Citation
Silvopasture as an Approach to Enhancing Phosphorus and Nitrate Retention in Pasturelands of Florida

Material Information

Title:
Silvopasture as an Approach to Enhancing Phosphorus and Nitrate Retention in Pasturelands of Florida
Creator:
MICHEL, GERARD-ALAIN
Copyright Date:
2008

Subjects

Subjects / Keywords:
Agriculture ( jstor )
Agroforestry ( jstor )
Manatees ( jstor )
Nutrients ( jstor )
Pastures ( jstor )
Phosphorus ( jstor )
Soil profiles ( jstor )
Soil water ( jstor )
Soils ( jstor )
Spodosols ( jstor )
Town of Suwannee ( local )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Gerard-Alain Michel. 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:
3/1/2007
Resource Identifier:
649814564 ( OCLC )

Downloads

This item has the following downloads:

michel_g ( .pdf )

michel_g_Page_028.txt

michel_g_Page_078.txt

michel_g_Page_009.txt

michel_g_Page_080.txt

michel_g_Page_114.txt

michel_g_Page_088.txt

michel_g_Page_003.txt

michel_g_Page_052.txt

michel_g_Page_103.txt

michel_g_Page_019.txt

michel_g_Page_122.txt

michel_g_Page_058.txt

michel_g_Page_031.txt

michel_g_Page_025.txt

michel_g_Page_076.txt

michel_g_Page_059.txt

michel_g_Page_001.txt

michel_g_Page_066.txt

michel_g_Page_060.txt

michel_g_Page_110.txt

michel_g_Page_096.txt

michel_g_Page_107.txt

michel_g_Page_087.txt

michel_g_Page_117.txt

michel_g_Page_051.txt

michel_g_Page_065.txt

michel_g_Page_015.txt

michel_g_Page_093.txt

michel_g_Page_071.txt

michel_g_Page_048.txt

michel_g_Page_127.txt

michel_g_Page_101.txt

michel_g_Page_016.txt

michel_g_Page_083.txt

michel_g_Page_020.txt

michel_g_Page_090.txt

michel_g_Page_026.txt

michel_g_Page_095.txt

michel_g_Page_007.txt

michel_g_Page_044.txt

michel_g_Page_017.txt

michel_g_Page_030.txt

michel_g_Page_082.txt

michel_g_Page_011.txt

michel_g_Page_062.txt

michel_g_Page_102.txt

michel_g_Page_112.txt

michel_g_Page_074.txt

michel_g_Page_099.txt

michel_g_Page_055.txt

michel_g_Page_047.txt

michel_g_Page_036.txt

michel_g_Page_077.txt

michel_g_Page_070.txt

michel_g_Page_033.txt

michel_g_Page_075.txt

michel_g_Page_081.txt

michel_g_Page_069.txt

michel_g_Page_124.txt

michel_g_Page_050.txt

michel_g_Page_072.txt

michel_g_Page_040.txt

michel_g_Page_106.txt

michel_g_Page_097.txt

michel_g_Page_089.txt

michel_g_Page_006.txt

michel_g_Page_091.txt

michel_g_Page_002.txt

michel_g_Page_049.txt

michel_g_Page_013.txt

michel_g_Page_084.txt

michel_g_Page_109.txt

michel_g_Page_008.txt

michel_g_Page_046.txt

michel_g_Page_118.txt

michel_g_Page_022.txt

michel_g_Page_023.txt

michel_g_Page_113.txt

michel_g_Page_027.txt

michel_g_Page_092.txt

michel_g_Page_125.txt

michel_g_Page_029.txt

michel_g_Page_057.txt

michel_g_Page_120.txt

michel_g_Page_054.txt

michel_g_Page_100.txt

michel_g_Page_067.txt

michel_g_Page_012.txt

michel_g_Page_004.txt

michel_g_Page_037.txt

michel_g_Page_119.txt

michel_g_Page_063.txt

michel_g_Page_039.txt

michel_g_Page_108.txt

michel_g_Page_126.txt

michel_g_Page_032.txt

michel_g_Page_038.txt

michel_g_Page_035.txt

michel_g_Page_115.txt

michel_g_Page_021.txt

michel_g_Page_045.txt

michel_g_Page_034.txt

michel_g_Page_086.txt

michel_g_Page_111.txt

michel_g_Page_010.txt

michel_g_Page_128.txt

michel_g_Page_116.txt

michel_g_Page_061.txt

michel_g_Page_085.txt

michel_g_Page_105.txt

michel_g_Page_123.txt

michel_g_Page_042.txt

michel_g_Page_043.txt

michel_g_Page_014.txt

michel_g_Page_018.txt

michel_g_Page_041.txt

michel_g_Page_064.txt

michel_g_Page_098.txt

michel_g_Page_130.txt

michel_g_Page_121.txt

michel_g_Page_073.txt

michel_g_Page_068.txt

michel_g_Page_005.txt

michel_g_pdf.txt

michel_g_Page_024.txt

michel_g_Page_079.txt

michel_g_Page_094.txt

michel_g_Page_129.txt

michel_g_Page_053.txt

michel_g_Page_104.txt

michel_g_Page_056.txt


Full Text











SILVOPASTURE AS AN APPROACH TO ENHANCING PHOSPHORUS AND
NITRATE RETENTION IN PASTURELANDS OF FLORIDA














By

GERARD-ALAIN MICHEL


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

Gerard-Alain Michel



































To my beloved wife Daphnee B. Michel, my precious son Alain-Christian Michel, my
adored mother Germaine D. Michel and to the loving memory of my dear father J.B.
Gerard Michel.
















ACKNOWLEDGMENTS

I am sincerely grateful to my maj or professor, Dr. Ramachandran P.K. Nair, for

accepting me as his student and providing me valuable guidance and academic support. I

am obliged to Dr. Vimala Nair, my co-advisor, for her important assistance and advice

during the course of my research. My sincere appreciation goes to Dr. Jerry Sartain for

his encouragement and dedicated help with the statistical analysis of my data. I would

like to thank Dr. Johannes Scholberg and Dr. Michael Bannister for their encouragement

and advice during my PhD studies.

Special thanks are extended to the study participants, Fred Clark (Alachua), Rudy

Garber (Manatee), Bruce Goff (Suwannee), and Harris Hill (Osceola), without whom this

study would have not been possible.

My appreciation also goes to several friends and collaborators for their assistance at

various stages of this study, Dawn Lucas, Dr. Samuel Allen, Dr. Myrlene Chrysostome,

the staff/students of the Soils Lab (SWSD, UF), Williams Reve (SWSD, UF), Solomon

Haile (SFRC, UF), and Brian Roth (SFRC, UF).

None of this would have been possible without the moral support of my beloved

wife, Daphnee B. Michel, my precious son Alain-Christian, and my dear mother,

Germaine Delaquis Michel. My appreciation goes to my two sisters, Joyce Michel

Dalpuppo and Abilene Michel Celestin, my uncle Jean-Claude Michel, and my godfather

Ronald Maximilien, for their encouragement throughout my graduate studies. Finally, I

wish to express my gratitude to the team of the Center for Subtropical Agroforestry.









I would like to acknowledge USDA IFAFS/CSREES (Initiative for Future

Agricultural and Food Systems grant through the Center for Subtropical Agroforestry

(CSTAF), University of Florida) for supporting this study.




















TABLE OF CONTENTS


page

ACKNOWLEDGMENT S ................. .............._ iv..._... ....

LI ST OF T ABLE S .........._.... .............._ ix...___. ....


LI ST OF FIGURE S .............. .................... xi


AB S TRAC T ......_ ................. ..........._..._ xiii..


CHAPTER

1 INTRODUCTION ................. ...............1.......... ......


Background ................. ...............1.......... ......
Problem Statement ................. ...............3.................
Organization of the Dissertation ................. ...............4................
Hypothesis and Obj ectives ................. ...............5.......... .....

2 REVIEW OF LITERATURE ................. ...............6.......... .....


Agroforestry and Silvopasture ............... .. ... ......... .. ........ ...... ... .. ......6
Historical Development of Agroforestry in the U. S. and Florida ................... ......6
Silvopasture: History, Characteristics and Potential .............. ....................1
Environmental Quality and Land Use Systems ................ ................ ......... .12
Florida Soils and their Management ................. ...............12........... ...
Fertilizer Use and Non-Point Source Pollution ................. ................. ...._14
Soil Phosphorus as a Pollutant .............. ...............15....
Nutrient Management in Silvopasture .....___.....__.___ .......____ ............1
Nutrient Cycling in Agroforestry Systems ................ ................ ......... .16
Soil Phosphorus in Pasturelands ........_................. ............_ ........1
New Concepts of P Management ................. ........._._ .......19..... ...

3 SOIL CHARACTERISTICS OF THE SILVOPSTURE AND TREELESS
PASTURE STUDY SITE S............... ...............26.


Introducti on ............ ..... .._ ...............26...
M materials and M ethods .............. ...............26....
Location of the Study Sites............... .... .. ... ... ..... ... .......2
Estimation of Nitrogen and Phosphorus Loads in the Pasture Soils ...................34












Maj or Tree and Grass Species at the Study Sites ................ .......................34
Soils of the Study Sites ................. ...............36...............
Soil sam pling ............... .... ...............36.......... ......
Soil-nutrient characterization .....___.....__.___ .......____ .............3
Soil texture and pH determination .............. ...............38....
Statistical Procedures............... ...............3
Results and Discussion .............. ...............39....
Soil Reaction (pH) ........._...... .... .. ......... .. ...............39...
Soil Particle Size Analysis (Soil Texture) .....__.___ ..... .._. __ ........_........40
Mehlich-1 Al and Fe Concentrations .............. ...............41....
Mehlich-1 Mg and Ca Concentrations .............. ...............43....
Conclusions............... ..............4


4 PHOSPHORUS AND NITROGEN CONCENTRATIONS UNDER A SLASH
PINE (Pinus elliottii) + BAHIAGRASS (Paspalum notatum) SILVOPASTURE
AND A TREELESS PASTURE............... ...............54


Introducti on ............ ..... .._ ...............54...
M materials and M ethods .............. ...............56....
Site Description .............. ...............56....
Statistical Procedures............... ..... .. ...............5
Water Soluble P and Mehlich 1-P Determination ......____ ...... ....__..........57
Nitrate and Ammonium-N Determination............... .............5
Results and Discussion ............... ...............58....
Water Soluble P Concentrations............... .............5
Mehlich 1-P Concentrations ................. ...............59................
Nitrate and Ammonium-N Concentrations .............. ...............61....
Conclusions............... ..............6


5 SOIL PHOSPHORUS STORAGE CAPACITY UNDER A SLASH PINE (Pinus
elliottii) + BAHIAGRASS (Paspalum notatum) SILVOPASTURE AND A
TREELESS PASTURE .............. ...............74....


Introducti on ............ ......_ ...............74....
Material and Methods ............ ..... ._ ...............77...
Site Description .............. ...............77....
Statistical Procedures............... ...............7
Calculations of Parameters ............... ......................._ .. ..... .....7
Soil P storage capacity calculated with Mehlich-1 ................ ....................77
Soil P storage capacity calculated with Single Point Isotherm Data............78
Results and Discussion .............. ...............80....

Phosphorus Saturation Ratio .............. ...............80....
Soil Phosphorus Storage Capacity............... ....... ...... ......... .......8
Estimation of Mehlich-1-based Soil P Storage Capacity (SPSC-M1) and Soil
P Storage Capacity-Single Point Isotherm (SPSC-ISO) ................. ...............82
Conclusions.............. .............84












6 SUMMARY AND CONCLUSIONS ................ ...............92................


APPENDIX: REGRESSION ANALYSIS .............. ...............95....


LIST OF REFERENCE S ................. ...............97................


BIOGRAPHICAL SKETCH ................. ...............116......... ......




































































V111

















LIST OF TABLES


Table pg

2-1 Agroforestry practices in the Southeastern U.S. ............. ...............9.....

2-2 Water soluble phosphorus (WSP) concentrations as affected by soil depth for a
native flatwoods site, a silvopastoral site, and an adjacent treeless pasture at
Ona, Florida, USA.CI ............ ...............23.....

3-1 Soil and climatic characteristics of the four farms selected for the study in four
counties of Florida............... ...............3

3-2 Description of pasture systems at the four study locations in Florida .................. ...3 1

3-3 Estimates of nutrient loads from beef cattle excreta at the four study sites .............34

3-4 Statistical model used in the studyc'................ ...............39..............

3-5 Mean values of Mehlich-1 Ca, Al, Mg, and Fe concentrations, pH, and percent
sand, silt and clay for soil profies at the center of the alley of the silvopasture
and a treeless pasture sites in Alachua Florida, USA ................. ........__. ........46

3-6 Mean values of Mehlich-1 Ca, Al, Mg, and Fe concentrations, pH, and percent
sand, silt and clay for soil profies at the center of the alley of the silvopasture
and a treeless pasture sites in Osceola Florida, USA. ............. .....................4

3-7 Mean values of Mehlich-1 Ca, Al, Mg, and Fe concentrations, pH, and percent
sand, silt and clay for soil profies at the center of the alley of the silvopasture
and a treeless pasture sites in Suwannee Florida, USA ................. ..........__ .....48

3-8 Mean values of Mehlich-1 Ca, Al, Mg, and Fe concentrations, pH, and percent
sand, silt and clay for soil profies at the center of the alley of the silvopasture
and a treeless pasture sites in Manatee Florida, USA. ............. .....................4

3-9 Mean values of Mehlich-1 Al and Fe concentrations and percent sand, silt and
clay for soil profies of the Ultisols and the Spodosols..........._.._.._ ......._.._.. .....50

4-1 Mean values of water soluble P (WSP), Mehlich-1 P (M-1 P), and nitrate and
ammonium-N concentrations for different soil layers of soil profies at the center
of the alley of the silvopasture and a treeless pasture sites in Alachua and
Suwannee Florida, USA. .............. ...............63....










4-2 Mean values of water soluble P (WSP), Mehlich-1 P (M1-P), nitrate, ammonium
concentrations of different soil layers for soil profiles at the center of the alley of
the silvopasture and a treeless pasture sites in Manatee and Osceola, Florida,
U S A .............. ...............64....

4-3 Mean values of Mehlich-1 extractable-P (M 1-P) concentrations for soil profiles
of the Spodosols (Osceola and Manatee) and Ultisols (Alachua and Suwannee)
in Florida. ............. ...............65.....

5-1 Mean values of phosphorus saturation ratio (PSR) and soil P storage capacity
(SPSC) for soil profiles at the center of the alley of the silvopasture and a
treeless pasture sites in Alachua, Suwannee, Manatee and Osceola, Florida, USA 85

5-2 Mean values of phosphorus saturation ratio (PSR) for soil profiles of the
Spodosols (Osceoala and Manatee) and Ultisols (Alachua and Suwannee) in
Florida. ............. ...............86.....

5-3 Cumulative soil P storage capacity (SPSC) for the 0-1 m soil depth at the center
of the alley of the silvopasture and a treeless pasture sites in Alachua,
Suwannee, Manatee and Osceola, Florida, USA. ............. .....................8

















LIST OF FIGURES


Figure pg

1-1 U. S. consumption of nitrogen and phosphate for 1960-2003. ............. ........._......2

3-1 Soil orders of Florida .............. ...............27....

3-2 Location of study sites in Florida. The Osceola and Manatee sites are on
Spodosols and the Alachua and Suwannee sites are on Ultisols. .............................29

3-3 Silvopastoral system at Osceola County, Florida............... ...............32

3-4 Treeless pasture at Manatee County, Florida ................. ........._.. ...._.._.. ..32

3-5 Profile of an Ultisol. ............. ...............33.....

3-6 Profile of a Spodosol. ............. ...............33.....

3-7 Grid sampling at the Suwannee location. Twenty four soil profiles were
collected at each field (silvopasture and treeless pasture) at each location, and
with 6 depths (0 5, 5 15, 15 30, 30 50, and 50 75 and 75 -100 cm) per
profile. The small circles, representing sampling points, illustrate the layout of
the grid sampling method used in each sites at each location. ................ ...............37

3-8 Aerial photo of the Alachua location ................. ...............51........... ..

3-9 Aerial photo of the Osceola location ................. ...............52........... ..

3 -10 Aerial photo of the Manatee location ................. ...............53.............

4-1 Water soluble P (WSP) concentration at different soil layers for silvopasture and
treeless pasture at the Alachua and Suwannee field sites featuring Spodosols.
Error bars indicate standard errors (n = 24). ............. ...............66.....

4-2 Water soluble P (WSP) concentration by depth in the silvopasture and treeless
pasture sites in Osceola and Manatee. Error bars designate standard errors............ 67

4-3 Mehlich-1 P concentration by depth in the silvopasture and treeless pasture sites
in Alachua and Suwanneee. Error bars designate standard errors. ..........................68

4-4 Mehlich-1 P concentration by depth in the silvopasture and treeless pasture sites
in Osceola and Manatee. Error bars designate standard errors. ............. ................69










4-5 Nitrate concentration level by depth in silvopasture and treeless pasture sites in
Alachua and Suwannee on Ultisols. Error bars designate standard errors. ..............70

4-6 Ammonium-N concentrations levels by depth in silvopasture and treeless
pasture sites in Alachua and Suwannee (Ultisols sites). Error bars designate
standard errors. ............. ...............71.....

4-7 Nitrate-N concentration for different soil depth for silvopasture and treeless
pasture systems for filed sites in Manatee and Osceola (Spodosols). Error bars
designate standard errors. .............. ...............72....

4-8 Ammonium-N concentrations levels for different soil depth in silvopasture and
treeless pasture sites in Manatee and Osceola on Spodosols. Error bars designate
standard errors ................ ...............73.................

5-1 Determination of P sorption capacity using a single point isotherm based on a
procedure by Nair et al. (1998). ............. ...............79.....

5-2 Relationship between phosphorus saturation ratio (PSR) and Water soluble P
(WSP) for surface soils (0-5 and 5-15 cm) of the study sites in Florida..................88

5-3 General trend in soil P storage capacity (SPSC) in the center of the alley of the
silvopasture and a treeless pasture sites in Florida ................. .......................89

5-4 Soil P storage capacity (SPSC) by depth at the center of the alley of the
silvopastures and treeless pasture sites in Osceola (error bars represent + one
standard error from the mean, n=24) ................. ...............90...............

5-5 Relationship between soil P storage capacity -single point isotherm (SPSC-ISO)
and soil P storage capacity-Mehlich-1 (SPSC-M1) for the surface (0 5 and 5 -
15 cm depths) of the study sites in Florida............... ...............91
















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

SILVOPASTURE AS AN APPROACH TO ENHANCING PHOSPHORUS AND
NITRATE RETENTION IN PASTURELANDS OF FLORIDA

By

Gerard-Alain Michel

December 2006

Chair: Ramachandran P.K. Nair
Cochair: Vimala D. Nair
Major Department: Forest Resources and Conservation

Phosphorus (P) and nitrate-nitrogen (N) loss from sandy soils that are predominant

in the 1.4 million ha of pastureland in Florida is a maj or cause of water pollution. We

hypothesized that soil P and N loss would be lower from silvopastoral systems than from

treeless pastures because soil P and N removal by the combined stand of trees and pasture

would be more than that of treeless pasture. Four slash pine (Pinus elliottii) + bahiagrass

(Paspalum notatum) silvopastoral systems located respectively in Osceola, Manatee,

Suwannee, and Alachua counties in Florida were selected. The Manatee and Osceola

soils are Spodosols, and the Alachua and Suwannee soils are Ultisols. Soil samples were

collected depth-wise from 0 to 100 cm. Soil P storage capacity (SPSC), the maximum

amount of P that can be safely applied to a soil before it becomes an environmental

concern, was calculated. Water-soluble P concentrations in the 0 5 cm soil layer were 4

and 10 mg kg-l for the silvopasture and treeless pasture in Suwannee, 11 and 23 mg kg-l

in Manatee, 7 and 11 mg kg-l in Osceola and in 6 and 11 mg kg-l in Alachua. Total SPSC









in the upper one meter depth were 342 and -60 kg ha-l respectively in the silvopasture and

treeless pasture in Suwannee. The corresponding values were 329 and 191 kg ha-l in

Manatee, 657 and 926 kg ha-l in Osceola, and -36 and -542 kg ha-l in Alachua. The

results suggest that P buildup within the soil profile was less in silvopastures than in

treeless pastures. Therefore silvopasture systems can be expected to provide a greater

environmental service in regard to water quality protection compared to treeless pastures

under comparable ecological settings. In a related investigation, the single point isotherm

method seemed to better estimate SPSC compared to the Mehlich-1 method. Although N

followed similar trends differences between the two pasture systems were less

conspicuous. This may be related to the difficulties in monitoring N, which is more labile

than P; the highly dynamic soil nitrogen levels might have confounded the results.















CHAPTER 1
INTTRODUCTION

Background

Phosphorus (P) and nitrogen (N) are critical to all forms of plant life. Nitrogen is an

essential component of basic biological building blocks such as amino acids and thus

their polymers: peptides and proteins. Nitrate-N, a negatively charged ion, is not

adsorbed by the negatively charged colloids that dominate most soils (Brady and Weil,

2002). As a result, it is susceptible to off-site losses via deep leaching (the downward

movement of nutrient below the effective root zone with percolating water) or in runoff

(overland and lateral subsurface flows). Nitrates (as in fertilizers) are highly soluble in

water and therefore move readily via leaching or runoff into the groundwater below.

According to Yaday (1997), 68% of NO3-N accumulation occurred outside the crop-

rooting zone and 20% of nitrate-N accumulation in the crop-root zone in the soil profile

moved into groundwater annually. Nitrate concentrations above 10 mg L1 in drinking

water may lead to the "blue baby syndrome" (United States Environmental Protection

Agency [USEPA], 2006).

Phosphorus is involved in many vital functions in plants such as energy storage and

transfer, photosynthesis, respiration, protein and nucleic acid synthesis, and ion transport

across cell membranes. Fertilization of crops represents the largest proportion of P used

in agriculture (Wood, 1998). Phosphorus and nitrogen fertilizers consumption in the U.S.

for 1960 to 2003 is shown in Figure 1-1(United States Department of

Agriculture/Economic Research Service [USDA/ERS], 2006a).














14,000.0

12,000.0

S10,000.0
g -*- Nitrogen
S8,000.0 -1 -=- Phosphate

8 6,000.0

S4,000.0

ti 2,000.0

0.0


Year (ended June 30)


Figure 1-1. U. S. consumption of nitrogen and phosphate for 1960-2003. Sources:
USDA/ERS (U. S. Department of Agriculture / Economic Research Service), 2006a.

In areas of intensive crop and livestock production, continual P applications as

mineral fertilizer and manure in excess of crop removal rates often result in P

accumulation (Sharpley et al., 1994; Tiessen, 1995). Phosphorus exists in agricultural

fields either in a dissolved form or as attached to soil particles. Phosphorus is often the

most limiting nutrient in surface water systems, and even small additions of P can have a

great influence on primary production in these waters (Correll, 1998). Concern about the

environmental fate of P has arisen from the evidence that land-applied animal wastes and

fertilizers can potentially promote eutrophication of water bodies (Allen, 1987;

Breeuwsma and Silva, 1992; Lemunyon and Daniels, 1997; Correll, 1998; Parry, 1998).

Phosphorus concentration above 0.02 mg L1 in lake water may result in increased alagal

growth in P-limited systems. Excessive algae growth will decrease dissolved oxygen










levels and reduced water transparency. This process is referred to as eutrophication

(Sharpley et al., 1999), and is directly associated with nutrient enrichment of surface

waters. Thus, even if P losses from agriculture were small from an agronomic point of

view, they may still constitute a serious potential environmental problem. Reducing

excess loadings of P and NO3-N into both surface- and ground-water is an important

environmental goal in the southeastern USA.

Problem Statement

Improved nutrient management strategies are required to maximize plant uptake of

nutrients and thus minimize potential losses by runoff and leaching in order to maintain

surface and ground water quality. The role of agroforestry systems, especially the effect

of trees within agroecosystems, on nutrient retention and/or soil storage is therefore of

interest. Due to their deep rooting systems, trees may provide a mechanism for improved

nutrient capture, by which tree-based systems act as a "safety net" for capturing nitrates

and other nutrients that would otherwise leach below the root zone of agronomic crops

(van Noordwijk et al., 1996; Rowe et al., 1999). Although this has been demonstrated in

tropical agroforestry systems (Seyfried and Rao, 1991; Horst 1995; Young, 1997;

Lehmann et al., 1999; Rowe et al. 1999; van Noordwijk et al., 1999; Rowe et al., 2005),

such nutrient movement studies in temperate agroforestry systems are rare, and whatever

little that has been done has focused on riparian buffers. Nevertheless, these few studies

have demonstrated the effectiveness of riparian buffer systems in reducing non-point

source pollution and thereby improving water quality (Schultz et al. 2000; Schultz et al.

2004). Little information is available on the environmental implications of other

agroforestry practices such as silvopasture in regard to their impact on surface and

groundwater quality.









Silvopasture, in the North American and other temperate-zone context, refers to

intentional combination of trees, forage plants and livestock into an integrated,

intensively-managed system (Nair, 1993; Garrett et al., 2000; Nair et al., 2004); trees in

such systems are managed for high-value sawlogs and at the same time an annual income

is generated from livestock grazing. Silvopasture is considered to be the most promising

agroforestry practice in the southeastern United States (Workman and Allen, 2004).

According to Zinkhan and Mercer (1997), silvopasture is the most common form of

agroforestry in the southern United States. Studies from across the southern pine region

(Louisiana, Mississippi, and Georgia) report the possibility of productive livestock

grazing while maintaining, or even improving, high value timber production (Workman

and Allen, 2004).

With the realization about its environmental amelioration potential and greater

appreciation of the societal value of ecosystem benefits such as carbon sequestration and

also providing unique wildlife habitats, the practice of silvopasture is now receiving

increased interest as an innovative and viable strategy for agroecosystem management

(Nair et al., 2005). Based on this, the focus of this study will be on the role of

silvopastoral systems in reducing P and N losses from the soil.

Organization of the Dissertation

The research is presented in five chapters. Chapter 2 is a review of the current

literature on agroforestry and silvopasture in the U.S. and Florida. This chapter outlines

the problems associated with land-use changes and their environmental impacts in

Florida, and discusses the development of new concepts of P management. Chapter 3

presents the main characteristics of the study sites in Florida in terms of the soil chemical

and physical characteristics as well as the plant species found in the two pasture systems










(treeless pasture and silvopasture). It also includes a detailed account of the soil sampling

methodology used in the study. Chapter 4 contains the results of the attempt to test the

safety net hypothesis by measuring the soil phosphorus and nitrate-N concentrations

throughout the soil profile for both the silvopasture and the treeless pasture sites at four

locations in Florida and for two representative soil orders (Spodosols and Ultisols). An

analysis of the "remaining soil P storage capacity" of the study sites at the four locations

with comparison between the treeless pastures and the silvopastures in the two soil orders

is the subj ect of Chapter 5. The final chapter (Chapter 6) includes the synthesis and

conclusion of the results and is followed by recommendations for future research needs.

Hypothesis and Objectives

The overall hypothesis of the study is that the presence of trees in silvopastoral

systems would enhance soil nutrient retention and/or crop removal of P and nitrates

compared to treeless pastures and thereby reduce potential loadings rates of surface

waters .

The study obj ectives are to

(i.) Quantify water soluble P (WSP), Mehlich-1 P, and NO3- and NH4' in the surface
and subsoil horizons in manure-impacted soil profiles of silvopastoral and treeless
pasture systems at four different locations (two soil orders) in Florida;

(ii.) Determine the relation between WSP and P saturation ratio (PSRhlu) in the surface
and subsoil horizons of the four study sites;

(iii.) Estimate the remaining soil P storage capacity (SPSC) of the two pasture systems
sites at the four locations;

(iv.) Assess differences in SPSC estimates by comparing two methods (Mehlich-1 and
single-point isotherm).















CHAPTER 2
REVIEW OF LITERATURE

Agroforestry and Silvopasture

Historical Development of Agroforestry in the U.S. and Florida

Drastic changes have occurred in land-cover use during the past 150 years in the

Eastern United States. Forests, woodlands and grasslands were severely affected by land

clearing for timber exploitation and agriculture from the middle 1800s to the 1930s

(Ware, 2002). Between 1973 and 2000, land cover statistics showed that forests remained

the predominant cover, despite the decline from a high of 28.5 % in 1973 to a low of 24.3

% in 2000. Agriculture saw a similar decline from 10.8 % to 8.0 % during the same

period (U.S Geological Survey [USGS], 2006).

Today, crop, pasture lands and woodlands represent 3.9, 4.0, and 3.4 millions

hectares, respectively, in Florida, Georgia, and Alabama (USDA/ERS, 2006b).

Agriculture and forestry combined represent the most important economic activity in the

states of Alabama and Georgia and the second (only to tourism) in Florida. In spite of this

dominant role of agriculture and forestry, substantial population growth has occurred,

causing an expansion of urban and developed land. Between 1970 and 2000, the

population of Florida increased by 235 percent, from 6.8 million to 16.0 million people

(NPG, 2006). At present, the State of Florida has over 17 million residents (Florida

Division of Emergency Management [FDEM], 2006). The urbanization of rural lands and

the subsequent increase in land prices constitute a striking testimony of the development

of suburbs and cities across rural areas of America. The main consequence of









urbanization is the diversion of food producing lands or lands reserved for forests and

natural habitats, to accommodate expanding populations (Cordell et al., 1998).

Agricultural intensification of croplands, forest and livestock systems has been very

articulated during recent decades. The mid-1900s brought about an unprecedented

increase in food and fiber yields per unit area resulting from a steady increase in the use

of chemical fertilizer, pesticides and irrigation while a decline in the rate of land clearing

was observed (Workman and Allen, 2004). By the 1970s, row crops generated production

surplus and higher incomes for farmers. During the 1980s, intense mechanization and

farm inputs augmented the number of farm loans, which led to an increase in farm debt

and eventual large-scale loss of farms to foreclosures (Fitchen, 1991). Today, excessive

input costs and low profit margins resulting from strong competition from corporate-run

farms and inexpensive foreign products are driving small farms to abandon farming

activities (Workman and Allen, 2004). Concerns regarding the effects of large-scale

transformation of ecosystems on the natural structure and function of land and water

resources have been raised and continue to be important research and development issues

(Lappe et al., 1998; Vitousek et al., 1997; Sampson and Hair, 1990; Savory, 1988).

A decline in industrial timberland acreage of more than 0.5 million hectares has

been observed in Florida and Georgia, since 1989 (Workman and Allen, 2004). Much of

this timberland is now owned by private corporations that are expected to carry on a

management plan for wood products (Conner and Hartsell, 2002). As urbanization

follows its course and an increasing number of people desire recreations in natural

settings, there will be a growing need for trees both within and outside forests (Long and

Nair, 1999; Leakey, 1998).










With an increase in population growth rate in the southern states and a decline in

industrial timberland, there have been growing concerns about the environmental

problems associated with potential fertilizers and pesticides pollution, soil erosion, pest

problems, and loss of biological diversity. Consequently, there is increased interest in the

use and adoption of land-management practices that increase both the aesthetic and

recreational value of lands while protecting and conserving the natural resource base

(Bliss et al., 1997; Teasley et al., 1997). Agroforestry, the intentional growing of trees

with crops, pasture and/or livestock, affords farmers with an alternative land-use practice

with potential for mitigating some of the environmental and economic setbacks

associated with modern agriculture (Nair, 1993). Agroforestry is now receiving increased

attention from researchers, landowners, government and private agencies in North

America (World Agroforestry Centre [ICRAF], 2000; Lassoie and Buck, 2000; Garrett et

al., 2000).

An essential characteristic of agroforestry is the wide array of land-management

options it offers for preserving natural resources and generating additional income.

Agroforestry systems, via the integration of trees with crops and/or livestock on the same

land, could provide numerous environmental services. These services include reduction

of soil erosion and nutrient losses, enhancement of inherent soil fertility, water infiltration

and groundwater recharge, protection against wind, snow, noise, and odor and other

nuisances, and creation of attractive and healthier landscapes (Ewel, 1999; Jordan, 1998,

Daily, 1997; Leakey, 1996). By diversifying the farm enterprises, agroforestry can also

help landowners to improve the resilience of their system and to minimize risks

associated with income loss from price fluctuation, crop failure or other unexpected





























































Source: AFTA (2006)


problems. Opportunities for intensifying the use of agroforestry practices, and the

benefits that result, are gradually rising in the southern U.S. (Workman and Allen, 2004).

Agroforestry in the Southeastern U. S. can be divided into five maj or practices or

land-use techniques (Merwin, 1997; Garrett et al., 2000; Nair, 2001): Alley Cropping,

Forest Farming, Riparian Forest Buffers, Silvopasture, and Windbreaks (Table 2-1).

Table 2-1. Agroforestry practices in the Southeastern U.S.
Silvopasture Silvopasture is the intentional combination of trees, forage plants
and livestock together as an integrated, intensively-managed
system. Silvopasture can provide profitable opportunities for
softwood or hardwood timber growers, forest landowners, and
livestock producers.


Alley cropping







Riparian Forest
Buffers





Forest Farming








Windbreaks


Alley cropping is the cultivation of food, forage or specialty crops
between rows of trees. It is a larger version of intercropping or
companion planting conducted over a longer time scale. Alley
cropping can provide profitable opportunities for row crop
farmers, hardwood timber growers, nut growers and Christmas tree
growers.

Riparian forest buffers are strips of trees, shrubs and grass planted
between cropland or pasture and surface water courses. Buffers
protect water quality, reduce erosion and flooding. Riparian forest
buffers can provide beneficial opportunities for row crop farmers,
ranchers, horticulturists, and dairy and livestock producers.

Forest farming is the intentional cultivation of edible, medicinal or
decorative specialty crops beneath native or planted woodlands
that are managed for both wood and understory crop production. It
does not include the gathering of naturally-occurring plants from
native forests, also known as wilderafting. Forest farming can
provide profitable opportunities for forest and woodland owners,
nut growers, sugar maple growers, and herb growers.

Windbreaks are linear plantings of trees and shrubs designed to
enhance crop production, protect people and livestock, and benefit
soil and water conservation. Windbreaks can provide valuable
opportunities for vine and tree fruit growers, row crop farmers,
livestock producers, and rural homeowners.









Silvopasture: History, Characteristics and Potential

Grazing in pine (Pinus spp.) forests in the Lower Costal Southern United States

dates back to around 1520, when cattle were first introduced in Florida by Ponce de Leon

from Spain (Lewis, 1983). In the early 1950s, pine trees were first planted in improved

pastures as part of the Conservation Reserve Soil Bank Program (Nowak and Long,

2003). At present silvopasture is becoming increasingly attractive to non-industrial

private forest landowners and livestock operators in southeastern United States who want

to diversify their enterprise (Nair et al., 2005).

Silvopasture has historically included use of shade trees in pasture grazed orchards

or woodlands, and as rangelands that include a managed tree or shrub component (Clason

and Sharrow, 2000; Robinson and Clason, 1997; Williams et al., 1997). In the Southeast,

silvopasture has generally included forest grazing with cattle, such as flatwoods

rangeland (Pearson, 1997), pine managed for turpentine and sawlogs mixed with forage

systems (Byrd et al., 1984; Cary, 1928), and tree pasture practices with pecan (Carya

illinoensis K. Koch) (Reid, 1991). Silvopastoral systems integrating tree, forage and

livestock production are the most common agroforestry practices found in the United

States and Canada with the largest area of grazed forests occurring in the southern and

southeastern United States (Clason and Sharrow, 2000). Burton and Scarfe (1991)

suggested associations with goats for both meat production and vegetation management.

According to Lewis and co-workers (Hart et al., 1970) the association of southern pine

production and beef cattle with improved pastures constitutes an opportunity for the

production of multiple products. Increased economic benefits may result from the

integration of forestry with ranching by helping buffer year-to-year income fluctuation

through the sale of forest products and hunting leases generated by the creation of









wildlife habitat. Studies on silvopasture in the Southeastern United States have reported

the economical viability of such enterprise (Lundgren et al., 1983; Clason, 1995; Grado et

al., 2001; Stainback and Alavalapati, 2004; Alavalapati et al., 2004). According to

Stainback et al. (2004), the combination of slash pine (Pinus elliotii) with cattle

production could become competitive with conventional ranching in Florida when

environmental costs and benefits are considered in the analysis.

Research on warm season forage under pines in South Georgia (that began in the

1940s) showed that Pensacola bahiagrass (Paspalum notatum) as the most shade tolerant

of the 23 grasses studied (Lewis and Pearson, 1987; Pearson, 1975). Several legume

species have been recognized as having promising potential for production under partial

shade (McGraw et al., 2001). In a silvopasture study comparing a double row system of

pines (3.1m between rows and 1.2 m between trees and 12.2 m wide alleys) to a single

row system (3.1 m x 3.7 m), the double row system yielded more forage and as much

wood (Lewis et al., 1985). At present the double row system remains the most popular

spacing for silvopasture across the region (Clason and Sharrow, 2000). Recently

developed varieties of bahiagrass, such as Tifton-9 and Argentine may perform as better

warm season forage component in silvopasture systems (Nowak and Blount, 2002).

Clason (1995) reported higher internal rate of return in silvopastoral practice (13%) than

managed timber (9%) or open pasture (6%) in the state of Louisiana. Grado et al. (2001)

reported that both silvopasture and pasture had positive cash flows with that for pasture

being highest during a short evaluation period. In Georgia, pine growth improved with

controlled grazing (Lewis et al., 1985), and loblolly-cattle-forage systems in the Coastal

Plain may have a 70% greater net present value per unit area (Dangerfield and Harwell,










1990) than a pure forestry operation It has been demonstrated in Alabama that when

mimosa (Albizia julibrissin) and leucaena (Leucaena leucocephala) are grown without

fertilizers, they can be cut for fodder at 6 to 8-week intervals (Bransby et al., 1996). A

livestock-forage-Chri stmas tree production system could be another viable silvopasture

mixture (Pearson et al., 1990). According to McGowan et al. (1999), there is also a

growing interest in goat production in this region. In the South, forest industry has used

goats as an alternative to using chemical or mechanical means of weed control in pine

plantations (Solaiman and Hill, 1991). Studies conducted in Arkansas and Alabama

suggest that goats can play a key role in vegetation reduction, especially kudzu (Pueraria

lobata), during site preparation for pine plantations (Pearson and Martin, 1991; Bonsi et

al., 1991).

Environmental Quality and Land Use Systems

Florida Soils and their Management

Florida soils are characterized by coarse soil textures and acidic conditions. P loading of

surface and ground waters following transport from dairy operations and heavily

manured pasturelands in these porous soils by P-rich animal manures constitute a serious

environmental concern .The Florida Phosphorus Index, a field-based indexing system

developed to rank the vulnerability level of farmlands as sources of P loss in surface

runoff water (University of Florida/Institute of Food and Agricultural Sciences

[UF/IFAS], 2006), take into account both phosphorus transport characteristics and

management practices. Management factors include soil test P, P application method, and

source and rate of P application (Nair and Graetz, 2004). Phosphorus based manure

applications are being increasingly incorporated into farm nutrient planning program

(Sharpley et al., 2004) because nitrogen-based manure applications have resulted in









continuous P-accumulation in soils. According to Sharpley et al. (2004), the planning

includes "the selection, timing, and implementation of source and transport best

management practices at Hield, farm, and watershed scales".

Some of the practices used for controlling erosion and runoff (the transport

factor) include, conservation tillage, contour farming, constructed wetlands, and

vegetative fi1ter strips. Conservation tillage is a management practice that provides

efficient and effective erosion control and it can improve soil properties and soil quality

(United States Department of Agriculture/Natural Resources Conservation Service

[USDA/NRCS], 1999).This practice can improve soil structure thereby resulting in

higher infiltration and percolation rates, increased water holding capacity and, enhanced

resistance to erosion. Vegetative fi1ter stripes are narrow rows of permanent vegetation

(usually grasses or shrubs) planted on the contour can be used to slow down runoff, trap

sediment, and eventually build up natural or living" terraces (Brady and Weil, 2002).

Constructed wetlands (CW) have been used for decades for the treatment of domestic

(Kadlec, 1985) and industrial wastewaters and agricultural runoff (Higgins et al., 1993).

These systems are useful for reducing suspended solids, ammonium-N, P, biochemical

oxygen demand and chemical oxygen demand in sewage (Tanner et al., 1995 and Von

Felde and Kunst, 1997).

Measures aimed at regulating the P accumulation involve attaining a suitable

balance between P imports and exports. Suitable management practices may include,

composting, improved livestock feeding, suitable rate, method, and timing of land

application of manure, along with physical and chemical treatments of manure

(Chrysostome, 2005). Drinking water treatment residuals (WTRs), non-hazardous









materials that are obtained at no costs from drinking-water treatment plants (Makris et al.,

2005), offer a cost-effective chemical soil amendment of P-impacted soils. These

materials reduce soluble P in soils with low P-sorption capacities, without affecting soil

fertility (Codling et al., 2002). Similar findings have also been reported by Novak and

Watts (2005).

Fertilizer Use and Non-Point Source Pollution

The U.S. Environmental Protection Agency (EPA, 1996) has identified agricultural

non-point source pollution as the maj or cause of stream and lake contamination that

prevents the attainment of the water quality goals established in the Clean Water Act. The

southeastern U. S. is home to a vast number of water bodies; Florida alone has over 7,800

lakes, about 10,000 km2 (4,000 sq. miles) of estuaries, and over 80,000 km (50,000

miles) of rivers and streams, including maj or water bodies such as Lake George, Lake

Okeechobee, the Everglades, and the St. Johns and Kissimmee rivers (Florida

Department of Environmental Protection [FDEP], 2003). In addition, the Floridian

aquifer system, underlying almost all of Florida and portions of Alabama, Georgia and

South Carolina, occupies a total area of about 260,000 km2 (100,000 sq. miles), and

supplies over 1 1.4 billion liters (3 billion gallons) of water per day for all uses throughout

the region (Johnson and Bush, 2002). There is increased awareness of non-point source

pollution of waters from agricultural chemicals in drainage and runoff water (Tiessen,

1995). This issue is particularly important in Florida and surrounding areas, as the karst

geography of the region is shaped by vast groundwater reserves that are sensitive to

nutrient build-up, and has led to a growing demand for information regarding the impact

of different land-use systems on surface water quality. Most of the common nitrogen and

phosphorus fertilizers are water-soluble. In the high rainfall areas, especially on sandy









soils, these nutrients can be leached into the groundwater and washed into waterways,

even after short periods (months and years, rather than decades) (Summers, 2002). In the

U. S., only 18% of the N-fertilizer inputs leaves farms with the produced crop, meaning

that on average, 174 kg ha-l of surplus N is left behind in croplands each year (Ecological

Society of America [ESA], 1998). Animal manures and fertilizers, pose a serious threat to

surface and groundwater quality in maj or river (Suwannee River and Lake Okeechobee)

basins of Florida. Elevated NO3-N concentrations have been measured in water from both

domestic and monitoring wells in the Suwannee River Basin (Katz et al., 1999),

exceeding the maximum contaminant level of 10 mg L^1 set by the U.S. Environmental

Protection Agency. Nitrate-N concentrations above this limit can cause

methemoglobinemia or "blue baby" syndrome in infants and elderly adults (Mueller and

Helsel, 1996). These increased nitrate concentrations in these waters have been attributed

to dairy wastes (Andrews 1994), poultry operations (Hatzell, 1995), and excessively high

fertilizer applications (Katz et al., 1999).

Soil Phosphorus as a Pollutant

In most U.S. watersheds, P export occurs mainly in surface runoff rather than in

subsurface flow. On the other hand, in some regions, particularly the Gulf coastal plains

and Florida, as well as in fields with subsurface drains, P can be transported in drainage

waters (Sharpley et al., 1999). Soils with high water infiltration rates and low nutrient

retention capacity, such as sandy soils and soils with low activity clays and low organic

matter content, are particularly prone to nutrient leaching (Lehmann and Schroth, 2003).

The Spodosols of Florida receiving significant loadings of animal manure, on the other

hand, are more prone to subsurface leaching of P (Nair and Graetz, 2002). The P in

organic manures has been shown to be more susceptible to leaching than that in inorganic









fertilizers (Eghball et al., 1996), and P leaching from grazed grassland tends to increase

when the stocking density is high, possibly due to a greater amount of soluble organic P

being released from dung (Beauchemin et al., 1999). The inj section of manure slurry also

greatly increases the leaching of P (Tunney et al., 1997). Phosphorus transported from

agricultural soils can accelerate eutrophication (Sharpley et al., 1994). This phenomenon

leads to increased costs and difficulty of water purification for drinking purposes (to

remove odor, turbidity, and color) and the replacement of high quality edible fish,

submerged macrophytic vegetation, and benthic organisms with coarse, rapid-growing

fish and algae, along with the control of proliferating noxious aquatic plants.

Nutrient Management in Silvopasture

Nutrient Cycling in Agroforestry Systems

Research conducted in tropical agroforestry systems has demonstrated the potential

for nutrients cycling by deep-rooting trees. According to Kang (1997), the presence of

woody species in an alley cropping production system has been shown "to contribute to

(1) nutrient recycling, (2) reduction in soil nutrient leaching losses, (3) stimulation of

higher soil faunal activities, (4) soil erosion control, (5) soil fertility improvement and (6)

sustained levels of crop production". Horst (1995) reported lower nitrate concentrations

in the soil solution and consequently reduced leaching under hedgerow intercropping

with Leucaena leucocephala and annual food crops than in the agricultural control

treatments in southern Benin. Lehmann et al. (1999) measured lower nutrient leaching

under an Acacia saligna- sorghum (Sorghum bicolor) intercrop than under pure sorghum

with runoff irrigation in northern Kenya. Das and Chaturved (2005) studied nutrient

concentrations in plant and soil and their corresponding cycling rates of cycling in poplar

(Populus deltoides)-based agroforestry systems at Pusa, Bihar, India. The authors









concluded that the poplar-based agroforestry system was more sustainable than other

single-crop systems in terms of in terms of soil nutrient status. A study conducted by

Tapia-Coral et al. (2005) in Central Amazonia, Brazil showed that the litter of the

agroforestry systems had lower carbon: nutrient ratios than the litter in the secondary

forest control, indicating a faster nutrient recycling in the agroforestry systems. Manezes

et al. (2002) found that the preservation of native trees or introduction of exotic tree

species in Cenchrus ciliarisiiiiiiiii~~~~~~~ pastures in semiarid northeastern Brazil significantly affects

microclimate and the dynamics of litter and soil nutrients, and may contribute to

enhanced nutrient cycling in these systems.

Other studies conducted in temperate settings have also shown similar results. In a

pecan (Carya illinoensis)-cotton (Gossypium hirsutum) alley cropping system in

northwestern Florida, Allen et al. (2004) found that N-leaching rates were lower in a

'non-barrier treatment' (competition treatment) compared to a 'barrier' (competition-

reduced treatment) treatment. Cumulative nitrate leaching below the 0.3 and 09 m depth

were 64 and 13 kg ha-l for non-barrier treatment, compared to 122 and 46 kg ha-l for

respective depths in barrier treatment. It appears that the tree roots captured N in the non-

barrier treatment, resulting in lower leaching rates below the root zone of the cotton crop.

Furthermore, tree water uptake, in addition to cotton water uptake in the non-barrier

treatment, may have decreased overall water drainage in comparison to the barrier

treatment, thereby influencing nutrient loadings rates. In the same study Allen et al.

(2004) reported a greater nitrification and mineralization rates in no-barrier treatment

than in a barrier treatment due to higher soil N, resulting from reduced nutrient uptake by

cotton. Udawatta et al. (2002) found in a paired watershed study conducted in









northeastern Missouri, consisting of agroforestry (trees plus grass buffer strips), contour

strips (grass buffer strips), and control treatments with a comn (Zea mays)-soybean

[Glycine max] rotation, that nitrate-N loss was reduced 24 and 37% by contour strip and

agroforestry treatments. Corresponding treatments also reduced total P loss by 8 and 17%

on contour strip and agroforestry watersheds.

Soil Phosphorus in Pasturelands

Pasturelands in Florida receiving significant load of manure are subj ect to

subsurface leaching of phosphorus due to the sandy nature of the soils. Many P

compounds present in manure are water soluble to begin with and have a potential to

move with rain or irrigation water (Gerritse, 1977). Furthermore, animals are known to

utilize P inefficiently in feed (only 30% is retained); therefore most of the P entering

livestock operations end up in manure (Sharpley and Moyer, 2000). In many areas of

concentrated animal production, manures are generally applied based on the crop N-

requirements to avoid groundwater quality problems associated with leaching of excess

N. This frequently leads to P build up in excess of the sufficiency level for optimal crop

yields, which in tumn can enhance the potential P runoff and leaching (Haygarth et al.,

1998; Heckrath et al., 1995; Sharpley et al., 1996). With time, the organic P will be

absorbed or hydrolyzed to form inorganic phosphate-containing compounds, which may

be subj ect to leaching and run-off as soluble P compounds (Zublena, 1995). Hedley et al.

(1982) found that the maj or portion of manure P was soluble in weak extractants such as

water and bicarbonate. Sharpley and Moyer (2000) assessed phosphorus forms in manure

for dairy, poultry and swine and P leaching potential using a simulated rainfall. Inorganic

P in the manures accounted for 63 to 92% of total P. Phosphorus leaching from the

different manures by simulated rainfall showed a close relationship to water soluble P









concentration. According to Sharpley et al. (2004), there is a general shift in soil P

chemistry with the application of manure from Fe-P and Al-P to Ca-P compounds,

resulting in a relatively greater extraction of P by the Mehlich-3 but a lower water

extractability of soil P. This shift has important environmental implications for soil P

testing. For example, the fact that Mehlich-3 P has been shown to overestimate potential

for P-losses in surface flow from heavily manured soils could be explained by the

extraction of Ca-P minerals that are insoluble in water.

New Concepts of P Management

Soil test P (STP) is the most widely used means to estimate the potential risk for P

leaching fom the soil (Maguire and Sims, 2002). Diagnostic soil test P values were

initially developed to evaluate the probability of obtaining a positive crop response to the

addition of a P fertilizer. However, it was not intended to be used as an environmental

indicator of the potential risk of P pollution. According to Sharpley et al. (1999),

additional calibration (beyond that conducted for fertility recommendations) are required

in order to also validate soil-test P values as an environmental indicator. Calibration

studies have resulted in development of environmentally-based threshold STP

concentrations and have been established for several regions. Sharpley and Tunney

(2000) have presented an array of examples of the use of such tests as guides for P

management recommendations to protect water quality. Heckrath et al. (1995) found that

P leaching losses were low when P extractable in 0.5 M sodium bicarbonate (Olsen-P)

was less than 60 mg kg-l and then increased rapidly when Olsen-P exceeded 60 mg kg-l

They referred to this value as a change-point, above which the potential for P loss

through leaching increased considerably. However, the change-point value was found to

be affected by experimental conditions, such as the range of STP values tested and the









soil to solution ratio for the P desorption parameter (Koopmans et al., 2002). Boirling

(2003) presented evidence that even if a change-point can be identified, it would be site-

specific. Hesketh and Brookes (2000) determined change-points in seven soils and found

variation from 10 to 119 mg kgl Olsen-P.

Water soluble P is another environmental soil testing parameter for assessing potential P

losses. Phosphorus extracted from soils using distilled deionized water (WSP) is expected

to simulate the rapid release of P to water better than a stronger chemical extractant (such

as agronomic soil P tests), since rainfall is very similar to distilled water (Moore et al.,

1998). This extraction should also maintain the soil pH within one unit of its original

value, which is a desirable attribute since P solubility is highly dependent upon soil pH

(Golterman, 1998; Sharpley, 1993). Pote et al., (1996) demonstrated, in a field study with

tall fescue the existence of a strong correlation between water soluble P with runoff P.

The concept of Degree of Phosphorus Saturation (DPS), was developed in the

Netherlands (van der Zee et al., 1987; Breeuwsma and Silva, 1992) and later applied in

other parts of the world has been used to evaluate the risk of P leaching from soils as

follows:


DPS = SreP x100
P sorption capacity

The DPS estimates how close the soil is to saturation and facilitates comparison

among soils with varying P sorption capacities. Different methods of estimating sorbed P

and P sorption capacity can be used to determine DPS. In the Netherlands and several

other countries, P extracted with ammonium oxalate has been used to estimate sorbed P,

and soil concentrations of both Fe and Al in order to estimate P sorption capacity

(Boirling, 2003) as follows:










DPS = Pox x00
(Feox +Alox)R

where a is an empirically derived sorption coefficient which has a value between 0 and 1.

This coefficient is derived from the following relationship

P sorption capacity = (Feox + Alox )a

A value for a of 0.5 was recommended by Beauchemin and Simard (1999); Breeuwsma

and Silva, 1992; Koopmans et al., 2003; Schoumans and Groenendijk, 2000 and Sims et

al., 2002 while Nair and Graetz (2002) reported a value of 0.55 for Spodosols in Florida.

The DPS normalizes P (oxalate, Mehlich-1 or -3) to extractable Fe and Al, and has been

related regionally to soil solution P concentration (Nair and Harris, 2004). Hooda et al.

(2000) suggested that the DPS, which relates ammonium oxalate-extractable P to the sum

of oxalate extractable Fe and Al (DPSox), is a good indicator of the soil's potential to

release P. Other researchers have suggested that DPSox can be a suitable tool for

predicting subsurface P lost since it has been shown to be closely correlated to P

concentrations in leachate waters (Leinweber et al., 1999 and Maguire and Sims, 2002).

A soil is P-saturated when 25% or more of PSC (Phosphorus Sorption Capacity)

has been used. This level of P sorption saturation will likely result in ground water P

concentrations above the 0. 1 mg P L standard used in the Netherlands (Breeuwsma et

al., 1995). Other studies conducted by Maguire et al. (1998) and McDowell and Sharpley

(2001) have shown a low potential for P losses via leaching when DPS is <20 to 25% and

sharp increase in P leaching above this DPS value. More recently, Nair and Graetz

(2002) showed that DPShln can be used as indicator of soluble P for both surface A

horizons and subsurface Bh (spodic horizons) of Spodosols in the Lake Okeechobee

basin. Nair et al. (2004) taking into consideration the change points, confidence intervals,









agronomic soil test values, and DPS values from other studies, recommended a threshold

DPShll of 30% for Florida sands. Values for DPSmlof 31 to 60% warrant caution with

regard to further addition of P to a land-use system, while DPShll values of > 60% imply

that the can potentially impair water quality. The Phosphorus Saturation Ratio (PSR) is

similar to DPS, but does not use a corrective constant (a) associated with the original

definition ofDPS (Breeuwma and Silva, 1992).

When used as environmental indicators of off-farm P loss, both STP and DPS

failed to provide a clear indication of the capacity of a soil to retain added P. Therefore,

soils with very low STP values may nevertheless have insufficient capacity to retain

added P, and soils with higher STP values may have capacity to retain additional P. Nair

and Harris (2004) recommended the use of a DPS-based calculation of the remaining soil

P storage capacity (SPSC) that consider risks arising from previous loading as well as

inherently low P sorption capacity. The SPSC of a soil is based on a threshold PSR of

0.15: SPSC = (0.15 PSR) x (Mehlich-1 Al + Mehlich-1 Fe) [Nair and Harris, 2004]

where (Mehlich-1 Al + Mehlich-1 Fe) is proportional to the P sorption capacity of a

given soil (van der Zee et al., 1987). According to Nair et al. (2004), the value of 0. 15

used in the SPSC calculation is the best approximation available to date of the PSR value

corresponding to the critical P solution concentration of 0. 10 mg litre-l P as proposed by

Breeuwsma and Silva (1992). The SPSC can be calculated by horizon (expressed as kg P

m-3) USing the bulk density and the thickness of that horizon or by depth (expressed as mg

P kg-l or kg P ha- ) [Nair and Harris, 2004]. The SPSC can be used for assessing the soil

status in regards to future environmental risk-free P additions. In other words, the SPSC









allows prediction of the amount of additional P that can be retained by a given soil

horizon or depth before the added P becomes an environmental concern.

Silvopasture and Environmental Quality

Nutrient movement studies in temperate agroforestry systems are rare and

whatever little that has been done has focused on riparian buffers. Nonetheless research

undertaken on silvopasture in Florida has shown great promise. Preliminary research

conducted on a silvopastoral site at a research farm at Ona, Florida, on flatwoods soils

(Spodosols) suggest that silvopastoral sites are less likely to retain nutrients within the

soil profile compared to an adj acent fertilized pasture with cattle grazing (Nair and

Graetz, 2004) (Table2-2).

Table 2-2. Water soluble phosphorus (WSP) concentrations as affected by soil depth for a
native flatwoods site, a silvopastoral site, and an adjacent treeless pasture at
Ona, Florida, USA.C
Depth Native flatwoods Silvopasture Pasture
cm (NSP, me kg~
0-5 1.59 (1.14)" 4.23 (1.34) 9.10 (3.26)
5-15 1.28 (1.17) 2.09 (1.36) 2.79 (2.20)
15-30 0.12 (0.02) 0.79 (0.39) 1.52 (0.94)
30-50 0.00 (0.00) 0.17 (0.14) 1.91 (1.25)
50-75 0.00 (0.00) 0.02 (0.02) 0.42 (0.40)
75-100 0.00 (0.00) 0.04 (0.03) 0.66 (0.52)
"Source: Nair and Graetz, 2004
QNumbers in parentheses are standard deviation values

In a study conducted at a University of Florida research station, Ona, Florida, for

which long term information on the management history of the sites is available, Nair and

Kalmbacher (2005), reported P concentrations in soil profiles obtained from each of the

pasture sites to be in the following order: pasture>silvopasture>native pasture. In the

same study the NH4 and NO3--N concentrations were found to be higher in the surface

horizon of the treeless pasture.









A substantial body of literature is available on the effectiveness of riparian buffers

in reducing nutrient losses from farms to water bodies, and on the potential offered by

these systems to reduce nutrient loss from agricultural lands. Hefting et al. (2005)

reported that plant production, nitrogen uptake, and N retention were significantly higher

in a forested buffer site compared to a herbaceous buffer site in riparian buffer zones in

study conducted in six European countries. Results of a field monitoring study done in

Tokachikawa watershed in Hokkaido, Japan, Cisadane, Cianten and Citamyang sub-

watersheds in Indonesia and Cauvery watershed in southern India, all demonstrated the

positive impact of forest buffer zones in reducing the influence of agricultural nutrients

and chemicals on surface stream waters (Anbumozhi et al., 2005). In a study conducted

in southern Illinois, USA, Blattel et al. (2005) found that ground water soluble P

concentrations were significantly reduced by 14 percent in the first 1.5 m of both mixed

hardwood forest and giant cane (Arundinaria gigan2tea) riparian buffers, and there was an

overall 28 percent reduction in water soluble P concentration by 12 m from the field

edge. Schoonover et al. (2005) showed that forest buffers significantly reduced incoming

dissolved NO3--N, dissolved NH4 -N, total ammonium-N, and total orthophosphate

concentrations in surface runoff within the 10.0 m riparian buffer by 97, 74, 68, and 78,

respectively. Furthermore, overall nutrient reductions for giant cane buffer strips were

100 percent for all three nutrients due to relatively high infiltration rates. Finally,

significant reductions of total NH4 N and total orthophosphate were also detected at a

distance of 3.3 m from the edge of the cane buffer compared to 6.6 m in the forest buffer.

In a study conducted in North Queen Island, Australia, McKergow et al. (2004) reported

a reduction in total N, total P and suspended sediment loads between 25 and 65% by the










planar slope grass buffer and within the first 15 m of the moderately convergent grass

buffer. Lee et al. (2003) showed, in a study in Bear Creek, Iowa, USA, that switchgrass

buffer (Panicum virgatum cy. Cave-n-Rock) removed 95% of the sediment, 80% of the

total-N, 62% of the nitrate-nitrate-N, 78% of the total-phosphorus, and 58% of the

phosphorus, whereas a switchgrass/woody buffer removed 97% of the sediment, 94% of

the total-N, 85% of the nitrate-N, 91% of the total-P, and 80% of the P in the runoff.

These findings provide clear evidence that agroforestry systems are effective in

reducing the soil P and nitrate-N store and would therefore reduce the amount of nutrients

likely to be lost from farmland. This phenomenon is of great importance because it

presents a mechanism to sustain current agricultural land use while achieving both

nutrient conservation and water quality protection.















CHAPTER 3
SOIL CHARACTERISTICS OF THE SILVOPSTURE AND TREELESS PASTURE
STUDY SITES

Introduction

A thorough understanding of the distribution of nutrients in different soil layers is

essential for developing effective nutrient management strategies that are economically

viable and environmentally sound. From that point of view, this study was undertaken to

investigate differences in soil characteristics between silvopasture and adj acent treeless

pasture systems.

Materials and Methods

Seven of the twelve soil orders are represented in Florida (Collins, 2003): Alfisols,

Entisols, Histosols, Inceptisols, Mollisols, Spodosols, and Ultisols. The sites selected for

this study represented two maj or soil orders of the state: Ultisols and Spodosols.

Spodosols are characterized by having undergone soil processes that translocate organic

matter and aluminum, or organic matter, aluminum, and iron, as amorphous materials

(Collins, 2003). Spodosols occur mostly on coarse-textured, acid parent materials subject

to ready leaching (Brady and Weil, 2002). The A and E horizons of Spodosols are sandy

and have poor sorption capacities while the Bh (spodic) and the Bw horizons have much

greater affinities for P retention (Mansell et al., 1991). According to Soil Survey Staff

(1996), Spodosols are characterized by a highly fluctuating water table situated between

the Bh and the A horizons during the summer rainy season, while it may go down to 125

cm during drier months. Due to low hydraulic conductivity of the lower soil layers,











rainfall that penetrates the soil during high water table conditions tends to move laterally,

transporting P to surface drainage ditches (Burgoa et al., 1991; Mansell et al., 1991).


Approximately 3.4 million hectares in Florida have been mapped as Spodosols (Collins,

2003).

Ultisols are characterized by the presence of an argillic horizon, enough moisture for


crops in most years, and a base saturation [proportion of chemical exchange sites on soil

particles that are occupied by basic cations (Ca2+, Mg2+, K Na )] less than 35 percent.

They developed in relatively warm and moist climates, such as northwest Florida, and

tend to be highly productive if managed properly. There are approximately 2.8 million

known hectares of Ultisols (Collins, 2003) in Florida. The distribution of Ultisols and


Spodosols in Florida is shown in Figure 3-1.


Figure 3-1. Soil orders of Florida (Source: Collins (2001)


Soil Orders
_ APlfisols
a Entisols
Histosols
Inceptisols
M~ollisols
SSpodosois
M Ultisols
SUroan Lana
M Water







50 0 50 100









Location of the Study Sites

The study was conducted on four farms that practice silvopasture in Florida: Mr.

Harris Hill's farm in Osceola County, Dr. Rudy Garber' s farm in Manatee County, the

Florida Sheriffs Youth Ranch in Suwannee County, and Mr. Fred Clark' s farm in

Alachua County (Table 3-1 and Figure3-1). The silvopastures consisted of a combination

of slash pine (Pinus elliottii) and bahiagrass (Paspalum notatum ), and the treeless

pastures consist ofbahiagrass (Figures 3-3 and 3-4). The tree planting configuration of

the silvopastures in Osceola and Manatee locations was a double row planting (3.1 m

between rows and 1.2 m between trees and 12.2 m wide alleys), whereas in Suwannee

and Alachua the configuration was a single row planting. The treeless pastures were

generally older than the silvopastures with the exception of the Suwannee location in

which both pasture systems were of the same age (Table 3-2). Generally no fertilizers

were applied to the pasture systems under study except for the treeless pasture in

Suwannee, which received a 19-5-19 formulation respectively at a rate of 336 kg ha-1

and at 224 kg ha-1 during 2003 and 2004 respectively. The soils at both Alachua and

Suwannee are Ultisols (Figure 3-5), whereas soils at Manatee and Osceola are Spodosols

(Figure 3-6). The Spodosols are poorly drained soils as compared to Ultisols that are well

drained. On average, the depth to the spodic horizon was 55 cm in the Spodosols (Table

3-1). Average annual precipitations range from 1233 to 1381 mm and the minimum water

table depth at the Osceola and Manatee field sites were 25 and 15 cm, respectively. Slope

percent vary from 0 to 8 percent at the study locations (Table 3-1).





























30 0 200 400 Kilometers


Site 3 (Suwannee)


Site 4 (Alachua)


Site 2


Us Census Bureau,
Reprojected data to Transverse Mercator,
WG S 1984, UTM Zone 17 N


Figure 3-2. Location of study sites in Florida. The Osceola and Manatee sites are on
Spodosols and the Alachua and Suwannee sites are on Ultisols.


A











Table 3 -1. Soil and climatic characteristics of the four farms selected for the study in four counties of Florida


Osceola County
280 9' N, 810 10' W
Central Florida

Immokalee: Sandy,
siliceous,
hyperthermic
Arenic Alaquods
Thick marine
deposits formed in
thick beds of
marine sands
0-2
25 102



Poorly drained


Manatee County
270 13' N, 820 8' W
Southwestern Florida

EauGallie: Sandy,
siliceous,
hyperthermic Alfic
Alaquods
Sandy and loamy
marine sediments in
Peninsula Florida

0-2
15 45



Poorly or very poorly
drained


Suwannee County
300 24' N, 83 0O' W
Florida Panhandle
Region
Blanton: Loamy,
siliceous, semiactive,
thermic Grossarenic
Paleudults
Sandy and loamy
marine or eolian
deposits


Alachua Cout
290 45' N, 820 33' W
North-central Florida

Kendrick: Loamy,
siliceous, semiactive,
hyperthermic Arenic
Paleudults
Thick beds of loamy
marine sediments


Location


Soil


Soil Series


Parent material




Slope (%)
Annual fluctuation
of the water table
depth (cm)

Drainage class




Depth to spodic
horizon (cm)
(spodosols)
Climate Avg. annual
precipitation (mm)
Avg. annual air
temperature (oC)


0-8


0-8


Somewhat
excessively to
moderately well
drained


Well drained; slow to
moderately slow
permeability


1233


1381


1366

20


1332

21


Source: Soil survey of Osceola (Readle, 1979), Sarasota (Hyde et al., 1991); Alachua (Thomas et al., 1982); USDA-NRCS (not
dated); World climate (2005); Florida Sheriffs Youth Ranch, 2003, unpublished; personal communication; and results of the study.











Table 3-2. Description of pasture systems at the four study locations in Florida
Pastoral Systems Osceola Manatee Suwannee Alachua


Silvopasture
Tree species

Pasture species

Tree planting
configuration (m)
Tree age (year)
Land area (ha)
Fertilization history
Prior land use history





Treeless pasture
Pasture species
Age (year)
Land area (ha)
Fertilization history


Slash pine
Pinus elliotii
Bahiagrass
Paspalum notatum
Double rows
3.1 x 1.2 x12.2
12
8.5
No fertilization
Florida Flatwoods
Treeless pasture for 15 years (Site
continuously fertilized and limed
when it was under treeless pasture
system)


Slash pine
Pinus elliotii
Bahiagrass
Paspalum notatum
Double rows
3.1 x 1.2 x12.2
12
20.2
No fertilization
Florida Flatwoods


Slash pine
Pinus elliotii
Bahiagrass
Paspalum notatum
Single rows
1.5 x 7.2
40
16.2
No fertilization
Agriculture






Bahiagrass
40
16.2
19-5-19 application at
336 kg ha-l in 2003
19-5-19 application at
224 kg ha in 2004
4.5 Mg hal of
dolomite applied
every 4 years since
1978
Agriculture


Slash pine
Pinus elliotii
Bahiagrass
Paspalum notatum
Single rows
1.5 x 3.0

28.3
No fertilization
Grazed naturally
occurring loblolly
pine forest with
grass


Bahiagrass
55
7.3
No fertilization








Agriculture (Corn
field)


Bahiagrass
45-50
6.1
No fertilization


Bahiagrass
30
3.0
No fertilization








Florida Flatwoods


Prior land use history Florida Flatwoods





































Figure 3-3. Yilvopastoral system at Usceola County, Florica


rigure s-4. Ireeless pasture at ivianatee county, rioriaa































Figure 3-5. Profile of an Ultisol. (Source: Collins, 2001)


: of a Spodosol. (Source: Collins, 2001)


rlgure s-0. rrc









Estimation of Nitrogen and Phosphorus Loads in the Pasture Soils

An estimation of nitrogen and phosphorus loads in the pasture sites was made

based on several assumptions (Table 3-3). According to Vicente-Chandlerl (966), a

mature cow produces 9. 1 tons (1 Mg = 1 ton = 1000 kg) of excreta yearly containing a

total of 71 kg N, 40 kg P205, and 59 kg K20. The total land grazed at the Manatee,

Suwannee, Osceola and Alachua farms were 259, 567, 445, and 259 hectares

respectively. Although grazing animals do not deposit urine and manure evenly across

the paddocks where they grazed, an even distribution of cattle manure was assumed in

this study (in the absence of a better way of presentation). Dalrymple (1994) reported

from a rotational grazing study that urine spots occupied 16.7 percent of the pasture while

manure spots occupied 18.8 percent, following 504 grazing days.

Table 3-3. Estimates of nutrient loads from beef cattle excreta at the four study sites
Location Farm size (ha) Head of Cattle Load estimates (kg ha-' yr- )
N P20s

Alachua 259 150 41 23
Suwannee 567 200 25 14
Manatee 259 150 41 23
Osceola 445 275 44 25

Phosphorus and nitrate loads from beef cattle were similar in both Alachua and the

Manatee sites (Table 3-3).The Suwannee sites had the lowest nutrient loads per unit area

than all four sites which is the result of a lower cattle density. The highest nutrient loads

per unit area were observed in the Osceola sites.

Major Tree and Grass Species at the Study Sites

Pintus elliottii (Slash pine). Slash pine is an important timber species in the

southeastern United States. Its strong, heavy wood is excellent for construction purposes.









Because of its high resin content, the wood is also used for railroad ties, poles, and piling

(Duncan and Duncan, 1988; McCune 1988; Lohrey and Kossuth, 1990)

Slash pine is a large, stately, heavily-branched, long-needled conifer whose fast

growth rate makes it capable of reaching 30.5 m in height and 0.9 to 1.2 m in diameter

(Gilman and Watson, 1994). Slash pine is self-pruning of its lower branches, it forms a

somewhat pyramidal canopy when young which evolves to an open, rounded canopy

over time. Mature trees generate a light, draped shade beneath. Aggressive root

competition takes place beneath pines, so the shrubs and lawn beneath and around the

canopy often require more frequent irrigation, particularly during the dry season. Pines

have some deep roots, except in poorly-drained soil where most roots proliferate near the

soil surface. The tap root is prominent in well-drained soil and can make pines difficult to

transplant in the wild. Slash pines grow well on a variety of acidic soils in either full sun

or partial shade. It does poorly in high pH soil and so is not recommended for calcareous

soils and where irrigation water has a high pH. Once established, it is more tolerant of

wet sites than most other pine species, and is highly drought-tolerant (Lohrey and

Kossuth, 1990)

Paspalum notatum (Bahingrass). Bahiagrass in an important forage species in

Florida where it occupies more land area than any other single pasture species, covering

an estimated 1 million hectare (Chambliss and Adjei, 2006). Most of this land area is

used not only for grazing but also for the production of hay, sod, and seed harvested from

pastures.

Bahiagrass is a warm-season perennial grown throughout Florida and in the Coastal

Plain and Gulf Coast regions of the southern United States. It is a native to South









America and is widely distributed in Argentina, Uruguay, Paraguay and Brazil.

Bahiagrass is adapted to climatic conditions throughout Florida and can be grown on

upland well-drained sands as well as the moist, poorly-drained flatwoods soil of

peninsular Florida (Chambliss and Adj ei, 2006); LSU, 1998). Bahiagrass is popular with

Florida ranchers because it tolerates a wider range of soil conditions than other improved

grasses, has the ability to produce moderate yields in soils of very low fertility, is easily

established from seed, withstands close grazing, and is relatively free from damaging

insects (except for mole crickets) and diseases.

Soils of the Study Sites

Soil sampling

Soil samples were collected during summer and fall 2004 at the four locations

(Figures 3-7 to 3-10) at the following sampling depths: 0 5, 5 15, 15 30, 30 50,

and 50 75 and 75 100 cm using a bucket soil auger (21 cm x 7 cm). A grid sampling

procedure was used (Figure 3-7). With eight plots, each with 24 sampling points and six

depths, there were a total of 1152 soil samples. Soil samples were taken in the alleys in

the silvopastoral systems. The spacing between the sampling points in both the regular

pasture and the silvopasture plots was based on the size of the plots. Soil Survey (hard

copies) and Soil Survey digital maps [Natural Resources Conservation Service (NRCS),

Soil Service Geographic (SSURGO)] of the area to be sampled were used to obtain

information on the soil series. A hand-held GPS unit (eTrexThi, GARMI~s, personal

navigator) was used to obtain the geographic coordinates of each the sampling point.













































Figure 3-7. Grid sampling at the Suwannee location. Twenty four soil profiles were
collected at each field (silvopasture and treeless pasture) at each location, and
with 6 depths (0 5, 5 15, 15 30, 30 50, and 50 75 and 75 -100 cm)
per profile. The small circles, representing sampling points, illustrate the
layout of the grid sampling method used in each sites at each location.

Soil-nutrient characterization

Mehlich-1 (0.0125MH2SO04 and 0.05M~HC1)-extractable Fe, Al, Ca, and Mg were

obtained using a 1:4 soil to double acid ratio (Mehlich, 1953). The metals in the

Mehlich-1 solutions were determined by atomic absorption spectrometry.









Soil texture and pH determination

Soil texture was determined by the hydrometer method following the analytic

procedure describe by Arshad et al. (1996). Soil pH (2:1) was assessed using the

analytical procedure described by Smith and Doran (1996).

Statistical Procedures

The statistical analyses were performed using SAS (PROC ANOVA) 8.2 for

Windows. The experimental design used was a 4-fold nested model (Table 3-4). The

statistical model used was the following:

TYkI U+a al p Tyk, ykl,
where

pu= grand mean

a, = effect of level i of location (Alachua, Suwannee, Osceola and Manatee)


fl, = effect of level j of treatment (Silvopasture and treeless pasture), nested

within level i of location

yOk = effect of level k of plot (24 sampling points), nested within level j of
treatment which is nested within level i of location

Syki = effect of level k of depth (6 depths), nested within level k of plot which is

nested in level j of treatment which is nested in level i of location










Table 3-4. Statistical model used in the study
Source Degree of
Freedom
Location 3
Treatment (location) 4
Plot (Treatment x location) 184
Depth (Treatment x location x plot) 960
Total 1151
4The hypotheses were tested using the ANOVA mean square for treatment (location) as the error term for
location, the ANOVA mean square for plot (treatment x location) as the error term for treatment (location),
and the ANOVA mean square for depth (treatment x location x plot) for plot (treatment x location).

Results and Discussion

Soil Reaction (pH)

Soil pH for the 0 5 cm depth for the silvopasture and treeless pasture in Alachua were

5.4 and 5.2 respectively. The corresponding values for the Suwannee, Manatee, and

Osceola field sites were 4.7 and 6.6; 4.8 and 4.4; 4.5 and 4.4 (Tables 3-5 to 3-8). The soil

pH was different between the two pasture systems for all soil layers at the Suwannee

location with the treeless pasture having the highest pH. Results from the Suwannee

location can be explained by the fact that the treeless pasture site has been continuously

limed at the rate of 4.5 Mg hal every 4 years since 1978. In Alachua, soil pH was only

found different at surface horizon with the treeless pasture site having lower pH values.

This is not consistent with the general perception that conifer (e.g., pines and firs) stand

tend to acidify the soil. Soil acidification is enhanced by most coniferous vegetation

because of the acidic nature of their resinous needles which tend to accumulate in a thick

O horizon (Brady and Weil, 2002). The difference in soil pH observed in Alachua maybe

attributed to some variation in the parent material across the farm (e.g. limestone

deposits) .

At the Manatee site soil pH was similar for both systems across all soil depths. In

Osceola differences in soil pH were only observed below 30 cm of depth with the treeless










pasture having the higher pH values. It is important to remember that soil samples were

only taken in the alleys in the silvopastures; therefore there could be differences in pH

near the pine trees. In general the soil pH in the two pasture systems at in all four

locations were in the desirable ranges for bahiagrass (<5.5 6.4) (Sartain, 2001)

throughout the soil profile of the sites.

Soil pH of in the similar range as been reported by Nair and Graetz (2002) for the A, E,

Bh, and Bw soil horizons of a beef pasture site (Spodosols) situated in the Okeechobee

County of Florida.

Soil Particle Size Analysis (Soil Texture)

The soils at Alachua, Suwannee, Osceola ,and Manatee were designated as Kendrick

(Loamy, siliceous, semiactive, hyperthermic Arenic Paleudults), Blanton (Loamy,

siliceous, semiactive, thermic Grossarenic Paleudults), Immokalee (Sandy, siliceous,

hyperthermic Arenic Alaquods), and EauGallie (Sandy, siliceous, hyperthermic Alfic

Alaquods) soil series, respectively (Table 3.1). For the Alachua field site, the percent

sand, silt, and clay for the 0 5 cm depth were 87, 4, and 9 and 92, 3, and 5% for the

treeless pasture and silvopasture, respectively. The corresponding values for Suwannee

were 94, 2, and 4% vs. 95, 3, and 3% compared to 94, 3, and 3% vs 95, 3, and 2%

(Manatee), and 92, 4, and 4% vs. 96, 2, and 2% in Osceola (Table 3-5 to 3-8). In general,

sand, silt and clay content was similar for both production systems across all soil depths.

In Suwannee the treeless pasture had a higher clay content in the upper soil layer. The

lack of differences in the soil texture between the two pastures at the four locations sites

is desirable since it facilitates the comparison of the two land use systems in terms of

there nutrient status. The clay content of the Ultisols at 75 100 cm depths was higher









compared to the Spodosols, which is consistent with the known characteristics of these

two soil orders (Collins, 2003).

Mehlich-1 Al and Fe Concentrations

Mehlich-1 Al concentrations at 0-5 cm depth for the silvopasture and treeless

pasture in Alachua were 436 and 479 mg kg-l respectively. The corresponding values

were 160 and 183 mg kg-l (Suwannee), 48 and 36 mg kg-l (Manatee) compared to 225

and 132 mg kg- in Osceola (Tables 3-5 to 3-8). Concentrations of Mehlich-1 extractable

Fe at the top 0 5 cm depth in the silvopasture and treeless pasture in Alachua were 28

and 43 mg kg- respectively. The corresponding values were 25 and 10 mg kg-l

(Suwannee), 18 and 18 mg kg-l (Manatee), while values at Osceola were 13 to 27 mg kg-l

(Tables 3-5 to 3-8). In Alachua, Mehlich-1 Al concentrations among the production

systems were different below 15 cm of depth. In contrast, Mehlichl-Fe concentrations

were different between the two pasture systems (Alachua) throughout the soil profile

from the top 0 5 cm to 75 100 cm depth. Differences in Al concentrations between the

two pasture sites within a location may be attributed to variation in clay content. The

higher the clay content of a soil the higher its Al concentration will be.

In Suwannee, there were generally no differences in Mehlich-1 Al and Fe

concentration among land use systems through out the entire soil profile. With the

exception of 0 5 and 75 100 depths, both Mehlich-1 Al and Fe concentrations were

not different between the pasture systems in Manatee. The level of these elements in the

soil was very low at the surface horizons in Manatee, which is a characteristic of

Spodosols. The Mehlich-1 Al content of the two pasture sites in Osceola showed no

difference throughout the soil profile. Iron concentration in the soil of the two pasture

systems was found to be different only at 5 15, 50 75, and 75 100 cm depths.









A strong relationship between Mehlich-1 Fe, and diagnostic tissue levels of Fe,

concentrations and a predictable responses to applied Fe has not being found, therefore

the University of Florida Extension Soil does not make fertilization requirement based

on extractable Fe (Sartain, 2001).

Average values for Mehlich-1 Al and Fe concentrations for the Ultisols and the

Spodosols were 315 and 110 mg kg-l and 27 and 19 mg kg- respectively, at depth 0 5

cm respectively. Results of the study showed that Ultisols generally had greater Mehlich-

1 Al concentration then Spodosols in the surface soil horizons (from 0 5 cm to 15 30

cm depth) [Table 3-9]. Below the 30 cm soil depth no significant difference in Al was

observed between the soil types. Ultisols had the highest Fe concentration at 15 30 and

75 100 cm. The higher Al concentrations observed in the Ultisols at the surface

horizons is an indication of the greater P retention capacity of this soils in the surface

horizons as compared to Spodosols since the amount of aluminum and iron oxides and

hydroxides in acid soils is a determinant factor in the amount of P that can be held by a

soil. At the deeper soil horizons no differences were observed between the two soil types

because the spodic horizon may have comparable levels of Al and Fe. These results are

consistent with Mansell et al. (1991) observations that the A and E horizons of Spodosols

are sandy and have poor sorption capacities while the spodic horizon have much greater

affinities for P retention. Nair and Graetz (2002) found Mehlich-1 Al and Fe

concentrations of 38.1i, 9.0, 385 and 635 and 13.2, 7.52, 46.1i, and 32.8, respectively, for

the A, E, Bh, and Bw soil horizons of a beef pasture site (Spodosols) of the Okeechobee

County of Florida.









Mehlich-1 Mg and Ca Concentrations

Mehlich-1 Mg concentrations for the silvopasture and treeless pasture in Alachua for the

0 5 cm depth were 132 and 223 mg kg- respectively. The corresponding values were

72 and 154 mg kg-l (Suwannee), 132 and 154 mg kg-l (Manatee), compared to 100 and

70 mg kg- in Osceola (Tables 3-5 to 3-8). In Alachua concentrations of Mehlich-1

extractable Ca in the top 0 5 cm depth for the silvopasture and the treeless pasture were

727 and 1006 mg kg- respectively. Values in Suwannee were 531 and 921 mg kg-l

compared to 1029 and 911 mg kg-l in Manatee, while values in Osceola were 663 to 596

mg kg-l (Tables 3-5 to 3-8). Similar Mehlich-1 Ca and Mg concentrations for the A, E,

Bh, and Bw soil horizons of a beef pasture site (Spodosols) in Okeechobee County (S-

Florida) of Florida were reported by Nair and Graetz (2002). Mehlich-1 Mg and Ca

concentrations were found to be different among the pasture systems in both Alachua and

Suwannee with the highest concentration found in the treeless pasture. Based on

UF/IFAS diagnostic soil test interpretations (Kidder et al., 2002) Mg concentration were

generally in the high range (>30 mg kg- ) in both the treeless pasture and the silvopasture

sites throughout the soil profile in Alachua. Soil Mg concentration was in the high range

in the treeless pasture throughout the soil profile and in the low range (<15 mg kg- ) in

the silvopasture below 5 cm of depth in Suwannee base on UF/IFAS interpretation

sy stem.

In general, Mehlich-1 Mg and Ca concentrations were similar for both production

systems in Manatee. The Mg concentration in two pasture sites was found to be in the

low range according to UF/IFAS interpretations. In Osceola Mehlich-1 Ca concentration

were also similar throughout sampled soil profile (up to 100 cm, which was the deepest

position studied), whereas Mehlich-1 Mg concentration was generally different. Soil Mg









concentration in the two pasture systems in Osceola was in the low range below 30 cm of

depth. In Manatee, the silvopasture and the treeless pasture sites had Mg concentration in

the high range at the surface 0 5 cm depth, whereas Mg concentration below 30 cm of

depth where in the low range.

No interpretation is made for Mehlich-1 extractable Ca levels because the

extraction method dissolves soil calcium compounds which are not readily available for

the plants. Generally, Ca levels tend to be adequate for turfgrass growth in Florida soils

due the fact that they are inherently high in Ca, have a history of Ca fertilization, or

receive Ca regularly through irrigation with high Ca water (Kidder et al., 2002).

Conclusions

The soils at the study sites typically had a rather coarse (sandy) texture, especially

in the upper layers, and in the absence ofliming may be prone to acidification and are

typical of the southeastern USA. Soil pH at the study sites were in the recommended

range for bahiagrass. In general the treeless pastures had higher pH values compared to

the silvopastures. Magnesium concentration levels in the treeless pasture sites were

generally higher than the silvopasture sites. Soil test values for agronomic crops were

"high" in Mg in the upper 0 5 cm at all four locations. Calcium concentration levels

were generally not different in the two pasture systems in Osceola and Manatee. Except

in Alachua where the treeless pasture had the highest Al levels of the two pasture

systems, no differences in Al concentrations were generally observed between the two

pastures. Iron concentrations were in general higher in the silvopasture as compared to

the treeless pasture in Alachua and Suwannee but this trend was not significant in

Osceola and Manatee. Soil Fe concentration level was found to be of lower magnitude at

all the four locations. The Ultisols had higher soil Al and Fe concentrations than the










Spodosols in the upper 30 cm of the soil profile. Higher clay content was also observed in

the deeper 75 100 cm depth in the Ultisols. The information presented here is critical to

the explanation of the results presented in the subsequent chapters.











































9 0.4415
8 0.4355
13 0.2511
16 0.3948
15 0.4091
15 0.5800


I


Table 3 -5. Mean values of Mehlich-1 Ca, Al, Mg, and Fe concentrations, pH, and percent sand, silt and clay for soil profiles at the
center of the alley of the silvopasture and a treeless pasture sites in Alachua Florida, USA.
Silvopasture Treeless pasture Silvopasture Treeless pasture Silvopasture Treeless P


pasture
Mg (mg kg- )
223*
126*
151*
216*
252*
239*


Ca (mg kg-')
1006*
920*
946*
969*
999*
1147*


Al (mg kg-')
479
563
531*
608*
660*
744*


Depth (cm)
0-5
5-15
15-30
30-50
50-75
75-100


P
0.0067
0.0001
0.0001
0.0001
0.0001
0.0001


P
0.3970
0.1338
0.0204
0.0004
0.0001
0.0001


0.0001
0.0001
0.0001
0.0001
0.0001
0.0001

P


0.4952
0.4225
0.3792
0.4173
0.3891
0.5366


Silvopasture Treeless pasture


P Silvopasture Treeless pasture


P Silvopasture Treeless
pasture
Sand (%)
0.0184 92 87
0.0061 91 86
0.7324 92 80
0.3964 91 76
0.0985 92 78
0.0632 89 79


Fe (mg kg-')
43*


Depth (cm)
0-5
5-15
15-30
30-50
50-75
75-100


pH (1:2)
5.2*
5.4*
5.8
5.8
5.7
5.6


0.0005
0.0001
0.0001
0.0001
0.0001
0.0001


Silvopasture Treeless pasture
Depth (cm) Silt (%)
0-5 3 4
5-15 4 5
15-30 7 7
30-50 4 7
50-75 4 7
75-100 4 7
[(*Indicates significant P values within a given


P Silvopasture Treeless pasture
Clay (%)


0.6851 5
0.4540 5
0.9931 1
0.4337 4
0.3258 5
0.4047 8
depth (p<0.05)]














Table 3 -6. Mean values of Mehlich-1 Ca, Al, Mg, and Fe concentrations, pH, and percent sand, silt and clay for soil profiles at the
center of the alley of the silvopasture and a treeless pasture sites in Osceola Florida, USA.
Silvopasture Treeless pasture Silvopasture Treeless pasture Silvopasture Treeless


pasture
Mg (mg kg-')
100* 70
24* 14
21* 5
9* 4
9* 4
8* 4


Ca (mg kg-')


Al (mg kg-')


Depth (cm)
0-5
5-15
15-30
30-50
50-75
75-100


P
0.0425
0.6351
0.0487
0.8618
0.4078
0.1296


P
0.2300
0.7845
0.6060
0.0810
0.7419
0.4600


P
0.0425
0.0346
0.0021
0.0003
0.0001
0.0004


663*
470
454
257
166
110


Silvopasture


Treeless pasture


Silvopasture Treeless pasture


Silvopasture


Treeless
pasture
)
92
92

94
96
95


Fe (mg kg-')


Depth (cm)
0-5
5-15
15-30
30-50
50-75
75-100


P
0.0903
0.0001
0.8071
0.0541
0.0320
0.0393


pH (1:2)


P
0.1257
0.7654
0.1221
0.0456
0.0175
0.0484


P
0.0143
0.3932
0.0761
0.5017
1.0000
0.3746


Sand (%


P
0.0542
0.1818
0.1383
0.5401
0.7798
0.4774


4.5a
4.6a
4.7a
4.6a
4.5a
4.4a


4.4a
4.5a
4.9a
4.9b
4.8b
4.7b


Silvopasture Treeless pasture
Depth (cm) Silt (%)
0-5 2 4
5-15 1 6
15-30 2 5
30-50 3 4
50-75 2 2
75-100 2 3
[(*Indicates significant P values within a given


Silvopasture Treeless pasture
P Clay (%)
0.1277 1 4*
0.1099 1 2
0.1807 1 3
0.6209 1 3
0.6690 2 2
0.5562 2 2
depth (p<0.05)]














Table 3 -7. Mean values of Mehlich-1 Ca, Al, Mg, and Fe concentrations, pH, and percent sand, silt and clay for soil profiles at the
center of the alley of the silvopasture and a treeless pasture sites in Suwannee Florida, USA.
Silvopasture Treeless pasture Silvopasture Treeless pasture Silvopasture Treeless


pasture
Mg (mg kg-')
154*
55*
28*
22*
15*
15*


Ca (mg kg-')
921*
689*
316*
232*
166*
132*


Al (mg kg-')
183
266
259
230
190*
151


Depth (cm)
0-5
5-15
15-30
30-50
50-75
75-100


P
0.0119
0.0001
0.0001
0.0001
0.0001
0.0001


P
0.2232
0.6034
0.6260
0.0078
0.0423
0.1581


P
0.0001
0.0001
0.0001
0.0001
0.0001
0.0041


Silvopasture


Treeless pasture


Silvopasture


Treeless pasture


Silvopasture


Treeless
pasture


Fe (mg kg-')
10
16
14
12
10
10


Depth (cm)
0-5
5-15
15-30
30-50
50-75
75-100


P
0.0001
0.0001
0.0052
0.0063
0.0043
0.0709


pH (1:2)
6.6*
6.1*
6.0*
6.0*
6.1*
6.1*


P
0.0001
0.0001
0.0001
0.0001
0.0001
0.0001


P
0.0133
0.1573
0.3268
0.1919
0.1548
0.5324


Sand (%)
94
94
94
94
95
94


P
0.0700
0.4226
0.2999
0.6020
0.7758
0.4641


Silvopasture Treeless pasture
Depth (cm) Silt (%)
0-5 3 2
5-15 3 3
15-30 3 4
30-50 4* 3
50-75 4 3
75-100 8 3
[(*Indicates significant P values within a given


Silvopasture Treeless pasture
P Clay (%)
0.2513 3 4*
0.6906 2 2
0.6414 2 3
0.3217 2 3
0.3057 2 3
0.3804 8 3
depth (p<0.05)]














Table 3 -8. Mean values of Mehlich-1 Ca, Al, Mg, and Fe concentrations, pH, and percent sand, silt and clay for soil profiles at the
center of the alley of the silvopasture and a treeless pasture sites in Manatee Florida, USA.
Silvopasture Treeless pasture Silvopasture Treeless pasture Silvopasture Treeless


asture
) P
154 0.1977
26* 0.0489
5 0.2427
3 0.3196
4 0.4792
8 0.5845


p
Mg (mg kg-'
132
18


Ca (mg kg-')
911
422
132
74
157
200*


Al (mg kg-')


Depth (cm)
0-5
5-15
15-30
30-50
50-75
75-100


P
0.2076
0.6161
0.0641
0.1158
0.2997
0.0029


P
0.0060
0.3668
0.1048
0.2783
0.4130
0.0002


1029
446
175
101
181


48*
19
14
56
144
246*


Silvopasture Treeless pasture


Silvopasture


Treeless pasture


Silvopasture


Treeless
pasture


Fe (mg kg-')


Depth (cm)
0-5
5-15
15-30
30-50
50-75
75-100


P
0.8605
0.2445
0.0001
0.8165
0.7010
0.4873


pH (1:2)


P
0.0001
0.1748
0.1731
0.0952
0.1956
0.0030


P
0.4073
0.4453
0.2137
0.7202
0.0299
0.0955


Sand (%)


4.8*
4.9
5.3
5.4
5.6
5.5


94 0.5124
89 0.3991
98 0.2929
98 1.0000
97 0.1296
97 1.0000


4.8
5.1
5.3
5.7
5.8*


Silvopasture Treeless pasture
Depth (cm) Silt (%)
0-5 3 3
5-15 2 10
15-30 1 1
30-50 2 1
50-75 2 2
75-100 2 2
[(*Indicates significant P values within a given


Silvopasture Treeless pasture
P Clay (%)
0.7758 2 3
0.3950 11
0.4929 1 1
0.7758 0 1
0.5425 0.8 1.2*
0.0513 1.5 1.2
depth (p<0.05)]










































0.2636
0.1023
0.3856
0.2633
0.2142
0.0048


Table 3 -9. Mean values of Mehlich-1 Al and Fe concentrations and percent sand, silt and clay for soil profiles of the Ultisols and the
Spodosols.
Ultisols Spodosols Ultisols Spodosols


Depth
(cm)
0-5
5-15
15-30
30-50
50-75
75-100


--------(mg kg ')-----
315* 110
395* 93
366* 110
345 289
319 272


--------(mg kg ')------


0.0224
0.0045
0.0143
0.7147
0.7011
0.8370

P

0.3242
0.5508
0.1980
0.1752
0.1597
0.0018


0.1702
0.1368
0.0416
0.0749
0.0803
0.0479


Ultisols Spodosols
-------------Sand (%)------------


Ultisols


Spodosols


Depth (cm)
0-5
5-15
15-30
30-50
50-75
75-100


------------Silt (%)-------------


0.7351
0.8862
0.0951
0.0428
0.0601
0.0286


Ultisols Spodosols
Depth (cm) ----------Clay (%)-----------
0-5 5 3
5-15 3 1
15-30 5 1
30-50 6 1
50-75 6 1
75-100 9* 2
[*Indicates significant P values (p<0.05)]



























































Figure 3-8. Aerial photo of the Alachua location





























































Figure 3-9. Aerial photo of the Osceola location



































I


Figure 3-10. Aerial photo of the Manatee location















CHAPTER 4
PHOSPHORUS AND NITROGEN CONCENTRATIONS UNDER A SLASH PINE
(Pinus elliottii) + BAHIAGRASS (Paspalum notatum) SILVOPASTURE AND A
TREELESS PASTURE

Introduction

Non-point pollution is a maj or source of phosphorus (P) and Nitrate-N to surface

waters of the United States (Carpenter et al., 1998). Phosphorus concentrations in excess

of 0.02 mg L^1 in lake water can lead to accelerated eutrophication (Sharpley et al., 1999).

Nitrate-N concentration in excess of 10 mg L^1 in drinking water may lead to the "blue

baby syndrome"(USEPA, 2006). Soil receiving large quantities of livestock manure can

accumulate significant amounts ofP and result in excessive P loss to surface and ground

water through surface runoff and subsurface leaching.

Studies in tropical agroforestry systems on nutrient-poor soils have shown that deep

roots of trees in agroforestry systems can absorb nutrients that have leached below the

rooting zone of associated agronomic crops, and subsequent deposition of litter and root

materials facilitate enhanced nutrient recycling and thereby improve the nutrient-use

efficiency in the system as a whole (Young, 1997). Furthermore, tree roots tend to be

long-lived and even in the upper soil layers they enhance nitrate and phosphorus retention

before and after the crop is planted and harvested (Jose et al., 2004). In essence, this

would enhance the overall resource-use efficiency of the system (van Noordwijk et al.,

2004). Although, the vertical distribution of a given species' root system in its natural

niche may follow a distinct pattern, leading to overlapping root distributions, some

species are able to respond to changes in nutrient supplies or restricting soil layers) (Jose









et al., 2004). This has been demonstrated by Wanvestraut et al. (2004) in a study

conducted in pecan (Calrya illinoensis) + cotton (Gossypium hirsutunt) alley cropping

system in northwestern Florida. They observed that cotton roots exhibited a degree of

plasticity, being limited by shallow soil strata and they also avoided a region of high

pecan root density deeper in the soil. In another study conducted at the same site, Allen et

al. (2003) reported a 34% reduction in nitrate nitrogen concentration in the soil solution

at 0.9 m depth when pecan and cotton roots were not separated by a root barrier

compared to the "root-barrier" treatment, suggesting the potential safety net role of pecan

roots. This safety net hypothesis can also be used to explain the sediment and nutrient

capture functions of riparian buffer strips. Riparian buffers have been presented as a tool

for removing non-point source pollutants from farmlands (Addy et al., 1999; Bharati et

al., 2002; Lee et al., 2003; Schultz et al. 2004). In a study conducted by Lee et al. (2003)

in central Iowa, Midwestern United States, a switch grass (Panicunt virgatunt)/woody

buffer treatment removed 90% of the sediment, 85% nitrate-N, and 80% of the P in the

runoff. The integration of the woody component in the buffer strip increased the uptake

efficiency of soluble nutrients by over 20%. Other applications of the safety-net include

phytoremediation of contaminated sites (Rockwood et al., 2004) and rehabilitation of

heavily fertilized agricultural soils in Florida soils (Nair and Graetz, 2004). Thus, while

the 'safety-net' concept is critical for enhancing nutrient use efficiency and improving

crop production under tropical conditions, it has a different application in the heavily

fertilized and excessively drained soils of Florida that have limited nutrient retention

capacity. The tree roots' ability to retrieve nutrients from deeper soil layers can be

exploited advantageously to increase the apparent nutrient retention in excessively









drained soils and thereby reduce the amount of nutrients that might otherwise be

transported to surface water through runoff and leaching and thereby cause non-point-

source pollution of these water bodies.

Water soluble P (WSP) a environmental soil testing method for environmental

assessment of available soil P, can better represent the rapid release of P to water than

stronger chemical extractant (such as agronomic soil P tests), since rainfall has similar

properties than distilled water (Moore et al., 1998). This extractant is also known to

maintain the soil pH within one unit of its original value, an important characteristic since

P solubility is strongly dependent on soil pH (Golterman, 1998; Sharpley, 1993).

Furthermore, a strong positive correlation exists between water soluble P and P in runoff

(Pote et al., 1996).

The obj ective of this study was to quantify water soluble P (WSP), Mehlich-1 P,

and nitrate and ammonium concentrations in surface and subsurface horizons in soil

profiles of silvopastoral and treeless pasture systems at four different locations of two soil

orders in Florida.

Materials and Methods

Site Description

The research was conducted on four slash pine (Pinus elliottii) + bahiagrass (Paslpalum

notatum) silvopastoral systems located in Osceola (280 9' N, 810 10' W), Manatee (270 13'

N, 820 8' W), Suwannee (300 24' N, 830 O' W), and Alachua (290 45' N, 820 33' W)

counties in Florida. A grid sampling procedure was adopted to collect 24 soil profiles

from each location during summer and fall 2004. The soil samples were collected by

depth: 0 5, 5 15, 15 30, 30 50, 50 75 and 75 100 cm. The soils in both









Manatee and Osceola counties are Spodosols, whereas those in Alachua and Suwannee

are Ultisols. Refer to Chapter 3 for more in-depth details of the site description.


Statistical Procedures

The statistical analyses were performed using SAS PROC ANOVA (SAS 8.2 for

Windows). The experimental design used was a 4-fold nested model (see Chapter 3 for

the description of the model).


Water Soluble P and Mehlich-1 P Determination

Water soluble P was determined by extracting the soil with distilled dionized water

using a 1:10 soil to water ratio. This mix was shaked with a mechanical shaker for 1 hour

and filtered through a 0.45-Cpm filter using vacuum filtration. Mehlich 1 (0.0125M~H2SO4

and 0.05M~HC1)-extractable P was obtained using a 1:4 soil to double acid ratio

(Mehlich, 1953). Phosphorus in solution was determined by an autoanalyzer (USEPA,

1993; Method 365.1) by the Murphy-Riley (1962) method.

Nitrate and Ammonium-N Determination

Nitrate and ammonium-N extracts were obtained using a 1:10 soil to KCL (2M)1

ratio. This mix was shaked on a mechanical shaker for 1 hour. The KCl suspension was

then filtered by vacuum filtration as described above. A membrane filter of 0.45Cpm was

used for the extraction. After extraction ammonium-N in the KCl extractant was

determined using a semi-automated colorimetric analysis (USEPA Method 350.1,

USEPA, 1993). Nitrate was determined using a semi-automated colorimetric analysis

(USEPA Method 353.2, USEPA, 1993). Only 6 out of the 24 soil profiles taken in each

sites were analyzed for nitrate and ammonium content. With eight sites, each with 6

sampling points and six depths, a total of 288 soil samples were analyzed.









Results and Discussion

Water Soluble P Concentrations

WSP concentrations at the 0 5 cm depth in the silvopasture and treeless pasture

on Alachua were 5.6 and 10.9 mg kg- respectively. Corresponding values in Suwannee

were 4. 1 and 10. 1 mg kg-l compared to 11.4 and 23.4 mg kg-l in Manatee, and values of

1.4 and 11.0 mg kg- in Osceola (Tables 4-1 and 4-2 and Figures 4-1 and 4-2).

Differences in WSP concentrations were found through out the soil profile (up to 100 cm)

between the two pasture systems in Alachua, Manatee and Suwannee. In Osceola, WSP

concentrations was found to be different between the two pasture systems at 0 5, 15 -

30, 30 50, and 75 100 depths. Water soluble P was generally higher in the treeless

pastures throughout the soil profile, indicating the safety net role of slash pine roots. Nair

and Graetz (2004) found that water soluble P concentrations in the surface soils of a

Spodosols varied throughout the soil profile in the order of pasture > silvopasture >

native flatwoods vegetation. They reported WSP concentrations of 9. 1, 4.2, and 1.6 mg

kg-l in the top 0 5 cm depth, respectively, in a treeless pasture, silvopasture, and native

flatwoods. Even in an Ultisol with substantial P retaining clay below 20 cm, there appears

to be movement of P to lower depths. Below the retentive clay layer the treeless pasture

contained more P in the soil than the silvopasture (Figure 4-2) suggesting removal of P

from the lower depths by the trees in the silvopasture systems. This observation suggests

that the likelihood of P moving out of the soil is greater under a treeless pasture than

under a tree-based system. These findings are consistent with that of a study conducted

by Friend et al. (2006) on a loblolly pine (Pinus taeda) stand. In this study dry poultry-

litter applications of approximately 5 Mg ha-l to mid-rotation loblolly pine stands would

be contained by the ecosystem and may result in significant increases in tree growth.









Although we would expect to find higher WSP concentrations in the Spodosols,

which have lower P retention capacity, especially in the surface A and E horizons as

opposed to the Ultisols, the results do not support this expectation (Table 4-2). These

findings could be explained by the hydrological dynamics of the Spodosols that have

seasonally and highly fluctuating water table. Spodosols are characterized by a highly

fluctuating water table situated between the Bh (a layer form by the accumulation of

organic matter and/or iron and aluminum oxides) and the A horizons during the summer

rainy season, which goes down to 125 cm during drier months (Soil Survey Staff, 1996).

Rainfall that infiltrate into the soil during high water table conditions can move laterally,

transporting P to surface drainage ditches (Burgoa et al., 1991; Mansell et al., 1991).

Under this condition, water soluble P, the most labile form of P, is prone to lost via

subsurface lateral flow before reaching the spodic horizon, where a P retention capacity

exi st..

Mehlich-1 P Concentrations

In Alachua, the Mehlich-1 P concentrations at 0 5 cm depth for the silvopasture

and treeless pasture were 97 and 104 mg kg- respectively. Corresponding values for the

Suwannee, Manatee, and Osceola fields were 9 and 11 mg kg- ,5 and 6 mg kg-l (Tables

4-1 and 4-2 and Figures 4-3 and 4-4). Mehlich-1 soil test values for both the silvopasture

and the treeless pasture in Alachua and the treeless pasture in Suwannee fell within the

high (31-60 mg kg- ) to the very high range (>60 mg kg- ) as suggested by UF/IFAS

Mehlich-1 soil-test interpretation for agronomic crops (Kidder et al., 2002). In contrast,

Mehlich-1 P concentrations in the silvopasture sites Suwannee and in both pasture

systems in Manatee and Osceola fell generally in the very low range. No difference in

Mehlich-1 P concentration was observed between the two pasture systems (a = 0.05) in










Alachua at 0 5, 5 15, and 15 30 cm depths. However, the treeless pasture had higher

P values at the 30 50, 50 75, and 75 100 cm depths. For the Suwannee site,

Mehlich-1 P values below 30 cm were lower for silvopasture systems, which is indicative

of the slash pine "nutrient-capture" effect.

In general, no difference was observed between the two pasture systems at the

Manatee and Osceola sites while the overall Mehlich-1 P levels at the sites was ten fold

lower compared to Alachua and Suwannee. Presence of perched water table above the

spodic soil layers enhances lateral P loss before it can reach the Bh horizon. In this case,

the low P retention capacity of the A and E horizons, will control overall P movement.

There is a need for further research to quantify the different nutrient pools in these

agroforestry systems. Establishing long-term on-farm runoff plots and in-site wells

monitoring will confirm the effectiveness of these agroforestry systems in reducing

nutrient loss.

Results from the current study showed that the Ultisols (Alachua and Suwannee)

had greater Mehlich-1 P concentration then Spodosols (Osceola and Manatee) in the

upper soil layers (up to 50 cm of depth) [Table 4.3]. This is consistent with the higher Al

and Fe concentrations in the upper soil layers of Ultisols (compared to Spodosols), which

increase the P sorption capacity of these soils (see Chapter 3). This is also consistent with

the characteristics of Spodosols, having a sandy A and E horizons with poor sorption

capacities while the spodic horizon has much greater capacity for P retention (Mansell et

al., 1991). Chen and Lena (2001) reported that P values in undisturbed Florida surface

soils were as follows: Hi stosol s>Incepti sol s>Molli sol s>Enti sol s>Alfi sol s> Spodosol s.










Average values for Mehlich-1 P concentrations from the present study for the

Alachua, Suwannee, Osceola and Manatee sites were, respectively, 100, 50, 10 and 6 mg

kg-l at depth 0 5 cm and 125, 7, 5, and 7 at depth 50 75 cm in the present study (Table

8-3). Nair and Graetz (2002) reported Mehlich-1 P concentrations of 14.9, 2.3, 5.2 and

8.1 respectively in the A, E, Bh, and Bw soil horizons of a beef pasture site (Spodosols)

in the Okeechobee County of Florida.

Nitrate and Ammonium-N Concentrations

In Alachua, nitrate concentrations at 0 5 cm soil depth were 0.7 and 4.2 mg kg-l

respectively, in the silvopasture and the treeless pasture. Corresponding values for

Suwannee were 1.9 and 10.8 compared to 0.8 and 0.0 in Manatee, and values of 2.0 and

0.2 for the Osceola site (Tables 4-1 to 4-2 and Figures 4-5 and 4-7). Ammonium

concentrations in the 0 5 cm soil depth were 13.5 and 16.7, respectively, in the

silvopasture and treeless pasture in Alachua, and 10.5 and 6.0 in Suwannee, 13.6 and 14.0

in Manatee, 2.3 and 2.2 in Osceola (Tables 4-1 to 4-2 and Figures 4-6 and 4-8). In

Suwannee higher nitrate-N concentration was found in the treeless pasture as compared

to the silvopasture at 0 5, 50 75, and 75 100 cm depths, whereas the ammonium-N

concentration was only different at 0 5 cm depth.

Contrary to our expectations, no nitrate- and ammonium-N concentrations were

generally observed among the pasture systems in Alachua, Osceola and Manatee. Nitrate

and ammonium-N levels were generally very low, sometimes below the detection level.

For a similar study in a pecan (Carya illinoensis) + cotton (Gossypium hirsutum) alley

cropping system on Ultisols in northwestern Florida, Allen et al. (2003) reported a 34%

reduction in nitrate concentration in the soil solution at 0.9 m depth when pecan and

cotton roots were not separated by a root barrier compared to the "root-barrier"










treatment., suggesting the potential safety net role of pecan roots. It is difficult to account

for nitrate in most soil. Nitrate, a negatively charged ion, is not retained by the adsorbing

complex of the soil and thus is very prone to leaching. Concentrations of these labile N

species are in constant flux. It is also important to bear in mind that most soil ammonium

will be transformed biologically to nitrate within 2 to 3 weeks at soil temperatures in the

750 to 900 F range (Boman and Obreza, 2002). Another avenue for potential nitrate losses

in soil with highly fluctuating water table such as Spodosols (Osceola and Manatee) is

denitrification. Under anaerobic conditions, nitrate is transformed into gaseous forms and

lost to the atmosphere. Also, the nitrate-N could have moved laterally from the soil via

subsurface flow.

Conclusions

In spite of the limitations of the study, the results show that silvopastures sites had

lower P levels in the soil profile than treeless pastures suggesting that there is a tree

effect. Silvopasture will increase the nutrient stock in the plant system following

enhanced total nutrient uptake by trees and associated forage species, compared to the

shallow-rooted forage species. The amount of P that could be transported to surface

water will thus be reduced in the silvopstoral system. Although a similar trend was

observed in the case on nitrogen, the differences between the two systems were less

striking, possibly because of the difficulties in measuring the highly dynamic soil

nitrogen levels.





Table 4-1. Mean values of water soluble P (WSP), Mehlich-1 P (M-1 P), and nitrate and ammonium-N concentrations for different
soil layers of soil profiles at the center of the alley of the silvopasture and a treeless pasture sites in Alachua and Suwannee
Florida, USA.


Alachua
Silvopasture Treeless pasture

WSP (mg kg-')
5.6 10.9*
2.8 8.5*
1.9 7.9*
1.7 7.3*
1.5 6.2*
1.4 6.7*
Alachua
Silvopasture Treeless pasture

NH4 (mg kg )
13.5 16.7
7.1 7.0
3.5 5.2
3.1 4.1
2.4 3.1
2.4 2.8
Suwannee
Silvopasture Treeless pasture


Alachua
Silvopasture Treeless pasture


Alachua
Silvopasture Treeless
pasture
NO3 (mg kg )
0.7 4.2
0.3 0.8
0.2 0.3
0.1 0.2
0.0 0.0
0.0 0.1
Suwannee
Silvopasture Treeless
pasture
M-1 P (mg kg )
8.6 90.6*
17.0 88.0*
11.0 72.5*
6.0 42.6*
3.4 19.7*
2.6 11.6*


P


0.0960
0.3358
0.4404
0.2271
0.1059
0.2345

P


0.0001
0.0001
0.0001
0.0001
0.0001
0.0001


M-1 P (mg kg )
104.1
125.8
108.6
125.4*
1 ~171.9*
i 200.4*
Suwannee
sture Treeless pasture


Depth (cm)
0-5
5-15
15-30
30-50
50-75
75-100


P
0.0001
0.0001
0.0001
0.0001
0.0001
0.0001


P
0.7011
0.1602
0.1669
0.0079
0.0001
0.0001




P
0.0001
0.0001
0.0001
0.0001
0.0001
0.0001


P
0.0149
0.3138
0.5804
0.0765
0.1599
0.3726


96.5
99.5
83.6
61.4
49.1
49.6


Silvopal


WSP (mg kg-')
4.1 10.1*
1.2 6.4*
0.6 4.6*
0.4 2.9*
0.3 1.7*
0.2 1.5*
Suwannee
Silvopasture Treeless pasture


NH4 (mg kg-')
6.0


Depth (cm)
0-5
5-15
15-30
30-50
50-75
75-100


P
0.1742
0.9443
0.0730
0.1282
0.0732
0.4612


Depth (cm) NO3 (mg kg ') P
0-5 1.9 10.8* 0.0028 10.5*
5-15 0.3 3.0* 0.0001 3.7
15-30 0.1 1.08* 0.0001 2.7
30-50 0.1 0.9* 0.0001 2.9
50-75 0.0 0.7* 0.0050 1.9
75-100 0.1 0.7* 0.0298 2.4
[(*Indicates significant P values within a given depth (p<0.05)]














































0.3369 3.0 2.5 0.4652
0.1876 2.1 2.1 0.1421
depth (p<0.05)]


Table 4-2. Mean values of water soluble P (WSP), Mehlich-1 P (M1-P), nitrate, ammonium concentrations of different soil layers for
soil profiles at the center of the alley of the silvopasture and a treeless pasture sites in Manatee and Osceola, Florida, USA.
Manatee Manatee Manatee


Treeless
pasture


Silvopasture Treeless pasture


Silvopasture Treeless pasture


Silvopasture


Depth (cm)
0-5
5-15
15-30
30-50
50-75
75-100




Depth (cm)
0-5
5-15
15-30
30-50
50-75
75-100


WSP (mg kg-1)
11.4
1.43
0.30
0.10
0.04
0.05

Silvopasture

NH4 (mg kg-1)
13.6
3.4
2.8
1.5
1.4
2.0

Silvopasture


P
0.0009
0.0419
0.0001
0.0001
0.0001
0.0002


M-1 P (mg kg-1
8.9
0.7
0.2
0.0
1.0
2.3

Silvopasture

WSP (mg kg-1)
6.8
1.4
1.0*
0.7*
0.6
0.9*


P
0.1796
0.0006
0.4311
0.1472
0.0095
0.0006




P
0.0352
0.9548
0.0208
0.0320
0.1360
0.0164


P
0.4956
0.0152
0.3807
0.3391


NO3 (mg kg-1)
0.8 0.0
0.8 0.0
0.5* 0.0
0.2* 0.0
0.7* 0.3
0.2 0.2
Osceola
Silvopasture Treeless
pasture
M-1 P (mg kg-1)
5.0 6.1
1.7 1.9
1.4* 0.5
3.9 1.0
7.1* 2.3
11.5* 2.7


P
0.3097
0.2583
0.0203
0.0282
0.0150
0.5636




P
0.1213
0.7390
0.0487
0.1009
0.0405
0.0002


23.4*
4.60*
0.60*
0.43*
0.33*
0.95*
Manatee
Treeless pasture


14.0
4.0
2.0
1.5
1.6
1.8
Osceola
Treeless pasture


10.9
1.5*
0.1
0.3
3.3*
8.1*
Osceola
Treeless pasture


11.0*
1.4
0.5
0.3
0.4
0.5
Osceola
Treeless pasture

2.2
10.5
5.8


P
0.9024
0.3212
0.3035
1.0000
0.3156
0.5706


Silvopasture
P NH4 (mg kg-1)
0.0068 2.3
0.6081 12.4
0.0786 5.0
0.1011 3.1


Depth (cm) NO3 (mg kg-1)
0-5 2.0* 0.2
5-15 0.8 0.6
15-30 1.7 0.5
30-50 3.0 0.3
50-75 0.3 0.2
75-100 0.4 0.1
[(*Indicates significant P values within a given






65


Table 4-3. Mean values of Mehlich-1 extractable-P (M 1-P) concentrations for soil
profiles of the Spodosols (Osceola and Manatee) and Ultisols (Alachua and
Suwannee) in Florida.
Depth Alachua Suwannee Manatee Osceola
(cm)
0-5 100a 50b 10c 6c
5-15 113a 52b 2c Ic
15-30 96a 42b Ic 0.1c
30-50 93a 24b 3c 0.2c
50-75 111a 12b 2b 5b
75-100 125a 7b 5b 7b
*Means separation by Waller-Duncan procedure. Mean values of soil
characteristics within a given depth followed by the same letter are not
different (p<0.05)











WSP concentration (mg kg l)


cl '

PA ~7C~


SSilvopasture (Alachua)
S- Treeless (Alachua)
SSilvopasture (Suwannee)
S- Treeless pasture (Suwannee)


I
'I '


3C~


I


100


Figure 4-1. Water soluble P (WSP) concentration at different soil layers for silvopasture and treeless pasture at the Alachua and
Suwannee field sites featuring Spodosols. Error bars indicate standard errors (n = 24).











WSP concentration (mg kg l)


0 5 10 15 20 25 30



10 Silvopasture (osceola)

-m Treeless pasture (osceola)
20
SSilvopasture (Manatee)

30 1 Treeless pasture (Manatee)


S 40

c~ 50 11Bh (Osceola)

IEJ I:IBh (Manatee)
60

70




90

100


Figure 4-2. Water soluble P (WSP) concentration by depth in the silvopasture and treeless pasture sites in Osceola and Manatee. Error
bars designate standard errors.






































100


150


200


250


10


20 -

30




50

60

70 -1 I

80i -


- Treeless pasture (Alachua)

-Silvopasture (Alachua)
Silvopasture (Suwannee)
- Treeless pasture (Suwannee)


100

Figure 4-3. Mehlich-1 P concentration by depth in the silvopasture and treeless pasture sites in Alachua and Suwanneee. Error bars
designate standard errors.


Mehlich-1 P concentration (mg kg l)










Mehlich-1 P concentration (mg kg l)


S- Treeless pasture (Manatee)
SSilvopasture (Manatee)
SSilvopasture (Osceola)
S- Treeless pasture (Osceola)


100'
Figure 4-4. Mehlich-1 P concentration by depth in the silvopasture and treeless pasture sites in Osceola and Manatee. Error bars
designate standard errors.














0



20

30

40

50

60


I I
II


-silvopasture (suwannee)
-* Treeless pasture (Suwannee)
-- Silvopasture (Alachua)
-* Treeless pasture (Alachua)


70 -


Figure 4-5. Nitrate concentration level by depth in silvopasture and treeless pasture sites in Alachua and Suwannee on Ultisols. Error
bars designate standard errors.


Nitrate concentration (mg kg" )


80

90





SSilvopasture (Alachua)
S- Treeless pasture (Alachua)
SSilvopasture (Suwannee)
S- Treeless pasture (Suwannee)


I~ /


100

Figure 4-6. Ammonium-N concentrations levels by depth in silvopasture and treeless pasture sites in Alachua and Suwannee (Ultisols
sites). Error bars designate standard errors.


Ammonium concentration(mg kg- )












Nitrate concentration (mg kg- )


O 1 2 3 4


Treeless pasture (Manatee)
10 Si'~S lvopasture (Manatee)
/ Silvopasture (Osceola)
20
Treeless pasture (Osceola)

so -11 /

40 I

50

60

70 -( II

so II

90

100

Figure 4-7. Nitrate-N concentration for different soil depth for silvopasture and treeless pasture systems for filed sites in Manatee and
Osceola (Spodosols). Error bars designate standard errors.


5













O 2 4 6 8 10 12 14 16 18



10 ~~ -

20-/
~~ Silvopasture (Mlanatee)
30 -1 / / Treeless pasture (Mlanatee)
/ ~ Silvopasture (Osceola)
S40 -1 ck Treeless pasture (Osceola)

50




70




90

100

Figure 4-8. Ammonium-N concentrations levels for different soil depth in silvopasture and treeless pasture sites in Manatee and
Osceola on Spodosols. Error bars designate standard errors


Ammonium concentration(mg kg- )














CHAPTER 5
SOIL PHOSPHORUS STORAGE CAPACITY UNDER A SLASH PINE (Pinus elliottii)
+ BAHIAGRASS (Paspalum notatum) SILVOPASTURE AND A TREELESS
PASTURE

Introduction

Phosphorus loss from farmlands has been identified as one of the main causes of

eutrophication of fresh water bodies. Phosphorus loading of water bodies associated with

runoff from operations and heavy manuring of pasturelands in soils is a serious

environmental concern. The incorporation of trees in pastures could enhance P storage in

the soil-plant system (Nair and Graetz, 2004; Nair and Kalmbacher, 2004; Jose et al.,

2004; Chapter 4 of the present study). Silvopasture, the intentional combination of trees,

forage plants and livestock into an integrated, intensively-managed system (Nair, 1993;

Garrett et al., 2000; Nair et al., 2004) is potentially a promising land-use system to

enhance the economic returns of forestry operations (Lundgren et al., 1983; Lewis et al,

1985; Dangerfield and Harwell, 1990; Clason, 1995; Grado et al., 2001; Stainback and

Alavalapati, 2004).

Studies on tropical agroforestry systems in nutrient-depleted soils have

demonstrated the capacity of deep rooted trees to absorb nutrients that have leached

below the rooting zone of agronomic crops, and recycle these nutrients through litterfall

and rootfall in the crop-root zone, thus improving the nutrient-use efficiency in the

system as a whole (Young, 1997). In addition, the longer lifespan of tree roots help

capture nutrients such as nitrate and phosphorus before and after the crop is planted and









harvested (Jose et al., 2004). This would enhance the overall resource-use efficiency of

the system (van Noordwijk et al., 1996).

Research conducted in temperate agroforestry systems have showed similar results.

In a study conducted in pecan (Carya illinoensis K. Koch)-cotton (Gossypium hirsutunt

L.) alley cropping system in northwestern Florida, Allen et al. (2003) reported a 34%

reduction in nitrate nitrogen concentration in the soil solution at 0.9 m depth when pecan

and cotton roots were not separated by a root barrier compared to the root barrier

treatment, suggesting the potential safety-net role of pecan roots. The nutrient

interception capacity of riparian buffer is another example of the safety-net concept.

Riparian buffer strips offer a means of removing non-point source pollutants from

farmlands (Addy et al., 1999; Bharati et al., 2002; Lee et al., 2003; Schultz et al., 2004).

Lee et al. (2003), in a study undertaken in central Iowa, Midwestern United States,

reported that a switch grass (Panicunt virgatunt)-woody buffer treatment removed 85%

nitrate-N, and 80% of the phosphate-P in the runoff. Including a woody component in the

buffer strip enhanced the uptake effectiveness of soluble nutrients by over 20%.

According to Anbumozhi et al. (2005) forested buffers reduced agricultural nutrient

losses from livestock activities in adj acent uplands in surface stream waters in the

Tokachigawa watershed in Japan. In another study conducted in several European

countries significantly higher N uptake from forested sites as compared to herbaceous

sites was observed (Hefting et al., 2005).

Despite the fact that the 'safety-net' concept is important to enhancing nutrient use

efficiency and improving crop production under tropical ecological settings, it has

another application in the heavily fertilized and sandy soils of Florida (Nair and Graetz,









2004) that have low nutrient retention capacities. The capacity of tree roots to capture

nutrients from the deeper soil horizons can enhance nutrient storage in the plant-soil

system and therefore reduce the amount of nutrients that might otherwise be transported

to ground and surface water through runoff and leaching and thus cause non-point-source

pollution of these water bodies.

Environmental indicators of P loss such as soil test P (STP) or degree of P

saturation (DPS) do not allow any assessment of the capacity of a soil to retain future

addition of P (Nair and Harris, 2004). To address this shortcoming, these authors

proposed a DPS-based calculation of the remaining "Soil P Storage Capacity (SPSC)".

This index takes into account both the risks attributed to prior loading and the inherent,

limited P sorption capacity of a soil. The SPSC offers a direct estimate of the amount of P

a given soil can hold before exceeding a threshold soil concentration, consequently,

resulting in water pollution.

Generally, the P sorption maximum of a soil is determined by using the Langmuir

adsorption equation (Nair et al., 1998), which is a complex and time-consuming method.

Another way of estimating the soil P storage capacity of a soil is to use a single point

isotherm (or one point) value. Simple measurements of a single-point isotherm were

shown to reasonably estimate P sorption capacities of soils (Harris et al., 1996; Indiati et

al., 2002; Nair and Graetz, 2002; Lu and O'Connor, 2004 ).

The objectives of this study were to (i) assess differences in SPSC estimates in the

surface and subsurface horizons in soil profiles of two pasture systems (silvopasture and

treeless pasture) at four locations including two different soil orders in Florida, and (ii)









assess differences in SPSC estimates using two methods (Mehlich-1 and single-point

isotherm).

Material and Methods

Site Description

An on-farm research was undertaken in Osceola (280 9' N, 810 10' W), Manatee (270 13'

N, 820 8' W), Suwannee (300 24' N, 830 O' W), and Alachua (290 45' N, 820 33' W)

counties in Florida. The silvopastoral systems consisted of a slash pine (Pinus elliottii)-

bahiagrass (Paspalum notatum) and the treeless pasture consisted only of bahiagrass. A

grid sampling procedure was used to collect 24 soil profiles from each site at each

location during summer and fall 2004. The soil samples were collected at six depths: 0 -

5, 5 15, 15 30, 30 50, 50 75 and 75 100 cm. The soils in both Alachua and

Suwannee are Ultisols, whereas those in Manatee and Osceola counties are Spodosols.

Refer to Chapter 3 for more in-depth details on the site description. Refer to chapter 3 for

an in depth description of the sites characteristics.

Statistical Procedures

The statistical analyses were performed using SAS PROC ANOVA (SAS 8.2 for

Windows). The experimental design used was a 4-fold nested model (see Chapter 3). The

regression analyses were performed using SAS PROC REG and PROC GLM (SAS 8.2

for Windows).

Calculations of Parameters

Soil P storage capacity calculated with Mehlich-1

The phosphorus saturation ratio (PSR) was computed as the molar ratio of Mehlich l-P

and Mehlichl-Fe and Al (Nair et al., 2004). The PSR formula is as follows:

PSR = (Mehlich-1 P + 31) + ((Mehlich-1 Fe + 55.8) + (Mehlich-1 Al + 27))









The soil phosphorus storage capacity (SPSC) was calculated as:

SPSC = (0.15 PSR) ((Mehlich-1 Fe + 55.8) + (Mehlich-1 Al + 27)) 31 (mg P kg- )

(Nair and Harris, 2004)

The SPSC, is thus an estimate of the amount of P that can safely be applied to a soil up to

a defined depth before the soil poses an environmental risk. The SPSC values were then

calculated on a kg ha-l basis (for a specified depth) assuming a bulk density of 1500 kg

m-3 (Nair and Harris, 2004). For the calculation of SPSC, the choice of PSR instead of

DPS was made solely for simplicity purpose. The SPSC of the soil was based on a

threshold PSR of 0. 15. According to Nair et al. (2004), the value of 0. 15 used in the

SPSC calculation is the best approximation available to date of the PSR value

corresponding to the critical P solution concentration of 0. 10 mg P L^1 as proposed by

Breeuwsma and Silva (1992).


Soil P storage capacity calculated with Single Point Isotherm Data

The single point isotherm was determined following the procedure described in

Figure 5-land using 4 sampling points from each study site. With eight plots, each with 4

sampling points and six depths, a total of 192 soil samples were used for the analysis.





2g air-dried soil



20 mL of 0.01M KCI solution containing 100 mg P L1



24-h equilibration period using a mechanical shaker



1-h suspension settling down



Vacuum Hiltration using 0.45 pm Hilter


i II
Ic~L~ii


Figure 5-1. Determination of P sorption capacity using a single point isotherm based on a
procedure by Nair et al. (1998).

The filtrate was then analyzed for P using the Murphy-Riley procedure (Murphy and

Riley, 1962). Following the single point isotherm determination, the phosphorus sorption

was determined by subtracting the P added to the P in solution. The PSR and SPSC

determined by the single point isotherm were then calculated using the following

formulas:

(i) PSR = Melhich-1 P + (P sorption + Mehlich-1 P)

(ii) SPSC = (0-15 PSR) (P sorption + Mehlich-1 P) (mg P kg l)

When the SPSC becomes negative, i.e. PSR > 0.15, the soil is considered to be a net

source of P source.









Results and Discussion

Phosphorus Saturation Ratio

Phosphorus saturation ratios for the 0 5 cm soil depth in the silvopasture and treeless

pasture in Alachua were 0.17 and 0.19, respectively. Corresponding values for the

Suwannee site were 0.05 and 0.93, compared to 0.15 and 0.22 (Manatee), and 0.06 and

0.45 in Osceola (Table 5-1). These findings suggest that the surface soils under treeless

pasture have a greater potential to release P irrespective of the location or the soil order of

the land use. Concentrations of Water Soluble P (WSP) (Chapter 4) also indicate higher

values in the surface soils for treeless pasture compared to silvopasture, therefore

supporting the PSR results. Results of the study showed differences between the

silvopasture and treeless pasture systems at the 50 75, 75 100 cm soil depth in

Alachua, with the treeless pasture having the highest PSR values. In Osceola PSR values

were similar between the two systems for all sampling depths. In Manatee and Suwannee,

differences were observed between the two pasture systems throughout the soil profile.

The silvopasture sites showed lower PSR values than the treeless pasture sites, indicative

of a tree effect. This is consistent with the results from the water soluble P and the

Mehlich-1 P analysis obtained at the same study sites (see Chapter 4 for details).

Results of the study showed lower PSR values in the Spodosols as compared to the

Ultisols from depth 15 30 to 50 75 cm (Table 5-2). The lower PSR values observed in

the Spodosol sites may be attributed to a low soil Mehlich-1 P associated with Spodosols

characterized by a highly fluctuating and perched water tables resulting in subsurface

lateral flow in the upper soil layers which have the low P sorption capacities. As a result,

appreciable P losses will occur since drainage water may not pass through the spodic









horizon which has a high P retention capacity and may function as a P filter (Campbell,

1995).

Relationships between WSP and PSR for the soil surface (0 5 and 5 15 cm depths)

of the study sites at the four locations are shown in Fig. 5-2. The regression analysis

showed that the cubic model had the highest r-squared (r2 = 0.33) [Appendix] but for

practical reason the linear equation was chosen (r2 = 0.31). The low r-square observed for

the relation is mainly due to site variations, in particular from Alachua sites which have

shown evidence of the presence of phosphatic materials. Figure 5.2 shows a possible

changing point at a PSR value between 0.05 and 0.10 where the WSP sharply increases.

Nair et al. (2004), indicated a PSR of 0. 15 as a threshold PSR at which a sharp increase in

WSP can occur in the A horizon and subsurface horizons in Florida Spodosols.

Soil Phosphorus Storage Capacity

Remaining soil P storage capacity at 0 5 cm depth for the silvopasture and

treeless pasture in Alachua were -14.3 and -13.6 kg hal respectively. Corresponding

values were 15.75 and -43.66 in Suwannee, 0.6 and -2.5 in Manatee, and 26.2 and 14. 1 in

Osceola (Table 5-1). Results of the study showed significant differences between the

silvopasture and treeless pasture systems at depth 50 75, 75 100 cm in Alachua with

the treeless pasture having the lowest SPSC values. In Osceola, no differences were

observed throughout the soil profile. In Manatee no differences were observed between

the two pasture systems at 30 50 and 50 75 cm depths. In Suwannee SPSC values

were consistently higher for silvopasture systems throughout the entire soil profile.

The highly negative SPSC found in the deeper soil horizons in Alachua sites may

be explained by the presence of phosphatic materials in the deeper sampling depths of the

sites (Table 5-1). Ramnarine (2003) identified phosphatic soils in Alachua County,









Florida. These soils showed an increase in total phosphorus (TP) with depth, which is

different for most nonphosphatic soils.

At the Suwannee site total SPSC to one meter depth in the silvopasture and treeless

pasture were 337 and -59 P kg ha-l respectively. Corresponding values for the Manatee

site were 329 and 191 kg P ha-l compared to 657 and 926 kg P ha-l in Osceola), and

values of -36 and -542 kg ha-l in Alachua (Table 5-3). Differences between the two

pasture systems were significant for the Alachua, Suwannee and Manatee sites (Table 5-

3).

In general, the capacity of the soil to store additional P was lower in the treeless

pasture sites as compared to silvopasture sites (Figure 5-3). These findings along with the

results of the water soluble P (WSP) analysis (Chapter 3) suggest that the chances for P to

move out of the soil are greater under treeless pasture than under silvopasure, a tree-

based system. The "bulge" in the graph for the Spodosol in Osceola (Figure 5-4)

corresponds to the sandy E horizon at the 30 50 cm depth where the storage capacity of

the soil (calculated from a low PSR resulting from a low detectable Mehlich-1 P for the

reason explained earlier) would result in erroneous calculations for SPSC.

Estimation of Mehlich-1-based Soil P Storage Capacity (SPSC-M1) and Soil P
Storage Capacity-Single Point Isotherm (SPSC-ISO)

Results from the study sites showed that SPSC values determined with Mehlich-1

were lower than that of the single point isotherm method (Figure 5-5). In other words,

soil P storage capacity was underestimated when the Mehlich-1 method was used, as

opposed to the single point isotherm. This is consistent with a study by Chrysostome

(2005), who found that the Mehlich-1 method underestimated SPSC compared to the

oxalate method. In this study Mehlich-1 Fe and Al concentrations and oxalate extractable









Fe and Al concentrations were respectively 13 and 129 mg kg-l vs. 268 and 327 mg kg- ,

respectively.

Even though Mehlich-1-based soil P storage capacity and SPSC-ISO give different

values, they are nevertheless correlated. The regression analysis showed that the cubic

model had the highest r-squared (r2 = 0.57) [Appendix] but for practical reason the linear

equation was chosen (r2 = 0.50). The relationship between SPSC-MI and SPSC-ISO

(Figure 5-5) for the soil surface (0 5 and 5 15 cm depths) of the study sites is given by

the following formula:

SPSC-MI = 0.65 (SPSC-ISO) + 29.77; r2 = 0.50, p <.0001

The general relationship indicates that the Mehlich-1 SPSC calculation method

underestimates the remaining P storage capacity. The results are consistent with the

findings of a research conducted by Chrysostome (2005) in which a comparison was

made between SPSC calculated using Mehlich-1 and oxalate (SPSCoxalate). The research

concluded that SPSCoxalate resulted in a more accurate estimation of the remaining

capacity of a soil to sorb P than Mehlich-1-based SPSC, because oxalate is a more

powerful extractant than Mehlich-1 and extracts most of the reactive Al and Fe present in

the soil and representing its P sorption capacity. Chrysostome (2005) recommended the

use of a correction factor for Mehlich-1-based SPSC to compensate for the

underestimation of the remaining capacity of the soil to adsorb additional P using this

extractant. Several researchers (Harris et al., 1996; Indiati et al., 2002; Nair and Graetz,

2002; Lu and O'Connor, 2004) have observed reasonable estimations of the maximum

soil P sorption capacity using the single point isotherm method and therefore, calculating










SPSC using this method is likely to be comparable to the oxalate determination. Further

research is needed to validate this statement.



Conclusions

The presence of trees in pastures may reduce the chances of future additions of P to move

out of the soil. This, in turn, suggests that silvopastoral systems would provide a greater

environmental service in regards to water quality protection compared to treeless pastures

under comparable ecological settings. The single point isotherm method seems to better

estimate the remaining Soil P Storage Capacity as compared to the Mehlich-1 method.














Table 5-1. Mean values of phosphorus saturation ratio (PSR) and soil P storage capacity (SPSC) for soil profiles at the center of the
alley of the silvopasture and a treeless pasture sites in Alachua, Suwannee, Manatee and Osceola, Florida, USA


0.0099*
0.0257*
0.0168*
0.0226*
0.0012*
0.0048*


Silvopasture


Alachua


Alachua
Pasture Treeless
pasture
SPSC (mg P kg-')


Suwannee
Treeless
pasture


Silvopasture Treeless pasture


Silvo


Depth (cm)
0-5
5-15
15-30
30-50
50-75
75-100


PSR
0.18
0.19
0.17
0.17
0.22*
0.23*
Suwannee
Treeless pasture


PSR
0.42
0.28
0.24
0.16
0.09
0.06
Manatee
,pasture Treeless
pasture
SPSC (mg P kg ')
-2.4


P
0.9504
0.3184
0.6331
0.2256
0.0025
0.0039


0.19
0.17
0.16
0.14
0.14
0.15


0.5904
0.4309
0.9199
0.3228
0.0005
0.0005


-14.3
-17.1
-16.8
7.4
6.7*
-4.4*


-13.6
-38.0
-30.6
-51.1
-207
-260
Manatee
Treeless


0.04
0.06
0.04
0.03
0.02
0.02

Silvo


0.6*
6*


Silvopasture


Silvopasture


Ip


pasture
PSR
0.21*
0.05*
0.01*
0.01*
0.02*
0.04*
Osceola
asture Treeless
pasture
PSC (mg P kg-')
14.1
41.2
92.1
406
270
227


SPSC (mg P kg-')
16* -44
44* -61
74* -60
81* -6
83* 52
73 57
Osceola
Silvopasture Treeless pasture


Depth (cm)
0-5
5-15
15-30
30-50
50-75
75-100


P
0.0001
0.0001
0.0001
0.0001
0.0211
0.1203


P
0.0194
0.0002
0.0200
0.0427
0.0195
0.0001


0.14
0.02
0.01
0.00
0.00
0.01

Silvo


0.0115
0.0171
0.0365
0.2653
0.2234
0.0002


31
108
175*


Depth (cm) PSR P S
0-5 0.02 0.04* 0.0885 26.2
5-15 0.01 0.01 0.7654 45.3
15-30 0.01 0.00 0.8633 69.3
30-50 0.01 0.00 0.1284 144
50-75 0.02 0.00 0.0637 229
75-100 0.02 0.01 0.0724 234
(*Indicates significant P values within a given depth (p<0.05)].


P
0.2491
0.8594
0.5666
0.0706
0.5318
0.8971










Table 5-2. Mean values of phosphorus saturation ratio (PSR) for soil profiles of the
Spodosols (Osceoala and Manatee) and Ultisols (Alachua and Suwannee) in
Florida.
Depth PSR
(cm)
Alachua Suwannee Osceola Manatee
0-5 0.18b* 0.48a 0.26ab 0.18b
5-15 0.17ab 0.27a 0.06bc 0.04c
15-30 0.16a 0.20a 0.01b 0.006b
30-50 0.15a 0.15a 0.009b 0.003b
50-75 0.17a 0.07b 0.01c 0.01c
75-100 0.17a 0.05b 0.03c 0.02bc
*Statistical analysis by Waller-Duncan procedure. Mean values of soil characteristics
of silvopasture and treeless pasture within a given depth followed by the same
letter are not significantly different (p<0.05).




Full Text

PAGE 1

SILVOPASTURE AS AN APPROACH TO ENHANCING PHOSPHORUS AND NITRATE RETENTION IN PA STURELANDS OF FLORIDA By GERARD-ALAIN MICHEL 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 Gerard-Alain Michel

PAGE 3

To my beloved wife Daphne B. Michel, my precious son Alain-Ch ristian Michel, my adored mother Germaine D. Michel and to the loving memory of my dear father J.B. Gerard Michel.

PAGE 4

ACKNOWLEDGMENTS I am sincerely grateful to my major professor, Dr. Ramachandran P.K. Nair, for accepting me as his student and providing me valuable guidance and academic support. I am obliged to Dr. Vimala Nair, my co-advisor, for her important assistance and advice during the course of my research. My sincere appreciation goes to Dr. Jerry Sartain for his encouragement and dedicated help with the statistical analysis of my data. I would like to thank Dr. Johannes Scholberg and Dr. Michael Bannister for their encouragement and advice during my PhD studies. Special thanks are extended to the study participants, Fred Clark (Alachua), Rudy Garber (Manatee), Bruce Goff (Suwannee), and Harris Hill (Osceola), without whom this study would have not been possible. My appreciation also goes to several friends and collaborators for their assistance at various stages of this study, Dawn Lucas, Dr. Samuel Allen, Dr. Myrlene Chrysostome, the staff/students of the Soils Lab (SWSD, UF), Williams Reve (SWSD, UF), Solomon Haile (SFRC, UF), and Brian Roth (SFRC, UF). None of this would have been possible without the moral support of my beloved wife, Daphne B. Michel, my precious son Alain-Christian, and my dear mother, Germaine Delaquis Michel. My appreciation goes to my two sisters, Joyce Michel Dalpuppo and Abilene Michel Celestin, my uncle Jean-Claude Michel, and my godfather Ronald Maximilien, for their encouragement throughout my graduate studies. Finally, I wish to express my gratitude to the team of the Center for Subtropical Agroforestry. iv

PAGE 5

I would like to acknowledge USDA IFAFS/CSREES (Initiative for Future Agricultural and Food Systems grant through the Center for Subtropical Agroforestry (CSTAF), University of Florida) for supporting this study. v

PAGE 6

TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................iv LIST OF TABLES .............................................................................................................ix LIST OF FIGURES ...........................................................................................................xi ABSTRACT .....................................................................................................................xiii CHAPTER 1 INTRODUCTION........................................................................................................1 Background...................................................................................................................1 Problem Statement........................................................................................................3 Organization of the Dissertation...................................................................................4 Hypothesis and Objectives...........................................................................................5 2 REVIEW OF LITERATURE.......................................................................................6 Agroforestry and Silvopasture......................................................................................6 Historical Development of Agroforestry in the U.S. and Florida.........................6 Silvopasture: History, Characteristics and Potential...........................................10 Environmental Quality and Land Use Systems..........................................................12 Florida Soils and their Management....................................................................12 Fertilizer Use and Non-Point Source Pollution...................................................14 Soil Phosphorus as a Pollutant............................................................................15 Nutrient Management in Silvopasture........................................................................16 Nutrient Cycling in Agroforestry Systems..........................................................16 Soil Phosphorus in Pasturelands..........................................................................18 New Concepts of P Management........................................................................19 3 SOIL CHARACTERISTICS OF THE SILVOPSTURE AND TREELESS PASTURE STUDY SITES.........................................................................................26 Introduction.................................................................................................................26 Materials and Methods...............................................................................................26 Location of the Study Sites..................................................................................28 Estimation of Nitrogen and Phosphorus Loads in the Pasture Soils...................34 vi

PAGE 7

Major Tree and Grass Species at the Study Sites................................................34 Soils of the Study Sites........................................................................................36 Soil sampling................................................................................................36 Soil-nutrient characterization.......................................................................37 Soil texture and pH determination...............................................................38 Statistical Procedures...........................................................................................38 Results and Discussion...............................................................................................39 Soil Reaction (pH)...............................................................................................39 Soil Particle Size Analysis (Soil Texture)...........................................................40 Mehlich-1 Al and Fe Concentrations..................................................................41 Mehlich-1 Mg and Ca Concentrations................................................................43 Conclusions.................................................................................................................44 4 PHOSPHORUS AND NITROGEN CONCENTRATIONS UNDER A SLASH PINE (Pinus elliottii) + BAHIAGRASS (Paspalum notatum) SILVOPASTURE AND A TREELESS PASTURE.................................................................................54 Introduction.................................................................................................................54 Materials and Methods...............................................................................................56 Site Description...................................................................................................56 Statistical Procedures...........................................................................................57 Water Soluble P and Mehlich 1-P Determination...............................................57 Nitrate and Ammonium-N Determination...........................................................57 Results and Discussion...............................................................................................58 Water Soluble P Concentrations..........................................................................58 Mehlich 1-P Concentrations................................................................................59 Nitrate and Ammonium-N Concentrations.........................................................61 Conclusions.................................................................................................................62 5 SOIL PHOSPHORUS STORAGE CAPACITY UNDER A SLASH PINE (Pinus elliottii) + BAHIAGRASS (Paspalum notatum) SILVOPASTURE AND A TREELESS PASTURE..............................................................................................74 Introduction.................................................................................................................74 Material and Methods.................................................................................................77 Site Description...................................................................................................77 Statistical Procedures...........................................................................................77 Calculations of Parameters..................................................................................77 Soil P storage capacity calculated with Mehlich-1......................................77 Soil P storage capacity calculated with Single Point Isotherm Data............78 Results and Discussion...............................................................................................80 Phosphorus Saturation Ratio...............................................................................80 Soil Phosphorus Storage Capacity.......................................................................81 Estimation of Mehlich-1-based Soil P Storage Capacity (SPSC-M1) and Soil P Storage Capacity-Single Point Isotherm (SPSC-ISO)..................................82 Conclusions.................................................................................................................84 vii

PAGE 8

6 SUMMARY AND CONCLUSIONS.........................................................................92 APPENDIX: REGRESSION ANALYSIS.......................................................................95 LIST OF REFERENCES...................................................................................................97 BIOGRAPHICAL SKETCH...........................................................................................116 viii

PAGE 9

LIST OF TABLES Table page 2-1 Agroforestry practices in the Southeastern U.S.........................................................9 2-2 Water soluble phosphorus (WSP) concentrations as affected by soil depth for a native flatwoods site, a silvopastoral site, and an adjacent treeless pasture at Ona, Florida, USA. Q .................................................................................................23 3-1 Soil and climatic characteristics of the four farms selected for the study in four counties of Florida....................................................................................................30 3-2 Description of pasture systems at the four study locations in Florida.....................31 3-3 Estimates of nutrient loads from beef cattle excreta at the four study sites.............34 3-4 Statistical model used in the study Q ..........................................................................39 3-5 Mean values of Mehlich-1 Ca, Al, Mg, and Fe concentrations, pH, and percent sand, silt and clay for soil profiles at the center of the alley of the silvopasture and a treeless pasture sites in Alachua Florida, USA...............................................46 3-6 Mean values of Mehlich-1 Ca, Al, Mg, and Fe concentrations, pH, and percent sand, silt and clay for soil profiles at the center of the alley of the silvopasture and a treeless pasture sites in Osceola Florida, USA...............................................47 3-7 Mean values of Mehlich-1 Ca, Al, Mg, and Fe concentrations, pH, and percent sand, silt and clay for soil profiles at the center of the alley of the silvopasture and a treeless pasture sites in Suwannee Florida, USA............................................48 3-8 Mean values of Mehlich-1 Ca, Al, Mg, and Fe concentrations, pH, and percent sand, silt and clay for soil profiles at the center of the alley of the silvopasture and a treeless pasture sites in Manatee Florida, USA..............................................49 3-9 Mean values of Mehlich-1 Al and Fe concentrations and percent sand, silt and clay for soil profiles of the Ultisols and the Spodosols............................................50 4-1 Mean values of water soluble P (WSP), Mehlich-1 P (M-1 P), and nitrate and ammonium-N concentrations for different soil layers of soil profiles at the center of the alley of the silvopasture and a treeless pasture sites in Alachua and Suwannee Florida, USA...........................................................................................63 ix

PAGE 10

4-2 Mean values of water soluble P (WSP), Mehlich-1 P (M1-P), nitrate, ammonium concentrations of different soil layers for soil profiles at the center of the alley of the silvopasture and a treeless pasture sites in Manatee and Osceola, Florida, USA..........................................................................................................................64 4-3 Mean values of Mehlich-1 extractable-P (M1-P) concentrations for soil profiles of the Spodosols (Osceola and Manatee) and Ultisols (Alachua and Suwannee) in Florida..................................................................................................................65 5-1 Mean values of phosphorus saturation ratio (PSR) and soil P storage capacity (SPSC) for soil profiles at the center of the alley of the silvopasture and a treeless pasture sites in Alachua, Suwannee, Manatee and Osceola, Florida, USA85 5-2 Mean values of phosphorus saturation ratio (PSR) for soil profiles of the Spodosols (Osceoala and Manatee) and Ultisols (Alachua and Suwannee) in Florida......................................................................................................................86 5-3 Cumulative soil P storage capacity (SPSC) for the 0-1 m soil depth at the center of the alley of the silvopasture and a treeless pasture sites in Alachua, Suwannee, Manatee and Osceola, Florida, USA.....................................................87 x

PAGE 11

LIST OF FIGURES Figure page 1-1 U.S. consumption of nitrogen and phosphate for 1960-2003....................................2 3-1 Soil orders of Florida ...............................................................................................27 3-2 Location of study sites in Florida. The Osceola and Manatee sites are on Spodosols and the Alachua and Suwannee sites are on Ultisols..............................29 3-3 Silvopastoral system at Osceola County, Florida.....................................................32 3-4 Treeless pasture at Manatee County, Florida...........................................................32 3-5 Profile of an Ultisol..................................................................................................33 3-6 Profile of a Spodosol................................................................................................33 3-7 Grid sampling at the Suwannee location. Twenty four soil profiles were collected at each field (silvopasture and treeless pasture) at each location, and with 6 depths (0 5, 5 15, 15 30, 30 50, and 50 75 and 75 cm) per profile. The small circles, representing sampling points, illustrate the layout of the grid sampling method used in each sites at each location..................................37 3-8 Aerial photo of the Alachua location.......................................................................51 3-9 Aerial photo of the Osceola location........................................................................52 3-10 Aerial photo of the Manatee location.......................................................................53 4-1 Water soluble P (WSP) concentration at different soil layers for silvopasture and treeless pasture at the Alachua and Suwannee field sites featuring Spodosols. Error bars indicate standard errors (n = 24).............................................................66 4-2 Water soluble P (WSP) concentration by depth in the silvopasture and treeless pasture sites in Osceola and Manatee. Error bars designate standard errors............67 4-3 Mehlich-1 P concentration by depth in the silvopasture and treeless pasture sites in Alachua and Suwanneee. Error bars designate standard errors...........................68 4-4 Mehlich-1 P concentration by depth in the silvopasture and treeless pasture sites in Osceola and Manatee. Error bars designate standard errors................................69 xi

PAGE 12

4-5 Nitrate concentration level by depth in silvopasture and treeless pasture sites in Alachua and Suwannee on Ultisols. Error bars designate standard errors...............70 4-6 Ammonium-N concentrations levels by depth in silvopasture and treeless pasture sites in Alachua and Suwannee (Ultisols sites). Error bars designate standard errors..........................................................................................................71 4-7 Nitrate-N concentration for different soil depth for silvopasture and treeless pasture systems for filed sites in Manatee and Osceola (Spodosols). Error bars designate standard errors..........................................................................................72 4-8 Ammonium-N concentrations levels for different soil depth in silvopasture and treeless pasture sites in Manatee and Osceola on Spodosols. Error bars designate standard errors..........................................................................................................73 5-1 Determination of P sorption capacity using a single point isotherm based on a procedure by Nair et al. (1998)................................................................................79 5-2 Relationship between phosphorus saturation ratio (PSR) and Water soluble P (WSP) for surface soils (0-5 and 5-15 cm) of the study sites in Florida..................88 5-3 General trend in soil P storage capacity (SPSC) in the center of the alley of the silvopasture and a treeless pasture sites in Florida...................................................89 5-4 Soil P storage capacity (SPSC) by depth at the center of the alley of the silvopastures and treeless pasture sites in Osceola (error bars represent one standard error from the mean, n=24)........................................................................90 5-5 Relationship between soil P storage capacity single point isotherm (SPSC-ISO) and soil P storage capacity-Mehlich-1 (SPSC-M1) for the surface (0 5 and 5 15 cm depths) of the study sites in Florida...............................................................91 xii

PAGE 13

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 SILVOPASTURE AS AN APPROACH TO ENHANCING PHOSPHORUS AND NITRATE RETENTION IN PASTURELANDS OF FLORIDA By Gerard-Alain Michel December 2006 Chair: Ramachandran P.K. Nair Cochair: Vimala D. Nair Major Department: Forest Resources and Conservation Phosphorus (P) and nitrate-nitrogen (N) loss from sandy soils that are predominant in the 1.4 million ha of pastureland in Florida is a major cause of water pollution. We hypothesized that soil P and N loss would be lower from silvopastoral systems than from treeless pastures because soil P and N removal by the combined stand of trees and pasture would be more than that of treeless pasture. Four slash pine (Pinus elliottii) + bahiagrass (Paspalum notatum) silvopastoral systems located respectively in Osceola, Manatee, Suwannee, and Alachua counties in Florida were selected. The Manatee and Osceola soils are Spodosols, and the Alachua and Suwannee soils are Ultisols. Soil samples were collected depth-wise from 0 to 100 cm. Soil P storage capacity (SPSC), the maximum amount of P that can be safely applied to a soil before it becomes an environmental concern, was calculated. Water-soluble P concentrations in the 0 5 cm soil layer were 4 and 10 mg kg -1 for the silvopasture and treeless pasture in Suwannee, 11 and 23 mg kg -1 in Manatee, 7 and 11 mg kg -1 in Osceola and in 6 and 11 mg kg -1 in Alachua. Total SPSC xiii

PAGE 14

in the upper one meter depth were 342 and -60 kg ha -1 respectively in the silvopasture and treeless pasture in Suwannee. The corresponding values were 329 and 191 kg ha -1 in Manatee, 657 and 926 kg ha -1 in Osceola, and -36 and -542 kg ha -1 in Alachua. The results suggest that P buildup within the soil profile was less in silvopastures than in treeless pastures. Therefore silvopasture systems can be expected to provide a greater environmental service in regard to water quality protection compared to treeless pastures under comparable ecological settings. In a related investigation, the single point isotherm method seemed to better estimate SPSC compared to the Mehlich-1 method. Although N followed similar trends differences between the two pasture systems were less conspicuous. This may be related to the difficulties in monitoring N, which is more labile than P; the highly dynamic soil nitrogen levels might have confounded the results. xiv

PAGE 15

CHAPTER 1 INTRODUCTION Background Phosphorus (P) and nitrogen (N) are critical to all forms of plant life. Nitrogen is an essential component of basic biological building blocks such as amino acids and thus their polymers: peptides and proteins. Nitrate-N, a negatively charged ion, is not adsorbed by the negatively charged colloids that dominate most soils (Brady and Weil, 2002). As a result, it is susceptible to off-site losses via deep leaching (the downward movement of nutrient below the effective root zone with percolating water) or in runoff (overland and lateral subsurface flows). Nitrates (as in fertilizers) are highly soluble in water and therefore move readily via leaching or runoff into the groundwater below. According to Yadav (1997), 68% of N0 3 -N accumulation occurred outside the crop-rooting zone and 20% of nitrate-N accumulation in the crop-root zone in the soil profile moved into groundwater annually. Nitrate concentrations above 10 mg L -1 in drinking water may lead to the blue baby syndrome (United States Environmental Protection Agency [USEPA], 2006). Phosphorus is involved in many vital functions in plants such as energy storage and transfer, photosynthesis, respiration, protein and nucleic acid synthesis, and ion transport across cell membranes. Fertilization of crops represents the largest proportion of P used in agriculture (Wood, 1998). Phosphorus and nitrogen fertilizers consumption in the U.S. for 1960 to 2003 is shown in Figure 1-1(United States Department of Agriculture/Economic Research Service [USDA/ERS], 2006a). 1

PAGE 16

2 0.02,000.04,000.06,000.08,000.010,000.012,000.014,000.01960196219641966196819701972197419761978198019821984198619881990199219941996199820002002Year (ended June 30)Primary nutrient content (1,000 nutrient tons) Nitrogen Phosphate Figure 1-1. U.S. consumption of nitrogen and phosphate for 1960-2003. Sources: USDA/ERS (U.S. Department of Agriculture / Economic Research Service), 2006a. In areas of intensive crop and livestock production, continual P applications as mineral fertilizer and manure in excess of crop removal rates often result in P accumulation (Sharpley et al., 1994; Tiessen, 1995). Phosphorus exists in agricultural fields either in a dissolved form or as attached to soil particles. Phosphorus is often the most limiting nutrient in surface water systems, and even small additions of P can have a great influence on primary production in these waters (Correll, 1998). Concern about the environmental fate of P has arisen from the evidence that land-applied animal wastes and fertilizers can potentially promote eutrophication of water bodies (Allen, 1987; Breeuwsma and Silva, 1992; Lemunyon and Daniels, 1997; Correll, 1998; Parry, 1998). Phosphorus concentration above 0.02 mg L -1 in lake water may result in increased alagal growth in P-limited systems. Excessive algae growth will decrease dissolved oxygen

PAGE 17

3 levels and reduced water transparency. This process is referred to as eutrophication (Sharpley et al., 1999), and is directly associated with nutrient enrichment of surface waters. Thus, even if P losses from agriculture were small from an agronomic point of view, they may still constitute a serious potential environmental problem. Reducing excess loadings of P and NO 3 -N into both surfaceand ground-water is an important environmental goal in the southeastern USA. Problem Statement Improved nutrient management strategies are required to maximize plant uptake of nutrients and thus minimize potential losses by runoff and leaching in order to maintain surface and ground water quality. The role of agroforestry systems, especially the effect of trees within agroecosystems, on nutrient retention and/or soil storage is therefore of interest. Due to their deep rooting systems, trees may provide a mechanism for improved nutrient capture, by which tree-based systems act as a safety net for capturing nitrates and other nutrients that would otherwise leach below the root zone of agronomic crops (van Noordwijk et al., 1996; Rowe et al., 1999). Although this has been demonstrated in tropical agroforestry systems (Seyfried and Rao, 1991; Horst 1995; Young, 1997; Lehmann et al., 1999; Rowe et al. 1999; van Noordwijk et al., 1999; Rowe et al., 2005), such nutrient movement studies in temperate agroforestry systems are rare, and whatever little that has been done has focused on riparian buffers. Nevertheless, these few studies have demonstrated the effectiveness of riparian buffer systems in reducing non-point source pollution and thereby improving water quality (Schultz et al. 2000; Schultz et al. 2004). Little information is available on the environmental implications of other agroforestry practices such as silvopasture in regard to their impact on surface and groundwater quality.

PAGE 18

4 Silvopasture, in the North American and other temperate-zone context, refers to intentional combination of trees, forage plants and livestock into an integrated, intensively-managed system (Nair, 1993; Garrett et al., 2000; Nair et al., 2004); trees in such systems are managed for high-value sawlogs and at the same time an annual income is generated from livestock grazing. Silvopasture is considered to be the most promising agroforestry practice in the southeastern United States (Workman and Allen, 2004). According to Zinkhan and Mercer (1997), silvopasture is the most common form of agroforestry in the southern United States. Studies from across the southern pine region (Louisiana, Mississippi, and Georgia) report the possibility of productive livestock grazing while maintaining, or even improving, high value timber production (Workman and Allen, 2004). With the realization about its environmental amelioration potential and greater appreciation of the societal value of ecosystem benefits such as carbon sequestration and also providing unique wildlife habitats, the practice of silvopasture is now receiving increased interest as an innovative and viable strategy for agroecosystem management (Nair et al., 2005). Based on this, the focus of this study will be on the role of silvopastoral systems in reducing P and N losses from the soil. Organization of the Dissertation The research is presented in five chapters. Chapter 2 is a review of the current literature on agroforestry and silvopasture in the U.S. and Florida. This chapter outlines the problems associated with land-use changes and their environmental impacts in Florida, and discusses the development of new concepts of P management. Chapter 3 presents the main characteristics of the study sites in Florida in terms of the soil chemical and physical characteristics as well as the plant species found in the two pasture systems

PAGE 19

5 (treeless pasture and silvopasture). It also includes a detailed account of the soil sampling methodology used in the study. Chapter 4 contains the results of the attempt to test the safety net hypothesis by measuring the soil phosphorus and nitrate-N concentrations throughout the soil profile for both the silvopasture and the treeless pasture sites at four locations in Florida and for two representative soil orders (Spodosols and Ultisols). An analysis of the remaining soil P storage capacity of the study sites at the four locations with comparison between the treeless pastures and the silvopastures in the two soil orders is the subject of Chapter 5. The final chapter (Chapter 6) includes the synthesis and conclusion of the results and is followed by recommendations for future research needs. Hypothesis and Objectives The overall hypothesis of the study is that the presence of trees in silvopastoral systems would enhance soil nutrient retention and/or crop removal of P and nitrates compared to treeless pastures and thereby reduce potential loadings rates of surface waters. The study objectives are to (i.) Quantify water soluble P (WSP), Mehlich-1 P, and NO 3 and NH 4 + in the surface and subsoil horizons in manure-impacted soil profiles of silvopastoral and treeless pasture systems at four different locations (two soil orders) in Florida; (ii.) Determine the relation between WSP and P saturation ratio (PSR M1 ) in the surface and subsoil horizons of the four study sites; (iii.) Estimate the remaining soil P storage capacity (SPSC) of the two pasture systems sites at the four locations; (iv.) Assess differences in SPSC estimates by comparing two methods (Mehlich-1 and single-point isotherm).

PAGE 20

CHAPTER 2 REVIEW OF LITERATURE Agroforestry and Silvopasture Historical Development of Agroforestry in the U.S. and Florida Drastic changes have occurred in land-cover use during the past 150 years in the Eastern United States. Forests, woodlands and grasslands were severely affected by land clearing for timber exploitation and agriculture from the middle 1800s to the 1930s (Ware, 2002). Between 1973 and 2000, land cover statistics showed that forests remained the predominant cover, despite the decline from a high of 28.5 % in 1973 to a low of 24.3 % in 2000. Agriculture saw a similar decline from 10.8 % to 8.0 % during the same period (U.S Geological Survey [USGS], 2006). Today, crop, pasture lands and woodlands represent 3.9, 4.0, and 3.4 millions hectares, respectively, in Florida, Georgia, and Alabama (USDA/ERS, 2006b). Agriculture and forestry combined represent the most important economic activity in the states of Alabama and Georgia and the second (only to tourism) in Florida. In spite of this dominant role of agriculture and forestry, s ubstantial population growth has occurred, causing an expansion of urban and developed land. Between 1970 and 2000, the population of Florida increased by 235 percent, from 6.8 million to 16.0 million people (NPG, 2006) At present, the State of Florida has over 17 million residents (Florida Division of Emergency Management [FDEM], 2006). The urbanization of rural lands and the subsequent increase in land prices constitute a striking testimony of the development of suburbs and cities across rural areas of America. The main consequence of 6

PAGE 21

7 urbanization is the diversion of food producing lands or lands reserved for forests and natural habitats, to accommodate expanding populations (Cordell et al., 1998). Agricultural intensification of croplands, forest and livestock systems has been very articulated during recent decades. The mid-1900s brought about an unprecedented increase in food and fiber yields per unit area resulting from a steady increase in the use of chemical fertilizer, pesticides and irrigation while a decline in the rate of land clearing was observed (Workman and Allen, 2004). By the 1970s, row crops generated production surplus and higher incomes for farmers. During the 1980s, intense mechanization and farm inputs augmented the number of farm loans, which led to an increase in farm debt and eventual large-scale loss of farms to foreclosures (Fitchen, 1991). Today, excessive input costs and low profit margins resulting from strong competition from corporate-run farms and inexpensive foreign products are driving small farms to abandon farming activities (Workman and Allen, 2004). Concerns regarding the effects of large-scale transformation of ecosystems on the natural structure and function of land and water resources have been raised and continue to be important research and development issues (Lapp et al., 1998; Vitousek et al., 1997; Sampson and Hair, 1990; Savory, 1988). A decline in industrial timberland acreage of more than 0.5 million hectares has been observed in Florida and Georgia, since 1989 (Workman and Allen, 2004). Much of this timberland is now owned by private corporations that are expected to carry on a management plan for wood products (Conner and Hartsell, 2002). As urbanization follows its course and an increasing number of people desire recreations in natural settings, there will be a growing need for trees both within and outside forests (Long and Nair, 1999; Leakey, 1998).

PAGE 22

8 With an increase in population growth rate in the southern states and a decline in industrial timberland, there have been growing concerns about the environmental problems associated with potential fertilizers and pesticides pollution, soil erosion, pest problems, and loss of biological diversity. Consequently, there is increased interest in the use and adoption of land-management practices that increase both the aesthetic and recreational value of lands while protecting and conserving the natural resource base (Bliss et al., 1997; Teasley et al., 1997). Agroforestry, the intentional growing of trees with crops, pasture and/or livestock, affords farmers with an alternative land-use practice with potential for mitigating some of the environmental and economic setbacks associated with modern agriculture (Nair, 1993). Agroforestry is now receiving increased attention from researchers, landowners, government and private agencies in North America (World Agroforestry Centre [ICRAF], 2000; Lassoie and Buck, 2000; Garrett et al., 2000). An essential characteristic of agroforestry is the wide array of land-management options it offers for preserving natural resources and generating additional income. Agroforestry systems, via the integration of trees with crops and/or livestock on the same land, could provide numerous environmental services. These services include reduction of soil erosion and nutrient losses, enhancement of inherent soil fertility, water infiltration and groundwater recharge, protection against wind, snow, noise, and odor and other nuisances, and creation of attractive and healthier landscapes (Ewel, 1999; Jordan, 1998, Daily, 1997; Leakey, 1996). By diversifying the farm enterprises, agroforestry can also help landowners to improve the resilience of their system and to minimize risks associated with income loss from price fluctuation, crop failure or other unexpected

PAGE 23

9 problems. Opportunities for intensifying the use of agroforestry practices, and the benefits that result, are gradually rising in the southern U.S. (Workman and Allen, 2004). Agroforestry in the Southeastern U.S. can be divided into five major practices or land-use techniques (Merwin, 1997; Garrett et al., 2000; Nair, 2001): Alley Cropping, Forest Farming, Riparian Forest Buffers, Silvopasture, and Windbreaks (Table 2-1). Table 2-1. Agroforestry practices in the Southeastern U.S. Silvopasture Silvopasture is the intentional combination of trees, forage plants and livestock together as an integrated, intensively-managed system. Silvopasture can provide profitable opportunities for softwood or hardwood timber growers, forest landowners, and livestock producers. Alley cropping Alley cropping is the cultivation of food, forage or specialty crops between rows of trees. It is a larger version of intercropping or companion planting conducted over a longer time scale. Alley cropping can provide profitable opportunities for row crop farmers, hardwood timber growers, nut growers and Christmas tree growers. Riparian Forest Buffers Riparian forest buffers are strips of trees, shrubs and grass planted between cropland or pasture and surface water courses. Buffers protect water quality, reduce erosion and flooding. Riparian forest buffers can provide beneficial opportunities for row crop farmers, ranchers, horticulturists, and dairy and livestock producers. Forest Farming Forest farming is the intentional cultivation of edible, medicinal or decorative specialty crops beneath native or planted woodlands that are managed for both wood and understory crop production. It does not include the gathering of naturally-occurring plants from native forests, also known as wildcrafting. Forest farming can provide profitable opportunities for forest and woodland owners, nut growers, sugar maple growers, and herb growers. Windbreaks Windbreaks are linear plantings of trees and shrubs designed to enhance crop production, protect people and livestock, and benefit soil and water conservation. Windbreaks can provide valuable opportunities for vine and tree fruit growers, row crop farmers, livestock producers, and rural homeowners. Source: AFTA (2006)

PAGE 24

10 Silvopasture: History, Characteristics and Potential Grazing in pine (Pinus spp.) forests in the Lower Costal Southern United States dates back to around 1520, when cattle were first introduced in Florida by Ponce de Leon from Spain (Lewis, 1983). In the early 1950s, pine trees were first planted in improved pastures as part of the Conservation Reserve Soil Bank Program (Nowak and Long, 2003). At present silvopasture is becoming increasingly attractive to non-industrial private forest landowners and livestock operators in southeastern United States who want to diversify their enterprise (Nair et al., 2005). Silvopasture has historically included use of shade trees in pasture grazed orchards or woodlands, and as rangelands that include a managed tree or shrub component (Clason and Sharrow, 2000; Robinson and Clason, 1997; Williams et al., 1997). In the Southeast, silvopasture has generally included forest grazing with cattle, such as flatwoods rangeland (Pearson, 1997), pine managed for turpentine and sawlogs mixed with forage systems (Byrd et al., 1984; Cary, 1928), and tree pasture practices with pecan (Carya illinoensis K. Koch) (Reid, 1991). Silvopastoral systems integrating tree, forage and livestock production are the most common agroforestry practices found in the United States and Canada with the largest area of grazed forests occurring in the southern and southeastern United States (Clason and Sharrow, 2000). Burton and Scarfe (1991) suggested associations with goats for both meat production and vegetation management. According to Lewis and co-workers (Hart et al., 1970) the association of southern pine production and beef cattle with improved pastures constitutes an opportunity for the production of multiple products. Increased economic benefits may result from the integration of forestry with ranching by helping buffer year-to-year income fluctuation through the sale of forest products and hunting leases generated by the creation of

PAGE 25

11 wildlife habitat. Studies on silvopasture in the Southeastern United States have reported the economical viability of such enterprise (Lundgren et al., 1983; Clason, 1995; Grado et al., 2001; Stainback and Alavalapati, 2004; Alavalapati et al., 2004). According to Stainback et al. (2004), the combination of slash pine (Pinus elliotii) with cattle production could become competitive with conventional ranching in Florida when environmental costs and benefits are considered in the analysis. Research on warm season forage under pines in South Georgia (that began in the 1940s) showed that Pensacola bahiagrass (Paspalum notatum) as the most shade tolerant of the 23 grasses studied (Lewis and Pearson, 1987; Pearson, 1975). Several legume species have been recognized as having promising potential for production under partial shade (McGraw et al., 2001). In a silvopasture study comparing a double row system of pines (3.1m between rows and 1.2 m between trees and 12.2 m wide alleys) to a single row system (3.1 m x 3.7 m), the double row system yielded more forage and as much wood (Lewis et al., 1985). At present the double row system remains the most popular spacing for silvopasture across the region (Clason and Sharrow, 2000). Recently developed varieties of bahiagrass, such as Tifton-9 and Argentine may perform as better warm season forage component in silvopasture systems (Nowak and Blount, 2002). Clason (1995) reported higher internal rate of return in silvopastoral practice (13%) than managed timber (9%) or open pasture (6%) in the state of Louisiana. Grado et al. (2001) reported that both silvopasture and pasture had positive cash flows with that for pasture being highest during a short evaluation period. In Georgia, pine growth improved with controlled grazing (Lewis et al., 1985), and loblolly-cattle-forage systems in the Coastal Plain may have a 70% greater net present value per unit area (Dangerfield and Harwell,

PAGE 26

12 1990) than a pure forestry operation It has been demonstrated in Alabama that when mimosa ( Albizia julibrissin ) and leucaena (Leucaena leucocephala) are grown without fertilizers, they can be cut for fodder at 6 to 8-week intervals (Bransby et al., 1996). A livestock-forage-Christmas tree production system could be another viable silvopasture mixture (Pearson et al., 1990). According to McGowan et al. (1999), there is also a growing interest in goat production in this region. In the South, forest industry has used goats as an alternative to using chemical or mechanical means of weed control in pine plantations (Solaiman and Hill, 1991). Studies conducted in Arkansas and Alabama suggest that goats can play a key role in vegetation reduction, especially kudzu (Pueraria lobata), during site preparation for pine plantations (Pearson and Martin, 1991; Bonsi et al., 1991). Environmental Quality and Land Use Systems Florida Soils and their Management Florida soils are characterized by coarse soil textures and acidic conditions. P loading of surface and ground waters following transport from dairy operations and heavily manured pasturelands in these porous soils by P-rich animal manures constitute a serious environmental concern .The Florida Phosphorus Index, a field-based indexing system developed to rank the vulnerability level of farmlands as sources of P loss in surface runoff water (University of Florida/Institute of Food and Agricultural Sciences [UF/IFAS], 2006), take into account both phosphorus transport characteristics and management practices. Management factors include soil test P, P application method, and source and rate of P application (Nair and Graetz, 2004). Phosphorus based manure applications are being increasingly incorporated into farm nutrient planning program (Sharpley et al., 2004) because nitrogen-based manure applications have resulted in

PAGE 27

13 continuous P-accumulation in soils. According to Sharpley et al. (2004), the planning includes the selection, timing, and implementation of source and transport best management practices at field, farm, and watershed scales. Some of the practices used for controlling erosion and runoff (the transport factor) include, conservation tillage, contour farming, constructed wetlands, and vegetative filter strips. Conservation tillage is a management practice that provides efficient and effective erosion control and it can improve soil properties and soil quality (United States Department of Agriculture/Natural Resources Conservation Service [USDA/NRCS], 1999).This practice can improve soil structure thereby resulting in higher infiltration and percolation rates, increased water holding capacity and, enhanced resistance to erosion. Vegetative filter stripes are narrow rows of permanent vegetation (usually grasses or shrubs) planted on the contour can be used to slow down runoff, trap sediment, and eventually build up natural or living terraces (Brady and Weil, 2002). Constructed wetlands (CW) have been used for decades for the treatment of domestic ( Kadlec, 1985 ) and industrial wastewaters and agricultural runoff ( Higgins et al., 1993 ). These systems are useful for reducing suspended solids, ammonium-N, P, biochemical oxygen demand and chemical oxygen demand in sewage ( Tanner et al., 1995 and Von Felde and Kunst, 1997 ). Measures aimed at regulating the P accumulation involve attaining a suitable balance between P imports and exports. Suitable management practices may include, composting, improved livestock feeding, suitable rate, method, and timing of land application of manure, along with physical and chemical treatments of manure (Chrysostome, 2005). Drinking water treatment residuals (WTRs), non-hazardous

PAGE 28

14 materials that are obtained at no costs from drinking-water treatment plants (Makris et al., 2005), offer a cost-effective chemical soil amendment of P-impacted soils. These materials reduce soluble P in soils with low P-sorption capacities, without affecting soil fertility (Codling et al., 2002). Similar findings have also been reported by Novak and Watts (2005). Fertilizer Use and Non-Point Source Pollution The U.S. Environmental Protection Agency (EPA, 1996) has identified agricultural non-point source pollution as the major cause of stream and lake contamination that prevents the attainment of the water quality goals established in the Clean Water Act. The southeastern U.S. is home to a vast number of water bodies; Florida alone has over 7,800 lakes, about 10,000 km 2 (4,000 sq. miles) of estuaries, and over 80,000 km (50,000 miles) of rivers and streams, including major water bodies such as Lake George, Lake Okeechobee, the Everglades, and the St. Johns and Kissimmee rivers (Florida Department of Environmental Protection [FDEP], 2003). In addition, the Floridian aquifer system, underlying almost all of Florida and portions of Alabama, Georgia and South Carolina, occupies a total area of about 260,000 km 2 (100,000 sq. miles), and supplies over 11.4 billion liters (3 billion gallons) of water per day for all uses throughout the region (Johnson and Bush, 2002). There is increased awareness of non-point source pollution of waters from agricultural chemicals in drainage and runoff water (Tiessen, 1995). This issue is particularly important in Florida and surrounding areas, as the karst geography of the region is shaped by vast groundwater reserves that are sensitive to nutrient build-up, and has led to a growing demand for information regarding the impact of different land-use systems on surface water quality. Most of the common nitrogen and phosphorus fertilizers are water-soluble. In the high rainfall areas, especially on sandy

PAGE 29

15 soils, these nutrients can be leached into the groundwater and washed into waterways, even after short periods (months and years, rather than decades) (Summers, 2002). In the U.S., only 18% of the N-fertilizer inputs leaves farms with the produced crop, meaning that on average, 174 kg ha -1 of surplus N is left behind in croplands each year (Ecological Society of America [ESA], 1998). Animal manures and fertilizers, pose a serious threat to surface and groundwater quality in major river (Suwannee River and Lake Okeechobee) basins of Florida. Elevated NO 3 -N concentrations have been measured in water from both domestic and monitoring wells in the Suwannee River Basin (Katz et al., 1999), exceeding the maximum contaminant level of 10 mg L -1 set by the U.S. Environmental Protection Agency. Nitrate-N concentrations above this limit can cause methemoglobinemia or blue baby syndrome in infants and elderly adults (Mueller and Helsel, 1996). These increased nitrate concentrations in these waters have been attributed to dairy wastes (Andrews 1994), poultry operations (Hatzell, 1995), and excessively high fertilizer applications (Katz et al., 1999). Soil Phosphorus as a Pollutant In most U.S. watersheds, P export occurs mainly in surface runoff rather than in subsurface flow. On the other hand, in some regions, particularly the Gulf coastal plains and Florida, as well as in fields with subsurface drains, P can be transported in drainage waters (Sharpley et al., 1999). Soils with high water infiltration rates and low nutrient retention capacity, such as sandy soils and soils with low activity clays and low organic matter content, are particularly prone to nutrient leaching (Lehmann and Schroth, 2003). The Spodosols of Florida receiving significant loadings of animal manure, on the other hand, are more prone to subsurface leaching of P (Nair and Graetz, 2002). The P in organic manures has been shown to be more susceptible to leaching than that in inorganic

PAGE 30

16 fertilizers (Eghball et al., 1996), and P leaching from grazed grassland tends to increase when the stocking density is high, possibly due to a greater amount of soluble organic P being released from dung (Beauchemin et al., 1999). The injection of manure slurry also greatly increases the leaching of P (Tunney et al., 1997). Phosphorus transported from agricultural soils can accelerate eutrophication (Sharpley et al., 1994). This phenomenon leads to increased costs and difficulty of water purification for drinking purposes (to remove odor, turbidity, and color) and the replacement of high quality edible fish, submerged macrophytic vegetation, and benthic organisms with coarse, rapid-growing fish and algae, along with the control of proliferating noxious aquatic plants. Nutrient Management in Silvopasture Nutrient Cycling in Agroforestry Systems Research conducted in tropical agroforestry systems has demonstrated the potential for nutrients cycling by deep-rooting trees. According to Kang (1997), the presence of woody species in an alley cropping production system has been shown to contribute to (1) nutrient recycling, (2) reduction in soil nutrient leaching losses, (3) stimulation of higher soil faunal activities, (4) soil erosion control, (5) soil fertility improvement and (6) sustained levels of crop production. Horst (1995) reported lower nitrate concentrations in the soil solution and consequently reduced leaching under hedgerow intercropping with Leucaena leucocephala and annual food crops than in the agricultural control treatments in southern Benin. Lehmann et al. (1999) measured lower nutrient leaching under an Acacia salignasorghum (Sorghum bicolor) intercrop than under pure sorghum with runoff irrigation in northern Kenya. Das and Chaturved (2005) studied nutrient concentrations in plant and soil and their corresponding cycling rates of cycling in poplar (Populus deltoides)-based agroforestry systems at Pusa, Bihar, India. The authors

PAGE 31

17 concluded that the poplar-based agroforestry system was more sustainable than other single-crop systems in terms of in terms of soil nutrient status. A study conducted by Tapia-Coral et al. (2005) in Central Amazonia, Brazil showed that the litter of the agroforestry systems had lower carbon: nutrient ratios than the litter in the secondary forest control, indicating a faster nutrient recycling in the agroforestry systems. Manezes et al. (2002) found that the preservation of native trees or introduction of exotic tree species in Cenchrus ciliaris pastures in semiarid northeastern Brazil significantly affects microclimate and the dynamics of litter and soil nutrients, and may contribute to enhanced nutrient cycling in these systems. Other studies conducted in temperate settings have also shown similar results. In a pecan (Carya illinoensis)cotton (Gossypium hirsutum) alley cropping system in northwestern Florida, Allen et al. (2004) found that N-leaching rates were lower in a non-barrier treatment (competition treatment) compared to a barrier (competition-reduced treatment) treatment. Cumulative nitrate leaching below the 0.3 and 09 m depth were 64 and 13 kg ha -1 for non-barrier treatment, compared to 122 and 46 kg ha -1 for respective depths in barrier treatment. It appears that the tree roots captured N in the non-barrier treatment, resulting in lower leaching rates below the root zone of the cotton crop. Furthermore, tree water uptake, in addition to cotton water uptake in the non-barrier treatment, may have decreased overall water drainage in comparison to the barrier treatment, thereby influencing nutrient loadings rates. In the same study Allen et al. (2004) reported a greater nitrification and mineralization rates in no-barrier treatment than in a barrier treatment due to higher soil N, resulting from reduced nutrient uptake by cotton. Udawatta et al. (2002) found in a paired watershed study conducted in

PAGE 32

18 northeastern Missouri, consisting of agroforestry (trees plus grass buffer strips), contour strips (grass buffer strips), and control treatments with a corn (Zea mays)soybean [Glycine max] rotation, that nitrate-N loss was reduced 24 and 37% by contour strip and agroforestry treatments. Corresponding treatments also reduced total P loss by 8 and 17% on contour strip and agroforestry watersheds. Soil Phosphorus in Pasturelands Pasturelands in Florida receiving significant load of manure are subject to subsurface leaching of phosphorus due to the sandy nature of the soils. Many P compounds present in manure are water soluble to begin with and have a potential to move with rain or irrigation water (Gerritse, 1977). Furthermore, animals are known to utilize P inefficiently in feed (only 30% is retained); therefore most of the P entering livestock operations end up in manure (Sharpley and Moyer, 2000). In many areas of concentrated animal production, manures are generally applied based on the crop N-requirements to avoid groundwater quality problems associated with leaching of excess N. This frequently leads to P build up in excess of the sufficiency level for optimal crop yields, which in turn can enhance the potential P runoff and leaching (Haygarth et al., 1998; Heckrath et al., 1995; Sharpley et al., 1996). With time, the organic P will be absorbed or hydrolyzed to form inorganic phosphate-containing compounds, which may be subject to leaching and run-off as soluble P compounds (Zublena, 1995). Hedley et al. (1982) found that the major portion of manure P was soluble in weak extractants such as water and bicarbonate. Sharpley and Moyer (2000) assessed phosphorus forms in manure for dairy, poultry and swine and P leaching potential using a simulated rainfall. Inorganic P in the manures accounted for 63 to 92% of total P. Phosphorus leaching from the different manures by simulated rainfall showed a close relationship to water soluble P

PAGE 33

19 concentration. According to Sharpley et al. (2004), there is a general shift in soil P chemistry with the application of manure from Fe-P and Al-P to Ca-P compounds, resulting in a relatively greater extraction of P by the Mehlich-3 but a lower water extractability of soil P. This shift has important environmental implications for soil P testing. For example, the fact that Mehlich-3 P has been shown to overestimate potential for P-losses in surface flow from heavily manured soils could be explained by the extraction of Ca-P minerals that are insoluble in water. New Concepts of P Management Soil test P (STP) is the most widely used means to estimate the potential risk for P leaching fom the soil (Maguire and Sims, 2002). Diagnostic soil test P values were initially developed to evaluate the probability of obtaining a positive crop response to the addition of a P fertilizer. However, it was not intended to be used as an environmental indicator of the potential risk of P pollution. According to Sharpley et al. (1999), additional calibration (beyond that conducted for fertility recommendations) are required in order to also validate soil-test P values as an environmental indicator. Calibration studies have resulted in development of environmentally-based threshold STP concentrations and have been established for several regions. Sharpley and Tunney (2000) have presented an array of examples of the use of such tests as guides for P management recommendations to protect water quality. Heckrath et al. (1995) found that P leaching losses were low when P extractable in 0.5 M sodium bicarbonate (Olsen-P) was less than 60 mg kg -1 and then increased rapidly when Olsen-P exceeded 60 mg kg -1 They referred to this value as a change-point, above which the potential for P loss through leaching increased considerably. However, the change-point value was found to be affected by experimental conditions, such as the range of STP values tested and the

PAGE 34

20 soil to solution ratio for the P desorption parameter (Koopmans et al., 2002). Brling (2003) presented evidence that even if a change-point can be identified, it would be site-specific. Hesketh and Brookes (2000) determined change-points in seven soils and found variation from 10 to 119 mg kg -1 Olsen-P. Water soluble P is another environmental soil testing parameter for assessing potential P losses. Phosphorus extracted from soils using distilled deionized water (WSP) is expected to simulate the rapid release of P to water better than a stronger chemical extractant (such as agronomic soil P tests), since rainfall is very similar to distilled water (Moore et al., 1998). This extraction should also maintain the soil pH within one unit of its original value, which is a desirable attribute since P solubility is highly dependent upon soil pH (Golterman, 1998; Sharpley, 1993). Pote et al., (1996) demonstrated, in a field study with tall fescue the existence of a strong correlation between water soluble P with runoff P. The concept of Degree of Phosphorus Saturation (DPS), was developed in the Netherlands (van der Zee et al., 1987; Breeuwsma and Silva, 1992) and later applied in other parts of the world has been used to evaluate the risk of P leaching from soils as follows: 100xcapacitysorptionPPSorbedDPS The DPS estimates how close the soil is to saturation and facilitates comparison among soils with varying P sorption capacities. Different methods of estimating sorbed P and P sorption capacity can be used to determine DPS. In the Netherlands and several other countries, P extracted with ammonium oxalate has been used to estimate sorbed P, and soil concentrations of both Fe and Al in order to estimate P sorption capacity (Brling, 2003) as follows:

PAGE 35

21 010x)Al(FePDPSoxoxox where is an empirically derived sorption coefficient which has a value between 0 and 1. This coefficient is derived from the following relationship )Al(Fecapacitysorption Poxox A value for of 0.5 was recommended by Beauchemin and Simard (1999); Breeuwsma and Silva, 1992; Koopmans et al., 2003; Schoumans and Groenendijk, 2000 and Sims et al., 2002 while Nair and Graetz (2002) reported a value of 0.55 for Spodosols in Florida. The DPS normalizes P (oxalate, Mehlich-1 or -3) to extractable Fe and Al, and has been related regionally to soil solution P concentration (Nair and Harris, 2004). Hooda et al. (2000) suggested that the DPS, which relates ammonium oxalate-extractable P to the sum of oxalate extractable Fe and Al (DPS Ox ), is a good indicator of the soils potential to release P. Other researchers have suggested that DPS Ox can be a suitable tool for predicting subsurface P lost since it has been shown to be closely correlated to P concentrations in leachate waters (Leinweber et al., 1999 and Maguire and Sims, 2002). A soil is P-saturated when 25% or more of PSC (Phosphorus Sorption Capacity) has been used. This level of P sorption saturation will likely result in ground water P concentrations above the 0.1 mg P L -1 standard used in the Netherlands (Breeuwsma et al., 1995). Other studies conducted by Maguire et al. (1998) and McDowell and Sharpley (2001) have shown a low potential for P losses via leaching when DPS is <20 to 25% and sharp increase in P leaching above this DPS value. More recently, Nair and Graetz (2002) showed that DPS M1 can be used as indicator of soluble P for both surface A horizons and subsurface Bh (spodic horizons) of Spodosols in the Lake Okeechobee basin. Nair et al. (2004) taking into consideration the change points, confidence intervals,

PAGE 36

22 agronomic soil test values, and DPS values from other studies, recommended a threshold DPS M1 of 30% for Florida sands. Values for DPS M1 of 31 to 60% warrant caution with regard to further addition of P to a land-use system, while DPS M1 values of > 60% imply that the can potentially impair water quality. The Phosphorus Saturation Ratio (PSR) is similar to DPS, but does not use a corrective constant () associated with the original definition of DPS (Breeuwma and Silva, 1992). When used as environmental indicators of off-farm P loss, both STP and DPS failed to provide a clear indication of the capacity of a soil to retain added P. Therefore, soils with very low STP values may nevertheless have insufficient capacity to retain added P, and soils with higher STP values may have capacity to retain additional P. Nair and Harris (2004) recommended the use of a DPS-based calculation of the remaining soil P storage capacity (SPSC) that consider risks arising from previous loading as well as inherently low P sorption capacity. The SPSC of a soil is based on a threshold PSR of 0.15: SPSC = (0.15 PSR ) x (Mehlich-1 Al + Mehlich-1 Fe) [Nair and Harris, 2004] where (Mehlich-1 Al + Mehlich-1 Fe) is proportional to the P sorption capacity of a given soil (van der Zee et al., 1987). According to Nair et al. (2004), the value of 0.15 used in the SPSC calculation is the best approximation available to date of the PSR value corresponding to the critical P solution concentration of 0.10 mg litre -1 P as proposed by Breeuwsma and Silva (1992). The SPSC can be calculated by horizon (expressed as kg P m -3 ) using the bulk density and the thickness of that horizon or by depth (expressed as mg P kg -1 or kg P ha -1 ) [Nair and Harris, 2004]. The SPSC can be used for assessing the soil status in regards to future environmental risk-free P additions. In other words, the SPSC

PAGE 37

23 allows prediction of the amount of additional P that can be retained by a given soil horizon or depth before the added P becomes an environmental concern. Silvopasture and Environmental Quality Nutrient movement studies in temperate agroforestry systems are rare and whatever little that has been done has focused on riparian buffers. Nonetheless research undertaken on silvopasture in Florida has shown great promise. Preliminary research conducted on a silvopastoral site at a research farm at Ona, Florida, on flatwoods soils (Spodosols) suggest that silvopastoral sites are less likely to retain nutrients within the soil profile compared to an adjacent fertilized pasture with cattle grazing (Nair and Graetz, 2004) (Table2-2). Table 2-2. Water soluble phosphorus (WSP) concentrations as affected by soil depth for a native flatwoods site, a silvopastoral site, and an adjacent treeless pasture at Ona, Florida, USA. Q Depth Native flatwoods Silvopasture Pasture cm WSP, mg kg -1 0-5 1.59 (1.14) Q 4.23 (1.34) 9.10 (3.26) 5-15 1.28 (1.17) 2.09 (1.36) 2.79 (2.20) 15-30 0.12 (0.02) 0.79 (0.39) 1.52 (0.94) 30-50 0.00 (0.00) 0.17 (0.14) 1.91 (1.25) 50-75 0.00 (0.00) 0.02 (0.02) 0.42 (0.40) 75-100 0.00 (0.00) 0.04 (0.03) 0.66 (0.52) Q Source: Nair and Graetz, 2004 Q Numbers in parentheses are standard deviation values In a study conducted at a University of Florida research station, Ona, Florida, for which long term information on the management history of the sites is available, Nair and Kalmbacher (2005), reported P concentrations in soil profiles obtained from each of the pasture sites to be in the following order: pasture>silvopasture>native pasture. In the same study the NH 4 + and NO 3 -N concentrations were found to be higher in the surface horizon of the treeless pasture.

PAGE 38

24 A substantial body of literature is available on the effectiveness of riparian buffers in reducing nutrient losses from farms to water bodies, and on the potential offered by these systems to reduce nutrient loss from agricultural lands. Hefting et al. (2005) reported that plant production, nitrogen uptake, and N retention were significantly higher in a forested buffer site compared to a herbaceous buffer site in riparian buffer zones in study conducted in six European countries. Results of a field monitoring study done in Tokachikawa watershed in Hokkaido, Japan, Cisadane, Cianten and Citamyang sub-watersheds in Indonesia and Cauvery watershed in southern India, all demonstrated the positive impact of forest buffer zones in reducing the influence of agricultural nutrients and chemicals on surface stream waters (Anbumozhi et al., 2005). In a study conducted i n southern Illinois, USA, Blattel et al. (2005) found that ground water soluble P concentrations were significantly reduced by 14 percent in the first 1.5 m of both mixed hardwood forest and giant cane ( Arundinaria gigantea) riparian buffers, and there was an overall 28 percent reduction in water soluble P concentration by 12 m from the field edge. Schoonover et al. (2005) showed that forest buffers significantly reduced incoming dissolved NO-N, dissolved NH-N, total ammonium-N, and total orthophosphate concentrations in surface runoff within the 10.0 m riparian buffer by 97, 74, 68, and 78, respectively. Furthermore, overall nutrient reductions for giant cane buffer strips were 100 percent for all three nutrients due to relatively high infiltration rates. Finally, significant reductions of total NHN and total orthophosphate were also detected at a distance of 3.3 m from the edge of the cane buffer compared to 6.6 m in the forest buffer. In a study conducted in North Queen Island, Australia, McKergow et al. (2004) reported a reduction in total N, total P and suspended sediment loads between 25 and 65% by the 3 4 + 4 +

PAGE 39

25 planar slope grass buffer and within the first 15 m of the moderately convergent grass buffer. Lee et al. (2003) showed, in a study in Bear Creek, Iowa, USA, that switchgrass buffer (Panicum virgatum cv. Cave-n-Rock) removed 95% of the sediment, 80% of the total-N, 62% of the nitrate-nitrate-N, 78% of the total-phosphorus, and 58% of the phosphorus, whereas a switchgrass/woody buffer removed 97% of the sediment, 94% of the total-N, 85% of the nitrate-N, 91% of the total-P, and 80% of the P in the runoff. These findings provide clear evidence that agroforestry systems are effective in reducing the soil P and nitrate-N store and would therefore reduce the amount of nutrients likely to be lost from farmland. This phenomenon is of great importance because it presents a mechanism to sustain current agricultural land use while achieving both nutrient conservation and water quality protection.

PAGE 40

CHAPTER 3 SOIL CHARACTERISTICS OF THE SILVOPSTURE AND TREELESS PASTURE STUDY SITES Introduction A thorough understanding of the distribution of nutrients in different soil layers is essential for developing effective nutrient management strategies that are economically viable and environmentally sound. From that point of view, this study was undertaken to investigate differences in soil characteristics between silvopasture and adjacent treeless pasture systems. Materials and Methods Seven of the twelve soil orders are represented in Florida (Collins, 2003): Alfisols, Entisols, Histosols, Inceptisols, Mollisols, Spodosols, and Ultisols. The sites selected for this study represented two major soil orders of the state: Ultisols and Spodosols. Spodosols are characterized by having undergone soil processes that translocate organic matter and aluminum, or organic matter, aluminum, and iron, as amorphous materials (Collins, 2003). Spodosols occur mostly on coarse-textured, acid parent materials subject to ready leaching (Brady and Weil, 2002). The A and E horizons of Spodosols are sandy and have poor sorption capacities while the Bh (spodic) and the Bw horizons have much greater affinities for P retention (Mansell et al., 1991). According to Soil Survey Staff (1996), Spodosols are characterized by a highly fluctuating water table situated between the Bh and the A horizons during the summer rainy season, while it may go down to 125 cm during drier months. Due to low hydraulic conductivity of the lower soil layers, 26

PAGE 41

27 rainfall that penetrates the soil during high water table conditions tends to move laterally, transporting P to surface drainage ditches (Burgoa et al., 1991; Mansell et al., 1991). Approximately 3.4 million hectares in Florida have been mapped as Spodosols (Collins, 2003). Ultisols are characterized by the presence of an argillic horizon, enough moisture for crops in most years, and a base saturation [proportion of chemical exchange sites on soil particles that are occupied by basic cations (Ca 2+ Mg 2+ K + Na + )] less than 35 percent. They developed in relatively warm and moist climates, such as northwest Florida, and tend to be highly productive if managed properly. There are approximately 2.8 million known hectares of Ultisols (Collins, 2003) in Florida. The distribution of Ultisols and Spodosols in Florida is shown in Figure 3-1. Figure 3-1. Soil orders of Florida (Source: Collins (2001)

PAGE 42

28 Location of the Study Sites The study was conducted on four farms that practice silvopasture in Florida: Mr. Harris Hills farm in Osceola County, Dr. Rudy Garbers farm in Manatee County, the Florida Sheriffs Youth Ranch in Suwannee County, and Mr. Fred Clarks farm in Alachua County (Table 3-1 and Figure3-1). The silvopastures consisted of a combination of slash pine (Pinus elliottii) and bahiagrass (Paspalum notatum ), and the treeless pastures consist of bahiagrass (Figures 3-3 and 3-4). The tree planting configuration of the silvopastures in Osceola and Manatee locations was a double row planting (3.1 m between rows and 1.2 m between trees and 12.2 m wide alleys), whereas in Suwannee and Alachua the configuration was a single row planting. The treeless pastures were generally older than the silvopastures with the exception of the Suwannee location in which both pasture systems were of the same age (Table 3-2). Generally no fertilizers were applied to the pasture systems under study except for the treeless pasture in Suwannee, which received a 19-5-19 formulation respectively at a rate of 336 kg ha-1 and at 224 kg ha-1 during 2003 and 2004 respectively. The soils at both Alachua and Suwannee are Ultisols (Figure 3-5), whereas soils at Manatee and Osceola are Spodosols (Figure 3-6). The Spodosols are poorly drained soils as compared to Ultisols that are well drained. On average, the depth to the spodic horizon was 55 cm in the Spodosols (Table 3-1). Average annual precipitations range from 1233 to 1381 mm and the minimum water table depth at the Osceola and Manatee field sites were 25 and 15 cm, respectively. Slope percent vary from 0 to 8 percent at the study locations (Table 3-1).

PAGE 43

29 Figure 3-2. Location of study sites in Florida. The Osceola and Manatee sites are on Spodosols and the Alachua and Suwannee sites are on Ultisols.

PAGE 44

Table 3-1. Soil and climatic characteristics of the four farms selected for the study in four counties of Florida Osceola County Manatee County Suwannee County Alachua County Location 28 0 9 N, 81 0 10 W Central Florida 27 0 13 N, 82 0 8 W Southwestern Florida 30 0 24 N, 83 0 0 W Florida Panhandle Region 29 0 45 N, 82 0 33 W North-central Florida Soil Soil Series Immokalee: Sandy, siliceous, hyperthermic Arenic Alaquods EauGallie: Sandy, siliceous, hyperthermic Alfic Alaquods Blanton: Loamy, siliceous, semiactive, thermic Grossarenic Paleudults Kendrick: Loamy, siliceous, semiactive, hyperthermic Arenic Paleudults Parent material Thick marine deposits formed in thick beds of marine sands Sandy and loamy marine sediments in Peninsula Florida Sandy and loamy marine or eolian deposits Thick beds of loamy marine sediments Slope (%) 0 2 0 2 0 8 0 8 Annual fluctuation of the water table depth (cm) 25 102 15 45 Drainage class Poorly drained Poorly or very poorly drained Somewhat excessively to moderately well drained Well drained; slow to moderately slow permeability Depth to spodic horizon (cm) (spodosols) 53 58 Climate Avg. annualprecipitation (mm) 1233 1381 1366 1332 Avg. annual air temperature ( O C) 22 22 20 21 30 Source: Soil survey of Osceola (Readle, 1979), Sarasota (Hyde et al., 1991); Alachua (Thomas et al., 1982); USDA-NRCS (not dated); World climate (2005); Florida Sheriffs Youth Ranch, 2003, unpublished; personal communication; and results of the study.

PAGE 45

31 Table 3-2. Description of pasture systems at the four study locations in Florida Pastoral Systems Osceola Manatee Suwannee Alachua Silvopasture Tree species Slash pine Pinus elliotii Slash pine Pinus elliotii Slash pine Pinus elliotii Slash pine Pinus elliotii Pasture species Bahiagrass Paspalum notatum Bahiagrass Paspalum notatum Bahiagrass Paspalum notatum Bahiagrass Paspalum notatum Tree planting configuration (m) Double rows 3.1 x 1.2 x 12.2 Double rows 3.1 x 1.2 x 12.2 Single rows 1.5 x 7.2 Single rows 1.5 x 3.0 Tree age (year) 12 12 40 8 Land area (ha) 8.5 20.2 16.2 28.3 Fertilization history No fertilization No fertilization No fertilization No fertilization Prior land use history Florida Flatwoods Treeless pasture for 15 years (Site continuously fertilized and limed when it was under treeless pasture system) Florida Flatwoods Agriculture Grazed naturally occurring loblolly pine forest with grass Treeless pasture Pasture species Bahiagrass Bahiagrass Bahiagrass Bahiagrass Age (year) 45-50 30 40 55 Land area (ha) 6.1 3.0 16.2 7.3 Fertilization history No fertilization No fertilization 19-5-19 application at 336 kg ha -1 in 2003 19-5-19 application at 224 kg ha -1 in 2004 4.5 Mg ha -1 of dolomite applied every 4 years since 1978 No fertilization Prior land use history Florida Flatwoods Florida Flatwoods Agriculture Agriculture (Cornfield)

PAGE 46

32 Figure 3-3. Silvopastoral system at Osceola County, Florida Figure 3-4. Treeless pasture at Manatee County, Florida

PAGE 47

33 Figure 3-5. Profile of an Ultisol. (Source: Collins, 2001) A E Bh Figure 3-6. Profile of a Spodos ol. (S ource: Collins, 2001)

PAGE 48

34 Estimation of Nitrogen and Phosphorus Loads in the Pasture Soils An estimation of nitrogen and phosphorus loads in the pasture sites was made based on several assumptions (Table 3-3). According to Vicente-Chandler (1966), a mature cow produces 9.1 tons (1 Mg = 1 ton = 1000 kg) of excreta yearly containing a total of 71 kg N, 40 kg P 2 O 5 and 59 kg K 2 O. The total land grazed at the Manatee, Suwannee, Osceola and Alachua farms were 259, 567, 445, and 259 hectares respectively. Although grazing animals do not deposit urine and manure evenly across the paddocks where they grazed, an even distribution of cattle manure was assumed in this study (in the absence of a better way of presentation). Dalrymple (1994) reported from a rotational grazing study that urine spots occupied 16.7 percent of the pasture while manure spots occupied 18.8 percent, following 504 grazing days. Table 3-3. Estimates of nutrient loads from beef cattle excreta at the four study sites Load estimates (kg ha -1 yr -1 ) Location Farm size (ha) Head of Cattle N P 2 O 5 Alachua 259 150 41 23 Suwannee 567 200 25 14 Manatee 259 150 41 23 Osceola 445 275 44 25 Phosphorus and nitrate loads from beef cattle were similar in both Alachua and the Manatee sites (Table 3-3).The Suwannee sites had the lowest nutrient loads per unit area than all four sites which is the result of a lower cattle density. The highest nutrient loads per unit area were observed in the Osceola sites. Major Tree and Grass Species at the Study Sites Pinus elliottii (Slash pine). Slash pine is an important timber species in the southeastern United States. Its strong, heavy wood is excellent for construction purposes.

PAGE 49

35 Because of its high resin content, the wood is also used for railroad ties, poles, and piling (Duncan and Duncan, 1988; McCune 1988; Lohrey and Kossuth, 1990) Slash pine is a large, stately, heavily-branched, long-needled conifer whose fast growth rate makes it capable of reaching 30.5 m in height and 0.9 to 1.2 m in diameter (Gilman and Watson, 1994). Slash pine is self-pruning of its lower branches, it forms a somewhat pyramidal canopy when young which evolves to an open, rounded canopy over time. Mature trees generate a light, draped shade beneath. Aggressive root competition takes place beneath pines, so the shrubs and lawn beneath and around the canopy often require more frequent irrigation, particularly during the dry season. Pines have some deep roots, except in poorly-drained soil where most roots proliferate near the soil surface. The tap root is prominent in well-drained soil and can make pines difficult to transplant in the wild. Slash pines grow well on a variety of acidic soils in either full sun or partial shade. It does poorly in high pH soil and so is not recommended for calcareous soils and where irrigation water has a high pH. Once established, it is more tolerant of wet sites than most other pine species, and is highly drought-tolerant (Lohrey and Kossuth, 1990) Paspalum notatum (Bahiagrass). Bahiagrass in an important forage species in Florida where it occupies more land area than any other single pasture species, covering an estimated 1 million hectare (Chambliss and Adjei, 2006). Most of this land area is used not only for grazing but also for the production of hay, sod, and seed harvested from pastures. Bahiagrass is a warm-season perennial grown throughout Florida and in the Coastal Plain and Gulf Coast regions of the southern United States. It is a native to South

PAGE 50

36 America and is widely distributed in Argentina, Uruguay, Paraguay and Brazil. Bahiagrass is adapted to climatic conditions throughout Florida and can be grown on upland well-drained sands as well as the moist, poorly-drained flatwoods soil of peninsular Florida (Chambliss and Adjei, 2006); LSU, 1998). Bahiagrass is popular with Florida ranchers because it tolerates a wider range of soil conditions than other improved grasses, has the ability to produce moderate yields in soils of very low fertility, is easily established from seed, withstands close grazing, and is relatively free from damaging insects (except for mole crickets) and diseases. Soils of the Study Sites Soil sampling Soil samples were collected during summer and fall 2004 at the four locations (Figures 3-7 to 3-10) at the following sampling depths: 0 5, 5 15, 15 30, 30 50, and 50 75 and 75 100 cm using a bucket soil auger (21 cm x 7 cm). A grid sampling procedure was used (Figure 3-7). With eight plots, each with 24 sampling points and six depths, there were a total of 1152 soil samples. Soil samples were taken in the alleys in the silvopastoral systems. The spacing between the sampling points in both the regular pasture and the silvopasture plots was based on the size of the plots. Soil Survey (hard copies) and Soil Survey digital maps [Natural Resources Conservation Service (NRCS), Soil Service Geographic (SSURGO)] of the area to be sampled were used to obtain information on the soil series. A hand-held GPS unit (eTrex TM GARMIN personal navigator) was used to obtain the geographic coordinates of each the sampling point.

PAGE 51

37 Tree-less pasture Silvopasture Suwannee Tree-less pasture Silvopasture Tree-less pasture Silvopasture Tree-less pasture Silvopasture Suwannee Figure 3-7. Grid sampling at the Suwannee location. Twenty four soil profiles were collected at each field (silvopasture and treeless pasture) at each location, and with 6 depths (0 5, 5 15, 15 30, 30 50, and 50 75 and 75 cm) per profile. The small circles, representing sampling points, illustrate the layout of the grid sampling method used in each sites at each location. Soil-nutrient characterization Mehlich-1 (0.0125M H 2 SO 4 and 0.05M HCl)-extractable Fe, Al, Ca, and Mg were obtained using a 1:4 soil to double acid ratio (Mehlich, 1953). The metals in the Mehlich-1 solutions were determined by atomic absorption spectrometry.

PAGE 52

38 Soil texture and pH determination Soil texture was determined by the hydrometer method following the analytic procedure describe by Arshad et al. (1996). Soil pH (2:1) was assessed using the analytical procedure described by Smith and Doran (1996). Statistical Procedures The statistical analyses were performed using SAS (PROC ANOVA) 8.2 for Windows. The experimental design used was a 4-fold nested model (Table 3-4). The statistical model used was the following: ijklijkijiijk Y where = grand mean i = effect of level i of location (Alachua, Suwannee, Osceola and Manatee) ij = effect of level j of treatment (Silvopasture and treeless pasture), nested within level i of location ijk = effect of level k of plot (24 sampling points), nested within level j of treatment which is nested within level i of location ijkl = effect of level k of depth (6 depths), nested within level k of plot which is nested in level j of treatment which is nested in level i of location

PAGE 53

39 Table 3-4. Statistical model used in the study Q Source Degree of Freedom Location 3 Treatment (location) 4 Plot (Treatment x location) 184 Depth (Treatment x location x plot) 960 Total 1151 Q The hypotheses were tested using the ANOVA mean square for treatment (location) as the error term for location, the ANOVA mean square for plot (treatment x location) as the error term for treatment (location), and the ANOVA mean square for depth (treatment x location x plot) for plot (treatment x location). Results and Discussion Soil Reaction (pH) Soil pH for the 0 5 cm depth for the silvopasture and treeless pasture in Alachua were 5.4 and 5.2 respectively. The corresponding values for the Suwannee, Manatee, and Osceola field sites were 4.7 and 6.6; 4.8 and 4.4; 4.5 and 4.4 (Tables 3-5 to 3-8). The soil pH was different between the two pasture systems for all soil layers at the Suwannee location with the treeless pasture having the highest pH. Results from the Suwannee location can be explained by the fact that the treeless pasture site has been continuously limed at the rate of 4.5 Mg ha -1 every 4 years since 1978. In Alachua, soil pH was only found different at surface horizon with the treeless pasture site having lower pH values. This is not consistent with the general perception that conifer (e.g., pines and firs) stand tend to acidify the soil. Soil acidification is enhanced by most coniferous vegetation because of the acidic nature of their resinous needles which tend to accumulate in a thick O horizon (Brady and Weil, 2002). The difference in soil pH observed in Alachua maybe attributed to some variation in the parent material across the farm (e.g. limestone deposits). At the Manatee site soil pH was similar for both systems across all soil depths. In Osceola differences in soil pH were only observed below 30 cm of depth with the treeless

PAGE 54

40 pasture having the higher pH values. It is important to remember that soil samples were only taken in the alleys in the silvopastures; therefore there could be differences in pH near the pine trees. In general the soil pH in the two pasture systems at in all four locations were in the desirable ranges for bahiagrass (<5.5 6.4) (Sartain, 2001) throughout the soil profile of the sites. Soil pH of in the similar range as been reported by Nair and Graetz (2002) for the A, E, Bh, and Bw soil horizons of a beef pasture site (Spodosols) situated in the Okeechobee County of Florida. Soil Particle Size Analysis (Soil Texture) The soils at Alachua, Suwannee, Osceola ,and Manatee were designated as Kendrick (Loamy, siliceous, semiactive, hyperthermic Arenic Paleudults), Blanton (Loamy, siliceous, semiactive, thermic Grossarenic Paleudults), Immokalee (Sandy, siliceous, hyperthermic Arenic Alaquods), and EauGallie (Sandy, siliceous, hyperthermic Alfic Alaquods) soil series, respectively (Table 3.1). For the Alachua field site, the percent sand, silt, and clay for the 0 5 cm depth were 87, 4, and 9 and 92, 3, and 5% for the treeless pasture and silvopasture, respectively. The corresponding values for Suwannee were 94, 2, and 4% vs. 95, 3, and 3% compared to 94, 3, and 3% vs 95, 3, and 2% (Manatee), and 92, 4, and 4% vs. 96, 2, and 2% in Osceola (Table 3-5 to 3-8). In general, sand, silt and clay content was similar for both production systems across all soil depths. In Suwannee the treeless pasture had a higher clay content in the upper soil layer. The lack of differences in the soil texture between the two pastures at the four locations sites is desirable since it facilitates the comparison of the two land use systems in terms of there nutrient status. The clay content of the Ultisols at 75 100 cm depths was higher

PAGE 55

41 compared to the Spodosols, which is consistent with the known characteristics of these two soil orders (Collins, 2003). Mehlich-1 Al and Fe Concentrations Mehlich-1 Al concentrations at 0 cm depth for the silvopasture and treeless pasture in Alachua were 436 and 479 mg kg -1 respectively. The corresponding values were 160 and 183 mg kg -1 (Suwannee), 48 and 36 mg kg -1 (Manatee) compared to 225 and 132 mg kg -1 in Osceola (Tables 3-5 to 3-8). Concentrations of Mehlich-1 extractable Fe at the top 0 5 cm depth in the silvopasture and treeless pasture in Alachua were 28 and 43 mg kg -1 respectively. The corresponding values were 25 and 10 mg kg -1 (Suwannee), 18 and 18 mg kg -1 (Manatee), while values at Osceola were 13 to 27 mg kg -1 (Tables 3-5 to 3-8). In Alachua, Mehlich-1 Al concentrations among the production systems were different below 15 cm of depth. In contrast, Mehlich1-Fe concentrations were different between the two pasture systems (Alachua) throughout the soil profile from the top 0 5 cm to 75 100 cm depth. Differences in Al concentrations between the two pasture sites within a location may be attributed to variation in clay content. The higher the clay content of a soil the higher its Al concentration will be. In Suwannee, there were generally no differences in Mehlich-1 Al and Fe concentration among land use systems through out the entire soil profile. With the exception of 0 5 and 75 100 depths, both Mehlich-1 Al and Fe concentrations were not different between the pasture systems in Manatee. The level of these elements in the soil was very low at the surface horizons in Manatee, which is a characteristic of Spodosols. The Mehlich-1 Al content of the two pasture sites in Osceola showed no difference throughout the soil profile. Iron concentration in the soil of the two pasture systems was found to be different only at 5 15, 50 75, and 75 100 cm depths.

PAGE 56

42 A strong relationship between Mehlich-1 Fe, and diagnostic tissue levels of Fe, concentrations and a predictable responses to applied Fe has not being found, therefore the University of Florida Extension Soil does not make fertilization requirement based on extractable Fe (Sartain, 2001). Average values for Mehlich-1 Al and Fe concentrations for the Ultisols and the Spodosols were 315 and 110 mg kg -1 and 27 and 19 mg kg -1 respectively, at depth 0 5 cm respectively. Results of the study showed that Ultisols generally had greater Mehlich1 Al concentration then Spodosols in the surface soil horizons (from 0 5 cm to 15 30 cm depth) [Table 3-9]. Below the 30 cm soil depth no significant difference in Al was observed between the soil types. Ultisols had the highest Fe concentration at 15 30 and 75 100 cm. The higher Al concentrations observed in the Ultisols at the surface horizons is an indication of the greater P retention capacity of this soils in the surface horizons as compared to Spodosols since the amount of aluminum and iron oxides and hydroxides in acid soils is a determinant factor in the amount of P that can be held by a soil. At the deeper soil horizons no differences were observed between the two soil types because the spodic horizon may have comparable levels of Al and Fe. These results are consistent with Mansell et al. (1991) observations that the A and E horizons of Spodosols are sandy and have poor sorption capacities while the spodic horizon have much greater affinities for P retention. Nair and Graetz (2002) found Mehlich-1 Al and Fe concentrations of 38.1, 9.0, 385 and 635 and 13.2, 7.52, 46.1, and 32.8, respectively, for the A, E, Bh, and Bw soil horizons of a beef pasture site (Spodosols) of the Okeechobee County of Florida.

PAGE 57

43 Mehlich-1 Mg and Ca Concentrations Mehlich-1 Mg concentrations for the silvopasture and treeless pasture in Alachua for the 0 5 cm depth were 132 and 223 mg kg -1 respectively. The corresponding values were 72 and 154 mg kg -1 (Suwannee), 132 and 154 mg kg -1 (Manatee), compared to 100 and 70 mg kg -1 in Osceola (Tables 3-5 to 3-8). In Alachua concentrations of Mehlich-1 extractable Ca in the top 0 5 cm depth for the silvopasture and the treeless pasture were 727 and 1006 mg kg -1 respectively. Values in Suwannee were 531 and 921 mg kg -1 compared to 1029 and 911 mg kg -1 in Manatee, while values in Osceola were 663 to 596 mg kg -1 (Tables 3-5 to 3-8). Similar Mehlich-1 Ca and Mg concentrations for the A, E, Bh, and Bw soil horizons of a beef pasture site (Spodosols) in Okeechobee County (S-Florida) of Florida were reported by Nair and Graetz (2002). Mehlich-1 Mg and Ca concentrations were found to be different among the pasture systems in both Alachua and Suwannee with the highest concentration found in the treeless pasture. Based on UF/IFAS diagnostic soil test interpretations (Kidder et al., 2002) Mg concentration were generally in the high range (>30 mg kg -1 ) in both the treeless pasture and the silvopasture sites throughout the soil profile in Alachua. Soil Mg concentration was in the high range in the treeless pasture throughout the soil profile and in the low range (<15 mg kg -1 ) in the silvopasture below 5 cm of depth in Suwannee base on UF/IFAS interpretation system. In general, Mehlich-1 Mg and Ca concentrations were similar for both production systems in Manatee. The Mg concentration in two pasture sites was found to be in the low range according to UF/IFAS interpretations. In Osceola Mehlich-1 Ca concentration were also similar throughout sampled soil profile (up to 100 cm, which was the deepest position studied), whereas Mehlich-1 Mg concentration was generally different. Soil Mg

PAGE 58

44 concentration in the two pasture systems in Osceola was in the low range below 30 cm of depth. In Manatee, the silvopasture and the treeless pasture sites had Mg concentration in the high range at the surface 0 5 cm depth, whereas Mg concentration below 30 cm of depth where in the low range. No interpretation is made for Mehlich-1 extractable Ca levels because the extraction method dissolves soil calcium compounds which are not readily available for the plants. Generally, Ca levels tend to be adequate for turfgrass growth in Florida soils due the fact that they are inherently high in Ca, have a history of Ca fertilization, or receive Ca regularly through irrigation with high Ca water (Kidder et al., 2002). Conclusions The soils at the study sites typically had a rather coarse (sandy) texture, especially in the upper layers, and in the absence of liming may be prone to acidification and are typical of the southeastern USA. Soil pH at the study sites were in the recommended range for bahiagrass. In general the treeless pastures had higher pH values compared to the silvopastures. Magnesium concentration levels in the treeless pasture sites were generally higher than the silvopasture sites. Soil test values for agronomic crops were high in Mg in the upper 0 5 cm at all four locations. Calcium concentration levels were generally not different in the two pasture systems in Osceola and Manatee. Except in Alachua where the treeless pasture had the highest Al levels of the two pasture systems, no differences in Al concentrations were generally observed between the two pastures. Iron concentrations were in general higher in the silvopasture as compared to the treeless pasture in Alachua and Suwannee but this trend was not significant in Osceola and Manatee. Soil Fe concentration level was found to be of lower magnitude at all the four locations. The Ultisols had higher soil Al and Fe concentrations than the

PAGE 59

45 Spodosols in the upper 30 cm of the soil profile. Higher clay content was also observed in the deeper 75 100 cm depth in the Ultisols. The information presented here is critical to the explanation of the results pres ented in the subsequent chapters.

PAGE 60

46 Table 3-5. Mean values of Mehlich-1 Ca, Al, Mg, and Fe concentrations, pH, and percent sand, silt and clay for soil profiles at the center of the alley of the silvopasture and a treeless pasture sites in Alachua Florida, USA. Silvopasture Treeless pasture Silvopasture Treeless pasture Silvopasture Treeless pasture P Depth (cm) Ca (mg kg -1 ) P Al (mg kg -1 ) P Mg (mg kg -1 ) 0-5 727 1006* 0.0067 436 479 0.3970 132 223* 0.0001 5-15 359 920* 0.0001 499 563 0.1338 48 126* 0.0001 15-30 215 946* 0.0001 429 531* 0.0204 32 151* 0.0001 30-50 169 969* 0.0001 359 608* 0.0004 32 216* 0.0001 50-75 132 999* 0.0001 285 660* 0.0001 39 252* 0.0001 75-100 182 1147* 0.0001 272 744* 0.0001 52 239* 0.0001 Silvopasture Treeless pasture P Silvopasture Treeless pasture P Silvopasture Treeless pasture P Depth (cm) Fe (mg kg -1 ) pH (1:2) Sand (%) 0-5 28 43* 0.0005 5.4 5.2* 0.0184 92 87 0.4952 5-15 27 43* 0.0001 5.6 5.4* 0.0061 91 86 0.4225 15-30 25 44* 0.0001 5.8 5.8 0.7324 92 80 0.3792 30-50 25 45* 0.0001 5.8 5.8 0.3964 91 76 0.4173 50-75 22 38* 0.0001 5.6 5.7 0.0985 92 78 0.3891 75-100 18 37* 0.0001 5.4 5.6 0.0632 89 79 0.5366 Silvopasture Treeless pasture P Silvopasture Treeless pasture Depth (cm) Silt (%) Clay (%) 0-5 3 4 0.6851 5 9 0.4415 5-15 4 5 0.4540 5 8 0.4355 15-30 7 7 0.9931 1 13 0.2511 30-50 4 7 0.4337 4 16 0.3948 50-75 4 7 0.3258 5 15 0.4091 75-100 4 7 0.4047 8 15 0.5800 [(*Indicates significant P values within a given depth (p<0.05)]

PAGE 61

Table 3-6. Mean values of Mehlich-1 Ca, Al, Mg, and Fe concentrations, pH, and percent sand, silt and clay for soil profiles at the center of the alley of the silvopasture and a treeless pasture sites in Osceola Florida, USA. 47 Silvopasture Treeless pasture Silvopasture Treeless pasture Silvopasture Treeless pasture Depth (cm) Ca (mg kg -1 ) P Al (mg kg -1 ) P Mg (mg kg -1 ) P 0-5 663* 596 0.0425 225 132 0.2300 100* 70 0.0425 5-15 470 529 0.6351 181 156 0.7845 24* 14 0.0346 15-30 454 336 0.0487 181 235 0.6060 21* 5 0.0021 30-50 257 246 0.8618 298 786 0.0810 9* 4 0.0003 50-75 166 204 0.4078 395 428 0.7419 9* 4 0.0001 75-100 110 169 0.1296 428 364 0.4600 8* 4 0.0004 Silvopasture Treeless pasture Silvopasture Treeless pasture Silvopasture Treeless pasture Depth (cm) Fe (mg kg -1 ) P pH (1:2) P Sand (%) P 0-5 13 27 0.0903 4.5a 4.4a 0.1257 97 92 0.0542 5-15 9 30* 0.0001 4.6a 4.5a 0.7654 98 92 0.1818 15-30 11 12 0.8071 4.7a 4.9a 0.1221 98 92 0.1383 30-50 5 11 0.0541 4.6a 4.9b 0.0456 95 94 0.5401 50-75 4 8* 0.0320 4.5a 4.8b 0.0175 96 96 0.7798 75-100 3 6* 0.0393 4.4a 4.7b 0.0484 97 95 0.4774 Silvopasture Treeless pasture Silvopasture Treeless pasture Depth (cm) Silt (%) P Clay (%) P 0-5 2 4 0.1277 1 4* 0.0143 5-15 1 6 0.1099 1 2 0.3932 15-30 2 5 0.1807 1 3 0.0761 30-50 3 4 0.6209 1 3 0.5017 50-75 2 2 0.6690 2 2 1.0000 75-100 2 3 0.5562 2 2 0.3746 [(*Indicates significant P values within a given depth (p<0.05)]

PAGE 62

48 Table 3-7. Mean values of Mehlich-1 Ca, Al, Mg, and Fe concentrations, pH, and percent sand, silt and clay for soil profiles at the center of the alley of the silvopasture and a treeless pasture sites in Suwannee Florida, USA. Silvopasture Treeless pasture Silvopasture Treeless pasture Silvopasture Treeless pasture Depth (cm) Ca (mg kg -1 ) P Al (mg kg -1 ) P Mg (mg kg -1 ) P 0-5 531 921* 0.0119 160 183 0.2232 72 154* 0.0001 5-15 78 689* 0.0001 253 266 0.6034 9 55* 0.0001 15-30 40 316* 0.0001 246 259 0.6260 6 28* 0.0001 30-50 36 232* 0.0001 184 230 0.0078 6 22* 0.0001 50-75 30 166* 0.0001 141 190* 0.0423 5 15* 0.0001 75-100 27 132* 0.0001 123 151 0.1581 6 15* 0.0041 Silvopasture Treeless pasture Silvopasture Treeless pasture Silvopasture Treeless pasture Depth (cm) Fe (mg kg -1 ) P pH (1:2) P Sand (%) P 0-5 25* 10 0.0001 4.7 6.6* 0.0001 95 94 0.0700 5-15 28* 16 0.0001 4.9 6.1* 0.0001 95 94 0.4226 15-30 19* 14 0.0052 5.1 6.0* 0.0001 94 94 0.2999 30-50 16* 12 0.0063 5.1 6.0* 0.0001 94 94 0.6020 50-75 14* 10 0.0043 5.1 6.1* 0.0001 94 95 0.7758 75-100 13 10 0.0709 5.0 6.1* 0.0001 84 94 0.4641 Silvopasture Treeless pasture Silvopasture Treeless pasture Depth (cm) Silt (%) P Clay (%) P 0-5 3 2 0.2513 3 4* 0.0133 5-15 3 3 0.6906 2 2 0.1573 15-30 3 4 0.6414 2 3 0.3268 30-50 4* 3 0.3217 2 3 0.1919 50-75 4 3 0.3057 2 3 0.1548 75-100 8 3 0.3804 8 3 0.5324 [(*Indicates significant P values within a given depth (p<0.05)]

PAGE 63

49 Table 3-8. Mean values of Mehlich-1 Ca, Al, Mg, and Fe concentrations, pH, and percent sand, silt and clay for soil profiles at the center of the alley of the silvopasture and a treeless pasture sites in Manatee Florida, USA. Silvopasture Treeless pasture Silvopasture Treeless pasture Silvopasture Treeless pasture Depth (cm) Ca (mg kg -1 ) P Al (mg kg -1 ) P Mg (mg kg -1 ) P 0-5 1029 911 0.2076 48* 36 0.0060 132 154 0.1977 5-15 446 422 0.6161 19 17 0.3668 18 26* 0.0489 15-30 175 132 0.0641 14 8 0.1048 4 5 0.2427 30-50 101 74 0.1158 56 15 0.2783 6 3 0.3196 50-75 181 157 0.2997 144 121 0.4130 5 4 0.4792 75-100 135 200* 0.0029 246* 129 0.0002 8 8 0.5845 Silvopasture Treeless pasture Silvopasture Treeless pasture Silvopasture Treeless pasture P Depth (cm) Fe (mg kg -1 ) P pH (1:2) P Sand (%) 0-5 18 18 0.8605 4.8* 4.4 0.0001 95 94 0.5124 5-15 16 14 0.2445 4.9 4.8 0.1748 98 89 0.3991 15-30 11* 6 0.0001 5.3 5.1 0.1731 98 98 0.2929 30-50 9 10 0.8165 5.4 5.3 0.0952 98 98 1.0000 50-75 62 55 0.7010 5.6 5.7 0.1956 97 97 0.1296 75-100 77 90 0.4873 5.5 5.8* 0.0030 97 97 1.0000 Silvopasture Treeless pasture Silvopasture Treeless pasture Depth (cm) Silt (%) P Clay (%) P 0-5 3 3 0.7758 2 3 0.4073 5-15 2 10 0.3950 1 1 0.4453 15-30 1 1 0.4929 1 1 0.2137 30-50 2 1 0.7758 0 1 0.7202 50-75 2 2 0.5425 0.8 1.2* 0.0299 75-100 2 2 0.0513 1.5 1.2 0.0955 [(*Indicates significant P values within a given depth (p<0.05)]

PAGE 64

50 Table 3-9. Mean values of Mehlich-1 Al and Fe concentrations and percent sand, silt and clay for soil profiles of the Ultisols and the Spodosols. Ultisols Spodosols Ultisols Spodosols Depth (cm) Al --------(mg kg -1 )----P Fe --------(mg kg -1 )-----P 0-5 315* 110 0.0224 27 19 0.1702 5-15 395* 93 0.0045 29 17 0.1368 15-30 366* 110 0.0143 26* 10 0.0416 30-50 345 289 0.7147 24 9 0.0749 50-75 319 272 0.7011 32 21 0.0803 75-100 323 292 0.8370 44* 20 0.0479 Ultisols Spodosols P Ultisols Spodosols P Depth (cm) -------------Sand (%)-----------------------Silt (%)------------0-5 92 95 0.3242 3 3 0.7351 5-15 92 94 0.5508 4 5 0.8862 15-30 90 96 0.1980 5 2 0.0951 30-50 89 96 0.1752 5* 2 0.0428 50-75 90 97 0.1597 4 2 0.0601 75-100 86 96* 0.0018 5* 2 0.0286 Ultisols Spodosols P Depth (cm) ----------Clay (%)----------0-5 5 3 0.2636 5-15 3 1 0.1023 15-30 5 1 0.3856 30-50 6 1 0.2633 50-75 6 1 0.2142 75-100 9* 2 0.0048 [*Indicates significant P values (p<0.05)]

PAGE 65

51 Figure 3-8. Aerial photo of the Alachua location

PAGE 66

52 Figure 3-9. Aerial photo of the Osceola location

PAGE 67

53 Silvopasture Tree-less pasture Figure 3-10. Aerial photo of the Manatee locatio n

PAGE 68

CHAPTER 4 PHOSPHORUS AND NITROGEN CONCENTRATIONS UNDER A SLASH PINE (Pinus elliottii) + BAHIAGRASS (Paspalum notatum) SILVOPASTURE AND A TREELESS PASTURE Introduction Non-point pollution is a major source of phosphorus (P) and Nitrate-N to surface waters of the United States (Carpenter et al., 1998). Phosphorus concentrations in excess of 0.02 mg L -1 in lake water can lead to accelerated eutrophication (Sharpley et al., 1999). Nitrate-N concentration in excess of 10 mg L -1 in drinking water may lead to the blue baby syndrome(USEPA, 2006). Soil receiving large quantities of livestock manure can accumulate significant amounts of P and result in excessive P loss to surface and ground water through surface runoff and subsurface leaching. Studies in tropical agroforestry systems on nutrient-poor soils have shown that deep roots of trees in agroforestry systems can absorb nutrients that have leached below the rooting zone of associated agronomic crops, and subsequent deposition of litter and root materials facilitate enhanced nutrient recycling and thereby improve the nutrient-use efficiency in the system as a whole (Young, 1997). Furthermore, tree roots tend to be long-lived and even in the upper soil layers they enhance nitrate and phosphorus retention before and after the crop is planted and harvested (Jose et al., 2004). In essence, this would enhance the overall resource-use efficiency of the system (van Noordwijk et al., 2004). Although, the vertical distribution of a given species root system in its natural niche may follow a distinct pattern, leading to overlapping root distributions, some species are able to respond to changes in nutrient supplies or restricting soil layers) (Jose 54

PAGE 69

55 et al., 2004). This has been demonstrated by Wanvestraut et al. (2004) in a study conducted in pecan (Carya illinoensis) + cotton (Gossypium hirsutum) alley cropping system in northwestern Florida. They observed that cotton roots exhibited a degree of plasticity, being limited by shallow soil strata and they also avoided a region of high pecan root density deeper in the soil. In another study conducted at the same site, Allen et al. (2003) reported a 34% reduction in nitrate nitrogen concentration in the soil solution at 0.9 m depth when pecan and cotton roots were not separated by a root barrier compared to the root-barrier treatment, suggesting the potential safety net role of pecan roots. This safety net hypothesis can also be used to explain the sediment and nutrient capture functions of riparian buffer strips. Riparian buffers have been presented as a tool for removing non-point source pollutants from farmlands (Addy et al., 1999; Bharati et al., 2002; Lee et al., 2003; Schultz et al. 2004). In a study conducted by Lee et al. (2003) in central Iowa, Midwestern United States, a switch grass (Panicum virgatum)/woody buffer treatment removed 90% of the sediment, 85% nitrate-N, and 80% of the P in the runoff. The integration of the woody component in the buffer strip increased the uptake efficiency of soluble nutrients by over 20%. Other applications of the safety-net include phytoremediation of contaminated sites (Rockwood et al., 2004) and rehabilitation of heavily fertilized agricultural soils in Florida soils (Nair and Graetz, 2004). Thus, while the safety-net concept is critical for enhancing nutrient use efficiency and improving crop production under tropical conditions, it has a different application in the heavily fertilized and excessively drained soils of Florida that have limited nutrient retention capacity. The tree roots ability to retrieve nutrients from deeper soil layers can be exploited advantageously to increase the apparent nutrient retention in excessively

PAGE 70

56 drained soils and thereby reduce the amount of nutrients that might otherwise be transported to surface water through runoff and leaching and thereby cause non-point-source pollution of these water bodies. Water soluble P (WSP) a environmental soil testing method for environmental assessment of available soil P, can better represent the rapid release of P to water than stronger chemical extractant (such as agronomic soil P tests), since rainfall has similar properties than distilled water (Moore et al., 1998). This extractant is also known to maintain the soil pH within one unit of its original value, an important characteristic since P solubility is strongly dependent on soil pH (Golterman, 1998; Sharpley, 1993). Furthermore, a strong positive correlation exists between water soluble P and P in runoff (Pote et al., 1996). The objective of this study was to quantify water soluble P (WSP), Mehlich-1 P, and nitrate and ammonium concentrations in surface and subsurface horizons in soil profiles of silvopastoral and treeless pasture systems at four different locations of two soil orders in Florida. Materials and Methods Site Description The research was conducted on four slash pine (Pinus elliottii) + bahiagrass (Paspalum notatum) silvopastoral systems located in Osceola (28 0 9' N, 81 0 10' W), Manatee (27 0 13' N, 82 0 8' W), Suwannee (30 0 24' N, 83 0 0' W), and Alachua (29 0 45' N, 82 0 33' W) counties in Florida. A grid sampling procedure was adopted to collect 24 soil profiles from each location during summer and fall 2004. The soil samples were collected by depth: 0 5, 5 15, 15 30, 30 50, 50 75 and 75 100 cm. The soils in both

PAGE 71

57 Manatee and Osceola counties are Spodosols, whereas those in Alachua and Suwannee are Ultisols. Refer to Chapter 3 for more in-depth details of the site description. Statistical Procedures The statistical analyses were performed using SAS PROC ANOVA (SAS 8.2 for Windows). The experimental design used was a 4-fold nested model (see Chapter 3 for the description of the model). Water Soluble P and Mehlich-1 P Determination Water soluble P was determined by extracting the soil with distilled dionized water using a 1:10 soil to water ratio. This mix was shaked with a mechanical shaker for 1 hour and filtered through a 0.45-m filter using vacuum filtration. Mehlich 1 (0.0125M H 2 SO 4 and 0.05M HCl)-extractable P was obtained using a 1:4 soil to double acid ratio (Mehlich, 1953). Phosphorus in solution was determined by an autoanalyzer (USEPA, 1993; Method 365.1) by the Murphy-Riley (1962) method. Nitrate and Ammonium-N Determination Nitrate and ammonium-N extracts were obtained using a 1:10 soil to KCL (2M) ratio. This mix was shaked on a mechanical shaker for 1 hour. The KCl suspension was then filtered by vacuum filtration as described above. A membrane filter of 0.45m was used for the extraction. After extraction ammonium-N in the KCl extractant was determined using a semi-automated colorimetric analysis (USEPA Method 350.1, USEPA, 1993). Nitrate was determined using a semi-automated colorimetric analysis (USEPA Method 353.2, USEPA, 1993). Only 6 out of the 24 soil profiles taken in each sites were analyzed for nitrate and ammonium content. With eight sites, each with 6 sampling points and six depths, a total of 288 soil samples were analyzed.

PAGE 72

58 Results and Discussion Water Soluble P Concentrations WSP concentrations at the 0 5 cm depth in the silvopasture and treeless pasture on Alachua were 5.6 and 10.9 mg kg -1 respectively. Corresponding values in Suwannee were 4.1 and 10.1 mg kg -1 compared to 11.4 and 23.4 mg kg -1 in Manatee, and values of 1.4 and 11.0 mg kg -1 in Osceola (Tables 4-1 and 4-2 and Figures 4-1 and 4-2). Differences in WSP concentrations were found through out the soil profile (up to 100 cm) between the two pasture systems in Alachua, Manatee and Suwannee. In Osceola, WSP concentrations was found to be different between the two pasture systems at 0 5, 15 30, 30 50, and 75 100 depths. Water soluble P was generally higher in the treeless pastures throughout the soil profile, indicating the safety net role of slash pine roots. Nair and Graetz (2004) found that water soluble P concentrations in the surface soils of a Spodosols varied throughout the soil profile in the order of pasture > silvopasture > native flatwoods vegetation. They reported WSP concentrations of 9.1, 4.2, and 1.6 mg kg -1 in the top 0 5 cm depth, respectively, in a treeless pasture, silvopasture, and native flatwoods. Even in an Ultisol with substantial P retaining clay below 20 cm, there appears to be movement of P to lower depths. Below the retentive clay layer the treeless pasture contained more P in the soil than the silvopasture (Figure 4-2) suggesting removal of P from the lower depths by the trees in the silvopasture systems. This observation suggests that the likelihood of P moving out of the soil is greater under a treeless pasture than under a tree-based system. These findings are consistent with that of a study conducted by Friend et al. (2006) on a loblolly pine (Pinus taeda) stand. In this study dry poultry-litter applications of approximately 5 Mg ha -1 to mid-rotation loblolly pine stands would be contained by the ecosystem and may result in significant increases in tree growth.

PAGE 73

59 Although we would expect to find higher WSP concentrations in the Spodosols, which have lower P retention capacity, especially in the surface A and E horizons as opposed to the Ultisols, the results do not support this expectation (Table 4-2). These findings could be explained by the hydrological dynamics of the Spodosols that have seasonally and highly fluctuating water table. Spodosols are characterized by a highly fluctuating water table situated between the Bh (a layer form by the accumulation of organic matter and/or iron and aluminum oxides) and the A horizons during the summer rainy season, which goes down to 125 cm during drier months (Soil Survey Staff, 1996). Rainfall that infiltrate into the soil during high water table conditions can move laterally, transporting P to surface drainage ditches (Burgoa et al., 1991; Mansell et al., 1991). Under this condition, water soluble P, the most labile form of P, is prone to lost via subsurface lateral flow before reaching the spodic horizon, where a P retention capacity exist.. Mehlich-1 P Concentrations In Alachua, the Mehlich-1 P concentrations at 0 5 cm depth for the silvopasture and treeless pasture were 97 and 104 mg kg -1 respectively. Corresponding values for the Suwannee, Manatee, and Osceola fields were 9 and 11 mg kg -1 ,5 and 6 mg kg -1 (Tables 4-1 and 4-2 and Figures 4-3 and 4-4). Mehlich-1 soil test values for both the silvopasture and the treeless pasture in Alachua and the treeless pasture in Suwannee fell within the high (31-60 mg kg -1 ) to the very high range (>60 mg kg -1 ) as suggested by UF/IFAS Mehlich-1 soil-test interpretation for agronomic crops (Kidder et al., 2002). In contrast, Mehlich-1 P concentrations in the silvopasture sites Suwannee and in both pasture systems in Manatee and Osceola fell generally in the very low range. No difference in Mehlich-1 P concentration was observed between the two pasture systems ( = 0.05) in

PAGE 74

60 Alachua at 0 5, 5 15, and 15 30 cm depths. However, the treeless pasture had higher P values at the 30 50, 50 75, and 75 100 cm depths. For the Suwannee site, Mehlich-1 P values below 30 cm were lower for silvopasture systems, which is indicative of the slash pine nutrient-capture effect. In general, no difference was observed between the two pasture systems at the Manatee and Osceola sites while the overall Mehlich-1 P levels at the sites was ten fold lower compared to Alachua and Suwannee. Presence of perched water table above the spodic soil layers enhances lateral P loss before it can reach the Bh horizon. In this case, the low P retention capacity of the A and E horizons, will control overall P movement. There is a need for further research to quantify the different nutrient pools in these agroforestry systems. Establishing long-term on-farm runoff plots and in-site wells monitoring will confirm the effectiveness of these agroforestry systems in reducing nutrient loss. Results from the current study showed that the Ultisols (Alachua and Suwannee) had greater Mehlich-1 P concentration then Spodosols (Osceola and Manatee) in the upper soil layers (up to 50 cm of depth) [Table 4.3]. This is consistent with the higher Al and Fe concentrations in the upper soil layers of Ultisols (compared to Spodosols), which increase the P sorption capacity of these soils (see Chapter 3). This is also consistent with the characteristics of Spodosols, having a sandy A and E horizons with poor sorption capacities while the spodic horizon has much greater capacity for P retention (Mansell et al., 1991). Chen and Lena (2001) reported that P values in undisturbed Florida surface soils were as follows: Histosols>Inceptisols>Mollisols>Entisols>Alfisols>Spodosols.

PAGE 75

61 Average values for Mehlich-1 P concentrations from the present study for the Alachua, Suwannee, Osceola and Manatee sites were, respectively, 100, 50, 10 and 6 mg kg -1 at depth 0 5 cm and 125, 7, 5, and 7 at depth 50 75 cm in the present study (Table 8-3). Nair and Graetz (2002) reported Mehlich-1 P concentrations of 14.9, 2.3, 5.2 and 8.1 respectively in the A, E, Bh, and Bw soil horizons of a beef pasture site (Spodosols) in the Okeechobee County of Florida. Nitrate and Ammonium-N Concentrations In Alachua, nitrate concentrations at 0 5 cm soil depth were 0.7 and 4.2 mg kg -1 respectively, in the silvopasture and the treeless pasture. Corresponding values for Suwannee were 1.9 and 10.8 compared to 0.8 and 0.0 in Manatee, and values of 2.0 and 0.2 for the Osceola site (Tables 4-1 to 4-2 and Figures 4-5 and 4-7). Ammonium concentrations in the 0 5 cm soil depth were 13.5 and 16.7, respectively, in the silvopasture and treeless pasture in Alachua, and 10.5 and 6.0 in Suwannee, 13.6 and 14.0 in Manatee, 2.3 and 2.2 in Osceola (Tables 4-1 to 4-2 and Figures 4-6 and 4-8). In Suwannee higher nitrate-N concentration was found in the treeless pasture as compared to the silvopasture at 0 5, 50 75, and 75 100 cm depths, whereas the ammonium-N concentration was only different at 0 5 cm depth. Contrary to our expectations, no nitrateand ammonium-N concentrations were generally observed among the pasture systems in Alachua, Osceola and Manatee. Nitrate and ammonium-N levels were generally very low, sometimes below the detection level. For a similar study in a pecan (Carya illinoensis) + cotton (Gossypium hirsutum) alley cropping system on Ultisols in northwestern Florida, Allen et al. (2003) reported a 34% reduction in nitrate concentration in the soil solution at 0.9 m depth when pecan and cotton roots were not separated by a root barrier compared to the root-barrier

PAGE 76

62 treatment., suggesting the potential safety net ro le of pecan roots. It is difficult to account for nitrate in most soil. Nitrate, a negatively charged ion, is not retained by the adsorbing complex of the soil and thus is very prone to leaching. Concentrati ons of these labile N species are in constant flux. It is also impor tant to bear in mind that most soil ammonium will be transformed biologically to nitrate within 2 to 3 weeks at soil temperatures in the 75 to 90 F range (Boman and Obreza, 2002). Another avenue for pot ential nitrate losses in soil with highly fluctuating water table such as Spodosol s (Osceola and Manatee) is denitrification. Under anaerobi c conditions, nitrate is transf ormed into gaseous forms and lost to the atmosphere. Also, the nitrate-N could have moved laterally from the soil via subsurface flow. Conclusions In spite of the limitations of the study, the resu lts show that silvopastures sites had lower P levels in the soil pr ofile than treeless pastures s uggesting that there is a tree effect. Silvopasture will increase the nut rient stock in the plant system following enhanced total nutrient uptake by trees and a ssociated forage species, compared to the shallow-rooted forage species. The amount of P that could be transported to surface water will thus be reduced in the silv opstoral system. Although a similar trend was observed in the case on nitrogen, the differe nces between the two systems were less striking, possibly because of the difficu lties in measuring the highly dynamic soil nitrogen levels.

PAGE 77

63 Table 4-1. Mean values of water soluble P (WSP), Mehlich-1 P (M-1 P), and nitrate and ammonium-N concentrations for different soil layers of soil profiles at the center of the alley of the silvopasture and a treeless pasture sites in Alachua and Suwannee Florida, USA. Alachua Alachua Alachua Silvopasture Treeless pasture Silvopasture Treeless pasture Silvopasture Treeless pasture P Depth (cm) WSP (mg kg -1 ) P M-1 P (mg kg -1 ) P NO 3 (mg kg -1 ) 0-5 5.6 10.9* 0.0001 96.5 104.1 0.7011 0.7 4.2 0.0960 5-15 2.8 8.5* 0.0001 99.5 125.8 0.1602 0.3 0.8 0.3358 15-30 1.9 7.9* 0.0001 83.6 108.6 0.1669 0.2 0.3 0.4404 30-50 1.7 7.3* 0.0001 61.4 125.4* 0.0079 0.1 0.2 0.2271 50-75 1.5 6.2* 0.0001 49.1 171.9* 0.0001 0.0 0.0 0.1059 75-100 1.4 6.7* 0.0001 49.6 200.4* 0.0001 0.0 0.1 0.2345 Alachua Suwannee Suwannee Silvopasture Treeless pasture Silvopasture Treeless pasture Silvopasture Treeless pasture P Depth (cm) NH 4 (mg kg -1 ) P WSP (mg kg -1 ) P M-1 P (mg kg -1 ) 0-5 13.5 16.7 0.1742 4.1 10.1* 0.0001 8.6 90.6* 0.0001 5-15 7.1 7.0 0.9443 1.2 6.4* 0.0001 17.0 88.0* 0.0001 15-30 3.5 5.2 0.0730 0.6 4.6* 0.0001 11.0 72.5* 0.0001 30-50 3.1 4.1 0.1282 0.4 2.9* 0.0001 6.0 42.6* 0.0001 50-75 2.4 3.1 0.0732 0.3 1.7* 0.0001 3.4 19.7* 0.0001 75-100 2.4 2.8 0.4612 0.2 1.5* 0.0001 2.6 11.6* 0.0001 Suwannee Suwannee Silvopasture Treeless pasture Silvopasture Treeless pasture Depth (cm) NO 3 (mg kg -1 ) P NH 4 (mg kg -1 ) P 0-5 1.9 10.8* 0.0028 10.5* 6.0 0.0149 5-15 0.3 3.0* 0.0001 3.7 4.3 0.3138 15-30 0.1 1.08* 0.0001 2.7 2.5 0.5804 30-50 0.1 0.9* 0.0001 2.9 2.2 0.0765 50-75 0.0 0.7* 0.0050 1.9 2.3 0.1599 75-100 0.1 0.7* 0.0298 2.4 2.0 0.3726 [(*Indicates significant P values within a given depth (p<0.05)]

PAGE 78

64 Table 4-2. Mean values of water soluble P (WSP), Mehlich-1 P (M1-P), nitrate, ammonium concentrations of different soil layers for soil profiles at the center of the alley of the silvopasture and a treeless pasture sites in Manatee and Osceola, Florida, USA. Manatee Manatee Manatee Silvopasture Treeless pasture Silvopasture Treeless pasture Silvopasture Treelesspasture Depth (cm) WSP (mg kg-1) P M-1 P (mg kg-1) P NO3 (mg kg-1) P 0-5 11.4 23.4* 0.0009 8.9 10.9 0.1796 0.8 0.0 0.3097 5-15 1.43 4.60* 0.0419 0.7 1.5* 0.0006 0.8 0.0 0.2583 15-30 0.30 0.60* 0.0001 0.2 0.1 0.4311 0.5* 0.0 0.0203 30-50 0.10 0.43* 0.0001 0.0 0.3 0.1472 0.2* 0.0 0.0282 50-75 0.04 0.33* 0.0001 1.0 3.3* 0.0095 0.7* 0.3 0.0150 75-100 0.05 0.95* 0.0002 2.3 8.1* 0.0006 0.2 0.2 0.5636 Manatee Osceola Osceola Silvopasture Treeless pasture Silvopasture Treeless pasture Silvopasture Treelesspasture Depth (cm) NH4 (mg kg-1) P WSP (mg kg-1) P M-1 P (mg kg-1) P 0-5 13.6 14.0 0.9024 6.8 11.0* 0.0352 5.0 6.1 0.1213 5-15 3.4 4.0 0.3212 1.4 1.4 0.9548 1.7 1.9 0.7390 15-30 2.8 2.0 0.3035 1.0* 0.5 0.0208 1.4* 0.5 0.0487 30-50 1.5 1.5 1.0000 0.7* 0.3 0.0320 3.9 1.0 0.1009 50-75 1.4 1.6 0.3156 0.6 0.4 0.1360 7.1* 2.3 0.0405 75-100 2.0 1.8 0.5706 0.9* 0.5 0.0164 11.5* 2.7 0.0002 Osceola Osceola Silvopasture Treeless pasture Silvopasture Treeless pasture Depth (cm) NO3 (mg kg-1) P NH4 (mg kg-1) P 0-5 2.0* 0.2 0.0068 2.3 2.2 0.4956 5-15 0.8 0.6 0.6081 12.4 10.5 0.0152 15-30 1.7 0.5 0.0786 5.0 5.8 0.3807 30-50 3.0 0.3 0.1011 3.1 3.3 0.3391 50-75 0.3 0.2 0.3369 3.0 2.5 0.4652 75-100 0.4 0.1 0.1876 2.1 2.1 0.1421 [(*Indicates significant P values within a given depth (p<0.05)]

PAGE 79

65 Table 4-3. Mean values of Mehlich-1 extractable-P (M1-P) concentrations for soil profiles of the Spodosols (O sceola and Manatee) a nd Ultisols (Alachua and Suwannee) in Florida. Depth (cm) Alachua Suwannee Manatee Osceola 0-5 100a 50b 10c 6c 5-15 113a 52b 2c 1c 15-30 96a 42b 1c 0.1c 30-50 93a 24b 3c 0.2c 50-75 111a 12b 2b 5b 75-100 125a 7b 5b 7b *Means separation by Waller-Duncan procedure. Mean values of soil characteristics within a given depth followed by the same letter are not different (p<0.05)

PAGE 80

0102030405060708090100051015202530WSP concentration (mg kg-1)Soil depth (cm) Silvopasture (Alachua) Treeless (Alachua) Silvopasture (Suwannee) Treeless pasture (Suwannee) 66 Figure 4-1. Water soluble P (WSP) concentration at different soil layers for silvopasture and treeless pasture at the Alachua and Suwannee field sites featuring Spodosols. Error bars indicate standard errors (n = 24).

PAGE 81

0102030405060708090100051015202530WSP concentration (mg kg-1)Soil depth (cm) Silvopasture (Osceola) Treeless pasture (Osceola) Silvopasture (Manatee) Treeless pasture (Manatee) Bh (Osceola)Bh (Manatee) 67 Figure 4-2. Water soluble P (WSP) concentration by depth in the silvopasture and treeless pasture sites in Osceola and Manatee. Error bars designate standard errors.

PAGE 82

0102030405060708090100050100150200250Mehlich-1 P concentration (mg kg-1)Soil depth (cm) Treeless pasture (Alachua) Silvopasture (Alachua) Silvopasture (Suwannee) Treeless pasture (Suwannee) 68 Figure 4-3. Mehlich-1 P concentration by depth in the silvopasture and treeless pasture sites in Alachua and Suwanneee. Error bars designate standard errors.

PAGE 83

010203040506070809010005101520Mehlich-1 P concentration (mg kg-1)Soil depth (cm) Treeless pasture (Manatee) Silvopasture (Manatee) Silvopasture (Osceola) Treeless pasture (Osceola) 69 Figure 4-4. Mehlich-1 P concentration by depth in the silvopasture and treeless pasture sites in Osceola and Manatee. Error bars designate standard errors.

PAGE 84

010203040506070809010005101520Nitrate concentration (mg kg-1)Soil depth (cm) Silvopasture (Suwannee) Treeless pasture (Suwannee) Silvopasture (Alachua) Treeless pasture (Alachua) 70 Figure 4-5. Nitrate concentration level by depth in silvopasture and treeless pasture sites in Alachua and Suwannee on Ultisols. Error bars designate standard errors.

PAGE 85

010203040506070809010005101520Ammonium concentration(mg kg-1)Soil depth (cm) Silvopasture (Alachua) Treeless pasture (Alachua) Silvopasture (Suwannee) Treeless pasture (Suwannee) 71 Figure 4-6. Ammonium-N concentrations levels by depth in silvopasture and treeless pasture sites in Alachua and Suwannee (Ultisols sites). Error bars designate standard errors.

PAGE 86

72 0102030405060708090100012345Nitrate concentration (mg kg-1)Soil depth (cm) Treeless pasture (Manatee) Silvopasture (Manatee) Silvopasture (Osceola) Treeless pasture (Osceola) Figure 4-7. Nitrate-N concentration for different soil depth for silvopasture and treeless pasture systems for filed sites in Manatee and Osceola (Spodosols). Error bars designate standard errors.

PAGE 87

73 0102030405060708090100024681012141618Ammonium concentration(mg kg-1)Soil depth (cm) Silvopasture (Manatee) Treeless pasture (Manatee) Silvopasture (Osceola) Treeless pasture (Osceola) Figure 4-8. Ammonium-N concentrations levels for different soil depth in silvopasture and treeless pasture sites in Manatee and Osceola on Spodosols. Error bars designate standard errors

PAGE 88

CHAPTER 5 SOIL PHOSPHORUS STORAGE CAPACITY UNDER A SLASH PINE (Pinus elliottii) + BAHIAGRASS (Paspalum notatum) SILVOPASTURE AND A TREELESS PASTURE Introduction Phosphorus loss from farmlands has been identified as one of the main causes of eutrophication of fresh water bodies. Phosphorus loading of water bodies associated with runoff from operations and heavy manuring of pasturelands in soils is a serious environmental concern. The incorporation of trees in pastures could enhance P storage in the soil-plant system (Nair and Graetz, 2004; Nair and Kalmbacher, 2004; Jose et al., 2004; Chapter 4 of the present study). Silvopasture, the intentional combination of trees, forage plants and livestock into an integrated, intensively-managed system (Nair, 1993; Garrett et al., 2000; Nair et al., 2004) is potentially a promising land-use system to enhance the economic returns of forestry operations (Lundgren et al., 1983; Lewis et al, 1985; Dangerfield and Harwell, 1990; Clason, 1995; Grado et al., 2001; Stainback and Alavalapati, 2004). Studies on tropical agroforestry systems in nutrient-depleted soils have demonstrated the capacity of deep rooted trees to absorb nutrients that have leached below the rooting zone of agronomic crops, and recycle these nutrients through litterfall and rootfall in the crop-root zone, thus improving the nutrient-use efficiency in the system as a whole (Young, 1997). In addition, the longer lifespan of tree roots help capture nutrients such as nitrate and phosphorus before and after the crop is planted and 74

PAGE 89

75 harvested (Jose et al., 2004). This would enhance the overall resource-use efficiency of the system (van Noordwijk et al., 1996). Research conducted in temperate agroforestry systems have showed similar results. In a study conducted in pecan (Carya illinoensis K. Koch)cotton (Gossypium hirsutum L.) alley cropping system in northwestern Florida, Allen et al. (2003) reported a 34% reduction in nitrate nitrogen concentration in the soil solution at 0.9 m depth when pecan and cotton roots were not separated by a root barrier compared to the root barrier treatment, suggesting the potential safety-net role of pecan roots. The nutrient interception capacity of riparian buffer is another example of the safety-net concept. Riparian buffer strips offer a means of removing non-point source pollutants from farmlands (Addy et al., 1999; Bharati et al., 2002; Lee et al., 2003; Schultz et al., 2004). Lee et al. (2003), in a study undertaken in central Iowa, Midwestern United States, reported that a switch grass (Panicum virgatum)-woody buffer treatment removed 85% nitrate-N, and 80% of the phosphate-P in the runoff. Including a woody component in the buffer strip enhanced the uptake effectiveness of soluble nutrients by over 20%. According to Anbumozhi et al. (2005) forested buffers reduced agricultural nutrient losses from livestock activities in adjacent uplands in surface stream waters in the Tokachigawa watershed in Japan. In another study conducted in several European countries significantly higher N uptake from forested sites as compared to herbaceous sites was observed (Hefting et al., 2005). Despite the fact that the safety-net concept is important to enhancing nutrient use efficiency and improving crop production under tropical ecological settings, it has another application in the heavily fertilized and sandy soils of Florida (Nair and Graetz,

PAGE 90

76 2004) that have low nutrient retention capacities. The capacity of tree roots to capture nutrients from the deeper soil horizons can enhance nutrient storage in the plant-soil system and therefore reduce the amount of nutrients that might otherwise be transported to ground and surface water through runoff and leaching and thus cause non-point-source pollution of these water bodies. Environmental indicators of P loss such as soil test P (STP) or degree of P saturation (DPS) do not allow any assessment of the capacity of a soil to retain future addition of P (Nair and Harris, 2004). To address this shortcoming, these authors proposed a DPS-based calculation of the remaining Soil P Storage Capacity (SPSC). This index takes into account both the risks attributed to prior loading and the inherent, limited P sorption capacity of a soil. The SPSC offers a direct estimate of the amount of P a given soil can hold before exceeding a threshold soil concentration, consequently, resulting in water pollution. Generally, the P sorption maximum of a soil is determined by using the Langmuir adsorption equation (Nair et al., 1998), which is a complex and time-consuming method. Another way of estimating the soil P storage capacity of a soil is to use a single point isotherm (or one point) value. Simple measurements of a single-point isotherm were shown to reasonably estimate P sorption capacities of soils (Harris et al., 1996; Indiati et al., 2002; Nair and Graetz, 2002; Lu and O'Connor, 2004 ). The objectives of this study were to (i) assess differences in SPSC estimates in the surface and subsurface horizons in soil profiles of two pasture systems (silvopasture and treeless pasture) at four locations including two different soil orders in Florida, and (ii)

PAGE 91

77 assess differences in SPSC estimates using two methods (Mehlich-1 and single-point isotherm). Material and Methods Site Description An on-farm research was undertaken in Osceola (28 0 9' N, 81 0 10' W), Manatee (27 0 13' N, 82 0 8' W), Suwannee (30 0 24' N, 830 0' W), and Alachua (29 0 45' N, 82 0 33' W) counties in Florida. The silvopastoral systems consisted of a slash pine (Pinus elliottii)bahiagrass (Paspalum notatum) and the treeless pasture consisted only of bahiagrass. A grid sampling procedure was used to collect 24 soil profiles from each site at each location during summer and fall 2004. The soil samples were collected at six depths: 0 5, 5 15, 15 30, 30 50, 50 75 and 75 100 cm. The soils in both Alachua and Suwannee are Ultisols, whereas those in Manatee and Osceola counties are Spodosols. Refer to Chapter 3 for more in-depth details on the site description. Refer to chapter 3 for an in depth description of the sites characteristics. Statistical Procedures The statistical analyses were performed using SAS PROC ANOVA (SAS 8.2 for Windows). The experimental design used was a 4-fold nested model (see Chapter 3). The regression analyses were performed using SAS PROC REG and PROC GLM (SAS 8.2 for Windows). Calculations of Parameters Soil P storage capacity calculated with Mehlich-1 The phosphorus saturation ratio (PSR) was computed as the molar ratio of Mehlich1-P and Mehlich1-Fe and Al (Nair et al., 2004). The PSR formula is as follows: PSR = (Mehlich-1 P 31) ((Mehlich-1 Fe 55.8) + (Mehlich-1 Al 27))

PAGE 92

78 The soil phosphorus storage capacity (SPSC) was calculated as: SPSC = (0.15 PSR) ((Mehlich-1 Fe 55.8) + (Mehlich-1 Al 27)) 31 (mg P kg -1 ) (Nair and Harris, 2004) The SPSC, is thus an estimate of the amount of P that can safely be applied to a soil up to a defined depth before the soil poses an environmental risk. The SPSC values were then calculated on a kg ha -1 basis (for a specified depth) assuming a bulk density of 1500 kg m -3 (Nair and Harris, 2004). For the calculation of SPSC, the choice of PSR instead of DPS was made solely for simplicity purpose. The SPSC of the soil was based on a threshold PSR of 0.15. According to Nair et al. (2004), the value of 0.15 used in the SPSC calculation is the best approximation available to date of the PSR value corresponding to the critical P solution concentration of 0.10 mg P L -1 as proposed by Breeuwsma and Silva (1992). Soil P storage capacity calculated with Single Point Isotherm Data The single point isotherm was determined following the procedure described in Figure 5-1and using 4 sampling points from each study site. With eight plots, each with 4 sampling points and six depths, a total of 192 soil samples were used for the analysis.

PAGE 93

79 20 mL of 0.01M KCl solution containing 100 mg P L-1 24-h equilibration period using a mechanical shaker 2g air-dried soil 1-h suspension settling down Vacuum filtration using 0.45 m filter Figure 5-1. Determination of P sorption capacity using a single point isotherm based on a procedure by Nair et al. (1998). The filtrate was then analyzed for P using the Murphy-Riley procedure (Murphy and Riley, 1962). Following the single point isotherm determination, the phosphorus sorption was determined by subtracting the P added to the P in solution. The PSR and SPSC determined by the single point isotherm were then calculated using the following formulas: (i) PSR = Melhich-1 P (P sorption + Mehlich-1 P) (ii) SPSC = (0-15 PSR) (P sorption + Mehlich-1 P) (mg P kg -1 ) When the SPSC becomes negative, i.e. PSR > 0.15, the soil is considered to be a net source of P source.

PAGE 94

80 Results and Discussion Phosphorus Saturation Ratio Phosphorus saturation ratios for the 0 5 cm soil depth in the silvopasture and treeless pasture in Alachua were 0.17 and 0.19, respectively. Corresponding values for the Suwannee site were 0.05 and 0.93, compared to 0.15 and 0.22 (Manatee), and 0.06 and 0.45 in Osceola (Table 5-1). These findings suggest that the surface soils under treeless pasture have a greater potential to release P irrespective of the location or the soil order of the land use. Concentrations of Water Soluble P (WSP) (Chapter 4) also indicate higher values in the surface soils for treeless pasture compared to silvopasture, therefore supporting the PSR results. Results of the study showed differences between the silvopasture and treeless pasture systems at the 50 75, 75 100 cm soil depth in Alachua, with the treeless pasture having the highest PSR values. In Osceola PSR values were similar between the two systems for all sampling depths. In Manatee and Suwannee, differences were observed between the two pasture systems throughout the soil profile. The silvopasture sites showed lower PSR values than the treeless pasture sites, indicative of a tree effect. This is consistent with the results from the water soluble P and the Mehlich-1 P analysis obtained at the same study sites (see Chapter 4 for details). Results of the study showed lower PSR values in the Spodosols as compared to the Ultisols from depth 15 30 to 50 75 cm (Table 5-2). The lower PSR values observed in the Spodosol sites may be attributed to a low soil Mehlich-1 P associated with Spodosols characterized by a highly fluctuating and perched water tables resulting in subsurface lateral flow in the upper soil layers which have the low P sorption capacities. As a result, appreciable P losses will occur since drainage water may not pass through the spodic

PAGE 95

81 horizon which has a high P retention capacity and may function as a P filter (Campbell, 1995). Relationships between WSP and PSR for the soil surface (0 5 and 5 15 cm depths) of the study sites at the four locations are shown in Fig. 5-2. The regression analysis showed that the cubic model had the highest r-squared (r 2 = 0.33) [Appendix] but for practical reason the linear equation was chosen (r 2 = 0.31). The low r-square observed for the relation is mainly due to site variations, in particular from Alachua sites which have shown evidence of the presence of phosphatic materials. Figure 5.2 shows a possible changing point at a PSR value between 0.05 and 0.10 where the WSP sharply increases. Nair et al. (2004), indicated a PSR of 0.15 as a threshold PSR at which a sharp increase in WSP can occur in the A horizon and subsurface horizons in Florida Spodosols. Soil Phosphorus Storage Capacity Remaining soil P storage capacity at 0 5 cm depth for the silvopasture and treeless pasture in Alachua were -14.3 and -13.6 kg ha -1 respectively. Corresponding values were 15.75 and -43.66 in Suwannee, 0.6 and -2.5 in Manatee, and 26.2 and 14.1 in Osceola (Table 5-1). Results of the study showed significant differences between the silvopasture and treeless pasture systems at depth 50 75, 75 100 cm in Alachua with the treeless pasture having the lowest SPSC values. In Osceola, no differences were observed throughout the soil profile. In Manatee no differences were observed between the two pasture systems at 30 50 and 50 75 cm depths. In Suwannee SPSC values were consistently higher for silvopasture systems throughout the entire soil profile. The highly negative SPSC found in the deeper soil horizons in Alachua sites may be explained by the presence of phosphatic materials in the deeper sampling depths of the sites (Table 5-1). Ramnarine (2003) identified phosphatic soils in Alachua County,

PAGE 96

82 Florida. These soils showed an increase in total phosphorus (TP) with depth, which is different for most nonphosphatic soils. At the Suwannee site total SPSC to one meter depth in the silvopasture and treeless pasture were 337 and -59 P kg ha -1 respectively. Corresponding values for the Manatee site were 329 and 191 kg P ha -1 compared to 657 and 926 kg P ha -1 in Osceola), and values of -36 and -542 kg ha -1 in Alachua (Table 5-3). Differences between the two pasture systems were significant for the Alachua, Suwannee and Manatee sites (Table 5-3). In general, the capacity of the soil to store additional P was lower in the treeless pasture sites as compared to silvopasture sites (Figure 5-3). These findings along with the results of the water soluble P (WSP) analysis (Chapter 3) suggest that the chances for P to move out of the soil are greater under treeless pasture than under silvopasure, a tree-based system. The bulge in the graph for the Spodosol in Osceola (Figure 5-4) corresponds to the sandy E horizon at the 30 50 cm depth where the storage capacity of the soil (calculated from a low PSR resulting from a low detectable Mehlich-1 P for the reason explained earlier) would result in erroneous calculations for SPSC. Estimation of Mehlich-1-based Soil P Storage Capacity (SPSC-M1) and Soil P Storage Capacity-Single Point Isotherm (SPSC-ISO) Results from the study sites showed that SPSC values determined with Mehlich-1 were lower than that of the single point isotherm method (Figure 5-5). In other words, soil P storage capacity was underestimated when the Mehlich-1 method was used, as opposed to the single point isotherm. This is consistent with a study by Chrysostome (2005), who found that the Mehlich-1 method underestimated SPSC compared to the oxalate method. In this study Mehlich-1 Fe and Al concentrations and oxalate extractable

PAGE 97

83 Fe and Al concentrations were respectively 13 and 129 mg kg -1 vs. 268 and 327 mg kg -1 respectively. Even though Mehlich-1-based soil P storage capacity and SPSC-ISO give different values, they are nevertheless correlated. The regression analysis showed that the cubic model had the highest r-squared (r 2 = 0.57) [Appendix] but for practical reason the linear equation was chosen (r 2 = 0.50). The relationship between SPSC-M1 and SPSC-ISO (Figure 5-5) for the soil surface (0 5 and 5 15 cm depths) of the study sites is given by the following formula: SPSC-M1 = 0.65 (SPSC-ISO) + 29.77; r 2 = 0.50, p <.0001 The general relationship indicates that the Mehlich-1 SPSC calculation method underestimates the remaining P storage capacity. The results are consistent with the findings of a research conducted by Chrysostome (2005) in which a comparison was made between SPSC calculated using Mehlich-1 and oxalate (SPSC oxalate ). The research concluded that SPSC oxalate resulted in a more accurate estimation of the remaining capacity of a soil to sorb P than Mehlich-1-based SPSC, because oxalate is a more powerful extractant than Mehlich-1 and extracts most of the reactive Al and Fe present in the soil and representing its P sorption capacity. Chrysostome (2005) recommended the use of a correction factor for Mehlich-1-based SPSC to compensate for the underestimation of the remaining capacity of the soil to adsorb additional P using this extractant. Several researchers (Harris et al., 1996; Indiati et al., 2002; Nair and Graetz, 2002; Lu and O'Connor, 2004) have observed reasonable estimations of the maximum soil P sorption capacity using the single point isotherm method and therefore, calculating

PAGE 98

84 SPSC using this method is likely to be comparable to the oxalate de termination. Further research is needed to validate this statement. Conclusions The presence of trees in pastures may reduce the chances of future additions of P to move out of the soil. This, in turn, suggests that silvopastoral systems would provide a greater environmental service in regards to water qual ity protection compared to treeless pastures under comparable ecological settings. The singl e point isotherm method seems to better estimate the remaining Soil P Storage Capac ity as compared to the Mehlich-1 method.

PAGE 99

85 Table 5-1. Mean values of phosphorus saturation ratio (PSR) and soil P storage capacity (SPSC) for soil profiles at the center of the alley of the silvopasture and a treeless pasture sites in Alachua, Suwannee, Manatee and Osceola, Florida, USA Alachua Alachua Suwannee Silvopasture Treeless pasture Silvopasture Treelesspasture Silvopasture Treeless pasture P Depth (cm) PSR P SPSC (mg P kg -1 ) P PSR 0-5 0.19 0.18 0.5904 -14.3 -13.6 0.9504 0.04 0.42 0.0099* 5-15 0.17 0.19 0.4309 -17.1 -38.0 0.3184 0.06 0.28 0.0257* 15-30 0.16 0.17 0.9199 -16.8 -30.6 0.6331 0.04 0.24 0.0168* 30-50 0.14 0.17 0.3228 7.4 -51.1 0.2256 0.03 0.16 0.0226* 50-75 0.14 0.22* 0.0005 6.7* -207 0.0025 0.02 0.09 0.0012* 75-100 0.15 0.23* 0.0005 -4.4* -260 0.0039 0.02 0.06 0.0048* Suwannee Manatee Manatee Silvopasture Treeless pasture Silvopasture Treelesspasture Silvopasture Treeless pasture P Depth (cm) SPSC (mg P kg -1 ) P PSR P SPSC (mg P kg -1 ) 0-5 16* -44 0.0001 0.14 0.21* 0.0194 0.6* -2.4 0.0115 5-15 44* -61 0.0001 0.02 0.05* 0.0002 6* 4 0.0171 15-30 74* -60 0.0001 0.01 0.01* 0.0200 7* 4 0.0365 30-50 81* -6 0.0001 0.00 0.01* 0.0427 31 9 0.2653 50-75 83* 52 0.0211 0.00 0.02* 0.0195 108 83 0.2234 75-100 73 57 0.1203 0.01 0.04* 0.0001 175* 81 0.0002 Osceola Osceola Silvopasture Treeless pasture Silvopasture Treelesspasture Depth (cm) PSR P SPSC (mg P kg -1 ) P 0-5 0.02 0.04* 0.0885 26.2 14.1 0.2491 5-15 0.01 0.01 0.7654 45.3 41.2 0.8594 15-30 0.01 0.00 0.8633 69.3 92.1 0.5666 30-50 0.01 0.00 0.1284 144 406 0.0706 50-75 0.02 0.00 0.0637 229 270 0.5318 75-100 0.02 0.01 0.0724 234 227 0.8971 (*Indicates significant P values within a given depth (p<0.05)].

PAGE 100

86 Table 5-2. Mean values of phosphorus saturation ratio (PSR) for soil profiles of the Spodosols (Osceoala and Manatee) and Ultisols (Alachua and Suwannee) in Florida. Depth (cm) PSR Alachua Suwannee Osceola Manatee 0-5 0.18b* 0.48a 0.26ab 0.18b 5-15 0.17ab 0.27a 0.06bc 0.04c 15-30 0.16a 0.20a 0.01b 0.006b 30-50 0.15a 0.15a 0.009b 0.003b 50-75 0.17a 0.07b 0.01c 0.01c 75-100 0.17a 0.05b 0.03c 0.02bc *Statistical analysis by Waller-Duncan procedure. Mean values of soil characteristics of silvopasture and treeless pasture within a given depth followed by the same letter are not significantly different (p<0.05).

PAGE 101

87 Table 5-3. Cumulative soil P storage capacity (SPSC) for the 0-1 m soil depth at the center of the alley of the silvopasture and a treeless pasture sites in Alachua, Suwannee, Manatee and Osceola, Florida, USA. Location Treatment SPSC (kg P ha -1 ) P Alachua Silvopasture -36 Treeless pasture -542* 0.0186 Manatee Silvopasture 329* Treeless pasture 191 0.0051 Osceola Silvopasture 657 Treeless pasture 926 0.2182 Suwannee Silvopasture 342* Treeless pasture -60 0.0001 [(*Indicates significant P values (p<0.05)]

PAGE 102

88 y = 29.21x + 0.74r2 = 0.30CV = 137.30102030405060708000.10.20.30.40.50.60.70.8PSRWSP (mg kg-1) Figure 5-2. Relationship between phosphorus saturation ratio (PSR) and Water soluble P (WSP) for surface soils (0-5 and 5-15 cm) of the study sites in Florida

PAGE 103

89 0102030405060708090100-150-100-50050100150SPSC (kg ha-1)Soil depth (cm) Silvopasture Treeless pasture Figure 5-3. General trend in soil P storage capacity (SPSC) in the center of the alley of the silvopasture and a treeless pasture sites in Florida.

PAGE 104

90 Figure 5-4. Soil P storage capacity (SPSC) by depth at the center of the alley of the silvopastures and treeless pasture sites in Osceola (error bars represent one standard error from the mean, n=24). 0102030-1000100200300400500SPSC (kg ha-1) 405060708090100Soil depth (cm) Silvopasture Treeless pasture Osceola

PAGE 105

91 y = 0.65x + 29.77r2 = 0.50CV = 228.8-200-150-100-50050100150200250300-250-200-150-100-50050100150SPSC-ISO (mg P kg-1)SPSC-M1 (mg P kg-1) Figure 5-5. Relationship between soil P storage capacity single point isotherm (SPSC-ISO) and soil P storage capacity-Mehlich-1 (SPSC-M1) for the surface (0 5 and 5 15 cm depths) of the study sites in Florida.

PAGE 106

CHAPTER 6 SUMMARY AND CONCLUSIONS Phosphorus (P) and nitrate-N loss from sandy soils that predominates pasturelands in Florida is a major cause of nutrient pollution of water bodies. Our overall hypothesis is that the loss of P to surface and ground water could be reduced in the silvopastoral systems as compared to treeless pastures. Presence of trees will enhance both nutrient retrieval from deeper soil layers and also increase nutrient accumulation capacity. As a result, silvopastoral systems can reduce nutrients losses and environmental impacts associated with farm operations. A series of studies was conducted in Osceola (28 0 9' N, 81 0 10' W), Manatee (270 13' N, 82 0 8' W), Suwannee (30 0 24' N, 83 0 0' W), and Alachua (29 0 45' N, 82 0 33' W) counties in Florida, investigating the safety net function of slash pine (Pinus elliottii) + bahiagrass (Paspalum notatum) silvopastoral systems on heavily manured and porous pastureland soils in Florida. The goal of these studies were to evaluate if the presence of trees on pastureland would decrease soil P levels and thereby increase the capacity of the soil to store additional P. Soil nitrate-nitrogen (NO 3 -N and NH 4 -N) and phosphorus (P) concentrations were measured in soil profiles of the two pasture production systems (silvopasture vs. treeless pasture) at six depths: 0 5, 5 15, 15 30, 30 50, 50 75 and 75 100 cm. The remaining soil P storage capacity, which represents the maximum amount of P that can be safely applied to a soil before it becomes an environmental concern, was then estimated. 92

PAGE 107

93 Soil characterization studies (Chapter 3) confirmed the coarse texture and acidic conditions of the soils at the study sites, typical of the southeastern USA. Ultisols had higher soil Al and Fe concentrations in the top 30 cm of depth, compared to Spodosols and they also had higher clay content in the deeper 75 100 cm soil layer. The safety-net-validation study (Chapter 4) showed that silvopastures sites contained less P in the soil profile than treeless pastures suggesting a tree effect on nutrient retention in soils. Silvopasture would possibly increase the nutrient stock in the plant system following enhanced total nutrient uptake by trees and associated forage species, compared to the shallow-rooted forage species. The amount of P that could be transported to surface water could thus be reduced in the silvopstoral system. Although a similar trend was observed in the case on nitrogen, the differences between the two systems were less striking, possibly because of the difficulties in capturing the highly dynamic soil nitrogen dynamics at a single point in time. Based on estimates of the remaining soil P storage capacity of the soils (Chapter 5) it is concluded that that the presence of trees appears to enhance remaining P storage capacity and thus may reduce potential P losses, particularly in Ultisols. This, in turn, would suggest that silvopastoral systems would provide a greater environmental service in regard to water quality protection compared to treeless pastures under similar ecological settings. The single point isotherm method provided a better estimate of the remaining Soil P Storage Capacity compared to the Mehlich-1 method. These studies indicate that silvopastoral systems offer a means of managing nutrient inputs for enhanced nutrient use efficiency and surface and quality groundwater preservation. These results are, however, of preliminary nature. Several unknown factors

PAGE 108

94 remain to be investigated in more detail. For example, the relative impact of these agroforestry systems may vary depending on a number of site-specific factors such as tree and crop species, system design and management practices. Actual absorption and retention capacity of both silvopastoral and treeless pasture systems should be compared using labeled isotope and other approaches. Similarly, the effect of relatively lower soil nutrient levels in silvopastoral system as compared to treeless pasture system on N and P leaching and run-off losses should be demonstrated using run-off samplers and sampling wells. Establishing long-term on-farm runoff plots and on-site wells for monitoring nutrient dynamics in this plant-soil-water continuum will help confirm the effectiveness of these agroforestry systems in reducing nutrient loss. Overall, the findings from this study offer a preliminary, yet significant contribution to our understanding of the nutrient-capture capacity of tree-based systems and their potential use in the remediation of P-impacted sites in Florida. These results provide a conceptual framework for additional research that may facilitate enhanced utilization of slash pine + bahiagrass silvopastural systems to provide environmental and ecological services. Such systems may afford non-industrial private landowners and livestock operators with a viable alternative for current land uses.

PAGE 109

APPENDIX REGRESSION ANALYSIS 1. Regression analysis of water soluble P and phosphorous saturation ratio for the top 0-15 cm depth. Parameter* Estimate Error t value P Intercept 0.74 0.18 4.08 0.0001 linear 29.21 1.31 22.35 0.0001 Parameter Estimate Error t value P Intercept 0.21 0.20 1.03 0.31 linear 43.86 2.90 15.11 0.0001 quadratic -41.30 7.33 -5.63 0.0001 Parameter Estimate Error t value P Intercept 0.43 0.22 1.97 0.0495 linear 31.54 5.71 5.53 0.0001 quadratic 35.19 31.41 1.12 0.2628 cubic -102.98 41.12 -2.50 0.0124 Parameter Estimate Error t value P Intercept 0.28 0.24 1.17 0.2438 linear 45.19 9.7 4.66 0.0001 quadratic -106.54 87.35 -1.22 0.2228 cubic 334.91 255.19 1.31 0.1896 quartic -395.70 227.60 -1.74 0.0824 *The dependent and the independent variables are respectively water soluble P and phosphorus saturation ratio. 2. R-squared selection method for soluble P and phosphorous saturation ratio for the top 0-15 cm depth. Number in model r-squared Variables in the model 1 0.31 Linear 2 0.32 Linear quadratic 3 0.33 Linear quadratic cubic 1 0.26 Exponential 1 0.21 Logarithmic 1 0.02 Inverse 95

PAGE 110

96 3. Regression analysis of soil P storage capacity-Mehlich-1 and soil P storage capacity-single point isotherm *Parameter Estimate Error t value P Intercept 29.77 2.74 10.86 0.0001 linear 0.65 0.05 12.83 0.0001 Parameter Estimate Error t value P Intercept 30.54 2.61 11.72 0.0001 linear 1.01 0.09 11.17 0.0001 quadratic 0.0025 0.00053 4.71 0.0001 Parameter Estimate Error t value P Intercept 27.32 2.60 10.50 0.0001 linear 1.18 0.095 12.41 0.0001 quadratic 0.008 0.0014 5.83 0.0001 cubic 0.000022 0.000005 4.30 0.0001 Parameter Estimate Error t value P Intercept 25.91 2.76 9.40 0.0001 linear 1.10 0.10 10.28 0.0001 quadratic 0.00097 0.0018 5.45 0.0001 cubic 0.000051 0.00002 2.56 0.0113 quartic 0.000000008 0.000000006 1.51 0.1317 *The dependent and the independent variables are respectively soil P storage capacity-Mehlich-1 and soil P storage capacity-single point isotherm 4. R-squared selection method for soil P storage capacity-Mehlich-1 and soil P storage capacity-single point isotherm. Number in model r-squared Variables in the model 1 0.47 Linear 2 0.52 Linear quadratic 3 0.56 Linear quadratic cubic 1 0.1231 Exponential

PAGE 111

LIST OF REFERENCES Addy, K. L., A. J. Gold, P. M. Groffman, and P. A. Jacinthe. 1999. Groundwater nitrate removal in subsoil of forested and mowed riparian buffer zones. Journal of Environmental Quality. 28:962-970. Anbumozhi, V., Radhakrishnan J., and Yamaji, E. 2005. Impact of riparian buffer zones on water quality and associated management considerations. Ecological Engineering. 24:517-523. Andrews, W.J. 1994. Nitrate in ground water and spring water near four dairy farms in north Florida, 1990-93: U.S. Geological Survey water-resources investigations report 94-4162. 63 p. Allen, L.H., Jr. 1987. Dairy-siting criteria and other options for wastewater managementon high water-table soils. Soil Crop Science Society of Florida. Proceedings. 47:108-127. Allen, S.C. 2003. Nitrogen dynamics in a pecan (Carya illinoensis K.Koch)-Cotton(Gossypium hirsutum L. Alley cropping systems in the Southern United States. PhD Dissertation. University of Florida. Allen, S.C., S. Jose, P.K.R. Nair, B.J. Brecke, P. Nkedi-Kizza, and C.L. Ramsey. 2004. Safety-net role of tree roots: evidence from a pecan (Carya illinoensis K. Koch)-cotton (Gossypium hirsutum L.) alley cropping system in the southern United States. Forest Ecology and Management. 192: 395-407. Allen, S.C., S. Jose, P.K.R. Nair, B.J. Brecke, V.D. Nair, D.A. Graetz, and C.L. Ramsey. 2005. Nitrogen mineralization in a pecan (Carya illinoensis K. Koch)-cotton (Gossypium hirsutum L.) alley cropping system in the southern United States. Biology and Fertility of Soils. 41: 28-37. Arshad, M.A, B. Lowery, and B. Grossman. 1996. Methods for assessing soil quality. Chapter 7: Physical tests for monitoring soil quality. Soil Science Society of America. Madison, WI, USA. p. 123-155. Alavalapati, J.R.R., R.K. Shrestha, A. Stainback, and J.R. Matta. 2004. Agroforestry development: An environmental economic perspective. Agroforestry Systems. 61: 299-310. 97

PAGE 112

98 Baker, J.B. [n.d.]. Alternative silvicultural systems -south. In: Silvicultural challenges and opportunities in the 1990's: Proceedings of the National Silvicultural Workshop; 1989 July 10-13; Petersburg, AK. Washington, DC: U.S. Department of Agriculture, Forest Service, Timber Management: 51-60. Baker, J.B., and O.G. Langdon. 1990. Pinus taeda L. Loblolly pine. In: Burns, R.M. and B.H. Honkala. Silvics of North America. Volume 1. Conifers. Agric. Handb. 654.Washington, DC: U.S. Department of Agriculture, Forest Service. p. 497-512. Beauchemin, S., R.R. Simard, and D. Cluis. 1996. Phosphorus sorption-desorption kinetics of soils under contrasting land use. Journal of Environmental Quality 5:1317-1325. Beauchemin, S., R.R. Simard. 1999. Soil phosphorus saturation degree: Review of someindices and their suitability for phosphorus management in Canada. Canadian Journal of Soil Science. 79:615-625. Bharati L., Lee, K.H., Isenhart, T.M., and Schultz, R.C. 2002. Soil-water infiltrationunder crops, pasture, and established riparian buffer in Midwestern USA. Agroforestry Systems. 56:249-257. Blattel, C.R ., Williard, K.W.J Baer, S.G ., Zaczek, J.J 2005. Abatement of ground waterphosphate in giant cane and forest riparian buffers. Journal of the American Water Resources Association. 41(2):301-307. Bliss, J.C., S.K. Nepal, R.T. Brooks Jr., and M.D. Larsen. 1997. In the mainstream: Environmental attitudes of mid-south NIPF owners. Southern Journal of Applied Forestry. 21:37-42. Boman, B., T. Obreza. 2002. Fertigation nutrient sources and application considerations for citrus. http://edis.ifas.ufl.edu/pdffiles/CH/CH18500.pdf verified 14 November 2006. Brling, K. 2003. Phosphorus sorption, accumulation and leaching. Effect of long term inorganic fertilization of cultivated soils. Doctoral thesis. Swedish University of Agricultural Sciences. 39 p. Bonsi, C., E. Rhoden, A. Woldeghebriel, P. Mount, S. Solaiman, R. Noble, G. Garis, C. McMahon, H. Pearson, and B. Cash. 1991. Kudzu-goat interactions a pilot study. pp. 84-90 In: G. Solaiman and W.A. Hill (eds.) 1991. Using goats to manage forest vegetation: A regional inquiry. Workshop Proceedings. Tuskegee University Agricultural Experiment Station, Tuskegee, AL. Brady, N.C., and R.R. Weil. 2002. The nature and properties of soils. Prentice-Hall, Inc.Upper Saddle River, NJ, USA. 960 p.

PAGE 113

99 Bransby, D.I., S.E. Sladden, D.D. Kee, and N.E. West. 1996. Forage yield response of mimosa (Albizia julibrissin) to harvest frequency. pp. 64-65. Rangelands in a sustainable biosphere. Proceedings of the Fifth International Rangeland Congress. Vol. 1. Society of Range Management, Salt Lake City, UT. Breeuwsma, A., and S. Silva. 1992. Phosphorus fertilization and environmental effects in The Netherlands and Po Region (Italy). Report 57. DLO The Winand Staring Centre, Wageningen, The Netherlands. Breeuwsma, A., J.G.A. Reijerink, and O.F. Schoumans. 1995. Impact of manure on accumulation and leaching of phosphate in areas of intensive livestock farming. p. 239-251. In: K. Steele (ed.) Animal waste and the land-water interface. Lewis Publishers-CRC, New York, NY. Brender, E.V., R.P. Belanger, and B.F. Malac. 1981. Loblolly pine. In: Choices insilviculture for American forests. Washington, DC: Society of American Foresters. p. 37-45. Burgoa, B., A.B. Bottcher, R.S. Mansell, and L.A. Allen, Jr. 1991. Distribution of residual soil phosphorus along transacts for three dairies in Okeechobee County, Florida. Soil Crop Science Society of Florida Proceedings. 50:137-144. Burton, N. L., and A. D. Scarfe. 1991. Angora goats in Alabama woodlands. pp. 78-83 In: G. Solaiman and W.A. Hill (eds.) 1991. Using goats to manage forest vegetation: A regional inquiry. Workshop Proceedings. Tuskegee University Agricultural Experiment Station, Tuskegee, AL. Byrd, N. A., C. E. Lewis, and H.A. Pearson. 1984. Management of southern pine forests for cattle production. USDA Forest Service, Southern Region, General Report R8-GR4. 22p. Carpenter, S., Caraco, N.F., Correll, D.L., Howarth, R.W., Sharpley, A. N., and Smith, V. H. 1998. Nonpoint pollution of surface waters with phosphorus and nitrogen. Issues in Ecology # 3. http://www.esa.org/science/Issues/FileEnglish/issue3.pdf verified 14 November 2006. Cary, A. 1928. On management of turpentine forests. Weekly Naval Stores Review, reprints, Savannah, GA. Chambliss, C. G., and M.B. Adjei. 2006. Bahiagrass: http://edis.ifas.ufl.edu/AA184 ,verified 14 November 2006. Chen, M., L. Q. Ma. 2001. Taxonomic and geographic distribution of total phosphorus in Florida surface soils. Soil Science Society of America Journal 65:1539 (2001). Chrysostome, M. 2005. Assessing the vertical movement of phosphorus inmanure-impacted soils. Phd dissertation, University of Florida.

PAGE 114

100 Clason, T.R. 1995. Economic implications of silvopastures on southern pine plantations.Agrofororestry Systems. 29:227-238. Clason, T.R., and S.H. Sharrow. 2000. Silvopastoral practices. pp. 119-147. In: H.E. Garrett, W.J. Rietveld, and R.F. Fisher (eds.) North American Agroforestry: An Integrated Science and Practice. American Society of Agronomy, Madison WI. Codling, E.E., R.L. Chaney and C.L. Mulchi. 2002. Biomass yield and phosphorusavailability to wheat grown on high phosphorus soils amended with phosphateinactivating residues. I. Drinking water treatment residue. Communications in SoilScience and Plant Analysis. 33: 1039-1061. Collins, M.E. 2003. Key to Soil Orders in Florida. Fact sheet SL-43. Soil and Water Science Department, Florida Cooperative Extension Service, Institute of Food and Agricultural Sciences, University of Florida. First published: August 1985. Revised: October 1997. Reviewed: September 2003. Available: http://edis.ifas.ufl.edu/SS113 verified 14 November 2006. Collins, M.E. 2001. Soils in Florida. Research brief. Soil and Water Science, Institute of Food and Agricultural Sciences, University of Florida. http://soils.ifas.ufl.edu/department/briefs/sws01-9pdf verified 14 November 2006. Conner, R.C., and A.J. Hartsell. 2002. Forest Area and Conditions. Chapter 16, pp. 357-401. In: D.N. Ware and J.G. Greis (eds.) Southern forest resource assessment. USDA Forest Service GTR SRS-53. Conner, W.H., N.L. Hill, E.M. Whitehead, [and others]. 2001. Forested wetlands of the Southern United States: a bibliography. USDA Forest Service, Southern Research Station, General Technical Report SRS-043. Cordell, H.K., J.C. Bliss, C.Y. Johnson, and M. Fly. 1998. Voices from southern forests. Trans. 63rd No. American Wildlife and Natural Resources Conference. 332-347. Correll, D.L. 1998. The role of phosphorus in the eutrophication of receiving waters: A review. Journal Environmental Quality. 27: 261-266. CSTAF (Center for Subtropical Agroforestry). 2003. Research results show benefits of agroforestry. CSTAF Newsletter Volume 3, No. 2. University of Florida, IFAS. http://cstaf.ifas.ufl.edu/NLs2.03.htm verified 14 November 2006. Daily, G. (ed.). 1997. Nature's services: Societal dependence on natural ecosystems. Island Press, Washington, DC. Dalrymple, R.L. 1994. Nutrient recycling. Agricultural News and Views. Vol. 12, No. 4. p.3. Dangerfield, C.W., and R.L. Harwell. 1990. An analysis of a silvopastoral system for the marginal land in the southeast United States. Agroforestry Systems. 10:187-197.

PAGE 115

101 Das D.K ., O.P. Chaturvedi 2005. Structure and function of Populus deltoids agroforestry systems in eastern India: 2 Nutrient dynamics. Agroforestry Systems. 65(3):223-230. Duncan, W. H., and M.B. Duncan. 1988. Trees of the southeastern United States. The University of Georgia Press, Athens, GA. 322 p. Eghball, B., G.D. Bindford, and D.D. Baltensperger. 1996. Phosphorus movement and adsorption in a soil receiving long-term manure and fertilizer applications. Journal of Environmental Quality. 25:1339-1343. ESA (Ecological Society of America). 1998. Issues in Ecology: Nonpoint Pollution of surface waters with phosphorus and nitrogen. Washington, DC. 14 pp. http://www.esa.org/science/Issues/FileEnglish/issue3.pdf verified 14 November 2006. Ewel, J.J. 1999. Natural systems as models for the design of sustainable systems of land use. Agroforestry. Systems. 45(1-3):1-21. FDEP (Florida Department of Environmental Protection). 2003. Surface Water. http://www.dep.state.fl.us/water/surfacewater/index.htm verified November 2006. Felker, P., C.T. Sorensson, D. Ueckert, P. Jacoby, E. Singer, and R. Ohm. 1998. Growth, cold-hardiness, protein content, and digestibility of 70 Leucaena seedlots on three sites in Texas, USA. Agroforestry Systems. 42:159-179. Fitchen, J.M. 1991. Endangered spaces, enduring places: Change, identity, and survival in rural America. Westview Press, Boulder, CO. FDEM (Florida Division of Emergency Management. 2006. National Flood Insurance Program. http://www.floridadisater.org/brm/nfip.htm verified 14 November 2006. Florida Sheriffs Youth Ranch. 2003. Forest Stewardship Management Plan. 12 p.Unpublished. Friend, A.L., Roberts, S.D., Schoenholtz, S.H., Mobley, J.A., Gerard, P.D., 2006. Poultry litter application to loblolly pine forests: Growth and nutrient containment. J. Environ. Qual. 35:837-848. Frossard, E., Condron, L.M., Oberson, A., Sinaj, S. & Fardeau, J.C. 2000. Processes governing phosphorus availability in temperate soils. Journal of Environmental Quality. 29:15-23.

PAGE 116

102 Garrett, H.E, W.B. Kurtz, L.E. Buck, M.A. Gold, L.H. Hardesty, J.P. Lassoie, H.A. Pearson, and J.P. Slusher. 1994. Agroforestry: An integrated land-use management system for production and farmland conservation. Prepared for USDA SCS 68-3A75-3-134, Resource Conservation Act Appraisal. Association for Temperate Agroforestry, Columbia, MO. Garrett, H.E., W.J. Rietveld, and R.F. Fisher (eds.). 2000. North American Agroforestry: An Integrated Science and Practice. American Society of Agronomy, Madison, WI. Gerritse, R.G., and Zugec, I. 1977. The phosphorus cycle in pig slurry measuredfrom 32 PO 4 distribution ratio. Journal of Agricultural Science Science (Cambridge). 88:101. Gilman F.E., and D.G. Watson. 1994. Pinus elliottii: Slash Pine. USDA Forest Service. Fact Sheet ST-463. 4 p. Golterman, H.L. 1988. Reflections on fractionation and bioavailability of sediment bond phosphate. Archive Hydrobiology. 30:1-4. Grado, S.C., C.H. Hovermale, D.G. St. Louis. 2001. A financial analysis of a silvopasture system in southern Mississippi. Agroforestry. Systems. 53:313-322. Harris, W.G ., R.D. Rhue, G. Kidder R.B. Brown, R. Littell 1996. Phosphorus retention as related to morphology of sandy coastal plain soil materials. Soil Science Society of America Journal of Science. 60 (5):1513-1521. Hart, R.H., R.H. Hughes, C.E. Lewis, and W.G. Monson. 1970. Effect of nitrogen and shading on yield and quality of grasses grown under young slash pines. Agronomy Journal. 62:285-287. Haygarth, P.M., L. Hepworth, and S.C. Jarvis. 1998. Forms of phosphorus transfer in hydrological pathways from soil under grazed grassland. European Journal of Soil Science. 49: 65-72. Hatzell, H.H. 1995. Effects of waste-disposal practices on ground-water quality at five poultry (broiler) farms in north-central Florida, 1992-1993: U.S. Geological Survey water-resources investigations report 954064. 35 p. Heckrath, G., Brookes, P.C., Poulton, P.R. & Goulding, K.W.T. 1995. Phosphorus leaching from soils containing different phosphorus concentrations in the Broadbalk experiment. Journal of Environmental Quality 24: 904-910. Hefting, M.M., Clement, J.-C., Bienkowski, P., Dowrick, D., Guenat, C., Butturini, A.,Topa, S., Pinay, G., and Verhoeven, J.T.A. 2005. The role of vegetation and litter in the nitrogen dynamics of riparian buffer zones in Europe. Ecological Engineering. 24: 465-482.

PAGE 117

103 Hedley, M.J., Stewart, J.W.B., and Chauhan, B.S. 1982. Changes in inorganic andorganic phosphorus fractions induced by cultivation practices and laboratory incubations. Soil Science Society of America Journal. 46:970-976. Hesketh, N., and Brookes, P.C. 2000. Development of an indicator for risk of phosphorus leaching. Journal Environmental Quality. 29:105-110. Higgins, M.J., Rock, C.A. Bouchard, R and Wengrezynek, B. 1993. Controlling agricultural runoff by use of constructed wetlands. In: G.A. Moshiri, Editor, Constructed Wetlands for Water Quality Improvement, Lewis Publishers, Boca Raton, FL (1993). pp. 359. Hooda, P.S., M. Moynagh, I.F. Svoboda, A.C. Edwards, H.A. Anderson, and G. Sym. 1999. Phosphorus loss in drainflow from intensively managed grassland soils. Journal of Environmental Quality. 28:1235-1242. Hooda, P.S., A.R. Rendell, A.C. Edwards, P.J.A Withers, M.N.Aitken, and V.W. Truesdale. 2000. Relating soil phosphorus indices to potential phosphorus release to water. Journal Environmental Quality. 29:1166-1171. Hyde, A.G., G. W. Hurt, and C.A. Wettstein. 1991. Soil survey of Sarasota County, Florida. United States Departement of Agriculture in cooperation with University of Florida Institute of Food and Agricultural Sciences and Agricultural Experiment Stations, Soil Science Department; and the Florida Department of Agriculture and Consumer Services. 147 p. Horst, W.J. 1995. Efficiency of soil-nutrient use in intercropping systems. In: Sinquet, A. and Cruz, P. (eds) Ecophysiology of tropical intercropping. Institut National de la Recherche Agronomique, Paris. pp. 197-211. ICRAF ( World Agroforestry Centre International). 2000. Trees of Change: Corporate Report 2000. International Centre for Research in Agroforestry, Nairobi, Kenya. Indiati, R ., Trujillo, I ., Gutierrez, F 2002. Characterization of P adsorption properties of some agricultural volcanic soils from Canary Islands, Spain. Tropical Agriculture. 79(4):254-259. Jose, S., Gillespie, A.R., and Pallardy, S.G. 2004. Interspecific interactions in temperate agroforestry. Agroforestry Systems. 61:237-255. Jordan, C.F. 1998a. Alley Cropping. pp. 14.1-14.5 In P.G. Cannon (ed.) Exploring agroforestry opportunities in the South. Southern Agroforestry Conference sponsored by the Natural Resources Conservation Service, the US Forest Service, and Alabama A&M University. Oct. 1998. Huntsville, AL.

PAGE 118

104 Johnson, R.H., and P.W. Bush. 2002. Summary of the hydrology of the Floridan aquifer system in Florida and in Parts of Georgia, South Carolina, and Alabama. Professional Paper 1403-A. Center for Coastal Geology, U.S. Geological Survey, U.S. Department of the Interior. http://sofia.usgs.gov/publications/papers/pp1403a/abstract.html verified 14 November 2006. Kadlec, R.H. 1985. Aging phenomena in wastewater wetlands. In: P.G. Godfrey,Editor, Ecological Considerations in Wetland Treatment of Municipal Wastewaters, Von Nostrand Reinhold, New York (1985). pp. 239-247. Kang, B.T. 1997.Alley cropping Soil productivity and nutrient recycling. Forest Ecology and Management. 91(1): 75-82. Katz, B.G., Hornsby H.D., Bohlke J.K. and Mokray M.F. 1999. Sources and chronology of nitrate contamination in springs, Suwannee River Basin, Florida: U.S. Geological Survey water-resources investigations report 99-4252. 54 p. Keeney, D.R., and D.W. Nelson. 1982. Chapter 33. Nitrogen-Inorganic Forms. In Methods of soil analysis. Part 2. Chemical and microbial properties. Second Edition. American Society of Agronomy, Inc. Wisconsin, USA. p 643-699. Kidder, G., Chambliss, C.G., and Mylavarapu, R. 2002. UF/IFAS Standardized Fertilization Recommendations for Agronomic Crops. http://edis.ifas.ufl.edu/pdffiles/SS/SS16300.pdf verified 14 November 2006) Koopmans, G.F., W.J. Chardon, J.Dolfing, O. O.enema, P. van der Meer, and W.H. van Riemsdiijk. 2003. Wet chemical and phosphorus-31 nuclear magnetic resonance analysis of phosphorus speciation in a sandy soil receiving long-term fertilizer or animal manure applications. Journal of Environmental Quality. 32:287-295. Lapp, F.M., J. Collins, and P. Rosset, with L. Esparza. 1998. World Hunger: 12 Myths. Grove Press/Earthscan. Lassoie, J.P., P. Huxley and L.E. Buck. 1994. Updating our ideas about agroforestry education and training. Agroforestry Systems. 28:5-19. Lassoie, J.P. and L.E. Buck. 2000. Development of agroforestry as an integrated land use management strategy. pp. 1-29 In: H.E. Garrett, W.J. Rietveld, and R.F. Fisher (eds.) North American Agroforestry: An Integrated Science and Practice. American Society of Agronomy, Madison WI. Lavelle, P., and A.V. Spain. 2001. Soil Ecology. Kluwer Academic Publishers, Dordrecht, The Netherlands. 654 p. Lehmann, J., and W. Zech.1998. Fine root turnover of hedgerow intercropping in Northern Kenya. Plant Soil. 198:19-31.

PAGE 119

105 Lehmann, J., Weigl, D., Droppelmann, K., Huwe, B. and Zech, W. 1999. Nutrient cycling in an agroforestry system with runoff irrigation in Northern Kenya. Agroforestry Systems. 43: 49-70. Lehmann, J., and G. Schroth. 2003. Nutrient leaching. Chapter 7. In: Trees, Crops and Soil Fertility: Concepts and Research Methods. CABI Publishing. 448 p. Lee, K. H ., Disinherit, T. M ., Schultz, R. C 2003. Sediment and nutrient removal in an established multi-species riparian buffer. Journal of Soil and Water Conservation. 58 (1): 1-8. Leinweber, P., R. Meissner, K.U. Eckhardt, and J. Seeger. 1999. Management effects on forms of phosphorus in soil and leaching losses. European Journal of Soil Science 50:413-424. Lemunyon, J.I., and T.C. Daniels. 1997. Phosphorus management for water qualityprotection: a national effort. In: J.T. Sims (ed.). Soil testing for Phosphorus:Enviromental Uses and Implications. SERA-IEG 17 USDA-CRES Regional Committee: Minimizing Agricultural Phosphorus losses for protection of Water Resource. Leakey, R.R.B. 1996. Definition of agroforestry revisited. Agroforestry Today. 8 (1): 5-6. Leakey, R.R.B., and A.J. Simons. 1997. The domestication and commercialization of indigenous trees in agroforestry for the alleviation of poverty. Agroforestry Systems. 38:65-176. Leakey, R.R.B. 1998. The use of biodiversity and implications for agroforestry. pp. 43-58 In D.E. Leihner, T.A. Mitschein, and P. Lang (eds) A Third Millennium for Humanity? The Search for Paths of Sustainable Development. Frankfurt, Germany. Lewis, C.E., G.W. Burton, W.G. Monson, and W.C. McCormick. 1983. Integrations of pines, pastures and cattle in south Georgia, USA. Agrofor. Syst. 1:277-297. Lewis, C.E., G.W. Tanner, and W.S. Terry. 1985. Double vs. single-row pine plantations for wood and forage production. Southern Journal of Applied Forestry. 9:55-61. Lewis, C.E., and H.A. Pearson. 1987. Agroforestry using tame pastures under planted pines in the southeastern United States. pp. 195-212 In H.L. Gholz (ed.) Agroforestry: Realities, Possibilities and Potentials. Nijhoff Junk, Dordrecht, The Netherlands. Long, A.J. and P.K.R. Nair. 1999. Trees outside forests: Agro-, community, and urban forestry. New Forests. 17:145-174. LSU (Lousiana State University). 1998. Bahiagrass Production and Management. http://www.lsuagcenter.com/NR/rdonlyres/82A096A6-7F6F-4306-853B-F49A1D98AFEF/2966/pub26992.pdf verified 14 November 2006.

PAGE 120

106 Lohrey, R .E., and S.V. Kossuth. 1990. Pinus elliottii Engelm. Slash pine. In: Burns, R.M.and B.H. Honkala. Silvics of North America. Volume 1. Conifers. Agric. Handb. 654. Washington, DC: U.S. Department of Agriculture Forest Service. p. 338-347. Lu, P ., and G.A. O'Connor 2004. Biosolids effects on phosphorus retention and release in some sandy Florida soils. Journal of Environmental Quality. 30(3):1059-1063. Lundgren, G.K., Conner J.R. and Pearson H.A. 1983. An economic analysis of forest grazing on four timber management situation. Southern Journal of Applied Forestry. 7:119-124. Maguire, R.O., R.H. Foy, and J.S. Bailey. 1998. Factors onfluencing the relationshipbetween Olsen and Calcium chloride extractable phosphorus for some Irish soils. P. 98-99. In: R.H. Foy and R. Dils (ed.) Practical and innovative measures for the control of agricultural phosphorus losses to water. OECD Sponsored Work-shop, Belfast. June 1998. Dep. Of Agric. For the Northern Ireland, Belfast. Maguire, R.O., and J.T. Sims. 2002. Soil testing to predict phosphorus leaching. Journal of Environmental Quality. 31:1601-1609. Makris K.C., W.G. Harris, G.A. OConnor, T.A. Obreza and H.A. Elliott. 2005.Physicochemical properties related to long-term phosphorus retention bydrinking-water treatment residuals. Environmental Science and Technology. 39:4280-4289. Mansell, R.S., S.A. Bloom, and B. Burgoa. 1991. Phosphorus transport with water flowin acid, sandy soils. p. 271-314. In B. Jacob an M.Y. Corapcioglu (ed.) Transport processes in porous media. Kluwer Academic Publ.. Dordrecht, The Netherlands. McCulley, R.D. 1950. Management of natural slash pine stands in the Flatwoods of south Florida and north Florida. Circular No. 845. Washington, DC: U.S. Department of Agriculture. 57 p. McCune, B. 1988. Ecological diversity in North American pines. American Journal of Botany. 75 (3):353-368. McGraw, R.L, N.E. Navarrete-Tindall, and J.W. Van Sambeek. 2001. Effects of shade on growth and nodulation of six native legumes with potential for use in agroforestry. p. 50 In Program abstracts, Seventh Biennial Conference on Agroforestry in North America. August 13-15, 2001, Regina, Sask., Canada. McGraw, R.L, N.E. Navarrete-Tindall, and J.W. Van Sambeek. 2001. Effects of shade on growth and nodulation of six native legumes with potential for use in agroforestry. p. 50 In Program abstracts, Seventh Biennial Conference on Agroforestry in North America. August 13-15, 2001, Regina, Sask., Canada.

PAGE 121

107 McDowell, R.W., and A.N. Sharpley. 2001. Phosphorus losses in subsurface flow beforeand after manure application to intensively farmed land. Science of the Total Environment. 278:113-25. McKergow, L.A., I. P. Prosser, R.B. Grayson, and D. Heiner. 2004. Performance of grass and rainforest riparian buffers in the wet tropics, Far North Queensland. 2. Water quality. Australian Journal of Soil Research. 42(4): 485-498. Mehlich, A. 1953. Determination of P, Ca, Mg, K, Na and NH 4. Soil Testing Div. Publ. 1-53. North Carolina Dep. Agric., Raleigh. Menezes, R.S.C ., I.H. Salcedo E.T. Elliott 2002. Microclimate and nutrient dynamics in a silvopastoral system of semiarid northeastern Brazil. Agroforestry Systems 56 (1): 27-38. Merwin, M. 1997. The Status, Opportunities, and Needs for Agroforestry in the United States. Association for Temperate Agroforestry (AFTA), University of Missouri, Columbia, MO. Mueller, D.K., and D.R. Helsel. 1996. Nutrients in the Nations waters-Too much of a good thing? U.S. Geological Survey Circular 1136. 26 p. Moore, P.A., B.C. Joern,, and T.L. Provin. 1998. Chapter 3: Improvements needed in environmental Soil testing for phosphorus. In: Sims, J. (ed.), Soil testing for phosphorus: Environmental uses and implications. Southern Coorperative Series, Bulletin No.389. USDA-CSREE Regional Committee. 43 p. Moreno, F., J.A. Cayuela, J.E. Fernndez, E. Fernndez-Boy, J.M. Murillo, and F. Cabrera. 1996. Water balance and nitrate leaching in an irrigated maize crop in SW Spain. Agricultural Water Management. 32:71-83. Mozafarri, M. and J.T. Sims. 1994. Phosphorus availability and Sorption in an Atlantic coastal plain watershed dominated by animal-based agriculture. Soil Science. 157:97-07. Murphy, J., J.P. Riley. 1962. A modified single solution method for the determination of phosphate in natural water. Analytica Chimica Acta. 27: 31-36. Nair, P.K.R. 1993. An Introduction to Agroforestry. Kluwer Acad. Publ., Dordrecht, Netherlands. Nair, P.K.R. 2001.Agroforestry: Integrating trees in crop-and livestock-production systems. In Tolba, M.K. (ed.), Our Fragile World (OFW): Challenges and Opportunities for Sustainable Development, Volumes I and II. Eolss Publishers, Oxford, UK. 2263 pp.

PAGE 122

108 Nair, P. K. R., M.E. Bannister, V.D Nair, J.R.R. Alavalapati, E. A. Ellis, S. Jose, and A.J.Long. 2005. Silvopasture in southeastern United States: More than just a new name for an old practice. In: Mosquera-Losada, M.R., McAdam, J., and Rigueiro-Rodrguez, A. (eds), Silvopastoralism and sustainable land management. CABI, Wallingford, U. K. Nair, P.K.R., M.E. Bannister, V.D. Nair, J.R.R. Alavalapati, E.A. Ellis, S. Jose, and A.J.Long. 2004. Silvopasture in southeastern United States: More than just a new name for an old practice. Silvopastoralism and Sustainable Management International Congress, Lugo, Spain, April 18-24, 2004. CAB International. Nair, V.D., and D.A. Graetz. 2002. Phosphorus saturation in Spodosols impacted by manure. Journal of Environmental Quality. 31:1279-1285. Nair, V.D., and D.A. Graetz. 2004. Agroforestry as an approach to minimizing nutrient loss from heavily fertilized soils: The Florida experience. In: Nair P.K.R, Rao M.R. and Buck L.E. (eds), New Vistas in Agroforestry: A compendium for the 1st world congress of agroforestry, Kluwer, Dordrecht, The Netherlands, 2004. Nair, V.D., D.A. Graetz., and D.O. Dooley. 2003. Phosphorus release characteristics of manure and manure-impacted soils. Food, Agriculture and Environment. 1(2): 217-223. Nair, V.D., D.A. Graetz, and K.R. Reddy. 1998. Phosphorus retention capacity of the spodic horizon under varying environmental conditions. Journal of Environmrntal Quality. 28:1308-1313. Nair, V.D., and R.S. Kalmbacher. 2005. Silvopasture as an approach to reducing nutrient load of surface water from farms. In: Mosquera Losada, M.R. McAdam, J., Riguero-Rodriguez, A. (ed.) 2005. p 272-274. Silvopastoralism and sustainable management. CAB international. Nair, V.D. and W.G. Harris. 2004. A capacity factor as an alternative to soil test phosphorus in phosphorus risk assessment. New Zealand Journal of Agricultural Research. 47:491-497. Nair, D., K. M. Portier, D. A. Graetz and M. L. Walker. 2004. An Environmental Threshold for Degree of Phosphorus Saturation in Sandy Soils. Journal of Environmental Quality. 33:107-113. NASS (National Agricultural Statistics Service). 2001. Derived from data at http://goveinfo.library.orst.edu/document/ag97/agcont.html. Nowak, J. and A. Blount. 2002. Silvopasture More than Cattle Grazing in Pine Plantations. Forest Landowner. 61 (2):10-13.

PAGE 123

109 Nowak, J., A. Long. 2003. Establishment of Integrated Timber, Forage and Livestock Silvopasture Systems in the Southeast a Review. Proceedings of Sod Based Cropping Systems Conference, North Florida Research and Education Center-Quincy, University of Florida, Feb. 20-21. NPG (Negative Population Growth). 2006. State population Facts. http://www.npg.org/states/fl.htm verified 14 November 2006. NRCS (Natural Resources Conservation Service). 1994. The phosphorus index: A phosphorus Assessment Tool. Technical Note. http://www.nrcs.usda.gov/technical/ECS/nutrient/pindex.html verified 14 November 2006. McGowan, C.H., L. Carter, S. Pancholy, and B.R. Phills. 1999. Statewide Goat Program: Program Plan. Florida Agricultural and Mechanical University, College of Engineering Sciences, Technology and Agriculture, Tallahassee, FL. McGraw, R.L, N.E. Navarrete-Tindall, and J.W. Van Sambeek. 2001. Effects of shade on growth and nodulation of six native legumes with potential for use in agroforestry. p. 50 In Program Abstracts, Seventh Biennial Conference on Agroforestry in North America. August 13-15, 2001, Regina, Sask., Canada. McKergow, L. A ., I. P. Prosser R. B. Grayson, D. Heiner 2004. Performance of grassand rainforest riparian buffers in the wet tropics, Far North Queensland. 2. Water quality. Australian Journal of Soil Research. 42 (4): 485-498. Parry, R. 1998. Agricultural phosphorus and water quality: a U.S. Environmental Protection Agency perspective. Journal of Environmental Quality. 27:258-260. Pearson, H.A. 1975. Exotic grass yields under southern pines. USDA Forest Service Southern For. Exper. Stat. Res. Note SO-201. New Orleans, LA. Pearson, H.A. 1997. Silvopastures: Forest grazing and agroforestry in the southern Coastal Plain. pp. 25-42 In: D. Henderson (ed.) Proc. Mid-South Conf. On Agroforestry Practices and Policies. West Memphis, AR. Aug 1991. Winrock International Inst. For Agric. Development. Morrilton, AR. Pearson, H.A., and A. Martin Jr. 1991. Goats for vegetation management on the Ouachita National Forest. pp. 59-73 In G. Solaiman and W.A. Hill (eds.) 1991. Using goats to manage forest vegetation: A regional inquiry. Workshop Proceedings. Tuskegee University Agricultural Experiment Station, Tuskegee, AL. Pearson, H.A., V.C. Baldwin, and J.P. Barnett. 1990. Cattle grazing and pine survival and growth in sub-terranean clover pasture. Agroforestry Systems. 10:161-168. Pote, D.H., T.C. Daniel, A.N. Sharpley, P.A. Jr. Moore, D.R. Edwards, D.J. Nichols. 1996. Relating extractable soil phosphorus to phosphorus losses in runoff. Soil Science Society of America Journal. 60:855-859.

PAGE 124

110 Ramnarine, R. 2003. Predicting phosphatic soil distribution in Alachua County, Florida. Masters degree thesis. University of Florida. Readle, E.L. 1979. Soil survey of Osceola County area, Florida. United States Departement of Agriculture in cooperation with University of Florida Institute of Food and Agricultural Sciences and Agricultural Experiment Stations, Soil Science Department. 151 p. Reid, W.R. 1991. Low input management systems for native pecans. pp. 140-158 In H.E. Garrett (ed.) Proc. Second Conference on Agroforestry in North America, August 1991, Springfield, MO. Robinson, J., and T. Clason. 1997. From a pasture to a silvopasture system. Agroforestry Notes AF Note 22. USDA National Agroforestry Center, Lincoln, NE. Rockwood, D.L., C.V. Naidu, D.R. Carter, M. Rahmani, T.A. Spriggs, C. Lin, G.R. Alker J.G. Isebrands, and S.A. Segrest. 2004. Short-rotation woody crops and phytoremediation: Opportunities for agroforestry? Agroforestry Systems. 61:51-63. Rowe, E.C., K. Hairiah, K.E. Giller, M. van Noordwijk, and G. Cadisch. 1999. Testing the safety-net role of hedgerow tree roots by 15 N placement at different soil depths. Agroforestry Systems. 43:81-93. Rowe, E.C ., M. Van Noordwijk, D. Suprayogo, G. Cadisch 2005. Nitrogen use efficiency of monoculture and hedgerow intercropping in the humid tropics. Plant and Soil. 268 (1):61-74. Sampson, R.N., and D. Hair. 1990. Natural resources for the 21st Century. Island Press, Washington, DC. Sartain, J.B. 2001. Soil testing and interpretation for Florida turfgrasses, http://edis.ifas.ufl.edu/pdffiles/SS/SS31700.pdf verified 14 November 2006 Sharpley, A.N. 1985. Depth of surface soil-runoff interaction as affected by rainfall, soilslope and management. Soil Science Society of America Journal. 49:1527-1534. Sharpley, A.N. 1993. Assessing phosphorus bioavailability in agricultural soils and runoff. Nutrient Cycling in Agroecosystems. 36:259-272. Sharpley, A.N., and B. Moyer. 2000. Phosphorus forms in manure and compost and their release during simulated rainfall. Journal of Environmental Quality. 29:1462-1469. Sharpley, A.N., and D. Beegle. 1999. Managing phosphorus and the environment. Cooperative Extension Service, Pennsylvania State University, University Park.

PAGE 125

111 Sharpley, A.N., and H. Tunney. 2000. Phosphorus research strategies to meet to meet agricultural and environmental challenges of the 21 st century. Journal of Environmental Quality. 29:171-181. Sharpley, A.N., R.W. McDowell and P.J.A. Kleinman. 2004. Amounts, Forms, and Solubility of Phosphorus in Soils Receiving Manure. Soil Science Society ofAmerica Journal. 68: 2048-2057. Sharpley, A.N., S.C. Chapra, R. Wedepohl, J.T. Sims, T.C. Daniel, and K.R. Reddy. 1994. Managing agricultural phosphorus for protection of surface waters: Issues and options. Journal of Environmental Quality. 23: 437-451. Sharpley, A.N., T. C. Daniel, J.T. Sims, J. Lemunyon, R. Stevens, and R. Parry. 1999. Agricultural phosphorus and eutrophication. USDA-ARS-149. 37 p. Sharpley, A.N., T.C. Daniel, J.T. Sims, and D.H. Pote. 1996. Determining environmentally sound soil phosphorus levels. Journal of Soil and Water Conservation. 51:160-166. Savory, A. 1988. Holistic resource management. Island Press, Washington, DC. Schoonover, J.E ., K.W. J. Williard, J. J. Zaczek, J. C. Mangun A.D. Carver 2005. Nutrient attenuation in agricultural surface runoff by riparian buffer zones in southern Illinois, USA. Agroforestry Systems. 64(2):169-180. Schoumans, O.F., and P. Groenendijk. 2000. Modeling soil phosphorus levels and phosphorus leaching from agricultural land in the Netherlands. Journal of Environmental Quality. 29:111-116. Schultz, R.C., J.P. Colletti, T.M. Isenhart, W.W. Marquez, and C.J. Ball. 2000. Riparian forest buffer practices. pp 189-281. In: Garrett H.E., Rietveld W.J. and Fisher R.F. (eds) North American Agroforestry: An Integrated Science and Practice. American Society of Agronomy, Inc. Madison, Wisconsin, USA. Schultz, R.C, T.M. Isenhart, W.W. Simpkins, and J.P. Colletti. 2004. Riparian Forest Buffers in Agroecosystems Structure, Function and Management Lessons Learned from the Bear Creek Watershed Project in Central Iowa, USA. Agroforestry System. 61:35-50. Seyfried, M.S., and P.S.C. Rao. 1991. Nutrient leaching loss from two contrasting. cropping systems in the humid tropics. Trop. Agric. (Trinidad). 68:9-18. Sinclair, F. 1999. A general classification of agroforestry practice. Agroforestry Systems. 46:161-180.

PAGE 126

112 Sims, T.J. 1998. Soil testing for Phosphorus: Environmental uses and implications. Minimizing agricultural phosphorus losses for protection of the water resource. Southern Cooperative Series, Bulletin No 389. USDA-CSEES Regional Committee. 43 p. Sims, T.J., R.O. Maguire, A.B. Leytem, K.L. Gartley, and M.C. Paulter. 2002. Evaluation of Mehlich 3 as an agri-environmental soil phosphorus test for the Mid-Antlantic United States of America. Soil Science Society of America Journal. 66:2016-2032. Smeck, N.E. 1985. Phosphorus dynamics in soils and landscapes. Geoderma. 36: 185-199. Smith, J.L., and J.W. Doran. 1996. Methods for assessing soil quality. Chapter 10: Measurement and Use of pH and electrical conductivity for soil quality Analysis. Soil Science Society of America. Madison, WI, USA. p. 169-201. Soil Survey Staff. 1996. Keys to soil taxonomy. U.S. Gov. Print. Office,Washington DC. Solaiman, G., and W.A. Hill. (eds.) 1991. Using goats to manage forest vegetation: A regional inquiry. Workshop Proceedings. Tuskegee University Agricultural Experiment Station, Tuskegee, AL. Stainback, G.A. and J.R.R. Alavalapati. 2004. Restoring longleaf pine through silvopasture practices: An economic analysis. Forest Policy and Economics. 6(3-4 ): 371-378. Stainback, G.A., J.R.R. Alavalapati, R.K. Shrestha, S. Larkin, G. Wong. 2004. Improving environmental quality in south Florida through silvopasture: an economic approach. Journal of Agricultural and Applied Economics. 36: 481-489. Summers, R. 2002. The environmental impact of nitrogen and phosphorus fertilizers in high rainfall areas. Department of Agriculture. Government of Western Australia. Farmnote No. 68. 2 p. Talsma, T, R.S. Mansell, and P.M. Hallam. 1979. Evaluation of porous cup soil water extractors: physical factors. Australian Journal of Soil Research. 17: 417-422. Tanner, C.C., J. S. Clayton, and M. P. Upsdell. 1995. Effect of loading rate and planting on treatment of dairy farm wastewater in constructed wetlands: Removal of oxygen demand, suspended solids and faecal coliforms. Water Resources. 29:17-26. Tapia-Coral, S.C ., Luizao FJ Wandelli E Fernandes E.C.M 2005. Carbon and nutrient stocks in the litter layer of agroforestry systems in central Amazonia, Brazil Agroforestry Systems. 65 (1): 33-42 2005. Teasley, R.J., J.C. Bergstrom, H.K. Cordell, S.J. Zarnoch, and P. Gentle. 1997. The use of private lands in the U.S. for outdoor recreation: Results of a nationwide survey. UGA Faculty Series 97-19. University of Georgia, Athens, GA. 102 p.

PAGE 127

113 Tiessen, H. 1995. Phosphorus in the global environment: Transfers, Cycles and Management. John Wiley & Sons Ltd, West Sussex, England. 462 p. Thomas, P.B., E. Cummings, and W.H. Wittstruck. 1982. Soil survey of Alachua County, Florida. United States Department of Agriculture, Soil Conservation Service. 117 p. Tunney, H., A. Breeuwsma, P.J.A. Withers, and P.A.I. Ehlert. 1997. Phosphorus fertilizer strategies: present and future. In: Tunney, H., O.T. Carton, P.C. Brookes, and A.E. Jhonston (eds), Phosphorus Loss from Soil to Water. CAB International, Wallingford, UK. pp. 117-203. Turner, B.L., N. Mahieu, and L.M. Condron. 2003. The phosphorus composition of temperate pasture soils determined by NaOH-EDTA extraction and solution 31P NMR spectroscopy. Org. Geochem. 34:1199-1210. Udawatta, R.P., J. J. Krstansky, G. S. Henderson, and H. E. Garrett. 2002. Agroforestry Practices, Runoff, and Nutrient Loss: A Paired Watershed Comparison. Journal of Environmental Quality. 31:1214-1225. UF/IFAS (University of Florida/Institute of Food and Agricultural Sciences). 2006. UF/IFAS Nutrient Management Series: Computational Tools for Field Implementation of the Florida Phosphorus Index: http://edis.ifas.ufl.edu/TOPIC_SERIES_Florida_Phosphorous_Index verified: 14 November 2006. USEPA (U.S. Environmental Protection Agency). 2006. List of drinking water contaminants and maximum contaminants levels: Inorganic chemicals. http://www.epa.gov/safewater/mcl.html verified 14 November 2006. USEPA (U.S. Environmental Protection Agency). 1993. Methods for the determination of inorganic substances in environmental samples. Cincinnati, Ohio, USA. USDA/NRCS (United States department of Agriculture/ National Resources Conservation Service). Official soil series description. Soil Survey Division. http://ortho.ftw.nrcs.usda.gov/cgi-bin/osd/osdname.cgi verified 14 November 2006. USDA/NRCS (United States department of Agriculture/ National Resources Conservation Service). 1999. Managing conservation tillage. Soil Quality-Agronomy. Technical note No.9. 8 p. USDA/ERS (U.S. Department of Agriculture / Economic Research Service). 2006a. http://www.ers.usda.gov/Data/FertilizerUse/ verified 14 November 2006. USDA/ERS (U.S. Department of Agriculture/Economic Research Service). 2006b. http://www.ers.usda.gov/statefacts/ verified 14 November 2006.

PAGE 128

114 USEPA (U.S. Environmental Protection Agency). 1996. Environmental indicators of water quality in the United States. EPA 841-7-96-002. USEPA Office of Water (4503F), U.S. Govt. Printing Office, Washington, D.C. USGS (U.S. Geological Survey). 2006. Land cover trends: Southern coastal plain. http://eros.usgs.gov/LT/regions/eco75.html verified 14 November 2006. van der Zee, S.E.A.T.M., L.G.J. Fokkink, and W.H. van Riems-dijk. 1987: A new technique for assessment of reversibly absorbed phosphate. Soil Science Society of America Journal. 51:599-604. van Noordwijk, M., G. Lawson, A. Soumar, J.J.R. Groot, and K. Hairiah. 1996. Root distribution of trees and crops: competition and/or complementarity. In Ong, C.K. and P. Huxley (eds), Tree-crop interactions: a physiological approach. pp. 319-364. CAB International, Wallingford, UK. van Noordwijk, M., G., Cadish, Ong, C.K. (eds). 2004. Below-ground Interactions in Tropical Agroecosystems: Concepts and Models with Multiple Plant Components. CABI Publishing, Wallingford, Oxfordshire, UK. 464 p. Vicente-Chandler, J. 1966. The role of fertilizers in hot humid tropical pastures. In Popenoe, H.L., and Blue, W.G. (eds). Soil Crop Science Society of Florida Proceedings Symposium: Fertilizers and use in the tropics. 26:328-360. Vitousek, P.M., H.A. Mooney, J. Lubchenco, and J.M. Melillo. 1997. Human domination of earth's ecosystems. Science. 277:494-499. Von Felde, K., and S. Kunst. 1997. Nitrogen and COD removal in vertical flow systems, Water Science and Technology. 35:79-85. Wanvestraut R., S. Jose., P.K.R. Nair. And B.J. Brecke. 2004. Competition for water in a pecancotton alley cropping system. Agroforestry Systems. 60: 167-179. Ware, D.N. 2002. Land Use. Chapter 6, pp.153-173 In: D.N. Ware and J.G. Greis (eds.). Southern Forest Resource Assessment. USDA Forest Service GTR SRS-53. Whitehead, D.C. Nutrient elements in grassland: Soil-plant-animal relationship. 2000. CABI Publishing, Wallingford, UK. 369 p. Williams, P.A., A.M. Gordon, H.E. Garrett, and L. Buck. 1997. Agroforestry in North America and its role in farming systems. pp. 9-84 In: A.M. Gordon and S.M. Newman (eds.) Temperate Agroforestry Systems. CAB International, Oxon, UK. World climate. 2005. http://www.worldclimate.com/ verified 14 November 2006.

PAGE 129

115 Workman, S.W., and S.C. Allen. 2004. The Practice and Potential of Agroforestry in the Southeastern United States. White paper. Center for Subtropical Agroforestry (CSTAF) School of Forest Resources and Conservation, Institute of Food and Agricultural Sciences, University of Florida. http://cstaf.ifas.ufl.edu/whitepaper.htm verified 14 November 2006. Wood, C.W. 1998. Chapter 1: Agricultural Phosphorus and Water Quality: An overview. A publication of SERA-IEG 17: Minimizing agricultural phosphorus losses for protection of the water resource. Available: http://www.soil.ncsu.edu/sera17/publications/sera17-1/chapter1-98.htm verified 14 November 2006. Yadav, S.N., 1997. Formulation and estimation of nitrate-nitrogen leaching from corn cultivation. Journal of Environmental Quality. 26: 808-814. Young, A. 1997. The effectiveness of contour hedgerows for soil and water conservation. Agroforestry Forum. 8:2-4. Zublena, J.P. 1995. Characteristics of Animal wastes and waste-amended soils: Nutrientmanagement planning beyond the farm boundary. In: K. Steel (ed.) Animal waste and the land-water interface. Lewis Publi-CRC Press, New York. Zinkhan, C.F., and D.E. Mercer. 1997. An assessment of agroforestry systems in the southern USA. Agroforestry Systems. 35: 303-321.

PAGE 130

BIOGRAPHICAL SKETCH Grard-Alain Michel, the third of three siblings, was born December 12, 1969, in Port-au-Prince, Haiti. Alain is a husband and a father of a 6 year-old boy named Alain-Christian Michel. In July 1996, he received his BS in agronomy from the School of Agricultural and Environmental Sciences of Quisqueya University, Port-au-Prince, Haiti. After working for 2 years in extension at Quisqueya University in a Kellogg-funded project of rural development called UNIR/UniQ, he won a Fulbright Scholarship to study in the United States, where he received, in 2000, his M.Sc. in forest resources management at the School of Environmental Science and Forestry of the State University of New York, Syracuse, New York. Back in Haiti, he was appointed faculty member at the School of Agricultural and Environmental Sciences of Quisqueya University, where he taught two courses. Less than two years later, in spring 2002, he obtained a research assistantship to start a PhD program in Agroforestry at the School of Forest Resources and Conservation of the University of Florida under the supervision of Dr. P.K.R. Nair. 116