, . .37.. 8r 1d . ~43. , r? 1.? is. wwnfl. 5. .o. w J... v. . .MIm \Jf hmw. u. .am In. . ti an. «Wu , $0 ”3...: “Tim... M... 4 t! . wit... 2:? .3 in” 3.. ._.,_ ‘ sl a : #3; gr :7 '1? g: {is .1922; 1 . L5,. .. , 21.5 diet l. '4 , V..l in, . 359?. V . 1... i. A Z a? {1.7. . .1 k... .H a : .6 II! is? 5.. 14‘1t In VI). in . l ‘ 1.... Ilia; .ltvflnll. fag fihWLuufWH 4 It. SF?!» I: la}. a: ‘41.! p This is to certify that the thesis entitled PHOSPHORUS SORPTION AND AVAILABILITY FROM MANURES FOR SELECTED MICHIGAN SOILS presented by Sarah Kathleen Marshall has been accepted towards fulfillment of the requirements for the MS. degree in Crop and Soil Sciences (24,.w4flzm Major Professor’s Signature Jagged 3; gas 1/ Date MSU is an Affirmative Action/Equal Opportunity Institution LIBRARY ‘ Michigan State University PLACE IN RETURN BOX to remove this checkout from your record. To AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE "“5 991.9223 9 5 H 6/01 c:/ClRC/DateDue.p65-p.15 PHOSPHORUS SORPTION AND AVAILABILITY FROM MANURES FOR SELECTED MICHIGAN SOILS By Sarah Kathleen Marshall A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Crop and Soil Sciences 2004 ABSTRACT PHOSPHORUS SORPTION AND AVAILABILITY FROM MANURES FOR SELECTED MICHIGAN SOILS By Sarah Kathleen Marshall A method was developed to determine if inorganic P (Pi) and total P (Pt) soIption to soil was different when potassium phosphate (KHzPO4), dairy slurry, or swine slurry was used as the P source. Sorption isotherms were constructed and compared. Pi sorption capacities for all soils decreased in the following order: dairy slurry > KHzPO4 2 swine slurry. Overall, Pi sorption strength of either manure slurry was less than KH2P04. For all soils, Pt sorption capacity for either manure slurry was greater than KH2P04 and sorption strength for dairy and swine slurries was reduced or unchanged. Desorption of Fe and Al from soil occurred when dairy and swine slurries were used as the P source; however, no apparent desorption occurred for KH2P04 solutions, this may be caused by organic acids in the manure solutions. The difference in Pi sorption between dairy slurry and KHzPO4 was moderately correlated to desorbed Fe and A1 for five soils however, this correlation did not exist for swine slurry and KH2P04. In an incubation experiment, soil was amended with dairy slurry, swine slurry, or fertilizer. Overall, incubation with any amendment reduced sorption capacity and strength for at least three soils. Relative P availability from the different P sources increased as follows: dairy < swine s fertilizer. The results from these studies suggest that P sorption characteristics depend on the P source and P availability coefficients for specific manure sources should be developed to provide better guidance when developing nutrient management recommendations. ACKNOWLEDGMENTS I would like to thank my advisor, Dr. Carrie A. M. Laboski, for persuading me and giving me the opportunity to pursue a Master’s degree. I am gratefiIl for your continual support and guidance while helping me attain the research goals set forth. Special thanks to Dr.’s Brian Teppen and Bill Northcott for serving on my committee and providing assistance throughout my degrees at Michigan State University. I would like to especially thank Dr. Brian Teppen for his assistance with data analysis. I would like to extend endless thanks to the members of the Soil Fertility Laboratory — you have all made this experience more enjoyable with your smiles, laughter and words of wisdom and encouragement. Special thanks to Cal Bricker, Brian Long and Gary Zehr for their technical assistance in countless times of desperation. I would like to especially thank Emily Sneller, Randy Opperman, and Gordon Shetler, for their continuous assistance, support, and entertainment. To my roommate Amy Guza thank you so much for always having a listening ear followed by great advice — without you I am not sure I would have made it through some days. I would like to extend my special thanks to present and past students and staff who have been near and dear to me and who have assisted me in one way or another to reach my goals. Finally, to my family and friends for their support, without it I would not be where I am today. A special thanks to my parents, Bob and Louise, and my brother Aaron and his wife Jamie and daughter Brenna, for always believing in me and supporting me through the good and not so good times during this journey. iii TABLE OF CONTENTS LIST OF TABLES vii LIST OF FIGURES ix CHAPTER 1 1 A method to determine inorganic phosphorus sorption from manure to soil Abstract 1 Introduction 3 Methods and Materials 5 Soil and Manure .............................................................................................................. 5 Sorption Procedure Determination ................................................................................. 7 Inorganic Phosphorus Sorption Experiment ................................................................. 10 Results and Discussion 12 Comparison of Matrices ................................................................................................ 12 Inorganic Phosphorus Sorption Experiment ................................................................. 12 Conclusions 21 References 23 CHAPTER 2 25 Sorption of inorganic and total phosphorus from dairy slurry, swine slurry, and potassium phosphate to selected Michigan soils Abstract 25 Introduction 27 Methods and Materials 31 Soil and Manure ............................................................................................................ 31 Phosphorus Sorption Isotherms .................................................................................... 33 Iron and Aluminum Solution Concentrations ............................................................... 36 Data Analysis ................................................................................................................ 36 Results and Discussion 38 Iron and Aluminum Solution Concentrations ............................................................... 38 Phosphorus Sorption Capacity ...................................................................................... 38 Phosphorus Sorption Strength ....................................................................................... 49 Organic Phosphorus Sorption ....................................................................................... 50 vi Conclusions 51 References 56 CHAPTER 3 59 Manure application influences on phosphorus sorption and availability of selected Michigan soils Abstract 59 Introduction 61 Methods and Materials 64 Soils and Manures ......................................................................................................... 64 Incubation ..................................................................................................................... 64 Laboratory Analysis ...................................................................................................... 66 Data Analysis ................................................................................................................ 67 Results and Discussion 68 Phosphorus Sorption Capacity ...................................................................................... 68 Phosphorus Sorption Strength ....................................................................................... 73 Phosphorus Availability ................................................................................................ 74 Conclusions 79 References 81 APPENDIX A 85 Determination of optimum rate of chloroform for microbial suppression in sorption solutions where manure is the P source APPENDIX B 94 Determination of effect of filtration method on inorganic P solution concentration APPENDIX C 97 Percent removal of inorganic P and total P from manure sorption solutions by centrifugation APPENDIX D 105 Determination of the optimum amount of potassium persulfate for complete total P digestion of manure slurry APPENDIX E 109 Relationship between Fe and Al in solution and desorbed Fe and Al APPENDIX F 117 Relationship between desorbed Fe or Al and the difference in inorganic P or total P sorption for slurry and KHzPO4 vi LIST OF TABLES Table 1.1. Soil characteristics and classification ................................................................. 6 Table 1.2. Characterization of dairy and swine manure slurries .......................................... 7 Table 1.3. Experimental determination of background electrolyte concentration required for sorption solutions when manure slurries were diluted with various concentrations of KCl ............................................................................................................................. 8 Table 1.4. Concentration of inorganic P (Pi) in each sorption solution when dairy slurry, swine slurry, and KH2P04 were used as P sources. ..................................................... 11 Table 1.5. Comparison of Langmuir parameters for standard and adjusted P sorption matrices ........................................................................................................................ 13 Table 1.6. Effect of inorganic P source on Langmuir parameters ..................................... 17 Table 2.1. Classification and characterization of selected surface soils ............................ 32 Table 2.2. Characterization of dairy and swine manure slurries ........................................ 33 Table 2.3. Phosphorus sorption solution characterization ................................................. 34 Table 2.4. Effect of P source on Langmuir parameters for selected soils ......................... 39 Table 2.5. Correlation of desorbed Fe or Al to the difference between inorganic P (Pi) sorption capacity from each slurry and KHZPO4. ........................................................ 47 Table 2.6. Correlation of desorbed Fe or A1 to the difference between total P (Pt) sorption capacity from each slurry and KH2P04. ...................................................................... 48 Table 3.1. Soil characterization and classification ............................................................. 65 Table 3.2. Nutrient characterization of dairy and swine manure slurries .......................... 65 Table 3.3. Effect of manure and fertilizer application on soil sorption capacity (b) and sorption strength (k) as estimated by the Langmuir equation ...................................... 69 Table 3.4. Relative availability of P from dairy and swine slurries compared to fertilizer ............................................................................................................................................ 75 Table A.1. Comparison of the ability of various rates of chloroform to suppress microbial activity for the 24 h equilibration period ..................................................................... 88 vii Table A2. Comparison of the efficacy of 0.16 ml of chloroform to 25 ml of solution to suppress microbial growth when added to different soils and P sources ..................... 90 Table B.1. Mean inorganic P concentration in solution after 24 h equilibration period with Oshtemo soil followed by one of three filtration treatments ............................... 95 Table C. 1. Effect of centrifugation on inorganic P (Pi) solution concentration ............... 100 Table C .2. Effect of centrifugation on total P (Pt) solution concentration ....................... 101 Table C.3. Percent inorganic P (Pi) and total P (Pt) removal by centrifugation for varying solution concentrations. For a given P source and P component (Pi and Pt), removal is compared over the range of solution concentrations .............................................. 101 Table C.4. Percent inorganic P (Pi) of total P (Pt) for centrifuged and uncentrifuged samples ....................................................................................................................... 102 Table D. 1. Average percent total P (Pt) recovery, of three replications, using 0.7 g potassium persulfate ................................................................................................... 107 Table E.1. Mean of four replications Fe and Al solution concentrations after equilibration with soil when KH2P04 was the P source .................................................................. 110 viii LIST OF FIGURES Figure 1.1. Sorption isotherms for CaClz and KC] sorption solution matrices for the Capac and Colwood 1 soils when KHZPO4 was used as the source of P ........................... 14 Figure 1.2. Sorption isotherms for CaClz and KCl sorption solution matrices for the Colwood 2 and Oshtemo soils when KHzPO4 was used as the source of P ................ 15 Figure 1.3. Sorption isotherms for CaClz and KC] sorption solution matrices for the Parkhill and Spinks soils when KH2P04 was used as the source of P. ........................ 16 Figure 1.4. Inorganic P (P,) sorption isotherms for the Capac and Colwood 1 soils when various concentrations and sources of P (dairy slurry, swine slurry, KH2P04) were equilibrated with soil in a 0.1 M KCl, pH 6 matrix ..................................................... 18 Figure 1.5. Inorganic P (Pi) sorption isotherms for the Colwood 2 and Oshtemo soils when various concentrations and sources of P (dairy slurry, swine slurry, KH2P04) were equilibrated with soil in a 0.1 M KCl, pH 6 matrix ............................................. 19 Figure 1.6. Inorganic P (Pi) sorption isotherms for the Parkhill and Spinks soils when various concentrations and sources of P (dairy slurry, swine slurry, KH2P04) were equilibrated with soil in a 0.1 M KCl, pH 6 matrix ..................................................... 20 Figure 2.1. Inorganic P (Pi) and total P (Pt) sorption isotherms for Capac soil ................. 40 Figure 2.2. Inorganic P (Pi) and total P (Pt) sorption isotherms for Colwood 1 soil ......... 41 Figure 2.3. Inorganic P (Pi) and total P (Pt) sorption isotherms for Colwood 2 soil ......... 42 Figure 2.4. Inorganic P (Pi) and total P (Pt) sorption isotherms for Oshtemo soil ............. 43 Figure 2.5. Inorganic P (Pi) and total P (Pt) sorption isotherms for Parkhill soil .............. 44 Figure 2.6. Inorganic P (Pi) and total P (Pt) sorption isotherms for Spinks soil ................ 45 Figure 2.7. Organic P (Po) sorption isotherms for Capac and Colwood l soils ................. 52 Figure 2.8. Organic P (Po) sorption isotherms for Colwood 2 and Oshtemo soils ............ 53 Figure 2.9. Organic P (Po) sorption isotherms for Parkhill and Spinks soils ..................... 54 ix Figure 3.1. Sorption isotherms for Colwood 1 and Colwood 2 soils for non-incubated, non-amended (PRE) and amended incubation treatments ........................................... 70 Figure 3.2. Sorption isotherms for Oshtemo and Parkhill soils for non-incubated, non- arnended (PRE) and amended incubation treatments .................................................. 71 Figure 3.3. Sorption isotherms for Spinks soil for non-incubated, non-amended (PRE) and amended incubation treatments ................................................................................... 72 Figure 3.4. Relative increase in Bray l-P soil test level per P applied during a six week incubation of selected Michigan soils with fertilizer, swine slurry, and dairy slurry. For a given soil, P sources with the same letter are not significantly different (or=0.05) ........... 75 Figure E. 1. Relationship between Fe and A1 in solution and desorbed Fe and Al for dairy and swine slunies for Capac soil. Relationship parameters are provided when significant ................................................................................................................... 111 Figure E.2. Relationship between Fe and Al in solution and desorbed Fe and Al for dairy and swine slurries for Colwood 1 soil. Regression parameters are provided when significant ................................................................................................................... 1 12 Figure E.3. Regression between Fe and A1 in solution and desorbed Fe and Al for dairy and swine slurries for Colwood 2 soil. Regression parameters are provided when significant ................................................................................................................... 113 Figure E.4. Relationship between Fe and A1 in solution and desorbed Fe and Al for dairy and swine slurries for Oshtemo soil. Regression parameters are provided when significant ................................................................................................................... 1 14 Figure E.5. Relationship between Fe and A1 in solution and desorbed Fe and Al for dairy and swine slurries for Parkhill soil. Regression parameters are provided when significant ................................................................................................................... 1 15 Figure E.6. Relationship between Fe and Al in solution and desorbed Fe and Al for dairy and swine slunies for Spinks soil. Regression parameters are provided when significant ...................................................................................................................... 1 16 Figure F. 1. Correlation of desorbed Fe or Al and the difference in P, sorption for manure slurries and KH2P04 for Capac soil where r is the correlation coefficient and P is the significance of the correlation .................................................................................... 118 Figure F2. Correlation of desorbed Fe or Al and the difference in Pi sorption for manure slurries and KHzPO4 for Colwood 1 soil where r is the correlation coefficient and P is the significance of the correlation .............................................................................. 119 Figure F3. Correlation of desorbed Fe or Al and the difference in Pi sorption for manure slurries and KH2P04 for Colwood 2 soil where r is the correlation coefficient and P is the significance of the correlation .............................................................................. 120 Figure F.4. Correlation of desorbed Fe or A1 and the difference in P, sorption for manure slurries and KH2P04 for Oshtemo soil where r is the correlation coefficient and P is the significance of the correlation .............................................................................. 121 Figure F .5. Correlation of desorbed Fe or Al and the difference in P, sorption for manure slurries and KH2P04 for Parkhill soil where r is the correlation coefficient and P is the significance of the correlation .............................................................................. 122 Figure F .6. Correlation of desorbed Fe or Al and the difference in Pi sorption for manure slunies and KH2P04 for Spinks soil where r is the correlation coefficient and P is the significance of the correlation .................................................................................... 123 xi Chapter 1 A method to determine inorganic phosphorus sorption from manure to soil Abstract Understanding the sorption characteristics of phosphorus (P) from manure and fertilizer is important in developing nutrient management recommendations that are environmentally sustainable. Most recent P sorption research has focused on the influence of past manure application on P sorption characteristics, while very little research has been conducted on the differences in sorption characteristics where manure is used as the P source in sorption solutions. The objective of this study was to develop a method to determine if inorganic P (Pi) sorption to soils is dependent upon the source of phosphorus: Potassium phosphate (KHzPO4), dairy slurry, or swine slurry. The six selected Michigan soils had no manure application history, soil test P < 30 mg kg", pH < 7, and had a range in soil texture. The sorption procedure developed deviates from the proposed procedure of Nair et a1. (1984) in that KH2P04 sorption solutions were prepared in a 0.1M KCl matrix at pH 6. Nonlinear regression analysis was used to compare the standard sorption procedure of Nair et a1. (1984) with the developed procedure. The two procedures, using KHZPO4 as the P source for sorption solutions, were similar for all soils. Nonlinear regression analysis was used to compare Pi sorption isotherms for the different P sources. For all soils, Pi sorption capacity is significantly greater for dairy slurry than swine slurry or KHZPO4. Pi sorption capacity for the swine slurry is significantly less than KH2P04 for four of the six soils. Pi sorption strength of both slurries was less than KH2P04; however, this trend was not always significant. Based on these results, it is expected that manure from different species may increase soil test P differently. Understanding how P sorption and subsequent agronomic and environmental P availability change with manure application is necessary for sound nutrient management recommendations. Introduction Phosphorus (P) sorption capacity and strength of a soil have been found to change with application of manure or fertilizer. Decreased P sorption capacity and P sorption strength have been found in soils treated with manure, effluents, or superphosphate (Mozaffari and Sims, 1994; Sharpley, 1996; Holford et al.,1997; Siddique and Robinson; 2003; Laboski and Lamb, 2004). Holford et a1. (1997) found that the magnitude of the decrease in sorption capacity and strength was dependent on the amount of manure or effluent applied over time. However, P sorption capacity has not always been found to decrease with manure application. Mozaffari and Sims (1994), Laboski and Lamb (2004), and Sharpley (1996) reported increases in P sorption capacity with manure application for at least one soil used in each study. The primary focus of recent P sorption research has been to determine if manure application can change soil P sorption characteristics. This has been done by constructing sorption isotherms after application of manure, either in an incubation study or in a field setting, using KH2P04 as the source of P in the isotherm characterization. Limited research has been conducted on P sorption where manure is the source of P in the sorption solutions. Bhat and O’Callaghan (1980) found that more inorganic P (Pi) sorbed to soil from solutions of pig slurry than KH2P04. However, the sorption solutions were not treated to inhibit microbial activity. Thus, it is likely that some of the observed increase in P sorption was actually microbial consumption of P. Additionally, the solution matrix, unbuffered 0.01 M KCl, had a low ionic strength and the pH was not buffered. This matrix may have permitted the pig slurry to control ionic strength and perhaps pH for some solutions over the range in P concentrations. A standard P sorption procedure was published by Nair et al. (1984) to provide a reproducible method that would allow for comparison of isotherm coefficients from different research. This procedure uses a range of low P concentration solutions, prepared from KHzPO4, in unbuffered 0.01 M CaClz. If this procedure is used when manure is the source of P in the sorption solutions, precipitation of calcium (Ca) phosphates would likely occur. Generally, manure has high Ca concentrations. The addition of Ca ions in the manure sorption solution matrix would likely cause the solution to be supersaturated with respect to calcium phosphate compounds. Furthermore, an unbuffered system would likely result in solutions with pH greater than 7, further increasing the possibility of calcium phosphate precipitation if the pH is not fixed. In addition, the ionic strength of the sorption solution matrix suggested by Nair et al. (1984) is likely less than the ionic strength of diluted manure slurries. If sorption solutions with manure as the P source are prepared in unbuffered 0.01 M CaClz, the ionic strength of the sorption solutions would increase as the concentration of P in solution increases because the ionic strength is being controlled by the amount of slurry used. Controlling ionic strength throughout the entire P concentration range of the sorption solutions is important because an increase in ionic strength has been shown to enhance P sorption (Ryden and Syers, 1975). The effects of ionic strength, cation species, and pH on P sorption have been studied previously. Ryden and Syers (1975) found that ionic strength and cation species both affected P sorption when four New Zealand soils were equilibrated for 40 h with P sorption solutions with varying cations and ionic strength. They suggest that if a rapid equilibration time is desired then a solution matrix of high ionic strength with a divalent cation should be utilized. To avoid precipitation of phosphate compounds when high P additions are used they suggest the use of a solution matrix of high ionic strength with cations of soluble phosphates such as sodium or potassium. The objectives of this study were to (i) determine a procedure to characterize how P, from dairy slurry and swine slurry sorbs to soil and (ii) compare how Pi sorption from dairy and swine slurries differ from the sorption of P from KHZPO4. The information obtained will aid in understanding the ability of soil to sorb P directly from manure and highlight potential implications for nutrient management. Methods and Materials Soil and Manure Six surface soils (0-0.15 In) used in this experiment were collected in southern, central Michigan and are described in Table 1.1. Selected soils had a range in soil texture, Bray l-P < 30 mg kg'l, and pH < 7. None of the selected soils had a history of manure application. These selection criteria were employed to find soils where initial desorption of P would be minimal and to minimize calcium phosphate precipitation. Soils were air dried, sieved (2 mm sieve), and characterized by measuring pH, organic matter, cation exchange capacity (CBC), and Bray l-P (Brown, 1998). Soil particle size was determined by the hydrometer method (Gee and Bauder, 1986) (Table 1.1). The liquid dairy slurry used in this experiment was collected from a short-term storage pit near the feeding facility. The pit was agitated before sample collection to insure sample homogeneity. The liquid swine slurry used in this experiment was collected from a slotted floor feeder house at time of cleaning. Once collected, manure .3888 owcmnoxolcomwmu + recs—new 2:253 288 3 v6 m 2 2. m6 5v 63:8 vacuum 35am . . . muaozumoccm 2:22 288 a as v _ N N am Nm N m n c .2228: .35388 €828 SSS—cer— =.§ m . . . ragga: 2.5 28:. S m w a N we m N v m .038“ £828 $88. 3:80 2:850 . . . zescmoucm 29C. 388 co? 0 w a E NL ov me _ o m m .958 .358 Same—-ofim N v _ U . . . =o=cmoccm 39p 23:. oo? o 2 v E NN cm Nv v m c m .0385. 6828 $8287th _ w _ U . . . became—O 3:3» 286 3 n Z 2“ Em Nm a N v 0 638388 6828 $8835..“ 2230 73 m8 73 £ch ................ c\o .............. .2 5m .80 Eu :5 2am 5%: ca: E Seagaso 85m :8 oanO In .mzom 83.5"... @8028 we coumutogofiano use coumoummmflu .: 03m... from each species was individually mixed and sieved (2 mm sieve) to facilitate pipetting by removing large residual particles. Manures were characterized by measuring percent moisture, percent solids, total N, total P, soluble salts (K, Ca, Mg, and Na), Al, and Fe (Peters, 2003) (Table 1.2). Manures were stored in polyethylene containers at 0 °C. Aliquots of manure from each species were thawed and stored at 4 °C as needed. Table 1.2. Characterization of liquid dairy and swine manure slurries. Species Moisture Solids N P K Ca Mg Na Al Fe % mg L'1 Dairy 95.3 4.7 3271 529 2228 1618 539 647 21 57 Swine 94.8 5.2 6542 1294 3810 1078 648 755 53 151 Sorption Procedure Determination The standard P sorption procedure of Nair et a1. (1984) used unbuffered KHzPO4 solutions in 0.01 M CaClz. The speciation model Phreeqc (Parkhurst, 1995) predicted that calcium phosphate compounds would precipitate at high P solution concentrations when manure chemistry was input into the model with the background solution matrix proposed by Nair et al. (1984) (unbuffered 0.01 M CaClz). To prevent precipitation, the sorption solution pH was fixed at the average soil pH of 6 and 0.1 M KCl was used as the background electrolyte. The electrolyte concentration was determined by comparing the conductivity of P solutions (0, 50, 75, and 100 mg P L'l) with dairy slurry or swine slurry as the P source that were prepared in 0.05, 0.07, or 0.1 M KCl (Table 1.3). The necessary molarity of KCl was determined to be 0.1 M KCl due to low variability of electrical conductivity between solutions prepared with 0 mg total P L'I up to 100 mg total P L'1 when manure was the P source. Table 1.3. Experimental determination of background electrolyte concentration required for sorption solutions when manure slurries were diluted with various concentrations of KCl. Manure Species Pf concentration Electrical Conductivity 0.05 M KCl 0.07 M KCl 0.1 M KCl mg P. L'1 1118 Dairy 0 4.3 5.7 7.9 50 8.2 6.1 7.5 75 9.0 6.1 7.5 100 8.4 6.4 7.9 Swine 0 4.3 5.7 7.9 50 9.1 6.4 7.9 75 9.2 6.7 8.1 100 10.5 7.1 8.5 * Approximate concentrations of total P (Pt) in each sorption solution. Other deviations from the Nair et al. (1984) method included the addition of chloroform and filtration method. Chloroforrn (in 0.75% ethanol; EM Science, guaranteed reagent; Gibbstown, NJ ) was added at a rate of 0.16 ml of chloroform to 25 ml of solution. This ratio is about half that proposed by Nair et al. (1984) but was experimentally determined to be adequate in suppressing microbial activity for the 24 h equilibration period (Appendix A). Nair et al. (1984) proposed the use of 0.45 pm filter paper for filtration of equilibrated solutions. However, this form of filtration is very time consuming and costly. Thus, 0.45 pm filtration units (MillexTM Millipore Corporation, Bedford, MA) and Whatman No. 1 filter paper (Whatman International Ltd, England) were tested to determine if there were differences in P, solution concentrations for theses filtration methods. Whatman No. 1 filter paper filtration was not different than filtration with 0.45 pm filter disc units. Thus, Whatman No. 1 filter paper was found to be is adequate for filtration of equilibrated solutions, for further details see Appendix B. To determine if variations in cation and ionic strength of the sorption solutions created sorption isotherms with different characteristics, two separate sorption experiments were conducted. First, soils were equilibrated with P solutions (ranging in concentration from 0 mg P L'1 to 50 mg P L'l) prepared from KHzPO4 in unbuffered 0.01 M CaClz at a 1:25 (w/v) soil to solution ratio with the addition of 0. 16 ml of chloroform to 25 ml of solution to inhibit microbial activity. In the second experiment, soils were equilibrated with P solutions (ranging in concentration from 0 mg P L'1 to 50 mg P L’l) prepared from KH2P04 in 0.1 M KCl, pH 6 at a 1:25 (w/v) soil to solution ratio with the addition of 0.16 ml of chloroform to 25 ml of solution to inhibit microbial activity. In the sorption experiments, soil was equilibrated with the solutions for 24 h in an end-over-end shaker. After equilibration, samples were centrifuged for 10 min at 2350 g and filtered through Whatman No. 1 filter paper. The amount of P sorbed was determined by the difference in the concentration of P initially added and the concentration of P in solution at the end of the equilibration period. Sorption isotherms were measured in triplicate. Sorption isotherms were fit to the Langmuir equation which is given by: bkC Q—1+kC (1) where Q is the amount of P sorbed to the soil (mg P kg'l), C is the amount of P in the solution after equilibration (mg P L'l), b is the sorption maxima (mg P kg'l), and k is the sorption strength (L mg"). Nonlinear regression was used to obtain the estimates and standard errors of b and k sorption parameters. (N LIN procedure; SAS Institute, Inc., 1999). Sorption isotherms were compared using nonlinear regression to determine if changing the solution matrix changed the isotherm characteristics. The procedure used was similar to comparing regression coefficients in linear regression (Cook and Weisberg, 1999; Freud and Littell, 2000; Laboski and Lamb, 2004). Inorganic Phosphorus Sorption Experiment Pi sorption experiments were conducted using three different sources of P: potassium phosphate (KHZPO4), dairy slurry, and swine slurry. Solution P, concentrations ranged fiom 0-75 mg P kg'1 for KH2P04, 0-70 mg Pi kg'1 for dairy slurry, and 0-60 mg Pi kg’1 for swine slurry. Solutions were prepared in 0.1 M KCl at pH 6 (Table 1.4). P solutions were added to soil at a 1:25 (w/v) soil to solution ratio. Chloroforrn was added, to inhibit microbial activity, at a ratio of 0.16 ml to 25 ml of solution. Samples were equilibrated for 24 h in an end-over-end shaker and centrifuged at 2350 g for 10 minutes then filtered through Whatman No. 1 filter paper. Isotherms were measured in triplicate for each combination of soil and P source. Daily differences in solution Pi concentrations caused by P cycling were noticed when initial P, values were determined from solutions stored at 4 °C. Therefore, for each replication initial P, values for each manure solution were determined in duplicate by placing 25 m1 of each sorption solution and 0.16 ml of chloroform in a 50 ml centrifuge tube (Corning Inc., Life Sciences; Acton, MA) and shaking for 24 h under the same conditions as the samples (Table 1.4). 10 Table 1.4. Concentration of inorganic P (Pi) in each sorption solution when dairy slurry, swine slurry, and KHzPO4 were used as P sources. Sorption Pi,r Solution Dairy Swine KH2P04 ............. mg L‘1 ------------- 1 O O O 2 1.3 1.9 2 3 3.4 4.5 5 4 6.8 8.8 10 5 10.0 13.4 15 6 13.3 17.3 20 7 20.5 25.1 30 8 33.0 41.9 50 9 49.3 60.3 75 10 69.8 -- -- f Average initial inorganic P (Pi) concentration as measured in solutions which were not equilibrated with soil but were shaken for 24 h with 0.16 ml of chloroform to 25 m1 of solution at the same time samples went through the batch equilibration process. Samples were analyzed for P; by the ascorbic acid colorimetric method at a wavelength of 882 nm (Frank et al., 1998). The amount of phosphorus sorbed was calculated as the difference between the concentration of Pi initially added and the concentration of Pi in solution at the end of the equilibration period. Sorption isotherms were fit to equation 1. Sorption isotherms were compared as previously described using nonlinear regression to determine if Pi sorption characteristics were different for the different P sources. 11 Results and Discussion Comparison of Matrices Sorption capacity for the two P sorption solution matrices were not significantly (or=0.05) different for all soils (Table 1.5, Figure 1.1, 1.2, and 1.3). Sorption strength for the two solution matrices was not significantly different for all soils except Oshtemo and Colwood 2. Overall, these two solution matrices yielded sorption isotherms that were not considerably different. Hence, a matrix of 0.1 M KCl at pH 6 was used to prepare manure sorption solutions and will provide results comparable to research using an unbuffered 0.01 M CaClz solution matrix. Inorganic Phosphorus Sorption Experiment For all soils, Pi sorption capacity (b) was significantly (o=0.05) greater for dairy slurry than swine slurry or KHzPO4 (Table 1.6, Figure 1.4, 1.5, and 1.6). The increase in P, sorption capacity observed when dairy slurry is the P source may be due to complexation of organic acids in the manure with Al and Fe from soil minerals giving rise to new P sorption sites on these complexes where hydroxyl groups are exchanged for phosphate (Appelt et al., 1975). P, sorption capacity from the swine slurry was significantly (o=0.05) less than that of KH2P04 for four of the six soils (Capac, Colwood 2, Oshtemo, and Spinks). P.- sorption capacity from the swine slurry was not significantly (o=0.05) different than KH2P04 for Colwood 1 and Parkhill soils (Figure 1.4 and Figure 1.6). The reduction in sorption capacity from swine manure compared to KH2P04 may be caused by preferential blocking of retention sites or displacement of sorbed P by anions of organic acids (Swenson et al., 1949; Kafltafi et al., 1998). 12 Table 1.5. Comparison of Langmuir parameters for standard and developed P sorption matrices. Sorption ------ Langmuir Parameters ----- Soil Series 8181121113? b k R2 mg kg'1 L mg'1 Capac each" 265a§ 0.2703 0.97 Kori 26la 0.201a 0.96 Colwood 1 CaClz 476a 0.217a 0.98 KCl 481a 0.1753 0.98 Colwood 2 CaClz 175a 0.070a 0.97 KC] 207a 0.038b 0.99 Oshtemo CaClz 106a 0.231a 0.96 KCl 109a 0.141b 0.97 Parkhill CaClz 329a 0.200a 0.96 CaClz 294a 0.1803 0.96 Spinks KCl 369a 0.31 1a 0.97 CaClz 363a 0.301a 0.97 I CaClz = standard matrix = unbuffered 0.01 M CaClz *KCI = developed matrix = 0.1 M KCl at pH 6.0 § Within a soil series, rows within a column followed by the same letter are not significantly different (or=0.05) from one another. 13 P sorbed (mg kg!) P sorbed (mg kg") 300 j 250 l 200 j 150 100 SOj Capac l.- v 0.01 M CaCl2 unbuffered O 0.1MKClatpH 6 LJII 20 30 40 50 60 Colwood 1 V 0.01 M CaCl2 unbuffered O 0.1 MKCI at pH 6 20 30 P in solution (mg L") 40 I I I I I T I I 7 r r I I I I Y 50 60 Figure 1.1. Sorption isotherms for CaClz and KC] sorption solution matrices for the Capac and Colwood 1 soils when KH2P04 was used as the source of P. 14 140 3 Colwood 2 P sorbed (mg kg") V 0.01 M CaCl2 unbuffered O 0.1 MKCl at pH 6 1L1144 ILII l ILLIJII - IrIIII IIII IIII IIII IIII 111! rfTr III I I I I I I I r 80 P sorbed (mg kg") 20 j v 0.01 M CaCl2 unbuffered O 0.1MKClatpH 6 ITI I I I I I I 171 I I I I I IT I I I l I I I T rj I I I I I I I I l I I I I 0 10 20 30 4o 50 60 70 P in solution (mg L") Figure 1.2. Sorption isotherms for CaClz and KC] sorption solution matrices for the Colwood 2 and Oshtemo soils when KH2P04 was used as the source of P. 15 80 350 300 Parkhill v , ALJILLJJ 250 £ 200 { 150 5 P sorbed (mg kg") 100 5 50 «j v 0.01 M Cac12 unbuffered ‘ O 0.1MKClatpH6 f I I I I I I 1 I W I I l I I I I I I I I I I I T I I I I f I I l I I I I 0 10 20 30 40 50 60 400 350 300 LJ4ILL1L11111 250 L11 200 IIIII 150 P sorbed (mg kg") 100 IIIILAIIIII v 0.01 M CaCl2 unbuffered O 0.1MKClatpH 6 I r I I I I I I I I I I I I I I I T T I I I I I T I I I r I I I I I I I 0 10 20 30 40 50 60 P in solution (mg L“) Figure 1.3. Sorption isotherms for CaClz and KC] sorption solution matrices for the Parkhill and Spinks soils when KH2P04 was used as the source of P. 16 70 Table 1.6. Effect of inorganic P (Pi) source on Langmuir parameters. ------ Langmuir Parameters ------ Soil Series Source b k R2 mg kg'1 L mg'l Capac KH2P04 261a'r 0.201 a 0.96 Dairy 32% 0.135b 0.95 Swine 1720 0.180ab 0.97 Colwood 1 KH2P04 481a 0.175a 0.98 Dairy 713b 0.107b 0.97 Swine 318C 0.175a 0.98 Colwood 2 KH2P04 205a 0.056a 0.99 Dairy 327b 0.049ab 0.98 Swine 187c 0.042b 0.99 Oshtemo KH2P04 109a O. 141 a 0.97 Dairy 144b 0.123a 0.95 Swine 1 14a 0.063b 0.96 Parkhill KH2P04 294a 0.1 80a 0.96 Dairy 374b 0.098b 0.96 Swine 191c 0.146ab 0.97 Spinks KH2P04 3633 0.301 a 0.97 Dairy 671b 0.011b 0.96 Swine 2720 0.225a 0.98 *Within a soil series, rows within a column followed by the same letter are not significantly different (o:=0.05) from one another. 17 350 { Capac 5 TO!) a A4 . co -4 5 a '3 I «‘3 J 3 i of _‘ 0 10 20 30 40 50 60 70 700 . Colwood 1 ‘ 600 j j ‘ 1 Too A 1 '5‘ L 50 a E I '0 a D .. '9 : g . a: J I Dairy 1 A Swine I . KH2P04 : 0 10 20 30 40 50 60 70 Pi in solution (mg 151) Figure 1.4. Inorganic P (Pi) sorption isotherms for the Capac and Colwood 1 soils when various concentrations and sources of P (dairy slurry, swine slurry, KHzPOa) were equilibrated with soil in a 0.1 M KCl, pH 6 matrix. 18 Colwood 2 ‘ ' 1 PLJ h—I M O l l Pi sorbed (mg kg") '5 O 50 3 f’ . 3 : O I Dairy . .’ A Swine « 0 _' e KH2P04 l ‘ 1 0 10 20 30 40 50 6O 70 250 . 4 Oshtemo j 200 1 3 "Log 150 . - on I . E. ‘1 . E 100 - e —8 - r— “ I I ‘ 8 a a: I ‘ 50 d v ’ I Dai ~ , ry I A Swine 0 J ’ . “121304 0 10 20 30 4o 50 60 7o Pi in solution (mg L") Figure 1.5. Inorganic P (Pi) sorption isotherms for the Colwood 2 and Oshtemo soils when various concentrations and sources of P (dairy slurry, swine slurry, KH2P04) were equilibrated with soil in a 0.1 M KCl, pH 6 matrix. 19 400 Parkhill | 350 300 250 200 150 Pi sorbed (mg kg") 100 50 LII:IILIIIIIIIIIIIIIIIIIIIIIIIIIIIIIJIILL IIILIIIIIIIIIIJIIIIIIIIIIIIIIIIIIIIrlIIII I T I I I T I r T I I I T I I I T l I I I I I I I I I I I I I I I r I I I ' I I 0 10 20 30 40 50 60 70 700 Spinks I 600 500 400 300 Pi sorbed (mg kg") 200 . I Barry A Swine O KH2P04 IJ_LIIL+LIIIIIJIIIIIIIllllllll 100 IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII LI I 1L1 O r r I I I TI l' I I I I I I I I I I I I I I l I I T T I I I I I I I I I I I I I 0 10 20 30 40 50 60 7O Pi in solution (mg L") Figure 1.6. Inorganic P (Pi) sorption isotherms for the Parkhill and Spinks soils when various concentrations and sources of P (dairy slurry, swine slurry, KH2P04) were equilibrated with soil in a 0.1 M KCl, pH 6 matrix. 20 Sorption strength (k) of the dairy slurry was significantly (a=0.05) less than KHZPO4 for Colwood 1 and Spinks soils (Table 1.6). For all other soils, P, sorption strength was not significantly different between dairy slurry and KH2P04. Sorption strength of swine slurry was significantly less than KH2P04 for Capac, Colwood 2, and Oshtemo soils. For all other soils, Pi sorption strength of swine slurry and KH2P04 were the same. Sorption strength of dairy slurry was not significantly different than swine slurry for Capac, Colwood 2, and Parkhill soils. Sorption strength of dairy slurry was significantly less than swine slurry for Colwood 1 and Spinks soils and was significantly greater for Oshtemo soil. Overall, Pi sorption strength of either manure slurry was less than KH2P04; however this trend was not always significant. This observed reduction in sorption strength may be caused by the interferences of other compounds in the manure slurry with sorption of P, at high strength sorption sites causing Pi to be sorbed to low strength sites (Holford et al., 1997). Conclusions Generally, soils had a greater sorption capacity for P,- from dairy slurry compared to KH2P04 or swine slurry. Manure slurries tended to have a lower Pi sorption strength than KH2P04 and therefore, may present an increased risk for P losses to runoff and leaching. Based on these data, it appears that the addition of swine slurry to soil will result in P being more available, as measured by soil test, than when P is added as KH2P04, because less P is sorbing to the soil and the sorption strength is weaker. On the other hand, availability of P from dairy slurry may be less than KH2P04 because more P, sorbs to soil from dairy slurry compared to KH2P04. Thus, manure from different animal species may increase soil test P levels differently. 21 Data from an incubation study with five of these soils support the conclusions above. Dairy slurry was always less available (as measured by soil test) than KH2P04 or swine slurry whereas, swine slurry was as available as KH2P04 for two soils and less available for the other three soils. See Chapter 3 for more details. The sorption isotherm data presented in this chapter may also explain why Laboski and Lamb (2003) found that swine slurry increased soil test P more than KH2P04 when incubated with soil at the same rate of total P. Understanding how P sorption and subsequent agronomic and environmental availability change with manure application is necessary to develop sound nutrient management recommendations. 22 References Appelt, H., N.T. Coleman, P.F. Pratt. 1975. Interactions between organic compounds, minerals, and ions in volcanic-ash-derived soils:H. Effects of organic compounds on the adsorption of phosphate. Soil Sci. Soc. Amer. Proc. 39:628-630. Bhat, K.K.S., and J .R. O’Callaghan. 1980. Behaviour in the soil of orthophosphate from pig slurry compared with that of KHzPO4. J. Agric. Sci. Camb. 94: 195-201. Brown, J .R. 1998. Recommended Chemical Soil Test Procedures for the North Central Region. North Central Regional Research Publication No. 221 (Revised). Missouri Agric. Exp. Stn. SB1001. Cook, RD. and S. Weisburg. 1999. Applied regression including computing and graphics. John Wiley&Sons, New York. Frank, K., D. Beegle, J. Denning. 1998. Phosphorus. In Recommended Chemical Soil Test Procedures of the North Central Region. North Central Region Research Pub No. 221 (revised) Missouri Agricultural Experiment Station. SBlOOl. Freud, R.J. and RC. Little. 2000. SAS system for Regression Ananlysis, 3rd Ed. SAS Institute, Inc., Cary, NC. Gee, G.W. and J .W. Bauder. 1986. Particle-size Analysis. p 383-411. In A.Klute (ed.) Methods of Soil Analysis Partl- Physical and Mineralogical Methods. American Society of Agronomy, Madison, WI. Holford, I.C.R., C. Hird, and R. Lawrie. 1997. Effects of animal effluent on the phosphorus sorption characteristics of soil. Aust. J. Soil Res. 35:365-73. Kaflcafi, U., B. Bar-Yosef, R. Rosenburg, and G. Sposito. 1998. Phosphorus adsorption by kaolinite and montmorillonite: H. Organic anion competition. Soil Sci. Soc. Am. J. 52:1585-1589. Laboski, CAM. and J .A. Lamb. 2003. Changes in soil test phosphorus concentrations after application of manure or fertilizer. Soil Sci. Soc. Am. J. 67: 544-554. Laboski, CAM. and J .A. Lamb. 2004. Impact of manure application on soil phosphorus sorption characteristics and subsequent water quality implications. Soil Sci. 169: 440- 448. Mozaffari, M. and J .T. Sims. 1994. Phosphorus availability and sorption in an atlantic coastal plain watershed dominated by animal-based agriculture. Soil Sci. 1 57 :97- 1 07. Nair, P.S., T.J. Logan, A.N. Sharpley, L.E. Sommers, M.A. Tabatabai, and TL. Yuan. 23 1984. Interlaboratory comparisons of a standardized phosphorus adsorption procedure. J. Environ. Qual. 13:591-595. Parkhurst, D.L., 1995, User's guide to PHREEQC--A computer program for speciation, reaction-path, advective-transport, and inverse geochemical calculations: US. Geological Survey Water-Resources Investigations Report 95-4227, 143 p. Peters, J .(ed). 2003. Recommended Methods of Manure Analysis. Bulletin A3769, University of Wisconsin Extension. Cooperative Extension Publishing, Madison, WI. 1-2/2003. Ryden, J .C. and J .K. Syers, 1975. Rationalization of ionic strength and cation effects on phosphate sorption by soils. J. Soil Sci. 26: 395-406. Sample, E.C., R.J. Soper, and G]. Racz. 1980. Reactions of Phosphate Fertilizers in Soils. In M. Stelly (ed.) The Role of Phosphorus in Agriculture. American Society of Agronomy, Madison, WI. SAS Institute, Inc. 1999. SAS 8.1. SAS Institute. Inc, Cary, NC. Sharpley, A.N. (1996). Availability of residual phosphorus in manured soils. Soil Sci. Soc. Am. J. 60:1459-1466. Siddique, MT and J .8. Robinson. 2003. Phosphorus sorption and availability in soils amended with animal manures and sewage sludge. J. Environ. Qual. 32:1114-1121. Sims, J .T. 1993. Environmental soil testing for phosphorus. J. Prod. Agric. 6:501-507. Swenson, R.M., C.V. Cole, and DH. Sieling. 1949. Fixation of phosphate by iron and aluminum and replacement by organic and inorganic anions. Soil Sci. 67:3-22. 24 Chapter 2 Sorption of inorganic and total phosphorus from dairy slurry, swine slurry, and potassium phosphate to selected Michigan soils Abstract Phosphorus (P) is essential for crop production but is of environmental concern with regard to surface water pollution. Little research has focused on the construction of P sorption isotherms where the P source is manure. The objective of this study was to determine and compare the sorption characteristics of inorganic P (Pi) and total P (P,) where the P source was dairy slurry, swine slurry, or potassium phosphate (KHZPO4). The six selected Michigan soils had no manure application history, soil test P < 30 ppm, pH < 7, and had a range in soil texture. Sorption solutions were prepared in 0.1 M KCl at pH 6 and equilibrated with soils at a 1:25 (w/v) soil to solution ratio for 24 h. The amount of Pi, Pt, iron (Fe) and aluminum (Al) remaining in solution were measured. For all soils, P.- and PI sorption capacity of dairy slurry was greater than KH2P04. Pt sorption capacity of swine slurry was greater than KHzPO4, while P, sorption capacity was less than KHzPO4. Overall, Pi and Pt sorption strengths of the manure slurries were less than or equal to KHZPO4. Desorbed Fe and Al were correlated to the differences in P, sorption between dairy slurry and KHzPO4 but were not correlated to the differences in Pi sorption between swine slurry and KH2P04. Desorption of Fe and Al from the soil surfaces may have been caused by organic acids. The greater soil sorption capacity for P from dairy slurry may suggest that P from dairy slurry would be less available, as 25 measured by soil test, than fertilizer P. The reduced P sorption capacity from swine slurry suggests that swine slurry P might be more available than fertilizer P. The results also suggest that dairy slurry having lower sorption strength than swine slurry and KHzPO4 may have a greater potential to release P into runoff waters and/or leach. Further investigation is needed to determine the potential for P to release to surface water runoff when these manures are applied to soil. 26 Introduction In 2000, the US. EPA reported that agriculture was the leading source of surface water pollution, affecting 18% of the assessed rivers, streams, and lakes. Agriculture contributed to approximately 45% of the reported water quality problems in impaired rivers, streams, and lakes (U .8. EPA, 2000). This is less than the 1994 estimate that agriculture contributed to 70% of the impairment in the water quality of rivers and lakes (US. EPA, 1994). However, further reducing pollution from agriculture is required to improve the quality of our waters. Studies have shown that excessive application of phosphorus (P) in the form of animal manure and/or inorganic fertilizer is responsible for the accumulation of soil P to levels in excess of crop requirements (Sharpley et al., 1984; Sims, 1993). Such soils present a risk to water quality if the P is transported to surface water and accelerates eutrophication. Eutrophication has the potential to impair water quality by restricting a water body’s use for fisheries, recreation, industry, and drinking purposes (Sharpley and Menzel, 1987; Sharpley et al., 1998). Phosphorus is strongly adsorbed by most soils, thus, loss of P in surface runoff is generally of greater concern than leaching (Sharpley et al., 1998). Runoff water contains P in both soluble (dissolved) and particulate (sediment bound) forms of P (Sharpley et al., 1998). Phosphorus sorption characteristics, such as sorption capacity and sorption strength, of a given soil likely affect the potential for P losses to surface waters both as soluble and particulate P. To reduce the environmental impact of P fiom agricultural soils, it is important to understand the mechanisms of P sorption (binding) to soil colloid surfaces. 27 The focus of recent P sorption research has been to construct P sorption isotherms, using solutions of KH2P04, to determine if soil P sorption characteristics change after application of manure or fertilizer compared to untreated soils. In general, P sorption capacity and strength have been found to change with the application of manure or fertilizer, with reductions in both properties more common than increases (Mozaffari and Sims, 1994; Sharpley, 1996; Holford etal., 1997; Siddique and Robinson, 2003; Laboski and Lamb, 2004). Manure application increased, reduced, or did not change sorption capacities of seven Minnesota soils (Laboski and Lamb, 2004). The authors also reported reduced or unchanged sorption strength for these soils when soils with a history of manure application were compared with the same soils to which no manure was applied (Laboski and Lamb, 2004). In another study, P sorption capacity and strength were reduced when soils were treated with manures from various species, sewage effluent, or superphosphate (Holford et al., 1997). The magnitude of the decrease in sorption capacity and strength depended on the amount of manure or effluent applied over time; with increased application rates causing a greater reduction in the sorption capacity and strength (Holford et al., 1997). The observed decreases in P sorption capacity have been suggested by several researchers to be due to anions of organic acids competing for P sorption sites (Swenson et al., 1949; Nagarjah et al., 1970; Kaflcafi et al., 1998). When organic amendments are added to soils, organic acids may be added as part of this amendment directly and/or as a result of microbial decomposition of the amendment. Reduction of P sorption can be caused by the preferential blocking of retention sites or by displacement of sorbed P by anions of these organic acids (Swenson et al., 1949; Kaflcafi et al., 1998). Nagarjah et al. 28 (1970) found the decrease in P sorption by organic acids is dependent on the stability of the complexes formed with Fe and Al. Inositol hexaphosphate (II-1P), a form of organic P in animal manures, was found to preferentially sorb to soil compared to inorganic P (Anderson et al., 1974). IHP also released orthophosphate, which was previously bound to soil, into solution when added to soil alone (Anderson et al., 1974). Additionally, as the concentration of added IHP increased, Fe and Al solution concentrations increased, whereas, the absolute amount of IHP sorbed decreased. This was probably caused by the formation of soluble complexes of Fe and A1 with IHP (Anderson et al., 1974). These studies suggest that organic P (P0) in manures may compete for binding sites on soil surfaces and potentially reduce inorganic P (Pi) sorption. The increase in sorption capacity observed in previously mentioned research is less well understood. Appelt et al. (1975) suggested increases in P sorption capacity may be caused by complexation of organic acids with Al and Fe from soil minerals giving rise to new P sorption sites on the complexes where hydroxyl groups are exchanged for phosphate. Whereas, others found that low molecular weight organic acids increase solubility of solid-phase Al in soils (Stumm, 1986; Fox et al., 1990). Organic acids influence the release of P and Al from surfaces in soils by formation of stable complexes with Al, which is dependent on the stability constant of the ligand (Fox etal., 1990). The dissolution of Al containing solids may give rise to new P sorption sites on the freshly scoured soil surfaces. Sorption strength estimated from the Langmuir equation was described by Holford et al. (1997) as an average strength of the total sorption sites which includes both high and low strength binding sites in the soil. Holford et al. (1997) suggested that P 29 initially sorbs to high strength sites followed by low bonding strength sites. They estimated that high strength sites represent approximately one-fourth of the total sites and have a bonding strength about 100 times greater than low strength sites (Holford et al., 1974) Some authors have reported a reduction in sorption strength with the application of amendments but few have given explanations as to why this reduction in sorption strength likely occurs (Mozaffari and Sims, 1994; Sharpley, 1996; Holford et al., 1997; Siddique and Robinson, 2003; Laboski and Lamb, 2004). Holford et al. (1997) suggested that the observed reduction in sorption strength with manure or effluent application history was likely caused by organic anion interaction with sorbing P anions; thus, lowering the sorption strength. Both Holford et al. (1997) and Laboski and Lamb (2004) concluded that P sorption strength may have a greater impact on P availability than P sorption capacity and is possibly the more environmentally important sorption parameter because the reduction in P sorption strength with application of amendments resulted in a greater amount of P in soil solution at any given level of P sorbed. Far less research has been conducted on the sorption of P where isotherms were determined using sorption solutions containing manure as the source of P. Bhat and O’Callaghan (1980) studied the P sorption characteristics of pig slurry compared to KH2P04. They found that more P, sorbed to soil from solutions of pig slurry than KH2P04. However, the sorption solutions were not treated to inhibit microbial activity. As a result, it is likely that some of the observed increase in P sorption was actually microbial consumption of P. The objective of this study was to determine and compare 30 sorption characteristics of total P (Pt), inorganic P (Pi), and organic P (Po) from dairy slurry, swine slurry, and KHzPO4. Methods and Materials Soil and Manure Six surface soils-(0-0.15 m) used in these experiments were collected in southern, central Michigan and are described in Table 2.1. Soils selected had no history of manure application, Bray l-P < 30 mg P kg", pH < 7, and varied in soil texture. These selection criteria were employed to reduce initial desorption of P and minimize calcium phosphate precipitation. Prior to experimentation, soils were air dried, sieved (2 mm sieve) and mixed thoroughly to ensure homogeneity. Soils were characterized by measuring pH, organic matter, cation exchange capacity, and Bray l-P (Brown, 1998). Soil particle size was determined by the hydrometer method (Gee and Bauder, 1986) (Table 2.1). The liquid dairy slurry used in this experiment was collected from a short-terrn storage pit near a feeding facility. The pit was agitated prior to sample collection to ensure sample homogeneity. The liquid swine slurry used in this experiment was collected from a slotted floor feeder house at time of cleaning. Once collected, manures were individually mixed and sieved (2 mm sieve) to facilitate pipetting by removing large residual particles. Manures were characterized by measuring percent moisture, percent solids, total N, total P, soluble salts (K, Ca, Mg, and Na), Al, and Fe (Table 2.2) (Peters, 2003). After collection and sieving, manures were stored in polyethylene containers at 0 °C. Aliquots of manure from each species were thawed and stored at 4 °C as needed. 31 .3633 owgzoxolcozmo + tag—am: 2:083 0.52.: 3 an m a 2 2. S. .358 223m 33% v mow am am mm Nam We mEozamoucm 2:22 288 Eco—Sm .Eomco: .o>._uou._Eom 6828 552-85 . . . . 282%: 2x5 285 2 m w a mm we m N v m .338 €828 $88. 8:80 2:850 . . . zeacaoucm 29C. 0.52: w a E S ow 3 L o m w 638a flux—E .mEanEm N coca—o0 2 4.2 .8 cm s. 2% 9m . zeacwoncm we? 28.: _ Basso 038a Bis bane—-05..“ . . . 23585 2:? p.85 E A : 2 R mm a N v c 38258 .855 $58.85 2&8 Tm: wE :3 93:8 ................ o\° .............. A: bum Emu ‘35 Em 28m bum—2 ONE 2 5.585320 85m zom 2530 mg 26m 85.35 3823 we :outhoSmaco use 20385320 ._.m 035. 32 Table 2.2. Characterization of dairy and swine manure slurries. Species Moisture Solids N P K Ca Mg Na Al Fe ......... % mg L'I Dairy 95.3 4.7 3271 529 2228 1618 539 647 21 57 Swine 94.8 5.2 6542 1294 3810 1078 648 755 53 151 Phosphorus Sorption Isotherms P sorption experiments. P sorption experiments were conducted using three different sources of P: KHZPO4, dairy slurry, and swine slurry. Sorption solutions were prepared on an approximate P, basis, using the nutrient characteristics of whole sieved manures, to contain a range of P concentrations between 0 and 100 mg P, L'1 (Table 2.2). Actual solution P, concentrations ranged from 0-75 mg P L '1 for KHzPO4, 0-70 mg P, L'1 for dairy slurry, and 0-60 mg Pt L'1 for swine slurry (Table 2.3). Solutions were prepared in 0.1 M KCl at pH 6 (See Chapter 1 for details). P sorption solutions were added to soil at a 1:25 (w/v) soil to solution ratio. Chloroforrn (in 0.75% ethanol; guaranteed reagent; EM Science, Gibbstown, NJ) was added to inhibit microbial activity at a rate of 0.16 ml of chloroform to 25 ml of solution (Appendix A). Solutions were equilibrated for 24 h in an end-over-end shaker. Daily differences in solution P, concentrations caused by P cycling were noticed when initial P,- solution concentrations were determined from solutions stored at 4 °C. Therefore, for each replication initial P, and P, values for each manure sorption solution were determined in duplicate by placing 25 ml of each sorption solution in a 50 ml centrifuge tube (Corning Inc., Life Sciences; Acton, MA) with no soil (Table 2.3). 33 Table 2.3. Phosphorus sorption solution characterization. / -------- Dairy ----------------- Swine --------- KHZPO4 S orption P,- as P, as S olution Pf P3 percent P5 P3 percent P Of Pt OfPt --- mg L'1 --- % --- mg L'1 --- % mg L'1 1 O 0 0 0 0 2 1.3 1.6 81 1.9 2.0 95 2 3 3.4 4.0 85 4.5 4.7 96 5 4 6.8 7.9 86 8.8 9.1 97 10 5 10.0 11.8 85 13.4 14.0 96 15 6 13.3 15.3 87 17.3 18.0 96 20 7 20.5 23.0 89 25.1 26.5 95 30 8 33.0 37.8 87 41.9 42.9 98 50 9 49.3 57.3 86 60.3 63.1 96 75 10 69.8 70.1 100 -- -- -- -- + Average (eight replications) initial inorganic P (P,) solution concentration of centrifiiged §a.mples. " Average (eight replications) initial total P (P,) solution concentration, of centrifuged samples. Solutions were shaken for 24 h under the same conditions as the samples. After equilibration, all samples, including those for measurement of initial P, and P, concentrations, were centrifuged at 2350 g for 10 minutes. It was assumed that similar amounts of manure P would be removed by centrifugation with and without the presence 0 f soil. The percent P, and P, removed from manure sorption solutions by centrifugation Was determined in a separate experiment, see Appendix C for details. The amount of P, as a percent of P, is similar across the solution range for a given slurry showing that centrifugation removed similar percents of P0 in these solutions (Table 2.3). P, analysis. After centrifugation, an aliquot of solution was reserved for P, digestion. The remaining solution was then filtered through Whatrnan No. 1 filter paper (Whatman International Ltd, England) (Appendix B) for P, sorption analysis. All 34 samples were analyzed for P, immediately after equilibration by the ascorbic acid colorimetric method at a wavelength of 882 nm (Frank et al., 1998). The amount of P, sorbed was determined by the difference between the concentration of P, initially added and the concentration of P, in solution at the end of the equilibration period. P, analysis. Reserved aliquots of unfiltered equilibrated sorption solutions were immediately stored at 4 °C until digestion for P, analysis. Samples were digested according to the persulfate oxidation method of Bender and Wood (2000) with some modifications. Aliquots of equilibrated samples (5 or 10 ml) were diluted to a final volume of 50 ml with distilled water. Diluted samples were then digested with 0.7 g potassium persulfate and 1 ml of 11 N H2804. The optimum potassium persulfate needed for complete digestion was determined in a separate experiment, see Appendix D for details. Samples were first digested for 1.25 h at approximately 100 °C, and then the hotplate was turned to the maximum setting, to increase boiling, until the samples were reduced to approximately 10 ml. Cooled samples were diluted with 30 ml of distilled water. After samples were diluted, approximately 5 m1 of 2 M NaOH was added to adjust the pH of the solution. Solutions were transferred to volumetric flasks and diluted to a final volume of 100 ml. Phosphorus in diluted digested samples was analyzed by the ascorbic acid colorimetric method at a wavelength of 882 nm (Frank et al., 1998). Digested standards of KH2P04 were used to produce the standard curve for determination 0f the concentration of P, in solution. The amount of P, sorbed was determined by the difference between the concentration of P, initially added and the concentration of P, in Solution at the end of the equilibration period. 35 P, analysis. The amount of Po sorbed was determined by the difference between the amount of P, and P, sorbed. Likewise, the amount of Po remaining in solution after equilibration was determine by the difference between the amount of P, in solution and the amount of P, in solution. Iron and Aluminum Solution Concentrations Aliquots of equilibrated solutions used for P, analysis also were analyzed for Fe and Al concentrations in solution to determine if manure slurry solutions caused dissolution of Fe and Al from soil colloid surfaces. Samples were spiked with 5 mg L'1 each of Fe and Alto achieve values above the 1 mg L'1 method detection limit. Spiked samples were analyzed for Fe and Al in solution using a direct current plasma spectrometer (DCP) (SpectraSpan V; Andover, MA). The amount of Fe and Al desorbed was calculated by the difference between the amount of Fe and Al in solution initially and the amount in solution after equilibration. Data Analysis P Sorption Isotherms. Nonlinear regression was used to fit P, and P, sorption data to the Langmuir equation which is given by: bkC Q_1+kC (1) where Q is the amount of P sorbed to the soil (mg P kg'l), C is the amount of P in the solution after equilibration (mg P L'l), b is the sorption maxima (mg P kg"), and k is the sorption strength (L mg'1 ). Nonlinear regression was used to obtain estimates and standard errors of the b and k parameters (N LIN procedure; SAS Institute, Inc., 1999). 36 Comparisons of b and k terms between two data sets were made using nonlinear regression in a procedure similar to comparing regression coefficients in linear regression (Cook and Weisberg, 1999; Freud and Littell, 2000; Laboski and Lamb, 2004). The P0 sorption isotherms did not fit the Langmuir equation. Relationship Between Phosphorus and Iron and Aluminum. As shown in Chapter 1, for a given soil there were significant differences in P, sorption capacities for the different P sources. However, P, sorption capacities for the dairy and swine slurries were not considerably different from one another and were greater than the sorption capacity of KHZPO4. In order to investigate whether those differences in either P, or P, sorption capacities were related to desorption of Fe or Al from soil surfaces, the following analysis was employed. First, because discrete data points for the three P sources at precisely the same solution P concentrations did not exist, Q values for KHzPO4 were computed by substituting the experimentally determined C values of a given slurry in the Langmuir equation for sorption from KHZPO4 for each data point from the manure slurries. These calculated KHzPO4- Q values were then subtracted from the experimentally derived Q values for the respective slurry to obtain the difference in P, or P, sorbed between the slurries and KHZPO4. Next, the Fe and Al desorbed during the equilibration of each slurry-soil sample were correlated with the differences in P, or P, sorption calculated in the first step. 37 Results and Discussion Iron and Aluminum Solution Concentrations For all soils, Fe and Al desorbed from soil constituents for both slurries with more desorption occurring in soils equilibrated with dairy slurry compared to swine slurry. (Appendix E). No apparent desorption of Fe and Al occurs when soils are equilibrated with KHZPO4 solutions (Appendix E). Difference in dissolution of Fe and Al for the two manure sources may be caused by differences in the concentrations and type of organic acids in the different manure species; Fox et al. (1990) found these factors impact solubility of solid phase A1 in soils. Organic acid type and concentration was not measured for the manures used in this study. However, differences in organic acid compositions between the manures could likely occur and be caused by differences in animal diet, physiology, and perhaps manure storage. Phosphorus Sorption Capacity For all soils, P, sorption capacity (b) was significantly (01 =0.05) greater for dairy slurry than swine slurry or KHZPO4 (Table 2.4, Figure 2.1, 2.2, 2.3, 2.4, 2.5, and 2.6). P, sorption capacity for the swine slurry was significantly less than that of KHzPO4 for Capac, Oshtemo, and Spinks soils, but was not significantly different than KHzPO4 for Colwood l, Colwood 2, and Parkhill soils. For all soils, P, sorption capacity was significantly greater for both dairy and swine slurries compared to KHZPO4 (Table 2.4). For Capac, Colwood l, and Spinks soils, P, sorption capacity was significantly greater for dairy slurry than swine slurry, but dairy slurry was not significantly different than swine slurry for all other soils. 38 Table 2.4. Effect of P source on Langmuir parameters for selected soils. ---- Langmuir Parameters ---- Soil Series Source b k R2 mg kg'1 L mg'1 Capac P, KHZPO4 261a" 0.201a 0.96 P, Dairy 32% 0.135b 0.95 P, Dairy 439C 0.087c 0.94 P, Swine 172d 0.180abc 0.97 P, Swine 297e 0.23 8a 0.95 Colwood l P, KH2P04 481a 0.175a 0.98 P, Dairy 713b 0.107b 0.97 P, Dairy 741b 0.08% 0.98 P, Swine 3180 0.175a 0.98 P, Swine 604d 0. 161a 0.97 Colwood 2 P, KH2P04 207a 0.03 8a 0.99 P, Dairy 362b 0.105a 0.98 P, Dairy 577c 0.016b 0.98 P, Swine 212a 0.027ab 0.99 P, Swine 676C 00le 0.96 Oshtemo P, KH2P04 1093 O. 141 a 0.97 P, Dairy 144b 0.1233 0.95 P, Dairy 181c 0.091b 0.98 P, Swine 1 14a 0.063c 0.96 P, Swine 181C 0.091 abc 0.91 Parkhill P, KH2P04 294a 0.180a 0.96 P, Dairy 374b 0.098b 0.96 P, Dairy 437c 0.081b 0.95 P, Swine 191d 0.146ab 0.97 P, Swine 394bc 0.1 15b 0.93 Spinks P, KHzPO4 363a 0.301 a 0.97 P, Dairy 671b 0.10% 0.96 P, Dairy 692b 0.081b 0.97 P, Swine 272c 0.225ac 0.98 P, Swine 517d 0.179c 0.97 * For a given soil, values within a column followed by the same letter are not significantly different at a=0.05. 39 P sorbed (mg kg") KHZPO4 3 Dairy Pt 3 Dairy Pi ‘: 0 10 20 30 40 50 60 7O 80 500 . 450 i 400 1" 350 2 A ‘ .32 300 E A 5 an : A : 5 250‘: A a 'U I 2 5.; 200 5 o E 8 3 A E a. 150 j j 1 : 100 -: o KH2P04 .3 50 _ A Swrne Pt _. f A Swine Pi 3 O 1 ': 0 10 20 30 40 50 60 70 80 P in solution (mg L") Figure 2.1. Inorganic P (P,) and total P (P,) sorption isotherms for Capac soil. 40 P sorbed (mg kg") 0 KHZPO4 j E] Dairy P‘ I I Dairy Pi 60 70 80 —A q Im d .2 . °0 ‘_ E . v '8 4 5’: 1 9‘ a . KHZPO4 j A Swine P I t A Swine Pi l I I II I I I I 0 10 20 30 40 50 60 70 80 P in solution (mg L'l) Figure 2.2. Inorganic P (P,) and total P (P,) sorption isotherms for Colwood 1 soil. 41 P sorbed (mg kg’l) P sorbed (mg kg“) 350 3 :1 3 300 § ' £ a '3 I 250 1 D { -( ~4 . 1| . : ll 1 200 El II , : - 3 150 j z T : g- : 1005 urf‘ 5 ,0 3 '4' ' o KH2P04 5 I . J D ' p I : .a '3 auy" : 0 “j I Dairy Pi j 10 20 30 40 50 6O 70 80 400 , , 350 € 5 1 : 300 “j j : A 1 250 i 5 200 i .j 100 € 21 ‘A a : A Swine Pt O 1 . : 3 A Swme Pi ; 10 20 30 40 50 60 70 80 P in solution (mg L“) Figure 2.3. Inorganic P (P,) and total P (P,) sorption isotherms for Colwood 2 soil. 180 , 160 140 120 100 so 60 40 i P sorbed (mg kg'l) 180 _ 160 - 140% 120 - 100- 80 - 60- 40 j P sorbed (mg kg") 205 A A KH2P04 Swine P, Swine Pi rermr l r r LJJAL '20 I l ' l v r v r - r r I 0 10 20 30 40 50 P in solution (mg Lil) I 60 1 70 Figure 2.4. Inorganic P (P,) and total P (P,) sorption isotherms for Oshtemo soil. 43 80 P sorbed (mg kg") 0 101,130, Dairy Pt Dairy Pi E 0 10 20 30 40 50 60 7O 80 450 , 350 5 5 To 300-: 2 ‘ A 0 i 00 “ 4 : A a“. 1 E0 250 1 ,A 1 E 200 -j ‘ a, . -j -E I 3 ‘ I 8 150 E ./ -I 9.. j A 3 10° 3 . o KHZPO. 3 50 j , A Swine Pt J 0 _§ -' A Swine r>i 0 10 20 30 40 50 60 70 80 P in solution (mg L") Figure 2.5. Inorganic P (P,) and total P (P,) sorption isotherms for Parkhill soil. 44 P sorbed (mg kg") P sorbed (mg kg") 700 , 600 -j 500 5 5 400 f 5 : 1 300 j ‘j 200 3 1 : KHZPO4 l 100 j Dairy Pt ‘1 : Dairy P, 1 0 l 1 7o 80 700 , . 600 -: «j i : 500 j j 400 l J i m ' ‘ 2 300 — _ : O 2 200 f 3 : o KH2P04 1 u 4 100 j A Swine Pt 1 f ; Swine Pi f o - - 'l' I I I I I I I 1 I I I I l f 0 10 20 3o 40 50 60 7o 80 P in solution (mg L") Figure 2.6. Inorganic P (P,) and total P (P,) sorption isotherms for Spinks soil. 45 For Capac, Colwood 2, Oshtemo, and Parkhill soils, P, sorption capacity was significantly (0: =0.05) greater than P, sorption capacity for dairy slurry. While for the Colwood 1 and Spinks soils, P, and P, sorption capacities for dairy slurry were not significantly different. P, sorption capacity of swine slurry was significantly greater than P, sorption capacity of swine slurry for all soils. As hypothesized by others, the reduction in P, sorption capacity observed by swine slurry compared to KHzPO4 may be due to preferential blocking of retention sites or displacement of sorbed P by anions of organic acids (Swenson et al., 1949; Kafl88 m0 :ouawtofifimno 88.802 .N.m 030,—. $2880 088008100sz .. . . . 282%: 2:83 288 E e m m a 8 m e s e .8er $88 08% . . . 83:80.0:m 2:02 0608 _ 8 v _ ON 0N 0m Nm N m m 0 €680: 030888 €808 £882-85 :5. m . . . 28398 2.5. are 2 m w 0 N we m N v m .258 €808 $882 8800 08850 . . . 20:88.5 083. 0608 003 0 w 0 E N_ ow we _ 0 m m .0208 €808 $882-08.; N 0 _ U . . . =008008m 89C. 0608 003 0 m _ v 3 NN cm NV 0 m o m .258 €808 $882-08m _ 0 _ U Tm; w8 _-w0_ e_080 ................ .x. .............. a: ham Emu >20 Em 08m 582 05 fi— 0088885 mow—0m =0m 08830 In 80805820 08 0088880880 mom .fim 030,—. 65 pipette to 75 g of soil. P sources were incorporated with soil by mixing with a spatula until all soil was moistened and the treatment was evenly distributed. Distilled water was added and incorporated to bring soils to 85% field capacity. Soils were packed to a bulk density of 1200 kg m’3. The liquid volume for the dairy treatment exceeded the water- holding capacity of the Oshtemo soil; thus the total treatment volume was applied over a two day period. Two replicates of each treatment were incubated at 21.5 i 1 °C for six weeks in covered containers with pinhole openings to allow for aeration while minimizing water loss. Soil moisture was maintained between 80 and 85% field capacity during the duration of the incubation. After incubation soils were air dried and sieved (2 mm sieve). Laboratory Analysis Extractable P for all treatments was measured in duplicate by Bray l-P soil extraction (Frank et al., 1998). Initial P sorption characteristics were measured for each soil prior to incubation. P sorption was measured in duplicate for each replication of the treated incubated soils after incubation. P sorption was measured using the procedure of Nair et al. (1984) with some modifications. Soils were equilibrated with KH2P04 solutions containing 0-50 mg P L’I at a 1:25 (w/v) soil to solution ratio with the addition of 0.14 ml of chloroform for microbial inhibition. Soil solutions were equilibrated for 24 h in an end-over-end shaker. After equilibration, solutions were centrifuged for 10 min at 2350 g and filtered through Whatman No. 1 filter paper (Whatman International Ltd, England). Filtrates were analyzed for inorganic P (Pi) by the ascorbic acid colorimetric method at a wavelength of 882 nm (Frank et al., 1998). Phosphorus sorbed was calculated as the difference between 66 the amount of P added initially and the amount remaining in the filtrate after equilibration. Data Analysis The Langmuir equation is given by: bkC Q-1+kC (1) where Q is the amount of P sorbed to the soil (mg P kg"), C is the amount of P in the solution afier equilibration (mg P L'l), b is the sorption maxima (mg P kg"), and k is the sorption strength (L mg’l). In this experiment, several of the data sets contained data points with negative sorption (desorption) and when nonlinear regression was used the NLIN procedure in SAS (SAS Institute, Inc., 1999) would not converge on the Langmuir equation. Therefore, it was necessary to correct these data sets for the amount of P initially sorbed, Q0, using the procedure outlined by Laboski and Lamb (2004). When Q, was not a significant parameter in the regression, it was set to zero. Corrected data sets were fit to equation 1 with nonlinear regression to obtain the estimates and standard errors of the estimates of the b and k parameters. For a given soil, nonlinear regression was used to compare b and k terms of the Langmuir equation between the incubated treatments and non-incubated, non-amended soil. The procedure used was similar to comparing regression coefficients in linear regression (Cook and Weisberg, 1999; Freud and Littell, 2000; Laboski and Lamb, 2004). The difference between the Bray l-P soil test of the incubated treated soils and the Bray l-P soil test of the control soil (no P added) after incubation is the increase in soil test P caused by application of manure or fertilizer. By subtracting the soil test P of the 67 control soil treatment, a correction was made for P mineralization that may have occurred during the incubation from rewetting dry soil. The increase in soil test P was then divided by the amount of total P applied, giving the increase in soil test P per P applied. This was necessary because the amount of P applied varied with treatment. Comparisons of the impact of the various soil amendments on the increase in soil test P per P applied were made by using analysis of variance (ANOVA) and Fisher’s Protected Least Significant Difference (LSD) for mean separation at the a=0.05 level. Relative P availability was calculated by dividing the increase in soil test P per P applied for each manure treatment by the increase in soil test P per P applied for the fertilizer treatment. The relative P availability of a manure treatment compared to fertilizer was considered significantly different than one if the ANOVA for the increase in soil test P per P applied for the two treatments was significantly different (SAS Institute, Inc., 1999). Relative P availability was then interpreted as follows: when the relative availability is equal to one, manure P is equally available as fertilizer P; when the relative availability is significantly greater than one, manure P is more available than fertilizer P; when the relative P availability is significantly less than one, manure P is less available than fertilizer P (Laboski and Lamb, 2003). Results and Discussion Phosphorus Sorption Capacity Sorption capacity was significantly reduced (a=0.05) by the application of dairy slurry for Colwood 1, Colwood 2, and Spinks soils, was unchanged for Oshtemo soil, and was significantly increased for Parkhill soil (Table 3.3, Figures 3.1, 3.2, and 3.3). 68 Table 3.3. Effect of manure and fertilizer application on soil sorption capacity (b) and sorption strength (k) as estimated by the Langmuir equation. Incubation Total P ------------ Langmuir Parameters ------------ Soil ID Treatment applied b k Q0§ R2 mg kg"l mg kg " L mg'1 mg kg '1 Colwoodl PRET 0 475a: 0.217a -- 0.98 Dairy 100 41 1b 0.126b -- 0.98 Fertilizer 100 41 Ob 0.1 12b -- 0.98 Swine 1 17 38% 0.1 13b -- 0.99 Colwood 2 PRE 0 178a 0.1213 21.04 0.97 Dairy 100 143b 0.094b 47.09 0.98 Fertilizer 100 14% 0.095b 71.05 0.96 Swine 1 17 161 ab 0.070b 61.29 0.94 Oshtemo PRE 0 106a 0.23 l a -- 0.96 Dairy 100 lOlab 0.210a 13.13 0.97 Fertilizer 100 92b 0.102b -- 0.96 Swine 1 17 95b 0.222a -- 0.96 Parkhill PRE 0 329a 0.200a -- 0.96 Dairy 100 388b 0.186a -- 0.95 Fertilizer 100 327a 0.162a -- 0.98 Swine 1 17 372ab 0.1643 -- 0.93 Spinks PRE 0 370a 0.31 1a -- 0.97 Dairy 100 324b 0.176b -- 0.98 Fertilizer 100 319b 0.1210 -- 0.98 Swine l 17 326b 0.161b -- 0.98 I Sorption characteristics for non-incubated non-amended soils. :Within a soil series, rows within a column followed by the same letter are not significantly different (oz=0.05). § Q, is an approximation of the amount of P sorbed to the soil prior to any experimentation. Q, was used to correct the data sets such that the NLIN procedure (SAS Institute, Inc., 1999) would converge on the Langmuir equation (Laboski and Lamb, 2004) 69 500 . . 450 : COIWOOd 1 .’ q 400 -i ———*"’"" ~i A 350': _: 34° 3005 -2 is" 1 2 v 250% 5 E 5 2 o 200‘; . "’ : : n. 150 j j 100 _‘ O PRE 5 : El dairy slurry 3 50 5 O fertilizer g f A swine slurry 3 O _ 1' I I I I I I I I I I I I I I — O 10 20 30 40 50 160 . 140-3 E 1205 .3 :~ 3 3 '3‘!) 100-: j °° : : g 80: j -0 . D . .. “E 605 E m u . O-I : 1 40 j 0 PRE : 2 CI dairyslurry 20: O fertilizer j i A swine slurry i 03 : I I r I r I I I I II I I I I f I 0 10 20 30 P in solution (mg L") l I I I I I 40 50 Figure 3.1. Sorption isotherms for Colwood 1 and Colwood 2 soils for non-incubated, non-amended (PRE) and amended incubation treatments. 70 160 1 1 Oshtemo 1 140 ‘7 j 120 5 —j A I O I '5.“ 10°; ___________ g 3 00 .4 . E 803 g g E : : 8 6O - .j 2. ~ . 40 1 o PRE 1 j C] dairy slurry 20 j 0 fertilizer 1 7’ A swine slurry O 7 4 - . . . . . .7 O 10 20 30 4O 50 500 3 . 450 _‘ Parkhlll “i 400-: A 350% E '52 300-: on 3 3 E. 250 : 1 s 2 a .e 200 1 1 8 E 1 9.. 150 1 1 1 PRE 1 1003 dairy slurry 50 «j fertilizer 5 0 j swine slurry 3 O 10 20 3O 40 50 P in solution (mg L'l) Figure 3.2. Sorption isotherms for Oshtemo and Parkhill soils for non-incubated, non- amended (PRE) and amended incubation treatments. 71 500 : Spinks : 400 3 - q 1 f3 300<3 3 E .. . :§ 200 d ‘ 2.. j I ‘ O PRE - 100 2 [3 dairy slurry : j 0 fertilizer : . A swine slurry . O - - II TTII1TIIIIIIIIIIIIfiIIIfIIII 0 10 20 30 40 50 P in solution (mg L'l) Figure 3.3. Sorption isotherms for Spinks soil for non-incubated, non-amended (PRE) and amended incubation treatments. Application of swine slurry significantly reduced sorption capacity in the Colwood 1, Oshtemo and Spinks soils, but did not change sorption capacity for Colwood 2 and Parkhill soils. Sorption capacity was unchanged by fertilizer application to the Parkhill soil but was significantly reduced for all other soils. Siddique and Robinson (2003) also found significant reduction in sorption capacity with the application of poultry litter, poultry manure, cattle slurry, municipal sewage sludge, or fertilizer at a rate of 100 mg P kg‘I during an incubation study. Field studies have also found a reduction in sorption capacity with manure and fertilizer application (Sharpley et al., 1993; Holford et al., 1997; Laboski and Lamb, 2004). The 72 reduction in P sorption capacity observed with application of manure slurry may be caused by microbial decomposition of soil organic matter and manure producing organic acids that compete with P for surface binding sites (Swenson etal., 1949; Nagarajah et al., 197 0; Kafltafi et al., 1998). The reduction in sorption capacity after application of fertilizer is not well understood but may be caused by desorption of initially bound P. However, this amount of P can theoretically be accounted for by correcting the data set using Q0. This procedure was used for the Colwood 2 and Oshtemo soils in this experiment. However, sorption capacity for the Colwood 2 soil was still reduced, thus Q0 may not be accounting for the total amount of P initially sorbed. If the amount of P initially sorbed is not accounted for then sorption capacity may be reduced because some binding sites were already occupied. There are far fewer studies that have reported an increase in sorption capacity with manure application (Mozaffari and Sims, 1994; Sharpley, 1996, Laboski and Lamb, 2004). Explanation of this phenomenon is limited. Laboski and Lamb (2004) found that one of seven soils had increased sorption capacity with manure application history. For this soil they found an increase in extractable Fe and Al compared to an adjacent soil in the same soil mapping unit with no manure history. There may have been more P binding sites in this manured soil; however, greater concentrations of extractable Fe and Al were found in other manured soils which had reduced sorption capacity. In the present study Fe and Al in the soils was not measured, thus these relationships can not be explored at this time. Phosphorus Sorption Strength 73 Dairy slurry or swine slurry application significantly reduced (o:=0.05) sorption strength in the Colwood 1, Colwood 2, and Spinks soils, but did not significantly reduce sorption strength in the Oshtemo and Parkhill soils. Fertilizer application significantly reduced sorption strength in the Colwood 1, Colwood 2, Oshtemo, and Spinks soils, but did not significantly reduced sorption strength in the Parkhill soil. For a given soil, application of either manure slurry or fertilizer reduced sorption strength compared to the sorption strength of the non-incubated, non-ammended soil. This supports other studies that have also found a reduction in sorption strength when manures and fertilizers were applied (Holford et al., 1997; Siddique and Robinson, 2003; Laboski and Lamb, 2004). Holford et al. (1997) suggested that the reduction in sorption strength with manure or effluent application history was likely due to organic anion interactions with sorbing P anions thus, lowering the sorption strength. Additionally, Holford et al. (1997) described sorption strength estimated from the Langmuir equation as an average of the total sorption sites which includes high and low strength binding sites in the soil. Holford et al. (1997) suggests that P initially sorbs to high strength sites followed by low bonding strength sites. Thus, reduced sorption strength with the addition of manure or fertilizer may also be caused by the binding of the applied P to lower strength sites because the high strength sorption sites may already be occupied. Phosphorus Availability For Colwood 1, Colwood 2, and Parkhill soils the increase in soil test P per P applied was significantly different (a=0.05) between all sources (Figure 3.4). This increase in soil test P per P applied for these soils was greatest for the fertilizer followed by swine slurry and dairy slurry, respectively. For the Oshtemo and Spinks soils, swine 74 Fertilizer 0.8 j [:1 Swine I - Dairy 0.6 -f 0.4 i 0.2 : Increase in 5011 test P per P applied . I . . . i I I . . .. . - I ~. rm: .9... Is . {4“}; ism"? :3 "g N o- O 0- O r! 5 . (I; TVE‘VF: .3“ 516‘} \- 01% I. . .318 .. .32: “II. ”I: ‘” _ I 0" '71:. - ‘1‘ I: N 0* O o‘ .I......il.........l.........l...... 1.1.... Colwoodl Colwood 2 Oshtemo Parkhill Spinks Soil Figure 3.4. Relative increase in Bray l-P soil test level per P applied during a six week incubation of selected Michigan soils with fertilizer, swine slurry, and dairy slurry. For a given soil, P sources with the same letter are not significantly different (o.=0.05). Table 3.4. Relative availability of P from dairy and swine slurries compared to fertilizer. Soil Dairy Swine Colwood 1 056* 0.95"“ Colwood 2 0.64* 0.90* Oshtemo 0.75* 1.01 Parkhill 0.45* 0.74* Spinks 0.69* 0.99 Mean 0.62 0.92 * Significantly different than fertilizer P (a=0.05). 75 slurry application increased soil test P per P applied as much as fertilizer application whereas, dairy slurry application increased soil test P significantly less than fertilizer or swine slurry application. Relative P availability from the different P sources were as follows: dairy slurry < swine slurry S fertilizer (Table 3.4). Relative P availability from swine slurry was significantly (oz=0.05) less than fertilizer P in the Colwood l, Colwood 2, and Parkhill soils but was as available as fertilizer P in the Oshtemo and Spinks soils. Relative P availability fi'om dairy slurry is significantly less than fertilizer P for all soils, with Parkhill soil having the lowest relative availability. The relative P availability of dairy slurry is less than the relative P availability of swine slurry for all soils. The low relative P availability of dairy slurry for the Parkhill soil may be explained by the sorption characteristics of this soil after incubation with the various amendments. This was the only soil where the sorption capacity increased after application of dairy or swine slurries compared to fertilizer. Therefore, the increase in sorption capacity, which was greatest for the dairy slurry amendment likely resulted in a low relative P availability. Dairy slurry application increased sorption capacity more than swine slurry application, though not significant, and had a lower relative P availability compared to swine slurry. It is also important to note that the sorption strength for this soil was unchanged by all amendments compared to the non-incubated, non-amended soil. The low relative P availability of this soil may also be caused by its soil characteristics (Table 3.1). This soil has high clay and organic matter contents and has a low Bray P-l soil test level. The low relative P availability is likely caused by its 76 greater sorption capacity and strength or binding affinity for P because of its low P soil test level. In addition, P sorption capacity is generally greater for soils with large concentrations of clay and or organic matter. Oshtemo and Spinks soils were the only soils where the relative P availability of swine slurry was not significantly different than fertilizer. For these soils the change in sorption capacity between the non-incubated, non-amended soil and the swine slurry or fertilizer amended soil was very similar. For all other soils, the relative P availability of swine slurry was significantly less than fertilizer and the magnitude of the change in sorption capacity upon treatment was much greater than the non-incubated, non-amended soils. Additionally, the Oshtemo soil was the only soil where fertilizer application was the only amendment that significantly reduced sorption strength. However, the sorption strength for Spinks soil was significantly reduced by all amendment treatments. The Oshtemo and Spinks soils were the sandiest (68 % and 76 % respectively) compared to the other soils (32-48 % sand). Therefore, the high relative P availability for these soils may be caused by the low sorptive properties of these soils. The difference in relative P availability observed for swine and dairy slurries may be explained by the sorption characteristics measured when the slurries or KH2P04 were used as the P source in a sorption experiment (Chapter 2). The P, sorption capacities for all soils when dairy slurry, swine slurry, and KH2P04 were used as the P source decreased in the following order: dairy slurry > KH2P04 2 swine slurry; with the magnitude of the change in P, sorption capacity between dairy slurry and KH2P04 being larger than the change in P.- sorption between swine slurry and KH2P04. Therefore, dairy slurry having a greater P, sorption capacity than KH2P04, when used as the P source, 77 likely would have a lower relative P availability whereas, swine slurry having a reduced or equivalent Pi sorption capacity compared to KH2P04 likely would have a lower or more similar relative P availability. The Pi sorption strength is less, however not always significant, for the dairy and swine slurries compared to KH2P04 when used as the P source during the sorption experiment. Additionally, dairy slurry had a lower, however not always significant, P, sorption strength than swine slurry for all soils except Colwood 2 and Oshtemo. This smaller Pi sorption strength observed when dairy slurry was the P source compared to swine slurry may suggest that dairy slurry would be more available than swine slurry. However, this is not what was shown in the present experiment. These findings do not concur with the conclusions made by Holford et al. (1997) and Laboski and Lamb (2004) that sorption strength may influence P availability more than sorption capacity when manure and fertilizer were applied to soils in a field setting. P sorption capacity was much greater when dairy slurry was used as the P source in sorption experiments compared to swine or KH2P04. Thus, in the present study it is likely that the increase in sorption capacity and reduction in sorption strength interact to affect P availability. Dairy slurry had a mean relative P availability of 0.62 (62 %) and P, was 79 % of P,. The mean relative P availability of swine slurry was 0.92 (92 %) and P, was 88 % of Pt. Thus, the relative P availability for a given slurry may be similar to but is not equivalent to percent P,. If percent P, was used to estimate relative P availability, then P availability would be overestimated for dairy slurry and underestimated for swine slurry based on the data in this study. Currently, land grant universities in the Midwest are recommending that P from manure be considered 50 % to 100 % available to plants in the 78 first year of application regardless of animal species (Jacobs, 1995; Busch et a1. 1997; Madison et al., 1998; Sutton et al., 2001; Beegle and Curran, 2004). Clearly this one size fits all approach is not appropriate to estimate P availability to crops grown on soils with soil test levels less than or equal to agronomically optimum. Conclusions Results of this study show that sorption characteristics can increase, decrease, or remain unchanged with manure or fertilizer application. In this study, at least half the soils had reduced sorption capacity and strength with a manure or fertilizer application. Dairy slurry application always increased soil test levels less than swine slurry or fertilizer application. These data show that the application of P from swine slurry is more effective in increasing soil test P than dairy slurry over a range of soil textures. These results are supported by other research conducted on these selected Michigan soils where Pi in dairy slurry sorbed to soil more than KH2P04 and P, in swine slurry sorbed to soil less than KH2P04. Greater sorption of P, from dairy slurry leads to lower P availability compared to swine slurry where less Pi sorption of swine slurry leads to similar or less relative P availability. As shown in this study, the relative availability of P, as measured by Bray l-P, from swine and dairy slurries, is 92 % and 62 % respectively compared to fertilizer. This suggests that P availability to plants should not be generalized for manure from all animal species. Rather, P availability coefficients for manures from different animal species should be developed to provide better guidance when developing nutrient management 79 recommendations. Further studies are needed in the field and laboratory to determine P availability from manure to plants in the first through third year after application. 80 References Beegle, DB. and W.S. Curran. 2004. The Agronomy Guide. The Pennsylvania State University. ARS-26. Brown, J .R. 1998. Recommended Chemical Soil Test Procedures for the North Central Region. North Central Regional Research Publication No. 221 (Revised). Missouri Agricultral Experiment Station. SBlOOl. Busch, D., L. Busman, P. Nesse. 1997. Developing a Manure Management Plan. A planning and record keeping guide. Minnesota Extension Services. B4-6957—D. Cook, RD. and S. Weisburg. 1999. Applied regression including computing and graphics. John Wiley&Sons, New York. Ebeling A.M., L.R. Cooperband, and LG. Bundy. 2003. Phosphorus source effects on soil test phosphorus and forms of phosphorus on soil. Commun. Soil Sci. Plant Anal. 34:1897-1917. Frank, K.D., D. Beegle, J. Denning. 1998. Phosphorus. In Recommended Chemical Soil Test Procedures of the North Central Region. North Central Region Research Pub. No. 221 (revised). Missouri Agriculture Experiment Station. SBlOOl. Freud, R.J. and RC. Little. 2000. SAS system for Regression Ananlysis, 3rd Ed. SAS Institute, Inc., Cary, NC. Gee, G.W. and J .W. Bauder. 1986. Particle-size Analysis. Pg 383-411. In A.Klute (ed.) Methods of Soil Analysis Partl- Physical and Mineralogical Methods. American Society of Agronomy, Madison, WI. Holford, I.C.R., C. Hird, and R. Lawrie. 1997. Effects of animal effluent on the phosphorus sorption characteristics of soil. Aust. J. Soil Res. 35:365-73. Jacobs, L.W. 1995. Uilization of Animal Manure for Crop Production Part II. Manure Application to Cropland. Michigan State University Extension. Bulletin MM-2. Kafkafi, U., B. Bar—Yosef, R. Rosenburg, and G. Sposito. 1998. Phosphorus adsorption by kaolinite and montrnorillonite: H. Organic anion competition. Soil Sci. Soc. Am. J. 52:1585-1589. Laboski, C.A.M. and J .A. Lamb. 2003. Changes in soil test phosphorus concentrations after application of manure or fertilizer. Soil Sci. Soc. Am. J. 67: 544-554. Laboski, C.A.M. and J .A. Lamb. 2004. Impact of manure application on soil phosphorus sorption characteristics and subsequent water quality implications. Soil Sci. 440-448. 81 Madison, F ., K. Kelling, L.Massie, and L.W. Good. 1998. Guidelines for Applying Manure to Cropland and Pastures in Wisconsin. University of Wisconsin Extension. A3392. Mozaffari, M. and J .T. Sims. 1994. Phosphorus availability and sorption in an atlantic coastal plain watershed dominated by animal-based agriculture. Soil Sci. 157297-107. Nagarajah, S., A.M. Posner, and J .P. Quirk. 1970. Competitive adsorption of phosphate with polygalacturonate and other organic anions on kaolinite and oxide surfaces. Nature (London) 228:83-85. Nair, P.S., T.J. Logan, A.N. Sharpley, L.E. Sommers, M.A. Tabatabai, and TL. Yuan. 1984. Interlaboratory comparisons of a standardized phosphorus adsorption procedure. J. Environ. Qual. 13:591-595. Peters, J .(ed). 2003. Recommended Methods of Manure Analysis. Bulletin A3769, University of Wisconsin Extension. Cooperative Extension Publishing, Madison, WI. I-2/2003. Rawls, W.J., D.L. Brakensiek, and K.E. Saxton. 1982. Estimation of soil water properties. Trans ASAE 25:1316-1320. SAS Institute, Inc. 1999. SAS 8.1. SAS Institute. Inc, Cary, NC. Sharpley, A.N. (1996). Availability of residual phosphorus in manured soils. Soil Sci. Soc. Am. J. 60:1459-1466. Sharpley, A.N., J .J . Meisinger, A. Breeuwsma, J .T. Sims, T.C. Daniels, and J .S. Schepers. 1998. Impacts of animal manure management on ground and surface water quality. P. 173-242. In J .L. Hatfield and BA. Stewart (ed.) Animal waste utilization: Effective use of manure as a soil resource. Ann Arbor Press, Chelsea, MI. Sharpley AN. and R.G. Menzel. 1987. The impact of soil and fertilizer phosphorus on the environment. Adv. Agron. 41 :297-324. Sharpley, AN. and I. Sisak. 1997. Differential availability of manure and inorganic sources of phosphorus in soil. Soil Sci. Soc. Am. J. 61 :1503-1508. Sharpley, A.N., S.J. Smith, and W.R. Bain. 1993. Nitrogen and phosphorus fate from long-term poultry litter applications to Oklahoma soils. Soil Sci. Soc. Am. J. 57:1131- 1137. Siddique, MT and J .8. Robinson. 2003. Phosphorus sorption and availability in soils amended with animal manures and sewage sludge. J. Environ. Qual. 32:1114-1121. 82 Sims, J .T. 1993. Environmental soil testing for phosphorus. J. Prod. Agric. 6:501-507. Sutton, A.L., D.D. Jones, B.C. Joern, D.M. Huber. 2001. Animal Manure as a Plant Nutrient Resource. Purdue University Cooperative Extension Resource. ID-lOl Swenson, R.M., C.V. Cole, and DH. Sieling. 1949. Fixation of phosphate by iron and aluminum and replacement by organic and inorganic anions. Soil Sci. 67:3-22. 83 APPENDICIES 84 APPENDIX A Determination of optimum rate of chloroform for microbial suppression in sorption solutions where manure is the P source Experiment 1 Objective The standard P sorption procedure proposed by Nair et al. (1984) suggests the use of 20 g L'1 of chloroform for microbial inhibition where KH2P04 is the P source of the sorption solutions. However, a different rate of chloroform would likely be needed when manure is the P source of the sorption solutions because there would be more microbial activity than when KHZPO4 is the P source. Therefore, an experiment was designed to determine the rate of chloroform needed to inhibit microbial activity during the 24 h equilibration period when manure is the P source. The objective was to suppress the microorganisms to the point where no change in P, could be detected between the chloroform treated equilibrated solution and a solution of the same concentration kept at 4 °C until analysis, in which it is assumed no net microbial activity occurs. Methods and Materials To determine if a rate of chloroform greater than 20 g L", as proposed by Nair et a1. (1984), was adequate for microbial suppression when manure is the source of the P for sorption solutions, chloroform (in 0.75% ethanol; guaranteed reagent; EM Science, Gibbstown, NJ) was added at a rate of 0 ml, 0.63 ml, or 1.25 ml to 25 ml of solution where dairy slurry or swine slurry is the P source. Solutions of approximately 85 75 mg total P L'1 were prepared in 0.1 M KCl at pH 6 for each manure species. The rates of chloroform added are equivalent to 0 g L'l, 36.9 g L'l, and 73.7 g L", respectively. Solutions were equilibrated for 24 h at room temperature (approximately 25 °C) in an end-over-end shaker to simulate the conditions under which sorption experiments would take place. After equilibration, solutions were centrifuged for 10 min at 2350 g and filtered through Whatman No. 1 filter paper. Aliquots of the dairy slurry and swine slurry solutions kept at 4 °C were centrifuged and filtered similarly prior to colorimetric analysis. All samples were diluted and analyzed using ascorbic acid color development at a wavelength of 882 nm (Frank et al., 1998). All treatments were run in duplicate. Results and Discussion When the 0.63 ml and 1.25 ml rates of chloroform were added to the slurry solutions in 50 ml centrifuge tubes (Corning Inc., Life Sciences; Acton, MA) for a 24 h equilibration period, the tubes became deformed. This is likely due to a reaction of the chloroform with the polypropylene tube. Therefore, these rates can not be used with the previously mentioned tubes for the sorption experiments. Thus, other lower rates of chloroform were investigated for effectiveness of microbial inhibition. Experiment 2 Objective The objective of this experiment was to determine if lower rates of chloroform, closer to the rate proposed by Nair et al. (1984), are effective in suppressing microbial activity in a sorption solution of greatest P concentration, where dairy and swine slurry 86 was the source of P, for the 24 h equilibration period. Effective microbial suppression was determined to be where no change in Pi could be detected between the equilibrated chloroform treated solution and a solution of the same concentration kept at 4 °C, in which it is assumed no net microbial activity occurs. Methods and Materials Chloroform was added at a rate of 0 ml, 0.16 ml, or 0.31 ml to 25 ml of solution where dairy slurry or swine slurry was the P source. Solutions of approximately 75 mg total P L'I were prepared in 0.1 M KC1 at pH 6 for each manure P source. The rates of chloroform added are equivalent to 0 g L", 9.2 g L'l, and 18.4 g L'l, respectively. Solutions were equilibrated for 24 h at room temperature (approximately 25 °C) in an end-over-end shaker to simulate the conditions under which sorption experiments would take place. After equilibration, solutions were centrifuged for 10 min at 2350 g and filtered through Whatman No. 1 filter paper. Solutions of dairy and swine slurry kept at 4 °C were centrifuged and filtered similarly just prior to colorimetric analysis. All samples were diluted and analyzed using ascorbic acid color development at a wavelength of 882 nm (Frank et al., 1998). All treatments were run in duplicate. Results and Discussion The addition of chloroform at a rate of either 0.16 ml or 0.31 ml to 25 ml of solution were not significantly different (a=0.05) in suppressing microbial activity for both manure species (Table A.1). For dairy slurry, addition of either 0.16 ml or 0.31 ml of chloroform to 25 m1 of solution was significantly better at reducing microbial activity 87 Table A. 1. Comparison of the ability of various rates of chloroform to suppress microbial activity for the 24 h equilibration period. ----- P Source ----- Rate of Chloroform Dairy Swine ml P, in Solution mg kg'1 0 (4 °C) 47.7621" 73.01a 0 (25 °C) 44.16b 69.10b 0.16 (25 °C) 46.64c 71.84ab 0.31 (25 °C) 46.70c 69.84ab LSD (a=0.05) 0.841 3.77 +Means within a row followed by the same letter are not significantly different as determined by Fisher’s Protected Least Significant Difference (LSD) at the a=0.05 level. than 0 ml of chloroform at 25 °C, however, both chloroform rates resulted in significantly lower amounts of P; in solution than the control at 4 °C. The addition of either 0.16 ml or 0.31 ml of chloroform to 25 ml of solution to the swine slurry solution resulted in statistically similar concentrations of P; in solution compared to the addition of no chloroform at either 4 °C or 25 °C, however, the concentrations of P; in solution were significantly different for the two treatments with no chloroform addition (25 °C or 4 °C). Overall, it appears that the addition of either 0.16 ml of chloroform to 25 ml of solution or 0.31 ml of chloroform to 25 ml of solution will give adequate microbial suppression for swine slurry solutions for the 24 h equilibration period as compared to the control at 4 °C. As for the dairy solution, it appears that these rates of chloroform may not be completely suppressing microbial activity as compared to the control at 4 °C, however, the treatment means are very similar. Thus, significant differences between means may be caused by the small number of replications. Because there was no difference in P; solution concentrations between these rates of chloroform for both 88 manure species chloroform at a rate of 0.16 ml of chloroform to 25 ml of solution will further be tested for its effectiveness in the suppression of microbial activity over a range of P solution concentrations where KHzPO4, dairy slurry, or swine slurry is the P source of the sorption solution. It is also important to determine if this rate of chloroform is adequate to suppress microbial activity when solutions are equilibrated with soil. Experiment 3 Objective A study was conducted to determine if the rate of 0.16 ml chloroform to 25 ml of solution would be sufficient for microbial suppression with the addition of soil. Degradation of the centrifuge tubes was also a concern after the observed visual deformation with high rates of chloroform. It was hypothesized that P, would be different, when KHzPO4 was used as the P source, without soil, with and without chloroform addition, if any small amount of tube deformation caused changes in the P sorptiveness of the centrifuge tubes. Methods and Materials For this experiment three soil variables (no soil, Parkhill, or Spinks) were equilibrated with sorption solutions prepared from each P source (dairy slurry, swine slurry, or KHzPO4) at a total P solution concentration of approximately 5 or 75 mg P L". Sorption solutions were prepared in 0.1 M KCl at pH 6. Solutions were added at 1:25 (w/v), soil to solution ratio, either with or without the addition of chloroform at a rate of 0.16 ml chloroform to 25 ml of solution. Samples were equilibrated for 24 h at 89 room temperature (approximately 25 °C) in an end-over-end shaker. Afier equilibration, samples were centrifuged at 2350 g for 10 min and filtered through Whatman No. 1 filter paper (Whatman International Ltd, England). P; in solution was measured by ascorbic acid color development at a wavelength of 882 nm (Frank et al., 1998). All treatments were run in triplicate with treatment means given in Table A2. Table A2. Comparison of the efficacy of 0.16 ml of chloroform to 25 ml of solution to suppress microbial growth when added to different soils and P sources. Chloroform rate Treatment“ 0 ml 0.16 ml P, in solution (mg L'l) None x D x 5 2.70a* 3.421» None x S x 5 4.79a 5.59b NonexeS 5.15a 5.13a None x D x 75 48.8a 50.2b None x S x 75 73.0a 73.6a None x F x 75 78.4a 78.9a Parkhill x D x 5 0.257a 0.899b Parkhill x S x 5 0.758a 1.78b Parkhill x F x 5 1.43a 1.54b Parkhill x D x 75 38.2a 37.9a Parkhill x S x 75 57.73 59.03 Parkhill x F x 75 63.3a 62.4a Spinks x D x 5 0.210a 0.436b Spinks x S x 5 0.5723 0.875b Spinks x F x 5 0.700a 0.867b Spinks x D x 75 27.6a 29.2b Spinks x S x 75 50.2a 52.0b Spinks x F x 75 58.7a 57.5a * Treatment = Soil x P source x approximate total P concentration of sorption solution in mg P L", where D= dairy slurry, S=swine slurry, and F= KHzPO4. iMeans within a row followed by the same letter are not significantly different as determined by Fisher’s Protected Least Significant Difference (LSD) at the (F005 level. 90 Results and Discussion When no soil was present where KH2P04 is the P source, at either solution P concentration, P, in solution was not significantly (a=0.05) different with or without chloroform, thus chloroform is not changing the properties of the tube such that P sorption is impacted (Table A.2). For both soils, the addition of chloroform resulted in a significantly (a=0.05) greater amount of P; in solution at the 5 mg P L'1 solution concentration where KHzPO4 was the P source. For both soils, at the 75 mg P L'1 solution concentration, where KH2P04 was the P source, the chloroform treatments were not significantly different (a=0.05). This may be because at the higher solution ' concentrations there is a large enough concentration of P that a change in solution Pi is not noticed when the microbes are suppressed. Whereas at the lower solution concentrations, there is a small enough amount of P that changes in P, solution concentrations are more noticeable. For both soils, the addition of chloroform resulted in a significantly (a=0.05) greater concentration of P, in solution at the 5 mg P L'1 solution concentration when dairy slurry or swine slurry was the P source. For the Spinks soil, the addition of chloroform resulted in a significantly greater concentration of P; in solution for the 75 mg P L'1 solution concentration when dairy slurry or swine slurry was the P source. In contrast, for the Parkhill soil, the concentration of P; in solution for the 75 mg P L'1 solution concentration was not significantly different with or without the addition of chloroform. The observed difference in microbial suppression between soils at the greater P solution concentration may be due to the differences in microbial communities between the soils. 91 Overall, the addition of chloroform at a rate of 0.16 ml to 25 ml of solution is sufficient in suppressing the activity of microorganisms. Conclusions Based on these results, the addition of chloroform at a rate of 0.16 ml to 25 ml of solution is sufficient in suppressing the activity of microorganisms for the 24 h equilibration period for all P sources. For precautionary purposes centrifuge tubes should only be used once for the batch equilibration process. 92 References Nair, P.S., T.J. Logan, A.N. Sharpley, L.E. Sommers, M.A. Tabatabai, and TL. Yuan. 1984. Interlaboratory comparisons of a standardized phosphorus adsorption procedure. J. Environ. Qual. 13:591-595. 93 APPENDIX B Determination of effect of filtration method on inorganic P solution concentration Objective The standard P sorption method published by Nair et al. (1984) proposed the use of 0.45 pm membrane filtration of equilibrated, centrifuged samples. However, this method of filtration is costly and time consuming. Therefore, the objective of this study was to determine if there was a difference in inorganic P (P,) concentration in equilibrated solutions when filtered through 0.45 pm filter units (MillexTM Millipore Corporation, Bedford, MA) or Whatman No. 1 filter paper (Whatman International Ltd, England). Methods and Materials Sorption solutions with 0-30 mg P L'1 in the form of KHZPO4 were prepared in unbuffered 0.01 M CaClz (Nair et al., 1984). Solutions were added to the Oshtemo soil at a 1:25 (w/v) soil to solution ratio. Solutions were equilibrated for 24 h in an end-over- end shaker. After equilibration, solutions were centrifuged for 10 min at 2350 g. Solutions were then filtered using one of the three filter treatments; (i)Whatman No. 1 filter paper, (ii) 0.45 um syringe filter unit, or (iii) unfiltered (control). All treatments were performed in triplicate. P, solution concentrations for each treatment were determined by analysis with ascorbic acid color development at a wavelength of 882 nm (Frank et al., 1998). Differences in P, in solution were determined by Fisher’s Protected Least Significant Difference (LSD) at the a=0.05 level. 94 Results and Discussion For all solution P concentrations, filtration treatments are not significantly (o:=0.05) different, except for the 30 mg P L'1 solution concentration where Whatman No. 1 filter paper had significantly greater P, in solution compared to the control and the 0.45 pm syringe filter unit (Table B. 1). These results suggest that Whatman No. 1 filter paper is adequate for filtration of equilibrated solutions. Kleinman et a1. (2002) found that the effect of filtration method (Whatman No.1 and 0.45 pm membranes) on water extractable P concentrations of manure varied with manure species but that Whatman No. 1 filter paper was adequate. Based on the results fi'om this experiment for sorption solutions using KHzPO4 as the P source and the findings of Kleinman et a1. (2002) for manure solutions, Whatman No. 1 filter paper was determined to be adequate for filtration of P sorption solutions for all P sources after centrifugation. Table B. 1. Mean inorganic P concentration in solution after 24 h equilibration period with Oshtemo soil followed by one of three filtration treatments. Filtration Method P solution Unfiltered Whatman No. 1 0.45 pm syringe concentration filter paper“ filter unit: 0 0.03321§ -0.0033a 0.00a 2 0.737a 0.787a 0.793a 4 2.22a 2.21a 2.21a 6 3.59a 3.83a 3.80a 10 7.61a 7.44a 7.37a 16 12.98a 13.13a 12.93a 20 16.92a 17.02a 16.67a 30 26.55a 27.36b 26.36a + Whatman International Ltd, England 3' MillexTM Millipore Corporation, Bedford, MA § Means across rows followed by the same letter are not significantly different (0:005). 95 References Frank, K., D. Beegle, J. Denning. 1998. Phosphorus. In Recommended Chemical Soil Test Procedures of the North Central Region. North Central Region Research Pub No. 221 (revised) Missouri Agricultural Experiment Station. SBlOOl. Kleinman, P.J.A., A.N. Sharpley, A.M. Wolf, D.B. Beegle, P.A. Moore, Jr. 2002. Measuring water-extractable phosphorus in manure as an indicator of phosphorus in runoff. Soil Sci. Soc. Am. J. 66:2009-2015. Nair, P.S., T.J. Logan, A.N. Sharpley, L.E. Sommers, M.A. Tabatabai, and TL. Yuan. 1984. Interlaboratory comparisons of a standardized phosphorus adsorption procedure. J. Environ. Qual. 13:591-595. 96 APPENDIX C Percent removal of inorganic P and total P from manure sorption solutions by centrifugation Objective Before inorganic P (Pi) and total P (Pt) analysis for sorption experiments can be conducted, centrifugation of equilibrated solutions is necessary to separate soil particles from solution. However, this poses a potential problem with Pt measurement because the heavier particles in the manure will likely also be removed from solution by centrifugation, resulting in a skewed Pt measurement. Thus, the objective of this study was to determine the percent Pi and Pt removal by centrifugation in manure sorption solutions. Methods and Materials Percent removal of Pi and Pt by centrifugation of manure sorption solutions was determined experimentally by comparing the Pi and P, concentrations in centrifuged and uncentrifuged samples. A representative set of sorption solutions for each manure source were prepared on a P. basis, using the nutrient content of whole manure, in 0.1 M KCl at pH 6, to obtain Pt concentrations of approximately 0, 5, 20, 50 and 100 mg P, L'1 for dairy slurry and O, 5, 15, 30, and 75 mg P, L’1 for swine slurry. Solutions were added to 50 ml centrifuge tubes (Corning Inc., Life Sciences; Acton, MA) with the addition of 0.16 ml of chloroform (in 0.75% ethanol; guaranteed reagent; EM Science, Gibbstown, NJ) to 25 ml of solution for microbial inhibition. Solutions were equilibrated for 24 h in 97 an end-over-end shaker at room temperature. After equilibration, one set of five replications of solutions for each species was centrifuged for 10 min at 2350 g while the other set of five replications of solutions was not centrifuged. Aliquots of each solution were reserved for Pt digestion prior to filtration. The remaining portion of each solution was filtered through Whatman No. 1 filter paper for P; analysis. Filtered solutions, centrifuged and uncentrifuged, were analyzed immediately for P; by the ascorbic acid colorimetric method at a wavelength of 882 nm (Frank et al., 1998). Samples were digested according to the persulfate oxidation method of Bender and Wood (2000) with some modifications. Aliquots of equilibrated samples (10 or 5 ml) were diluted to a final volume of 50 ml with distilled water. Diluted samples were then digested with 0.7 g potassium persulfate and 1 ml of 11 N H2804 (Appendix D). Samples were first digested for 1.25 h at approximately 100 °C, and then the hotplate was turned to the maximum setting to increase boiling until the samples were reduced to approximately 10 ml. Cooled samples were diluted with 30 ml of distilled water. After samples were diluted, approximately 5 ml of 2 M NaOH was added to adjust the pH of the solution. Solutions were transferred to volumetric flasks and diluted to a final volume of 100 ml. Phosphorus in diluted, digested samples was analyzed by the ascorbic acid colorimetric method at a wavelength of 882 nm (Frank et al., 1998). Digested standards of KHZPO4 were used to produce the standard curve for determination of the concentration of P, in solution. 98 Data Analysis Differences in Pi and Pt solution concentrations in centrifirged and uncentrifuged solutions were determined by Fisher’s Protected Least Significant Difference (LSD) at a=0.05. Percent Pi and Pt removal by centrifugation was determined on an uncentrifuged basis. Variation in Pt removal for different solution concentrations was determined by Fisher’ Protected Least Significant Difference (LSD) at a=0.05. Results and Discussion Centrifugation had no effect on P; solution concentration for all solution concentrations for swine slurry (Table C. 1). Whereas, for dairy slurry the uncentrifuged samples had significantly greater P, in solution for the 20 and 50 mg P.- L'I solution concentrations than centrifuged samples. All other solution concentrations (0, 5, and 100 mg P; L") for dairy slurry had similar Pi solution concentrations for centrifuged and uncentrifuged samples. All solution concentrations, except 0 mg L", for both dairy and swine slurries, had significantly greater Pt in solution for the uncentrifuged samples compared to centrifuged samples (Table C.2). Centrifugation did not remove P, differently over range of solutions for either dairy or swine slurries (Table C.3). For dairy slurry, Pt removal was similar for the 20, 50, and 100 mg P. L" solutions and was significantly (or=0.05) greater for these solutions compared to the 5 mg P, L" solution. For swine slurry, Pt removal was similar in the 5, 15, and 30 mg P, L" solutions, whereas P, was removed by centrifugation at a significantly greater percentage for the 75 mg Pt L'I solution as compared these solution concentrations. P,- as a percent of PI is fairly consistent over the range of P solution 99 concentrations for centrifuged and uncentrifuged samples for both species (Table C .4). Thus, there was a general consistency in the composition of P (Pi and P0) in solution over the range of sorption solutions. Table C. 1. Effect of centrifirgation on inorganic P (P,) solution concentration. P Source Solution ------ Mean P; in solution ------ P value: concentration]r Uncentrifuged Centrifuged mg Pt L" mg L'1 ------------ Dairy . 0 0.02 0.02 0.4714 5 3.24 3.25 0.8597 20 13.21 13.06 0.0241 50 34.16 33.21 0.0398 100 65.69 65.70 0.1853 Swine 0 0.02 0.03 0.3972 5 4.20 4.24 0.1248 15 13.53 13.51 0.7639 30 25.06 25.54 0.2676 75 60.93 61.39 0.5844 l Approximate total P (P,) concentration of sorption solutions based on the nutrient content of whole manure. : Significance for comparison of mean inorganic P (P,) in solution for uncentrifuged and centrifuged samples. 100 Table C.2. Effect of centrifugation on total P (P,) solution concentration. P Source Solution , ----- Mean P, in solution ----- P value: concentration Uncentrifuged Centrifuged mg P. L" -------------- mg L'1 ----------- Dairy 5 4.47 4.04 0.0295 20 17.99 15.52 <0.0001 50 46.57 38.99 0.0001 100 90.59 76.98 <0.0001 Swine 5 4.66 4.44 0.0004 15 14.88 14.07 <0.0001 30 29.35 26.79 <0.0001 75 74.85 62.09 <0.0001 l Approximate total P (P,) concentration of sorption solutions based on the nutrient content of whole manure. * Significance for comparison of mean total P (P,) in solution for uncentrifuged and centrifuged samples. Table C.3. Percent inorganic P (P,) and total P (P,) removal by centrifugation for varying solution concentrations. For a given P source and P component (Pi and Pt), removal is compared over the range of solution concentrations. P Source Solution Mean Pi Mean Pt concentrationf removal removal mg P, L" % % Dairy 5 -0.13 9.8 20 1.13 13.7 50 2.74 16.3 100 0.94 15.0 LSD (a=0.05) NsI 2.14 Swine 5 -0.87 4.6 15 0.12 5.4 30 -1.9 8.7 75 -0.77 17.1 LSD (oz=0.05) NS 3.93 l Approximate total P (Pt) concentration of sorption solutions based on the nutrient content of whole manure. * NS=Not significant at a=0.05. 101 Table C.4. Percent inorganic P (P,) of total P (P,) for centrifuged and uncentrifuged samples. P Source Solution P; as a percent P; as a percent concentration,r of P, of Pt uncentrifuged centrifuged mg P. L" % % Dairy 5 72 80 20 73 85 50 73 88 100 73 85 LSD (0.05) Ns" 2.14§ P value 0.7990 <0.0001 Swine 5 90 95 15 91 96 30 85 94 75 81 98 LSD (0.05) 2.30 NS P value <0.0001 0.3721 l Approximate Pt concentration of sorption solutions based on the nutrient content of whole manure. 3' NS= Not significant at a=0.05 level. §Means, for a given manure species, within a column were separated using Fisher’s Protected Least Significant Difference test (a=0.05). Conclusions No Pi is removed by centrifugation. Some Pt is removed by centrifirgation; however, the percent is similar over most solution concentrations. Removal of Pt from solution is a problem when determining Pt sorption characteristics. However, a method could not be determined to separate the soil particles and not preferentially separate out a portion of the P in the manure. It is unknown if the portion of organic P (Po) that is removed by centrifirgation is reactive in P sorption experiments. It was assumed that similar amounts of P0 are removed by centrifugation after manure is equilibrated with and without soil. Thus, if a similar amount of P0 is not measured in solution before and after 102 equilibration then it is unnecessary to know the exact amount removed by centrifugation because P sorption is calculated as the difference between the initial and final solution concentrations. 103 References Bender, MR. and CW. Wood. 2000. Total phosphorus in residual materials. In Methods of P analysis for soils, sediments, residuals, and waters. Pierzynski, G.M. (ed). Southern Cooperative Series Bulletin No. 396. North Carolina State University ISBN: 1-58161-369-2. Frank, K.D., D. Beegle, J. Denning. 1998. Phosphorus. In Recommended Chemical Soil Test Procedures of the North Central Region. North Central Region Research Pub. No. 221 (revised). Missouri Agriculture Experiment Station. 881001. 104 Appendix D Determination of the optimum amount of potassium persulfate for complete total P digestion of manure slurry Objective Preliminary research showed that 0.5 g potassium persulfate, as proposed by Bender and Wood (2000), was not providing complete digestion of diluted manure solutions. Thus, the objective of this study was to determine the optimum amount of potassium persulfate needed to completely digest diluted solutions of manure and measure total P (P,) in the manure solutions. Methods and Materials To determine Pt recovery, dairy and swine phosphorus (P) sorption solutions of low, moderate, and high P concentration were digested with varying rates of potassium persulfate. Diluted manure solutions were prepared on a Pt basis in 0.1 M KCl at pH 6, to obtain Pt concentrations of approximately 10, 30, and 100 mg P, L". Theoretical P concentrations of the diluted manure solutions were determined based on P content of whole manure which was determined by digestion and analysis of dry ashed samples (Peters, 2003). Solutions were digested using a persulfate oxidation method with some modifications (Bender and Wood, 2000). The amount of potassium persulfate used varied from 0.5 g to 0.7 g in 0.05 g increments to determine the optimal amount needed. 105 The temperature and duration of boiling were also adjusted in an attempt to increase recovery. Preliminary results (data not shown) had shown the optimum amount of potassium persulfate needed was 0.7 g when first digesting at a temperature of approximately 100 °C for 1.25 h, and then turning the hotplate to the maximum setting, to increase boiling, until the sample had reduced to approximately 10 ml. The amount of sulfuric acid (1 ml of 11 N H2804) suggested in the method of Bender and Wood (2000) was sufficient for digestion. To determine repeatability of this adjusted method, samples were digested in three separate runs using the same sorption solutions as previously mentioned. Sorption solutions were digested in triplicate within each run. Pt solution concentration was determined by the ascorbic acid colorimetric method at a wavelength of 882 nm (Frank et al., 1998). Results and Conclusions For P sorption solutions with the greatest P concentrations, recovery was improved when 0.7 g potassium persulfate was used in addition to reducing the amount of sorption solution from 10 ml to 5 ml. The average P. recovery, over all runs, for dairy and swine slurries was 92% and 98.3%, respectively (Table D.1). The recovery was consistent between runs and was similar at low, moderate and high P solution concentrations for a given manure. The recovery for dairy slurry is a little lower than desired but if P in this slurry can not be digested it is likely not reactive with soil in sorption experiments either. If a similar amount of organic P (P0) is not digested in initial and equilibrated samples then poor recovery does not affect P sorption results. 106 Table D.1. Average percent total P (P,) recovery, of three replications, using 0.7 g potassium persulfate. Experimental Theoretical Average P Source Run P concf P conct Recovery mg L" mg L'1 % Dairy 1 9.27 10 92.7 27.68 30 92.3 89.25 100 89.3 2 9.32 10 93.2 27.97 30 93.2 89.32 100 89.3 3 9.38 10 93.8 27.86 30 92.9 91.49 100 91.5 Average 9.32 10 93.2 27.8 30 92.8 90.0 100 90.0 LSD (0.05) 1.02 Swine 1 9.88 10 98.8 29.19 30 97.3 73.93 75 98.6 2 9.83 10 98.3 28.87 30 96.4 73.93 75 98.6 3 9.92 10 99.2 29.43 30 98.1 74.44 75 99.2 Average 9.88 10 98.8 29.2 30 97.3 74.1 75 98.8 LSD (0.05) NS § f Average experimental Pt concentration of sorption solutions after digestion using 0.7 g potassium persulfate and one ml of 11 N H2804. +Theoretical Pt concentration of sorption solutions based on the nutrient content of whole manure. §NS= Not significant at a=0.05 level. 107 References Bender, MR. and CW. Wood. 2000. Total phosphorus in residual materials. In Methods of P analysis for soils, sediments, residuals, and waters. Pierzynski, G.M. (ed). Southern Cooperative Series Bulletin No. 396. North Carolina State University ISBN: 1-58161-369-2. Frank, K.D., D. Beegle, J. Denning. 1998. Phosphorus. In Recommended Chemical Soil Test Procedures of the North Central Region. North Central Region Research Pub. No. 221 (revised). Missouri Agriculture Experiment Station. SB1001. Peters, J .(ed). 2003. Recommended Methods of Manure Analysis. Bulletin A3769, University of Wisconsin Extension. Cooperative Extension Publishing, Madison, WI. I-2/2003. 108 APPENDIX E . ' -— Relationship between Fe and Al in solution and desorbed Fe and Al The following graphs illustrate the relationship between Fe and Al in solution and Fe and Al desorbed from soil for dairy and swine slurries for each soil. Regression equations and parameters are provided for each relationship when significant. Further details of the data analysis are provided in Chapter 2. In general, Fe and Al desorbed from soil constituents for both slurries with more desorption in soils equilibrated with dairy slurry compared to swine slurry. Furthermore, the relationship between A1 concentrations in solution and desorbed Al were stronger compared to similar relationships for Fe. No apparent desorption of Fe and Al occurs when soils are equilibrated with KHZPO4 solutions with a range in P concentrations (Table 13.1.). 109 Table E.1. Mean of four replications Fe and Al solution concentrations after equilibration with soil when KHzPO4 was the P source. Soil P solution Mean Al in P value for Mean F e in P value for solutron H0: A1 = 0 solutron Ho: Fe = 0 mg L'I mg L'1 mg L" Capac 0 -0.050 0.6729 0.043 0.7314 20 0.020 0.9162 0.050 0.8075 75 0.280 0.1772 0.200 0.2692 Colwood l 0 0.227 0.0439 0.103 0.5423 20 0.027 0.8262 -0.197 0.1058 75 0.470 0.0537 0.250 0.0624 Colwood 2 0 0.183 0.3897 0.027 0.8941 20 0.177 0.3283 0.020 0.8495 75 0.397 0.0610 0.217 0.2044 Oshtemo 0 -0.150 0.6381 -0.087 0.6395 20 0.140 0.0454 -0.090 0.6560 75 0.010 0.9599 0.013 0.9258 Parkhill 0 -0.387 0.1572 0.067 0.4773 20 0.103 0.7960 0.040 0.8346 75 0.210 0.4011 0.327 0.0853 Spinks 0 -0.263 0.6180 -0.243 0.5681 20 0.240 0.3503 -0.07 0.6868 75 0.457 0.2135 0.290 0.3414 110 AL web .8328 E on .- N- 4. O- p p p 8—1 _OO0.0vnm - - SOHR ON a 3.3527; OO- m 0 - O? W p - ON- \m, an - c an”... 053m - om v - v- c- 9. p r L C”! . _OO0.0vum . OO- NOOnNm a NOOINOOTHN .Osm . ON- m.- D: .O w, an - ON Oma- NCEQ - ow O O .ancEw_m can? 325:. 0.8 wages—«an nommmocwom .28 830 c8 moths—O 053m 98 EEO a8 2 use on Bahama 28 .8328 E _< No on 5953 Enmcotflom ._.m oSwE A L Nev 8:28 E _< a. N O N- 4. O- . . . _ . ON- SSO um mvdu NM - oo- V NON + 352N- n a m- . O4- m m - ON- .m- a. 1 o (n. ON a. N O N- O. O- . . . . . O:- _OO0.0 v um - ONT hwdn NM - o2- V 32 ZOOONJN .ON- m - OO- m- - Ov- N - ON- m . 0 GM... - ON (1 ban a - O4 OO 111 A71— 35 5.538 E on w c Q N o N- p p h b 8~I o a cm. H 1 81 m C I. - O? m- p o o - ON- Ow o x. c we... 4 o m.- 053m 0 - 0N ov w c v N o N- i P p h ONfll SGQGVHK - 2:- _..._-.N....fia 1.. N. .- - OO- m - O..- m. - ON- w Go I 0 Ga” . 1 ON In“. C ban on - 3 .Ocaocamm 5:3 32.65 2O £2253; commmohwom do» _ @8330 no.“ mot-BE 053m 28 .33 .8.“ _< 28 on 388% ES cows—om E _< 8 on 50.33 QEONSEN—om .N.m 95me AL mp5 8:28 E _< O 4 N O N- p b - p OWI OOOOOum . - SOHR .OO . . i» w ONT-x: N u . s fow- o w - ON- w . m L. «I O «on. . . .ON 33% O 4 N O N- O4 . p p - 8N! _OO0.0vnm - «OOHNm O2 BOT-£3 -nN .OO_- Om- O o o o - Om EEO OO_ (,BII 3w) Paqms IV 112 o— A c v 1— ?: 5.528 E on N c I QWI EEO _OO0.0 v n m ONO n Na OO.O + NON:- n N (fix 3w) Paqws 9.1 (.331 Bur) paqros 9:1 aqua—mama 5:3 BEE-a 0.3 E30883 568E3— .=8 N week—co No.“ mot-SE 053m 98 .33. New 2 28 on 388% can cows—om 5 _< .5 om c853 gammy—mom .m.m 2&5 AL Nev 8.528 a _< O_ O O a. N O N- b p - Li _ Owl O .. 81 W . . Os m N 0 fi - 3. W;- w an ”of. - O «04. o - (H 05% a ON ON O. O O a. N O N- . . b b . 8M1 386 v n ON T enm- . OOOuNm .OON- W. 32138-; N - of. we. - OO_- w,- - OO- W o I c ”I ESQ o T an O2 113 A.-.“ we: coca—om E on M N — C —r N: - - . . . OO_- ONOOOnm . - ON- NO. "NO - a min + 5&2- " N o - 8 m - O? w- W 7 ON- \I m r o a“. - ON (.- 0 053m T O4 OO M N — O —1 N1 . . . . . OO- 0 BOO-ovum . o . .. NvOuNm Ora mm.m+xmm.:u> m - ON- m N 1 O N . a a o 0 o - ON .m- o - (1 ESQ SN 4:35:me 5:3 BBS-a 0.8 320883 comma-83m £8 08850 8O mom-cam 05.5. 98 3:6 he _< 98 um won-.88 28 5:28 E _< .8 on 5053 aimeozafim .vd oasmi A.-.“ was cows—em E _< N P _ b O b SNOOum BOHR N_.O+ Oak-OT H» OOT ESQ _OOO.O v n m NNO u Nu ONO: xmm.mN- n in r Ow- OO- I O?- 1 ON- f 1 ON I Ov OO (13» gm ) [”qu IV (13’! 3w ) Paqms Iv 114 N— b: 95 sous—8 E on O. O O 4 N O N- s b p p p _ p F AVG—I .VO.O.OuO. . - ON- .NO-n-Nm a. O...N+§.:-u. .OO- m. - m. 0 .- CV m. o o r ON- Ow .O a. o (u. o. - 053m coo. 8 O4 N. O. O O 4 N O N- s . . . . . . . OO.- O Soodvum 33 u... .OO.-u.. NO... + saw:- u N m - OO- w- p .O m x 8. . . .OO (- ESQ O O .EmoOEwmm :23 3253 2.. £808.23 commmohwom .=oO £533 .8 mot-.2... 053m 33 EEO 8O 2 OS. on 39.82. OS. .8528 E _< .8 on .8222. E55323— .m.m charm .2. Nev 8.3.8 a. .< N. O. N O .- N O N- I p . p p p p p VI OOOOOu... . - N- NNOuNm v OONIOONNJN - N- m o m- - .- m- . w o - O .m- N s.- a - . ( 053m N N. O. O O O N O N- I . . . . . . . OON- .OOOOvnN. - OON- 80"... v 30. .LROONJ.A - OON- e1. - O2- m m. - 8.. .m- w Om- aw x- - O co..- ERQ .OO OO. 115 N— 52 ms: 5:28 2 on O. O O N. N O N- v- N N N N N N N 8—1 386nm - cw- .NO n N. % OON + #2.- u N . OO- O . .2. m o o - ON- .m .O m. o (M. O. - 023m coco ON ON. O. O O N. N O N- .- N N N N N N N cm—l O .OOOOVHK ONO "N”. - 874. ON... +NOO.N.-uN m - OO- w D. - O m X 3. . o - OO (- >25 .255ch 5:3 5255 0.... 555.89 585.5% 28 22am 5.: 55.... 2:38 O5 .35 5.: 2 O5 on: 59.8% :5 5:28 2 _< 5 on 5035: QEONSOEBN .O.m 0.5m...— f we. 8.5.8 a. .< N. O. O O N. N O N- .- N N N N N N F VI ONOOOnm NNOuNN. . .N v OONIOONNJN N- H . w 0 -.-m . w 0 Go a .. ........ a -. ( 023m N N. O. N O O N O N- 4- N N N N N N N 8M! .OOOOvnm - OON- 8O"... v E.O.+NNO.ON-HN .OON-m. NOS-m m TOE-w wen.- «W x -O ”on. .050 .8 OO. 116 APPENDIX F .L— Relationship between desorbed Fe or Aland the difference in inorganic P or total P sorption for slurry and KHZPO... The following graphs illustrate the relationship between the difference in P- sorption capacity between each slurry and KHzPO4. Where the correlation is significant, the correlation coefficient and the level of significance are given for each relationship. Further details of data analysis are provided in Chapter 2. In general, the difference between P; sorbed from dairy slurry and KHZPO4 were moderately to strongly correlated to desorbed Fe or Al for all soils except Capac and Colwood 1 soils. Whereas, the difference between P- sorption from swine slurry and KHzPO4 was moderately correlated to desorbed Al for Capac soil (r = 0.43) only and was not significantly correlated to desorbed Fe for any of the soils. In general, the difference between P- sorbed from dairy slurry and KHZPO4 as well as the difference between P- sorbed from swine slurry and KHzPO4 were not significantly correlated to desorbed Fe or Al (Table 2.6). 117 ON. we. see 0.. 55:20:50 0:: .5 0052228 05 O. m :5 20205000 5:20.50 05 2 .. 05:3 :8 850 5.: 5&va— :5 805.8 0.55:. 5.: 5:88 5 2 0055:? 0:: :5 _< 5 0 m 5:580: .5 5:20.50 2...: 059.: o: ov ON O ON- Ov- O:- Ow- OO_- ON:- N N N N N N N N 8—l o NOm _.O H k o oo o o N-NOHN - ON- 5 o o. o o - O:- o o o o - Ov- - ON- . o coo. - O 023m .0 ON O: OV ON O ON- ON». o:- Ow- OOT ON:- N N P N N N N P ON... 0 o hmOOO H NN o o o ONO- u .N - .0 O” 0.. O. O o o o a o o o - O . . ON 0 O - Ov o o - o: N25 0 o Om (1.8)[ 8m) paqros id at aouarayrq (1-8x 8m) paqros id ur aouarayrq .79. we. Oats .< O: ov ON O ON- O».- O:- Ow- OOT ONT N N N N N N N N 8—0 :ONOO u k s N . N...O u . . ON- C . O O on o - o:- o o o oo o - Ov- - ON- oo’ o - o 8.8 .. . ON O: O»: ON O ON- 3». O:- Ow- OOT ON:- N N N N N N N N ON! wOOOO n m o . - n N O. O O. C 3 O I O o o o O O O O o o o o 5 - ON 5 O O O Q - ov o - O: O O 0 N25 Ow a a! male-mo (l_8)[ 8w) paqros ’ (I_8)[ Bur) paqros !d u; aouarayrq 118 1. 50:20:50 0:: .5 00:50:58 0:: 2 m :5 5205000 5:20:50 0:: 2 .. 05:3 :8 : :00350 5: 5&sz :5 82.5.8 0.555 5.: 5:98 5 2 0055.:2: 0:: :5 _< 5 0 n: :0:..80: .5 5:22.50 .N...: 0.52”: pa: OE: 5:58 0": OO_ Om O Om- OOT O2- OON- OmN- N N N N N N Sfil O mn:w.O HR - O:- . . . H.— 0.“ 8 O - ON.- “ O - COT - Ox- 00 o - OO- CCO 0 .- CV1 - ON- . 0‘8 023m - O ON OO_ Om O Om- OOT O2- OON- N N N P N on- 0 0 0 N3:.O um - O l 0 0 NO. H: a, O ‘ .~ . C C .- CW 0 o - OO— O - Om— b... c Q . T SN 0 ON (13" Bur) paqros id 11; 90110.10}er (1.8x Bur) paqros ‘d In aouarayrq .79. we: O88... .< OO: Om O Om- GO:- Om _ - OON- OmN- N N N N N N oc—l O :ON—.Ofla~ - O3- 0 o ONO".N - Na . - ON. 0 - - O O OO— - Ow- . s o - OO- 2 - N o . ON. - ON- .‘au 2.8 . O ON OO— Om O Ow- OOT O2.- OON- N N N N N on- :VOOO H m 0 0 0000 o va-H .N - O O O 0 ‘OO‘O'. 0. 0 0 o T On 0 o - OO— 0 0 - Ow— .CQQ o - OON 0 0 OmN ([_8)[ 8m) paqros !d u; 93119.19le (1-8)1 8w) paqros Ed at 90119.10}er 119 00:20:50 0:: .5 00:00:56 0:: 2 m :5 5208000 5:20:50 0:: 2 -_ 05:3 :00 N 000350 5.: 00.35: :5 0.2-5:0 055:: 5.: 5:050 d 2 00:80:05 0:: :5: _< 5 0 m 00:58: .5 5:20:50 6...: 05mm: p-00: we: 8&8 0.: 9.0: we: 8&8 2 OO— O Om- OOT Om.- OON- OmN- OO— Om O Om- OO _ - Om _ - OON- OmN- N N N N N N WM! '00. N N N N N N WMI a . I u: . u I. o. NNOOO-o: - ON- 0 .. OONNO nan - ON- mm 0 mm O n .: m 0 ON O I m - N. m - N. a on 0 m. 00 w m. .0 .ON--d o 0 .ON- .4 O O S 0 Km: 0 o 0 - 2. m 00 o N 2. m 000 m... - f m - OT \wl OT ) s ... m .- W... X. . ‘ .- WI XL ocmam ’0 cor Omnmgm 0’ 5. O (x O (u- OO- Ow O Om- OO _ - Om _- OON- OmN- OO— Om O Om- OO_- Om T OON- OmN- - . . . - . ON- a . . . . _ . ON- a . .. as: .o y 1.. aim-M ; . mm O- u 0 m . . . ON O - a c. .a. - ON a 5. o. - ON w I ‘00 m. I ‘ 0 m. 0 O O. O T O? l..d O 0.. - CV Md . S C m 50 0 - O: m.- 00 r OO m- . O D- . O m- o - Ow WI 0 - Ow MW 0 O 0 - On 0 o - 3 N53 0 OO: cum >200 0 OO: cum ON: (- ON: (- 120 50:20:50 0:: .5 00:50:56 0:: 2 m :5 520550 5:20:50 0:: 2 : 05:3 :00 050500 5.: «0“:va— :5 02520 0555 5.: 5:800 :N: 2 0000:0.:.::: 0:: :5 _< 5 0 N: 5:58: 5 5:20:50 .3: 053:: 2.0: we: Bea 0.: OO O: ON O ON- Ov- OO- Ow- OOT N N N N N N N GMI 0 o OmahOum - . . o OOOn: - N 0000 0 0 0 - ON- C 0 00 T m:- C 0 $0 - O:- o o - m- 0 0 000 - O 0 05:5 - O. O: O: O: ON O ON- Ov- Oo- Ow- OOT N N F N P N N °_I 0 0 0 0 h: :0.0 Ml NvO- u : - O 0 0 o 0. o o o a... - O: O O O O O 0 fi ON C O O O . - ON C 0 O - O:. 05 Om (03" 3w) paqlos ’d u; 00110103“ (fix 3w) paqms ’d U! ”mama p-00: 0:: O 8&8 :< O: o: ON O ON- O:- OO- Ow- OO _ - N N N N N N N cml O 0 WvaO H K - mN- . N: .O u : do. - ON- C 0 r m:- m - O:- N , m- . O - O 0 053m - n O— OO ON. ON O ON- OV- OO- Ow- OO:- N N N L N N N o“! 0 0 0 0 MOOOO HQ NWO- H .: r O o u o 00 fl 0 ‘ 0 - O: O O O . o . - ON C O O O o - Om . C EEO o o o - Ow O V) (1-3)! 3m) paqws ’d U! swam-m (03" But) paqlos Ed u; aouanygq 121 50:20:50 0:: 5 0:050:56 0:: 2 O: :5 520::000 5:20:50 0:: 2: 05:3 :0: :25: 5.: «0:25: :5 025:: 0555 5.: 50:05: :N: :2 00:0:0:.::: 0:: :5 :< :0 0.: 5:500: .5 5:20:50 6...: 052.: :0: we: OE... 0: OO: Om O Om- OO:- Om : - OON- Om N- OOm- N N . N N N N aN—l G ROOOHO: m . N . :OOOu: - OO:-m I O. . 0 OO::- “O. I 0 m. r 8: [40 C O S . - O:- m p .- ON: \m) a.” - O 3 023m 00 0d. ON (- OO: Om O Om- OO:- Om :- OON- Om N- 8».- . . . . _ L . O:- a OOOOOuO: - m 00 - ON a 00. :vO-n : m 0 0 lo 0 - O m o I» o . - ON m. -..d O - O? W- C C .- 8 m. o - O: m, o 8 5:5 - OO: can ON: (- :79: we: 5:8 _< OO: OO O OO. OO:- OO:- OON- OON- OON- - . . . . . . ON:- a :mmmOnO: n: ~ «OOH: .8: m 0 - ON- a a m. u 8: I..d a m . O:- m. r ON- \w) . .O a 023m 0 afl ON ( OO: OO O Om- OO:- O:- OON- OON- OON- . . . _ . _ . O:- a ONOOOuO: - - w v... .u . ONO-u : ON m C C C‘ C .- o a O O O ‘0 o o 0 -ON m. .d o - O:. m o o - OO N p . - ON Ow 0 E50 - OO::M. ONT-u 122 .TLH: .FA LII-Iii 50:20:50 0:: .5 0050552: 0:: 2 m :5 520850 5:20:50 0:: 2: 05:3 :8 85:5 5.: NO:N:0: :5 0.05:: 0555 5.: 5:550 _: 2 0055::5 0:: :5 _< 5 0.: 5:58: .5 5:20:50 0.: 052-: :TNO: 05: 5:8 0: :9: we: :38... :< OO: O: O OO. OO:- Om:- OON- OON- OON- OO: OO O Om- OO:- Om:- OON- OON- OON- . . . . . . . ON:- 0 . _ . . _ . . ON:- a nah—Out m Ow:v.OHN: mu 0 Q o mNOu: T OO:-m m 20": - OO:- m ' .... .ON. a 0 :ON- a .. . m. : m. - O0- Md T O:- ....d . S S . - O:- m- a . O:- m- 9 - ON- N . ON- .m- fio" T O emu. ‘ - O an 0226 0. Owen. 0526 0 mm ON (- ON ( OO: O: O 8. OO:- O2- OON- OON- OON- OO: OO O Om- OO:- O:- OON- OON- OON- N N N N N N N a?! g F N N N N N L! CY Ifl OOOOOuO: - - u: ONOOOH: - - u: « ..... 3O-" ._ ON m a...” . u . ONO-u : ON m 0 0 3000 - O m 0 0 .0 0 - O x 0.3 . . TON m. ...... . . -ON m. ha -...a .- Ov S - CV S o m 0 m . . - OO m . . - OO m- p. D. . :ON W O: w. no 0 3 :55 T OO: Om. Nn59 - 02%.». ON: (- ON:( 123 MICHIGAN STATE UNIVERSI‘YY LIBRARIES llHIHWIHHII)1||IHMINIIIIWUN!lllHlHlHllllHlHl 3 1293 02504 5125