<%BANNER%>

Phosphorus Fate and Transport in Wastewater Applied to Rapid Infiltration Basins

Permanent Link: http://ufdc.ufl.edu/UFE0024468/00001

Material Information

Title: Phosphorus Fate and Transport in Wastewater Applied to Rapid Infiltration Basins
Physical Description: 1 online resource (105 p.)
Language: english
Creator: Moura, Daniel
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: phosphorus, rib, soil, spsc, wastewater
Soil and Water Science -- Dissertations, Academic -- UF
Genre: Soil and Water Science thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Water reuse is one answer to the problem of water scarcity that present and future generations will face. Wastewater applied to land can intentionally recharge aquifers for future reuse. Regulations related to phosphorus (P) applications to soils (P index in Florida) are likely to regulate aquifer recharge systems because wastewater is a P source. While regulations need to be based on science, environmental impacts due to continuous application of wastewater P to soils have received limited research. A careful determination of the fate and transport of wastewater P on a full scale land treatment system can provide useful information in the development of P indices for land application of wastewater and in the determination of the 'life span' of the study sites. The Conserv II system, located west of Orlando, FL, promotes groundwater recharge with a commonly used method called surface spreading using rapid infiltration basins (RIBs). Each RIB at Conserv II has different qualitative infiltration rate characteristics ('good' or 'not so good'), age (young, middle aged and old) and number of cells (1 to 4). We selected four distinct cells, each one within a different RIB type, for study. We attempted to field validate the soil P storage capacity (SPSC) for predicting the maximum amount of P that can be safely applied to the cells. The concept assumes that P in a soil is predominantly associated with iron (Fe) and aluminum (Al) oxides and it relies on a site-specific threshold P saturation ratio (PSR). The threshold is normally defined by the change point that corresponds to the PSR value above which water extractable P (WEP) increases rapidly. A collective change point for the studied cells could not be determined in the field, but a change point was well defined in a laboratory study that mimicked long-term wastewater P application to soils. The change point of ~ 0.1 found in the laboratory study agrees with the literature. Two approaches were conducted to investigate why the change point was found in the laboratory study but not in the field: 1) a P fractionation study combined with a chemical modeling software and 2) a kinetics study of P retention rate, along with field infiltration rate measurements. Results from the fractionation study showed that in most cases soil P association was with Fe and Al oxides. Chemical modeling revealed that P can potentially associate with calcium (Ca) and magnesium (Mg) in old cells, as suggested by P sequential extraction data from surface samples of a particular cell. Results from the second approach revealed that < 50% of equilibrium P adsorption is reached because wastewater-P flows rapidly through surface soils in the cells. Therefore the SPSC concept could not be applied to the Conserv II sites because either equilibrium conditions were not met in any of the studied cells or because Ca-P and Mg-P (rather than Fe- and Al-P) associations dominated P removal in the surface of one cell. Groundwater data show that at least one of the old cells is failing to reduce wastewater P concentrations. Changes in management practices are suggested to enhance soil P retention and reduce potential impacts of wastewater-P on groundwater quality.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Daniel Moura.
Thesis: Thesis (M.S.)--University of Florida, 2009.
Local: Adviser: Silveira, Maria L.
Local: Co-adviser: O'Connor, George A.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2009-11-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024468:00001

Permanent Link: http://ufdc.ufl.edu/UFE0024468/00001

Material Information

Title: Phosphorus Fate and Transport in Wastewater Applied to Rapid Infiltration Basins
Physical Description: 1 online resource (105 p.)
Language: english
Creator: Moura, Daniel
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: phosphorus, rib, soil, spsc, wastewater
Soil and Water Science -- Dissertations, Academic -- UF
Genre: Soil and Water Science thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Water reuse is one answer to the problem of water scarcity that present and future generations will face. Wastewater applied to land can intentionally recharge aquifers for future reuse. Regulations related to phosphorus (P) applications to soils (P index in Florida) are likely to regulate aquifer recharge systems because wastewater is a P source. While regulations need to be based on science, environmental impacts due to continuous application of wastewater P to soils have received limited research. A careful determination of the fate and transport of wastewater P on a full scale land treatment system can provide useful information in the development of P indices for land application of wastewater and in the determination of the 'life span' of the study sites. The Conserv II system, located west of Orlando, FL, promotes groundwater recharge with a commonly used method called surface spreading using rapid infiltration basins (RIBs). Each RIB at Conserv II has different qualitative infiltration rate characteristics ('good' or 'not so good'), age (young, middle aged and old) and number of cells (1 to 4). We selected four distinct cells, each one within a different RIB type, for study. We attempted to field validate the soil P storage capacity (SPSC) for predicting the maximum amount of P that can be safely applied to the cells. The concept assumes that P in a soil is predominantly associated with iron (Fe) and aluminum (Al) oxides and it relies on a site-specific threshold P saturation ratio (PSR). The threshold is normally defined by the change point that corresponds to the PSR value above which water extractable P (WEP) increases rapidly. A collective change point for the studied cells could not be determined in the field, but a change point was well defined in a laboratory study that mimicked long-term wastewater P application to soils. The change point of ~ 0.1 found in the laboratory study agrees with the literature. Two approaches were conducted to investigate why the change point was found in the laboratory study but not in the field: 1) a P fractionation study combined with a chemical modeling software and 2) a kinetics study of P retention rate, along with field infiltration rate measurements. Results from the fractionation study showed that in most cases soil P association was with Fe and Al oxides. Chemical modeling revealed that P can potentially associate with calcium (Ca) and magnesium (Mg) in old cells, as suggested by P sequential extraction data from surface samples of a particular cell. Results from the second approach revealed that < 50% of equilibrium P adsorption is reached because wastewater-P flows rapidly through surface soils in the cells. Therefore the SPSC concept could not be applied to the Conserv II sites because either equilibrium conditions were not met in any of the studied cells or because Ca-P and Mg-P (rather than Fe- and Al-P) associations dominated P removal in the surface of one cell. Groundwater data show that at least one of the old cells is failing to reduce wastewater P concentrations. Changes in management practices are suggested to enhance soil P retention and reduce potential impacts of wastewater-P on groundwater quality.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Daniel Moura.
Thesis: Thesis (M.S.)--University of Florida, 2009.
Local: Adviser: Silveira, Maria L.
Local: Co-adviser: O'Connor, George A.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2009-11-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024468:00001


This item has the following downloads:


Full Text

PAGE 1

PHOSPHORUS FATE AND TRANSPORT IN WASTEWATER APPLIED TO RAPID INFILTRATION BASINS By DANIEL RODRIGUES MOURA A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORI DA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2009 1

PAGE 2

2009 Daniel Rodrigues Moura 2

PAGE 3

To my Me, Pai and Irmo 3

PAGE 4

ACKNOWLEDGMENTS First I would like to acknowledge the Almighty God, the in ventor of the reason and the reasoning, the creator of the scien ce and the scientist. I want to thank my two advisors, Dr. Maria L. Silveira and Dr. George A. OConnor. Dr. S ilveira accepted me as her student even without knowing me. She was always available to help a nd her suggestions in my proposal, reports and thesis were always very useful. I do not know any professor who spends so much time with his/her students as Dr. OConnor. I thank him for reading all my written work (even when correcting my writing was an arduous task!) and for his constant commitment to teaching. I learned several valuable lessons from him, such as how to be more organized and to be always prepared and ahead of time. Next, my thanks go to Dr. William R. Wise for accepting to be part of my committee and for his valuable suggestions throughout my research. I appreciate all the support provided by the Water Conserv II administration. Sr. Project Manager Mr. Phil Cross opened their doors to us at Conserv II for all necessary analysis and Operations Chief Mr. Glenn Bu rden provided us with all th e necessary data and spent a considerable amount of time showing us to the study sites. I am very grateful for being able to work with so many outstanding researchers and friends from the Environmental Soil Chemistry group: Augustine, Jaya, Liz, Sampson, Manmeet, Matt, and Xiaolin. Especially, I woul d like to thank Sampson: he thought me all the basic lab procedures, he helped me in the field (with Matt, augering down to 4 m!), he ran the samples in the ICP, and assisted considerably in the interpretation of my data. I also r eally appreciate Matts help in several laboratory analys es and for his advice and rides to Georgia. I also want to thank my office mate Sylvia Lang for her friendship an d her important input in my manuscript. She was able to make my writing process less boring and fun sometimes. 4

PAGE 5

Also, I want to acknowledge several professors that helped me during various stages of my research: Dr. Harris, Dr. Jawitz, and Dr. Nair. Thanks go to Dr. Josan and Dr. Jango for helping me with the interpretati on of my data. In addition, I tha nk Dr. Rafael Muoz-Carpena and Eban Bean for providing me the necessary apparatus to measure infiltration rates in the field. I appreciate all the time Eban spent explaining how to use the instruments and how to analyze the results. I thank James Colee for he lping with some of my statistical analysis. I also appreciate the help of several tutors from the UF writi ng center that helped review my writing. I am extremely grateful for all the encour agement and care provided by my relatives. I would especially like to thank my Uncle Luiz fo r being invested in my education since middle school. Without his support I would not have been ab le to finish this degree. In addition, I thank all my friends in Gainesville, including my roommates (Johnny and Brad) and the Brazilian community (Gabriel and Eduardo) and others. The support provided by my friends in Brazil, including Nayara, was very important also. I want to thank Emily, Carol, and Mr. and Mrs. Roberts for accepting me as part of their own family. Lastly, and most importantly, I wish to th ank my parents, Magno and Malu for having taught, supported and loved me through my brief exis tence. I am eternally gr ateful to my brother Tiago from whom I had the most encouragemen t since the beginning to the end of my MS studies. 5

PAGE 6

TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........8 LIST OF FIGURES.........................................................................................................................9 ABSTRACT...................................................................................................................................14 CHAPTER 1 INTRODUCTION................................................................................................................. .16 Hypothesis and Research Objectives......................................................................................22 Study Approach......................................................................................................................23 2 MATERIALS AND METHODS...........................................................................................24 Study Area..............................................................................................................................24 Soil Sampling.................................................................................................................. ........26 Soil Analyses..........................................................................................................................27 Infiltration Rate Determination...............................................................................................2 8 Wastewater Data................................................................................................................ .....30 PSR Study...............................................................................................................................31 Kinetics Study.........................................................................................................................32 Phosphorus Fractionation Study.............................................................................................33 Chemical Equilibrium Modeling............................................................................................34 Standard QA/QC Protocols.....................................................................................................34 Statistical Analysis........................................................................................................... .......35 3 RESULTS AND DISCUSSION.............................................................................................41 Ground Water Characterization..............................................................................................41 Soil Analysis...........................................................................................................................42 Phosphorus Saturation Ratio (PSR).................................................................................42 Water Extractable Phosphorus (WEP)............................................................................44 Change Point Determination...........................................................................................44 PSR Study...............................................................................................................................45 Phosphorus Fractionation Study.............................................................................................46 Soil pH....................................................................................................................................47 Chemical Equilibrium Modeling............................................................................................49 Kinetics Study.........................................................................................................................50 Infiltration Rate Field Measurements..................................................................................50 Comparison between Data from the Ki netics Study and from the Field................................52 6

PAGE 7

4 CONCLUSIONS.................................................................................................................. ..69 Recommendations...........................................................................................................71 Further Investigations......................................................................................................73 APPENDIX A SAMPLING SCHEME FOR TREATMENT SAMPLES......................................................75 B VARIABILITY OF KSAT AT DIFFERENT DEPTHS........................................................80 C CUMULATIVE WASTEWATER VOLUME APPLIED TO THE CELLS OVER TIME.......................................................................................................................................82 D DISTRIBUTION OF TOTAL PHOSPHOR US CONCENTRATIONS OVER TIME FOR TREATMENT WELLS.................................................................................................84 E COMPARISON BETWEEN PSR VALUES FROM SAMPLES TAKEN NEAR AND FAR FROM THE OUFLOWS...............................................................................................88 F PLOTS OF PSR AS A FUNCTION OF WEP.......................................................................91 G ADDITIONAL IMPORTANT FIGURES AND TABLES....................................................94 H INFILTRATION RATES AS FUNCTION OF TIME...........................................................96 LIST OF REFERENCES.............................................................................................................100 BIOGRAPHICAL SKETCH.......................................................................................................105 7

PAGE 8

LIST OF TABLES Table page 2-1 Selected characteristics of tertiary wa stewater at Conserv II and typical constituent concentrations in irrigation water......................................................................................38 2-2 Total concentration of cations and anions in tertiary wastewater at Conserv II................38 2-3 Water quality guidelines for irrigation salinity and infiltration problems.......................39 2-4 Selected characteristics of Candler A horizon...................................................................39 2-5 Classes of selected RIBs....................................................................................................39 2-6 Elevation differences betw een the top of the soil profile for treatment and control sample sites................................................................................................................... .....39 3-1 Comparison between fits with three models in the PSR study..........................................60 3-2 Phosphorus fraction distribution........................................................................................62 3-3 Phosphorus fraction distribution of surf ace samples from cell 6-21-C (old and not so good)............................................................................................................................64 3-4 pH values for selected treatment a nd control samples from each studied cell. Treatment pH values corresponding to depths of 30 cm are averages..............................66 3-5 Comparison between pH and concentrat ions of P extracted by NaOH and HCl...............67 3-6 Infiltration rate results from fi eld measurements at good RIBs.....................................67 G-1 Results from the chemical modeling exercise...................................................................95 8

PAGE 9

LIST OF FIGURES Figure page 2-1 Study area................................................................................................................. ..........37 2-2 Typical Conserv II cell during: A) wet and B) drying cycles............................................37 2-3 Infiltration rate measurement from a cont rol site using a double-ring infiltrometer.........40 3-1 Average treatment and control PSR value distribution throughout the soil profile for soils from cell 2-3-A (young and good).........................................................................55 3-2 Average treatment and control PSR value distribution through the soil profile for soils from cell 4-3-B (middle aged and good)................................................................55 3-3 Average treatment and control PSR value distribution through the soil profile for soils from cell 6-21-C (Old and not so good).................................................................56 3-4 Average treatment and control PSR value distribution through the soil profile for soils from cell 7-6-B (middle aged and good)................................................................56 3-5 Average PSR distribution through the soil profile for treated soils from cells 2-3-A (young and good) and 7-6B (old and good)..............................................................57 3-6 Average treatment and control WEP dist ribution throughout the soil profile for soils from cell 2-3-A (young and good).................................................................................57 3-7 Average treatment and control WEP dist ribution through the so il profile for soils from cell 4-3-B (middle aged and good)........................................................................58 3-8 Average treatment and control WEP dist ribution through the so il profile for soils from cell 6-21-C (Old and not so good).........................................................................58 3-9 Average treatment and control WEP dist ribution through the so il profile for soils from cell 7-6-B (middl e aged and good)........................................................................59 3-10 The relationship between WEP and PSR values for treatment and control samples from cell 2-3-A (young and good).................................................................................59 3-11 Phosphorus adsorption data (fitted by the LangmuirFreundlich model) for control samples from: A. cell 6-21-C (old and not so good) at a depth of 270cm and B. cell 7-6-B (old and good) at a depth of 300cm.....................................................................60 3-12 WEP as a function of PSR for treatment and control field samples, and for samples used in the laboratory PSR study for cell 7-6-B (old and good)....................................61 3-13 WEP as a function of PSR for treatment and control field samples, and for samples used in the laboratory PSR study for cell 6-21-C (old and not so good).......................61 9

PAGE 10

3-14 Percentage of phosphorus distribution among the various fractions for treatment soils, relative to the su m of all P fractions.........................................................................63 3-15 Percentage of phosphorus distribution am ong the various fractions for control soils, relative to the sum of all P fractions..................................................................................63 3-16 Percentage of phosphorus distribution among the various fractions for treatment surface samples from cell 6-21-C (old and not so good), relative to the sum of all P fractions..............................................................................................................................64 3-17 Change in P sorption with shaking time for the sample BR2 from cell 4-3-B (middle aged and not so good) at 300 cm depth..........................................................................65 3-18 Change in P sorption with shaking time fo r the control sample from cell 6-21-C (Old and not so good) at 90 cm depth.....................................................................................65 3-19 Observed averages of P sorption as a func tion of shaking time for the control sample from cell 6-21-C (Old and not so good) at 90 cm depth and fitted non-linear model....68 A-1 Sampling scheme for cell 2-3-A........................................................................................76 A-2 Sampling scheme for cell 4-3-B........................................................................................77 A-3 Sampling scheme for cell 6-21-C......................................................................................78 A-4 Sampling scheme for cell 7-6-B........................................................................................79 B-1 Distribution of Ksat values as a functi on of depth for Candler soil samples in the state of Florida...................................................................................................................80 B-2 Distribution of Ksat values as a function of depth for Candl er soil samples taken close to Orange County, FL...............................................................................................80 B-3 Distribution of infiltration rate values for Candler soil samples taken in the Orange Country (FL) area..............................................................................................................81 C-1 Cumulative wastewater applied to ce ll 2-3-A during the cells life span......................82 C-2 Cumulative wastewater applied to ce ll 4-3-B during the cells life span.......................82 C-3 Cumulative wastewater applied to ce ll 6-21-C during the cells life span.....................83 C-4 Cumulative wastewater applied to ce ll 7-6-B during the cells life span.......................83 D-1 Well 2-01 First treatment monitoring well data for RIB 2-3 (young and good)........84 D-2 Well 2-02 Second treatment monitoring well data for RIB 2-3 (young and good).....84 D-3 Well 2-03 Third treatment monitoring well data for RIB 2-3 (young and good).......85 10

PAGE 11

D-4 Well 4-01 First treatment monitoring we ll data for RIB 4-3 (middle aged and not so good)............................................................................................................................85 D-5 Well 4-02 Second treatment monitoring well data for RIB 4-3 (middle aged and not so good)....................................................................................................................86 D-6 Well 6-23 Treatment monitoring well data for RIB 6-21 (old and not so good)........86 D-7 Well 7-08 Treatment monitoring well data for RIB 7-6 (old and good).....................87 E-1 Average PSR distribution through the soil pr ofile for soil samples taken near (within 3 meters) and distant (in th e cell edges) from the cell 2-3-A (young and goodRIB) outflows..............................................................................................................................88 E-2 Average PSR distribution through the soil pr ofile for soil samples taken near (within 3 meters) and distant (in the cell edges) from the cell 4-3B (middle aged and not so good RIB) outflows..........................................................................................................89 E-3 Average PSR distribution through the soil pr ofile for soil samples taken near (within 3 meters) and distant (in the cell edges) from the cell 6-21-C (old and not so good RIB) outflows.....................................................................................................................89 E-4 Average PSR distribution through the soil pr ofile for soil samples taken near (within 3 meters) and distant (in th e cell edges) from the cell 7-6-B (old and goodRIB) outflows..............................................................................................................................90 F-1 PSR against WEP for treatment and control samples from cell 2-3-A (young and goodRIB)........................................................................................................................91 F-2 PSR against WEP for treatment and c ontrol samples from cell 7-6-B (old and goodRIB)........................................................................................................................91 F-3 PSR against WEP for treatment and contro l samples from cell 6-21-C (old and not so good RIB)....................................................................................................................92 F-4 PSR against WEP for treatment and cont rol samples from cell 4-3-B (middle aged and goodRIB).................................................................................................................92 F-5 PSR against WEP for treatment samples ta ken within 15 m from effluent outflows........93 F-6 PSR against WEP for treatment samples ta ken more than 15 m away from effluent outflows..............................................................................................................................93 G-1 Phosphorus concentration in the wastewater since 1987...................................................94 G-2 Trend of wastewater pH changes since 1987.....................................................................95 H-1 Infiltration rate as a function of cumulative time from Cell 2-3-A (young and goodRIB)........................................................................................................................96 11

PAGE 12

H-2 Infiltration rate as a function of cumula tive time from Cell 4-3-B (middle aged and not so good RIB)............................................................................................................96 H-3 Infiltration rate as a function of cumulati ve time from Cell 6-21-C (old and not so good RIB).........................................................................................................................97 H-4 Infiltration rate as a function of cu mulative time from Cell 7-6-B (old and goodRIB)........................................................................................................................97 H-5 Infiltration rate as a function of cumula tive time from a control sample taken from Cell 2-3-A (young and goodRIB)...................................................................................98 H-6 Infiltration rate as a function of cumula tive time from a control sample taken from Cell 6-21-C (old and not so good RIB)..........................................................................98 H-7 Infiltration rate as a function of cu mulative time from Cell 7-6-B (old and goodRIB)........................................................................................................................99 H-8 Infiltration rate as a function of cu mulative time from Cell 7-6-B (old and goodRIB)........................................................................................................................99 12

PAGE 13

LIST OF ABBREVIATIONS AIC Akaikes information criterion Al Aluminum Ca Calcium CaCO3 Calcium carbonate EC Electrical conductivity DPS Degree of phosphorus saturation FDEP Florida Department of Environmental Protection Fe Iron K Potassium Mg Magnesium Mn Manganese NRCS National Resource Conservation Service P Phosphorus PSR Phosphorus saturation ratio QA/QC Quality assurance/quality control RIB Rapid infiltration basin R2 Coefficient of determination SAS Statistical analysis software SPSC Soil phosphorus storage capacity TP Total phosphorus USEPA United States Environmental Protection Agency USDA United States Department of Agriculture WEP Water-extractable phosphorus WTR Water treatment residual 13

PAGE 14

Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Masters of Science PHOSPHORUS FATE AND TRANSPORT IN WASTEWATER APPLIED TO RAPID INFILTRATION BASINS By Daniel Moura May 2009 Chair: Maria Silveira Cochair: George A. OConnor Major: Soil and Water Science Water reuse is one answer to the problem of water scarcity that present and future generations will face. Wastewater applied to land can intentionally recharge aquifers for future reuse. Regulations related to phosph orus (P) applications to soils (P index in Florida) are likely to regulate aquifer recharge systems because wast ewater is a P source. While regulations need to be based on science, environmental impacts due to continuous application of wastewater P to soils have received limited research. A careful determination of the fate and transport of wastewater P on a full scale land treatment syst em can provide useful information in the development of P indices for land application of wastewater and in the determination of the life span of the study sites. The Conserv II syst em, located west of Orlando, FL, promotes groundwater recharge with a commonly used method called surface spreading using rapid infiltration basins (RIBs). Each RIB at Conserv II has different qualitative infiltration rate characteristics (good or not so good), age (young, middle aged and old) and number of cells (1 to 4). We selected four distinct cells, each one within a different RIB type, for study. We attempted to field validate the soil P storag e capacity (SPSC) for predicting the maximum amount of P that can be safely applied to the cells. The concept assumes that P in a soil is 14

PAGE 15

15 predominantly associated with iron (Fe) and alumin um (Al) oxides and it relies on a site-specific threshold P saturation ratio (PSR). The threshol d is normally defined by the change point that corresponds to the PSR value above which wate r extractable P (WEP) increases rapidly. A collective change point for the studied cells coul d not be determined in the field, but a change point was well defined in a labor atory study that mimicked long-te rm wastewater P application to soils. The change point of 0.1 found in the laborat ory study agrees with the literature. Two approaches were conducted to investigate why the change point was f ound in the laboratory study but not in the field: 1) a P fractionation st udy combined with a chemical modeling software and 2) a kinetics study of P retention rate, along w ith field infiltration rate measurements. Results from the fractionation study showed that in most cases soil P associati on was with Fe and Al oxides. Chemical modeling revealed that P can potentially associate with calcium (Ca) and magnesium (Mg) in old cells, as suggested by P sequential extraction data from surface samples of a particular cell. Results from the second a pproach revealed that <50% of equilibrium P adsorption is reached because wastewater-P fl ows rapidly through surfa ce soils in the cells. Therefore the SPSC concept could not be applie d to the Conserv II sites because either equilibrium conditions were not me t in any of the studied cells or because Ca-P and Mg-P (rather than Feand Al-P) associations dominated P re moval in the surface of one cell. Groundwater data show that at leas t one of the old cells is failing to reduce wastew ater P concentrations. Changes in management practices are suggested to enhance soil P retent ion and reduce potential impacts of wastewater-P on groundwater quality.

PAGE 16

CHAPTER 1 INTRODUCTION Water scarcity is probably one of the biggest challenges that present and future generations face. Despite the fact that the total volume of renewable fresh water is several times greater than needed by the worlds population, only about 30% is accessible for human use (Metcalf & Eddy, 2007). Assuming no shift in water consumption patte rn, nearly 50% of the worlds population is projected to live in water-stressed (water supplies less than 1700 m3 per capita per year) river basins by 2025 (World Resource Institute, 2000). Fresh water supply shortages are caused by a variety of factor s: population growth, increased demand for water, unequal distribution of fresh water in the world, climate change, and degradation of existing water s upplies. Probably the most prominent approach to increase fresh water availability is by reusi ng wastewater. Although some people still oppose wastewater reuse, humans have long been reusing water: ancient populations used wastew ater for crop irrigation and for recharging water bodies (OConnor et al., 2008). Indications of irrigation using wastewater in the Minoan civilization in an cient Greece extend back approximately 3000 years (Asano and Levine, 1996). During the nineteenth century, the introduction of large-scale wastew ater carriage systems led to an unplanned water reuse: wastewater di scharged in water bodies ended up being used for drinking purposes. Contemporary water treatme nt was minimal, and led to catastrophic epidemics of waterborne diseases in big cities like London (Asa no and Levine, 1996). The United States began planning wa ter reuse programs early in the 20th century. California pioneered water reuse by approving re use regulations in 1918. In 1940, chlorinated wastewater was used for steel processing, and in 1960 urban reus e programs started in Colorado and Florida. Currently, the WateReuse Association estimates that nearly1010 L d-1 of municipal 16

PAGE 17

wastewater is reused, and reuse is growing at about 15 percent per year in the United States (Metcalf & Eddy, 2007). The states of Florida and Calif ornia lead the country in term s of volumes of water reused. Florida has many challenges in terms of water av ailability. Despite the apparent abundance of water (1,270 mm of rainfall per ye ar), Florida has the fourth larg est population in the country and has a projected population of 20 million by 2020. Howe ver, Florida was motivated to become the nations water reuse leader to reduce eutrophica tion of surface waters, rather than to increase water availability for the expandi ng population (Metcalf & Eddy, 2007). Prior the 1960s, discharge of secondary treat ed wastewater into surface waters was a common practice. Regulations crea ted in the 1960s, however, limite d the quality and quantity of the wastewater discharged in Florida surface wa ter bodies significantly. As a result, a move towards land application of wastewater st arted in the 1970s (Metcalf & Eddy, 2007). For example, as a result of a court decision in 1979, the City of Orlando and Orange County, Florida, were obligated to stop discharging municipal ef fluent into Shingle Creek by 1988. The city and the county joined forces and decided to create a groundwater recharge system named Conserv II (USEPA, 2004). The Conserv II system was not the first such system in the United States: groundwater recharge with reclaimed water has been practiced in the country for more than 40 years (Metcalf & Eddy, 2007). One of the most common and oldest technologies for groundwater recharge is called surface spreading using rapid infiltration basins (RIBs), which is specifically the system used at Conserv II. Rapid infiltration basins (R IBs) require less previous wastewater treatment and relatively small land area and maintenance requirements when compared with other groundwater recharging systems (USEPA, 2004). 17

PAGE 18

The RIB system has two main purposes: to provide long-term water storage and to improve the reclaimed water quality (Metcalf & Eddy, 2007). The system, and other wastewater land treatment systems, can attain a degree of nu trient removal for wast ewaters that would be more expensive to attain using biological treatment systems (H u et al., 2005). In addition, the RIB system has the advantage of storing water in the watershed for future reuse (e.g. irrigation) during seasonal water shortages (USEPA, 2006). Rapid infiltration basins (RIBs) are constructed on hi ghly permeable soils (e.g. sandy loam, loamy sand, and fine sands) to accomplish high hydraulic infiltration rates (USEPA, 2004). Rapid movement of wastewater through sandy so ils can limit soil remediation of wastewater constituents, such as phosphorus (P), because th e contact time between soil and wastewater can be less than the time required fo r maximum adsorption to occur. The presence of P in water supplies has no dire ct human health conse quences, but P is the limiting factor for eutrophication in many fresh water systems, so P removal from wastewater is necessary before effluent discharge to water bodies (USEPA, 2006). In addition, P accumulation in the soil may limit the life e xpectancy of wastewater land tr eatment systems (Hu et al., 2005; USEPA, 2006; Zhang et al., 2007) Phosphorus is removed by ad sorption and precipitation, and the soil is the major player in the treatment process (USEPA, 1981; United States Army Corp of Engineers, 1982; USEPA, 2006). The removal pro cesses occur when P reacts with aluminum (Al), iron (Fe) and manganese (Mn) (either as diss olved ions, as hydrous oxid es, or as oxides) to form insoluble hydroxy phosphates. Precipitates of various calcium (Ca) or magnesium (Mg) phosphate minerals can also form, or P can become adsorbed to the iron impurities on the surface of carbonates, or on clay mine rals. Ideal soil char acteristics in RIB systems include rapid infiltration rates of water (> 5 cm h-1) and low content (< 10 % by weight) of expanding18

PAGE 19

contracting clays (USEPA, 2006). High clay conten t reduces infiltration rates, but provides additional adsorption capacity for ions in the wastewater, mainly P (USEPA, 2004; USEPA, 2006). The soils P removal capacity is finite, but cations in treated munici pal wastewater can add to the systems P sorption capacity. Hu et al. (2 005) found that the P so rbing capacity increased with long-term wastewater appl ication due to wastewater calcium additions and subsequent precipitation of Ca-P compounds. Woodard et al. ( 2007) suggested the same effect due to dairy effluent-borne Al. However, the phenomenon is not expected to occur in all wastewater land treatment systems, because soil pH and the ch emical composition (e.g., Ca concentration) of various wastewaters differ. Extensive research has been done to quan tify P-source impacts on water quality. Although soil testing alone does not answer all questions about P losses from soils, soil test P (STP) is often used to determine if the P concentra tions in a given soil are sufficient to pose environmental risks. Soil test P has been broadl y researched because it correlates well with subsurface P losses and is relatively inexpe nsive to perform (Maguire and Sims, 2002). Examples of STPs that have been used for es timating labile P are water extractable P (WEP), Olsen P, Mehlich-3 P, and CaCl2-extractable P (Maguire and Sims, 2002; Hooda et al., 2000; Sims et al., 2002). In the context of the present thesis, we define labile P as the fraction of adsorbed P that is rapidly desorbed from soil particles and thus, readily available for plant (Pierzynski, et al., 2000), and microbial uptake (Cross and Schlesinger, 1995) and for chemical reactions in soil solution. Another measure used for assessing P loss risk s is the soil degree of phosphorus saturation (DPS), which is calculated as the ratio of acid ammonium oxalate extract able P to the sum of 19

PAGE 20

acid ammonium oxalate extractable iron and aluminum in a soil and multiplied by an empirical constant based on soil type (Nai r et al., 2004). Other researchers have suggested the use of Mehlich-3 instead of acid ammonium oxalate as th e extractant in the D PS determination due to practical reasons (Maguire and Sims, 2002; Si ms et al. 2002; Shober and Sims, 2007). Although DPS correlates well to P concentration in leachate waters, it fails to quantitatively indicate the P storage capacity of a soil (Oladeji, 2006). To overcome the limitation of the DPS concep t, Nair and Harris ( 2004) suggested the use of a new approach, the soil phosphorus storage cap acity (SPSC), as a direct estimate of the amount of P a soil can sorb before exceeding a threshold soil equilibrium concentration. The SPSC concept can be used to predict the maximum amount of P that can be safely applied to a soil (the amount of P leaching out of the soil and reaching a receiving water body is below a designated threshold). The SPSC is determined from equation 1-1: SPSC (mg kg-1) = (threshold PSR-calculated PSR) x (Alox + Feox) x 31, (1-1) where (Alox+Feox) is oxalate extr actable aluminum and iron (mmol kg-1), respectively, and represents a soils P sorption cap acity. The phosphorus saturation ra tio (PSR) is calculated from equation 1-2: PSR (phosphorus saturation ratio ) = [(Pox)/(Alox + Feox)], (1-2) where Pox represents oxalate extractable P concentration in mmol kg-1 of soil (Nair and Harris, 2004). The SPSC concept assumes a threshold PSR value that corresponds to a critical (threshold) P solution concentration, which Nair and Harris (2004) originally propos ed was 0.15 for Florida sandy soils. The rationalization behind choosing the threshold value is that the 0.15 value corresponds to the critical P so lution concentrat ion of 0.10 mg L-1, acceptable for streams by 20

PAGE 21

USEPA (USEPA, 1986), and for shallow ground waters in the Netherlands (Breeuwsma et al., 1995). One crucial variable to be acc ounted for in the threshold PSR value determination is the change point that corresponds to the PSR value above which WEP increases rapidly. Above the change point, P is believed to be sorbed on lowenergy sites from which P can be easily released (Heckrath et al., 1995) and soil P loss is drasti cally amplified (Nair et al., 2004; Maguire and Sims, 2002). Below the change point, P is believed to be held on high-en ergy sites (Heckrath et al., 1995) and not readily released. Nair et al. (2004) found a cha nge point PSR of 0.10 for soils from all depths, indicating that the value can be used independently of the soil depth studied. The SPSC equation proposed by Nair and Harris (2004) can be modified when necessary because the threshold PSR value is site specific (depending on soil characteristics and acceptable P solution concentration). Positive values of SPSC mean that a P-source can be safely land applied; negative values mean that P additions can result in P losses from the soil that can unacceptably contaminate groundwater. The approach has been validated by several researcher s, using manure and biosolids as the P sources (Ola deji, 2006; Chrysostome et al., 2007). However, validation of the SPSC concept with municipal wastewater ap plied to land has not been accomplished. Regulations pertinent to land application of wa stewater have to be considered. According to federal law (USEPA and USDA, 1999), each state Natural Resources Conservation Service (NRCS) must address P in nutrient management practice standards (code 590). Among three strategies to manage P in agri culture, Florida chose to use the P indexing tool (P index), which considers several landscape and management factors demonstrated to affect P loss from fields by 21

PAGE 22

runoff and leaching (Oladeji, 2006). The wastewater applied in RIBs is clearly a P-source, and thus, P index applies to RIB operations. Wastewater P land application regulations need scientific basis, but environmental impacts due to continuous application of wastewater P to soils have received limited research (OConnor et al., 2008), particularly in th e case of RIB systems. A careful determination of the fate of wastewater P on a full scale system would provide useful information in the development of P indices for land application of wa stewater and in the determina tion of the life span of the studied land treatment sites. Hypothesis and Research Objectives Previous studies have successfully validated the SPSC concept for bios olids and fertilizer amended soils. The concept needs to be validated fo r wastewater P, specifically when applied to RIBs, because different P sources have different potential to negatively impact the environment. Additionally, factors representing the soils and op eration of RIBs are of ten unique and require a good understanding of the system to determine the P loss risk associated with the systems. Hypothesis 1: Equilibrium between wastewater P and soil is not fully attained in RIBs because infiltration rates typically far exceed the time necessary to reach P adsorption equilibrium. Hypothesis 2: SPSC predicts the life expectancy of RIBs before P breakthrough to groundwater at environmentally significant concentrations. Hypothesis 3: The change point PSR value of 0.10 (Nair et al., 2004) is a pplicable to land application of wastewater. Hypothesis 4: Long-term wastewater a pplication to the RIBs in creases soil P lability. Hypothesis 5: Long-term wastewater a pplications do not alter th e phosphorus species in the RIBs (treatment soils) as compared to phosphorus species in the soils surrounding the RIBs (control soils). Objective 1: Estimate the degree (percentage) of e quilibrium adsorption between P in wastewater and soil that is attained du ring wastewater application to the RIBs. 22

PAGE 23

Objective 2: Determine the P accumulation in the RIBs and evaluate the capability of the SPSC concept to predict the P breakthrough. Objective 3: Establish the change point that co rresponds to the PSR value above which WEP increases rapidly. Objective 4: Compare water extractable P (WEP) resu lts from treatment and control soils. Objective 5: Characterize the various P fractions pr esent in treatment and control soils. Study Approach The study approach consisted of field sampli ng by depth at four RIBs at Conserv II, located west of Orlando, FL. Soil samples were an alyzed to determine the threshold PSR value and ultimately the SPSC of the studied RIBs at the Conserv II site. Infiltration rates were determined for selected RIBs. In addition, a lab experiment was conducted to determine the kinetics of P adsorption to and release from sele cted soil samples. Las tly, a P fractionation study and chemical equilibrium modeling were conducted to reveal the influence of the wastewater on the soil P speciation on each of the studied RIBs. 23

PAGE 24

CHAPTER 2 MATERIALS AND METHODS Study Area The Water Conserv II land treatment system, located west of Orlando, FL (28.493N, 81.620 W), is the largest system of its type in the world (Figure 2-1). In this 3,000 hectare system, citrus is irrigated with approximately 68 million liters of reclaimed wastewater per day (after tertiary treatment and disinfection). Recl aimed wastewater in excess of citrus needs (approximately 46 million liters per day) is disposed in 72 rapid infiltration basins (RIBs) with a total area of 1,507 hectares (Parsons et al., 2001 ; McFarland et al., 2007; Orange County Growth Management Department, 2006). The size of the basins varies from 0.4 ha to 4 ha. Surface soil with low permeability is excavated and used to construct confining berm s around the basins. Algal growth (due to high solar incidence and high nutrient content in the reclaimed water) can reduce the water infiltration rate. Drying cycles (Figure 2-2) are needed to desiccate the soil and th e algae, recovering the infiltration rate (Metcalf & Eddy, 2007). Anothe r common practice for im proving the infiltration rate in the RIBs is tilling the soil at the e nd of each drying cycle (Sumner and Bradner, 1996). At Conserv II, the tillage depth is approximately 5 cm, and tillage occurs at the end of each drying cycle. Properties of the tertiary wastewater land appl ied at the Conserv II are shown in Table 2-1. The wastewater characteristics are similar to typical consti tuent concentrations found in irrigation water, with the exception of potassium. All waters are ne utrally charged and the sum of positively charged ions equals the sum of negatively charged ions (expressed in meq L-1). In terms of wastewater characterization, the compar ison between cations and anions concentrations is the most fundamental quality assurance/qua lity control (QA/QC) pr ocedure. Differences 24

PAGE 25

between cation and anion analyt ical results greater than 5% should be questioned (Idaho Department of Environmental Quality, 2007). To tal cation and anion concentrations and the percentage difference are presented in Table 2-2. Differences greater th an 5% may be due the fact that dissolved organic car bon (DOC) was not considered in the total anion concentration determinations. Guidelines for evaluation of water quality for irrigation are given in Table 2-3. The problems considered (salinity and infiltration) are related to longterm influences of irrigation on crops and soil properties, respectively. By compar ing Tables 2-1 and 2-3, we conclude that the wastewater applied to the RIBs at Conserv II poses slight to moderate infiltration problems on soils. When very pronounced, infiltration proble ms can be a limiting factor to RIB operations because high infiltration rates ( > 5 cm h-1) are required for the systems (USEPA, 2006). State rule Chapter 62-610, Florida Administrative Code (FAC) regul ates land application of reclaimed wastewater in the state of Florida. Part IV of Chapter 62-610.510, FAC regulations state that at a minimum, pre-application wast e treatment shall result in a reclaimed water meeting secondary treatment and basic disinfecti on levels prior to spreading into the rapid infiltration basins. In addition, nitrate concentr ations in the applied effluent cannot exceed 12 mg L-1. The regulation does not include limits for P concentrations (Florida Department of Environmental Protection, 1999). Because the wast ewater from Conserv II receives tertiary treatment and disinfection, and the average nitrate concentration is 4.82 mg L-1(Table 2-1), there are no legal restrictions on applications to the RIBs. According to the Web Soil Survey from the Natural Resource Conservation Services (NRCS), the soil at Conserv II is mapped as a Candler fine sand (hyperthermic, uncoated lamellic Quartzipsamments). The saturated hydra ulic conductivity (Ksat) for the RIB area is 25

PAGE 26

reported as 33 cm h-1 (Natural Resources Conservation Service, 2008). The Web Soil Survey makes no distinction for saturated hydraulic conductivity (Ksat) values among depths and different RIBs (Natural Resour ces Conservation Service, 2008). Conserv II (Candler) soils are similar to typical beach sand in terms of Ksat (Brady and Weil, 2002). The average bulk density (1/3 bar) for the Conserv II soil equals 1.53 and 1.58 g cm-3 for depths of 0 to 30 cm and deeper than 30cm, respectively (Natural Resources C onservation Service, 2008). Values of Ksat and bulk density are both in accordance with Fares and Alva (1999), who studied another sample of the Candler soil. Other characte ristics of Candler soil are pres ented in Table 2-4. Using the equation developed by Nair and Harris (2004) and the data from Table 2-4, the SPSC of the Candler soil is estimated as 34 mg kg-1. Conserv II operators qualitatively classify the RIBs based on observed wastewater infiltration rates: rapidly infiltrating RIBs are co nsidered good; others are considered not so good. We decided to classify the RIBs by age as well, because age influences the total P loading applied to a given RIB, and the RIBs expected SPSC (remaining capacity to retain P). We designated old RIBs as those operating since 1986, whereas middle aged and young RIBs started operating in 2002, and 2005, resp ectively. We selected four distinct cells within each RIB type for study (Table 2-5). Soil Sampling Eleven soil samples were obtained from the interior of the selected cells at 30 cm depth intervals to 300 cm using a hand auger. Wastewat er was distributed to the cells through two outflows. The two outflows emitted wastewater at similar rates and the wastewater P was assumed to spread outwards, creating concentric circles of approximately equal soil P concentrations. The final distribution of the wa stewater throughout the cells was not uniform, and areas near the outflows were expected to re ceive more wastewater (and, consequently, more 26

PAGE 27

P) than areas away from the outflows. Therefore, the areas near the outf lows were sampled more intensively. Four soil samples were taken at set distances (within 20 m) from the outflows in each half of the cell. There was always one sa mple point in the middle of the cell, and two sample points as far as possible from the outflows, near the cell edges (labeled MIDD and MIDU). The spatial distribution of samples take n from each cell is diagrammed in Appendix A. Control soil samples (no wastewater applied) we re obtained from outside each cell to allow comparison of P characteristics, at a given depth, in soils affected (treatments) or not affected (controls) by wastewater additions. For each cell, 3 control samples were obtained in increments of 30 cm to a total depth that corresponded to 300 cm below the cell interior surface level. The control samples were obtained at horizontal increments of 15 m from the cell edge. During cell construction, the site was excavated so that the top of the soil profile within the cell was below the top of the soil profile outside the cells. We used the excavation depth of the cell to determine the corresponding depth to begi n control sampling. We used a clinometer to determine the elevation difference between the ce ll surface and the surrounding area (outside the cell). Table 2-6 summarizes the vertical differen ces between the surface of various cells and the surrounding areas. Negative numbers i ndicate that the top of the so il profile in the control areas was lower than the surface leve l of the corresponding cells. Soil Analyses Soil samples were analyzed for oxalate extract able P, Al, and Fe and water extractable P (WEP). The oxalate extractable analysis followed the proced ure of Schoumans (2000): 0.5 g of air-dried soil reacted with 30 mL of 0.2 M ammonium oxalate + oxalic acid solution (pH 3) shaken (200 rpm) for 4 hours in the dark. Th e suspensions were filtered through a Whatman number 42 membrane filter paper, and analyzed for oxalate-extractable P, Fe, and Al by inductively coupled argon plasma (ICAP). 27

PAGE 28

The water extractable P determination followed the Self-Davis et al. (2000) procedure. Two grams of air dried samples were shaken (200 rpm) for 1 hour with 20 mL of distilled deionized (DDI) water. We then centrifuged (894 x g) the samples for 5 minutes and filtered (0.45 m) the supernatant. Phosphorus concentrati ons in solution were determined using the ascorbic acid method (Murphy and Riley, 1962). We selected five surface samples and two deep (180 cm or 210 cm and 300 cm soil depth) samples to evaluate soil pH and EC distribut ion through the soil profile. We selected more samples of surface soil than soil deeper in the profile, because a P fractionation study suggested that the pH values in surface so ils of certain cells were very different from native soils. Two control samples (one surface and one deep sample) from each cell were also analyzed for pH and EC. Soil pH was determined in a suspension composed of 1:2 soil to distil led, de-ionized water (DDI) ratio. Measurements were taken directly in the suspension usi ng a pH meter after 30 minutes of static equilibration followed by thre e minutes of suspension stirring (Thomas, 1996). After pH measurements, the suspensions were filt ered and the filtrate analyzed for EC using a conductivity meter. Infiltration Rate Determination The infiltration rate of each of the studied cells was determined using a double-ring infiltrometer, following a standard test method (American Society for Testing and Materials, 2003). The method consists of driving two cylinders of different sizes, one in side the other, into the ground with care so as to minimally disturb the soil. The inner and outer rings, made of galvanized steel, had approximately internal area of 179 cm2 and 540 cm2. The two rings were partially filled with clean water and the head (depth of water) was kept equal and constant in both cylinders, using a Mariotte siphon (American Society for Testing and Materials, 2003). 28

PAGE 29

Measurements of the infiltration rate pertain to water in the inner ring, where the water is assumed to travel vertically through the soil (Metcalf & Eddy, 2007). The volume of water added to the inner ring to maintain the constant head was used as a measure of the volume of water infiltrating the soil (American Society for Testing and Materials, 2003). The outer ring served to promote one dimensional, vertical flow beneath the inner ring (American Society for Testing and Materials, 2003). For each cell, we measured infiltration rate s at two locations (treatment sites): each measurement within 3 m from each wastewater outf low. The infiltration rate was assumed to be equal to the saturated hydraulic conductivity (a t least for the soils first 30 cm) and described how fast the wastewater moves downward through the top soil. Additionally, we measured the infiltration rate at two control sites outside each cell. The ideal scenario would have been to excavate the control sites to the same elevati on as the surface soil of the corresponding cell, because as explained before, the RIBs top so ils were excavated during the construction. However, deep excavation was laborious and time-consuming and, ultimately, deemed unnecessary. Saturated hydraulic conductivity (Ksat) values for the Candl er soil at various depths obtained from the Florida Soil Survey (Natural Resources Conservation Service, 2008) were compared with our measured values. The variability of Ksat values at di fferent depths (Natural Resources Conservation Service, 2008) was smaller than the variability between sites determined by our preliminary analysis (Appendix B). We c oncluded that spatial va riation was the most important factor to be considered rather than variation due to de pth. Thus, control sites were not excavated to the same elevation of the correspon ding cell soil surface. Instead, we excavated the 29

PAGE 30

control sites to 30 cm to remove plants, r oots and excess organic matter and made Ksat determinations on the cleared soil (Figure 2-3). Infiltration rates were determined after the wa ter inflow reached steady-state. Because we wanted to determine the infiltra tion rate under saturated conditions, the smallest value, after at least three repetitions of similar measurements, was adopted as the value for the sampling site. In all cases, we applied water to the rings for at least 20 minutes before measurements began. Readings were taken within 3 or 5 minutes, de pending on the infiltration rate of the sampling location. Comparisons between results from treat ment and control sites were hypothesized to reveal the long-term influence of the wastewater on the infiltration rate of the studied sites. We used the infiltration rate results to estimate the contact time between wastewater-P and soil to estimate the degree of P adsorption equilibrium reached during wastewater flow through the cells. Wastewater Data The Conserv II manager provided a complete ch aracterization of the wastewater that has been applied to the cells since 2003 (Table 21) and the volume of the wastewater applied (Appendix C). We also decided to measure disso lved organic carbon (DOC) in the wastewater because DOC can compete for reactive adsorption sites with P in soils. Lane (2002) cited by Silveira et al. (2006) for instance, found that DOC reduced the capacity of an aluminum rich material water treatment residual to adsorb P. Dissolved organic carbon was determined from a single wastewater sample collected in November 2008. Dissolved organic carbon was determined on filtered (0.6 m) wastewater adjusted to pH 2.0 with ultra-pure 2 M HCl. Total organic carbon (TOC) was measured on a Shimadzu T5000 Total Organic Carbon Analyzer (Mann a nd Wetzel, 1995) and identified as DOC. 30

PAGE 31

PSR Study One of the objectives of the soil analysis was to determine the relationship between WEP and PSR for each cell. However, the range of PSR values found in the soil samples may not be large enough to explain completely the relationship between WEP and PSR. Greater PSR samples can be obtained by appl ying P solutions of various c oncentrations to control soil samples, mimicking what might happen in the field after several years of P wastewater additions. We ran batch equilibrium experiments with control samples from cell 6-21-C and cell 7-6B. Phosphorus solutions were added to increase the soil P loads. Assu ming that the sum of oxalate extractable Fe and Al concentrations of the soil is a good indi cator of the P sorption capacity of the soil (Maguire et al., 2001), we chose soils with relatively small and large Alox + Feox concentrations. The control sample from ce ll 6-21-C was taken at the depth of 270 cm and had Alox + Feox concentration of 813 mg kg-1. The control sample from cell 7-6-B was taken at the depth of 300 cm and had Alox + Feox concentration of 312 mg kg-1. The objective of the study was to create soil samples in equilibrium w ith a variety of P solution concentrations and a large range of PSR values. Lu and OConnor (1999) developed a P sorption isotherm for a Candler soil and correlated P adsorbed with equilibrium P c oncentration, using 8 ini tial P concentrations that ranged from 0 to 10 mg P L-1. For the Candler sample used, the equi librium P concentration was about 1.6 mg L-1 when the initial P concentration was equal to 10 mg L-1 (Lu and OConnor, 1999). The value of 1.6 mg L-1 is similar to the average P concentrati on of the Conserv II wastewater (1.67 mg L1). The following P concentrations: 0, 0.4, 0.8, 1.6, 3.2, 6.4, 9.6, and 12.8 mg P L-1 were used in the current study. The experiment followed the same procedure used by Lu and OConnor (1999): soil solution rati o of 1:10 and reaction time of 5 days. The suspensions were then 31

PAGE 32

centrifuged (894 x g), filtered (0.45 um) and P con centration in the liquid phase was determined using the ascorbic acid method (Murphy and Riley, 1962). We plotted the P adsorbed by the samples as a function of equilibrium P concen tration and fit the data with three models (Langmuir, Freundlich, and the three-paramete r LangmuirFreundlich models) using an excel spreadsheet developed by Bo lster and Hornberger (2007). From the P concentrations in solution (batch equilibration data), we estimated PSR and WEP values for each sample. All P adsorbed by each soil sample was assumed to be oxalate extractable. We considered that the new oxalate extractable P of each samp le was equal to native adsorbed P, plus newly adsorbed P. Because P additions are not expected to change the soils Feox and Alox content, we assumed that Feox and Alox of each sample was the same before and after the experiment. Knowing the concentrations of Pox, Feox and Alox, we were able to calculate the PSR for each sample. The soil:solution ratio and matrix (water) us ed for WEP determinations (Schoumans, 2000) and in the PSR experiment were the same. Thus, the amount of P in solution for each data point in the batch equilibration study was assumed to equal the WEP value of each soil sample. Kinetics Study Phosphorus adsorption kinetics were dete rmined by laboratory analysis. The study consisted of a batch technique where the effect of contact time between soil samples and a P solution was determined. The P solution concentration was 1.67 mg L-1 (similar to the average P concentration of the wastewater effluent applied to the RIBs at Conserv II). The matrix of the P solution was DDI water. We mixed two soil samples (BR2, obtained fr om cell 4-3-B at a depth of 300 cm and Control 3 from cell 6-21-C at a depth of 90 cm) w ith the P solution at a ratio of 1:10. The soils used had the lowest PSR values among the treatment and control samples, respectively. 32

PAGE 33

Mixtures were shaken for: 1 min, 5m in, 10min, 15 min, 30min, 60min, 5h, 10h, 24h, 48h, and 5d. The time periods of 24h and 5d were expect ed to represent 70% and 100%, respectively, of equilibrium between a Candler soil and a P solution (Lu and O'Connor, 1999). We compared the infiltration rate results with the kinetics st udy data and estimated the degree (percentage) of equilibrium between the soil and P solution achieve d during wastewater infiltration in the RIBs. Phosphorus Fractionation Study A P fractionation study was conducted to determin e if the wastewater altered significantly the percentage of various P fractions in each cell. Soil P fractions were determined from twenty samples, five samples from each of the four cells, and analyzed in triplicate. Two soils had large PSR values, one had a small PSR value, and two we re control samples (one from the same level as the cell surface and the second from a 3 m depth). Using a sequential extraction procedure (Chang et al., 1984), we determined the amount of: i) soluble and exchangeable P using 1 M KCl ii) Feand Al-bound P and organic P (Po) using 0.1 M NaOH, iii) Caand Mgbound P using 0.5 M HCl, and iv) residual P in a 6 M HCl digestion. For the first three extractions (using KCl, Na OH, and HCl) a 1:20 soil/solution ratio was used. First we added 20mL of 1 M KCl to a 50ml centrifuge tube containing 1g of soil. After shaking the tubes for 2 h, the solution was centrifuged (894 x g) for 5 minutes and the supernatant decanted and filtered (0.45 m). In sequence, we extrac ted the residuals from the first extraction after adding 20 mL of a 0.1 M NaOH solution to the tubes and shaking for 17 h. After centrifuging and filtering (0.45 m), we started the third extr action by adding 20 mL of 0.5 M HCl solution to the 50 mL tubes. After shakin g the tubes for 24 h, the solution was centrifuged and filtered (0.45 m). The sum of the three fractions re portedly identifies inorganic P, though NaOH can extract some organic P (Silveira et al., 2006). 33

PAGE 34

Residual materials from the NaOH extraction we re transferred to a 50 mL beaker and oven dried for 24 h. Total P concentrati on present in the residual materi al represented residual P. In addition, we independently determined total P (TP) concentrations in the same samples used for P sequential extractions. Total P for both residual P and independently determined TP were determined by ashing and digesting each of the twenty samples, according to Andersen (1976). For both sequential extraction and TP determinati ons, soluble reactive P (SRP) concentrations in the filtrates were determined using the ascorbic acid me thod (Murphy and Riley, 1962). Difference between independently determined TP and the sum of all P fractions were within 15%. In addition, difference between TP de termined from a standard reference material (Standard reference material 1547, National Ins titute of Standards and Technology) and the certified values agreed within 15%. Chemical Equilibrium Modeling Phosphorus speciation in the wastewater-s oil system was modeled using MINEQL+ chemical equilibrium software, version. 4.5 (E nvironmental Research Software, 2002). The soil pH buffering capacity is expected to be several or ders of magnitude greater than the effluent pH buffering capacity; thus, we used measured soil pH values as the pH for our modeled system. However, we assumed no change in wastewater constituent concentrations (Table 2-1) after contact with soil. The ionic strength was estim ated by the model based on the wastewater constituent concentrations (Table 2-1). Standard QA/QC Protocols Standard curves were individually developed for all chemical analyses. Method blanks, certified standards, 5% matrix spike (accuracy), and 5% of duplicates (precision) of the analyzed set were appropriately used. R ecoveries from spikes and duplicates were within 5% of the 34

PAGE 35

expected result. We reran all the analyses that had unacceptable results in terms of the standard QA/QC protocols (Kennedy et al., 1994). Statistical Analysis The relationship between PSR and WEP was estim ated using a split-line model (Nair et al., 2004). The model consists of two linear relationshi ps with significantly different slopes on each side of the change point (McDowell and Shar pley, 2001). We used the equations suggested by Nair et al. (2004): WEP = a0 + b0 PSR, if PSR PSR change point; (2-1) WEP = a1 + b1 PSR, if PSR > PSR change point; (2-2) Where bo = [(a1 a0) + b1 PSR change point]/ PSR change point. (2-3) The SAS software package (SAS Institute 2001) was used in the change point determinations. Additionally, time series anal ysis, using the GLM procedure in SAS (SAS Institute, 2001) and a Tukey test ( P 0.05) were used to determine when P chemical equilibrium was reached during the kinetics study (data not shown). Each studied cell was defined as a single ex perimental unit. Since only one cell of each type was selected, no statistical comparisons between control and treatment samples and among cells were performed. We compared models used to fit the P adsorption data (PSR study) by performing an F test ( P 0.05) and using Akaikes Informa tion Criterion (AIC), where the model with the lowest AIC was considered to be the most likely to be correct (Bolster and Hornberger, 2007). A non-linear plateau model was fit to the data for the kinetics study using the SAS software package (SAS Institute, 2001). The coefficient of determination was calculated using the equation (2-4). 35

PAGE 36

R2 = 1 Sum of Square Error (2-4) Sum of Squares of Model 36

PAGE 37

Figure 2-1. Study area (Google Earth, 2009; Water Conserv II, 2009) Figure 2-2. Typical Conserv II cell du ring: A) wet and B) drying cycles. 37

PAGE 38

Table 2-1. Selected characteristi cs of tertiary wastewater at Conserv II (data from January 2003 to December 2007) and typical constituent concentrations in irrigation water (Food and Agriculture Organization of th e United Nations, 1985). Wastewat er data provided by ConservII personnel Parameter Average Concentration (mg L-1) Standard Deviation Average Concentration (meq L-1) Usual range in irrigation water Bicarbonate 117.43 11.26 1.92 0 10 meq L-1 Boron 0.15 0.02 0.04 0 2 mg L-1 Calcium 43.93 2.66 2.19 0 20 meq L-1 Chloride 79.39 5.66 2.24 0 30 meq L-1 Iron 0.13 0.06 0.01 DNP Manganese 0.01 0.02 0.00 DNP Magnesium 2.85 0.63 0.23 0 5 meq L-1 Nitrite 0.95 0.58 0.02 DNP Nitrate 4.82 0.93 0.08 0 10 mg L-1 Phosphorus 1.67 0.48 0.16 0 2 mg L-1 Potassium 14.01 1.75 0.36 0 2 mg L-1 Sodium 60.37 3.16 2.63 0 40 meq L-1 Sulfate 32.56 3.58 0.68 0 20 meq L-1 Total Nitrogen 8.36 0.96 NA DNP COD 33.16 15.36 NA DNP Electrical Conductivity 0.633 0.023 NA 0-3 pH ** 6.88 0.10 NA 6.0 8.5 Sodium adsorption ratio (SAR) ** 2.38 NA NA 0 15 dS m-1, ** dimensionless, NA non applicable, DNP data not provided, meq L-1 = milliequivalent per litre Table 2-2. Total concentration of cations and anions in tertia ry wastewater at Conserv II Total concentration of cations (meq L-1) 5.46 Total concentration of anions (meq L-1) 5.10 Difference (%) 6.58 Dissolved organic carbon not included 38

PAGE 39

Table 2-3. Water quality guideline s for irrigation salinity a nd infiltration problems (Food and Agriculture Organization of the United Nations, 1985) Degree of Restriction on Use Potential Irrigation Problem Units None Slight to Moderate Severe Salinity ECw dS m-1 < 0.7 0.7 3.0 > 3.0 (or) TDS mg L-1 < 450 450 2000 > 2000 Infiltration (Evaluate using ECw and SAR together) SAR = 0 3 and ECw = > 0.7 0.7 0.2 < 0.2 = 3 6 = > 1.2 1.2 0.3 < 0.3 = 6 12 = > 1.9 1.9 0.5 < 0.5 = 12 20 = > 2.9 2.9 1.3 < 1.3 = 20 40 = > 5.0 5.0 2.9 < 2.9 Table 2-4. Selected characteris tics of Candler A horizon (data from OConnor et al., 2002; Lu and OConnor, 1999; and Natural Res ources Conservation Service, 2008) Parameter Value Pox (mmol kg-1)* 1 Feox (mmol kg-1)* 3.54 Alox (mmol kg-1)* 10.5 PSR** 0.07 Sand (%) 93 Clay (%) 1.5 Organic Matter (%) 1.9 pH 5.3 *Pox, Feox, and Alox, are, respectively, ammonium oxalate extractable phosphorus, iron and aluminum. **PSR refers to Phosphorus Saturation Ratio Table 2-5. Classes of selected RIBs Cell ID Class Number of infiltration cells RIB7-6-B Old and good 3 RIB6-21-C Old and not so good 4 RIB4-3-B Middle aged and not so good 2 RIB2-3-B Young and good 2 Table 2-6. Elevation differences between the top of the soil profile for treatment and control sample sites. Cell level difference 1 (cm) level difference 2 (cm) level difference 3 (cm) 2-3-A 75 75 75 4-3-B 75 -15 -15 6-21-C 90 90 90 7-6-B 90 90 90 Level difference 1 refers to the difference in surface level of the profile between treatment and first control sample sites Level difference 2 refers to the differen ce in surface level of the pr ofile between treatment and second control sample sites Level difference 3 refers to the difference in surface level of the profile between treatment and third control sample sites 39

PAGE 40

Figure 2-3. Infiltration rate measurement from a control site using a doub le-ring infiltrometer. 40

PAGE 41

CHAPTER 3 RESULTS AND DISCUSSION Ground Water Characterization The Conserv II manager provided groundwater qu ality data for a well (well 6-30) deemed to best represent unimpacted ground water conditions in the Conser v II area. Other wells at the sight are also considered background wells, but water quality in the wells (e.g., well 6-23) is believed altered by P contributions from other RI Bs or other operations (e.g. agriculture, golf courses) up gradient. Well 6-30 is up gradient from the monitoring we ll (6-23) adjacent to RIB 621, and is not expected to be influenced by this, or other, RIBs. We averaged and compared groundwater P concentration data collected between 1986 and 2008 from wells 6-30 and 6-23, both of which represent shallow groundwater. Th e average and median water depths for well 630 were 3.9 m and 3.7 m, respectively, and for well 6-23 4.2 m and 4.0 m, respectively, from 1986 to 2008. We refer to well 6-30 as a backgr ound (unimpacted) well and to well 6-23 as a treatment well (well that is influenced by the RIB treated with P wastewater). Average groundwater P concentrat ions for treatment (0.49 mg P L-1) and background monitoring wells (0.05 mg P L-1) are smaller than the average wastewater P concentration (1.67 mg P L-1) (data not shown). As expected, the averag e P concentration for the background well is less than the P concentration of the treatment monitoring well. The background monitoring well seems to represent local native groundwater quality because the average P concentration in the background well is approximately 10 times smaller th an the concentration in the treatment well. The average P concentration in the unimpact ed background monitoring well is less than the critical P solution con centration (0.10 mg P L-1) guideline established by USEPA for streams (USEPA, 1986) and than the acceptable P level for shallow ground waters in the Netherlands 41

PAGE 42

(Breeuwsma et al., 1995). The P concentration in the background monitoring well exceeded the 0.10 mg P L-1 threshold only in 10% of the measurements over 22 years. We also obtained groundwater quality data fr om other treatment m onitoring wells adjacent to each studied RIB and plotted P concentrations over time (Appendix D). For wells adjacent to old RIBs, there was a trend for P concentrati on to increase after 1998 and 1991 (wells 6-23 and 7-08). The increases probably reflect movement of P from the RIBs to groundwater as a result of long-term P additions. Phosphorus concentrations fo r the well adjacent to cell 7-6-B (Figure D-7) were 1.67 mg L-1 (the average wastewater P concentr ation) since 1994, suggesting minimum P removal from the wastewater by the soil. We detected little or no tendency for P to increase over time for the other wells, probably because insuffic ient P has been applied to the middle aged and new RIBs to cause extensive P leaching through the soil to groundwater. Soil Analysis Phosphorus Saturation Ratio (PSR) Average PSR values were generally greater in the treatment than in the control soils (Figures 3-1, 3-2, 3-3, and 3-4). The increases in oxalate-extractable P concentrations in treated soils were expected because wastewater had been applied to the cells from 3 to 22 years. In addition, the variability among the PSR values (evi denced by the error bars ) is less for control soils than for treatment soils. Application of wa stewater was not uniform throughout a cell; soil regions near outflows of the ce lls received much more wastewater than regions at cell perimeters. Thus, over time, the soils near the outflows accumulated more P than further away, increasing the overall variability of P accumulati on across the cell. The treatment and control soils initially had similar and uniform PSR values, but over time the treatment soils developed greater PSR spatial variability because of the uneven wastewater P distribution. 42

PAGE 43

Cells 2-3-A and 7-6-B (good RIBs) yielded ex pected PSR results (Figures 3-1 and 3-4, respectively): extensive wastewat er P application has increased the PSR values at all depths relative to the controls. Due to the high infiltration rate char acteristic of good RIBs, most of the wastewater infiltrates in th e areas near the outflows. The PSR values of soils sampled from near the outflows (within 3 m) we re greater than the values for samples far from the outflows (near the cell edges) (Appendix E) The data demonstrate that th e wastewater is not uniformly distributed throughout the cell and, thus, parts of the cell receive more ef fluent-P than others. Cell 7-6-B received greater cumulative P wast ewater volume than cell 2-3-A because cell 7-6-B is an old RIB, whereas cell 2-3-A is a new operation (Appendix C). The PSR values were always greater for treatment samples taken from the older cell 7-6-B than for treatment samples taken from the newer cell 2-3-A (Figure 3-5). Cell 4-3-B (Figure 3-2) had si milar PSR values as cells 23-A and 7-6-B (Figure 3-1 and Figure 3-4 respectively): wastewat er application has increased th e average soil P concentration and, consequently, the average so il PSR (Figure 3-2). However, c onsidering the error bars, the difference between treatment and control samp les might not be statistically different. Results for cell 6-21-C were not expected (Figur e 3-3). Despite the increase in average soil PSR after long-term effluent-P application (com parison between treatmen t and controls), the increase (especially for deeper samples) was not very pronounced (Figure 3-3). One possible explanation is that the forms of P that have accumulated in cell 6-21-C are less extractable by oxalate acid as compared to soil P from other ce lls (e.g. cell 2-3-A). The hypothesis was tested in the P fractionation and modeling studies. For cells 4-3-B and 6-21-C, the PSR values of soils sampled from near the outflows (within 3 m) were very similar to values for sa mples far from the outflows (near the cell edges) 43

PAGE 44

(Appendix E). The data suggest th at the wastewater is more e qually distributed throughout the not so good cells than throughout good cells. For certain depths, PSR values for samples far from the outflows exceeded PSR values for samples near the outflows. Water Extractable Phosphorus (WEP) A distinction between WEP values for treatments and control samples was observed in all the studied cells (Figures 3-6, 37, 3-8, and 3-9). The distinction is even clearer for old cells (621-C and 7-6-B): for both cells, WEP values in tr eatment samples were always greater than WEP values in controls. We conclude that wastewater-P application to the studied cells has a potential to increase the soil WEP, and the increase depends on the cells age. Water extractable P (WEP) is well correlated with P leaching from soils (Zhang et al., 2002). Thus, long-term wastewater P application to the studied cells has increased the potential of P leaching from the soils, with possible movement to groundwater, mainly in old cells. Change Point Determination Water extractable P (WEP) values for soil samples were plotted against determined PSR values for each cell (Appendix F). We separately plotted WEP versus PSR for samples taken within 15 m and more than 15 m from the outfl ows (Figures F-5 and F-6, respectively). The main objective of plotting WEP versus PSR is to determine the change point. The relationship between PSR and WEP values was estimated for each cell using a splitline model. To determine the change point, two di stinct linear relationships with significantly different slopes on each side of the change point are necessary (e.g., Figure 3-10). For all cells, except cell 2-3-A (Figure 3-10), no change point was found. For cell 2-3-A, the change point was estimated as 0.08 and two linear equations, wi th different slopes, were determined: WEP = -0.764 + 34.7*PSR, for PSR less than 0.08 (3-1) WEP = -4.05 + 75.7*PSR, for PSR greater than 0.08 (3-2) 44

PAGE 45

However, the relationship between PSR and WE P for cell 2-3-A could nearly as well be described as a linear relationship (R2 = 0.37). The determined change point value is similar to the change point PSR of 0.10 suggested by Nair et al. (2004) for surface horizons of Flor ida sandy soils. The PSR value of 0.08 corresponds to a WEP of 2.0 mg P kg-1 (using equations 1 or 2), or a so lution P concentration of 0.2 mg P L-1. The 0.2 mg P L-1 concentration is greater than the critical P concentration (0.1 mg P L-1) suggested by Nair et al. (2004) and the average P concentrat ion in the background monitoring well (0.05 mg P L-1). Values of WEP for all the cells except cell 2-3-B inexplicably failed to change with increasing PSR increments. One of the hypotheses for the unexpected result is that long-term wastewater applications to the cells have change d the soil P chemistry (e.g. changed the pH or added significant amounts of cations like calcium). Another hypothesis is that the soilwastewater system is not in equilibrium in term s of P adsorption. As a result, we conducted an experiment that mimicked the long-term P applic ation to the cells without interference of other chemicals to test the first hypothesis. To test the second hypothesis, we determined the degree (percentage) of P equilibrium in the cells by laboratory and field measurements. PSR Study The plots for P adsorption as a function of equilibrium P concentration for the control sample from cell 6-21-C at a depth of 270 cm an d for the control sample from cell 7-6-B at a depth of 300 cm are presented in Figure 3-11. Th e data fit best a Langmuir-Freundlich adsorption model. For both samples, the Langmuir -Freundlich model had the greatest R2 and the smallest AIC values among the three models evaluated (Table 3-1). The P adsorption rate decreased as the equilibrium P concentration increased, approach ing zero when P adsorption was close to the maximum value. 45

PAGE 46

The adsorption plots had expected results: the amount of adsorbed P increased as P concentration in solution increased until a predicted maximum value (62.4 mg P kg-1 and 56 mg P kg-1 for samples from cells 6-21-C and 7-6-B, resp ectively) that was not reached in the study. The PSR experiment enabled the determina tion of expanded WEP and PSR relationships for control soils from cells 7-6-B (Figure 3-12) and 6-21-C (Figure 3-13). The additional values for WEP as a function of PSR, along with results obtained from the initia l analysis of the soil samples of each cell, are include d in the plots. The additional va lues obtained from the PSR lab experiment are different from the results from th e field treatment soil analysis. Results from the PSR study seem to agree with the literature (eg. Na ir et al., 2004): for a given increment of PSR, WEP values increase dramatically after a PSR of approximately 0.1 (change point). There were at least two differences between the conditions in the PSR study and in the field. First, although approximately the same P concen trations were used in both situations, in the PSR study, distilled de-ionized (DDI) water was used as the P solution matrix, whereas the wastewater applied to the RIBs contain several other constituents (Table 2-1). Another study (P fractionation study) was conducted to estimate the formation/presence of various P fractions present in the soils of each st udied cell. The difference between P species found in control and treatment soil samples can reveal the influence of the wastewater constituents on soil P retention. The second distinction refers to the fact that under field conditions, P reactions are probably not in equilibrium (wastewater flows to o rapidly through the soil). The hypothesis is addressed later in the chapter. Lu and OConnor (1999) showed that 5 days was required to attain P adsorption equilibrium in the Candler soil. Phosphorus Fractionation Study The sum of all P fraction concentrations was similar to the TP independently determined, with the mass balance recovery va rying from 84% to 109%. The percentage of KCl-P, considered 46

PAGE 47

to represent the labile P fraction, increased 1.5 to 10 fold from control to treatment samples (Table 3-2). Long-term wastewater application to the cells is expect ed to increase the percentage of labile P in the soil. The percentages of P extracted by each solu tion is shown in Table 3-2, and Figures 3-14 and 3-15 and illustrate the difference in P fractionation between treatments and controls. Phosphorus concentrations were greater in the sodium hydroxide (NaOH), followed by the residual, the hydrochloric acid (HCl), and finally the potassi um chloride (KCl) fractions. Assuming that most of the organic P is represente d by the residual fraction, the majority of P was in inorganic forms. For most of the sample s analyzed, NaOH-P was th e dominant P fraction, suggesting P association with iron and aluminum oxides (Otani and Ae, 1997) as the primary mechanism of P removal from the wastewater. The exception was the surface sample BL from the cell 6-21-C that had 76% of P extracted by HCl-P. The unexpected result prompted running the same P fractionation procedure for five su rface samples from cell 6-21-C, including the sample BL (Table 3-3 and Figure 3-16). For 3 of the 5 samples, HCl-P was the pre dominant fraction. For the other two samples, NaOH-P was the dominant fraction, but the percentage of HCl-P was greater th an in the first run. Cell 6-21-C is in an old RIB and we hypothesize that long-term wastewat er application changed the P adsorption chemistry, at least in the top 30 cm of the RIB. Another measurement pertinent to the hypothesis is the soil pH, due to its crucial role in P speciation in soils. Soil pH Long-term wastewater application to RIBs appears to have increased soil pH (Table 3-4). In most cases, pH values for treatment soil sample s are greater than pH values for control soil samples. Average wastewater pH (6.88) is greater than all the measured pH values of native soils (Table 3-4) and long-term wastewat er application to the cells is expected to increase soil pH. In 47

PAGE 48

some cases, the pH of treatment soils was even greater than the average wastewater pH. One possible explanation is that some alkaline co mpounds (e.g., calcium carbonate) accumulated in the soil (mainly in the surface horizons) after pr ecipitating out from the wastewater causing an increase in soil pH. However, as the modeling exercise di d not predict CaCO3 formation, CaCO3 may have simply been present in the material le ft on the surface during RIB construction. In fact, we observed the presence of rock fragments (possibly CaCO3) in the surface of old cells, especially cell 6-21-C. To evaluate the presence of CaCO3 in the surface of cell 6-21-C, we ran an x-ray analysis in a single sample (BL2 30c m) but we were unable to detect the presence of CaCO3 or Ca-P minerals. The inability of the x-ray analysis to detect a chemical does not imply absence of the compound. When the concentrations are too low (<1%) or if the chemical is in a non-crystalline form, the compound may not be dete cted (Josan et al., 2005). We then added a weak acid (0.1 M HCl) to the sample under the micros cope and we detected fizzing, which indicated possible presence of CaCO3 in the sample. The greatest pH values correspond to old RI Bs; middle aged and young RIBs always had pH values below 7 (Table 3-4). Although RIB 6-21 has received 8% less effluent than RIB 4-3 (Appendix C), RIB 4-3 began operating in 2003, whereas RIB 6-21 began operation in 1986, when wastewater quality (including greater pH) was different (Appendix G). Application of a wastewater with greater pH might have increased the soil pH in old RIBs more than the soil pH from RIBs that began operation more recently. Table 3-5 compares soil pH with the amount s of P extracted in the NaOH-P and HCl-P fractions. When HCl-P was the dominant P fracti on, soil pH was > 7. Phosphorus is more likely to exist as Caand Mg-P forms under alkaline conditions and to be extracted by HCl. However, in other cases, even with pH values > 7, the dominant fraction remained NaOH-P. The 48

PAGE 49

percentage of HCl-P fraction was always greater fo r treatments than for controls from old RIBs, suggesting an effect of pH and long-term wastew ater application on the formation of Ca and Mgbound P. Phosphorus speciation depends not only on the soil pH, but also on the soil and wastewater chemical composition. Chemical Equilibrium Modeling As each cell has different soil pH values, we expected to find different percentages of P species for each cell simulation. For cells 7-6-B a nd 6-21-C, about 95% of the P was predicted to exist in the mineral hydroxylapatit e, and remainder as soluble H2PO4 and HPO4 -2. The predictions suggested possible form ation of Ca-P or Mg-P minerals in the two cells, as indicated by the P fractionation study for surface samples of cell 6-21-C. However, for cell 7-6-B, the P fractionation study indicated that P association with Fe and Al dominated, rather than Ca-P or Mg-P association. Fragments of calcium carbonate (CaCO3) observed in the cell 6-21-C soil surface, probably due to natural geological formati on, are likely to be one of the reasons why CaP formation was more pronounced in the cell 6-21-C than in the cell 7-6-B. Data generated through the modeling exercise are presented in Table G-1. In the simulations for the other two cells, al most all (>98%) the P present in the soilwastewater system was predicted to be soluble forms. The mode ling results for cells 2-3-A and 4-3-B confirmed the results from the P fractionati on study: P association with Ca and Mg in the two cells is unlikely due to the low pH. However, the difference in average pH between all the cells is less than one unit. The pH range that we are dealing with in th e present study is very unstable in terms of P speciation, because pH ch anges (e.g. between cells 4-3-B and 6-21-C) as small as 0.8 units, drastically altered the predic ted P removal mechanism in the systems, as observed using the chemical e quilibrium model (Table G-1) 49

PAGE 50

Kinetics Study Plots of P adsorbed as a function of reacti on time are shown in Figure 3-17 for cell 4-3-B and in Figure 18 for cell 6-21-C. Time series anal ysis suggests that the treatment sample from cell 4-3-B reached P sorption equilibrium in 5 minutes, whereas the control sample from cell 621-C reached equilibrium in 2 days. As the sample from cell 6-21-C took the longest time to reach equilibrium, wastewater flow through the cell likely represents less P ad sorption, compared with the sample from cell 43-B. To calculate degree of equilibrium adsorptio n between P in wastewater and soil that is attained during wastewater applic ation to each cell, we used kinetic data for cell 6-21-C (more conservative result). Kinetic data for cell 4-3-B likely overestimate P adsorption in the field in most cases due to extremely fast P adsorption kinetics. Infiltration Rate Field Measurements Field measurements of inf iltration rates were conducte d on 10/27/08 and 11/14/08. We intended to determine infiltration rates for the trea tments on the first visit and for the controls on the second visit. However, on the second vis it we took additional measurements from the treatment cell 7-6-A to confirm previous results. Due to time constraints, we were able to determine the infiltration rate of only one control sample from cell 6-21-C and no control samples from cell 4-3-B. The distribution of infi ltrations rates, measured each three or five minutes are presented in Appendix H. Steady-state infiltration rate results are pres ented for good and not so good cells in Tables 3-6 and 3-7, respectively. The terminolog ies good and not so good are qualitative and reflect Conserv II operators obs ervations of wastewater infilt ration in the cells. Cell 2-3-A (good) had the greatest average treatment inf iltration rate among all the cells. Unexpectedly, 50

PAGE 51

treatment samples from cell 7-6-A (good) had sim ilar or even smaller infiltration rates than not so good cells (4-3-B and 6-21-C). During wastewater application, several factors can affect water infilt ration through the soil, including textural discontinuities and water table depth. During field sampling, we observed a thin clay layer present at 2.7 to 3 m in the not so good cells. When a textural discontinuity occurs this deep in th e soil profile, infiltration measuremen ts using a double-ring infiltrometer at the surface are likely minimally affected. However, the field infiltration rate can be very different after days of wastewater discha rge in the RIBs. Deep water ta bles can become shallower and drastically reduce the infiltration rate. Infiltrati on rates measured on surface soils, however, may not reflect the effect of raised water tables unless the water tabl e is very close to the surface. Results in Tables 3-6 and 3-7 represent infiltra tion rates only of shallow soil depths, and not expected to be affected by water table depth. The use of the qualitative classification of each RIB as good and not so good remains necessary to more accurately identify field conditions. Infiltration rates measured in control soils we re always greater than rates measured in treatment soils. There are several possible explan ations: 1) Control and treatment measurements were taken at different depths reflecting differ ent soil conditions. The explanation is unlikely because spatial variations in rate measurements we re much more important than variations due to depth (Appendix B). 2) Chemical constituents in the wastewater modify soil structure by promoting clay dispersion. This explanation is also unlikely as wastewater Na concentrations (and SAR values) suggest little dispersion eff ects and only slight to moderate infiltration limitations, particularly on the very sandy soils (T able 2-1). 3) Algal growth in the cells may reduce the infiltration rates. Similar problems have been reported in other RIB systems (Metcalf & Eddy, 2007; USEPA, 2006). Periodic soil tillage on Conserv II may not be sufficient in 51

PAGE 52

frequency or depth to restore native infiltration rate conditions. In fact, Sumner and Bradner (1996) reported that, in a RIB system similar to Conserv II operations, long-term incorporation of the cake (formed by deposition of fine material s and algal growth in the soil surface) within the upper profile reduced the cell infiltration cap acity. 4) Long-term tillage can cause fine particles to migrate to a deeper layer possibl y developing a deep clogg ing layer (USEPA, 2004). If enough wastewater were applie d to the cells, the clogging laye r could reduce the wastewater infiltration rates. Comparison between Data from the Kinetics Study and from the Field The main goal of the infiltration rate meas urements was to estimate the contact time between the soil and the wastewater-P and to co mpare the estimate with kinetic data generated for P adsorption. However, there are clear diff erences between the condi tions of the kinetics study and the field conditions. The kinetics st udy was a batch (vigorous shaking of a high solution to solid ratio suspension) study that maximizes solid/solute contact, homogeneity, and reaction rates. Flow in the RIBs is continuous, likely approa ching ideal piston-flow, with maximal flow in large pores and the possibility of heterogeneous mixing and contact of solutes with P-retention surfaces. To compare results from both studies, we had to assume that the hydraulic regimes were similar and that the wastewater (P) movement in the field was homogeneous. The contact time between the wastewater and so il was established as the time necessary for the wastewater to travel a give n vertical distance in the fiel d. The chosen distance was 30 cm since the double-ring infiltrometer best represents the infiltration rate of the surface soil. We calculated the contact time (i.e. travel time) for each cell, because each has different infiltration rates. The infiltration rates were averaged for treatment samples from each cell (Tables 3-6 and 3-7). Cells 2-3-A, 4-3-B, 6-21-C, and 7-6-B ha d the following average treatment infiltration 52

PAGE 53

rates, respectively: 78.5 cm h-1, 59.6 cm h-1, 34.6 cm h-1, and 16.0 cm h-1. To determine the contact time we used the following equation: I = D T-1 (3-3) where I is the infiltrati on rate, T is the travel time, D is the vertical distance, adopted as 30cm. We obtained the following contact times for cells 2-3-A, 4-3-B, 6-21-C, and 7-6-B, respectively: 22.9, 30.2, 52.0, and 112 minutes. The re lationship between P adsorbed by the soil and shaking time was well explained (R2 = 0.92) by a non-linear plateau equation: P adsorbed = 3.92+7.80*(1-e(-0.12*shaking time)) (3-4) Where P adsorbed is in mg kg-1 and shaking time in hours. We determined the P adsorbed for each cell: 4.27, 4.38, 4.69, and 5.49 mg kg-1 for cells 23-A, 4-3-B, 6-21-C, and 7-6B, respectively, and the maximum P adsorbed by the soil by averaging the adsorption values after 2 days, when the system reaches equilibrium (Figure 3-19). Using equation 3-4, we found that the maximum P adsorbed using a P solution of 1.67 mg P L-1 was equal to 11.7 mg kg-1. The value agrees with the resu lts from the PSR study: average P adsorbed after 5 days of contact betw een soil and a P solution of 1.6 mg P L-1 was equal to 11.6 mg kg-1 and 11.9 mg kg-1 for the samples from cells 6-21-C and 7-6-B, respectively (data not shown). We determined the degree (percentage) of P adsorption equilibrium reached in each cell by comparing the estimated P adsorbed for each cell with the maximum value of 11.7 mg P kg-1. The percent P adsorption equilibrium reached was 37%, 37%, 40%, and 47% for cells 2-3-A, 43-B, 6-21-C, and 7-6-B, respectively. At least for surface horizons, the wastewater apparently flows by the soil much faster (is in contact with soil much less) than the minimum time required 53

PAGE 54

for adsorption equilibrium. A reduction in the in filtration rates could increase the contact time between soil and wastewater P. The increase in contact time can potentially enhance the capacity of the soil to remove P from the wastewater, because as presented in Figure 3-15, before equilibrium, the longer the contact time th e greater the P adsorption to the soil. 54

PAGE 55

Figure 3-1. Average treatment (n=11) and cont rol (n=3) PSR value distribution throughout the soil profile for soils from cell 2-3-A (young and good). Soil surface of the cell 2-3A represents zero depth. Error bars re present standard deviations for each corresponding depth. Figure 3-2. Average treatment (n=11) and control (n=3) PSR value distribution through the soil profile for soils from cell 4-3-B (middle ag ed and good). Soil surface of the cell 43-B represents zero depth. Error bars re present standard deviations for each corresponding depth. 55

PAGE 56

Figure 3-3. Average treatment (n=11) and control (n=3) PSR value distribution through the soil profile for soils from cell 6-21-C (Old and not so good). Soil surface of the cell 621-C represents zero depth. Error bars re present standard deviations for each corresponding depth. Figure 3-4. Average treatment (n=11) and control (n=3) PSR value distribution through the soil profile for soils from cell 7-6-B (Old and good). Soil surface of the cell 7-6-B represents zero depth. Error bars represen t standard deviations for each corresponding depth. 56

PAGE 57

Figure 3-5. Average PSR distribu tion through the soil profile for treated soils from cells 2-3-A (young and good) (n=11) and 7-6-B (old a nd good) (n=11). Error bars represent standard deviations for each corresponding depth. Figure 3-6. Average treatment (n=11) and c ontrol (n=3) WEP distri bution throughout the soil profile for soils from cell 2-3-A (young and good). Soil surface of the cell 2-3-A represents zero depth. Error bars represent standard deviations for each corresponding depth. 57

PAGE 58

Figure 3-7. Average treatment (n=11) and c ontrol (n=3) WEP distribution through the soil profile for soils from cell 4-3-B (middle ag ed and good). Soil surface of the cell 43-B represents zero depth. Error bars re present standard deviations for each corresponding depth. Figure 3-8. Average treatment (n=11) and c ontrol (n=3) WEP distribution through the soil profile for soils from cell 6-21-C (Old and not so good). Soil surface of the cell 621-C represents zero depth. Error bars re present standard deviations for each corresponding depth. 58

PAGE 59

59 Figure 3-9. Average treatment (n=11) and c ontrol (n=3) WEP distribution through the soil profile for soils from cell 7-6-B (Old and good). Soil surface of the cell 7-6-B represents zero depth. Error bars represen t standard deviations for each corresponding depth. WEP = -0.764 + 34.7 PSR Change point (PSR 0.08) WEP = -4.1 + 76 PSR Figure 3-10. The relationship be tween WEP and PSR values for treatment and control samples from cell 2-3-A (young and good).

PAGE 60

Figure 3-11. Phosphorus adsorption data (fi tted by the LangmuirFreundlich model) for control samples from: A. cell 6-21-C (old and not so good) at a depth of 270cm (F-test P < 0.0001 and R2 = 0.990) and B. cell 7-6-B (old and good) at a depth of 300cm (F-test P = 0.00077 and R2 = 0.968). Each data point represents a si ngle measurement (Bolster and Hornberger, 2007). 60 Table 3-1. Comparison between fits w ith three models in the PSR study Cell where the sample was taken from 7-6-B 6-21-C Model R2 AIC R2 AIC Langmuir 0.947 78.9 0.920 99.3 Freundlich 0.923 88.6 0.893 106.5 Langmuir-Freundlich 0.968 68.6 0.990 51.1

PAGE 61

Figure 3-12. WEP as a function of PSR for treatme nt and control field samples, and for samples used in the laboratory PSR study for cell 7-6-B (old and good). Figure 3-13. WEP as a function of PSR for treatme nt and control field samples, and for samples used in the laboratory PSR study for cell 6-21-C (old and not so good) 61

PAGE 62

Table 3-2. Phosphorus fraction distribution. 62 BDL = Below the detection limit, MS = Missing sample

PAGE 63

Figure 3-14. Percentage of phosphorus distribu tion among the various fractions for treatment soils, relative to the sum of all P fractions. Figure 3-15. Percentage of phosphorus distribution among the various fractions for control soils, relative to the sum of all P fractions. 63

PAGE 64

Table 3-3. Phosphorus fraction dist ribution of surface samples from cell 6-21-C (old and not so good). Sequentially extracted P Sum of all Independent KCl-P NaOH-P HCl-P Residual-P fractions Total P Average Percentage of Average Percentage of Average Percentage of Average Percentage of determination ID (mg kg-1) the total (%) (mg kg-1) the total (%) (mg kg-1) the total (%) (mg kg-1) the total (%) (mg kg-1) (mg kg-1) BL2 18.4 4.31 29.0 6.81 365 85.5 30.7 7.20 443 427 BL 8.94 6.70 58.9 44.1 40.4 30.3 37.7 28.2 146 134 MIDR 7.07 3.20 127 57.4 16.2 7.34 66.2 29.9 216 221 BR1 20.5 6.43 72.1 22.7 233 73.2 51.1 16.1 377 318 BR2 21.5 4.45 87.3 18.0 398 82.2 50.3 10.4 557 484 64 Figure 3-16. Percentage of phosphorus dist ribution among the various fractions for trea tment surface samples from cell 6-21-C (old and not so good), relative to the sum of all P fractions. 64

PAGE 65

16.0 16.3 16.6 16.9 020406080100120 Shaking time (h)P adsorbed (mg kg-1) Figure 3-17. Change in P sorption with shak ing time for the sample BR2 from cell 4-3-B (middle aged and not so good) at 300 cm depth. Error bars repr esent the standard deviation for each corresponding shaking time. 2.0 4.0 6.0 8.0 10.0 12.0 0 20406080100120 Shaking time (h)P adsorbed (mg kg-1) Figure 3-18. Change in P sorption with shaki ng time for the control sample from cell 6-21-C (Old and not so good) at 90 cm depth. Error bars repres ent the standard deviation for each corresponding shaking time. 65

PAGE 66

Table 3-4. pH values for selected treatmen t and control samples from each studied cell. Treatment pH values corresponding to depths of 30 cm are averages (n=5). Treatment Samples Control Samples Cell Depth (cm) ID pH Cell Depth (cm) ID pH 2-3-A 30 Average 6.15 2-3-A 75 CTRL1 6.75 2-3-A 120 MIDU 5.59 2-3-A 375 CTRL1 6.46 2-3-A 180 BL 6.65 2-3-A 300 BL 6.78 4-3-B 30 Average 6.03 4-3-B 75 CTRL1 5.95 4-3-B 180 BL 6.28 4-3-B 375 CTRL1 5.88 4-3-B 270 MIDR 6.42 4-3-B 300 BR2 6.69 4-3-B 300 BL 6.28 6-21-C 30 Average 7.09 6-21-C 90 CTRL1 5.86 6-21-C 210 BL 7.00 6-21-C 390 CTRL1 6.48 6-21-C 210 BR1 6.85 6-21-C 300 BL 6.91 7-6-B 30 Average 7.05 7-6-B 90 CTRL1 6.76 7-6-B 180 BR 7.13 7-6-B 390 CTRL1 6.59 7-6-B 300 BL1 7.13 66

PAGE 67

Table 3-5. Comparison between pH and concen trations of P extracted by NaOH and HCl. NaOH-P HCl-P Depth Average Percentage of Average Percentage of Cell (cm) ID pH (mg kg-1) the total (%) (mg kg-1) the total* (%) 6-21-C 30 BL2 7.79 29.0 6.81 365 85.5 6-21-C 30 BR1 7.13 72.1 22.7 233 73.2 6-21-C 30 MIDR 7.05 127 57.4 16.2 7.34 6-21-C 30 BL 6.85 58.9 44.1 40.4 30.3 6-21-C 30 BR2 7.05 87.3 18.0 398 82.2 6-21-C 210 BL 7.00 198 90.7 12.5 5.72 6-21-C 210 BR1 6.85 72.5 69.7 3.97 3.82 6-21-C 90 CTRL1 5.86 51.0 54.9 2.49 2.68 6-21-C 390 CTRL1 6.48 129 89.7 6.64 4.61 7-6-B 30 MID 7.09 102 57.5 12.2 6.87 7-6-B 180 BR 7.13 73.3 63.3 5.29 4.56 7-6-B 300 BL1 7.13 73.5 65.0 4.25 3.75 7-6-B 90 CTRL1 6.76 35.4 47.2 2.12 2.83 7-6-B 390 CTRL1 6.59 114 62.1 2.37 1.29 4-3-B 30 MIDL 5.95 60.2 61.7 4.01 4.11 4-3-B 270 MIDR 6.42 328 61.3 0.429 0.080 4-3-B 300 BR2 6.69 44.7 20.3 0.116 0.053 4-3-B 75 CTRL1 5.95 68.4 57.3 19.1 16.0 4-3-B 375 CTRL1 5.88 54.6 57.6 3.24 3.42 2-3-A 30 BL 6.76 44.5 59.1 3.04 4.04 2-3-A 180 BL 6.65 55.4 76.5 3.07 4.24 2-3-A 120 MIDU 5.59 28.1 50.7 0.650 1.17 2-3-A 75 CTRL1 6.75 35.5 52.0 3.82 5.60 2-3-A 375 CTRL1 6.46 39.2 62.5 2.61 4.15 Relative to independe ntly determined TP. Table 3-6. Infiltration rate results from field measurements at good RIBs. Cell 2-3-A 7-6-A Sample ID BL BR CTRL2 BL* BR* BL** BR** CTRL2 CTRL3 Infiltration Rate 88.6 68.3 129 10.1 16.3 9.20 28.5 36.8 70.2 (cm h-1) BL and BR refer to treatment samples taken with 3 m from the RIBs effluent outflows. CTRL 2 and CTRL 3 refer to control samples taken within 30 m and 45 m from the edge of the corresponding RIB, respectively. and ** = measurements taken on 10/27/08 and 11/14/08, respectively. 67

PAGE 68

Table 3-7. Infiltration rate results from fi eld measurements at not so good RIBs. Cell 4-3-B 6-21-C Sample ID BL BR BL BR CTRL2 Infiltration Rate 26.5 92.7 29.0 40.2 97.7 (cm h-1) BL and BR refer to treatment samples taken with 3 m from the RIBs effluent outflows. CTRL 2 refers to control samples taken within 30 m from the edge of the corresponding RIB. 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 0.0020.0040.0060.0080.00100.00120.00 Shaking time (h)P adsorbed (mg kg-1) P < 0.0001 R2 = 0.92 P adsorbed = 3.9191+7.7976*(1-e(-0.1174*shaking time)) Figure 3-19. Observed averages of P sorption as a function of shaking time for the control sample from cell 6-21-C (Old and not so good) at 90 cm depth and fitted non-linear model. Error bars represent the standa rd deviation for each corresponding shaking time. 68

PAGE 69

CHAPTER 4 CONCLUSIONS The Conserv II site is an unique study site because wastewater has been applied for many years in a full scale system. Despite the fact that field conditions are difficult to model, the results from the long-term fiel d study may provide useful information in the development of rules for land application of wastes under P-ba sed management. Even if the results are not directly used for developing new regulations, our fi ndings can be used to judge the practicality of current regulations. We hypothesized (Hypothesis # 1) that equilib rium between wastewater P and soil is not reached in RIBs because of the high infiltrati on (wastewater flow) rates. Comparison between results from the kinetics study and from field infiltration rate measur ements demonstrated that, at least for surface horizons, <50% of P adsorption equilibrium is likely reached for the cells. Therefore, we fail to reject the Hypothesis # 1. The lack of adsorption equilibrium suggests that the P adsorption capacity of soils at Conserv II is under-utilized. The Conserv II RIB systems rely on groundwater monitoring to determine when P wastewater application should be terminated. The approach is not the most appropriate because the impact is identified after P has become a potential environmenta l problem. Soil phosphorus storage capacity (SPSC) could be used as a tool to determine (prior to environmental impacts) the additional P that can be safely applied to the cells. Our study sought to validate the SPSC concept under RIB conditions. A PSR threshold of 0.08 was identified for cell 2-3-A, but there was no change point indicate d for other Conserv II RIBs. The SPSC concept is not applicable to the Conserv II RIBs because PSR and SPSC concepts assume equilibrium conditions betw een P-sources and soils. As established by infiltration rate measurements and the kinetics study, the investigated systems are not likely in 69

PAGE 70

equilibrium. Another possible expl anation is that, at least for su rface horizons of cell 6-21-C, the average soil pH becomes alkaline and thus, th e P removal mechanism is not dominated by P association with Al and Fe. Consequently, the SPSC concept is not applicable to Conserv II RIBs, and we reject Hypothesis # 2. Miller (2008) emphasized that although the PSR concept can be applied to a variety of biosolids amende d soils, the PSR concept is not appropriate for alkaline stabilized biosolids because P lability is controlled by P association with Ca instead of Fe and Al. The PSR laboratory study showed that, without the interferen ce of wastewater constitutes and under equilibrium conditions, the change point PSR is approximately 0.10. As the change point at equilibrium was almost the same as the value found by Nair et al. (2004), we fail to reject Hypothesis # 3: when the SPSC concept is applicable for land application of wastewater (for systems in equilibrium), the change point should be equal to 0.10. The SPSC remains as a very useful tool to determine the maximum P that a given soil can receive before posing environmental risks, including the case of using wastewater as the sour ce of P, but only if equilibrium conditions are met. The Hypothesis #4 was that long-term wastewater application to the RIBs increases soil P lability. Wastewater application increased soil WE P values in all cells. The increase is more pronounced in old RIBs due to greater P loading. In addition, the per centage KCl-P for the samples analyzed in the P fractionation study incr eased with increasing wa stewater application. Elliott et al. (2002) found the KCl-P correlated with P leaching from biosolids and chicken manure amended soils. The correlation is expected to be also true for municipal wastewater amended soils, as in the case of Conserv II soils. Therefore, we fail to reject the hypothesis. 70

PAGE 71

Phosphorus wastewater applicati on to the RIBs will have to ceas e at some point to protect groundwater quality. Results from monitoring wells adjacent to the cell 7-6-B (old and good) suggest that the soil is failing to remove P fr om the wastewater. The determinations of the maximum additional P that can be safely applie d to each cell and/or changes in management practices are urgent. Hypothesis # 5 was that long-term wastewater applications do not alter P species inside the RIBs (treatment soils) as compared to soils in the surrounding area (contro l soils). Results from the P sequential extraction study showed that in th e surface soils of one cell, the P species were different from controls. However, we do not ha ve enough evidence to prove that changes were caused by long term wastewater applications. Ther efore, at this point, we fail to reject the hypothesis. Recommendations A common practice at Conserv II is to prioritize wastewater application to good RIBs. For instance, even though cells 7-6-B and 6-21-C are the same age, 9.3 x 109 L of wastewater has been applied to cell 7-6-B (goodRIB), whereas 84 % less wastewater (1.5 x 109 L) has been applied to the not so good cell 6-21-C. Not so good RIBs can be more efficient at P removal from wastewater (due to more contact time between wastewater -P and soil adsorption sites) than good RIBs. The slower infiltration rates can be the result of accumulation of a thin layer of clay at depths of 270 and 300cm as observed in not so good cells. Clay particles, cont aining Fe and Al oxides, can increase the P removal efficiency from the wast ewater. We recommend that wastewater should be applied in greater volumes to not so good RIBs instead of good RIBs. Another advantage of not so good RIBs is a better (more uniform) wastewater distribution than in good RIBs. Comparison be tween PSR values from samples within three 71

PAGE 72

meters from the outflow and PSR values for sample s near the edges of the cells confirmed that P is better distributed throughout not so good cells than in good cells. Th erefore, the volume of soil that can potentially adsorb P from the wastew ater is much greater in not so good than in good RIBs. The use of not so good RIBs has a disadvantage when compared to good RIBs: larger areas would be required for a system based on not so good RIBs only, because the volume that infiltrates pe r unit area and time is smaller in n ot so good than in good RIBs. All the investigated cells had only two outfl ows operating in an area greater than 0.4 ha. Two outflows, mainly in the case of good RIBs, do not appear to be suffic ient to distribute the wastewater in a fashion that takes full advantag e of the soil adsorption capacity. Some cell (soil) regions may only rarely come in contact with th e wastewater, and thus, the overall cell P removal capacity is reduced. State regulations affirm that provisions shall be made in the design to ensure reasonably uniform distribution of reclai med water across the entire bottom area of rapid infiltration basins (Florida Department of Environmental Protection, 1999).We suggest that RIBs to be constructed (or modified) to have as many outflows as practical. The disadvantage is that the insertion of more outflows will increa se the cells construction costs. The use of sprinklers (USEPA, 2006) might be a more cost-eff ective solution to better distribute wastewater application to the cells. Some studies have revealed that concomitant application of P-sources with P-sorbing amendments, like water treatment residuals (WTR s), can reduce soil P solubility by increasing the long-term soil P retenti on capacity (Agyin-Birikorang et al., 2007; OConnor et al., 2002, Silveira et al., 2006). When incorporated into th e soil, WTRs are even effective for reducing the solubility of legacy P (Silveira et al., 2006). In the case of RIBs, incorporation can increase the contact time between wastewater and WTR, incr easing the P adsorption. Therefore, WTR can be 72

PAGE 73

applied prior wastewater disc harge to increase the life e xpectancy of RIBs. Costs of transportation and incorporation and availability of the WTR can limit the use of WTR in RIB systems. Further Investigations Several issues about P fate a nd transport in RIBs remain unresolved. The studied system is not in equilibrium (in terms of P adsorption) and thus, the SPSC concept is not applicable. Further studies should consider th e kinetics of the system in the determination of the maximum P load to be safely applied to RIBs. Further investigations shoul d include the development of a modified SPSC equation, applicable to high pH soils. The modified e quation should account for P removal mechanisms other than adsorption to Fe and Al oxides (e.g. precipitation of Ca-P and Mg-P) when applied to high pH soils. For the case of Conserv II, the modified equation would have to account the concentration of cations like Ca and Mg not only in the soil, but al so in the wastewater. In which case, chemical speciation modeling could be useful. A major limitation of the study refers to the fact that our sampling scheme was limited to the first three meters of the soil profile. For most RIBs, the water table is much deeper and deeper samples are needed to de termine the extent of soil P satu ration. Other limitations include the use of chemical equilibrium model for predicting P speciation in the soil. The model is a first approximation and thus, should be used carefully; predictions of presence/absence of P species by the model might not be observed in the field, where the system is much more complex than the model assumes. In addition, we used th e model without consid ering the soil solution constituents (we just considered the wastewater constituents). Lastly, we were not able to make statistical comparisons between cells because just one experimental unit per RIB class was studied. Future investigati ons should consider studying at 73

PAGE 74

least two cells per RIB type. For the fractiona tion study, only two deep samples per cell were analyzed. A more detailed characterization of th e P speciation in the RIBs, considering different depths and distances from the outflows, is necessary. 74

PAGE 75

75 APPENDIX A SAMPLING SCHEME FOR TREATMENT SAMPLES

PAGE 76

76 MIDU BL BL1 BL2 MIDL MID BR BR1 BR2 MIDR MIDD 3 m 17 m 27 m 37 m 37 m 20 m Figure A-1. Sampling scheme for cell 2-3-A.

PAGE 77

77 BL BL1 BL2 MIDL 17 m 10 m 3 m MIDU MID BR BR1 MIDR 19 m BR2 MIDD 30 m 13 m 23 m Figure A-2. Sampling scheme for cell 4-3-B.

PAGE 78

78 BL BL1 BL2 MIDL 21 m 10 m 3 m MIDU MID BR BR1 MIDR 23 m BR2 MIDD 23 m 23 m 23 m Figure A-3. Sampling scheme for cell 6-21-C.

PAGE 79

MID BR BR1 BR2 MIDR MIDD 40 m 25 m 20 m BL BL1 BL2 MIDL 20 m 10 m 3 m MIDU Figure A-4. Sampling scheme for cell 7-6-B. 79

PAGE 80

APPENDIX B VARIABILITY OF KSAT AT DIFFERENT DEPTHS 0 50 100 150 200 0 50100150200 Depth (cm)Ksat (cm h-1) Figure B-1. Distribution of Ksat values as a function of depth for Candler soil samples in the state of Florida (Natural Resour ces Conservation Service, 2008). 0 50 100 150 200 0 50100150200 Depth (cm)Ksat (cm h-1) Figure B-2. Distribution of Ksat values as a function of depth for Candler soil samples taken close to Orange County, FL (Natural Resources Conservation Service, 2008). 80

PAGE 81

0 50 100 150 200 01 Sampling siteInfiltration rate (cm h -1)2 Cell 2-3-A Cell 4-3-B Cell 6-21-C Cell 7-6-B Figure B-3. Distribution of infiltration rate valu es for Candler soil samples taken in the Orange Country (FL) area. Sampling sites number 1 and 2 refer to samples taken within 3 meters from left-hand side and ri ght-hand side outflows, respectively. 81

PAGE 82

APPENDIX C CUMULATIVE WASTEWATER VOLUME APPLIED TO THE CELLS OVER TIME 0 250,000,000 500,000,000 750,000,000 1,000,000,000 2/15/20059/3/20053/22/200610/8/20064/26/200711/12/2007 dateVolume of Wastewater (L) Figure C-1. Cumulative wastewat er applied to cell 2-3-A during the cells life span. 0 500,000,000 1,000,000,000 1,500,000,000 2,000,000,000 2/5/20023/12/20034/15/20045/20/20056/24/20067/29/2007 dateVolume of Wastewater (L) Figure C-2. Cumulative wastewat er applied to cell 4-3-B during the cells life span. 82

PAGE 83

0 500,000,000 1,000,000,000 1,500,000,000 2,000,000,000 12/27/19862/4/19913/15/19954/23/19996/1/20037/10/2007 dateVolume of Wastewater (L) Figure C-3. Cumulative wastewat er applied to cell 6-21-C dur ing the cells life span. 0 2,500,000,000 5,000,000,000 7,500,000,000 10,000,000,000 12/27/19862/4/19913/15/19954/23/19996/1/20037/10/2007 dateVolume of Wastewater (L) Figure C-4. Cumulative wastewat er applied to cell 7-6-B during the cells life span. 83

PAGE 84

APPENDIX D DISTRIBUTION OF TOTAL PHOSPHORUS CONCENTRATIONS OVER TIME FOR TREATMENT WELLS 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.83/20/2003 6/28/2003 10/6/2003 1/14/2004 4/23/2004 8/1/2004 11/9/2004 2/17/2005 5/28/2005 9/5/2005 12/14/2005 3/24/2006 7/2/2006 10/10/2006 1/18/2007 4/28/2007 8/6/2007 11/14/2007 2/22/2008dateTotal phosphorus (mg L -1) P concentration = 1.67 mg L-1 Figure D-1. Well 2-01 First treatment monito ring well data for RIB 2-3 (young and good). 1.67 mg L-1 represents the average wastewater P concentration. 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.83/20/2003 6/28/2003 10/6/2003 1/14/2004 4/23/2004 8/1/2004 11/9/2004 2/17/2005 5/28/2005 9/5/2005 12/14/2005 3/24/2006 7/2/2006 10/10/2006 1/18/2007 4/28/2007 8/6/2007 11/14/2007 2/22/2008dateTotal phosphorus (mg L -1) P concentration = 1.67 mg L-1 Figure D-2. Well 2-02 Second treatment monitoring well data for RIB 2-3 (young and good). 1.67 mg L-1 represents the average wast ewater P concentration. 84

PAGE 85

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.83/20/2003 6/28/2003 10/6/2003 1/14/2004 4/23/2004 8/1/2004 11/9/2004 2/17/2005 5/28/2005 9/5/2005 12/14/2005 3/24/2006 7/2/2006 10/10/2006 1/18/2007 4/28/2007 8/6/2007 11/14/2007 2/22/2008dateTotal phosphorus (mg L -1) P concentration = 1.67 mg L-1 Figure D-3. Well 2-03 Third treatment monito ring well data for RIB 2-3 (young and good). 1.67 mg L-1 represents the average wastewater P concentration. 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.812/6/1999 6/23/2000 1/9/2001 7/28/2001 2/13/2002 9/1/2002 3/20/2003 10/6/2003 4/23/2004 11/9/2004 5/28/2005 12/14/2005 7/2/2006 1/18/2007 8/6/2007 2/22/2008dateTotal phosphorus (mg L -1) P concentration = 1.67 mg L-1 Figure D-4. Well 4-01 First treatment monitoring well data for RIB 4-3 (middle aged and not so good). 1.67 mg L-1 represents the average wastewater P concentration. 85

PAGE 86

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.812/6/1999 6/23/2000 1/9/2001 7/28/2001 2/13/2002 9/1/2002 3/20/2003 10/6/2003 4/23/2004 11/9/2004 5/28/2005 12/14/2005 7/2/2006 1/18/2007 8/6/2007 2/22/2008dateTotal phosphorus (mg L -1) P concentration = 1.67 mg L-1 Figure D-5. Well 4-02 Second tr eatment monitoring well data for RIB 4-3 (middle aged and not so good). 1.67 mg L-1 represents the average wastewater P concentration. 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.83/29/1986 8/11/1987 12/23/1988 5/7/1990 9/19/1991 1/31/1993 6/15/1994 10/28/1995 3/11/1997 7/24/1998 12/6/1999 4/19/2001 9/1/2002 1/14/2004 5/28/2005 10/10/2006 2/22/2008dateTotal phosphorus (mg L -1) P concentration = 1.67 mg L-1 Figure D-6. Well 6-23 Treatment monitoring well data for RIB 6-21 (old and not so good). 1.67 mg L-1 represents the average wastewater P concentration. 86

PAGE 87

0 0.5 1 1.5 2 2.5 3 3.511/14/1984 3/29/1986 8/11/1987 12/23/1988 5/7/1990 9/19/1991 1/31/1993 6/15/1994 10/28/1995 3/11/1997 7/24/1998 12/6/1999 4/19/2001 9/1/2002 1/14/2004 5/28/2005 10/10/2006dateTotal phosphorus (mg L -1) P concentration = 1.67 m g L -1 Figure D-7. Well 7-08 Treatment monitoring well data for RIB 7-6 (old and good).1.67 mg L-1 represents the average wastewater P concentration. 87

PAGE 88

APPENDIX E COMPARISON BETWEEN PSR VALUES FROM SAMPLES TAKEN NEAR AND FAR FROM THE OUFLOWS Figure E-1. Average PSR distribu tion through the soil profile for so il samples taken near (within 3 meters) (n=2) and distant (in the cell ed ges) (n=2) from the cell 2-3-A (young and goodRIB) outflows. Soil surface of the cell 2-3-A represents zero depth. Error bars represent standard deviations for each corresponding depth. 88

PAGE 89

Figure E-2. Average PSR distribu tion through the soil profile for so il samples taken near (within 3 meters) (n=2) and distant (in the cell edge s) (n=2) from the cell 4-3-B (middle aged and not so good RIB) outflows. Soil surfac e of the cell 4-3-B represents zero depth. Error bars represent standard devi ations for each corresponding depth. Figure E-3. Average PSR distribu tion through the soil profile for so il samples taken near (within 3 meters) (n=2) and distant (in the cell ed ges) (n=2) from the cell 6-21-C (old and not so good RIB) outflows. Soil surface of the cell 6-21-C represents zero depth. Error bars represent standard devi ations for each corresponding depth. 89

PAGE 90

Figure E-4. Average PSR distribu tion through the soil profile for so il samples taken near (within 3 meters) (n=2) and distant (in the cell ed ges) (n=2) from the cell 7-6-B (old and goodRIB) outflows. Soil surface of the cell 7-6-B represents zero depth. Error bars represent standard deviations for each corresponding depth. 90

PAGE 91

APPENDIX F PLOTS OF PSR AS A FUNCTION OF WEP Figure F-1. PSR against WEP for treatment a nd control samples from cell 2-3-A (young and goodRIB). Figure F-2. PSR against WEP for treatment and control samples from cell 7-6-B (old and goodRIB). 91

PAGE 92

Figure F-3. PSR against WEP for treatment and c ontrol samples from cell 6-21-C (old and not so good RIB). Figure F-4. PSR against WEP for treatment and control samples from cell 4-3-B (middle aged and goodRIB). 92

PAGE 93

Figure F-5. PSR against WEP for treatment sample s taken within 15 m from effluent outflows. Figure F-6. PSR against WEP for treatment sample s taken more than 15 m away from effluent outflows (We omitted three data points (0.872, 5.52), (0.674, 7.43), (0.792, 9.31) that had PSR greater than 0.6 to focus result data with PSR small than 0.3). 93

PAGE 94

APPENDIX G ADDITIONAL IMPORTANT FIGURES AND TABLES 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00 11.00 12.00 1/1/1987 6/23/199212/14/1997 6/6/2003 11/26/2008 dateP concentration (mg L -1 ) Average P concentration I = 4.61 mg L-1 Avera g e P concentrationII= 1.75 m g L-1 Figure G-1. Phosphorus concentration in the wa stewater since 1987. Average P concentration I and II refer to periods from 1987 to 1992 and from 1993 to 2008 respectively. 94

PAGE 95

R2 = 0.3143 6.65 6.75 6.85 6.95 7.05 7.15 7.25 7.35 12/3/1986 5/25/1992 11/15/1997 5/8/2003 10/28/2008 DatepH Figure G-2. Trend of wastewater pH changes since 1987. Table G-1. Results from the chemical mode ling exercise. pH of 6.09, 6.16, 7.01, and 7.07 refer to the mean soil pH from cells 2-3-A, 4-3-B, 621-C, and 7-6-B respectively. pH of 6.89 refers to the average wastewater pH from 2003 to present. pH 6.09 6.16 6.89 7.01 7.07 CaH2PO4 + 1.4% 1.4% 0.0% 0.0% 0.0% CaHPO4 (aq) 2.0% 2.4% 0.0% 0.0% 0.0% MgHPO4 (aq) 0.0% 1.0% 0.0% 0.0% 0.0% H2PO4 86% 84% 5.3% 2.6% 1.9% HPO4 -2 8.8% 10% 3.4% 2.3% 1.9% Hydroxylapatite 0.0% 0.0% 90% 94% 96% 95

PAGE 96

APPENDIX H INFILTRATION RATES AS FUNCTION OF TIME 0 15 30 45 60 75 90 105 120 0.000.200.400.600.801.00 Midpoint Time (h)Infiltration rate (cm h -1) BR BL Figure H-1. Infiltration rate as a function of cumulative time from Cell 2-3-A (young and goodRIB). Midpoint time refers to the time in between the start and the end of each individual measurement. BR refers to samples taken within 10 m from the right hand side outflow and BL refers to samples ta ken within 10 m from the left hand side outflow. 0 15 30 45 60 75 90 105 120 0.000.200.400.600.801.00 Midpoint Time (h)Infiltration rate (cm h -1) BR BL Figure H-2. Infiltration rate as a function of cumulative time from Cell 4-3-B (middle aged and not so good RIB). Midpoint time refers to the time in between the start and the end of each individual measurement.BR refers to samples taken within 10 m from the right hand side outflow and BL refers to sa mples taken within 10 m from the left hand side outflow. 96

PAGE 97

0 15 30 45 60 75 90 105 120 0.000.200.400.600.801.00 Midpoint Time (h)Infiltration rate (cm h -1) BR BL Figure H-3. Infiltration rate as a function of cumulative time from Cell 6-21-C (old and not so good RIB). Midpoint time refers to the time in between the start and the end of each individual measurement. BR refers to samples taken within 10 m from the right hand side outflow and BL refers to samples ta ken within 10 m from the left hand side outflow. 0 15 30 45 60 75 90 105 120 0.000.200.400.600.801.00 Midpoint Time (h)Infiltration rate (cm h -1) BR BL Figure H-4. Infiltration rate as a function of cumulative time from Cell 7-6-B (old and goodRIB). Midpoint time refers to the time in between the start and the end of each individual measurement. BR refers to samples taken within 10 m from the right hand side outflow and BL refers to samples ta ken within 10 m from the left hand side outflow. 97

PAGE 98

0 20 40 60 80 100 120 140 160 180 0.000.100.200.300.400.500.60 Midpoint Time (h)Infiltration rate (cm h -1) Figure H-5. Infiltration rate as a function of cumulative time from a control sample taken from Cell 2-3-A (young and goodRIB). Midpoint time refers to the time in between the start and the end of each individual measurement. 0 20 40 60 80 100 120 140 160 180 0.000.100.200.300.400.500.60 Midpoint Time (h)Infiltration rate (cm h -1) Figure H-6. Infiltration rate as a function of cumulative time from a control sample taken from Cell 6-21-C (old and not so good RIB). Mi dpoint time refers to the time in between the start and the end of each individual measurement. 98

PAGE 99

0 20 40 60 80 0.000.100.200.300.400.500.60 Midpoint Time (h)Infiltration rate (cm h -1) BL BR Figure H-7. Infiltration rate as a function of cumulative time from Cell 7-6-B (old and goodRIB). Midpoint time refers to the time in between the start and the end of each individual measurement.BR refers to sample s taken within 10 m from the right hand side outflow and BL refers to samples ta ken within 10 m from the left hand side outflow. 0 20 40 60 80 0.000.100.200.300.400.500.60 Midpoint Time (h)Infiltration rate (cm h -1) CTRL2 CTRL3 Figure H-8. Infiltration rate as a function of cumulative time from Cell 7-6-B (old and goodRIB). Midpoint time refers to the time in between the start and the end of each individual measurement.CTRL2 refers to co ntrol samples taken within 30m from the edge of the RIB and CTRL3 refers to contro l samples taken within 45m from edge of the RIB. 99

PAGE 100

LIST OF REFERENCES Agyin-Birikorang, S., G.A. O'Connor, L.W. J acobs, K.C. Makris, and S.R. Brinton. 2007. Longterm phosphorus immobilization by a drinking water treatment residual. J. Environ. Qual.36: 316-323. American Society for Testing and Materials. 2003. ASTM D3385-03. ASTM International, West Conshohocken, PA. Andersen, J.M. 1976. An ignition method for de termination of total phosphorus in lake sediments. Water Research 10:329-331. Asano, T. and A. D. Levine. 1996. Wastewater reclamation, recycling and reuse: Past, present, and future. Wat. Sci. Tech. 33:1-14. Bolster, C. H., and G.M. Hornberger. 2007. On th e Use of Linearized Langmuir Equations. Soil Sci. Soc. Am. J. 71:1796. Brady, N.C., and R.R. Weil. 2002. The nature and properties of soils. 13th ed. Prentice Hall, New Jersey. Breeuwsma, A., J.G.A Rijerink, and O.F. Schou mans. 1995. Impact of manure on accumulation and leaching of phosphate in areas of intensive livestock farming. p.239. In K. Steele (ed.) Animal waste and the landwater in terface. Lewis Publ.CRC Press, New York. Chang, A.C., A.L. Page, J.E. Warneke, and E. Grgurevic. 1984. Sequential extraction of soil heavy metals following a sludge application. J. Environ. Qual. 13:33-38. Chrysostome, M., V.D. Nair, W.G. Harris, a nd R.D. Rhue. 2007. Laboratory validation of soil phosphorus storage capacity predictions for use in risk assessment. Soil Sci. Soc. Am. J. 71:1564-1569. Cross, A.F., and W.H. Schlesinger. 1995. A lite rature review and eval uation of the Hedley fractionation: applications to the biogeoch emical cycle of soil phosphorus in natural ecosystems, Geoderma 64:197. Elliott, H.A., G.A. OConnor, and S. Brinton. 2002. Phosphorus leaching from biosolidsamended sandy soils. J. Environ. Qual. 31:681-689. Environmental Research Software. 2002. MINEQL+ Version 4.5. Available at http://www.mineql.com/ (verified Feb. 2009). Fares, A. and A. K. Alva. 1999. Estimation of citrus evapotranspiration by soil water mass balance. Soil Science 164:302-310. Florida Department of Environmental Protection. 1999. Reuse of Reclaimed Water and Land Application. Chapter 62-610, Florida Ad ministrative Code. Tallahassee, FL. 100

PAGE 101

Food and Agriculture Organization of the Unite d Nations. 1985. Water quality for agriculture. In R.S. Ayers and D.W. Westcot (ed.). FAO Irriga tion and drainage paper 29, Rev. 1. FAO, Rome. Google Earth. 2009. Google earth 5.0. Available at http://earth.google.com (verified Mar. 2009). Heckrath, G., P. C. Brookes, P. R. Poult on and K. W. T. Goulding. 1995. Phosphorus leaching from soils containing different phosphorus conc entrations in the br oadbalk experiment. J Environ. Qual. 24:904-910. Hooda, P. S., A. R. Rendell, A. C. Edwards, P. J. A. Withers, M. N. Aitk en and V. W. Truesdale. 2000. Relating soil phosphorus indices to potential phosphorus release to water. J Environ. Qual. 29:1166-1171. Hu, C., T. C. Zhang, Y. H. Huang, M. F. Dahab and R. Surampalli. 2005. Effects of long-term wastewater application on chemical propertie s and phosphorus adsorption capacity in soils of a wastewater land treatment syst em. Environ. Sci. Technol. 39:7240-7245. Idaho Department of Environmental Quality. 2007. Guidance for reclamation and reuse of municipal and industrial wastewater. Available at http://www.deq.state.id.us/water/perm its_forms/permitting/guidance_reuse_0907.pdf (verified Apr. 2008). Josan, M. S., V. D. Nair, W. G. Harris and D. Herrera. 2005. Associated release of magnesium and phosphorus from active and abandoned dairy soils. J Environ. Qual. 34:184-191. Kennedy, V.H., A.P. Rowland, J. Parrington.1994. Qu ality assurance for soil nutrient analysis. Soil Sci. Plant Anal. 25:1605-1627. Lu, P. and G. A. O'Connor. 1999. Factors affecting phosphorus reactions in soils: Potential sewage sludge effects. Soil Crop Sci. Soc. Fla. Proc. 58:66-71. Maguire, R. O., R. H. Foy, J. S. Bailey a nd J. T. Sims. 2001. Estimation of the phosphorus sorption capacity of ac idic soils in Ireland. European Journal of Soil Science 52:479-487. Maguire, R. O. and J. T. Sims. 2002. Soil tes ting to predict phosphor us leaching. J Environ. Qual. 31:1601-1609. Mann, C. J. and R. G. Wetzel. 1995. Dissolved orga nic carbon and its utilization in a riverine wetland ecosystem. Biogeochemistry 31:99-120. McDowell, R. W. and A. N. Sharpley. 2001. A pproximating phosphorus release from soils to surface runoff and subsurface drainage. J Environ. Qual. 30:508-520. McFarland, M.J., M.A. Sanderson, A.M.S. McFa rland. 2007. Wastewater and reclaimed water irrigation. p. 754-789. In G.J. Hoffman, R.G. Evans, M.E. Jensen, D.L. Martin, R.L. Elliott (eds). Design and operation of farm irrigati on systems. American Society of Agricultural and Biological Engineers. St. Joseph, Mich. 101

PAGE 102

Metcalf & Eddy. 2007. Water reuse. Issues, t echnologies, and applications. McGraw-Hill Publisher, New York. Miller. M.L. 2008. Characterizing the long-term lability of biosolids-phosphorus. MS Thesis. p.85. University of Florida, Gainesville, Florida. Murphy, J., and J.P. Riley. 1962. A modified sing le solution method for the determination of phosphate in natural water. Anal. Chim. Acta. 27:31-36. Nair, V.D., and W.G. Harris. 2004. A capacity factor as an alternative to soil test phosphorus in phosphorus risk assessment. New Z ealand J. Agri. Res. 47:491-497. Nair, V. D., K. M. Portier, D. A. Graetz and M. L. Walker. 2004. An Environmental Threshold for Degree of Phosphorus Saturation in Sa ndy Soils. J Environ. Qual. 33: 107-113. Natural Resources Conservation Servi ce. 2008. Candler Series. Available at http://websoilsurvey.nrcs.usda.gov/app/WebSoilSurvey.aspx (verified April, 2008) OConnor, G. A., D. Sarkar, D. A. Graetz, and H.A. Elliott. 2002. Characterizing forms, solubilities, bioavailabilitie s, and mineralization rates of phosphorus in biosolids, commercial fertilizers, and manures (Phase 1). Water Environment Research Foundation. OConnor, G. A., H. A. Elliott and R. K. Bastian. 2008. Degraded water reuse: An overview. J Environ. Qual. 37:S-157-168. Otani, T., and N. Ae. 1997. The status of inor ganic and organic phosphorus in some soils in relation to plant availability. So il Sci.Plant Nutr. (Tokyo) 43:419. Oladeji, O.O. 2006. Management of phosphorus s ources and water treatm ent residuals (WTR) for environmental and agronomic benefits. PhD Thesis. p.271. University of Florida, Gainesville, Florida. Orange County Growth Management Department. 2006. Orange county comprehensive plan evaluation and appraisal report. Available at www.ocfrd.org/NR/rdonlyres/eedmkpw b5mearhhgnrx776yrx7gkadcxernf2xouosjrfgv352p 7y4zpz4yp3tz3y3hpbqakpctfeyd3jw2eg4cls4h/EARFinal.pdf (verified Jan. 2008). Parsons, L.R., K.T. Morgan, T.A. Wheaton, and W.S. Castle. 2001. Wastewater and reclaimed water disposal problem or potential reso urce? Proc. Fla. Stat e Hort. Soc. 114:97-100. Pierzynski, G.M., J.T. Sims, and G.F. Vance. 1994. Soils and environmental quality. CRC Press, Boca Raton, FL. SAS Institute. 2001. SAS online doc. Version 8.1. SAS Inst., Cary, NC. 102

PAGE 103

Schoumans, O.F. 2000. Determining the degree of phosphate saturation in non-calcareous soils. p. 31. In G.M. Pierzynski (ed.) Methods of phosphorus analysis for soils, sediments, residuals, and waters. Southern Corp. Seri es Bull. 396. Coop. Ext. Ser., North Carolina State Univ., Raleigh, NC. Self-Davis, M.L., P.A. Moore Jr., and B.C. Jo ern. 2000. Determination of wateran/or dilute salt-extractable phosphorus. In G.M.Pierzynski (ed). Methods for P analysis. Shober, A. L. and J. T. Sims. 2007. Integrating Phosphorus Source and Soil Properties into Risk Assessments for Phosphorus Loss. Soil Sci. Soc. Am. J. 71: 551-560. Silveira, M. L., M. K. Miyittah and G. A. O'Connor. 2006. Phosphorus release from a manureimpacted spodosol: Effects of a water trea tment residual. J. Environ. Qual. 35:529-541. Sims, J. T., R. O. Maguire, A. B. Leytem, K. L. Gartley and M. C. Pautler. 2002. Evaluation of mehlich 3 as an agri-environmental soil phosphorus test for the mid-Atlantic United States of America. Soil Sci. Soc. Am. J. 66:2016-2032. Sumner, D. M. and L. A. Bradner. 1996. Hydrau lic characteristics and nutrient transport and transformation beneath a rapid infiltration basin, reedy creek improvement district, Orange County, Florida. Water-Resources Investiga tions Report 95-4281. U.S. Geological Survey, Tallahassee, Florida. Thomas, G.W. 1996. Soil pH and soil acidity. In Methods of Soil Analysis. Part 3. Chemical Methods. P. 475. J.M. Bighman, (ed.) Soil Scie nce Society of America, ASA, Madison, Wisconsin. United States Army Corp of Engineers. 1982. E ngineering and design Process design manual for land treatment of muni cipal wastewater. USACE EM 1110-1-501. Department of Army, Washington, D.C. USEPA. 1981. Process design manual Land trea tment of municipal wastewater. U.S. EPA 625/1 (COE EM 1110-501). Center for Environmental Research Information, U.S. EPA, Cincinnati, OH. USEPA. 1986. Quality criteria for water. U.S. EPA440/5-86-001. Office of water regulations and standards, Washington, D.C. USEPA and USDA. 1999. Unified national strate gy for animal feeding operations. U.S. Government Printing Office, Washington, D.C. Available at http://www.epa.gov/npdes/pubs/finafost.pdf (verified Feb. 2008). USEPA. 2004. Guidelines for water reuse. U.S. EPA/625/R-04-108. Technology Transfer and Support Division, U.S. EPA, Cincinnati, OH. USEPA. 2006. Process design manual land treatmen t of municipal wastewat er effluents. U.S. EPA/625/R-06-016. Land Remediation and Po llution Control Divi sion, U.S. EPA, Cincinnati, OH. 103

PAGE 104

104 Water Conserv II. 2009. Plant 1. Available at http://www.waterconservii.com/new-dc.JPG (verified Mar. 2009). Woodard, K. R., Sollenberger L. E ., Sweat L. A., Graetz D. A., Nair V. D., Rymph S. J., Walker L. and Joo Y. 2007. Phosphorus and other soil co mponents in a dairy effluent sprayfield within the central Florida ridge. J. Environ. Qual. 36:1042-1049. World Resource Institute. 2000. World map. Available at http://earthtrends.wri.org /images/maps/2-4_m_WaterSupp ly2025_lg.gif (verified Jan. 2009). World Resource Institute, Washington, DC. Zhang, M. K., Z. L. He, D. V. Calvert, P. J. Stoffella, Y. C. Li, and E. M. Lamb. 2002. Release potential of phosphorus in Florida sandy soils in relation to phosphorus fractions and adsorption capacity. J. Envir on. Sci. and Health, Part A. 37:793-809. Zhang, T. C., M.F. Dahab, G.S. Nunes, C. H u, and R. Surampalli. 2007. Phosphorus fate and transport in soil columns loaded interm ittently with influent of high phosphorus concentrations. Water Environ. Res. 79:2343-2351.

PAGE 105

BIOGRAPHICAL SKETCH Daniel R. Moura was born in Petropolis, Brazil, in 1984. Son of Magno Daniel de Mello Moura and Maria de Lourdes Rodr igues Moura and brother of Tia go Rodrigues Moura, he lived in his hometown until 2002, when he moved to another state to start his undergraduate studies at the University of Vicosa. Daniel finished his B.S. in environmental engineering in August 2007 and one week after his graduation, he moved to Ga inesville, Florida, join ing the University of Floridas Soil and Water Science Department. He conducted his MS studies under the supervision of Drs. Maria L. S ilveira and George A. OConnor, and he graduated in the spring 2009. 105