<%BANNER%>

Evaluating Agronomic and Environmental Impacts of Phosphorus Fertilization of Low Input Bahiagrass Pastures in Florida

Permanent Link: http://ufdc.ufl.edu/UFE0042001/00001

Material Information

Title: Evaluating Agronomic and Environmental Impacts of Phosphorus Fertilization of Low Input Bahiagrass Pastures in Florida
Physical Description: 1 online resource (137 p.)
Language: english
Creator: Obour, Augustine
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: bahiagrass, cow, low, phosphorus, spodosols
Soil and Water Science -- Dissertations, Academic -- UF
Genre: Soil and Water Science thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Environmental concerns over potential phosphorus (P) losses and their impacts on water quality have prompted several revisions in P fertilizer recommendations for bahiagrass (Paspalum notatum Flu umlautgge) pastures in Florida. The current University of Florida Institute of Food and Agricultural Sciences (UF/IFAS) fertilizer recommendations require both soil and tissue testing for bahiagrass pastures, but there are limited data to support the new recommendations. Greenhouse and field experiments (clipping and grazing studies) were conducted at UF/IFAS Range Cattle Research and Education Center in Ona, FL from 2006 to 2008 to evaluate the effects of the revised UF/IFAS fertilizer recommendations on forage yield, nutritive value and the potential impacts on water quality in bahiagrass pastures growing on a local Spodosol. The greenhouse study data indicated that dry matter yields (DMY) and tissue P concentration increased with P application. The critical minimum bahiagrass tissue P concentration below which DMY was impacted was 1.3 g kg-1. In the 2-yr field grazing study, bahiagrass herbage mass, herbage accumulation rates, crude protein (CP), and in vitro digestible organic matter (IVDOM) were not affected by P application. However, tissue P concentrations increased from 1.9 to 2.2 g kg-1 as P fertilization increased from 0 to 10 kg P ha-1. Mehlich-1 soil extractable P and leachate-P concentrations were not affected by P application. A field clipping study evaluated three N rates (0, 56 and 112 kg ha-1) and four P rates (0, 5, 10 and 20 kg P ha-1) effects on forage production, P leaching and effects of water table fluctuations on P fluxes from the spodic layer. In 2007, bahiagrass showed no response to P, however, there was a linear increase in DMY as P rates increased in 2008. Phosphorus additions had no effects on soil Mehlich-1 and leachate-P concentrations. Leachate-P concentrations in lysimeters above the spodic horizon varied seasonally, with spikes coinciding with periods of high rainfall and rising water tables. Leachate-P concentration at depths below the spodic horizon remained relatively constant (0.02 mg L-1) during the entire growing season. Greatest soil-P availability as determined with anion exchange membranes occurred in August when the water table rose. Phosphorus mass balance computations based on input and outputs and soil P content in the top soil (Ap horizon) were negative for all treatments. However, when soil P concentrations within the subsoil (Bh horizon) were included in the P mass balance, the overall P mass balances were positive for all treatments. Collectively, the study results suggest that tissue testing can be used to predict established bahiagrass response to P fertilization, and that minimal P fertilization can improve bahiagrass yield with no impacts on water quality. Fluctuating water table conditions cause upward fluxes of P from the Bh horizon, which increases soil P bioavailability. It is therefore imperative that P levels in the Bh horizon be considered in bahiagrass nutrient management programs in cow-calf pastures established on Spodosols in Florida.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Augustine Obour.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Silveira, Maria L.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2012-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0042001:00001

Permanent Link: http://ufdc.ufl.edu/UFE0042001/00001

Material Information

Title: Evaluating Agronomic and Environmental Impacts of Phosphorus Fertilization of Low Input Bahiagrass Pastures in Florida
Physical Description: 1 online resource (137 p.)
Language: english
Creator: Obour, Augustine
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: bahiagrass, cow, low, phosphorus, spodosols
Soil and Water Science -- Dissertations, Academic -- UF
Genre: Soil and Water Science thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Environmental concerns over potential phosphorus (P) losses and their impacts on water quality have prompted several revisions in P fertilizer recommendations for bahiagrass (Paspalum notatum Flu umlautgge) pastures in Florida. The current University of Florida Institute of Food and Agricultural Sciences (UF/IFAS) fertilizer recommendations require both soil and tissue testing for bahiagrass pastures, but there are limited data to support the new recommendations. Greenhouse and field experiments (clipping and grazing studies) were conducted at UF/IFAS Range Cattle Research and Education Center in Ona, FL from 2006 to 2008 to evaluate the effects of the revised UF/IFAS fertilizer recommendations on forage yield, nutritive value and the potential impacts on water quality in bahiagrass pastures growing on a local Spodosol. The greenhouse study data indicated that dry matter yields (DMY) and tissue P concentration increased with P application. The critical minimum bahiagrass tissue P concentration below which DMY was impacted was 1.3 g kg-1. In the 2-yr field grazing study, bahiagrass herbage mass, herbage accumulation rates, crude protein (CP), and in vitro digestible organic matter (IVDOM) were not affected by P application. However, tissue P concentrations increased from 1.9 to 2.2 g kg-1 as P fertilization increased from 0 to 10 kg P ha-1. Mehlich-1 soil extractable P and leachate-P concentrations were not affected by P application. A field clipping study evaluated three N rates (0, 56 and 112 kg ha-1) and four P rates (0, 5, 10 and 20 kg P ha-1) effects on forage production, P leaching and effects of water table fluctuations on P fluxes from the spodic layer. In 2007, bahiagrass showed no response to P, however, there was a linear increase in DMY as P rates increased in 2008. Phosphorus additions had no effects on soil Mehlich-1 and leachate-P concentrations. Leachate-P concentrations in lysimeters above the spodic horizon varied seasonally, with spikes coinciding with periods of high rainfall and rising water tables. Leachate-P concentration at depths below the spodic horizon remained relatively constant (0.02 mg L-1) during the entire growing season. Greatest soil-P availability as determined with anion exchange membranes occurred in August when the water table rose. Phosphorus mass balance computations based on input and outputs and soil P content in the top soil (Ap horizon) were negative for all treatments. However, when soil P concentrations within the subsoil (Bh horizon) were included in the P mass balance, the overall P mass balances were positive for all treatments. Collectively, the study results suggest that tissue testing can be used to predict established bahiagrass response to P fertilization, and that minimal P fertilization can improve bahiagrass yield with no impacts on water quality. Fluctuating water table conditions cause upward fluxes of P from the Bh horizon, which increases soil P bioavailability. It is therefore imperative that P levels in the Bh horizon be considered in bahiagrass nutrient management programs in cow-calf pastures established on Spodosols in Florida.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Augustine Obour.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Silveira, Maria L.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2012-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0042001:00001


This item has the following downloads:


Full Text

PAGE 1

1 EVALUATING AGRONOMIC AND ENVIRONMENTAL IMPACTS OF PHOSPHORUS FERTILIZATION OF LOW INPUT BAHIAGRASS PASTURES IN FLORIDA By AUGUSTINE K. OBOUR 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 2010

PAGE 2

2 2010 Augustine K. Obour

PAGE 3

3 T o God be the glory, great things H e has done.

PAGE 4

4 ACKNOWLEDGMENTS I express my profound gratitude to my super visor, Dr. M.L Silveira for her mentorship, encouragement and candid criticisms during my graduate program. I am heartily thankful to Dr. G.A OConnor for act ing as my oncampus advisor and all the precious time he spent in helping shape my research ideas. Much appreciation goes to my other committee members, (Dr. J. Jawitz, Dr. L. E. Sollenberger and Dr. J.M.B Vendramini) for their suggestions and immense contribution to my research. Special thanks go to Cindy Holley, Bill, Curly, and all the farm crew at t he Range Cattle Research and Education Center who contributed in various ways to the conduct of the study. I am highly indebted to fellow graduate students and friends from the Environmental Soil Chemistry group: Daniel, Jaya, Liz, Manmeet, Matt, Sampson a nd Xiaolin. Especially, I would like to thank Xiaolin for his as sistance in field sampling and being a great companion in the frequent long hours of commuting from Gainesville to Ona. I will also like to show my gratitude to my parents, Mr and Mrs. Bekoe for making me who I am today. Finally, I owe my deepest gratit ude to my family; my wife Anna and my son, Obour Jr., for all their prayers, love and support.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................. 4 LIST OF TABLES ............................................................................................................ 8 LIST OF FIGURES .......................................................................................................... 9 ABSTRACT ................................................................................................................... 10 CHAPTER 1 INTRODUCTION AND LITERATURE REVIEW ..................................................... 12 Introduction ............................................................................................................. 12 Hypotheses ............................................................................................................. 14 Objectives ............................................................................................................... 14 Trends of Fertilizer Management in Cow Calf Pastures in Florida .......................... 15 Bahiagrass Fertilization in South Florida ................................................................. 17 Tissue testing in Nutrient Management ................................................................... 21 Environmental Impacts of P Fertilization on Cow ca lf Pastures in Florida .............. 22 Phosphorus Movement in Florida Spodosols .......................................................... 28 2 PREDICTING BAHIAGRASS REPONSE TO PHOSPHORUS FERTILIZATIO N USING TISSUE TESTING ...................................................................................... 30 Introduction ............................................................................................................. 30 Materials and Methods ............................................................................................ 32 Experiment Design and Setup .......................................................................... 32 Statistical Analyses .......................................................................................... 34 Results and Discussion ........................................................................................... 35 Bahiagrass Dry Matter Yield ............................................................................. 35 Tissue P and N Concentrations ........................................................................ 36 Phosphorus Uptake and Recovery ................................................................... 38 Root Mass Accumulation .................................................................................. 39 Summary and Conclusions ..................................................................................... 40 3 LOW RATES OF PHOSPH ORUS FERTILIZATION EFFECTS ON FORAGE PRODUCTION AND WATER QUALITY OF GRAZED BAHIAGRASS PASTURES ............................................................................................................ 46 Introduction ............................................................................................................. 46 Material and Methods ............................................................................................. 48 Experimental Setup .......................................................................................... 48 Forage Sampling .............................................................................................. 49 Soil Samplin g Protocol and Analysis ................................................................ 51

PAGE 6

6 Water Quality Monitoring .................................................................................. 52 Statistical Analysis ............................................................................................ 52 Results and Discussion ........................................................................................... 53 Climatological Data .......................................................................................... 53 Herbage Mass and Accumulation Rate ............................................................ 53 Herbage Nutritive Value ................................................................................... 54 Plant Tissue P Concentration ........................................................................... 54 Soiltest P Concentrations ................................................................................ 57 Leachate P and NO3N Concentration .............................................................. 58 Summary and Conclusions ..................................................................................... 59 4 AGRONOMIC AND ENVIRONMENTAL IMPACTS OF PHOSPHORUS FERTILIZATION OF LOW INPUT BAHIAGRASS PASTURES .............................. 66 Introduction ............................................................................................................. 66 Materials and Methods ............................................................................................ 68 Experiment Setup ............................................................................................. 68 Forage Dry Matter Yield, Crude Protein and Tissue P Analysis ....................... 68 Soil Analysis ..................................................................................................... 69 Water Quality Monitoring .................................................................................. 70 Statistical Analysis ............................................................................................ 70 Results and Discussion ........................................................................................... 71 Climatological Data .......................................................................................... 71 Bahiagrass Dry Matter Yield ............................................................................. 71 Tissue P Concentration and P Uptake ............................................................. 73 Crude Protein Concentration and N Uptake ..................................................... 74 Soil Phosphorus Concentrations ...................................................................... 75 Leachate Phosphorus Concentration ............................................................... 76 Summary and Conclusions ..................................................................................... 78 5 FLUCTUATING WATER TABLE EFFECT ON PHOSPHORUS RELEASE AND AVAILABILITY FROM A FLORIDA SPODOSOL .................................................... 82 Introduction ............................................................................................................. 82 Materials and Methods ............................................................................................ 84 Experimental Design and Setup ....................................................................... 84 Measuring Soil Phosphorus Bioavailabil ity ....................................................... 85 Soil Sampling, Analysis and Phosphorus AdsorptionDesorption Study .......... 88 Calculation of Sorption Parameters .................................................................. 90 Monitoring Leachate P Concentration and P Release From the Spodic Layer ............................................................................................................. 91 Results and Discussions ......................................................................................... 92 Soil Phosphorus Concentration and P Sorption Study ..................................... 92 Rainfall and Water Table Fluctuations .............................................................. 95 Changes in Leachate P Concentration ............................................................. 95 Seasonal Variation in Soil P Availability ........................................................... 97 Summary and Conclusions ..................................................................................... 98

PAGE 7

7 6 PHOSPHORUS MASS BALANCE IN A TYPICAL FLORIDA SPODOSOL .......... 104 Introduction ........................................................................................................... 104 Materials an d Methods .......................................................................................... 108 Study Site ....................................................................................................... 108 Crop P Uptake ................................................................................................ 108 Soil P .............................................................................................................. 109 Water Quality Monitoring ................................................................................ 109 Phosphorus Mass Balance Calculations ........................................................ 110 Statistical Analysis .......................................................................................... 112 Results and Discussion ......................................................................................... 113 Crop P Uptake ................................................................................................ 1 13 Changes in Soil Phosphorus Concentration ................................................... 113 Rainfall Distribution, Water Table Fluctuations and Drainage ........................ 114 Phosphorus Leaching and Runoff L osses ...................................................... 115 Phosphorus Mass Balance ............................................................................. 116 Summary and Conclusions ................................................................................... 117 7 SUMMARY AND CONCLUSIONS ........................................................................ 122 Summary .............................................................................................................. 122 Conclusions .......................................................................................................... 124 LIST OF REFERENCES ............................................................................................. 126 BIOGRAPHICAL SKETCH .......................................................................................... 137

PAGE 8

8 LIST OF TABLES Table page 2 1 Bahiagrass cumulative dry matter (DM) yield as affected by N and P application rates. ................................................................................................ 41 2 2 Bahiagrass tissue P concentration, P uptake, and P recovery as affected by N and P application rates. .................................................................................. 41 2 3 Effect of N and P application rates on bahiagrass root mass. ............................ 42 3 1 Monthly maximum and minimum temperatures and rainfall at the Range Cattle Research and Education Center (RCREC), Ona in 2007 and 2008. ........ 61 3 2 Year x month interaction effect on herbage mass, accumulation rate, crude protein (CP), in vitro digestible organic matter (IVDOM), and tissue P concentration (TKP) on bahiagrass pastures with three P fertilization levels. .... 62 3 3 Tissue P concentration and leachate nitrate concentration as affected by P application rate. .................................................................................................. 63 3 4 Effect of feeding, open, and shaded regions on soil extractable P concentration before and after treatment application. ......................................... 63 4 1 Monthly maximum and minimum temperatures and rainfall at the Range Cattle Research and Education Center (RCREC), Ona, FL in 2007 and 2008. .. 79 4 2 Cumulative bahiagrass dry matter yields as affected by P application rates and year. ............................................................................................................ 79 4 3 Tissue P concentration and P uptake as affected by P application rates on bahiagrass.. ........................................................................................................ 80 4 4 Mehlich 1 soil P concentration at various depths as affected by year.. .............. 80 4 5 LeachateP concentrations at the various depths as affected by year and P application rate. .................................................................................................. 80 5 1 Average Mehlich1 soil Al, Fe, and P concentrations at various depths in the controls at the end of the study in 2008. ............................................................. 99 5 2 Phosphorus sorption characteristics of the Bh horizon at the study site. ............ 99 6 1 Mehlich 1 soil P concentration at various depths as affected by year. ............ 119 6 2 Estimated P mass balances for various depths in 2007and 2008 as affected by fertilizer P, atmospheric deposition, soil available P and P loss through leaching and runoff. .......................................................................................... 119

PAGE 9

9 LIST OF FIGURES Figure page 2 1 Bahiagrass tissue P concentration as affected by harvest frequency.. ............... 43 2 2 Bahiagrass cumulative yield as affected by N rate and harvest frequency.. ....... 43 2 3 Bahiagrass tissue N concentration as affected by N rate and harvest frequency.. .......................................................................................................... 44 2 4 Relationship between bahiagrass DM yield and tissue P concentration. ............ 44 2 5 Bahiagrass tissue P uptake as affected by N rate and harvest frequency.. ........ 45 3 1 Leachate P concentration at various sampling dates as affected by P application rate in (a) 2007and (b) 2008. ............................................................ 64 3 2 Leachate NO3N concentrations as a function of sampling date in (a) 2007, and (b) 2008. ...................................................................................................... 65 4 1 Leachate P concentrations as affected by soil depth and sampling date in ( a) 2007 and (b) 2008. ............................................................................................. 81 5 1 Rainfall distribution and water table depth in (a) 2007 and (b) 2008. ................ 100 5 2 Phosphorus sorpt ion isotherm for the Bh horizon soil at the study site. ........... 101 5 3 LeachateP concentration in control plots as affected by soil depth and sampling date in (a) 2007 and (b) 2008. ........................................................... 102 5 4 Relationship between water table depth and leachate P concentration at the 15 and 30 cm lysimeters in the 2yr study. ....................................................... 103 5 5 In situ soil P availability measured by anion exchange membranes in 2008.. .. 103 6 1 Rainfall and water table depth in (a) 2007 and (B) 2008. ................................. 120 6 2 Estimated daily drainage below 45 cm rooting depth in (A) 2007 and (B) 2008.. ............................................................................................................... 121

PAGE 10

10 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Par tial Fulfillment of the Requirements for the Degree of Doctor of Philosophy EVALUATING AGRONOMIC AND ENVIRONMENTAL IMPACTS OF PHOSPHORUS FERTILIZATION OF LOW INPUT BAHIAGRASS PASTURES IN FLORIDA By Aug us tine K. Obour August 2010 Chair: Maria L. Silvei ra Major: Soil and Water Science Environmental concerns over potential phosphorus (P) losses and their impact s on water quality have prompted several revisions in P fertilizer recommendations for bahiagrass ( Paspalum notatum Fl gge) pastures in Florida. The current University of Florida Institute of Food and Agricultural Sciences (UF/IFAS) fertilizer recommendations require both soil and tissue testing for bahiagrass pastures but there are limited data to support the new recommendations. G reenhouse and f ield experiments (clipping and grazing studies) were conducted at UF/IFAS Range Cattle Research and Education Center in Ona, FL from 2006 to 2008 to evaluate the effects of the revised UF/IFAS fertilizer recommendations on forage yield, nutritive value and the potential impacts on water quality in bahiagrass pastures growing on a local Spodosol. The greenhouse study data indicated that dry matter yields (DMY) and tissue P concentration increased with P application. The critical minimum bahiagrass tissue P c oncentration below which DMY was impacted was 1.3 g kg1. In the 2 yr field grazing study bahiagrass herbage mass, herbage accumulation rates, crude protein (CP), and in vitro digestible organic matter ( IVDOM ) were not affected by P application. However, tissue P concentrations

PAGE 11

11 increased from 1.9 to 2.2 g kg1 as P fertilization increased from 0 to 10 kg P ha1. Mehlich 1 soil extractable P and leachateP concentrations were not affected by P application. A field clipping study evaluated three N rates (0, 56 and 112 kg ha1) and four P rates (0, 5, 10 and 20 kg P ha1) effects on forage production, P leaching and effects of water table fluctuations on P fluxes from the spodic layer In 2007, bahiagrass showed no response to P, however, there was a linear increase in DMY as P rates increased in 2008. Phosphorus additions had no effects on soil Mehlich1 and leachateP concentrations. LeachateP concentrations in lysimeters above the spodic horizon varied seasonally, with spikes coinciding with periods of high rainfall and rising water tables. LeachateP concentration at depths below the spodic horizon remained relatively constant (0.02 mg L1) during the entire growing season. Greatest soil P availability as determined with anion exchange membranes occurred in August when the water table rose Phosphorus mass balance computations based on input and outputs and soil P content in the top soil (Ap horizon) were negative for all treatments. However, when soil P concentrations within the subsoil (Bh horizon) were in cluded in the P mass balance, the overall P mass balances w ere positive for all treatments. Collectively, the study results suggest that tissue testing can be used to predict established bahiagrass response to P fertilization, and that minimal P fertilizat ion can improve bahiagrass yield with no impacts on water quality. Fluctuating water table conditions cause upward flux es of P from the Bh horizon, which increases soil P bioavailability It is therefore imperative that P levels in the Bh horizon be considered in bahiagrass nutrient management programs in cow calf pastures established on Spodosols in Florida.

PAGE 12

12 CHAPTER 1 INTRODUCTION AND LIT ERATURE REVIEW I ntroduction Bahiagrass ( Paspalum notatum Fl gge) is the most widely used planted perennial forage g rass in Florida (Chambliss and Adjei, 2006). It is well adapted to sandy soils, tolerates low soil fertility, low pH, intermittent wet conditions and produces reasonable forage yields in dr oughty soils. Bahiagrass pastures occupy about 2 million of the 5 m illion ha of improved pasture in S outh Florida (Muchovej and Mullahey, 2000) and predominantly grown on S podosols, which constitute over 3.4 million ha in Florida (Collins, 2003). Most foragebased beef cattle systems in Florida rely on bahiagrass pastures as the major source of energy and protein for most of the growing season. Although mineral supplementation is common in Floridas cow calf system, production of adequate bahiagrass forage of high quality is critical for the success of the cow calf industr y in the state. The cow calf industry plays a major role in Floridas economy. In 2006, the Florida cattle industry ranked 12th in beef cows and 18th in total cattle nationally (FDACS, 2006). The top ten beef cattle counties in Florida are Okeechobee (10%), Osceola (7%), Highlands (6%), Polk (6%), Hardee (5%), Hendry (5%), DeSoto (5%), Glades (4%), Hillsborough (4%), and Manatee (3%) (FDACS, 2006). Approximately 55% of the state cow calf production is concentrated in the environmentally sensitive wetland ecosystems in S outh Florida. These watersheds are threatened by nutrients losses, especially nitrogen (N) and phosphorus (P) from agricultural activities from adjacent lands (Reddy et al., 1999). Dairy and beef cattle operations are regarded as major cont ributor s of P to Lake Okeechobee in S outh Florida (Bottcher et al., 1995).

PAGE 13

13 Though P losses from dairy operations is greater than from beef ranches, the latter occupy about 65% of the major land use in the l akes watershed (Allen et al., 1982) and may be targeted by regulatory agencies to reduc e P losses. Due to the extensive acreage of bahiagrass in cow calf operations and potential environmental impacts associated with nutrient loss, fertilizer management of bahiagrass pastures has received considerable at tention by state regulatory agencies, livestock farmers and researchers in recent years. Bahiagrass response to P fertilization has been less consistent than response to N fertilization While some studies have shown that bahiagrass can respond to P appl ication rates of 6 to 24 kg P ha1 when Mehlich1 soil test P is below 10 mg kg1 (McCaleb et al., 1966; Payne and Rechcigl, 1989; Rechcigl et al., 1992), others report no yield response to P fertilization (Ibrikci et al., 1994, Ibrikci et al., 1999; Rechc igl et al., 1992). The lack of yield response to P application was attributed to availability of P from the Bh horizon, which is accessible to bahiagrass roots (Ibrickci et al., 1994; Ibrikci et al., 1999; Rechcigl and Bottcher, 1995). The hypothesized sup ply of P from the Bh horizon to bahiagrass pastures has not been fully proven. Additionally, the S podosols in Florida are subject to fluctuating water table conditions which may have a significant effect on P fluxes from the spodic horizon and consequently affect P availability The discrepancies in the literature regarding bahiagrass response to P fertilization may also be due to the inability of routine soil testing (sampling only the top 15 cm of the soil) alone to accurately predict forage P requiremen ts. Therefore additional testing procedures for nutrient management programs are needed that will predict when bahiagrass needs additional P without contributing to offsite loss of P to

PAGE 14

14 the environment. The combination of plant tissue and soil testing recommended by the current UF/IFAS bahiagrass P recommendations for established pastures (Mylavarapu et al., 2007) is one approach to addressing the problem. However, there are limited field data to validate the approach. Additionally, the impacts of P fertilization at the recommended rates on water quality have not been fully investigated. Field research is necessary to determine the minimum amount of P required to sustain bahiagrass yields without negatively impact ing water quality The goal is to find a sati sfactory balance between forage production, nutritive value and environmental quality. G reenhouse and field studies were conducted to determine the minimum amount of P fertilizer required to sustain bahiagrass yield and forage quality in cow calf operations that simultaneously minimize P impacts on water quality Additionally, the effects of fluctuating water table conditions experienced in S outh Florida on P availability to established bahiagrass pastures and on P losses to the environment were investigat ed under field conditions. Hypothes e s Hypothesis 1: Minimal P fertilizer applications will sustain bahiagrass yield without negative impacts on water quality Hypothesis 2: Phosphorus availability to bahiagrass and P losses to the environment will increas e under fluctuating water table conditions O bjectives Objective 1: Determine the minimum amount of P application required to produce satisfactory bahiagrass yields. Objective 2: Determine the minimum critical bahiagrass tissue P concentration below which DM yields are negatively impacted. Objective 3: Investigate P leaching following P application. Objective 4: Determine the effects of fluctuating water table on soil P availability

PAGE 15

15 Objective 5: Determine P mass balance for low input bahiagrass pastures. Trends of Fertilizer Management in Cow Calf Pastures in Florida Cattle were first introduced to Florida in 1521 by the Spanish and by early 1700 there were four distinct cattle raising centers around Tallahassee, Gainesville, St. Augustine, and along the St. Johns River (Yarlett, 1985). The early ranches produced beef primary for the Spanish military garrisons. Largescale cattle raising started in the late 1890s in Florida. The introduction of planted pastures around 1921 was a significant boost to the c attle industry in Florida, since cattle grazing was primarily based on cleared forest, prairie lands, common bermudagrass [ Cynodon dactylon (L.) Pers.] and highway right sof way (Yarlett, 1985). The earlier pasture improvement programs involved seeding carpetgrass ( Axonopus affinis Chase) or common bahiagrass ( Paspalum notatum Flgge) fertilized with 2000 kg ha1 raw rock phosphate (Hodges and McCaleb, 1959). The first grazing research trial on pasture establishment on Spodosols was conducted at the Range Cattle Research Station at Ona in 1941 using carpetgrass and bahiagrass. Fertilizer rates used in the study were 34 kg ha1 of N, P2O5, and K2O and 1800 and 2000 kg ha1 of rock phosphate and lime, respectively (Hodges and McCaleb, 1959).The highly productive planted pasture species introduced in Florida at that time included Pangola digitgrass ( Digitaria eriantha Steud ), Pensacola bahiagrass, bermudagrass, and stargrass ( Cynodon nlemfluencis Vanderyst). Nutrient requirements for these species were generally greater than the indigenous grasses and greater fertilizer inputs were needed to achieve their maximum production potential. In the early 1950s, fertilizer recommendation rates were the same for all grasses, and consisted of approximately 81 kg h a1 of N and 54 kg ha1 of P2O5 and K2O applied as a fertilizer mix

PAGE 16

16 (Hodges and McCaleb, 1959). Increased knowledge about the nutrient requirements of the various pasture species revealed that greater fertilizer rates increased pasture productivity and res ulted in greater animal productivity and overall pasture carrying capacity. Accordingly, numerous fertilizer trials were conducted during the 1950s to 1970s involving relatively high N, P, and K fertilizer applications rates (Blue and Gammon, 1955; Jeffers 1955; Hodges and McCaleb, 1959; Hodges et al., 1976; Mislevy and Hodges, 1976). Research focused on avoiding crop and animal P deficiency, which was often experienced in range cattle operations. Moreover, the relatively low cost of fertilizer also favored high application rates. In the late 1970s, increased fertilizer cost and evidence that maximum yields under grazing conditions could be obtained at lower fertilizer rates than those used for hay fields led scientist to consider reducing fertilizer input s in pastures (Blue, 1988; Hodges et al., 1976). Blue (1988) evaluated bahiagrass response to N fertilizer rates of 0 to 200 kg N ha1 while P and K rates were chosen to provide a N:P:K ratio of 4:1:2. Similarly, Hodges et al. (1976) conducted grazing trials on tropical grass legume mixtures at reduced fertilizer rates of 112, 24, and 52 kg ha1 of N, P, and K, respectively. These early research efforts were the basis for the development of forage crop fertilizer recommendations (134 kg N ha1, 48 kg P ha1, and 75 kg kha1) in the 1980s and early 1990s (Whitty et al., 1977). However, because of increasing cost and concerns about the impacts of fertiliz er nutrients on water quality, fertilizer recommendations for forage crops in Florida have undergone several modifications during the last decade. Bahiagrass fertilization, in particular, has been a topic of agronomic and environmental importance due to the extensive area of planted

PAGE 17

17 bahiag rass pastures in the state and potential environmental impacts associated with nutrient loss from fertilized pastures Current fertilizer management strategies for forage crops, including bahiagrass, are aimed at balancing agronomic nutrient requirement s of crops while reducing the risks of nutrient accumulation in soils and subsequent transport to water bodies Bahiagrass Fertilization in S outh Florida Bahiagrass like most warm season grasses responds very well to N fertilization. Application of N to bahiagrass pastures increased yields from 3.34 Mg ha1 for the control to 10.3 Mg ha1 in plots receiving 270 kg N ha1 (Beaty et al 1960). Similarly, Blue (1970) found bahiagrass dry matter yields of 3 to 4 Mg ha1 when no N was applied and 12 Mg ha1 when N was applied at 224 kg N ha1. Burton et al. (1997) showed that increas ing N application from 56 to 448 kg N ha1 increased bahiagrass herbage yield by 155%; however, there was no response to P or potassium (K) application. In a multilocational 3 yr study conducted in nine counties in S outh Florida, Sumner et al. (1991) found that annual N fertilization at 67 kg ha1 sustained adequate forage yield in low input cow calf systems in Florida. Although bahiagrass responded to P or K fertilization, Sumner et al. (1991) indicated that P and K applications were not economical when c ompared to N application alone Bahiagrass response to P fertilization has been less consistent than response to N application ; therefore whether P application is warranted continues to be an important agronomic and environmental issue for established bahi agrass pastures. Bahiagrass nutrient management has undergone several reviews due to concerns regarding the environmental impacts of pasture fertilization in Florida. In 1990, an ad hoc committee reviewed bahiagrass fertilization management recommendations for established

PAGE 18

18 pastures in Florida (Hanlon et al., 2006). The target soil pH was identified as 5.5 and lime application was recommended whenever the soil pH recommendations were based on the grower selected N option and Mehlich1 soil test interpretation. For example, when a grower adopted the high N option (180 kg N ha1), the rec ommendation was to apply P at a rate of 19.8 kg P ha1. For the medium N option (112 kg N ha1), P was suggested at 12.3 kg P ha1 when soil test P 1. For the low N option (56 kg ha1) only used for grazed pastures, no P application was suggest ed (Kidder et al., 2000). The suggested P rates were based on increasing N application rates to insure adequate plant nutrition and forage yield. In 1996, the state was divided into North and South designations (dividing line roughly north and south, respectively, of I 4 near Orlando) for nutrient management purposes. Based on available literature at that time [Sumner et al. (1991), Rechcigl et al. (1992) and Rechcigl et al. (1995)], the target soil pH values were set at 5.5 and 5.0 for the North and South regions of the State, respectively. Soil testing was not recommended for S outh and C entral Florida because bahiagrass showed no economic response to P and K fertilization (Sumner et al., 1991). The current P fertilizer recommendation for established bahiagrass in Florida is based on both soil and plant tissue P testing (Mylavarapu et al., 2007). Phosphorus is only recommended when tissue P concentrations are below 1.5 g kg1 and soil test P levels are very low or low (< 15 mg kg1 Mehlich1 P). Conflicting reports in the literature on bahiagrass response to P fertilizer based on soil testing alone prompted this new recommendation. Research has shown that bahiagrass can respond to P application rates in the range of 6 to 24 kg P ha1 when Mehlich1 soil test P is below 10 mg kg1

PAGE 19

19 (McCaleb et al., 1966; Payne and Rechcigl, 1989; Rechcigl et al., 1992). In a 2yr field study on Immokalee fine sand, Rechcigl et al. (1992) reported a quadratic response of bahiagrass yields to P application. Maximum yields of 13.5 (1988) and 14.7 Mg ha1 ( 1989) were obtained at an applied P rate of 24 kg P ha1. All plots in the study received N application at 120 kg N ha1. Ibrikci et al. (1992) investigat ed P fertilization on a Myakka fine sand in Ona, FL with an initial Mehlich1 P concentration of 6.9 mg kg1 and N rate of 120 kg ha1. B ahiagrass responded only to P application rates of 1. In contrast, Rechcigl et al. (1995) showed that bahiagrass yields were not affected by the addition of P, and attributed the lack of response to the ability of bahiagrass roots to obtain P from the s podic horizon, which typically contains significant amounts of soil test P. These findings were contradictory to earlier reports by the same authors (Rechcigl and Bottcher, 1995; Rechcigl et al., 1992). Similarly, Ibrikci et al. (1999) concluded that a lac k of bahiagrass yield response to P application was due to P from the Bh horizon being available for plant uptake. However, a critical assessment of the data from Ibrikci et al. (1999) revealed that bahiagrass yield responded to P application in the second year of the study. Despite the suggestion that subsurface (spodic) horizons play an important role in bahiagrass grown in Florida S podosols, the hypothesized P supply from the Bh horizon has not been fully investigated. Important questions regarding how much P is available for bahiagrass uptake and how long the Bh horizon can provide adequate P to sustain bahiagrass pastures remain unanswered. Sigua et al. (2004) reported about 21% declines in soil P concentrations in the top 25 cm in a 12yr study in gra zed bahiagrass pastures receiving only N at the USDA ARS Subtropical Agricultural

PAGE 20

20 Research Station in Brook sville, FL. The initial Mehlich 1 P concentration i n the top 25 cm in 1988 was 71 mg kg1 and by the year 2000 had decreased to 56 mg kg1. The soils used in study were Ultisols which do not represent typical Florida Spodosols used in cow calf operations in Florida with general lower soil test P levels in the Ap horizon. The discrepancies in the literature regarding bahiagrass response to P fertilizat ion may be partially due to the inability of soil testing alone to accurately predict forage P requirements. Additional testing for nutrient management programs are needed that will predict when bahiagrass requires additional P while avoiding offsite loss of P to the environment. Tissue and soil testing suggested by the current UF/IFAS bahiagrass P recommendations for established pastures (Mylavarapu et al., 2007) may be one way to resolve the problem. According to the new fertilizer recommendations, there is no need for tissue testing and P application if the Mehlich1 soil test P is medium (1630 mg kg1) or high (3160 mg kg1). When soil test P is low/very low (< 10 and 1015 mg kg1, respectively) and tissue P Howe ver, if soil test P is low/very low and tissue P < 0.15%, 12.3 kg P ha1 is recommended for the low and medium N options (Mylavarapu et al., 2007). There are limited field data to validate the new recommendations. Additionally, the impacts of P fertilization at the recommended rates on water quality have not been fully investigated. Field research is necessary to determine the m inimum amount of P required to maintain bahiagrass yield without negative impacts on t he environment. The goal is to find a satisfa ctory balance between forage production, nutritive value and environmental quality.

PAGE 21

21 Tissue testing in Nutrient Management Plant tissue analysis is used widely as a diagnostic tool for assessing the nutrient requirement of crops (Ulrich, 1952; Smith, 1962; Tyner, 1946; Mallarino and Higashi, 2009). Unlike soil analysis which relates soil extracted nutrients to plant response, plant analysis usually give a direct indication of nutrient availability to the crop. The application of plant tissue analysis to pl ant nutrition revolves around the concept of a critical nutrient concentration in the plant determined from calibration curves. The critical tissue concentration of a particular nutrient is defined as the nutrient concentration corresponding to 90% of maxi mum yield (Ulrich and Hills, 1973). The latter authors developed a general response curve relating yield to nutrient concentration and divided the curve into three zones (deficient, transition and adequate zones). Within the transition zone is the critical tissue concentration, plants with tissue nutrient concentrations above the critical concentration are adequately supplied with nutrients whereas lower nutrient concentrations are considered deficient and such plants are expected to respond to fertilizati on. Nutrient concentrations i n plant tissue are affected by a number of factors. These includes plant part used for the analysis and physiological growth stage ( Bates, 1971, Mallarino and Higashi, 2009, Jones, 1970 ), supply of nutrients, mobility of par ticular nutrient in the plant, soil moisture, temperature (Follett and Reichman, 1972; Nielson et al., 1960; Power et al., 1963), and seasonality (Grings et al., 1996; Walsh and Bir rell, 1987). Plant tissue analysis can therefore be used as a diagnostic t ool to provide understanding of nutritional status of many crops. The critical tissue nutrient concentrations of several crops have been investigated by scientists worldwide but studies on pastures are limited. In Coastal bermudagrass [ Cynodon dactylon (L ) Pers .] critical N is 18 to 22 g kg1 (Burton, 1954,

PAGE 22

22 Adams et al 1967), P ranged from 2.0 to 2.6 g kg1 (Fisher and Caldwell, 1959) and K 15 to17 g kg1 (Jordon et al., 1966). For P angola digit grass ( Digitaria eriantha Ste ud) critical tissue N wa s 12 to 16 g kg1 (Harris et al., 1968), P ranged from 1.2 to 1.6 g kg1 ( L ittle et al., 1959) and K 12 to 14 g kg1 (Plucknett and Fox, 1965). In St. Augustinegrass [ Stenotaphrum secon datum (Walt.) Kuntze], the critical tissue P concentration reported to support adequate turfgrass quality and growth was 1.8 g kg1 (Liu et al., 2008). Although the use of tissue testing as a nutrient management tool is not new, its application to bahiagrass P fertilization has recently been incorporated in bahiagrass fertil ization programs in Florida. There is not much information on critical bahiagrass tissue P concentration in low input cow calf systems in Florida It is therefore imperative for field investigations to determine the minimum critical bahiagrass tissue P con centration below which yields are reduced. Unlike most nutrient management studies that focused on critical tissue P concentrations for optimum production, the current study is focused on determining the minimum critical bahiagrass tissue P concentration b elow which yields are reduced. Environmental I mpacts of P F ertilization on C ow calf P astures in Florida The assessment of potential water quality issues associated with P application to bahiagrass pastures growing on Florida S podosols represents the secon d component of this research project. O ff site movement of N and P from agricultural lands has been cited as a major contributor to eutrophication in Lake Okeechobee in S outh Florida ( Fonyo and Flaig, 1995; Flaig and Havens, 1995; Havens et al., 2003). Bog gess et al. (1995) estimated that about 51% of P inputs into the Lake Okeechobee basin are from P fertilizer; thus, P application to pasturelands in regions near the l ake watershed is strictly regulated. The Green Swamp and the Okeechobee Basin are designated as P -

PAGE 23

23 limited by legislation, and P based nutrient budgets are required irrespective of the nutrient source (Nair and Graetz, 2004). This means P applications are based on the crop P requirements to reduced soil P accumulation. Between 1973 and 1988, t he P concentration in the water of Lake Okeechobee increased by 25%. In the same time frame, cattle numbers in the counties surrounding the lake increased by more than 900 cows yr1 and were estimated to contribute about 40% of the P load to the lake (Bogg ess et al., 1997). The concentration of pelagic total P increased from near 50 g L1 in the mid 1970s to over 100 g L1 in the late 1990s (Havens et al., 2003). Due to the increased P concentration in the lake, the Florida Department of Environmental Pro tection established a total maximum daily load of 400 Mg of P per year to reach a target P concentration of 0.04 mg P L1 in the pelagic zone of the Lake by the year 2015 (Pant and Reddy, 2002). Boggess et al. (1995) estimated the total net P imports (sum of P imported minus P exported in agricultural products) into the Lake Okeechobee watershed from land use activities to be 2380 Mg P yr1. However, due to changes in land use and adoption of P best management practices in the Lakes watershed, net P import s ha ve decreased from 2380 to 1717 Mg P yr1 (Hiscock et al, 2003). Net P imports estimated by Hiscock et al.(2003) identified improved pastures as the largest P importer (558 Mg P yr1) compared to 545 Mg P yr1 for row crops, 458 Mg P yr1 for dairy, and 151 Mg P yr1 for residential areas. Comparing these estimates with earlier studies by Boggess et al ( 1995), there have been about 60 and 45% reductions in P imports from dairy and improved pastures, respectively. Over the same period,

PAGE 24

24 import estimates from row crops and residential areas have greatly increased, an excess of 200 and 650%, respectively. Managing P inputs in the Okeechobee watershed is complex due to current trends in land use. Changing land use patterns, especially urbanization and its potential impacts on water quality are important issues that should receive attention in discussions related to water quality programs in Florida. Currently, Florida ranks 4th in population behind California, Texas, and New York in the USA. It is estimated that Floridas population growth rate between 1990 and 2000 was 23.5% (U.S. Census Bureau, 2000). In Okeechobee County, the average population growth rate during the same period was ~20%. Meanwhile, the number of beef cattle in the state decreased from 1.06 to 0.9 million from 1990 to 2009, which represents a reduction of ~ 11% (USDA, National Agricultural Statistics Services, 2009). The number of beef cattle in Okeechobee County remained relatively constant during the same period (~ 70,000 to 80,000 head). This suggests that continuing urbanization of South Floridas sensitive ecosystems may pose serious challenges to the success of water quality programs in the state. Thus, comprehensive basinscale approaches that consider all the various P sources are necessar y for the development of long term alternative strategies to meet P loading goals. There are limited reported field data on the impacts of bahiagrass pastures used for cow calf operations in S outh Florida on water quality Rechcigl et al. (1992) exa mined the effects of reducing P application rates on bahiagrass forage production, soil P levels and water quality on an established grazed bahiagrass pasture that had not received fertilizer application for >6 yr in Okeechobee County. Total P concentrations in

PAGE 25

25 runoff water collected from five storm events in the study ranged from 0.4 to 0.8 mg L1 for the control, 1.0 to 1.9 mg L1 for treatments that had 12 kg P ha1 applied, and 1.2 to 3.5 mg L1 for plots that received 48 kg P ha1. Runoff P losses were reduced by 60% when P fertilization was reduced from 48 to 12 kg ha1. However, the reported runoff P concentrations for all the P rates (including the control) were above the total P standard of 0.35 mg P L1 for surface water established by the South Fl orida Water Management District (SFWMD, 1989). The authors presented no information on the initial soil test P levels, and there w ere no water quality data for the first year of the study. Study results may not be applicable to low input systems where fert ilizer P rates are lower than the amounts used in Rechcigl et al. (1992). Sigua et al. (2006) evaluated the changes in soil P levels and water quality of lakes associated with foragebased cattle operations in South Florida. Pastures used for the study were rhizoma peanut ( Arachis glabrata Benth.) mixed with bahiagrass and bermudagrass [ Cyn o don dactylon (L.) Pers. ] The pastures were fertilized annual ly with 17.2 kg P and 56 kg K ha1 from 1988 to 2002 with no N fertilizer application. The pastures were st ocked rotationally with rest periods of 2 to 4 wk based on forage mass during the spring, and harvested for hay in late summer. Between 1993 and 2002, total N (TN) and total P (TP) concentrations, and trophic state indices (TSI) of lakes associated with these pastures were evaluated. The TSI was calculated using water transparency, chlorophyll a content, TN, and TP concentrations. Water quality from the calculated TSI in Spring Lake and Lake Lindsey was good (within 30 46 TSI), but that of Lake Bystere was rated as only fair (55 TSI), based on Florida water quality standards used in the study (Brezonik, 1984). Total P concentrations for the lakes studied in 1993

PAGE 26

26 and 2002, were respectively, 0.08 and 0.34 mg L1 for Lake Bystere, 0.02 mg L1 in both years for Lake Lindsey and, 0.19 and 0.01 mg L1 for Spring Lake. Despite similar agriculture land use in Spring Lake and Bystere Lakes shores, water quality rating in the lat t er wa s only fair. Urbanization may be a contributory factor to water quality impairment in the Bystere Lake. Major land uses in Lake Lindsey watershed from 1988 to 2000 were 38% forest, cropland and pastures 34%, wetlands 13% and urban 13%. Similarly, land uses in Lake Bsytere shore within the study period were 49% cropland and pasture, fores t 18%, wetlands 11% and urban area 22%, while in Spring Lake S hore the dominant land uses were 46% cropland and pastureland, 34% forest, 9% wetland and 11% urban areas. Urban land use from 1988 to 2000 constitutes 22% of major land uses within 500 m of the shore of Lake Bystere, compared to 13 and 11% urban land use within the shores of Lake Lindsey and Spring Lake, respectively (Sigua et al, 2006). The study demonstrates that properly managed foragebased cattle operations may not pose serious environmental threats. Grass legume mixture systems investigated by Sigua et al. (2006) are not common in S outh Florida because of poor legume persistence and the high cost of establishment and management, and thus Siguas results are not directly applicable to the lo w input cow calf systems of South Florida Notwithstanding, the results indicated that other land uses had a greater impact on water quality than pasture P fertilization. Sigua et al. (2010a) recently evaluated P levels in plants, soils, surface water, and shallow groundwater in bahiagrass cow calf pastures in Florida. The pastures had been fertilized annually with 77 kg N, 10 kg P, and 37 kg K ha1 for 4 years and stocked rotationally with a 2to 4 wk rest interval. Soil samples were collected in the spr ing of

PAGE 27

27 2004 to 2006 from four landscape positions (top slope, middle, bottom and seep area) at 0 to 20 20 to 40, 40 to 60, and 60to 100cm depths. Forage mass and surface and groundwater P concentration were also determined for the four landscape positions. Results showed a net gain for soil P of 2.6 kg P ha1; however, management practices such as rotational stocking and proper fertilization had negligible effects on water quality In a similar study, Sigua et al. (2010b) reported total N concentr ation in shallow ground water beneath bahiagrass pastures ranged from 0.6 to 1.5 mg L1. Average NO3N concentrations ranged from 0.4 mg L1 to 0.9 mg L1 and average runoff NO3N concentration was 1.0 mg L1. Fertilization at a rate of 77 kg ha1, which i s the UF/IFAS fertilizer recommendations for established bahiagrass pastures in Florida, had no detrimental impacts on water quality. Cap e ce et al. (2007) evaluated the effects of stocking rate on water quality in cow calf pastures in S outh Florida. The s tudy was conducted for 5 yr (19982003) on summer pastures planted with bahiagrass, and winter semi native pastures (mixture of native grasses and bahiagrass). The summer pastures were fertilized with 50 kg N ha1 annually, but no P fertilizer was applied. The winter pastures had no fertilizer application. The experimental site for the summer pastures had received approximately 20 kg P ha1 yr1 for 20 to 30 yr until 1987, when P application was discontinued. The stocking rates used were control (no cattle) 1.3, 1.0, and 0.6 animal units ha1 for the summer pastures and 2.1, 1.6 and 0.9 animal units ha1 for the winter pastures. Over the 5yr study, cattle stocking rate had no effect on nutrient (TKN, TP, NO3N and NH4N) concentrations or loads in surface runoff. Pasture type significantly affected total P concentrations in runoff (0.63 for summer pastures and 0.15 mg L1 for winter pastures).

PAGE 28

28 The authors attributed the differences to the years of prior P fertilizer application on the summer pastures and li kely soil P accumulation. The initial Mehlich1 soil P concentration in the top 0 to 5 cm of the summer pasture was 40.1 mg kg1, wh ereas that of the winter pasture was 14.5 mg kg1. I nitial soil P concentration s at the top 5 to 10 cm soil depth was 13.7 and 7.1 mg kg1 for the summer and winter pastures, respectively. The historical P application rate on the summer pasture experimental site was approximately 63% greater than the current UF/IFAS suggested P application rate of 12.3 kg P ha1, which suggests that when P is over applied, pasture P fertilization can have negative long term effects on water quality. Limited research is available that truly represents a typic al cow calf system in Florida. Thus, a better understanding of the agronomic and environm ental impacts with regards to P fertilization of low input bahiagrass systems in S outh Florida is a critical need for the success of water quality programs in the state. Phosphorus Movement in Florida S podosols The S podosols of Florida have a unique hydr ological cycle, characterized by a high water table (~30 cm) located between the Bh and the Ap horizons during the summer rain y season and cm soil depth during the drier months (Soil Survey Staff, 1996). The sandy nature of the upper horizons provides limited nutrient holding capacity, and leaching or subsurface flow can be the predominant pathway for nutrient loss (Allen, 1988). Additionally, surface runoff can transport nutrients off site during storm events that result in significant flooding. The spodic layer has high sorptive capacity for P leached from the overlying layers (Yucan, 1966, Nair et al., 2004). However, wh en the P storage capacity is exceeded in P impacted soils, the spodic horizon may release P into solution (Nair et al, 2004, Villapando and Graetz, 2001, Nair

PAGE 29

29 et al., 1998). The impact of the Bh horizon on P liability in low input cow calf systems is not k nown. The Bh horizon is typically high in Al and organic matter and can a d sorb P leached from the overlying layers. However, P sorption by soils can be reversible with time (Rhue and Harris, 1999), and the release of some fractions of the P sorbed by the B h can occur. Under high water table conditions, for instance, P can potentially be released from the Bh and subsequently be carried to surface horizons or lost to the environment (Nair et al., 1999). We hypothesize that fluctuating water table conditions t ypically experienced in S outh Flori da may influence P release from the Bh horizon, availability to the plants, and potential for transport. Information about the P transport dynamics under fluctuating water table conditions on P supply to bahiagrass and loss to the environment could be very useful in developing meaningful P best management practices for forage systems in S outh Florida.

PAGE 30

30 CHAPTER 2 PREDICTING BAHIAGRAS S REPONSE TO PHOSPHO RUS FERTILIZATION US ING TISSUE TESTING I ntroduction Bahiagrass ( Paspal um notatum Fl gge) occupies approximately 2 million ha in Florida and represents the most widely used planted forage grass in the state (Adjei and Rechcigl, 2002). Bahiagrass pastures are grown predominantly on low fertility Spodosols that are highly susceptible to nutrient losses. Because of the extensive acreage of bahiagrass pastures and the potential environmental impacts associated with nutrient transport, bahigrass nutrient management has been a focus of state regulatory agencies and livestock producers. Phosphorus fertilization, in particular, has been an important concern from agronomic, economic, and environmental perspectives. Phosphorus fertilization can incr ease bahiagrass dry matter (DM) yield s, but can also be an important nonpoint pollution s ource that negatively affect s water quality. Phosphorus fertilization of bahiagrass and soil testing calibrations for P fertilizer recommendations continue to be important topics in Florida Previous studies o f bahiagrass response to P reveal conflicting results. Some studies have shown bahiagrass response to P application (McCaleb et al., 1966; Rechcigl et al., 1992), but others report no yield response (Ibrikci et al., 1992; Rechcigl et al., 1995; Burton et al., 1997; Ibrikci et al., 1999). The lack of yield response to P is often attributed to the ability of bahiagrass roots to obtain P from spodic horizons (Ibrikci et al., 1994; Ibrikci et al., 1999). The spodic horizon of Florida Spodosols typically contains significant amounts of available P compared to the Ap horizon. Rechcigl and Bottcher (1995) and Ibrikci et al. (1999) reported that P held in the Bh horizon can be accessed by established bahiagrass pastures. However, the hypothesis that the Bh horizon supplies P to

PAGE 31

31 bahiagrass pastures has not been fully tested. Important questions relative to the effect of fluctuating water table on P supply to bahiagrass also remain unanswered. Sigua et al. (2006) suggested that soil P concentrations can decrease significantly in bahiagrass pastures that are both grazed and hayed. Numerous producer observations in Florida indicated that bahiagrass yields were declining in areas that received no P fertilization for several years. Under these circumstances, P fertilization may improve overall pasture health and forag e production. The conflicting reports in the literature on bahiagrass response to P shows that routine soil testing alone is not a good indicator of P requirements of bahiagrass pastures growing on Spodosols Routine soil test ing focuses on the top 15 cm of the soil profile, which may not reflect the actual soil P available for bahiagrass uptake. The roots of established bahiagrass plants have can grow as deep as 90 cm (Rechcigl et al., 1992), which means that P in the Bh horizon can be assimilated by the plants. The challenge to agronomists and soil scientists has been to develop additional tools to better manage soil fertility in bahiagrass pastures and identify areas where P fertilization is required to maintain satisfactory forage production. Plant ti ssue analysis in combination with soil testing may be a useful diagnostic tool to predict bahiagrass nutrient requirements. Tissue analysis can be especially important for perennial pastures where fertilization may occur at any time during the growing seas on (Martin and Matocha, 1973). Although using plant nutrient analysis is not new, information about the critical tissue P levels for bahiagrass is limited. Based on a comprehensive review of the literature on P fertilization of bahiagrass, Silveira et al. (2007) suggested 1.5 g kg1 as the critical P level below which bahiagrass production is

PAGE 32

32 severely reduced. The critical tissue nutrient concentrations of several crops have been investigated. In Coastal bermudagrass [ Cynodon dactylon (L ) Pers .], the cri tical tissue N concentration was 18 to 22 g kg1 (Burton, 1954, Adams et al 1967), P ranged from 2.0 to 2.6 g kg1 (Fisher and Caldwell, 1959) and K ranged from 15 to17 g kg1 (Jordon et al., 1966). For P angola digit grass ( Digitaria eriant h a Ste ud) t he critical tissue N was 12 to 16 g kg1 (Harris et al., 1968), P ranged from 1.2 to 1.6 g kg1 ( L ittle et al., 1959) and K rang ed from 12 to 14 g kg1 (Plucknett and Fox, 1965). In a greenhouse study, Liu et al. (2008) reported 1.8 g kg1 as the critical minimum tissue P concentration of Floratam St. Augustinegrass [ Stenotaphrum secondatum (Walt.) Kuntze] to sustain adequate turfgrass quality and growth. There is little information on critical bahiagrass tissue P concentration in low input cow calf systems in Florida. Most studies evaluated bahiagrass response to N and P application rates much greater than rates currently used by forage producers in Florida. Unlike previous studies that focused on maximum production, the focus of this study was to identify the critical tissue P concentration that can sustain forage production with minimum N and P inputs. The objectives of the study were to i) identify the minimum critical bahiagrass tissue P concentration below which bahiagrass DM yield is impacted, ii) investigate the effect of N fertilization and plant maturity on tissue P concentrations, and iii) determine P fertilization effects on bahiagrass DMY and root mass production under different N fertility levels Materials and Methods Experiment Design an d Setup The study was conducted in the greenhouse at the Range Cattle Research and Education Center in Ona, Florida. Pots of size 10cm diameter by 41 cm height were

PAGE 33

33 filled with 12 kg of a commercial builders sand with negligible Mehlich 1 P (< 0.1 mg P kg1) concentrations. The pH of the sand was 5.9. Treatments consisted of three N rates (0, 50, and 100 kg N ha1), five P rates (0, 5, 10, 20, and 30 kg P ha1) and two clipping intervals (28 and 56 d). The P application rates corresponded to 0, 0.5, 1 2 and 3times the University of Florida Institute of Food and Agricultural Sciences (UF/IFAS) recommended P fertilizer application rates. Nitrogen and P were applied as ammonium nitrate and potassium phosphate, respectively. Fertilizer rates per experimental unit were calculated on a weight basis. Nitrogen and P fertilization was repeated after every 56 d. Treatments were applied in triplicate in a factorial arrangement using a completely randomized design. The experiment was conducted in the fall of 2006 and repeated in the summer of 2007. Greenhouse temperature was maintained at an average 25oC with a range of 18 to 33oC. Pots were watered daily, but no leaching was allowed to minimize nutrient losses. Bahiagrass was seeded in propagation trays and transplanted to pots 3 wk after germination. At the initiation of the study ( 9 Nov 2006 and 5 June 2007), each pot received 10 g of a micronutrient mix (F503G micromix) containing 24 g kg1 of B, and Cu, 144 g kg1 of Fe, 60 g kg1 of Mn, 0.6 g kg1 of Mo, and 56 g kg1 of Zn. Potassium (as KCl) was applied at 50 kg K2O ha1 at 56d intervals. Bahiagrass was clipped at 5cm stubble height at either 28or 56 d intervals for a total of 168 d each year. Harvested samples were weighed fresh, and subsamples were ovendried at 60oC for 48 hr for DM determination. Ovendried samples were ground to pass through a 1mm mesh screen in a Wiley mill (Model 4, Thomas Wiley, Laboratory Mill, Thomas Scientific, Swedesboro, NJ). H arvested forage samples were digested using t he total

PAGE 34

34 Kjeldahl procedure (McKenzie and Wallace, 1954) and N and P in the digest s were analyzed on a Seal AQ2 discrete auto analyzer (Seal Analytical Inc., Maquon WI, USA) using the total Kheldahl N and P procedures (USEPA, 1993). At termination of the study, root mass was determined. Roots were separated from the sand by gentle washing with water. Root samples were ovendried at 60oC for 48 h for root mass determination. Nitrogen and P uptake was calculated as the product of tissue N and P tissue conc entration and DM yield for each pot and harvest. Phosphorus recovery was defined as: P recovery (%) = P uptake (fertilized plot) P uptake (control plot) 100 ( 2 1) P applied where P uptake was calculated as the product of tissue P concentration and DM yield for each plot and harvest. Statistical Analyses Statistical analyses were performed using Proc Mixed (SAS, 1999). Treatments were the factorial arrangement of three N rates, five P rates, and two frequencies of harvest replicated thr ee times in a completely randomized design. Nitrogen and P rates, frequency of harvest, and year were considered fixed effects, with replicates and their interactions considered as random effects. The PDIFF test of the LSMEANS procedure was used to compare means. Treatments and their interactions were considered different when F test P values were < 0.05. The response of DMY as a function of tissue P concentration was determined with a non linear model using the NLIN procedure of the SAS software (SAS Instit ute, 1999) as: DMY = a + bX if X 2 2 ) DMY = P if X > C (2 3 )

PAGE 35

35 where DMY is bahiagrass dry m atter yield (g m2) and X is tissue P concentration (g kg1); a (intercept), b (linear coefficient), C (critical tissue P concentration, which occurs at the inter section of the linear response and the plateau lines), and P is plateau yield. The constants a b, C and P are obtained by fitting the model to the data Results and Discussion Bahiagrass Dry Matter Y ield Because year showed no effect on bahiagrass DM yields ( P = 0.7), data represent the average s of 2006 and 2007 DM yields The interaction between N and P rates significantly ( P = 0.0003) affected bahiagrass DM yield (Table 2 1). There was a linear increase in bahiagrass DM yield in response to P application when N was applied at 50 and 100 kg ha1. Compared to the zero control, P application at t he lowest rate (4.4 kg P ha1) increased bahiagrass DM yields by 33 and 61% for the treatments receiving 50 and 100 kg N ha1, respectively. No differences in DM yields among the various P rates were observed when no N was appli ed The results agree with d ata from Rechcigl and Bottcher (1995), who reported that bahiagrass responded quadratically to P application rates in a 2 yr field study with maximum yield being attained at the 24 kg P ha1. Bahiagrass DM yield averaged 10.6 and 14.1 Mg ha1 for the 0 and 24 kg ha1 P rate, respectively (~25% yield increase). Ibrikci et al (1992) reported bahiagrass yield response to the addition of 17 kg ha1 of P but no response to greater P ap plication rates Similarly, McCaleb et al. (1966) observed bahiagrass DM yield response to annual application of 24 kg P ha1. In a greenhouse study, Rodulfo and Blue (1970) observed that bahiagrass did not respond to P additions above 16 kg P ha1. Rhoads et al. (1997) indicated bahiagrass response to P application rates up to 84 kg ha1 when N was applied at rates ranging from 84 to

PAGE 36

36 336 kg ha1. However, unlike our results, the latter authors observed no interactions between N and P rates. Differences in bahiagrass DM yield response to P rates between our study and those in the l iterature are likely due to the contrasting initial soil P concentrations. For instance, the study by Rodulfo and Blue (1970) was carried out on a Leon fine sand that contained substantial levels of extractable P (48 mg NH4OAc extractable P kg1) Although our data indicated that bahiagrass may respond to P application rates as high as 30 kg P ha1, the results may not apply across a wide range of soils and environments and under field conditions Further studies on different soil types and environmental co nditions should be conducted on the field scale to determine adequate P application rates under the various N fertilization options currently recommended for established bahiagrass pastures in Florida. Numerous studies have shown that lower P rates (< 20 k g P2O5 ha1) may be sufficient to maintain adequate forage growth. Bahiagrass DM yield was affected by the interaction of N rate and harvest frequency (Fig ure 21). Interaction occurred because when N rates were zero, there was no effect of harvest interval, but when N was applied, yield was greater for the 56d frequency. Tissue P and N C oncentrations Tissue P concentration increased in response to P and N application rates (Table 2 2 ). There were linear and quadratic effects on tissue P concentration as a function of P rate when N was applied at the 0 and 50 kg ha1 rates. For the 100 kg ha1 N rate, tissue P increased linearly as P rates increased. Bahiagrass tissue P concentration ranged from 0.7 (for the 100 kg N ha1, zero P treatment) to 3.8 g kg1 ( for the 100 kg N ha1, 30 kg P ha1 treatment). Similar results were observed by Rodulfo and Blue (1970). In a greenhouse study, these authors observed t issue P concentrations ranging

PAGE 37

37 from 0.7 to 0.10 g kg1 when P was applied at 0 to 16 mg P kg1. Rechci gl and Bottcher (1995) also observed a quadratic response of tissue P concentration to P application. However, tissue P concentrations reported by Rechcigl and Bottcher (1995) w ere relatively greater (1.6 to 3.3 g kg1) than observed in our study probably due to the greater initial soil P concentrations in the previous work Rechcigl et al. (1992) also reported greater tissue P concentrations (~1.6 g kg1) for control plots (no P applied) which may explain the lack of bahiagrass yield response to applied P. I n absence of P fertilizer increasing N rates resulted in lower tissue P concentrations. In general, tissue P concentrations were above the proposed critical limit of 1.5 g kg1 (Silveira et al., 2007) when P was applied at the recommended UF/IFAS rate of 10 kg P ha1. There was a significant ( P < 0.0001) effect of harvest frequency on tissue P concentrations. Averaged across N and P rates, harvesting at 28 d resulted in tissue P concentrations of 2.1 g P kg1 compared to 1.6 g P kg1 for the 56d treat ment This response is probably due to the generally greater yield for the 56vs. the 28 d harvest frequency. Phosphorus additions had no effect on bahiagrass tissue N concentrations. Averaged across P rates, tissue N concentrations ranged from 18 to 20 g kg1. Lack of tissue N response to P application observed in this study is consistent with other reports in the literature (Rechcigl et al., 1992; Rechcigl and Bottcher, 1995; Rechcigl et al., 1 995).Unlike P application, the interaction between N rate a nd harvest interval affected tissue N concentrations. Regardless of the N rate, harvest interval of 56 d resulted in smaller tissue N concentrations than 28d harvest interval (Figure 2 2 ). Differences in tissue N as a function of harvest interval were greater when N was applied at 50 and 100 kg ha1 rates than for the zero N control.

PAGE 38

38 There was correlation between tissue P and bahiagrass DM yield (Figure 23). The curve that describes the relationship between tissue P and DM yield showed a broad transition zone between adequacy and deficiency. The coefficient of determination indicated that 86% of the yield variability could be explained by tissue P concentration. Based on the linear plateau model, there was a linear increase in DM yield when tissue P concentrations increased from 0.02 to 1.3 g kg1. Maximum yields of 478 g m2 were obtained when tissue P concentrations were 1.3 g kg1 ( 0.2) Tissue P co ncentrations above 1.3 g kg1 had no significant effect on bahiagrass DM yields. These results support t he selection of 1.5 g kg1 as the critical tissue P concentration for bahiagrass, as s uggested by Silveira et al. (2007) and suggest this level is a useful indicator to avoid bahiagrass DM yield reduction due to limiting levels of P. This is further subst antiated by previous studies that showed no yield response to P application in bahiagrass pastures where tissue P concentration was above 1.5 g kg1. Despite low soil test P concentration (< 10 mg Mehlich1 extractable P kg1 ), Sumner et al. (1991) report ed no bahiagrass DM yield response to P application; however initial tissue P concentration was 2.7 g k g1 Similarly, lack of bahiagrass response to P application was observed when tissue P concentration was 2.5 g kg1, even though soil test indicated low extractable P (Rechcigl et al., 1995). Phosphorus Uptake and R ecovery Phosphorus uptake was affected by the N x P interaction (Table 22 ). When N was a pplied at 50 and 100 kg ha1 rates, there were linear and quadratic effects of P rate on P uptake. Compared to control plots (no P applied), the highest P rate ( 30 kg P ha1 ) resulted in a 12and 15fold increase in P uptake in response to P application when N was applied at 50 and 100 kg ha1, respectively. No significant differences in P

PAGE 39

39 uptake in res ponse to P rate were observed when N was omitted. Burton et al. (1997) observed greater P uptake as N rates increased from 56 to 224 kg ha1. These authors suggested that when N is applied, bahiagrass can take up more P than required for optimum growth. St anley and Rhoads (2000) also observed an increase in P uptake when N was appl ied at rates up to 168 kg ha1. When N w as applied at the 100 kg ha1 rate, the 56d harvest interval resulted in greater P uptake comp ared to the 28d interval (Figure 24). Thi s is largely due to the greater bahiagrass DM yield observed for the treatments harvested at 56d intervals (Figure 2 1). Conversely, no effect of N rate and harvest interval was observed on P uptake when N was applied at rates < 100 kg ha1. Phosphorus recovery increased linearly as P rate increased when N was applied (Table 2 2). No effect of P rates on P recovery was observed when no N was applied. Compared to the 5 kg ha1 P rate, the 30 kg P ha1 resulted in an increase in P recovery of 113 and 122% for the 50 and 100 kg N ha1 rates, respectively. At any given P rate, increas ing N from 50 to 100 kg N ha1 showed no effect on P recovery with the exception of the treatments receiving 30 kg P ha1. Rechcigl and Bottcher (1995) observed a quadratic respons e in P recovery as a function of P application rate. After 1 yr of P application, P recoveries were 41% (contr ol), 63% (6 and 12 kg P ha1 application rates) and 31% (24 kg P ha1 application rate). Root Mass Accumulation Root mass was affected by N rate x P rate interaction. Phosphorus had no effect on root mass for the treatments receiving no N (Table 2 3). When N was applied at the 50 and 100 kg ha1 rates, root mass increased linearly as P rate increased. Similarly, Rodulfo and Blue (1970) also observ ed an increase in bahiagrass root mass in

PAGE 40

40 response to P application. At any given P rate, there were linear and quadratic effects of N on root mass. Summary and Conclusions The data indicated that the critical minimum tissue P concentration of bahiagrass was 1.3 g kg1 ( 0.2) Bahiagrass is not expected to respond to P fertilization when tissue P concentration is above this concentration. This result is consistent with the UF/IFAS recommend ation that suggests 1.5 g kg1 as the minimum critical tissue P co ncentration for bahiagrass pastures grown in Florida. The results also support the use of tissue analysis in combination with soil test as a management tool to guide proper P fertilization in bahiagrass pastures in Florida. Information relative to tissue P concentration can also allow unnecessary P applications to be avoided and minimize the negative impacts of P fertilization on water quality. Further investigation should be conducted at field scale to determine the minimum P application rates that will result in satisfactory forage production and yet protect the environment

PAGE 41

41 Table 21. Bahiagrass cumulative dry matter (DM) yield as affected by N and P application rates. Data represent averages of 2 yr ( fall 2006 and summer 2007). P rate N rate (kg ha 1 ) Po lynomial contrast 0 50 100 k g ha 1 g m 2 0 16 274 310 L** 5 23 364 498 L***, Q* 10 16 491 533 L***, Q* 20 23 561 606 L***, Q* 30 22 610 782 L***, Q** SE 1.4 38 38 Polynomial contrast NS L** L* L = linear, Q =quadratic ; NS = not significant ; SE = standard error; = P P 0.01;*** = P Table 22. Bahiagrass tissue P concentration, P uptake and P recovery as affected by N and P application rates. Date represent averages over the 2 yr (fall 2006 and summer 2007). P rate N rate (kg ha 1 ) 0 50 100 kg ha 1 Tissue P (g kg 1 ) 0 1.1 0.8 0.7 5 1.3 0.9 0.9 10 1.5 1.2 1.6 20 2.0 1.9 2.5 30 2.2 3.8 3.7 SE 0.2 0.2 0.2 Polynomial contrast L***, Q* L***, Q* L*** P uptake (g m 2 ) 0 0.02 0.12 0.11 5 0.02 0.36 0.38 10 0.02 0.36 0.36 20 0.05 0.81 0.87 30 0.05 1.4 1.7 SE 0.02 0.05 0.05 Polynomial contrast NS L***, Q* L***, Q* P recovery (%) 5 0.5 15 18 10 0.3 19 19 20 1.1 26 29 30 0.8 32 40 SE 1.8 1.9 2.0 Polynomial contrast NS L*** L** L = linear, Q = quadratic; NS = not significant; SE = standard error; = P P 0.01;*** = P

PAGE 42

42 Table 23. Effect of N and P application rates on bahiagrass root mass. Data represent averages of 2 yr (fall 2006 and summer 2007). P rate N rate (kg ha 1 ) Polynomial contrast 0 50 100 k g ha 1 g m 2 0 56 477 503 L** Q* 5 75 635 866 L***, Q* 10 60 862 802 L***, Q* 20 73 982 901 L***, Q* 30 66 948 982 L***, Q** SE 6 55 55 Polynomial contrast NS L** L* L = linear, Q =quadratic; NS = not significant; SE = standard error; = P P 0.01; *** = P

PAGE 43

43 Figure 21. Bahiagrass tissue P concentration as affected by harvest frequency. Data represent averages across N rate, P rate, a nd year Means followed by the same letter are not significantly different ( P > 0.05). Error bars represent one standard error Figure 22. Bahiagrass cumulativ e yield as affected by N rate and harvest frequency. Data represent averages across P rate and year. Mea ns followed by the same letter within N rate are not significantl y different ( P > 0.05). Error bars represent one standard error.

PAGE 44

44 Figure 23. Bahiagrass tissue N concentration as affected by N rate and harvest frequency Data represent averages across P rate and year. Means followed by the same letter within N rate are not significantly different ( P > 0.05). Error bars represent one standard error. Figure 24. Relationship between bahiagrass DM yield and tissue P concentration. Data represent average across N rates, P rates and year.

PAGE 45

45 Figure 25. Bahiagrass tiss ue P uptake as affected by N rate and harvest frequency Data represent averages across P rate and year. Means followed by the same letter within N rate are not significantly different ( P > 0.05). Error bars represent one standard error.

PAGE 46

46 CHAPTER 3 LOW RA TES OF PHOSPHORUS FERTILIZATION EFFECTS ON FORAGE PRODUCTION AND WATER QUALITY OF GRAZED BA HIAGRASS PASTURES I ntroduction Phosphorus fertilization of forage crops i s of importance to both environmental regulatory agencies and livestock producers in Flori da. Fertilized planted pastures and dairy operations are regarded as major contributors of water quality impairment of Lake Okeechobee in South Florida. Conflicting reports in the literature suggest that, despite significant research efforts on pasture fer tilization, bahiagrass ( Paspalum notatum Flgge) response to P fertilization is not fully understood. Research has shown that bahiagrass can respond to P application rates in the range of 6 to 24 kg P ha1 when Mehlich 1 soil test P is below 10 mg kg1 (Mc Caleb et al., 1966; Rechcigl et al., 1992). Rechcigl et al. (1992) conducted a 2yr field study on Immokalee fine sand and reported a quadratic response of bahiagrass yields to P application. Maximum yields were obtained at 24 kg P ha1. Ibrikci et al. (19 92), investigating P fertilization on a Myakka fine sand at Ona, FL with an initial Mehlich1 P concentration of 7 mg kg1, showed that bahiagrass responded only to P application rates of 1. In contrast, Rechcigl et al. (1995) showed that bahi agrass yields were not affected by the addition of P, and attributed the lack of response to the ability of bahiagrass roots to obtain P from the spodic horizon, which typically contains significant amounts of available P. Similarly, Ibrikci et al. (1999) showed bahiagrass did not respond to P application due to availability of P from the Bh horizon for plant uptake. The discrepancies in the literature regarding bahiagrass response to P fertilization are partly due to the inability of typical soil testing alone to accurately predict forage P requirements. Although most Florida sands exhibit very low P concentrations

PAGE 47

47 in the surface horizons, adequate concentrations are often found at deeper soil depths. However, because soil tests typically examine only the top 15 cm of the soil profile, the total available soil P pool present at deeper soil depths may not be reflected. Bahiagrass roots grow beyond 85 cm (Rechcigl et al., 1992) and can access P at deeper soil depths Therefore routine soil testing at 15 cm m ay not reflect P levels available for bahiagrass root uptake and may poorly predict bahiagrass P requirements. Both t issue and soil testing are recommended in the current University of Florida Institute of Food and Agricultural Sciences (UF/IFAS) P guideli nes for established pastures (Mylavarapu et al., 2007) in an attempt to resolve the problem. According to the revised fertilizer recommendations for established bahiagrass, tissue testing is recommended when soil test P is low/very low (< 10 and 1015 mg k g1, respectively). Phosphorus fertilizer is not recommended when Mehlich1 soil test P is medium (1630 mg kg1) or high (3160 mg kg1). When soil test P is low/very low and tissue P kg1, no P application is required. However, if soil test P is low/very low and tissue P < 1.5 g kg1, 12.3 kg P ha1 is recommended for the low and medium N options (Mylavarapu et al., 2007). Findings of the greenhouse studies (Chapter 2) confirm that tissue testing can be use to predict bahiagrass response to P f ertilization and that bahiagrass dry matter yield ( DMY ) is negatively impacted when tissue P concentration is less than 1.3 g kg1. There are no published grazing studies to validate the greenhouse findings and the new P fertilizer recommendations for esta blished grazed bahiagrass pastures in Florida. In addition, the impacts of P fertilization at the recommended rates on water quality have not been fully investigated. The objectives of this study were to (i) investigate the effects

PAGE 48

48 of the new UF/IFAS P fer tilization rates on forage production and nutritive value, and (ii) evaluate the impacts of new recommendations on soil test P and water quality in bahiagrass pastures grazed by beef heifers ( Bos sp ) M aterial and M ethods Experimental S etup The study was conducted on established bahiagrass pastures at the University of Florida Range Cattle Research and Education Center at Ona, FL (27o26N, 82o55W, 24 m alt.) on a Pomona fine sand (sandy, silicious, hyperthermic, Ultic Alaquods) from 22 May to 8 Oct 2007 and 2008. The soil series is a typical Florida S podosol consisting of a sandy Ap horizon, an eluted E horizon, and a spodic (Bh) horizon that is dark in color, high in Al (and sometimes Fe) and high in organic matter (OM) con centrations. The depth of the Ap horizon at the study site is 0 to 15 cm, and a very thick E horizon extends from 15 to 60 cm. The Bh horizon extends from 60 cm to 75 80cm. The Pom o na series is characterized by a high water table at depth of 25 cm for 1 to 3 mo during the rainy season and at depths of less than 100 cm for more than 6 mo during the lower rainfall months in an average year (Soil Survey Staff, 1984). Although the soils are poorly drained, permeability is rapid in the Ap and E horizons wh ereas the Bh and Btg horizons have moderate permeability. The Ap and E horizons have very low P sorption capacity and P added to the soil can easily leach into the Bh horizon which has high affinity for P due to presence of high Al and OM concentrations The soil is naturally acidic and generally poor in fertility (Soil S urvey Staff, 1986) Use in planted pasture s typically require s fertilizer inputs and liming. The i nitial soil pH was 5.1, and Mehlich1 extractable nutrient P, K, Mg, and Ca concentrations in the Ap (0 to 15 cm depth) horizon were 9.1 (P), 50 (K) 49 (Mg), and 413 (Ca) mg kg1. Prior to

PAGE 49

49 the initiation of the study, pastures were limed with 1 Mg ha1 of dolomitic lime (Aglime Sales, Inc., Lake Wales, FL) in March 2007 with the objective to increase the soil pH to 5.5. Tre atments consisted of three annual P fertilization rates (0, 5, and 10 kg P ha1) arranged in a completely randomized design with four replicates. Phosphorus fertilization rates corresponded to 0, onehalf of, and equal to the UF/IFAS P fertilizer applicati on rates for established grazed bahiagrass ( Mylavarapu et al 2007). Pastures also received 56 kg N ha1 and 47 kg K ha1, which corresponded to the University of Florida low N option typically used for grazed pastures. Fertilizers were applied to 1.2 ha established Pensacola bahiagrass pastures on 22 May 2007 and 2008. Twenty four crossbred (Angus sired on crossbred cows) yearling heifers (338 30 kg live weight; LW ) were randomly distributed in 12 pastures (2 heifers per pasture). Pastures were stocke d continuously using a fixed stocking rate. Forage Sampling Forage samples were collected just prior to the initiation of the study and every 14 d thereafter during a 2 yr study period for herbage mass (HM) determination. Double sampling was used to deter mine herbage mass (Santillan et al., 1979). The indirect measure was settling height of a 0.25m2 aluminum disk, and the direct measure involved hand clipping all herbage 2 cm above soil level. For calibration of the disk, t hree sites per experimental unit were chosen that represented the range of herbage mass present in the pastures. At each site, disk settling height was measured and the forage clipped. Clipped forage was dried at 60C for 48 h in a forcedair oven to constant weight for DM determination. Actual HM was regressed on disk height to develop a calibration equation using the PROC REG procedure of SAS (SAS Institute

PAGE 50

50 Inc., 1988 ). The regression equation was used to predict pasture HM using the average disk settling height from 20 measurements for each pasture at each sampling period. Because the pastures were stocked continuously, a cage technique was used to measure herbage accumulation rate (HAR). Three 1m2 cages were placed in the pasture at the initial sampling date. Placement sites were chos en where the disk settling height was the same (1cm) as that of the pasture average. Disk settling height was recorded at a specific site and the cage placed. After 14 d, the cage was removed and the new disk settling height recorded. Herbage accumulation was calculated as the change in estimated herbage mass during the 14 d. At the end of each 14d period, cages were moved to new locations on the pasture with a current average disk settl ing height. Handplucked forage samples were taken from each pasture prior to the initiation of the study and every 14 d thereafter for herbage crude protein (CP), total P, and in vitro digestible organic matter ( IVDOM ) determination. The objective was to represent the diet consumed by the grazing animal, and the technique involved removing the top 10 cm of herbage at approximately 30 sites randomly chosen in each experimental unit. Herbage was composited across sites, dried at 60C for 48 h in a forcedair oven to constant weight and ground in a Wiley mill (Model 4, Thomas Wiley Laboratory Mill, Thomas Scientific, and Swedesboro, NJ) to pass a 1mm stainless steel screen. Samples were analyzed for IVDOM using the twostage technique described by Tilley and Terry (1963) and modified by Moore and Mott (1974). Tissue N and P concentrations were determined using the total Kjedahl digestion procedure (McKenzie and Wallace, 1954). Digested samples were analyzed on a Seal AQ2 discrete auto

PAGE 51

51 analyzer (Seal Analytical Inc., Maquon WI, USA) using the standard EPA method 353.2 (USEPA, 1 993). Crude protein was estimated by multiplying N concentration by 6.25. Nitrogen and P uptake were calculated as the product of tissue N and P concentration and DMY for each plot and harvest. Soil Sampling Protocol and Analysis To account for possible spatial variation associated with soil test P distribution, each experimental unit was divided into three sections for soil sampling purposes: feeding, open, and shade areas. The sections were defined according to the distance from feeding/water and shad ed locat ions. The feeding area (Feeding ) corresponded to 30 m from the fence line and included feeding and water troughs. The shaded area (Shade) corresponded to the portion of the pasture occupied by trees. The open area (Open) corresponded to the area in t he pasture between the Feeding and Shade areas. Ten soil core samples were taken from the Ap (0to 15cm) and Bh (60to 75cm) horizons from each of the three sections of each pasture and combined into a composited sample. A total of 72 composite samples (12 pastures x 3 sections x 2 depths = 72) were collected at each sampling event. Soil samples were collected prior to the initiation of the study and after each growing season (November) in 2007 and 2008. Soil samples were air dried, crushed and sieved through a 2mm stainless steel screen and analyzed for Mehlich1 (M1) soil test P (Mehlich, 1953) and water extractable P (Self Davis et al., 2000). Phosphorus concentrations were measured colorimetrically on a Seal AQ2 discrete analyzer (Seal Analytical Inc., Maquon, WI, USA) using EPA method 365.1 (USEPA, 1993). Soil pH was determined using 1:2

PAGE 52

52 soil/deionized water (m/v basis) with a glass electrode (My lavarapu and Kennelley, 2002). Water Quality Monitoring Leachate P was monitored with two ceramic suct ion cup lysimeters (Soil Moisture Equipment Corp. Santa Barbara, CA,) installed in the center of each pasture at 60and 90 cm depths. The 60and 90cm lysimeter depths were intended to monitor water quality at and below the spodic horizon, respectively. Leachate samples were collected after rainfall events > 10 mm. In 2007, a total of eight leachate events were sampled from June to October wh ereas in 2008 six leachate events were sampled Samples were collected within 2 to 24 h of rainfall events (dependi ng on the soil moisture conditions) using a hand vacuum pump (~60 kPa) and stored at 4oC until analysis. Leachate samples were filtered (0.45P concentration on a Seal AQ2 discrete auto analyzer (Seal Analytical Inc., Maquon WI, USA) using standard E PA Method 365.1 (USEPA, 1993). Statistical Analysis Response variables were herbage mass, accumulation rate, and nutritive value, and concentrations of plant tissue P, soil P and leachate P. Data were analyzed using PROC MIXED of SAS (SAS Institute Inc., 1996) with P fertilization levels (main plot), year (subplot), and month as fixed effects. Year was considered fixed because year effects and interactions with year were of interest. Replicate and its interactions were considered random effects. Months were analyzed as repeated measures and means were compared using PDIFF (SAS Institute, 1996). Treatments were considered different when P P > 0.05). The means reported are least squares means.

PAGE 53

53 Results and Discussion Climatological Data Total rainfall recorded in 2007 (992 mm) and 2008 (1124 mm) was below the 67yr average (1363 mm) (Table 3 1). In general, seasonal rainfall patterns were similar among years and reflecte d longterm trends with higher rainfall amounts in May through October. However, rainfall was greater in 2008 than 2007, especially during the months of May through October, which correspond to the growing season in South Florida (Table 3 1). Greater prec ipitation in 2008 especially in June and August when the amount of precipitation exceeded the 67yr average, may have favored bahiagrass growth compared to 2007. Average minimum and maximum temperatures were similar in 2007 and 2008. Herbage Mass and Accum ulation Rate T here was no effect of P fertilization rate on HM and HAR. Average HM and HAR across treatments were 4.4 Mg ha1 and 43 kg ha1 d1, respectively. There was a year x month interaction on HM and HAR (Table 3 2). The interaction occurred becaus e HM increased from May to September but subsequently decreased in October 2007. In 2008, HM increased from May to August and decreased in September. No difference in HM between September and October 2008 was observed. The increase in HM from spring to sum mer is expected because of more favorable rainfall and temperature conditions for forage growth in summer (Table 3 1). Conversely, bahiagrass HM and HAR tends to decrease in the fall because of shorter day length (Sinclair et al., 2003). Greater HM in 2008 compared to 2007 was attributed to greater rainfall in 2008 (Table 3 1). Herbage accumulation rate was similar from June to September in 2007, with a subsequent decrease in September to October. In 2008, HAR was similar from June to

PAGE 54

54 August, followed by a sharp decrease in September, and an increase in October. Limited rainfall in September 2008 likely caused the reduction in HAR (Table 3 2). Annual HAR was generally greater in 2008 than in 2007. Herbage Nutritive Value There was no effect of P fertilizati on rates on CP and IVDOM concentrations. Average CP and IVDOM concentrations across treatments were 94 and 500 g kg1, respectively. There was a year x month interaction on CP, IVDOM, and TKP concentrations. In 2007, CP increased from May to July, and then decreased from July to October (Table 3 2). In 2008, CP concentrations increased from August to September. The greatest CP concentration in September 2008 was due to the decreased HAR during this month. The increase in CP concentrations from May to July i n both years was likely in response to the N fertilization, which occurred in May each year. Forage CP concentrations below 62 g kg1 are considered deficient in meeting the protein requirements of the grazing animal (Minson, 1990). Hence, the monthly CP concentrations observed in the study were above the minimum level to maintain the performance of grazing animals. In vitro digestible organic matter concentrations were similar from May to July, but they decreased from July to October 2007. In 2008, IVDOM i ncreased from May to June and decreased from June to October. The greater IVDOM concentrations in the beginning of the growing season were also likely related to the spring N fertilization. The decrease in IVDOM after July occurred because of the reductio n in new tissue growth and appearance of reproductive stems. Plant Tissue P Concentration Tissue P concentration increased linearly with increasing P application rate (Table 33 ). Average tissue P concentration during the 2yr study was 1.9 g kg1 for th e

PAGE 55

55 control and 2.2 g kg1 for the treatments that received 10 kg P ha1. There was significant year x month interaction on tissue P concentration. In 2007, tissue P concentration increased from 1.8 g kg1 in May to 2.1 g kg1 in July. Similarly, in 2008, ti ssue P concentration was 1.5 and 2.4 g kg1 for the same months (Table 32). The reported tissue P concentrations in the study are consistent with previous findings in South Florida. Adjei et al. (2006) reported tissue P concentrations ranging from 1.5 to 3.0 g kg1 in bahiagrass plots receiving 0 to 20 kg P ha1 harvest1. Rechcigl et al. (1992) also reported tissue P concentrations ranging from 1.6 g kg1 for control plots to 3.3 g kg1 for treatments receiving 48 kg P ha1. In general, tissue P concentrations for all treatments were above the reported critical tissue P concentration of 1.5 g kg1 for established bahiagrass pastures in Florida (Silveira et al., 2007; Mylavarapu et al., 2007). The recommended dietary P requirements by the National Research Council (1996) for beef cattle are 1.7 and 2.3 g P kg 1 for growing and lactating cows, respectively. The pastures in the current study will therefore supply adequate P in the diet of grazing animals. Despite low soil test P values in the Ap horizon, tiss ue P levels, especially in the control treatments (1.9 g kg1), indicate that bahiagrass obtained sufficient P from subsurface horizons. The root system of fully established bahiagrass has been shown to be active and growing at depths greater than 80 cm (R echcigl et al., 1992), although the majority of the roots were found within the top 15 cm of the soil (Rechcigl et al., 1992; Ibrikci et al., 1999). Therefore the depth ( volume) of soil available for potential P assimilation by bahiagrass roots extends bey ond the top 15 cm that is routinely used in soil testing programs. In the current study, soil test P in the Ap horizon (top 15 cm) was

PAGE 56

56 9.1 mg kg1 (~16 kg ha1) which is low according to the Mehlich soil test interpretations i n Florida (Mylavarapu et al., 2007). However, the soil test P in the Bh horizon was 28 mg kg1 (~84 kg P ha1) about five fold greater than the amount of P in the Ap horizon and may serve as a significan t source of P for bahiagrass whose roots can grown into the Bh horizon. Therefore i f the P levels in the entire profile are considered (both surface and subsoil), the soil will have an adequate supply of P to maintain bahiagrass forage production. This possibly explains the lack of bahiagrass HM and HAR response to P fertilizer in the current study although the surface soil test values were low. The results are consistent with a 4 yr study conducted on an established Pensacola bahiagrass pasture on similar soils where bahiagrass DMY showed no response to P fertilization as high as 53 kg P ha1 (Rechcigl et al., 1995). The results from this study imply that routine soil testing considering P levels in the top 15 cm poorly reflects the amount of soil and P available for potential uptake by established bahiagrass pastures with well developed rooting system. Unlike soil test calibration curves which relate soil extracted nutrients to plant response, plant tissue analysis can give a direct indication o f P availability to the crop. Tissue analysis has been widely used as a diagnostic tool for assessing nutrient requirement of many crops (Ulrich, 1952; Smith, 1962; Tyner, 1946; Liu et al., 2008; Mallarino and Higashi, 2009). Although the concept of tissue testing as a nutrient management tool is not new, its application to bahiagrass P fertiliz ation has only recently been incorporated in bahiagrass fertilization programs in Florida. According to the new recommendations, when tissue P concentration is above the critical limit of 1.5 g kg1, P fertilization is not recommended even if the soil test is low or very low in P

PAGE 57

57 (Mylavarapu et al., 2007). Therefore based on the new UF/IFAS fertilizer recommendation for bahiagrass pastures, the soils at the study site were probably not P deficient and no P fertilizer would be recommended. Soil test P C once ntrations Phosphorus fertilization had no effect on M 1 P concentrations in the Ap ( P = 0.85) and Bh ( P = 0.17) horizons during the 2 yr study. However, s oiltest P concentrations in the Ap ( P = 0.0001) and Bh ( P = 0.02) horizons were affected by the samp ling location in the pasture. Before treatment application in 2007 and 2008, soil test P concentrations in the Ap horizon were approximately threefold greater in the feeding area than the open and shade areas (Table 3 4). Based on the current M 1 soil tes t interpretations for agronomic crops in Florida (Mylavarapu et al., 2007), soil P concentrations in the Ap during both years were very low for the open and shade areas, but medium for the feeding area. Similarly, M 1 P concentrations in the Bh horizon wer e at least 1.5fold greater in the feeding area than the open and shade areas of the pasture. Sampling location also affected water extractable P (WEP) values in the Ap ( P = 0.0001) and Bh ( P = 0.01) horizons (Table 3 4). Water extractable P concentrations were generally greater in the feeding area than the open and shade areas. The WEP concentrations in the feeding area were about 50 and 40% greater than WEP concentration is the open and shade areas, respectively The average WEP concentration in the Bh horizon was 4.1 and 1.9 mg kg1 in the feeding and open areas, respectively. The high soil test P levels observed in the feeding area compared to shading and open areas in the pasture is likely due to cattle spending more time in feeding and shaded regions of the pasture compared to the open areas. Grazing cattle recycle

PAGE 58

58 nutrients (N, K and P) back to the pasture via excreta (Dubeux et al., 2009) Generally, about 70 to 90% of the N, P and K ingested are recovered in the excreta and urine (Haynes and Will iams, 1993) and are important factors in nutrient cycling in grazing lands. However, the uneven distribution of nutrient s by grazing cattle creates nutrient accumulation in shade, watering and feeding locations in pasture (Dubeux et al., 2009). These nutr ient enriched areas may lose nutrient s to the environment. Although nutrient contribution of grazing cattle can be significant, the impact of grazing cattle on P accumulation was not significant in the current study. The soil test P levels of surface and s ubsoil after 2 yr of the study w ere similar to the initial P concentrations (Table 3 4). The results are consistent with the findings of Capece et al. (2007) who reported that stocking rates ranging from 0.8 to 1.7 animal unit s ha1 had no effect on soil P accumulation and runoff P concentration in bahiagrass pastures. Applying P at the low rates used in the study did not increase residual soil P concentrations. SoilP concentrations in the Ap and Bh horizons following the study were similar to the initial soilP concentrations, which suggests no soil P accumulation during the 2yr study. Leachate P and NO3N Concentration There was a year x P level x sampling date interaction on leachateP concentrations ( P = 0.002). There was no difference ( P = 0.4) bet ween leachateP concentrations at 60 and 90 cm during the experimental period. In 2007, leachateP concentrations among treatments were similar during the first five sampling events ( Figure 31 ). However, after 25 August, there was a gradual increase in le achateP concentrations in treatments that received 5 kg P ha1 compared to the control and 10 kg P ha1 treatments. Average leachateP concentrations on 1 July were similar for all treatments (0.02 mg L1) but increased to 0.11 mg L1 on 25 September for the 5 kg P

PAGE 59

59 ha1 treatment (a nearly six fold increase). Similarly, in 2008, the 5 kg P ha1 treatment had greater leachateP concentrations in September compared to the other treatments. LeachateP concentrations were not different between the control and the 10 kg P ha1 treatments over the 2yr study period. Lower leachateP concentrations in the 10 kg P ha1 treatments compared to the 5 kg P ha1 suggest P held in the sub soil rather than fertilizer P might have accounted for the differences among the tre atments. Some of the P held in the Bh can be released and transported to the surface during high water table conditions ( Chapter 4 and 5) hence the reported differences in leachateP concentrations among the treatments is likely due to release of P from t he Bh horizon. There was a month x year effect on NO3N concentrations (Figure 32 ). In 2007, leachate NO3N concentrations were similar across sampling dates, ranging from 0.05 mg L1 in early July to 0.04 mg L1 in October. In 2008, however, the averag e NO3N concentration was 0.03 mg L1 in May, increased to 0.08 mg L1 in September, and decreased again in October. The greater rainfall amounts in 2008 might have caused the relatively higher leachate NO3N concentrations in 2008 compared to 2007. Increased levels of P fertilization were associated with decreases in leachate NO3N concentrations ( P = 0.05) (Table 3 3). For the 2yr study, the average NO3N concentration was 0.06 mg L1 for the control and 0.04 mg L1 for the treatments receiving 10 kg P ha1. The NO3N concentrations observed in the study were below the maximum of 10 mg L1 allowable in drinking water. Summary and C onclusions The P fertilization levels used in this study had no effects on bahiagrass HM and HAR but increased tissueP conce ntration and P uptake. Tissue P concentrations of all treatments and across sampling periods were above the recommended critical tissue P

PAGE 60

60 concentration for bahiagrass pastures. Applying P fertilizer to bahiagrass had no effect on soil P concentrations in t he Ap and Bh horizons over the experimental period. However, the position of feeding troughs open space and shaded regions in the pasture affected soil test P concentrations LeachateP concentrations were not affected by P application rates Leachate NO3N concentrations i n all treatments were below the minimum acceptable threshold for drinking water quality standards. The results of this study support the revised UF/IFAS fertilizer recommendations that if tissue P concentration is above the critical limi t of 1.5 g kg1, P fertilization is not recommended even if surface soil test values are low or very low in P.

PAGE 61

61 Table 31 Monthly maximum and minimum temperatures and rainfall at the Range Cattle Research and Education Center (RCREC), Ona in 2007 and 2008. Month Temperature Rainfall 2007 2008 Min Max Min Max 2007 2008 67 yr average -----------------------o C --------------------------------------------mm -------------------Jan. 13 23 14 22 38 23 54 Feb. 12 22 16 23 16 39 66 M ar. 17 25 17 24 51 57 79 Apr. 18 28 20 27 41 8 62 May 22 30 22 31 10 70 94 Jun. 23 30 25 31 203 249 221 Jul. 26 32 24 31 152 195 212 Aug. 26 32 26 31 208 250 211 Sep. 25 31 25 30 169 141 186 Oct. 24 29 18 28 52 42 78 Nov. 19 26 16 25 2 19 49 Dec. 15 24 14 22 52 31 52 Total 992 1124 1363 Source: Seller s, 2006.

PAGE 62

62 Table 3 2. Year x month interaction effect on herbage mass, accumulation rate, crude protein (CP), in vitro digestible organic matter (IVDOM), and tissue P co ncentration (TKP) on bahiagrass pastures with three P fertilization levels. Month Treatment / Year May June July Aug Sept Oct SE Herbage mass ---------Mg ha 1 ---------2007 2.1 d 2.4 cd 2.6 c 3.7 b 4.2 a 3.4 b 0.2 2008 4.5 d 5.2 c 6.2 b 7.1 a 5.9 b 6.1 b 0.2 P value < 0.01 < 0.01 < 0.01 < 0.01 < 0.01 < 0.01 Herbage mass accumulation rate ---------kg ha 1 d 1 ---------2007 -36 a 41 a 43 a 38 a 11 b 6 2008 -53 a 64 a 65 a 29 b 57 a 6 P value 0.01 < 0.01 < 0.01 0.20 < 0.01 Herbage CP ---------g kg 1 ---------2007 106 b 119 a 115 a 97 c 80 d 71 e 2 2008 79 c 111 a 100 a 80 c 87 b 70 d 2 P value <0.01 <0.01 <0.01 <0.01 <0.01 0.75 Herbage IVDOM ---------g kg 1 ---------2007 555 a 560 a 554 a 501 b 445 c 426 d 9 2008 520 b 553 a 509 b 448 d 491 c 457 d 9 P value <0.01 0.43 <0.01 <0.01 <0.01 < 0.01 Herbage tissue P ---------g kg 1 ---------2007 1.8 b 2.0 a 2.1 a 2.0 a 1.9 a 1.7 b 0.08 2008 1.5 d 2.2 a 2.4 a 1.8 c 2.0 b 1.7 c 0.08 P value <0.01 0.16 <0.01 0.02 0.26 0 .47 Monthly means within a year were compared using PDIFF (SAS Institute, 1996). Means not followed by the same letter are different ( P >0.05) P value of year effect within month

PAGE 63

63 Table 33. Tissue P concentration and leachate nitrate c oncentration as affected by P application rate Data represent averages across year and month. P rate Tissue P NO 3 N kg ha 1 g kg 1 mg L 1 0 1.9 b 0.06 a 5 2.1 ab 0.05 b 10 2.2 a 0.04 c SE 0.06 0.003 Means within a column followed by same letter( s) are not different ( P < 0.05). Data for NO3N represent average across year, depth, and sampling date. SE standard error Table 3 4. Effect of feeding, open, and shaded regions on soil extractable P concentration before and after treatment applicati on. Mehlich 1 P Water extractable P Position Initial After 2 years Initial After 2 years mg kg 1 Ap horizon (0 15 cm) Feeding 16.8 a 14.8 a 10.9 a 7.6 a Open 4.8 b 3.1 b 6.1 b 3.8 b Shaded 5.8 b 5.9 b 6.1 b 4.5 b SE 2.7 2.1 1.5 0.9 Bh horizon (60 75 cm) Feeding 42 a 43 a 3.5 a 4.1 a Open 26 ab 28 ab 1.5 a 1.9 b Shaded 15 b 18 b 1.3 a 2.4 b SE 11 8.2 1.1 0.7 Means followed by same letter(s) in a column are not significantly different at P Data represent av erages across P rates.

PAGE 64

64 Figure 31. Leachate P concentration at various sampling dates as affected by P application rate in (a) 2007 and (b) 2008. Data represent averages across sampling depth since sampling depth was not significant ( P = 0.4).

PAGE 65

65 Figure 32 Leachate NO3N concentrations as a function of sampling date in (a) 2007, and (b) 2008. Data represent averages across P rates and sampling depth.

PAGE 66

66 CHAPTER 4 AGRONOMIC AND ENVIRO NMENTAL IMPACTS OF PHOSPHORUS FERTILIZATION OF LOW INPUT BAHIAGRASS PASTURES Introduction Bahiagrass ( Paspalum notatum Flgge) is the most widely planted perennial forage grass in Florida, occupying about 2 million ha in the state (Muchovej and Mullahey, 2000). Bahiagrass pastures are predominantly grown on S podosols and are well adapted to the sandy soils of Florida. The grass tolerates low soil fertility conditions, low pH, and intermittent wet conditions and produces reasonable forage yields in droughty conditions. Most foragebased cow calf ( Bos sp.) syste ms in Florida rely on bahiagrass pasture as the major source of energy and protein for most of the growing season, therefore adequate production of bahiagrass forage is critical for the success of the cow calf industry in the state. Approximately 80% of c ow calf production in Florida is concentrated in environmentally sensitive areas in Central and South Florida (FDCAS, 1998), where P transport from agricultural activities represents a serious threat to water quality (Reddy et al., 1999). The primary cow calf production region in the state, the Lake Okeechobee watershed, faces significant problems associated with nonpoint source pollution of surface waters by agricultural P. Although dairy and beef operations have been suggested as the major contributors of P to Lake Okeechobee watershed (Burgoa et al., 1991), potential P losses from low input cow calf systems are relatively smaller than other intensive agricultural operations such as confined animal feeding operations and intensive row crop farming. However, because of the large acreage occupied by grazinglands in Florida, P transport from beef ranching operations can potentially have major impacts on water quality (Allen, 1988). Therefore, effective P management

PAGE 67

67 strategies that balance productivity and env ironmental protection are required to support the continued success of the cow calf industry and water quality programs in the state. In recent years, concerns about the environmental impacts of P fertilization on bahiagrass pastures have prompted several revisions in the University of Florida bahiagrass fertilization recommendations (Hanlon et al., 2008). The current P fertilizer recommendation for established bahiagrass in Florida is based on both soil and plant tissue P testing (Mylavarapu et al., 2007).Phosphorus is recommended only when tissue P concentrations are below 1.5 g kg1 and soil test P levels are very low or low (< 15 mg kg1 Mehlich 1 P). However, there are limited field data to validate the revised P fertilization recommendations for esta blished bahiagrass pastures. Most previous studies on bahi agrass P fertilization focused on maximizing bahiagrass yields using relatively high rates of fertilizer P. Thus, despite significant literature on bahiagrass response to P fertilization, there is limited information on strategies that balance productivity with environmental implications. Field research is necessary to determine the minimum amount of P required to maintain bahiagrass yield and pasture persistence without adversely affecting water q uality. This is particularly important in Florida where the hydrology in combination with environmental conditions and soil characteristics accelerate nutrient transport of agricultural P to surface water. The objectives of the study were to (i) investigate bahiagrass response to reduced P fertilization rates and (ii) evaluate the potential effects of bahiagrass P fertilization on soiltest P concentrations and water quality.

PAGE 68

68 Materials and Methods Experiment Setup The experiment was conducted on an establ ished bahiagrass field (27o23 N, 81o56 W) on Smyrna sand (sandy, siliceous, hyperthermic Aeric Alaquods). Plot size was 12.2 x 12.2m, with a 3.1 m alley between plots. Treatments were a factorial combination of three N application rates (0, 56, and 112 kg N ha1 yr1) and four P rates (0, 5, 10, and 20 kg P ha1 yr1 ) arranged in a completely randomized design with three replications, for a total of 36 plots. The 56 and 112 kg N ha1 rates correspond to the recommended University of Florida Institute of Food and Agricultural Sciences (UF/ IFAS) low and medium bahiagrass N options, respectively (Mylavarapu et al., 2007). Phosphorus application rates correspond to 0, 0.5, 1 and 2times the UF/ IFAS P fertilizer application rate of 10 kg P ha1. Each plot also received a basal annual application of 47 kg K ha1 and 0.5 kg ha1 of micronutrient mix (F503G micromix) containing 24 g kg1 of B and Cu, 144 g kg1 of Fe, 60 g kg1 of Mn, 0.6 g kg1 of Mo, and 56 g kg1 of Zn. Plots received lime prior to the init iation of the study t o raise the soil pH to 5.5. The experiment was conducted for 2 yr (2007 and 2008), and treatments were applied in May of each year. Forage Dry Matter Yield, Crude Protein and Tissue P A nalysis The plots were harvested at 28d interval s from June to October each year to determine dry matter ( DMY ) tissue N and P concentrations, and N and P uptake. During each harvest, two 0.9x 6.1 m forage strips were harvested from each plot to a 7.5 cm stubble height using a forage harvester. The re maining herbage was mowed with a flail harvester at the same stubble height. Samples were weighed fresh, and subsamples weighed and oven dried at 60oC for 48 h for DMY determination. Dried

PAGE 69

69 samples were ground to pass a 1mm mesh screen in a Wiley mill. Cumulative DMY was calculated as the sum of individual harvests for each year. Tissue N and P were determined using th e Kjeldahl digestion procedure (McKenzie and Wallace, 1954). Briefly, 0.2 g of plant material and 2 g Kjeldahl digestion mixture were diges ted in 5 ml concentrated sulfuric acid at 365oC for 3 to 4 hr. Digested samples were diluted to 100 ml and analyzed on a Seal AQ2 discrete auto analyzer (Seal Analytical Inc., Maquon WI, USA) using the Kjeldahl N and P procedures (USEPA, 1993). Crude prot ein (CP) concentration was calculated as percent N multiplied by 6.25 (Stewart et al., 2007). Nitrogen and P uptake were calculated as the product of tissue N and P concentration and DMY for each plot and harvest. Soil Analysis Prior to treatment applicat ion, and at the end of each growing season, five composite soil core samples were taken from the Ap (015 cm), E (16 30 cm) and Bh (31 60 cm) horizons of each plot. The composited s oil samples were air dried, crushed, and sieved through a 2 mm stainless s teel screen and analyzed for Mehlich1 P (Mehlich, 1953) and WEP (Page et al., 1982). Mehlich1 extractable P concentration was determined by equilibrating 5 g of soil and 20 mL of Mehlich1 solution (0.0125 M H2SO4 and 0.05 M HCl) for 5 min on a reciprocating shaker and then filtering through a Whatman # 42 filter paper. Phosphorus concentrations were measured colorimetrically on a Seal AQ2 discrete analyzer (Seal Analytical Inc., Maquon WI, USA) using standard methods for P analysis (USEPA, 1993). Water e xtractable P was determined by equilibrating 2 g of air dried soil and 20 mL of DDI water. The suspension was shaken on an orbital shaker at a rate of 200 strikes min1 at room temperature for 1 h. After

PAGE 70

70 shaking, the samples were centrifuged at 3200 X g for 10 min, and the supernatant was 1 P. Water Quality Monitoring Water quality was monitored on plots receiving annual application of 56 kg N ha1 and 0, 5, or 10 kg P ha1. The N rate of 56 kg ha1 represents the UF/IFAS low N option which is the one used most commonly in cow calf systems in Florida. The plot s were isolated hydrologically by berms and ditches. Due to the relatively flat topography of the landscape, coupled with the sandy nature of the soil, leaching and subsurface runoff are expected to be the predominant pathways of nutrient losses. The berms and dense ground cover are expected to further reduce the risks of nutrient transport via surface runoff. Five suctioncup lysimeters ( Soil Moisture Equipment Corp. Santa Barbara, CA.) were installed in the center of each plot at 15, 30 60 90 and 150 cm depths. The 15and 30 cm lysimeters were located above the spodic horizon, whereas lysimeters at 60, 90 and 150cm were below the spodic horizon. Leachate samples were collected after rainfall events > 10 mm. A total of twelve leachate samples were collected per year in 2007 and 2008. Samples were collected within 2 to 24 h of rainfall (depending on the soil moistu re conditions) using a hand vacuum pump (~60 kPa) and stored at 4oC until analysis. Lea chate samples were filtered through a 0.45analyzed for orthoP concentration using a S eal AQ2 discrete auto analyzer (Seal Analytical Inc., Maquon WI, USA) using standard methods for P analysis (USEPA, 1993). Statistical Analysis Statistical analyses were performed using Proc Mixed (SAS, 1999). Nitrogen and P rates and year were considered fixed effects, with replicates and their interactions

PAGE 71

71 co nsidered as random effects. Year was considered a fixed effect because of the potential for carryover effects of treatments from Year 1 to Year 2. Year was included in the model as a subplot treatment in a split plot arrangement, with the treatment combinations being the main plots. The PDIFF test of the LSMEANS procedure and single degree of freedom orthogonal contrasts were used to compare means. Treatments and their interactions were considered different when F test P values were < 0.05. Interactions not discussed in the results and discussion sections were not significant ( P > 0.05). The means reported are least squares means. Results and Discussion Climatological Data Total rainfall recorded in 2007 and 2008 was below the 67yr average (Table 41 ). Rai nfall in 2008 was greater than 2007, especially during the months of May through October, which correspond to the growing season in South Florida. Minimum and maximum temperatures were relatively similar during the 2yr study. Bahiagrass Dry Matter Yield There was significant ( P = 0.03) interaction between P application and year on cumulative bahiagrass DMY (Table 4 2). Although P application had no effects on forage DMY in 2007, bahiagrass DMY increased linearly in response to increasing P application in 2008. Compared to the zero P control, bahiagrass DMY increased ~ 4, 10, and 19% when P was applied at rates of 5, 10, and 20 kg P ha1, respectively, in 2008. Results observed in 2008 are consistent with those reported by Rechcigl et al. (1992) who observed a 25% yield increase in bahiagrass plots grown on a Spodosol and receiving 24 kg P ha1 compared to control treatments. In that study, N application rates were similar to our study (120 kg N ha1) and the bahiagrass yield response was linear

PAGE 72

72 for P rates up to 24 kg ha1. However, Rhoads et al. (1997) showed that bahiagrass grown on an Ultisol in North Florida receiving 336 kg N ha1 respond ed to P application rates as high as 84 kg P ha1. There was also a linear response of bahiagrass DMY to N application rates in both years. Cumulative DMY in 2007 ranged from 4.9 to 8.4 Mg ha1 for the 0 and 112 kg N ha1 treatments, respectively, whereas in 2008 DMY was 5.1 and 9.9 Mg ha1 for the same treatments. More rainfall in 2008 likely caused greater DMY in that year, and it also may have increased response to P fertilizer, resulting in a linear DMY response to P applications. Although addition of P increased bahiagrass DMY in 2008, there w ere no significant differences between treatments that received 5 and 10 kg P ha1, suggesting that half the recommended UF/ IFAS P application rate can be used to maintain bahiagrass production in low input systems B ahiagrass response to P application can be variable depending on year and environmental conditions which may explain the apparent discrepancies in the literature relative to the effects of P fertilization on bahiagrass pastures in South Florida. For instance, whereas Rechcigl et al. (1995) showed that P application had no effect on bahiagrass DMY, other studies by the same researchers (Rechcigl et al., 1992; Rechcigl and Bottcher, 1995) using similar P application rates, showed that bahiagrass DMY increased quadratically to P application with maximum yield obtained at 24 kg P ha1. Ibrikci et al. (1992) showed that bahiagrass response to P may also depend on the application rate. Bahiagrass responded to P rates of 17 kg P ha1, beyond which there was no response to ad ditional P (up to 68 kg P ha1).

PAGE 73

73 The poor correlation between P application rates and bahiagrass DMY (R2 = 0.13 to 0.23) observed by Rechcigl and Bottcher (1995), indicates that other factors such as climate conditions may have affected bahiagrass response to fertilizer application. Ibrikci et al. (1999) observed no response to P in Year 1 of a 2yr s tudy, but P application increased bahiagrass DMY in Year 2. Although the authors did not explain the lack of response in Year 1, they showed that P application increased root length density which may have enhanced P uptake over time and, consequently, incr eased DMY in Year 2. Tissue P Concentration and P Uptake Tissue P concentration was affected by P application rate ( P = 0.002) and year ( P < 0.0001). Tissue P concentration increased linearly in response to P application rates from 0 to 20 kg P ha1(Tab le 4 3). Tissue P varied from 2.1 (no P added) to 2.5 g kg1 (20 kg P ha1). These values were well above the proposed critical tissue P concentration of 1.5 g kg1 (Silveira et al., 2007), which probably explains the lack of DMY response to P in 2007 and the relatively nominal response observed in 2008. The linear increase in tissue P concentration as a function of P application rate suggests luxury P consumption by bahiagrass. The higher tissue P concentrations, despite low Mehlich I soil P in the Ap hori zon, support the suggestion that bahiagrass growing on Spodosols accesses P from below the Ap (Ibrikci et al., 1994). Tissue P was greater in 2007 (2.5 g kg1) than in 2008 (2.2 g kg1). This response was likely due to the greater yields obtained in 2008 w hich resulted in greater dilution of tissue P. Unlike results reported in the greenhouse study (chapter 2) N rates had no effect ( P = 0.07) on tissue P concentrations in the field clipping study Differences were likely due to the much

PAGE 74

74 smaller range in N application rates used in field study (0 to 112 kg N ha1) compared to the greenhouse study (up to 300 kg N ha1). Tissue P concentrations observed in this study are consistent with previous research conducted on similar soils in South Florida. Adjei et al. (2006) reported tissue P concentrations ranging from 1.5 to 3.0 g kg1 in bahiagrass plots receiving 0 to 20 kg P ha1 harvest1. Similarly, Rechcigl et al. (1992) reported tissue P concentrations ranging from 1.6 g kg1 for unfertilized plots to 3.3 g kg1 for treatments receiving 48 kg P ha1. Bahiagrass P uptake responded linearly to P applicat ion rates (Table 4 3). The increase in P applicati on rates from 0 to 20 kg ha1 increased bahiagrass P uptake by ~ 36%. Similarly, N application increased bahi agrass P uptake linearly ( P < 0.0001). On average, N applications of 0, 56, and 112 kg ha1 resulted in P uptak e of 12, 17, and 20 kg P ha1. Newman et al. (2009a, b) studying bahiagrass P uptake when growing on a Spodosol also reported that N fertilization enhanced bahiagrass P uptake. These authors reported bahiagrass P uptake of 13 to 32 kg P ha1 in plots receiving N application rates of ~ 250 to 600 kg N ha1 yr1. Greater P uptake in those studies compared to current study was likely due to the differ ences in initial soil P concentrations (Mehlich1 P concentrations in the Ap horizon = 29 mg kg1 ) and greater N fertilization regimens. Crude Protein Concentration and N Uptake There was a year effect on bahiagrass CP concentration ( P < 0.0001), with gr eater average CP in 2007 (106 g kg1) than in 2008 (87 g kg1) Differences were likely due to variation in environmental conditions and the lower bahiagrass DMY in 2007. Crude protein concentration increased linearly with increasing N application but was unaffected by P application ( P = 0.9). Adjei et al. (2000) and Obour et al. (2009)

PAGE 75

75 reported similar lack of response of CP to P application. Averaged across P rates and year, CP concentration varied from 92 g kg1 for the control to 99 and 105 g kg1, resp ectively, for the treatments that received 56 and 112 kg N ha1. These CP values are consistent with values reported by others for warm season grasses in Florida (Adjei et al., 2000; Muchovej and Mullahey, 2000; Obour et al. 2009). Bahiagrass N uptake was affected by N rate ( P < 0.0001) and year ( P = 0.008). Bahiagrass N uptake varied from 74 to 151 kg N ha1 for the treatments receiving 0 and 112 kg N ha1, respectively. Nitrogen uptake was greater in 2007 (116 kg N ha1) than in 2008 (108 kg N ha1), ref lecting the greater tissue N concentrations in 2007. Phosphorus addition had no effect on bahiagrass N uptake ( P = 0.2). Soil Phosphorus Concentrations Phosphorus applications to bahiagrass had no effect on Mehlich1 soil P concentrations in the Ap, E, and Bh horizons. However, there was a year effect on soil test P concentrations (Table 44 ). After 2 yr of P application, Mehlich1 soil P concentration in the Ap horizon were lower than the initial conditions and were considered very low, according to the current soil test P interpretations for agronomic crops in Florida (Mylavarapu et al., 2007). Similarly, Mehlich1 P concentrations in the E and Bh horizons found at the end of 2007 and 2008 growing season were either smaller or similar to the initial conditions (Table 4 4). This suggests that P application at rates similar to P removal will likely have no impacts on soil test P concentrations. Regardless of treatment, P concentration in the Bh horizon was always greater than in the Ap and E horizons (Tabl e 4 4 ). The substantial amount of P held in the Bh horizon may be plant available (Ibrikci et al., 1994) and likely contributed to bahiagrass P uptake, masking treatment effects. There was no effect of P application on WEP

PAGE 76

76 concentration ( P = 0.75) at all d epths Water extractabl e P concentration averaged across treatments after 2yr of P fertilization were 3.6, 0.4, and 2.8 mg kg1 for the Ap, E, and Bh horizons, respectively. Numerous studies have shown that WEP is a reliable indicator of P loss potential (Kleinman et al., 2002; Ehlert et al., 2003). Results from the current study indicate that addition of relatively low rates of P fertilizer for 2yr did not increase the risks associated with P transport, even when P was applied up to 2fold the recommende d rate (20 kg P ha1 ). Extensive bahiagass roots, adequate ground cover, and removal of harvested forage (i.e., no grazing and associated P return in excreta) likely increased fertilizer utilization efficiency and P removal and minimized the potential for soil P accumulation. Leachate Phosphorus Concentration There was a year x P rate x sampling depth interaction for leachateP concentration (Table 4 5). In 2007, leachateP concentration measured in the lysimeters placed at 15 cm w as not different between the control and the 10 kg P ha1 treatment, but the concentration for the 10 kg P ha1 rate was greater than the treatments receiving 5 kg P ha1 (Table 4 5). For the lysimeters at a 30cm depth, the greatest P concentration of 0.9 mg L1 was recorded for the control treatment. In 2008, increased leachateP concentrations at 15cm depth were observed in the treatments receiving 10 kg P ha1. However, leachate P at the 30cm depth was similar for the control and the tr eatments receiving 10 kg P ha1, which suggests that limited P leaching from the upper soil layers occurred. Regardless of the P application rate, leachate P concentrations found in lysimeters at 60, 90 and 150cm depths were similar in both years. Thus, P application had no effect on leachate P concentration below the spodic horizon. In

PAGE 77

77 general, leachateP concentrations in the lysimeters below the spodic horizon (> 60 cm) were less than in the lysimeters installed at upper depths. This trend suggests there was limited P movement below the Bh horizon during the 2yr study. These results are also consistent with the Mehlich1 and WEP soil data that showed no effect of P fertilization on soil P concentration in the Bh horizon (Table 4 4). There was seasonal variation in leachat e P during the 2 yr study (Figure 41) but the interaction season x P rate was not significant ( P = 0.3). In 2007, leachateP concentrations in all five lysimeters were similar during the first four sampling events (Figure 4 1 a). However, after 4 July, there was a gradual increase in leachateP concentration measured in the upper two lysimeters (15 and 30 cm) but this did not occur in the lysimeters below the spodic horizon. In 2007, there was a large spike in leachateP concentration in the 15and 30cm lysimeters in September and October (Fig. 1a). For instance, leachateP concentration in the 15cm lysimeter increased from 0.03 mg L1 on June 4 to 0.76 mg L1 in September, representing a 25fold increase in P concentration compared to the initial sampling period. A similar trend was observed in 2008, with the greatest spike in leachateP concentrations in the upper lysimeters occurring in August and September (Figure 41 b). I n both years, the times of large spikes in soil solution P concentration in the 15and 30c m lysimeters coincided with periods of high rainfall and high water table conditions at the experimental site. The high water table might have contributed to P fluxes from the spodic horizon into the surface layers, thus increasing P concentration in the l ysimeters above the spodic layer. The fluctuating water table conditions experienced in Florida may affect P release from the spodic layer hence contributing

PAGE 78

78 significantly to P fluxes from the spodic layer (Pant et al., 2002). Further studies are needed t o evaluate the influence of fluctuating water table on P bioavailability to bahiagrass pastures growing on Spodosols in Florida and potential losses of P to the environment. Summary and Conclusions Treatments spanning a range of low levels of P fertilizat ion were targeted in this study because of the critical need to minimize P loss to surface water while maintaining vigorous bahiagrass stands. Across the range of N rates most often used on bahiagrass swards in Florida, increasing P application from 0 to 20 kg ha1 had no effect on DMY in one year and increased DMY by 19% in the second year. There was no effect of P application on residual soil P concentration. Leachate P was generally not affected by P application. The only exception occurred in 2008 where leachate P concentration at the 15cm depth was greater for the 10 kg P ha1 compared to the other treatments. However, leachate P at the 30cm depth was either lower or similar for the 10 kg P ha1 compared to the control in 2007 and 2008, respectively. Also the results of the study suggested that limited P leaching occurred at depths > 30 cm. T he study showed that lower P rates can maintain bahiagrass production with limited impacts to the water quality.

PAGE 79

79 Table 41. Monthly maximum and minimum temperatures and rainfall at the Range Cattle Research and Education Center (RCREC), Ona, FL in 2007 and 2008. Month Temperature Rainfall 2007 2008 Min Max Min Max 2007 2008 67 yr average -----------------------o C -------------------------------------------mm -------------------Jan. 13 23 14 22 38 23 54 Feb. 12 22 16 23 16 39 66 Mar. 17 25 17 24 51 57 79 Apr. 18 28 20 27 41 8 62 May 22 30 22 31 10 70 94 Jun. 23 30 25 31 203 249 221 Jul. 26 32 24 31 152 195 212 Aug. 26 32 26 31 208 250 21 1 Sep. 25 31 25 30 169 141 186 Oct. 24 29 18 28 52 42 78 Nov. 19 26 16 25 2 19 49 Dec. 15 24 14 22 52 31 52 Total 992 1124 1363 Source: Sellers, 2006. Table 42. Cumulative bahiagrass dry matter yields as affected by P application rates and year. Data represent average across three N rates and three replicates (n = 9) P rates Year 2007 2008 kg ha 1 Mg ha 1 0 6.5 6.9 5 7.0 7.2 10 6.6 7.6 20 6.8 8.2 SE 0.3 0.3 Polynomial con trast NS L** NS = not significant; L = linear; ** = P

PAGE 80

80 Table 43. Tissue P concentration and P uptake as aff ected by P application rates on bahiagrass. Data represent average across three N rates, 2 yr, and three replicates (n =18) P rate Tissue P P uptake kg ha 1 g kg 1 kg h a 1 0 2.1 13.9 5 2.3 15.7 10 2.4 17.3 20 2.5 18.9 SE 0.06 0.7 Polynomial contrast L*** L*** L= linear; *** P 0.0001; SE = standard error P uptake = tissue P concentration x DM yield Table 44. Mehlich1 soil P concentration at various depths as a ffected by year. Data represent averages across three N rates, four P rates and three replicates (n = 36). Depth Initial 2007 2008 SE P value mg kg 1 Ap horizon (0 15 cm) 4.0 b 6.3 a 3.0 c 0.4 < 0.0001 E horizon (16 30 cm) 2.3 a 0.9 b 1.0 b 0.3 0.0005 Bh horizon (31 60 cm ) 46 a 32 b 51 a 4.2 0.0001 Means within a soil depth followed by the same letter are not different using the LSMEASN/PDIFF procedure (P > 0.05). SE = standard error Table 4 5. LeachateP concentrations at the various depths as affected by year and P application rate. Data represent average across 12 sampling events for each year. Depth 2007 2008 P rate (kg ha 1 ) 0 10 20 0 10 20 C m mg L 1 15 0.4ab 0.07b 0.5a 0.2b 0.09b 0.5a 30 0.9a 0.05c 0.3b 0.3a 0.10b 0.4a 6 0 0.01a 0.07a 0.009a 0.005a 0.03a 0.03a 90 0.1a 0.02a 0.01a 0.009a 0.03a 0.11a 150 0.03a 0.008a 0.006a 0.009a 0.01a 0.02a SE 0.2 0.1 Means within a soil depth and year followed by the same letter (s) are not different using the LSMEASN/PDIFF procedure. SE = standard error

PAGE 81

81 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 4-Jun 11-Jun 18-Jun 25-Jun 2-Jul 9-Jul 16-Jul 23-Jul 30-Jul 6-Aug 13-Aug 20-Aug 27-Aug 3-Sep 10-Sep 17-Sep 24-Sep 1-Oct Leachate P concentration (mg L-1) 15 cm 30 cm 60 cm 90 cm 150 cm a. 2007 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 21-May 28-May 4-Jun 11-Jun 18-Jun 25-Jun 2-Jul 9-Jul 16-Jul 23-Jul 30-Jul 6-Aug 13-Aug 20-Aug 27-Aug 3-Sep 10-Sep 17-Sep 24-Sep Sampling date Leachate P concentration (mg L-1) b. 2008 Figure 41.Leachate P concentrations as affected by soil depth and sampling date in (a) 2007 and (b) 2008. Errors bars represent one standard error. Date represent average across P rates.

PAGE 82

82 CHAPTER 5 FLUCTUATING WATER TA BLE E FFECT ON PHOSPHORUS RELEASE AND AVAILABILITY FROM A FLORIDA SPODOSOL I ntroduction Spodosols are the dominant soil series in Florida, constituting over 3.4 million ha in the state (Collins, 2003). Typical Florida Spodosols are characterized by a surface Ap horizon (0 to 15 cm), followed by an elluviated E horizon ( 1 5 to 60 cm), and a spodic (Bh) horizon ( 3 0 to 120 cm). The thickness es of the E and Bh horizons are highly variable and distinguish the Spodosols soil series. The Ap and E horizons are highly perm eable while the Bh has moderate permeability (Soil Survey Staff, 1996). The S podosols of Florida have a unique hydrological cycle, characterized by a high water table (~30 cm) located between the Bh and the Ap horizons during the summer raining season and cm soil depth during the drier months (Soil Survey Staff, 1996). The sandy nature of the upper horizons provides limited nutrient holding capacity, which can favor nutrient losses via leaching or subsurface flow (Allen, 1988). The spodic horizon has high P sorbing capacity and can retain P leached from the overlying horizons (Yucan, 1966, Nair et al., 2004). Graetz et al. (1999) investigated P concentration and distribution in the various depths in Florida Spodosol s under dairy and beef cattle production. Mehlich1 P concentrations in the Ap horizon ranged from 12 to 60 mg kg1 for beef and dairy holding areas, respectively, and P concentrations in the Bh horizon were 29 and 57 mg kg1 for the same areas. Phosphorus concentration in the E horizon P concentration was 33 mg kg1 for dairy holding areas and 1.7 mg kg1 for beef pastures. Other investigators have reported greater Mehlich1 P concentrations in Bh horizons compared to Ap horizons on pastures in Florida (Newman et al., 2009; O bour et al., 2 009; Rechcigl et al., 1992). Using u ranium isotope technology, Zeilinski et

PAGE 83

83 al (2006) reported that residual soil P in the upper 30 cm of the soil profile of a planted pasture in South Florida is derived from P fertilizer sources. Hence the source of P in the Bh horizon is likely due to accumulation of previous P fertilizer leached from the Ap and E horizons. Even though Bh horizons often exhibit s relatively high affinity for P, the P sorption capacity can be exceeded in P impacted soils, and the spodic horizon may release P into solution (Nair et al, 2004, Villapando and Graetz, 2001, Nair et al., 1998). Even when the sorption capacity is not exceeded, P can be reversibl y desorbed with time (Rhue and Harris, 1999). Thus, release of some portion of the P so rbed by the Bh is possible and may increase the amount of P in the soil solution. Fluctuating water table conditions experience d in summer in Florida also have a significant effect on P dynamics in the Bh horizon. High water table depths and flooding can i ncrease the amount of P released into solution due to increases in labil ity of sorbed P (Terry et al., 1980; Martin et al., 1997; Wright et al., 2001). This phenomenon is often attributed to reduction and dissolution of Fe (III) phosphates and hydrolysis of Al phosphates (Ponnamperuma, 1972). Although Al does not undergo reduction, flooding causes increases in soil solution pH which affects Al hydrolysis with subsequent releases of P from Al P minerals (Lindsay, 1979). When acid soils are saturated, the pH increased to about 7 which affect Al hydrolysis and release of portions of P from Al P minerals. In a study on a Florida histosol, Terry et al. (1980) observed a 20fold increased in soilsolution P concentration when the water table depth was 35 cm compa red to when water table depth was 90 cm. Nair et al. (1999) showed that the P retention capacity of the spodic horizon of a Myakka sand (sandy, siliceous, hyperthermic Arenic

PAGE 84

84 Alaquods), Immokalee (sandy, siliceous, hyperthermic Aeric Alaquods) and Pomello (sandy, siliceous, hyperthermic Typic Haplonthods) were reduced by 75, 70, and 60%, respectively under high water table conditions. Similarly, the equilibrium P concentration (EPC O ) of these soils increased by approximately 30% when the soils were flooded (Nair et al., 1999). Thus, under high water table conditions the potential of P release and movement above the Bh horizon is expected to increase. Phosphorus released from the Bh under rising water table conditions can be carried to surface horizons increasing the amount of P available for plant uptake and potential loss to the water bodies Although high water table conditions can have a significant effect on P dynamics in soils, there are few field studies on the effects of fluctuating water table condit ions on P fluxes and availability in Spodosols in Florida. The purpose of this study was to evaluate the effects of fluctuating water table conditions experienced in the summer months in Florida, on (i) P release from the spodic horizon into the soil solution and (ii) soil P availability as determined by anion exchange membranes. Materials and Methods Experimental Design and Setup The study was conducted on an established bahiagrass field at the UF/IFAS Range Cattle Research Center (REC) in Ona (27o 26'N, 82 o 55'W) on Smyrna sand (sandy, siliceous, hyperthermic Aeric Alaquods) in the summer of 2007 and 2008. The soil series is a typical Florida Spodosol soil with a sandy Ap horizon, E horizon, and the spodic (Bh) horizon which is dark in color, high in Al ( 592 mg kg1 ) and Fe (60 mg kg1 ) content compared to the Ap and E horizons The depth of the Ap horizon at the study site was 0 to 15 cm, the E horizon depth was 15 to 30 cm and the Bh depth was 30 to

PAGE 85

85 60 cm. The Bh horizon occurred at 30 cm and the thickn ess of the spodic layer varied from 15 to 30 cm. The Smyrna series is characterized by fluctuating water table conditions, with the water table depth at less than 45 cm for 1 to 4 mo and at depths of 30 to 100 cm for more than 6 mo in an average year. Duri ng the summer months, the water table briefly rises above the soil surface (Soil Survey Staff, 1984) resulting in periodic inundation of the land surface during the rainy season (June through October). The experiment was conducted on established bahiagras s pasture that had not received P fertilization for 15yr prior Non fertilized P treatment plots were used in this study to avoid the confounding effects of P fertilizer additions on P fluxes in the soil. Each plot received annual application of N and K a t 56 and 47 kg ha 1 respectively; and 0.5 kg ha1 of micronutrient mix (F503G micromix) containing 24 g kg 1 of B, and Cu, 144 g kg 1 of Fe, 60 g kg1 of Mn, 0.6 g kg 1 of Mo, and 56 g kg 1 of Zn. Plot size w as 12.2 m x 12.2 m, with a 3.1m alley in between plots. Plots were hydrologically isolated by berms and ditches (0.6 m deep and 0.3 m wide) Due to the relatively flat topography of the landscape, coupled with the sandy nature of the soil and the berms leaching and subsurface runoff are expected to be the predominant pathways of nutrient losses.The large plot size, a grassed alley way and drainage ditches along the perimeter of plots reduced the risks of surface crosscontamination from adjacent plots. Measuring Soil Phosphorus Bioavailability Measuring soil P availability under high water conditions using traditional soil test methods (Mehlich, Oslen, Morgan, etc ) may be difficult since they are mostly done using dried soils. Besides, soil test does not accurately quantify available soil P because the amount and forms of P extracted depends on the strength of the chemicals used in the extraction procedure. Hence soil test P may include forms that are not available for

PAGE 86

86 plant uptake. Anion exchange membranes (AEM) have been used to measure soil nutrient availability directly in the field (Mangiafico and Guillard, 2006; Cooperband and Logan, 1994; Weih, 1998; Meason and Idol, 2008; Nyiraneza et al., 2009) or in laboratory settings (Cooperband et al., 1999; Qian et al., 1992; Shekiffu and Semoka, 2007; J onhson et al., 2005). The AEM technology mimics P uptake by plant roots and has often been used as P availability index (Amer et al., 1955; Abrams and Jarrell, 1992; Wright et al., 2001). The AEM acts as an exchanger with soil solids, competing for dissolv ed species in the soil solution (Cooperband and Logan, 1994). For instance, when AEMs initially saturated with Clions are buried in the soil, H2PO4 in soil solution is exchanged for Clions adsorbed on the surface of the AEM. The ability of the AEM to function properly as exchange sites will depend on exchange capacity of the membrane material, the length of time AEM are deployed in the field and the soil nutrient retention capacity ( Cooperband and Logan, 1994; Qian and Schoenau, 2002; Giblin et al., 1994). Soil nutrient availability measurements u sing AEM in the field can be done over varied time scales; one day (Shekifu and Semoka, 2006), bi weekly (Mangiafico and Guillard, 2006; ) monthly (Weih, 1998; Wright et al., 2001;) or longer periods (240 days ) (Ziadi et al., 1999). However, longer deployment of AEMs in the field has been shown to give lower estimates of nutrient availability compared to shorter residence periods (Giblin et al, 1994). Therefore for better estimates of P availability, Qian and S choenau (2002) suggested AEM should not be buried for more than 2 weeks in the field. Anion exchange membranes are reusable (Cooperband and Logan, 1994), and the same AEM can be deploy ed to the field several times over the growing season. This makes AEM su itable for field measurements where estimation of spatial and temporal changes

PAGE 87

87 in nutrient availability is paramount. The technique has been successfully used to measure seasonal patterns in nutrient availability over the growing season (Nyiraneza et al., 2009; Johnson et al., 2005; Giblin et al., 1994) In the current study, AEM were buried in the field to determin e in situ soil P bioavailability in the summer of 2008 following the method of Cooperband and Logan (1994). The AEM s used (product no. 55164, B .D.H; Poole, U.K) were polystyrene sheets with quaternary ammonium groups attached. The AEM were prepared by cutting into 2 x 6 cm strips, rinsing in deionized water and presaturated with Clusing 1 M NaCl solution for 24 hr. Nylon fishing net was sewn to each membrane and a colored flag attached to the other end for easy recovery in the field. After saturation with Cl-, the AEM were washed in deionized water and transported to the field in water containing bottles. The membranes were then inserted in the soil at 15 cm depth by opening a vertical slit with the aid of a hand trowel, and gently sliding the membranes into the slit using a pair of tweezers. The slits were then closed by firmly pressing the soil to ensure proper soil membrane contact. Two AEM s w ere deployed in each plot. The membranes were collected bi weekly and new membranes reinstalled at a different location on each plot. After collection, large aggregates adhering to the membrane were removed and the membranes return to the laboratory in deionized water filled plastic bottles. Individual AEM membranes were placed in 50mL centrifuge tubes and extracted wit h 15 mL 1 M NaCl solution by shaking in a reciprocating shaker for 1h (Cooperband and Logan, 1994). The extracts were analyzed for P calorimetrically on a Seal AQ2 discrete analyzer (Seal Analytical Inc., Maquon WI, USA ) using EPA

PAGE 88

88 method 365.1 (USEPA, 1993). Used AEM s were rinsed in deionized water soaked in 1 M NaCl, and reused. Soil Sampling, Analysis and Phosphorus Adsorption Desorption Study Prior to treatment application and after each growing season, five composite soil core samples were taken from depths of 0 to 15 cm (Ap horizon), 15 to 30 cm (E horizon), and 30 to 60 cm (Bh horizon) to determine soil P concentrations Additional soil samples were also taken at 15cm depth increments to 150 cm to determine the soil P profile at the end of the study period in 2008. All soil samples were air dried, crushed and sieved through a 2 mm stainless steel screen extracted using the Mehlich1 [standard] soil test (Mehlich, 1953). ExtractableP Fe and Al concentrations w ere determined by equilibrating 5 g of soil and 20 ml of Mehlich1 solution (0.0125 M H2SO4 and 0.05 M HCl) for 5 min on a reciprocating shaker and then filtering through What man # 42 filter paper. Phosphorus analysis of the filtrate was done c o lorimetrically using the acid molybdate method (Murphy and Riley, 1962). The Al and Fe concentrations were determined using atomic absorption spectroscopy. Phosphorus saturation ratio (P SR) as a measure of the sorption capacity of the soil was calculated according to E quation 1 (Nair et al., 2004). PSR = M1 P/ (M1 Al + M1 Fe) ( 5 1 ) where M1P, M1 Al and M1F e correspond to Mehlich1 extractable P, Al and Fe concentrations respectively, expressed in moles. Soil P storage capacity (SPSC), which is the amount of P a soil can a d sorb before exceeding a threshold soil equilibrium concentration, was also calculated as;

PAGE 89

89 SPSC = (0.15 PSR) x (M1 Al + M1 Fe) ( 5 2 ) where (M1 Al + M1 Fe) is proportional to P sorption capacity of a given soil (Nair and Harris, 2004). The chosen value of 0.15 is an approxi mate PSR value corresponding to a critical P solution concentration of 0.10 mg L 1 P for environmental quality impairment (Nair et al., 2004). A P adsorption and desorption studies were performed on soil samples collected from the 30 to 60 cm depth (Bh h orizon) using a batch equilibration technique (Belmont et al., 2009; Pant and Reddy, 2001; Reddy et al., 1998). Seven P concentrations (0, 0.02, 0.25, 0.5, 1, 5, 10 mg P L1) as KH2PO4 were prepared in 0.01M CaCl2 solution. Two grams of soil were weighed i n triplicate into a 50mL centrifuge tubes and 20 mL of solution containing the various P concentrations in 0.01M CaCl2 were added to obtain soil/solution ratio of 1:10. The tubes were equilibrated for 24h by shaking on a mechanical shaker at 25oC After 24 h equilibration, the samples were centrifuged at 8000 x g for 5 minutes and the supernatant filtered through a 0.45m membrane filter. The filtrate was analyzed for orthoP using the acid molybdate method (Murphy and Riley, 1962). Phosphorus not recovered in solution was assumed to be a b sorbed by the soil. A P desorption study followed immediately after the adsorption experiments. Soil residue from the adsorption study was equilibrated with P free 0.01M CaCl2 solutions. To ensure soil/solution ratio of 1:10, the centrifuge tubes containing residual soil from the adsorption study were reweighed and 0.01M CaCl2 solution added to attain the initial weight before filtration in the adsorption study. The tubes were then shaken on a mechanical shaker for 24h at 25oC The tubes were centrifuged at 8000 x g for 5 minutes and desorbed P measured as described in the adsorption study. The amount of

PAGE 90

90 P retained by the soil was assumed to be the difference in the amount of P adsorbed and P recovered in solution foll owing desorption study. Calculation of Sorption Parameters The total amount of P adsorbed by the soil was calculated as: S= S +So, ( 5 3 ) Where S = total P adsorbed by the soil (mg kg1) S = amount of added P adsorbed by the soil (mg kg1) So= initial P adsorbed by the soil (mg kg1) The initial adsorbed P (So) was determined by the method of least squares fit using the equation below and S values measured at low equilibrium concentrations (C): S = KC So ( 5 4) Where K is the linear adsorption coefficient (L kg1) and C is the solution P con centration after 24 h equilibration (mg L1). A plot of S vs C is linear with K as slope and So as intercept. The equilibrium P concentration (EPCO) which describes the P concentration in solution where net absorption and desorption equal zero (S= 0) can also be estimated from the above relationship. Hence by substituting the values of S and C in the above equation, EPC O = S o /K (Reddy et al., 1998; Belmont et al., 2009). The Langmuir adsorption equation was used to estimate P sorption maximum and sorption coefficient as: C/S = 1/(KSmax) + C/Smax (5 5 )

PAGE 91

91 Where K is the Langmuir sorption coefficient and Smax is the maximum sorption capacity of the sediment. Plotting C/S v s. C gives a linear equation with slope as 1/Smax and intercept equals 1/K x Smax. Monitoring Leachate P Concentration and P Release F rom the Spodic Layer Five suction cup lysimeters (referred to as lysimeters herein) were installed at 15, 30, 60 90 and 150cm depths on each plot. The lysimeters (Soil Moisture Equipment Corp. Santa Barbara, CA. USA) consisted of a PVC tube with an external diameter of 2.2 cm and a porous ceramic cup with a length of 7.5 cm mounted on one end. To install the lysimeters a metal rod of the same diameter as the lysimeter was hammered into the soil and lysimeters were inserted into the hole after removing the rod, and the hole sealed with moist soil. The procedure should minimize disturbance of the soil around the lysimeter and ensure good contact of the lysimeter with the soil. The shallowest two lysimeters (15and 30 cm depths) were located above the spodic (Bh) horizon, whereas the remaining lysimeters (60, 90, and 150 cm) were below the Bh horizon. Soil solution was collected after each rainfall event greater than 10 mm. Samples were collected within 2 to 24 h of each event (depending on the soil moisture conditions) using a hand vacuum pump (~60 kPa) and stored at 4oC until they were analyzed for orthoP on a Seal AQ2 discrete analyzer (Seal Analytical Inc., Maquon WI, USA ) using EPA method 365.1 (USEPA, 1993). A pressure transducer was installed at the center of the experimental site to monitor changes in water table depth over the growing season. Statistical Analysis Statistical analyses were performed using Proc Mixed (SAS, 1999). Year sampling date and depth were considered fixed effect, with replicates and their

PAGE 92

92 interactions considered as random effects. The PDIFF test of the LSMEANS procedure and single degree o f freedom orthogonal contrasts were used to compare means. Treatments and their interactions were considered different when F test P values were < 0.05. Results and Discussions Soil Phosphorus Concentration and P S orption Study There was no effect of year ( P = 0.16) or year x depth interaction ( P = 0.15) on Mehlich 1 soil P concentration. However, sampling depth had a significant ( P < 0.0001) effect on soil test P concentration. Averaged across sampling dates (year), P concentration was 3.3 mg kg1 in the Ap horizon, 2.0 mg kg1 in the E horizon and 30.8 mg kg1 in the Bh horizon. Similar to P, Mehlich1 Al concentration was not affected by year ( P = 0.96) or year x depth ( P = 0.8). Also year ( P = 0.59) and year x depth ( P = 0.92) had no effect on Fe concentration. In general, Al and Fe concentrations were greater in the Bh horizon compared to the Ap and E horizons (Table 5 1) Average Mehlich1 Al, Fe and P concentrations at various depths after 2 yr are shown in Table 5 1. The average soil P concentration in the Ap horizon (015 cm) was 3.2 mg kg1, 2.6 mg kg1 in the E horizon and 46.8 mg kg1 in the Bh horizon was. Below the Bh horizon, soil test P concentration s ranged from 16.1 mg kg1 in the 4 5 to 60 cm depth to 10.4 mg kg1 in the 6 0 to 90cm depth. In general, soil test P in the Bh was tenfold greater than in the Ap horizon and threefold greater than the layers below the Bh (Table 5 1). Similarly, Al and Fe concentrations in Bh horizon were 14and 2.5 fold respectively, greater than Al and Fe concentrations in the Ap horizon. The sum of the measured Al and Fe concentrations in a given depth increment represent the P sorption

PAGE 93

93 capacity of that horizon. Therefore the Bh horizon has greatest sorption capacity for P, and P leached from the Ap and E hor izons is expected to accumulate in the Bh. There were significant differences in SPSC values among horizons, with the Bh horizon having the greatest SPSC value (Table 5 1). The E horizon had the lowest SPSC value (3.2 mg kg1), reflecting the relatively small Al and Fe concentrations present compared to the other horizons (Table 5 1 ). The SPSC is used to estimate the capacity of the soil to retain additional applied P. A negative SPSC value means additional P input may result in P movement via surface runo ff or subsurface leaching into the environment. The SPSC values were positive at all depths, indicating the soils at the experimental site were not severely impacted by previous P additions. The reported SPSC values are indicative of the P sorption capacit ies of the various horizons. It has been reported that the Ap and E horizons of Spodosols in Florida have low P sorption capacities while the Bh horizon has high P sorption capacity (Yucan, 1966, Mansell et al., 1991; Nair et al., 2004; Nair et al., 1998), hence the Bh horizon can retain more P compared to the upper horizons. The higher values of SPSC for the Bh horizon means it can act as a sink for P leached from the overlying A and E horizons. However, due to the high water table conditions experienced during the summer months, it is possible that a fraction of the P held in the Bh can be desorbed. Once released into solution, the desorbed P can be transported to the upper horizons under high water table conditions. High water table depth has been reported to decrease the P retention capacity of the Bh horizon in Florida (Nair et al 1999), hence the soil at the study site has a great potential to releas e P held in the Bh due to fluctuating water table conditions experienced at the study site (Fig ure 51)

PAGE 94

94 The sorption isotherm obtained from the Bh horizon soils in the study fits a typical Langmuir curve (F igure 52). The equilibrium P concentration (EPCO) calculated using linear regression analysis of the linear portion of the isotherm (Fig ure 52) was 0.04 mg L1. Thus at soil solution P concentration of < 0.04 mg L1, the Bh horizon can serve as a P source and release soluble P into solution. For instance, when P concentrations of 0 and 0.02 mg L1 w ere added to the Bh soils in the sorption study, there was no P adsorption by the soils. Instead the soil released 0.04 and 0.06 mg L1 P, respectively into the solution (Table 5 2). However, when P concentrations exceeded the EPCO value, the Bh horizon acted as sink for the added P. In general, at higher P concentration (P concentration of 1 mg L1 or above), > 90% of the P added was adsorbed; at lower concentration (P concentration of 0.25 or 0.5 mg L1) about 80% of the added P was adsorbed. The desorption study showed that when P was added at 5 and 10 mg L1 the fraction of P desorbed after 24 hr equilibration was 8 and 9%, respectively of the amount of adsorbed P. However, about 24 and 13% of the adsorbed P were desorbed when 0.25 and 0.5 mg L1 P respectively was added. The P adsorption maximum for the Bh horizon according to the Langmuir isotherm is 156 mg P kg1. Although the Bh horizon has high P sorption capacity, the data from the desorption study showed that some of P held in Bh horizon can be released into solution and become plant available or transported to other horizons. Additionally, the P holding capacity of the Bh horizon of Spodosols are reportedly reduced und er high water table conditions (Nair et al., 1999), so the fluctuating water table conditions in the study site may result in ev en more P release from the Bh into solution.

PAGE 95

95 Rainfall and Water Table Fluctuations Rainfall distribution patterns at the study site were similar in 2007 and 2008 (Fig ure 51). Rainfall amounts were lower in May to the middle of July, increas ed after the m iddle of July to October and declin ed thereafter. W ater table depth mirrored the rainfall distribution pattern with rising water table depths occurring during periods of greater rainfall amounts in August and October. In 2007, the water table depth was at 125cm below the soil surface from May through June, increas ed to 60 cm in July and by 14 September, the water table was at the soil surface. Similarly in 2008, the water table was below 116 cm o n 1 May but by 18 July, the water table dept h was at the soil surface (Figure 51b). Although the water table depth followed similar patterns in the 2 yr, the water table was above 45 cm (spodic layer depth) for 122 d in 2008 compared to only 75 d in 2007. The difference is due to the relatively wetter conditions in 2008 compared to 2007 (Figure 5 1). The difference in the length of saturation period between 2007 and 2008 is important because it may affect P adsorption and desorption processes in the Bh horizon and subsequent P release and availability. Changes in Leachate P Concentration In the 2 yr, there w as significant seasonal variation in leachateP concentration with soil depth (Fig ure 53). Leachate P concentrations in lysimeters positioned above the Bh horizon (15 and 30 cm) increased with rising water table depth in the months of July through September. However, leachateP concentration in lysimeters positioned below the Bh horizon (60, 90 and 150 cm) remained relatively constant (0.02 0.01 mg L1) during the entire growing season in 2007 and 2008. In 2007, leachateP concentrations in all five lysimeters w ere similar during the early sampling events except 11 June when there was a spike in leachateP concentration in the 30cm lysimeter

PAGE 96

96 (Figure 5 3a). However, there was a spike in soil solution P concentration in the 15and 30cm lysimeters in September and October. LeachateP concentration in the 15cm lysimeter was 0.02 mg L1 on 4 June but increased to 0.98 mg L1 in September, approximately a 49fold increase in P concentration compared to the init ial sampling period. Similarly in 2008, spikes in soil solution P concentration in the upper lysimeters occurred in August and September ( Figure 53b). The period of spikes in leachateP concentration in the 15and 30cm lysimeters coincided with periods of high rainfall and high water table conditions in the experimental site. There was a quadratic relationship between water table depth and leachateP concentration in the 15and 30cm lysimeters with R2 of 0.89 and 0.98, respectively (F igure 54). This means that changes in water table depth accounted for approximately 94 and 99% of the changes in leachateP concentrations in the 15and 30cm lysimeters, respectively over the study period. The relationship between water table depth and leachateP conce ntration in lysimeters below the Bh horizon was not significant ( P > 0.05). The increase in water table depth and flooding has been reported to increase labile pools of P, thus increasing P availability (Wright et al., 2001; Martin et al., 1997). Nair et al. (1999) showed that the P retention capacity of the Bh horizon is reduced by 70% under high water table conditions. Hence the high water table conditions contributed to P fluxes from deeper horizons (Bh horizon) into the surface layers, and increased P c oncentration in the lysimeters above the spodic layer. Hence, fluctuating water table conditions experienced in Florida contribut e significantly to P fluxes from the spodic layer.

PAGE 97

97 Seasonal Variation in Soil P Availability In situ s oil P availability (as determined by AEMs) deceased in the early part of July but then spike d in August when the water table depth increased. Soil P availability increased from 3.2 g cm2 in June to 9.2 g cm2 in August (Figure 5 5) due to changes in water table depth. This re presents approximately 188% increase in P availability between June and August when the water table depth increased. This increase was likely due to the high water table conditions increasing P availability. The P availability data agree with the leachateP data which showed that P concentrations in the upper lysimeters (15 and 30 cm) increased with rising water table depth in August and September. Although the Smyrna series may not be the dominant Spodosol in Florida, the observations in the study should be applicable to Spodosols and the wide array of soils in different environments that are subjected to alternate wetting and drying conditions. Wright et al. (2001) in a mesocosms study using soils collected from an Inceptisol on a floodplain forest in Ge orgia USA showed that P availability measured using AEM increased with flooding compared to when the soils were drained. In a field study on Pineda sand (Loamy, siliceous, active, hyperthermic Arenic Glossaqualfs) in S outh Florida, Capace et al. (2007) indicated that soil P availability measured using AEM varied seasonally with greater spikes in July through August. Other investigators have observed increased P availability when soils are flooded (Young et al., 2001; Seng et al., 2006; Terry et al., 1980; S hekifu and Semoka, 2006; Surridge et al., 2007; Berryman et al., 2009). Based on the current study, we conclude that the high water table conditions experienced in Florida may affect P release and availability to bahiagrass pastures growing on Spodosols.

PAGE 98

98 S ummary and Conclusions In the 2 yr study P concentrations in lysimeters above the Bh horizon (15 and 30 cm) increased with increasing water table depth in the months of August and September. However, P concentrations in lysimeters below the Bh horizon (60, 90 and 150 cm) remained relatively constant (0.02 mg L1) during the entire growing season in 2007 and 2008. The period of greater spikes in leachate P concentration in the 15and 30cm lysimeters coincided with periods of high water table conditions. Similarly, soil P availability increased when the water table approached the surface horizons in August. Our results showed that rising water table depth increases P release from the Bh horizon which can be transported to the upper layers of the soil. Als o the high water table conditions experienced in the summer months in Florida caused upward flux of P from the Bh horizon which increases soil P bioavailability.

PAGE 99

99 Table 51. Average Mehlich1 soil Al Fe, and P concentrations at various depths in the controls at the end of the study in 2008. Soil depth P Al Fe SPSC cm mg kg 1 0 15 3.6 1.7 42 16 26 12 6 3 1 5 30 2.1 1.2 22 4 15 12 3 2 3 0 45 46.8 27.6 611 104 65 3 58 40 4 5 60 16.1 15.1 433 108 48 25 57 18 6 0 90 10.4 12.6 182 11 38 26 22 11 91 150 15.1 8.1 201 57 17 9 19 3 V alues are means and standard deviations. Data represent average of 3 replications. SPSC = soil phosphorus storage capacity Table 52. Phosphorus sorption characteristics o f the Bh horizon at the study site. P added P in solution P adsorbed P desorbed mg L 1 mg kg 1 0 0.033 0.33 0.4 0.02 0.031 0.11 0.6 0.25 0.036 2.1 0.5 0.5 0.045 4.5 0.6 1 0.044 9.4 0.6 5 0.225 47.7 3.8 10 0.934 90.7 8.2

PAGE 100

100 Figure 51. Rainfall distribution and water table depth in (a) 2007 and (b) 2008.

PAGE 101

101 Figure 52. Phosphorus sorption isotherm for the Bh horizon soil at the study site.

PAGE 102

102 Figure 53. LeachateP concentration in control plots as affec ted by soil depth and sampling date in (a) 2007 and (b) 2008. Data are average of three replicates. Error bars represent one standard error.

PAGE 103

103 Figure 54. Relationship between water table depth and leachate P concentration at the 15 and 30 cm lysimeters in the 2yr study Figure 55. In si tu soil P availability measured by anion exchange membranes in 2008. Data are averages of three replicates. Error bars represent one standard error.

PAGE 104

104 CHAPTER 6 PHOSPHORUS MASS BALANCE IN A TYPICAL FLO RIDA SPODOSO L I ntroduction Phosphorus balances and budgets are quantitative tools that are often used to represent P flows in an agricultural production system. Nutrient mass balance is defined as the differences between nutrient inputs and exports in a particular agr icultural operation system. Ideally, P inputs into farming systems should equal outputs. When inputs greatly exceed outputs from a given production system, the risk of nutrient losses to surface and groundwater increases For instance, imbalances in soil N and P and subsequent accumulation in soils have been cited as the major contribution to eutrophication of lakes, rivers, streams and coastal waters worldwide (Carpenter et al., 1998). Achieving a balance between nutrient inputs and outputs is the key to mitigating nutrient related environmental risks associated with agricultural productions systems. Proper P balance in cow calf pastures is imperative for sustainable forage production and to reduce potential P loss to water bodies. Numerous researchers in the USA and elsewhere have applied nutrient mass balance approaches to evaluate environmental performance of various agricultural production systems (Webb et al., 2004; Watson and Atkinson, 1999, Dou et al., 1998; Oenema et al., 2003; Gentry et al., 1998). Fertilizer and animal feeds are typically the major nutrient inputs in livestock operations. In low input cow calf system in Florida, inorganic N and P fertilizers are usually applied at relatively small rates compared to other agricultural operations. P hosphorus fertilizer is applied at rates of 30 kg P2O5 ha1 when Mehlich1 soil test P is low. However for established bahiagrass pasture, if tissue P concentration is > 1.5 g kg1, P fertilization is not required even when soil test P is low.

PAGE 105

105 The imports of P through feeding supplementary fortified molasses range from 0.36 to 1.65 kg P ha1 for low and high stocking rate pastures, respectively (Swain et al., 2007). Phosphorus is also imported with mineral salts at rates that ranged from 0.21 to 0.66 kg P h a1 for low and high stocking rates respectively (Swain et al., 2007). These low P inputs suggest most of the crop P uptake is supplied from soil P reserves indicating the significance of residual soil P for forage productivity in low input systems in Florida. There are three major approaches in computing P budgets; (i) soil surface balance (ii) farm gate balance and (iii) the soil system balance (Oenema et al., 2003; van Eerdt and Fong, 1998). The soil surface balance consider s the inputs of P in fertiliz er and organic residues entering the soil and P output from crop removal. In the farm gate budgets, P balance is calculated from the inventory of all P sources entering the farm and outputs of P leaving the farm gate. Both the soil surface and farm gate budgets ignore internal P fluxes and potential fate of P in the environment. The soil system approach considers P inputs and outputs together with P fluxes and changes in P stocks in the soil. Nutrient budgets can be calculated at the plot and farm scale (Koelsch, 2005; Swain et al., 2007), watershed or regional scale (Hiscok et al., 2003; David and Gentry, 2000; Spears et al; 2003; Gentry et al; 2009), national and continental scale (Stoorvogel et al., 1993) and even global scale (Liu et al., 2008). Greater variations in climate, soil type, farming systems, crop types, variability in fertilizer inputs by different farmers and topographical differences makes larger scale mass balance estimates less reliable. Field and farm scale estimates provide a better appreciation for the nutrient fluxes and balances in the soil. For instance, P budget estimated for beef cattle pastures in the

PAGE 106

106 Lake Okeechobee watershed in Florida showed a net P gain (import) of 3.04 kg ha1 yr1 (Hiscok et al., 2003). However, P budgets es timated on experimental pastures at a farm level showed a net P export (loss) ranging from 0.43 to 1.78 kg P ha1 for low (0.8 animal units ha1) and high (1.7 animal units ha1) stocking rate pastures respectively (Swain et al., 2007). The choice of a P balance method and scale depends on the purpose of the study. In any case, that the boundary conditions, inputs and outputs of P mass balance must be clearly defined for proper interpretation of results. Despite the extensive area of planted pastures used for livestock production in Florida, there are limited data on P mass balance for low input cow calf system. Phosphorus budgets on experimental bahiagrass pastures that were grazed in S outh Florida showed net P exports (loss) of 0.43 to 1.8 kg ha1 (Swain et al., 2007). The results suggest loss of P from the cow calf pastures via sale of calves from the farm, but give no indication of P fertility status of the production system. The available P stocks of the soil will determine P depletion in the soil and the ability of the soil to supply adequate P to meet forage crop requirements. In low input agriculture production systems, nutrient deficits are often offset by nutrients derived from soil reserves (Smaling and Braun., 1996; Ranger and Turpault, 1999). Hence incorporation of available soil P in the overall P budgets is especially important in cow calf pastures in Florida where P input s are very low. Rao et al. (1982) calculated total N and P balance within the top 15cm plow layer of a sandy soil in India. The N and P budgets were computed as the nutrient output in crop removal minus the sum s of soil N and P content. Over the 7yr study, there were negative P budgets of 54 kg ha1 for both the control treatments and when N was applied, but without P ferti lization. There was a net

PAGE 107

107 gain of 173 kg P ha1 for treatments that received complete N, P, and K fertilization. Other investigators have also reported P budgets where soil available P is considered as an input (Raupach et al., 2001; Prasad and Sinha, 1981; Sharma et al., 1987). Most bahiagrass pastures in Florida are grown on Spodosols. Soils typically exhibit A p and E horizons with limited P holding capacity Sub surface (Bh) horizon s often contains greater Al, and Fe concentrations and retain P Althoug h forage crops can access P from the Bh horizon (Rechcigl and Bottcher, 1995), soil test ing for established pastures focus only on P concentrations in the top 15 cm (Ap horizon). Because P concentration in the Bh horizon can be as much as threefold greater than in the Ap horizon (Graetz et al., 1999; Newman et al., 2009; O bour et al., 2009; Rechcigl et al., 1992), such soil test s may not represent the total P pool available for plant uptake. Phosphorus management strategies for perennial pastures should als o consider P stored in the Bh horizon. In our study, P mass balance was defined as total available P content in the soil after adjusting for input s from fertilizer, and atmospheric deposition, and outputs through harvested forage, leaching and runoff. T he P mass balance approach was used here as a tool to estimate the fluxes of P in typical Florida Spodosol. We hypothesized that the P held in the Bh horizon contributes to the net P supply in Spodosols. The objectives of this study were to determine P mas s balance on a typical Florida Spodosol and to evaluate the potential impacts of fluctuating water table on P flux from the spodic horizon to the overall plant P supply in low input pastures.

PAGE 108

108 Materials and Methods Study Site The study was conducted at UF/ IFAS Range Cattle Research Center (REC) in Ona (27o26'N, 82o55'W) on established bahiagrass field in the summer of 2007 and 2008. Treatments were three P rates (0, 5 and 10 kg P ha1) arranged in a completely randomized design with three replications. Each plot also received a basal annual application of at 56 kg N ha1 and 47 kg K ha1, plus 0.5 kg ha1 of micronutrient mix. Treatments were applied once in May of each year. Plots were 12.2 m x 12.2 m, with a 3.1 m alley in between plots. Plots were hydrologically isolated by berms and using ditches ( 0.6 m deep and 0.3 m wide) Due to the relatively flat topography of the landscape, coupled with the sandy nature of the soil, leaching and subsurface runoff were expected to be the predominant pathw ay for offs ite P loss Runoff samplers were also installed at the lowest elevation of each plot to collect any runoff generated during storm events. Crop P Uptake Forage was harvested at 28d intervals from June to October of each year. Dry matter yield, tissue P concentrations, and P uptake were determined at each harvest. During each harvest, two 0.9x 6.1 m forage strips were harvested from each plot to a 7.5 cm stubble height using a forage harvester with a 0.9m swath width. The remaining herbage was clean mow ed with a flai l harvester. Harvested samples were weighed fresh and sub samples oven dried at 60oC for 48 hr and weighed for DMY determination. Dried samples were ground to pass a 1mm mesh screen in a Wiley mill ( Model 4, Thomas Wiley Laboratory Mill, Tho mas Scientific, and Swedesboro, NJ ).

PAGE 109

109 Tissue P was determined using the total Kjeldahl digest procedure (McKenzie and Wallace, 1954). Briefly, 0.2 g of plant material and 2 g Kjeldahl digestion mixture were digested in 5 ml concentrated sulfuric acid at 365oC for 3 to 4 hr. Digested samples were diluted to 100 ml and analyzed on a Seal AQ2 discrete analyzer (Seal Analytical Inc., Maquon WI, USA ) using the standard Kjeldahl P procedure (USEPA, 1993). Phosphorus uptake was calculated as the product of tissue P concentration and DMY for each plot and harvest. Soil P Five composite d soil core samples were each taken from the Ap (015 cm), E (1 5 30 cm), and Bh (3 0 60 cm) horizons at the initiation of the study and at the end of each growing season (November 2006 and 2007). Soil samples were air dried, crushed, and sieved through a 2 mm stainless steel screen and analyzed for Mehlich1 soil test P (Mehlich, 1953). Phosphorus was analyzed colorimetrically on a Seal AQ2 discrete analyzer (Seal Analytical Inc., Maquon WI, USA ) using EPA method 365.1 (USEPA, 1993). Five undisturbed soil core samples were also collected from each horizon for bulk density determination. Water Quality Monitoring Five suction cup lysimeters ( Soil Moisture Equipment Corp. Santa Barbara, CA. USA ) were installed in the center of each plot at 15, 30 60 90 and 150cm depths. The 15 and 30 cm lysimeters were installed above the spodic horizon; lysimeters at 60, 90, and 150 cm were below the spodic horizon. Leachate samples were collect ed after rainfall event s > 10 mm. A total of 12 sampling events occurred in 2007 and 2008. Samples were collected within 2 to 24 h of rainfall ( depending on the soil moisture conditions) using a hand vacuum pump (~60 kPa) and stored at 4oC until analysis.

PAGE 110

110 Leachate samples were analyzed for orthoP concentration colorimetrically on a Seal AQ2 discrete analyzer (Seal Analytical Inc., Maquon WI, USA ) according to EPA method 365.1 (USEPA, 1993). A pressure transducer was installed at the center of the experimental site to monitor water table fluctuation over the 2yr study. Phosphorus Mass Balance Calculations The major P inputs and outputs considered in this study were: (i) mineral P fertilizer, (ii) plant uptake, and (iii) P leached below the rooting zone. The depth of 45 cm was chosen as the rooting depth because about 80 to 90% of bahiagrass roots are within the top 45 cm of the soil (Rechcigl et al., 1992). Additionally, the Bh horizon which has high affinity for P occurs within 45 cm, P leached beyond this depth is assumed to be lost to the groundwater. Although P transport via runoff was included in the P mass balance calculations runoff P was not expected to be significant because of the flat topography, sandy nature of the soils, and dense grass cover Similarly, atmospheric deposition of P in south Florida (0.4 to 0.7 kg P ha1 yr1) is negligible compared to other P inputs (Izuno et al., 1991; Ahn and James, 2001). Though P stored in the soil is not an input, it is significant component of P cycling in permanent pastures systems (Oyanarte et al., 1997) and soil test P is credited in the overall P budget of perennial grassland systems. The volume of water drainage below 45 cm was estimated from the water balance equation, D = P ET W S (Izuno, 1987) ( 6 1 ) where; D = drainage below 45 cm P = precipitation ET = evapotranspiration

PAGE 111

111 W S = change in water storage In typical Florida Spodosols the water table fluctuates between 120 cm and the soil surface T he change in water storage can be estimated using changes in water table depth because drainage through the soil recharges the shallow aquifer. Hence, the change in water storage ( W S ) = x Sy ( 6 2 ), where, is change in water table depth and Sy is the specific yield. Th e specific yield is the volume of water that a saturated soil will yield by gravity per unit area of soil per unit change in water table depth (Marshall and Holmes, 1988). It is eq uivalent to the drainage porosity (the amount of drainable pore space available for water flow under saturated conditions). The Sy used in the study was 0.24 which represent the average Sy value reported for sandy soils in Florida (Said et al., 2005). The soils at the study site were Smyrna sand (sandy, siliceous, hyperthermic Aeric Alaquods) which is a typical Florida Spodosol with the water table depth fluctuating between 120 cm and the soil surface, and temporary rising to the soil surface in the raini ng season (June through October). C hange in water table depth can be used to estimate the amount of drainage because the water table depth can only be reduced by drainage and evapotranspiration. Drainage was calculated daily and values summed to yield tot al drainage for each month. The monthly mass of P leached for each treatment was calculated by multiplying the average monthly soil solution P concentration measured in the 60 cm lysimeter by the total drainage for the month. The seasonal mass of P leached was computed as the sum of total mass es of monthly P leached. It was assumed that negligible leaching of P occurs when the water table was above 45 cm, hence, drainage volumes included only days when the water table was below 45 cm.

PAGE 112

112 Runoff losses were es timated for periods of high water table conditions when the water table was close or above the soil surface. Runoff volume was estimated as; R = ( P ET ) APS (Sands, 2001) ( 6 5 ) where; P and ET are the same as defined previously and APS = available pore space for water infiltration and R = runoff volume. ASP = depth of water table to the surface x Sy (Sands, 2001) ( 6 6 ) When the water table was above the soil surface, runoff volume equal to the difference between rainfall and evapotranspiration. Runoff P load was calculated by multiplying the volume of runoff by P concentration measured in runoff samplers. There were three significant runoff events in 2007 (9/14/07, 9/24/07 and 10/4/07) and four events in 2008 (7/18/08, 8/6/08, 8/17/08 and 10/2/08). Runoff P loads were summed to obtain total seasonal P lost through runoff. Daily precipitation and evaporation data were obtained from the Florida Aut omated Weather Network (FAWN) situated at the UF/IFAS Range Cattle REC, Ona, FL (27o26'N, 82o55'W ). The P mass balance was then calculated separately for soil depths of 15 cm and 45 cm as; Change in storage = (Deposition +fertilizer P + soil P) (P uptak e + P leached+ Runoff P) ( 6 7 ) Statistical Analysis Statistical analyses for ANOVA for all responses were done using PROC MIXED pro cedure of SAS (SAS Institute Inc., 1999). P hosphorus rates and year were considered as fixed effects and replicates and their interactions considered random effects. When year was significant, means were calculated separately for each year when year was not significant, means were averaged across year s. The LSMEANS

PAGE 113

113 procedure and associated PDIFF were used for mean comparisons. Interaction and treatments effects were considered significant when F test P values were <0.05. Results and Discussion Crop P Uptak e The application of P had no significant ( P = 0.2) effect on P uptake. Similarly, year ( P = 0.25) and P rate x year ( P = 0.8) had no effect on P uptake. Phosphorus uptake was 13.8 kg ha1 for the control, 16.8 for the 5 kg ha1 treatments and 16.2 kg ha1 for the 10 kg ha1 treatments The relatively high P uptake from the control plots ( compared to treatments that received fertilizer application) indicates the soil at the study site supplied adequate available P despite low Mehlich1 soil test P concentr ations in the top 15 cm (Ap horizon). Over the 2yr study, native soil P supplied more than 83% of the P requirement of the crop (P uptake of control as a percentage of P uptakes of the fertilized plots). Bahiagrass roots can access P held in the Bh horizon (Rechcigl et al., 1992) and possibly explains the lack of P uptake response to P fertilizer application in the current study. Changes in Soil Phosphorus Concentration Application of P fertilizer had no significant ( P = 0.45) effect on Mehlich1 soil P concentration at the various depths. However, there was significant year x depth interaction effect ( P = 0.01) on Mehlich1 soil P concentration .The initial Mehlich1 P concentrations were 3.1 mg kg1 in the Ap, 4.5 mg kg1 in the E and 36.1 mg kg1 in the in Bh horizons. After a year of treatment application, residual soil P in the Ap horizon increased to 5.3 mg kg1 but decreased to the initial levels by the end of study period in 2008 (Table 6 1). Mehlich 1 P concentrations in the E horizons in 2007 and 2008 were lower than the initial P levels (Table 6 1) Similarly Mehlich 1 P concentrations in the Bh

PAGE 114

114 at the end of the study period in 2008 were not significantly different from the initial levels, indicating that the reduced fertilizer P applications rat es used in the current study has no effects on residual P accumulation in the soil. Rainfall Distribution, Water Table Fluctuations and Drainage The overall annual rainfall distribution patterns were similar in 2007 and 2008 (Fig ure 6 1 ), with smaller ra infall amounts in May to the middle of July, peaking after middle of July to October and declining thereafter. Water table depth followed rainfall distribution patterns with rising water table depths coinciding with periods of greater precipitation in August and October. During the early part of the growing season in May 2007, the water table depth was at 120 cm below the soil surface, rising to 65 cm in July and by August and September, the water table was above the soil surface. Similarly in 2008, the wat er table was below 120 cm in May at 40 cm in July and at the soil surface in August through October. Although the water table depth followed similar patterns in both years the water table was above 45 cm (spodic layer) for 122 d in 2008 but only for 75 d in 2007. Also the water table rose earl ier in 2008 compared to 2007. I n 2007, the water table was above the spodic layer by 25 August in 2008, the water table was above the spodic by 1July ( Figure 61 b). Daily drainage below the rooting depth of 45 c m varied seasonally in both years Drainage was generally low in May representing the drier early part of the growing season than in July through October (usual wet period). In 2007, the l east drainage ( 3.3 mm) occurred on 5 May wh ereas the greatest drainage (84.4 mm ) occurred on 30 June following 88.1 mm of rainfall (Figure 62a). Similarly in 2008, the l east drainage (4.2 mm) occurred on 19 May wh ereas the greatest drainage ( 53.9 mm ) occurred on 30

PAGE 115

115 June following 58.7 mm of rainfall (Fig ure 62b). Cumulative drainage was greater in 2007 (493.3 mm) compared to 200 8 (387.3 mm) The difference is due to the water table rising above 45 cm earlier in 2008 than 2007 when the water table rose above the spodic layer in late August. Phosphorus Leaching and Runoff L osses Phosphorus application rate had no significant ( P = 0.5) effect on the total amount of P leached below the Bh horizon. Similarly, year ( P =0. 07) and the interaction of P rate x year ( P = 0.3) had no effect on the amount of P leached. During the 2yr study, average P mass leached below 45 cm was 0.08 kg ha1 for the control, 0.08 kg ha1 for the 5 kg ha1 treatment, and 0.09 kg ha1 for the 10 kg P ha1 treatments There are few studies that quantified P leached from grasslands and differences in s oil type, hydrology and environmental conditions complicates the comparisons of the results from the current study with those observed by others. Notwithstanding, the data are consistent with the findings of Izuno et al. (1991) who reported values of 0.43 kg P ha1 and 0.72 kg P ha1 as P leached from sugarcane and cabbage plots on Histosols in South Florida. Since the amount of P leached from treatments that received P applications were similar to the control, applying P at the low rates used in the current study may not pose threat to water quality R unoff P loads were not significant affected by P rates ( P = 0.1), but the interaction of P rate x year was significant ( P = 0.04). Runoff P loads in 2007 were 1.3 kg ha1 for the control, and 1.2 kg ha1 fo r the 10 kg ha1 treatment. The same treatments had runoff P loads of 1.3, and 2.5 kg ha1 in 2008, respectively (Table 6 2). The estimated runoff P loads from the study are consisted with previous studies in south Florida (Boggess et al., 1995; Cape ce et al.; 2007). Boggess et al (1995) estimated that

PAGE 116

116 the annual runoff P load for planted pastures in the Lake Okeechobee watershed in south Florida was 1.5 kg P ha1 yr1. Cap e ce et al. (2007) studying the effects of stocking rates on soil P accumulation and w ater quality on a cattle ranch in south Florida, reported runoff P loads ranging from 0.08 to 3.8 kg ha1 for control treatments (no cattle or fertilizer application). Similarly, Rechcigl et al. (1992) reported runoff P loads ranging from 0.9 kg ha1 for c ontrol plots, 1.1 kg ha1 when 12 kg ha1 P was applied, and 2.4 kg ha1 when 48 kg P ha1 was applied. Phosphorus Mass Balance Inorganic fertilizer was the main P input and plant uptake was the major output of P in our experimental site (Table 6 2). The amount of inorganic P deposited from the atmosphere ( ~6 % of total P input), P loss through runoff and leaching ( ~10 % of total P output) were relatively insignificant compared to plant uptake ( ~ 90% of total P output). In 2007, P mass balance estimated including soil P stored within the top 15 cm (Ap horizon) w as 10.8 kg ha1 for the control and 0.8 kg ha1 for the 10 kg ha1 treatments. The same treatments had P mass balance of 6.6 and 7.2 kg ha1 in 2008. The negative P mass balance suggests that regardl ess of P application rates, P removal from harvested forage exceeded P inputs through fertilizer, atmospheric deposition and the soil P held in the Ap horizon. It was therefore expected that crops would likely respond to P fertilization. Contrary to our ex pectations, there was no P uptake response to P fertilization indicating that P reserves in the subsoil probably contributed to bahiagrass P uptake. These findings are consistent with the soil test P data which showed that the reduced fertilizer P applicat ions rates used in the current study has no effects on residual P accumulation in the soil over the 2yr study.

PAGE 117

117 When P storage within 0 to 45 cm ( include Bh horizon) was included in the overall P budget, P mass balance was positive regardless of treatment and P application rate (Table 6 2). The P budget estimates using soil test P concentrations within 45 cm suggest the soil will supply adequate P for plant uptake. P erennial grasses has extensive and deep rooting systems (bahiagrass roots were found be low 75 cm at the study site), and the P held in the Bh horizon can be assimilated by bahiagrass roots ; thus P application may not be warranted. T he lack of bahiagrass P uptake response to P fertilization, suggest that P in the Bh horizon is available for bahia grass uptake and is a significant source of P to low input perennial pastures growing on Spodosols. N utrient management of bahiagrass pastures growing on Spodosols should therefore consider the P levels in the Bh horizon. Summary and Conclusions The appli cation of P had no significant effect on bahiagrass P uptake in the 2yr study. The amount of P loss through leaching and runoff were insignificant compared to P uptake in the overall mass balance computations. The P held in the Ap horizon had no effect on the overall P balance. However, when soil P held within the top 45 cm of the soil was included in the P budget, mass balances w ere positive for all treatments. Traditionally, soil testing focuses on the top 15 cm of the soil and for this study site woul d have characterized the soils as P deficient. When P concentration within the entire top 45 cm of the soil was considered, the soil was P sufficient and no P fertilization would be needed. This is evident by no P uptake response of bahiagrass to P fertilization, suggesting that the P held in the Bh horizon is available for bahiagrass uptake. We conclude that P held in the Bh horizon is a significant supply of P to low

PAGE 118

118 input pastures growing on Spodosols in Florida and should be considered in nutrient manag ement programs for perennial grasses

PAGE 119

119 Table 6 1. Mehlich1 soil P concentration at various depths as affected by year. Data represent average across three P rates and three replicates (n = 9). Depth Initial 2007 2008 SE P value cm mg kg 1 Ap horizon (0 15 cm) 3.1 5.3 3.0 0.7 0.003 E horizon (1 5 30 cm) 4.5 1.2 1.4 1.4 0.09 Bh horizon (3 0 60 cm ) 36.1 27.4 58. 3 10.5 0.02 SE = standard error Table 6 2. Estimated P mass balances for various depths in 2007 and 2008 as affected by fertilizer P, atmospheric deposition, soil available P and P loss through leaching and runoff. P rate Soil P Fertilizer P Deposition Runoff P P leached P uptake P storage 2007 P mass balance ( kg ha 1 ) at 15 cm ( Ap horizon) 0 5.7 0 0.41 1.34 0.1 15 10.8 5 6.2 5 0.41 1.29 0.09 18 7.5 10 6.4 10 0.41 1.23 0.09 16 0.8 SE 2.5 0.3 0.02 1.9 2.7 P mass balance at 45 cm (Bh horizon) 0 70 0 0.41 1.3 0.1 15 53.9 5 103 5 0.41 1.6 0.09 18 89.5 10 107 10 0.41 1.6 0.09 16 100.1 SE 19 0.3 0.02 1.9 19.5 2008 P mass balance at 15 cm ( Ap horizon) 0 7.2 0 0.41 1.3 0.06 13 6.6 5 9.2 5 0.41 2.0 0.07 16 2.9 10 15.5 10 0.41 2.5 0.09 16 7.2 SE 2.5 0.3 0.02 2.9 2.7 P mass balance at 45 cm (Bh horizon) 0 55 0 0.41 1.3 0.06 13 42.1 5 80 5 0.41 2.0 0. 07 16 67.6 10 82 10 0.41 2.5 0.09 16 73.0 SE 19 0.3 0.02 2.9 19.5

PAGE 120

120 Figure 61 Rainfall and water table depth in (a) 2007 and (B) 2008.

PAGE 121

121 Figure 62. Estimated daily drainage below 45 cm rooting depth in (A) 2007 and (B) 2008. Total drainag e for 2007 was 493.3 mm and 387.3 mm in 2008.

PAGE 122

122 CHAPTER 7 SUMMARY AND CONCLUSI ONS Summary Phosphorus management in low input bahiagrass ( Paspalum notatum Fl gge) pastures remains a major challenge of agronomic and environmental importance in Florida. Concerns over potential P losses and environmental impacts on water quality have prompted several revisions in the P fertilizer recommendations for bahiagrass pastures in Florida. The current UF/IFAS fertilizer recommendations suggest both soil and tissue tes ting for bahiagrass pastures. However there are limited data on impacts of the revised recommendation on bahiagrass forage production and potential impacts on water quality. G reenhouse and field experiments were conducted at UF/IFAS Range Cattle Research and Education Center (RCREC) in Ona from 2006 to 2008 to evaluate the effects of the new UF/IFAS fertilizer recommendations on forage yield, nutritive value, and the potential impacts on water quality in bahiagrass pastures growing on a Spodosol. The obje ctive of the greenhouse study was to indentify the critical minimum bahiagrass tissue P concentration below which bahiagrass dry matter yield (DMY) i s reduced. The data from the greenhouse study showed a positive correlation between DMY and tissue P concentration and that the critical minimum bahiagrass tissue P concentration was 1.3 g kg1 ( 0.2) The critical tissue P concentration reported in the study is consistent with the suggested UF/IFAS value of 1.5 g kg1. This means bahiagrass tissue P concentra tion can be reduced as low as 1.3 g kg1 without significant negative impacts on forage DMY Alth ough tissue P concentration was

PAGE 123

123 greater when sampling was done at 28 d compared to 56 d, tissue P concentrations at 56 d were above the critical minimum P conc entration. A field grazing study was conducted to investigate the effects of the revised UF/IFAS fertilizer recommendation on bahiagrass yield, nutritive value, and potential impacts of reduced P fertilization on water quality at N and P rates typically used in cow calf pastures. The results of the 2yr study showed that herbage mass, herbage accumulation rates, crude protein (CP), and in vitro digestible organic matter ( IVDOM ) were not affected by P applicatio n. However, t issue P concentrations increased from 1.9 to 2.2 g kg1 as P fertilization increased from 0 to 10 kg P ha1. Mehlich 1 soil extractable P and leachateP concentrations were not affected by P application. The adequate tissue P concentrations in the control treatments (1.9 g kg1) despite low soil test P values in surface horizon indicates that bahiagrass obtained sufficient P from the subsoil. Soil test P concentration in the Bh horizon was threefold greater than the Ap horizon. H ence, if the P levels in the entire soil profile are consi dered (both surface and subsoil), the soil will have adequate supply of P to maintain bahiagrass forage production. The reported tissue P concentration for the control treatments was 2 6 % greater than the suggested UF/IFAS critical bahiagrass tissueP conce ntration This possibly explains the lack of bahiagrass herbage mass and herbage accumulation rates responses to P fertilizer application in the current study although the surface soil test values were low. The objectives of the field clipping study (Cha pter 4) were to (i) investigate bahiagrass response to reduced P fertilization rates, (ii) evaluate the potential effects of bahiagrass P fertilization on soil test P concentrations and water quality, (iii) determine

PAGE 124

124 fluctuating water table effects on P release and soil P availability and ( iv) determine P mass balance for low input bahiagrass systems. The clipping study results showed that bahiagrass DMY response to P depended on year. In 2007, bahiagrass showed no response to P, whereas there was a linear increase in DMY with P rates in 2008. Phosphorus additions had no effects on soil Mehlich1 and leachateP concentrations. LeachateP concentrations in lysimeters positioned above the spodic horizon varied seasonally, with spikes coinciding with periods o f high rainfall and rising water tables. LeachateP concentrations at depths below the spodic horizon remained relatively constant (0.02 mg L1) during the entire growing season. Great est soil P availability ( index ed with anion exchange membranes ) occurred in August when water tables rose. Phos phorus mass balance estimates based on input and outputs and soil test P con centrations of the Ap horizon (15 cm) were negative for all treatments. However, when soil P held within the subsoil (Bh horizon) was included in the P mass balances as available P, the overall P mass balances were positive for all treatments. Conclusions Results from the greenhouse and field studies suggest that tissue testing is a useful tool in predicting bahiagrass response to P fertilizat ion. When tissue levels are above the critical tissue P concentration, P fertilization is not justified because there will be no yield response ( as shown in the grazing study ) However, if plant tissue levels are below the critical tissue P concentration a nd soil P levels are low, addition of small amounts of P fertilizer (10 kg P ha1) can improve bahiagrass yields with no impacts on water quality ( as demonstrated in the field clipping study ) The study also showed that fluctuating water table conditions p romotes release and upward flux of P from the Bh

PAGE 125

125 horizon, which increases soil P bioavailability. The P mass balance data suggests that P held in the Bh horizon is a significant source of P to low input bahiagrass pastures growing on Spodosols and that the P levels in the Bh horizon should be considered in bahiagrass nutrient management programs in cow calf pastures established on Spodosols in Florida. Though the Smyrna series used in the field clipping study may not be the dominant Spodosols soil series i n Florida, the observations in the study may be applicable to other Spodosols in the state and to soils in different environments that are subjected to alternate wetting and drying conditions. However, since this work was conducted over a relatively short time period, additional work is warranted on the effects of high water table condi tions experienced elsewhere in Florida on P release and availability to bahiagrass pastures growing on Spodosols .

PAGE 126

126 LIST OF REFERENCES Abrams, M.M.,and W.M. Jarrell. 1992. Bioavailability index for phosphorus using ion exchange resinimpregnated membranes. Soil Sci. Soc. Am. J. 56:15321537. Adams, W.E., A.W. White, R. D. McCreery, and R. N. Dawson. 1967. Coastal bermudagrass forage production and chemical composition as influenced by K source rate and frequency of application. Agron. J. 59:247250. Adjei, M. B., and J.E. Rechcigl. 2004. Interactive effect of lime and nitrogen on bahiagrass pastures. Soil Crop Sci. Soc. Fla. Proc. 63:5256. Allen, L.H., Jr., J.M. Ruddell, G.J. R utter and P. Yates, 1988. Land use effects on Taylor Creek water quality. p.6777. In E.G. Kruse, C.R. B urdick and Y.A. Yousef (Eds.), Environmentally Sound Water and Soil Management. American Society of Civil Engineers, New York. Ahn, H. and T.S. James. 2 001. Variability, uncertainty and sensitivity of phosphorus deposition load estimates in south Florida. Water, Air and Soil Pollu. 126:3751. Amer, F., D.R. Bouldin, C.A. Black, and F.R. Duke. 1955. Characterization of soil phosphorus by anion exchange resin adsorption and 32P equilibration. Plant Soil 6:391408. Bates, T.E. 1971. Factors affecting critical nutrient conc entrations in plants and their evaluation. A review. Soil Sci. 112:116130. Beaty, E.R., R.A. McCreery, and J.D. Powell. 1960. Resp onse of bahiagrass to nitrogen fertilization. Agron J. 52:453455. Belmont, M.A, J.R. White and K. R .Reddy. 2009. Phosphorus sorption and potential phosphorus storage in sediments of Lake Istokpoga and the upper chain of Lakes Florida, USA. J. Enviorn. Qual.38: 987 996. Berryman E.M., R.T. Venterea, J.M. Baker, P.R. Bloom and B. Elf. 2009. Phosphorus and greenhouse gas dynamics in a drained calcareous wetland soils in Minnesota. J. Environ. Qual. 38:21472158. Blue, W.G. 1970. Fertilizer nitrogen uptake by Pensacola bahiagrass from Leon fine sand, a Spodosol. Proc XI Int. Grassland Congress, Surfers Paradise, Queensland, Australia. p. 389392. Brezonik, P.L. 1984. Trophic state indices: Rational for m ultivariate approaches. p. 441445. In Lake and reservoir manag ement. EPA 440/584 00111. USEPA, Washington, DC.

PAGE 127

127 Boggess, W.G., G. Johns, and C. Meline. 1997. Economic impacts of water quality programs in the Lake Okeechobee watershed of Florida. J. Dairy Sci. 80:26822691. Boggess, W.G., E.G. Flaig and R.C. Fluck. 19 95. Phosphorus budget basin relationships for Lake Okeechobee tributary basins. Ecol. Eng. 5:143162. Bottcher, A.B, T.K. Tremwel and K.L. Campbell. 1995. Best management practices for water quality improvement in the Lake Okeechobee watershed. Ecol. Engi n. 5:341356. Burgoa, B., R.S. Mansell, G.J. Sawka, P. Nkedi Kizza, J.C. Capece and K.L. Campbell, 1991. Spatial variability of depth to Bh horizon in Florida Haplaquods using groundpenetrating radar. Proc. Soil Crop Sci. Soc. Fla., 50: 125130. Burton, G.W., R.N. Gates, and G.J. Gascho. 1997. Response of Pensacola bahiagrass to rates of nitrogen, phosphorus and potassium fertilizers. Soil Crop Sci. Soc. Fla. Proc. 56:31 35. Burton, G. W. 1954. Coastal bermudagrass Georgia Agr. Exp. Sta. Bull. # S. 2. Br ye, K.R., T.W. Andraski, W. M. Jarrell, L.G. Bundy and J.M. Norman. 2002. Pphosphorus leaching under restored tallgrass prairie and corn agroecosystems. J. Enviorn. Qual. 31:769 781. Capece, J.C., K L. Campbell, P.J. Bohlen, D. A. Graetz ,and K.M. Portier. Soil phosphorus, cattle stocking rates, and water quality in subtropical pastures in Florida, USA. Rangeland Ecol. Manage. 60:1930. Carpenter, S. R., N. F. Caraco, D. L. Correll, R. W. Howarth, A. N. Sharpley, and V. H. Smith. 1998. Nonpoint pollution of surface waters with phosphorus and nitrogen. Ecol. Appl. 8:559 568. Chambliss, C.G., and M.B. Adjei. 2006: Bahiagrass. University of Florida Coop. Ext. Serv. Gainesville, FL. Collins, M.E. 2003. Keys to soil orders in Florida. University of Florida Coop. Ext. Serv. Gainesville, FL. http://edis.ifas.ufl.edu/S s113 Cooperband, L.R., and T.J. Logan. 1994. Measuring in situ changes in labile soil phosphorus with anion exchange membranes. Soil Sci. Soc. Am. J. 58:105114. Cooperband, L.R., P.M. Gale, and N.B. Comerford. 1999. Refinement of the anion exchange membrane method for soluble phosphorus measurement. Soil. Sci. Soc. Am. J 63:58 64. David, M.B., and L.E. Gentry. 2000. Anthropogenic inputs of nitrogen and phosphorus and riverine export for Illinois, USA. J. Environ. Qual. 29:494508.

PAGE 128

128 Dou, Z., L.E. Lanyon, J.D. Ferguson, R.A. Kohn, R.C. Boston, and W. Chalupa. 1998. An integ rated approach of managing nitrogen on dairy farms: Evaluating farm performance using the dairy nitrogen planner. Agron. J. 90:573 581. Flaig, E.G., and K.E. Havens. 1995. Historical trends in the Lake Okeechobee ecosystem: I. Land use and nutrient loading Hydrobiologie 107:124. Florida Department of Agriculture and Consumer Services. 2006. Agriculture Floridas economic engine. Florida Agricultural Statistical Directory, Tallahassee. Florida Department of Agriculture and Consumer Services. 2007. Fertil izer Florida consumption requirement. FY 20062007. Fisher, F.L., and A.G. Caldwell. 1959. The effect of continued use of heavy rates of fertilizer on forage production and quality of coastal bermudagras s. Agron. J. 51:99102. Follett, K.F., and G.A. Reichman. 1972. Soil temperature, water and phosphorus effects upon barley growth. Agron. J. 64:3639. Fonyo, C., and E. Flaig. 1995. Phosphorus budgets for Lake Okeechobee tributary basins. Ecol. Eng. 5:209227. Gentry, L.E., M.B. David, F.E. Below, T.V. Royer, and G.F. Mclsaac. 2009. Nitrogen mass balance of a tiledrained agriculture watershed in east central Illinois. J Environ. Qual. 38:1841 1847. Gentry, L.E., M.B. David, K.M. Smith, and D.A. Kovac ic. 1998. Nitro gen cycling and tile drainage nitrate loss in a corn/soybean watershed. Agric. Ecosyst. Environ. 68:8597. Giblin, A.E., J.A. Laundre, K.J. Nadelhoffer, and G.R. Shaver. 1994. Measuring nutrient availability in Arctic soils using ion exchange resins: A field test. Soil Sci. Soc. Am. J. 58:1154 1162. Graetz, D.A., V.D. Nair, K.M. Portier and R.L. Ross. 1999. Phosphorus accumulation in manure impacted Spodosols of Florida. Agric. Ecosyst. Environ. 75:3140. Grings, E.E., M.R. Haferkamp, R.K. Heitschmidt, and M.G. Karl. 1996. Mineral dynamics in forages of the Northern Great Plains. J. Range Mange. 49:234240. Hanlon E.A, R. Mylavarapu, and I.V. Ezewa: 2006. Development of bahiagrass fertilization recommendations: 19902005. University of Florida Coop. Ext. Serv. Gainesville, FL. http://edis.ifas.ufl.edu/SS456 Harris, H.C., V.N. Schroder, and R.L. Silman. 1968. Nutrie nt deficiency effects on yield and chemical composition of plants grown on Leon fine sand. Flor ida Agri c. Exp. Sta. Tech. Bull. 725.

PAGE 129

129 Havens, K.E., R.T. James, T.L. East, and V.H. Smith. 2003. N:P ratios, light limitation and cynobacterial dominance in a subtropical lake impacted by nonpoint source nutrient pollution. Environ. Pollut 122:379390. Hiscock, J.G., C.S. Thourot, and J. Zhang. 2003. Phosphorus budget land use relationships for the northern Lake Okeechobee watershed, Florida. Ecol. Eng. 21:6374. Hodges, E.M., F.M. Peacock, H.L. Champman,Jr ., and F.G. Martin. 1976. Grazing trials with tropical grasses and legumes in Peninsular Florida. Soil Crop Sci. Soc. Fla. Proc. 35:84 86. Ibrikci, H., E.A. Hanlon, and J.E. Rechcigl. 1999. Inorganic phosphorus and manure effects on bahiagrass production on a spodosol. Nutr. Cycling Agroeco. 54:259266. Ibrikci, H., E.A. Hanlon, and J.E. Rechcigl. 1992. Initial calibration and correlation of inorganic phosphorus soil test methods with bahiagrass field trial. Commun. Soil Sci. Plant Anal. 23:2569 2579. Ibrikci, H., N.B. Comerford, E.A. Ha n lon, and J.E. Rechci gl. 1994. Phosphorus uptake by bahiagrass from spodosols: modeling of uptake from different horizons. Soil Sci.Soc. Am. J. 58:139 143. Izuno, F.T., C.A. Sanchez, F.J. Coale, A.B. Bottcher and D.B. Campbell. 1991. Phosphorus concentration in drainage waters in Everglades agricultural area. J. Environ. Qual. 20:608619. Izuno, F.T. 1987. Water Budgeting for High Water Table Soils. Circ. #769 (Revised in 2005) Univ. of Florida Coop.Ext. Serv. Gainesville, FL. Jones, J.B. Jr. 1970. Distribution of element s in corn leaves. Commu. Soil Sc. Plant Anal. 1:2734. Jones, J.B. Jr. 1967. Interpretation of plant analsysis for several agronomic crops. p. 49 58. In Soil testing and plant analysis, Part 1. SSSA Special Publi. Series No. 2. Soil Sci. Soc Amer. Madison, WI. Johnson,D.W., P.S.J. Verburg, and J.A. Arnone. 2005. Soil extraction, ion exchange resin, and ion exchange membrane measures of soil mineral nitrogen during incubation of a tallgrass prairie soil Soil Sci. Soc. Am. J. 2005 69:260265. Jordon, C.W., C .E. Evans, and R.D. Ross. 1966. Coastal bermudagrass response to application of P and K as related to P and K levels in the soil. Soil Sci. Soc. Amer. Proc. 30:477480.

PAGE 130

130 Kidder, G., E.A. Hanlon, and C.G. Chambliss. 2000. IFAS standardized fertilizer recomme ndations for agronomic crops. Fact Sheet SL129 (revised). Univ. of Florida Coop. Ext. Service, Gainesville, FL. Koelsch, R. 2005. Evaluating livestock system environmental performance with wholefarm nutrient balance. J. Environ. Qual. 34:149155. Koelsc h, R., and G. Lesoing. 1999. Nutrient balance on Nebraska livestock confinement systems. J. Anim. Sci. 77:63 71. Koopmans, G.F., W. J. Chardon, P. A. I. Ehlert, J. Dolfing, R. A. A. Suurs, O. Oenema, and W. H. van Riemsdijk. 2004. Phosphorus availability f o r plant uptake in a phosphorus enriched noncalcareous sandy soil J. Environ. Qual. 33:965975. Lindsay, W.L. 1979. Chemical equilibria in soils. John Wiley & Sons, New York. Little, S., J. Vicente, and F. Albruna. 1959. Yield and protein content of irrig ated napiergrass, guineagrass and pangolagrass as affected by nitrogen fertilization. Agron. J. 51:111113. Liu, Min, J.B. Sartain, L.E. Trenholm, and G.L. Miller. 2008. Phosphorus requirement of St. Augustinegrass grown in sandy soils. Crop Sci. 48:11781186. Liu Y., G. Villalba R.U. Ayres and H. Schroder. 2008. Global phosphorus flows and environmental impacts from a consumption perspective. J. Indust. Eco. 12:229247. Mackoviak, C.L., A.R. Blount, E.A. Hanlon, M.L. Silveir a, M.B. Adjei, and R.O. M yer. 2008. Getting the most out of bahiagrass fertilization. Univ. of Florida Coop.Ext. Serv. Gainesville, FL. http://edis.ifas.ufl.edu/SS469 Mallarino, A.P., and S.L. Higashi. 2009. Assessment o f potassium su pply for corn by analysis of plant parts. Soil Sci. Soc. Am. J. 73:21772183. Mangiafico, S.S., and K. Guillard. 2006. Anion exchange membrane soil nitrate predicts turfgrass color and yield. Crop Sci. 46:569577. Marshall, T.J., and J.W. Holmes.1988. Soil physics 2nd Ed., Cambridge University Press, New York Martin, W.E., and J.E. Matocha. 1973. Plant analysis as an aid in fertilization of forage crops. p. 393426. In L.M. Walsh and J.D. Beaton (eds) Soil testing and plant analysis. Revised ed. SSSA Madison, WI. Martin, H.W., D.B. Ivanoff, D.A. Graetz and K.R. Reddy. 1997. Water table effects on histosol drainage water carbon, nitrogen and phosphorus. J. Environ. Qual. 26:10621071.

PAGE 131

131 McCaleb, J.E., C.L. Dantzman, and E.M. Hodges. 1966. "Response of pangolagrass and Pensacola bahiagrass to different amounts of phosphorus and potassium." Soil Crop Sci. Soc. Fla. Proc. 26:249256. McKenzie, H.A., and H.S. W allace. 1954. The Kjeldahl determination of nitrogen a critical study of digestion conditions temp erature, catalyst, and oxidizing agent. Aust. J. Chem. 7:55 70. Meason, D.F., and T.W. Idol. 2008. Nutrient sorption dynamics of resin membranes and resin bags in a tropical forest. Soil Sci. Soc. Am. J. 72:18061814. Mehlich, A. 1953. Determination of P, Ca, Mg, K, Na, NH4. Soil Testing Div. Pub.153, Nor th Carolina Dept. Agric. Raleigh, NC. Mylavarapu, R., D. Wright, G. Kidder and C. G. Chambliss. 2007. UF/IFAS standard fertilization recommendations for agronomic crops. Fact Sheet SL129. Univ. of Florid a Coop. Ext. Service, Gainesville, FL. Mylavarapu, R.S., and D.E. Kennelley. 2002. UF/IFAS Extension Soil Testing Laboratory (ESTL) analytical procedures and traini ng manual. Series in Soil Sci. No.1248. Florida Coop. Ext. Service, Gainesville, FL. Muchov ej, R.M., and J.J. Mullahey. 2000. Yield and quality of five bahiagrass cultivars in southwest Florida. Soil Crop Sci. Soc. Fla. Proc. 59:82 84. Murphy, J., and J.P. Riley. 1962. A modified single solution method for the determination of phosphate in natur al waters. Anal.Chem. Acta 27:3136. Nair, V.D., R.R. Villapando and D.A. Graetz. 1999. Phosphorus retention capa city of the spodic horizon under varying environmental conditions. J. Environ. Qual. 28:13081313. Nair, V.D., D.A. Graetz, and K.R. Reddy. 1998. Dairy manure influences on phosphorus retention capacity of Spodosols. J. Environ. Qual. 27:522527. Nair, V.D., K.M. Portier, D. A. Graetz, and M. L. Walker. 2004. An environmental threshold for degree of phosphorus saturation in sandy soils. J. Environ. Qual. 33:107113. Nair, V.D., and D. A. Graetz. 2004. Agroforestry as an approach to minimizing nutrient loss from heavily fertilized soils: The Florida experience. Agroforestry Syst. 61:269279. National Research Council (NRC). 1996. Nutrient requirem ents of beef cattle. 7th ed. National Academy Press, Washington, D. C.

PAGE 132

132 Newman, Y.C., S. Agyin Birikorang, M.B. Adjei, J.M. Scholberg, M.L. Silveira, J.M.B. Vendramini, J.E. Rechcigl and L.E. Sollenberger. 2009 a Nitrogen fertilization effect on phosphorus remediation potential of three perennial warm season forages. Agron. J. 101:12431248. Newman, Y.C., S. Agyin Birikorang, M.B. Adjei, J.M. Scholberg, M.L. Silveira, J.M.B. Vendramini, J.E. Rechcigl and L.E. Sollenberger. 2009b. Enhancing phosphorus phytor emedation potential of two warm season perennial grasses with nitrogen fertilization. Agron. J. 101:13451351. Nielsen, K.F., R.L. Halstead, A.J. Maclean, R.M. Holme s, and S.J. Bourget. 1960. The influence of soil temperature on the growth and mineral composition of oats. Can. J. Soil Sci. 40:255262. Nyiraneza, J., A. N Dayegamiye, M.H. Chantigny and M.R. Laverdire. 2009. Variations in corn yield and nitrogen uptake in relation to soil nitrogen attributes and nitrogen availability indices. Soil Sci. Soc. Am. J. 73 :317 327. Obour, A.K., M.L. Silveira, M.B. Adjei, J.M.B. Vendramini, and J.E. Rechcigl. 2009. Cattle manure application strategies effects on bahiagrass yield, nutritive value, and phosphorus recovery. Agron. J. 101:10991107. Oenema, O., H. Kro s, and W.de Vries. 2003. Approaches and uncertainties in nutrient budgets: Implications for nutrient management and environmental policies. Eur. J. Agron. 20:316. Oyanarte, M., G. Besga, M. Rodriguez, M. Domingo, and A.G. Sinclair. 1997. Balance pasture f ertilization in Basque country: 1. Phosphorus and potassium budgets on dairy farms. Pank, H.K., and K.R. Reddy. 1998. Phosphorus sorptio n characteristics of estuarine sediments under different redox conditions. J. Environ. Qual.30:1474 1480. Pannamperuma, F.N. 1972. The chemistry of submerged soils. Advan. Agron. 26: 2995. Pant H.K., and K.R. Reddy. 2002. Potential internal loading of phosphorus in a wetland constructed in agricultural land. Water Res. 37:965972. Plucknett, D.L., and R.L. Fox. 1965. Effec ts of phosphor us fertilization on yields and composition of pangolagrass and Desmdium intortum. Proc. IX. Int. Grassland Congress, Sao Paulo, Brazil. p.15251529. Power, J.F., D.L. Grunes, W.O. Willis, and G.A. Reichma n. 1963. Soil temperature and phosphor us effects upon barley growth. Agron. J. 55:389392. Prasad, B., and N.P. Sinha. 1981. Balance sheet of soil phosphorus and potassium as influence by intensive cropping and fertilizer use. Plant and Soil 60:187193.

PAGE 133

133 Ranger, J. and M.P. Turpault. 1999. Input output nutrient budgets as a diagnostic tool for sustainable forest management. Forest Eco. Manage. 122:139154. Rao, A.S., P.G.Babrekar and A.B.Ghosh. 1982. Total nitrogen and phosphorus balance sheet in a typic ustochrept soil under intensive cropping and fertilizer use. Plant and Soil. 68:125129. Raupach, M.R., J.M. Kirby, D.J. Barrett and P.R. Briggs. 2001. Balances of water, carbon, nitrogen and phosphorus in Australian landscapes: I project description and results. CSIRO Land and Water technical report # 40/01. Rechcigl, J.E., P. Mislevy, and H. Ibrikci. 1995. Response of established bahiagrass to broadcast lime and phosphorus. J. Prod. Agric. 8:249253. Rechcigl, J.E., G.G. Payne, A.B. Bottcher, and P.S. Porter. 1992. Reduced P application on bahi agrass and water quality. Agronomy J. 84:463468. Rechcigl, J. E., and A.B. Bottcher. 1995. Fate of phosphorus on bahiagrass ( Paspalum notatum ) pastures. Ecol. Engin. 5:247259. Reddy, K.R., E. Lowe, and T. Fontaine. 1999. Phosphorus in Florida ecosystems : Analysis of current issues. pp .111141. In K.R. Reddy, G.A. OConnor and C.L. Schelake (eds.), Phosphorus biogeochemistry in subtropical ecosystems. Lewis Publishers, Boca Raton, FL. Reddy, K.R., G.A. OConnor and P.M. Gale. 1998. Phosphorus sorption c apacities of wetlands soils and stream sediments impacted by dairy effluent. J. Environ. Qual. 27:438447. Reddy, K.R., E.G. Flaig and D.A. Graetz. 1996. Phosphorus storage capacity of uplands, wetlands, and streams of the Lake Okeechobee watershed, Flori da. Agric. Ecosyst. Environ. 59:203 216. Rhoads, F.M., R. L. Stanley, Jr., and E.A. Hanlon. 1997. Response of bahiagrass to N, P, and K on an Ultisol in North Florida. Soil Crop Sci. Soc. FL. Proc. 56: 79:83. Rodulfo, S., and W.G. Blue. 197 0. The availabil ity to forage plants of accumulated phosphorus in Leon fine sand. Soil Crop Sci. Soc. Fla. Proc. 30: 167 174. Rhue, R. D., and G. W. Harris. 1999. Phosphorus sorption/desorption reactions in soils and sediments. pp.187203. In K. R. Reddy, G. A. OConnor and C. L. Schelake (eds.), Phosphorus biogeochemistry in subtropical ecosystems. Lewis Publishers, Boca Raton, FL. Sands, G. 2001. Soil water concepts. University of Minnesota Coop.Ext. Serv. St. Paul, MN. Bull. # 07644S

PAGE 134

134 Said, A. M. Nachabe, M. Ross and J. Vomacka. 2005. Methodology for estimating specific yield in shallow water environment using continuous soil m oisture data. J. Irrg. Drain. Egin. 131: 533538 SAS Institute. 1999. SAS/STAT guide for personal computers. Version 6. SAS Inst.,Cary, NC. S eng, V., R.W. Bell and I.R. Willett. 2006. Effects of lime and flooding on phosphorus availability and rice growth on two acidic lowland soils. Commu. Soil Sci. Plant Anal. 37:313336. Sharma, K.N., D.S.Rans and A.L. Bhandari. 1987. Influence of growing various crops in five cropping sequence on the changes in phosphorus and potassium content of soil. J. Agric. Sci. 109:281284. Shekiffu, C.Y., and J. M. R. Semoka. 2006. E valuation of iron oxide impregnated paper method as an index of phosphorus availability in paddy soils of Tanzania. Nutr.Cycl. Agroecosyst. 77:169177 Sigua, G.C., M. J. Williams, S. W. Coleman, and R. Starks. 2006. Nitrogen and phosphorus status of soil and trophic state of lakes associated with foragebased beef cattle operations in Florida. J. Environ. Qual. 35:240252. Sigua, G.C., M. J. Williams, and S. W. Coleman. 2004. Levels and changes of soil phosphorus in the subtropical beef cattle pastur es. Commun. Soil Sci. Plant Anal. 35:975990. Sigua, G.C., R.K. Hunnard, and S.W. Coleman. 2010a. Quantifying phosphorus levels in soils, plants, surface water, and shallow groundwater associated with bahiagrass based pastures. Enviorn. Sci. Pollut. Res. 17:210 219. Sigua, G.C., R.K. Hunnard, S.W. Coleman, and M. Williams. 2010b. Nitrogen in soils, plants, surface water and shallow groundwater in a bahiagrass pasture of southern Florida, USA. Nutr. Cycl. Agroecosyst. 86:175187. Silveira, M.L., J.M. Vendram ini, L.E. Sollenbe rger, C.L. Mackowiak, and Y.C. Newman. 2007. Tissue analysis as a nutrient management tool for bahiagrass pastures. Univ. of Florida Coop. Ext.Ser. Gainesville SL252. Smaling, E.M.A., and A.R. Braun. 1996. Soil fertility research in sub saharan Africa: New dimensions, new challenges Commun Soil Sci. Plant Anal. 27:365386. Smith, P.F. 1962. Mineral analysis of plant tissue. Ann. Rev. of Plant Physiol. 13:81108. Spears, R.A., A.J.Young, and R A. Kohnt. 2003. Wh ole farm phosphorus balance on western dairy farms. J. Dairy Sc. 88:688695.

PAGE 135

135 Soil Survey Staff. 1996. Keys to soil taxonomy. U.S. Gov. Print. Office, Washington, DC Soil Survey Staff. 1984. Soil survey of Hardee County, Florida. USDA Soil Conservation Service, U.S. Government Pr inting Office, Washington DC. South Florida Water Management District (SFWMD). 1989. Surface water improvement and management (SWIM) plan for Lake Okeechobee. West Palm Beach, FL. Stanley, R.L., and F.M. Rhoads. 2000. Bahiagrass production, nutrient uptak e, and soil test P and K. Soil and Crop Sci. Soc. Fla. Proc. 59:159163. Stoorvogel, J.J., Smaling, E.M.A., Janssen, B.H., 1993. Calculating soil nutrient balances in Africa at different scales: I. Supra national scale. Fertilizer Research 35:227 235. Sumn er, S., W. Wade, J. Selph, J. Southwell, V. Hoge, P. Hogue, E. Jennings, P. Miller, and T. Seawright. 1991. Fertilization of established bahiagrass pasture in Florida.Cir. 916. Univ. of Florida Coop. Ext.Ser. Gainesville. Surridge, B.W.J, A.L.Heathwait and A.J. Baird. 2007. The release of phosphorus to Porewater and surface water from river riparian sediments. J. Environ. Qual. 36:15341544. Swain, H.M., P.J. Bohlen, K.L. Campbell, L.O. Lollis, and A.D. Steinman. 2007. Integrated e cological and economic analysis of ranch management systems: An example from south central Florida. Rangeland Eco.Mang.60:111. Terry, R.E., G.J. Gascho, and S.F.Shih. 1980. Effect of depth of water table on the quality of in the Everglades Agricultural Area. In Proc. 6th Peat Congress, Duluth, MN. Tyner, E.H. 1946. The relation of corn yields to leaf nitrogen, phosphorus, and potassium content. Soil Sci. Soc. Am. Proc. 11:317323. Ulrich, A. 1952. Physiological bases for assessing the nutri tional requirements of plants. Ann. Rev. of Plant Physiol. 3:207228. Ulrich, A., and F.J. Hills. 1973. Plant analysis as an aid in fertilizing sugar crops: Part 1. sugar beets. p. 271288. In L. M. Walsh and J. D. Beaton (eds) Soil testing and plant analysis. Revised ed. SSSA ,Madison, WI. U. S. Environmental Protection Agency (USEPA), 1993. Methods for the determination of inorganic substances in environmental samples, U SEPA 600/R 93/100. Method 353. 2. U. S. Environmental Protection Agency (USEPA) 1983. Methods for chemical analysis of waters an d wastes, USEPA 600/479020. Method 351.2.

PAGE 136

136 Vallapando, R.R., and D.A. Graetz. 2001. Water table effects on phosphorus reactivity and mobility in a dairy manureimpacted spodosol. Ecol. Engin. 18:7789. v an Eerdt, M.M., P.K.N. Fong. 1998. The monitoring of nitrogen surpluses from agriculture. Environ. Pollut. 102:227233. Walsh, G. L., and H. A. Birrell. 1987. Seasonal variation i n the chemical composition and nutritive value of five pasture species in southwestern Victoria. Aust. J. Exp. Agric. 27:807816 Watson, C. A., and D. Artkinson. 1999. Using nitrogen bugets to indicate nitrogen use efficiency and losses from whole farm systems: A comparison of three methodological approaches. Nutr. Cycl. Agroecosyst. 53:259267. Watts, P.J., E.A. Gardnes, R.W. Tuc ker, and K.D. Casey. 1994. Mass balance approach to design of nutrient management systems at cattle feedlots. p. 27 33. In D.E. Storm and K.G. Casey (ed.) Proc. of the Great Plains Animal Waste Conf. on Confined Animal Production and Water Quality: Balanci ng Animal Production and Environ., Denver. 19 21 Oct. 1994. Great Plains Publ. 151. Natl. Cattlemens Assoc., Englewood, CO. Webb, J., S. Ellis, R. Harrison, and R. Thorman. 2004. Me asurement of N fluxes and soil N in two arable soils in the UK. Plant Soi l 260:253270. Weih, M. 1998. Seasonality of nutrient availability in soils of subarctic mountain birch woodlands, Swedish Lapland. Arctic and Alpine Res. 30:1925. Whitty, E.B., D.W. Johnson, G. Kidder, C.G. Chambliss, D.L. Wright and J.J. Street. 1977. F ertilization of field and forage crops. Agronomy Fact no.70. University of Florida Coop. Ext. Serv. Gainesville. Wright, R.B., B.G. Lockaby and M.R. Walbridge. 2001. Phosphorus availability in an artificially flooded southeastern floodplain forest soil. So il Sci. Soc. Am. J. 65:12931302. Yucan, T. L. 1966. Characteristics of surface and spodi c horizons of some Spodosols. Soil Crop Sci. Soc. Fla. Proc. 26:163174. Young, E.O., and D.S. Ross. 2001. Phosphate release from seasonal flooded soils: A laboratory microcosm study. J. Environ. Qual. 30:91101. Ziadi, N., R.R. Simard, G. Allard, and J. Lafond. 1999. Field evaluation of anion exchange membranes as a N soil testing method for grasslands. Can. J. Soil Sci. 79:281 294. Zielinski, R.A., W.H. Orem, K.P. Si mmons, and P.J. Bohlen. 2006. Fertilizer derived uranium and surfer in rangland soil and runoff; a case study in central Florida. Water, Air, Soil Pollut. 176:163183.

PAGE 137

137 BIOGRAPHICAL SKETCH Augustine is a native of Ghana, born some 30 years ago to M r and Mrs. Bekoe. He enrolled in Sunyani Secondary School where he earned his GCE O and A Level certificates in 1994 and 1997, respectively. Augustine taught at Berekum secondary school as part of his national service requirement as an integrated science tutor from 1997 to 1998. He had his BSc degree in Agriculture from the Kwame Nkrumah University of Science and Technology (KNUST) Kumasi, Ghana in May 2002. Before joining the Doctor of Plant Medicine program at the University of Florida in August 2004, he work ed as a Teaching and Research Assistant at the Biotechnology Unit of the Crop Science Department of KNUST. In the spring of 2005, Augustine joined the Agronomy Department to pursue a Maste r of Science degree with Dr. M. B. Adjei (of Blessed memory) at the R ange Cattle Research and Education Center in Ona as his mentor. Mr Obour graduated with an M S c degree in Agronomy (Forage Production and Management) and a minor in soil science in spring 2007. With strong interest in soil science, Augustine decided to pur sue a PhD degree in s oil and w ater sciences supervised by Dr M.L. Silveira. His res earch is focused on developing nutrient management strategies that can simultaneously improved agronomic yields of forage grasses, and yet reduced offsite loss of nutrients to the environment. Augustine is married to Anna and has two children Augustine, Jr., and Brian