Citation
Management of Phosphorus Sources and Water Treatment Residuals (WTR) for Environmental and Agronomic Benefits

Material Information

Title:
Management of Phosphorus Sources and Water Treatment Residuals (WTR) for Environmental and Agronomic Benefits
Creator:
OLADEJI, OLAWALE OLUSEGUN
Copyright Date:
2008

Subjects

Subjects / Keywords:
Agricultural economics ( jstor )
Greenhouses ( jstor )
Manure ( jstor )
Phosphorus ( jstor )
Soil samples ( jstor )
Soil science ( jstor )
Soil water ( jstor )
Soils ( jstor )
Surface runoff ( jstor )
Water treatment ( jstor )
City of Boca Raton ( local )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Olawale Olusegun Oladeji. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
12/31/2007
Resource Identifier:
658230831 ( OCLC )

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

1 MANAGEMENT OF PHOSPHORUS SOURCE S AND WATER TREA TMENT RESIDUALS (WTR) FOR ENVIRONMENTAL AND AGRONOMIC BENEFITS By OLAWALE OLUSEGUN OLADEJI A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2006

PAGE 2

2 Copyright 2006 by Olawale Olusegun Oladeji

PAGE 3

3 To my wife, ‘Bunmi, and my ch ildren, Faith, ‘Tobi, and ‘Dara.

PAGE 4

4 ACKNOWLEDGMENTS Successful completion of my graduate studies represented by this dissertation was made possible by a remarkably large group of people, and I deeply regret that I lack the space to thank them all. My first word of thanks is to my advisor, Dr. George A. OÂ’Connor. Without his consistent motivation and commitment to excellen ce, my research in this field could not have been accomplished. Frequent discussions with h im helped shape my dissertation and clarify needed ideas. Dr. Jerry B. Sartain, my co-advi sor, was always there to assist and make me realize what needed to be done. I appreciate his support, en couragement, and his patient explaining of difficult concepts to me over and over again. I would like to thank Drs. Tom A. Obreza, Vi mala D. Nair, Ramon C. Littell, and JeanClaude J. Bonzongo for serving on my advisory committee, sharing with me their time, knowledge and expertise whenever it was need ed, and for tolerating my inconveniencies. This research was done with the financia l support from Water Environment Research Foundation. I really appreciate th e assistance and hope the resu lts will enhance a healthy environment, which is our common goal. I am also significantly indebt ed to a number of other pe ople. Foremost among them are Mrs. OÂ’Connor and Mrs. Sartain who acclimati zed me and enhanced my stay in US. My colleagues Sampson, Scott, Sarah, Liz, Jaya, Matt, Drs. Maria L. Silveria, and Luis Alleoni have been true friends and comrades during our mutual exploration of a field that once seemed so strange. Being part of the soil chemistry group was a source of pride and honor. I have always felt fortunate to be part of this group and am gr ateful especially for their genuine friendship and support. I benefited from both collaboration and the pleasant work environment offered by members and visitors in the group. They were a real team, as well as family for me.

PAGE 5

5 I thank all members of a large family cal led the Soil and Water Science Department. Surely, it is an opportunity to be a member of the team, and I quite appreciate how everyone imparted my life. I would like to thank my friends in Gainesvill e, Nigeria, and other places too numerous to mention, whose friendship and support helped me research and then write this dissertation. I quite appreciate the support of my bible study group: the Akinyemis, Adesogans, Fayigas, AgynBirikorangs, and others for their prayers and en couragement. To the Adegbembos, for giving me a car that was so helpful during my study, I say thanks and God bless you. I would like to thank my parents, whose unwavering support and confidence in me has been a beacon, all of my life. They taught me to learn a lesson from anyone, and every situation by paying attention and listening and never retreat ing from pursuing a goal. Special thanks go to my brothers and sisters for their love, help a nd support, and for standing in during my absence. For those who stayed with and encouraged my family in Nigeria during my absence, especially the Oluyeduns, Akinwandes, Adeoyes, Aworindes, and others, I thank you all. I thank God for my loving, beautiful, and de dicated wife, ‘Bunmi, and the wonderful children, Faith, ‘Tobi, and ‘Dara, who are a constant source of inspiration. Surely, they are the reason that I have come to this point in my lif e with the level of satis faction, completeness, and joy I feel. I thank them for being my partner, c onfidant, travel compani on, motivator, best friend, and the love of my life. They were the undisclosed source of energy and strength that I relied on during challenging times. Most importantly, I again thank God for everything. Surely he is the unlimited, unchangeable, and faithful God.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ........10 LIST OF FIGURES................................................................................................................ .......13 ABSTRACT....................................................................................................................... ............17 CHAPTER 1 INTRODUCTION AND LITERATURE REVIEW..............................................................19 Introduction................................................................................................................... ..........19 Phosphorus Losses and Availability to Plants.................................................................20 Phosphorus Application Rates and P Losses...................................................................23 Agronomic and Environmental Soil P Thresholds..........................................................24 Water Treatment Residuals (WTR) as Soil Amendments...............................................28 Hypotheses and Research Objectives.....................................................................................33 Study Approach................................................................................................................. .....34 Dissertation Format............................................................................................................ ....35 2 CHARACTERIZATION OF AMENDMENTS AND SOILS USED IN THE STUDY.......37 Introduction................................................................................................................... ..........37 Materials and Methods.......................................................................................................... .38 Amendments and Soil Selection......................................................................................38 Amendments and Soil Analysis.......................................................................................39 Results and Discussion......................................................................................................... ..42 Amendments (P-sources and WTR) Characterization....................................................42 Soil Characterization.......................................................................................................46 3 AGRONOMIC IMPACTS OF LAND APPLIED WTR AND DIFFERENT PSOURCES........................................................................................................................ ......49 Introduction................................................................................................................... ..........49 Materials and Methods.......................................................................................................... .52 Experimental Procedure..................................................................................................52 Soil and Plant Analysis....................................................................................................55 Statistical Analysis........................................................................................................... .......56 Results and Discussion......................................................................................................... ..57 Soil pH and EC During the Study...................................................................................57 Soil Phosphorus During the Study..................................................................................58 Soil P and Sorption Properties as Affected by WTR......................................................68 Plant P Uptake, P Concentrations and Dry Matter Yield................................................76

PAGE 7

7 Relative Phosphorus Phytoavailability (RPP).................................................................84 Aluminum Toxicity.........................................................................................................88 Summary and Conclusions.....................................................................................................90 4 FIELD VALIDATION OF AGRONOMI C IMPACTS OF LAND APPLIED WTR AND DIFFERENT P-SOURCES...........................................................................................92 Introduction................................................................................................................... ..........92 Materials and Methods.......................................................................................................... .93 Experimental Procedure..................................................................................................93 Soil Samplings and Analysis...........................................................................................95 Plant Samples and Analysis.............................................................................................96 Statistical Analysis........................................................................................................... .......97 Results and Discussion......................................................................................................... ..97 Impacts of WTR and P-source Rates on Soil P...............................................................97 Impacts of P-Sources on Soil P.....................................................................................103 Impacts of Treatments on Soil Sorption Properties.......................................................105 Impacts of Treatments on Plants...................................................................................110 Relative Phosphorus Phytoavailability (RPP)...............................................................115 Aluminum Toxicity.......................................................................................................122 Summary and Conclusions...................................................................................................123 5 EVALUATION OF SOIL TEST METHODS FOR FLORIDA SANDS TREATED WITH VARIOUS P-SOURCE S AND WATER TREATMENT RESIDUAL (WTR).......125 Introduction................................................................................................................... ........125 Materials and Methods.........................................................................................................1 27 Glasshouse Experiment.................................................................................................127 Field Experiment...........................................................................................................128 Statistical Analysis........................................................................................................... .....129 Results and Discussion......................................................................................................... 129 Soil Test P and WTR Treatments..................................................................................129 Soil Test P and Plant Response.....................................................................................133 Summary and Conclusions...................................................................................................140 6 APPLICATION RATE OF WATER TR EATMENT RESIDUAL (WTR) FOR AGRONOMIC AND ENVIRONMENTAL BENEFITS.....................................................142 Introduction................................................................................................................... ........142 Materials and Methods.........................................................................................................1 45 Results and Discussion......................................................................................................... 146 Soil Phosphorus Storage Capacity (SPSC) in the Glasshouse Study............................146 Soil Phosphorus Storage Capacity (SPSC) in the Field Experiment.............................149 Soil Phosphorus Storage Capacity (SPSC ) and Plant Growth in the Glasshouse Study.......................................................................................................................... 150 Summary and Conclusions...................................................................................................159

PAGE 8

8 7 EFFECTS OF A WATER TREATMENT RESIDUAL (WTR) ON RUNOFF AND LEACHATE PHOSPHORUS LOSSES...............................................................................161 Introduction................................................................................................................... ........161 Materials and Methods.........................................................................................................1 63 Rainfall Simulation Experiment....................................................................................163 Leachate and Runoff Analyses......................................................................................165 Statistical Analysis........................................................................................................... .....166 Results and Discussion......................................................................................................... 167 Runoff and Leachate pH and EC...................................................................................167 Runoff and Leachate P Forms and Concentrations.......................................................168 Runoff and Leachate Bioavailable P Concentrations....................................................171 Forms of Runoff and Leachate Phosphorus Losses......................................................173 Effect of P-Sources, P-Source Rates and WTR on BAP Losses...................................176 Summary and Conclusions...................................................................................................183 8 A METHODOLOGY TO ACCOUNT FO R P RELEASE POTENTIAL FROM DIFFERENT SOURCES OF P: FLORIDA P INDE X AS A CASE STUDY.....................185 Introduction................................................................................................................... ........185 Materials and Methods.........................................................................................................1 87 Statistical Analysis........................................................................................................... .....188 Results and Discussion......................................................................................................... 188 Runoff and Leachate P as Affected by P-Sources.........................................................188 Relative P Losses and P-source Coefficients................................................................192 Summary and Conclusions...................................................................................................200 9 FIELD VALIDATION OF ENVIRONMENTAL IMPACTS OF LAND APPLIED PSOURCES AND WATER TREATM ENT RESIDUAL (WTR).........................................201 Introduction................................................................................................................... ........201 Materials and Methods.........................................................................................................2 02 Statistical Analysis........................................................................................................... .....206 Results and Discussion......................................................................................................... 207 Groundwater Aluminum Concentrations and WTR......................................................207 Groundwater P Concentrations and WTR.....................................................................209 Soil P Sorption Indices and Gr oundwater P Concentrations.........................................214 Summary and Conclusions...................................................................................................216 10 SENSITIVITY ANALYSIS OF THE DRAFT FLORIDA P INDEX.................................217 Introduction................................................................................................................... ........217 Materials and Methods.........................................................................................................2 20 Results and Discussion......................................................................................................... 226 Sensitivity of the Draft Florida P Index Model to the Variables...................................226 Sensitivity of the Draft Florida P Index Model to P Management................................233 Sensitivity of the Draft Florida P Inde x Model to P-Source and P-Source Rate..........234

PAGE 9

9 Summary and Conclusions...................................................................................................235 11 MANAGEMENT OF PHOSPHORUS SOURCES AND WATER TREATMENT RESIDUALS (WTR) FOR ENVIRONM ENTAL AND AGRONOMIC BENEFITS........238 APPENDIX A PUBLICATIONS ARISING FROM THE DISSERTATION.............................................244 B OTHER RELEVANT TAB LES AND FIGURES...............................................................246 LIST OF REFERENCES......................................................................................................254 BIOGRAPHICAL SKETCH................................................................................................271

PAGE 10

10 LIST OF TABLES Table page 2-1 General chemical properties of amendments used for the studies.....................................43 2-2 Phosphorus characteristics of th e amendments used for the studies..................................46 2-3 Immokalee soil general chemical propert ies measured in 2001 from Kirton Ranch.........47 2-4 General chemical properties of the Imm okalee soil used for glasshouse and rainfall simulation studies............................................................................................................. ..48 3-1 Effects of P-source, sour ce application rates, and WTR on iron strip P values of soil samples taken during the glasshouse study........................................................................62 3-2 Effects of P-source, sour ce application rates, and WTR on Mehlich-1P values of soil samples taken during the glasshouse study........................................................................64 3-3 Effects of P-sources, source applicati on rates, and WTR on Total recoverable P values of soil samples take n during the glasshouse study.................................................66 3-4 Soil Total recoverable Al and Fe concen trations as affected by water treatment residual (WTR) and the P-source applica tion rates during th e glasshouse study..............72 3-5 Effects of P-source and source application rates on plant dry matter yields.....................81 3-6 Relative agronomic yield (RAY) va lues of the different treatments.................................83 3-7 Relative P phytoavailability (RPP) values fo r the different P-sources at each P-source rate by point estimate......................................................................................................... 85 3-8 Relative P phytoavailability (RPP) values for the different P-s ources by regression........87 3-9 Effect of aluminum water treatment re sidual on bahiagrass and ryegrass Al, Ca, and Mg concentrations.............................................................................................................. 89 4-1 Effects of water treatment residual on wa ter extractable P and ir on strip P values of A-horizon soils sampled between June 2003 and December 2004....................................98 4-2 Percent water extractable P (PWEP) va lues of soil samples taken during the glasshouse and field studies, as affect ed by WTR at the two P-source rates...................103 4-3 Effects of P-sources, P-source rates, a nd water treatment residual (WTR) rates, on plant P uptake values of bahiagrass harvested in 2004....................................................112 4-4 Relative P phytoavailability values for th e different P-sources during the field study...116

PAGE 11

11 4-5 Relative P phytoavailability values for the different P-sources from field and glasshouse studies............................................................................................................1 17 4-6 Stepwise regression of rela tive P phytoavailability values of biosolids with some of the biosolids properties....................................................................................................120 4-7 Summary of regression parameters for the variables selected by the stepwise regression of relative P phytoa vailability values of bios olids with some of the Psources properties............................................................................................................. 120 4-8 Effect of aluminum water treatment re sidual on bahiagrass Al concentrations and uptake during the field study............................................................................................122 5-1 Summary of phosphorus ex traction procedures used......................................................128 5-2 Correlations between varying measures of soil test P and plant parameters in the glasshouse study............................................................................................................... 134 5-3 Coefficients of determination (r2) and other regression parameters obtained by relating various soil test P values agains t plant data from the glasshouse study.............135 5-4 Coefficients of determination (r2) and other regression parameters obtained by plotting various soil test P values ag ainst each other (glasshouse study)........................137 5-5 PearsonÂ’s correlation coefficients between the different measures of soil test P of Ahorizon soils and plant data for 2003 and 2004 (field study)..........................................139 5-6 Coefficients of determination (r2) values obtained between varying soil test P measures for the two planting seasons during the field study.........................................140 6-1 Total and oxalate extractable phosphorus , aluminum and iron in water treatment residuals (WTR) used in some recent studies..................................................................143 6-2 Observed SPSC values of time zero soils at 0, 1, and 2.5% WTR and calculated amounts of WTR needed to achieve 0 mg SPSC kg-1.....................................................159 7-1 Masses of P forms lost in runoff and leachates................................................................175 7-2 ANOVA table of the effect of P-s ource, P-source rates, and WTR on runoff bioavailable P (BAP), leachate BAP, and Total BAP mass losses..................................177 7-3 Effects of P-source and P-source rate s on runoff and leachates BAP mass loss.............178 7-4 Effect of WTR and P-sources on leachate BAP mass loss..............................................180 7-5 Effect of WTR and P-sources rate on leachate BAP mass loss.......................................181 8-1 Masses of P lost in runoff and leachat es during rainfall simulation experiment.............189

PAGE 12

12 8-2 Regressions of total bioavailable P (B AP) loss with P applied for each P-source..........192 8-3 Total P and bioavailable P (BAP) losses a nd some indices of the P-sources solubility..193 8-4 Pearson correlation coefficients and p -values between P loss (bioavailable P (BAP) and total P (TP)) and various P-s ource solubility coefficients........................................195 8-5 Regressions of total bi oavailable P (BAP) mass loss with P applied for all the Psources in rainfall simulation experiment........................................................................196 8-6 Regressions of mass of P loss with P a pplied for P-sources used in a leaching study with two Florida soils.......................................................................................................19 6 8-7 Trends of draft Florida P Index scores (using different P source coefficients) and BAP loss....................................................................................................................... ....198 8-8 Percentage water extractable P (PWE P) values of some P-sources and the corresponding P-source coefficien ts based on the PWEP values....................................199 10-1 The draft Florida P Index worksheet...............................................................................218 10-2 Input values for each variable used in nominal range sensitivity analysis of draft Florida P Index................................................................................................................ .225 10-3 Sensitivity coefficients and swings and normalized values for each variable in the draft Florida P Index........................................................................................................22 8 10-4 Draft Florida P Index scores of plots treate d with four P-sources at two P rates in the field.......................................................................................................................... ........235

PAGE 13

13 LIST OF FIGURES Figure page 1-1 Illustration of agronomic and environmental threshold concepts.....................................25 3-2 Oxalate extractable P values of time zero and time final soil samples taken during the glasshouse study............................................................................................................... ..65 3-3 Trends of soil extractable P values as affected by time and WTR treatments...................67 3-4 Effects of WTR and P-source rates on (a) water extractable P (WEP)and (b) Iron strip extractable P (ISP) values of soil samples taken during the glasshouse study..........69 3-5 Effects of WTR and P-source rates on Mehl ich 1P and Total recoverable P values of soil samples taken during the glasshouse study.................................................................71 3-6 Effects of WTR and P-source rates on de gree of P saturations (DPS) values of soil samples taken during the glasshouse study........................................................................73 3-7 The water extractable P (WEP) values as affected by degree of P saturation (DPS) of soil samples taken during the study...................................................................................74 3-8 Soil phosphorus storage capacity values of time zero samples taken during the glasshouse study for the different treatments.....................................................................75 3-9 Effects of water treatment residual (W TR) and P-source rates on plant P uptake during the glasshouse study...............................................................................................77 3-10 Effects of water treatment residual (W TR) rates and P-source rates on plant P concentrations................................................................................................................. ...79 3-11 Effects of water treatment residual (W TR) and P-source rates on plant dry matter yields......................................................................................................................... .........80 4-1 Effects of water treatment residual (W TR) rates and P-source rates on Mehlich 1 P (M-1P) values of A-horizon soils at each of the sampling periods....................................99 4-2 Effects of WTR rates and P-source rates on Total P (TP) values of A-horizon soils at each of the sampling periods............................................................................................100 4-3 Effects of WTR rates and P-source rate s on water extractable P (WEP), Iron strip P(ISP), and degree of P sorption (DPS) valu es of A-horizon soils at each of the sampling periods..............................................................................................................1 02 4-4 Trends of soil water extract able P and Total recoverable P values for treatments with different P-sources at the two P rates in the absence of water treatment residual...........104

PAGE 14

14 4-5 Effects of WTR rates and P-source rate s on degree of P saturation (DPS) values of A-horizon soils at each of the sampling periods..............................................................105 4-6 Soil water extractable P (WEP) values as affected by degree of P saturation (DPS) values of soil samples taken between June 2003 and December 2004............................107 4-7 Soil phosphorus storage capacity (SPSC, mg kg-1) values of subsurface (E and Bh horizons) samples from the field study (June 2003 – Dec. 2004)...................................108 4-8 Soil phosphorus storage capacity (SPSC, mg kg-1) values of A-horizon (0-5cm) samples from the field study (June 2003 – D ec. 2004) as affected by the treatments.....109 4-9 Plant dry matter yields (Mg ha-1), yield-weighted P concentrations (g kg-1), and P uptake (kg ha-1) from 2003 and 2004 harvests.................................................................111 4-10 Effects of water treatment residual (W TR) rates and P-source rates on plant DM yield, P concentration and P uptake.................................................................................114 4-11 The relative P phytoavailability (RPP) as predicted by the Total P, NaOH-P, and %solids values of the biosolids plotted against the observed RPP..................................121 5-1 Extractable P values in samples take n during the glasshouse study as affected by application rates of P-sources a nd water treatment residual (WTR)...............................130 5-2 Relationships between extractable P valu es of samples taken during the glasshouse study and the effects of wate r treatment residual (WTR)................................................131 5-3 Effects of P-source rates and water treatment residual (WTR) on extractable P values of A-horizon (0-5cm) soil samples taken during the field study.....................................132 5-4 Regression of total P uptake and extractable P values of soil samples at planting of the grasses.................................................................................................................... ....136 5-5 Regression of water extractable P (WEP ) and iron strip P (ISP) values of soils sampled at planting of the three croppings......................................................................138 5-6 Relationships between extractable P valu es of soil samples taken during the field study and the effects of wate r treatment residual (WTR)................................................139 6-1 Soil phosphorus storage capacity values for the different treatments in time zero samples taking during glasshouse study..........................................................................147 6-2 Soil phosphorus storage capacity values at different rates of all P-sources and WTR during the glasshouse study over time.............................................................................148 6-3 Soil phosphorus storage capacity values of A-horizon (0-5cm) samples from the field study.......................................................................................................................... .......150

PAGE 15

15 6-4 Soil P storage capacity and water extract able P values of soil samples obtained during the glasshouse study.............................................................................................151 6-5 The soil phosphorus stor age capacity (SPSC, mg kg-1) and water extractable P (WEP, mg kg-1) values of soil samples obtained from the A-horizon (0-5cm) during the field study................................................................................................................ ...152 6-6 Plant dry matter yields and soil P storag e capacity values as a function of plant P concentrations for first bahiagrass cr op, ryegrass, and second bahiagrass crop..............155 6-7 Soil P storage capacity values as a func tion of plant P concentrations in samples taken during glasshouse and field experiment.................................................................157 7-1 National P Research proj ect (a) runoff box design and (b ) box design modified to collect runoff and leachate simultaneously......................................................................164 7-2 Leachate flow-weighted mean soluble reac tive P and Total P concentrations for the various treatments............................................................................................................1 68 7-3 Runoff flow-weighted P concentrat ions for the various treatments................................170 7-4 Flow-weighted runoff and leachate bioava ilable P (BAP) concentrations for the various treatments............................................................................................................1 72 7-5 Proportions of total P loss as bioa vailable P (BAP) from each treatment.......................176 7-6 Effect of WTR on runoff BAP mass loss.........................................................................179 7-7 Total mass of bioavailable P (BAP) lost during the thre e rain events.............................182 7-8 Total bioavailable P (BAP) loss with, versus without, WTR..........................................183 8-1 Bioavailable P (BAP) lost as affected by the P-sources at the two application rates......191 8-2 Regression of bioavailable P (BAP) loss with P Index score obtained using varying measures of P source coefficients....................................................................................197 9-1 Deep and shallow well positions under ground in the field study...................................203 9-2 Plot layouts and water sample rs locations in the field study...........................................204 9-3 One of the experimental blocks (rep licates) showing the water sampler and the telemetry installed on each of the 17 rows of plots in the field study.............................204 9-4 Flooding at Kirton Ranch (about 200 mete r away from experimental plots) as observed after hurricane Jeanne on September 26th, 2006...............................................206 9-5 Trends of Al concentrations in gra b, shallow well, and deep well water samples taken during the study......................................................................................................208

PAGE 16

16 9-6 Aluminum concentrations in water sa mples taken during the study as affected by surface applied WTR.......................................................................................................209 9-7 Regression of ortho-P concentrations in groundwater samples reanalyzed by three other laboratories with values obtained fr om the Analytical Research Laboratory.........210 9-8 Trends of ortho-P concentrations for th e various treatments in water samples taken during the field study.......................................................................................................21 2 9-9 Trends of total dissolved P concentrations for the various treatments in shallow, and deep well water samples taken during the study..............................................................213 9-10 Total and ortho-P concentra tions of water samples taken during the study as affected by WTR application.........................................................................................................214 9-11 Relationships between soil P sorption indices and shallow well water P concentrations................................................................................................................. .215 10-1 Number of map units associated with undrained runoff, draine d runoff and leaching potential categorized into very low, lo w, medium, high, and very high in Florida soils.......................................................................................................................... ........224 10-2 Spider diagram of variables in the draft Florida P Index.................................................226 10-3 Tornado diagram of variables in the draft Florida Pindex.............................................229 10-4 Nominal range sensitivity analysis matrix of the draft Florida P Index..........................231 10-5 Draft Florida P Index scores associated with a 50% reduction in each input factor when other factors are he ld at baseline values.................................................................233

PAGE 17

17 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy MANAGEMENT OF PHOSPHORUS SOURCE S AND WATER TREA TMENT RESIDUALS (WTR) FOR ENVIRONMENTAL AND AGRONOMIC BENEFITS By Olawale Olusegun Oladeji December 2006 Chair: George A. OÂ’Connor Cochair: Jerry B. Sartain Major Department: Soil and Water Science Land co-application of different sources of phosphorus (P) and aluminum water treatment residuals (Al-WTR) has potential as a best mana gement practice to reduce environmental hazard associated with excess soil P in low P-sorbing coastal plain sands. Accu rate knowledge of how P-sources, source application ra tes, and WTR affect soil P lo ss and agronomic returns can enhance sound management of the wastes in watersheds for agronomic and environmental benefits. Agronomic and environmental impacts of P-sources, source appli cation rates, and WTR were studied using four P-sources at two application ra tes and an Al-WTR in glasshouse, rainfall simulation and field studies. Applying P-sources at nitrogen (N) based rates will meet plant nutrient needs, while co-applying th e P-sources with WTR to 0 mg kg-1 soil phosphorus storage capacity (SPSC) will improve P sorption properties and reduce P hazard associated with N-based rates to that observed at P-based rates. Surface applied WTR effectively reduced P concentrations of groundwater samples of P treat ed plots below those obs erved in control plots without increasing either groundwat er Al concentrations or induc ing plant Al phytotoxicity. Soil soluble P and P losses associated with applying moderate water so luble P-sources were minimal. Thus, environmental P hazards associated with hi gh application rates (N-b ased) of P-sources can

PAGE 18

18 be managed by either applying the P-s ources with WTR to attain 0 mg kg-1 SPSC, or using a moderate water soluble P-source. The P-sources differ in P lo ss potentials and in relative P phytoavailability (RPP). Coefficients based on PWEP values of the P-sources are suggested to account for P loss potentials of diffe rent P-sources. Properties of P-sources, such as Total P, NaOH-P, and %solids, could affect the RPP of biosolids. Further studies will be needed to identify properties that could account for manure RPP. Sensitivity analysis of the drafted Florida P Index model indicates that all nine variables in the model are important, and all variables fell into either medium or higher impact categories. Studies are needed into all variables in the P Index and the use of continuous ratings for the variab les where possible. Use of more than 3 variables to account for wide sp ectrum of P-sources is recomme nded, and coefficients based on PWEP values of the sources should be considered.

PAGE 19

19 CHAPTER 1 INTRODUCTION AND LITERATURE REVIEW Introduction A major issue confronting agriculture in the USA is the environmental challenge of land applied phosphorus (P). The animal industry pr oduces nearly 160 million Mg (dry matter) of manure annually, which was estimated to resu lt in about 2 million Mg annual P excretion (Sweeten, 1992). In addition, approximately 132 billi on liters of waste wate r is treated daily in the US, resulting in an estimated 6.3 million Mg of biosolids production yearly. The mass of biosolids produced is projected to be 7.4 million dry Mg yr-1 by 2010 (United States Environmental Protection Agency [USEPA], 1999) . At an average P concentration of 25 kg P Mg-1, biosolids represent 159,000 Mg P yr-1, of which ~40% (63,500 Mg P yr-1) is land applied in USA alone. The decreased availability of landfills in the USA (from ~8000 in 1988 to 3090 in 1996) suggests that land applica tion of biosolids will increase to 48% by 2010 (USEPA, 1999; Mullins et al., 2005). Land application of manures and biosolids resi duals to meet crop nutrient needs is a major beneficial and economical method of disposal and accomplishes nutrient recycling. However, land application of the residuals based on crop N requirement (N-based rates) usually supplies P in excess of crop needs due to the lower N:P rati o in the materials than needed by the plants (Reddy et al., 1980; Pierzynski, 1994; Shober and Sims, 2003). At the N-based rates, organic sources of P can supply >5 times crop needs. Exce ss soil P is not harmful to plants, but offsite migration of P to surface waters is a concern, as P is the limiting nutrient for eutrophication of most freshwaters (Elliott et al., 2002a). Over 33% of US rivers, lakes, wetlands, and estuaries were reported degraded due to agricultural pract ices, and P losses from agricultural land have

PAGE 20

20 been implicated as one of the main causes of reduced water quali ty (USEPA, 2000; Boesch et al., 2001). Meeting the criteria for water quality w ithout compromising the agronomic benefits of residuals land application requi res proper management of the P-sources. This is especially important in the sandy soils of Florida that so rb P poorly and surround P-sensitive water bodies (Harris et al., 1996). Phosphorus Losses and Availability to Plants Phosphorus losses and availability to plants can vary with applied P-source, P-source rates, soil sorption properties, and management practices. The solub ility, bioavailability and transport potential of P varies among biosolids, manures a nd fertilizer types (Brand t et al., 2004; Leytem et al., 2004; Elliott et al., 2005). Wide variability in total P (TP) concentrations has been reported in biosolids (Keeney and Walsh, 1975; Dow dy et al., 1976; Sommers, 1977) and manures (Sommers and Sutton, 1980). Over 70% of the TP in residuals occurs in inorganic P forms (Gerritse and Vriesema, 1984; Sharpley and Moyer, 2000; Dentel et al., 2002). Organic forms, if present in significant amount in fresh manures, ra pidly mineralize to inorganic forms on storage (Peperzak et al., 1959; Gerritse, 1981; Crouse et al., 2002). Inorganic P forms are also readily available for uptake by algae and aquatic plants and represent immediate risks to the water quality. This explains why studi es on organic sources of P of ten focus on the reactions of inorganic P forms with soil components. Nutrient management to reduce P losses fr om residuals amended fields necessitates understanding and accounting for differences in th e phytoavailability of P in various P-sources. Accurate estimates of P phytoavailability may he lp tailor manure and biosolids applications to plant needs and thereby minimi ze the buildup of bioavailable P, which can degrade sensitive aquatic systems. Neutral ammoni um citrate extraction (Associat ion of Official Analytical

PAGE 21

21 Chemists (AOAC), 1995) is used to determine the guaranteed plant available P in commercial mineral fertilizer. However, Elli ott et al. (2005), showed that th e extractant failed to quantify biosolids P phytoavailability. A recent study by Zvomuya et al. (2006) suggests that the P availability of soil-applied composted and noncomposted manures can be predicted from the water extractable P and the tota l P concentrations. Phosphorus availability from manure and composts has been assessed using crop uptak e in pot, laboratory incubation, and field experiments (Eghball et al., 2002; Eghball et al., 2005). A 50% e ffectiveness of biosolids-P relative to fertilizer-P was suggested in the USEPA proce ss design manual (USEPA 1995), while 40% was recommended in Ontario, Canada regulati ons (Ontario Ministries of the Environment and Energy (OMEE), 1996). However, research indi cates P phytoavailability can vary from ~0% to 100% of fertilizer P (de Haan 1981; Smith et al., 2002; OÂ’Connor et al., 2004). Biosolids-P phytoavailabilty depends on the waste water treatm ent process used, which dictates the forms and solubility of P in the sources. OÂ’Connor et al. (2004) classified biosolids-P into three phytoavailability categories: high (>75% as available as fertilizer -P), moderate (25% to 75%), and low (<25%), based on a 4-month greenhouse study. The data need to be validated for longer periods, and preferably, in the fi eld. A longer study period (at leas t 1 year) is necessary because the P that is not bioavailable in the short term may ultimately be released by various biochemical processes. Understanding the contribution of P-sources to P loss potentials is also critical to mitigate P loss from agriculture to the environment. Coef ficients suggested to access P loss potentials of different P-sources include the Phosphorus Satu ration Index (PSI) (Ellio tt et al., 2002b), water extractable P (WEP) (Brandt et al., 2004; Wolf et al., 2005), Phosphorus Source Coefficient (PSC) (Leytem et al., 2004; Elliott et al., 2006), and varying measures of soil test P. The PSI was

PAGE 22

22 developed as an index of P-sources solubility and relates well with the P losses from varying biosolids P-sources (Elliott et al ., 2002b), especially for biosolids whose P solubility is controlled by Fe and Al. The index is calculated as ratio of a mmonium oxalate extractable P (Pox) to the sum of oxalate extractable Fe and Al (Feox, Alox): PSI = ([Pox] / [Alox+Feox]) (concentrations in mmol kg-1 biosolids) Conceptually, PSI is the molar ratio of to tal sorbable P to the sum of Al and Fe components capable of P fixation. A PSI value < 1 indicates excess P sorpti on sites in materials and minimal labile P available for leaching, wher eas PSI value > 1 indicates greater biosolids-P than the P sorption capacity of the materials. El liott et al. (2002b) found th at biosolids with PSI ~1.1 exhibited no appreciable P leaching, whereas a biosolids cake (PSI = 1.4) and its pelletized form (PSI = 1.3) exhibited significant leaching losses. The biosolids PSI appears useful for identifying biosolids with potential to enrich l eachate P when applied to low P-sorbing soils. However, soil with even a modest P-sorption cap acity may mask the differences in biosolids-P leachability. Also, the PSI is less useful to acco unt for the P leachability in the whole spectrum of the organic sources of P, es pecially materials rich in Ca, and Mg where P solubility is controlled by the Ca and Mg (e.g., manures). The phys ical state of the materials such as % solids and particle size can also affect the source P solub ility and the potential for P loss or availability to plants. OÂ’Connor and Sarkar (1999) indicated that pelletization of biosolids by heat drying resists degradation and reduced P release, and agree with Smith et al. (2002) that P availability is reduced by thermal drying of biosolids. Thus, there is a need for an index (or indices) of the Psources that integrates both physical and chemical properties to account for P lability.

PAGE 23

23 The importance of P-source lability to soil P loss is well appreciated by some of the P indices being developed by most States in US, in cluding Florida. The P-source coefficients and other variables in the P Index were assigned based on professiona l judgments of the scientists developing the model, and need to be validate d. While some states (e.g., Pennsylvania) have validated the P-source coefficients used in the P Index with experimental data, most P Indices including the Florida P Index, are yet to be validated. Determini ng the sensitivity of the draft Florida P Index to the P-source coefficients and other variables in the model will also be a necessary step towards validation of the model. Se nsitivity analysis not on ly indicates the impact of each variable (including P-source) on the P Index score, but can assist in identifying areas of priority in future research on P loss potential as estimated by the P Index. Improving the Florida P Index will ensure more accurate assessment of vulnerability of Florida landscapes to P loss and enhance management to reduce P losses to the environment. Phosphorus Application Rates and P Losses Accumulation of P in surface soils beyond levels needed for optimal crop yields often results when organic sources of P are applied at N-based rates (Reddy et al., 1980; Pierzynski, 1994; Peterson et al., 1994; Maguire et al., 2000). Soil P loads in excess of those needed for optimum crop production increase the potential for nonpoint P loss through runoff and leaching. Rechcigl et al. (1992) showed a strong correlation between P fertilizer surfac e-applied to bahiagrass ( Paspalum notatum Flugge) pastures and dissolv ed reactive P (DRP) in surface runoff. Thus, fertilizer-P was identifi ed as a major contributor to a 310 km2 algae bloom in Lake Okeechobee, FL in 1986. Runoff P concentrations decreased 33 to 66%, and total P losses in runoff decreased 17 to 78%, by reducing fertilizer-P application rates. Biosolids applied at Nbased rates to corn (Zea mays) added from 93 to 294 kg P ha-1 of which only ~25 kg P ha-1 was

PAGE 24

24 removed by the plants (Stehouwer et al., 2000). Studies in the past 30 years consistently implicate N-based rates with soil P build up and P loss to the environment (Kelling et al., 1977; Reddy et al., 1980; Kick, 1981; Pierzynski, 1994; Pe terson et al., 1994; Magui re et al., 2000). As P-sources differ in solubility, the environmental h azard of all P-sources is not expected to be the same. For example, P-sources of lower solubil ity might be applied at N-based rates without excessive P losses to the environment. This hypothesis, however, needs to be tested using suitable procedures such as rain fall simulations or field studies. Agronomic and Environmental Soil P Thresholds Several researchers have attempted to identify the critical soil test P (STP) levels above which P delivery to water bodies is unacceptabl e. These environmental STP thresholds are commonly based on readily available, agronomic soil P testing procedures. The environmental thresholds can be established based on the rationale that soil P in excess of crop requirements is susceptible to release in runoff and draina ge waters. However, equating agronomic and environmental thresholds can be inadequate as the processes by which crops access soil P are different from those that determine the susceptib ility of source-P to solubilization by subsurface leaching or surface runoff (Kleinman et al., 2000). Plants can solubilize soil water-insoluble P compounds and enhance P uptake by the production of organic acids in root exudates. Thus, estimates of adverse water quality effects of the so il P levels need not be directly inferable from the crop response (Sharpley et al., 2003). A developing consensus amon g researchers is that it is possible to maintain STP at levels that optimize crop yields, while minimizing the risk of offsite P transport (Higgs et al., 2000). Studies show that dissolved runoff P (DRP) is linearly related with STP in the topsoil (Pote et al., 1999; McDowell and Sharpley, 2001; Andr aski and Bundy, 2003). When a sufficiently

PAGE 25

25 wide range of STP levels considered, however, the relationship becomes curvilinear (Fig. 1-1) due to saturation of P sorption sites on the so il (McDowell and Sharpley, 2001; Elliott et al., 2004). The curvilinear relationship between DR P and STP can be described by exponential models, or by a simple split-line model that defi nes two linear sections, w ith a change in slope occurring at the so-called “cha nge point” (Elliott et al., 2004). The change point identifies the STP beyond which environmentally significant amounts of added P are expected to be mobilized by rainwater because soil P-sorbing sites have become sufficiently filled. Below the change point, much of the added P is retained by the soil (Fig. 1-1). Figure 1-1. Illustration of agrono mic and environmental threshold concepts (Elliott et al., 2004). The change point is typically ~3-4 times th e agronomic threshold. In Pennsylvania, no additional P is recommended for crop fer tilization beyond STP le vels of 50 mg kg-1 P (MehlichAgronomic optimum ~3-4 x agronomic optimum Soil Test P (STP) DRP (mg L-1) Change point Agronomic optimum (STP level sufficient for crop production added P not economically justified). Environmental threshold (STP level where soil P sorbing sites are largely filled added P potentially available for export)

PAGE 26

26 3), considered sufficient for crop producti on (Agronomy Guide, 2002). The environmental threshold, however, is considered to be 200 mg kg-1 STP (Sharpley et al., 2001). The state of Maryland defined 75 mg kg-1 Mehlich-1 P, which is three times the agronomic critical level, as the environmental STP threshold (Coale et al., 2002). Texas uses 200-500 mg kg-1 Bray-1 P (depending on site and watershed characteristics) as environmental soil test thresholds, beyond which no further P applications are allowed un til changes that could lower the siteÂ’s P loss potential. In Florida, Mehlic h-1 P values above 30 mg kg-1 are considered high from an agronomic standpoint, and a value above 60 mg kg-1 is considered very high (Kidder et al., 2002). Elliott et al. (2004) postulated that, on aver age, the environmental threshold is between three to four times the agr onomic optimum. Thus, reducing the soil test P by 75% with amendment (e.g., WTR) can minimize environmental hazard and optimize the agronomic benefits. Various chemical extractions (Mehlich I a nd III, Olsen, Bray, water or 0.01 M CaCl2, anion exchange resin, Fe-oxide st rip, etc) are suggested as envi ronmental soil tests for estimating labile P (Gartley and Sims, 1994; Simard et al., 1995). The relationship be tween extractable soil P concentration and dissolved P co ncentration in runoff water is not unique and varies with soil type (Sharpley, 1995) and the P-source. Thus, in risk assessment of P desorption for a range of soils, parameters related to an intensity factor (solution P concentration) and capacity factor (P sorption capacity) should be c onsidered (Beauchemin and Simard, 1999). The degree of P saturation (DPS) of soil surface is a promising variable to predict this risk (Sharpley, 1995; Beauchemin et al., 1996; Pote et al., 1996; Provin, 1996; Nair et al., 2004). The concept of DPS integrates both intensit y and capacity factors, as it measures the intensity of P accumulation while describing the poten tial of P to desorb from soil matrix into

PAGE 27

27 soil solution (Sharpley, 1995). Various ways of estimating this parameter could be good indicators of a soil‘s potential to release envir onmentally significant amounts of P. Relationships between water-soluble P (deionized water) and de gree of P saturation estimated from oxalate extracts (DPSox) gave a change point at 20% DPSox for manure impacted surface and subsurface Florida sands (Nair et al., 2004). The degree of P saturation calculated from 0.2M oxalate extractable P, Fe, and Al (DPSox) is also closely related to P concentrations in leachate wa ters (Leinweber et al., 1999; Maguire and Sims, 2002), suggesting that DPSox can be a suitable tool for predicting subsurface P losses. Soils with DPSox of >25% contributed to gr ound water pollution by P in the Netherlands (Breeuwsma et al., 1995). The 25% value corresponds to 0.15 mg total P L in ground water in the Netherlands. Values for DPSox of >30% in topsoils ha ve been identified as a threat to water quality degradation in Mid-At lantic U.S. soils (Paulter and Sims, 2000), and associated with increased P losse s in runoff (Pote et al., 1996). The University of Delaware rates soil s with Mehlich-1 P values >50 mg P kg-1 as excessive (Paulter and Sims, 2000). Relationshi p between M-1P values and DPSox, indicated M-1P concentration of 30 mg P kg-1 corresponds to a DPSox value of 22%, whereas a 60 mg P kg-1 value corresponds to a DPSox value of 28% (Nair et al., 2004). The study by Nair et al. (2004) suggest that DPSox value of 25% corresponds to 50 mg P kg –1 identified by Paulter and Sims (2000) to be excessive. The differences in the thre shold DPS values determined could result from using different -values in the calculation of the DPS. Paulter and Sims (2000) used an alpha value of 0.68 to calculate DPS, whereas Nair et al. (2004) used a value of 0.50. The alpha value for Spodosols of the Lake Okeechobee basin in Flor ida is 0.55 (Nair and Graetz, 2002), which is

PAGE 28

28 close to the 0.5 values used by Nair et al. (2004). Nair et al. (2004) also observed a change point at DPSox values of 16-24% (95% confidence interval) in Florida soils. Another index that could identify environmenta l thresholds is the soil P storage capacity (SPSC) values suggested by Nair and Harris ( 2004). The SPSC concept is an improvement on the DPS as it quantitatively indicates the P stor age capacity of a soil ( how much P could be safely added to a soil volume). While the enviro nmental threshold of the SPSC term is known to be at phosphorus saturation ratio (PSR) value of 0.15 (Breeuwsma and Silva, 1992; Nair and Harris, 2004), the agronomic thre shold has not been considered. The agronomic threshold is expected to be below the environmental thresh old and, hence, should be environmental friendly as a basis for P-sources land application. Phosphor us rates based on agronomic thresholds will be economically justified, as it will ensure applying the P-sources to meet plant needs and will keep the soil solution below the environmental threshold. Water Treatment Residuals (WTR) as Soil Amendments Numerous studies have been conducted in Florid a over the years utilizing a wide variety of amendments, amendment rates, soils, P-sources , and P loss mechanisms to identify best managements practices to reduce negative envi ronment P impacts on the aquatic systems (e.g., Allen, 1988; Anderson et al., 1995; Alcordo, et al., 2001; Matichenkov et al., 2001). There is increasing interest in using soil amendments to counter excess soil P and, hence, reduce dissolved P in runoff and leachate from manureand biosolids-amended soils. Recent work by OÂ’Connor and colleagues (OÂ’C onnor and Elliott, 2001; OÂ’Connor et al., 2002a) has shown water treatment residual (WTRs) to be eff ective soil amendments to immobilize and manage the excess soil P in Florida soils. The residuals, WTRs, are Al and or Fe rich waste products of municipal water treatme nt. The aluminum and iron salts added during

PAGE 29

29 water treatment hydrolyze to form amorphous meta l oxides that sorb organic matter, color, turbidity, phosphorus (P) and other wastewater constituents. Commonly, WTRs are land-filled; however, as landfill space becomes less available a nd more costly, land application is considered as a method of beneficial recycling WTRs that can also address P related water quality concerns. Studies by Moore and colleagues (Moore and M iller, 1994; Shreve et al., 1995; Moore et al., 2000) document effective cont rol of P solubility by Al ad ded to poultry manure. OÂ’Connor and Elliott (2001) also co-applied Alwater treatm ent residual (Al-WTR) with several biosolids, fertilizer, and two manures. They demonstrated almost complete control of P leaching through amended Florida sands initially low in P, regard less of P-source because soluble P levels were dramatically reduced in the soil/amendment mixtures. Laboratory studies (OÂ’Connor et al., 2002a) also showed that Al-WTRs adsorb large amounts of P, and that poorly Psorbing Florida soils could adsorb significantly more P followi ng amendment with modest amounts of Al-WTRs. The soil P retained by Al-WTR is irreve rsibly bound, barring unrealistic changes in environmental conditions (pH < 4) that dissolv e the WTR solid. Iron-based WTRs, or salts, can also effectively sorb P, but are subject to P release under reducing c ondition (Ann et al., 2000). Aside from P solubility control, other pot ential benefits of la nd applied WTR are: increased plant available nutrients (e.g., nitrogen and total organi c C) (Lin, 1988; Dempsey et al., 1989; Elliott et al., 1990; Elliott and Dempsey, 1991) , and increased aggregate stability, water retention, aeration, and drainage capacity (El-Swaify and Em erson, 1975; Rengasamy et al., 1980; Bugbee and Frink, 1985). The amorphous hydr ous oxides in WTR may also increase cation exchange capacity of coarse-textured soils (American Society of Civil Engineers et al., 1996), while alkaline stabilized WTR e.g., stabilized with CaCO3) can in addition serve as a liming agent. Such enormous benefits of land applied WTR and other similar BMP may need to

PAGE 30

30 be compensated for in P management tools such as Florida P Index. However, potential negative impact of the residuals (WTR ) needs to be evaluated before making a case for such compensation. Potential negative impacts of Al-WTR la nd application could include excessive immobilization of plant-available soil P and Al toxicity. Heil and Barbarick (1989) noted severe P-deficiency symptoms associated with 25 g WTR kg-1 soil planted to sorghum-sudangrass [ Sorghum bicolor (L.) Moench Sorghum X drummondii (Steudel) Millsp . Chase]. A decreased P concentration in blue grama ( Bouteloua gracilis (H.B.K.) Lag. ex Steud.) was found by Ippolito et al. (1999) when the rate of WTR was increased. Rengasamy et al. (1980) reported reduced P uptake in maize ( Zea mays L.) with WTR addition, while Ellio tt and Singer (1988) and Bugbee and Frink (1985) found reduced P concentrations in tomato ( Lycopersicon esculentum L.) and lettuce ( Lactuca sativa L.) grown in WTR-amended potting media. However, in a study by Naylor and Carr (1997), an Al-WTR (116 g Al kg-1) amendment reduced exchangeable P level in the soils, but did not limit plant growth. Brow n and Sartain (2000) also reported reduction in leaching P from a USDA green soil profile amended with 25 g kg-1 by weight with Fe-based WTR while maintaining adequate plant P uptake. The data suggest that WTRs can reduce P solubility in high P soils without inducing a P deficiency. Adding air-dried Al-WTR to soils at rates of 2 and 20 g kg-1 improved aggregation, but the hi gh application ra te decreased germination and decreased P uptake by maize, zea mays (Rengasamy et al., 1980). Yields of fescue grass ( festuca ovina ‘glauca’ ) grown in the greenhouse decr eased with increasing AlWTR application rates (0, 10, 20, and 40 g kg-1) to soil and the trend contributed to reductions in plant-available P that were corrected with suppl emental P fertilizer (Lucas et al., 1994). The possibility of reducing crop yield as a result of P deficiency following application of WTR calls

PAGE 31

31 for in-depth study into the appli cation rate of the am endment that will be environmentally and agronomically beneficial. A better understanding of the change in ch emistry of the soil as a result of WTR application and impacts on the pl ants is needed as the WTRs can affect soil reaction (pH), solubility of P, adsorption of P and speciation and distribution of other chemicals (notably Al). Soluble aluminum has been implicated as the most common source of phytotoxicity in acid soils (Arkin and Taylor, 1981), and a common yieldlimiting factor in acid soils. Aluminum is a phytotoxic element when present at excess conc entrations in soluti on. Cornell Recommends (1992) suggest that Morgan soil test aluminum values in the range of about 1 to 50 mg kg-1 are normal, with higher values being excessive, bu t not necessarily phytotoxic. Many soils exceed 50 mg kg-1 soil test aluminum and continue to remain productive. Aluminum c oncentrations can be sufficiently high in acid soils with pH values of <5.5 to be toxic to plants (Brady and Weil, 2002). The aluminum species (Al3+) responsible for the phytotoxic e ffect is often a small fraction of the total aluminum in the so il solution. Alum-treated litter or alum hydrosolids (similar to WTR) have neutral or alkaline pH, and Al exist as insoluble Al oxides, which should not release toxic Al or produce acidity in so il or aqueous systems (Peters a nd Basta, 1996). However, plants have mechanism (including releasin g of exudates) to assess soil nut rients and the impact of the applied Al-rich WTR on plants should be evalua ted, especially in Florida soils with pH < 5.5. Changes in basic soil properties th at affect nutrient availability to plants as a result of WTR application demand a better understa nding of the chemistry and suita bility of soil test methods for the available P that correlate well with plan t uptake. Inadequacy of STP as an estimate of plant response could lead to incorrect P manage ment for agronomic benefits. Reports of plant response as a function of soil test P methods applied to WTRs or WTR-amended soils yield

PAGE 32

32 conflicting results. Basta et al. (2000) evaluated th ree Al-WTRs as soil substitutes and the ability of soil tests to predict P adequacy for bermudagrass (c ynodon) . Soil tests indicated P deficiency for two of the WTRs and a contro l soil, and P concentrations in tissue grown on the unfertilized WTRs and soil were below adequate leve ls. Fertilization (50, 100 and 200 mg P kg-1) increased Bermuda grass yield and tissue P concentrations for the soil, but not for the WTRs. Watersoluble P and Olsen P were useful in predicting the ab ility of WTRs to support growth, but not P adequacy, while Mehlich-3 P (M3P) soil test overes timated plant available P in WTRs due to the dissolution of P adsorbed by amorphous Al. Water extracts were judged adequate to predict P adequacy in WTR-amended soil (Basta et al., 2000). However, these findings need to be investigated using additional so ils and different P-sources. Cox et al. (1997) conducted a greenhouse study to determine Al-WTR effects on inorganic forms of P and availability to wheat ( Triticum aestivum L.) in a thermic Aquic Hapludult. Of the inorganic P fractions st udied, loosely-bound (1 M NH4Cl-extractable) P was the best predictor of P availability in Al-WTR amended soil, but Mehlich-1 P (M-1P) was also a good indicator. However the suitability of M-1P may need to be studied especially in Florida sands. Another method to assess plant available P in WTR amen ded soils is the iron oxide filter paper method, sometimes referred to as " strip P" or the "Pi soil test " (Sharpley, 1991; Sharpley, 1993 a, b; Chardon et al., 1996; Pote et al ., 1996; Menon et al., 1997). The pr inciple involves an Fe-oxide strip acting as an "infinit e sink" for the P that can be desorbed from a soil and, thus, measures the potential of a soil to continue to release P to plants. Pote et al. (1996) found the method accurately predicted the quantity of P susceptible to runoff, and was better than most agronomic soil P tests. Sharpley (1993a) also reported that Fe-oxide "strip P" was a good indicator of the biological availability of P in runoff waters to algae.

PAGE 33

33 OÂ’Connor et al. (2002b) utilized a greenhouse study to determine the bioavailability of biosolidsand manure-P as compared with ferti lizer-P. Water extractable P and iron strip P were identified as potential P test methods for labi le P in biosolids and WTR amended soils where Mehlich 1 test failed. The suitabil ity of various soil tests is exp ected to vary with soil, soil reactions, P-sources, and P forms and could be mo re complex in WTR treated soils. Fertilizer P requirements can differ in WTR-amended and una mendeds soils, so careful selection of soil testing methods is necessary. Hypotheses and Research Objectives A good understanding of agronomic and environm ental impacts of P-sources, P or source application rates and P sorbing amendments su ch as WTR is necessary to derive best management practices (BMP) for P-sensitive areas in Florida. Conflicting reports of the impact of the different amendments on the plant nutrient uptake and yield could be resolved through the use of an appropriate soil test methods, es pecially when WTR is applied. Also, the environmental benefits of WTR application could be optimized, without compromising the agronomic importance, if applied at a rate to ta rget only the P in excess of plant needs. Thus it was hypothesized that: I. P-based rates of different organic sources of P, without WTR, optimize P uptake. II. N-based rates of different organic sources of P, with WTR, optimize P uptake. III. Indices exist to account for P phytoava ilability of different P-sources IV. Suitable soil test methods exist to access P bioa vailability in Florida sands receiving organic sources of P and WTRs.

PAGE 34

34 V. Amendment rates selected in (I and II) that optimize P uptake also minimize P leaching and runoff. VI. Indices exist to account for P loss potentials of different P-sources In summary the objectives of this study were to evaluate the environmental and agronomic impacts of different P sources and WTR and to determine the rate of P-sources and WTR that optimize environmental and agronomic benefits. The specific objectives of the studies are: I. Determine the rates of WTR and organic sour ces of P that optimize plant P uptake while minimizing environmental P hazards. II. Evaluate the impact of selected amendments (W TR and organic sources of P) rates on leaching and runoff P. III. Validate the expected impacts of selected amendments (WTR and organic sources of P) rates on P uptake and P loss in field settings. IV. Identify suitable soil test methods for P bioa vailability in soils am ended with different Psources and WTR. V. Evaluate the sensitivity of the draft Florida P Index model to P-source coefficients and other variables in the model. Study Approach Three (3) major experiments were ca rried out to test the hypotheses: a glasshouse study a rainfall simulation study and a field study The main objective of the glasshouse study was to study the agronomic impacts of Psources and WTR treatments. The rainfall simula tion experiment was designed to evaluate the impacts of organic sources of P, P-source app lication rates, and WTR on leached and runoff P.

PAGE 35

35 The field experiment was used to validate the results of the glasshouse and rainfall simulation experiments. The three studies used the same four P-sour ces, which are Boca Raton biosolids, Pompano biosolids, poultry manure, and Triple Super Phosphate (TSP). The sources were chosen to represent varying types of P-sour ces (manure, biosolids, and mine ral source) and to include high water soluble P biosolids (Boca Raton bioso lids), and medium water soluble P biosolids (Pompano biosolids). The material s were applied at low and hi gh rates (P-based and N-based rates in the glasshouse and field study). Two levels of WTR (0 and 20 Mg WTR ha-1, 0 and 1% oven dry basis, respectively) were surface-applied in the field and rainfall simulation studies. In addition to the 0 and 20 Mg WTR ha-1, a 50 Mg WTR ha-1 (or 2.5%) rate was used in the glasshouse study, and the materi als were soil-incorporated. Both field and glasshouse studies used bahiag rass as the test plant, while ryegrass was planted during the cool season in the glasshouse study. The rain fall simulation study did not include plants, as the National Phosphorus Research Protocol (National Phosphorus Research Project, 2001) was adopted. Details of the studi es are included in the subsequent chapters. Dissertation Format This dissertation is written as reports from the three major studies to address specific objectives of the study and in manuscript format intended for journal publications. Chapter 2 reports on the characteristics of the amendments and soils used in the three studies. Chapters 3 and 4 evaluate the agronomic impacts of the tr eatments in the glasshouse and field settings, respectively and thus tested hypothesis I, II and I II. An attempt to identify suitable soil test methods for WTR-treated Florida sands and test hypothesis IV wa s made using data from the glasshouse and the field study, and is reported in Chapter 5. The basi s for application rates of the

PAGE 36

36 WTR to ensure optimum agronomic benefits is discussed in Chapter 6. Chapter 7 reports on impacts of P-sources, WTR, and P rates on P losses (environmental impacts) in a rainfall simulation study and hence tested hypothesis V. Chapter 8 is the report on evaluation of coefficients that could account for P loss potential of different P-sources when land applied as tested by hypothesis VI. The impact of WT R on P losses was validated using ground water samples collected from the field study and are reported in Chapter 9. The Florida P Index is expected to assist in P-source management in Florida soils. Thus, a sensitivity analysis of the drafted Florida P Index is reported in Chap ter 10. Overall summary and conclusions are presented in Chapter 11.

PAGE 37

37 CHAPTER 2 CHARACTERIZATION OF AMENDMENTS AND SOILS USED IN THE STUDY Introduction Compositions of organic sources of P could be important from both environmental and agronomic perspectives, as they are expected to a ffect P availability to plants or P release to water via runoff and leaching. The chemical co mpositions (forms and amount of P and other elements) of the amendments used could assi st in understanding P reaction chemistry and availability after addition to so ils. Organic sources of P usually contain lower Total P, and soluble P concentrations, and a wi der variety of chemical elements, than mineral P-sources. The variation in elemental compositi ons of biosolids depends on th e treatment processes and the quality of the influent wastewater (OÂ’Connor et al., 2004). Manure composition also depends on the source (Sommers and Sutton, 1980; Sims a nd Wolf, 1994; Duo et al ., 2001), but is less variable than biosolids. The soils being amended with P-sources can also determine the extent of P losses. Florida soils are dominated by Spodosols, which occur al ong the hydrologic conti nuum from the uplands to aquatic systems. Most Spodosols are sandy, with water tables fluctua ting between the spodic subsurface horizon and the surface A-E horizon (S oil Survey Staff, 1996). Low P-sorbing, sandy surface soils with high water tables tend to prom ote rapid surface and subsurface flows, which enhance P transport (Allen, 1988). Substantial accumulation of P often occurs in the Al-rich spodic horizons, which can serve as sources or sinks for P, dependi ng on its degree of P saturation. Phosphorus, especially when organic s ources are added at N-based rates, can easily accumulate in, and exceed the sorption capacity of the surface soil and promote P transport to adjacent water bodies.

PAGE 38

38 This chapter describes the chemical and physic al characterization of the P-sources, WTR, and soils used in studies reported in subsequent chapters. Materials and Methods Amendments and Soil Selection Phosphorus sources used for the glasshouse, fi eld and rainfall simu lation studies were: Boca Raton biosolids Pompano biosolids Poultry manure Triple Super Phosphate (TSP) The sources chosen represent varying types of P-sources (manure, biosolids, and mineral fertilizer), and materials with different water soluble P concentrations. The two biosolids (high water soluble-P, Boca Raton biosolids, and mo derate water soluble-P, Pompano biosolids) represent biosolids that could supply excessi ve P when land-applied. All P-sources were obtained from Florida. The two biosolids, Po mpano and Boca Raton, were obtained from the cities of Pompano Beach and Boca Raton, FL, respectively. Poultry manure was obtained from Tampa Farms in Indiantown, FL., a large egg-la ying operation. Triple supe r phosphate (TSP) is a typical P mineral fertilizer a pplied to Florida crops. The WT R was obtained from a domestic water treatment plant in Bradenton, Fl. The site for the field study, Kirton Ranch, is a cattle pasture located on the eastern border of Okeechobee County, eleven k ilometers northeast of Okeec hobee, north of the Lake Okeechobee. The soil is Immokal ee fine sand, a typical Florida Spodosol. The Immokalee soil is classified in the Arenic Alaquods taxonomic group, and has distinct A, E and Bh horizons. Initial

PAGE 39

39 samples to characterize the soil were taken from the first 5 cm of the A horizon, the middle of the E horizon (~ 20-30 cm from the soil surface); a nd the Bh horizon (75-90cm) before amendment application. The top soil (0-15cm) of another sample of the Immokalee fine sand was used for the glasshouse experiments and the rainfall simula tion experiments. Bulk samples were collected from the Southwest Florida Research and Educa tion Center (SWFREC), wh ich is 3 km north of Immokalee, Florida. Amendments and Soil Analysis The soil samples from the field and the bulk so il taken for the glasshouse and the rainfall simulation studies were air-drie d, thoroughly mixed, and sieved (<2mm) before analysis. Both the soils and the amendments were analyzed for Total P, Fe, Al, Ca, Mg by inductively coupled argon plasma (ICAP) spectrometry, following digestion according to EPA Method 3050A (USEPA, 1986). Oxalate-extractab le P, Fe, Al, Ca, and Mg were determined by ICAP after extraction with solutions of 0.1 M oxalic acid pl us 0.175 M ammonium oxalate (pH = 3.0) at a 1: 60 solid:solution ratio, following the procedures of Schoumans (2000). The suspensions were equilibrated for 4 h in the dark with con tinuous shaking, centrifuged, filtered through a 0.45-µm filter, and analyzed for P, Fe, Al, Ca, a nd Mg by ICAP within 24 h after extraction. The Degree of P saturation (DPSox) values of the soils were calculated from oxalate extractable P Fe, and Al as: DPSox = [(Pox)/ (Alox + Feox)] where Pox, Alox, and Feox are oxalate extractable P, Al, a nd Fe concentrations in mmoles and = 0.55 (Nair et al., 2004).

PAGE 40

40 The soil P storage capacity (SPSC) values were also determined from the Pox, Alox, and Feox values of the soils as: SPSC (mg kg-1) = [(0.15-PSR)*(Alox + Feox)]*31 Where PSR = Phosphorus saturation ratio = [(Pox)/(Alox + Feox)] Biosolids PSI values were calculated using the same formula as for soil PSR values, but utilized Pox, Alox, and Feox of the biosolids. The amendment P storage capacity (APSC) values of the P-sources and the WTR (corresponding to SPSC) were calculated as: APSC (mg kg-1) = [(0.15-PSI)*(Alox + Feox)]*31 Where PSI = Phosphorus sorption index = [(Pox)/(Alox + Feox)] Total C and N of the amendments were de termined by combustion at 1010 °C using a Carlo Erba NA-1500 CNS analyzer. Total C was dete rmined on representative soil samples. Soil reaction (pH) was determined on fresh materials (1 :2 solid or soil:solution ratio). Percent solids were determined by drying materials at 105 °C (A merican Public Health Association, American Water Works Association, and Water Envir onmental Federation (APHA/AWWA/WEF), 1995) to constant weight. Soils and amendments were analyzed for elec trical conductivity (EC), Mehlich-1 P, Water extractable-P (WEP) and Iron strip-P (ISP). Wate r-extractable P was determined by extracting each soil sample with water at a 1:10 soil to wa ter ratio (1:200 ratio for amendments) for 1 h, and determining P on the filtrate collected after pa ssing through a 0.45-µm filter (Self-Davis et al., 2000). Iron strip-P determination i nvolved extracting solids in a cen trifuge tube containing a strip of filter paper coated with Fe-oxide (a strong adsorbent for P) in 0.01M CaCl2 (Chardon et al., 1996). The suspension was shaken with the Fe-strip paper for 16 h, and the P sorbed by the Feoxide on the filter paper was extracted by 0.1M H2SO4. Mehlich 1-P was determined by shaking

PAGE 41

41 the samples with 0.0125 M H2SO4 in 0.05 M HCl solution at a ratio of 1:4 soil:solution ratio for 5 minutes (Hanlon et al., 1997). Extracts we re immediately filtered through Whatman No 42 filter paper and analyzed co lorimetrically by the Murphy and Riley method (1962). Watersoluble-P and Iron strip-P concentrat ions were determined colorimetrically in each of the extracts with the Murphy and Riley (1962) procedure. Fractionation of the forms of P in the sources involved sequen tial extraction of the samples with KCl, NaOH, and HCl solutions, in that order (modified fr om Chang et al., 1983). The sequential extraction starte d by shaking the materials with 30 mL KCl solution for 2 h, after which the solution was centrifuged, filtered (0.45 µm), and analyzed for soluble reactive P (SRP) by Murphy and Riley (1962) colorimetric procedure. The residuals from the first extraction step were then extracted with 30 mL of 0.1M Na OH overnight for 17 h, filtered, and SRP measured in the extract. The last step of the extraction i nvolved shaking the residua ls from step 2 with 0.5M HCl for 24 h and analyzi ng the extract for SRP. Th e KCl fraction, considered exchangeable P, represents the readily availa ble P forms to plants. The NaOH –extractable P represents the Feand Al-bound P fraction that can buffer the soluble P forms, while the HCl extractable P is the Ca and Mg-bound P that can be important in soil reaction of manure and some biosolids amended soils. The sum of the three fractions (KCl-, NaOH-, and HClextractable P), is usually define d as inorganic P (Sui et al., 1999; O’Connor et al., 2004), though NaOH can extract some organic P. Standard QA/QC protocols were observed during the sample co llection, handling and chemical analysis. For each set of samples dur ing chemical analysis, a standard curve was constructed (r2 > 0.998). Method reagent blanks were a ppropriately used, as well as certified standards. A 5% matrix spike of the set was used to determine the accuracy of the data obtained

PAGE 42

42 (recovery ranging from 96-103%), and another 5% of the set to determine the precision of the measurements (duplicates). Analyses that did not satisfy the QA/QC pr otocol were rerun. Results and Discussion Amendments (P-sources and WTR) Characterization All organic sources of P had pH values of ~7.6 (Table 2-1), whereas TSP was slightly acidic (pH of 5.9); all pH values fell within th e typical pH range for soils (Bohn et al., 1985). Total P and WEP concentrations of the organic sources were greatest in Boca Raton biosolids (TP = 47.3 g kg-1; WEP = 5.52 g kg-1), representative of the hi gh end of P concentrations spectrum in biosolids produced nationally (K irkham, 1982; USEPA, 1995). The Boca Raton material is produced via “high rate activated sl udge” process similar to a biological P removal (BPR) process and the greater total P concentr ation and P lability were, thus, expected (O’Connor et al., 2004). The TP concentrations of biosolids produced nationally can vary from <1g kg-1 to >140g kg-1 dry weight basis (Keeney and Walsh, 1975; Dowdy et al., 1976; Sommers, 1977), but typically are 10 g kg-1 to 50 g kg-1 (Kirkham, 1982). The TP concentrations of forty-one biosolids used by Brandt et al. (2004) ranged between 3 g kg-1 and 40 g kg-1. Thus, the Boca Raton biosolids represents the greater so luble P member of the biosolids likely to be land-applied in FL, whereas the Pompano biosolids (TP = 26.2 g kg-1; WEP = 1.12 g kg-1) represents the moderate water soluble P members in the spectrum. Total Al + Fe concentrations for the Boca Raton biosolids (33 g kg-1), and the Pompano (42 g kg-1) biosolids are common values for biosolids not stabilized with Fe or Al salts (O’Connor et al., 2004). Manure had the least total P concentration (25.3 g kg-1), but a greater portion of the manure P was water soluble P than in the biosolids. The Boca Raton biosolids contained nearly five

PAGE 43

43 times more soluble WEP (5.52 g kg-1) than the Pompano biosolids (1.16 g kg-1), and about the same (4.57 g kg-1) as the poultry manure. However, among th e organic sources, th e percent of TP that is water extractable (PWEP) was great est in manure (18%) followed by Boca Raton biosolids (11%), and least in Pompano biosolids (4%). Table 2-1. General chemical properties of am endments (P-sources and the water treatment residual (Al-WTR)) used for the glasshous e, rainfall simulation, and field studies. <---------------------P-source-------------------> Parameters (units) Poultry manure Boca Raton Biosolids Pompano Biosolids TSP Al-WTR pH 7.7 7.6 7.6 5.9 5.5 C (g kg-1) 320 347 366 199 N (g kg-1) 27.0 ± 0.3† 50.4 ± 0.4† 43.1 ± 0.6† 0.7 C:N 11.9 6.9 8.5 % Solids 25.1± 0.1† 13.4 ± 0.04† 15.4 ± 0.1† 100 62.5 ± 2.2† WEP (g kg-1) 4.57± 0.16† 5.52 ± 0.18† 1.16 ± 0.08† 175 0.03 §PWEP (%) 18.1 11.7 4.43 83.7 0.65 Total P (g kg -1) 25.3 ± 0.3† 47.3 ± 2.3† 26.2 ± 0.2† 209 2.7 ± 0.7† Total Al (g kg -1) 0.9 ± 0.1† 9.3 ± 0.4† 9.2 ± 0.4† 10.0 98.7 ± 5.4† Total Fe (g kg -1) 1.5 ± 0.1† 24.3 ± 0.8† 32.8 ± 0.4† 15.7 6.1 ± 0.1† Total Ca (g kg -1) 102 ± 3† 27.5 ± 1.1† 47.0 ± 0.5† 137 1.5 ± 0.1† Total Mg(g kg -1) 5.8 ± 0.2† 10.0 ± 0.5† 4.1 ± 0.1† 6.2 0.40 ± 0.02† ‡Oxalate Ca (g kg -1) 0.06 ± 0.0† 0.05 ± 0.0† 0.33 ± 0.01† Oxalate Mg (g kg -1) 4.2 ± 0.1† 9.7 ± 0.0† 3.7 ± 0.2† 0.36 ± 0.01† Oxalate P (g kg -1) 12.7 ± 0.0† 34.0 ± 0.9† 20.4 ± 0.1† 186 2.3 ± 0.0† Oxalate Fe (g kg -1) 0.7 ± 0.0† 19.4 ± 0.5† 24.7 ± 0.2† 11.0 4.8 ± 0.3† Oxalate Al (g kg -1) 0.2 ± 0.0† 8.9 ± 0.6† 9.2 ± 0.0† 6.9 95.1 ± 1.3† ¶PSI 1.44 ± 0.02† 0.7 ± 0.02† 0.02 ± 0.0† ††APSC (g P kg -1) -12.6 -30.9 -16.8 -184 14.5 † Means of three samples ± standard deviation ‡ 0.2 M oxalate extractable § Percentage water extractable P (PWEP) = [(WEP mg kg-1) / (Total P mg kg-1)]*100 ¶ Phosphorus Saturation Inde x = [oxalate-P / oxalate-F e + oxalate-Al (in moles)] †† Amendments P saturation capac ity = (0.15 – PSI)* [oxalate-Fe + oxalate-Al (in moles)]*31 The PWEP values of the organic sources can reflect the chemical compositions, especially the total and oxalate P, Al and Fe concentrations . The Boca Raton biosolids, a BPR material with associated greater P and smaller Al and Fe conc entrations, had greater P solubility, and hence PWEP than Pompano biosolids (with smaller P a nd greater Al and Fe concentrations). The Al

PAGE 44

44 and Fe concentrations were smalle r in manure than in the two bios olids, but the Ca concentration of poultry manure was greater than in the biosolids, reflecting Ca-ri ch additives in animal feeds (Barnett, 1994). Thus, P chemistry in manure is likely controlled by Ca-P compounds that are more soluble at the soil pH (p H ~5.5) than the Fe and Al-P compounds that tend to dominate biosolids P chemistry. The greater oxalate P and sm aller oxalate Fe and Al concentrations of the Boca Raton biosolids resulted in a greater PSI = 1.44 and, hence, greater P solubility expected than in the Pompano biosolids ( PSI = 0.7). A PSI of 1.1 was identifi ed as a critical value in a glasshouse leaching study by Elliott et al. (2002b). Biosolids with PSI >1.1 resulted in much greater P leaching than those with PSI values <1 .1. Biosolids PSI values are consistent with the WEP and PWEP values (Boca Raton biosolids PSI = 1.44, WEP = 5.52 g kg-1, PWEP = 11%; Pompano biosolids PSI = 0.7, WEP = 1.16 g kg-1, PWEP = 4%), and suggest greater mobility of P in the Boca Raton biosolids than in Pompano biosolids. The PSI index is not applicable to manure and TSP, where P chemistry is expected to be controlled by Ca rather than Al and Fe (Elliott et al., 2002b). Total and oxalate Al and Fe concentrations were smaller in manure than in the two biosolids, but the Ca concentration was greater than in the biosolids. The WEP and PWEP values could serve as relati ve measures of solubility of th e P-sources, irrespective of the P chemistry dominance. The variations in the two m easures of P solubility of the P-sources support the need for a coefficient that relates the P losse s to the P-source solubility, irrespective of the P chemistry dominance. Total P and N concentrations were greatest in Boca Raton biosolids and least in poultry manure. The Al-WTR was slightly acidic (pH = 5.5) and dominated by Al (157 g kg-1), more than 90% (145 g kg-1) of which was amorphous (0.2M ammonium oxalate extractable; McKeague et al., 1971). The total Al value (157 g kg-1) was close to the range for typical Al-WTR (50150 g

PAGE 45

45 kg-1, ASCE, 1996). The WTR also contain significant P and Fe concentrations, ~90% of which were amorphous. Previous studies based on WEP and PWEP of WTR materials, showed that most of the P was not soluble, and the material se rved as a sink rather than as a source of P (O’Connor et al., 2002a). The very small PSI (0.02) of the WTR established its great P sorption capability. Also the positive A PSC value of the WTR (21g kg-1) identifies the material as P sink, in contrast to the negati ve values (-12.6 to -184 g kg-1) in the biosolids, manure, and TSP. Oxalate extractable P, Fe, and Al are usually associated with the amorphous phase of the particles. Oxalate extractable Fe + Al concen tration in the Boca Raton biosolids was 28 g kg-1, and 34 g kg-1 for the Pompano biosolids, well within the typical range (10-80 g kg-1) for biosolids (O’Connor et al., 2000). The sum of inor ganic sequential P fractionation values was ~ 70% of total P (Table 2-2), was close to the ox alate-P values, and typica l of biosolids produced nationally (Wolf and Baker, 1985). The NaOH-P (measure of Feand Al-associated forms) was the dominant fraction for the Boca Raton, and Pompano biosolids, as well as the Al-WTR (Table 2-2). The readily available P pool, KCl-P, varied among the di fferent P-sources and was approxi mately one third of the total inorganic P of the Boca Raton biosolids and the poultry manure (“high soluble P” sources), but only 6% of the total inorganic P of the Pompano biosolids (“moderate soluble P” source). The P sorption properties of the WTR resulted in a ve ry low KCl-P value, which represented only 0.3% (19 mg kg-1) of the total P, thus establishing the materi al as a P sink rather than as a P-source. Chemical characteristics of the native Immokal ee fine sand collected from the field site are given in Table 2-3. The three soil horizons are acidic (pH 5.5), and relatively low in organic carbon, ranging from 17 g kg-1 in the Bh horizon to 3 g kg-1 in the E horizon and 12 g kg-1 in the

PAGE 46

46 A horizon. The top soil was Pdeficient (M-1P < 10 mg kg-1), however the Bh had greater plant available P (M-1P = 12 mg kg-1). Table 2-2. Phosphorus characteristics of the amen dments (P-sources and water treatment residual (Al-WTR) used for the studies (values expressed in g kg-1) † Means of three samples ± standard deviat ion (for all data except % organic-P) ‡ Iron strip extractable P § Water extractable P ¶ Mehlich 1 extractable P Soil Characterization Total P values ranged from 7.9 mg P kg-1 (E horizon) to 24.5 mg P kg-1 for both the A and Bh horizons. The oxalate-P values of the Bh horizon (23.8 mg kg-1) were, however, greater than in the A horizons (10 mg kg-1). The E horizons contained the l east oxalate-P values (3.8 mg kg1). More than 95% of Bh-horizon P is amorphous , whereas <50% amorphous P is observed in A and E-horizons. This was similar to the trend of the oxalate-extractable Al values, which showed that the Bh horizon had the grea test amount (970 mg kg-1), versus 55 for the A and 16 mg kg-1 for the E horizon (Table 2-3). Greater oxalate extractable Fe (62 mg kg-1) was found in the A horizon than in the Bh horizon (39 mg kg-1, Table 2-3). Sequentially extracted P Source ‡ISP §WEP KCl-PNaOH-PHCl-P Inorganic P Total P Organic P (% of TP) Poultry Manure †3.90 ± 0.12 4.57 ± 0.16 3.91 ± 0.01 0.1 ±0.00 7.1 ±0.2 11.1 ±0.5 19.1 ±0.2 37 Boca Raton Biosolids 6.76 ± 0.02 5.52 ± 0.18 9.15 ± 0.02 11.7 ±0.07 8.2 ±0.8 29.1 ±0.7 34.7 ±0.3 4 Pompano Biosolids 2.30 ± 0.02 1.16 ± 0.08 1.14 ± 0.03 9.1 ±0.55 7.4 ±0.08 17.6 ±0.14 24.5 ±0.05 24 Bradenton Al-WTR 0.29 ± 0.04 0.03 ± 0.00 0.019 ± 0.00 4.0 ±0.11 0.4 ±0.04 4.4 ±0.14 5.6 ±0.04 28

PAGE 47

47 Table 2-3. Immokalee soil ge neral chemical properties† measured in 2001 from Kirton Ranch field study Horizons A (0-10cm) E (~ 20 -30 cm) Bh (75-85cm) pH 5.5 5.9 5.1 Total P (mg kg -1) 24.5 ± 5.46 7.9 ± 4.6 24.5 ± 11.3 Total Al (mg kg -1) 106 ± 23.3 33.6 ± 7.55 1280 ± 770 Total Fe (mg kg -1) 101 ± 39 38.5 ± 6.21 94.8 ± 23.2 Oxalate‡ P (mg kg -1) 10.0 ± 2.95 3.76 ± 2.76 23.8 ± 13.7 Oxalate Al (mg kg -1) 54.8 ± 6.74 15.8 ± 5.35 970 ± 418 Oxalate Fe (mg kg -1) 61.7 ± 8.52 13.0 ± 5.98 39.0 ± 5.08 Mehlich 1-P (mg kg -1) 7.03 ± 6.54 1.88 ± 1.33 12.1 ± 12.8 KCl-P (mg kg -1) 3.87 ± 0.78 0.77 ± 0.32 3.31 ± 0.52 NaOH-P (mg kg -1) 9.87 ± 1.23 5.49 ± 1.02 21.3 ± 3.2 HCl-P (mg kg -1) 3.78 ± 1.02 3.14 ± 0.88 4.56 ± 1.10 Fe-strip-P (mg kg -1) 9.6 ± 0.11 6.3 ± 0.12 16.4 ± 0.25 DPS§ (%) 20.6 29.7 4.16 SPSC¶ (mg P kg-1) 4.56 0.04 147 † Means of six samples ± standard deviation ‡ 0.2 M oxalate extractable §Degree of Phosphorus Saturation = [oxalate P / oxalate Fe + oxalate Al (in moles)]*100 ¶ Soil P saturation capacity (SPSC) = [(0.15-PSR)*(Alox + Feox)]*31 Where PSR (Phosphorus Saturation Ratio) = [(Pox)/(Alox + Feox)] The small DPS value of Bh-horizon soil ( 4.16%) reflects the greater amorphous Al concentration. The DPS value of the A-horizon (20.6%) was also below the 25% threshold DPS suggested for Florida soils (Nair et al., 2004) indi cating that the surface horizon is not impacted with excess P. Despite the low P concentra tions of the E-horizon, the DPS exceeded the threshold value because of relatively low Al a nd Fe concentrations and, hence, low P sorption. The data suggest that P leached into the E-hor izon would move freely through. As expected, soil phosphorus storage capacity (SPSC) values va ry inversely with DPS values, and the interpretation is similar to that offered for D PS values. The top soil used for the glasshouse and the rainfall simulation studies had low extractabl e P values (M-1P, WEP and ISP), which were similar to values for the sample taken from the A-horizon in the field (Table 2-4).

PAGE 48

48 Table 2-4. General chemical properties of the Immokalee soil (0-5cm) used for glasshouse and rainfall simulation studies. Parameters (units) Value pH 5.5 EC (µs cm-1) 323 C (g kg -1) †12.0 ± 0.1 Mehlich 1-P (mg kg -1) 6.40 ± 0.35 WEP (mg kg -1) 2.88 ± 0.19 Fe-strip-P (mg kg -1) 3.11 ± 0.48 Total P (mg kg -1) 24.5 ± 2.1 Total Al (mg kg -1) 88.4 ± 4 Total Fe (mg kg -1) 107 ± 20 Total Ca (mg kg -1) 449 ± 8 Total Mg (mg kg -1) 36 ± 3 Oxalate P (mg kg -1) 22.6 ± 1.6 Oxalate Al (mg kg -1) 49.8 ± 3.6 Oxalate Fe (mg kg -1) 96.0 ± 5.3 Oxalate Ca (mg kg -1) 34.8 ± 2.5 Oxalate Mg (mg kg -1) 20.4 ± 1.8 DPS‡ (%) 41.0 SPSC§ (mg P kg-1) -6.05 † Means of three samples ± standard deviation (for all data except pH, EC, DPS, and SPSC) ‡Degree of Phosphorus Saturation [oxalate-P / 0.5*[oxalate-Fe + oxalate-Al (in moles)]*100 §Soil P saturation capacity = (0.15 –PSR¶)* [oxalate-Fe + oxal ate-Al (in moles)]*31 ¶Phosphorus Saturation Ratio = [oxalate-P /(oxalate-Fe + oxalate-Al (in moles)] Soil M-1P of <10 mg kg-1 is considered very low for agronomic crops, including bahiagrass (Kidder et al., 2002). The low P makes th e soil suitable for the P response experiment, and for testing impacts of different P-sources an d WTR application rates on plants and P losses. Plant response to added P and other treatments should be easily identifi ed in an initially P deficient soil. The pH 5.5 coincides with the socalled “target” pH for bahiagrass, thus, making it suitable for the growth of bahiagrass (Hanlon et al., 1990). Contributions of the native soil to P losses are expected to be negligible, making tr eatment impacts on the runoff and leachate P more pronounced and more easily identified in the rainfall simulation study.

PAGE 49

49 CHAPTER 3 AGRONOMIC IMPACTS OF LAND APPLIE D WTR AND DIFFERENT P-SOURCES Introduction Land application of organic amendments is supported by USEPA 40 CFR Part 503 (USEPA, 1995) and other environmental agencies worldwide as long as the amendments are applied at agronomic rates based on crop N-re quirements (N-based). However, the N-based application of manures and biosol ids often supplies P to soil in excess of that removed by plants. The excess P accumulates in the soil (Pierzynski, 1994; Maguire et al., 2000), and is often subject to offsite migration to surface wate rs. Amendment applicati on rates based on crop phosphorus needs (P-based) dictate substantiall y lower P-source application rates and less potential for P loss. The lower rates, however, are economically unattractive because they require supplemental N-fertilizer application, larger di sposal areas to accommodate the same amount of amendments, and higher cost to transport the mate rials to additional land from outside sensitive watersheds. Phosphorus pollution of waters is a major con cern in Florida and other coastal plain soils with low-P retention capacities. Low P-retention, coupled with the characteristic flat topography and interception of shallow ground waters by disc harge systems, favors th e eventual entry of leached P to surface water bodies. Thus, contro l of excess soil soluble P resulting from amendments applications that ex ceed plant needs and in P-impacted soils is very important in Florida. Among the measures being suggested to reduce environmental P losses is the use of Alrich water treatment residuals (WTR) to increase affinity for soluble P. Possible negative impacts of WTR land application include excessive immobilization of plant-available soil P and Al

PAGE 50

50 toxicity. Severe P-deficiency symptoms were noted by Heil and Barbarick (1989) with 25 g WTR kg-1 soil planted to sorghum-sudangrass [ Sorghum bicolor (L.) Moench Sorghum X drummondii (Steudel) Millsp. Chase]. Ippolito et al. ( 1999) also found decreased P concentrations in blue grama ( Bouteloua gracilis (H.B.K.) Lag. ex Steud.) with increasing WTR rates. Rengasamy et al. (1980) reported reduced P uptake in maize (Zea mays L.) with WTR addition, while Elliott and Singer (1988) a nd Bugbee and Frink (1985) found reduced P concentrations in tomato (Lycopersicon esculent um L.) and lettuce (Lactuca sativa L.) grown in WTR-amended potting media. Exch angeable P was measured in soils amended with sewage biosolids and WTR (Naylor and Ca rr, 1997). The Al-WTR (116g Al kg-1) amendment reduced exchangeable P level in the soils, but did not li mit plant growth, suggesting that WTRs may be useful for reducing P solubility in high P soils without inducing a P defici ency. Rengasamy et al. (1980) also reported improved aggregation by adding air-dried Al-WTR to soils at rates of 2 and 20 g kg-1, but the high application rate decrea sed germination and P uptake by maize, zea mays . Yields of fescue grass ( Festuca ovina ‘glauca’) grown in the greenhouse decreased with increasing Al-WTR application rates (0, 10, 20, and 40 g WTR kg-1) to soil and attributed to reductions in plant-available P due to excessive P immobilization. The de ficiency was corrected with supplemental P fertilizer (Lucas et al., 1994 ). The possibility of P deficiency, reduced crop yield, and Al phytotoxicity following land applicati on of WTR, calls for an in-depth study into the best management of the amendment that will not induce negative agronomic impacts. Solubility of WTRs rich in amorphous Al and, hence, dissolution of P, adsorption of P and speciation and distribution of other chemicals ca n be affected by soil pH. Soluble aluminum is the most common source of phytotoxicity in acid soils (Arkin and Taylor , 1981), and Al toxicity is one of the yield-limiting factors identified in ac id soils. The Al toxic ity in soil could cause

PAGE 51

51 shallow rooting, drought susceptibility and poor soil nutrients usage by plants. The aluminum species, responsible for the phytotoxicity, Al3+, typically is a small fraction of the total aluminum in the soil solution. Corn ell Recommends (1992) suggest that a soil test (e.g., Morgan) aluminum value in the range of about 1 to 100 kg/ha is norm al, with higher values being excessive, but not necessarily phytotoxic. Indee d, many soils exceed 100 kg ha-1 soil test (exchangeable) aluminum and continue to remain productive. Aluminum con centration can be sufficiently high in acid soils with pH values < 5.5 to be toxic to plants. Alum -treated litter or alum hydrosolids (similar to WTR) have neutral or alkaline pH, and the result ing insoluble Al oxides do not release toxic Al or produce acidity in soil or aqueous systems (P eters and Basta, 1996). The lack of Al toxicity needs to be confirmed for other soils, in cluding Florida soil s with pH values 5.5. Another aspect of organic amen dments usage that needs consideration is the effectiveness as P-sources for agronomic benefits. A 50% e ffectiveness of biosolids-P compared with fertilizer-P was suggested by USEPA proce ss design manual (USEPA, 1995), while 40% was recommended by Ontario, Canada regulations (O MEE, 1996). Short-term studies have shown biosolids-P phytoavailability can vary from ~0% to 100% depending on properties of the sources (de Haan, 1981; Hani et al., 1981; Smith et al., 2002; OÂ’Connor et al., 2004). Additional study is needed for longer periods in Florida sands and in the absence of leaching. The longer period of the study will allow determining how the relative phytoavailability changes with time (residual effects), and differences in P accessible by plants from each P-sources over time will be prevented when no leaching is allowed. The purpose of this study was to evaluate agronomic impacts of various P-sources and WTR applied to Florida sands. Results are e xpected to enhance consistent and accurate

PAGE 52

52 fertilization decisions for Florida sands receivi ng different sources of P and WTR and help avoid reduced crop yields due to excessive P immob ilization and Al toxicity. We hypothesized that: 1. (a) P-based rates of differen t organic sources of P, withou t WTR, optimize agronomic yield and P uptake. (b) N-based rates of different organic sources of P, with WTR, optimize agronomic yields and P uptake. 2. Amendment rates (P-sources and WTR rates) se lected in (1) that optimize P uptake also minimize soil soluble P. 3. Organic sources of P vary in P bioavailability 4. Land application of Al-WTR incr eases plant Al concentrations. The main objective of the study was to eval uate the agronomic impacts of different Psources and WTR, and the specific ob jectives of the studies were to: 1) Determine the rates of WTR and organic sour ces of P that optimize agronomic benefits, while minimizing soil soluble P that could pose environmental hazards. 2) Evaluate the impacts of selected amendments (WTR and organic sources of P) on soil Psorption properties. 3) Determine the relative P phytoavailabilities of different P-sources. 4) Evaluate the impacts of Al-WTR on plant Al concentrations. Materials and Methods Experimental Procedure Each of the four P-sources (poultry manure, Bo ca Raton biosolids, Pompano biosolids, and TSP) was applied to the P-deficient Florida soil at two rates (N and P plant requirement basis). Each treatment also received WTR app lications at 3 rates (0, 10 and 25 g kg-1 oven dry basis). Thus, the glasshouse pot experiment was a 4X2X 3 factorial experiment with 1 control and

PAGE 53

53 arranged in randomized complete block design wi th 3 replicates. Soil ( 8.5 kg) and appropriate amounts of the amendments were weighed and thoroughly mixed in a polythene bag. Water was added to bring the mixture to field capacity, and the treated soils allowed to equilibrate for one week with daily mixing. Weights of P-sources needed to supply the equivalent of 44 kg total P ha-1 (P-based application rate) and 179 kg plan t available nitrogen (PAN) ha-1 (N-based rate), as recommended for bahiagrass (Kidder et al., 1998) , were calculated from total P and N concentrations of the Psources. Mineralization rates of 40% of total N in biosolids and 60% of total N in manure were assumed in the calculation, based on previous experience in simila r studies (O’Connor and Sarkar, 1999; O'Connor et al., 2004). Deficits in PAN (for the P-based rates) between the N provided by various P-sources app lied and the target PAN levels were calculated and supplied by split (monthly) applications of NH4NO3. Twice the P applied in P-ba sed rates (88 kg total P ha -1) was used as P supplied by TSP at N-based rate. The intent was to fix the P supplied at the Pbased rates, whereas the P supplied by the N-base d rates varied with P-so urces and all treatments received equal amounts of N (179 kg PAN ha-1). An amount (1.8 g, equivalent to 444 kg ha -1) of potassium-magnesium sulfate ("Sul-Po-Mag": 22 % S, 18% K, and 11% Mg) was added to each treatment to provide adequate and uniform S, K, a nd Mg. Amounts of NH4NO3 needed to supplement the control treatment to 179 kg PAN ha –1 were also added to en sure that all pots had adequate readily available N. Samples of the amended soils were taken afte r the one week of equilibration in June 2004 for analysis (Time zero samples). The remaini ng soil was placed in a plastic pot (6.5 X 103 cm3) and planted with first bahiagra ss at a depth of 3 mm and seed ing rate of 7g per pot. The soil surface of each pot was covered with filter paper to reduce ev aporation, and moistened daily,

PAGE 54

54 until seed germination. After germination, the pa pers were removed and soil wetted daily, and returned to initial weight week ly. Water with pH adjusted to 5.0 was used throughout the study to eliminate poor growth of ba hiagrass in high pH soil. There was non-uniform germination of the bahiagrass despite the caref ul nurturing described, and mi ssing areas of the pots were reseeded after one week. Because of the prob lems encountered during establishment of the plants, the first harvest was done 36 days after removing the filter pape rs, whereas subsequent harvests occurred monthly. Harvesting was at a height of 5 cm above soil surface with scissors or electric clippers. Cuttings were placed in a pre-weighed labeled pa per bag for drying to constant weight at 65 0C, and dry matter (DM) weight determined as the differences between the dried paper bags with cuttings and pre-weighed empt y bag. After each harvest, plant pots were weighed and watered as necessary after adding th e supplemented N (split applied) as necessary to return to initially determined pot weight s. The randomized pots were shifted by a position twice weekly to minimize positioning advantage in the glasshouse. All treatments were supplied wi th adequate N and other nutrient elements except P throughout the study and each pot planted with past ure grasses continuously for fifteen months (bahiagrass during the warm season and ryegrass during the cold season). The order of grass planted was bahiagrass ( paspalum notatum Fluggae) for six months (between June 2004 and December 2004), ryegrass ( Lolium perenne L.) for five months (between December 2004 and May 2005), and a second bahiagrass cropping for four months (between May 2005 and September 2005). The extended growing season a llowed studying the residual effect of the WTR, the P-sources and source application rate s on agronomic P use efficiency. Also, mining the soil P for extended periods will ensure P deficiency, which is often accompanied by Al toxicity (if it is an issue) esp ecially when soil pH is below 5.5.

PAGE 55

55 A total of four monthly cuttings were obtaine d from the first bahiagrass cropping. After the fourth harvest of the first bahiagrass cropping, soil samples were taken in December 2004 from the center of each pot using an auger of 5 cm diameter and extending to the bottom of the pot. The hole created was filled with time zero soil preserved for that purpose. Each pot was then planted with ryegrass (3g seed pot-1), a cool season grass. Continuous planting is necessary to evaluate the residual effects of the amendments. Soil obtained afte r the fourth bahiagrass harvest served as time zero soil for the ryegrass cropping for analysis purpose. The management of the ryegrass was the same as for the bahiagrass, and the grass was harvested three times (approximately monthly). Additional soil samples were taken in May 2005 after the ryegrass final harvest. Previous samp ling points had been marked to prevent sampling the same spot twice. The hole created was again filled with soil preserved from the initial time zero soil samples of each treatment. Another ba hiagrass crop was then planted with similar management for four months (3 harvests) betw een July and September 2005. Final soil samples were taken in September 2005 and the experiment terminated. No leaching was allowed throughout the study so as to mi nimize P and other nutrient losse s and to enhance studying the long term P phytoavailability of the different P-sources. Soil and Plant Analysis All sets of soil samples taken during the study (in June 2004, December 2004, May 2005 and Sept. 2005) were analyzed for pH and EC (1:2 solid:solution), total r ecoverable P, Al and Fe (USEPA, 1986) and 0.2M oxalate extractable P, Al and Fe (Schoumans, 2000). The same procedures used for the initial soil characteri zation (Chapter 2) were employed. Extractable P determinations (Mehlich-1 P, WEP and ISP) we re also measured as described in Chapter 2.

PAGE 56

56 Dried monthly plant cuttings weighed for dry matter determinations were ground in a Wiley mill with stainless steel blades to pa ss a 20mm-mesh sieve and stored in airtight polyethylene containers. Ground plant samples were ashed, treated with 6M HCl, and brought to final volume with distilled water as described by Plank (1992). Phosphorus in diluted digests was determined colorimetrically (Murphy a nd Riley, 1962). The plant uptake (kg P ha-1) was calculated as the product of P con centrations and dry matter weights. Weighted means of plant P concentrations were obtained by di viding the total P uptake by the total dry matter weight for all the harvests of each cropping. Aluminum, Ca, Mg, and Fe concentrations in the diluted plant digests of the N-based TSP treatments were determined usi ng ICAP to evaluate the impact of the applied WTR on the Al and elemental concentrations of the plants. Standard QA/QC protocols were observed during the sample co llection, handling and chemical analysis. For each set of samples dur ing chemical analysis, a standard curve was constructed (r2 > 0.998). Method reagent blanks were a ppropriately used, as well as certified standards. A 5% matrix spike of the set was used to determine the accuracy of the data obtained and another 5% of the set was used to determin e the precision of the measurements (recovery ranging from 96 – 103%). Analyses that did no t satisfy the QA/QC pr otocol were rerun. Statistical Analysis Soil and plant data were analyzed by an alysis of variance (ANOVA), using the GLM procedure in SAS (SAS Institute, 1999). The means were separated by single degree of freedom contrast procedures or the Tuke y method. Regressions of soil extr actable P with plant yields, P concentrations, and P uptakes were done using SAS. All statistical analysis tests were done using a significance level of 5%.

PAGE 57

57 Results and Discussion Soil pH and EC During the Study The pH of soil sampled during the study varied with the treatments and time (Appendix Fig. 3-1). Manure treatments were slightly basi c (pH values = 7.1 to 7.4) at the N-based rate (11.1 Mg ha-1) but slightly acidic (pH values = 5.6 to 6.6) at the P-based rate (1.74 Mg ha-1 of manure). The greater pH in manure treatments wa s also noted by O’Connor et al. (2000) when poultry manure was applied at 12 Mg ha-1 to a similar soil. The soil pH of treatments receiving other P-sources (biosolids and TSP) were acidic (4.9 -6.6) and similar to the pH of the control (4.9 – 5.7). Thus, the biosolids and TSP treatment s had less impact on the soil pH than manure treatments at the N-based rate. The greater soil pH values in treatments receiving poultry manure relative to other P-sources at both Pand N-based rates ma y have resulted from calcium carbonate-containing additives in the poultry feeds. Application rates of different P-sources also affected the soil pH. Greater soil pH at Nbased than P-based rate was observed by contrast s for all the organic sources treatments at time zero, whereas TSP treatments showed the opposite tre nd, i.e., lower pH values at N-based than at P-based rates. The greater soil pH as a result of manure application at N-ba sed rate than in other P-sources could reduce the growth of the acid tolerant bahiagrass, which has a target pH of 5.5 (Kidder et al., 2002). Soil pH increased with increasing WTR rates fo r each of the P-sources applied at both Nand P-based rates. The WTR, though lower in Ca concentration, had a su bstantial amount of Mg, and lime added during WTR produc tion could also raise the soil pH. Other studies (Bugbee and Frink, 1985; Codling et al., 2002) reported similar pH increases with WTR or alum sludge applied to soils. The greater soil pH at N-based ra tes than P-based rates at each of the WTR rates

PAGE 58

58 was apparently due to greater ma terial loads at N-based rate. Th e materials applied at N-based rates (11.1, 10.4, and 8.66 Mg ha-1 for manure, Boca Raton bioso lids and Pompano biosolids, respectively) were more than six times the ma sses applied at the P-based rates (1.74, 1.24, and 1.68 Mg ha-1 for manure, Boca Raton biosolids and Pomp ano biosolids, respectively). Thus, both WTR rate and rate of P-sources increased soil pH. Generally, soil EC values values were greater at N-based rates than P-based rates for all Psources tested. Also, similar EC values were obs erved for different P-sources at P-based rate treatments, but at N-based rate, soils treated with organic sources of P have greater EC values than mineral P-source treated soils. The soil EC values at P-based rate of all P-sources, irrespective of amount of WTR added, were similar to the control throughout the study. Generally, the EC values were greater at the Nbase rates with WTR than in the control. The greatest soil EC values of each sampling period was observed at N-based rate of manure with 2.5% WTR (Appendix Table B-2), whic h could result from the greater soil load of the material rich in residues from poultry feeds additives especially in this study with no leaching allowed. However, the EC values were below 800 µs cm-1, and hence, fall within the tolerance range of the test plants (Brady and Weil, 2002) . Bahiagrass can tolerate 7500 µs cm-1 (Bogdan, 1977), while ryegrass is tolerant to EC value not greater than 8000 µs cm-1 (Brady and Weil, 2002). Most importantly, there were no sa lt effects observed on the growth of the plants throughout the study. Soil Phosphorus During the Study Both measures of soil soluble P, WEP and I SP, exhibited similar trends with applied treatments, and clearly reflected WTR treatment effects. Soil WEP was affected by P-source, source application rates, a nd WTR rates (Fig. 3-1).

PAGE 59

59 Figure 3-1. Effects of P-source, source application rates, and WT R rates on water extractable P (WEP) values of soil samples taken during the glasshouse study. Note the differences in scales of y-axis. (Treatments with the sa me letter are not different at 5% significant level by Tukey test) (a) June 04 samples 0 10 20 30 40 50 ControlMan-PMan-NBoca-PBoca-NPomp-PPomp-NTSP-PTSP-N TreatmentsWEP (mg kg-1) 0% WTR 1% WTR 2.5% WTR gh fg gh gh gh gh gh gh h gh ghgh h gh fg fg ef fgh e bc a b cd b de (b) Dec. 04 samples 0 5 10 15 20 25 ControlMan-PMan-NBoca-PBoca-NPomp-PPomp-NTSP-PTSP-N TreatmentsWEP (mg kg-1) a b b bc bcd bcd cde ef cdef cde cdefdef cdef ef def f f ef f ef f f ef cdef def (c) May 05 samples 0 5 10 15 20 25 ControlMan-PMan-NBoca-PBoca-NPomp-PPomp-NTSP-PTSP-N TreatmentsWEP (mg kg-1) a bbc dede ede e cde de e de de cde bcde cde bcde bcde bcde bcde de de de de bcd (d) Sept. 05 samples 0 5 10 15 20 25 ControlMan-PMan-NBoca-PBoca-NPomp-PPomp-NTSP-PTSP-N TreatmentsWEP (mg kg-1) b b b b b b bb b bb b b b b b b b b b a ab b b b

PAGE 60

60 Adding WTR reduced soil WEP values of sa mples taken throughout the study (June Â’04, Dec. Â’04, May Â’05 and Sept.Â’05) for all P-sources and at both source appl ication rates. In the absence of WTR, the absolute WEP values were gr eater at the N-based than at the P-based rate of all P-sources. However, the pr esence of WTR resulted in similar soil WEP values for soils at both source rates. Thus, the potential hazards of excess soil soluble P in soils amended at Nbased rates could be reduced to that observ ed at P-based rates by WTR addition. The only exception to this WTR effect was the time zero sample at 1%WTR, where soil WEP values was greater at N-based than P-based rate of the T SP and indicates 1% WTR ma y not be sufficient to sorb and mask the excess P at the greater P rate of the highly soluble TSP treatments. The greater solubility of TSP makes it accessible to sorption especially at higher ra tes of WTR and, hence, resulted in similar WEP values of the two a pplication rates of TSP at 2.5% WTR. Also in samples taken at time zero, WEP values of soils treated with manure and Boca Raton biosolids were greater at N-based rates than at P-based rates at all three le vels of WTR. Thus, the excess P associated with applying organic sources of P at higher rates (N-based) was not totally masked by the added WTR. The greater P solubilities of the two P-sources applied at N-based rates may require more than 2.5% WTR to reduce the so luble P values to those at P-based rate. Sources of P also affected soil WEP values during the study. The e ffect of P-sources on soil WEP values could be isolated by consider ing P-based treatments (without WTR) in which equal P loads were applied from the different Psources. At the P-based rate, and in the absence of WTR, time zero soil WEP values were grea ter in the TSP treatment than the biosolids treatments and in Boca Raton biosolids treatment than in Pompano biosolids treatment. The trend of the soil WEP values tracked well with the so lubility of the P-source s as indicated by their WEP and PWEP values, which were greater in TSP than in biosolids and in Boca Raton

PAGE 61

61 biosolids than in Pompano biosolids. Thus, the soil soluble P values depend on solubility of the P-sources applied. However, the effects of Psources on WEP values were not observed when WTR was applied at P-based rates. At N-based rates, the soil WEP values were greater in Boca Raton biosolids treatment than in manure, Pompano, or TSP treatments, reflectin g the greater P loads of Boca Raton biosolids (201 mg kg-1) than other P-sources (75 – 125 mg kg-1). Apart from manure and Boca Raton biosolids at N-based rate, the time final soil WE P values were similar fo r the different organic source of P, irrespective of Psource application rate and rate of WTR (Fig. 3-1). The WEP values of Pompano biosolids (N-based, without WTR) were similar to, or lower than, values observed at P-based rate of ot her P-sources without WTR. This suggests that moderate water soluble P-source such as Pompano biosolids could be applied at N-based rates without greater P hazards than observed at the Pbased rate of other P-sources. The effects of treatments on soil ISP values were similar to the WEP trends, and the slight difference could be traced to more P being extracted as ISP than WEP (Table 3-1). For instance, in time zero soils, ISP determination wa s able to differentiate between N-, and P-based rates in all the P-sources (incl uding Pompano biosolids) at 1% and 2% WTR, which were shown to have similar WEP values. Thus, soil ISP indica tes more P is available for plants at N-based rates than at the P-based rate, whether WTR is a dded or not. Also contrary to the trend of the WEP data, ISP values were greater in Boca Rat on biosolids treatments than Pompano biosolids treatments at P-based rate (without WTR) throughout the study. The data reflected greater soluble P and, hence, plant available P expect ed from Boca Raton biosolids than Pompano biosolids treatments. The data suggest that the IS P technique predicts bioavailability better than the WEP determination. The ISP has been show n to be a good measure of bioavailable P

PAGE 62

62 (Sharpley, 1993a, b). Apart from the few differen ces in the two measures of time zero soil P, both WEP and ISP values were generally similar for the P-based treatments for the different Psources when 2.5%WTR was applied. They also both agree that, at N-based rates and at all levels of WTR, soil P is greater in Boca Ra ton biosolids than in other P-sources. Table 3-1. Effects of P-source, source applicatio n rates, and WTR on iron st rip P (ISP) values of soil samples taken during the glasshous e study. All concentr ation values are expressed in mg kg-1 soil. <------------Sampling periods†-------------> P-source P-source rate WTR rate (Oven dry %) June ‘04 Dec. ‘04 May ‘05 Sept. ‘05 Control --3.11 h 3.87 g 1.35 c 1.82 h 0 9.04 ef 8.16 efg 2.23 c 6.19 efgh 1 6.04 fgh 4.91 fg 2.12 c 2.92 fgh P-based 2.5 3.08 gh 4.37 g 1.78 c 2.37 gh 0 23.3 c 25.5 b 11.7 bc 23.8 b 1 16.3 d 13.1 cde 8.12 bc 11.2 cde Manure N-based 2.5 11.7 e 9.26 defg 7.77 bc 7.32 defgh 0 7.37 efg 8.00 efg 2.90 bc 6.04 efgh 1 4.09 fgh 5.24 fg 2.24 c 3.26 fgh P-based 2.5 3.79 gh 4.28 g 1.90 c 2.27 h 0 52.4 a 32.7 a 26.0 a 31.1 a 1 35.9 b 17.5 c 10.4 bc 15.7 c Boca Raton Biosolids N-based 2.5 20.9 cd 11.1 cdef 11.8 bc 9.15 cdefg 0 4.34 h 6.33 fg 3.46 bc 4.34 fgh 1 3.32 h 4.54 fg 2.78 bc 2.57 fgh P-based 2.5 2.72 h 4.00 g 2.12 c 2.01 h 0 11.9 e 15.0 cd 14.0 b 13.2 cd 1 7.96 efg 8.38 defg 4.75 bc 6.42 efgh Pompano Biosolids N-based 2.5 5.43 fgh 6.08 fg 3.79 bc 4.10 fgh 0 10.4 e 7.43 efg 2.69 bc 5.47 efgh 1 4.47 fgh 4.68 fg 1.78 c 2.70 fgh P-based 2.5 3.02 h 3.98 g 1.80 c 2.00 h 0 20.8 cd 11.2 cdef 5.33 bc 9.26 cdef 1 12.4 e 5.54 fg 3.24 bc 3.58 fgh TSP N-based 2.5 6.97 fgh 4.83 fg 8.10 bc 2.83 fgh †Means (n = 3) of treatments during the same sampling period follow by the same letter are not different at 5% significa nt level by Tukey test. Also, similar to the soil WEP data, there were reductions of time zero soil ISP values with addition of WTR at both Pand N-based rate s. The only exception was Pompano biosolids

PAGE 63

63 treatments at P-based rate. However, in so il samples taken between Dec. 2004 and Sept. 2005 when the study was terminated, similar ISP values were observed in all P-sources treatments at the P-based rate, irrespective of WTR rate. The si milar soil ISP values show the ability of the extractant and, hence, the potential of plants to desorb some WTR-sorbed P especially when P is limiting, as observed during the study of treatments residual effects at P-based rates. On the other hand, at the N-based rate, soil ISP decreased with increased WTR for all P-sources in most cases. These data suggest that not all WTR-sorbed P is accessible to the iron stri p extractant, hence, if the extractant adequately predict plant availabl e P, not all WTR-sorbed P will be accessible to the plants. The soil M-1P values were either similar or increased with increasing WTR, and indicate the solubilising effect of the acidic extractant (T able 3-2). The acidic extractant (pH < 2) releases some WTR-sorbed P, reflected in the similar or greater M-1P values with increasing WTR rates. Greater soil M-1P values at Nbased rates than P-based rates of organic source treatments are observed in all soil samples and in TSP at time zero (but not in subsequent samples), reflecting the greater P loads at Nthan P-base d rates (Table 3-2). Sim ilar M-1P values were observed for the different P-source s at P-based rate (due to sim ilar added P), but reflected the trend of P added from different P-sources at N-based rates. The M-1P values were greatest in Boca Raton biosolids and least in TSP treatments. The P load at the N-based rate for TSP (88 kg P ha-1) was smaller than the P loads from other sources, (280 kg ha-1 from Manure, 370 kg ha-1 from Boca biosolids and 233 kg ha-1 from Pompano biosolids) wh ich may explain the observed smaller M-1P in the TSP than in the organic sour ces of P treatments. Soil M-1P reflects the soil P load, but is insensitive to the WTR-sorbed P.

PAGE 64

64 Table 3-2. Effects of P-source, source applicatio n rates, and WTR on Mehlich-1P (M-1P) values of soil samples taken during the glasshous e study. All concentration values are expressed in mg kg-1 soil. P-source Psource rate WTR ( %) June ‘04 Dec. ‘04 May ‘05 Sept. ‘05 Control --†6.40 g 2.93 f 2.24 i 1.40 h 0 15.8 fg 14.3 ef 5.93 ghi 4.63 gh 1 23.7 efg 14.7 ef 13.6 fghi 10.3 fgh Pbased 2.5 29.7 efg 19.2 e 20.2 fg 15.5 fg 0 63.5 bcd 67.5 abc 54.0 bc 51.4 bcd 1 69.5 b 67.5 abc 59.6 ab 55.1 bc Manure Nbased 2.5 69.3 b 66.1 bc 57.7 b 47.9 cde 0 23.7 efg 10.5 ef 6.92 ghi 3.81 gh 1 24.8 efg 14.9 ef 14.1 fghi 11.3 fgh Pbased 2.5 31.8 efg 18.4 ef 21.5 ef 16.4 fg 0 164 a 67.9 abc 55.7 bc 48.1 cde 1 147 a 69.6 ab 64.5 ab 68.4 a Boca Raton Biosolids Nbased 2.5 148 a 82.7 a 72.3 a 61.9 ab 0 16.6 fg 8.77 ef 7.50 fghi 4.16 gh 1 20.5 efg 11.9 ef 11.4 fghi 10.9 fgh Pbased 2.5 26.9 efg 17.9 ef 16.0 ef 13.6 fgh 0 66.5 bc 48.8 d 35.8 de 35.0 e 1 64.3 bc 49.6 d 50.2 bcd 40.9 de Pompano Biosolids Nbased 2.5 65.2 bc 52.2 cd 41.7 cd 45.4 cde 0 23.5 efg 10.7 ef 5.01 hi 4.16 gh 1 24.4 efg 11.6 ef 10.4 fghi 12.9 fgh Pbased 2.5 24.7 efg 15.1 ef 17.3 fgh 14.8 fg 0 46.8 bcde 16.0 ef 11.2 fghi 7.86 fgh 1 37.6 def 17.1 ef 17.1 fgh 15.6 fg TSP Nbased 2.5 41.2 cdef 19.9 e 16.3 fghi 18.9 f †Means (n = 3) of treatments during the same sampling period follow by the same letter are not different at 5% significa nt level by Tukey test The 0.2M oxalate extractable P (Ox-P) measur es the sum of soil soluble and amorphous oxide-sorbed P forms, including a part of the WTRsorbed P. The trends of Ox-P values of time zero and time final soil samples were generally si milar (Fig. 3-2). As expected, soil Ox-P values were greater for N-based treatment than for P-based treatment reflecting the different P loads, and were similar for the P-based treatments with similar P loads. However, at N-based rate, the trend was for greater Ox-P values in biosolids than in manure treate d soils, in organic source of P

PAGE 65

65 treatments than TSP treatment, and similar Ox-P values in the two biosolids treatments. The trend also tracked well with the trend of appliedP from the different P-sources at N-based rates, as earlier explained. The soils showed increasing Ox-P values with increasing WTR, due to P sorbed and contribution of WTR to soil P. Figure 3-2. Oxalate extractable P (0.2M) values of (a) time zero and (b) time final soil samples taken during the glasshouse study Soil Total recoverable P is a measure of the soil P load, and includes the applied-P and P from natural sources. The trend of the Total recoverable P values was similar for soils sampled at different times during the study with few differences (Table 3-3). Soil Total recoverable P values were greater for N-based treatment than for Pbased treatments of the organic source of P (a) June 2004 samples 0 40 80 120 160 200 240 ControlMan-PMan-NBoca-PBoca-NPomp-PPomp-NTSP-PTSP-NP source and their application rateOx-P (mg kg-1) 0% WTR 1% WTR 2.5% WTR Contrast (P sources) N-based P-based Man vs Biosolids * NS Organic vs TSP * NS Boca vs Pompano NS NS Contrast: (N-based vs P-based) Manure * Boca * Pompano * TSP NS Polynomial effect of WTR Linear * Quadratic NS (b) September 2005 samples 0 40 80 120 160 200 240 ControlMan-PMan-NBoca-PBoca-NPomp-PPomp-NTSP-PTSP-NP source and their application rateOx-P (mg kg-1) 0% WTR 1% WTR 2.5% WTR Contrast (P sources) N-based P-based Man vs Biosolids * NS Organic vs TSP * NS Boca vs Pompano NS NS Contrast: (N-based vs P-based) Manure * Boca * Pompano * TSP NS Polynomial effect of WTR Linear * Quadratic *

PAGE 66

66 throughout the study. The least amount of P applied as organic sour ces of P at N-based rates was six times the applied-P at P-based rates, which explains the greater Total recoverable P at Nbased than at P-based rates. However, in TSP amended soils, the soil Total recoverable P was greater at N-based than P-based rates only in time zero soils at 0 and 1% WTR. Table 3-3. Effects of P-sources, source applicatio n rates, and WTR on Total recoverable P values of soil samples taken during the glasshous e study. All concentration values are expressed in mg kg-1 soil. Sampling periods† P-source P-source rate WTR rate (Oven dry %) June ‘04 Dec.04 May ‘05 Sept. ‘05 Control --23.6 k 22.0 k 28.3 h 12.2 j 0 40.0 jk 43.3 hijk 38.3 fgh 19.6 hij 1 53.0 ghijk 56.3 ghijk 58.2 fgh 44.2 ghi P-based 2.5 77.9 efg 78.7 efghi 82.2 efgh 45.1 ghi 0 125 bcd 111 cdef 133 bcde 78.2 cde 1 128 bcd 133 bc 142 bcd 85.9 cd Manure N-based 2.5 153 b 152 bc 144 bc 103 bc 0 46.0 ijk 35.5 jk 36.1 gh 17.9 ij 1 61.6 fghij 52.9 ghijk 60.7 fgh 40.9 ghi P-based 2.5 82.1 efgh 75.8 efghij 93.7 cdefg 61.7 defg 0 201 a 88.2 defg 131 bcde 77.0 cdef 1 156 b 170 ab 163 ab 121 ab Boca Raton Biosolids N-based 2.5 221 a 207 a 212 a 147 a 0 48.7 hijk 35.4 defg 38.6 gh 20.0 hij 1 67.8 fghij 48.2 ghijk 61.0 fgh 36.4 ghij P-based 2.5 73.5 fghij 73.0 efghij 84.6 defgh 49.7 fg 0 116 cde 89.9 jk 67.5 fgh 55.9 efg 1 133 bc 113 cde 128 bcde 82.3 cde Pompano Biosolids N-based 2.5 131 bc 126 cd 133 bcde 98.5 bc 0 54.7 ghijk 35.5 jk 32.3 h 17.3 ij 1 49.5 ghijk 50.4 ghijk 54.8 fgh 41.3 ghi P-based 2.5 48.3 efg 69.2 fghij 81.6 efgh 50.1 fg 0 75.3 fghi 40.6 ijk 40.8 fgh 18.7 ij 1 77.5 fgh 66.6 ghij 68.6 fgh 46.9 gh TSP N-based 2.5 88.8 def 84.5 defgh 98.5 cdef 57.9 efg †Means (n = 3) of treatments during the same sampling period follow by the same letter are not different at 5% significa nt level by Tukey test. Similar to the oxalate-P data, the effects of the P-sources were not seen in P-based treatments on the soil Total recoverable P values because of similar quantity of P applied (44 kg

PAGE 67

67 ha-1) from the different P-sources. On the other ha nd, in N-based treatments, Total P values were greatest in Boca Raton biosolids and least in TSP treatments. Thus, comparing the P-sources treatments at N-based rates requires normalizing for differences in the soil P loads. Soil Total recoverable P increased with WTR at the two rates for the differen t P-sources due to P contained in the WTR. The soil P load was increased by about 50% and more than 100% by the added P from WTR at 1% and 2.5% WTR rates, resp ectively. The estimated increase in soil P concentrations by the WTR, were 27 mg P kg-1 and 67.5 mg P kg-1 for 1% and 2.5% WTR treatments, respectively. Figure 3-3. Trends of soil water extractable P (WEP) and Total r ecoverable P (TP) values as affected by time and WTR treatments. (Tre atments within the same sampling period in (a) and (c) or within same WTR rate in (b) and (d) with same letter are not different at pvalue of 0.05 by Tukey test. (a) 0 2 4 6 8 10 12 14 16 Jun-04Dec.-04May-05Sep-05 Sampling periodWEP (mg kg-1) 0% WTR 1% WTR 2.5% WTR c b a b b b b a c b a a (b) 0 2 4 6 8 10 12 14 16 012.5 WTR rates (%)WEP (mg kg-1) June '04 samples December '04 samples May '05 samples September '05 samples a b c d a b c c a b c c (c) 0 20 40 60 80 100 120 140 Jun-04Dec.-04May-05Sep-05 Sampling periodTP (mg kg-1) 0% WTR 1% WTR 2.5% WTR c b b a a a a b b b c c (d) 0 20 40 60 80 100 120 140 012.5 WTR rates (%)TP (mg kg-1) June '04 samples December '04 samples May '05 samples September '05 samples ns b a b b ns

PAGE 68

68 The average soil Total recoverable P values in the absence of WTR were reduced between June and December 2004, but stabilized through th e end of the study. The reduction could result from greater P uptake by plants in the absence of WTR. However, in the presence of WTR (1% or 2.5%), the average Total recoverable P values were similar for the different sampling periods (Fig. 3-3d). The trend indicates negligible P upt ake by plants relative to soil P loads in WTR treated soils, but greater P uptak e by the first cropping in soils without WTR treatments. The P uptake in the absence of WTR was twice the P up take in the presence of WTR at the two Psource rates of almost all P-sources. Another ex planation of the variability of time zero soil P values is the inability to achieve thorough mixi ng even with the efforts of daily mixing during the one week incubation period. This observati on was also supported by the June 2004 soil P load data, which were similar for 0% and 1% WT R treatments within the first six months of the study, but lower at 0% WTR than at 1% WTR afterwards (Fig. 3-3c). The soil soluble P, as measured by soil WEP valu es, decreased with time in the absence of WTR (Fig. 3-3b). However, in presence of WTR, the WEP values decreased only within the first year (June 2005 to May 2006), and were similar thereafter (between May and September 2005). The effect of increasing WTR rates was also ob vious in the first six months, where soil WEP values decreased linearly as WTR increased fr om 0% to 2.5%. However, in samples taken in May 2005, and afterwards, soil WEP values were similar for the 1% and 2.5% WTR treatments (Fig. 3-3a). Soil P and Sorption Properties as Affected by WTR The effects of WTR treatments on soil soluble P values at the two P-source rates during the study are summarized in Fig. 3-4. Both WEP and ISP determinations reflected the sorbing effect

PAGE 69

69 of WTR, as both measures of soil P were reduced with increasing WTR rates. The two soil test methods are environmental soil test methods, a nd ISP has been shown to be a good measure of soil bioavailable P (Sharpley, 1993a, b). Figure 3-4. Effects of WTR and P-source rates on (a) water extractable P (WEP)and (b) Iron strip extractable P (ISP) values of soil samples taken during the glasshouse study. Treatments within the same sampling period with same letter ar e not different at p value of 0.05 by Tukey test. (a) Soil water extractable P (WEP) 0 5 10 15 20 25 30 35 40 June '04Dec. '04May '05Sept. ,05Sampling PeriodWEP (mg kg-1) N based (0% WTR) N based (1% WTR) N based (2.5% WTR) Control P based (0% WTR) P based (1% WTR) P based (2.5% WTR) a b c cd b d c d d a b c cd d a b bcd cd bc cd d b b a b b b b (b) Soil iron strip extractable P (ISP)0 5 10 15 20 25 30 35 40 June '04Dec. '04May '05Sept. ,05Sampling PeriodISP (mg kg-1) N based (0% WTR) N based (1% WTR) N based (2.5% WTR) Control P based (0% WTR) P based (1% WTR) P based (2.5% WTR) a b d c d c e f a b c d d a b b c c c c d d a b c c d ef

PAGE 70

70 Their reduction with increasing WTR rates establ ished the capability of WTR to reduce soil soluble P and, hence, P loss potential. Within th e one week of amendments equilibration, the WEP values of N-based treatments with 2.5% WTR were similar to WEP values of P-based treatments without WTR. Greater ISP values at N-base rates (with 2.5% WTR) than at P-based (without WTR) indicated more read ily available P even at 2.5% WTR. Thus, within a week of application, the potential hazards as sociated with N-based rates coul d be reduced to that with Pbased rates, while maintaining sufficient available P for plants. Soil M-1P values were either similar at each rate of the P-sources for the three WTR rates or increased with increasing WTR rates (Fig. 35a). The acidic extractant (pH<2) apparently solubilizes some of the WTR-sorbed P that will not be available at the soil specific pH, and thus, could not distinguish sorbed P by WTR and the readily available P. The trend of Total recoverable P values with treatments was similar at each sampling period throughout the study (Fig. 3-5b). The trend was increasing Total recoverable P with WTR at each of the application rate of the P-source. Thus, though solubl e P was reduced in presence of WTR, the P was retained in the soil (as show n by the greater soil TP values) and suggests improvements in the P sorption properties of the sandy soil. Soil total recoverable or oxalate extractable Al and Fe concentrations can indicate sorption characteristics of a soil. The soil Al co ncentrations were affected by the amounts of WTR applied (Table 3-4). Similar Al concentra tions were observed for N-based and P-based rates at each WTR rate throughout the study indicat ing negligible contributions of the P-sources to the soil Al concentrations. However, soil Al concentration increased with increasing WTR rate. In the absence of WTR, the soil Al con centrations of N-based and P-based rates were

PAGE 71

71 similar to the baseline concentrations observed in control. Unlike Al, the soil Total recoverable Fe concentrations were affected by the contributions from both sources of P rate and WTR. Figure 3-5. Effects of WTR and P-source rates on (a) Mehlich 1P and (b) Total recoverable P values of soil samples taken during the gl asshouse study. (Treatments within the same sampling period with same letter are not different at p -value of 0.05 by Tukey test. (a) Soil Mehlich 1 extractable P (M-1P) 0 20 40 60 80 100 120 140 160 180 June '04Dec. '04May '05Sept. ,05Sampling PeriodM-1P (mg kg-1) N based (0% WTR) N based (1% WTR) N based (2.5% WTR) Control P based (0% WTR) P based (1% WTR) P based (2.5% WTR) a a a d c bc b b ab a e d cd c b a a e d c e c c d d b a a (b) Soil Total recoverable P (TP) 0 20 40 60 80 100 120 140 160 180 June '04Dec. '04May '05Sept. ,05Sampling PeriodTP (mg kg-1) a f e d c e b b a a a c c c b b b f f e e e d d d c c c

PAGE 72

72 The greatest Fe concentrations were observed at the N-based ra tes with 2.5% WTR, while the concentrations at the P-based rate without WTR were similar to the control treatments throughout the study (Table 3-4). Gr eater Fe (than Al) concentra tions in the P-sources could result in the observed greater soil Fe at N-based rates than at P-based rate. Table 3-4. Soil Total recoverable Al and Fe c oncentrations as affect ed by water treatment residual (WTR) and the P-source applic ation rates during the glasshouse study. Total recoverable Al (mg kg-1) Total recoverable Fe (mg kg-1) P-source rate WTR (%) †June ‘04 Dec. ‘04 May ‘05 Sept. ‘05 June ‘04 Dec. ‘04 May ‘05 Sept. ‘05 0 124c 152c 235c 95.7c 136cd 140c 150bc 110b 1 491b 483b 528b 375b 153bc 154bc 172ab 114b N-based 2.5 1007a 976a 1072a 850a 184a 187a 199a 143a 0 107c 86.5c 111c 81.4c 108e 114d 112d 82.6c 1 427b 404b 511b 425b 133d 140c 130cd 117b P-based 2.5 956a 1020a 1129a 845a 165b 171ab 169ab 114b Control 0 88.4c 57.1c 97.0c 68.4c 107e 109d 103d 77.3c †Means (n = 12) of treatments during the same sampling period follow by the same letter are not different at 5% significa nt level by Tukey test The contributions of the soil P, Fe, and Al concentrations to the soil sorption properties could be summarized into indices of soil P sorp tion, such as Degree of P Saturation (DPS), or Soil P Storage Capacity (SPSC). The DPS, as the na me implies, is an index of soil P sorption site saturation and, thus, a measure of capability of the soil to ho ld and prevent loss of P through runoff and leaching. Large DPS values suggest limite d soil capability to re tain P. Soil DPS is calculated as the ratio of the soil 0.2M oxalate extractable P to the corresponding 0.2M oxalate Fe and Al and assuming an value of 0.55 (recommended for Flor ida soils by Nair et al., 2004). DPS = (Pox)/ (Alox + Feox) (Equation 3-1) Where Pox, Alox, and Feox are 0.2M Oxalate extractable P, Al, and Fe (expressed in moles), respectively.

PAGE 73

73 The DPS values of all soils that received P application (without WTR), exceeded 25%. Coapplying P-sources at the P-based rates along w ith 1% WTR and at N-based rates along with 2.5% WTR, resulted in DPS values below the critical value of ~25% (Fig. 36). Nair et al. (2004) observed a changed point at 20% DPS, ag ronomic high M-1P values (30 mg P kg-1) at 22% DPS, and very high M-1P values (60 mg P kg-1) at 28% DPS. The critical M-1P value (50 mg P kg-1, Paulter and Sims, 2000), is equivalent to 25% DPS. 0 25 50 75 100 125 150June '04Dec. '04May '05Sept. ,05Sampling PeriodDPS (%) N based (0% WTR) N based (1% WTR) P based (0% WTR) Control N based (2.5% WTR) P based (1% WTR) P based (2.5% WTR) a bc a a a bcd b cd cd d c c cd cd b e de b cd cd e de b b b b b b Figure 3-6. Effects of WTR and P-source rates on degree of P saturations (DPS) values of soil samples taken during the glasshouse study. (Treatments within the same sampling period with by the same letter are not differe nt at 5% significant level by Tukey test). Dotted line represents 25 % threshold DPS value. The data show that the P-sorption properties of sandy, low P-sorbing Florida soils could be enhanced with WTR, and that the amount of WTR needed is dictated by the P-source and source

PAGE 74

74 application rates. The reduction in DPS values with WTR addition also explains the reduction in soil soluble P with WTR addition. A plot of WEP values against the DPS values of all the soils sampled during the study shows increasing WEP values with increasing so il DPS (Fig. 3-7). The in creased soil sorption sites (indicated by reduced DPS), reduces soil sol uble P. The variation of soil DPS values at the same P application rate and WTR rate reflects differences in compositions of P-sources, especially P, Al and Fe concentrations, and sugge sts the need to account for these variations in determining the rates of WTR to achieve similar soil DPS. 0 10 20 30 40 50 050100150200 Soil DPS (%)Soil WEP (mg kg-1) 0% WTR 1% WTR 2.5% WTR Figure 3-7. The water extractable P (WEP) values as affected by degree of P saturation (DPS) of soil samples taken during the study. The soil phosphorus storage cap acity (SPSC) is another inde x suggested recently by Nair and Harris (2004) to quantify the amount of P a so il can sorb before exceeding the threshold soil equilibrium P concentration. The SPSC is calculated as:

PAGE 75

75 SPSC (mg P kg-1) = (0.15 – PSR)* (Alox + Feox)*31 (Equation 3-2) PSR = P saturation ratio = (Pox)/(Alox + Feox) (Equation 3-3) Where Pox, Alox, and Feox are 0.2M Oxalate extractable P, Al, and Fe, respectively. The 0.15 value used in SPSC calculation corresponds to the cr itical solution P concentration of 0.10 mg L-1 (threshold) proposed by Breeuswsm a and Silva (1992). Zero SPSC indicates a soil PSR value of 0.15 and solution P concentrations 0.10 mg L-1. The greater the SPSC value, the more applied-P a soil can retain (sorb). Generally, SPSC values of time zero soil samples increased with increasing WTR rates (F ig. 3-8) due to added Al. The SPSC was more negative at N-based rates than at P-based rates due to greater added P at N-based rate, which increased saturation of the P sorption sites. -175 -150 -125 -100 -75 -50 -25 0 25 50 75 100Control Manure P Boca P Pompano P TSP P Manure N Boca N Pompano N TSP NTreatmentSPSC (mg kg-1) 0%WTR 1%WTR 2.5%WTR Contrasts: P-based (without WTR) vs Control P-based (1% WTR) vs Control P-based vs N-based P-based (2.5% WTR) vs P-based (without WTR) N-based (with 2.5% WTR) vs Control *** *** *** *** *** Figure 3-8. Soil phosphorus stor age capacity (SPSC, mg kg-1) values of time zero samples taken during the glasshouse study for the different treatments. (Treatments ending in P, and N, are P-based and N-based rates of the sources, respectively). The SPSC values at the P-based rates of the sources, without WTR, were close to 0 mg P kg-1 confirming the P-based rate as environmentally friendly (Fig. 3-8). However, N-based rates

PAGE 76

76 without WTR (and even at 1% WTR) resulted in negative SPSC values. The greatest negative SPSC value was observed in Boca Raton biosolids, N-based treatment. Ne gative values of SPSC indicate that a soil contains excess P, and suggest that N-ba sed rates could cause negative environmental impacts (Reddy et al., 1980; Pier zynski, 1994; Peterson et al., 1994; Maguire et al., 2000). With addition of WTR, the SPSC valu es of the N-based rates of all P-sources increased and were positive for most P-sources (except Boca Raton bios olids) at 2.5% WTR. Thus, application of WTR creates P sorption si tes for soil soluble P in excess of 0.10 mg L-1 at the N-based rates. Addition of WTR increases the storage capac ity at both the P-based and Nbased rate. However, SPSC is greater at P-base d than at N-based rate s when equal amounts of WTR are applied. Different amounts of WTR will be required to bring soils treated with different P-sources at N-based rate s to equal SPSC values. Plant P Uptake, P Concentrations and Dry Matter Yield Plant uptake of P was affected by the P-s ource, P-source rates and WTR rates. The P uptake of each cropping, and the Total P uptake (s um of the uptakes from the three croppings), were greater at the N-based rates than at the P-based rates at each level of WTR (Fig. 3-9). Addition of WTR reduced P uptake at each P-so urce rate. The P uptake was greater in the absence of WTR than at 1% or 2.5% WTR, an d most WTR-sorbed P wa s not accessible by the plants. However, greater P uptake was observed in ryegrass and in the second bahiagrass crop at N-based rates with 1% and 2.5% WTR than at Pbased rates without WTR. The greater P uptake, indicates that applying WTR to N-based rates reduces P uptake, but not below the P requirements of the plants, which is expected to be optimum at P-based rate (with no WTR). We, therefore, fail to reject hypothe sis I and II and concl ude that P-based rate s (without WTR) and Nbased rates (with WTR) both resu lt in minimized soil soluble P and optimized plant P uptake.

PAGE 77

77 0 10 20 30 40 50 60 70 80 Bahiagrass (1)RyegrassBahiagrass (2)Total (The three plants) Planting periodP uptake (kg ha-1) N-based (0%WTR) N-based (1%WTR) N-based (2.5%WTR) P-based (0%WTR) P-based (1%WTR) P-based (2.5%WTR) Control a c c b d d d a b b c dd d a b b c c cdd a b c bc d dd Figure 3-9. Effects of water treatment residua l (WTR) and P-source rates on plant P uptake during the glasshouse study. Treatments w ithin the same cropping with the same letter are not differe nt at 5% significant level by Tukey test. The greater P uptake of the three individual cr oppings and the total P uptake for the entire growing season at P-based rate without WTR than at the P-based rate in the presence of WTR indicate it is not advisable to apply WTR to P-based rate. In pres ence of WTR, the uptake of the two bahiagrass croppings and the total P uptake va lues at the P-based rate were not different from those observed in the control treatment . Similar P uptake was observed for 1 and 2.5% WTR treatments in the three croppi ngs at P-based rate and in th e first bahiagrass and ryegrass croppings at Nbased rates. Thus, increasing th e WTR rates from 1% to 2.5% did not further reduce P uptake. The trends of the P concentrations of the tw o bahiagrass croppings were similar to the P uptake data. Plant P concentrations were smaller in the presence, th an in the absence, of WTR at

PAGE 78

78 each application rate and greater in Nthan Pbased treatments. Ryegrass P concentrations were also reduced with increasing WTR rate at each Psource rate (Fig. 3-10). Th e P concentrations at the N-based rate, with WTR, were similar to t hose at the P-based rate without WTR for ryegrass and even greater at N-based rates with WTR than P-based rates without WTR for the second bahiagrass crop. The similar plant P concentrations suggest that applicat ion of WTR, even at 2.5%, will not reduce plant P concentrations be low the optimum observed at P-based rate without WTR. Plant P concentrations in the seco nd bahiagrass cropping were greater than in the first cropping in WTR-amended soils, and vice versa in treatments without WTR. The data suggest that P-sorbed by the WTR could be assessed by the plants over time, when plants are in the deficiency stage. The greater P concentrations of the second bahiagrass crop could also be due to “Steenbjerg effect”, where plant nutrient concentr ations increase when plants are subjected to nutrient deficiency (Steenbjerg, 1951; Bates, 1971). Nutrient defi ciency destroys potential for growth, but plants continue to accumulate the nutrient, resulting in greater nutrient concentrations during deficiency periods (Ulr ich and Hills, 1967; Jones, 1967; Bates, 1971). Though the P concentrations of the grasses we re reduced when WTR was applied, the least P concentration observed during the two bahiagrass croppings (1.73 g kg-1) was above the 1.3 g kg-1 mean P concentration reported for bahiagrass (USDA, 1996). Also, the least P concentration observed in ryegrass samples (1.82 g kg-1) was above the 1 g kg-1 regarded as sufficient for ryegrass (Hylton et al., 1965). The reductions in P concentrations are consistent with other studies that show plants grow n in potting media treated with WTR had lower P concentrations (Bugbee and Frink, 1985; Elliott a nd Singer 1988). Ippolito et al. (1999) also reported decreasing P concentration, but increasing plant yi elds with increasing WTR rates.

PAGE 79

79 0 1 2 3 4 5 6 Bahiagrass (1)RyegrassBahiagrass (2)Weighted (over the three plants) Planting periodP concentrations (g kg-1) N-based (0%WTR) N-based (1%WTR) N-based (2.5%WTR) P-based (0%WTR) P-based (1%WTR) P-based (2.5%WTR) Control a c c b d d d a b c bc d d d a b b cdc d e a b c b d d d Figure 3-10. Effects of water treatment residual (WTR) rates and P-source rates on plant P concentrations. Treatments within the same planting with the same letter are not different at 5% significa nt level by Tukey test. Yield reduction was observed in the first bahiagrass cropping with increasing addition of WTR (Fig. 3-11). Reduced yields may have partly resulted from i rregular growth of the grass due to germination difficulties encountered during the grass establishment. However, the yield of the ryegrass and the second bahiagrass croppings at N-based and P-based rates with WTR were similar to the yields of similar treatments in th e absence of WTR. Thus, the yields of the plants were not affected even at 2.5% WTR at both N-based and P-based rates. The yields of the ryegrass and the second ba hiagrass croppings were greater at N-based than at P-based rates. The N-based rates not only supplied adequate nutrients, but enhanced yields for longer periods than the P-based rates. Total plant dry matter yields from the three croppings at N-based rates (even w ith WTR) were greater than yi elds at P-based rates (in the

PAGE 80

80 absence of WTR). However, the greater yields of N-based treat ments may be not be economical considering the greater negative P impact of the high P rate on the environment. Thus, P-based rates (without WTR) and N-based rates with WTR optimize plants dry matter (DM) yields. 0 2 4 6 8 10 12 14 16 Bahiagrass (1)RyegrassBahiagrass (2)Total (The three plants) Planting periodYield (Mg ha-1) N-based (0%WTR) N-based (1%WTR) N-based (2.5%WTR) P-based (0%WTR) P-based (1%WTR) P-based (2.5%WTR) Control a a a bc bc b bc c c b b b a b b b b a a a a b bc c d d d a Figure 3-11. Effects of water treatment residual (WTR) and Psource rates on plant dry matter yields. Treatments within the same planting period with the same letter are not different at 5% significa nt level by Tukey test. The yields of the plants were also affected by the P-sources (Table 3-5). For the P-based rates (equal P loads), the yields of the two bahiagrass croppings were similar for the different organic source of P, but greater than the yields obtained from TSP treatment. Ryegrass yields from the TSP treatments were similar to yields from organic sources of P treatments. Similarly, at the N-based rates, the yields of ryegrass and the second bahiagrass cropping from organic sources of P treatments were greater than yields from the TSP treatments. The data indicate that organic sources are not inferior to the mineral fertilizer in terms of the agronomic returns

PAGE 81

81 (yields). Rather, the organic sources, which act as slow release P fe rtilizer, promote more efficient utilization of the added P. Table 3-5. Effects of P-source and source ap plication rates on plant dry matter yields† (Mg ha-1). Rate Plant Source of P / Contrast N-based P-based Contrast N-based vs P-based Manure 4.55† 6.52 * Boca Raton biosolids 7.53 5.97 * Pompano biosolids 6.65 5.91 * TSP 6.11 5.62 * Manure vs Biosolids * NS Organic vs Mineral source NS * First Bahiagrass cropping Boca Raton vs Pompano * NS Manure 4.38 3.36 * Boca Raton biosolids 4.06 2.95 * Pompano biosolids 3.75 3.06 * TSP 3.30 3.14 NS Manure vs Biosolids NS * Organic vs Mineral source * NS Ryegrass cropping Boca Raton vs Pompano * NS Manure 3.99 3.17 * Boca Raton biosolids 5.54 3.23 * Pompano biosolids 4.28 3.39 * TSP 2.83 2.97 NS Manure vs Biosolids * NS Organic vs Mineral source * * Second Bahiagrass cropping Boca Raton vs Pompano NS NS † Means of nine samples * Significant at P = 0.05% NS not significant at P = 0.05% The yields of first the bahiagrass cropping we re greater at N-based rates than at P-based rates, except for soils amended with manure at the N-based rate. The N-based manure treatment took extra time to establish thus resulted in the lower yields observed at N-based than P-based

PAGE 82

82 rates (Table 3-5). The greater pH of soils treated with manure at N-based rates (pH>7) than the targeted pH (5.5) also likely reduced bahiagrass yields for the N-based treatment. Lower yields of bahiagrass at high, than at low, rates of manure were also reported by OÂ’Connor et al. (2005). The yields reported at low P rate were twice the yield at a high P rate, and were also attributed to a greater soil pH at high P rate than the so called target pH for bahiagrass (pH = 5.5). Ryegrass and second bahiagrass croppings yields , were greater at N-based than P-based rate of the sources, except in TSP treated soil where similar yields, were observed for the two rates. The similar yields of P-based rates of TSP to that observed at high P-rate establishe d that the P-based rate optimized plant yields and can serve as the basis for comparing the yields from other treatments. The relative agronomic yield (RAY) term wa s obtained for each croppings by relating the yield obtained from each treatment to the greatest yield, and expressed as percentage (Table 3-6). The TSP treatments did not give the greatest RAY for the three croppings (individually or collectively). However, the P-based TSP treatment could still serve as basis for comparison of the agronomic benefits, as it proportionately repr esents sufficient readil y available P and N for optimal plant growth. The RAY values of treatments with WTR are ei ther greater than or similar to RAY values of the P-based rate of TSP without WTR. This indicates no reduction in yield below the optimum expected when WTR is co-applied with organic sour ce of P at either of the two P-source rates. Possible reductions in yield were indicated by lower RAY values in some WTR-amended treatments than in non-WTR treatments for P-ba sed rate. The smaller RAY values obtained in manure treatments at N-based rate of the firs t bahiagrass cropping (<77%) may be connected with greater soil pH values (> 7) in manure treatments which are not favorable for growth of the

PAGE 83

83 acidic soil loving bahiagrass as earlier explaine d. The same manure treatm ent resulted in the greatest RAY values for ryegrass, which is more tolerant to high soil pH. Table 3-6. Relative agronomi c yield (RAY) values of th e different treatments (%) P-source Rate WTR (%) Bahiagrass (1)Ryegrass Bahiagrass (2) Combined† Control 69 55 58 65 0 87 75 58 79 1 78 75 57 74 Pbased 2.5 74 74 53 71 0 63 100 68 77 1 54 98 76 75 Manure Nbased 2.5 50 94 68 69 0 83 69 64 77 1 70 63 57 67 Pbased 2.5 66 65 52 64 0 100 85 98 100 1 91 92 96 97 Boca Raton biosolids Nbased 2.5 86 94 100 96 0 80 70 62 75 1 71 68 55 68 Pbased 2.5 66 65 62 68 0 84 81 85 87 1 80 85 74 83 Pompano biosolids Nbased 2.5 80 84 68 81 0 77 68 61 73 1 71 68 50 66 Pbased 2.5 59 74 47 61 0 83 73 60 77 1 69 76 44 66 TSP Nbased 2.5 72 72 46 67 † RAY based on sum of DM yields for the three croppings The combined RAY values were based on to tal DM yields obtained from all three croppings, and trends in RAY values agreed with the trends for individual cropping. The data suggest that both P-based and N-ba sed rates of different P-sources together with WTR optimized agronomic yields. Also, a P-based rate of poultry manure is advised if bahiagrass is to be grown. An N-based rate of the Ca-rich poultry manure w ill result in greater soil pH than tolerated by bahiagrass and will reduce the yield.

PAGE 84

84 Relative Phosphorus Phytoavailability (RPP) Relating extractable P to the ability of organic sources to supply plants with P has not been successful (Beegle, 2005, Elliott et al., 2005). The P -release characteristics of biosolids were compared to plant P uptakes by McLaughlin (1988) . The potential for P release to plants was not only underestimated by the extractants, but the ti me P-release characteristics were also poorly predicted. The relative P phytoavailability (RPP) is a co mparative measure of the availability of phosphorus from P-sources to plan ts. Other studies used regressi on parameters to calculate the RPP (McLaughlin and Champion, 1987; OÂ’Connor et al., 2004). However, both regression methods and point estimates were used in this study, and treatments without WTR were used. To study how RPP is affected by the plants, Psources, and P-source rates, point estimates of the RPP were calculated for each experimental unit. The point estimate is appropriate for treatments where all plant nutrients are adequate a nd treatments have equal soil P loads. Both Nand P-based rates (without WTR) satisfy the co ndition of adequate nutrients. However, the assumption of equal soil P loads is satisfied on ly by the P-based rates for the first bahiagrass croppings and the subsequent croppi ngs (if P-applied is adjusted for P uptake by previous crop). The N-based rates also can be nor malized to satisfy the condition of equal P loads if expressed as P uptake per unit P load. Thus, the use of P uptake per unit P load will account for the differences in the soil P loads from the differe nt P-sources at N-based rates. The P uptake per unit P load obtained from each P-source is then rela ted to similar ratio obtained for TSP to arrive at the RPP expressed in percentage for each source as in Equation 3-4. (P uptakesource/P appliedsource) RPP = --------------------------------------* 100 (Equation 3-4) P uptakeTSP/ P appliedTSP

PAGE 85

85 The TSP treatment served as the basis for comparison with other P-sources because the mineral source of P was expected to give optim um agronomic performance. Fertilizer-P is readily available P and other ne eded nutrients were adequately supplied for optimum growth. Analysis of variance (ANOVA) indicated P-source, P-source rate and pl ant cropped affects RPP. The RPP values obtained at each cropping and at Pand N-based rates by point estimate are shown in Table 3-7. The point estimates indicate that RPP values can be affected by the source application rates. At P-based rate s, the RPP values were similar for all the P-sources especially for the ryegrass and second bahiagrass croppings, indi cating that the plants were as efficient at taking up P from organic sources as from the mineral P-source. Table 3-7. Relative P phytoavailab ility (RPP) values for the diffe rent P-sources at each P-source rate by point estimate. All RPP values are expressed in %. Source rate P-source Bahiagrass (1) Ryegrass Bahiagrass (2) †Average Manure 22 c 55 b 36 b 38 c Boca Raton biosolids 70 b 58 b 75 a 67 b Pompano biosolids 42 c 66 b 73 a 60 b N-based TSP 100 a 100 a 100 a 100 a Manure 94 ab 107 a 124 a 109 ab Boca Raton biosolids 116 a 111a 123 a 117 a Pompano biosolids 79 b 91a 100 a 90 b P-based TSP 100 ab 100 a 100 a 100 ab † Average of RPP of the three croppings. ‡Means (n = 3) of treatments during the same cropping and at the same P-source rate follow by the same letter are not different at 5% significant level by Tukey test The RPP of all the P-sources at the P-based ra tes were, therefore, > 75% and categorize all P-sources into high RPP. At the N-based rate, the RPP values obtained by point estimate were 75%. The RPP values of manure and Pompano bi osolids are similar at planting of first bahiagrass and ryegrass but greater for Pompano biosolids than for manure at planting of the

PAGE 86

86 second bahiagrass crop. The RPP of manure was greater at cropping of rygrass than at either of the two bahiagrass croppings. The reduced RPP of manure during bahiagrass croppings reflect the low tolerance of the grass to high pH associated with manure treatment. Thus, apart from the impacts of P-source composition suggested fr om other studies (McLaughlin and Champion, 1987; OÂ’Connor et al., 2004), plant tolerance to pH could also a ffect the RPP (as observed in manure treated soils). Few effects of most biosolids on soil pH ar e expected, and plant tolerance to pH may not be an issue as regards RPP of biosolids. However, there may be some special cases, as in Tarpon Spring N-viro biosolids (p H = 11.9) used in a study by OÂ’Connor et al., (2000), which also reduced bahiag rass growth and its RPP. The regression slope-ratio procedure can define RPP values better than point estimates as it integrates impacts of both source a nd P rates to the RPP values. The slope-ratio procedure is also widely used to determine P bioavailability (McLaughlin and Champion, 1987; Cromwell, 2002; OÂ’Connor et al., 2004) and, hence will enhance comparison of the results. To obtain the regression RPP estimates, the P uptake from each cropping and the total P uptake for the three croppings were regressed with applied P at 0-ra te (control), P-based rates, and N-based rates (adjusted for P uptake of previous croppings). The regression slopes obtained for each P-source were compared with the slope of TSP treatment (the greatest slope) and e xpressed as percentage as in Equation 3-5. RPP = [(Slope source) / (Slope TSP)] * 100 (Equation 3-5) The RPP obtained for each P-source by the sl ope-ratio is shown on Table 3-8. The RPP values obtained (30-100%) were within the ranges from other studies (McLaughlin and Champion, 1987; OÂ’Connor et al., 2004; Pritchar d, 2005). The ranges of RPP values obtained by combining the three croppings P uptake and by using uptake from each individual cropping are 30-55% for manure, 68-82% for Boca Raton bioso lids and 41-58% for Pompano biosolids. The

PAGE 87

87 RPP values of Boca Raton biosolids for the tw o bahiagrass croppings (82%) were within the range of “high” phytoavailability materials id entified by O’Connor et al. (2004). The reduced RPP value for ryegrass may be due to the crop rath er than the source P-lability. The “high rate activated sludge” process by which the Boca Rat on material is produced apparently allows similar P availability (indicated by its high WEP) as in biological P removal (BPR) processes. Table 3-8. Relative P phytoavail ability (RPP) values for the di fferent P-sources by regression. Cropping Regression parameter / RPP Manure Boca Raton biosolids Pompano biosolids TSP Slope 0.076 0.209 0.105 0.256 r2 0.71 0.99 0.92 0.99 p -value 0.16 <0.01 0.04 <0.01 First Bahiagrass RPP (%) 30 82 41 100 Slope 0.068 0.083 0.068 0.123 r2 0.94 0.97 0.96 0.99 p -value 0.03 0.01 0.02 <0.01 Ryegrass RPP (%) 55 68 55 100 Slope 0.042 0.095 0.067 0.116 r2 0.74 0.98 0.94 0.98 p -value 0.14 0.01 0.02 0.01 Second Bahiagrass RPP (%) 36 82 58 100 Slope 0.178 0.345 0.223 0.419 r2 0.83 0.99 0.95 0.99 p -value 0.08 <0.01 0.03 <0.01 †Combined RPP (%) 43 82 53 100 RPP Category Moderate (25-75%) High (>75%) Moderate (25-75%) † RPP based on sum of P uptake from the three croppings The RPP value of Pompano biosolids (41-58%) is similar to biosolids categorized as “moderate” phytoavailability and is characteristic of the RPP va lues of most conventionally processed biosolids. The available P of some ma nures has been reported to be about equal to fertilizer-P (Agronomy Guide, 2002), but availability depends on the type of manure and plant grown. The lower RPP value obtained for manure in this study mainly results from high soil pH values (pH > 7.0) associated w ith the high poultry manur e application rates and reduced growth

PAGE 88

88 of the acid-tolerant bahiagrass. The reduced growth of bahiagrass at high ra te of manure resulted in poor regression and reflected in lower RPP in bahiagrass than in ryegrass. The RPP values of the manure treatments were greater for ryegrass (55 %), which is more tolerant of basic soil pH values than for bahiagrass (30 -36%) while Boca Raton biosolids shows the opposite. The reduced RPP of Boca Raton biosolids for ryegra ss cropping may also refl ect tolerance of the grass to some of the biosolids properties. Howeve r, this study is not sufficient to make a more definite reason or conclusion a bout reasons for the observed tr end. The RPP values obtained from manure treatments by individual or comb ined croppings categorize the amendments into medium RPP. The combined data (based on su m of uptake from the three croppings) gave RPP values similar to those obtained from the individual croppings and categorized manure and Pompano biosolids into the moderate RPP cat egory and Boca Raton biosolids into high RPP category. Thus the residual effect indicated RPP of organic sources did not change with time in Florida sand. The RPP values obta ined at first bahiagrass cro pping (short term) is similar to values got from the cumulative data and suggested adequacy of the short time studies to evaluate P phytoavailability from organic source. Aluminum Toxicity The impact of the applied Al-WTR on Al con centrations of the plan ts and the potential for Al toxicity was studied using TSP treatments at N-based rates. The applied Al-WTR did not affect plant Al concentrations in any cropping. Plant Al concen trations in treatments that received no WTR were similar to those wh ere WTR was applied, even at 25 g WTR kg-1 (Table 3-9). The plant Al concentrations (21 – 80 mg kg-1) are within the range of common (non toxic) plant Al concentrations (10 – 1000 mg kg-1) reported by Pais and Jone s (1997) and the average concentration of 73 mg kg-1 reported in other studies for bahiagrass (USDA, 1996; Arthington,

PAGE 89

89 2002; Arthington et al., 2002). The phytotoxic species (Al3+) is expected in so il solution with pH 4.0 (Kennedy and Cooke, 1982). Thus, at the obse rved soil pH of > 5.2, the less toxic Al species (Al(OH)2+, Al(OH)2 +, Al(OH)3) predominate in the WTR amended soils. The soil solution Al3+concentration is not expected to exceed 50 µg L-1 at pH > 5.2 (Kennedy and Cooke, 1982). This concentration is ten times lower than the 0.5 mg L-1 expected to cause toxicity problems with plant root (Sar tain, 2005). In addition, comple xation of Al with organic compounds from the organic sources of P could further reduce the free Al3+. Table 3-9. Effect of aluminum water treatmen t residual (Al-WTR) on bahiagrass (first and second croppings) and ryegrass Al , Ca, and Mg concentrations. Plant WTR rate (%) / Contrast Al (mg kg-1) Ca (g kg-1) Mg (g kg-1) 0 59.2 5.94 4.42 1 66.8 4.97 3.76 2.5 80.4 5.78 4.18 Linear NS NS NS First Bahiagrass Quadratic NS NS * 0 60.8 4.28 4.51 1 45.6 4.82 4.48 2.5 51.5 4.92 4.30 Linear NS * NS Ryegrass Quadratic NS NS NS 0 23.3 2.04 4.05 1 21.9 1.99 4.28 2.5 25.8 2.33 4.42 Linear NS NS NS Second Bahiagrass Quadratic NS NS NS †Average Concentrations of South FL Bahiagrass ‡73 4.30 3.30 *Significant polynomial effect of WTR at p -value of 5%. NSNon-significant polynomia l effect of WTR at p -value of 5%. †Arthington, 2002. ‡USDA, 1996. The lack of Al toxicity is further establishe d by the similar and adequate plant Ca and Mg concentrations measured (Table 3-9). The anta gonistic effects of increased Al uptake are expected to reduce the concentra tions of other cations especially Ca in the plants. In addition,

PAGE 90

90 and most importantly, no symptoms of Al toxic ity were observed in any treatment throughout the study. The plant growth was good, and the well-dev eloped roots indicated no Al toxicity. The plant mineral concentrations are also similar to concentrations reported in other studies (Hylton et al., 1965; Arthington, 2002). Summary and Conclusions Land application of Al-WTR has potential as a BMP to reduce the environmental hazard associated with excess soil P, without negativ e agronomic impacts. The P-sources, source application rates, and WTR affected varying meas ures of soil P values (WEP, ISP, M-1P, TP and Ox-P). Within a short time of application of WTR, the DPS and SPSC values of low P-sorbing sandy soil were improved and soil soluble P was reduced. Applying WTR, even at 2.5%, to Nbased rates treatments did not redu ce plant yields in most cases. Ho wever, at P-based rates, WTR rate >1% reduced yields. Plant P concentrations were reduced by application of WTR, but the P concentrations remained greater than the 1 g kg-1, reported to be sufficient for ryegrass, and the 1.3 g kg-1 reported for bahiagrass, even in treatments that received 2.5% WTR at P-based source rates. The Al concentrations of the plants we re not affected by th e added WTR, and other mineral concentrations were within normal ranges. The potential environmental hazard associ ated with excess P-loads from N-based addition of P-sources can be reduced with a pplication of WTR wit hout negative agronomic impacts. As much as 2.5% WTR can be applied to N-based rates treatments, but less than 1% WTR is advised if the need arises to apply the amendment at P-based rates. An alternative to WTR addition is to apply lower water soluble-P ma terials such as Pompano biosolids, at N-based rates.

PAGE 91

91 The organic sources of P varied in relative P phytoavailability (RPP) values especially at high (N-based) P rates, with Boca Raton biosol ids being as readily available as mineral Psources. Lower RPP values were observed in manure treatments due to the greater soil pH associated with its application, especially at Nbased amendment rates. Apart from the properties of the sources known to affect the P-solubil ity, source application ra tes, the plant grown (especially plant tolerance to resulting soil cond itions after treatments) could affect the RPP.

PAGE 92

92 CHAPTER 4 FIELD VALIDATION OF AGRONOMIC IM PACTS OF LAND APPLIED WTR AND DIFFERENT P-SOURCES Introduction Off-site P movement in coastal plain soils is facilitated by the low P binding characteristics of the soils (Harris et al., 1996; Hansen et al., 2002). Applied P readily saturates the sorption capacity of such soils, and P is often lost into nearby sensit ive water bodies. One method being suggested to reduce P losses from such soils is the applicatio n of chemical amendments to improve P sorption and retention by the soils. La boratory and rainfall si mulation studies have demonstrated the ability of water treatment resi duals (WTR) to reduce soil soluble P in Florida soils (OÂ’Connor et al., 2002a; Elli ott et al., 2002b; Elliott et al., 2005). A glasshouse study using Florida surface soil confirmed that co-applyi ng WTR with different P-sources could reduce excess soil soluble P while optimizing plant grow th and P uptake (Chapter 3). Thus, co-applying WTR with P-sources should reduce edge-of-field P losses without negative agronomic impacts. Uncertainties about Al phytotoxi city of the Al rich residua ls make land application of WTR unattractive to some (Anderson et al., 1995) . However, Peters and Basta (1996) used drinking water treatment alum hydrosolids to redu ce bioavailable P in soil s with no increase in extractable Al after application. Results from the glasshouse study described in Chapter 3 also indicated no negative imp acts of the Al-WTR on plan ts Al concentrations, but the data need to be validated in the field. Another concern with the land application of organic source s of P is the extent of availability of source-P for plant uptake. Co mparative studies indica ted that biosolids-P availability to plants can vary from <10% to 100%, relative to TSP, and depended on the method of biosolids preparation, which changed biosol ids chemical properties (OÂ’Connor et al., 2004).

PAGE 93

93 The glasshouse study (Chapter 3) also identified differences in relative P phytoavailability of two biosolids, and poultry manure. The inferences from glasshouse studies need to be validated in field settings, as environmental factors (edaphic and climatic) can modify nutrient concentrations in plants (Bates, 1971). Thus, to ensure consistent and accurate amen dment utilization decisions, data from a field study were used to validate the agronomic impact s of the same P-sources and WTR noted in the glasshouse study. It was hypothesized that: 1. Both P-based rates of differe nt organic sources of P (without WTR) and N-based rates (with WTR) optimize plant P uptake in the fiel d as observed in the glasshouse study. 2. Amendment rates (P-sources and WTR rates) se lected in (1) that optimize P uptake also minimize soil soluble P in the field. 3. Organic sources of P vary in P ph ytoavailability in a field setting 4. Land application of Al-WTR w ill not induce greater plant Al concentrations than unamended soils. The main objective of the field study was to validate the agronomic impacts of the Psources and WTR in a real world setting. Materials and Methods Experimental Procedure The field experiment, sponsored by South Florida Water Management District in cooperation with a private comp any and UF-IFAS, was at Kirton Ranch near Okeechobee, FL. The study site is a ranch dominated by improved pa stures, and is located on the eastern border of Okeechobee County, eleven kilometers northeast of Okeechobee, north of the Lake Okeechobee.

PAGE 94

94 Soil at the field site is Immokalee fine sand, a typical Spodosol, classified in the Arenic Alaquods taxonomic group, with distinct A, E, and Bh horizons. The experiment was a 4X2X2 factorial, with one control in randomized complete block design and 3 replicates. Control received neit her P-source nor WTR tr eatments. The 51 plots (20.7m x 95m each) were arranged in three blocks of 17 plots each. The factors were: four Psources (Boca Raton biosolids, Pompano bioso lids, poultry manure and TSP), two application rates of each P-sour ce (N-based, 179 kg N ha-1 and P-based, 39.6 kg P ha-1), and two application rates of WTR (0 and 1% oven dry basis). Thus, th e treatments were the same four P-sources, at two application rates each, as in the glasshouse study, but two (rath er than three) rates of WTR were tested and treatments were surface applie d (rather than incoporated throughout the rooting zone). A maximum of 1% WTR applic ation rate was tested in the fi eld due to limited availability of the material. The test plan t was established bahiagrass ( paspalum notatum Fluggae). Dry matter yields and P concentrations were determ ined for each of the 2 growing years (2003 and 2004). The P-source application rates of 39.6 kg P ha-1 (equivalent to 80 lbs P2O5 acre-1) and 179 kg PAN ha-1 (equivalent to 160 lbs N acre-1) were based on the IFAS recommended rates for bahiagrass hay (Kidder et al., 2002 ). The TSP fertilizer N-based rate used to mimic N-based amendment applications was 128 kg P ha-1 (260 lbs P2O5 acre-1). The value represented the average rate of P applied when biosolids or ma nure are applied at N-ba sed rates. Based on the analysis of the materials at time of applica tion, the N-based applica tion rates supplied 175 kg P ha-1 (357 lbs P2O5 acre-1) in Boca Raton biosolids treatments, 123 kg P ha-1 (251 lbs P2O5 acre-1) in Pompano biosolids treatments, and 81 kg P ha-1 (166 lbs P2O5 acre-1) in poultry manure treatments. Ammonium nitrate was applied to the plots that received P-base d rates of the organic

PAGE 95

95 sources and TSP plots, to equal the N supplied in treatments at N-base d rates of the organic sources, and to isolate P as the critical plant nut rient. The nitrogen fertilizer was split applied (22.7 kg per plot (50 lbs per plot) at the start of the experiment and the remainder after the first harvest (2 months later). The amendments we re applied only once during the study (May 2003), even though the study extended through Decem ber 2004. The Al-WTR (1 % by weight) was applied first, on May 9-13, 2003. The two bioso lids and manure were applied from May 13-14, 2003, while the TSP fertilizer was applied on May 19, 2003. Soil Samplings and Analysis Soil samples were taken to characterize the s ite before treatment application in May 2003. The established pasture (bahiagrass) was mo wed, but the hay was not removed before amendments were surface applied. Amendments and initial soil samples were analyzed as explained in Chapter 2. Three soil samples per plot from A (0-5cm), E (~ 20-30 cm from the top) and the first 10 cm of the Bh horizon horizons were taken in June 2003 (1 month), January 2004 (8 months), and December 2004 (19 months) after treatment application. Additional soil samples were taken from the A-horizon (0-15cm) in Ma rch and December 2004 to determine the impacts of the surface applied treatments on P chem istry throughout the A horizon. Four hurricanes (Charley, Frances, Ivan and Jea nne) impacted Florida within 44 days (August 15 and September 25) in 2004. All soil samples taken during the field st udy were analyzed for pH and EC (1:2 solid:solution), total recoverabl e P, Al, and Fe (USEPA, 1986) a nd 0.2M oxalate extractable P, Al, and Fe (Schoumans, 2000). Other parameters measured included Mehlich-1 P, WEP, and ISP. All analyses were carried out using the same procedures as in initial soil characterization (Chapter 2) and as in gl asshouse study (Chapter 3).

PAGE 96

96 The Degree of P saturation (DPSox) of th e soil was calculated as in Equation 4-1. DPSox (%) = [(Ox-P)/ (Ox-Al + Ox-Fe)]*100 (Equation 4-1) All concentrations are in mmoles and = 0.55 (Nair et al., 2004). Soil P storage capacity (SPSC) was also calculated from the oxalate extractable P, Al, and Fe as in Equation 4-2. SPSC (mg P kg-1) = (0.15 – PSR)* (Alox + Feox)*31 (Equation 4-2) PSR = P saturation ratio = (Pox)/ (Alox + Feox) (Equation 4-3) Where Pox, Alox, and Feox are 0.2M oxalate extractable P, Al, and Fe, respectively. Plant Samples and Analysis Plants were harvested twice (July and Oc tober) in 2003, and four times (July, August, October, and November) in 2004 from each plot. The grass cuttings were obtained by laying out a 1m by 1m frame on each plot, and all the gra ss within the frame cut with hand shears to a height of 5cm above the ground surface. Bahiagra ss cuttings were then placed in pre-weighed bags, and dried for several days to constant we ight at 65°C. Dried materials were afterwards weighed for yield determinations, and sub-samp les ground in a Wiley mill with stainless steel blades to pass a 20-mesh sieve. The ground plant samples were stored in airtight polyethylene containers for chemical analysis. Sub-samples of ground plant materials were ashed, treated with 6M HCl, and brought to final volume with dist illed water as descri bed by Plank (1992). Colorimetry was used for P determination in the diluted digests (Murphy and Riley, 1962), and Al content determined in digests by ICAP. The plant P and Al uptake values (expressed as kg ha1) were obtained as the pr oduct of concentration (mg kg-1 plant) and dry matter yields (kg plant ha-1).

PAGE 97

97 Statistical Analysis Soil and plant data were analyzed by an alysis of variance (ANOVA) as randomized complete block experiments using the GLM pr ocedure and the means separated by contrast procedures in SAS (SAS Institute, 1999). Th e Tukey test was also used to compare the treatments means, including cont rol treatments, where necessary using one factor (treatment) model: Yij = µ + i + ij; where i is effect of ith treatment and ij is the error terms. All statistical analysis tests were done at a si gnificance level of 5%. The correl ations between soil extractable P and plant parameters and other st atistical tests were done using co rrelation procedure in SAS and Excel software. Results and Discussion Impacts of WTR and P-source Rates on Soil P The subsurface horizons were not affected by the surface applied treatments; rather, the soil P values (WEP, ISP, M-1P, Total P, and Ox alate P) of E and Bh horizon reflected natural variability of the site (data not shown). Thus , the surface applied amen dments had no impact on the chemical properties of the subsurface E and Bh -horizons. In the event of P leaching, P likely moved freely through the E-horizon. Further disc ussions focus on changes in the A-horizon, where treatment effects were observed. Both WEP and ISP values were consistently reduced by WTR addition in all the A-horizon samples (0-5 and 0-15cm) throughout the study (Tab le 4-1). The WEP is a good estimate of soil soluble P, and ISP was developed to estimate bi ologically available P (Menon et al., 1997). The effectiveness of surface applied WTR at reducing so luble P levels of the su rface soil will ensure

PAGE 98

98 reductions in bioavailable P loss from the soil and, hence, should prevent eutrophication of the nearby aquatic systems. Analysis of variance (ANOVA) indicated the M1P values of soil samples taken from 0-5 cm depth was not affected by P-s ource, source rate, or WTR. The M-1P values were similar for plots with and without WTR at th e two P-source rates in soil samples taken from 0-5 cm between June 2003 and December 2004, and in samples taken from 0-15 cm in March 2004, but greater than in control (Fig. 4-1). The general trend of the absolute M-1P values, especially in 0-15 cm samples was: greatest value at N-based w ith WTR, and least in control treatment. Table 4-1 Effects of water treatment residual (W TR) on water extractable P (WEP) and iron strip P (ISP) values of A-horizon soils samp led between June 2003 and December 2004. All concentration values are expressed in mg kg-1 soil. <---------A-horizon (0-5cm)-------> Soil P WTR rates June 03 Jan. 04 Dec. 04 March 04 Dec. 04 0% 25.7† a 14.8 a 16.2 a 11.8 a 12.1 a WEP 1% 6.23 b 4.68 b 5.13 b 5.55 b 4.71 b 0% 29.7 a 20.8 a 18.5 a 13.7 a 15.2 a ISP 1% 16.7 b 9.74 b 14.0 b 8.95 b 11.1 b †Means of 24 samples. Each measure of soil solubl e P (WEP or ISP) of samples taken during the same period with similar letter are not diffe rent at 5% significan ce level by Tukey test. The similar M-1P values in WTR treated a nd untreated soils likely reflected partial solubilization of the WTRsorbed P by the acidic extractant, and results were consistent with those observed in the glasshouse study.

PAGE 99

99 0 10 20 30 40 50 60 70June '03 (0-5cm Depth) Jan. '04 (0-5cm Depth) Dec. '04 (0-5cm Depth) March '04 (015cm Depth) Dec. '04 (015cm Depth) Sampling periodsSoil M-1P (mg kg-1) N-based (+1%WTR) P-based (+1%WTR) N-based P-based Control a a a c ab a ab a ab b a ab ab ab ab b a a ab ab abc c bc c ab Figure 4-1 Effects of water treatment residual (W TR) rates and P-source rate on Mehlich 1 P (M1P) values of A-horizon soils at each of the sampling periods (n = 12). Treatments within the same sampling period with th e same letters are not different at p = 0.05 by Tukey test. Total recoverable P (TP) values, a measure of soil P loads, were greater in WTR treated than untreated soils (Table 4-2) and reflect both P added as part of the WTR and the WTRsorbed P. The difference between TP of WTR treated and untreated plots (>100 mg kg-1) was greater than the estimated increase in soil TP attributable to P in th e added WTR (27 mg P kg-1), which suggest P retention in WTRtreated soils and losses in untreat ed plots. The greater P loads in WTR treated than in untreated soils agrees with results of the glasshouse study (Chapter 2), and earlier findings of the capabil ity of the WTR to sorb and retain soil P (OÂ’Connor et al., 2000).

PAGE 100

100 0 50 100 150 200 250 300 350 400 June '03 (0-5cm Depth) Jan. '04 (0-5cm Depth) Dec. '04 (0-5cm Depth) March ,04 (015cm Depth) Dec. '04 (0-15cm Depth)Sampling periodsSoil TP (mg kg-1) N-based (+1%WTR) P-based (+1%WTR) N-based P-based Control a ab a b c ab a ab b a a c d a b b b bc cd dbc c ab ab c Figure 4-2 Effects of WTR rates and P-source rates on Total P (TP) values of A-horizon soils at each of the sampling periods. (Treatments w ithin the same sampling period with the same letters are not different at p = 0.05 by TukeyÂ’s test). Analysis of variance (ANOVA) indicated that neither the applica tion rates, nor the Psources affected soil soluble P values (WEP a nd ISP) throughout the study. Values of WEP and ISP were similar for the two rates (N-and P-based) in the absence of WTR, which may be due to loss of soluble P in proportion to the initial leve ls. Similarly, soil soluble P, as indicated by WEP and ISP values, were not different for the two P-source application rate s when WTR was applied, but were consistently smaller than in the abse nce of WTR (Fig. 4-3a,b). The similar soluble P measures at the two P-source rates established th at the effect of app lication rates could be

PAGE 101

101 masked by adding WTR to the soils. The soil WEP va lues of the N-based rate treatments were reduced when WTR was added, and were similar to the soluble P value observed in the control. The observed reductions in soil soluble P values (Fig. 4-3a and b) and increases in soil TP values (Fig. 4-2) when WTR is applied are inte rnally consistent and can be explained by the reduction of soil DPS values with addition of WTR throughout the study (Fig. 4-3c). In the Ahorizons (0-5 cm), DPS values we re reduced to below 25%, identified as critical for Florida soils by Nair et al. (2004). The sorpti on and retention of th e excess added P by WTR was reflected in greater soil TP values in WTR treated than in un treated soils (Fig. 4-2). Thus, as observed in the glasshouse study, the P hazard associated with th e N-based P-source application rates could be reduced, below that of the environmenta l friendly P-based ra te by adding WTR. In the absence of WTR, absolute WEP values were reduced more between June 2003 and January 2004 than between January and December 2004. However, the soil WEP values were similar at all sampling periods for soil treated w ith WTR. Thus, in the ab sence of WTR, the low sorbing soil lost substantial soil soluble P, and total soil P values decreased with time. In the presence of WTR, soil WEP values were stable at ~5 mg kg-1 with time. Also, ISP values of the A horizon soil samples (0-5cm and 0-15cm) were generally reduced with time except in WTR treated soils. The proportion of Total P that is water extrac table (percentage water extractable or PWEP) was reduced by WTR at both rate s in samples taken during the glasshouse and field studies. Added WTR reduced soil PWEP values below 10% in all the samples at both WTR rates during glasshouse and field studies (Table 4-2). Soil PWEP was greater in the absence of WTR than in the presence of WTR in field and the glasshouse study.

PAGE 102

102 Figure 4-3 Effects of WTR rates and P-source rates on (a) water extractable P (WEP), (b) Iron strip P (ISP), and (c) degree of P sorption (D PS) values of A-horizon soils at each of the sampling periods. (Treatments within the same sampling period with the same letters are not different at p = 0.05 by TukeyÂ’s test). (a) Soil water extractable P (WEP)0 5 10 15 20 25 30 June '03 (0-5cm Depth) Jan. '04 (0-5cm Depth) Dec. '04 (0-5cm Depth) March ,04 (015cm Depth) Dec. '04 (0-15cm Depth)Sampling periodsSoil WEP (mg kg-1) N-based P-based Control N-based (+1%WTR) P-based (+1%WTR) a b ab b b a b ab a a c c ab a a b b bb a a bb a a (b) Soil iron strip P (ISP)0 5 10 15 20 25 30 35 40 June '03 (0-5cm Depth) Jan. '04 (0-5cm Depth) Dec. '04 (0-5cm Depth) March ,04 (015cm Depth) Dec. '04 (0-15cm Depth)Sampling periodsSoil ISP (mg kg-1) N-based P-based Control N-based (+1%WTR) P-based (+1%WTR) a b ns ab b b bb a a ab ns ns (c) Soil degree of P saturation (DPS)0 25 50 75 100 125 June '03 (0-5cm Depth) Jan. '04 (0-5cm Depth) Dec. '04 (0-5cm Depth) March ,04 (015cm Depth) Dec. '04 (0-15cm Depth)Sampling periodsSoil DPS (%) N-based P-based Control N-based (+1%WTR) P-based (+1%WTR) a a b a ab a c a b b b d ab b b b c c c a a b b a a a b b

PAGE 103

103 Table 4-2. Mean (n = 12) percent water extract able P (PWEP) values of soil samples taken during the glasshouse and field studies, as affected by WTR at th e two P-source rates. All PWEP values are expressed in %. 0% WTR 1%WTR 2.5% WTR Study Sampling depth Sampling date Nbased Pbased Nbased Pbased Nbased Pbased June ‘04 ‡16.9 a 15.6 a 9.00 b 6.40 c 4.40 cd 3.10 d Dec. ‘04 22.3 a 17.8 a 5.10 b 4.60 b 3.10 b 3.00 b May ‘05 11.4 a 5.30 b 1.94 b 1.77 b 1.40 b 1.14 b Glasshouse †N/A Sept. ‘05 4.74 a 4.13 a 1.48 b 1.44 b 0.99 b 0.77 b June ’03 12.8 a 13.9 a 1.56 b 2.20 b N/A N/A Jan. ’04 11.3 a 13.2 a 1.90 b 1.98 b N/A N/A 0-5 cm Dec. ’04 13.3 a 16.9 a 2.63 b 3.58 b N/A N/A Mar. ‘05 10.1 b 17.3 a 3.95 c 6.03bc N/A N/A Field 0-15 cm Dec. ‘05 15.3 a 18.9 a 4.80 b 5.73 b N/A N/A †Not applicable. ‡Means follow by same letter in each row (sam e sampling period) are not different at 5% significance level by Tukey test. Impacts of P-Sources on Soil P Analysis of variance indicated that soil solubl e P values (WEP and ISP) were not affected by the sources of P. However, soil Total recovera ble P values differed for the different P-sources (ANOVA Table). Impacts of the P-sources on the soil P is bett er studied in the absence of WTR, where Psources and application rates are is olated as variables affecting m easures of soil P. Soil soluble P (WEP) values were similar for the different Psources at both source ap plication rates throughout the study (Fig. 4-4a).

PAGE 104

104 Figure 4-4. Trends of (a) soil wate r extractable P (WEP) and (b) To tal recoverable P (TP) values for treatments with different P-sources at the two P rates in the absence of water treatment residual (WTR). (Treatments w ithin the same sampling period with the same letters are not different at p = 0.05 by TukeyÂ’s test). (a) Water extractable P (WEP)0 5 10 15 20 25 30 35 40 June '03 (0-5cm Depth)Jan. '04 (0-5cm Depth)Dec. '04 (0-5cm Depth)March ,04 (0-15cm Depth)Dec. '04 (0-15cm Depth)TreatmentsWEP (mg kg-1) Boca Raton biosolids (N-based) TSP (N-based) Poultry manure (N-based) Pompano biosolids (N-based) Boca Raton biosolids (P-based) TSP (P-based) Poultry manure (P-based) Pompano biosolids (P-based) Control ns ns ns ns (b) Total P (TP)0 50 100 150 200 250 300 June '03 (0-5cm Depth)Jan. '04 (0-5cm Depth)Dec. '04 (0-5cm Depth)March ,04 (0-15cm Depth)Dec. '04 (0-15cm Depth)TreatmentsTP (mg kg-1) a a ab ab ab ab ab ab b a ab ab ab ab ab ab ab b a ab abc bc bc bc bc cbc a b c c c cdc de e a ab abc bc bcd bcd cd abc d ns

PAGE 105

105 Also, Total P values for samples taken in June 2003 and January 2004 (0-5 cm) were similar for different P-sources, and only differe nt from control treatment. However, samples taken in March and December 2004 from A-horizon (0 -15cm depth) indicate d greatest soil TP in Boca Raton biosolids an d least in control. Impacts of Treatments on Soil Sorption Properties The degree of phosphorus saturation (DPS) is an index of how saturated the P sorption sites of soil are and, hence, a measure of soil cap ability to retain and pr event P losses from the soil through runoff and leaching. Soils with large D PS values indicate the soil sorption sites are saturated and suggest littl e capability of th e soil to retain additional P. 0 25 50 75 100 125June '03 (0-5cm Depth) Jan. '04 (0-5cm Depth) Dec. '04 (0-5cm Depth) March ,04 (0-15cm Depth) Dec. '04 (0-15cm Depth)Sampling periodsSoil DPS (%) N-based P-based Control N-based (+1%WTR) P-based (+1%WTR) a a b a ab a c a b b b d ab b b b c c c a a b b a a a b b Figure 4-5 Effects of WTR rates and P-source rates on degree of P saturation (DPS) values of Ahorizon soils at each of the sampling peri ods. Treatments within the same sampling period with the same letters are not different at p = 0.05 by TukeyÂ’s test.

PAGE 106

106 Source of P did not affect the soil DPS va lue throughout the study. The effects of WTR on soil DPS values at each P-source rate are show n in Fig. 4-5 for the samples between June 2003 and December 2004. The DPS values of all soil s not treated with WTR exceed 25%, whereas most WTR amended soils have DPS below the criti cal value of ~ 25% sugge sted for Florida soils (Nair et al., 2004). The reduced soil DPS values in the presence of WTR are consis tent with the observations from the glasshouse study, and conf irm the ability of the WTR to enhance the sorption properties of the sandy, low P-sorbing Florida soils. The increasing soil sorption sites indicated by reduced DPS values are consistent with reduced WEP values (Fig. 4-6). The soil WEP values are expected to indi cate potential for soil P loss through runoff (Pote et al., 1999; Vadas et al., 2006), and leachate (Kleinman et al., 2000). The WEP values were lower than 10 mg kg-1 in soils receiving WTR, but 10 mg kg-1 in soils without WTR, throughout the study (Fig. 4-6). Als o, a change point can be identified by CateNelson (1977) type of approximation at ~25% DPS. Above 25% DPS, most soil WEP values were 10mg kg-1, but the WEP values were lowered than 10 mg kg-1 below 25% DPS. The WEP values were also shown to relate we ll with DPS values by Nair et al. (2004) in Florida soils. A threshold (change point) value of ~25% was also observed by Breeuwsma et al. (1995) in sandy soils of the Netherlands, while ~ 20% was identified as the threshold DPS value for Florida soils by Nair et al. (2004). The relationship be tween M-1P values and DPSox values indicated that DPSox values of 22% and 28% are equivalents to 20 mg kg-1 M-1P (agronomic high) and 50 mg kg-1 M-1P (very high) values, respectivel y. Irrespective of the P-source application rates, the DPS values of treat ments receiving no WTR we re greater than 25% throughout the two year study. This suggests that applying any P-sour ce, even at P-based rates,

PAGE 107

107 could increase the risk of P loss. However, applying WTR could mask the effects of both the source and rate of P applied, and, hence, dramatically reduce P loss potential. 0 5 10 15 20 25 30 35 400102030405060708090100110120130140150160170180DPS (%)WEP (mg kg-1) June 03 No WTR (0-5cm) June 03 With WTR (0-5cm) June 03 Control (0-5cm) Jan 04 No WTR (0-5cm) Jan 04 With WTR (0-5cm) Jan 04 Control (0-5cm) March 04 No WTR (0-15cm) March 04 With WTR (0-15cm) March 04 Control (0-15cm) Dec. 04 No WTR (0-5cm) Dec. 04 With WTR (0-5cm) Dec. 04 Control (0-5cm) Dec. 04 No WTR (0-15cm) Dec. 04 With WTR (0-15cm) Dec. 04 Control (0-15cm) Figure 4-6. Soil water extractable P (WEP) values as affected by degree of P saturation (DPS) values of soil samples taken between June 2003 and December 2004 (Vertical dotted line locates the 25% DPS environmental threshold. Soil phosphorus storage capacity (SPSC) valu es of samples taken in June 2003, January 2004 and December 2004 from the three soil horizons are shown in Fig. 4-7 (E and Bh horizons) and Fig. 4-8 (A-horizons). The minimal impact of the surface applied treatments on soils in the subsurface horizons is established by the similar SPSC values of the E and Bh horizons for all treatments, irrespective of the sampling period (F ig. 4-7). The SPSC values of the E-horizon soil samples are similar and negative (or approximately zero). Such values indi cate near saturation of P sorption sites on the soils and inability to retain added P (Nair and Harris, 2004). The SPSC values of the Bh horizon samples were similar and positive (~147 mg kg-1) for all treatments.

PAGE 108

108 Positive SPSC values, an indication of soil capacity to retain added P, are expected of the Al-rich Bh-horizon, and the similarity of the values fo r the different treatments over time reflects the minimal effect of the surface-applied treatments. Figure 4-7. Soil phosphorus stor age capacity (SPSC, mg kg-1) values of subsurface (E and Bh horizons) samples from the field study (June 2003 – Dec. 2004). Note the differences in scales of the 2 figures. (There were no significant treatments effects (P-source, P rate or WTR) at p = 0.05 on SPSC values of E and Bh horizon soil sampled at any time during the study (ANOVA)). (a) E-horizon SPSC -20 -15 -10 -5 0Controls Manure-N Manure-N+WTR Manure-P Manure-P+WTR Boca-N Boca-N+WTR Boca-P Boca-P+WTR Pomp-N Pomp-N+WTR Pomp-P Pomp-P+WTR TSP-N TSP-N+WTR TSP-P TSP-P+WTRTreatmentsSoil SPSC (mg kg-1) June 2003 samples January 2004 samples December 2004 samples (b) Bh-horizon SPSC 0 50 100 150 200Controls Manure-N Manure-N+WTR Manure-P Manure-P+WTR Boca-N Boca-N+WTR Boca-P Boca-P+WTR Pomp-N Pomp-N+WTR Pomp-P Pomp-P+WTR TSP-N TSP-N+WTR TSP-P TSP-P+WTRTreatmentsSoil SPSC (mg kg-1)

PAGE 109

109 Figure 4-8. Soil phosphorus stor age capacity (SPSC, mg kg-1) values of A-horizon (0-5cm) samples from the field study (June 2003 – D ec. 2004) as affected by the treatments. (Treatments ending in P, and N, are P-ba sed and N-based rates of the sources, respectively). Treatments with the same letters are not different at p = 0.05 by Tukey test. (a) June 2003 (0-5cm) samples -200 -100 0 100 200 300 400 500 600Controls Manure-N Manure-P Boca-N Boca-P Pom pano-N Pom pano-P TSP-N TSP-PP sources and their application rates Soil SPSC (mg kg-1) Without WTR With WTR a ab abcde bcde de de de e abcd de de bcde abc de bcde de bcde (b) January 2004 (0-5cm) samples -200 -100 0 100 200 300 400 500 600Controls Manure-N Manure-P Boca-N Boca-P PompanoN PompanoP TSP-N TSP-PP sources and their application rates Soil SPSC (mg kg-1) Without WTR With WTR a ab bc abc abc bc c c c c c c c c c abc abc (c) December 2004 (0-5cm) samples -200 -100 0 100 200 300 400 500 600Controls ManureN ManureP Boca-N Boca-P Pompan o-N Pompan o-P TSP-N TSP-PP sources and their application rates Soil SPSC (mg kg-1) Without WTR With WTR a a bcd bcde bcde bc bc bc def def def ef def f cdef def cdef

PAGE 110

110 The impacts of the surface applied treatments were obvious in the SPSC values of samples from A-horizons (0-5cm). Similar to observati ons in the glasshouse st udy, soil samples from plots amended with P-sources without WTR had negative SPSC values, and SPSC values for the N-based treatments were more negative than fo r P-based treatments (Fig 4-8). Thus, when the soils are amended with P without WTR, the P-st orage capacity was reduced more (more negative SPSC value) at N-based rates than at P-based rates. Similar to the trends of the DPS data, the SPSC values (and, hence, P sorption properties) were improved by addition of WTR. The more positive the SPSC value, the greater the P sorption capacity. Impacts of Treatments on Plants Plant dry matter yields, yield-we ighted P concentrations, and P uptake data for the 2003 and 2004 harvests are summarized in Figure 4-9. The gr eater variability in th e dry matter yields of 2003 than in 2004 likely reflected natu ral, nutritional variations in the fields before treatment applications, and before some tr eatments took effect. The yields of the two years were similar (for each treatment) even though harvesting was done two times in 2003 compared to four times in 2004 and treatments were applied onl y in 2003, at the start of the study The yield-weighted P concentrations were reduced with addition of WTR for each Psource, and source application rate during both growing seasons. The reduced plants P concentrations agree with the results from the glasshouse stud y. Other studies also indicated reductions in plant P concentrations with addi tion of WTR (Heil and Ba rbarick, 1989; Elliott and Singer, 1988), and suggest that not all P sorbed by WTR is accessible by the plant. However, all plant P concentrations in the fiel d study were above the critical le vel, as indicated by the uniform yields.

PAGE 111

111 Figure 4-9. Plant dry matter yields (Mg ha-1), yield-weighted P concentrations (g kg-1), and P uptake (kg ha-1) from 2003 and 2004 harvests (Err or bar represent one standard deviation (n =3)). Plant P uptake in 2003 was not affected by the source of P or P-sour ce rates, but the P uptake was reduced in plots treated with WTR. In 2004 harvests, P uptake was not affected by WTR treatments at the N-based rates of manure or TSP (Table 4-3), but reduced P uptake by WTR at N-based rates of the two biosolids, and at P-based rate of all P-sources were observed. The trend of P uptake supports the findings from the glasshouse study th at applying WTR to Pbased treatments will reduce the P uptake , irrespective of th e P-source applied. (a) Plant dry matter yields 20030 2 4 6 8 10 12 ControlManureBocaPompanoTSP Source of PDry matter yield (Mg ha-1) N-based no WTR N-based WTR P-based no WTR P-based WTR Control (b) Plant dry matter yields 20040 2 4 6 8 10 12 ControlManureBocaPompanoTSP Source of PDry matter yield (Mg ha-1) (c) Plant P concentration 20030 1 2 3 4 ControlManureBocaPompanoTSP Source of PP concentration (g kg-1) (d) Plant P concentration 20040 1 2 3 4 ControlManureBocaPompanoTSP Source of PP concentration (g kg-1) (e) Plant P uptake 20030 5 10 15 20 25 30 ControlManureBocaPompanoTSP Source of PP uptake (kg ha-1) (f) Plant P uptake 20040 5 10 15 20 25 30 ControlManureBocaPompanoTSP Source of PP uptake (kg ha-1)

PAGE 112

112 Table 4-3. Effects of P-sources, P-source rates, and water treatment residual (WTR) rates, on plant P uptake (kg P ha-1) values of bahiagrass harvested in 2004. ------------Rate-----------Source of P WTR Rate / Contrast N-based P-based Contrast N-based vs. P-based 0% 12.4† 14.8 NS 1% 9.7 9.9 NS Manure Contrast NS * 0% 13.0 18.5 * 1% 10.9 9.4 NS Boca Raton biosolids Contrast * * 0% 16.3 14.5 NS 1% 10.0 10.1 NS Pompano biosolids Contrast * * 0% 17.1 13.1 * 1% 12.3 9.8 NS TSP Contrast NS * Manure vs. Biosolids NS NS Organic vs. Mineral * NS Contrast (At 0% WTR) Boca Raton vs. Pompano * NS Manure vs. Biosolids NS NS Organic vs. Mineral * NS Contrast (At 1% WTR) Boca Raton vs. Pompano * NS † Means of three samples * Significant at p = 0.05 by contrast NS not significant at p = 0.05 by contrast In addition to the effect of WTR, P uptake was also affected by the source of P, and Psource application rates in 2004 (Table 4-3). In the absence of WTR, P uptake from TSP treated plots was greater at N-based rates than at P–based rates, which reflects the greater initial P loads of the N-based rates. Phosphorus was applied only once (in 2003) and, at th e P-based rate, is not sufficient for growth in the following year (2004 ). However, at the N-based rate, the excess P applied in the 2003 was apparently sufficient to sustain plant growth through 2004. Plant P uptake values were similar at the two P rates fo r all sources of P when WTR was applied. Thus, similar to the effects on the soil soluble P values, WTR apparently masked the effects of Psource rates on plant P uptake, irre spective of the P-source applied.

PAGE 113

113 Phosphorus uptake values were also similar for treatments with different P-sources at each of the two levels of WTR at P-based rate. In the N-based treatments, P uptake was greater in TSP-treated than organic source of P-treated soils at each level of WTR. In the absence of WTR, and at the N-based rate, greater P uptake was obs erved in Pompano than in Boca Raton biosolids treated plots. The greater loss of soluble P fr om the high water soluble P Boca Raton biosolids treated soils, than from the moderate water solubl e P Pompano biosolids trea ted soils, could have resulted in greater P uptake from Pompano bios olids than from Boca Raton biosolids treatments. The glasshouse study also indicated that, in the absence of WTR, Pompano biosolids could mimic slow release fertilizer as regards P releas e to plants. The greater P uptake in Boca Raton biosolids than in Pompano biosolids treatments at N-based rate (and at 1%WTR) suggests a portion of the WTR-sorbed P from the high water soluble P Boca Raton biosolids is accessible by the plants. The yield-weighted P concentrations in 2003 harv ests were greater at N-based rates than Pbased rates, and greater in the absence than in the presence of WTR. However, in the 2004 harvests, there were greater plant P concentrations in the absence than in the presence of WTR. Greater plant P concentrations in WTR treated than untreated soil in 2004 suggested that a portion of P retained in the soil by the WTR is accessed by the plants. Despite the reduction in plant P concentrations by added WTR, plant yields were not affected in either growing seasons (Fig. 4-10). The plant yields of 2003 and 2004 were also not affected by source of P, nor the source application rates. This indicates sufficien t plant P concentrations for optimum growth of the plant.

PAGE 114

114 Figure 4-10. Effects of water tr eatment residual (WTR) rates a nd P-source rates on plant DM yield, P concentration and P uptake. (Treatments within the same sampling period with the same letters are not different at p = 0.05 by Tukey test). In summary, applying WTR increased P sorpti on of the low P-sorbi ng soil and, thereby, reduced soil soluble P (Fig. 4-10). Plant yields we re not affected by the reduced soil soluble P, (a) Plant dry matter yield 0 2 4 6 8 10 12 14 20032004Total DM Sampling periods DM (Mg ha-1) N-based P-based Control N-based (+1%WTR) P-based (+1%WTR) nsns ns (b) Plant P concentration 0.0 0.5 1.0 1.5 2.0 2.5 3.0 20032004weighted over 2003 and 2004 Sampling periods P concentration (g kg-1) ab a a a b b b b a a a a ab ab b (c) Plant P uptake 0 5 10 15 20 25 30 35 20032004Total P uptake Sampling periods P uptake (kg ha-1) ns a a a a b b abc ab bc c

PAGE 115

115 nor by the reduced plant P concen trations (Fig. 4-10). Thus, th e readily desorbable P and P potentially available for loss to runoff and leachate were reduced by amendment with WTR. Improving the P sorption property of a soil with WT R is expected to have no negative effects on the yield, though plant P uptake may be reduced. Thus, similar to the observations in the glasshouse study, we fail to rej ect the null hypothesis, and conc lude that applying WTR to Nbased rates of P-sources reduces negative envi ronmental P impacts and optimizes the agronomic benefits (as in P-based rates without WTR). Plant and soil data from both glasshouse a nd field study established that problems of excessive soluble P associated w ith high P rate (N-based) can be controlled without affecting bahiagrass yields. Relative Phosphorus Phytoavailability (RPP) The observed reductions in soil soluble P with time could be due to P uptake, increased P retention by WTR, and P loss. Thus, the relative ava ilability of P from orga nic sources to plants, compared with mineral fertilizer, is better studie d in the absence of WTR and where P losses can be quantified. The differential P loss in the abse nce of WTR was minimal in the first growing season compared to the 2004 growing season as i ndicated by the trend of soil TP (Fig. 4-2). Thus, the 2003 data with minimal P loss were rega rded as most appropriate for comparing the Psources. The “relative P phytoavailability” (RPP) was cal culated for each P-source at P-based rates without WTR. The P-based rates supplied equal am ounts of P from the different P-sources to the plants (unlike the N-based rate), and all plots were supplied with adequate nitrogen. The P loss was accounted for by averaging time zero and time final total soil P conc entrations for the 2003 growing season and P uptake calculated per unit of average soil P for each source. The P uptake

PAGE 116

116 per unit of soil P from plots treated with differe nt P-sources was then divided by similar ratio obtained for the TSP treatment to arrive at RPP expressed in percentage for each P-source. The different P-source treatments were related to the TSP treatment because it contained sufficient readily available mineral P, and was expect ed to give optimum agronomic performance Thus, RPP were calculated as in Equation 4-4. (P uptakesource/Average soil P conc.source) RPP = ------------------------------------------------* 100 (Equation 4-4) P uptakeTSP/ Average soil P conc.TSP The RPP values obtained for the organic sour ces of P ranged from 55% for poultry manure to 85% in Boca Raton biosolids (Table 4-4). Most biosolids evaluated by de Haan (1981) were between 36 and 90% as phytoava ilable as fertilizer P. Table 4-4. Relative P phytoavailab ility (RPP) values for the diffe rent P-sources during the field study. Source of P RPP (%) Manure 55 Boca Raton biosolids 85 Pompano biosolids 59 TSP 100 O’Connor et al. (2004) rated the relative P phytoavailabil ity of various biosolids (compared to TSP), and suggested three classes: high (similar to TSP, RPP >75%), moderate (RPP = 25 – 75%), and low (RPP <25%). Values of RPP differed with biosolids processing, total Fe, and Al concentrations, and %solids conten t. Results from both th e glasshouse and field studies (Table 4-5) show the R PP of Boca Raton biosolids compares favorably well with the RPP for TSP, and the biosolids would be regard ed as a high RPP organic source. Boca Raton biosolids are produced via a process similar to biological P removal process (BPR), which O’Connor et al. (2004) suggest ed should have high RPP valu es. Pompano biosolids have

PAGE 117

117 characteristics (Table 2-1) that suggests it should have moderate RPP (Total Al + Fe concentration = 42 g kg-1; PSI = 0.7). Biosolids of similar properties (e.g., Tarpon Spring cake; Total Al + Fe concentration = 42 g kg-1) studied by O’Connor et al . (2004) also fell into moderate P phytoavailability category. Table 4-5. Relative P phytoavailab ility (RPP) values for the diffe rent P-sources from field and glasshouse studies. All RPP va lues are expressed in %. <----------------G lasshouse study-----------------> P-source Bahiagrass (1) Ryegrass Bahiagrass (2) †Combined Field study RPP ‡Category Manure 30 55 36 43 55 Moderate (25-75%) Boca Raton biosolids 82 68 82 82 85 High (>75%) Pompano biosolids 41 55 58 53 59 Moderate (25-75%) TSP 100 100 100 100 100 High (>75%) † RPP based on sum of P uptake from the three cropping. ‡ Categorized according to O’Connor et al. (2004). The data collected from both glasshouse and field studies (Table 47) suggest that both Pompano biosolids and manure have RPP represen tative of the moderate category. The range of the values of moderate RPP P-source category is consistent with 50% suggested for organic source of P by USEPA guidelines (USEPA, 1995). A 67% relative e ffectiveness of the biosolids used in a recent study by Pritchard (2005) was also reported. The RPP values reported for the varying biosolids studied by O’Connor et al. (2004) ranged between <10% and 100% RPP, compared to TSP. The range of values is sim ilar to the 10 – 100% range reported by de Haan (1988). McLaughlin and Champion (1987) observed >90% RPP for an anaerobically digested biosolids, compared to monocalci um phosphate fertilizer (MCP).

PAGE 118

118 The Quebec Canada regulatory agency proposed s hort term estimates of P availability from biosolids (%Pavailable) can be calculated using Equation 4-5 (C entre de référence en agriculture et agroalimentaire du Québec (CRAAQ), 2003; Québ ec Ministère de l’Environnement (MENV) 2003): %Pavailable = 70-{concentration (Altotal + 0.5 Fetotal [mg kg-1]) 20,000}/2000 (Equation 4-5) The equation has not been field validated, and is based on the following assumptions: For biosolids not treated with Al or Fe salts, P availabil ity is about 70% that of mineral fertilizers. Manure P availability also ranges from 65-80% compared with mineral fertilizer. With increasing Al and Fe content, biosolids P availability is assumed to decrease linearly to an arbitrary value of 5% for an Al + 0.5 Fe content of 150,000 mg kg-1. An Al + 0.5 Fe content of 20,000 mg kg-1 represents the backgr ound content of biosolids with no Al or Fe salt added. (Note that Al and Fe both have valences of +3, and atomic weights 27 and 56, respectively and, hence, Al is equivalent to 0.5 Fe). Using Equation 4-5, the %Pavailable obtained for the organic sour ce of P used in this study are 79% (poultry manure), 69% (Boca Raton bi osolids), and 67% (Pompano biosolids). The %Pavailable obtained did not agree with the observed RPP, rather it has an inverse relationship with the P-source RPP values measured. The %Pavailable obtained, tracked well (negative correlation) with the Al + 0.5 Fe concentrations (1.65, 21.5, and 25.6 g kg-1 for poultry, Boca Raton biosolids, and Pompano biosolids, respectively) of the P-sources which are the only variables used for the estimation. An attempt was made to estimate the RPP of the twelve (12) biosolids used by O’Connor et al. (2004) us ing Equation 4-5. The calculated %Pavailable values were also very different from the observed RPP values. Th us, the RPP may not be adequately estimated

PAGE 119

119 from the %Pavailable equation suggested by Quebec Canada re gulatory agency without accounting for other properties of the P-sources. Data from the study carried out by OÂ’Connor et al. (2004) with 12 biosolids were pooled with data from this study to id entify properties of bios olids that could affect and account for their RPP values. Manure was not included since only one manure type (poultry manure) was used in this study, which is insufficient to make infere nce about the wide spectrum of manure types. Properties of biosolids determined included Total P, Fe, Al, Ca, Mg, % solids, % organic matter, and C:N. The forms of P in the materials were also characterized by se quential extraction into KCl-P, NaOH-P, and HCl-P fractions. Varying measur es of biosolids extractable P (citric acid P, M-1P, WEP, Oxalate-P), percentage water extrac table P (PWEP), oxalate extractable Fe, Al, and PSI were also determined. Properties of biosolid s that could account for RPP variability were identified by stepwise regression of the obser ved RPP values with the properties of the Psources. Altogether, 14 biosolids were considered (including Bo ca Raton biosolids and Pompano biosolids from this study). Among the twenty variables used for the regr ession, only three variables (Total P, NaOHP, and %solids) were identified by the stepwise re gression as affecting the RPP values (Table 4-6 and 4-7). The results indicat ed that Total P and NaOH-P, could account for ~90% (r2 = 0.9; p<0.05) of the variability in RPP values, and 95 % of the variability could be accounted for by including %solids of the P-sources. Total-P concen tration of the P-sources was identified as the most important variable that could account for the RPP variability. Over 70% of the variability (r2 = 0.74) in RPP values could be accounted fo r by the Total P concen tration. Inclusion of NaOH-P with Total-P concentra tion improved the regression (r2 = 0.90).

PAGE 120

120 Table 4-6. Stepwise regression of relative P phyto availability (RPP) valu es of biosolids with some of the biosolids properties (incl udes data from OÂ’Conn or et al., 2004) Step Variable Estimate p -value (regression parameters) r2 p -value (regression equation) Intercept -115 0.0114 1. Total P entered Total P 5.89 0.0015 0.73 0.0015 Intercept -63.7 0.0555 Total P 5.17 0.0005 2. NaOH-P entered NaOH-P -2.08 0.0113 0.90 0.0003 Intercept -42.2 0.1049 Total P 4.64 0.0004 NaOH-P -1.84 0.0070 3. % Solids entered % Solids -0.24 0.0398 0.95 0.0002 Table 4-7. Summary of regression parameters for the variables selected by the stepwise regression of relative P phytoa vailability (RPP) values of biosolids with some of the P-sources properties Regression step Variable entered Number of variables in the model Partial r2 Model r2 p -value 1 Total P 1 0.74 0.74 0.0015 2 NaOH-P 2 0.16 0.90 0.0113 3 % Solids 3 0.05 0.95 0.0398 The NaOH extractable P represents the Fe and Al-bound P fraction, which may not be readily available to the plants. The negative regr ession coefficients associated with the NaOH-P support the negative impact of the P form on RPP value. Thus , the equation suggests Total-P less NaOH-P of the source could accoun t for 90% of the variability in RPP estimation. The NaOH-P term accounts for differences in RPP estimate from the source Total-P by considering Fe and Al composition of the sources, and impr oves the estimate of RPP values (r2 = 0.90). The Al and Fe concentrations of the biosolids were also not ed as important properties by OÂ’Connor et al.

PAGE 121

121 (2004). The important role of Al and Fe concentr ations in biosolids P lability could explain why the two variables were accounted for in an attempt by Quebec Canada regulatory agency to estimate %Pavailable of biosolids using Equation 45 (CRAAQ, 2003; MENV 2003). Another source variable identified as important by the regression analysis was %solids, a physical property of the source that reflec ts the extent of dewatering us ed in producing the biosolids. Increasing the %solids by pelletiz ing biosolids has been reported to reduce their RPP. The RPP of Largo and Baltimore cakes were reported redu ced when pelletized by OÂ’Connor et al. (2004). Smith et al. (2002) also indicated that heat dryi ng significantly reduced bi osolids P availability. The present study is not suffici ent to recommend estimating RPP of biosolids from the regression equation, but serves to identify P-source properties that could be the focus of further study to arrive at a robust pred iction equation for RPP. The study shows that the RPP could be adequately predicted from the source physic al and chemical properties (Fig. 4-11). y = 0.96x + 0.61 r2 = 0.920 20 40 60 80 100 120 140 020406080100120140 Observed RPP (%)Predicted RPP (%) Figure 4-11. The relative P phytoavailability (RPP) as predicted by the Total P, NaOH-P, and %solids values of the biosolids plotted against the observed RPP.

PAGE 122

122 However, the study was largely limited to bios olids applied to a P-deficient Florida sandy soil. Studies with other organic sources especi ally manure will be n eeded, as only one manure type (poultry manure) was used in this study, whic h is insufficient to make inference about the wide spectrum of manure types. Also, as hinted by O’Connor et al. ( 2004), effects of applying different organic sources to soils with adequate P or large P-retention capacities may be masked. Aluminum Toxicity Plant Al concentrations and uptake for each of the harvests in the 2003 and 2004 growing seasons are summarized in Table 4-8. Table 4-8. Effect of aluminum water trea tment residual (Al-WTR) on bahiagrass Al concentrations and uptak e during the field study WTR Treatments Planting Season Parameters Sampling period Without WTRWith WTR July 19.3 b† 77.2 a Al concentrations (mg kg-1) October 18.8 b 141 a July 75.6 b 316 a October 55.3 b 375 a 2003 Al uptake (g ha-1) Total 131 b 691 a July 41.8 a 70.7 a August 13.6 a 12.3 a October 5.75 a 8.53 a Al concentrations (mg kg-1) November 32.3 b 123 a July 103 a 178 a August 22.7 a 16.5 b October 6.99 a 9.72 a November 30.0 b 147 a 2004 Al uptake (g ha-1) Total 163 b 352 a † Means (n = 24) within the same sampling period (ro w) with similar letter are not different at 5% significance level by Tukey test. The greater plant Al concentrations observed in WTRtreated plots th an in untreated plots in the 2003 cropping season (Table 4-8) could have resulted from contamination by the surface applied WTR adhering to harvested bahiagrass. The impact of the contaminated pasture on

PAGE 123

123 grazing animals is still being investigated. Th e contamination makes the 2003 data unsuitable (rather 2004 data may be more suitable) to validate the impact of Al-WTR on plant Al concentrations. Similar to the observations in the glasshouse, the plant Al concentrations from the 2004 cuttings were similar in treatments with a nd without WTR. November plant samples show greater Al concentrations in WTR-amended th an in unamended soils, likely due to cross contamination of treatments induced by the hurricane activities. The similar plant Al concentrations in WTR-treated and untreated trea tments is expected because insoluble Al oxides (Al-WTR) are not expected to release toxic Al c oncentrations or to produce acidity in soils or aqueous systems with pH above 5.2 (Peter s and Basta, 1996; Codling et al 2002). Summary and Conclusions Results of the field study ar e consistent with most findings from the glasshouse study. Reductions in soil Total P concentrat ions with time in soils treated with different P-sources were observed, especially in N-ba sed treatments when no WTR was added, and suggest that substantial soil P losses occurr ed. In presence of WTR, soil DPS values were reduced, SPSC values increased, and soil soluble P measures (W EP and ISP) as well as P loss potential were reduced. Thus, similar to the resu lts from the glasshouse study, th e field data suggest that the added WTR reduced the environmental P hazards a ssociated with the excess P supplied into the soil especially at N-based rates of P-sources. The reduced soil so luble P values (WEP and ISP) were not reflected in reduced yields during the two growi ng seasons. Yield-weighted P concentrations were reduced in WTR-amended tr eatments, but plant growth was not. Also, plant Al concentrations were similar in WTR-trea ted and untreated soils. Thus, amending soils, especially Florida sands, with WTR could be a best management prac tice (BMP) to reduce the

PAGE 124

124 hazards associated with excess P from mineral a nd organic sources of P la nd-applied at high rate such (N-based) wit hout fear of reduction in plan t yields or Al phytotoxicity. The organic sources of P varied in RPP values in the field in a similar manner as the values observed in the glasshouse study. The field study RPP values of the Pompano biosolids and poultry manure agreed with the expected moderate phytoavailable biosolids class determined in the glasshouse study, and the Boca Raton biosolids RPP values from both studies were classified as high. Properties identified to affect the R PP values of biosolids are Total P concentration, NaOHP and %solids. These properties could be the focus of further study into estimating RPP values.

PAGE 125

125 CHAPTER 5 EVALUATION OF SOIL TEST METHODS FOR FLORIDA SANDS TREATED WITH VARIOUS P-SOURCES AND WATER TREATMENT RESIDUAL (WTR) Introduction Amorphous, hydrous oxide-rich water treatment residuals (WTR) can control excessive soluble P in coarse-textured low sorbing coastal soils, like the abundant Florida sands. Studies have shown that various types of WTRs can be effective soil amendments in Florida soils (Brown and Sartain, 2000; OÂ’Connor and Elliott, 2001; OÂ’Connor et al., 2002a). Changes in basic soil properties that can affect P availability to pl ants by land applying WTR, demand evaluating the suitability of soil te st methods for assessing plant available P, and plant response (Basta et al., 2000). A good agronomic soil test me thod extracts a soil nutrient pool that is representative of th at available to plants and that is well correlated with the plant nutrient uptake and other plant growth responses. Soil test methods for P (STP) were developed based on expected P forms in the soils. In most soils, P is associated with either Al, Fe in acid to neutral soils (pH < 7) and with Ca in calcareous soils (Hed ley and McLaughlin, 2005). Other factors that determine the degree to which P is bound in a soil incl ude the types of Fe, Al, or Ca compounds (amorphous or crystallin e), amounts of P present in the soil, and soil properties, e.g., organic matter content, mineralogy, and pH. T hus, a STP method for a particular region is determined by the predominant soil P sorption pr ocesses expected, as affected by regional soil physical and chemical characterist ics. The acidic extractant, Mehlic h-1 (M-1P) soil test is used extensively in the Southern and Mid-Atlantic States due to th e predominance of acidic, highly weathered, low CEC soils in the re gion. The interpretations of the STP vary with soil properties. For example, Mehlich 1P (M-1P) values of 20-25 mg P kg-1 are considered optimum for plant growth in sandy soils, but 10 mg P kg-1 is considered adequate for plants in fine-textured soils

PAGE 126

126 (Kamprath and Watson, 1980). In Florida, M-1P values < 15 mg kg-1 are regarded as low, > 30 mg kg-1 is considered high, and values > 60 mg kg-1 are considered very high, from an agronomic standpoint (Kidder et al., 2002). Various reports have indicated the inadequ acy of traditional soil test P methods as measures of plant response to P in WTR-am ended soils (OÂ’Connor et al., 2002; Basta et al., 2000). Basta et al. (2000) evaluated three Al-WTRs as soil substitute s and the ability of soil tests to predict P adequacy for Bermudagrass (c ynodon) . They found no yield or tissue P response, even though both M3-P and Olsen P predicte d P responses. The acidic extractant (M3-P), overestimated plant available P in WTRs by dissolving P sorbed by amorphous Al. Water extractable P (WEP), which accesses the most lab ile forms of soil P, was suggested as predictor of plant available P in WTR (Basta et al., 2000). Both water-soluble P and Olsen P were reported useful to predict the ability of WTRs to suppor t plant growth, but not P adequacy. Cox et al. (1997) conducted a greenhouse study to determine Al-WTR effects on inorga nic forms of P, and availability to wheat ( Triticum aestivum L.) in a thermic Aquic Hapludult. Of the inorganic P fractions studied, loosely-bound (1 M NH4Cl-extractable) P was a better predictor of plant P availability in Al-WTR amended soil than M-1P. Another method which could assess plant avai lable P in WTR amended soils is the iron oxide filter paper method, sometimes referred to as iron strip P (ISP) or the "Pi soil test" (Sharpley, 1991; Sharpley, 1993a, b; Chardon et al., 1996; Pote et al., 1996; Menon et al., 1997). In principle, the Fe-oxide strip acts as an "inf inite sink" for desorbable soil P and measures the potential of a soil to continue to release P to plants. The ISP differs fr om other soil tests that chemically extract P from soils. The Fe oxide co ated filter paper strip sorbs P from solution,

PAGE 127

127 facilitating desorption of readily available P fr om soil colloids. Sharpl ey (1993b) reported ISP was a good indicator of biological availabil ity of P in runoff waters to algae. A greenhouse study conducted by OÂ’Connor et al . (2002b) suggested water extractable P (WEP) as a potential soil test method for labi le P in WTR-amended soils, where Mehlich 1 (M1P) failed. Fertilizer P requirements can differ in WTR-amended and unamended soil, so careful selection of soil testing methodology is necessary. The suitability of various soil tests is expected to depend on soil, soil reactions, and P forms, bu t could be even more complex when WTR is coapplied with different P-sources. We hypothesized th at there exist a suitable agronomic soil test method for P in Florida soils amended with va rious sources of P and WTR. Accordingly, the main objective of this study was to identify a suit able soil test method for P bioavailability in Florida soils amended with di fferent P-sources and WTR. Materials and Methods Data from the glasshouse (Chapter 3) and the field study (Chapter 4) were used to evaluate the suitability of soil test P (STP) methods as predictors of plant res ponse in Florida sands. Glasshouse Experiment The experiment involved continuous grow ing of pasture grasses (bahiagrass ( paspalum notatum Fluggae), ryegrass ( Lolium perenne L.), and second bahiagrass crop) in succession. Crops were grown in a Florida sand amended with four P-sources at tw o P-source rates, and three rates of WTR. Details of the experimental procedures are given in Chapter 3. The soil samples taken following harvest of each grass were analyzed for varying measures of plant available P (M-1P, WEP and ISP). In addition, time zero soil samples were analyzed for ammonium acetate (pH 4.8) extractable P (AA-P) to check for its possible improvements over

PAGE 128

128 other STP methods. The AA-P meth od involves extracting soil sa mples with a mixture of 0.7N NH4OAc and 0.5N CH3COOH solution at 1:5, soil:solution ratio. The mixtures were shaken for 30 minutes, centrifuged, and filtere d (0.45µm) before analysis (P age et al., 1965, Sartain, 1979). Table 5-1 summarizes the methods used for the soil analysis, and details of the analysis procedures are explained in Chapters 2 and 3. Th e plant dry matter yields were determined and samples were analyzed for P concentration (Chapt er 3). Phosphorus uptak e was calculated as the product of DM and P concentrations for each ha rvest, and yield-weighted P concentrations obtained by dividing the total P uptak e by the total dry matter weight. Table 5-1. Summary of phosphorus extraction procedures used Method Extractant Soil:Solution ratio Shaking time Reference Water extractable P (WEP) Distilled water 1:10 60 min. Sharpley and Moyer, 2000 Mehlich 1 P (M-1P) 0.05M HCl + 0.0125M H2SO4 1:4 5 min. Sims, 2000 Iron strip P (ISP) FeO paper + 0.01M CaCl2 1:40 16 h Chardon et al., 1996 Ammonium acetate P (AA-P ) 0.7M NH4OAc + 0.5M CH3COOH 1:5 30 min. Sartain, 1979; Page et al.,1965 Field Experiment Data from the field experiment (Chapter 4) were used to validate the glasshouse experiments, and provided additional data for the soil test validation effort. The dry matter yields, P concentrations, and P uptake of the test plant (established bahiagrass) were determined from the harvests of the 2003 and 2004 growing seasons. Soils sampled from the A (0-5cm) in June 2003 and in January 2004 served as measures of soil P for the 2003 and 2004 growing seasons. Additional soil samples (0-15 cm) were taken in March 2004 to better account for A horizon contribution to P supply, a nd were also related to plant parameters of the 2004 growing

PAGE 129

129 season. All soil samples were analyzed for WEP, ISP, and M-1P. Details of the experimental procedures and analysis are pr ovided in Chapters 4 and 2. Statistical Analysis Normal probability plots and residuals of th e data were studied to ensure the samples satisfied the assumptions of normality, consta nt variance, and independence. PearsonÂ’s correlation and regression analysis between soil extractable P values and plant responses, and other statistical tests, were done using SAS (SAS Institute, 1999). Graphical representations were done using Excel software. Results and Discussion Soil Test P and WTR Treatments Generally, the extractable P values (WEP, ISP and M-1P) for the same soil samples differed. The greatest values were observed as M-1P and the least as WEP. Apart from the control treatment, M-1P values at time zer o of first bahiagrass crop exceeded 15 mg kg-1, indicating sufficient soil P for plant growth in al l the treatments (Kidder et al., 2002). Thus, the growth (DM yield) response of the first bahiagra ss cropping was expected to be unaffected if the plants could assess the same soil P pool as M-1P, including some of the WTR-sorbed P. Stanley and Rhoads (2000) reported no first year response of bahiagrass to P fert ilization if the M-1P values exceed 16 mg kg-1, and 39 mg kg-1 initial STP was shown sufficient for two years growth. A better response was expected in the second bahiagrass cropping because soil M-1P values across the treatments at planting of th e grass ranged from low (< 15 mg kg-1) to very high (>60 mg kg-1).

PAGE 130

130 Water extractable P, an expected good test of the readily available P, indicated lower P in the soils amended with WTR than in soils without WTR across P-sources throughout the study. The lower soluble P in soils receiving WTR th an in those without WTR was confirmed by the low soil ISP values in WTR treatments (Fig. 5-1). Figure 5-1. Extractable P [(a)wate r extractable P (WEP), (b) iron strip P (ISP), (c) Mehlich 1 P (M-1P), and (d) ammonium acetate P (AA-P )] values in samples taken during the glasshouse study as affected by application rates of P-sources and water treatment residual (WTR). Note the different scale for M-1P Both WEP and ISP are measures of soil soluble P (readily available P) and are also regarded as good indices of environmental P hazard (Menon et al., 1997; Pote et al., 1999; Kleinman et al., 2005; Vadas et al., 2006). (a) Soil water extractable P (WEP) 0 5 10 15 20 25 30 35 40N based (0% WTR) N based (1% WTR) N based (2.5% WTR) ControlP based (0% WTR) P based (1% WTR) P based (2.5% WTR)WTR and P sources rates WEP (mg kg-1) Time Zero (June 04) After Bahiagrass 1 (Dec. 04) After Ryegrass (May. 05) After Bahiagrass 2 (Sept. 05) (b) Soil ISP0 5 10 15 20 25 30 35 40N based (0% WTR) N based (1% WTR) N based (2.5% WTR) ControlP based (0% WTR) P based (1% WTR) P based (2.5% WTR)WTR and P sources rates ISP (mg kg-1) Time Zero (June 04) After Bahiagrass 1 (Dec. 04) After Ryegrass (May. 05) After Bahiagrass 2 (Sept. 05) (c) Soil M-1P0 20 40 60 80 100 120 140 160 180N based (0% WTR) N based (1% WTR) N based (2.5% WTR) ControlP based (0% WTR) P based (1% WTR) P based (2.5% WTR)WTR and P sources rates M-1P (mg kg-1) Time Zero (June 04) After Bahiagrass 1 (Dec. 04) After Ryegrass (May. 05) After Bahiagrass 2 (Sept. 05) (d) Ammonium acetate P of soil sampled in June 20040 5 10 15 20 25 30 35 40N based (0% WTR) N based (1% WTR) N based (2.5% WTR) ControlP based (0% WTR) P based (1% WTR) P based (2.5% WTR)WTR and P sources rates AA-P (mg kg-1)

PAGE 131

131 The trends in AA-P values for samples taken in June were similar to trends in WEP and ISP values (Fig. 5-1). The greater, or similar, M-1P values for treatments with increasing amounts of WTR at each rate of the P-sources (Tab le 5-2 and Fig. 5-1d) indicate that the acidic M 1-P extractant (pH<2) releases some of the P sorbed by the WTR. The trends of pooled data (samples taken th roughout the glasshouse study) of the various STP methods are shown in Fig. 5-2. Both WEP and ISP values reflected the effects of WTR, and the trend of extractable P for both measures of P was: 0% WTR > 1% WTR > 2.5% WTR (Fig. 5-2a). Unlike soil WEP and ISP values, the effect s of WTR treatments could not be distinguished by M-1P (Figs. 5-2b and 5-2d). Figure 5-2. Relationships between water extractab le P (WEP) and (a) iron strip P (ISP), (b) Mehlich 1 P (M-1P), (c) ammonium acetate P (AA-P) and (d) between M-1P and ISP values of samples taken during the gla sshouse study and the effects of water treatment residual (WTR). The AA-P tested on time zero soils improved iden tification of the WTR treatments (Figs. 5-1c and 5-2c), but was not better than WEP or ISP. (a) ISP vs WEP for all samples taken during the glasshouse study0 10 20 30 40 50 60 01020304050 WEP (mg kg-1)ISP (mg kg-1) 0% WTR 1% WTR 2.5% WTR (b) WEP vs M-1P for all samples taken during the glasshouse study0 50 100 150 200 01020304050 WEP (mg kg-1)M-1P (mg kg-1) 0% WTR 1% WTR 2.5% WTR (c) AA-P vs WEP of time zero samples taken during the glasshouse study 0 10 20 30 40 01020304050 WEP (mg kg-1)AA-P (mg kg-1) 0% WTR 1% WTR 2.5% WTR (d) ISP vs M-1P for all samples taken during the glasshouse study0 50 100 150 200 0102030405060 ISP (mg kg-1)M-1P (mg kg-1) 0% WTR 1% WTR 2.5% WTR

PAGE 132

132 Figure 5-3. Effects of P-source ra tes and water treatment residual (WTR) on (a) water extractable P (WEP), (b) Iron strip P (ISP), and (c) Me hlich 1P (M-1P) values of A-horizon (05cm) soil samples taken during the field study. Data from the field experiment exhibited trends similar to those observed in the glasshouse. The soluble P measures, especially WE P values were greater in the absence of, than in the presence of, WTR (Fig. 5-3). Similar, or gr eater, M-1P values were observed in treatments with or without WTR for each application rate in A-horizon (0-5cm) samples taken during the (b) ISP in A-Horizon (0-5cm) 0 10 20 30 40 50 60 June '03Jan. '04Dec. '04 Sampling periodsISP (mg kg-1) a ab ns b b ab ns (c) M-1P in A-Horizon (0-5cm) 0 10 20 30 40 50 60 June '03Jan. '04Dec. '04 Sampling periodsM-1P (mg kg-1) a ab ns b a a a ab b ab ab (a) WEP in A-Horizon (0-5cm) 0 10 20 30 40 50 60 June '03Jan. '04Dec. '04 Sampling periodsWEP (mg kg-1) N based P based Control N based+WTR P based+WTR a ab c c c a a a b b a a ab b b

PAGE 133

133 study. Thus, as observed in the glasshouse study, M-1P could not dist inguish between WTR treated and untreated soils in the field. Data from both the gl asshouse and field experiments suggest WEP and ISP as good measures of soil so luble P in WTR-treated soils, and both were better than the traditiona l agronomic STP (M-1P). Soil Test P and Plant Response The suitability of an agronomic soil test me thod is judged by the degree of relationship between extractable nutrient values and plant responses. Thus, the suitability of soil test extractant could be established by the correla tion and regression relati onships between the STP measured and plant responses such as DM, P concentrations, and P uptake. Generally, low correlation coefficients of plant yields with the varying measures of soil P values were observed. Sufficient plant available P, especially in the first bahiagrass cropping probably contributed to the muted responses obs erved, as yield was not limited by the soil P in most treatments. The plant tissue concentrations also indicated sufficient P available for plant growth, even in the ryegrass (s econd) cropping. The P concentrations in all of the treatments exceeded 1.0 g kg-1 recommended for ryegrass by Hylton et al. (1965). The soil P values correlated better with plant P concentrations and P uptake than with dry matter yields (Table 53). Both WEP and ISP measures correlated better with the first bahiag rass crop responses than did M-1P values (Table 5-2). However, M-1P correlated better with the ry egrass and the second bahiagrass cropping responses, than with first bahi agrass parameters. Also, the correlation of M1P values with P concentrations and P uptak e compared favorably well with WEP and ISP values after the first bahiagrass cropping. The ammonium acetate P at time zero did not improve the correlation of WEP and ISP values with first bahiagrass cr op response, but the combined values correlated better than M-1P (Table 5-2).

PAGE 134

134 Table 5-2. Correlations between varying measures of soil test P and plant parameters in the glasshouse study Soil P Plant Dependent variable §WEP ¶ISP ††M-1P Ammonium Acetate P Dry Matter 0.413† 0.394 0.377 0.359 0.0002‡ 0.0005 0.0008 0.0016 P concentration 0.860 0.801 0.561 0.722 <.0001 <.0001 <.0001 <.0001 P uptake 0.849 0.797 0.596 0.700 Bahiagrass (First) <.0001 <.0001 <.0001 <.0001 Dry Matter 0.333 0.574 0.845 0.0035 <.0001 <.0001 P concentration 0.772 0.821 0.556 <.0001 <.0001 <.0001 P uptake 0.725 0.857 0.718 Ryegrass <.0001 <.0001 <.0001 Dry Matter 0.526 0.669 0.716 <.0001 <.0001 <.0001 P concentration 0.595 0.596 0.585 <.0001 <.0001 <.0001 P uptake 0.669 0.723 0.701 Bahiagrass (Second) <.0001 <.0001 <.0001 †Correlation coefficient (r) ‡pvalue §Water extractable P ¶Iron strip P ††Mehlich 1P

PAGE 135

135 Table 5-3. Coefficients of determination (r2) and other regression parame ters obtained by relating various soil test P valu es against plant data from the glasshouse study Plant Dependent variable Independent variable r2 CV p -values ‡WEP 0.18 17 0.0008 §ISP 0.18 17 0.0007 ††AA-P 0.21 17 0.002 †Dry matter ¶M-1P 0.13 18 0.007 WEP 0.74 30 <0.0001 ISP 0.64 35 <0.0001 AA-P 0.52 41 <0.0001 P concentration M-1P 0.32 49 <0.0001 WEP 0.72 41 <0.0001 ISP 0.64 47 <0.0001 AA-P 0.49 56 <0.0001 Bahiagrass (First) P uptake M-1P 0.36 63 <0.0001 WEP 0.14 16 0.0038 ISP 0.48 12 <0.0001 Dry matter M-1P 0.72 9 <0.0001 WEP 0.6 25 <0.0001 ISP 0.67 23 <0.0001 P concentration M-1P 0.31 33 <0.0001 WEP 0.53 35 <0.0001 ISP 0.73 26 <0.0001 Ryegrass P uptake M-1P 0.52 35 <0.0001 WEP 0.32 22 <0.0001 ISP 0.57 18 <0.0001 Dry matter M-1P 0.57 18 <0.0001 WEP 0.35 23 <0.0001 ISP 0.36 23 <0.0001 P concentration M-1P 0.34 23 <0.0001 WEP 0.45 40 <0.0001 ISP 0.52 38 <0.0001 Bahiagrass (Second) P uptake M-1P 0.49 39 <0.0001 WEP 0.38 12 <0.0001 ISP 0.50 11 <0.0001 AA-P 0.50 11 <0.0001 Dry matter M-1P 0.51 11 <0.0001 WEP 0.81 20 <0.0001 ISP 0.69 25 <0.0001 AA-P 0.51 32 <0.0001 P concentration M-1P 0.24 40 <0.0001 WEP 0.82 26 <0.0001 ISP 0.75 31 <0.0001 AA-P 0.58 40 <0.0001 Total P uptake M-1P 0.34 50 <0.0001 †Quadratic model was used for plant dry matter yi elds, while simple linear regression were used for P concentrations and P uptake ‡Water extractable P; §Iron strip P; ¶Mehlich 1P; ††Ammonium acetate P

PAGE 136

136 At initial planting, treatments effects were likely minimal, but th e proportions of soil P extracted by both WEP and ISP still related well with plant available P, whereas M-1P failed. The coefficients of determination obtained fr om regressions of plant dry matter yields, P concentrations and P uptakes with varying meas ure of STP also showed that the WEP and ISP value predictabilities were better than M-1P values, especially in first bahiagrass cropping (Table 5-3). However, with time, the regressions w ith M-1P values improved. Soil M-1P may be inadequate to assess plant response to P in soils newly treated with WTR, whereas the WEP or ISP soil tests are better. However, with time (>5 months), the predictability of plant responses by M-1P improved (Table 5-3). The glasshouse study by Cox et al. (1997) also reported M-1P as a good indicator of P availability to wheat ( Triticum aestivum L.) in WTR-treated soils. Cox et al. (1997) did not evaluate the WEP and ISP methods. Figure 5-4. Regression of total P uptake (sum for the three croppings) and average (a) water extractable P (WEP), (b) iron strip P (ISP), (c) ammonium acetate P (AA-P ), and (d) Mehlich 1 P (M-1P) values of soil samples at planting of the grasses. (b) y = 2.18x + 12.1 r2 = 0.75 0 20 40 60 80 100 120 0510152025303540 ISP (mg kg-1)P uptake (kg ha-1) CV = 31% p-value < 0.0001 (c) y = 2.80x + 7.69 R2 = 0.58 0 20 40 60 80 100 120 05101520253035 AA-P (mg kg-1)P uptake (kg ha-1) CV = 40% p -value < 0.0001 (d) y = 0.39x + 16.9 r2 = 0.34 0 20 40 60 80 100 120 0102030405060708090100110 M-1P (mg kg-1)P uptake (kg ha-1) CV = 50% p -value < 0.0001 (a) y = 3.29x + 12.9 r2 = 0.82 0 20 40 60 80 100 120 051015202530 WEP (mg kg-1)P uptake (kg ha-1) WTR Treatments 0% WTR 1% WTR 2.5% WTR CV = 26% p-value < 0.0001

PAGE 137

137 Regression of total P uptake (sum of P uptak e for the three croppings) with averages of STP values at planting of the three grasses established WEP (r2 = 0.82) as a good predictor of plant response in WTR-amended so ils (Fig. 5-4). The ISP method (r2 = 0.75) was also successful, whereas M-1P failed. The WEP and ISP values c ould be used interchangeably, as both were closely related (r2 = 0.65 – 0.93) with each other throughout the study (Table 5-4 and Fig. 5-5). Table 5-4. Coefficients of determination (r2) and other regression parameters obtained by plotting various soil test P values ag ainst each other (glasshouse study). Dependent variable Dependent variable Independent variable InterceptSlope r2 CV p -values §ISP ‡WEP 1.16 1.22 0.9326 <0.0001 †M-1P WEP 19.3 3.64 0.5755 <0.0001 M-1P ISP 13.1 3.22 0.7245 <0.0001 M-1P ¶AA-P 4.88 6.59 0.6650 <0.0001 WEP AA-P 3.23 1.41 0.7057 <0.0001 At planting of first bahiagrass crop ISP AA-P 4.20 1.89 0.7845 <0.0001 ISP WEP 2.02 1.29 0.6746 <0.0001 M-1P WEP 20.6 2.01 0.1375 <0.0001 At planting of ryegrass crop M-1P ISP 9.75 2.35 0.4759 <0.0001 ISP WEP 1.95 1.47 0.6565 <0.0001 M-1P WEP 21.2 2.42 0.1475 <0.0001 At planting of second bahiagrass crop M-1P ISP 15.8 2.04 0.3366 <0.0001 †Mehlich 1P ‡Water extractable P §Iron strip P ¶Ammonium acetate P The greater suitability of WEP and ISP as agronomic soil tests for P than M-1P was demonstrated in the greater correlation coefficien t (r) for WEP and ISP values than for M-1P in the field study (Table 5-5). Overall, the correla tion coefficients were small (r <0.7) because of the less impact of the treatments on the plant responses. The established, deep-rooted bahiagrass accessed nutrients in the lower E and Bh horizons (w hich were not affected by the treatments), in

PAGE 138

138 addition to those in the A-horizon (Ibrikci et al., 1994). This phenomenon has made identifying a suitable soil test method to predict bahiagrass response to P unsuccessful (Ibrikci et al., 1992, Rechcigl et al., 1992), and soil te sting (M 1-P) is not recommende d for bahiagrass pastures in central and south Florida (Kidder et al., 2002). ISP = 1.19WEP + 2.9 r2 = 0.72 0 10 20 30 40 50 60 01020304050 WEP (mg kg-1)ISP (mg kg-1) 0% WTR 1% WTR 2.5% WTR Figure 5-5. Regression of water extractable P (WEP) and iron st rip P (ISP) values of soils sampled at planting of the three croppings.. Irrespective of the smaller correlation coeffici ents, values of WEP and ISP correlated (P <0.05) with plant P uptake and P concentrati on throughout the study, even when M-1P did not (Table 5-5). Thus, the field da ta support the observations from the glasshouse study that both WEP and ISP could be used in WTR-amende d soil as soil tests, whereas M-1P fails. The field study also confirmed that WEP a nd ISP methods clearly separate the WTRtreated and untreated soils, but M-1P does not (Fig. 5-6).

PAGE 139

139 Table 5-5. Pearson’s correlation coefficients betw een the different measures of soil test P of Ahorizon soils (0-5cm) and plant da ta for 2003 and 2004 (field study). June 2003 soil P and 2003 plant data Jan. 2004 Soil P and 2004 plant data Plants Data §WEP ¶ISP ††M-1P WEP ISP M-1P 0.216† 0.249 0.024 0.173 0.146 0.117 Dry matter 0.1279‡ 0.0777 0.8621 0.2243 0.3063 0.4098 0.448 0.420 0.068 0.623 0.286 0.102 P uptake 0.001 0.0021 0.6339 <.0001 0.0416 0.4733 0.581 0.483 0.086 0.647 0.261 0.035 P concentration <.0001 0.0003 0.5484 <.0001 0.0636 0.8069 † Correlation coefficient (r) ‡ p -value § Water extractable P ¶Iron strip P †† Mehlich 1P Figure 5-6. Relationships between water extractable P (WEP) and iron strip P (ISP) of (a) June 2003, (b) January 2004, (c) December 2004; an d between Mehlich 1 P (M-1P) and Total P (TP) values of (d) June 2003, (e) January 2004, (f) December 2004 soil samples taken during the field study and the effects of water treatment residual (WTR) (a) June 2003 samples0 20 40 60 80 100 0102030405060 WEP (mg kg-1)ISP (mg kg-1) No WTR With WTR (d) June 2003 samples0 100 200 300 400 500 020406080100 M-1P (mg kg-1)TP (mg kg-1) No WTR With WTR (b) January 2004 samples0 20 40 60 80 100 120 0510152025 WEP (mg kg-1)ISP (mg kg-1) No WTR With WTR (e) January 2004 samples0 100 200 300 400 500 600 0102030405060 M-1P (mg kg-1)TP (mg kg-1) No WTR With WTR (c) December 2004 samples0 10 20 30 40 50 01020304050 WEP (mg kg-1)ISP (mg kg-1) No WTR With WTR (f) December 2004 samples0 50 100 150 200 250 050100150 M-1P (mg kg-1)TP (mg kg-1) No WTR With WTR

PAGE 140

140 The WEP and ISP methods better related with each other than with M-1P values (Table 56). Thus, the two soil test P (WEP and I SP) can be used interchangeably. Soil WEP determination is easier and si mpler than ISP. The good relationship found between the two STP methods indicates a good estimate of the ISP (bioav ailable P) could be achieved from the simpler WEP values in WTR-amended soil. Table 5-6. Coefficients of determination (r2) values obtained between varying soil test P measures for the two planting seasons during the field study. Soil Tests Depth SlopeIntercept r2 †M-1P vs. §ISP (All) 0-5cm and 0-15cm 0.24 28.2 0.02 M-1P vs. ‡WEP (All) 0-5cm and 0-15cm -0.05 32.9 <0.01 ISP vs. WEP (All) 0-5cm and 0-15cm 0.97 5.43 0.58 ISP vs. WEP (June 2003) 0-5cm 0.82 10.0 0.72 ISP vs. WEP (Jan 2004) 0-5cm 1.37 1.79 0.39 ISP vs. WEP (Dec 2004) 0-5cm 0.78 7.82 0.56 ISP vs. WEP (March 2004) 0-15cm 0.83 4.08 0.78 ISP vs. WEP (Dec 2004) 0-15cm 1.09 3.93 0.57 †Mehlich 1P ‡Water extractable P §Iron strip P Summary and Conclusions Strongly acidic extractants, including M-1P , are not suitable as measures of plant response to P in WTR-amended soils. Both wate r extractable P (WEP) and Iron strip P (ISP) methods applied to soils sample d at planting of bahiagrass ( paspalum notatum Fluggae ), ryegrass ( Lolium perenne L.), and a second bahiagrass crop dis tinguished the treatments into WTRtreated and untreated soils. Correl ations of the dry matter yields , P concentrations, and P uptake of the first bahiagrass crop were also better with WEP and ISP values than with M-1P values, but regression of plant responses with M-1P improve d after the first cropping. Total plant P uptake correlated better with WEP (r2 = 0.82***) and ISP (r2 = 0.75***) than with M–1P (r2 = 0.34***).

PAGE 141

141 Data from the field study al so support WEP and ISP as better STP methods than M-1P in WTR-treated soils. Both WEP and ISP are r ecommended as STP methods for Florida soils treated with WTR.

PAGE 142

142 CHAPTER 6 APPLICATION RATE OF WATER TREATMEN T RESIDUAL (WTR) FOR AGRONOMIC AND ENVIRONMENTAL BENEFITS Introduction Land application of Al water treatment residual (Al-WTRs) can serve as a best management practice (BMP) to reduce environmen tal hazards associated with excessive soil phosphorus (P) loads. The Al-WTR can increase so il P retention and, thereby, decrease offsite P loss to water bodies (O’Connor et al., 2002a; Da yton et al., 2003; Novak and Watts, 2004). However, over application of the residuals can le ad to excessive immobilization of soil P and induce plant P deficiencies. Thus, knowing the corre ct amount of WTR to land apply is critical. Determining the appropriate application rate s of WTR is complicat ed by variations in chemical properties of the residu als as influenced by the source of water, treatment chemicals and processings used by treatment plants (O’Con nor et al., 2004). The WTRs not only vary in total Al concentrations, but al so other chemical properties th at affect the sorption capacity, including other elemental concen trations (e.g., Fe and P) and me tal oxides forms (amorphous and crystalline). The WTRs used in a recent study by Makris (2004) had Al concentrations that ranged between 37 and 103 g kg-1 for Al-WTRs and between 1.5 – 9.8 g kg-1 for Fe-WTRs and varying P and Fe concentrations (Table 6-1). Twenty–one Al-WTRs used in a study by Dayton et al. (2003) also widely varied in total Al (14.7 -177 g kg-1), Fe (5.02 – 49.9 g kg-1) and P (0.20 – 4.04 g kg-1) concentrations (Table 6-1). In a batch equilibration study by Dayton et al . (2003) to examine the components of WTR that could contribute to P sorption pr operties, oxalate extractable Al (Alox) correlated with the linearized Langmuir Pmax values. The sorption capacities of various WTRs were also shown by O’Connor et al. (2002a) to depend on the oxalate ex tractable Al, Fe, and P concentrations of the

PAGE 143

143 WTRs. Haustein et al. (2000) compared the abilities of two Al rich materials to reduce runoff P from excessively P-impacted fields. Materi al with greater Al concentration (46.7 g kg-1) applied at both 9 and 18 Mg ha-1 decrease runoff P below those of control plots throughout the 4-month experimental periods. However, at the same rates (9 and 18 Mg ha-1), material with lower Al concentration (15.9 g kg-1) decrease the runoff P concentrat ions for only 1month. Pautler and Sims (2000) reported a signi ficant relationship (r = 0.61, p -value = 0.01) between P sorption and amorphous Al and Fe concentrations of soils. El liott et al. (2002b) sugge sted that the phosphorus saturation index (PSI) de termined from 0.2M oxalate extractab le P, Al and Fe concentrations (Pox, Alox and Feox, respectively) was useful for determini ng WTR application rates. The soils and the P-sources that can be co-applied with WTR can also vary in Pox, Alox and Feox. Thus, the compositional variability of soils, P-sources, and WTRs need to be accounted for in determining the amount of WTR to be applied to a soil. Table 6-1. Total and oxalate extractable phospho rus, aluminum and iron in water treatment residuals (WTR) used in some recent studies <-------------Total (g kg-1)----------> < -------------Ox alate (g kg-1) ----------> Study P Al Fe P Al Fe Makris, 2004 0.80 – 3.1 37.0 103 5.70 -20.7 0.50 – 2.98 29.0 – 91.0 2.30 – 5.80 Dayton et al., 2003 14.7 177 5.02 -49.9 0.30– 5.14 1.33 – 48.7 0.43 – 7.14 O’Connor et. al., 2005 1.91 -2.79 78.1 -145 2.97 -5.33 0.61 -3.02 73.7 -109 0.78 – 3.23 Application rates of WTR used in most stud ies are often based on arbitrary dry weight amendments:soil ratio, with little account taken of the chemical composition of the materials in arriving at the WTR rates (Peter s and Basta, 1996; Basta and St orm, 1997; Gallimore et al.,

PAGE 144

144 1999; Ippolito et al., 1999; Brown and Sartain, 2000; Haustein et al., 2000; Codling et al., 2002; Dayton et al., 2003; Novak and Watts, 2004). App lication of WTR based on dry weight (or soil:amendment ratio can result in excessive or inadequate immobiliza tion of soil soluble P depending on the amount and reactivity of Al and or Fe added in the WTRs. Based on the consensus among researchers, the STP could be ma intained at levels that optimize crop yields and still minimize the risk of offsite P tran sport (Higg et al., 2000). However, this agroenvironmental optimal need to be determined a nd its suitability as a basis for WTR application rate evaluated. Potential indices of environmental P losses are the degree of phos phorus saturation (DPS) for soils, and the phosphorus saturation index (P SI) for amendments (P-sources and WTR). Both DPS and PSI are calculated as ratios of Pox to the sum of Alox and Feox of the soil and amendment, respectively, but with -value (which depends on soil characteristics) included in the denominator for DPS calculation (van der Zee, et al., 1987; Breeuwsma and Silva, 1992; Nair et al., 2004). A recent study by Nair and Harris (2004) recommended determining the soil phosphorus storage capacity (SPSC) values rather than DPS as an index to predict th e amount of P a soil can sorb before exceeding a threshold soil equilibrium concentration. The SPSC values indicate the risk arising from P loadings as well as inherent P sorption capacity of the soil. The SPSC values range from negative values (for highly P-impacted soil) to positive values (for less P-impacted soils). Zero SPSC values represen ts the value at which the soil PSR is at the threshold value (0.15) related to a soil solu tion concentration of 0.1 mg L-1 (Nair and Harris, 2004). Application of WTR, if base d on SPSC values, could target only the excess P that poses environmental threats and is not expected to nega tively impact the P pools needed to meet plant

PAGE 145

145 P requirement. However, there is need to dete rmine the agronomic threshold SPSC value above which plant yields are negatively impacted. The SPSC-based rates of WTR were hypothesized to result in similar soluble P concentrations and pl ant yields, irrespective of soil P loads and sources of P. Further, it was hypothesized that there exist an SPSC value above which the plant yields are reduced. The objective of this study was to ev aluate the impact of SPSC-based Al-WTR application rates on plant yields and P concentrations, and to id entify the agro-environmental SPSC threshold. Materials and Methods Data from the glasshouse (Chapt er 3) and the field experiment (Chapter 4) were used for the study. The SPSC values were calculated from the soil oxalate extractable P, Fe, and Al concentrations as: SPSC = (0.15 – PSR)* (Alox + Feox) Equation (6-1) PSR = P saturation ratio = (Pox)/(Alox + Feox) Equation (6-2) Where Pox, Alox, and Feox are 0.2M oxalate extractable P, Al , and Fe concentrations of the soil expressed in mmoles, respectively. The 0.15 value used in the SPSC calculation was the threshold PSR value suggested by Nair and Harris (2004) for Florid a soils. The index, SPSC, was expr essed as equivalent mg P kg1 by multiplying the SPSC value calculated in Equation 6-1 by 31 (the atomic mass of P) as: SPSC (mg P kg-1) = (0.15 – PSR)* (Alox + Feox)*31 Equation (6-3) The relationship between the soil WEP and SPSC values was used to determine the environmental thresholds, while critical plant P concentrations were identified using Cate-Nelson method (Cate and Nelson, 1971).

PAGE 146

146 Results and Discussion Soil Phosphorus Storage Capacity (SPSC) in the Glasshouse Study The soil samples taken at time zero of first bahi agrass are expected to give a better picture of how the SPSC values are affected by the treat ments, as the samples are not affected by the uptake by previous crops. For all P-sources, at both application rates, SPSC values increased with increasing WTR rates due to the added Al (F ig. 6-1). The SPSC values were lower at the high P application rate (N-based rate) reflecting greater P:Al+Fe ratio than at the low rate (Pbased rate). The greater added P at N-based rates obviously provi ded (saturated the P sorption sites) more excess P than in P-based rates. Th e SPSC values at higher P loads (N-based rate) were negative for the four P-sources in the ab sence of WTR, which indicates P added exceeded soil P storage capacity. This establishes th e N-based rates of the P-sources as not environmentally friendly without WTR. The variation in the magnitude of the SPSC va lues at P-based rates (where equal P loads was applied) without WTR, reflects the differe nces in the P-source chemical compositions (especially Al, Fe, and P, as summarized by PSI). The differences in the SPSC values suggest soils amended with P-sources of different PSI valu es will need different amounts of Al and or Fe added as WTR to achieve equal soil SPSC values. Addition of WTR increases the P storage capac ity at either P-based or N-based rates. However, SPSC values were greater at the P-ba sed, than at the N-based, rates when an equal amount of WTR is applied for each P-source. Applying equal amounts of WTR also gave different SPSC values for different P-sources at either the Por N-based rates. Thus, varying amounts of WTR will be needed depending on the source PSI to achieve equal SPSC value for soils treated with various P-sour ces. The variations of SPSC valu es for different P-sources at

PAGE 147

147 each application rate and WTR indicate effects of the P, Al and Fe composition (which vary for different P-sources) on the soil. The variations could be accounted for by applying P-sources and WTR based on the desired SPSC value. -175 -150 -125 -100 -75 -50 -25 0 25 50 75 100Control Manure P Boca P Pompano P TSP P Manure N Boca N Pompano N TSP NTreatmentSPSC (mg kg-1) 0%WTR 1%WTR 2.5%WTR Contrasts: P-based (without WTR) vs Control P-based (1% WTR) vs Control P-based vs N-based P-based (2.5% WTR) vs P-based (without WTR) N-based (with 2.5% WTR) vs Control *** *** *** *** *** Figure 6-1. Soil phosphorus stor age capacity (SPSC, mg kg-1) values for the different treatments in time zero samples taking during glasshouse study. (Treatments ending in P, and N, are P-based and N-based rates of the sources, respectively). Values of SPSC increased with WTR addition in all samples taken during the study due to reduction in soil P:Al+Fe ratio (or PSR)(Fig. 6-2) . The SPSC values at th e P-based rate without WTR were close to 0 mg P kg-1, confirming the rate (P-based) as environmentally friendly. Zero SPSC value is equivalent to PSR value of 0.15, wh ich is the environmental threshold. However, the N-based rates without WTR resulted in negati ve SPSC values even at 1% WTR and, in some cases, at 2.5%WTR. Obviously, the SPSC values at the two rates increased with increasing WTR rates.

PAGE 148

148 The negative values of SPSC at the N-based rates of P-sources indicate that soil is receiving excess P, and agrees with other studies that N-based rates load soil with excess P that could cause negative environmental impact (Reddy et al., 1980; Pierzynski , 1994; Peterson et al., 1994; Maguire et al., 2000). However, with addi tion of WTR, the SPSC values of the N-based rates were increased and even become positive for some P-sources at 2.5% WTR (Fig. 6-1 and 62). The applied WTR obviously creates P sorptio n sites for the soil soluble P in excess of Pbased rates. -150 -100 -50 0 50 100 150 June '04Dec. '04May '05Sept. ,05 Planting periodSPSC (mg kg-1) N-based (0%WTR) N-based (1%WTR) P-based (0%WTR) Control P-based (1%WTR) N-based (2.5%WTR) P-based (2.5%WTR) bc b a aa a b b ab b abc b b bc d cd bcd cd d de e e e c d c c bc Figure 6-2. Soil phosphorus stor age capacity (SPSC, mg kg-1) values at different rates of all Psources and WTR during the glasshouse study over time. Treatments within the same sampling period with the same letters are not different at p = 0.05 by Tukey test.

PAGE 149

149 Soil Phosphorus Storage Capacity (SPSC) in the Field Experiment The SPSC values of samples taken in J une 2003, January 2004 and December 2004 from the soil A horizons during the field study are sh own in Fig. 6-3. Soils from subsurface horizons are less affected by the surface applied treatm ents, and had similar SPSC values for all treatments at the E and Bh horizons. The SPSC va lues of the E-horizon soil samples are similar, negative, and approximately zero indicating satura tion with P and inability to hold added P. The SPSC values of the Bh horizon were positive and also similar (~147 mg kg-1) for all treatments. Positive SPSC values, an indication of soil capacity to hold added P, are expected of an Al-rich Bh horizon. The positive SPSC values of Bh horiz ons agree with finding s by other researchers that noted high P retention capaci ty of the spodic horizon in spodos ols compared with surface A and E horizons (Mansell et al., 1991 ; Nair et al., 1998; Nair et al., 2004). The similarity of the values for the different treatments and at th e different sampling periods shows the spodic horizons are less affected by the surface applied tr eatments; differences in values likely reflects natural variability. The impacts of the surface applied treatments were obvious in the SPSC values of samples from A horizons (0-5cm). Soil samples from pl ots amended with P-sour ces without WTR have negative SPSC values, and SPSC values for the Nbased rates were more negative than for Pbased rates (Fig 6-3). Treatments receiving WT R had greater SPSC values than equivalent treatments without WTR. Thus, the field results confirm glasshouse results that SPSC values increased with addition of WTR a nd decreased with P added to the soil. The SPSC values of the time zero soil samples were affected by chemical pr operties of the P-sources, as observed in the glasshouse study (Fig. 6-4a).

PAGE 150

150 Figure 6-3. Soil phosphorus stor age capacity (SPSC, mg kg-1) values of A horizon (0-5cm) samples from the field in (a) June 2003 as affected by the different treatments and (b) June 2003 – Dec. 2004 as affected by P rates and WTR. (Treatments in (a) ending in P, and N, are P-based and N-based rates of the sources, respectively). Treatments in (a) or within the same sampling period in (b) with the same letters are not different at p = 0.05 by Tukey test. Soil Phosphorus Storage Capacity (SPSC) a nd Plant Growth in the Glasshouse Study The absolute values of plant yields and P c oncentrations varied with P-sources, P-source application rates and the amounts of WTR added. For most of the P-sources and at either application rate, plant yields a nd P concentrations values were greatest in the absence of WTR and least with 2.5% WTR. In most cases, the sma llest plant yields and P concentrations were observed at 2.5% WTR applied to Pbased rates for each P-source. (a) June 2003 (0-5cm) samples -200 -100 0 100 200 300 400 500 600Controls Manure-N Manure-P Boca-N Boca-P Pom pano-N Pom pano-P TSP-N TSP-PP sources and their application rates Soil SPSC (mg kg-1) Without WTR With WTR a ab abcde bcde de de de e abcd de de bcde abc de bcde de bcde (b) Samples (0-5cm) taken during the study -100 -50 0 50 100 150 200 250 300 June '03Jan. '04Dec. '04 Sampling periodSoil SPSC (mg kg-1) N-based P-based Control N-nased (with WTR) P-based (with WTR) a a a ab c c b b c bc bc b bc a ab

PAGE 151

Figure 6-4. Soil P storag e capacity (SPSC, mg kg-1) and water extractable P (WEP, mg kg-1) values of soil samples obtained during the glasshouse study in (a) June 2004, (b) May 2005, (c) September 2005, and (d) averaged over all dates. Note the difference in y-axis scales for the June 2004 sampling. (a) June 2004 sample 0 5 10 15 20 25 30 35 40 45 50 -250 -200 -150 -100 -50 0 50 100 150 200 Soil SPSC (mg kg-1)Soil WEP (mg kg-1) 0%WTR 1%WTR 2.5%WTR (b) May 2005 sample 0 5 10 15 20 25 -250 -200 -150 -100 -50 0 50 100 150 200 Soil SPSC (mg kg-1)Soil WEP (mg kg-1) 0%WTR 1%WTR 2.5%WTR (c) September 2005 sample 0 5 10 15 20 25 -250 -200 -150 -100 -50 0 50 100 150 200 Soil SPSC (mg kg-1)Soil WEP (mg kg-1) 0%WTR 1%WTR 2.5%WTR (d) Average of samples taken during the study 0 5 10 15 20 25 -250 -200 -150 -100 -50 0 50 100 150 200 Soil SPSC (mg kg-1)Soil WEP (mg kg-1) 0%WTR 1%WTR 2.5%WTR

PAGE 152

152 Soil soluble P, indicated by the WEP values, increased with decreasing (more negative) soil SPSC values in the glasshouse study. However, the rate of change in the WEP values was greater below (negativ e SPSC) than above (positive SPSC ) zero-SPSC, suggesting a change point at zero SPSC value (Fig. 6-4) . Similar trends were obtained fr om the field experiment data, which also indicated a change point at zero soil SPSC value (Fig. 6-5). Thus, application rates of the P-sources to the zero soil SPSC value is accompanied by minimal soil soluble P and could be environmentally friendly. However, below zero SPSC value (negative va lues resulting from either greater soil P or smaller Fe+Al), there could be concerns for greater P loss from the soils due to increasing soil soluble P. The zero SPSC va lue is equivalent to PSR of 0.15, suggested to be an environmental thres hold (Breeuwsma and Silva, 1992; Nair and Harris, 2004). 0 10 20 30 40 50 60 -400 -200 0 200 400 600 800 1000 Soil SPSC (mg kg-1)Soil WEP (mg kg-1) June 2003 (No WTR) June 2003 (with WTR) Jan. 2004 (No WTR) Jan. 2004 (with WTR) Dec. 2004 (No WTR) Dec. 2004 (with WTR) Figure 6-5. The soil phosphorus storage capacity (SPSC, mg kg-1) and water extractable P (WEP, mg kg-1) values of soil samples obtained from the A horizon (0-5cm) during the field study.

PAGE 153

153 Most WTR-amended soils in both the glasshous e and the field studies had greater than zero (positive) soil SPSC values. However, so me soils amended with WTR still had negative SPSC values, indicating insufficient added WTR. This is expected because of the variations in the chemical compositions (Al, Fe, and P concentra tions) in the P-sources in addition to the rate of P applications. Thus, both the glasshouse and the field studies show that the amount of WTR needed to achieve zero soil SPSC value depends on the compositions and application rates of the applied P-sources. Zero SPSC values (corresponding to PSR value of 0.15) have been suggested (Breeuwsma and Silva, 1992; Nair and Harris, 2004) as a co nservative environmental threshold, and could be recommended as agro-environmental threshold ba sed on the rationale that agronomic threshold is below environmental threshold. Hence, no negative agronomic impact is expected at environmental threshold, which is expected to be greater (3 ti mes) than agronomic threshold. The processes by which crops access soil P are diffe rent from those that determine susceptibility to solubilization by subsurface leaching or surf ace runoff (Kleinman et al., 2000). Plants can solubilize soil water-inso luble P compounds and enhance P uptake by organic acids produced in root exudates. Thus, STP can be maintain at le vels that optimize crop yields while minimizing the risk of offsite P transport (Higgs et al., 2000), and WTR application can be based on agronomic threshold. The range of P concentrations for th e first bahiagrass crop (1.5 and 6.4 g kg-1) contains the critical P concentra tion value of ~2.0 g kg-1 identified by Cate-Nelson (1977) type of approximation (Fig. 6-6). The critical (agronomic threshold) P concentration can be defined as the concentration above which th ere is no plant yield response to increased P concentration. Below a P concentration of 2.0 g kg-1, the first bahiagrass crop dry matter yield was reduced and,

PAGE 154

154 little or no response in the plant DM yield to increasing P concentrations was observed above 2.0 g kg-1 P concentration. Kincheloe et al. (1987) indicated tissue P concentrations of 2.1 to 4.0 g kg-1 are within sufficiency range for gra ss production. Hence, the 1.6 to 1.7 g kg-1 P concentrations observed in pastures (which include ba hiagrass) by Adjei et al. (2000) were considered limiting. Ryegrass P concentrations ranged between 1.8 and 6.3 g kg-1. Ryegrass dry matter yield did not respond to increasing plants P concentrati on (Fig. 6-6) because the P concentration range observed were greater than the 1 g kg-1 critical value suggested for the plant by Hylton et al. (1965). The critical value could not be clearly identified in the second bahiagrass crop (Fig. 6-6). Previous studies have indicated that a reduc tion in yield-producing capability of organic amendments P (by 20-70%) in the next season fo llowing initial fresh P-source application, and continued decline in subsequent seasons (Bolland and Gilkes , 1990). The second bahiagrass was cropped between 12 and 15 months after treatment application and in a pot which had previously and continuously cropped for 11 months. Thus, P deficiency can explain the little response of second bahiagrass crop to increasing plant P conc entration. The greater P concentrations during P deficiency periods indicate a “Steenbjerg effect”, which is increasing plant nutrient concentrations during nutrient deficiency (St eenbjerg, 1951; Bates, 1971). As earlier explained, the nutrient deficiency destroys potential for gr owth, but the plants continue to accumulate the nutrient (Ulrich and Hills, 1967; Jones, 1967; Bates, 1971) The first bahiagrass cropping, which gave plan t P concentrations range that includes 2 g kg-1 critical value, is free of the Steenbjerg effect and can be used to locate the agronomic critical SPSC value.

PAGE 155

155 Figure 6-6. Plant dry matter yields a nd soil P storage capacity (SPSC, mg kg-1) values as a function of plant P concentrations for firs t bahiagrass crop (a, d), ryegrass (b, e) and second bahiagrass crop (c, f). Among the six common statistical models availa ble to relate soil te st P (STP) to plant yields (Cate-Nelson, linear plat eau, quadratic plateau, quadratic, and exponential Mitscherlich type equations), Cate-Nelson method was sele cted as the best for guiding fertilization (a) Bahiagrass 0 2 4 6 8 10 0123456P concentration (g kg-1)Dry matter (Mg ha-1) 0% WTR 1% WTR 2.5% WTR (d) Bahiagrass -200 -150 -100 -50 0 50 100 150 0123456P concentration (g kg-1)SPSC (mg kg-1) 0% WTR 1% WTR 2.5% WTR (b) Ryegrass 0 2 4 6 8 10 0123456P concentration (g kg-1)Dry matter (Mg ha-1) 0% WTR 1% WTR 2.5% WTR (e) Ryegrass -200 -150 -100 -50 0 50 100 150 0123456P concentration (g kg-1)SPSC (mg kg-1) 0% WTR 1% WTR 2.5% WTR (c) Second Bahiagrass 0 2 4 6 8 10 0123456P concentration (g kg-1)Dry matter (Mg ha-1) 0% WTR 1% WTR 2.5% WTR (f)Second Bahiagrass -200 -150 -100 -50 0 50 100 150 200 0123456P concentration (g kg-1)SPSC (mg kg-1) 0% WTR 1% WTR 2.5% WTR

PAGE 156

156 recommendations (Mallarino and Blackmer, 1992). Th e soil SPSC values at planting of the first bahiagrass crop decreased (greater negative values ) with increasing tissue P concentrations, and the identified critical P concentrations (2 g kg-1 for first bahiagrass) was located at zero soil SPSC value by Cate-Nelson method (Fig. 6-6). The glasshouse study indicates that by applyi ng either a P-source or WTR (or both) to attain a zero SPSC value will ensu re sufficient P concentrations in the plant for growth without negative environmental impacts. Si milarly, zero SPSC value also en sured plant P concentrations of 2 g kg-1 in the field study (Fig. 6-7). An SPSC valu e of zero could serve not only as agroenvironmental threshold, but also as basis for determining the ra tes of WTR to be applied. The Psources can be applied at any ra te without negative environmenta l impact if sufficient WTR is applied to achieve an SPSC va lue of zero. Applying WTR to atta in a zero soil SPSC value will keep the soil soluble P below the change point and above the optimum plant P concentration. Application rates of WTR based on desired soil SPSC values will ensure applying the amount needed for optimum plant growth with no fear of excessive P immobilization. The SPSC values and amendment P storage capacity (APSC) values (synonymous to SPSC) can be used to determine the amount of WTR needed to be applied to a P impacted soil or co-applied with the P-sources. Th e SPSC-based rate will not only account for the P, Al, and Fe concentrations in the residuals and the soil, but the threshold soil P is also considered in the calculation. Thus, the WTR rate based on desired SPSC value will ensure a soil P level below the environmental threshold as well as suffi cient P level to meet plant needs.

PAGE 157

157 Figure 6-7. Soil P storage capacity (SPSC) values as a function of plant P concentrations in samples taken during (a) glasshouse study (T ime zero soil vs. first bahiagrass crop P concentrations) and (b) field experiment (Time zero SPSC of 0-5cm soil samples vs. yield-weighted P concentrations for the 2003 and 2004 harvests. (a) Glasshouse Experiment (First bahiagrass)-200 -150 -100 -50 0 50 100 150 0123456 P concentration (g kg-1)SPSC (mg kg-1) 0% WTR 1% WTR 2.5% WTR (b) Field Experiment-200 -100 0 100 200 300 400 500 0123456 P concentration (g kg-1)SPSC (g kg-1) With WTR Without WTR

PAGE 158

158 The amendment P storage capacity (APSC) of the WTR and the P-sources could be estimated by modifying Equation 6-1 to include PSI instead of PSR. Thus, APSC of the Psources (APSCsource) and WTR (APSCWTR) can be calculated as in Equation 6-4. APSC (mg P kg-1) = [(0.15-PSI)*(Alox + Feox)]*31 Equation 6-4 Where PSI = Phosphorus sorption index = [(Pox)/(Alox + Feox)] The amount of WTR to be added could then be determined from Equation 6-5 as: SPSCsoil* Weightsoil+ APSCsource* Weightsource+ APSCWTR* WeightWTR = 0 Equation 6-5 The SPSC value of the soil and APSC value of P-source and the WTR could be estimated from their chemical compositions. The weight of the P-sources is known from the application rate and the weight of soil could be determined from the land area to depth of impact (depending on AM; 15 cm depth if incorporated, or 5 cm wh en surface applied) and the soil bulk density. Thus, the only unknown in Equation 6-5 is the we ight of WTR, which can be determined by substituting the known values into the equation. Equation 6-5 can be used to calculate amount of WTR needed to achieve a particular soil SPSC value under any given condition. For example, to decide on the amount of WTR needed to increase a highly P-impacted soil SPSC value to 5, the equation is used without P-sources parameters (since no P-source is added) and the formula equated to five (5) instead of zero. Based on the SPSC and APSC values, the amount of WTR needed to be applied to a P impacted soil or co-applied with the P-sources can be determined. The SPSC value observed in time zero soil samples of each treatment at the th ree rate of WTR are shown on Table 6-2. Also included are the amounts of WTR needed to achieve 0 mg SPSC kg-1 at the two rates of the four P-sources. At the P-based rate, a pplying at least 1% WTR gave SPSC values greater than zero in all the P-sources, which indicated more WTR was applied than necessary. However, at N-based rates more than 1%, but less than 2.5% WTR, is needed by manure, Pompano and TSP

PAGE 159

159 treatments; and greater than 2.5% WTR is need ed by Boca Raton biosolids to achieve 0 mg SPSC kg-1. Table 6-2. Observed SPSC values (mg kg-1) of time zero soils at 0, 1, and 2.5% WTR and calculated amounts of WTR need ed to achieve 0 mg SPSC kg-1 when co-applied with the four P-sources at the two Psource rates (glasshouse study). WTR Rate P-source rate P-source 0% † 1% ‡ 2.5%¶ WTR (g) needed to achieve 0 mg SPSC kg-1 Manure -25 29 87 25 Boca Raton biosolids -26 5.5 75 90 Pompano biosolids -21 0.3 60 58 P-based TSP -12 39 96 15 Manure -78 -50 53 140 Boca Raton biosolids -158 -132 -9.4 726 Pompano biosolids -64 -15 43 294 N-based TSP -84 -2.3 23 27 †0 g of WTR applied per pot ‡115g of WTR applied per pot ¶287 g of WTR applied per pot Summary and Conclusions Applying P-sources at Pbased rates result s in soil SPSC values close to 0 mg kg-1. However, when the P-sources were applied at Nbased rates, the SPSC values were negative and the magnitude depended on Pox, Alox and Feox of the P-sources. Simila rly, co-application of equal amounts of the same WTR with different P-source s will result in different soil SPSC values, reflecting different chemical compositions (Alox and Feox) of the P-sources. Application of different WTRs at the same dry weight basis c ould result in negative agronomic and or environmental impact de pending on WTR and P-sources Al, P and Fe composition. A WTR with greater oxalate Al and Fe concentrations will result in greater SPSC

PAGE 160

160 values and, hence, lower soil soluble P con centration more than a WTR with lower Al concentrations. Application rates of WTR base d on desired soil SPSC value will ensure applying the amount needed for optimum plant growth wi th no problem of excessive plant available P immobilization. The zero soil SPSC value was identif ied as the critical point above which the plant P concentrations can be sufficiently redu ced to reduce plant yields and below which the soil soluble P and, hence, potential P loss may increase. Amendment P st orage capacity (APSC), an equivalent term of SPSC for the P-sources, needed for the calculation of WTR rate was also suggested. This study shows application rates of the P-sources and WTR to “ZERO” SPSC values will optimize both agronomic and environmental benefits of the residuals.

PAGE 161

161 CHAPTER 7 EFFECTS OF A WATER TREATMENT RESIDUAL (WTR) ON RUNOFF AND LEACHATE PHOSPHORUS LOSSES Introduction Phosphorus losses from agricultu ral land have been implicated as one of the main causes of reduced water quality in the USA (USEPA , 2000; Boesch et al., 2001). An adequate understanding of pathways for P loss from ag ricultural fields would enhance management techniques to minimize the loss. Studies of P loss have focused primarily on movement of P via the soil surface (runoff) with less attention to su bsurface (leachate) loss pathways. The focus on runoff was based on the assumption that most soils contain sufficient P sorbing oxides to maintain subsurface soil solution P concentratio ns below euthrophication thresholds (0.01 to 0.05 mg L-1, Sims et al., 1998). However, subsurface leach ing of P could be equally as important as runoff P inputs into surface waters in areas with shallow ground waters and sandy soils with little P sorbing capacity (Eghball et al., 1996; Sims et al., 1998; Novak et al., 2000; Elliott et al., 2002a). Such areas and soils are common in Florid a and other coastal plai n regions of the US. Approximately 3.4 million hectares in Florida have been mapped as Spodosols with sandy texture and poor P sorption capacities in A and E horizons (Collins, 2003). In Florida, the spodosols are characterized by high water tables located between the Bh and the A horizons during the summer rainy season, which recede to 125 cm during drier months (Soil Survey Staff, 1996). Lateral water movement of rainfall that infiltrates the soil duri ng the high water table season can transport P to surface drainage ditc hes (Burgoa et al., 1990; Mansell et al., 1991). Thus, P loss evaluations in such soils must account for both runoff and leaching. The phosphorus loss through runoff and leaching can also vary with applied P-source, as the solubility, bioavailability and transport pot ential of P varies among biosolids, manures, and

PAGE 162

162 fertilizer types (Sharpley and M oyer, 2000; Brandt et al., 2003; Leyt em et al., 2004; Elliott et al., 2005). Aside from availability of P in the P-sour ce, the amount of P applied could also influence the amount of P loss. An example is nitrogen (N)-based nutrient management of the organic source of P, which could enhance P loss. As th e rate considers only nitrogen, P is often over applied. The large soil P loads that accompa ny N-based amendment rates can cause soil P accumulation to levels above those needed for optimum crop production (Reddy et al., 1980; Pierzynski, 1994; Magu ire et al., 2000). The forms of phosphorus that promote organi sm growth, referred to as bioavailable P (BAP), include dissolved P and bioavailable particulate P (Sharpley, 1993a,b; Myers and Pierzynski, 2000). Dissolved react ive P results from desorption, di ssolution, and extraction of P from soil amendments including biosolids, manure, or recently applied P fertilizer (Daniel et al., 1998; Sauer et al., 1999; Sharpley et al., 1999). Bioavailable partic ulate P includes a portion of P bound to soil particles or to organic matter that enters surface water bodies and is subsequently made available for aquatic organisms. Studies on environmental P losses based on water soluble P, without accounting for bioavaila ble particulate P, may be inad equate as both dissolved forms and portions of colloidal P forms could promote eutrophication. Iron strip ex tractable P (ISP) has been shown as a good measure of BAP (Sharpley, 1993a, b; Myers and Pierzynski, 2000). In an incubation study with runoff as the sole s ource of P, the growth of P-starved selenastum capricornutum was strongly related (r2 = 0.96) to the runoff ISP (Sharpley, 1993a). The amount of P extracted from runoff by the Fe-oxide strips adequately estimated the BAP content of the runoff that was potentially available for uptake by freshwater organisms. Reduction of soil soluble P following applica tion of water treatment residual (WTR) has been reported (Peters and Basta, 1996; Bast a and Storm, 1997; OÂ’Connor and Elliott, 2001;

PAGE 163

163 OÂ’Connor et al., 2002a; Elliott et al., 2002b; Elliott et al., 2005; Novak and Watts, 2005). Surface application of WTR has been shown to reduc e runoff P (Peters and Basta, 1996; Basta and Storm, 1997; Dayton et al., 2003). Also, leachate P was reduced when WTR was incorporated into the soil (Elliott et al., 2002b; OÂ’Connor et al., 2002a; Novak and Watts, 2004; Dayton and Basta, 2005). Surface co-application of P-sources and WTR could be a practical way of applying these residuals to an established pasture. Howeve r, the impact of the su rface applied sources of P and the WTR on P loss through both runoff and l eaching needs to be quantified. The objective was to determine the effects of surface applied P-sources and an Al-WTR on runoff and leaching P losses in a rainfall simulation study. Materials and Methods Rainfall Simulation Experiment The same P-sources used in the glasshouse and field study (poultry manure, Boca Raton biosolids, Pompano biosolids, and TSP) were surface applied at rates equivalent to 56 kg P ha-1 and 224 kg P ha-1 to represent the low and high soil P loads typical of P-based and N-based amendments application rates, re spectively. The soils also received Al-WTR surface applied at 0 or 1% (22.4 Mg ha-1 oven dry basis) on top of the applie d P-sources. Soils without any added Psource, but with and without WTR, were included as controls. The rainfall simulation was carri ed out as prescribed in th e National Phosphorus Research Project indoor runoff box protocol (National Phosphorus Research Project, 2001). The protocol specifies the dimension of runo ff boxes (100 cm long, 20 cm wide and 7.5 cm deep). However, the box design was modified to quantify leaching of P in addition to runoff P by adding a second

PAGE 164

164 box (no soil and water tight) under the first in a double-decker design (F ig. 7-1). This design allowed collection of runoff and leachate simultaneously. Figure 7-1. National P Research project (a) runoff box design and (b) box design modified to collect runoff and leachate simultaneously. The top boxes were packed with 5 cm de pth of soil to a bulk density of 1.4 g cm-3. Treatments were surface applied as uniformly as possible (P-source first, and then WTR) a day after wetting the soil to near saturation. The treate d soils were all leveled with the lower edge of the boxes. Three boxes each (3 replicates) of th e eighteen (18) treatments were prepared, and rainfall events conducted 3, 5 and, 7 days (representing 1st, 2nd and 3rd rainfall events, respectively) after the initial wetting on each of the treatments . The rainfall intensity, 7.1 cm h-1 1 00 cm 7.5 cm 7.5 cm 7.5 cm 2 0 cm R u n o ff out l et L eac h ate out l et (a) Runoff box (b) Modified box to collect runoff and leachates 2 0 cm R u n o ff bo x (uppe r dec k ) Leachate box (lower deck) R u n o ff bo x 5 c m (dept h o f so il ) 5 c m (dept h o f so il ) Soil (Treatment)

PAGE 165

165 (equivalent to a 10-y, 24-h rain), was applied fr om a height of 3 m above the soil in boxes slanted at 3% slope. The flow ra te was measured before each si mulation rain event to ensure a flow rate of 210 mL sec-1 stated in the protocol. Also the uniformity of the rain intensity in simulation area was ensured and calibrated before each rain event. Thirty (30) minutes of runoff generated during each rainfall event were collect ed from each soil box at the down-slope end of each box and the volumes recorded. In addition, l eachate was collected continuously during each rainfall event. Representative, well mixed samples (250 mL) of runoff and leachate were collected for analysis. Another sub-sample of the runoff was filtered (0.45um) before P analysis for dissolved P determination. Leachate and Runoff Analyses Soluble reactive P (SRP) concentrations were determined on the filtered runoff (R-SRP) and the leachate (L-SRP) samples colorimetric ally with the Murphy and Riley method (1962). The Iron-strip P concentrations (a measure of BAP) in the runoff and the leachates were determined by shaking 50 mL samples (diluted with 30 mL of dei onized water) with Feimpregnated (0.65 M FeCl3 in 0.6 M HCl) filter paper. The P adsorbed was then extracted with 50mL of 0.1 M H2SO4 and analyzed colorimetric ally (van der Zee, 1987). Total phosphorus concentrations were measured on the filtered runoff (total dissolved P, TDP) and the leachate (LTP) samples after dige sting 10 mL of the samples with 0.5 mL 6N H2SO4 and 0.15g of potassium persulfate in an auto clave for 1 h (Pote and Daniel 2000a and b). Total P in unfiltered runoff samples (TRP) was determined by digesting 5 mL of the samples with 1mL of 6N H2SO4 and 0.3g of potassium persulfate on a digestion block and then diluted by adding 10 mL of water. All digested samples we re analyzed for P colorimetrically (Murphy and Riley, 1962). Particulate phosphorus (PP) concentration was calcul ated by subtracting TDP from

PAGE 166

166 the total P (TP) of each sample. Dissolved organic P (DOP) was assumed to be the difference between SRP and TDP. Leachate and runoff pH a nd EC were also determined on each sample. Flow-weighted P concentrations were calcu lated for the runoff and the leachate by summing the products of the P concentrations and vol umes for the three rain events and dividing by the total volume of the events. The runoff and leachate P losses (mg) were calculated as the product of flow-weighted concentrations (mg L-1) and the total runoff and leachate volumes (L). Masses of TP and BAP loss (mg) were determined by summing the masses of runoff and the leachate P loss of each form. Statistical Analysis Normal probability plots and residuals of th e data were studied to ensure the samples satisfied the assumptions of normality, constant variance and independence. The assumptions of normality and constant variance were violated by the leachates and runoff TP, BAP, and SRP concentrations and P losses, and log transforma tion was shown to be appropriate by Box Cox to correct the violations (SAS Institute, 1999). The da ta were log-transformed to normalize the data and stabilize the variance. Analysis of Vari ance (ANOVA) was performed on the various forms of runoff and leachate P loss (concentrations a nd masses) data (or the transformed data as applicable) using PROC GLM to determine signi ficance treatment effects (SAS Institute, 1999). The data were analyzed as a RCBD using the model: Yijkl = µ + i + j + k + ij + ik + jk + ijk + ijkl; where i is effect of ith P-source (i = manure, Boca Raton biosolids, Pompano biosolids, and TSP); j effect of jth source rate (j = P, and N-based rates); k effect of kth WTR rate (k = 0, and 1%) and ot her terms are the 2-way ( ij, ik, and jk), and 3-way ( ijk) interactions, and error ( ijkl) terms. To compare the treatments means including the control, the 18 treatments were analyzed usi ng one factor (treatment) model: Yij = µ + i + ij; where i is

PAGE 167

167 effect of ith treatment and ij is the error terms. When si gnificance was indicated by ANOVA, means multiple comparisons by Tukey test were pe rformed at 0.05 significance level using SAS. Results and Discussion Runoff and Leachate pH and EC The pH values of the leachates ranged from 5.55 for TSP (at low rate with WTR) to 7.10 for manure (with or without WTR). Runoff pH valu es followed similar trends, with the lowest value (6.32) associated with the low rate of T SP, and the greatest pH value (7.32) observed in manure applied at the high rate with WTR. The TSP had minimal impact on pH values in runoff and leachate, with pH values similar to the values for control treatments. The pH valu es were also similar for treatments with and without WTR for each P-source. The minimal imp acts of TSP and WTR on runoff and leachates pH were expected because the P-source pH values were similar to the range of pH values (pH = 5.5 – 5.9) of soil used for the study (Table 2-1) . Greater runoff and leach ate pH values were observed in manure treatments (at the high applicati on rate), owing to the large Ca concentration and alkalinity of the material. The EC values of leachate and runoff followed si milar trends as the pH (i.e., greatest in manure, high rate treatments, and least in TSP, low rate treatments) and could similarly be explained. Novak and Watts (2005) also reported minimal impacts of WTR on soil pH and EC values at rates as great as 6 % WTR. Greates t EC and pH values were observed in manure amended soils during the glasshous e study, likely as a result of poultr y feed additives, which can increase the EC of manure.

PAGE 168

168 Runoff and Leachate P Forms and Concentrations The P concentrations in runoff and leachate (flo w-weighted over the three rainfall events) are summarized in Figs. 72 and 7-3, respectively. Figure 7-2. Leachate flow-weighted mean (a) sol uble reactive P and (b) To tal P concentrations for the various treatments. (Treatments bars capped by the same letters are not different at p = 0.05 by Tukey) Soluble P dominated total P concentrations in leachate (Fig. 7-2 a) in all treatments (~85% in TSP, and > 60% in the organic sources). In the absence of WTR, the leachate flow-weighted TP concentrations were greater in TSP treatments than in the organic source treatments, similar (a) Leachate soluble reactive P (L-SRP) 0 10 20 30 40 50 60 70TSPChick ManBoca RatonPompanoControlSources of PSRP (mg L-1) High rate WTR High rate +WTR Low rate WTR Low rate + WTR a b cde cd c c ef def def ef def def f ef def f f f (b) Leachate total P (LTP) 0 10 20 30 40 50 60 70TSPChick ManBoca RatonPompanoControlSources of PLTP (mg L-1) High rate WTR High rate +WTR Low rate WTR Low rate + WTR a b def cde cd c fg efg efg efg defg fg def g g fg efg fg

PAGE 169

169 in manure and Boca Raton biosolids treatments, and greater in Boca Rat on biosolids than in Pompano biosolids treatments (F ig. 7-2 b). Similar trends were noted for the leachate flowweighted soluble reactive P concen trations (Fig. 7-2a), and refl ected the P-source solubility. The flow-weighted TP and SRP concentratio ns in leachates were reduced by surface applied WTR for both application rates of TSP, but the reductions of the TP concentrations in organic sources of P treatments by WTR were not significant (at p = 0.05). Greater TP and SRP concentrations in leachates were observed in the manure and Boca Raton biosolids treatments applied at N-based rates in the absence of WTR, but the concentr ations were not greater than those for the P-based rate of TSP. Thus, the P con centrations at high rate of organic sources of P were not greater than observed at the P-based ra te of TSP. Applying the moderate water soluble organic source of P (Pompano bios olids) resulted in similar P concentrations in leachates, irrespective of the applic ation rate. The leachate TP and SRP concentrations were small at low rates of organic sources of P a nd similar to concentrations obse rved in control treatments. In addition, the leachate TP, and SRP concentrations, at N-based rates of the organic sources of P (in the presence of WTR), were similar to cont rol treatments. Thus, hazards of greater soil P concentrations could be managed by either applying the P-sources at P-based rates, or at N-based rates in combination with WTR. Moderate water soluble P-sources could also be applied at Nbased rates, without inducing gr eater soil P concentrations th at can negatively impact the environment. Runoff soluble reactive-P concen trations (R-SRP) were similar to runoff total dissolved-P concentrations (R-TDP) for most of the treatmen ts (Fig. 7-3). The R-TDP was mostly inorganic; dissolved organic P (DOP) concentrations accounte d for less than 5% of the runoff total P (TRP) concentrations in any treatment. Other studies also indicated inorgani c P as dominant soluble

PAGE 170

170 fraction in the runoff and leachates (Heckrath et al., 1995; Turner and Haygarth, 2000). Total P concentrations (Fig. 7-3) of the runoff (TRP) we re greater than the runoff soluble P (R-SRP and R-TDP) concentrations due to th e significant contribution of particulate P (PP) to the P loss from the surface applied P-sources. Particulate P concen trations in runoff accounted for >80% of TRP concentrations for the organic sources and about 60% for TSP treatments. The proportion of TRP concentrations as particulate P of the organi c sources follows the trend Pompano biosolids (~90%) > Boca Raton biosolid s (~85%) > Manure (~80%). Figure 7-3. Runoff flow-weighted mean (a) solubl e reactive P (b) Total P (c) Total dissolved P concentrations and (d) relative P forms c oncentrations for the various treatments (Treatments followed by the same letters are not different at p = 0.05 by Tukey) (a) Runoff Soluble Reactive P (R-SRP) 0 5 10 15 20 25 30 35 40TSPChick ManBoca RatonPompanoControlSources of PSRP (mg L-1) High rate WTR High rate +WTR Low rate WTR Low rate + WTR a b fg g bc efg d efg de def efg fg fg g de cd defg efg (b) Total runoff P (TRP) 0 5 10 15 20 25 30 35 40TSPChick ManBoca RatonPompanoControlSources of PTRP (mg L-1) High rate WTR High rate +WTR Low rate WTR Low rate + WTR a a bc b bc bc bcd ef cde g g bc bcd def bcde ef fg a (c) Runoff Total Disolved P (R-TDP) 0 5 10 15 20 25 30 35 40TSPChick ManBoca RatonPompanoControlSources of PTDP (mg L-1) High rate WTR High rate +WTR Low rate WTR Low rate + WTR a b fg g efg d efg de def efg fg fg g de cd defg efg ab (d) Relative runoff P forms concentrations 0% 20% 40% 60% 80% 100%TSP(H-) TSP(H+) TSP(L-) TSP(L+) Man(H-) Man(H+) Man(L-) Man(L+) Boca(H-) Boca(H+) Boca(L-) Boca(L+) Pomp(H-) Pomp(H+) Pomp(L-) Pomp(L+) Control(-) Control(+)TreatmentsRelative P form (%) Runoff total dissolve P Dissolve organic P Runoff Particulate P

PAGE 171

171 The runoff soluble P concentrations (R-SRP and R-TDP) and their percentages of TRP concentrations in the absence of WTR tracked well with the PWEP values of the different organic source of P. The order (PWEP values in parentheses) was manure (18%) > Boca Raton biosolids (11%) > Pompano biosolids (4%). Runo ff soluble P for the TSP treatments was much less than in the organic sources despite the larg e TSP PWEP value (84%) because most of the P lost appeared in leachate. Runoff and Leachate Bioavailable P Concentrations The greater particulate P observed in th e runoff may confound the estimation of bioavailable P in the aquatic system. Sedime ntation losses of the particulate P will reduce effective bioavailability of the particulate P in the lakes compar ed to TP measured in the lab (Young and DePinto, 1982; Effler et al., 2002). Thus to account for the portion of the particulate P that is bioavailable along with the soluble P, bioavailability of runoff and leachate P (BAP) were estimated using iron strip P method. The flow-weighted BAP concentrations (Fig. 74) followed similar trends for the different treatments as the trends of flowweighted SR P and TP concentrations in the runoff and the leachate. The BAP concentrations were greater in l eachate than runoff, especially for fertilizer P. This is consistent with greater leachate P loss in dicated to be significant in Florida soils (Izuno et al., 1991). The BAP concentrations were reduced by adding WTR at both application rates of manure and Boca Raton biosolids (in runoff) and T SP (in leachate). Similar to the leachate TP and SRP concentrations in the absence of WTR, the leachate BAP concentr ations values at the N-based rates of organic source of P were sm aller than at the P-based rate of TSP.

PAGE 172

172 Figure 7-4. Flow-weighted (a) runof f and (b) leachate bioavailable P (BAP) concentrations for the various treatments (Treatments bars capped by the same letters are not different at p = 0.05 by Tukey). Both leachate and runoff BAP concentrations were affected by the source of P. The runoff BAP concentrations at P-based rates were si milar in the manure and Boca Raton biosolids treatments, but greater than BA P concentrations in Pompano biosolids treatments. At the Nbased rate, the BAP concentration trends we re: manure > Boca Raton biosolids > Pompano biosolids, which agrees with the solubility of the organic sources. Leachate BAP concentrations were greater in TSP treatments than in the orga nic source treatments due to the greater solubility (a) Runoff 0 5 10 15 20 25 30 35 40 TSPManureBoca biosolidsPompano biosolidsControl TreatmentsBAP concentration (mg L-1) High rate WTR High rate + WTR Low rate WTR Low rate + WTR a ab bc c efg efg def hi cd de efgh def efgh hi fgh hi i efgh (b) Leachate 0 5 10 15 20 25 30 35 40 TSPManureBoca biosolidsPompano biosolidsControl TreatmentsBAP concentration (mg L-1) High rate WTR High rate + WTR Low rate WTR Low rate + WTR a cd b c de def efg ghi ghi ghfgh ghi ghi hi ghi hi i i ghi

PAGE 173

173 of TSP. The trends of the leachate BAP concentr ations at both applicatio n rates of the organic sources were similar to the trends observed in runoff at P-based rates. The BAP concentrations were greater in manure and Boca Raton biosolids treatments than in the Pompano treatments. Thus, the BAP concentrations in leachate track ed well with P-source WEP values: TSP (WEP = 175 g kg-1) > Manure (WEP = 4.6 g kg-1) Boca Raton biosolids (WEP = 5.5 g kg-1) > Pompano biosolids (WEP = 1.2 g kg-1). Forms of Runoff and Leachate Phosphorus Losses To account for runoff and leachate volumes, ma ss of total P losses (runoff and leachate) were evaluated as the products of concentrations (in runoff and in leachate) and their volumes. There were greater P losses (SRP, TDP, BAP, PP and total P) in the first rain event than in subsequent events in both runoff and leachate. Other studies also documented decreasing TP and dissolved P in runoff with successive rainfall ev ents following surface applications of P-sources (Sharpley, 1997; Penn and Sims, 2002; Sims et. al ., 2003; Elliott et. al., 2005). The first rainfall event accounted for ~70% of the cumulative sol uble and TP losses (runoff plus leachate) from TSP treated soil collected over the three rain events. About 40% of the cumulative P losses (depending on the P-source) also occurred in th e first rainfall event from the organic source treated soils. Similar to the trends of weighted P concentr ations, fertilizer P loss was primarily in the leachate, but substantial amounts of soluble P also appeared in leachates from the organic source of P treatments. The masses of leachate P loss were greater in TSP treatments (147 – 746 mg) than in organic source treatments (20 – 126 mg), whereas the runoff TP losses were greater in organic source treatments (41 – 492 mg) than in TSP treatments (16 – 42 mg). The greater runoff P losses from organic sources were due to larg e amounts of particulate P, which dominated P

PAGE 174

174 losses in runoff. Particulate P loss in TSP treated soils was much less (Table 7-1). In runoff, P losses (especially BAP and other soluble reactive P forms) at th e N-based application rate of manure and Boca Raton biosolids were smaller when WTR was co-applied. Both TP and SRP losses in leachates from TSP treatments at the N-based rates were reduced in the presence of WTR. However, the particulate P and dissolv ed organic P (DOP) masses were similar for treatments with and without WTR for all the P-sources. The masses of P lost as BAP in the leachates and runoff followed similar trends as the soluble and total P losses. The masses of BAP loss in TSP treatments were greater in leachate (133 – 536 mg) than runoff (5 – 23 mg). The total masses of BAP loss were also greater in the absence of WTR (30 – 548 mg) than in the presence of WTR (17 – 464 mg). The N-based manure treatment, without WTR, resulted in the greatest masses of runoff P losses as TP, SRP TDP, PP and BAP. The runo ff TP, PP and BAP mass losses from Boca Raton biosolids treatment at the N-based rate were si milar to loses from manure applied at the same rate. However, the runoff P losses as TP, SRP, TDP, DOP and PP from Pompano biosolids (applied at the N-based rate) were less th an from the Boca Raton biosolids and manure treatments and similar to the P losses from cont rol treatment (Table 7-1). Thus, the expected greater P hazard associated with N-based rates ma y not be true for all organic sources of P. Applying a moderate water so luble P-source like Pompano biosolids may pose minimal environmental threat (with respect to P in the runoff), even at a N-based rate. The absolute values of the BAP losses were generally greater than the SRP losses in most cases, especially in the runoff where a significan t portion of particulate P was bioavailable. The proportion of total P mass losses that is bioavailable was greater in TSP than in organic source of P treated soils, and greater in the absence than in the presence of WTR (Fig. 7-5).

PAGE 175

175 Table 7-1. Masses of P forms lost in runoff and leachates. <-------------------Runoff P (mg)--------------------> <-------Leachate P (mg)-----> P-source Rate (kg ha-1) Mass of P applied (mg) WTR (%) Soil P load mass‡‡ (mg) TP† SRP‡ TDP§ DOP¶ PP# BAP†† TP† SRP‡ PP# BAP†† Total P loss §§Total BAP loss Percentage of applied P lost as BAP (%) TSP 224 1680 0 2048 27.3cd 11.8cdefg 13.0cdef 1.20b 14.4c 11.9efg 746a 706a 40.2ab 536a 774 a 548a 32.6ab TSP 56 420 0 788 41.7cd 15.9cde 17.6cde 1.64b 24.1 c 23.1cde 147cd 134 c 13.8b 133b 189d 156bc 37.1a Manure 224 1680 0 2048 492a 96.0a 104a 7.94a 388a 146a 66.3ef 63.1de 3.18b 43.3de 558bc 190b 11.3def Manure 56 420 0 788 91.5bcd 18.7cd 20.0cd 1.38b 71.5cd 48.4bc 83.8def 65.5de 18.2b 64.1cd 175d 113c 26.9bc Boca 224 1680 0 2048 391a 45.3b 47.4b 2.14b 343a 88.6ab 126cde 114cd 12.1b 97.0bc 516c 186b 11.1def Boca 56 420 0 788 65.9bcd 7.96defg 8.52def 0.56b 57.3bc 22.1cde 35.7f 30.8e 4.92b 31.2efg 193de 53.3d 12.7de Pompano 224 1680 0 2048 111bcd 5.98efg 6.82def 0.84b 104bc 23.5cde 40.1f 34.6e 5.45b 39.3de 151de 62.8d 3.74efgh Pompano 56 420 0 788 77.7bcd 2.24fg 2.65f 0.41b 75.0bc 11.3efg 46.5ef 34.3e 12.2b 33.1ef 124de 44.3def 10.5defg Control 0 0 0 368 8.97d 0.99fg 1.16f 0.17b 7.80c 3.30h 29.2f 25.1e 4.02b 26.1efgh 38.1e 29.5fg TSP 224 2370 1 2738 33.8cd 12.9cdef 13.4cdef 0.50b 20.5c 15.4ef 620b 524b 95.2a 448a 653ab 464 a 19.6cd TSP 56 1110 1 1478 15.6d 4.42efg 5.20ef 0.74b 10.4c 5.20gh 171c 162c 8.18b 140b 186d 146bc 13.2de Manure 224 2370 1 2738 453a 47.2a 58.2b 11.0a 395a 90.3ab 51.4ef 38.8e 12.5b 23.6fghi 505c 114c 4.81efgh Manure 56 1110 1 1478 48.9bcd 5.80efg 6.06def 0.27b 42.8bc 12.7def 51.1ef 30.8e 20.4b 34.3ef 100de 47.0de 4.23efgh Boca 224 2370 1 2738 157b 22.6c 24.3c 1.74b 132b 35.2bcd 35.7f 30.8e 4.92b 28.9efg 192d 64.2d 2.71fgh Boca 56 1110 1 1478 79.6bcd 6.69defg 7.06def 0.36b 72.6bc 15.1de 35.7f 23.9e 11.9b 30.3efg 115de 45.4def 4.09efgh Pompano 224 2370 1 2738 137bc 4.16efg 4.94ef 0.78b 132b 15.5de 22.9f 21.1e 1.80b 15.9hi 160de 31.5efg 1.33gh Pompano 56 1110 1 1478 41.4cd 1.38fg 3.17f 1.79b 38.2bc 4.67fgh 23.0f 21.0e 2.08b 19.1ghi 64.4de 23.8gh 2.14fgh Control 0 690 1 1058 9.10d 0.38g 1.24f 0.86b 7.90c 0.91i 20.2f 18.5e 1.66b 16.4i 29.3e 17.3h 2.51fgh †Total P ‡ Soluble reactive P §Total dissolved P ¶Dissolved organic P #Particulate P ††Bioavailable P (ISP) ‡‡Calculated as P added (source) + soil P (control) + WTR P (when applied) §§Total BAP loss is sum of runoff and leachate BAP mass loss) Phosphorus loss from treatments followed by the same le tters (in the same column) are not different at p = 0.05 by Tukey test.

PAGE 176

176 0% 20% 40% 60% 80% 100%TSPHTSPH+ TSPLTSPL+ ManHManH+ ManLManL+ BocaHBocaH+ BocaLBocaL+ PompHPompH+ PompLPompL+ CNCN+TreatmentsForms of P loss (%) Other forms of P loss BAP loss Figure 7-5. Proportions of total P loss as bioavailable P (BAP) from each treatment Effect of P-Sources, P-Source Rates and WTR on BAP Losses The BAP mass loss through runoff, leachate, and the Total BAP loss were all shown by ANOVA to be affected by the source of P, P-s ource rates and WTR (Table 7-2). The impacts of P-sources on runoff and leachate BAP loss were obs erved at Nand P-base d rates. Greater BAP loss was observed at the two rates in the TSP tr eatments than in the organic source of P, reflecting the greater BAP concen trations of the high solubleP mineral source than in the organic source of P (Table 7-3). However, the runoff BAP mass loss was similar in the TSP to the P loss in organic source of P at the P-based ra te and even greater in the organic source of P than in the TSP at the N-based rate. The greater solubility of Boca Raton biosolids was reflected

PAGE 177

177 in its greater runoff and leachate BAP loss than observed in Pompano biosolids except in leachate at P-based rate. The gr eater solubility of manure than biosolids treatments is also consistent with the greater BAP mass loss observe d in poultry manure treatment than in biosolids treatments especially at N-based in runoff and P-based in leachate. Table 7-2. ANOVA table of the effect of P-source, P-source rates, and WTR on runoff bioavailable P (BAP), leachate BAP, and Total BAP mass losses. Source of variation Leachate-P loss Runoff-P lossTotal P loss P-Source *** *** *** P-source rate *** *** *** WTR *** ** * P-Source * P-source rate *** *** *** P-Source *WTR * NS *** P-source rate *WTR ** NS NS P-Source * P-source rate *WTR NS NS * *** indicates significance at p <0.001 ** indicates significance at p <0.01 * indicates significance at p <0.05 Application rate also affect ed the BAP losses. Greater BA P mass losses were observed at N-based than at P-based rates in the runoff from organic source of P treatments, and in leachate from TSP treatment (Table 7-3). Runoff BAP losses were also reduced by WTR (Fig. 7-6). The runoff loss was greater in absence than in presence of WTR (Fig. 7-6). The impact of WTR was also observed in the leachate BAP mass loss. The WTR reduced leachate BAP loss in organic source of P treatments,

PAGE 178

178 but not in the TSP treatments (Table 7-4) and al so at N-based rates but not at P-based rates (Table 7-4). Table 7-3. Effects of P-source and P-source ra tes on runoff and leachates BAP mass loss (mg). P loss P-source N-based rateP-based rate Contrasts Nvs. P-Based rate Poultry manure 119 30.6 ** Boca Raton biosolids 61.9 18.6 ** Pompano biosolids 19.5 7.98 * TSP 13.7 14.1 NS Contrasts Organic vs. mineral source *** NS Manure vs. Biosolids *** NS Runoff Boca Raton vs. Pompano biosolids ** * Poultry manure 33.5 49.2 NS Boca Raton biosolids 62.9 30.7 NS Pompano biosolids 27.6 26.1 NS TSP 492 137 *** Contrasts Organic vs. mineral source *** *** Manure vs. Biosolids NS * Leachate Boca Raton vs. Pompano biosolids * NS *** indicates significance at p <0.001 by contrast ** indicates significance at p <0.01 by contrast * indicates significance at p <0.05 by contrast NS indicates nonsignificance at p <0.05 by contrast

PAGE 179

179 Runoff BAP loss (mg)0 10 20 30 40 50 Without WTRWith WTR WTR treatmentsBAP loss (mg) a b Figure 7-6. Effect of WTR on r unoff BAP mass loss (Treatments w ith the same letters are not different at p = 0.05 by Tukey). Most of the BAP loss from TSP was in leac hate and unaffected by WTR addition on top of applied TSP. Mixing of the WTR with TSP and/ or incorporation of both materials with soil would likely enhance WTR contact with, and retention of, soluble P and reduce the P loss (Elliott et al., 2002b; Novak and Watts, 2004 ; Dayton and Basta, 2005; S ilveira et al., 2006). Others studies have shown WTR to reduce P losses in surface runoff (Gallimore et al., 1999; Basta and Storm 1997; Hausetein et al., 2000, Dayton et al., 2003) and in leachate when WTR and Psources were co-incorporated with soil (Elliott et al., 2002; Novak and Watts, 2004). This study also shows that WTR, surface applied with P-so urces, reduces BAP in both runoff and leachate.

PAGE 180

180 Table 7-4. Effect of WTR and P-sour ces on leachate BAP mass loss (mg). P-source 0% WTR 1% WTR Contrasts 0% vs. 1% WTR Poultry manure 53.7 28.9 ** Boca Raton biosolids 64.1 29.7 ** Pompano biosolids 36.2 17.5 *** TSP 334 295 NS Contrasts Organic vs. mineral source *** *** Manure vs. Biosolids NS NS Boca Raton vs. Pompano biosolids * ** *** indicates significance at p <0.001 by contrast ** indicates significance at p <0.01 by contrast * indicates significance at p <0.05 by contrast NS indicates nonsignificance at p <0.05 by contrast Greater leachate BAP mass loss wa s also observed in the absen ce than in the presence of WTR at N-based rates and, thus, establishes the effectiveness of surface applied WTR at reducing leachate BAP loss (Table 7-5). The gr eater BAP losses observe d at N-based than Pbased rates in the absence of WTR was also elim inated when WTR is applied (Table 7-5). This supports the earlier observation that WTR additi on can reduce and even eliminate the effect of excess P hazard associated with high P loads. Analysis of variance showed that log transf ormed total masses of BAP loss were affected by P-sources, which was involved in a 3-way inte raction with WTR and the application rates. The effects of all the treatments (including control) on Total mass of BAP loss was compared (Fig. 7-7). Total BAP loss (runoff + leachate) was greater at N-based, than at P-based, rates for all P-sources, except in Pompano biosolids trea tment (Fig. 7-7). The minimal Total BAP loss in the moderate soluble-P Pompano biosolids could obs cure the impact of application rate in the treatment.

PAGE 181

181 Table 7-5. Effect of WTR and P-sources rate on leachate BAP mass loss (mg). P-source rate 0% WTR 1% WTR Contrasts 0% vs. 1% WTR N-based 179 129 ** P-based 65.4 56.0 NS Contrasts N-based vs. P-based ** NS *** indicates significance at p <0.001 by contrast ** indicates significance at p <0.01 by contrast * indicates significance at p <0.05 by contrast NS indicates nonsignificance at p <0.05 by contrast Also, P-sources affected the Total BAP mass loss, a nd the trends of absolute values of total BAP losses at each of the application rates suggested greater BAP losses from P-sources with greater PWEP values. The general order of total ma ss of BAP loss at the two rates (PWEP in parentheses) was: TSP (84%) > manure (18% ) > Boca Raton biosolids (12%) > Pompano biosolids (4%). Thus, there are differences in the environmen tal hazards among the P-sources, and the tendency is for sources with low PW EP values to lose less BAP than high PWEP materials. Total BAP mass losses were also smaller in the presence than in the absence of WTR for all organic sources of P at high rates, in manure and Pompano at low rate, and also in control treatments (Fig. 7-8). In the ab sence of WTR, total BAP losses from each of the organic sources applied at N-based rates were not greater than the P loss from TSP applied at a P-based rate. Thus, the hazards of excess P from applying the organic source of P at N-based rates is not greater than observed at P-based rates of mineral fertilizer. Als o, total BAP loss at the high rate of Pompano biosolids was similar to BAP losses at the low rates of Boca Raton biosolids and manure treatments. This indicates hazards of ex cess P from applying the organic source of P at

PAGE 182

182 N-based rates could be reduced to that observe d at their P-based rate s by applying moderate water soluble P-sources such as Pompano bioso lids. Total BAP losses were further reduced by WTR application especially in the organic sources. 0 100 200 300 400 500 600 700 800TSPH* TSPL ManH ManL BocaH BocaL PompH PompL CNTreatmentsP loss (mg) Without WTR With WTR bc b b c d defd def gh h a a bc c de d efg fg Figure 7-7. Total (runoff +leachate) mass of bioa vailable P (BAP) lost during the three rain events. *Treatments ending with ‘H’ and ‘L’ repr esents high and low application rate of the P-sources, respectively. (Treatments w ith the same letters are not different at p = 0.05 by Tukey test) Thus, this study indicates P loss in Florida sa nds amended at high rates of organic sources of P is smaller than at P-based rates of fertilizer P. Also the h azards of excess P at N-based rates could be managed to that observed at P-based ra tes by either applying m oderate water soluble Psources at N-based rate or co-applyin g the sources at high rates with WTR. In the absence of WTR, ~35% of P applied was lost as BAP in TSP treatments and 12.7% in organic sources of P treatments (Table 7-1) . In the presence of WTR, the BAP loss as a

PAGE 183

183 percentage of P applied was reduced to ~15% (T SP) and ~3% (organic sources of P). Masses of BAP lost in the presence of WTR is less than an averaged ~75% of BAP losses in the absence of WTR (Fig. 7-8). y = 0.74x 7.49 r2 = 0.82 0 100 200 300 400 500 600 0100200300400500600700800900 BAP loss without WTR (mg)BAP loss With WTR (mg) Figure 7-8. Total bioavailable P (BAP ) loss with, versus without, WTR Summary and Conclusions Previous studies demonstrated the ability of WTR to reduce P loss in runoff. This study showed that both runoff and leachate P losses fr om surface applied P-sources can be reduced by surface applied WTR. The measured runoff P losse s agree with previous studies that showed significant runoff P loss in the first rain event. Sim ilarly, the first rainfall event resulted in greater leaching P losses than subsequent events. Leachate from the first rain event accounted for more than 70% of total P loss from TSP treatment and ~40% (depending on the P-source) from

PAGE 184

184 organic sources treatments during th e three rainfall events. The BAP concentrations in leachates tracked well with WEP va lues: TSP (WEP = 175 g kg-1) > Manure (WEP = 4.6 g kg-1) Boca Raton biosolids (WEP = 5.5 g kg-1) > Pompano biosolids (WEP = 1.2 g kg-1). The masses of TP losses from TSP treatments we re greater in leachate (147 – 746 mg) than in runoff (16 – 42 mg) from the sandy soil. Ho wever, runoff TP losses were greater in the organic sources (41 492 mg) than in the TSP ( 16 – 42 mg) treatments due to greater particulate P (which dominated the runoff P losses) in orga nic source than the TSP treatments. The masses of BAP losses in TSP treatments were substantia lly greater in leachate (133 – 536 mg) than in runoff (5 – 23 mg). In addition to the reported ability of WTR to reduce P in surface runoff or in leachate when incorporated with the soil, the study shows that surface applied WTR reduced P losses in leachates as well as in runoff. The masses of BA P and TP loss were similarly affected by the Psources and followed the same trend as PWEP of the sources: TSP > Manure > Boca Raton biosolids > Pompano biosolids. The trends of the TP losses agree with the flow-weighted BAP concentrations and show that the P hazards associated with applying organic sources of P at N-based rates are smaller than for fertilizer P applied at Pbased rates. The study suggests that environmental P hazards associated with high applicati on rates (N-based) of P materi als can be managed by either applying the sources with WTR, or by using a moderate water soluble P-sources (e.g., Pompano biosolids). The total BAP mass losses at the N-based rate of TSP were greater than at the P-based rate. Most of the loss was in leachate and was unaffected by WTR placed on top of applied TSP. Mixing of the WTR with TSP and/ or incorporation of both mate rials with soil would likely enhance WTR retention of soluble P and reduce the BAP loss.

PAGE 185

185 CHAPTER 8 A METHODOLOGY TO ACCOUNT FOR P R ELEASE POTENTIAL FROM DIFFERENT SOURCES OF P: FLORIDA P INDEX AS A CASE STUDY Introduction Concerns over impacts of agricultural wate rsheds on water quality degradation resulting from accelerated eutrophication have elicited vari ous initiatives. The national initiatives include the Unified Strategy for Animal Feeding Oper ation issued by USDA and USEPA (USEPA, 1999), which requires each state NRCS to address P in nutrient management practice standards (Code 590). The three strategies outlined in code 590 for managing P in agricultural operations are: (1) soil test crop response strategy, (2) environmental soil P threshold strategy, and (3) P Index. Most states, including Flor ida, opted for the P Indexing tool (P Index), which considers multiple landscape and management factors demonstr ated to affect P loss to water bodies. The P Index addresses both P-source and tr ansport factors, as P loss re quires coexistence of labile Psource and viable transport pathway. The original P Index containe d five ‘source’ and four ‘tra nsport’ factors, designed to identify vulnerable sites where P loss reduc tion should be focused (Lemunyon and Gilbert, 1993). The draft Florida P Index similarly describes the relative risk of P movement from a given field using nine variables that are known to govern P losses (Graetz et al., 2004). One of the nine variables identified to affect P losses is the P-s ource, which is assigned a weighting coefficient to distinguish P-source lability, ba sed on professional judgments of the scientists developing the approach. The solubility, bioavailability, and transpor t potential of P vari es among biosolids, manures, and fertilizer types (Brandt et al., 2004; Leytem et al., 2004; Ellio tt et al., 2005). This fact is not well appreciated in most state P indi ces being developed because variations in P losses

PAGE 186

186 due to P-sources are not well accounted for. The draft Florida P Index currently uses a single coefficient (0.05) for both P-fertilizers and all kind s of manures, and a single value (0.015) for all kinds of biosolids (Graetz et al., 2004). Assigning a lower coefficien t to biosolids than fertilizer and manures is based on evidence that the Fe and Al contents of biosolids affect P solubility in biosolids-amended soils (Elliott et al., 2002a). However, fundamental differences in the behavior of P-sources warrant additional differentiation of P-sources. Biosolids, for example, vary in Fe and Al concentrations and, hence, P solubil ity depending on method of production (OÂ’Connor et al., 2004). Biosolids produced via biological P removal (BPR) process can mimic fertilizer P with regards to P lability. Hence, P loss is expect ed to be greater in BPR biosolids amended soils than in soils amended with biosolids with high Fe and Al concentrations, or in BPR biosolids supplemented with Al. Biosolids vary widely in susceptibility to P solubilization by water (Brandt et al., 2004), and loss in surface runoff a nd subsurface drainage (Penn and Sims, 2002; Elliott et al., 2002a; Sims et al., 2003; Elliott et al., 2005). Similarly, manures vary in P form and solubility depending on animal source, animal di ets, storage and handlin g practices (Barnett, 1994; Leinweber, 1996; Kleinman et al., 2005; Wolf et al., 2005; Vadas and Kleinman, 2006). Sharpley and Moyer (2000) showed that P forms and P release to leachates vary widely with different manure sources. The wide variability in P-s ource solubility has resulted in the suggestion of continuous, rather than discrete, coefficients to account for P availability of the P-sources (Elliott et al., 2006). Elliott et al. (2006) recently proposed a more refined algorithm for the estimation of Psource coefficients (PSC) based on correlations of runoff dissolved P and WEP values (of multiple applied manures and biosolids) genera ted from seven published rainfall simulation studies. However, the PSC values developed excl usively on runoff P loss data may be inadequate

PAGE 187

187 in Florida and other coastal plain soils where leac hing losses of P can be significant. Thus, it was hypothesized that there exist better coefficients to account for P-source potential to P loss in Florida sands with significant leachate P loss. The objective of this study was to determine a methodology that could measure the impacts of P-sour ces on P losses in Florida soils in a rainfall simulation study. Materials and Methods Data from a rainfall simulation experiment carried out as prescribed in the National Phosphorus Research Project indoor runoff box protoc ol, but with leaching and runoff P quantified was used for the study. The details of th e rainfall simulation procedures and analysis of runoff and leachates are give n in Chapter 7. Only data for treatments without WTR (four Psources at two rates and a control) were used in this study. Thus, the experiment could be described as a 4 by 2 factorial experime nt in randomized complete block design. The rainfall simulation results were compared with data from a glasshouse column leaching study that use two Florida soils (Ell iott et al., 2002a) to validate the findings. The glasshouse study involved 126 columns packed with tr eated 15 cm of A-horizons of either the moderate P-sorbing Candler soil (hyperthermic, uncoated Typic Quartzipsamments) or the low P-sorbing Immokalee series, overly ing 28 cm of E-horizon of the Myakka series. Each of the top soils was treated with ten P-sources (includi ng 8 biosolids, poultry manure, and TSP), and planted with bahiagrass. Columns were leached afte r each of four grass harvests, and total P in the leachates over the 4 month gr owing season were determined. Th e water extractable P (WEP), total P (TP), and other properties of the P-s ources were also determined. Details of the experiment are given in Elliott et al. (2002a). Th e column study data are suitable to validate the result from rainfall simulation study because more Psources (ten) and soils (2 ) were used than in

PAGE 188

188 the rainfall simulation study. The leaching study, also in a different way account for the total P loss (as leachate without runoff), while the modified rainfall simulation study accounted for total P loss as sum of runoff and leachates P losses. Statistical Analysis Normal probability plots and residuals of th e data were examined to ensure the data satisfied the assumptions of normality, cons tant variance and independence. Where the assumptions were violated, appropriate tr ansformations were applied using Box Cox transformation (SAS Institute, 1999) to normalize the runoff and leachate P concentrations and P loss data, and stabilize the va riance. Analysis of Variance (ANOVA) was performed on the various forms of runoff and leachate P loss (concen trations and masses) da ta (or the transformed data) using PROC GLM to determ ine significance treatment effect s (SAS Institute, 1999). When significance was indicated by ANOVA , the Tukey method was used to separate the means at 0.05. Simple linear regressions (ordinary least sq uare) were used to model the relationship between TP loss and P-source rates (or P-source adjusted rates) using PROC REG in SAS (SAS Institute, 1999). Correlations of the P-source coe fficients with the P losses were obtained using PROC CORR in SAS (SAS Institute, 1999). Results and Discussion Runoff and Leachate P as Affected by P-Sources Masses of TP loss in runoff and leachate and their sum are shown in Table 8-1. The masses of total P loss in the runoff are similar for the organic source of P treatments at each application rate, but greater than runoff P loss in TSP treatments at the two rates. However, the greater

PAGE 189

189 leachate P losses in TSP treatments (Table 8-1) resulted in greater total (runoff plus leachate) masses of P loss (TP). Table 8-1. Mean masses (n = 3) of P lost in runoff and leachates during rainfall simulation experiment Runoff P loss Leachate P loss †Total P loss ‡ Total BAP loss P-source P-source Rate (kg ha-1) <---------------mg-----------------------> Percentage of applied P lost as BAP (%) 224 27.3 c 746 a 774 a 548 a 32.6 a TSP 56 41.7 c 147 b 189 bc 156 b 37.1a 224 492 a 66.3 bcd 558 a 190 b 11.3 b Poultry manure 56 91.5 abc 83.8 bcd 175 c 113 bc 26.9 a 224 391 ab 126 bc 516 ab 186 b 11.1 b Boca Raton biosolids 56 65.9 bc 35.7 cd 193 c 53.3 cd 12.7 b 224 111 abc 40.1 cd 151 c 62.8 cd 3.74 b Pompano biosolids 56 77.7 abc 46.5 d 124 c 44.3 d 10.5 b †Total P loss is sum of runoff and leachate P mass loss. ‡Total bioavailable P (BAP) loss is the sum of runoff and leachate BAP masses lost. Phosphorus loss from treatments followed by the same letters (in the same column) are not different at p = 0.05 by Tukey (statistical analys is based on log-transformed data) The TP losses in manure and Boca Raton biosol ids treatments at the high P rate were similar to the losses in TSP treatments, but greater than losses at low P ra tes of all P-sources and than losses from both rates of the Pompano biosolids treatments. Thus, the TP loss from moderate water soluble P Pompa no biosolids (even at the N-base d application rate) was similar to losses for other sources applied at P-based rates. Similar (or smaller) BAP loss occurred in Po mpano biosolids treatme nts at high P loads than at low P loads of the other P-sources (Fig. 8-1). At the high P rate, greater masses of total

PAGE 190

190 BAP loss (leaching plus runoff) were observed in TSP treatments than in manure and Boca Raton biosolids treatments. This is expected of the high soluble inorganic source of P. The organic sources generally contain lower concentr ations of TP and soluble P compared to the mineral P-source. Similar to TP loss data, BAP losses from bot h manure and Boca Raton biosolids treatments were greater than in Pompano biosolids treatmen t. Thus, contrary to the similar coefficients assigned in the draft Florida P Index to fer tilizer and manure, the BAP losses from TSP treatments were greater than from manure treatme nts. Also, greater BAP losses were observed in the high water soluble P Boca Raton biosolids treatment than in moderate water soluble P Pompano biosolids treatment. In addition, contrary to the different assigned coefficients in the draft Florida P Index, similar P losses were observed in manure and Boca Raton biosolids treatment. Thus P losses from some biosolids, especially high water soluble P biosolids, were comparable to the greater P losses expected from manures treatments. Total BAP losses of the 8 treatments (4 P-sour ces, each at 2 rates) varied as: TSP (at high rate) > manure and Boca Rat on biosolids (at high rate) TSP (at low rate) Manure (at low rate) > Boca Raton biosolids (at low rates) Pompano biosolids (at high rate) Pompano biosolids (at low rate) control (Fig. 8-1c). The P hazard at the N-based rate of the organic sources was about the same as the P-based rate of TSP, and the P hazard of the moderate soluble P biosolids applied at the N-based rate was no mo re than the P-based rate of other P-sources.

PAGE 191

191 Figure 8-1. Bioavailable P (BAP) lost in (a) runof f, (b) leachate, and (c) total (runoff + leachate) as affected by the P-sources at the two a pplication rates. (Treatments followed by the same letters are not different at p = 0.05 by Tukey test; †Treatments with labels ending in “H” and “L” indicate high and low application rates of the P-source, respectively) (a) Runoff BAP loss0 20 40 60 80 100 120 140 160† T S P H T S P L M a n u r e H M a n u r e L B o c a H B o c a L P o m p a n o H P o m p a n o L C o n t r o lTreatmentsP loss (mg) d cd ab cd cd bc a d e (b) Leachate BAP loss0 100 200 300 400 500 600† T S P H T S P L M a n u r e H M a n u r e L B o c a H B o c a L P o m p a n o H P o m p a n o L C o n t r o lTreatmentsP loss (mg) a b de cd e de e bc e (c) Total BAP loss0 100 200 300 400 500 600† T S P H T S P L M a n u r e H M a n u r e L B o c a H B o c a L P o m p a n o H P o m p a n o L C o n t r o lTreatmentsP loss (mg) a bc b c d b e de d

PAGE 192

192 Relative P Losses and P-source Coefficients The regressions of total mass of BAP loss as a function of app lication rate for each of the four P-sources are shown in Table 8-2. Loading ra te accounted for at least 70% of the variability (r2 > 0.7, CV < 60%) in mass of BAP loss from any of the P-sources. However, the coefficients of determination (r2) and the slopes varied with the different organic sour ces of P, indicating that equal P loadings result in varying P loss for diffe rent P-sources. Generally, the total mass of BAP loss increased with P load for all sources (positive slope). Table 8-2. Regressions of total bioavailable P (BAP) loss with P applied for each P-source (at zero intercept). P-source Slope r2 CV (%) p -value TSP 2.47 0.90 48 <0.0001 Poultry manure 0.92 0.90 40 <0.0001 Boca Raton biosolids 0.84 0.92 40 <0.0001 Pompano biosolids 0.31 0.74 57 0.0014 The rate of increase in BAP loss per unit increase in P lo ad was ~2.5 for TSP, ~0.9 for manure, 0.8 for Boca Raton biosolids, and 0.3 for Pompano biosolids. The BAP loss from soils treated with Boca Raton biosolids was about three times greater th an the loss from soils treated with Pompano biosolids for each unit increase in applied P. The BAP loss from soil amended with Pompano biosolids at the high P rate (224 kg P ha-1) was similar to the BAP loss from Boca Raton biosolids amended soil at the low P rate (56 kg P ha-1). The P loss from soil amended with TSP was ~2.5 times greater than the P loss from manure. The results are not consistent with the draft Florida P Index that assigns the same 0.05 coefficient for fertilizer and manures and 0.015

PAGE 193

193 for all biosolids. However, the results agree wi th the assumed greater lability of manure-P than biosolids-P in the draft Florida P Index. The BAP losses from manure treatments were greater than from Boca Raton and Pompano biosolids treatments at either application rate. Parameters that could account for variable P losses include PWEP , WEP, WEP-based PSC proposed by Elliott et al. (2006), and the P-source coefficients in the draft Florida P Index (FPSC). The ranking of total (runoff + leachate) masses of TP and BAP losses at the two application rates agree with the ranking of the PWEP values of the sources, (PWEP values in parentheses): TSP (84%) > manure (18%) > Boca Raton biosolids (12 %) > Pompano biosolids (4%) (Table 8-3). The WEP value of manure (4.57 mg kg-1) was less than the value for Boca Raton biosolids (5.52 mg kg-1), but P losses (BAP and TP) we re greater in manure treatments than in the Boca Raton biosolids treatments. Thus, values of PWEP could be superior to WEP as indices of organic source of P solubi lity and as predictors of P loss. Table 8-3. Total P (TP) and bioavailable P (B AP) losses and some indi ces of the P-sources solubility <---Possible P-source coefficients--> Total BAP loss (mg) Total P loss (mg) Source WEP (g kg-1) PWEP¶(%) FPSC§PSC† High rate Low rate High rate Low rate TSP 175 84 0.05 1.00 (1.0) 548 a 156 a 774 a 189 a Poultry Manure 4.57 18 0.05 0.46 (0.9) 190 b 113 a 558 b 175 a Boca Raton biosolids 5.52 11 0.015 0.55 (0.8) 186 b 53.2 b 516 b 193 a Pompano biosolids 1.16 4 0.015 0.12 (0.4) 62.8 c 44.3 b 151 c 124 a †P-source coefficients calculated from the P-source WEP (g kg-1) as:PSC = 0.102 x WEP 0.99 (Elliott et al., 2006). Values in the parenthesi s are the corresponding Mid-Atlantic region PSC for the sources. ‡P-source coefficients from the draft Florida P Index (Gra etz et al., 2004). §Percentage water extractable P Phosphorus loss from treatments followed by the same letters (in the same column) are not different at p = 0.05 by Tukey (statistical analys is based on log-transformed data)

PAGE 194

194 The PSC values related well with the ranki ngs of mass of P losses (BAP and TP), especially for biosolids (Table 8-3). The PSC is a characteristic of th e P-source developed in Pennsylvania as an indicator of relative solubi lity and accounts for the proportion of total P applied to a field that is potenti ally subject to loss with draina ge water (Weld et al., 2000; Coale et al., 2005). The PSC is calculated from an em pirical relationship us ing WEP values (g kg-1) of the sources as: PSC = 0.102 x WEP 0.99 The relationship was derived from correlations of runoff dissolved P from several studies with the P-source WEP values (Elliott et al., 200 6). The masses of TP and BAP losses for the various P-sources are given along with the PSC va lues calculated from the P-source WEP values and other indices in Table 8-3. The Mid-Atlantic region PSC values for use in P Index site evaluation are expressed on a relative scale from zero to one: 1.0 (inorganic P fertilizer and swin e manure), 0.8 (other manures, BPR biosolids) 0.5 (alum-treated manure) and 0.4 (all other biosolids) (Coale et al., 2005). The Pennsylvania PSC values (related to TSP) are 1.0 (swine slurry), 0.9 [poultry (layer), turkey, duck, and dairy (liquid) manure], 0.8 [poultry (broiler), beef and dairy (bedded pack) manure and BPR biosolids], 0.4 (alkaline st abilized biosolids), 0.3 (conventi onally stabilized and composted biosolids) and 0.2 (heat-dri ed and advanced-alkaline stabilized biosolids). The relationships between the total BAP lo sses and the various indices of P-source solubility were modeled. Masses of BAP losses co rrelated better with PWEP values than with the PSC and FPSC values especially in the Florid a sand, where leaching is an issue (Table 8-4). Regression of masses of total BAP loss with application rates gave poor relationships (r2 < 0.30) with high variability (CV = 95%, Table 8-5). The poor relationship indicates that application rate

PAGE 195

195 alone is not sufficient to account for P losses where different Psources are applied. When the differences in solubility of the P-sources were accounted for by multiplyi ng the application rates with PWEP, PSC, or FPSC values, the relationships improved (r2 > 0.50). Elliott et al. (2005) also reported improved prediction of runoff P concentrations when application rates were adjusted with PSC values (rates multiplied by PSC values). However, accounting for P-source solubility with PWEP values resulted in better coefficients of determination (r2 = 0.81) and lower variability (CV = 49%) than when FPSC (r2 = 0.54, CV = 76%) or PSC values (r2 = 0.77, CV = 54%) were used. The data indicate not only the need for a coefficient to account for difference in P-source solubility, but that PW EP is superior to PSC and FP SC, especially in cases where leachate P losses are significant, as in Florida soils. Table 8-4. Pearson correlation coefficients and p -values between P loss (bioavailable P (BAP) and total P (TP)) and various P-s ource solubility coefficients. Form of P loss Rate WEP §FPSC ¶ PSC # PWEP 0.501† -0.266 0.135 0.048 -0.153 Runoff 0.007‡ 0.1802 0.087 0.812 0.445 0.367 0.714 0.439 0.641 0.711 Leachate 0.0596 <0.0001 0.008 0.0003 <0.0001 0.517 0.650 0.542 0.665 0.679 Based on rainfall simulation experiment BAP (Immokalee soil) Total 0.0057 0.0002 0.004 0.0002 0.0001 0.258 0.694 0.476 0.445 0.635 Total P (Candler soil) Leachate 0.3016 0.0014 0.046 0.064 0.005 0.3610 0.679 0.466 0.648 0.709 Based on data from Elliott et al., 2002 Total P (Immokalee soil) Leachate 0.141 0.0019 0.052 0.004 0.001 †Correlation coefficient (r) ‡p -value §P-source coefficients from Florid a P Index (Graetz et al., 2004). ¶P-source coefficients calculated from the P-source WEP (g kg-1) (Elliott et al., 2006). #Percentage water extractable P

PAGE 196

196 Table 8-5. Regressions of total bioavailabl e P (BAP) mass loss with P applied (and FPSC†, PWEP‡ and PSC§ adjusted P applied) for all the P-sources in rainfall simulation experiment. Independent variables Intercept Slope r2 CV (%) p -value Rate 36.6 0.94 0.27 95 0.0001 Rate*FPSC 30.9 30.3 0.54 76 <0.0001 Rate*PSC 13.5 2.11 0.77 54 <0.0001 Rate*PWEP 57.2 0.03 0.81 49 <0.0001 † P-source coefficients from the draf t Florida P Index (Graetz et al., 2004). ‡ Percentage water extractable P § P-source coefficients calculated from the P-source WEP values (g kg-1) (Elliott et al., 2006). The best regression of P loss with PWEP adju sted rates was also obtained for data from another study (Elliott et al., 2002a) that focuse d on P leaching. The study used another Florida soil (Candler series) in additi on to the Immokalee soil used in the rainfall simulation study (Table 8-6). The predictions for the two soils were better when application rates were adjusted with PWEP values (r2 = 0.85, 0.79 and CV = 59, 86%) than with FPSC values (r2 = 0.72, 0.36 and CV = 81, 150%) and PSC values (r2 = 0.74, 0.56 and CV = 79, 123%). Table 8-6. Regressions of mass of P loss with P applied and FPSC†, PWEP‡ and PSC§-adjusted P loads for P-sources used in a leaching study with two Florida soils (data from Elliott et al., 2002a used). <-----------Candler so il----------> <--------Immokalee soil ---------> Independent variables r2 CV (%) p -value r2 CV (%) p -value Rate 0.14 142 0.148 0.03 183 0.5270 Rate*FPSC 0.72 81 <0.0001 0.36 150 0.0146 Rate*PSC 0.74 79 <0.0001 0.56 123 0.0008 Rate*PWEP 0.85 59 <0.0001 0.79 86 <0.0001 † P-source coefficients from the draf t Florida P Index (Graetz et al., 2004). ‡ Percentage water extractable P § P-source coefficients calculated from the P-source WEP values (g kg-1) (Elliott et al., 2006).

PAGE 197

197 Using data from the rainfall simulation study, P Index scores were calculated for the four P-sources at each of the two rates (P-based and N-based) assuming similar transport variables, FIVs, and AM, with no waste water applied. The scor es were calculated using the draft Florida P Index worksheet, and P-sources accounted for by th e three different approaches (PSC, FPSC, and PWEP) as in Table 8-7. Ranking of BAP losses was more consistent with P Index score obtained using PWEP, than with either PSCor FPSC-based P Inde x scores. The observed Total BAP losses were regressed with the calculated P Index scores obta ined using each of the three approaches (PSC, FPSC, and PWEP) as in Fig. 8-2. Regression also show P Index scores obtained using PWEP better estimated BAP loss, than scores obtai ned using either PSC or FPSC (Fig. 8-2). BAP loss = 1.59P-Index (PWEP) 18.1 r2 = 0.96 BAP loss = 1.86 P-Index (FPSC) 63.9 r2 = 0.61 BAP loss = 1.31P-Index (PSC) 58.1 r2 = 0.910 100 200 300 400 500 600 0100200300400500 P-Index scoreBAP loss (mg) Using PWEP Using PSC Using FPSC Figure 8-2. Regression of bioava ilable P (BAP) loss with P Inde x score obtained using varying measures of coefficients (Percentage water extractable P (PWEP); P-source coefficients (PSC); Florida P I ndex source coefficients (FPSC)).

PAGE 198

198 Table 8-7. Trends of draft Florida P Index scor es (using different P-s ource coefficients; FPSC†, PWEP‡ and PSC§) and bioavailable P loss from th e four P-sources at two P rates during the rainfall simulation study <-Manure-> Boca Raton biosolids <--TSP----> Factors Variable Pbased Nbased Pbased Nbased Pbased Nbased Pbased Nbased P rate (lb P2O5 acre-1) 115 458 115 458 115 458 115 458 FPSC 0.05 0.05 0.015 0.0150.015 0.015 0.05 0.05 PSC 0.0460.0460.055 0.0550.012 0.012 0.10 0.10 PWEP 0.0180.0180.011 0.0110.004 0.004 0.0840.084 P-source multiplier (FPSC) 5.7 22.9 1.7 6.9 1.7 6.9 5.7 22.9 P-source multiplier (PSC) 5.3 21.0 6.3 25.4 1.4 5.4 11.5 45.8 P-source multiplier (PWEP) 2.1 8.2 1.3 5.0 0.5 1.8 9.6 38.5 FIV@ 7ppm M-1P 0.35 0.35 0.35 0.35 0.35 0.35 0.35 0.35 Application method 6 6 6 6 6 6 6 6 P-source Waste water (not applied) 0 0 0 0 0 0 0 0 Soil Erosion 1 1 1 1 1 1 1 1 Runoff potential 2 2 2 2 2 2 2 2 Leaching potential 4 4 4 4 4 4 4 4 Transport Potential to reach water body1 1 1 1 1 1 1 1 P Index score using FPSC 97 234 65 106 65 106 97 234 P Index score using PSC 93 219 102 254 62 94 142 417 P Index score using PWEP 67 117 61 91 54 65 128 359 Bioavailable P loss (mg) 113 190 53 186 44 63 156 549 †P-source coefficients from the draft Florida P Index (Gra etz et al., 2004). ‡Percentage water extractable P §P-source coefficients calculated from the P-source WEP values (g kg-1) (Elliott et al., 2006). To enhance comparing the wide spectrum of P-source solubilities, PWEP, as a continuous coefficient, could be used. The PWEP could be converted to a P-source co efficient in the range of 0 to 0.10 (as in FPSC) by multiplying the PWEP by 10-3. Thus, TSP with PWEP of 85%

PAGE 199

199 would be assigned a source coefficient of 0.085. The PWEP can be calculated or taken as default values from the average of the PWEP values fo r different P-sources re ported by Brandt et al. (2004) as in Table 8-8. The proposed PWEP-b ased PSC values would substitute for the coefficients (0.05, 0.015 and 0.1) in the “P application source a nd rate” section of the draft Florida P Index. This approach not only adequately accounts for differences in the P-sources, but also satisfies calls for the use of continuous co efficients to account for the P-source impacts (Elliott et al., 2006). Table 8-8. Percentage water extractable P (PWEP) values of some P-sources (calculated using data from Brandt et al., 2004) and the corresponding Psource coefficients based on the PWEP values at the 0 – 0.1 range in the draft P Index. P-source Type of P-source § FPSC †PWEP (%) ‡ PWEP-based source coefficient (PWEP-SC) Aerobically digested cake 0.015 2.75 0.003 Anaerobically digested cake 0.015 2.21 0.002 Biological P removal 0.015 13.9 0.013 Alkaline stabilized cake 0.015 7.39 0.007 Composted 0.015 3.04 0.003 Heat dried 0.015 0.48 0.0005 BPR heat dried 0.015 11.3 0.011 BPR N-vitro 0.015 0.21 0.0002 Biosolids Unstabilized 0.015 10.4 0.010 Dairy manure 0.050 51.8 0.052 Poultry manure (Layer) 0.050 20.4 0.020 Manures Poultry manure (Broiler) 0.050 20.9 0.021 Fertilizer TSP 0.050 85.2 0.085 †Based on average of PWEP values obtained from Br andt et al., 2004 ‡Based on the PWEP and the 0 0.1 range of P-sour ce coefficients in the draft Florida P Index (Graetz et al., 2004) §P-source coefficients from Florid a P Index (Graetz et al., 2004).

PAGE 200

200 Summary and Conclusions Land application of different P-sources result ed in varying environmental P losses because of differences in P-source water so lubility. The three coefficients suggested in the draft Florida P Index (coefficient = 0.05 for fertilizer and ma nure, 0.015 for biosolids, and 0.10 for waste water) are insufficient to account for th e wide variability in organic s ources of P solubility. Masses of both TP and BAP losses from various P-sources applied at N-based and P-based rates followed similar trends with P-source percentage water extractable P (PWEP) values in studies that accounted for leachate P loss. The trend of the P losses (PWEP values in parentheses) was: TSP (84%) > manure (18%) > Boca Raton biosolids ( 12%) > Pompano biosolids (4%). Regressions of BAP loss with application rate (r2 = 0.27) were improved by accounting for P-source solubility differences with: the Florida P Index coefficients (r2 = 0.54), P-source coefficients (PSC) values suggested by Elliott et al. (2006) (r2 = 0.77), and PWEP (r2 = 0.81). Similar improvements in the P loss model with PWEP and other coefficients were observed for data from another study that meas ured P leached from two Florida soils amended with eight biosolids, poultry manure and TSP. The r2 values observed in th e regressions of P lost with P application rate were im proved by using coefficients in th e draft Florida P Index and PSC, but the improvement was better s till with PWEP values. Use of coefficients based on PWEP of the P-source is suggested as an alternative to coefficients curren tly used in the draft Florida P Index, and default values for PWEP-b ased coefficients are suggested.

PAGE 201

201 CHAPTER 9 FIELD VALIDATION OF ENVI RONMENTAL IMPACTS OF LAND APPLIED P-SOURCES AND WATER TREATMENT RESIDUAL (WTR) Introduction Phosphorus added in excess of forage uptake a nd a soilÂ’s P sorption capacity can leach to shallow groundwaters (Liu et al., 1997). This is pa rticularly important in Florida low Psorbing sands. The dominant soil group in Florida, Spodo sols, is characterized by high water tables located between the Bh and the A horizons, es pecially during the summer rainy season. Phosphorus contamination of the shallow groundw aters can be conveyed to surface waters via drainage ditches (Burgoa et al ., 1991; Mansell et al., 1991). A chemical fractionation study by Graetz and Nair (1995) indicated th at about 80% of total P (TP) in the A-horizon of spodosols is leachable. Variables associated with P-sources that affect the amounts of P leached are the composition of the sources, source application rates, and the use of P-sorbing materials such as water treatment residuals (WTR). Studies have indicated the sorption propertie s of low P sorbing soils can be improved by applying WTR (Peters and Basta, 1996; Basta and Storm, 1997; OÂ’C onnor and Elliott, 2001; OÂ’Connor et al., 2002a; Elliott et al., 2002b; El liott et al., 2005; Novak and Watts, 2005). Reductions in P concentrations in runoff occur following surface applicat ion of WTR (Peters and Basta, 1996; Basta and Storm, 1997; Dayton et al ., 2003), and of P in leachates when WTR is incorporated into the soil (E lliott et al., 2002b; OÂ’Connor et al., 2002a; Novak and Watts, 2004; Dayton and Basta, 2005). Surface co-application of the P-sources and WTR reduced both leachate and runoff P concentrations in a ra infall simulation experiment (Chapter 7). Most studies that evaluated impacts of the P-sources and WTR additions on soluble P in Florida soils have been either laboratory inc ubations, column leaching studies, indoor rainfall

PAGE 202

202 simulation studies, or other procedures that control confounding vari ables, and are not representative of real world fi eld and landscape conditions. For exam ple, use of packed rainfall simulation boxes with disturbed soil and infiltration limited by th e 5 cm depth and the nine 5mm-diameter drain holes may not adequately simulate P transport in natural landscapes. The hydrology of bare, packed, sieved soils likely diffe rs from field soils with undisturbed structure, horizonation, and plant coverage. The need for re alistic data prompted the validation of some studies with field runoff studies using either natural or simula ted rainfall (Gascho et al., 1998; McDowell and Sharpley, 2001). Kleinman et al. (2004) evaluated the use of packed boxes to simulate P transport from agricultural soils a nd reported practical but limited comparability of the soil box data to field plot data . There is also a need to validate the suitabi lity of the modified (double deck) rainfall simulation box design (descr ibed in Chapter 7) to measure P loss in agricultural soils where transport is dominated by leaching and other subsurface mechanisms. The objectives of this study were to valid ate the impacts of P-sources and WTR on P loss in a natural field setting and to evaluate the suitability of the modified rainfall simulation box design to measure P loss in soils where leaching is significant. Materials and Methods Data from the field experiment described in Chapter 4 were used for this study. The study involved surface application of 4 P-sources at two ra tes, with or without WTR. In addition to the soil and plant samples collected, ground water and surface water samples were collected. Two wells, one shallow (< 0.9 m) and one deep (~3 m) , were installed in the center of each plot 1.2 m apart. The shallow and deep wells were lo cated above, and below, the spodic horizon, respectively (Fig. 9-1). The deep wells yielded samples all year round, in cluding the dry winter periods when the water table falls be low the shallow well sampling depth.

PAGE 203

203 Figure 9-1. Deep and shallow well positions under ground in the field study The wells were constructed of PVC pipe (5 cm diameter), and were surrounded by a 20 cm diameter casing of the same material. The surf ace water sampling system included an electronic controller, data logger, and telemetry system to ensure collection of flow-weighted, composite water samples. Each of the experimental pl ots contained one sampling scheme for surface, shallow, and deep ground waters samples (Fig. 9-2) . The wells were located at the center of each plot, while the electronic contro ller was located at the edge of each plot along the drainage ditch. Figure 9-3 shows the 17 rows of plots (one block or rep) and the surface wa ter sampler with data logger and telemetry system. Spodic horizon E horizon Deep well Soilsurface Shallow well A horizon 1.2m

PAGE 204

204 Figure 9-2. Plot layouts and water samplers locations in the field study Figure 9-3. One of the experime ntal blocks (replicates) show ing the water sampler and the telemetry installed on each of the 17 rows of plots in the field study. Legend Plotberm Shallowditch Drainage ditch Shallow0.9mwell Deep 3 m well Flume and sampler

PAGE 205

205 Initial groundwater samples were collected fr om the deep and shallow wells (but no grab water samples – surface runoff) on March 19, 2003, before treatment application. The Al-WTR (1 % by weight) was applied first, on May 9-13, 2003. The two bioso lids and manure were applied from May 13-14, 2003, while the TSP fertilizer was applied on May 19, 2003. Additional water samples were co llected from the shallow wells five times in 2003 (June 24, August 19, October 3, November 10 and December 17) and four times in 2004 (August 11 and 30, and September 8 and 30). Deep well samples we re collected eight times in 2003 (June 13 and 27, July 17, Aug. 19, Sept. 19, Oct. 20, Nov. 10 and Dec. 22) and seven times in 2004 (March 23, April 28, May 21, June 23, July 13, Aug. 19, Sept. 30). Surface water (grab) samples were collected following periods of high rainfall. Four hurricanes (Charley, Fr ances, Ivan and Jeanne) impacted Florida within 44 days (August 15 and September 25) in 2004 (Fig. 9-4). All water samples were analyzed for orthophosph ate P (ortho-P), total dissolved P (TDP), and total aluminum concentrations. Orthophosphat e P concentrations were determined on the filtered samples colorimetrically with the Mu rphy and Riley method (1962). Total dissolved phosphorus (TDP) and aluminum concentrations were measured on the filtered water samples after digesting 10 mL of th e samples with 0.5 mL 6N H2SO4 and 0.15g of potassium persulfate in an autoclave for 1h (Pote and Daniel 2000a an d b). Digested samples were analyzed for P colorimetrically (Murphy and Riley, 1962), while ICAP was used to determine Al concentrations. All samplings and analysis we re performed in accordance with the Florida Department of Environmental Protection’s sta ndard operating procedures, to minimize sampling and handling contamination (Florida Departme nt of Environmental Protection, 2002a and b).

PAGE 206

206 Figure 9-4. Flooding at Ki rton Ranch (about 200 me ter away from experimental plots) as observed after hurricane Jeanne on September 26th, 2006. Statistical Analysis Normality, constant variance and independence tests were carried ou t on the data and, whenever assumptions of normality and cons tant variance were violated, appropriate transformation power was determined by Box Cox using SAS (SAS Institute, 1999). Analysis of Variance (ANOVA) was performed on all the data using PROC GLM to determine significant treatment effects (SAS Institute, 1999). Wh en significance was indicated by ANOVA, means multiple comparisons by Tukey test were perf ormed using SAS, at 0.05 significant level. The Cate-Nelson method was used to identify the change point in the relationship between groundwater P and the soil sorption indices. About 200 meters away from Kirthon Ranch experimental plots after hurricane Jeanne (September 26, 2004)

PAGE 207

207 Results and Discussion Groundwater Aluminum Concentrations and WTR The Al concentrations of all the water samp les obtained during the study were unaffected by WTR rate, P-source, or P-source application rates. The ranges of Al concentrations in samples obtained after treatment app lication were 0.4 to 1.2 mg L-1, and 0.7 to 2.7 mg L-1 for the shallow and deep wells, respectively. The ranges compared well with the concentration ranges in samples obtained before treatments a pplication (0.6 to 2.4 mg L-1 for shallow wells and 1.0 to 3.8 mg L-1 for deep wells) on March 19, 2003. The trends of Al concentrations in samples obtained before and after treatment applications (2003-2004) are shown in Fig. 9-5. Generally, the Al concentrations were greater in the deep wells than in the shallow wells, which could result from contributio ns of organic Al species to Al solubility in the spodic horizon and not from the surface-applied treatments. Nil sson and Bergkvist (1983) studied Al chemistry in a Swedish podzols, and reported greater total Al concentrati ons (95 to 115 µM L-1 or 2.6 to 3.1 mg L-1) in leachate samples below th e Bh-horizon, than below the A-horizon (3.3 to 47 µM L-1 or 0.09 to 1.3 mg L-1). Thus, surface-applied Al-WTR increas ed soil total Al concentrations of surface soils, but did not affect Al concentratio ns in shallow or deep ground waters or runoff (grab) samples (Fig. 9-6). The ground water Al data were consistent with the results obtained from studying agronomic impacts of applied WTR in the field and the glasshouse, where plant Al concentrations were the same in WTRamended and unamended soils. Thus, WTR can be safely used to enhance the P-sorption capacity of Florida soils and reduce soluble P losses without increasing soluble Al c oncentrations in water or Al concentrations in plants.

PAGE 208

208 Figure 9-5. Trends of Al concentr ations surface grab (a and b), sh allow well (c and d) and deep well (e and f) water samples taken during the study. (a) Grab Al (2003)0.0 0.5 1.0 1.5 2.0 7090110130150 Days after treatments applicationAl concentration (mg L-1) N-based N-based (with WTR) Control P-based P-based (with WTR) (b) Grab Al (2004) 0.0 0.5 1.0 1.5 2.0 480490500510 Days after treatments applicationAl concentration (mg L-1) N-based N-based (with WTR) Control P-based P-based (with WTR) (c) Shallow Al (2003) 0.0 0.5 1.0 1.5 2.0 -1000100200 Days after treatments applicationAl concentration (mg L-1) N-based N-based (with WTR) Control P-based P-based (with WTR) (d) Shallow Al (2004) 0.0 0.5 1.0 1.5 2.0 460470480490500510520 Days after treatments applicationAl concentration (mg L-1) N-based N-based (with WTR) Control P-based P-based (with WTR) (e) Deep well Al (2003) 0 1 2 3 4 5 6 -100-50050100150200250300 Days after treatments applicationAl concentration (mg L-1) N-based N-based (with WTR) Control P-based P-based (with WTR) (f) Deep well Al (2004) 0 1 2 3 4 5 6 250300350400450500550 Days after treatments applicationAl concentration (mg L-1) N-based N-based (with WTR) Control P-based P-based (with WTR)

PAGE 209

209 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Initial shallow Shallow (2003) Shallow (2004) Initial Deep well Deep well (2003) Deep well (2004) Grab sample (2003) Grab sample (2004) SampleAl concentration (mg L-1) no WTR with WTR Figure 9-6. Aluminum concentrations in water samples taken during th e study as affected by surface applied WTR (n = 24, error bars represent one standard error). Groundwater P Concentrations and WTR The groundwater P concentrations indicated gr eater ortho-P than total dissolved P (TDP) in some cases, an anomaly that is still bei ng investigated (South Florida Water Management District, 2003). However, the analysis was done in a certified laboratory (Analytical Research Laboratory, UF (ARL)) with precision and accu racy ensured (5% duplicate and QC check samples and recoveries of 95 110%). Reagent and method blanks analyz ed also indicated no contamination. Other quality control measures in cluded matrix spikes and continuing calibration standards. Some samples were send to other cer tified laboratories (UF -IFAS Southwest Florida Research Laboratory (SWFRL); Wetland Bioge ochemistry Laboratory, UF; and Lee County

PAGE 210

210 Laboratory) for ortho-P duplicat e analysis. Total P could not be reanalyzed by the other laboratories because the laboratories are not sufficiently equipped for the analysis. Figure 9-7. Regression of ortho-P concentrations results in groundwater samples reanalyzed by three other laboratories (UF -IFAS Southwest Florida Rese arch Laboratory (SWFRL); Wetland Biogeochemistry Laboratory, UF; a nd Lee County Laboratory) with values obtained from Analytical Research Laboratory. (a) Analytical Research Laboratory, UF (ARL) VS Southwest Florida Research Laboratory (SWFRL) Y= 0.77X + 0.39 r2 = 0.91 0 2 4 6 8 0123456789 Ortho-P (mg L-1)--by ARLOrtho-P (mg L-1)--by SWFREC (b) Analytical Research Laboratory, UF (ARL) VS Lee County Laboratory Y = 0.71X + 0.07 r2 = 0.95 0 1 2 3 4 5 6 7 8 012345678910 Ortho-P (mg L-1)--by ARLOrtho-P (mg L-1)--by Lee County (c) Analytical Research Laboratory, UF (ARL) VS Wetland Laboratory, UF Y = 0.96X 0.08 r2 = 0.99 0 1 2 3 4 5 012345 Ortho-P (mg L-1)--by ARLOrtho-P (mg L-1)--by Wetland Lab

PAGE 211

211 Comparison of duplicate results indicates consis tently greater ortho-P concentration values of samples analysed by ARL than other laboratories (Fig. 9-7). However, the results from ARL have similar trend as in other laboratories (r2 > 0.9), which suggests the error may be systematic. Thus, inferences about the absolu te values of the P concentrati ons may be limited, but trends of the treatment effects could still be studied. Analysis of variance indicated that P concentr ations of water samples taken from deep and shallow wells were affected by the surface-ap plied WTR, but not by P-source or P-source application rate throughout the study (2003-2004). There were no effects of P-source, P application rate, or WTR addition on P concentr ations in the surface water grab samples collected during the tw o-year experiment. Trends of the water P concentrations for each P-source rate treatment (with and without WTR) for all sampling periods in 2003 and 2004 are shown on Fig. 9-8. The grab samples had similar P concentrations in samples obtained from plots treated at N-based and P-based rates, and whether WTR was applied or not. The observed sim ilar grab water P concentrations of samples from WTR treated and untreated plots are incons istent with effectiveness of WTR at reducing runoff P concentrations observe d during rainfall simulation study. The inconsistency could not be explained, but may result from the differences in material (P-sources and WTR) application in the two studies. The P-sources were applied fi rst followed by WTR (on t op) during the rainfall simulation study, but the other way round (WTR first then the P-source) in the field. Though the effect of WTR was not observed in the grab samples, the added WTR reduced P concentrations in groundwater samples throughout the study. Bo th ortho-P and TDP concentrations were greater in the absence, than in the presence, of WT R at both P-source rates (Fig. 9-8 and 9-9). Surface appl ying WTR at 1% reduced groundwat er P concentrations at both

PAGE 212

212 P-based and N-based source rates below P concentr ations observed in control treatment. Shallow well P concentrations for both P-source rates were similar in the presence of WTR in almost all the samples taken during the study. Thus, the added WTR masked the effect of P-source rates on ground water P concentrations. Figure 9-8. Trends of ortho-P conc entrations for the various treatm ents in surface grab (a and b), shallow well (c and d) and deep well (e and f) water samples taken during the study. (Note: Data for “pre-application” shallow well samples are pending in the lab as of the time of this report, and no surface gr ab samples were taken until 130 days after treatment application). (a) Grab ortho-P (2003)0 1 2 3 4 5 6 7 708090100110120130140150 Days after treatments applicationP concentration (mg L-1) N-based N-based (with WTR) Control P-based P-based (with WTR) (b) Grab ortho-P (2004)0 1 2 3 4 5 6 7 480485490495500505510 Days after treatments applicationP concentration (mg L-1) N-based N-based (with WTR) Control P-based P-based (with WTR) (c) Shallow ortho-P (2003)0 1 2 3 4 5 6 7 050100150200250 Days after treatments applicationP concentration (mg L-1) N-based N-based (with WTR) Control P-based P-based (with WTR) (d) Shallow ortho-P (2004)0 1 2 3 4 5 6 7 450460470480490500510520 Days after treatments applicationP concentration (mg L-1) N-based N-based (with WTR) Control P-based P-based (with WTR) (e) Deep Ortho-P (2003)0 1 2 3 4 5 6 7 -100-50050100150200250 Days after treatments applicationP concentration (mg L-1) N-based N-based (with WTR) Control P-based P-based (with WTR) (f) Deep Ortho-P (2004)0 1 2 3 4 5 6 7 250300350400450500550 Days after treatments applicationP concentration (mg L-1) N-based N-based (with WTR) Control P-based P-based (with WTR)

PAGE 213

213 Figure 9-9. Trends of total dissolv ed P concentrations for the various treatments in shallow (a and b), and deep well (c and d) water samples taken during the study The impact of WTR on P concentrations of samples from deep wells was not as pronounced as in shallow wells (Fig 9-9). Th e WTR effects were likely confounded by the Alrich spodic horizon above the sampling point of deep wells. The spodic horizon sorbs P, as does WTR. Samples taken before treatment applicatio n have similar P concentrations, and the grab samples were not affected by the WTR (Fig 9-10). However, all ground water samples taken from deep and shallow wells contained greater P concentrations in the absence, than in the presence, of WTR throughout the study. The effe ctiveness of WTR at reducing soluble P was reflected in the water P concentrations, even in 2004 samples, with gr eater hurricane activity. (a) Shallow Total-P (2003)0 1 2 3 4 5 6 7 -100-50050100150200250 Days after treatments applicationP concentration (mg L-1) N-based N-based (with WTR) Control P-based P-based (with WTR) (b) Shallow Total-P (2004)0 1 2 3 4 5 6 7 460465470475480485 Days after treatments applicationP concentration (mg L-1) N-based N-based (with WTR) Control P-based P-based (with WTR) (c) Deep Total P (2003)0 1 2 3 4 5 6 7 -100-50050100150200250300 Days after treatments applicationP concentration (mg L-1) N-based N-based (with WTR) Control P-based P-based (with WTR) (d) Deep Total P (2004)0 1 2 3 4 5 6 7 250300350400450500 Days after treatments applicationP concentration (mg L-1) N-based N-based (with WTR) Control P-based P-based (with WTR)

PAGE 214

214 Figure 9-10. Total dissolved and Ortho-P concentr ations of water samples taken during the study as affected by WTR application. Treatmen ts bar within the same sampling period capped by the same letters are not different at p = 0.05 by Tukey test Soil P Sorption Indices and Gr oundwater P Concentrations Studies on acidic sandy soils have shown good co rrelations between soil soluble P and the degree of P saturation (DPS) values, which are calc ulated from oxalate extractable P, Fe, and Al of soils (Beek, 1979; Nair and Graetz, 2002). Anot her index of soil P so rption is the soil P (a) Ortho phosphate0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5Shallow (2003) Shallow (2004) Initial Deep well Deep well (2003) Deep well (2004) Grab sample (2003) Grab sample (2004)SampleP concentration (mg L-1) no WTR with WTR a a aa b b b b ns ns ns (b) Total P0.0 0.5 1.0 1.5 2.0 2.5 3.0 Initial shallow Shallow (2003) Shallow (2004) Initial Deep well Deep well (2003) Deep well (2004) Grab sample (2003) Grab sample (2004)SampleP concentration (mg L-1) no WTR with WTR a b ns b bb ns a aans ns

PAGE 215

215 storage capacity (SPSC), which is also calculated from the extractab le P, Fe, and Al as explained in previous chapters (Chapter 3, 4, and 6). Data from both gla sshouse and field studies showed that soil sorption of P increased with WTR app lication, and that WTR effects were correlated with changes in DPS and SPSC values. Figure 9-11. Relationships between soil P sorptio n indices (Degree of P saturation (DPS), and soil P storage capacity (SPSC)) and shallo w well water ortho-P (SOP), and Total dissolved P (STP) concentrations. Soil DPS values were reduced below the enviro nmental threshold, with accompanied reduction in soil soluble P, when WTR was applied. Also, SPSC values were increased in all P-sources (a) DPS VS SOP0 2 4 6 8 10 12 14 0255075100125150175200225 DPS (%)P concentration (mg L-1) Control 2003 Control 2004 P-based no WTR 2003 P-based WTR 2003 N-based no WTR 2003 N-based WTR 2003 N-based no WTR 2004 N-based WTR 2004 P-based no WTR 2004 P-based WTR 2004 r2 = 0.27 (b) SPSC VS SOP 0 2 4 6 8 10 12 14 -400 -200 0 200 400 600 800 1000 SPSC (mg kg-1)P concentration (mg L-1) Control 2003 Control 2004 P-based no WTR 2003 P-based WTR 2003 N-based no WTR 2003 N-based WTR 2003 N-based no WTR 2004 N-based WTR 2004 P-based no WTR 2004 P-based WTR 2004 r2 = 0.29 (c) DPS VS STP 0 2 4 6 8 10 12 14 0255075100125150175200225 DPS (%)P concentration (mg L-1) Control 2003 Control 2004 P-based no WTR 2003 P-based WTR 2003 N-based no WTR 2003 N-based WTR 2003 N-based no WTR 2004 N-based WTR 2004 P-based no WTR 2004 P-based WTR 2004 r2 = 0.12 (d) SPSC VS STP 0 2 4 6 8 10 12 14 -400 -200 0 200 400 600 800 1000 SPSC (mg kg-1)P concentration (mg L-1) Control 2003 Control 2004 P-based no WTR 2003 P-based WTR 2003 N-based no WTR 2003 N-based WTR 2003 N-based no WTR 2004 N-based WTR 2004 P-based no WTR 2004 P-based WTR 2004 r2 = 0.16

PAGE 216

216 treatments, and at both P-source application ra tes following addition of WTR. The reduction in soil soluble P concentrations with increasing so il P sorption properties noted earlier (Fig. 4-6) was confirmed by this study (Fig. 9-12). Generall y, the relationships be tween the groundwater P concentrations and the soil sorption i ndicies (DPS and SPSC) were very poor (r2 < 0.3). However, change points were easier located by the Cate Nelson procedure at ~0 mg kg-1 SPSC, than at 25% DPS thresholds. Thus, the data suppo rt the contention of Nair and Harris (2004) that SPSC value is a better indicator of environmenta l P hazard than DPS. The SPSC values better assesses the capacity of a soil to retain P and, thereby, redu ce groundwater P concentrations. Summary and Conclusions The greater soil total Al concen trations of surface soils follow ing surface application of AlWTR did not increase Al concentrations in surf ace water, or shallow and deep groundwaters. The data support the contention th at WTR can be safely used to enhance P sorption capacity of Florida sandy soils without increas ing groundwater Al concentration. Surface-applied WTR reduced groundwater P co ncentrations in all the P-source and Psource rate treatments to values below those observed in the cont rol treatment. The impact of WTR on P concentrations of deep well sample s was pronounced despite the confounding effect of the Al-rich spodic horizon above the sampling point. The P concentrations of water from the shallo w wells related better with SPSC values than DPS values, supporting the use of SPSC values as measures of soil capacity to retain P and protect against environmental P hazard.

PAGE 217

217 CHAPTER 10 SENSITIVITY ANALYSIS OF THE DRAFT FLORIDA P INDEX Introduction Phosphorus source, P-source rate, and sorption properties of the soils are among variables shown by this study to affect P losses in Florid a sands. Thus, appropriate management to reduce P losses requires estimates of P loss potential from landscapes using a model that is sensitive to the variables. The phosphorus index (P Index) is a site-specific, qua litative vulnerability assessment model being developed by states (i ncluding Florida) for P management plans to reduce P losses and address water quality (USE PA, 1999). The Florida P Index will help to determine whether organic sources of P should be applied at N-based or P-based rates and other suitable site specific nutrient management syst ems that could be employed. However, there is need to study the sensitivity of the Florida P Inde x model to the variables identified to affect P losses. Basically, the concept of the P Index is that management of agricultural P should target the critical point at which P-source and transport factors overlap (G burek and Sharpley, 1998). Thus, the P Index developed by all the states in the US identifies “source” and “transport” variables that could account for P losses. The original P Index by Lemunyon and Gilbert, (1993) contained five source and three transport variables, each a ssigned five discrete ratings. The draft Florida P Index differs slightly, and contains four variable s related to site and tr ansport (Table 10-1, Part A) and five source variables (Tab le 10-1, Part B). The transport va riables (i.e. soil erosion, runoff potential, leaching potential, and po tential to reach water body) are assigned discrete ratings to describe the magnitude of each variable. App lication method and P-source applied are also

PAGE 218

218 assigned discrete ratings, but other source variable s (fertility index value, P application rate, and waste water application) have continuous ratings. Table 10-1. The draft Florida P Index worksheet Part A: Transport potential due to site and transport characteristics Site and Transport characteristics Phosphorus transport rating Value Soil Erosion No surface outlet 0 < 5 T/A a 1 5-10 T/A 2 10-15 T/A 4 > 15 T/A 8 Runoff Potential Very Low 0 Low 1 Medium 2 High 4 Very High 8 Leaching Potential Very Low 0 Low 1 Medium 2 High 4 Very High 8 Potential to reach water body Very Low 0 Low 1 Medium 2 High 4 Total for part A: Site and Transport b a T/A = tons per acre. b if the sum of part A is 0 (zero), then change the sum to 1 (one). Part B: Transport potential due to phosphorus source management Phosphorus source Management Phosphorus Loss Rating Value Fertility Index Value Soil Fertility Index X 0.025 ( __ ppm P X 2 X 0.025) c P Application Source and Rated 0.05 X (__ lbs P2O5/acre) for fertilizer, manure, or compost 0.015 X ( lbs P2O5/acre) for biosolids 0.1 X ( lbs P2O5/acre) for waste water Application Method No Surface Outlet or solids incorporated immediately or injected 0 Applies via Irrigation or solids incorporated within 1 day of application 2 Solids incorporated within 5 days of application e 4 Solids not incorporated within 5 days of application 6 Waste Water Application 0.20 X acre inches/acre/year Sum for Table 2: Phosphorus Source c From soil test (Mehlich 1) results. d Initial evaluation should be N-based rates e Solids include fertilizers, composts, biosolids, and manure and other animal wastes P Index score = (site and transport ratings)* (Phosphorus source rating) The P Index score (i.e. overall P loss vulne rability rating) is obtained by multiplying transport and source total values as: I = (RSE + RRP + RLP + RPWB)*(RFIV + RAM + RWWA + RPAS*RPSAR) Equation 10-1

PAGE 219

219 Where I = P Index score; RSE, RRP, RLP, RPWB, RFIV, RAM, RWWA, RPAS, RPSAR are site ratings for soil erosion, runoff potential, leaching potential, pot ential to reach water body, fertility inde value, application method, waste water applica tion, P-source applied, an d P application rate, respectively. The Florida P Index is a linear model, as th e relationship between the P Index score and the variables is first order. Prior to implementi ng the Florida P Index, the model should be tested, and the impact of each variable on P loss evaluated in the field. However, the first step should be a sensitivity analysis to evaluate the variab les included in the P Index and to ensure the conformity of the P Index model to the intentio ns of the developers (Brandt and Elliott, 2005). Sensitivity analysis is a tool that can help th e modeler and the users understand the importance of variable inputs on the computed outputs. The sens itivity analysis can be classified based on the model type (as mathematical, st atistical, and graphical ) or based on capability, rather than the methodology, of a specific technique. For example, deterministic analysis, using mathematical method such as nominal range sensitivity, can be employed to evaluate model variables sensitivity. Alternatively, an anal yst may perform a probabilistic an alysis, using either frequentist or Bayesian frameworks, in which case statistical-based sensitivity analysis methods can be used. The nominal range sensitivity analysis (NRSA) employed in this study is a mathematical method that assesses the sensitivity of model out put to the range of variation of each input variable. The NRSA involves ca lculating and studying the change in the output within possible ranges of the input values. Graphical methods, which visually represent sensit ivity in the form of graphs, charts, or surfaces, are compatible with NRSA and can be used to complement the results. The NRSA addresses only a potentially small portion of the possible space of input values, as interactions among i nputs are not captured (Cullen a nd Frey, 1999). Thus, the NRSA

PAGE 220

220 will determine the degree to which P Index score changes per unit change in the impact variables. The sensitivity of each input variable is determined by the slope (i.e. changes in output score with unit change in input values) obtained fr om the plot of P Index scores with increasing values of each variable. If a sma ll change in a variable value results in a relatively large change in the P Index score, the P Index is said to be sensitive to that variable. The NRSA is an important method for assessing the quality of a m odel, model robustness, and reliability of the model analysis. The sensitivity analysis can also help direct future studies. Variables to which the model is relatively sensitive could require fu rther characterization, while possible causes of insensitivity to other variables may need to be investigated. Th e objective of the study is to utilize sensitivity analysis to evaluate the suitabil ity of draft Florida P Index model as a tool for P management in different landscapes. Materials and Methods The nominal range sensitivity analysis pro cedure used by Brandt and Elliott (2005) was adapted for this study. Nominal range sensitivity analysis can evaluate the effect exerted on model outputs by individually varying only one of the model inputs across the entire range of plausible values, while holding all other inputs at nominal or baseline va lues (Cullen and Frey, 1999). The difference in the model output due to the ch ange in the input variable is referred to as the “sensitivity” or “swing weight” of the model to that particular input variable (Morgan and Henrion, 1990). The sensitivity analysis can be repeated for any number of individual model inputs and is most valid when applied to a lin ear model such as the draft Florida P Index. The Florida P Index is a linear model, as it can be represented by firs t order linear equation as: I = (RSE + RRP + RLP + RPWB)*(RFIV + RAM + RWWA + RPAS*RPSAR)

PAGE 221

221 Where I = P Index score The sensitivity of the P Index score (Si) with respect to variable “ i ", when all other input variables are kept cons tant at baseline ( Xi) can be computed as: Si = I = I Xi X Equation 10-2 Xi = condition where input variables are at baseline except variable i The variables in the P Index model are a mixt ure of continuous and discrete variables, which differ from each other in units. To enha nce variable comparisons using similar units, dimensionless sensitivity coefficients (Si ’) were computed by expressing each variable as percentage of the resp ective baseline value. Si ’ = I(X i + X i ) – I(X i ) 100*[((Xi + Xi)/(Xi)] Equation 10-3 where I(Xi) and I(Xi + Xi) are P Index scores for base line and perturbed values, respectively (Brandt and Elliott, 2005). The sens itivity of the variables in the P Index are compared using the slope (Si) obtained for each variable (Eschenbach, 1992). The greater the slope obtained for a variable, the more sens itive the P Index score to the variable. The relative importance of each input can be rank-ordered based upon the magnitude of calculated sensitivity measures, provided the range s assigned to each sensitive input are accurate. Thus, the study requires realistic domain limits (ra nges) and baseline conditions for each variable input. Below is a summary of the basis and justificati on of parameter selections for each of the nine input variables. Phosphorus Application Source (PAS): To account for P-sources , the P Index contains three PAS values (0.05 for fertilizer, manure or composts, 0.015 for biosolids and 0.1 for waste water). Thus, the range of PAS values in th e P Index, 0.015 – 0.1, is used as the domain limits, while the median value (0.05) represents the PAS baseline condition.

PAGE 222

222 Phosphorus Source Application Rate (PSAR): The PSAR values are selected based on UF/IFAS standardized fertiliz ation recommendation for agronomi c crops (Kidder et al., 2002). The minimum value was taken to be zero, repres enting a condition in which no P is applied. The maximum UF/IFAS recommended P-source rates are 175 lbs P2O5 acre-1 yr-1 (P-based) and 210 lbs PAN acre-1 (N-based). Typically, ~224 kg P ha-1 (450 lbs P2O5 acre-1) is assumed to represent P load at N-based rates of organic amendmen ts (Stehouwer et al., 2000). Hence, 450 lbs P2O5 acre-1 was chosen as the upper limit of the domain. The 225 lbs P2O5 acre-1 is half of the 450 value used as representating P lo ads at N-based rates and was chos en to represent the baseline value. Application Method (AM): The draft P Index rates P application method from zero (when the solids are incorporated immediately or inje cted) to 6 (when the solids are not incorporated within 5 days of application). Thus, the range 06 is used as the domain limit. In Florida, the common management practice is surface applicati on without incorporation when the amendment are applied to established pastur e (rating = 6), or inco rporated within 1-5 day when applied to other agronomic crops (rating 2). Thus, the medi an value of the two ratings assigned to these practices is “4” and is used as the baseline value for AM. Fertilizer Index Value (FIV): The FIV is calculated from the soil test P (STP) as: FIV = STP*2*0.025 The variable is the STP, a nd as a continuous variable, could range from zero to maximum values determined from the fi eld where the P Index is applie d. An upper limit of 300 ppm was used here based on a field survey study of poten tial sites for P Index use (personal discussion with Dr. V.D. Nair). Thus, the range 0-300 ppm ST P, which is equivalent to FIV values of 0 – 15, was used as the domain. The M-1P STP interp retation used for agronomic crops are: very

PAGE 223

223 low (< 10 ppm), low (10 – 15 ppm), medium (16 – 30 ppm), high (31 – 60 ppm) and very high (> 60 ppm). Based on the median of high STP, a value of 40 ppm was used as baseline STP (equivalent to FIV value of 2). Waste Water Application (WWA): Sewage waste water or effl uent is land applied in Florida to meet irrigation needs. In studies on land application of waste water in Florida, the application rate ranges fro m 400 mm (25 acre inches yr-1) to 1250 mm (~ 50 acre inches yr-1) (Maurer et al., 1995; Parsons et al., 2001). The P concentration ranges from 0.01 mg L-1 to 10 mg L-1, with average of 3.88 mg L-1. Thus, the range of WWA values used for the study was 0 acre inches yr-1 (when no water is applied) to 50 acre inches yr-1. The commonly used rate in Florida, (400 mm, equivalent to 25 acre inches yr-1; Parsons et al., 2001) was used as the baseline value. The P rate (P load) at the 50 acre inches yr-1 and at maximum concentration (10 mg L-1) falls within the range of PSA R used (0 450 lbs P2O5 acre-1). Phosphorus Transport Potential due to Site and Transport Characteristics: The transport component of the P Index consists of f our variables: Soil erosion (SE), runoff potential (RP), leaching potential (LP), and potential to reach water body (PWB). The four variables are each assigned discreet categorical ratings that serve as the dom ain limits (range) in this study. Thus, the domain limits are 0 to 8 (for SE, RP, and LP), and 0 to 4 (for PWB), and are the ranges of the ratings assigned to the variables. Data of runoff and leaching poten tial ratings for Florida soil survey map units (summarized by Hurt et al., 2006), were used to select baselin e values representative of Florida soils. The number of map units associated with undrained runoff, drained runoff, and leaching potential (categorized into very low, lo w, medium, high, and very high), in Florida soils for all counties

PAGE 224

224 pooled together is shown in Fig. 10-1a. Since dr ained and undrained runoff is not distinguished in the P Index, the two variables are pooled to obtain the values for runoff potential (Fig. 10-1b). Figure 10-1. Number of map units associated with (a) undrained runo ff, drained runoff and leaching potential and (b) reclassified to r unoff and leaching potenti al rated very low, low, medium, high, and very high in Florida soils. (a) 0 500 1000 1500 2000 2500 LeachingDrained runoffUndrained runoffNo of Map units very low low medium high very high (b) 0 500 1000 1500 2000 2500 3000 3500 4000 LeachingRunoffNo of Map units very low low medium high very high Mean = Medium Median = Medium Mode = Medium Baseline = Medium Mean = High Median = High Mode = Very high Baseline = High

PAGE 225

225 The baseline values are selected for each transport variable (Table 10-1), based on the distribution of runoff and leachi ng potential categories for all counties in Florida. The summary statistics of the map units indicates the “medium” category (rating = 2) as the mean, median, and mode of the leaching potential. Al so, the mean and median of runoff potential is “high” (rating = 4), but its mode is “very high” (rating = 6). Ba sed on the summary statistics, the baseline values used are medium (rating = 2) for leaching potenti al and high (rating = 4) for runoff potential. From the description of the PWB rating criter ia, the “low” category (r ating = 1) describes the situation in which P in runoff can be atte nuated by flow through a wetland, buffer strip or overland treatment area. A “low” rating (value = 1) describes common situ ations in agricultural areas and was selected as the baseline value for PWB. The SE factor was indicated to be <5T/A in most counties in FL by Hurt et al. (2006), and identified baseline value for SE as ”. Table 10-2 summarizes the input values (domain limits a nd the baseline) for each of the nine variables used in the NRSA. Table 10-2. Input values for each variable used in nominal range sensitivity analysis of draft Florida P Index Domain limits Variable Unit MinimumMaximum Baseline condition Fertility index value 0 15 2 P-source 0.015 0.1 0.05 P-source rate lbs P2O5 0 450 225 Application method 0 6 4 Waste water 0 4 2 Soil erosion 0 8 1 Runoff potential 0 8 4 Leaching potential 0 8 2 Potential to reach water body 0 4 1

PAGE 226

226 Results and Discussion Sensitivity of the Draft Florida P Index Model to the Variables The sensitivity of the P Index variables is compared using the slope obtained for each variable in the spider plot (Fig 10-2). The sc ore for base condition (178) fell within the high Pindex rating, which dictates Pbased management. The high baseli ne score suggests that most Florida landscapes will be rated high by the draft P-index. The steep er the slope of a variable in the spider plot (sensitivity coefficient), the mo re sensitive the P Index score to the variable. y (PSR) = 0.9x + 88 y (WWA) = 0.4x + 138 y (PWB) = 0.22x + 155 y (AM) = 0.32x + 146 y (FIV) = 0.16x + 162 y (SE) = 0.22x + 155 y (LP) = 0.45x + 133 y (PSA) = 0.9x + 88 y (RP) = 0.89x + 890 25 50 75 100 125 150 175 200 225 250 275 300 325 350 0100200300400500600700800 Factor perturbation (% of base condition)P index score FIV: Fertility index value PSA: P application source AM: Application method WWA: Waste water application SE: Soil erosion RP: Runoff potential LP: Leaching potential PWB: Potential to reach water body PSR: P source rate Base condition PI = 178 Figure 10-2. Spider diagram of vari ables in the draft Florida P Index.

PAGE 227

227 Figure 10-2 indicates that bot h PAS and PSAR have the greatest sensitivity coefficients (0.90) and, hence, exert the greatest impact on the P Index score. A 100% increase in the PSA or PSAR baseline ratings increase the P Index score by 90 points. The lo west sensitivity coefficient (0.16) was observed for the FIV regression. Incr easing the soil test P from 40 ppm to 80 ppm only increases the P Index score by 16. Thus, the draf t Florida P Index is not especially sensitive to the STP comparared to other variables. Re ducing STP from 40 ppm to 0 ppm only reduces the P Index score by 16. As the impact appears counteri ntuitive, the result may need to be verified by experimental data. The nine va riables are categorized into si x sensitivity coefficient groups: 0.90 (PSA and PSAR), 0.89 (RP), 0.45 (LP), 0.40 ( WWA), 0.32 (AM), 0.22 (SE and PWB), and 0.16 (FIV). Some questions posed by the sensitiv ity analysis results are obvious and should be addressed before validation with experimental data. For example, the sensitivity coefficient of AM was greater than that of FIV. The AM may be more important than FIV in soils where subsurface runoff flow is not an issue. Howeve r in Florida sandy low P-sorbing surface soils with extensive subsurface flow, AM may not af fect P loss as much as FIV. An important application of the sensitivity coefficients is aidi ng nutrient managers in identifying the impacts of varying management practices. For example, a mana gement practice to attain zero SE (and every other variable at baseline) will only reduce th e P Index score by 22. However, if instead of managing SE, the AM is improved by ensuring amendments or P-sources are immediately incorporated (rating = 0) instead of waiting 2 days or more to incorporate the solids (rating = 4), the P Index score is reduced by 32. Thus, P-source incorporation can change the field rating from a P Index score of 178 (P-based management) to 146, which allows N-based management. However, the question still remains. Does incorpor ation make so much difference in Florida soils with regards to P loss? Result from rainfall simu lation study (Chapter 7) indicated greater P loss

PAGE 228

228 through leaching though the P-sources were surf ace applied (and not incoporated). Thus, application method may have little impact in Florida soils where greater P losses are through leachate. However, field data are needed to test the impact of placement method on P loss. An important observation from the spider plot is that both the coeffi cients (slope) and the length of the regression lines (r ange of P Index score covered by each variable) vary. Thus, a variable may have a greater coefficient, but co ver a smaller P Index score range. A good example is PSA, which has a larger sensitivity coefficien t, but a smaller range than either LP or SE. The range of P Index scores covered by each va riable is called the swing, and calculated as: Swing = I(Xi max) – I(Xi min) Equation 10-4 Where I(Xi max) and I(Xi min) are the P Index scores at maximum and minimum values, respectively, of the ith variables (keeping all other variables at baseline). The calculated swings (also called tornado sw ing), and sensitivity coefficients for each variable are shown on Table 10-3. Table 10-3. Sensitivity coefficients and swings a nd normalized values for each variable in the draft Florida P Index Input factor Spider Sensitivity Coefficient Tornado Swing Normalized Sensitivity Coefficient (%) Normalized Swing (%) Fertility index value 0.16 120 18 67 P-source 0.90 153 100 85 P-source rate 0.90 180 100 100 Application method 0.32 48 36 27 Waste water 0.40 80 44 44 Soil erosion 0.22 178 24 99 Runoff potential 0.89 178 99 99 Leaching potential 0.45 178 49 99 Potential to reach water body 0.22 89 24 49

PAGE 229

229 The table also contains the normalized sens itivity coefficients and normalized tornado swings that express each sensit ivity coefficient and swing value as a percentage of the observed greatest values of the sensitivity coefficients and the swings, respectively. The swing is represented graphically as tornado swing in Fi g. 10-3. The tornado swing indicates the range of impact each variable has on the P Index score if a ll other variables are left at baseline value. 50100150200250300350 Application method Potential to reach water body Waste water Fertility index value P source Soil erosion Leaching potential Runoff potential P source rate P index score Base condition PI = 178 300 ppm 8 4 50 acre inches yr-10.015 0 450 lbs P2O58 0 lbs P2O50 ppm 0 0 6 0.1 0 0 6 0 8 0 0 Figure 10-3. Tornado diagram of variab les in the draft Florida Pindex Both PSAR and RP have the greatest impact, and are variables that can each reduce the P Index score to < 100 when other variables are at baseline values. Changes in the SE, FIV, and PWB are not capable of reducing the score belo w 150 (forcing P-based management), if other variables are at baseline. When other variables are at base line values, the minimum score achievable by reducing SE and PWB is 156 while FIV can only reduce the score to 162. Thus, the P Index score that will allow N-based manage ment (< 150 P Index score) is not achievable by singularly managing SE, PWB, or FIV.

PAGE 230

230 The FIV (or soil test P), which has the lowest sensitivity coefficient, has a greater swing (i.e impact on the P Index score) than AM, PWB, and WWA. Thus, the FIV (or STP), though not as sensitive as other variable s, could singularly increase the score to 282. However, if other variables are at baseline, the P Index score can not fall below 162 by simply reducing the STP. Among the transport variables, RP has the grea test sensitivity coe fficient, but not the greatest impact. The limit to its impact results fr om constraints by the domain limits allowed in the draft P Index. Similar constraints on the domain also limit the impacts of other variables with discrete ratings. Thus, LP and SE both have si milar impacts on P Index score (swing = 178) as RP. However, unlike with RP, both LP and SE can not be each managed to reduce the P Index score below 90, when other variables are at baseline. The PSA is also limited by the domain limit, and can swing the P I ndex score between 115 and 268. PWB can swing the score between 156 and 245, and the most limited swing was observed in AM (146 – 194). Management involving two or more variables can lower the score below the lower limits of the swing for individual variables. For exam ple, if no waste water is applied, the score is reduced from the 178 (base condition value) to 9 8. However, the P Index score can be further lowered to <80 if biosolids are also used instead of manure as the P-source, and lowered further, if the biosolids are incorporated immediately following application. The effects of the variable sensitivity co efficients and swing on the P Index can be categorized by the matrix of the normalized coefficients and the swing (Fig. 10-4). For comparison, normalized values of related vari ables from the Pennsylvania P Index and the original Lemunyon and Gilbert P Index (1993), as calculated by Brandt and Elliott (2005), were included in Fig. 10-4.

PAGE 231

231 0 10 20 30 40 50 60 70 80 90 100 0102030405060708090100 Normalized sensitivity coefficient (%)Normalized swing (% ) FL P index Lemunyon and Gilbert P index (1993) Pensylvania P index (Brandt and Elliott, 2005) STP PSA PSR AM WWA SE RP LP STP PSR AM SE RP CD FIV PSR AM RP PWB SE PSAPSA: P source applied PWB: Potential to reach water PSR: P source rate FIV: Fertilit y index value WW: Waste water A M: Application method SE: Soil erosion LP: Leaching RP: Runoff potential CD: Contributin g STP: Soil test P Figure 10-4. Nominal range sensitivity analys is matrix of the draft Florida P Index. The variables are grouped by se nsitivity and P Index score im pacts into high (normalized coefficient and/or swing > 67%), intermediate (normalized coeffi cient and/or swing > 33%, but < 67%), and low (normalized coefficient and/or sw ing < 33%) categories. Five variables in the draft Florida P Index, PSAR, PSA, RP, LP, a nd SE fell into the hi gh impact category. The remaining four variables, FIV, AM, WWA, and PW B, fell into the intermediate category. None of the variables in the draft Florida P Inde x fell within the low category, emphasizing the importance of all nine variables. The PSAR was categorized as a high impact variable in all three P-Indices (Florida, Pennsylvania, and the original Lemunyon and G ilbert P Index). The PSA is included in the Pennsylvania and Florida P Index, but not in th e original Lemunyon and G ilbert P Index, and is

PAGE 232

232 categorized as a high impact variable in both the Pennsylvania index an d draft Florida P Index. The high impact observed justifies the inclusion of the P-source variable in the P-Indices of both Florida and Pennsylvania. The high impact of RP reflects the effect of high water table conditions in Florida that promote substantial ru noff. Other important va riables (high category) in Florida are LP (included in Florida P Inde x, but not in the Pennsylvania or in original Lemunyon and Gilbert P Indices) and SE (included in Florida and Pensylvani a P Indices, but not in original Lemunyon and Gilbert P Index). The high impact category of SE results from the greater swing given to th e variable in the draft Florida P I ndex. Thus, the domain (0-8) may need to be revised (lowered), e.g., to 0-4, or as jus tified by experimental data. Application method fell into the medium category in the Florida P I ndex, unlike in the Pennsylvania and original Lemunyon and Gilbert P-Indices, where the variable was categorized as hi gh. The lesser impact of AM in the Florida than in the Pennsylvan ia and Lemunyon and Gilb ert P-Indices may be related to the local conditions of Florida soils. Surface applicati on, or incorporation, of P-sources may have little impact on P loss in Florida sa nds with high water tables and significant subsurface movement. However, in the greater P-sorbing Pennsylva nia soils, incorporation of the P-source may significantly reduce P loss as compared to surface applicatio n, and result in greater impacts of AM. The FIV (or soil test P), a high impact variable in the original Lemunyon and Gilbert PIndex, was categorized as a medium impact variab le in both the Florida and Pennsylvania P-Indices. The ranking is consiste nt with research studies indicating reduced impact of STP on P losses when other variables are considered (Pierson et al., 2001; Eghball and Gilley, 2001; DeLaune et al., 2002; Brandt and E lliott, 2003). PWB (or “contributing distances” in Pennsylvania P Index), are also categorized as medium impact variables in both the Florida and Pennsylvania P-Indices.

PAGE 233

233 Sensitivity of the Draft Florida P Index Model to P Management Apart from assisting the mode ler in studying the impacts of the variables on the P Index score, another practical use of sensitivity an alysis is evaluating P management strategies. Evaluation of impact management strategies on P losses is illustrated the following hypothetical question. How much can the P Inde x score be reduced if the mana gement strategy is to reduce the rating of any one variable by half? Figure 10-5 shows how much the score could be reduced when each variable is reduced by 50%, keeping all other variables at baseline. 120130140150160170 Fertility index value Potential to reach water body Soil erosion Application method Waste water Leaching potential Runoff potential P source P source rate P index score Base condition PI = 178 40 ppm 12.5 acre inches yr-14 1 4 0.5112 lbs P2 O51 2 0.05 0.025 0.5 20 ppm 225 lbs P2O51 2 25 acre inches yr-12 Figure 10-5. Draft Florida P Inde x scores associated with a 50 % reduction in eac h input factor when other factors are held at baseline values Reduction of PSAR from baseline 225 lbs P2O5 (110 kg P ha-1) to 112 lbs P2O5 (55 kg P ha-1; P-based) can reduce the PI score from 178 baseline value to 133. However, better Pmanagement could be achieved by applying biosolids at 225 lbs P2O5 (P Index score = 115) than

PAGE 234

234 applying manure at 225 lbs P2O5 (P Index score = 178). Reduction of RP from high to medium rating gave the greatest reducti on in P score (32). However, th e cost of reducing RP can be prohibitive. Using biosolids instead of manure as a management practice could be easier, more economical, and reduces the score to as low as 115. Sensitivity of the Draft Florida P Inde x Model to P-Source and P-Source Rate The second hypothetical case examines the effe cts of different P-s ources and P-source application rates on the P Index score. Kirton ra nch field study data were used to study the sensitivity of the draft Florid a P Index model to P-source a nd PSAR. The field study involved applying four P-sources at two ra tes (P-based and N-based) to pl ots in a landscape with similar SE, RP, LP, PWB, FIV, and AM, with no wast e water applied (Table 10-4). The P-source coefficients in the draft Florid a Index and the PWEP-based coeffi cients recommended in Chapter 8 were each used to compute the P index scores. Applying manure or TSP at a P-based rate yiel ded a score of 83, and indicated the field had a medium P vulnerability category. However, the sa me field was categorized as low vulnerability when biosolids are applied at P-based rate. A pplying biosolids at N-based rates increases the field rating back into the medium category. If manure is applied at an N-based rate, the field reaches a very high vulnerability rating and reme dial action is required. Using the PWEP-based coefficients, the P Index values categorized P-ba sed rates of all organic sources of P to low P loss vulnerability. The low category rating is consis tent with low P harzards associated with the P-based rate and support the PWEP as a better coeffi cient than the source coefficients in the draft P Index. However, at the N-based rate, the scor es range from 77 (medium category) for Pompano biosolids to 177 (high category) for TSP. The se nsitivity analysis can serve as a very useful nutrient management tool for farmers, nutrient managers, and regulators.

PAGE 235

235 Table 10-4. Draft Florida P Index scor es of plots treated with four P-sources at two P rates in the field <----Manure---> Boca Raton biosolids Pompano biosolids <------TSP--------> Factors Variable P-based Nbased Pbased N-based Pbased N-based P-based N-based P rate (lbs P2O5) 80 573 80 646 80 556 80 180 P-source coefficient (FPSC) 0.05 0.05 0.015 0.015 0.015 0.015 0.05 0.05 P-source coefficient (PWEP-based) 0.046 0.046 0.055 0.055 0.012 0.012 0.084 0.084 P-source multiplier (FPSC) 4 28.7 1.2 9.7 1.2 8.3 4 9 P-source multiplier (PWEP) 1.44 10.3 0.96 7.8 0.32 2.2 6.72 15.1 FIV@ 7ppm M-1P 0.35 0.35 0.35 0.35 0.35 0.35 0.35 0.35 Application method (not incorporated) 6 6 6 6 6 6 6 6 P-source Waste water (not applied) 0 0 0 0 0 0 0 0 Soil Erosion 1 1 1 1 1 1 1 1 Runoff potential 2 2 2 2 2 2 2 2 Leaching potential 4 4 4 4 4 4 4 4 Transport Potential to reach water body 1 1 1 1 1 1 1 1 P Index value 83 280 60 128 60 117 83 123 P Index value (PWEP-based) 70 150 66 127 60 77 105 172 Generalized interpretation of P Index result Medium Very high Low MediumLow Medium MediumMedium Summary and Conclusions A sensitivity analysis of the Florida P Index id entified all nine variables in the model as important. The variables fell into either the medium (FIV, AM, WWA, and PWB) or high (PSA,

PAGE 236

236 PSAR, RP, LP, and SE) impact category. None of the variable s is redundant or impart low impacts. Experimental data are needed to asse ss the relative impacts of the nine variables on P loss. Study is also needed to evaluate the consis tency of the variable im pacts on P loss, and their impacts on P Index scores. Most variable categories are consistent with other state P-Indices (Pennsylvania and the original Lemunyon and Gilbert P Index). The PSA and PSAR are cat egorized as high impact variables, and SE, PWB, and FI V categorized as medium imp act variables by both Florida P Index and Pennsylvania P Index. Immediate research addressing high impact va riables is recommended. The wide range of P-sources available for land applic ation, which have been reported to differ in solubility and P losses, may be better accounted for by more than the current three ratings (0.1 for waste water, 0.05 for manure and TSP, and 0.015 for biosolids) . The use of PWEP-based coefficients recommended in Chapter 8 should be considered. Im pacts of WWA could also be integrated into the P-sources variable, rather than being repeated as a variable under WWA amounts. Other questions raised by the sensitivity analysis that need to be vali dated by experimental data are: Is soil erosion so important in Florida th at a wider swing should be assigned and should erosion be categorized as high impact variable , unlike in other P-Indices (Pennsylvania and the original Lemunyon and Gilbert P Index), wher e the variable is categorized as medium or low impact? Is application method, categorized as having onl y a medium impact in Florida, consistent with the high impact categorization in the original Lemunyon and Gilbert P Index? The use of continuous, rather than discrete ratings should be considered, where possible. Continuous variables provide smoother model output and avoid subjectivity inherent to assigning ratings to a particular variable (Elliott et al., 2006). Continuous ratings could be used for SE by multiplying the SE values in tons per acre by 0.533. The factor, 0.533 will give equivalent ratings (0 – 8) for erosio n between 0 and 15 tons acre-1 (as in Florida P Inde x) and greater ratings

PAGE 237

237 when erosion is 15 tons acre-1. Other variables that could be assigned continuous ratings are PSA (Elliott et al., 2006), and PWB. However, assigning continuous ratings to these variables should be based on data from studies on Florida soils.

PAGE 238

238 CHAPTER 11 MANAGEMENT OF PHOSPHORUS SOURCE S AND WATER TREA TMENT RESIDUALS (WTR) FOR ENVIRONMENTAL AND AGRONOMIC BENEFITS Co-application of WTR with different P-sources has poten tial as a BMP to reduce the environmental hazard associated with excess soil P in low P-sorbi ng coastal plain sands, without negative agronomic impact. Understanding how diffe rent P-sources, source a pplication rates, and WTR affects soil P loss and agronomic returns w ill enhance sound management of the wastes for agronomic and environmental benefits. Thus th e objective of the study was to evaluate the environmental and agronomic impacts of different P sources and WTR and to determine the rate of P-sources and WTR that optimize e nvironmental and agronomic benefits. Impacts of P-sources, source application rates, and WTR on P loss and availability to plants were evaluated in glass house and rainfall simulation stud ies and validated using results from a 2-year field study with similar treatments . Within a week of inco rporating the P-sources and WTR with a low P-sorbing sandy soil in the glasshouse study, the degree of P saturation (DPS) was reduced and, in most cases, was below the 25% threshold value suggested for Florida soils (Nair et al., 2004). The capacity of the soil to retain P and prevent P migration and loss as measured by soil P storage capacity (SPSC) valu es was also increased. Both DPS and SPSC values showed dramatic improvement in the P sorption property of the soil following amendment with WTR. Similar improvements in the soil DPS and SPSC values were observed in all the surface A-horizon samples obtained from the fiel d sampled periodically in the 2-year study. However, the impacts of the WTR on soil DPS and SPSC values depend on the application rates of the P-sources and WTR. Generally, greater amounts of WTR are required for N-based Psource application rates than for P-based applicati on rates to achieve similar sorption properties.

PAGE 239

239 239 Improved soil P-sorption was accompanied by reduced soil soluble P and groundwater P concentrations. Both WEP and ISP values of so il samples taken during the glasshouse study and samples obtained from the A-horizon in the field study were reduced in WTR treatments compared to untreated soils. Th e reduction in soil soluble P following addition of WTR was reflected in reduced losses of P in runoff and leachate in the rainfall simulation study. Also and most importantly, the P concentrat ions of groundwater samples obtained in the field were smaller in the WTR treatments than in untreated so ils. The impacts of surface-applied WTR were observed in groundwater samples taken both abov e and below the spodic horizon during the field study, regardless of P-source or so urce application rate, and P con centrations were less than those measured in the control treatments. The great er soil total Al concen trations of the Al-WTR treatments neither increased groundwate r Al nor plant Al concentrations. The P-sources and source application rates a ffected soil soluble P concentrations and P losses. Generally, greater soil WEP values and, he nce soluble P concentrations, were observed at N-based rates than P-based rates in the glass house study. Also, there were greater P losses from N-based treatments than P-based treatments in th e runoff, leachate, and TP (runoff + leachate) of each P-source in the absence of WTR. However, the P loss from the TSP treatments applied at a P-based rate was similar or gr eater (depending on the source) than the losses observed at Nbased rates of the organic sources of P. Thus, the P hazard from applying organic sources at Nbased rates was lower than, or si milar to, that observed at P-base d rate of the mineral P-source. Also, the P loss from applying a moderate water soluble P biosolids (Pompano biosolids) at Nbased rates was not greater than the P loss from a high water solubl e P organic source at P-based rates. Thus, P losses can be c ontrolled (without applying WTR) by using lower water soluble P materials. The collective results of the study s uggest that environmental P hazards associated

PAGE 240

240 240 with high application rates (N-bas ed rate) of P materials can be managed by either applying the P-source with WTR, or by using lower water solu ble P-sources. The masses of bioavailable P (BAP) and TP loss were similarly affected by the P-sources and followed the same trend as PWEP values of the sources (PWEP values in parentheses): TSP (84%) > manure (18%) > Boca Raton biosolids (12%) > Pompano biosolids (4 %). Regression of BAP loss with source application rate (r2 = 0.27) was improved by accounting for Psources solubility differences with the draft Florida P Index coefficients (r2 = 0.54), P-source coefficients (PSC) values suggested by Elliott et al. (2006) (r2 = 0.77), and PWEP values of the sources (r2 = 0.81). Use of a coefficient based on PWEP of the P-source is suggested as a means of differentiatin g P loss potentials of different P-sources, and is recommen ded as alternative to coefficien ts currently used in the draft Florida P Index. The three coefficients suggested in the draft Florida P Index (coefficient = 0.05 for fertilizer and manure, 0.015 for all biosolids, and 0.10 for waste water), were insufficient to account for the wide variability in the P-sources. Default values for PWEP-based coefficients for different types of P-sources are suggested. The agronomic impacts of the three manageme nts approaches identified to reduce P loss (improving the soil sorption capacity by applicati on of WTR, applying the P-sources at P-based rates, and use of lower water soluble P amendments) were test ed. The applied Al-rich WTR, neither increased the groundwater Al concentrati ons, nor increased plant Al concentrations. Data from both the field and the glasshouse studies in dicate no greater Al conc entrations in plants grown in WTR-amended soils than in unamende d soils. Amending soil with WTR could be a best management practice (BMP) to reduce th e hazard associated with excess P from landapplied mineral and organic sources of P, even at high P loads associated with N-based source

PAGE 241

241 241 application rates. The soil P hazard will be reduced, if not eliminated, without negative agronomic impacts such as reduced plants yields or Al toxicity. Applying WTR, even at 2.5%, to N-based rate s treatments did not reduce plant yields in most cases. However, at P-based rates of P-s ource application, WTR of more than 1% reduced yields. The plant P concentrations were reduced by application of WTR, but at the N-based source application rate, with WTR, the P con centrations were sufficient for optimum plant growth. The plant yields at N-ba sed rates with WTR were similar or greater than observed at the P-based rate of TSP without WTR in th e glasshouse and th e field studies. A recommendation to apply WTR on a fixed oven dry basis could result in negative agronomic and or environmental impact s depending on WTR and the Al, P and Fe concentrations of the P-sources . Application rates of WTR ba sed on a desired soil SPSC value ensure applying the amount of WTR needed to re duce excess P, while providing sufficient P for optimum plant growth. A SPSC value of zero (0 mg kg-1) was identified as the critical point, above which plant P concentrations can be reduced sufficiently to reduce plant yields, and below which the potential for P loss increases. Results from the glasshouse and fi eld studies show that the environmental hazard associated with excess P loads from N-based source applications are controlled by applicatio n of WTR without negative agronomic impacts. The P-sources differed in potential for P lo ss and relative P phytoava ilability (RPP). The RPP value of the moderate water soluble P, Pompano biosolids fell into the moderate phytoavailable biosolids categor y, whereas the high water solubl e P Boca Raton biosolids RPP fell into the high category along w ith TSP. The RPP value of th e moderate water soluble P, Pompano biosolids, and poultry manure also fell into the moderate rela tive phytoavailability category. Properties of manure that can account fo r the RPP could not be investigated because

PAGE 242

242 242 only one manure type (poultry manure) was used in this study. Results fro m the field study were consistent with the glasshouse studies in cl assifying P-sources into RPP categories. The organic sources of P varied in RPP values in the field in a similar manner as the values observed in the glasshouse study. The field study RPP values of the Pompano biosolids and poultry manure agreed with the expected moderate phytoavailable biosolids class determined in the glasshouse study, and the Boca Raton biosolids RPP values from both studies were classified as high. A method proposed by Quebec Canada regulatory agency (CRAAQ, 2003; MENV 2003), which estimate % P availability values of orga nic sources of P from an empirical equation that accounts for select P-sour ce characteristics, was not valid ated by either the greenhouse or the field data. Properties identified to account for the RPP values of biosolids are Total P concentration, NaOH-P and %solids. These properties could be th e focus of further study into estimating RPP values of biosolids from their prop erties. Also further studies with varying types of manure will be needed to identify prope rties that could account for manure RPP. Both the P-source and P transport potentials affect P loss from a watershed. The importance of the two factors on P loss is addr essed in the P-Indices being developed by 47 states in the US. Sensitivity anal ysis of the draft Florida P In dex was carried out to study the impacts of the variables and identify areas wh ere future studies on P losses should be focused. The sensitivity analysis indicated that all nine variables in the FL model are important and fell into either the medium (FIV, AM, WWA, a nd PWB) or high (PSR, RP, LP, PSA, and SE) impact categories. None of the variables is redundant or of low impact. The variable categories are also consistent with other states P-indi ces (Pennsylvania and the original Lemunyon and Gilbert P Index).

PAGE 243

243 243 Peculiar local conditions were well considered in the draft Florida P Index. However, contrary to widely-held expect ations of greater leaching pot ential in Florida soils, runoff potential was identified as more important. Th e importance of runoff likely results from the high water table conditions that characterizes many FL soils, and reduces the vertical movement (leaching). However, on a relative scale (compared to other states), leaching is also important in Florida soils. Recommendations were made for st udies into all the variab les in the P Index and the use of continuous ratings for the variables wh erever possible. Also the use of more than 3 variables to account for the wide spectrum of P-source solubilities is recommended, including source coefficients based on PWEP of the sources.

PAGE 244

244 APPENDIX A PUBLICATIONS ARISING FROM THE DISSERTATION Oladeji, O.O., G.A. O’Connor, and S.R. Brinton. Effects of water treatment residual (WTR) on runoff and leachate phosphorus losses in a Flor ida sand (In preparation for publication in Journal of Environmental Quality). Oladeji, O.O., J.B. Sartain, and G.A. O’Connor. (2006). Agronomic impact of land applied water treatment residuals (Submitted for publicati on in Soil Crop Science Society of Florida Proceedings). Oladeji, O.O., G.A. O’Connor, and H.A. Elliott. Basis for estimating P-sources coefficients for the Florida P Index (In preparation for publicat ion in Journal of Environmental Quality). Oladeji, O.O., J.B. Sartain, and G.A. O’Connor. Application rates of phos phorus (P) sources and water treatment residuals (WTR ) for agronomic and environmental benefits (In preparation for publication in Journal of Environmental Quality). Oladeji, O.O., J.B. Sartain, and G.A. O’Connor . Land application of water treatment residual (WTR): Agronomic P efficiency and alum inum phytotoxicity (In preparation for publication in Communications in So il Science and Plant Analysis). Oladeji, O.O., J.B. Sartain, and G.A. O’Connor. Evaluation of soil test methods for Florida sand treated with different P-sources and WTR (In preparation for publication in Communications in Soil Science and Plant Analysis). Oladeji, O.O., G.A. O’Connor, H.A. Elliott, and J. B. Sartain. Sensitivity analysis of Florida P Index (In preparation for publication in J ournal of Soil and Water Conservation). Oral Presentations Oladeji, O.O., G.A. O’Connor, and J.B. Sartain. 2006. Effect of water treatment residuals (WTR) on runoff and leachate phosphorus losses in a Florida sand. Presented at the 2006 Annual meeting of the Southern Branch of Ag ronomy Society of America in Orlando, FL. Oladeji, O.O., G.A. O’Connor, and J.B. Sartain. 2006. Basis for estimating P-source coefficients for the Florida P Index. Presented at Soil a nd Crop Science Society of Florida annual meeting 2006. Oladeji, O.O., J.B. Sartain, and G.A. O’C onnor. 2006. Application rate of water treatment residual (WTR) based on agro-environmental threshold. Presented at Soil and Crop Science Society of Florida annual meeting 2006. Oladeji, O.O., G.A. O’Connor, and J.B. Sart ain. 2006. A methodology to account for P release potential from different sources of P. Pres entation at Soil Scien ce Society of America annual meetings, 2006.

PAGE 245

245 Oladeji, O.O., J.B. Sartain, and G.A. O’Connor . 2005. Agronomic impact of land applied water treatment residuals: Soil test methods and appl ication rates. Presented at Soil and Crop Science Society of Florida annual meeting, 2005.

PAGE 246

246 APPENDIX B OTHER RELEVANT TABLES AND FIGURES Appendix Figure B-1. The pH of soil sa mples taken during the glasshouse study (a) June 04 samples0 1 2 3 4 5 6 7 8 ControlMan-PMan-NBoca-PBoca-NPomp-PPomp-NTSP-PTSP-N P source and their application ratepH 0% WTR 1% WTR 2.5% WTR (b) Dec. 04 samples0 1 2 3 4 5 6 7 8 ControlMan-PMan-NBoca-PBoca-NPomp-PPomp-NTSP-PTSP-N P source and their application ratepH 0% WTR 1% WTR 2.5% WTR (c) May 05 samples0 1 2 3 4 5 6 7 8 ControlMan-PMan-NBoca-PBoca-NPomp-PPomp-NTSP-PTSP-N P source and their application ratepH 0% WTR 1% WTR 2.5% WTR (d) Sept. 05 samples0 1 2 3 4 5 6 7 8 ControlMan-PMan-NBoca-PBoca-NPomp-PPomp-NTSP-PTSP-N P source and their application ratepH 0% WTR 1% WTR 2.5% WTR

PAGE 247

247 Appendix Figure B-2. The EC values (µs cm-1) of time zero and time final soil samples taken during the glasshouse study. (* significan t at 5%; NS is non significant at 5%) (a) June 04 samples 0 200 400 600 800 ControlMan-PMan-NBoca-PBoca-NPomp-PPomp-NTSP-PTSP-N P source and their application rateEC 0% WTR 1% WTR 2.5% WTR Contrast (P sources) N-based P-based Man vs Biosolids NS NS Organic vs Minerals P * NS Boca vs Pompano NS NS Contrast: ( N-based vs P-based) Manure * Boca * Pompano * TSP * Note: WTR did not affect the EC (b) Sept. 05 samples 0 200 400 600 800 ControlMan-PMan-NBoca-PBoca-NPomp-PPomp-NTSP-PTSP-N P source and their application rateEC 0% WTR 1% WTR 2.5% WTR Contrast (P sources) N-based P-based Man vs Biosolids NS NS Organic vs Minerals P * NS Boca vs Pompano NS NS Contrast: ( N-based vs P-based) Manure * Boca * Pompano * TSP * Polynomial effect of WTR Linear * Quadratic NS

PAGE 248

248 0 25 50 75 100 125 0246810121416 Planting periods (months)Relative P phytoavailability (%) TSP (Mineral P source) Boca Raton biosolids Pompano biosolids Poulry manure Appendix Figure B-3. Effects of time on relative P phytoavailability (RPP) of the different Psources during the glasshouse study. 0 1 2 3 4 5 6 7 8TSPHTSPH+ TSPLTSPL+ ManHManH+ ManLManL+ BocaHBocaH+ BocaLBocaL+ PompHPompH+ PompLPompL+ CNCN+TreatmentspH Leachate Runoff Appendix Figure B-4. Runoff and leachate pH values of the treatments across the three runs and replicates during the rainfall simulation experiment (n = 6, error bars represent one standard error)

PAGE 249

249 0 500 1000 1500 2000 2500 3000TSPHTSPH+ TSPLTSPL+ ManHManH+ ManLManL+ BocaHBocaH+ BocaLBocaL+ PompHPompH+ PompLPompL+ CNCN+TreatmentsEC µs cm-1) Leachate Runoff Sediment Ladden (Runoff) Appendix Figure B-5. Runoff and leachate EC values of the treatments across the three runs and replicates during the rainfall simulation experiment (n = 6, error bars represent one standard error)

PAGE 250

250 Appendix Table B-1. Elect ric conductivity (µs cm-3) values of soil samples taken during the glasshouse study. Sampling periods P-source Psource rate WTR rate (0ven dry %) June ‘04 Dec. ‘04 May ‘05 Sept. ‘05 Control --324 defg 211 cd 249 b 261 def 0 315 defg 265 abcd 258 b 279 def 1 348 bcdefg 323 abcd 324 ab 331 bcdef P based 2.5 380 abcdefg 394 abcd 245 b 340 bcdef 0 424 abcdefg 573 abc 317 ab 438 abcdef 1 458 abcdef 646 a 518 ab 540 ab Manure N-based 2.5 528 a 599 ab 776 a 634 a 0 310 defg 301 abcd 223 b 278 def 1 270 g 258 bcd 153 b 226 f P based 2.5 323 defg 301 abcd 407 ab 343 bcdef 0 489 abc 318 abcd 587 ab 464 abcde 1 439 abcdef 325 abcd 534 ab 432 abcdef Boca Raton Biosolids N-based 2.5 499 ab 557 abc 523 ab 526 abc 0 301 fg 189 cd 323 ab 269 def 1 308 efg 228 bcd 205 b 247 ef P based 2.5 333 cdefg 348 abcd 342 ab 344 bcdef 0 418 abcdefg 415 abcd 300 ab 377 bcdef 1 465 abcde 546 abc 392 ab 467 abcd Pompano Biosolids N-based 2.5 472 abcd 491 abcd 291 ab 418 abcdef 0 336 bcdefg 257 bcd 261 b 284 def 1 299 fg 387 abcd 232 b 306 def P based 2.5 311 defg 302 abcd 187 b 266 def 0 391 abcdefg 159 d 314 ab 288 def 1 360 bcdefg 422 abcd 179 b 320 cdef TSP N-based 2.5 362 bcdefg 282 abcd 620 ab 376 bcdef Means (n = 3) of treatments during the same sampling period follow by the same letter are not different at 5% significance level

PAGE 251

251 Appendix Table B-2. Bahiagrass and ryegrass yields, P concentrations, and P uptake valu es from different treatments during the glasshouse study. First Bahiagrass Ryegrass Second Bahiagrass P-source Rate WTR ( %) Dry Matter (Mg ha-1) P concn. (g kg-1) P uptake (kg ha-1) Dry Matter (Mg ha-1) P concn. (g kg-1) P uptake (kg ha-1) Dry Matter (Mg ha-1) P concn. (g kg-1) P uptake (kg ha-1) Control 5.66 ± 0.25† 1.94 ± 0.12 11.0 ± 0.7 2.49 ± 0.12 2.03 ± 0.15 5.07 ± 0.50 3.27 ± 0.46 1.59 ± 0.11 5.24 ± 0.96 0 7.12 ± 1.18 3.03 ± 0.65 21.2 ± 2.4 3.38 ± 0.05 2.78 ± 0.13 9.41 ± 0.51 3.28 ± 0.07 2.65 ± 0.08 8.71 ± 0.45 1 6.35 ± 0.60 1.79 ± 0.19 11.4 ± 1.7 3.37 ± 0.18 2.11 ± 0.08 7.12 ± 0.57 3.21 ± 0.18 2.57 ± 0.15 8.20 ± 0.11 Pbased 2.5 6.08 ± 0.22 1.73± 0.04 10.5 ± 0.5 3.32 ± 0.21 1.82 ± 0.07 6.09 ± 0.61 3.02 ± 0.26 1.98 ± 0.16 5.93 ± 0.47 0 5.13 ± 0.85 3.84± 0.24 19.8 ± 4.1 4.50 ± 0.04 4.33 ± 0.09 19.4 ± 0.4 3.86 ± 0.31 2.76 ± 0.35 10.4 ± 0.71 1 4.39 ± 0.58 1.73± 0.41 7.75 ± 2.86 4.41 ± 0.19 2.72 ± 0.32 12.0 ± 1.6 4.29 ± 0.44 2.39 ± 0.20 10.3 ± 1.6 Manure Nbased 2.5 4.12 ± 0.66 1.87± 0.12 7.67 ± 1.03 4.23 ± 0.13 2.30 ± 0.26 9.77 ± 1.30 3.83 ± 0.08 2.50 ± 0.13 9.55 ± 0.35 0 6.77 ± 1.32 3.90± 0.20 26.2 ± 4.3 3.11 ± 0.12 2.94 ± 0.26 9.10 ± 0.54 3.59 ± 0.09 2.23 ± 0.10 7.98 ± 0.18 1 5.71 ± 0.79 1.83± 0.27 10.3 ± 0.5 2.82 ± 0.04 2.03 ± 0.11 5.72 ± 0.36 3.19 ± 0.08 2.63 ± 0.41 8.34 ± 1.16 Pbased 2.5 5.43 ± 1.18 1.77± 0.17 9.58 ± 1.85 2.93 ± 0.18 1.99 ± 0.21 5.87 ± 0.92 2.92 ± 0.10 1.99 ± 0.13 5.79 ± 0.31 0 8.17 ± 0.62 6.43± 0.12 52.6 ± 0.6 3.83 ± 0.12 5.98 ± 0.75 23.0 ± 3.2 5.56 ± 0.38 4.38 ± 0.55 24.5 ± 4.2 1 7.42 ± 0.92 3.60± 0.43 27.0 ± 6.6 4.12 ± 0.36 4.19 ± 0.39 17.5 ± 3.10 5.43 ± 0.35 3.45 ± 0.16 18.6 ± 0.6 Boca Biosolids Nbased 2.5 7.01 ± 1.06 2.80± 0.51 19.8 ± 5.9 4.22 ± 0.24 3.09 ± 0.45 13.1 ± 2.32 5.65 ± 0.39 3.26 ± 0.24 18.4 ± 1.8 0 6.51 ± 1.05 2.74± 0.14 17.8 ± 3.0 3.14 ± 0.08 2.69 ± 0.22 8.48 ± 0.85 3.52 ± 0.10 2.18 ± 0.23 7.62 ± 0.60 1 5.80 ± 0.51 1.82± 0.12 10.6 ± 0.6 3.08 ± 0.21 2.11 ± 0.05 6.49 ± 0.50 3.12 ± 0.20 2.01 ± 0.29 6.21 ± 0.87 Pbased 2.5 5.40 ± 0.39 1.67± 0.15 8.98 ± 0.70 2.94 ± 0.12 1.89 ± 0.19 5.58 ± 0.75 3.51 ± 0.01 1.95 ± 0.27 6.85 ± 0.92 0 6.86 ± 0.62 4.43± 0.48 30.2 ± 2.2 3.62 ± 0.17 5.13 ± 0.19 18.5 ± 0.9 4.81 ± 0.46 3.57 ± 0.24 17.0 ± 1.2 1 6.56 ± 1.38 2.47± 0.13 16.3 ± 3.8 3.83 ± 0.04 3.09 ± 0.30 11.8 ± 1.0 4.19 ± 0.17 3.20 ± 0.19 13.5 ± 1.2 Pompano Biosolids Nbased 2.5 6.54 ± 0.76 2.36± 0.08 15.4 ± 2.0 3.79 ± 0.16 2.67 ± 0.23 10.1 ± 1.2 3.86 ± 0.17 3.32 ± 0.32 12.7 ± 0.9 0 6.29 ± 0.37 3.60± 0.29 22.6 ± 1.3 3.07 ± 0.11 6.28 ± 3.75 8.87 ± 1.34 3.43 ± 0.34 2.06 ± 0.10 7.03 ± 0.53 1 5.77 ± 0.36 1.82± 0.17 10.5 ± 0.8 3.04 ± 0.10 2.20 ± 0.18 6.65 ± 0.36 2.85 ± 0.24 2.18 ± 0.20 6.16 ± 0.53 Pbased 2.5 4.82 ± 0.64 1.55± 0.19 7.38 ± 0.19 3.31 ± 0.28 1.90 ± 0.24 6.34 ± 1.07 2.64 ± 0.03 1.98 ± 0.16 5.23 ± 0.49 0 6.79 ± 1.18 5.31± 0.68 36.0 ± 7.1 3.28 ± 0.07 3.59 ± 0.43 11.7 ± 1.4 3.38 ± 0.40 2.68 ± 0.12 9.07 ± 1.19 1 5.67 ± 1.11 1.99± 0.23 11.1 ± 1.7 3.40 ± 0.20 2.57 ± 0.08 8.73 ± 0.54 2.50 ± 0.13 2.35 ± 0.19 5.83 ± 0.31 TSP Nbased 2.5 5.88 ± 0.94 1.95± 0.16 11.4 ± 0.9 3.23 ± 0.20 2.40 ± 0.10 7.77 ± 0.72 2.60 ± 0.51 1.95 ± 0.22 4.97 ± 0.86 † means (n = 3) + Standard deviation

PAGE 252

252 Appendix Table B-3. Varying measures of so il test P at planting of th e first bahiagrass crop (Jun e 2004), the ryegrass crop (De cember 2004), and the second bahiagrass crop(Ma y 2005) during the glasshouse study <----------------‡WEP----------------> <--------------------§ISP-----------------> <----------------¶M-1P-------------------> P-source P rate WTR rate (%) June 2004 Dec. 2004 May 2005 June 2004 Dec. 2004 May 2005 June 2004 Dec. 2004 May 2005 Control --†2.88±0.19 3.01 ± 0.33 1.14 ± 0.07 3.11 ± 0.48 3.87 ± 0.23 1.35 ± 0.22 6.40 ± 0.35 2.93 ± 0.58 2.24 ± 0.39 0 6.35±1.58 8.47 ± 0.56 1.82 ± 0.16 9.04 ± 0.68 8.16 ± 0.43 2.23 ± 0.61 15.8 ± 0.1 14.38 ± 0.35 5.93 ± 1.14 1 3.54±0.18 2.23 ± 0.26 1.21 ± 0.11 6.04 ± 0.36 4.91 ± 0.22 2.12 ± 0.20 23.7 ±5.3 14.70 ± 1.23 13.6 ± 0.6 Pbased 2.5 2.73±0.44 1.90 ± 0.12 0.97 ± 0.03 3.08 ± 0.72 4.37 ± 0.13 1.78 ± 0.20 29.7 ± 1.8 19.18 ± 1.22 20.2 ± 2.1 0 18.7±0.78 11.47 ± 1.17 4.99 ± 1.63 23.3 ± 1.8 25.52 ± 3.26 11.7 ± 3.8 63.5 ± 8.1 67.53 ± 2.35 54.0 ± 3.9 1 11.3±0.7 5.08 ± 1.07 2.83 ± 0.07 16.3 ± 0.6 13.09 ± 0.64 8.12 ± 0.79 69.5 ± 8.0 67.53 ± 3.58 59.6 ± 2.4 Manure Nbased 2.5 5.78±0.30 3.37 ± 1.00 2.88 ± 1.35 11.7 ± 0.3 9.26 ± 0.42 7.77 ± 2.19 69.3 ± 11.7 66.08 ± 2.16 57.7 ± 1.9 0 6.53±0.29 4.43 ± 0.83 2.06 ± 0.17 7.37 ± 0.83 8.00 ± 1.07 2.90 ± 0.30 23.7 ± 6.6 10.49 ± 1.73 6.92 ± 0.78 1 3.80±0.45 1.43 ± 0.32 1.15 ± 0.14 4.09 ± 0.53 5.24 ± 0.16 2.24 ± 0.11 24.8 ±7.6 14.97 ± 0.61 14.1 ± 0.2 Pbased 2.5 2.63±0.32 1.17 ± 0.34 0.97 ± 0.05 3.79 ± 0.47 4.28 ± 0.38 1.90 ± 0.11 31.8 ±7.1 18.36 ± 2.58 21.5 ± 1.2 0 41.4±2.3 21.73 ± 0.27 17.8 ± 1.5 52.4 ± 5.1 32.79 ± 3.79 26.0 ± 1.1 164 ±13 67.94 ± 4.79 55.7 ± 3.9 1 20.8±1.0 6.51 ± 0.44 3.19 ± 0.33 35.9 ±2.0 17.55 ± 0.83 10.4 ± 1.0 147 ±8 69.55 ± 2.35 64.5 ± 0.3 Boca Raton Biosolids Nbased 2.5 15.4±0.6 3.65 ± 0.41 1.74 ± 0.03 20.9 ±1.9 11.07 ± 0.47 11.8 ± 4.7 148 ±18 82.71 ± 7.24 72.3 ± 1.2 0 3.94±0.41 5.14 ± 1.39 2.05 ± 0.01 4.34 ± 0.84 6.33 ± 0.52 3.46 ± 0.52 16.6 ±1.7 8.77 ± 0.42 7.50 ± 1.01 1 2.59± 0.32 1.40 ± 0.29 1.03 ± 0.10 3.32 ± 0.18 4.54 ± 0.22 2.78 ± 0.83 20.5 ±1.6 11.87 ± 1.61 11.4 ± 0.8 Pbased 2.5 1.96± 0.05 1.26 ± 0.26 0.93 ± 0.05 2.72 ± 0.19 4.00 ± 0.10 2.12 ± 0.38 26.9 ±1.6 17.93 ± 1.20 16.0 ± 0.7 0 6.37± 0.40 9.04 ± 1.67 4.70 ± 0.58 11.9 ±0.6 15.05 ± 2.59 14.0 ± 4.6 66.5 ±4.9 48.75 ± 6.92 35.8 ± 5.0 1 3.77± 0.65 6.67 ± 1.50 1.49 ± 0.04 7.96 ± 0.13 8.38 ± 0.53 4.75 ± 0.72 64.3 ±12.3 49.55 ± 4.84 50.2 ± 6.3 Pompano Biosolids Nbased 2.5 2.63± 0.34 3.88 ± 1.36 1.20 ± 0.07 5.43 ± 0.18 6.08 ± 0.19 3.79 ± 0.51 65.2 ±7.2 52.19 ± 2.01 41.7 ± 3.8 0 12.2 ± 4.1 8.61 ± 0.40 1.81 ± 0.29 10.4 ± 1.11 7.43 ± 0.67 2.69 ± 0.40 23.5 ± 5.7 10.73 ± 1.60 5.01 ± 0.72 1 3.68± 1.75 4.60 ± 0.51 1.11 ± 0.06 4.47 ± 0.66 4.68 ± 0.24 1.78 ± 0.19 24.4 ± 6.5 11.63 ± 0.52 10.4 ± 0.3 Pbased 2.5 1.95± 0.51 4.67 ± 0.19 0.94 ± 0.15 3.02 ± 0.51 3.98 ± 0.13 1.80 ± 0.33 24.7 ± 6.9 15.08 ± 1.79 17.3 ± 2.7 0 19.4 ± 2.7 11.42 ± 1.29 4.16 ± 0.75 20.8 ± 0.5 11.16 ± 1.11 5.33 ± 0.39 46.8 ± 15.6 16.00 ± 0.41 11.2 ± 2.2 1 8.45 ±1.56 4.45 ± 0.84 1.34 ± 0.05 12.4 ± 1.5 5.54 ± 0.39 3.24 ± 0.28 37.6 ± 6.5 17.13 ± 0.60 17.1 ± 1.5 TSP Nbased 2.5 3.90± 0.40 4.19 ± 0.40 1.32 ± 0.41 6.97 ± 0.6 4.83 ± 0.43 8.10 ± 6.29 41.2 ± 0.5 19.97 ± 3.67 16.3 ± 5.4 † means (n = 3) + Standard deviation ‡ Water extractable P § Iron strip P ¶ Mehlich 1P

PAGE 253

253 Appendix Table B-4. Degree of phosphorus saturati on (DPS) values of soil samples between June 2003 and December 2004 during the fi eld study. (All values are in %) Treatments June 2003 (0-5cm) Jan 2004 (0-5cm) Dec. 2004 (0-5cm) March 2004 (0-15cm) Dec 2004 (0-15cm) Controls ‡62 ab 52 ab 8 gh 180 a 79 abc Manure-N, no WTR 117 ab 95 a 33 cde 185 a 57 d Manure-P, no WTR 88 ab 55 ab 25 def 186 a 61 cd Manure-N, WTR 31 ab 18 ab 10 fgh 43 c 14 e Manure-P, WTR 15 b 14 ab 6 gh 30 c 14 e Boca-N, no WTR 94 ab 65 ab 76 a 43 c 93 a Boca-P, no WTR 50 ab 65 ab 38 bcd 78 bc 82 ab Boca-N, WTR 33 ab 19 ab 21 efg 26 c 13 e Boca-P, WTR 19 ab 41 ab 11 fgh 31 c 14 e Pompano-N, no WTR 137 a 86 ab 52 b 52 c 77 abcd Pompano-P, no WTR 57 ab 60 ab 37 bcde 58 c 72 bcd Pompano-N, WTR 13 b 13 b 7 gh 31 c 6 e Pompano-P, WTR 17 b 10 b 4 h 25 c 7 e TSP-N, no WTR 127 ab 86 ab 43 bc 211 a 72 bcd TSP-P, no WTR 79 ab 59 ab 42 bcd 158 ab 73 bcd TSP-N, WTR 21 ab 16 ab 9 gh 39 c 19 e TSP-P, WTR 24 ab 25 ab 14 fgh 42 c 18 e ‡Means (n = 3) of treatments during the same sampling period follow by the same letter are not different at 5% significa nt level by Tukey test.

PAGE 254

254 LIST OF REFERENCES Adjei, M.B., C.S. Gardner, D. Mayo, T. Seaw right, and E. Jennings. 200 0. Fertilizer treatment effects on forage yield and quality of tropical pasture grasses. Soil Crop Sci. Soc. Florida Proc. 59:32-37. Agronomy Guide. 2002. College of Agricultural Sciences, Penn State University, University Park, PA. Alcordo, I.S., J.E. Rechcigl, and R.C. Littell. 2001. Greenhouse evaluation of lime and gypsum materials for reducing phosphorus leaching in a Spodosol. Final report to the South Florida Water Management District. (No. C-10201). Allen, L.H., Jr. 1988. Dairy-siting criteria and ot her options for wastewater management on high water-table soils. Soil Crop Sci. Soc. Florida Proc. 47:108-127. American Society of Civil Engineers, Ameri can Water Works Association and the USEPA. 1996. Technology transfer handbook: Management of water treatment plant residuals. ASCE and AWWA, New York. Anderson, D.L., O.H. Tuovinen, A. Faber, and I. Ostrokowski. 1995. Use of soil amendments to reduce soluble phosphorus in dairy soils. Ecol. Eng. 5:229-246. Andraski, T.W. and L.G. Bundy. 2003. Relationship between phosphorus levels in soil and in runoff from corn production system s. J. Environ. Qual. 32:310-316. Ann, Y., K.R. Reddy, and J.J. Delfino. 2000. Influence of chemical amendments on phosphorus immobilization in soils from a constructed wetland. Ecol. Eng. 14:157-167. AOAC. 1995. Official method of AOAC international. Chapter 3 Fertilizer. In Cunniff, P. (ed.) Official analytical chemists in ternational, Arlington, Virginia. APHA/AWWA/WEF. 1995. Standard methods for th e examination of water and wastewater (19th edition). A.D., L.S. Clesceri, and A.E. Greenberg (eds.). pp 2-58. Arkin, G.F., and Taylor, H.M. 1981. Modifying the root environment to reduce crop stress. ASAE Monograph No. 4, American Society of Agricultural Engin eers, St. Joseph, MI. Arthington, J.D. 2002. Essential trace minerals for grazing cattle in Florida. Document AN086 of series of the Animal Science Department, Florida cooperative exte nsion service, IFAS University of Florida (Available at http://edis.ifas.ufl.edu/BODY_AN086 ) (verified Oct. 2006) Barnett, G.M. 1994. Phosphorus forms in animal manure. Bioresource Tech. 49:139-147.

PAGE 255

255 Basta, N.T., and D.E. Storm. 1997. Use of al um residuals to reduce nutrient runoff from agricultural land treated with animal manures to protect surface wate r quality. p. 14-28. In Proc. of water residuals and biosolids management:Approaching the year 2000. Water Environ. Federation/Am. Water Works Assoc. Jo int Specialty Conf., Philadelphia, PA. 3-6 Aug.1997. Water Environment Federation, Arlington, VA. Basta, N.T., R.J. Zupancic and E.A. Dayton. 2000. Evaluating soil tests to predict Bermuda grass growth in drinking water treatment residual with phosphorus fertiliz er. J. Environ. Qual. 29:2007-2012. Bates, T. E. 1971. Factors affecting critical nutrient concentrations in plants and their evaluation:A review. Soil Sci. 112:116-130. Beauchemin, S. and R.R.Simard. 1999. Soil phos phorus saturation degree:Review of some indices and their suitability for P management in Quebec, Canada. Can. J. Soil Sc. 79:615625. Beauchemin, S., R.R. Simard, and D. Cluis. 1996. Phosphorus sorption-desorption kinetics of soil under contrasting land uses. J. Environ. Qual. 25:1317-1325. Beegle, D.B., J.L. Weld, W.J. Gburek, J.A. Kleinman, A.N. Sharpley, and C. Kogelmann. 2005. The Pennsylvania phosphorus index version 1. User documentation. Available on the PA Nutrients Managements web site: http://panutrientmgmt.cas.ps u.edu/pdf/rp_p_index_guidance1005.pdf (verified Oct. 2006) Beek, J. 1979. Phosphate retention by soil in relation to waste disposal. PhD. Thesis. p.162 Agric. Univ., Wageningen, The Netherlands. Boesch, D.F., R.B. Brinsfield, and R.E. Magnien. 2001. Chesapeake Bay eutrophication: Scientific understandi ng, ecosystem restoration, and challe nges for agriculture. J. Environ. Qual. 30:303-320. Bogdan, A.V. 1977. Tropical pasture and fodder plants. Longman Inc., New York. pp. 205-212. Bohn, H.L., B.L. McNeal, and G.A. OÂ’Connor. 1985. Soil chemistry. 2nd ed. John Wiley & Sons, New York. Bolland, M.D.A., R.J.Gilkes. 1990. Cultivation redu ces fertilizer residual effectiveness and affects soil testing for availabl e phosphorus. Fert. Res. 24: 33-46. Brady, N.C., and R.R. Weil. 2002. The nature and properties of soils. 13th ed. Prentice Hall, New Jersey. Brandt, R.C., and H.A. Elliott. 2003. Phosphorus runoff losses from surface-applied biosolids and dairy manure. In: Proceedings of the 2003 Joint Residuals and Biosolids Management Conference. Water Environment Federati on. February 19-22, 2003, Baltimore, Maryland.

PAGE 256

256 Brandt, R.C., and H.A. Elliott. 2005. Sensitivity analysis of the Pennsylvania phosphorus index for agricultural recycling of municipal biosolids. J. Soil Water Conser. 60:209-219. Brandt, R.C., H.A. Elliott, and G.A. OÂ’Connor. 2004. Water extractable phosphorus in biosolids: Implications for land-based recyc ling. Water Environ. Res. 76:121-129. Breeuwsma, A., and S. Silva. 1992. Phosphorus fe rtilization and environm ental effects in the Netherlands and the Po Region (Italy). Re p. 57. Agric. Res. Dept., The Winand Staring Centre for Integrated Land, Soil, and Water Res., Wageningen, the Netherlands. Breeuwsma, A., J.G.A. Reijerink, and O.F. Schoumans. 1995. Impact of manure on accumulation and leaching of phosphate in area s of intensive livestock farming. p. 239251. In K. Steele (ed.) Animal waste and the la nd-water interface. Lewis Publ.-CRC, New York. Brown, E., and J.B. Sartain. 2000. Phosphorus rete ntion in United States Golf Association (USGA) greens. Soil Crop Sci. Soc. Florida Proc. 59:112-117. Bugbee, G.J., and C.R. Frink. 1985. Alum sludge as a soil amendment: Effects on soil properties and plant growth. Bull. 823. Connec ticut Agric. Exp. Stn., New Haven. Burgoa, B., A.B. Bottcher, R.S. Mansell, and L.A. Allen. Jr. 1991. Distributions of residual soil phosphorus along transacts for three dairies in Okeechobee County. Florida. Soil Crop Sci. Soc. Fla. Proc. 50:137-144. Burgoa, B., R. S. Mansell, and D. Rhue. 1990. Effects of solids concentration upon nonequilibrium phosphorus sorption in aqueous soil suspensions. Soil Crop Sci. Soc. Fla. Proc. 49:60-66. Cate, R.B.,Jr., and L.A. Nelson. 1971. A simple st atistical procedure fo r partitioning soil test correlation data into two classes. Soil Sci. Soc. Am. Proc. 35:658-660. Chang, A.C., A.L. Page, F.H. Sutherland, and E. Grgurevic, 1983. Fractionation of P in sludge affected soils. J. Environ. Qual. 12:286-290. Chardon, W. J., R. G. Menon, and S. H. Chie n.1996. Iron oxide impregnated filter paper (Pi test): A review of its development and me thodological research. Nutr. Cycl. Agroecosys. 46:41-51. Coale, F. J., J. T. Sims, and A.B. Leytem. 2002. Accelerated deployment of an agricultural nutrient tool:The Maryland phosphorus site index. J. Envi ron. Qual. 31:1471-1476. Coale, F. J., T. Basden, D. B. Beegle, R. C. Bra ndt, H. A. Elliott, D. J. Hansen, P. Kleinman, G. Mullins, and J. T. Sims. 2005. Development of regionally-consistent phosphorus source coefficients for use in phosphorus index ev aluations in the Mid-Atlantic region MidAtlantic Water Program MAWQP# 05-04.

PAGE 257

257 Codling, E.E., R.L. Chaney, and C.L. Mulchi. 2 002. Biomass yield and P av ailability to wheat grown on high phosphorus soils amended with p hosphate inactivating residues.I. Drinking water treatment residue. Commun. So il Sci. Plant Anal. 33:1039-1061. Collins, M.E. 2003. Key to soil orders in Florida (SL-43). Fact sheets of the Soil and Water Science Department, Florida Cooperative Ex tension Service, Institute of Food and Agricultural Sciences, University of Florida. Cornell Recommends for Field Crops. 1992. New York State College of Agriculture and life sciences, Cornell University, Ithaca, NY. Cox, A.E., J.J. Camberato, and B.R. Smith. 1997. Phosphate availability and inorganic transformation in an alum sludge-affect ed soil. J. Environ. Qual. 26:1393-1398. CRAAQ (Centre de référence en agriculture et agroalimentaire du Québec). 2003.Guide de référence en fertilisation. 1ère édition 300 p. Cromwell, G.L. 2002. Approaches to meeting the animal’s phosphorus requirements. p.61-76. In Proc. Minnesota Nutri. Conf ., Minneapolis-St. Paul, MN. Crouse, D.A., R.L. Sierzputowska-Gracz, R.L. Mikkelsen and A.G. Wollum. 2002. Monitoring phosphorus mineralization from poultry manure using phosphatase assays and phosphorus31 nuclear magnetic resonance spectrosc opy. Commun. Soil Sci. Plant Anal. 33:12051217. Cullen, A.C., and H.C. Frey. 1999. Probabilistic techniques in exposure assessment. Plenum Press: New York. Daniel, T.C., A.N. Sharpley, and J. L. Lemunyon. 1998. Agricultural phosphorus and eutrophication: A symposium overview. J. Environ. Qual. 27:251-257. Dayton, E. A. and N. T. Basta. 2005. Use of dri nking water treatment residua l as a potential best management practice to reduce phosphorus ri sk index scores. J. Environ. Qual. 34:21122117. Dayton, E.A., N.T. Basta, C.A Jakober, J.A. Ha ttey. 2003. Using treatment residuals to reduce phosphorus in agricultural runoff. J. Am. Water Works Assoc. 95:151-158. de Hann, S. 1981. Sewage sludge as a phosphate fertilizer. In T. W. G. Hucker & G. Catroux, (eds.), Phosphorus in sewage sludge and anim al waste slurries: proceedings of the EEC Seminar organised jointly by the CEC and the Institute for Soil Fertility, Haren (Gr.) and held at Groningen, Netherlands, 12-13 June 1980, D. Reidel Publishing Co., Boston, Mass. DeLaune, P.B., P.A. Moore, Jr., D.K. Carman, A. N. Sharpley, B.E. Haggard, and T.C. Daniel, 2002. Using the phosphorus index to reduce phosphorus runoff from soils in the Eucha/Spavnaw watershed. USDA-Agricult ural research service project No.445. Department of Crop, Soil, and Environmen tal Sciences. University of Arkansas. Fayetteville, Arkansas.

PAGE 258

258 Dempsey, B.A., J. DeWolfe, D. Hamilton, Y . Lee, R. Liebowitz, and H.A. Elliott. 1989. Land application of water plant sludges. p. 537-543. In Proc. 44th Purdue Industrial Waste Conf., Purdue Univ., West Lafayette, IN. 9-11 May 1989. Le wis Publ., Chelsea, MI. Dentel, S. K., and J.T. Mah. 2002. Chemical tr eatment effects on phosphorus availability in sludges. In H. H. Hahn, E. Hoffman and H. Od egaard, (eds.), Chemical Water and Wastewater Treatment VII, IWA Publishing, London, pp. 339-348. Dowdy, R.H., R.E. Larson, and E. Epstein. 1976. Sewa ge sludge and effluent use in agriculture. p. 118-153. In Land application of waste material s. Soil Conserv. Soc.Am. Ames, Ankeny, Iowa. Duo, Z., D.T. Galligan, R.D. Allshouse, J.D. Toth, C.F.Ramberg, Jr., and J.D. Ferguson. 2001. Manure sampling for nutrient analysis: Variabil ity and sampling efficacy. J. Environ. Qual. 30:1432-1437. Effler, S.W . , S.M. O'Donnell, D.A. Matthews, C.M. Matthews , D.M. O'Donnell, M.T. Auer, and E.M. Owens. 2002. Limnological and loadi ng information and a phosphorus total maximum daily load (TMDL) analysis fo r Onondaga Lake. Lake Reserv. Manag. 18:87108 Eghball, B., and J.E. Gilley. 2001. Phosphorus ri sk assessment index evaluation using runoff measurements. J. Soil Water Conser. 56:202-206. Eghball, B., B.J. Wienhold, J.E. Gilley, and R.A. Eigenberg. 2002. Mi neralization of manure nutrients. J. Soil Water Conser. 57:470-473. Eghball, B., B.J.Wienhold, B.L. Woodbury, and R.A. Eigenberg. 2005. Plant availability of phosphorus in swine slurry and cattle feedlot manure. Agron. J. 97:542-548. Eghball, B., G.D. Binford, and D.D. Baltenspe rger. 1996. Phosphorus movement and adsorption in a soil receiving long-term manure and fertilizer applicat ion. J. Environ. Qual. 25:13391343. Elliott, H.A., and B.A. Dempsey. 1991. Agronomic e ffects of land application of water treatment sludges. J. Am. Water Works Assoc. 83:126-131. Elliott, H.A., and L.M. Singer. 1988. Effect of water treatment sludge on growth and elemental composition of tomato shoots. Comm un. Soil Sci. Plant Anal. 19:345-354. Elliott, H.A., B.A. Dempsey, D.W. Hamilton, and J.R. DeWolfe. 1990. Land application of water treatment sludges: Impact and mana gement. Am. Water Works Assoc. Res. Foundation, Denver. Elliott, H.A., G.A. O'Connor, and S. Brint on. 2002a. Phosphorus leaching from biosolidsamended sandy soils. J. Environ. Qual. 31:681-689.

PAGE 259

259 Elliott, H.A., G.A. O'Connor, P. Lu, and S. Brinton. 2002b. Influence of water treatment residual on phosphorus solubility and leaching. J. Environ. Qual. 31:1362-1369. Elliott, H.A., R. C. Brandt, P.J.A. Kleinma n, A.N. Sharpley, and D. B. Beegle. (2006). Estimating source coefficients for phosphorus site indices. J. E nviron. Qual. 35:2195-2201. Elliott, H.A., R.C. Brandt, and G.A. O'Connor. 2005. J. Environ. Qual. 34:1632-1639. El-Swaify, S.A., and W.W. Emerson. 1975. Changes in the physical propert ies of soil clays due to precipitated aluminum and iron hydroxides: I. Swelling and aggr egate stability after drying. Soil Sci. Soc. Am. Proc. 39:1056-1063. Eschenbach, T.G. 1992. Spider plots versus torna do diagrams for sensitivity analysis. Interfaces 22:40-46. Florida Department of Environmental Prot ection. 2002a. Department of environmental protection standard operating procedures for laboratory activities, DEP-SOP-002/01. Florida Department of Environmental Protecti on. 2002b. Requirements for field and analytical work . DEP-QA-002/02. Available at: http://www.doh.state.fl.us/lab/ PDF_Files/WaterCert/qa00202.doc (verified Oct. 2006) Gallimore, L.E., N.T. Basta, D.E. Storm, M.E. Payton, R.H. Hunke, and M.D. Smolen. 1999. Use of water treatment residuals to reduce nutrients in surface runoff from agricultural land. J. Environ. Qual. 28:1474-1478. Gartley, K. L., and Sims, J. T. 1994. Phosphor us soil testing: Envi ronmental uses and implications. Commun. Soil Sci. Plant Anal.25:1565-1582. Gascho, G.J., R.D. Wauchope , J.G. Davis, C. C. Truman, C.C. Dowler , J.E. Hook, H.R. Sumner, and A.W. Johnson. 1998. Nitrate-nitrog en, soluble, and bioavailable phosphorus runoff from simulated rainfall after fertilizer application. Soil Sci. Soc. Am. J.62: 17111718. Gburek, W.J., and A.N. Sharpley. 1998. Hydr ologic controls on phosphorus loss from upland agricultural watersheds. J. Environ. Qual. 27:267-277. Gerritse, R.G. 1981. Mobility of phosphorus from pig slurry in soils. p.347-369. In T.W.G. Hucker and G. Catroux (ed.) phosphorus in sewage sludge and animal waste slurries. Reidel, Dordrecht. The Netherland. Gerritse, R.G., and R. Vriesema. 1984. Phosphate di stribution in animal slurries. J. Agric. Sc. (Cambridge) 102:150-161.

PAGE 260

260 Graetz, D.A., V.D. Nair, W.G. Harris, G. Kidde r, K.L. Campbell, R.S. Mylavarapu, R.D. Rhue, S.P. Boetger, G.W. Hurt, W.G. Henderson, W. R. Reck, N. Watts, P.B. Deal, and W.D. Tooke. 2004. The Florida phosphorus index: A phosphorus risk assessment tool. 190FLAGRFH, 8/04. Hani, H, S.K. Gupta, and O.J.Furrer. 1981. Avai lability of phosphorus fractions in sewage sludge. In T. W. G. Hucker & G. Catroux, (eds.), Phosphorus in sewage sludge and animal waste slurries: proceedings of the EEC Se minar organised jointly by the CEC and the Institute for Soil Fertility, Haren (Gr.) a nd held at Groningen, Ne therlands, 12-13 June 1980, D. Reidel Publishing Co., Boston, Mass. Hanlon, E.A., G. Kidder, and B.L. McNeal. 1990. Soil, container media, and water testing: Interpretations and IFAS standardized fer tilization recommendati ons. Circ. 817. Florida Coop. Ext. Serv., Gainesville. Hanlon, E.A., J.S. Gonzalez, and J.M. Bartos. 1997. Mehlich 1 extractabl e P. Ca, Mg, Mn, Cu, and Zn. IFAS Extension Soil Testing Labor atory (ESTIL) and Analytical research Laboratory (ARL) Chemical proced ures and training manual. pp 18-20. Hansen, N.C., T.C. Daniel, A.N. Sharpley, a nd J.L. Lemunyon. 2002. The fate and transport of phosphorus in agricultural systems. J. Soil Water Conserv. 57:408-417. Harris, W.G., R.D. Rhue, G. Kidder, R.B. Brow n, and R.C. Littell. 1996. Phosphorus retention as related to morphology of sandy Coastal Plain so il materials. Soil Sci. Soc. Am. J. 60:15131521. Haustein, O.K., T.C. Daniel, D.M. Miller, P. A. Moore, Jr., and R.W. McNew. 2000. Aluminumcontaining residuals influence high-phosphorus soils and runoff water quality. J. Environ. Qual. 29:1954-1959. Heckrath, G., P.C. Brookes, P.R. Poulton, and W.T. Gooulding. 1995. P hosphorus leaching from soils containing different phos phorus concentrations in the Broadbalk experiment. J. Environ. Qual. 24:904-910. Hedley, M., and M.J. McLaughlin. 2005. Reaction of phosphate fertilizers and by-products in soils. In Sims et al. (eds). Phosphorus: Agri culture and the environment. pp. 181-252. ASA-CSA-SSA. Agronomy monograph no 46. Heil, D.M., and K.A. Barbarick. 1989. Water tr eatment sludge influence on the growth of sorghum-sudangrass. J. Environ. Qual.18:292-298. Higgs, B., A.E. Johnston, J.L. Salter, and C.J. Dawson. 2000. Some aspects of achieving sustainable phosphorus use in agricu lture. J. Environ. Qual. 29:80-87. Hurt, G.W., R.S. Mylavarapu, and S.P. Boetge r. 2006. UF/IFAS Nutrient management series: Computational tools for field implementa tion of the Florida phosphorus index. CIR 1337 Soil and Water Sci., Coop. Ext. Serv., IFAS, Gainesville, FL.

PAGE 261

261 Hylton, L.O. Jr., A. Ulrich, D.R. Cornelius, a nd K. Okhi. 1965. Phosphorus nutrition of Italian ryegrass relative to growth, moisture content and mineral constituents. Agronomy Journal 505-508 Ibrikci, H., E.A. Hanlon, and J.E. Rechcigl. 1992 . Initial calibration and correlation of inorganic –P soil test methods with a Bahiagrass fiel d trial. Commun. Plant Soil Anal. 23:2569-2579. Ibrikci, H., N.B. Comerford, E.A. Hanlon, and J.E. Rechcigl. 1994. Phosphorus uptake by Bahiagrass from spodosols: Modeling of uptake from different horizons. Soil Sc. Soc. Am. J. 58:139-143. Ippolito, J.A., K.A. Barbaric k, and E.F. Redente. 1999. Co-a pplication effects of water treatment residuals and biosolids on two range gra sses. J. Environ. Qual. 28:1644-1650. Izuno, F.T., C.A.Sanchez, F.J.Coale, A.B. Bottcher, and D.B. Jones. 1991. Phosphoruus concentrations in drainage waters in the Ev erglades agricultural area. J. Environ. Qual. 20:608-619. Jones, J.B. 1967. Interpretation of plant analysis for several agronomic crops. In G.W. Hardy et al. (eds). Soil test and plant analysis Part II. pp. 49-58. Soil Sci. Soc. Amer. Special Pub. No.2. Kamprath, E.J., and M.E. Watson. 1980. Conven tional soil and tissue tests for assessing phosphorus status of soils. p. 433-469. In F.E. Khasawneh et al. (ed.) The role of phosphorus in agriculture. ASA, CSSA, and SSSA, Madison. WI. Keeney, D.R., and L.M. Walsh. 1975. Guidelines fo r the application of waste water sludge to agricultural land in Wisconsin. Tech. Bull. 88:2-39. Wisconsin Dep. Nat. Resources, Madison. Kelling, K.A., L.M. Walsh, D.R. Keeney, J.A. Ryan, and A.E.Peterson. 1977. A field study of the agricultural use of sewage sludge: II. Eff ect on soil N and P. J. Environ. Qual. 6:345352 Kennedy, R. and G. Cooke. 1982. Control of lake phosphorus with aluminum sulfate: Dose determination and application techni ques. Water Resour. Bull. 18:389-395. Kick, H. 1981. Phosphorus balance after the application of sewa ge sludge. p. 255-270. In T.W.G. Hucker, and G. Catroux (ed.) Phosphorus in sewage sludge and animal waste slurries. Proc. EED seminar, Groningen, The Netherla nds. 112-13 June 1980. D. Reidel Publ. Co. Drodrecht, The Netherlands. Kidder, G., C.G. Chamblis, and R. Mylava rapu. 2002. UF/IFAS standard fertilization recommendations for agronomic crops. SL129. Soil and Water Sci., Coop. Ext. Serv., IFAS, Gainesville, FL. Kincheloe, S., R.A. Hoynum, F.J. Reed, P.J. Bergeau, H.D. Morris, H.F. Reetz, V.W. Case, and M.E. Sevey. 1987. Efficiency fertilizer use. Inter. Min. and Chem. Corp., Mundelein, IL.

PAGE 262

262 Kirkham, M.B. 1982. Agricultural use of phos phorus in sewage sludge. Adv. Agron. 35:129-163. Kleinman, P. J. A., A. M. Wolf, A. N. Sharple y, D. B. Beegle, and L. S. Saporito. 2005. Survey of water extractable phosphorus in livestoc k manures. Soil Sci. Soc. Am. J., 69:701-708. Kleinman, P.J.A., A.N. Sharpley, T.L. Veith, R.O. Maguire, and P.A. Vadas. 2004. Evaluations of phosphorus transport in surface runoff from packed soil boxes. J. Environ. Qual. 33:1413-1423. Kleinman, P.J.A., R.B. Bryant, W.S. Reid, A. N. Sharpley, and D. Pimentel. 2000. Using soil phosphorus behavior to identify environm ental thresholds. Soil Sci. 165:943-950. Leinweber, P. 1996. Phosphorus frac tions in soils from an area with high density of livestock population. Z. Pflanzeneraehr. Bodenkd. 159:251-256. Leinweber, P., R. Meissner, K.U. Eckhardt, a nd J. Seeger. 1999. Management effects on forms of phosphorus in soil and leaching losses. Eur. J. Soil Sci. 50:413-424. Lemunyon, J.L., and R.G. Gilbert. 1993. The con cept and need for a phosphorus assessment tool. J. Prod. Agr. 6:483-496. Leytem, A.B., J.T. Sims, and F.J. Coale. 2004. Determination of phosphorus source coefficients for organic phosphorus sources: Laboratory studies. J. Environ. Qual. 33:380-388. Lin, S. 1988. Effects of alum sludge app lication on corn and soybeans. p. 299-309. In Environ. Eng.: Proc. of the Joint CSCE-ASCE Natl. Conf., Vancouver, BC, Canada. 13-15 July 1988. Can. Soc. of Civil Eng., Montreal, in c onjunction with Environ. Canada and the Univ. of British Columbia. Liu, F.H., C.C. Mitchell, J.W. Odom, D.T. H ill, E.W. Rochester. 1997. Swine lagoon effluent disposal by overland flow: Effects on fora ge production and uptake of nitrogen and phosphorus. Agron. J. 89: 900-904. Lucas, J.B., T.A. Dillaha, R.B. Reneau, J.T. Novak, and W.R. Knocke. 1994. Alum sludge land application and its effect on plant growth. J. Amer. Water Works Assoc. 86:75-83. Maguire, R.O., and J.T. Sims. 2002. Soil testing to predict phosphorus leaching. J. Environ. Qual. 31:1601-1609. Maguire, R.O., J.T. Sims, and F.J. Coale. 2000. Phosphorus fractionation in biosolids-amended soils: relationship to solu ble and desorbable phosphorus. Soil Sci. Soc. Am. J. 64:20182024 Makris, K.C. 2004. Long-term stability of so rbed phosphorus by drinki ng-water treatment residuals: mechanisms and implications. PhD Di ssertation. pp. 9-90. University of Florida, Gainesville, Fl.

PAGE 263

263 Mallarino, A.P., and A.M. Blackmer. 1992. Compar ison of methods for determining critical concentrations of soil test phosphorus for corn. Agron. J. 84:850-856. Mansell, R. S., S. A. Bloom, and P. Nked i-Kizza. 1995. Phosphorus transport in Spodosols impacted by dairy waste. Ecol. Eng. 5:281-299. Mansell, R.S., S.A. Bloom, and B. Burgoa. 1991. Phosphorus transport with water flow in acid, sandy soils. p. 271-314. In : B. Jacob and M.Y. Corapciogl u (ed.) Transport processes in porous media. Kluwer Academic Publ., Dordrecht, the Netherlands. Matichenkov, V.V., D.V. Calvert, and E.A. Bo charnikova. 2001. Effect of Si fertilization on growth and P nutrition of bahiagrass. So il Crop Sci. Soc. Florida Proc. 60:30-36. Maurer, M. A., F. S. Davies, and D. A. Graetz. 1995. Reclai med wastewater irrigation and fertilization of mature ‘Redbl ush’ grapefruit trees on Spodosols in Florida. Journal of the Am. Soc. Hort. Sci., 120:394-402. McDowell, R.W., and A.N. Sharpley. 2001. Appr oximating phosphorus release from soils to surface runoff and subsurface drainage. J. Environ. Qual. 30:508-520. McKeague, J.A., J.E Brydon, and N.M. Miles, 1971. Differentiation of forms of extractable iron and aluminum in soils. Soil Sci. Soc. Am. Proc. 35:33-38. McLaughlin, M J., and L.Champion. 1987. Sewage sludge as a phosphorus amendment for sesquioxide soils. Soil Sci. 143:113-119. McLaughlin, M. J. 1988. A comparison between shor t-term chemical and biological methods to measure phosphorus availability in sludge-ame nded sesquioxide soils. South African J. Plant Soil. 5:137-141. Menon, R. G., S. H. Chien, and W. J. Chardon. 1997. Iron oxide impregnated filter paper (Pi test): II. A review of its applic ation. Nutr. Cycl. Agroecosys. 47:7-18. MENV (Québec, Ministère de l’ Environnement). 2003. Literature review and recommendations concerning the impact of iron and aluminum salts on soil phosphorus availability. Final report: Beneficial use of municipal bi osolids in agriculture – Available at http://www.weao.org/committees/biosolid s/pdf/Al%20_Fe_Final_Report%20_English_%2 0020603.pdf (verified 10 Jan. 2006). Moore, P.A., Jr., and D.M. Miller. 1994. Decrea sing phosphorus solubility in poultry litter with aluminum, calcium, and iron amendments. J. Environ. Qual. 23:325-330. Moore, P.A., Jr., T.C Daniel, and D.R. Edwards. 2000. Reducing phosphorus runoff and inhibiting ammonia loss from poultry manure w ith aluminum sulfate. J. Environ. Qual. 29:37-49. Morgan, M.G., and M. Henrion. 1990. Uncertainty : A guide to dealing with uncertainty in quantitative risk and policy analysis. Cambridge University press: Cambridge, NY.

PAGE 264

264 Mullins, G., B. Joern, and P. Moore. 2005. By-pr oduct phosphorus: sources, characteristics, and management. In Sims et al. (eds) Phosphorus: Agri culture and the environment. pp. 829879. ASA-CSA-SSA. Agronomy monograph no 46. Murphy, J., and J.P. Riley. 1962. A modified sing le solution method for the determination of phosphate in natural water. Anal. Chim. Acta 27:31-36. Myers, R.G., and G.M. Pierzynski. 2000. Using the iron oxide method to estimate bioavailable phosphorus in runoff. In: Methods of phosphorus analysis. G.M. Pierzynski. North Carolina State University. pp 98-102. Nair V.D, K.M. Portier, D.A, Graetz, and M. L. Walker. 2004. An envi ronmental threshold for degree of phosphorus saturation in sandy soils. J.Environ. Qual. 33:107-113 Nair, V. D., D. A. Graetz, and K. R. Re ddy. 1998. Dairy manure influences on phosphorus retention capacity of Spodosols. J. Environ. Qual. 27:522-527. Nair, V.D., and D.A. Graetz. 2002. Phosphorus sa turation in Spodosols impacted by manure. J. Environ. Qual. 31:1279-1285. Nair, V.D., and W.G. Harris. 2004. A capacity factor as an alternative to soil test phosphorus in phosphorus risk assessment. New Z ealand J. Agri. Res. 47:491-497. National Phosphorus Research Project. 2001. Nationa l research project for simulated rainfallsurface runoff studies. Available at http://www.sera17.ext.vt.edu/Docu ments/National_P_protocol.pdf (verified 10 Jan. 2006). Naylor, L.M., and J.S. Carr. 1997. Exchangeab le phosphorus in soils amended with water treatment residuals, biosolids, and aluminum-rich residues. In Proc. Conf. Water residuals biosolids management: Approaching the ye ar 2000. Water Envir on. Fed., Philadelphia, PA. Nilsson, S I, and B. Bergkvist. 1983. Aluminum chemistry and acidification processes in a shallow podzol on the Swedish West Coast. Water Air Soil Pollu. 20:311-330. Novak, J.M., and D.W. Watts. 2004. Increasing the phosphorus sorption capacity of southeastern Coastal Plain soils using water trea tment residuals. Soil Sci. 169:206-214. Novak, J.M., and D.W. Watts. 2005. An alum-b ased water treatment residual can reduce extractable phosphorus concentrations in three phosphorus-enriched coas tal plain soils. J. Environ. Qual. 34:1820-1827. Novak, J.M., D.W. Watts, P.G. Hunt, K. C. Stone, and M.H. Johnson. 2000. Phosphorus movement through a coastal plain soil af ter a decade of inte nsive swine manure applications. J. Environ. Qual. 29:1310-1315. OÂ’Connor, G.A., and H.A. Elliott. 2001. Co-app lication of biosolids and water treatment residuals. Final report. FL Dept . Environ. Protection. (No. S 3700 206296).

PAGE 265

265 OÂ’Connor, G.A., D. Sarkar, H.A. Elliot, a nd D.A. Graetz. 2000. Characterizing forms solubilities, bioavailabilities and minerali zation rates of P in biosolids, commercial fertilizers and manures, WERF report. 99-PUM-2T. pp. 1-77. OÂ’Connor, G.A., D. Sarkar, S.R. Brinton, H.A. E lliott, and F.G. Martin. 2004. Phytoavailability of biosolids-phosphorus. J. Environ. Qual. 33:703-712. OÂ’Connor, G.A., H.A. Elliott, and P. Lu. 2002a . Characterizing water treatment residuals phosphorus retention. Soil Crop Sci. Soc. Florida Proc. 61:67-73. OÂ’Connor, G.A., H.A. Elliott, D. Sarkar, a nd D.A. Graetz. 2002b. Characterizing forms, solubilities, bioavailabilities and minera lization rates of phosphorus in biosolids, commercial fertilizers and manures. Interim report submitted to Water Environment Research foundation. OÂ’Connor, G.A., S. Brinton, and M.L. Silvei ra. 2005. Evaluation and selection of soil amendments for field testing to reduce P losse s. Soil Crop. Sci. Soc. Florida Proc. 64:2234. O'Connor, G.A., and D. Sarkar. 1999. Fate of land applied residua ls-bound phosphorus. DEP WM 661. Florida Environ. Protection Agency, Tallahassee. OMEE. 1996 Guidelines for the utilization of biosolids and other wastes on agricultural land OMEE and OMAFRA, Toronto. Available at http://www.gov.on.ca/MBS/english/government/ogm.html . (verified Oct. 2006) Page, N.R., G.W. Thomas, H.F. Perkins, and R. D. Rouse. 1965. Procedures used by state soil testing laboratories in the southern region of the United States. Southern Cooperative Series Bull. No. 102. Pais, I., and J.B. Jones. 1997. Handbook of Trace Elements pp 1-77 Parsons, L. R., T. A. Wheaton, and W. S. Castle . 2001. High application rates of reclaimed water benefit citrus tree growth and fruit production. HortSci. 36:1273-1277 Paulter, M.C., and J.T. Sims. 2000. Relations hips between soil test phosphorus, soluble phosphorus, and phosphorus saturation in Delawa re soils. Soil Sci. Soc. Am. J. 64:765773. Penn, C.J., and J.T. Sims. 2002. Phosphorus forms in biosolids-amended soils and losses in runoff: Effects of wastewater treatmen t process. J. Environ. Qual. 31:1349-1361. Peperzak, P., A.G. Caldwell, R.R. Hunziker , and C.A. Black. 1959. Phosphorus fractions in manures. Soil Sci. 87:293-302. Peters, J.M., and N.T. Basta. 1996. Reduction of excessive bioavailable phosphorus in soils by using municipal and industrial wa stes. J. Environ. Qual. 25:1236-1241.

PAGE 266

266 Peterson, A.E., P.E. Speth, R.B. Corey, T.H. Wri ght, and P.L. Schleet. 1994. Effect of twelve years of liquid digested sludge applic ations on soil phosphorus levels. pp. 237-248. In C.E. Clapp et al. (ed.) Sewage sludge: Land utiliz ation and the environment. ASA, CSSA, and SSSA, Madison, WI. Pierson, S.T., M.L. Cabrera, G.K. Evanylo, H.A. Kuykendall, C.S. Hoveland, M.A. McCann, and L.T. West. 2001. Phosphorus and ammonium concentrations in surface runoff from grasslands fertilized with broiler litter. J. Environ. Qual 30:1784-1789. Pierzynski, G.M. 1994. Plant nutrient as pects of sewage sludge. p. 21-27. In C.E. Clapp et al. (ed.) Sewage sludge: Land utilization a nd the environment. SSSA, Madison WI. Plank, C.O. 1992. Plant analysis reference proce dures for the southern region of the United States. Southern Coop. Ser. Bull. 368. Pote, D.H., and T.C. Daniel. 2000a. Analysis of dissolved reactive phosphor us in water samples. p 91-93. In: Pierzynski, G.M. (eds). Methods of phosphorus analysis for soils, sediments, residuals, and waters. Southern cooperative series bulletin no 396. A publication of SERAIEG 17 (A USDA-CSREES Regional comm ittee minimizing agricultural phosphorus losses for protection of the water resources). Pote, D.H., and T.C. Daniel. 2000b. Analysis of total phosphorus and total dissolved phosphorus in water samples. p 94-97 In: Pierzynski, G.M. (eds). Methods of phosphorus analysis for soils, sediments, residuals, and waters. S outhern cooperative series bulletin no 396. A publication of SERA-IEG 17 (A USDA-CS REES Regional committee minimizing agricultural phosphorus losses for prot ection of the water resources). Pote, D.H., T.C. Daniel, A.N. Sharpley, P.A. M oore, Jr., D.R. Edwards, and D.J. Nichols. 1996. Relating extractable soil phosphorus-to-phosphorus losses in runoff. Soil Sci. Soc. Am. J. 60:855-859. Pote, D.H., T.C. Daniel, D.J. Nichols, A.N. Shar pley, P.A. Moore, Jr., D.R. Edwards, and D.M. Miller. 1999. Relationship between phosphorus levels in three ultisols and phosphorus concentrations in runoff. J. Environ. Qual. 28:170-175. Pritchard, D.L. 2005. Phosphorus bioavailability fr om land-applied biosolids in south-western Australia. A PhD thesis. pp 100-147 Cur tin University of Technology, Bentley Provin, T. L. 1996. Phosphorus retention in I ndiana soils. Ph.D. thesis. pp 222. Purdue University, West Lafayette, IN. Rechcigl, J.E, G.G. Payne, A.B. Bottch er, and P.S. Porter. 1992. Reduced phosphorus application on bahiagrass and wa ter-quality. Agron. J. 84: 463-468. Reddy, K.R., M.R. Overcash, R. Khaleel, and P.W. Westerman. 1980. Phosphorus adsorptiondesorption characteristics of two soils utilized for disposal of animal wastes. J. Environ. Qual. 9:86-92.

PAGE 267

267 Rengasamy, P., J.M. Oades, and T.W. Hancoc k. 1980. Improvement of soil structure and plant growth by addition of alum sludge. Co mm. Soil Sci. Plant Anal. 11:533-545. Sartain, J.B. 1979. Mobility and extractability of phosphorus applied to the surface of Tifway bermudagrass turf. Soil Crop Sci. Soc. Florida Proc. 39:47-50. Sartain, J.B. 2005. Soil fertility (SOS 6136) c ourse note. pp. 58. Soil and Water Science Department, University of Florida. SAS Institute. 1989. SAS/STAT user's guide. Version 6. 4th ed. SAS Inst., Cary, NC. Sauer, T.J., T.C. Daniel, P.L.A. Moore, Jr., K.P. Coffee, D.J. Nichols, and C.P. West. 1999. Poultry litter and grazing animal waste eff ects on runoff water quality from tall fescue plots. J. Environ. Qual. 28:860-865. Schoumans, O.F. 2000. Determination of the degr ee of phosphate saturation in non-calcareous soils. p. 31-34. In: Pierzynski, G.M. (ed). Methods of phosphorus analysis for soils, sediments, residuals, and waters. Southe rn cooperative seri es bulletin no 396. A publication of SERA-IEG 17 (A USDA-CS REES Regional committee minimizing agricultural phosphorus losses for prot ection of the water resources). Self-Davis, M.L., P.A. Moore, and B.C. Joern. 2000. Determination of wate rand/or dilute saltextractable phosphorus. In Methods for phosphorus analys is. G.M. Pierzynski (ed). Sharpley, A.N. 1991. Soil phosphorus extracted by Fe-Al-oxide-impregnated filter paper. Soil Sci. Soc. Am. J. 55:1038-1041. Sharpley, A.N. 1993a. An innovative approach to estimate bioavailable phosphorus in agricultural runoff using iron oxide-impregna ted paper. J. Environ. Qual. 22:597-601. Sharpley, A.N. 1993b. Assessing phosphorus bioavailabi lity in agricu ltural soils and runoff. Fert. Res. 36:259-272. Sharpley, A.N. 1995. Dependence of runoff phosphorus on extractable soil phosphorus. J. Environ. Qual. 24:920-926. Sharpley, A.N. 1997. Rainfall frequency and nitrogen and phosphorus in runoff from soil amended with poultry litter. J. Environ. Qual. 26:1127-1132. Sharpley, A.N., and B. Moyer. 2000. Phosphorus forms in manure and compost and their release during simulated rainfall. J. Environ. Qual. 29:1462-1469. Sharpley, A.N., J.L. Weld, D.B. Beegle, P.J.A. Kleinman, W.J. Gburek, P.A. Moore, Jr., and G. Mullins. 2003. Development of phosphorus indices for nutrient management planning strategies in the United States . J. Soil Water Conserv. 58:137-144.

PAGE 268

268 Sharpley, A.N., R. W. McDowell, J.L. Wel d, and P.J.A. Kleinman. 2001. Assessing site vulnerabaility to phosphorus loss in an agri cultural watershed. J. Environ. Qual. 30:20262036. Sharpley, A.N., T.C. Daniel, J.T. Sims, J. Lemunyon, R. Stevens, and R. Parry. 1999. Agricultural Phosphorus and Eutrophication. USDA-ARS. National Technical Information Serv. (NTIS), Port Royal Rd., Springfield, VA. Shober, A.L., and J.T. Sims. 2003. Phosphorus rest rictions for land application of biosolids: Current status and future tre nds. J. Environ. Qual. 32:1955-1964. Shreve, B.R., P.A. Moore, Jr., T.C. Daniel, D. R. Edwards, and D.M. Miller. 1995. Reduction of phosphorus in runoff from field-applied poultr y litter using chemical amendments. J. Environ. Qual. 24:106-111. Silveira, M.L., M.K. Miyittah, and G.A. OÂ’ Connor. 2006. Phosphorus release from a manure impacted spodosol: Effects of water treatmen t residuals. 2006. J. Environ. Qual. 35:529541. Simard, R.R., Cluis, D., Gangbazo, G., and Beau chemin, S. 1995. Phosphorus status of forest and agricultural soils from a watershed of high animal density. J. Environ. Qual. 24:10101017. Sims, J.T. 2000. The role of soil testing in environmental risk assessment for phosphorus. p.5781. In A.N. Sharpley (ed.) Agriculture and phosphorus management: The Chesapeake Bay. CRC Press, Boca Raton, FL and Lewis Publ., New York. Sims, J.T., and D.C. Wolf. 1994. Poultry waste management: Agriculture and environmental issues. Adv. Agron. 52:1-83. Sims, J.T., F. Coale, G. Evanylo, J.White, and A. Leytem. 2003. Field and on-farm evaluation of the effects of biosolids on phosphorus in soils and runoff. Final Project Report. Metropolitan Washington Council of Governments. Washington, D.C. Sims, J.T., R.R. Simard, and B.C. Joern. 1998. Phosphorus loss in agricultural drainage: Historical perspective and current research. J. Environ. Qual. 27:277-293. Smith, S.R., D.M. Bellett-Travers, R. Morris, and J.N.B. Bell. 2002. Fertilizer value of enhanced-treated and conventiona l biosolids products. Paper 10. In Proc. CIWEM Conf.Biosolids: The risk and benefits, an upda te on the latest research. London. 9 Jan. 2002. Chartered Inst. of Water and Environ. Manage., London. Soil Survey Staff. 1996. Keys to soil taxonomy. U.S. Gov. Print. Office, Washington DC. Sommers, L.E. 1977. Chemical composition of sewage sludges and analysis of their potential use as fertilizers. J. Environ. Qual. 6:225-232.

PAGE 269

269 Sommers, L.E., and A.L. Sutton. 1980. Use of waste materials as sources of phosphorus. p. 515545. In F.E. Khasawneh et al. (eds.) The role of phosphorus in agricu lture. ASA, CSSA, and SSSA, Madison. WI. South Florida Water Management District. 2003. Land applicati on of residuals and chicken manure in the Lake Okeechobee watershed: Phosphorus considerations. C-11653 Quarterly progress report no 1 (May-July, 2003). Stanley, R.L., and F.M. Rhoads. 2000. Bahiagrass production, nutrient up take, and soil-test P and K. Soil Crop Sci. Soc. Fl. Proc. 59:159–163. Steenbjerg, F. 1951. Yield curves and chemical plant analysis . Plant and Soils 3:97-109. Stehouwer, R.C., A.M. Wolf, and W.T. Dotie. 2000. Chemical monitoring of sewage sludge in Pennsylvania: Variability and application uncertainty. J. Environ. Qual. 29:1686-1695. Sui, Y., M.L. Thompson, and C. Shang. 1999. Fractionation of phosphorus in a Mollisol amended with biosolids. Soil Sci. Soc. Am. J. 63:1174-1180. Sweeten, J.M. 1992. Livestock and poultry wast e management: A national overview. p.4-15. In J. Blake et al. (ed.) Nationa l livestock, poultry and aquacu lture waste management. Am. Sov. Agric. Eng., St. Joseph, MI. Turner, B.L., and M.P. Haygarth. 2000. Phosphorus forms and concentrations in leachate under four grassland soil types. Soil Sci. Soc. Am. J. 64:1090-1099. Ulrich, A. and F.J. Hills. 1967. Principl es and practices of plant analysis. In G.W. Hardy et al., (eds). Soil test and plant analysis Part II. pp. 11-24. Soil Sci. Soc. Amer. Special Pub. No.2. USDA. 1996. Role of plants in waste managements. In Agricultural waste managements field handbook. AWMFH, rev. 1, July 1996 pp. 6-1 to 6-24. USEPA. 1986. Test methods for evaluating so lid waste. 3rd ed. EPA/530/SW-846. . NTIS, Springfield, VA.Document 955-001-0000-1. Superi ntendent of Documents, U.S. Gov. Print. Office, Washington, DC. USEPA. 1995. Process design manual. Land app lication of sewage sludge and municipal septage. EPA / 625/R-95/001, Office of res earch and development, Cincinnati, OH. USEPA. 1999. Unified national strategy for an imal feeding operations. March 9, 1999. U.S. Government Printing Office, Washington, D.C. USEPA. 2000. The quality of our nation's wate r. A summary of the national water quality inventory: 1998 Report to Congress. EPA841-S-00-001. USEPA, Washington, DC. Vadas, P. A., and P. J. A. Kleinman. 2006. Eff ect of methodology in estimating and interpreting water-extractable phosphoru s in animal manures. J. Environ. Qual., 35:1151-1159.

PAGE 270

270 Van der Zee, S.E.A.T.M., L.G.J. Fokkink, a nd W.H. van Riemsdjil. 1987. A new technique for assessment of reversibly ad sorbed phosphate. Soil Sci. Soc. Amer. J. 51:599-604. Weld, J.L., D.B. Beegle, J.T. Sims, J.L. Lemuny on, A.N. Sharpley, W.J. Gburek, F.C. Coale, and A.B. Leytem. 2000. Comparison of state appr oaches to the development of phosphorus indices. p. 329. In 2000 Annual meetings abstracts. AS A, CSSA, and SSSA. Madison, WI. Wolf, A.M., and D.E. Baker. 1985. Criteria for land spreading of sludge s in the northeast: Phosphorus. Pages 39-41 In D.E. Baker, D.R. Bouldin, H.A. Elliott, and J.R. Miller (eds.) Criteria and recommendations fo r land application of sludges in the northeast. Penn. Agric. Exp. Stn. Res. Bull. No851. The Pennsylvani a State University, University Park. Wolf, A.M., P. J. A. Kleinman, A. N. Shar pley, and D. B. Beegle. 2005. Development of a water extractable phosphorus test for manure: An interlaboratory study. Soil Sci. Soc. Am. J. 69:695 700. Zvomuya, F., B.L. Helgason, F.J. Larney, H.H. Janzen, O.O. Akinremi, and B.M. Olson. 2006. Predicting phosphorus availability from soilapplied composted and non-composted cattle feedlot manure. J. Environ. Qual., 35:928-937. Young, T.C., and J.V. Depinto. 1982. Algal-availability of partic ulate phosphorus from diffuse and point sources in the lower greatlakes basin. Hydrobi ologia 91: 111-119.

PAGE 271

271 BIOGRAPHICAL SKETCH Olawale O. Oladeji was born in Igangan, O yo state, Nigeria. He received a BSc in chemistry (1988), MSc in analytical chemistry (1993), and MSc in agronomy (soil science; 2002) all from University of Ibadan. He has worked as a teacher in high schools, as a quality control chemist in an industry, and as research laboratory supe rvisor. Olawale resigned his position as laboratory supervisor at the International Institut e of Tropical Agriculture for graduate study at the University of Florida in August 2003. He jo ined University of Florida, Department of Soil and Water Science, for doctoral study in soil chemistry and nutrient management under the supervision of Drs. George A. OÂ’Connor and Jerry B. Sartain in Fall 2003 and is scheduled to graduate in Fall 2006.