, and ref is soil columns without plan ts (reference columns). In order to validate the SUM system and to better assess the effect of competition on N uptake between cropping systems and plant components, we also used stable 15N isotopes. Two weeks before the last sampling, 15N 10% of atom enrichment was applied as Ca(15NO3)2 4H2O (Promichemical, LLC, El Sobrante, CA) together with regular Ca(NO3)2 *4H2O (Fisher Scientific Inc. Hampton, NH) amended w ith a modified N-depleted Hoagland-Arnon solution (Maust and Williamson, 1994). Nitrogen isotopic ratios of tissue material were determined with a N analyzer linked to a Tracer Mass Isotope Ratio Mass Spectrometer (Model DeltaPlus XL, Thermo Finnigan MAT, Bremen, Germany). Fertilizer-N uptake by the plant (Nabs) was calculated using the appro ach outlined by Cabrera and Kissel (1989): Nabs = Ntissue [(a-c)/(b-c)] (Equation 2) where, Ntissue is the overall N tissue concentration, a= atom % 15N abundance in plant; b= atom % 15N abundance in fertilizer; c= atom % 15N abundance in control plants (0.3663 % abundance). Fertilizer recovery (%Nrec) was then calculated as follows: %Nrec = (Nabs/Napp)*100 (Equation 3)

PAGE 116

100 where, Napp = Fertilizer applied and Nabs is the plant N uptake from the last Nfertilizer application event. Final Plant Sampling After the last leaching cycle (Septemb er 29, 2005) all the treatments were destructively harvested. Bermudagrass and pere nnial peanut total biomass was separated into shoots and roots, while citrus biomass wa s separated into stems, leaves and roots. Leaf area from different treatments was dete rmined using a Licor leaf area meter (LI3000, Li-cor; Lincoln, NE). Root systems were carefully excavated and washed above a 1-mm sieve to remove soil and organic debris In mixed systems, citrus roots were separated from ground cover roots and root s were placed into plastic bags and refrigerated (4 oC) until further processing. Roots were scanned using a Winrhizo scanner and software (Regent Instruments, Quebec C ity, Canada) and root scans were used to calculate root length. After sample processing, tissues were oven dried at 65 C for 72 hours until constant weight and dry wei ghts were recorded. Dried tissues were ground in a Wiley mill through a 1-mm screen, and a thoroughl y mixed 4-g portion of each grinding was subsequently stored in scintillation vials. Ground tissues were digested using a wet-acid Kjeldahl digestion (Gallaher et al., 1975). After digestion, samples were diluted, filtered, and analyzed for total Kjeldahl N at the UF-IF AS Analytical Research Lab (University of Florida, Gainesville, FL) using EP A method 351.2 (Jones and Case, 1991). Statistical Analysis It was hypothesized that the clipping of groundcovers affected N uptake, thus we segregated uptake of these species into “pre-clipped” and “post-clipped” cycles. Nitrogen uptake was thus clustered for different seasons (winter, spring, and summer) and different

PAGE 117

101 groundcover growth cycles (“pre-clipped” vs. “post-clipped”). Anal ysis of variance was performed on all data using St atistical Analysis Systems (SAS) software (SAS Inst. Inc., 2002). Dry weight, N concentration and N accumulation in different tissues were evaluated using analysis of variance proce dure utilizing Proc Mixe d of the Statistical Analysis Systems (SAS) software (SAS In st. Inc., 2002). If signi ficant interactions occurred between season, treatment, groundc over growing cycle, and/or sampling time, specific effects were tested and shown separately. Species were assumed to compete for N if the sum of the N uptake of monoculture systems was greater than that of the mixed systems, and we used statistical contrast between the sums of the monocrop systems vs the mixed system to assess if competition occurred between species for N and water uptake. A multiple regression model using the Proc Mixed procedure of (SAS Inst. Inc., 2002) was developed to capture the overall N uptake dynamics as affected by sampling time, season, growth cycles, ET, cumulative temperature (Degree-days = DD), solar radiation, and the quadratic effects of time, DD, solar radiation, the c ubic effect of time, the interaction treatment by season, treatment by cycle, treatment by DD, treatment by solar radiation, and treatment by time. Type 1 tests of fixed effect SAS output were used to select significant variable s in the regression model. Re sponse variables for the plant sampling parameters such as dry weight, r oot length, stem diameter, leaf area, N concentration, and N accumulation in different tissues were evaluate d through analysis of variance procedures utilizing Proc GLM of the Statistical Analysis Systems (SAS) software (SAS Inst. Inc., 2002). The LSMEANS procedure adjusted by Tukey test (pvalue=0.05) was used for mean separation.

PAGE 118

102 Results Environmental conditions includi ng mean, maximum and minimum soil temperatures, and solar radiation throughout th e experimental period are shown in Figs. 4.2 and 4.3. The Soil-N Uptake Monitoring (SUM) sy stem performed well and overall N recovery from non-planted reference column s was consistently high (97.2 5.1% ) and SUM-based N uptake values also closely matched 15N uptake values (Fig. 4.4). Calculated annual soil N mineralization rate of the upper 20-cm was on the order of 40 kg N ha-1 yr-1, which represented less than 2% of to tal N applied biweekly to each column. Nitrogen uptake for different treatments varied according to species, season and growth (clipping) cycle (Fig. 4.5). Across th e seasons, N uptake was highest for citrus + bermudagrass (C+BG) system followed by be rmudagrass (BG) and citrus+perennial peanut (C+PP), while uptake was lowest for PP and citrus (Table 4.1). Uptake was greatest during the summer and lowest duri ng the winter when N uptake was reduced by 81, 61, 55, 61, 58, and 61% for the citrus, BG, C+BG, PP, C+PP systems, respectively. For the C+BG system, competition between system components was significant during the summer period when overall N uptake rates were greatest. Water uptake was greatest for mixed cropping systems across all seasons and overall uptake was highest during the summer, intermediate for spring and lowest during the winter season. Citrus and bermudagrass and citrus and perennial peanut competed for water uptake during the spring a nd summer. Overall reduction in water uptake in winter in comparison with summer were 66, 56, 64, 93, and 70 % for the Citrus (C), BG, C +BG, PP, and C+PP, and BG systems, respectively.

PAGE 119

103 Clipping reduced N uptake for the BG and C+BG systems by 56 and 34%, respectively, while N uptake for PP was not si gnificantly affected by clipping (Table 4.2). Competition for N uptake between citrus and bermudagrass occurred during the regrowth cycle, while presence of PP did not appear to significantly reduce citrus N uptake. Clipping reduced water use for all systems and reductions in water use were greater (47-60%) in mono-cropped groundcovers compar ed to a 32-39% decrease in mixed systems. Significant competition for wate r occurred between citrus and full size groundcovers. The 15N technique demonstrated that bermuda grass N uptake was reduced by citrus in the C+BG system, while N uptake of citrus di d not significantly affect that of PP in the mixed system (C+PP). It thus appears that significant competition occurred in the C+BG system, but not in the C+PP (Tables 4.3 and 4.4) which verified the results obtained with the SUM system (Table 4.1). Nitrogen uptake efficiency (NUE), defined as crop-N uptake divided by the amount of N supp lied and determined either through 15N technique or SUM-based system, was greatest for the mixed systems and BG (Table 4.3). The N uptake regression model (r2 =0.845; p<0.05) was in cluded to assess N uptake as a function of season, growth cycl e, radiation, cropping system, and ET. These main effects besides interactions of radiat ion x treatment, and season x treatment were significant. The DD10, quadratic, and cubic effects of time and radiation were nonsignificant and therefore these components were not included in the model (Table 4.5). Based on the results of this regression model, more specific analyses were explored for different environmental parameters for sp ecific crop components. However, the only significant regression models found were fo r bermudagrass system for the growth

PAGE 120

104 cycle*radiation and DD10, terms both resulting in r2 values greater th an 0.76 (Figs. 4.6 and 4.7). Groundcover Biomass Production and N Accumulation Overall shoot biomass production was greatest for bermudagrass while its productivity was highest during the summer (T able 4.6). Perennial peanut produced 30 to 70% less biomass compared to bermudagrass but typically had 63 to 100% higher tissue N concentrations. As a result, overall N in groundcovers was similar across all systems and overall growth and N accumulation by groun dcovers was not affected by presence of citrus trees. Final Citrus and Groundcover Growth and N Accumulation End-of-season root length values for citr us were reduced by the presence of PP (Table 4.7). However, none of the other ci trus growth parameters were significantly affected by groundcovers. Total fibrous root length was greatest for bermudagrass followed by citrus, while values were lowest for perennial peanut (Tables 4.7 and 4.8). Bermudagrass DW was reduced by 23% in mixed system as compared to monocropped bermudagrass, but this reduction wa s not significant. Similarly PP DW was not significantly affected in the mixed sy stem. Bermudagrass accumulated more biomass at the end of the season but it had lower N tissue concentrations, therefore its total N content was not significantly different for BG and PP based systems (Table 4.8). The largest fraction of 15N was accumulated in shoots for all cropping systems except in citrus, where most of the 15N occurred in roots, followed by leaves and stems (Table 4.9).

PAGE 121

105 Discussion Estimated soil N mineralization rate was simila r to values reported in the literature for Florida sandy soils (Dou et al., 1997). Calc ulated N uptake followed a cyclic pattern (Fig 4.4) which was probably related to clipping cycle, se asonal temperatures changes (Fig 4.2) and citrus growth flushes (Bevi ngton and Castle, 1985). Si milar patterns were observed by Van Auken (1994) and Belesky and Fedders (1995). During the summer season, N uptake was the greatest, which was related to higher temperatures and evapotranspiration rates du ring this period (Table 4.1, Figs 4.2 and 4.4) which drive both active and passive N upt ake. A similar N uptake dependence on temperature was reported in citrus by Kato et al. (1982) and Sc holberg et al. (2002). Similarly, a higher N uptake in summer in comparison with winter season was also reported in citrus by Legaz and Primo-Mill o (1984) and Mooney and Richardson (1992). The higher N uptake for bermudagrassbased systems (BG and C+BG) in comparison with the other systems was pr obably related with the high demand of bermudagrass for water and N, as reported by Cohn et al. (1989) a nd Fernandez et al. (2002). Bermudagrass and citrus competed for N uptake during the summer season which may be due to a higher N demand of this C4 grass which tends to be most evident under high light and temperatures conditions prevai ling in summer (Allen, 1994; Paterson et al., 1996). The lack of competition for N uptake be tween PP and citrus th roughout the entire uptake period may be related to PP being a C3 species with a relatively low N demand in comparison with bermudagrass (Table 4.6). Moreover, since PP is also a leguminous species, for which N limitations enhances N fi xation (Miranda et al., 2003), this will render this system less prone to potential N competition.

PAGE 122

106 The increase in water use of mixed sy stems may be due to the higher water demands associated with complementar y canopy structures in comparison with monocrops. But in contrast with N uptake dynamics, PP and BG had similar demands for water and they competed with citrus for wa ter during spring and summer seasons. This is in agreement with reports by Wright et al (2003) who also observed water competition between BG and citrus and Firth et al. (2003) who worked with a perennial peanut and banana production system. Overall ET values were lower than those reported by Boman (1994) and Syvertsen and Smith (1996) in citrus and this may be due to differences in tree age, container size, a nd environmental conditions. Lower DW accumulation of perennial peanut compared to bermudagrass translated into greater water use efficiency for the gras s species which may be related to inherent superior metabolic trait to fix CO2 of C-4 crops in high temperature/radiation environments. Although PP was not clipped during the winter season, its poor growth and sparse canopy due to partial dormancy at low temp eratures could explain this lack of competition for water between PP and citr us during the winter period. Lack of competition between citrus and bermudagrass in winter was probably related to lower temperatures during this period, which re duced overall ET and the potential for competition between system components. Overall water use was greatest in summer for all the species due to high temperatures during this period (Fig. 4.2), whic h tends to increase the moisture gradient between the stomatal cavity l eaves and the bulk air (Taiz and Zieger, 2002). Low water

PAGE 123

107 use during the winter time ma y also be related to citrus and perennial peanut having limited growth during this time (winter dormancy). As consequence of clipping, N and water uptake were decreased after biomass removal for all the species. The reducti on in canopy volume reduced both crop water demand and assimilate supply for subsequent up take (Craine et al., 2005). Similar results of nutrient uptake declining after clipping have been obs erved by Van Auken (1994) and by Belesky and Fedders (1995) in bermudagrass. Nitrogen competition between citrus a nd bermudagrass was most obvious during the pre-clipped cycle possibly due to in creased water and nutrient demand of the bermudagrass. The reduction in N uptake of bermudagrass due to the presence of citrus may be related to shading and reduced a ssimilate supply which may hamper N uptake (Jiang et al., 2004; Tegg and Lane, 2004). Bermudagrass, be ing a C4 species, has a high light saturation point (Jiang et al., 2005; Paterson et al., 1996) and thus is susceptible to shading. Perennial peanut, being a C3, was probably less affected by partial shading (Newman et al., 2005). Absence of obvious competition for N uptake for perennial peanut-based systems may also be due to an increase in N fixati on, high N application rates, and limited overall crop N demand compared to BG-based systems. However, it could be argued that under lower fertility levels and poor nodulation c onditions, PP may compete with citrus under water and/or nutrient limited c onditions, as suggested by Crai ne et al. (2005) and Booth et al. (2003). Although water wa s replenished every 2-3 days system components still competed for the same resources with this effect being most pronounced when demand was greatest (during the pre-clipped cycle).

PAGE 124

108 According to the 15N study, citrus reduced bermuda grass N uptake and growth. It appeared that in the C+BG system, berm udagrass was most affected by competition, which may be a competition for light rather than for N as was discussed above. While citrus, which is a C3 with its canopy grow ing above the ground cover, would not be affected by potential competition for light. As a result, citrus growth parameters were not influenced significantly by bermudagrass, which could be expl ained by a potential allelopathic effect of certain citrus species on bermudagrass in confined environments as reported by Alsaadawi and Alrubeaa (1985). This result was unexpected and contradicted our field observati ons and the findings of other authors, who found se rious constraints due to the presence of bermudagrass as a weed in citrus groves (Wright et al., 2003) or in other crops (Cohn et al., 1989; Fernandez et al., 2002; Guglielmini and Sa torre, 2004). Since N may not have been severely limiting in our experimental set-up, additional resear ch is needed to elucidate the reciprocal effects of BG and citrus competition for pr oduction settings where N is more limiting. In contrast with the C+BG, there was no pronounced competition for N uptake in the C+PP system. Although citrus did not affect perennial peanut N uptake and its growth characteristics, PP may reduce water availabil ity in mixed systems when water is more limiting as was reported by Firth et al. (2003). Nitrogen use efficiency estimated through 15N study revealed highest NUE for BGbased systems (BG, C+BG), which is in ag reement with reports by Wiedenfeld et al. (1999) and Stork and Jerie (2003). These auth ors reported that intercropping citrus and another fruit species with grasses, increas ed overall N uptake while reducing N leaching in the systems, thereby providing sound enviro nmental alternatives to grove management.

PAGE 125

109 Since the 15N results closely correlated with uptake trends based on the SUM procedure (Fig. 4.7), it is c oncluded that the SUM system can be successfully used to evaluate short-term (1-2 wk) N uptake dynamics. The SUM system approach can thus provide a valuable and cost-effective alternative to the 15N technique (Scholberg, 2001). Contrary to what was exp ected, there was no direct co rrelation between water and N competition, except in summer between BG a nd citrus, which could be explained due to PP competing for water demands (Firth et al., 2003) but not for N since it is a leguminous crop. On the other hand, it turned out that evapotranspi ration rate was a good predictor for predicting crop N uptake, which is in agreement with other research finding (Syvertsen and Smith, 1996, Cerezo et al., 1999 ). In the current study, cropping system, season, ET, radiation, and growing cycle and their interactions accounted for a large fraction (r2 = 0.845) of the overall observed variation in N uptake. This may be related to large seasonal changes in temperature whic h may drive both active and passive uptake. Similar findings were obtained by Kato et al. (1982); Legaz an d Primo-Millo (1984); Scholberg et al. (2002). Alternatively, radia tion and sink capacity/demands may also have an appreciable effect on N uptake capacity as was reported by Legaz and Primo-Millo (1988), Mooney and Richardson (1992), Va n Auken and Bush (1990), Belesky, and Fedders (1995). Groundcover Biomass Production and N Accumulation Bermudgrass having the highest DW during both years and both cropping systems (C+BG, BG) was likely relate d to the high photosynthetic pr oduction capacity of this C4 species (Paterson et al., 1996), which also translated in to a relatively high N uptake capacity in order to synthesize proteins and other molecules to maintain high growth rates (Woodward et al., 2002). Bermudagrass N concen trations were simila r to those reported

PAGE 126

110 by Johnson et al. (2001) and Woodw ard et al. (2002). Overall N tissue concentrations of perennial peanut were greater than th ose for bermudagrass but observed PP N concentrations were similar to those re ported by Ocumpaugh (1990); Saldivar et al. (1990); Terill et al. (1996); and Venuto et al. (2000). Although PP is a leguminous N fixing crop, symbiotic N fixation under our experimental conditions was probably low due to the high supply rates of external Nsupply rates (Slattery et al., 2001; Singh and Usha, 2003). However, its leguminous metabolic pathway which allows PP to transport and store important amounts of N appeared to stay intact although the efficiency of N fixation was probably very low (OÂ’Hara, 2001). Nitrogen accumulation in both BG and PP was similar despite the differences in N c oncentration in tissues, due to the higher DW obtained in BG. Final Citrus and Groundcover Growth and N Accumulation Although there was a significant competition for N uptake between citrus trees and bermudagrass, overall growth of citrus was similar among different treatments at the end of the growing season. Since BG and citrus competition was only detected during the summer season, the competition may not have been prolonged and/or severe enough to result in significant reduction in overall ci trus growth. In contrast with annual crops, citrus trees have appreciable internal N reserves and relativ ely slow growth rates. As a result, remobilization of N from stem to othe r organs thus may allow trees to continue normal metabolism and citrus growth as suggested by Legaz and Primo-Millo (1988). Therefore, a prolonged period of limited N suppl y may be required before reductions in citrus growth due to competition for N upt ake with groundcovers can be discerned. Although there was a reduction in citrus root length in C+PP system, this did not translate into a significant reduction in tree weig ht, leaf area, or stem diameter. It may be

PAGE 127

111 possible that either root length was not limiting N uptake or that this limitation was overcome by increased uptake per unit root leng th or by increased N translocation from storage organs (Legaz and Primo-Millo, 1988) However, it may be possible that over time, the reduction in root length would result in reduced growth under water limited conditions. Since NUE for C+PP system wa s 84% (value derived from Table 4.3), compared to 97% for the C+BG system, N competition may also become more obvious for C+PP system over time (Craine et al., 2005). It thus may be pertinent to also look at competition for N uptake for systems with lower external N inputs under field conditions. However, one limitation of this research appr oach is that it probably would result in increased root competition within these micr ocosms, whereas under field conditions, root systems may explore different parts of the so il and “complement each other” rather than “compete with each other”. The high N allocation to the ci trus root system was in contrast with results for PP and BG systems in which the highest proportion of N was allocated to shoots. It may be possible that there was a ci trus root flush during the 15N application or that in the absence of an active shoot flush, an appreciable fract ion of the N is stored in roots prior to translocation to the shoots. These findings are similar to the one s reported by Legaz and Primo-Millo (1988) and Mooney and Richards on (1992), but different from the results from Feigenbaum et al. (1987) which may be related to differences in tree age and time of N application. Conclusions The SUM-based N uptake system appeared to work well and overall N recovery from reference columns was consistently hi gh while uptake rates matched those obtained via the 15N technique. Nitrogen uptake followed cy clic patterns as related to plant

PAGE 128

112 species, cropping system, growing cycle, ET, solar radiation and soil temperature. Nitrogen uptake was greatest for bermudagrassbased systems, while values were similar PP and citrus systems. According to the 15N study, N uptake for bermudagrass was reduced by citrus (possibly due to shading) in the citrus+BG system, and competition for N uptake did occur during the summer mont hs. Perennial peanut N uptake was not reduced by citrus in the citrus+PP system, and no obvious competition for N uptake occurred between these two species. Simila r conclusions were obtained using the SUMbased technique, implying that our soil N-uptake-system (SUM) is well-suited to monitor both short-term and long-term N uptake and co mpetition between citrus, PP, and BG over time. Nitrogen uptake was significantly reduced after clipping and increased during the pre-clipped cycle, which was probably re lated to increased sink capacities of groundcovers. Water uptake was greatest for the mixed systems and bermudagrass. Citrus and bermudagrass competed for water uptake during the spring and summer seasons. On a field scale, frequent mowing may thus re duce potential competition for water and N uptake between groundcovers and citrus. Howe ver, under our experimental conditions, the competition between citrus and BG fo r water and N did not significantly affect overall citrus tree growth characteristics, whereas presence of c itrus trees did reduce bermudagrass N concentration. Nitrogen uptake and growth ch aracteristics of PP were not affected by citrus in the citrus+PP system. On the other hand, PP appeared to compete with citrus only for water uptake and also reduced citrus root length, but this was not tr anslated into reduced tree growth. Additional research is required to eluc idate the long term effect of citrus root

PAGE 129

113 length reduction on N uptake and th e effect of BG on citrus gr owth characteristics under field conditions. Nitrogen use efficiency was greatest for mixed systems and bermudagrass. Groundcovers such as PP and BG, may func tion as a “catch crop” reducing potential N leaching. Some of the N accumulated in gr asses growing in row middles may be internally recycled and at a later point be re leased and re-utilized, which would facilitate more efficient N use. Alternatively, under N limiting conditions, presence of grasses near trees may hamper tree establishment in newl y planted groves. It is concluded that perennial peanut may be less prone to compete with citrus trees and also may be a more suitable cover crop for row middles cover since it will also generate extra farm income.

PAGE 130

114 Table 4.1. Effect of cropping system on citrus, bermudagrass, and perennial peanut (PP) N and water uptake for three different seasons during 2005. Season Cropping system Winter Spring Summer -----------------------mg N column-1 14d-1------------------------1) Citrus (C) 69 c B† 171 d B 364 c A 2) Bermudagrass (BG) 253 b B 654 b A 649 b A 3) Citrus + BG 394 a C 768 a B 882 a A 4) Perennial peanut (PP) 120 bc B 188 d B 304 c A 5) Citrus + PP 249 bc C 412 c B 592 b A Contrasts Syst1 + Syst2 vs. Syst3 NS NS Syst1 + Syst4 vs. Syst5 NS NS NS -----------------------L H20 column-1 14d-1---------------------1) Citrus 2.0 a C 4.4 b B 5.9 ab A 2) Bermudagrass (BG) 1.9 a C 3.6 b BC 4.3 b A 3) Citrus + BG 2.7 a C 6.0 a B 7.5 a A 4) Perennial peanut (PP) 0.3 ab C 3.4 b B 4.4 b A 5) Citrus + PP 2.0 a C 5.8 a B 6.6 a A Contrasts (Syst1 + Syst2)/2 vs. Syst3 NS ** ** (Syst1 + Syst4)/2 vs. Syst5 NS ** ** † Means within the same column or row, lo wer case letter within the same column and uppercase letters in the same row, followed by the same letter, do not differ statistically by the LSMEANS adjusted by Tukey test (P<0.05). NS, *, **, not significant, significan t at P= 0.05 and 0.01, respectively.

PAGE 131

115 Table 4.2. Effect of cropping system on citrus, bermudagrass, and perennial peanut (PP) N and water uptake for two different growth cycles during 2005. Cycle Cropping system Pre-clipped Post-clipped -----------------mg N column-114d-1-----------------1) Citrus (C) 202 b† 2) Bermudagrass (BG) 722 a A 316 b B 3) Citrus + BG 823 a A 540 a B 4) Perennial peanut (PP) 246 b A 162 c A 5) Citrus + PP 473 a A 362 b A Contrasts Syst1 + Syst2 vs. Syst3 NA Syst1 + Syst4 vs. Syst5 NS NA ---------------L H20column-114d-1--------------1) Citrus (C) 4.1 b 2) Bermudagrass (BG) 4.7 ab A 1.9 b B 3) Citrus + BG 6.7 a A 4.1 a B 4) Perennial peanut (PP) 3.6 b A 1.9 b B 5) Citrus + PP 5.7 a A 3.9 a B Contrasts (Syst1 + Syst2)/2 vs. Syst3 ** NA (Syst1 + Syst4)/2 vs. Syst5 ** NA † Means within the same column or row, lo wer case letter within the same column and uppercase letters in the same row, followed by the same letter, do not differ statistically by the LSMEANS adjusted by Tukey test (P<0.05). NS, *, **, not significant, significan t at P= 0.05 and 0.01, respectively.

PAGE 132

116 Table 4.3. Comparison of eff ect of cropping system on citrus, bermudagrass (BG), and perennial peanut (PP) N uptake at th e final harvest (end of the growing period) using 15N and SUM techniques. Cropping system 15N SUM mg N column-114d-1 mg N column-114d-1 1) Citrus 473 b† 439 b 2) Bermudagrass (BG) 818 a 772 a 3) Citrus + BG 965 a 978 a 4)Perennial peanut (PP) 562 b 548 b 5) Citrus + PP 840 a 748 a Contrast Syst1 + Syst2 vs. Syst3 * Syst1 + Syst4 vs. Syst5 NS NS † Means within the same column followed by th e same letter do not di ffer statistically by the LSMEANS adjusted by Tukey test (P<0.05). NS, *, **, not significant, significan t at P= 0.05 and 0.01, respectively. Table 4.4. Nitrogen accumulation by ci trus and ground covers based on 15N results. Cropping system Citrus Ground cover Total Bermudagrass -----------------mg N column-114d-1-----------------Monocrop Citrus + BG 473 A† 818 A 1291 A Mixed – Cit/BG 408 A 557 B 965 B Perennial peanut -----------------mg N column-114d-1-----------------Monocrop Citrus + PP 473 A 562 A 1035 A Mixed – Cit/PP 266 A 574 A 840 A † Means within the same column followed by th e same letter do not differ statistically by the LSMEANS adjusted by Tukey test (P<0.05).

PAGE 133

117 Table 4.5. Overview of parameters for N uptake regression model. Specific parameters Value or coefficient Intercept -1840 ** Time 53.5 *** Winter Season 618 *** Spring Season 384 *** Summer season 0 Radiation 3.2 *** ET 9.9 System 1 1250 ** System 2 895 ** System 3 -18.0 NS System 4 425 NS System 5 0 Clipped cycle (Syst 1) 0 Clipped cycle (Syst 2) 178 ** Clipped cycle (Syst 3) 315 ** Clipped cycle (Syst 4) 156 ** Clipped cycle (Syst 5) 0 Time*system1 -46.2 ** Time*system2 -9.6 NS Time*system3 19.4 NS Time*system4 -7.4 NS Time*system5 0 Radiation*system1 -3.0 Radiation*system2 -3.1 ** Radiation*system3 0.03 NS Radiation*system4 -2.3 Radiation*system5 0 Winter*system1 -686 ** Winter*system2 -269 NS Winter*system3 171 NS Winter*system4 -25.2 NS Winter*system5 0 Spring*system1 -543 *** Spring*system2 -278 Spring*system3 27.8 NS Spring*system4 -193 NS Spring*system5 0 Summer*systems 0 NS, *, **, *** not significant, significan t at P= 0.05 and 0.01, P<0.0001 respectively. System(S)1=Citrus; S2= C+PP; S3= C+BG; S4=PP; S5=BG

PAGE 134

118 Table 4.6. Effect of cropping system on berm udagrass and perennial peanut (PP) shoot dry weights (DW), nitrogen concentration (Nconc) and nitrogen accumulation (Naccum) for different growing seasons during 2004 and 2005. Dry Weight N concentration N accumulation Season Season Season Cropping Winter Spring Summer Fall Winter Spring Summer Fall Winter Spring Summer Fall System --------g column-1 ----------------------g N kg-1 -----------------------g N column-1 ------------2004 Bermuda NA NA 62.2aA† 18.8 aB NA NA 15.5 bA 20.3 abA NA NA 0.9aA 0.4aB Citrus + Bermuda NA NA 63.2 aA 19.7 aB NA NA 15.0 bA 18.4 abA NA NA 0.9aA 0.4aB PP NA NA 18.8 bA 13.5 aA NA NA 31.0 aA 34.8 aA NA NA 0.5aA 0.4aA Citrus + PP NA NA 19.9 bA 2.7 bB NA NA 30.0 aA 23.4 aA NA NA 0.5aA 0.1aA 2005 Bermuda 16.0aB 36.8 abA 51.3 aA NA 19.8 aA 18. 1 bA 19.4 bA NA 0.3 a B 0.7aAB 1.0aA NA Citrus + Bermuda 18.1aB 41.3 aA 43.3abA NA 20.2 aA 15.9 bA 18.7 bA NA 0.4 a B 0.7aAB 0.8aA NA PP ND 15.8 abA 30.3 bA NA ND 30.8 aA 31.7 aA NA ND 0.5 aA 0.8aA NA Citrus + PP ND 20.3 aA 39.9 bA NA ND 25.7 aA 28.9 aA NA ND 0.5 aA 1.1aA NA † Means within the same column or row, lower case letter within the same column and uppercase letters in the same row, followed b y the same letter, do not differ statistically by the LSMEANS adjusted by Tukey test (P<0.05). NA= non applicable; ND= non determined

PAGE 135

119Table 4.7. Effect of cropping system on citrus root dry weight (DW), root length, stem dry weight, diameter (Diam), leaf dry we ight, leaf area (LA), total dry weight, root nitrogen accumulation (Naccum), stem N accumulation, leaf N accumulation, and total N accumulation at the end of the growing season. Root Stem Leaf Total N accumulation Cropping system DW Length DW Diam DW Area DW Root Stem Leaf Total g tree-1 cm tree-1 g tree-1 mm g tree-1 cm2 col-1g tree-1 ------------g N column-1 ------------Citrus 124 a† 42318 a 109 a 11.4 a 75 a 5900 a 309 a 1.4 a 0.9 a 1.9 a 4.1 a Citrus + Bermuda 95 a 36105 a 86 a 9.5 a 40 a 3795 a 220 a 0.8 a 0.7 a 1.6 a 3.0 a Citrus + PP 86 a 21997 b 90 a 11.5 a 56 a 4940 a 231 a 1.0 a 0.7 a 1.2 a 2.8 a † Means within the same column followed by the same letter do no t differ statistically by the LSMEANS adjusted by Tukey test (P<0.05). Table 4.8. Effect of cropping system on bermudagrass (BG) and perenni al peanut (PP) root dry wei ght (DW), root length, shoot dr y weight (DW), leaf area (LA), root n itrogen concentration (Nconc), shoot n itrogen concentration, root nitrogen accumulation (Naccum), shoot N accumulation, and total N accumulation at the end of the growing season. Root Shoot Total Nconc Naccum Cropping system DW Length DW LA DW Roots Shoot s Roots Shoots Total g col-1 cm g col-1 cm2 col-1 g col-1 ----g N kg-1 --------------g N column-1 ----------BG 129 a† 94,646 a 90 a 1309 b 219 a 8.5 b 16.3 b 1.0 a 1.5 a 2.5 a Citrus + BG 92 ab 62,946 ab 76 ab 912 b 168 ab 8.0 b 14.7 c 0.7 a 1.1 b 1.8 a PP 42 bc 15,996 b 62 b 4401 a 104 b 17.2 a 20.2 a 0.7 a 1.3 ab 2.0 a Citrus + PP 51 bc 14,240 b 69 b 5135 a 120 b 16.3 a 19.4 a 0.8 a 1.3 ab 2.1 a † Means within the same column followed by the same letter do no t differ statistically by the LSMEANS adjusted by Tukey test (P<0.05).

PAGE 136

120 126 Table 4.9. Percentage of N distribution in different tissues for the diverse cropping systems. Cropping system Tissue N distribution (%) Citrus Roots 45 a Stems 14 b Leaves 41 a Citrus + Perennial Peanut Cit-roots 13 b Cit-stem 6 b Cit-leaves 13 b PP-roots 16 b PP-shoots 52 a Citrus + Bermudagrass Cit-roots 14 b Cit-stem 8 b Cit-leaves 20 b BG-roots 12 b BG-shoots 46 a Perennial peanut Roots 23 b Shoots 77 a Bermudagrass Roots 24 b Shoots 76 a † Means within the same column followed by the same letter do not differ statistically by the LSMEANS adjusted by Tukey test (P<0.05).

PAGE 137

121 Vacuum main line Water supply main line Quick disconnect couplings PVC End-cap Irrigation emitter PVC column Eye nut for inserting weighing bars Wooden support frame Cock valve Reinforced Wooden table Leachateline High vacuum bottle for collecting leachate Figure 4.1. Overview of soil-N uptake monitoring (SUM) system.

PAGE 138

122 0 5 10 15 20 25 30 35 40 45 1/21/20054/23/20057/23/200510/22/2005 DateTemperature C MinAverage Max average Soil Temp Figure 4.2. Minima, maxima, and soil averag e temperature during the experimental period. 0 5 10 15 20 25 01/20/0504/22/0507/23/0510/22/05 DateSolar Radiation (MJ m-2day-1) Figure 4.3. Solar radiation in the gree nhouse during the experimental period.

PAGE 139

123 y = 72+ 0.947x r2 = 0.873***0 200 400 600 800 1000 1200 020040060080010001200 SUM-based N uptake (mg N column-1)15N based N uptake (mg N column-1) Data 1:1 line Lin-Mod. Figure 4.4. Regression between SUM-based N uptake and 15N based N uptake (*** Coefficient of determination significant at level <0.001).

PAGE 140

124 0 200 400 600 800 1000 1200 1/21/20054/23/20057/23/200510/22/2005 DateN uptake (mg N column-1) CIT CIT+PP CIT+BERM PP BERM Figure 4.5. Nitrogen uptake dynamics for di fferent cropping systems across time.

PAGE 141

125 y = -586+12.9x-0.026x2r2 = 0.898** y = 693-9.4x+0.038x2 r2 = 0.851*0 200 400 600 800 1000 1200 050100150200250Cumulative uptake temperature (CdN)N uptake (mg N column-1) Pre-clipped Post-clipped Q-M (Pre-clip.) Q-M (Post-clip.) Figure 4.6. Nitrogen uptake as a function of cumulative uptake temperature during 14day pre-clipped vs. post-clipped upta ke period for bermudagrass mono-crop (*, ** coefficient of determination for quadratic regression model (QM) were significant at 0.05 and 0. 01 level respectively). y = -2853+33.2x-0.073x2r2 = 0.761* y = -738+7.4x0.011x2r2 = 0.757+0 200 400 600 800 1000 1200 0100200300400 Radiation (MJ m-2)N uptake (mg N column-1) Pre-clipped Post-clipped Q-M (Pre-clip.) Q-M (Post-clip.) Figure 4.7. Nitrogen uptake as a function of cumulative radiation during the 14-day preclipped vs. post-clipped uptake pe riod for bermudagrass mono-crop (*, + coefficient of determination for quadratic regression model (QM) were significant at 0.05 and 0.10 level, respectively).

PAGE 142

126 CHAPTER 5 EFFECTS OF ANNUAL AND PERENNIA L COVER CROPS ON SOIL AND CITRUS TREE CHARACTERISTICS, CITRUS TR EE ROW GROUND COVER, AND CITRUS YIELD AND QUALITY Introduction Cover crops (CC) can improve inherent soil fertility via increased carbon sequestration (Sainju et al., 2003). This in turn will increase soil organic matter (Conceicao et al., 2005; Fageri a et al., 2005; Ding et al., 200 6), soil water and nutrient holding capacities, and internal nutrient cyc ling. Roots of CC will reduce the risk of nutrient leaching during fallow periods and ma y also enhance nutrient recycling from deeper soil layers (Vos and van der Pu tten, 2001, Kristensen a nd Thorup-Kristensen, 2004). Leguminous CC may also add supplem ental N via symbiotic nitrogen fixation (Ramos et al., 2001, Perin et al., 2004). Howeve r, potential improvements in inherent soil fertility/quality will depend on the interactive effects of soil type, climatic conditions, and management on biomass accumulation and s ubsequent breakdown of organic matter (Cherr et al., 2006). The transition from conventional to orga nic production will impact agroecosystem management. Pronounced changes in soil chem ical, physical, and bi ological properties may occur during this conversion thereby also indirectly affecting insects, nematodes, diseases and weed dynamics (Ngouajio and McGiffen, 2002). Adequate weed control is one of the most challengi ng tasks during this conversi on due to changes in weed population dynamics, which will require implementation of alternative weed management strategies (Bond and Grundy, 2001; Lanini et al., 1994).

PAGE 143

127 Cover crops used during transition to organic production may influence soil chemical properties, including soil pH. Howe ver, the effect of CC on soil pH is not always consistent and depends on CC species initial soil conditions environment, and crop management. In general, CC tend to have little or no effect on soil pH (Waring and Gibson, 1994, Bloodworth and Johnson, 1995, Tian et al., 1999). On the other hand, Chaves et al. (1997) working in a degraded so il in Parana, Brazil, reported that leucaena ( Leucaena leucocephala L.), cowpea ( Vigna unguiculata L. Walp.), mucuna (Mucuna atropurpureum ( L.) DC), and crotalaria (Crotalaria ochroleuca L.) residues increased soil pH. Similarly, Espindola et al. (2005) reported that siratro ( Macroptilium atropurpureum L.), a CC species, reduced Al content and also increased soil pH, and base saturation, while perennial peanut and tropical kudzu ( Pueraria phaseoloides L.) did not affect soil pH. Use of blue lupin (Lupinus angustifolius L.) as a CC resulted in the greatest pH increase followed by radish (Raphanus sativus cv. Rufus) and black oat (Avena strigosa Schreb L.), while millet ( Panicum miliaceum L.) did not affect soil pH (Meda et al., 2001). Contrary to these results, Ikpe et al. (2003) working in an acidic Ultisol in Nigeria reported that Tephrosia candida (L.) used as a fallow CC reduced soil pH, and Ca, while soil Al content increased thereby exacerbati ng the soil acidification problem. Cover crops can enhance soil physical propert ies by shielding the soil surface from sunlight, wind, and the physical impact of raindrops, thereby reducing soil erosion and soil organic matter losses (Sainju et al., 2003) Cover crops also increase biological activity in the root zone thus enhancing the fo rmation of more stable soil aggregates and macropores while reducing soil compaction a nd soil bulk density. As a result, they

PAGE 144

128 improve soil structure (Kremer and Li, 2003) water infiltration, and root penetration (Justes et al., 1999). Cover crops can also increase soil C a nd N, two principal components regulating soil biological activity (Wagger et al., 1998; Abawi and Widmer, 2000). As a result, CC can increase the presence of beneficial organi sms that suppress pest organisms such as nematodes (Macchia et al., 2003; Wang et al., 2006). Moreover, they may also suppress soil pathogens via alleloch emicals (Bailey and Lazarovits, 2003) or impact weed population dynamics by increasing soil microbial diversity (Jordan et al., 2000, Kremer and Li, 2003). Weeds compete with crops for water, nutri ents, and light. They are potential hosts for pests and diseases and can also interfer e with soil tillage, irrigation, and harvest operations (Liebman and Davis, 2000). As a resu lt, they increase labor requirements and production costs. Cover crops (CC) may suppress weeds by either reducing resource availability (Ngouajio and Mennan, 2005), or inhibiting weed growth via allelopathy (Reberg-Horton et al., 2005; Fennimore a nd Jackson, 2003). Access to light, nutrients, water and soil as affected by crop management may impact weed persistence (Wright et al., 2003; Ngouajio and Mennan, 2005) and in consequence the composition of weed flora in citrus groves (Shrestha et al., 2002). Since many weed seeds require a high temp erature gradient to germinate, cover crops may suppress weeds by attenuating soil temperature gradient s (Aflakpul et al., 1998; Steinmaus et al., 2000; Le on et al., 2004). Moreover, th ey also can affect the red/far red ratio at the soil su rface (Gallagher and Cardina, 1998; Thomas et al., 2006), and/or change soil NO3-N status, which trigger weed seed germination (Alboresi et al.,

PAGE 145

129 2005; Bidwell et al., 2006; Perez-Fernandez et al., 2006). Cover crop residues may alter microbial soil ecology and increase soil microbi al diversity resulting in enhanced weed seed predation by soil microorganisms, and decr eased seed vigor (Gallagher et al., 1999; Ngouajio and McGiffen, 2002). Citrus growth characteristics including tr ee height, trunk diameter, fruit yield and fruit quality can be influenced by the potenti al competition for resources between citrus and weeds (Chen et al., 2004) or cover crops (Aiyelaagbe, 2001). However, there is no information available on the effect of CC on soil quality, tree row cover, and citrus growth, yield, and quality in organic citrus groves. The overall objectives of this study were to 1) determine changes in soil pH, C, and N content as affected by perennial and a nnual cover crop treatments; 2) quantify the effects of CC treatments on soil nematode popula tions; 3) evaluate changes in citrus tree row cover as affected by cover crop treatments; and 4) assess the effects of CC treatments on citrus height, diameter, leaf N, yield, and quality. The following hypotheses were being tested for newly planted organic citrus systems: 1) annual and perennial CC increase soil quality over time; 2) cover crops reduce pest nematode populations in organic citrus fields; 3) cover crop treatments in the row middles also affect weed growth in the tree rows; and 4) cover crops did not affect citrus growth characteristics (height, diameter, leaf N, yield, and quality) significantly. Materials and Methods Set-up and Experimental Design The overall treatments as well as site and experimental design are more fully described in Chapters 2 and 3. The study wa s conducted at the Plant Science Research and Education Unit in Citra, Florida (29.68 N, 82.35 W). Soil types at the experimental

PAGE 146

130 site were Candler and Tavares fine sand (~ 97-98% sand in the upper 15 cm). The initial soil pH ranged from 4.8 to 5.1 and soil carbon content was 5.4 g C kg-1. Two one-hectare blocks were planted with ‘Hamlin ’ and ‘Navel’ orange varieties [ Citrus sinensis (L.) Osb.] both grafted on Swingle citrumelo ( C. paradisi Macf. x P. trifoliata (L.) Raf.) during the summer of 2002 and spring 2003, respectively. Two separate experiments were conduc ted. The main emphasis of the first experiment (planted with ‘Hamlin’) was on pe rennial cover crop (per ennial peanut) while for the second block (planted with ‘Navel ’) the main focus was on annual cover crops (see Chapters 2 and 3 for more detailed in formation on the experimental designs and methodology used). A randomized complete bl ock design was used in both studies. The first study, hereafter referred to as the perennial CC study, included four different groundcover treatments in the 4.2-mwide row middles: 1) Annual cover crop (ACC), sunnhemp ( Crotalaria juncea L.) and/or cowpea ( Vigna unguiculata L. Walp) planted in summer, crimson clover ( Trifolium incarnatum L.) and/or rye ( Secale cereale L.) and in 2004-05 a triple mix of rye+crimson clover+radish ( Raphanus sativus cv. Rufus) planted in fall (non-perenniating cover crop); 2) Perennial pea nut (PP) planted in spring (PPsp); 3) Crimson clover planted in spring (2002) and PP planted in summer. The following years perennial peanut was over-s eeded with crimson clover in fall (PPsu-os); 4) Fallow in spring (2002) and PP planted in summer (PPsu). Treatments were replicated four times and a summary of these treatme nts was presented in Table 3.1. Each plot consisted of a total area of 18.9 m x 27.0 m and plots contai ned three row middles and two tree-rows of five trees each.

PAGE 147

131 The second study, hereafter referred to as the annual CC study, utilized annual CC treatments that were replicated four times. Cover crops were planted in June (summer CC) and October/November (winter CC). Cove r crop treatments were outlined in Tables 2.1 (2002 and 2003) and 2.2 (2004 and 2005). Each CC treatment plot consisted of a total area of 6.1 m x 27 m straddling a tree-row of five citrus trees. During 2002, only a grass fallow was used as the control while starti ng in 2003, a tillage fallow was also included as an experimental treatment. After initia l establishment, grass fallows and perennial peanut plots were mowed at 4-wk intervals throughout the spring, summer, and fall, after each sampling. Tillage fallows were tilled twice a year before CC planti ng. Similar methods were used in both experiments for crop maintenance and fo r collection of soil, agronomic, and pest data. In December 2001 four-composite soil samples were taken from both experiments (perennial and annual CC) and analyzed for pH, organic matter, macro and micronutrients and results were used to es timate lime and K applications (see appendix C). During the spring of 2002, lime (2.5 Mg ha-1) was applied to the entire area while chicken manure (2.5 Mg ha-1) was applied to the entire s econd production block (Study 2) and lightly incorporated into the soil via rototilling. During subsequent years, chicken manure was applied exclusively to a 1.8-m wi de strip straddling the tree rows. Manure was applied during early spring and applica tion rates were based on estimated manure N mineralization and N concentration followi ng N recommendations for newly-planted trees (Tucker et al., 1995). Due to the low soil K content (2.4 mg kg-1 Mehlich-1), an Organic Materials Review Institute (OMRI) approved non-synthetic (mined) K2SO4 fertilizer (SQM North America Corp., Atlanta, GA) was applied to the entire area at a

PAGE 148

132 rate of 45 kg K2O ha-1 prior to planting of winter CC to enhance their growth. Due to a build-up of residual soil P (Mehlich-1 P > 100 mg P kg-1), use of chicken manure was discontinued after 2004. Starting 200 5, an OMRI-approved 9-0-9 (N, P2O5, K2O) natural fertilizer derived from feather-meal a nd potassium sulfate (N ature Safe, Griffin Industries, Cold Spring, KY) was app lied to tree rows following IFAS Nrecommendations for newly-planted trees (T ucker et al., 1995). In order to try to overcome bermudagrass invasion in tree rows ‘ Iron Clay’ cowpea was planted in the tree rows in July 2003. Since ‘Iron Clay’ tended to be rather “viny” and grew into trees it was substituted with ‘Cream-40’, a more compact cowpea cultivar, which was planted in June 2004. However, ‘Cream-40’ was not vigorous enough and due to its short life cycle it also did not effectively suppr ess bermudagrass. So in 2005 we reverted back to ‘Iron Clay’, which was planted in July in the tree rows. Data Collection and Measurements Soil Soil samples of no less than 300 g were removed from each treatment for two different plot sections: row middles and tree rows on two sample dates per year: May (at the end of winter cover crops) and November (at the end of summer cover crops) in 2003, 2004, and 2005. Samples were collected from the top 15 cm from different areas within each plot and composite samples consisting of ten soil cores (2.5 cm diameter x 15 cm deep) were thoroughly mixed. Samples were air-dried, sifted th rough a 2-mm screen, mixed and stored in paper bags until furthe r analysis. Soil pH was determined for all treatments at all sample dates using the pr ocedure of the UF-IFAS Analytical Research Lab (University of Florida, Gainesville, FL). A mixture of 20 g soil was stirred with 40 g

PAGE 149

133 of pH-neutral DDI water and allowed to equilibrate for 20 minutes. The soil pH was measured using a pH probe (Model Accumet AR5 0, Fisher Scientific Inc. Hampton, NH). A representative subsample of 1 g soil was digested using an acid digestion (Gallaher et al., 1975), then samples were dilute d, filtered, and analyzed for total Kjeldahl N at the UF-IFAS Analytical Research Lab (Uni versity of Florida, Gainesville, FL) using EPA method 351.2 (Jones and Case, 1991). Soil organic matter was determined using dichromate oxidation method (Walkley and Black, 1934). Nematodes In 2004 and 2005, composite soil samples fr om each treatment were taken and stored in zip-lock ba gs, refrigerated at 4 oC and used for nematode analyses. After thorough mixing of the aggregate sample, a 100-cm3 subsample was removed for nematode extraction using a sieving a nd centrifugation procedure (Jenkins, 1964). Extracted nematodes were identified to genus and counted under an inverted microscope. Weed Growth Dynamics The effect of perennial CC treatment on species dominance in tree rows was determined by assessing weed canopy cover in tree rows. Percentage s of ground cover in the tree row were estimated visually using 7 numerical classes: 1 (0-1%), 2 (2-5%), 3 (625%), 4 (26-50%), 5 (51-75%), 6 (76-95%), and 7 (96-100%) using the Daubenmire canopy-coverage method (Daubenmire, 1959) as modified by Bailey and Poulton (1968). The midpoints of the classes were then used to register cover percentage and the percentage average of the two tree-rows per treatment was us ed for statistical analysis. Weed canopy cover within the tr ee row was estimated three times a year in April, July, and October (spring, summer a nd fall) for the ten most dominant weed species. During the summer and fall the coverage species were similar, and the six most common species

PAGE 150

134 in this group were reported together. In cont rast, only four ubiquitous weed species were observed during the spring season. Citrus Tree Performance Citrus tree height and diameter were determined for each tree in both studies two times a year, in May (spring) and November (fall). Trunk diameter was measured always at 20-cm above ground height using a digital caliper ( Model Number CD-6" CS Mitutoyo Corporation, Japan ). Fruit from the perennial CC study were harvested in January 2005 from the corresponding 10 trees per treatment in each block, and juice quality (degree Brix and total titratable acidity) was dete rmined according to the methods approved for Florida citrus quality test s (Wardowski et al., 1995). Diagnostic citrus leaves were samp led per treatment once a year (AugustSeptember) by picking 20 leaves per tree from four-to-six month old spring flush (Tucker et al., 1995). Leaves were placed in plastic bags, washed with DI water to remove soil and dust particles then oven-dried at 65 C for 72 hours until constant weight. Afterwards, tissues were ground in a Wiley mill through a 1-mm screen, and a thoroughly mixed portion (ca. 4 g) was subsequently st ored in scintillation vials. Ground tissue was digested using a wet-acid Kjeldahl digest ion (Gallaher et al., 1975). After digestion, samples were diluted, filtered, and analy zed for total Kjeldahl N at the UF-IFAS Analytical Research Lab (U niversity of Florida, Gain esville, FL) using EPA method 351.2 (Jones and Case, 1991). Data Analysis Since annual cover crops were rotated in the second production block (Study 2), soil pH, C, N, citrus tree diameter and hei ghts were averaged across annual cover crop treatments and contrasted with perennia l peanut, grass fallow, and tillage fallow

PAGE 151

135 treatments. Nematode data were log transformed prior to analysis: y = log10(x+1), where y = log-transformed data point and x= nema tode population count in a single sample. Analysis of variance was performed using Proc Mixed of the St atistical Analysis Systems (SAS) software (SAS Inst. Inc., 2002). Soil pH, C, N, C:N ratio, transformed nematode counts, percentage of coverage per species, citrus tree height and diameter, citrus leaf N, fruit yield and quality were the variables analyzed. If significant interaction occurred between year, season, location, and tr eatment, specific effects where tested and shown separately. The LSMEANS procedure ad justed by Tukey test (P<0.05) was used to compare either treatment or season means. Results Soil pH The CC treatments in study 1 (perennial CC study) did not affe ct soil pH (Table 5.1). Soil pH was similar between the tree row and row middles and in generally also not affected by season, except that during the spring of 2004 the annual CC treatment had a higher pH in both row middles and tree rows (Table 5.1). Similarly, for annual CC study (study 2), there was no effect of CC system, season and sampling location on soil pH. The only exception was a higher soil pH for the PP treatment during 2003 (Table 5.2). Soil C, N, and C:N ratio Similar to pH trends, neither perennial nor annual CC treatments affected soil C and N (Tables 5.3 and 5.4). A similar trend was observed for C:N ratio, except for the perennial CC study during 2005 (Table 5.3). Compared to the PPsu treatment, the C:N ratio in the row middles was 10, 20, and 29% lower for the PPsu-os, PPsp, and ACC treatments, respectively. Soil N was higher in tree rows in both annual and perennial CC studies, except for annual CC study in 2003. For the perennial CC study, soil N was 31,

PAGE 152

136 38, and 36% higher in tree rows compared to the row middles in 2003, 2004, and 2005, respectively. Corresponding increases for th e annual CC study were 39 and 33% during 2004 and 2005. For the perennial CC study, during 2004 the C:N ratio was significantly 26% higher for row middles in comparison with tree rows for PPsp treatment (Table 5.3). The overall carbon:nitrogen ratio was 19, 25, 40, and 48% higher in row middles in comparison with tree rows for ACC, PPsp, PPsu-os, and PPsu, respectively. In the annual CC study, the C:N ratio in row middles increas ed by 27% in 2004, the lower C:N ratio in the tree rows was probably related to th e chicken manure app lication (Table 5.4). Contrary to the expectations, C and N values in the tillage fallow treatment were similar to the values in the other treatments. Soil Nematodes Since location had no significant effect on nematode numbers, effects of seasons and years on nematode populations are pres ented for average response across position (Tables 5.5 and 5.6). For the perennial CC study, ring nematodes ( Criconemoides sp .) counts were about 4 times gr eater in spring in comparis on with fall 2004, while in 2005 spiral nematodes ( Helicotylenchus sp. ) increased by 338% between spring and fall (Table 5.5). For lesion nematodes ( Pratylenchus sp .), which can be economically important pests in citrus, were detected in fall in both years of this fiel d study. However, population levels of other nematodes were similar across different seasons. For the annual CC study, le sion, ring, and stubby root ( Paratrichodorus sp .) nematode counts were highest during at leas t one fall season (Table 5.6). On the other hand, pin nematode ( Paratylenchus sp .) populations followed a diffe rent trend; its initial population was about 12 times higher in spring th an in fall 2004. In general, we consider

PAGE 153

137 that for a perennial crop such as citrus these nematode numbers appear to be rather low. The citrus nematode ( Tylenchulus semipenetrans ), one of the most important pest nematodes in citrus, was not detected in our organic citrus grove (Tables 5.5 and 5.6). Tree Row Ground Cover in P erennial Cover Crop Study Overall there was no treatment effect on tree-row ground c overs (Table 5.7). Crabgrass ( Digitaria ciliaris (Retz) Koel) and co mmon bermudagrass ( Cynodon dactylon (L.) Pers.) were the most dominant weed sp ecies and they accounted for more than 75% of the ground coverage during the summer of 2003. Overall ground cover decreased during the fall of 2003 in comparison with summer season. During the spring of 2004, bermudagrass, bahiagrass ( Paspalum notatum Fluegge) and red sorrel ( Rumex acetosella L.) accounted for 46% of the tree row soil area coverage, whereas during the summer bermudagrass was the main weed species, whil e bermudagrass, crabgrass, and (planted) cowpea were the most abundant species during the fall. During the spring of 2005, the predominant species were bermudagrass and re d sorrel with these tw o species accounting for 75% of weed ground cover. The overall groundcover/weed community was dominated by bermudagrass and ‘Iron-Clay’ cowpea, the latter was planted during the summer 2005, and greatly reduced bermudagra ss population. This tre nd persisted during the fall of 2005 with ‘Iron Clay’ cowpea, a nd bermudagrass predominating in the tree row. Citrus Tree Growth Characteristics, Citrus Leaves N, and Fruit Quality For the perennial CC study, tree height, and diameter were significantly reduced by annual cover crop treatment in fall 2002 and spring 2003 (Table 5.8). However, after the fall of 2003, tree height and diameter were similar across all CC treatments. For the annual CC study, tree height and diameter were not affected by treatments (Table 5.9).

PAGE 154

138 Similarly, there was no treatment effect on c itrus leaf N concentration, with values ranging from 26.1 to 26.7 g N kg tissue-1 during the three years of study (2003-2005) for the perennial CC study. Nitrogen tissue values were 25.2 to 26.4 g N kg tissue-1 for the annual CC study during 2004 and 2005, resp ectively (data not shown). For perennial CC study, trees produced very few fruits during 2004 and fruits were removed to stimulate vegetative growth. Du ring 2005, fruits were harvested for the first time and yield, degree Brix (Brix), acidity, and the Brix/acid rati o were not affected significantly by treatm ents (Table 5.10). Discussion Soil pH In general, soil pH was not affected by CC treatments similar to the results reported by Waring and Gibson (1994), Bloodworth a nd Johnson, (1995), Tian et al. (1999). Overall soil pH was increased from 4.8-5.1 ( 2001) to about 6.4 in 2004, which may be related to the application of chicken manure. Similar results were reported by Hue and Licudine (1999), Eneji et al. (2002), Materechera and Mkhabe la (2002), and Mubarak et al. (2003). This may be due to the relativel y high Ca content of chicken manure and/or release of organic molecules/ anions with strong affinity for Al. Because of continued applications of chicken manure from 2002 to 2004, Mehlich-1-extractable-P values reached “very high” soil test cat egory (> 100 mg Mehlich-1 P kg soil-1) similar to the findings reported by Mubarak et al. (2003) a nd O’Hallorans et al. (1997), thus increasing the potential for surface and groundwater pol lution through runoff and leaching. Use of Nature Safe, a 9-0-9 (N, P2O5, K2O) formulation, prevented additional soil P build up but this product is 2-3 times more expensive. However, since it is a feather-meal based

PAGE 155

139 product its N release seems to be more gradua l which renders it a more suitable N source for young citrus trees. The gradual increase in soil pH in both tree rows and row middles was probably related to application and/or lateral disper sal of chicken manure associated with soil tillage and/or run off similar to the findi ngs reported by Sistani et al. (2003). The decrease in soil pH for sunnhemp plots in the perennial CC study in 2004 may be related to tillage-induced mineralization, resulting in increased nitrification rates and subsequent release of H+ ions (Myrold, 1999, Fortuna et al., 2003). However, this trend was not consistent across years. The increase of pH in fall for the annual CC study for the PP treatment, may be related to the effect of PP tissue residues on soil pH (Espindola et al., 2005). But again this effect was not consistent. Soil C, N, and C:N Ratio Soil C increased significantly from 5.4 g C kg-1 soil in 2001 (Appendix C) to about 8.3 g C kg-1 soil in 2005 (about 35% increase) in bot h perennial and annual CC studies, which may be related to addition of chicke n manure and cover crop residues. However, soil C was not affected by CC treatments, whic h was similar to the results obtained using crimson clover, hairy vetch, and wheat (Bl oodworth and Johnson, 1995; Mubarak et al., 2003). However, other authors reported that use of CC and/or organic amendments enhanced soil organic matter (Koutika et al., 2001; Wright et al., 2003; Agele et al., 2005; Tian et al., 2005; Conceicao, et al., 2005; Fa geria, et al., 2005; Ding et al., 2006). It may be possible that inherent soil variability and addition of high application rates of organic amendments to the tree row masked treatment ef fects. The overall initial increase in soil C could be attributed to a combination of factors. These may include soil protection and soil organic matter accumulation underneath th e sod of perennial crops and/or weeds,

PAGE 156

140 high carbon sequestration rates of annual CC such as sunnhemp, and the dispersal of chicken manure applied to the tree rows due to tillage and runoff (Mubarak et al., 2003) which could explain the relatively high soil C and N values in the tillage fallow. Typical C content in chicken manure ranges 300410 g C kg-1 (Obreza and OzoresHampton, 2000; Sistani, 2003) and total annual application rates were on the order of 10 Mg ha-1. Although its C:N ratio is low (9.1 to 12.4) and most of the N is readily available (Sistani, 2003), the bulk of th e material may consist of more recalcitrant co mponents that decompose more gradually over time and m odulate the build up of soil organic matter (Wagger et al., 1998). Overall total soil N was greater in tree rows in comparison with row middles because of the relatively high N content in ch icken manure and its direct application on the tree rows where it was applied. Nitrogen va lues in chicken manure range from 30 to 37 g N kg-1 (Obreza and Ozores-Hampton, 2000; Sistan i, 2003), which could explain the higher soil N values in tree rows compared to the row middles, simila r to the results of increased soil N due to chicken manur e reported by Eneji et al. (2002) The change in C:N ratio in row mi ddles in 2005 (after four years of experimentation) may be related to a gradua l build up of recalcitrant material in sod based (PP) systems. This was the first indi cation that systems started to diverge. The “sluggish” response to CC treatments may be related to the high annual SOM losses in coarse sandy soil combined with high soil temp eratures and high intensity rainfall events that tend to disperse/erode organic residues across the fi eld. As a result, continuous application of large amounts of soil organic amendments ma y be required to detect soil organic matter accumulation (Mubarak et al., 2003; Cookson et al., 2005). The higher

PAGE 157

141 C:N ratio obtained in PPsu was related probably to the higher biomass accumulation in this treatment and formation of a rhizomat ous and more recalcitrant root mat under a perennial system (See chapter 3). Similar resu lts were reported by Sainju et al. (2003) and Saldivar et al. (1992a). Nematode Counts The increase in lesion nematode during the fall in annual CC study may be related to a higher citrus fibrous root starch concentration in fall which stimulates nematode proliferation, as was suggested by Duncan et al. (1998). However, both citrus blocks showed very low levels of lesion nematode (0-5 nematodes per 100 cm3 soil) and numbers remained well below the critical hi gh levels of infestation (OÂ’Bannon et al., 1972). It appeared that neither the annual nor perennial CC assayed resulted in a proliferation of lesion nematode populations or any other pest nematode of citrus. Other nematode populations were also low and levels were so low to negatively affect citrus growth (Ferguson, 1984; Duncan and C ohn, 1990). Although some CC such as alyceclover, lablab, and mungbean may functi on as host to certain nematodes (Reddy et al., 1986b), the inherent divers ity in microfauna and fl ora of soils under organic management (Mader et al., 2002, Berkelmans et al., 2003) may have reduced nematode populations. Similar findings were docume nted by Wang et al. (2006) who reported reduction in phytophagus nemat odes using organic inputs. Tree Row Ground Cover in P erennial Cover Crop Study The lack of effect of CC treatment in row middles on tree-row weed ground cover, was hardly surprising since both regions fo rmed two distinct management zones. Moreover, repeated tillage and planting cowp ea in the tree rows may also have masked potential more subtle effects. Bermudagrass and crabgrass were the initial predominant

PAGE 158

142 weeds in tree rows, but crabgrass was reduced over time possibly due to the competition with bermudagrass and cowpea (Craine et al., 2005). Dominance of bermudagrass may be related to soil inversion with rototilli ng as suggested by Guglielmini and Satorre, (2004), since this results in a dispersal of aerial stolons and subterranean rhizomes. Based on field observations, up to 6 or 7 rototi lling passes within a 3-month period were required in order to exhaust bermudagrass re serves and destroy pr opagule structures. The increase in weed cover percentage (55%) of bermudagrass during the spring of 2005 was related to the use of ‘Cream-40’ cowp ea in summer 2004. This compact and earlymaturing cowpea variety lost its canopy domi nance quickly (in a bout 6 weeks). As a result, bermudagrass gained a competitive edge and increased its bi omass, reserves, and reproductive structures and its population de nsity peaked during the spring 2005. When ‘Iron Clay’ cowpea was planted, bermudagrass population was reduced to near 35% of ground coverage, showing the effectiveness of this cowpea cultivar in suppressing weeds in the trees rows, similar to the findings reported by Budhar and Tamilselvan (2003), Ram et al. (2003), and Tamado and Milberg (2004). Citrus Tree Performance Since sunnhemp was planted close to the tree row during 2002, its height (> 2 m) combined with the small size of the tree (<0.75 m) resulted in pa rtial shading of the citrus trees during the summer of 2002, thereby hamperi ng initial citrus development. In studies by Yamanishi and Hasegawa (1995), Dhyani a nd Tripathi (1999) a nd Aiyelaagbe (2001) intercropping sweet orange with species th at shaded citrus trees also significantly decreased tree height and diameter. But over time, citrus trees under annual CC treatment recovered to normal height and diameter. Th is resulted from planting sunnhemp further away from the tree trunk (> 2 m) and the trees also gradually becoming taller. Similar

PAGE 159

143 findings about effect of shade on citrus we re reported by Cohen et al. (2005). During 2005, there were high populations of locusts in the field and it was observed that citrus canopy damage in plots bordered by sunnhemp was greatly reduced. Similarly, some organic growers use sunnhemp as a wind break to prevent the spread of citrus canker. Nitrogen concentration in ci trus leaves was not affected by CC treatments, and values fell within the “adequate” range as documented by Tucker et al. (1995). Similarly Wright et al. (2003) reported that CC did not affect orange leaf nutrient content. Citrus yields during the first harvest were low co mpared to values reported by Whitney et al. (1994) for adult citrus trees, but these lower yields are expl ained by the fact the ‘Hamlin’ variety does not produce economically until its fifth or sixth year after planting (Ferguson, 1995). Degree Brix and acidity fe ll within the normal range as stated by Wardowski et al. (1995), and va lues were similar to thos e of Berger et al. (1996). Conclusions In general, perennial and annual CC treat ments did not affect soil pH, and soil C and N content nor the soil C:N ratio during th e initial 3 years of this study. Continuous application of chicken manure increased soil pH and soil organic matter but the corresponding build-up of soil P concentratio ns was deemed undesirable because it can contribute to surface and groundwater pollu tion through runoff and leaching. For this reason, chicken manure should be alternated with other nutrient source s that are low in P and leguminous cover crops may play a role in restoring P-imbalances. Nematode numbers appeared to be rather low, which may be related to CC occurrence and potential suppression by eventual high soil biodiversity typical of integr ated organic cover crop/citrus systems. Cover crop treatments of row middles did not affect weed growth dynamics in the tree row. However, plan ting cowpea in the tree row did decrease

PAGE 160

144 bermudagrass and crabgrass populations in tr ee rows due to light competition. Planting tall cover crops such as sunnhe mp near young citrus trees redu ced initial tree growth but this was remediated by planting sunnhemp furt her away from the tree row and planting a more compact crop like ‘Iron Clay’ cowpea in the remainder of the tree middles. Cover crop treatments did not affect citrus leaf N, fruit yield, and quality. Although the benefits of CC on soil quality and tree performance were not obvious, this is hardly surprising si nce both soil and tree systems are well-buffered and relatively slow to respond to changing environmental c onditions. Moreover, high initial application rates of chicken manure probably masked more subtle responses of system parameters as affected by treatments. However, it is expect ed that over time and with reduced use of external inputs such as chicken manure, effects on more sensitive parameters as affected by treatments such as particulate orga nic matter, may become more evident.

PAGE 161

145 Table 5.1. Effect of year, season, location, a nd treatments on soil pH for the perennial cover crop study during 2003-2005. Location Row middles Tree rows Season Season Treatment Spring Fall Spring Fall --------------------------------2003 ----------------------------------ACC§ 5.89 6.16 6.11 6.47 PPsp † 5.87 5.82 5.81 6.22 PPsu-os ‡ 5.93 6.31 6.36 6.23 PPsu 5.98 6.05 6.16 6.25 --------------------------------2004 -----------------------------------ACC 6.70 A 5.90 B 6.78 A 6.03 B PPsp 6.39 6.21 6.52 6.27 PPsu-os 6.52 6.26 6.73 6.26 PPsu 6.44 6.21 6.44 6.30 --------------------------------2005 -----------------------------------ACC 6.21 6.07 6.25 6.45 PPsp 6.04 6.50 6.09 6.32 PPsu-os 6.14 6.40 6.22 6.68 PPsu 6.30 6.72 6.21 6.21 § Annual cover crops: crimson clover, sunnhemp and cowpea. † Perennial peanut (PP) planted in spring. ‡ PP planted in summer, the following years PP was over-seeded with crimson clover in fall. PP planted in summe r. Absence of letters within the same column or row indicate no significant differe nces (P<0.05). Means within the same row followed by the same upper case letter do not differ statistically by the LSMEANS adjusted by Tukey test (P<0.05).

PAGE 162

146 Table 5.2. Effect of year, season, location, an d treatments on soil pH for the annual cover crop study during 2003-2005. Location Row middles Tree rows Season Season Treatment Spring Fall Spring Fall --------------------------------2003 -----------------------------------ACC§ 5.82 6.12 5.81 6.12 Perennial Peanut 5.62 B 6.27 A 5.55 B 6.22 A Grass fallow 5.58 5.91 5.58 6.12 Tillage Fallow 5.68 6.04 5.61 5.98 --------------------------------2004 -----------------------------------Annual CC 5.79 6.25 6.24 6.34 Perennial Peanut 5.71 6.09 6.19 6.20 Grass fallow 6.34 6.59 6.31 6.28 Tillage Fallow 5.82 6.25 6.20 6.34 --------------------------------2005 -----------------------------------Annual CC 6.11 6.25 6.18 6.37 Perennial Peanut 6.04 6.40 6.20 6.23 Grass fallow 6.45 6.73 6.19 6.39 Tillage Fallow 6.09 6.47 6.22 6.41 § Annual cover crops. Absence of letters with in the same column or row indicate no significant differences (P<0.05). Means within the same row followed by the same upper case letter do not differ statistically by th e LSMEANS adjusted by Tukey test (P<0.05).

PAGE 163

147 Table 5.3. Effect of year, location, and treatm ent on soil C, N, and C:N for the perennial cover crop study during 2003-2005. Location Row middles Tree rows Row middles Tree rows Row middles Tree rows C N C:N Treatment --------------g kg-1 soil ----------------------------------------2003 ACC§ 6.4 6.5 0.36 B 0.59 A 18.4 11.4 PPsp † 6.5 7.2 0.37 B 0.58 A 18.7 13.0 PPsu-os ‡ 5.8 6.4 0.35 B 0.46 A 17.5 14.4 PPsu 5.5 6.2 0.35 B 0.45 A 16.3 14.0 2004 ACC 8.2 9.3 0.42 B 0.70 A 19.8 13.4 PPsp 8.2 8.6 0.36 B 0.54 A 21.4 A 15.8 B PPsu-os 6.8 7.8 0.33 B 0.54 A 20.9 14.4 PPsu 6.7 7.2 0.34 B 0.56 A 19.6 13.3 2005 ACC 7.2 8.7 0.42 B 0.61 A 17.7 b 14.4 PPsp 7.9 8.9 0.42 B 0.63 A 19.8 b 14.9 PPsu-os 8.1 8.1 0.36 B 0.60 A 22.5 ab A 13.4 B PPsu 8.5 7.5 0.34 B 0.58 A 24.8 a A 12.9 B § Annual cover crops: crimson clover, sunnhemp and cowpea. † Perennial peanut (PP) planted in spring. ‡ PP planted in summer, the following years PP was over-seeded with crimson clover in fall. PP planted in summe r. Absence of letters within the same column or row indicate no significant diffe rences (P<0.05). Means within the same column followed by the same lower case letter and means within the same row followed by the same upper case letter do not differ st atistically by the LSMEANS adjusted by Tukey test (P<0.05).

PAGE 164

148 Table 5.4. Effect of year, lo cation, and treatment on soil C, N, and C:N ratio for the annual cover crop study during 2003-2005. Location Row middles Tree rows Row middles Tree rows Row middles Tree rows C N C:N Treatment ----------------g kg-1 soil ---------------------------------------2003 ACC§ 8.6 8.9 0.41 0.49 20.8 18.4 Grass Fallow 8.2 8.8 0.38 0.48 21.7 18.7 2004 ACC 10.0 11.0 0.42 B 0.65 A 23.7 A 17.3 B Perennial peanut 9.3 10.0 0.37 B 0.59 A 25.1 A 18.9 B Grass Fallow 9.4 11.1 0.40 B 0.65 A 23.7 A 17.6 B Tillage fallow 9.1 11.1 0.35 B 0.64 A 25.9 A 17.8 B 2005 ACC 8.6 9.7 0.46 B 0.65 A 19.0 15.1 Perennial peanut 7.9 9.8 0.39 B 0.58 A 20.7 17.1 Grass Fallow 8.0 9.4 0.42 B 0.65 A 19.4 15.9 Tillage fallow 8.2 9.3 0.37 B 0.57 A 20.1 15.6 § Annual cover crops. Absence of letters with in the same column or row indicate no significant differences (P<0.05). Means within the same row followed by the same upper case letter do not differ statistically by th e LSMEANS adjusted by Tukey test (P<0.05).

PAGE 165

149 Table 5.5. Number of plant-para sitic nematode for the pere nnial cover crop study during 2004 and 2005. Nematode 2004 2005 Spring Fall Spring Fall ----------------Nematodes 100 cm-3 soil -----------------Xiphinema (Dagger) 0 A 0 A 0 A 1 A Hemicriconemoides 0 A 2 A 0 B 4 A Pratylenchus (Lesion) 0 A 1 A 0 A 2 A Longidorus 0 A 0 A 0 A 0 A Paratylenchus (Pin) 1 A 0 A 2 A 1 A Criconemoides (Ring) 27 A 8 B 14 A 16 A Meloidogyne (Root-knot) 1 B 5 A 24 A 16 A Helicotylenchus (Spiral) 8 A 10 A 8 B 28 A Paratrichodorus (Stubby root) 0 A 3 B 1 A 1 A Means within the same row and year followed by the same upper case letter do not differ statistically by the LSMEANS adju sted by Tukey test (P<0.05).

PAGE 166

150 Table 5.6. Number of plant-parasitic nemat ode for the annual cover crop study during 2004 and 2005. Nematode 2004 2005 Spring Fall Spring Fall ----------------Nematodes 100 cm-3 soil -----------------Xiphinema (Dagger) 0 A 0 A 0 A 0 A Hemicriconemoides 0 A 0 A 0 A 0 A Pratylenchus (Lesion) 0 B 3 A 1 B 5 A Longidorus 0 A 0 A 0 A 0 A Paratylenchus (Pin) 36 A 3 B 16 A 7 A Criconemoides (Ring) 25 A 16 A 3 B 18 A Meloidogyne (Root-knot) 0 A 1 A 2 B 6 A Helicotylenchus (Spiral) 0 A 0 A 0 A 1 A Paratrichodorus (Stubby root) 1 B 3 A 1 A 0 A Means within the same row and year followed by the same upper case letter do not differ statistically by the LSMEANS adju sted by Tukey test (P<0.05).

PAGE 167

151Table 5.7. Percentages of ground cover in the tree row for most co mmonly observed weed species as affected by year and season i n tree rows for the perennial cover crop study during 2003-2005. 2003 2004 2005 Species Spring Summer Fall Spring Summer Fall Spring Summer Fall Paspalum notatum 1 c A 7 a A 11 b A 7 b A 5 c A 4 c A 4 b A 4 b A Cynodon dactylon 32 b A 7 a B 26 a A 36 a A 36 a A 55 a A 35 a B 35 a B Vigna unguiculata† 0 c B 15 a A 0 c B 5 b B 16 b A 0 c B 46 a A 46 a A Cyperus spp. 4 c A 0 A 0 c A 3 b A 1 c A 1 c A 7 b A 8 b A Digitaria ciliaris 45 a A 18 a B 0 c B 1 b B 21 b A 1 c A 3 b A 2 b A Richardia scabra 4 c A 0 A 1 c A 1 b A 2 c A 0 3 b A 2 b A Gnaphalium spp 0 0 2 c A 0 B 0 B 3 c A 0 A 0 A Linaria canadensis 0 0 4 c A 0 B 0 B 4 c A 0 A 0 A Rumex acetosella 0 0 9 b A 0 B 0 B 20 b A 0 B 0 B Triodanis perfoliata 0 0 3 c A 0 B 0 B 5 c A 0 A 0 A † Cowpea planted in tree rows in July 2003, and June 2004 and 2005. Iron Clay cv. planted in 2003 and 2005 and ‘Cream-40’ plant ed in 2004. Means within the same column followed by the same lower case letter and means within the same row followed by the same upper case letter do not differ stat istically by the LSMEANS adju sted by Tukey test (P<0.05).

PAGE 168

152 Table 5.8. Effect of year, s eason, and treatments on tree hei ght and trunk diameters for ‘Hamlin’ oranges (perennial cover crop study) during 2002-2005. Height Diameter Season Season Treatment Spring Fall Spring Fall -----------m ------------------------mm --------------2002 ACC§ -0.75 b -10.2 b PPsp † -0.82 a -11.2 a PPsu-os ‡ -0.83 a -11.1 a PPsu -0.81 a -10.9 a 2003 ACC 0.93 b B 1.64 A 16.0 b B 27.3 A PPsp 1.02 a B 1.75 A 17.6 a B 30.9 A PPsu-os 0.97 a B 1.59 A 17.3 a B 29.3 A PPsu 0.99 a B 1.65 A 17.2 a B 28.9 A 2004 ACC 1.85 B 2.16 A 32.7 B 46.1 A PPsp 1.83 B 2.21 A 36.5 B 50.5 A PPsu-os 1.74 B 2.09 A 33.2 B 46.7 A PPsu 1.86 B 2.16 A 35.5 B 47.7 A 2005 ACC 2.28 B 2.61 A 51.6 B 61.3 A PPsp 2.50 B 2.61 A 55.6 B 62.8 A PPsu-os 2.23 B 2.48 A 51.5 B 60.1 A PPsu 2.31 B 2.55 A 54.0 B 60.9 A § Annual cover crops: crimson clove r, sunnhemp and cowpea. † Perennial peanut (PP) planted in spring. ‡ PP planted in summer, the following y ears PP was over-seeded with crimson clover in fall. PP planted in summer. Absence of letters within the same column indicate no significant differences (P<0.05). Means within the same column followed by the same lower case letter and means within the same row followed by the same upper case letter do not differ statistically by the LSMEANS adjusted by Tuke y test (P<0.05). Trunk di ameter was measured at 0.2 m high.

PAGE 169

153 Table 5.9. Effect of year, s eason, and treatments on tree hei ght and trunk diameters for ‘Navel’ oranges (annual c over crop study) during 2003-2005. Height Diameter Season Season Treatment Spring Fall Spring Fall -----------m ------------------------mm --------------2003 ACC§ 0.71 B 0.91 A 11.3 B 14.2 A Perennial Peanut 0.70 B 0.95 A 11.0 B 14.9 A Grass fallow 0.70 B 0.97 A 12.0 B 16.3 A Tillage fallow 0.71 B 0.89 A 11.3 B 14.4 A 2004 ACC 1.09 B 1.39 A 19.4 B 29.4 A Perennial Peanut 1.14 B 1.43 A 23.7 B 31.9 A Grass fallow 1.14 B 1.49 A 21.4 B 31.0 A Tillage fallow 1.06 B 1.35 A 18.8 B 27.4 A 2005 ACC 1.45 B 1.68 A 33.3 B 39.1 A Perennial Peanut 1.54 B 1.73 A 34.0 B 39.3 A Grass fallow 1.49 B 1.70 A 35.7 B 41.7 A Tillage fallow 1.39 B 1.56 A 32.0 B 38.1 A § Annual cover crops. Absence of letters within the same column indicate no significant differences (P<0.05). Means within the same row followed by the same upper case letter do not differ statistically by the LSMEANS adjusted by Tukey test (P<0.05). Trunk diameter was measured at 0.2 m high.

PAGE 170

154 Table 5.10. Effect of cover crop treatment on citrus yield and fruit quality (degree Brix and acidity) for the perennial cover crop study during 2005. Treatment Yield Brix Acid Brix/Acid Mg ha-1 Degrees % ACC§ 4.1 11.9 1.1 10.8 PPsp † 3.9 11.9 1.1 10.8 PPsu-os ‡ 3.4 11.5 1.0 11.4 PPsu 3.6 12.1 1.0 12.0 § Annual cover crops: crimson clover, sunnhemp and cowpea. † Perennial peanut (PP) planted in spring. ‡ PP planted in summer, the following years PP was over-seeded with crimson clover in fall. PP planted in summer. Absence of letters within the same column indicate no significant differences (P<0.05).

PAGE 171

155 CHAPTER 6 SUMMARY AND CONCLUSIONS Citrus is one of the most important agricultural crops in Florida. However, during the past decade increased international competition, continuous urban development, diseases, and more stringent environmental re gulations have greatly affected the citrus industry. In contrast, organi c agriculture is the fastes t increasing segment of US agriculture, with organic sa les increased by 20% annually since 1990. However, despite the rapid growth of organic agriculture, in formation on organic production in general and organic citrus in particular is scarce and there is a lack of pertinent production guidelines. Although growers transiti oning to organic production may benefit from price premiums, they thus also face many challenges, incl uding development of cost-effective weed management strategies. Florida organic citrus growers emphasized that weed control was the most critical factor for growers to be successful duri ng the transition to organic production. Annual and perennial cover crops may constitute an environmentally sound approach for improved weed management in or ganic systems. However, current practices for establishment and management of perenni al cover crops (pere nnial peanut) were developed for conventional forage production and are not appropriate for its use as a cover crop in organic citrus. Res earch to assess the effectiveness of CC is thus needed to provide a scientific basis for appropriate produ ction guidelines for organic citrus growers. To accomplish this an interdisciplinary rese arch program was developed to assess the effectiveness of different c over crops in suppressing weeds in newly established citrus

PAGE 172

156 groves and their effect on soil quality and tree growth. The objectives of this dissertation were to i) determine growth characteristics of annual cover crops a nd their effectiveness in suppressing weeds in newly planted organic citrus groves (Chapter 2); ii) evaluate the effect of planting time on initial perennial peanut (PP) establishment, growth, and dry matter production, and its effectiveness on weed suppression compared to annual CC systems (Chapter 3); iii) quantify the e ffect of perennial peanut and common bermudagrass on citrus N and water uptake unde r controlled conditions (Chapter 4); and iv) assess changes in soil quality, weed popul ation dynamics in tree rows, initial citrus growth, yield, and fruit quality as affected by annual and perennial cover crop treatments (Chapter 5). To accomplish these objectives, two long-term field experiments and one greenhouse study were conducted between 2002 and 2005. Two one-hectare blocks were planted with ‘Hamlin’ and ‘Navel’ orange varieties [ Citrus sinensis (L.) Osb.] both grafted on swingle citrumelo ( C. paradisi Macf. x P. trifoliata (L.) Raf.) during the summer of 2002 and spring 2003, respectively. Th e studies were conducted at the Plant Science Research and Education Unit in Citr a, Florida (29.68 N, 82.35 W) at a certified organic research site on repres entative citrus soils (Candl er and Tavares fine sand). Annual Cover Crop Study The main emphasis in the field experiment planted with ‘Nav el’ variety was to determine the growth characte ristics of annual cover crops (ACC) and their effectiveness in suppressing weeds. A randomized complete block design was used with four blocks, each containing different annual CC treatments which were planted twice a year (summer vs. winter season). Grass and tillage fallows were included for comparison.

PAGE 173

157 Annual cover crops appear to provide th e most cost-effective method for managing weeds in organic citrus systems. Cover crop dry matter, N accumulation, and weed suppression by annual CC varied depending on plant species and season. In general summer CC had the highest biomass a nd N accumulation. Sunnhemp, hairy indigo, cowpea, and alyceclover performed consiste ntly well in terms of biomass production, N accumulation, and weed suppression. Although pigeon pea generally accumulated adequate amounts of biomass and N, its w eed suppression capacity was not always consistent. In terms of winter CC, the trip le mix of radish-rye-crimson clover performed best, had the highest biomass production, and was most effec tive in suppressing weeds. Total biomass (CC + weeds) averaged 9.7 and 4.0 Mg ha-1 for the best summer and winter cover crops, respectively. Corresponding values for total N accumulation were 174 and 69 kg N ha-1. Although weeds may compete for re sources, they also may provide ecological services and our work showed th e complementary role of weeds in nutrient retention and recycling. Thr oughout the course of the study, use of selected CC provided excellent weed control, which was superior to other methods including repeated tillage. Use of twoor three-component winter CC mixes resulted in higher DW and N accumulation and more effective weed suppression compared to monocrops, due to the synergistic and complementary inte raction among system components. Perennial Cover Crop Study The main emphasis of the field experiment planted with ‘Hamlin’ variety was use of perennial cover crop (perenni al peanut) to evaluate the effect of planting time of perennial peanut on its initial establishment, growth, and dry matter production. We also determined its effectiveness in suppressi ng weeds compared to annual CC systems. A randomized complete block design was used, with four repetitions and four treatments 1)

PAGE 174

158 Annual cover crop (ACC), sunnhemp and/or cowpea planted in summer, crimson clover and/or rye planted in fall (t his treatment was included fo r comparison); 2) Perennial peanut (PP) planted in spring (PPsp); 3) Crimson clover plante d in spring (2002) and PP planted in summer, the following years pere nnial peanut was over-seeded with crimson clover in fall (PPsu-os); 4) Fallow in spring (2002) and PP planted in summer (PPsu). In general establishment of PP was very slow. For the spring pl anting stands were poor due to a lack of adequate soil moistu re, while competition with weeds and grasses resulted in slow establishment for all tr eatments. Under our production settings, planting PP after the onset of the rainy season resulted in better initi al stands and more effective weed control. Initial weed suppression by PP was very poor to poor, which was related to its slow initial growth and high weed pressu res. Over-seeding PP w ith crimson clover in fall reduced PP vigor and its effec tiveness in suppressing weeds. Compared to PP, annual CC (ACC) provided mu ch better weed control, especially when species were used that have allelopath ic properties (rye) and/ or retain adequately dense canopies for prolonged periods of tim e (rye and sunnhemp). For both PP and ACC, weed biomass typically was inversely related to DW accumulation of either PP or ACC due to competition for light, water and nut rients. Presence of leguminous CC increased overall N content but weeds also contribu ted to enhanced N retention and nutrient cycling. Citrus, Perennial Peanut, and Bermudagra ss Competition for Nitrogen and Water A greenhouse experiment was conducted at the Agronomy and Physiology facilities at the University of Florida in Gainesville between April 2004 and September 2005. A Soil-N Uptake Monitoring (SUM) system was used to determine N and water uptake dynamics of citrus (CIT), perennial peanut (PP), and comm on bermudagrass (BG)

PAGE 175

159 and to assess the potential for competition for water and N uptake between citrus and species used for groundcover. Treatments in cluded 1) Citrus (CIT ); 2) Bermudagrass (BG); 3) Citrus + bermudagrass (CIT + BG); 4) Perennial pea nut (PP); 5) Citrus + Perennial peanut (CIT + PP); we also used N amended bare soil (reference) columns to assess crop N uptake. Treatments were rep licated four times using a randomized complete block design. The SUM-based N uptake system appeared to work well and overall N recovery from reference columns was consistently hi gh while uptake rates matched those obtained via the 15N technique. Nitrogen uptake followed cy clic patterns as related to plant species, cropping system, growing cycle, ET, solar radiation and soil temperature. Nitrogen uptake was greatest for bermudagrassbased systems, while were similar PP and citrus systems. According to the 15N study, N uptake for bermudagrass was reduced by citrus in the citrus+BG system, and co mpetition for N uptake did occur during the summer months. Shading of bermudagrass by citrus trees was probably the functional mechanism by which N uptake of bermudaggr ass was reduced. Perennial peanut N uptake was not reduced by citrus in the citrus+PP system since PP being a C3 plant has lower light requirements and is less susceptibl e to shading. This cr op can also sustain N uptake via increased symbiotic N fixation when N availability is being reduced. As a result, no obvious competition for N uptake occurred between citr us and PP. Similar conclusions were obtained us ing the SUM-based technique, implying that our soil Nuptake-system (SUM) is well-suited to mon itor N uptake and competition between citrus, PP, and BG. Nitrogen uptake was significan tly reduced after clipping and increased during the pre-clipped cycle related to the N sink capacities of groundcovers.

PAGE 176

160 Water uptake was greatest for the mixed systems and bermudagrass. Citrus and bermudagrass competed for water uptake during the spring and summer seasons. On a field scale, frequent mowing of groundcovers may thus be required to minimize the risk of potential competition for water under wa ter-limiting conditions. However, under our experimental conditions, the competition betwee n citrus and BG for water and N did not significantly affect overall citrus tree growth, whereas presence of citrus trees did reduce bermudagrass N concentration, possibly due to shading. Nitrogen uptake and growth ch aracteristics of PP were not affected by citrus in the citrus+PP system. On the other hand, PP appeared to compete with citrus only for water uptake and also reduced citrus root length, but this did not translate into reduced tree growth. Additional research is required to eluc idate the long term effect of citrus root length reduction on N uptake and th e effect of BG on citrus gr owth characteristics under field conditions. Nitrogen use efficiency was greatest for mixed systems and bermudagrass. Groundcovers such as PP and BG, showed the potential to act as catch crop reducing N leaching. Some of the N accumulated in gr asses growing in row middles may be internally recycled and at a later point be re leased and re-utilized, which would facilitate more efficient N use. Alternatively, under N limiting conditions, presence of grasses near trees may hamper tree establishment in newl y planted groves. It is concluded that perennial peanut may be less prone to compete with citrus trees and also may be a more suitable cover crop for row middles cover since it will also generate extra farm income.

PAGE 177

161 Effect of Cover Crops on Soil Characteri stics, Tree Row Cover and Citrus Growth and Yield In general, perennial and annual CC treatme nts did not affect soil pH, soil C and soil N content, nor the soil C:N ratio during th e initial 3 years of th is study. Continuous application of chicken manure increased soil pH and soil organic matter but the corresponding build-up of soil P concentra tions was deemed undesirable because it potentially can contribute to surface a nd groundwater pollution through runoff and leaching. For this reason, chicken manure should be alternated with other nutrient sources that are low in P and leguminous cover crops may play a role in restoring P-imbalances. Nematode numbers appeared to be rather low, which may be related to CC occurrence and potential suppression by eventual high soil biodiversity typical of integrated organic cover crop/citrus systems. Cover crop treatm ents of row middles did not affect weed growth dynamics in the tree row. However, pl anting cowpea in the tree row did decrease bermudagrass and crabgrass populations in tree rows due to competition for light. Planting tall cover crops such as sunnhemp near young citrus trees reduced initial tree growth but this was remediated by planting s unnhemp further away from the tree row and planting a more compact crop like ‘Iron Cl ay’ cowpea in the remainder of the tree middles. Cover crop treatments di d not affect citrus leaf N, fruit yield, and quality. Although the benefits of CC on soil quality and tree performance were not obvious, this is hardly surprising si nce both soil and tree systems are well-buffered and relatively slow to respond to changing environmental c onditions. Moreover, high initial application rates of chicken manure probably masked more subtle responses of system parameters as affected by treatments. However, it is expect ed that over time and with reduced use of

PAGE 178

162 external inputs such as chicken manure, eff ects on more sensitive parameters as affected by treatments such as particulate orga nic matter, may become more evident. Implications of the Research Effective weed management strategies in or ganic citrus are cri tical for successful transition from conventional to organic syst ems. Integration of annual and perennial cover crops in organic citrus groves will provide growers with environmentally sound alternatives for improved weed management. The use of annual summer CC such as ‘iron clay’ cowpea and/or sunnhemp in row middl es, and alyceclover and/or ‘Iron Clay cowpea in citrus rows during the summer season are viable strategies for weed management, while in winter season we propose the use of a system consisting of a tree strip planted with a mixture of crimson clove r and black oat as winter CC due to their compact canopy and low probability of comp etition with citrus trees for light. Intercropping rye with crimson clover and radi sh would be desirable for row middles in winter. Increased N accumulation in CC-based systems during the summer season may provide benefits to subsequent CC crops and/ or citrus trees via mineralization. Use of continuous CC sequence may also reduce poten tial nutrient losses due to leaching Continuous growth of CC combined with reduced tillage may also enhance C sequestration and N cycling and retention in the soil. Augmented soil organic matter is considered a desirable characteristic of sust ainable systems. In organic systems, this approach may also foster the development of soils that can enhan ce natural suppression of weeds, soil borne diseases, and insect populations, all of them through the mechanism of increased soil organic matter and soil mi crobial diversity. As a result, such an

PAGE 179

163 approach may be environmentally frie ndly and cost-effective due to reduced requirements of external inputs. In terms of PP system, since it is very important to ensure a clean and weed-free seed bed for PP planting by using repeated tillage followed by CC crops such as sunnhemp/‘Iron Clay’ cowpea and winter ry e may be beneficial to reduce weed population in organic systems for a minimum of one year prior to planting perennial peanut. We also propose the use of an inte grated management system with PP being planted in early summer in row middles follo wing repeated rototilli ng of a winter rye CC crop. Annual compact self-reseeding CC can be planted near young trees, complemented with manure or natural fertilizer amendment a pplied to the tree rows only. Once the trees are 5-6 years old and the perenni al peanut is established, sh eep can be introduced in the system to graze the row middles. Future Research Recommendations Focusing on the system as an integrated unity of complementary component whole, the incorporation of a small animal compone nt (sheep) in the citrus grove may be desirable. Animals may provide both a dditional income and may enhance weed suppression and nutrient recy cling in citrus groves. Since using a mix of cover crops rendered superior results, additional research is warranted to provide guidelines for optimal mi xes and mixing ratios. In this case it would be pertinent to look at different system performance parameters including C and N sequestration in plant and soil and potent ial N leaching, and the effects of residue management on soil physical, chemical, and biological properties, weed suppression, nematodes and beneficials as indicators of the health of the overall ecosystem.

PAGE 180

164 Additional research is needed to enhance our understanding of N cycling in organic citrus systems, with special reference to N immobilization, mineralization, and crop N uptake as affected by CC and/or organic amendm ents. Use of lysimeter and/or resin trap studies, will also be critical to monito r potential leaching and environmental risks associated with different management pract ices. Finally, cost-benefit analysis may be required to evaluate what pr oduction practices would be mo st cost-effective on a farm scale.

PAGE 181

165 APPENDIX A ANALYSES OF VARIANCE FOR PERENNIAL PEANUT STUDY Table A.1. Analyses of variance for perenni al peanut (PP) shoot dry weight (DWPP), N accumulation in PP shoots (Nacc-PP), PP leaf area index (LAIPP), number of PP shoots per square meter (shoo t#), Weed dry weight (DWWD), N accumulation in weeds (Nacc-WD), and Cover crop weed index (CCWI). Perennial Peanut Weeds DWPP Nacc-PP. LAIPP Shoot# DWWD NaccWD CCWI Year (Yr) *** *** *** *** *** *** *** Sampling (Spl) *** *** *** *** *** *** ** Treatment (Trtm) *** ** *** *** *** *** *** Yr* Spl *** *** *** *** *** *** *** Yr Trtm *** *** *** *** *** *** *** Spl*Trtm *** ** *** *** *** *** *** Year Spl* Trtm *** *** *** NS *** *** *** Significant at the 0.05 level. ** Significant at the 0.01 level. *** Significant at the 0.001 level. NS = not significant .

PAGE 182

166 APPENDIX B ANALYSES OF VARIANCE FOR EFFECTS OF PERENNIAL PEANUT ( Arachis glabrata Benth.) AND COMMON BERMUDA GRASS ( Cynodon dactylon L.) ON NITROGEN AND WATER UP TAKE OF CITRUS Table B.1. Analyses of variance for the e ffect of ground covers on N and water uptake. Effect Nitrogen Water Significance *** Season *** *** Sampling (time) *** *** Cycle *** *** Treatment *** ** Season*sampling ** ** Season*cycle *** *** Season*treatment *** *** Sampling*cycle ** *** Sampling*treatment *** *** Cycle*treatment ** *** Season*sampling*cycle ** *** Season*sampling*treatment NS ** Season*cycle*treatment *** *** Sampling*cycle*treatment *** NS Season*sampling*treatment*cycle NS NS NS= not significant *, **, *** Significant at level P= 0.05, 0.01, and <0.0001, respectively.

PAGE 183

167 Table B.2. Analyses of variance for the eff ect of cropping system on bermuda grass and perennial peanut shoot dry weight (D W), nitrogen concentration (Nconc) and nitrogen accumulation (Naccum). Effect DW Nconc Naccum Significance Year ** ** NS Season ** ** ** Cycle ** ** ** Treatment ** *** NS Year*season ** ** NS Year*cycle ** ** Year*treatment ** NS Season*cycle ** ** ** Season*treatment *** Cycle*treatment ** ** Year*season*cycle ** *** NS Year*season*treatment ** ** Year*cycle*treatment ** NS NS Season*cycle*treatment ** *** NS Year*season*cycle*treatment NS NS NS NS= not significant *, **, *** Significant at level P= 0.05, 0.01, and <0.0001, respectively.

PAGE 184

168 APPENDIX C INITIAL SOIL CONDITIONS OF EXPE RIMENTAL SITE, DECEMBER 2001 (SOIL ANALYSES RESULTS) Figure C.1. Initial soil conditions at the e xperimental site in December 2001 (soil analyses results from Analytical Re search Lab. IFAS, Gainesville, FL.)

PAGE 185

169 APPENDIX D ANALYSES OF VARIANCE FOR ANNUAL AND PERENNIAL COVER CROPS ON SOIL AND CITRUS TREE CHARACTERIS TICS, CITRUS TREE ROW GROUND COVER, AND CITRUS YIELD AND QUALITY Table D.1. Analyses of variance for the effect of perennial cover crops (PCC) and annual cover crops (ACC) on soil pH. Effect PCC ACC Significance Year *** *** Season *** Location NS NS Treatment NS NS Year*season *** ** Year*location NS ** Year*treatment NS *** Season*location NS Season*treatment NS NS Location*treatment NS NS Year*season*location *** Year*season*treatment *** NS Year*location*treatment NS Season*location*treatment NS Year*season*location*treatmentNS NS NS= not significant *, **, *** Significant at level P= 0.05, 0.01, and <0.0001, respectively.

PAGE 186

170 Table D.2. Analyses of variance for the effect of perennial cover crops on soil carbon (C), nitrogen (N) and C:N ratio. Effect C N C:N Significance Year *** ** *** Season NS *** Location *** *** Treatment NS NS ** Year*season ** ** NS Year*location NS NS ** Year*treatment NS NS NS Season*location NS NS Season*treatment NS NS NS Location*treatment NS NS NS Year*season*location NS ** Year*season*treatment NS NS NS Year*location*treatment NS NS NS Season*location*treatment NS NS NS Year*season*location*treatmentNS NS NS NS= not significant *, **, *** Significant at level P= 0.05, 0.01, and <0.0001, respectively. Table D.3. Analyses of variance for the eff ect of annual cover crops on soil carbon (C), nitrogen (N) and C:N ratio. Effect C N C:N Significance Year *** ** ** Season *** NS NS Location *** *** *** Treatment NS NS NS Year*season NS ** NS Year*location NS ** NS Year*treatment NS NS NS Season*location NS NS NS Season*treatment NS NS NS Location*treatment NS NS NS Year*season*location NS NS Year*season*treatment NS NS NS Year*location*treatment NS NS Season*location*treatment NS NS NS Year*season*location*treatmentNS NS NS NS= not significant *, **, *** Significant at level P= 0.05, 0.01, and <0.0001, respectively.

PAGE 187

171 Table D.4. Analyses of variance for the effect of perennial cover crops (PCC) and annual cover crops (ACC) on soil nematode populations. Effect PCC ACC Significance Year *** NS Season *** ** Location NS NS Species *** *** Year*season NS ** Year*location NS NS Year*Species *** *** Season*location NS NS Season*Species *** *** Location*species NS NS Year*season*location NS NS Year*season*species *** *** Year*location*species NS NS Season*location*species NS NS Year*season*location*species NS NS NS= not significant *, **, *** Significant at level P= 0.05, 0.01, and <0.0001, respectively. Table D.5. Analyses of variance for the effect of perennial cover crops on tree-row cover. Effect Significance Year *** Season *** Treatment NS Species *** Year*season ** Year*treatment NS Year*Species *** Season*treatment NS Season*Species *** Treatment*species NS Year*season* treatment NS Year*season*species *** Year* treatment*species NS Season* treatment*species Year*season* treatment *species NS NS= not significant *, **, *** Significant at level P= 0.05, 0.01, and <0.0001, respectively.

PAGE 188

172 Table D.6. Analyses of variance for the effect of perennial cover crops on citrus tree height (Height) and diameter (Diam). Effect Height Diam -------------------------------Si gnificance --------------------Year *** *** Season *** *** Treatment * Year*season *** *** Year*treatment * Season*treatment * Year*season*treatment NS NS NS= not significant *, **, *** Significant at level P= 0.05, 0.01, and <0.0001, respectively. Table D.7. Analyses of variance for the effect of annual cover crops on citrus tree height (Height) and diameter (Diam). Effect Height Diam -------------------------------Si gnificance --------------------Year *** *** Season *** *** Treatment NS NS Year*season ** *** Year*treatment NS NS Season*treatment NS NS Year*season*treatment NS NS NS= not significant *, **, *** Significant at level P= 0.05, 0.01, and <0.0001, respectively. Table D.8. Analyses of variance for the effect of perennial cover crops (PCC) and annual cover crops (ACC) on nitrogen ci trus leaf concentration. Effect ACC PCC -------------------------------Si gnificance --------------------Year NS NS Treatment NS NS Year*treatment NS NS NS= not significant *, **, *** Significant at level P= 0.05, 0.01, and <0.0001, respectively.

PAGE 189

173 LIST OF REFERENCES Abawi, G.S., and T.L. Widmer. 2000. Impact of soil health management practices on soilborne pathogens, nematodes and root diseases of vegetable crops. Appl. Soil Ecol. 15:37-47. Aflakpul, G.K., P.J. Gregory, and R.J. Fr oud-Williams. 1998. Effect of temperature on seed germination rate of Striga hermonthica (Del) Benth. Crop Prot. 17:129-133. Agele, S.O., B.S. Ewulo, and I.K. Oyewusi. 2005. Effects of soil management systems on soil physical properties, microbial bioma ss and nutrient distribution under rainfed maize production in a humid rainforest al fisol. Nut. Cycling Agroecosyst. 72:121134. Aguilar, V. 2001. Selective weed and ground cover management in a coffee plantation with shade trees in Nicaragua. Doctoral Th esis Swedish University of Agricultural Sciences. Acta Universitatis Agriculturae Sueciae, Agraria 269. Uppsala, Sweden. Aiyelaagbe, I. O. 2001. Productivity of an intercropped sweet orange orchard in southwestern Nigeria. Biol. Ag ric. Hortic. Int J. 18:317-325. Akemo, M. C., E.E. Regnier, and M.A. Be nnett. 2000. Weed suppression in spring-sown rye ( Secale cereale )-pea ( Pisum sativum ) cover crop mixes. Weed Technol. 14:545549. Alboresi, A., C. Gestin, M.T. Leydecker, M. Bedu, M. Meyer, and H.N. Truong. 2005. Nitrate, a signal relieving seed dorma ncy in Arabidopsis. Plant Cell Environ. 28:500-512. Allen, L.H., Jr. 1994. Carbon dioxide increase : Direct impacts on crops and indirect effects mediated through anticip ated climatic changes. P 425-459. In Boote, K.J., J. M. Bennet, and T. Sinclair (eds) Physio logy and determination of crop yield. ASA, CSSA, and SSSA, Madison, WI. Alsaadawi, I.S., and A.J. Alrubeaa. 1985. Allelopathic effects of Citrus aurantium L. I. vegetational patterning. J. Chem. Ecol. 11:1515-1526. Altieri, M. 1995. Agroecology. The Science of Sustainable Agriculture. Westview Press, Boulder, CO. Altieri, M. 1999. The ecological role of biodiversity in agro ecosystems. Agric. Ecosyst. Environ. 74:19-31.

PAGE 190

174 Altieri, M., and D. Letourneau. 1982. Vegetati on management and biological control in agroecosystems. Crop Prot. 1:405-430. Altieri, M., and C. Nichols. 2003. Soil fertility and insect pests: harmonizing soil and plant health in agroecosystems. Soil Tillage Res. 72:203-211. Athearn, K. 2004. Efficacy of environmental la beling: an economic analysis with two examples from Florida agriculture. Ph.D dissertation. University of Florida, Gainesville, FL. Bailey, A.W., and C.E. Poulton. 1968. Plant communities and environmental relationships in a portion of the Tillamook Burn, northwest Oregon. Ecology 49: 114. Bailey, K. L., and G. Lazarovits. 2003. Suppres sing soil-borne diseases with residue management and organic amendments. Soil Tillage Res. 72:169-180. Balkcom, K., and D.W. Reeves. 2005. Sunn-he mp utilized as a legume cover crop for corn production. Agron. J. 97:26-31. Baltensperger, D.D., G.M. Prine, and R.A. Dunn. 1986. Root-knot nematode resistance in Arachis glabrata Peanut Sci. 13:78-80. Barberi, P., and M. Mazzoncini. 2001. Ch anges in weed community composition as influenced by cover crops and management systems in continuous corn. Weed Sci. 49: 491-499. Bauer P.J., and W. Reeves. 1999. A comparison of winter cereal sp ecies and planting dates as residue cover for cotton grow n with conservation tillage. Crop Sci. 39:1824-1830. Belesky, D.P., and J.M. Fedders. 1995. Wa rm-season grass productivity rate as influenced by canopy management. Agron. J. 87:42-48. Benech-Arnold, R.L., R.A. Sanchez, F. Forcella, B.C. Kruk, and C.M. Ghersa. 2000. Environmental control of dormancy in weed seed banks in soil. Field Crop Res. 67:105-122. Berger, H., J. Opazo, S. Orellana, and L. Ga lletti. 1996. Potassium fe rtilizers and orange postharvest quality. Proc. Int. Soc. Citriculture 2:759-761. Berkelmans, R., H. Ferris, M. Tenuta, a nd A.H. Van Bruggen. 2003. Effects of long-term crop management on nematode trophic levels other than plant feeders disappear after 1 year of disruptive soil ma nagement. Appl. Soil Ecol. 23:223-235. Bevington, K.B., and W. S. Castle. 1985. A nnual root growth pattern of young citrus trees in relation to shoot growth, soil temp eratures, and soil water content. J. Amer. Soc. Hortic. Sci. 110:840-845.

PAGE 191

175 Bidwell, S., P.M., Attiwill, and M.A. Ad ams. 2006. Nitrogen availability and weed invasion in a remnant native woodland in urban Melbourne. Aust. Ecology. 31:262270. Blackshaw, R., J. Moyer, R. Doram, and L. Boswell. 2001. Yellow sweet clover, green manure, and its residues effectively s uppress weeds during fallow. Weed Sci. 49:406-413. Bloodworth, L.H., and J.R. Johnson. 1995. Cover crops and tillage effects on cotton. J. Prod. Agric. 8:107-112. Boman, B.J. 1994. Evapotranspiration by young Flor ida flatwoods citrus trees. J. Irrig. Drain. Eng. 120: 80-88. Bond, W., and A.C. Grundy. 2001. Non-chemical weed management in organic farming systems. Weed Res. 41:383-405. Booth, B.D., S.D. Murphy, and C.J. Swanton. 2003. Weed Ecology. CABI publishing, Wallinford Oxon, UK. Bradshaw L., and W. Lanini. 1995. Use of pe rennial cover crops to suppress weeds in Nicaragua coffee orchards. In t. J. Pest Manag. 41:185-194. Brandsaeter L., and J. Netland. 1999. Winter a nnual legumes for use as cover crops in row crops in northern regions: I Fiel d experiments. Crop Sci. 39:1369-1379. Buckles, D., and B. Triomphe. 1999. Adoption of mucuna in the farming systems of northern Honduras. Agroforest. Syst. 47:67-91. Budhar, M.N., and N. Tamilselvan. 2003. Weed dynamics, yield and economics in sorghum (Sorghum bicolor) plus pulse intercropping systems under rainfed condition. Indian J. Agron. 48:93-95. Caamal-Maldonado, J., J. Jimnez, A. Torres, and A. Anaya. 2001. The use of allelopathic legume cover and mulches species for weed control in cropping systems. Agron. J. 93:27-36. Cabrera, M.L., and D.E. Kissel. 1989. Review and simplification of calculations in 15N tracer studies. Fe rt. Res. 20:11-15. Calkins, J., and B. Swanson. 1995. Comparison of conventional and alternative nursery weed management strategies. Weed Technol. 9:761-767. Canudas, E.G., K.H. Quesenberry, L.E. Sollenberger, and G.M. Prine. 1989. Establishment of two cultivars of rhizoma peanut as affected by weed control and planting rate. Trop. Grass. 23:162-170.

PAGE 192

176 Carsky, R.J., B. Oyewole, and G. Tian. 2001. Ef fect of phosphorus application in legume cover crop rotation on subsequent maize in the savanna zone of West Africa. Nutr. Cycling Agroecosyst. 59:151-159. Cerezo, M., P. Garcia-Agustin, and E. Pr imo-Millo. 1999. Influence of chloride and transpiration on net 15N-uptake rate by citrus. Ann. Bot. 84:117-120. Chacon, J.C., and S. R. Gliessman. 1982. Use of the “non-weed” concept in traditional tropical agroecosystems of south-eas tern Mexico. Agro-Ecosyst. 8:1-11. Chaves, J.C., M.A. Pavan, and A. Calegari. 1997. Input of dry matter and nutrients to the soil from cover plants cultivated between rows of perennial crops and their effects on soil reaction. Arq. Biol. Tecn. 40:47-55. Chen, X., J. Tang, Z. Fang, and K. Shimiz u. 2004. Effect of weed communities with various species numbers on soil features in a subtropical citrus orchard ecosystem. Agric. Ecosys. Environ. 102:377-388. Cherr, C.M., J.M.S. Scholberg, and R. McSorley. 2006. Green manure approaches to crop production: a synthesis. Agron. J. 98:302-319. Clement, C.R., and J. DeFrank. 1998. The us e of ground cover during the establishment of heart-of-palm plantations in Hawaii. HortScience 33:814-815. Cohen, S., E. Raveh, Y. Li., A. Grava, and E.E. Goldschmidt. 2005. Physiological responses of leaves, tree growth and fru it yield of grapefruit tree under reflective shade screens. Sci. Hortic. 107:25-35. Cohn, E.J., O.W. Van Auken, and J.K. Bu sh. 1989. Competitive interaction between Cynodon dactylon and Acacia smallii seedlings at different nutrient levels. Am. Midl. Nat. 121:265-272. Coleman, R.G., 1995. Perennial peanut-potential looms larger in citrus. Citrus Ind. 76: 43-44. Conceicao, P.C., T.J. Amado, J. Mielnicz uk, and E. Spagnollo. 2005. Soil organic matter and other attributes as indi cators to evaluate soil quali ty in conservation systems. Rev. Bras. Ciencia Solo. 29:777-788. Cookson, W.R., D.A. Abaye, P. Marschner, D.V. Murphy, E.A. Stockdale, and K.W. Goulding. 2005. The contribution of soil or ganic matter fractions to carbon and nitrogen mineralization and microbial co mmunity size and structure. Soil Biol. Biochem. 37:1726-1737. Craine, J.M., J. Fargione, and S. Sugita 2005. Supply pre-emption, not concentration reduction, is the mechanism of competiti on for nutrients. New Phytologist 166:933940.

PAGE 193

177 Creamer, N.G., and K.R. Baldwin. 2000. An eval uation of summer cover crops for use in vegetable production systems in Nort h Carolina. HortScience 35:600-603. Daniel, J.B., A.O. Abaye, M.M. Alley, C. W. Adcock, and J.C. Maitland. 1999. Winter annual cover crops in a Virginia no-til l cotton production system: I. Biomass production, ground cover, and nitrogen a ssimilation. J. Cotton Sci. 3:74-83. Datamonitor, 2005. Organic food in the United States. Industry Profile 12:1-17. Daubenmire, R.F. 1959. A canopy-coverage method. Northwest Sci. 33:43-64. Dhyani, S.K., and R.S. Tripathi. 1999. Tree gr owth and crop yield under agrisilvicultural practices in north-east Indi a. Agrofor. Syst. 44:1-12. Dimitri, C., and C. Greene. 2002. Recent grow th patterns in the U.S. organic foods market. Agriculture Information Bulletin No. 777. USDA, ERS.Washington, D.C. Ding, G.W., X.B. Liu, S. Herbert, J. Nova k, D. Amarasiriwardena, and B.H. Xing. 2006. Effect of cover crop on soil or ganic matter. Geoderma 130:229-239. Dou, H., A.K. Alva, and B.R. Khakural. 1997. Nitrogen mineralizati on from citrus tree residues under different production c onditions. Soil Sci. Am. J. 61:1226-1232. Dunavin, L.S. 1990. Cool-season forage crops seeded over dormant rhizoma peanut. J. Prod. Agric. 3:112-114. Dunavin, L.S. 1992. Florigraze rhizoma pea nut in association with warm-season perennial grasses. Agron. J. 84:148-151. Duncan, L. W., and E. Cohn. 1990. Nematode pa rasites of citrus. P p. 321-346 in Luc M., R.A. Sikora, and J. Bridge, eds. Plant parasitic nematodes in subtropical and tropical agriculture. CAB Intern ational, Wallingford, U. K. Duncan, L.W., R.N. Inserra, and D. Dunn. 1998. Seasonal changes in citrus fibrous roots starch concentration a nd body length of female Pratylenchus coffeae Nematropica 28:263-266. Dyck E., M. Liebman, and M.S. Enrich. 1995. Cr op-weed interference as influenced by a leguminous or synthetic fertilizer nitr ogen source: I. Doublecropping experiments with crimson clover, sweet corn, and la mbsquarters. Agric. Ecosyst. Environ. 56:93-108. Eneji, A.E., T. Honna, S. Yamamoto, T. Saito, and T. Masuda. 2002. Nitrogen transformations in four Japanese soils following manure plus urea amendment. Com. Soil Sci. Plant Anal. 33:53-66.

PAGE 194

178 Espindola, J.A., J.G. Guerra, D.L. Almeid a, M.G. Texeira, and S. Urquiaga. 2005. Evaluation of perennial herbaceous legumes with different phosphorus sources in a Brazilian ultisol. Renew. Agric. Food Sys. 20:56-62. Fageria, N.K., V.C. Ballgar, and B.A. Baile y. 2005. Role of cover crops in improving soil and row crop productivi ty. Commun. Soil Sci. Plant Anal. 36:2733-2757. Feigenbaum, S, H. Bielorai, Y. Erne r, and S. Dasberg. 1987. The fate of 15N labelled nitrogen applied to mature ci trus trees. Plant Soil 97:179-187. Fennimore, S., and L. Jackson. 2003. Organic amendment and tillage effects on vegetable field weed emergence and seed banks. Weed Technol. 17:42-50. Ferguson, J. 1984. Occurrence and control of citrus nematodes. Citrus Ind. 23:17-20. Ferguson, J. 1995. Florida, your dooryard citrus guide. University of Florida. SP178. Cooperative extension servic e, IFAS, Gainesville FL. Fernandez, O.N., O.R. Vignolio, and E.C. Requesens. 2002. Competition between corn ( Zea mays ) and bermuda grass ( Cynodon dactylon ) in relation to the crop plant arrangement. Agronomie 22:293-305. Firth, D.J., G.G. Johns, and R.D. Whalle y. 2003. Glasshouse and field studies on the effects of groundcovers on banana and m acadamia growth and water relations. Aust. J. Exp. Agric. 43:1245-1254. Fischler, M., and C.S. Wortmann. 1999. Green manures for maize-bean systems in eastern Uganda: Agronomic performance a nd farmersÂ’ perceptions. Agrofor. Syst. 47:123-138. Florida Agricultural Statis tics Service. 2006. Citrus summary 2004-05. Florida Dept. Agric. And Consumer Serv. Tallaha ssee, FL. [Online] available at http://www.nass.usda.gov/Statistics_by_Sta te/Florida/Publicat ions/Citrus/cs/200405/cs0405all.pdf verified 1 June 2006). Fortuna, A., R.R. Harwood, G.P. Roberts on, J.W. Fisk, and E.A. Paul. 2003. Seasonal changes in nitrification potential associat ed with application of N fertilizer and compost in maize systems of southwest Michigan. Agric. Ecosyst. Env. 97: 285293. Freire, M.J., C.A. Kelly-Bagazo, and K.H. Quesenberry. 2000. Establishment, yield, and competitiveness of rhizoma perennial peanut germplasm on a flatwood soil. Soil Crop Sci. Soc. Florida Proc. 59:68-72. French, E.C., and G.M. Prine. 1991. Perennial peanut establishment guide. SS-AGR-35. Florida Cooperative Extension Service, IFAS University of Florida, Gainesville, FL.

PAGE 195

179 French, E.C., G.M. Prine, and A.R. Blount. 2001. Perennial peanut: An alternative forage of growing importance. SS-AGR-39. Florid a Cooperative Extens ion Service, IFAS, University of Florida. Gainesville, FL. French, E.C., G.M. Prine, W.R. Ocumpa ugh, and R.W. Rice. 1994. Regional experiences with forage Arachis in the United Stat es. p. 169-186. In Kerridge, P.C., and B. Hardy (eds.) Biology and agronomy of forage Arachis Centro Internacional de Agricultura Tropical, Cali, Colombia. Futch, S. 2005. Maximizing weed control in Fl orida citrus. Doc. HS-1007. Series of the Horticultural Sciences Department, Fl orida Cooperative Extension Service, Institute of Food and Agricultural Scienc es, University of Florida. [Online] available at http://edis.ifas.ufl.edu/HS237 (verified 3 March 2006). Gallaher, R.N., C.O. Weldon, and J.G. Futr al. 1975. An aluminum block digester for plant and soil analysis. Soil Sc i. Soc. Amer. Proc. 39:803-806. Gallagher, R.S., and J. Cardina. 1998. Phytoc hrome-mediated amaranthus germination I: Effect of seed burial and germina tion temperature. Weed Sci. 46:48-52. Gallagher, R.S., E.C.M. Fernandez, and E. L. McCallie. 1999. Weed management through short-term improved fallows in tropical agroecosystems. Agroforest. Syst. 47: 197221. Gallandt, E.R., M. Liebman, and D.R. Huggins. 1999. Improving soil quality: implications for weed manageme nt. J. Crop Product. 2:95-121. Gliessman, S. 1998. Agroecology: Ecological processes in Sustainable Agriculture. Sleeping Bear Press, Chelsea, MI. Guazzelli, L., F.S. Davies, J.J. Ferguson, and W.S. Castle. 1995. Nitrogen nutrition and growth of ‘Hamlin’ orange nursery tr ees on ‘Swingle’ ci trumelo rootstock. HorTechnology 5:147-151. Guglielmini, A.C., and E.H. Satorre. 2004. The effect of non-inversi on tillage and light availability on dispersal and spatial growth of Cynodon dactylon Weed Res. 44:366-374. Hue, N. V., and D.L. Licudine. 1999. Amelioration of subsoil acidity through surface application of organic manur es. J. Environ. Q. 28:623-632. Igual, J.F., J.S. Ricardo, and M.V. Sergio. 2002. Feasibility of the organic production of citrus fruit within a new EU agri-envir onment and rural development policy. Agric. Tropica Subtropica 35:65-72. Ikpe, F.N., L.G. Owoeye, and M.P. Gichur e. 2003. Nutrient recy cling potential of Tephrosia candida in cropping systems of southeastern Nigeria. Nut. Cycl. Agroecos. 67:129-136.

PAGE 196

180 Jenkins, W.R. 1964. A rapid centrifugal flot ation technique for separating nematodes from soil. Plant Di sease Reporter 48:692. Jeranyama, P.O., B. Hesterman, S.R. Waddington, and R.R. Harwood. 2000. Relayintercropping of sunnhemp and cowpea in to a smallholder maize system in Zimbabwe. Agron. J. 92:239-244. Jiang, Y.W., R.N. Carrow and R.R. D uncan. 2005. Physiological acclimation of seashore paspalum and bermudagrass to low light. Sci.Hortic. 105:101-115. Jiang, Y.W., R.R. Duncan, and R.N. Carrow. 2004. Assessment of low light tolerance of seashore paspalum and bermudagrass. Crop Sci. 44:587-594. Johnson, C.R., B.A. Reiling, P. Mislevy, and M.B. Hall. 2001. Effects of nitrogen fertilization and harvest date on yield, digestibility, fiber, and protein fractions of tropical grasses. J. Anim. Sci. 79:2439-2448. Johnson, S.E., L.E. Sollenberger, and J.M. Bennett. 1994. Yield and reserve status of rhizoma peanut growing under shade. Crop Sci. 34:757-761. Jones, J.B., and V.W. Case. 1991. Sampli ng, handling, and analyzing plant tissue samples. p. 389-415. In R.L. Westerman (ed.). Soil testing and plant analysis. 3rd ed. Book Series no 3. Soil Sci. Soc. Am. Madison, WI. Jordan, N., J. Zhang, and S. Huerd. 2000. Arbus cular-mychorrhizal fungi: potential roles in weed management. Weed Res. 40:397-410. Juskiw, P.E., J.H. Helm, and D.F. Salmon. 2000. Forage yield and qua lity for monocrops and mixtures of small grain cereals. Crop Sci. 40:138-147. Justes, E., M. Bruno, and B. Nicolardot. 1999. Comparing the effectiveness of radish cover crop, oilseed rape volunteers and o ilseed rape residues incorporation for reducing nitrate leaching. Nutr. Cycling Agroecosyst. 55:207-220. Kabir, Z., and R.T. Koide. 2002. Effect of au tumn and winter mycorrhizal cover crops on soil properties, nutrient uptake and yield of sweet corn in Pennsylvania, USA. Plant Soil 238:205-215. Karpenstein-Machan, M., and R. Stuelpnage l. 2000. Biomass yield and nitrogen fixation of legumes monocropped and intercroppe d with rye and rotation effects on a subsequent maize crop. Plant Soil 218:215-232. Kato, T., S. Kubota, and S. Bambang. 1982. Uptake of 15N-nitrate by citrus trees in winter and repartitioning in spring. J. Japan. Soc. Hortic. Sci. 50:421-426. Kennedy, A. 1998. Soil microorganisms for w eed management. J. Crop Prod. 2:123-138.

PAGE 197

181 King, A.P., and A. Berry. 2005. Vineyard 15N, nitrogen and water status in perennial clover and bunch grass cover crop systems of CaliforniaÂ’s central valley. Agric, Ecosyst. Environ. 109:262-272. Koutika, L.S., S. Hauser, and J. Henrot. 2001. Soil organic matter assessment in natural regrowth, Pueraria phaseoloides and Mucuna pruriens fallow. Soil Biol. Biochem. 33:1095-1101. Kremer, R., and J. Li. 2003. Developing weed -suppressive soils through improved soil quality management. Soil Tillage Res. 72:193-202. Kristensen, K.L., and K. T horup-Kristensen. 2004. Root growth and nitrate uptake of three different catch crops in deep soil layers. Soil Sci. Soc. Am. J. 68:529-537. Lanini, W.T., F. Zalom, J. Marois, and H. Ferris. 1994. Researchers find short-term insect problems, long-term weed problems. Calif. Agric. 48:27-33. Lea-Cox, J.D., and J.P. Syvertsen. 1996. Ho w nitrogen supply affects growth and nitrogen uptake, use efficiency, and loss from citrus seedlings. J. Amer. Soc. Hortic. Sci. 121:105-114. Legaz, F., and E. Primo-Millo. 1984. Infl uence of flowering, summer, and autumn flushes on the absorption and distribution of nitrogen compounds in citrus. Proc. Int. Soc. Citriculture 1:224-233. Legaz, F., and E. Primo-Millo. 1988. Absorptio n and distribution of nitrogen-15 applied to young orange trees. p. 643-661. In Proceedings of the sixth international citrus congress. Goren R. and Mendel K. (Eds). Balaban publishers, Margraf Scientific books. Philadelphia. PA. Leon, R.G., A.D. Knapp, and M.D. Owe n. 2004. Effect of temperature on the germination of common waterhemp ( Amaranthus tuberculatus ), giant foxtail ( Setaria faberi ), and velvetleaf ( Abutilon theophrasti ). Weed Sci. 52:67-73. Liebman, M., and A. Davis. 2000. Integration of soil, crop and weed management in lowexternal-input farming systems. Weed Res. 40:27-47. Liu, F., and N. OÂ’Connell. 2002. Off-site moveme nt of surface-applied simazine from a citrus orchard as affected by irrigati on incorporation. Weed Sci. 50:672-676. Macchia, E.T., R. McSorley, L.W. Duncan, and J.S. Syvertsen. 2003. Effects of perennial peanut ( Arachis glabrata ) ground cover on nematode communities in citrus. J. Nematol. 35:450-457. Mader, P., A. Fliebach, D. Dubois, L. Gunst, P. Fried, and U. Niggli. 2002. Soil fertility and biodiversity in organic farming. Science 296:1694-1697.

PAGE 198

182 Mafongoya, P.L., and B.H. Dzowela. 1999. Bi omass production of the tree fallows and their residual effect on maize in Zimbabwe. Agrofor. Syst. 47:139-151. Marschner, H. 2003. Mineral nu trition of higher plants. Academic Press, San Diego, CA. Materechera, S.A., and T.S. Mkhabela. 2002. The effectiveness of lime, chicken manure and leaf litter ash in ameliorating acid ity in a soil previously under black wattle ( Acacia mearnsii ) plantation. Biores. Technol. 85:9-16. Matheis, H., and R. Victoria. 2005. Cover cr ops and natural vegetation mulch effect achieved by mechanical management with la teral rotary mower in weed population dynamics in citrus. J. Environ. Sci. Health 40:185-190. Maust, B.E., and J.G. Williamson. 1994. N itrogen nutrition of containerized citrus nursery plants. J. Am. Soc. Hortic. Sci. 119:195-201. McDonald, L.M., P. Wright, and D.A. MacLeod. 2001. Nitrogen fixation by lablab ( Lablab purpureus ) and lucerne ( Medicago sativa ) rotation crops in an irrigated cotton farming system. Aust. J. Exp. Ag. 41:219-225. McNeal, B.L., C.D. Stanley, W.D. Graham P.R. Gilreath, D. Downey, and J.F. Creighton. 1995. Nutrient-loss tre nds for vegetable and citrus fields in west-central Florida: I. Nitrate. J. Environ. Qual. 24:95-100. McSorley, R. 2001. Multiple cropping systems for nematodes management: A review. Soil Crop Sci. Soc. Fla. Proc. 60:132-142. Meda, A.R., M.E. Cassiolato, M.A. Pava n, and M. Miyazawa. 2001. Alleviating soil acidity through plant organic compounds Braz. Arch. Biol Technol. 44:185-189. Miranda, C.H., A. Vieira, and G. Cadisc h. 2003. Determination of biological nitrogen fixation by the forage groundnut ( Arachis spp .) using the 15N natural abundance technique. Revista Brasileir a de Zootecnia 32:1859-1865. Mooney, P.A., and A.C. Richardson. 1992. Seasona l trends in the uptake and distribution of nitrogen in Satsuma mandarins. Proc Int. Soc. Citriculture 2:593-597. Morra, M. J., and J.A. Kirkegaard. 2002. Isothi ocyanate release from soilincorporated brassica tissues. Soil Biol. Biochem. 34:1683-1690. Mubarak, A.R., A.B. Rosenani, A.R. Anuar, a nd D. Siti. 2003. Effect of incorporation of crop residues on maize-groundnut sequence in the humid tropics. II. Soil physical and chemical properties. J. Plant Nutr. 26:2343-2364. Muir, J. 2002. Hand-plucked forage yield a nd quality and seed production from annual and short-lived perennial warm-season legum es fertilized with composted manure. Crop Sci. 42:897-904.

PAGE 199

183 Mullahey, J.J., R.E. Rouse, and E.C. French. 1994. Perennial peanut in citrus grovesan environmentally sustainable agricultural system. p.479-483. In Campbell (ed.) Proceedings Second Conference Environm entally Sound Agriculture, ASAE, St. Joseph, MI. Mullen, B.F., I. K. Rika, D.A. Kaligis, and W.W. Stur. 1997. Performance of grasslegume pastures under coconuts in Indonesia. Exp. Agric. 33:409-423. Muraro, R.P., Roka, F.M., and R.E. R ouse. 2003. Budgeting costs and returns for southwest Florida citrus production, 2002-03. EDIS FE 434, Florida Cooperative Extension Service, IFAS, Universi ty of Florida, Gainesville, FL. Myrold. D. 1999. Transformations of Nitrogen. p 259-294. In Sylvia, D., J. Fuhrmann, P. Hartel, and D.A. Zuberer (eds). Principl es and applications of soil microbiology. Prentice Hall Inc. Upper Saddle River, NJ. Nelson, L. 2004. Organic FAQs. What is organic farming? Nature 428:796. Newman, Y.C., L.E. Sollenberger, and K. J. Boote. 2005. Temperature and carbon dioxide effects on nutritive value of rh izoma peanut herbage. Crop Sci. 45:316-321. Ngouajio, M., and M.E. McGiffen. 2002. Go ing organic changes weed population dynamics. HorTechnology 12:590-596. Ngouajio, M., and H. Mennan. 2005. Weed populations and pickling cucumber ( Cucumis sativus ) yield under summer and winter cove r crop systems. Crop Prot. 24:521-526. Norsworthy, J.K., L. Brandenberger, N. Bu rgos, and M. Riley. 2005. Weed suppression in Vigna unguiculata with a spring-seeded brassi caceae green manure. Crop Prot. 24:441-447. OÂ’Bannon, J.H., J.D. Radewald, and A.T. Tome rlin. 1972. Population fluctuation of three parasitic nematodes in Florida citrus. J. Nematol. 4:194-199. OÂ’Hallorans, J.M., M.A. Munoz, and P.E. Marquez. 1997. Chicken manure as an amendment to correct soil acidity and fertil ity. J. Agric. Univ. Puerto Rico 81:1-8. OÂ’Hara, G.W. 2001. Nutritional constraints on root nodule bacteria affecting symbiotic nitrogen fixation. Aust. J. Exp. Agric. 41:417-433. Obreza, T.A., and M. Ozores-Hampton. 2000. Management of organic amendments in Florida citrus production systems. So il Crop Sci. Soc. Fla. Proc. 59:22-27. Ocumpaugh, W.R. 1990. Production and nutritive value of Florigraze rhizoma peanut in a semiarid climate. Agron. J. 82:179-182. Odhiambo, J., and A. Bomke. 2001. Grass and legume cover crop effects on dry matter and nitrogen accumulation. Agron. J. 93:299-307.

PAGE 200

184 Organic Farming Research Foundation. 2006. To foster the improvement and widespread adoption of organic farming practices. Sa nta Cruz, CA. [Online] available at http://www.ofrf.org (verified 6 June 2006). Ortega, J.A., L.E. Sollenberger, K.H. Quesenberry, J.A. Cornell, and C.S. Jones. 1992. Productivity and persistence of rhizoma p eanut pastures under different grazing managements. Agron. J. 84:799-804. Paterson, E. E.A. Rattray, and K. K illham. 1996. Effect of elevated CO2 concentration on C-partitioning and rhizosphere C-flow for three plant species. Soil Biol. Biochem. 28:195-201. Perez-Fernandez, M.A., E. Calvo-Magro, J. Montanero, and J.A. Oyola. 2006. Seed germination in response to chemicals: Eff ect of nitrogen and pH in the media. J. Environ. Biol. 27:13-20. Perez-Nieto, J., E. Valdes-Velarde, M. E. Hernandez-San Roman, and V. OrdazChaparro. 2005. Rain, runoff, and soil er osion in shaded coffee agroforestry systems. Agrociencia 39:409-418. Perin, A., R.H. Silva, S. Urquiaga, J.G. Marinho, and P.R. Cecon. 2004. Producao of fitomass, acumulo de nutrients e fixacao bi ologica de nitrogenio por adubos verdes em cultivo isolado e consorciado. Pesq. Agrop. 39:35-40. Pool, R.M., R.M. Dunst, and A.N. Laks o. 1990. Comparison of sod, mulch, cultivation, and herbicide floor management practic es for grape production in northern vineyards. J. Am. Soc. Hortic. Sci. 115:872-877. Prine, G.M., L.S. Dunavin, J.E. Moore, and R.D. Roush. 1981. ‘Florigraze’ rhizoma peanut-a perennial forage legume. Fl a. Agric. Exp. Stn. Circ. S-275. IFAS, University of Florida.Gainesville, FL. Prine, G.M., L.S. Dunavin, J.E. Moore, and R.D. Roush. 1986. Registration of ‘Florigraze’ rhizoma peanut. Crop Sci. 26:1084-1085. Putnam, A. 1988. Allelochemicals from plants as herbicides. W eed Technol. 2:510-518. Qi, A., T.R. Wheeler, J.D. Keatinge, R.H. Ellis, R.J. Summerfield, and P.Q. Craufurd. 1999. Modeling the effects of temperature on the rates of seed ling emergence and leaf appearance in legume cove r crops. Exp. Agric. 35:327-344. Quinones, A., J. Banuls, E.P. Millo, and F. Legaz. 2003. Effects of 15N application frequency on nitrogen uptake efficiency in citrus trees. J. Plant Phys. 106:14291434. Radosevich, S. 1995. Weed Ecol ogy: implications for management. J. Wiley. New York, NY.

PAGE 201

185 Ram, B., G.R. Chaudhary, A.S. Jat, and M. L. Jat. 2003. Effect of integrated weed management and intercropping systems on growth and yield of pearl millet ( Pennisetum glaucum ). Indian J. Agron. 48:254-258. Ramos, M.G., M.A. Villatoro, S. Urqui aga, B.J. Alves, and R.M. Boddey. 2001. Quantification of the contribution of biol ogical nitrogen fixation to tropical green manure crops and the residual benefit to a subsequent maize crop using 15N-isotope techniques. J. Biotechnol. 91:105-115. Reberg-Horton, S.C., J.D. Burton, D.A. Dane hower, G. Ma, D.W. Monks, J.P. Murphy, N.N. Ranells, J.D. Williamson, and N.G. Creamer. 2005. Changes over time in the allelochemical content of ten cultivars of rye ( Secale cereale L.). J. Chem. Ecol. 31:179-193. Reddy, K., A.R. Soffes, and G.M. Prine. 1986a. Tropical legumes for green manure. I Nitrogen production and the effects on succeeding crop yields. Agron. J. 78:1-4. Reddy, K., A.R. Soffes, G.M. Prine, and R.A. Dunn. 1986b. Tropical legumes for green manure. II Nematode populations and th eir effects on succeeding crop yields. Agron. J. 78:5-10. Reeves, D.W. 1994. Cover crops rotations. p125-172. In Crops Residue Management, J.L. Hatfield and B.A. Stewart (eds.) Advances in soil science, crops residue management. Lewis Publishers, Boca Raton, FL. Reganold, J. P., J.D. Glover, P.K. Andrews, and H.R. Hinman. 2001. Sustainability of three apple production syst ems. Nature 410:926-929. Rice, R., L.E. Sollenberger, K.H. Quese nberry, G.M. Prine, and E.C. French. 1996. Establishment of rhizoma perennial peanut with va ried rhizoma nitrogen and carbohydrate concentrations. Agron. J. 88:61-66. Roder, W., S. Maniphone, and B. Keoboulapha. 1998. Pigeon pea for fallow improvement in slash-burn systems in th e hills of Laos? Agroforest. Syst. 39:45-57. Roe, N.E., P.J. Stofella, and H.H. Brya n. 1994. Growth and yield of bell pepper and winter squash grown with organic and livi ng mulches. J. Amer. Soc. Hortic. Sci. 119:1193-1199. Ross, S.M., J.R. King, R.C. Izaurralde, and J.T. OÂ’Donovan. 2001. Weed suppression by seven clover species. Agron. J. 93:820-827. Rouse, R.E., and J.J. Mullahey. 1997. Perennial peanut groundcover in citrus orchard row middles and discussion of potential environm ental benefits. Proc. Fla. St. Hortic. Soc. 110:79-82. Rouse, R.E., R.M. Muchovej, and J.J. Mullahey. 2001. Guide to using perennial peanut as a cover crop in citrus. EDIS HS-805. University of Florida. Gainesville, FL.

PAGE 202

186 Ruiz, T.M., R. Ramos-Santana, and A. Soto mayor-Rios. 2000. Establishment of rhizoma perennial peanut ( Arachis glabrata ) under irrigation at two semiarid sites in the Caribbean. J. Agric. Univ. P. R. 84:105-114 Sainju, U.M., T. Terril, S. Gelaye, a nd B. Singh. 2003. Soil aggregation and carbon and nitrogen pools under rh izoma peanut and perennial w eeds. Soil Sci. Soc. Am. J. 67:146-155. Saldivar, A.J., W.R. Ocumpaugh, R.R. Gilder sleeve, and G.M. Prine. 1992a. Total nonstructural carbohydrates a nd nitrogen of Florigraze rhizoma peanut. Agron. J. 84:439-444. Saldivar, A.J., W.R. Ocumpaugh, R.R. Gildersleeve, and G.M. Prine. 1992b. Growth analysis of ‘Florigraze’ rhizoma peanu t: shoot and rhizoma dry matter production. Agron. J. 84:444-449. Saldivar, A.J., W.R. Ocumpaugh, R.R. Gild ersleeve, and J.E. Moore. 1990. Growth analysis of ‘Florigraze’ rh izoma peanut: forage nutritive value. Agron. J. 82:473477. SAS Institute Inc. 2002. The SAS system fo r windows 9.0. SAS Inst. Inc., Cary, NC. Scholberg, J.M.S., L.R. Parsons, T.A. Wheat on, B.L. McNeal, and K. T. Morgan. 2002. Soil temperature, nitrogen concentration, and residence time affect nitrogen uptake efficiency in citrus. J. Environ. Qual. 31:759-768. Scholberg, J.M.S., L.R. Parsons, T.A. Wheat on, and K. T. Morgan. 2001. Procedures for determining the effects of environmental conditions on plant n itrogen uptake: An alternative approach. Soil Crop Sci. Soc. Fla Proc. 60:40-49. Schomberg, H., and D.M. Endale. 2004. Cove r crop effects on nitr ogen mineralization and availability in conservation tillage cotton. Biol. Fertil. Soils 40:398-405. Scientific Congress on Organic Agricultur e Research (SCOAR). 2003. NOP issues for research and extension. [Online] available at http://cari.unl.edu/2003abstractpr esentation/Porter%20Proceedings.pdf (verified 1 March 2006). Shrestha, A., S.Z. Knezevic, R.C. Roy, B.R. Ball-Coelho, and C.J. Swanton. 2002. Effect of tillage, cover crop and crop rotation on the composition of weed flora in a sandy soil. Weed Res. 42:76-87. Singh, B., and K. Usha. 2003. Nodulation and sy mbiotic fixation of cowpea genotypes as affected by fertilizer nitroge n. J. Plant Nutr. 26:463-473. Sistani, K.R., G.E. Brink, S.L. McGo wen, D.E. Rowe, and J.L. Oldham. 2003. Characterization of broiler cake and br oiler litter, the by-products of two management practices. Biores. Technol. 90:27-32.

PAGE 203

187 Slattery, J.F., D.R. Coventry, and W.J. Sl attery. 2001. Rhizobial ecology as affected by the soil environment. Aust. J. Exp. Agric. 41:289-298. Sooby, J. 2003. State of the States. Organic farming systems research at land grant institutions 2001~2003. 2nd ed. A publication of Or ganic Farming Research Foundation (OFRF), Santa Cruz, CA. Staples, T.E., K.C. Van Rees, and C. Van Kessel. 1999. Nitrogen competition using 15N between early successional plants and planted white spruce seedlings. Can. J. For. Res. 29: 1282-1289. Steinmaier, N., and A. Ngoliya. 2001. Potentia l of pasture legumes in low-external-input and sustainable agriculture (LEISA): I. Results from green manure research in Luapula Province, Zambia. Exp. Agric. 37:297-307. Steinmaus, S.J., T.S. Prather, and J.S. Holt. 2000. Estimation of base temperature for nine weed species. J. Exp. Bot. 343:275-286. Stork, P.R., and P.H. Jerie. 2003. Initial studies of the growth, nitr ogen sequestering, and de-watering potential of perenni al grass selections for use as nitrogen catch crops in orchards. Aus. J. Agric. Res. 54:27-37. Sustainable Agriculture Research and Edu cation Program (SARE) 2006. United States Department of Agriculture (USDA). [Online] available at http://wsare.usu.edu/about/index.cfm?sub=hist_concept (verified 6 June 2006). Swisher, M.E., P. Monaghan, and J. Fergus on. 1994. A profile of FloridaÂ’s commercial organic citrus growers. EES-108, Florid a Cooperative Extension Service, IFAS, University of Florida, Gainesville, FL. Syvertsen, J.P., and M.L. Smith. 1996. Nitroge n uptake efficiency and leaching losses from lysimeter-grown citrus trees fertilized at three nitrogen rates. J. Amer. Soc. HortScience. 121:57-62. Taiz, L., and E. Zieger. 2002. Plant physio logy. Sinauer Associates, Inc., Sunderland, MA. Tamado, T., and P. Milberg. 2004. Control of parthenium ( Parthenium hysterophorus ) in grain sorghum ( Sorghum bicolor ) in the smallholder farming system in eastern Ethiopia. Weed Technol. 18:100-105. Teasdale, J.R., T.E. Devine, J.A. Mosjidis, R.R. Bellinder, and C.E. Beste. 2004. Growth and development of hairy vetch cultivars in the northern United States as influenced by planting and harvesting date. Agron. J. 96:1266-1271. Tegg, R.S., and P.A. Lane. 2004. A comparison of the performance and growth of a range of turfgrass species under shade. Aust. J. Exp. Agric. 44:353-358.

PAGE 204

188 Terril, T.H., S. Gelaye, S. Mahotiere, E.A. Amoah, S. Miller, R.N. Gates, and W.R. Windham. 1996. Rhizoma peanut and alfalfa productivity and nutrient composition in central Georgia. Agron. J. 88:485-488. Thomas, W.E., I.C. Burke, J.E. Spear s, and J.W. Wilcut. 2006. Influence of environmental factors on slender amaranth ( Amaranthus viridis ) germination. Weed Sci. 54:316-320. Tian, G., and B.T. Kang. 1998. Effects of so il fertility and fertilizer application on biomass and chemical composition of leguminous cover crops. Nut. Cycl. Agroecosyst. 51:231-238. Tian, G., B.T. Kang, G.O. Kolawole, P. Idinoba, and F.K. Salako. 2005. Long-term effects of fallow systems and lengths on crop production and soil fertility maintenance in West Africa. Nut. Cycl. Agroecosys. 71:139-150. Tian, G., G.O. Kolawole, B.T. Kang, and G. Kirchhof. 2000. Nitrogen fertilizer replacement indexes of legumes cover crops in the derived savanna of West Africa. Plant Soil 224:287-296. Tian, G., G.O. Kolawole, F.K. Salako, and B.T. Kang. 1999. An improved cover cropfallow system for sustainable management of low activity clay soils of the tropics. Soil Sci. 164:671-682. Troiano, J., and C. Garretson. 1998. Movement of simazine in runoff water from citrus orchard row middles as affected by mech anical incorporati on. J. Environ. Qual. 27:488-494. Tucker, D.P., A.K. Alva, L.K. Jackson, and T.A. Wheaton (eds.). 1995. Nutrition of Florida citrus trees. SP-169. Uni v. of Florida, Gainesville, FL. United States Department of AgricultureAgricultural Marketing Service [USDA-AMS]. 2000. The National Organic Program. Program of Standards. [Online] available at http://www.ams.usda.gov/nop/NOP/standards/FullText.pdf (Verified 6-1-06). Valencia, E., A. Sotomayor-Rios, and S. Torres-Cardona. 1997. Establishment and effect of cutting interval on yield and nutritive value of rhizoma perennial peanut in northwestern Puerto Rico. J. Agric. Univ. P. R. 81:19-30. Valencia, E., M.J. Williams, C.C. Chase, L.E. Sollenberger, A.C. Hammond, R.S. Kalmbacher, and W.E. Kunkle. 1999. Manage ment effects on herbage yield and botanical composition of rhizoma peanut -mixed grass associations. Agron. J. 91:431-438. Van Auken, O.W. 1994. Changes in compe tition between a C-4 grass and a woody legume with differential herbi vory. SouthWest. Nat. 39:114-121.

PAGE 205

189 Van Auken, O.W., and J.K. Bush. 1990. Influe nce of light levels, soil nutrients, and competition on seedling growth of Baccharis neglecta (Asteracea). Bull. Torrey Bot. Club 117:438-444. Venuto, B.C., W. Elkins, and D. Redfearn. 2000. Soil fertility eff ects on growth and nutrient uptake of rhizoma pea nut. J. Plant Nutr. 23:231-241. Vos, J., and P. van der Putten. 2001. Field observations on nitrogen catch crops. III. Transfer of nitrogen to the succeeding main crop. Plant Soil 236:263-273. Vyn, T. J., G. Faber, K.J. Janovicek, and E.G. Beauchamp. 2000. Cover crop effects on nitrogen availability to corn following wheat. Agron. J. 92:915-924. Wagger, M.G., M. L. Cabrera, and N.N. Ranells. 1998. Nitrogen a nd carbon cycling in relation to cover crop re sidue quality. J. Soil Water Conserv. 214:210-218. Walkley, A., and I.A. Black. 1934. An examina tion of Degtjareff method for determining soil organic matter and a proposed modifi cation of the chromic titration method. Soil Sci. 37:29-38. Wang, K.H., R. McSorley, A. Marshall, and R.N. Gallaher. 2006. Influence of organic Crotalaria juncea hay and ammonium nitrate fertilizers on soil nematode communities. Appl. Soil Ecol. 31:186-198. Wardowski, W., J. Whigham, W. Grierson, a nd J. Soule. 1995. Quality tests for Florida citrus. Bul. SP99. University of Florida. Gainesville, FL. Waring, S.A., and T.A. Gibson. 1994. The soil fertility effects of leguminous ley pastures in northeast Thailand .2 Effects on soil phys ical and chemical parameters. Field Crops Res. 39:129-137. Weinbaum, S.A., M.L. Merwin, and T.T. Muraoka. 1978. Seasonal variation in nitrate uptake efficiency and distribution of abso rbed nitrogen in non-bearing prune trees. J. Amer. Soc. Hortic. Sci. 103:516-519. Weston, L. 1996. Utilization of allelopathy fo r weed management in agroecosystems. Agron. J. 88:860-866. Whitworth, J. 1995. The ability of some c over crops to suppress common weeds of strawberry fields. J. Sust. Agric. 7:137-145. Whitney, J.D., T.A. Wheaton, W.S. Castle and D.P Tucker. 1994. Optimizing orange grove factors for fruit production an d harvesting. Trans. ASAE 37:365-371. Wiedenfeld, B., L.B. Fenn, S. Miyamoto, D. Swietlik, and C. Marlene. 1999. Using sod to manage nitrogen in orchard floors. Commun. Soil Sci. Plant Anal. 30:353-363.

PAGE 206

190 Williams, M.J. 1993. Planting date and preplant tillage effects on emergence and survival of rhizoma perennial peanut. Crop Sci. 33:132-136. Williams, M.J. 1994a. Growth characteristics of rhizoma peanut and nitrogen-fertilized bahia grass swards. Agron. J. 86:819-823. Williams, M.J. 1994b. Reproductive-resource allo cation in rhizoma peanut. Crop Sci. 34: 477-482. Williams, M.J., C.A. Kelly-Begazo, R.L. Stanle y, K. H. Quesenberry, and G.M. Prine. 1997. Establishment of rhizoma peanut: inte raction of cultivar, planting date, and location on emergence and rate of cover. Agron. J. 89:981-987. Williams, M.J., E. Valencia, and L.E. Sollenberger. 2002. No-till establishment of rhizoma peanut. Agron. J. 94:1350-1354. Woodward, K.R., E.C. French, L.A. Sweat, D. A. Graetz, L.E. Sollenberger, B. Macoon, K.M. Portier, B.L. Wade, S.J. Rymph, G.M. Prine, and H.H. Van Horn. 2002. Nitrogen removal and nitrate leaching for forage systems receiving dairy effluent. J. Environ. Qual. 31: 1980-1992. Wortmann, C.S., B.D. McIntyre, and C.K. Kaizzi. 2000. Annual soil improving legumes: agronomic effectiveness, nutrient uptake, nitrogen fixation and water use. Field Crops Res. 68:75-83. Wright, G.C., W.B. McCloskey, and K. C. Taylor. 2003. Managing orchard floor vegetation in flood-irrigated citr us groves. HorTechnology 13:668-677. Yamanishi, O.K., and K. Hasegawa. 1995. Trunk strangulation responses to the detrimental effect of heavy shade on fruit size and quality of to sa-buntan pomelo. J. HortScience. 70:875-887. Yao, Q., H.H. Zhu, J.Z. Chen, and P. Chri stie. 2005. Influence of an arbuscular mycorrhizal fungus on competition for phos phorus between sweet oranges and a leguminous herb. J. Plant Nutr. 28:2179-2192. Zekri, M., and Obreza, T. 2003. Macronutrien ts deficiencies in citrus: Nitrogen, phosphorus, and potassium. [Online] available at http://edis.ifas.ufl.edu/pdffiles/SS/SS42000.pdf (verified 6 June 2006).

PAGE 207

191 BIOGRAPHICAL SKETCH Jose Linares was born in San Cristobal, Ven ezuela. He received a bachelor’s degree in agronomy at “La Universidad del Tachir a” and a Master of Science degree in agronomy at “La Universidad Central de Ve nezuela” in Maracay, Ve nezuela. Since his early career, he has been interested in agroecology and sustainable agriculture. He worked in the Ecological Center for High Lands in the Andes in Venezuela from 1989 to 1992 on enhancing the sustainability of lo cal agriculture, including organic farming systems, and participated in extension programs with local small farmers. In 1992, he joined the University of Tachira (“La Univ ersidad del Tachira”) as a lecturer in a program focusing on agroecology, crop nutriti on, small farm operations, and improved use of local resources. Jose plans to return to his university in T achira and continuing his teaching responsibilities and a trans-disc iplinary research program pertaining to agroecology (cover crops) In this manner, he aims to pa rticipate in the training of local students and also assist small farmers in ma king better and more sustainable use of local resources.