Citation
The Effect of Alum on Phosphorus Sequestration, Macrophytes, Mineralogy, and Microbial Community in a Municipal Wastewater Treatment Wetland

Material Information

Title:
The Effect of Alum on Phosphorus Sequestration, Macrophytes, Mineralogy, and Microbial Community in a Municipal Wastewater Treatment Wetland
Creator:
BROWN, LYNETTE MALECKI
Copyright Date:
2008

Subjects

Subjects / Keywords:
Aluminum ( jstor )
Biomass ( jstor )
Cells ( jstor )
Constructed wetlands ( jstor )
Dosage ( jstor )
Nutrients ( jstor )
Phosphorus ( jstor )
Soil science ( jstor )
Soils ( jstor )
Wetlands ( jstor )
City of Orlando ( local )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Lynette Malecki Brown. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
5/31/2011
Resource Identifier:
660020693 ( OCLC )

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

EFFECT OF ALUM ON PHOSPHORUS SEQUESTRATION, MACROPHYTES, MICROBIAL COMMUNITY, AND MI NERALOGY IN A MUNICIPAL WASTEWATER TREATMENT WETLAND By LYNETTE MALECKI BROWN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2007

PAGE 2

Lynette Malecki Brown

PAGE 3

To my parents who have given me so much. You have helped me become the woman I am today.

PAGE 4

iv ACKNOWLEDGMENTS I would like to acknowledge my advi sor and co-chair, Dr. John White, for all of his guidance, patience, and understa nding during the last four years and thank him for all of his traveling and hard work to help me not only in the field, but also in life. I would also like to thank my co-chair Dr. K. Rame sh Reddy, and committee members Dr. Willie Harris, Dr. Jim Sickman, and Dr. Mark Brenner for their insightful input and support. Appreciation is expressed to the City of Orlando for funding this research, and in particular Mark Sees, the Orlando Easterly Wetland manager, for all of his technical assistance as well as field help over the y ears. Without his creativity and handyman expertise several of my studies would have been in jeopardy. I thank Sue (Simon) Lindstrom, Isabela Clarett Torres, and Cece (Concha) Kennedy for not only being my Newell Hall offi ce mates but close friends as well. I am extremely grateful for their patient listening, honest advice, and true friendship over the years. I also thank my past roommates C ourtney Peppler and Mega n Gualatunia for their continued friendship regardless of time or di stance. I must also remember my departed friend Ben Skulnick who inspired me to not only embrace soil science, but life itself, and will always remain in my heart. I would also like to thank a ll of the students who helped me in the field including Sue Lindstrom and Chakesha Martin who put in long hours helping me establish my mesocosm study, and Erin (Bostic) Corstanj e who helped in the initial and final mesocosm samplings. Appreciation is e xpressed to Yu Wang and Gavin Wilson who

PAGE 5

v helped me on so many occasions, as well as Maverick Leblanc, Brett Marks, Jeremy Conkle, and Dave Gambrell of Louisiana State Un iversity for all of their wonderful assistance with laboratory work and analysis . I would also like to acknowledge my husband Eric A. Brown who helped on numer ous occasions in the field while I was establishing and sampling my mesocosm study. He has always been willing to listen, and has gotten me through many tough times over the pa st four years. I thank him for always bringing a smile to my face in any situation. Finally I thank my parents, Daniel and Pa uline Malecki, for all of their help and company during my weekly sampling trips, as well as their love and support in my graduate school endeavors. Not many parents would face the harsh elements of a hot summer day in a Florida wetland just to spend time with their da ughter and learn about wetland biogeochemistry, but both did, and for that I am ever grateful. I want to also thank my sister and brother-in-law Valerie and John Breun for their understanding while I have been a poor, overworked, stressed gradua te student, unable to make many trips to visit. I have truly apprecia ted their willingness to drive down to Florida on numerous occasions, and hope to see them, and my nephew Ayden much more often in future years.

PAGE 6

vi TABLE OF CONTENTS page ACKNOWLEDGMENTS................................................................................................. iv LIST OF TABLES............................................................................................................... x LIST OF FIGURES...........................................................................................................xv ABSTRACT................................................................................................................... xviii CHAPTER 1 LITERATURE REVIEW.............................................................................................1 Constructed Treatment Wetlands................................................................................. 1 Phosphorus in Wetlands............................................................................................... 2 Wetland Macrophytes................................................................................................... 3 The Problem..................................................................................................................4 Aluminum Sulfate (Alum)............................................................................................ 5 Concerns in Alum Application..................................................................................... 8 Effect of Alum on Aquatic Macrophytes....................................................................10 Effect of Alum on Mineralogy................................................................................... 12 Effect of Alum on Microbes and P Cycling............................................................... 13 Alum Forms and Alternatives..................................................................................... 14 Alum Dose Determination and Effectiveness............................................................. 16 Hypotheses and Objectives......................................................................................... 18 Site Description..........................................................................................................18 2 RESTORATION OF PHOSPHORUS SEQUESTRATION IN TREATMENT WETLAND SOIL USING ALUMIN UM-CONTAINING AMENDMENTS.......... 23 Introduction................................................................................................................. 23 Materials and Methods...............................................................................................28 Site Description...................................................................................................28 Field Sampling, Laboratory Set-Up and Analysis............................................... 30 Statistical Analysis..............................................................................................33 Results and Discussion...............................................................................................34 Soil Physicochemical Characteristics.................................................................. 34 Soil Microbial Characteristics............................................................................. 37 Soil Phosphorus Forms........................................................................................ 39

PAGE 7

vii Water Column Results......................................................................................... 44 Soluble Reactive Phosphorus – Incubation Study........................................ 44 Soluble Reactive Phosphorus – Re-Spiking Study......................................47 Water Column pH........................................................................................49 Water Column Soluble Aluminum............................................................... 52 Conclusions.................................................................................................................54 3 INFLUENCE OF ALUM ON WATER QUALITY AND AQUATIC MACROPHYTES: A ME SOCOSM STUDY............................................................ 56 Introduction................................................................................................................. 56 Materials and Methods...............................................................................................61 Site Description...................................................................................................61 Mesocosm Establishment.................................................................................... 64 Experiment Initiation........................................................................................... 65 Statistical Analysis...................................................................................................... 67 Results and Discussion...............................................................................................68 Water Quality Characteristics.............................................................................. 68 pH ................................................................................................................. 68 Dissolved Oxygen........................................................................................ 69 Conductivity.................................................................................................70 Dissolved Organic Carbon...........................................................................70 Dissolved Inorganic Nitrogen...................................................................... 72 Total Dissolved Nitrogen.............................................................................73 Total Kjeldahl Nitrogen............................................................................... 74 Soluble Reactive Phosphorus....................................................................... 75 Total Dissolved Phosphorus.........................................................................76 Total Phosphorus..........................................................................................77 Particulate Phosphorus................................................................................. 78 Macrophyte Characteristics................................................................................. 79 Emergent Growth Study............................................................................... 79 Destructive Biomass and Growth................................................................. 82 Total Carbon, Nitrogen, Phosphorus Concentrations and Storage...............83 Other Macronutrient and Micr onutrient Concentrations.............................. 87 Conclusion..................................................................................................................89 4 INFLUENCE OF ALUM ON TR EATMENT WETLAND SOIL PHYSICOCHEMICAL AND MICR OBIAL CHARACTERIS TICS........................ 91 Introduction................................................................................................................. 91 Materials and Methods...............................................................................................93 Site Description...................................................................................................93 Mesocosm Establishment.................................................................................... 95 Experiment Initiation........................................................................................... 97 Laboratory Analysis............................................................................................ 97 Statistical Analysis...................................................................................................... 98 Results and Discussion...............................................................................................99

PAGE 8

viii Soil Physicochemical Characteristics.................................................................. 99 Soil Microbial Characteristics........................................................................... 104 Soil Phosphorus Forms...................................................................................... 106 Mass Balance of Phosphorus.............................................................................110 Conclusions...............................................................................................................112 5 SPATIAL AND TEMPORAL EFFE CTS OF CONTINUOUS LOW-DOSAGE ALUM ON FIELD SOIL CHARACTERISTICS.................................................... 114 Introduction............................................................................................................... 114 Materials and Methods.............................................................................................116 Site Description.................................................................................................116 Transect Establishment...................................................................................... 118 Experiment Initiation......................................................................................... 119 X-ray Diffraction Analysis................................................................................121 Statistical Analysis....................................................................................................122 Results and Discussion.............................................................................................122 Soil Physicochemical Characteristics................................................................ 122 Bulk Density and LOI................................................................................122 Soil pH........................................................................................................125 Soil Total Phosphorus................................................................................ 126 Total Calcium and Magnesium..................................................................130 Total Iron.................................................................................................... 131 Soil Aluminum Ch aracterization....................................................................... 132 Total Aluminum......................................................................................... 132 Amorphous Al umin um ............................................................................... 133 Crystalline Alumin um ................................................................................ 136 Soil Microbial Characteristics........................................................................... 137 Microbial Biomass Phosphorus..................................................................137 Soil Oxygen Demand.................................................................................141 Potentially Mineralizable Phosphorus........................................................ 142 Soil Phosphorus Forms...................................................................................... 143 Soluble Phosphorus....................................................................................144 Calcium and Magnesium-bound Phosphorus............................................. 147 Aluminum and Iron–bound Phosphorus.................................................... 148 Organically-bound Phosphorus..................................................................150 Residual Phosphorus..................................................................................153 X-Ray Diffraction..............................................................................................154 Conclusions...............................................................................................................154 6 SPATIAL AND TEMPORAL EFFE CTS OF CONTINUOUS LOW-DOSAGE ALUM ON FIELD WATER QUAL ITY AND AQUATIC MACROPHYTE CHARACTERISTICS..............................................................................................156 Introduction............................................................................................................... 156 Materials and Methods.............................................................................................158 Site Description.................................................................................................158

PAGE 9

ix Pre-experiment Water Quality Monitoring....................................................... 160 Transect Establishment...................................................................................... 161 Experiment Initiation......................................................................................... 162 Statistical Analysis....................................................................................................165 Results and Discussion.............................................................................................165 Water Quality Characterization......................................................................... 165 pH ............................................................................................................... 165 Dissolved Oxygen...................................................................................... 166 Dissolved Organic Carbon.........................................................................167 Soluble Reactive Phosphorus..................................................................... 169 Total Dissolved Phosphorus.......................................................................171 Total Phosphorus........................................................................................173 Particulate Phosphorus............................................................................... 174 Typha spp. Characterization.............................................................................. 176 Plant and Leaf Densities............................................................................. 176 Biomass...................................................................................................... 177 Relative Growth Rate.................................................................................180 Total Carbon and Nitrogen Concentrations............................................... 181 Total Phosphorus Concentrations...............................................................186 Metal Concentrations................................................................................. 189 Conclusions...............................................................................................................191 EFFECTS OF ALUM ON A MUNICIPAL WASTEWATER TREATMENT WETLAND: A SUMMARY.................................................................................... 193 APPENDIX A CORE STUDY RAW DATA................................................................................... 197 Lake Alum Appli cation Summary ............................................................................ 197 Water Column Analyses...........................................................................................198 B MESOCOSM STUDY RAW DATA....................................................................... 202 C FIELD STUDY SPATIAL DATA........................................................................... 208 LIST OF REFERENCES.................................................................................................212 BIOGRAPHICAL SKETCH...........................................................................................236

PAGE 10

x LIST OF TABLES Table page 2-1 Mean soil physicochemical characterizat ion data for the 0-5 cm depth of cores taken from the Orlando Easterly W etla nd, (n=3) 1 standard deviation................ 35 2-2 Mean soil physicochemical characterizat ion data for the 5-10 cm depth of cores taken from the Orlando Easterly Wetla nd, (n=3) 1 standard deviation................ 35 2-3 Mean soil microbial characterization data for the 0-5 cm depth of cores taken from the Orlando Easterly W etland, (n=3 ) 1 standard deviation................................... 38 2-4 Mean soil microbial characterization data for the 5-10 cm depth of cores taken from the Orlando Easterly W etland, (n=3 ) 1 standard deviation................................... 38 2-5 Mean organic phosphorus derived from inorganic phosphorus fr actionation data for the 0-5 cm depth interval of cores taken from the Orlando Easterly Wetland, (n=3) 1 standard deviation..............................................................................................40 2-6 Mean inorganic phosphorus fractionation data for the 0-5 cm depth interval of cores taken from the Orlando Easterly Wetland, (n=3) 1 standard deviation....... 40 2-7 Mean 1M HCl-extractable metals for the 0-5 cm depth interval of cores taken from the Orlando Easterly Wetland, (n=3 ) 1 standard deviation................................... 41 2-8 Mean organic phosphorus derived from inorganic phosphorus fr actionation data for the 5-10 cm depth interval of cores ta ken from the Orlando Easterly Wetland, (n=3) 1 standard deviation.................................................................................... 42 2-9 Mean inorganic phosphorus fractionation data for the 5-10 cm depth interval of cores taken from the Orlando Easterly Wetland, (n=3) 1 standard deviation....... 42 2-10 Mean 1M HCl-extractable metals for th e 5-10 cm depth interval of cores taken from the Orlando Easterly Wetland, (n=3) 1 standard deviation.......................... 44 2-11 Mean soluble reactive phosphorus flux from constructed wetland soil under anaerobic water colum n conditions, (n=6) 1 standard deviation..........................47 2-12 Mean water column data collected ov er 14 days in soil cores incubated under anaerobic conditions, (n=18) 1 standard deviation............................................... 51

PAGE 11

xi 3-1 Water quality characteristics of in flow water and treated outflow leaving me socosms...............................................................................................................71 3-2 Emergent macrophyte growth characteri stics within me socosms at the Orlando Easterly Wetland. Values are means 1 standard deviation (n = 9)...................... 80 3-3 Scirpus californicus shoot and Typha spp. leaf characteristics within me socosms at the Orlando Easterly Wetland. Values are means 1 standard deviation (n = 9 plants).......................................................................................................................80 3-4 Biomass and relative growth rates (RGR ) based on total dry weight of plants harvested from the Orlando Easterly W etla nd mesocosms. Values are means 1 standard deviation (n=3)...........................................................................................83 3-5 Carbon, nitrogen, and phosphorus tissue c oncentrations in aquatic macrophytes harvested from the Orlando Easterly Wetla nd mesocosms. Values are means 1 standard deviation (n=3)...........................................................................................84 3-6 Carbon, nitrogen, and phosphorus storage in aquatic macrophytes harvested from the Orlando Easterly Wetland mesocosms. Values are means 1 standard deviation (n=3)......................................................................................................... 84 3-7 Mean tissue concentrations of metals in aquatic macrophytes harvested from the Orlando Easterly We tland mesocosms. Valu es are means 1 standard deviation (n=3)......................................................................................................................... 88 4-1 Mean soil physicochemical characterizat ion data for the 0-5 cm depth of Orlando Easterly We tland mesocosms. Values are means 1 standard deviation (n=3)... 100 4-2 Mean soil physicochemical characterizat ion data for the 5-10 cm depth of Orlando Easterly W etland mesocosms. Values are means 1 standard deviation (n=3)... 101 4-3 Mean soil microbial characterization data for the 0-5 cm depth in Orlando Easterly Wetland me socosms. Values are means 1 standard deviation (n=3)................. 104 4-4 Mean soil microbial characterizati on data for the 5-10 cm depth of Orlando Easterly W etland mesocosms. Values are means 1 standard deviation (n=3)... 105 4-5 Mean inorganic phosphorus fractionation data from the 0-5 cm layer of Orlando Easterly We tland mesocosms. Values are means 1 standard deviation (n=3).... 106 4-6. Mean organic phosphorus derived from inorganic phosphorus fractionation data from the 0-5 cm layer of Orlando Easter ly Wetland mesocosms. Values are means 1 standard deviation (n=3).................................................................................. 107 4-7 Mean inorganic phosphorus fractionation data from the 5-10 cm layer of Orlando Easterly We tland mesocosms. Values are means 1 standard deviation (n=3).... 107

PAGE 12

xii 4-8 Mean organic phosphorus derived from inorganic phosphorus fractionation data from the 5-10 cm layer of Orlando Easterly Wetland mesocosms. Values are means 1 standard deviation (n=3)....................................................................... 107 5-1 Soil physicochemical characterization da ta for the 0-5 cm depth in the Orlando Easterly W etland. Values are means 1 standard deviation (n=3)...................... 123 5-2 Soil physicochemical characterization da ta for the 5-10 cm depth in the Orlando Easterly W etland. Values are means 1 standard deviation (n=3)...................... 124 5-3a Soil nutrient data for the 0-5 cm dept h in alum-treated cell 10 of the Orlando Easterly W etland. Values are means 1 standard deviation (n=3)...................... 127 5-3b Soil nutrient data for the 0-5 cm depth in control cell 9 of the Orlando Easterly Wetland. Values are means 1 standard deviation (n=3). .................................... 127 5-4a Soil nutrient data for the 5-10 cm dept h in alum-treated cell 10 of the Orlando Easterly W etland. Values are means 1 standard deviation (n=3)...................... 128 5-4b Soil nutrient data for the 5-10 cm depth in control cell 9 of the Orlando Easterly Wetland. Values are means 1 standard deviation (n=3). .................................... 128 5-5a Aluminum characterization for the 0-5 cm depth interval in alum -treated cell 10 of the Orlando Easterly Wetland. Values are means 1 standard deviation (n=3).. 134 5-5b Aluminum characterization for the 0-5 cm depth interval in control cell 9 of the Orlando Easterly We tland. Values are means 1 standard deviation (n=3)........ 134 5-6a Aluminum characterizati on for the 5-10 cm depth interv al in the alum -treated cell 10 of the Orlando Easterly Wetland. Valu es are means 1 standard deviation (n=3)....................................................................................................................... 135 5-6b Aluminum characterization for the 5-10 cm depth interval in control cell 9 of the Orlando Easterly W etland. Values are means 1 standard deviation (n=3)........ 135 5-7 Microbial characterization of the 0-5 cm surface soil in cell 9 and 10 of the Orlando Easterly Wetland. Values are means 1 standard deviation (n=3)...................... 139 5-8 Microbial characterization of the 5-10 cm subsurface soil in cell 9 and 10 of the Orlando Easterly Wetland. Values are means 1 standard deviation (n=3)........ 140 5-9 Mean inorganic phosphorus fractionation da ta from the 0-5 cm soil layer in cell 9 and 10 of the Orlando Ea sterly Wetland. Values are means 1 standard deviation (n=3)....................................................................................................................... 145 5-10. Mean inorganic phosphorus fractionation data from the 5-10 cm soil layer in cell 9 and 10 of the Orlando Easterly Wetland. Values are means 1 standard deviation (n=3)....................................................................................................................... 146

PAGE 13

xiii 5-11 Mean organic phosphorus derived from the inorganic phosphorus fractionation data from the 0-5 cm soil layer in cell 9 and 10 of the Orlando Easterly Wetland. Values are means 1 standard deviation (n=3)................................................................. 147 5-12 Mean organic phosphorus derived from the inorganic phosphorus fractionation data from the 5-10 cm soil layer in cell 9 and 10 of the Orlando Easterly Wetland. Values are means 1 standard deviation (n=3)..................................................... 152 6-1 Plant density and biomass allocation of Typha spp. plants harvested from cell 9 and 10 in the Orlando Easterly W etland. Valu es are means 1 standard deviation (n=6)....................................................................................................................... 178 6-2 Biomass of Typha spp. plants on a dry weight basis, harvested from cell 9 and 10 in the Orlando Easterly W etland. Values are means 1 standard deviation (n=6).. 179 6-3 Relative growth rates ba sed on total dry weight of Typha spp. harvested from cell 9 and 10 of the Orlando E asterly Wetland. Values are means 1 standard deviation (n=6)....................................................................................................................... 180 6-4 Carbon, nitrogen, and phosphorus tissue c oncentrations in Typha spp. harvested from cell 9 and 10 of the Orlando Easterly W etland. Values are means 1 standard deviation (n=6)....................................................................................................... 182 6-5 Total phosphorus, carbon, and nitrogen ratios in Typha spp. harvested from cell 9 and 10 of the Orlando E asterly Wetland. Values are means 1 standard deviation (n=6)....................................................................................................................... 184 6-6 Nitrogen and phosphorus resorption effici ency and nutrient use efficiency of Typha spp. harvested from cell 9 and 10 of th e Orlando Easterly W etland. Values are means 1 standard deviation (n=6)....................................................................... 185 6-7 Live tissue Typha spp. me tal concentrations, on a dry weight basis, in cell 9 and 10 of the Orlando Easterly Wetland............................................................................ 190 7-1 Summary of variables imp acted by alum application in a constructed wastewater treatment wetland during a tr i-scale study from 2004-2006.................................. 195 A-1 Literature review of previous alum applications to aquatic ecosystems. ............... 197 A-2 Raw soluble reactive phosphorus data for anaerobic core incubation study.......... 198 A-3 Raw soluble reactive phosphorus data for anaerobic core spiking study............... 201 B-1 Raw mesocosm water column pH data fo r 20 weeks in 2004 afte r planting, prior to alum additio n.......................................................................................................... 202 B-2 Raw outflow soluble reactive phosphorus (mg L-1) data for mesocosms nine weeks prior to alum addition............................................................................................. 203

PAGE 14

xiv B-3 Raw outflow total kjel dahl nitrogen data (mg L-1) for mesocosms nine weeks prior to alum addition......................................................................................................204 B-4 Raw outflow dissolved organic carbon (mg L-1) data for mesocosms 9 weeks prior to alum addition......................................................................................................205 B-5 Initial characterization of soil used in me socosm establishment (n=5)................. 206 C-1 Spatial coordinates for all samples collected within cell 9 and 10 of the Orlando Easterly We tland (UTM Zone 17, NAD83)........................................................... 208 C-2 Soluble reactive phosphorus collected in 2005 from replic ate transects in cell 9 and 10 of the Orlando Easterly W etland....................................................................... 210 C-3 Soluble reactive phosphorus collected in 2006 from replic ate transects in cell 9 and 10 of the Orlando Easterly W etland....................................................................... 211

PAGE 15

xv LIST OF FIGURES Figure page 1-1 The Orlando Easterly Wetland 1995 digi tal orthoquad (Orang e County, Florida).. 20 2-1 Site map of Orlando Easterly Wetland, Christ ma s, Florida. Soils collected in cell 10 (star), dominated by Typha spp. Arrows indicate water flow paths................... 29 2-2 Changes in soluble reactive phosphorus concentration in the water colum n under anaerobic conditions at treatment dosag e (a) 36.0, (b) 18.0, and (c) 9.0 mg Al m-2 for soil cores from the Orlando Easterly Wetland (n=6).......................................... 45 2-3 Changes in soluble reactive phosphorus concentration in the water colum n during weekly P spiking at treatment dosag e (a) 36.0, (b) 18.0, and (c) 9.0 mg Al m-2 for soil cores from the Orlando Easterly Wetland (n=3)............................................... 48 2-4 Changes in water column pH under anaer obic co nditions at treatment dosage (a) 36.0, (b) 18.0, and (c) 9.0 mg Al m-2 for soil cores from the Orlando Easterly Wetland (n=3)..........................................................................................................50 2-5 Changes in water column soluble al uminum concentration under anaerobic conditions at treatm ent dosage (a ) 36.0, (b) 18.0, and (c) 9.0 mg Al m-2 for soil cores from the Orlando Easterly Wetland (n=6)...................................................... 53 3-1 Site map of Orlando Easterly Wetland, Christ ma s, Florida. Plants collected in cell 1 ( Scirpus californicus ), cell 10 (Typha spp.), and cell 13 ( Najas guadalupensis ). Mesocosm location designated by star..................................................................... 63 3-2 Water column pH in Orlando Easterly Wetland mesocosms (n=3)......................... 68 3-3 Water column dissolved oxygen conten t in Orlando Easterly Wetland subm erged aquatic vegetation (SAV) and emergent aquatic vegetation (EAV) mesocosms (SAV n=3, EAV n=6)...............................................................................................69 3-4 Water column dissolved oxygen con centration in Orlando Easterly We tland mesocosms (n=9) and inflow water......................................................................... 71 3-5 Water column ammonium-nitrogen concen trations in Orlando Easterly Wetland me socosms (n=9) and inflow water......................................................................... 72

PAGE 16

xvi 3-6 Total dissolved nitrogen concentrations in the water colum n of Orlando Easterly Wetland submerged aquatic vegetation (SAV) , emergent aquatic vegetation (EAV) mesocosms (SAV n=3, EAV n=6), and inflow water.............................................. 73 3-7 Total kjeldahl nitrogen concentrations in the water column of Orlando Easterly Wetland m esocosms (n=9) and inflow water........................................................... 74 3-8 Water column soluble reactive phosphorus concentrations in Orlando Easterly Wetland m esocosms (n=3) and inflow water........................................................... 75 3-9 Water column total dissolved phosphor us concentrations in Orlando Easterly Wetland me socosms (n=9) and inflow water........................................................... 76 3-10 Total phosphorus concentrations in th e water column of Or lando Easterly W etland submerged aquatic vegetation (SAV) a nd emergent aquatic vegetation (EAV) mesocosms (SAV n=3, EAV n=6), and inflow water.............................................. 78 3-11 Particulate phosphorus concentration in the water column of Orlando Easterly Wetland subm erged aquatic vegetation (SAV) , emergent aquatic vegetation (EAV) mesocosms (SAV n=3, EAV n=6), and inflow water.............................................. 79 3-12 Unimodal growth pa ttern exhibited in Typha spp. leaves in Orlando Easterly Wetland me socosms (n=9)....................................................................................... 81 3-13 The number of live plants in each Or lando Easterly We tland mesocosm throughout the experiment (n=3 standard error)...................................................................... 81 4-1 Site map of Orlando Easterly Wetland, Christ ma s, Florida. Plants collected in cell 1 ( Scirpus californicus ), cell 10 (Typha spp.), and cell 13 ( Najas guadalupensis ). Mesocosm location designated by star..................................................................... 95 4-2 Inorganic and organic phosphorus forms in the surface and subsurface soil of alumtreated Orlando Easterly We tland m esocosms at the start and end of the study. Values are means (mg kg-1) 1 standard deviation (n=9) with percent of total P pool in parenthesis.................................................................................................. 109 4-3 Phosphorus mass balance in the submer ged aquatic vegetation me socosms after three months. Values are percent of tota l mesocosm P, with mean value (g) 1 standard deviation in parenthesis (n=3). Inflow and outflow P values are in g d-1.111 5-2 Map of study transects established in cell 9 and 10 of the Orlando Easterly Wetland, Christma s, Florida................................................................................... 119 6-1 Site map of Orlando Easterly Wetland, Christmas, Florida. Ty pha spp...............159 6-2 Map of study transects established in cell 9 and 10 of the Orlando Easterly Wetland, Christma s, Florida................................................................................... 162

PAGE 17

xvii 6-3 Water column pH of (a) alum-treated cell 10, and (b) control cell 9 in the Orlando Easterly We tland (inflow n=1, outflow n=2)......................................................... 166 6-4 Water column dissolved oxygen content in cell 9 and 10 of the Orlando Easterly Wetland (inflow n=1, outflow n=2). ......................................................................167 6-5 Water column dissolved organic C concentrat ions in (a ) control cell 9 transects, (b) alum-treated cell 10 transects, (c) inflow and outflow weirs of the Orlando Easterly Wetland (n=2)........................................................................................................168 6-6 Water column soluble reactive phosphorus concentrations in (a ) alum -treated cell 10, and (b) control cell 9 of the Orlando Easterly Wetland (n=2)......................... 170 6-7 Water column dissolved reactive phosphorus concentrations in (a ) alum-treated cell 10, and (b) control cell 9 of the Orlando Easterly Wetland (n=2). ........................ 172 6-8 Water column total phosphorus concentra tions in (a) alum-treated cell 10, and (b) control cell 9 of the Orla ndo Easterly W etland (n=2)............................................ 174 6-9 Water column particulate phosphorus con centrations in (a) alum-treated cell 10, and (b) control cell 9 of the Orlando Easterly W etland (n=2)............................... 175 B-1 Scirpus californicus leaf length to dry weight regr ession harvested from cell 2 in the Orlando Easterly W etland (n=30).................................................................... 207 B-2 Typha spp. leaf length to dry weight regr ession harvested from cell 10 in the Orlando Easterly We tland (n=12 plants, 69 leaves)............................................... 207 C-1 Field study transects in control cell 9 and alum-treated ce ll 10 of the Orlando Easterly W etland. .....................................209

PAGE 18

xviii Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy THE EFFECT OF ALUM ON PHOSPHORUS SEQUESTRATION, MACROPHYTES, MINERALOGY, AND MICROBIAL COMMUNITY IN A MUNICIPAL WASTEWATER TREATMENT WETLAND By LYNETTE MALECKI BROWN May 2007 Chair: Dr. K. Ramesh Reddy Co-chair: Dr. John R. White Major: Soil and Water Science Constructed wastewater treatment wetlands are used for secondary and tertiary treatment of nutrients su ch as phosphorus (P) prior to discharge into surface waters. Over time, the P treatment capacity of constructed wetlands may decline. Little research has been done on methods to restore the treatm ent capacity of these aging wetlands. One method to increase the P binding capacity of we tland soils is the addition of aluminum (Al)-containing amendmen ts such as alum (Al2(SO4)3H2O). Phosphorus immobilization in treatment we tlands was investigated using a fiveweek laboratory core study, three-month me socosm study, and full-scale one-year field study. Results suggest a continuous or s easonal low-dosage alum application to treatment wetlands significantly improves removal efficiencies of soluble reactive P, total P and nutrients associated with organic matter (carbon and nitrogen). However, a significant increase in amorphous Al within the surface soil was associated with lower

PAGE 19

xix soil pH. The increased Al and reduced pH both negatively impacted soil microbial biomass as well as activity. These findings we re true in short-term studies to varying degrees, while the long-term field study was inconclusive. Plant growth and nutrient bioavailability were relatively unimpaired by alum application in emergent systems, while in submerged aquatic vegetation (SAV) systems, biomass and calcium tissue concentrations significantly decreased while Al concentrations significantly increased, suggesting Al toxicity. A P mass balance suggests alum impacts in SAV-dominated systems are predominantly associated with vegetation, while in emergent macrophyte systems the soil will be most impacted, dire ctly corresponding to the largest P storage compartments in each. Total P in the surface soil was five times higher after the oneyear field alum application compared to the control. Application of alum or Al-containing am endments to soil proximal to the outflow regions of treatment wetlands will provide an effective management tool to maintain discharge concentrations within permitted values during inefficient wetland treatment times. However, alum application does result in increased particulate P within the water column which could raise total suspended solids in wetland discharges. To minimize negative impacts, the minimum dosage necessary for effective treatment should be applied, and usage constrained to the wint er season during low treatment efficiency.

PAGE 20

1 CHAPTER 1 LITERATURE REVIEW Wetlands are acknowledg ed as unique ecosystems with multiple functions and values including biodiversity and habitat conservation, water quality improvement, and hydrologic benefits. Natural wetlands have b een used as discharge points for wastewater and sewage since 1912 (Kadlec a nd Knight, 1996) and this pract ice continues even today. With increasing knowledge of how wetlands function to improve water quality and a growing interest in ecological engineering, constructed wetlands started to be built in the 1970s. The first constructed wetland treatmen t system in North America was constructed in 1973 by Brookhaven National Laboratory (Sma ll and Wurm, 1977). Since that time, constructed wetland designs have improve d and a variety can be found including individual surface or subsurface flow wetla nds to more complicated systems that incorporate cascade aeration and ornamental ponds. A variety of wetland plants have been used in these systems from submerge d aquatics to emergents. The most common vegetation includes Phragmites spp., Scirpus spp., and Typha spp. (Denny, 1997; Knight, 1997). Constructed Treatment Wetlands Constructed wastewater treatme nt wetlands are a relatively lo w-cost alternative increasingly being used in de veloping countries to provide primary wastewater treatment, while in industrial nations of Europe and North America, treatment wetlands are being used for tertiary treatment of nutrients pr ior to discharge into surface waters. In 1999 more than 200 communities in the United Stat es reported use of constructed wetlands for

PAGE 21

2 wastewater treatment (Reagin, 2002). Several countries also use constructed treatment wetlands for treatment of urban runoff, stor mwater, animal and agricultural wastewater, industrial wastewater, acid mine leachate, a nd even landfill leachat e (Bulc et al., 1997; Denny, 1997; Lakatos et al., 1997; Mander and Mauring, 1997; Schreijer et al., 1997; Schutes et al., 1997; Clarke and Baldwin, 2002; White et al., 2004; Hadad et al., 2006). The performance of most constructed wa stewater treatment wetlands is based on their nutrient (phosphorus (P) and nitrogen (N )) reduction capacity, as well as reduction in biological oxygen demand (BOD) and susp ended solids. Nutrient reduction is accomplished by physical settlement, plant and microbial uptake, and denitrification in the case of N (Healy and Cawley, 2002). Cons tructed wetlands can remove up to 80% of the N (Knight et al., 1993) entering the sy stem depending on the residence time of the wetland, which allows particulate organic N to settle, and temperature, which affects denitrification rates and microbial activ ity (Bachand and Horne, 2000; Healy and Cawley, 2002). Denitrification is an anaerobic process where n itrate or nitrite is used as the terminal electron acceptor. The possibl e end products resulting from denitrification include bacterial redu ction of nitrate (NO3 -) to nitrous oxide (N2O) or dinitrogen gas (N2), both potential gaseous losses of N, diffusing up and out into the atmosphere (Christian, 1989). Phosphorus in Wetlands Phosphorus does not have a gaseous form that can typically be released from wetlands, instead cycling betw een the biota, soil, and wa ter. The adsorption and retention of P in freshwater wetland soils is controlled by in teractions of redox potential, pH, Fe, Al, Mg, and Ca minerals (Kaggwa et al., 2001). Phosphorus removal under oxic conditions is usually attributed to its binding with ferric iron (Fe3+), forming insoluble

PAGE 22

3 complexes (Upchurch et al., 1974). Howe ver under anoxic conditions typical of wetlands, Fe3+ is reduced to ferrous iron (Fe2+) with liberation of P to the overlying water column (Lee et al., 1977). Aluminum (A l), calcium (Ca), and magnesium (Mg), however, can form insoluble compounds, binding P under both aerobic and anaerobic conditions, affected instead by the pH of the system. Aluminum tends to bind P at pH 68 (Cooke et al., 1993b), while calcium has been found to bind P at pHs above 8 (Diaz et al., 1995; Gomez et al., 1999). Faulkner and Richardson (1989) reported that the most important P retention mechanism in wetland soils is ligand exchange, where phosphate displaces water or hydroxyls from the surface of Fe and Al hydrous oxides to form complexes within the coordinati on sphere. Phosphorus is also taken up initially by plants and microbes. The uptake capacity of emer gent plants ranges from 30 to 150 kg P ha-1 yr-1, however, 35-75% of it is eventually released back into the water column if the plants are not harvested (Gumbricht, 1993; White et al., 2000). Wetland Macrophytes Wetland plants are a critical part of cons tructed wetlands. They serve to distribute and reduce current v elocities which increases sedimentation of suspended solids and reduces erosion and resuspen sion (Pettecrew and Kalff, 1992). The stems and leaves submerged in the water column provide surf aces for biofilms of algae, protozoa, and bacteria while the roots and rhizomes provi de a substrate for a ttached growth of microorganisms as well (Gumbricht, 1993). The biofilms above and below ground are responsible for most of the microbial cyc ling that occurs in wetlands (Brix, 1997). Additionally, wetland macrophytes have adapted to anaerobic soil conditions allowing for the transport of oxygen into the roots and rhizosphere to support aerobi c respiration and oxidize phytotoxic reduced compounds such as Fe2+, Mn2+, and S2(Allen et al.,

PAGE 23

4 2002). Some plants release enough oxygen in to the root zone to support aerobic microbial activity and this can represent up to 90% of the total oxygen entering the wetland soil (Reddy et al, 1989; Bodelier et al., 1996). Oxygen release rates from submerged aquatics range between 0.5-5.2 g m2 d-1 (Sand-Jensen et al., 1982; Caffrey and Kemp, 1991). This oxygen release influences the biogeochemical cycles by affecting the redox status of the soil, stimulating aer obic decomposition of organic matter, and sustaining nitrifying bacteria (Barko et al., 1991; Brix, 1997). Cattails ( Typha spp.) are one of the dominant macrophytes used in treatment wetlands. Davis (1984) conduc ted an extensive study of Typha plant nutrient fluxes in Water Conservation Area 2A of the Florida Everglades. Typha had a life cycle ranging between 12-96 weeks. The P and N concentra tions of the leaves, roots, and rhizomes were analyzed seasonally and concentrations were found to be lower in the spring during flowering, and higher in the fall. The a nnual uptake of P through leaf production was 0.77-3.65 g m-2 while N uptake was 11.1-29.3 g m-2. However most of the nutrients acquired during leave growth were lost befo re death, and nutrient retention was only 1728% P and 37-71% N. The nutrients leached fr om the dying leaves are released back to the water column or translocated to the roots and rhizomes (Davis, 1984). The Problem Over time, the P rem oval capacity of constructed treatment wetlands may therefore decline. Little research has been done on met hods to restore treatment capacity of older constructed wetlands since most treatme nt wetlands in use today are young, and restoration has not been necessary until now (Ann et al., 2000; Hunter et al., 2001; Stevens et al., 2002; Jamieson et al., 2003; Wa ng et al., 2006). However, extensive research has been done on lake rehabilita tion techniques and thei r applicability in

PAGE 24

5 treatment wetlands are now being tested. Se veral restoration opti ons include dredging, nutrient inactivation / precip itation, biotic harvesting, and sediment exposure and desiccation to restore ecological function (Dunst et al., 1974). Chemical amendments for nutrient inactivation have several advantages over other restoration methods including the ease of their application and relatively low cost . However, their effectiveness in wetlands to inactivate P, their longevity, and effect on the flora and fauna are issu es that need to be more fully addressed. Aluminum Sulfate (Alum) The chemical am endment used most often for P inactivation in lakes and coagulation in the wastewat er treatment industry is Al2(SO4)3H2O (alum). When alum is added to the water column it disso ciates, forming aluminum ions (Al3+) that are immediately hydrated, forming a hexaaquoaluminum ion (Al(H2O)6 3+) which is moderately acidic (Omoike and Vanloon, 1999) . Through several rapid hydrolytic reactions an insoluble, gelatinous, poor ly crystalline aluminum hydroxide (Al(OH)3) floc is formed (Ebeling et al., 2003) which has high P adsorption properties with a surface area greater than 600 m2 g-1 (Huang et al., 2002). The alum floc passes through the wate r column removing particulates and inorganic P (Connor and Martin, 1989). It then settles to the soil surface where it sorbs and retains P within the molecule’s lattice, inert to changes in redox (Cooke et al., 1993b). This floc acts to retard P releas e from the sediment to the water column (Peterson et al., 1973; Connor and Martin, 1989), not only bi nding mobile P but also P bound to Fe (Rydin and Welch, 1998). Once all Al adsorption sites are occupied, P diffusing upward will replenish the available s ites on the Fe (Rydin et al, 2000). The floc then consolidates with the soils where it can continue to sorb and retain P for several

PAGE 25

6 years as indicated by increased TP and Al-P in soil core profiles (Narf, 1990; Welch and Schrieve, 1994; Rydin and Welch, 1998; Rydin et al., 2000). As the Al(OH)3 floc migrates downward through the soil via sedime ntation and resuspension however, it loses its effectiveness as a physical barrier to P fluxing out of the surf icial soil layer (Connor and Martin, 1989). Additionally the P binding sites become saturated over time, rendering the floc unavailable to bind futu re migrating P layer (Connor and Martin, 1989). The size of the floc formed is directly re lated to the alum dose (Chakraborti et al., 2003). Immobilization of 1 mg of PO4 3theoretically requires 0.28 mg of Al3+, however, the alum floc also binds organic matter, re ducing its P treatment efficiency, requiring an increased alum dosage (Van Hullebusch et al ., 2002). The Al hydrol ysis involved in the floc formation is a complexation reacti on involving hydroxyl ions , and therefore is temperature dependent. The concentrati on of hydroxyl ions changes due to the temperature dependence of the ion product of wa ter, in turn increasi ng Al solubility at colder temperatures due to decreased hydr oxide concentrations, unless the pH is increased (Van Benschoten and Edzwald, 1990; Exall and Vanloon, 2000). Additionally, the increased viscosity of cooler water hamp ers sedimentation of the coagulant resulting in smaller, less dense flocs (Morris a nd Knocke, 1984; Hanson and Cleasby, 1990), and may influence the kinetics and equilibrium of hydrolysis and metal hydroxide formation (Knocke et al., 1986; Matsui et al., 1998). The complexation of P by Al coagulants invol ves three types of interactions. First, a precipitate may form (AlPO4) resulting from charge neutralization by ligand competition (Ratnaweera et al., 1992). The pr ecipitation is hypothesized by some to be

PAGE 26

7 governed by the integration of Al-OH-Al and Al-PO4-Al linkages into an Al hydroxyphosphate rather than th e precipitation of Al(OH)3 or AlPO4 (Hsu, 1975). Second, there are chemical complexations that result in soluble comple xes. Finally, there is adsorption of phosphate ions onto the Al(OH)3 surface through outer sphere complexation (Ratnaweera et al., 1992). Th e sorption of phosphates occurs by replacing the –OH2 or –OH groups of the hydrous Al oxides (Rajan et al., 1974). The PO4-Al interaction is primarily dependent on the charge density of the Al species, thus the cation charge can be increased by lowering the OH/Al ratio or pH to induce more efficient P removal (Lijklema, 1980; Diamadopoulos and Benedek, 1984; Boisvert, 1997). If, however, the OH/Al ratio falls below 2.12 th e coagulation-flocculation-sedimentation process does not occur (B oisvert et al., 1997). The controlling factor in the effectiveness a nd toxicity of alum is, therefore, the pH of the system. Alum itself has a pH of approximately 2.4 (Beecroft et al., 1995; Lind, 2003) and therefore tends to decr ease the pH of the system it is added to. As long as the pH of the system remains between 6 and 8, insoluble polymeric aluminum hydroxide (Al(OH)3) will dominate and P inactivation results (May et al., 1979). If the pH decreases to between 4 and 6 soluble interm ediates will occur, releasing bound P (Cooke et al., 1993a). Below pH 4 soluble Al3+ dominates which may result in Al toxicity (Cooke et al., 1993b), and at pH 8 or greater the aluminate ion (Al(OH)4 -) dominates due to its amphoteric nature, re leasing bound P and increasing soluble Al (Cooke et al, 1993a). Aluminate, similar to Al3+, is associated with Al toxi city in plants (Kinraide, 1990, Eleftheriou et al., 1993; Ma et al., 2003). The most reactive, toxic Al forms are Al3+, AlOH2+, and Al(OH)2 + (Pyrzy ska et al., 2000) and Al13+ (Kinraide, 1990).

PAGE 27

8 Concerns in Alum Application Aluminum toxicity is a critical concern in alum applications to lakes due to its effects on the biota. Aluminum has been found to be toxic to fish at concentrations as low as 0.1 mg L-1 at a pH below 6 or above 10 (Baker, 1982; Neville, 1985; Ramamoorthy, 1988). The mechanism of toxicity has been attributed to the inability of fish to maintain osmoregulatory balance and respiratory problems associated with the gills (Driscoll and Schecher, 1990). Other c ontinuous exposure studies have found that even low doses of Al(OH)3 resulted in chronic effects on fish (Freeman and Everhart, 1971; Kane and Rabeni, 1987; Gundersen et al., 1994). Impacts of Al toxicity have been noted in macroinvertebrate and phytoplankt on populations as well (Moffet, 1979; Lamb and Bailey, 1981; Welch et al., 1982; Barbiero et al., 1988; Narf, 1990; Smeltzer, 1990; Smeltzer et al., 1999). Howeve r, as long as the pH of the system in question remains above 6, typically there should be no nega tive effects on the biota from the alum treatment. It is common to observe an imme diate drop in pH after alum treatment (down to a pH between 4 and 5) but the pH of lakes usually recover in less than one hour (Welch et al., 1982). The effectiveness of alum can also be co mpromised if external nutrient loading is not reduced prior to application. The Eau Galle Reservoir located in Wisconsin was dosed with alum in 1986 to reduce the inte rnal P loading, however, the treatment was ineffective, lasting less than one year (B arko et al., 1990). The ineffectiveness was attributed to elevated external P loading due to unusually high precipitation during the treatment year. Additionally, the high inflow resulted in high sedi ment deposition rates, burying the alum floc, which in turn shortene d its treatment effec tiveness. Foy (1985) had similar results in White Lough due to sedimentation.

PAGE 28

9 The abundance of submersed macrophytes can also affect the alum treatment longevity. Three lakes that contained thick stands of macrophytes in Washington (Wapato Lake, Long Lake, and Pattison Lake South) were treated with alum and the applications failed to reduce P levels (W elch et al., 1982; Entranco, 1986; Entranco, 1987; Welch and Schrieve, 1994). It was suggest ed that the thick stands of macrophytes such as Myriophyllum , Ceratophyllum, and Elodea canadensis reduced the effectiveness of alum. The plants may intercept the set tling alum floc, resulti ng in irregular alum coverage on the sediment surface that may ha ve led to the treatments ineffectiveness (Welch and Schrieve, 1994; Welch and Cooke, 1999). Additionally, P was released from senes cence and decomposition of the plants in Pattison Lake South, continuing internal P lo ading because the plant roots penetrated below the alum layer at the sediment surf ace (Welch and Schrieve, 1994). In Wapato Lake, sediment P release occurred due to a plant-mediated increase to pH 10 (Entranco, 1986). Another problem with macrophytes in alum treated lakes is that the increased water clarity tends to stimulate plant growth. The increased biomass can result in even greater internal loading from senescence (Entranco, 1986). However, not all plants contribute to internal P loading. Some such as Egeria densa in Long Lake senesce slowly and do not increase growth in res ponse to increased water clarity (Welch and Schrieve, 1994). Additionally, an alum/ calcium hydroxide (Ca(OH)2) treatment was used successfully by the St. Johns River Wate r Management District to treat 12,000 acres of wetlands in the Harris Chain of Lakes region (Hoge, 2003), and preliminary studies by DB Environmental, Inc. at the Orlando East erly Wetland (OEW) have shown alum to be an effective treatment as well (Auter et al., 2003).

PAGE 29

10 Effect of Alum on Aquatic Macrophytes It is important to de termine how the alum floc affects the availability of nutrients for the aquatic plants. Available P is an e ssential nutrient for aqua tic vegetation and the alum floc which settles to the soil floor ma y not only deprive the plant of P but the alum may have adverse effects on the plants. The Al concentration in plants is typically very low (0.02% of the total weight) (Hutchinson, 1945). Aluminum toxicity in plants is related to the activity of the Al3+ ion, having several different effects on the plant. The most notable result of Al toxi city in upland plants is the inhibition of root elongation and respiration (Schier, 1985; Jarvis and Hatch, 1986) resulting in root s that are thickened, stubby, brittle, and often ineffici ent in nutrient absorption. There are five possible mechanisms by which Al toxicity af fects the cellular function within plants (Taylor, 1989). First, Al may disrupt the stru cture and function of the plasma membrane which serves as a wall between the cytosol and external environment. The Al has been found to inhibit golgi development, inactivate membranebound enzymes, and affect membrane permeability (Matsumoto and Yamaya, 1986; Bennet et al., 1987; Zhao et al., 1987). A sec ond possible mechanism of Al toxicity is the inhibition of ATP (Viola et al., 1980) and DNA synthesis as well as mitosis. Aluminum has been found to localize in th e nucleus of several different plants and is thought to bind to the phosphate in DNA, increasing the ri gidity of the double helix resulting in chromatin aggregation (Matsumoto et al., 1977b; Matsumoto, 1988). Aluminum toxicity also disrupts cellular function by inhibiting cell elongation. The Al binds to the free carboxyls of pectin resulting in cross-linking of the molecules, decreasing the cell wall elasticity, which in turn inhibi ts root elongation (Klimashevskii and Dedov, 1975; Matsumoto et al ., 1977a). Aluminum stress also results in disruption

PAGE 30

11 of mineral nutrition. The disruptions could arise from reduction in mycorrhizal association (Entry et al., 1987). Some hypothesize the toxic e ffects of Al are due to Alinduced phosphate deficiency, since the symp toms look similar and Al stressed plants have increased P concentrations in the roots and decreased P concentr ations in the shoots (Thornton et al., 1987; Cumming et al., 1986; Jarvis and Hatch, 1986, Tan and Binger, 1986). Aluminum also interferes with the absorption and transport of Ca and Mg in plants, resulting in reduced Ca concentrations in the roots an d shoots of Al stressed plants (Baligar et al., 1987; Bennet et al., 1987; Thornton et al., 1987). Welch et al. (1982) found alum treatment of a lake in Washington resulted in plant growth inhibition. They assumed there would be an increase in macrophyte biomass due to the increased transparency of the lake, how ever the biomass either stayed constant or decreased from pre-treatment values. A study by Kaggwa et al. (2001) looked at the effects of alum sludge discharge from a wate r treatment plant into a natural swamp in Uganda. Initial biomass of 5 x 5 m plots was determined by clear cutting plants ( Phragmites, Cladium , and Cyperus ) at their base. Shoot productivity and re-growth were then monitored by harvesting 1 x 1 m plots every 3 weeks, separating the shoots into leaves, roots, and stems. There was no effect on the plant biomass, but productivity was hindered, suggesting the alum sludge may affect the root system, reducing nutrient availability to the plants. Root biomass produc tion is normally more sensitive to Al than shoot biomass production (Zhang and Taylor, 19 88). Additionally, the P concentration in plant tissue tends to increase with biomass, but Kaggwa et al. (2001) found there to be no change, indicating the P supply to the plants may be inhibited by the Al. Goransson and Eldhuset (1991) determined P bound to Al limits the growth of terre strial plants and

PAGE 31

12 causes physiological stress. On the other ha nd, Carignan and Kalff ( 1979) indicated that alum should not interfere in the growth of rooted aquatic macrophytes because they are able to access the high P content in the subs urface sediment. Peltier and Welch (1969, 1970) found that the amount of nutrients (N and P) in the sediment affected plant growth more so than the nutrient concentration in the water column. Effect of Alum on Mineralogy There ma y also be changes in the mineralogy of the soil, due to alum application, that could affect the nutrient availability to plants. Alum inum forms both crystalline and poorly crystalline oxides, hydroxides, and oxyhydr oxides which are able to strongly sorb essential plant nutrients (Huang et al., 2002). The primary Al mineral found in soils is gibbsite (Al(OH)3) followed by boehmite (AlOOH) which is less common. Other rare Al minerals include bayerite (Al(OH)3), nordstrandite (Al(OH)3), doyleite (Al(OH)3), diaspore (AlOOH), and corundum (Al2O3). Rapid precipitation in the soil environment typically produces bayerite and/or nordstrandite while slow crystallization produces gibbsite. Aluminum may al so form variscite (AlPO4) which is the stable solid phase if phosphate precipitates at pH 5 to 7 (Jiang and Graham, 1998). Burrows (1977) indicated crystallization after alum application may take over a year to complete, as larger units of polymeric Al(OH)3 are formed. The identification of Al minerals can be determined via x-ray diffraction (XRD). In oriented mounts of the clay fraction, gi bbsite has a strong diffr action peak at 4.85 and a weaker peak at 4.37 (Hsu, 1989; Kmp f et al., 2000). The peaks do not change with potassium saturation and it dehydroxylat es at 300 C, thus the peak collapses. Gibbsite, bayerite, nordstrandite, and doyleite all have their primary diffraction peak between 4.75-4.85 , thus identification require s careful measurements of the positions

PAGE 32

13 and intensities of the additional peaks (Chao et al., 1985). Boehmite is identified by a peak at 6.11 which is often broad since it o ccurs as small crystals (Tettenhorst and Hoffman, 1980). There is no XRD met hod for measuring the amount of poorly crystalline Al hydroxide in soils (Bertsch and Bloom, 1996). The edges of gibbsite and boehmite, as well as the surface of poorly crystalline Al hydroxide contain undercoordinated oxygen atoms which are the source of pH-dependent variable charge and serve as active sites for sorption reactions with transition metals, organic acids, and oxyanions such as phosphate (Goldberg et al., 1996; Huang et al., 2002 ). Phosphate has a strong affinity for Al, displacing hydroxyls from surface sites to form surface Al complexes, therefore its sorption to Al hydr oxides remains strong at all pHs, decreasing its bioavailability (Hsu, 1989). Effect of Alum on Microbes and P Cycling A similar area of research that has not been addressed, even in lake studies, is the effect of alum on the m icrobial populations in the soil. Microbes ar e generally sensitive to soil acidity and soluble Al (Robert, 1995). The microbial biomass has the potential of being a sensitive indicator of changes in the soil dynamics (Powlson and Jenkinson, 1980) due to alum application since there is a close relationshi p between microbial biomass nutrients and levels of mineralizable nutrients available in the soil (Jenkison and Ladd, 1981). An alum study by Connor and Martin (1989) on a shallow New Hamshire lake suggested that the dissolved oxygen con centrations within the lake may have been affected by a suppression in activity or reduction in populatio n of BOD producing organisms. Bacterial cells are colloidal par ticulates and can theref ore be aggregated by the alum floc (Eriksson and Axberg, 1981). This bacterial re moval may delay the reestablishment of essential nutr ient cycling (Bulson et al., 19 84). Therefore, the size and

PAGE 33

14 activity of the microbes need to be assesse d to fully understand the nutrient cycling within an alum treated wetland. The effect of alum on the microbial-mediated nutrient cycling of P within the soil is also unknown. It has been noted that over time the alum floc migrates downward in the soil profile due to sediment deposition (Lewandowski et al., 2003). This may result in differing microbi al activity, biomass, and nutrient cycling above and below the floc layer. Alum Forms and Alternatives Once alum is selected as the treatm ent for P immobilization, the form of alum to be used must be determined. There are both dry (powder or pellets) and liquid forms of alum available. Alum residual formed as a by-product in the potable water treatment process can also be used and can often be obtained free of ch arge. Hoge (2003) recommended that dry alum or alum residual be used in dry situati ons such as drought or drawdown, and liquid alum be used in wet s ituations such as ponds and lakes. Hoge (2003) also noted that when liquid alum was used in wetlands the deposits on the vegetation could not be washed off suggesti ng that perhaps alum pellets may provide better results with less damage to the plants. Recently polyaluminum chloride (PAC) has been developed for use in the water treatment process (Viraraghavan and Wi mmer, 1988). Polyaluminum chloride (Aln(OH)mCl(3n-m)) is a partially hydrolyzed aluminum chloride solution which has recently been found to provide stronger, faster settling flocs than alum (Boisvert and Jolicoeur, 1999). That is because of the partial elimination of the polymerization process that occurs when alum is added to water. Polyaluminum chloride instead contains preformed, highly charged polymers such as the dimer Al2(OH)2 4+ and the Al13 species AlO4Al12(OH)24(H2O)12 7+, inducing rapid coagulation (K oether et al., 1997). The Al13

PAGE 34

15 species is considered to be the most stab le and effective species in water treatment (Bottero et al., 1980). A study by Shen a nd Dempsey (1998) found that 50% of the polymers converted to Al(OH)3 precipitate after 60 days a nd the rest were converted to more inert species. This sugge sts the initially formed stab le polymers are transitional, slowly converting to more st able species during aging. Another benefit of PAC is the elimination of the need for pH adjusting chemicals since PAC does not decrease the pH as much as alum does (Lind, 2003) and works in a wider pH range (Jiang and Graham, 1998). A dditionally, PAC is affected much less by temperature than alum, thus it would benefit re storation efforts in colder climates (Van Benschoten and Edzwald, 1990; Koether et al., 1997; Jiang and Graham, 1998). Furthermore, in the manufacturing process, the Al becomes more polymerized resulting in higher cation charge than alum, thus incr easing its P binding cap acity so that lower doses can achieve equivalent treatment effi ciency (Jiang and Graham, 1998; Lind, 2003). Low basicity (0-16.7%) PAC (Al2(OH)Cl5) is best for P removal while high basicity PAC is used for turbidity removal (Lind, 2003). Basicity is determined by the OH/Al ratio (Boisvert et al., 1997). PAC has not yet b een used for enviro nmental restoration however, due to its high cost, 2.5 times more than alum (Viraraghavan and Wimmer, 1988). A low-cost alternative to PAC is partially-neutralized aluminum sulfate (PNAS) also called polyaluminum hydroxysulfate (P AHS) developed by Koether et al. (1993a; 1997). Partially-neutralized aluminum sulf ate is a solution formed by adding powdered calcium carbonate to concentrated alum in a 0.75:1 molar ratio for 50% basicity, and only adds 10% cost to that of alum (Beecroft et al ., 1995; Koether et al., 1997). ). Partially-

PAGE 35

16 neutralized aluminum sulf ate contains several Al species including monomers [Al(OH)]2+, small octahedrally coordinated Al species, and larger polymers [Al13([AlO4(Al12(OH)24H2O)12]7+] (Koether et al., 1993b). The pH of PNAS is 3.3-3.4 depending on whether a concentrated or dilu te solution is used (Beecroft et al., 1995) which is less acidic than alum and preferre d when adding Al to a natural system. Partially-neutralized aluminum sulfate like PAC is affected much less by temperature than alum (Koether et al., 1997; Exall and Vanloon, 2000), thus benefiting restoration efforts in colder climates, and has been used in wastewater and drinking water treatment, but has not yet been used fo r environmental restoration. Alum Dose Determination and Effectiveness After the form of alum is selected, the dos e to be applied f or P inactivation needs to be determined. Several methods of dose dete rmination are available. Ideally, the dose should be based on the total amount of mobile P in the soil to be inactivated (Rydin et al., 2000). Assuming the duration of alum’s P binding capacity is related to the concentration of Al(OH)3 in the sediment, Kennedy and Cooke (1982) established that the maximum dose of alum should result in less than 50 g L-1 of dissolved Al in the water column to prevent adverse effects on the biota. Another method, used in lake dose determ ination, is based on alkalinity. Water samples of differing alkalinitie s are titrated with alum to a pH of 6.0 (Kennedy and Cooke, 1982). Dose determination can also be based on the internal P loading of the system (Kennedy, 1978; Kennedy et al., 1987). Assuming a target period of control and stoichiometric ratio of 1.0 between the alum-P complexes, the dose would equal the average summer internal load times the target period and that value is then doubled to account for errors. A study by Rydin and Welch (1998) determined that the amount of

PAGE 36

17 alum used should be based on the concentration of Fe-bound P since this entire pool will be depleted if enough Al is used, increasing th e burial of the mobile surface sediment P subject to release under anoxic conditions. Rydin and Welch (1999) developed a ne w method for dose determination based on the direct measurement of mobile P (loosely-so rbed P and Fe-P) in the top 4 to 10 cm of retrieved sediment cores and a 100:1 ratio of Al added to Al-P formation. This ratio reflects competition for binding sites, thus if a 1:1 ratio is used for dose determination the amount of alum calculated to inactivate the P may be greatly underestimated. The U. S. Army Corps of Engineers is examining the Rydin and Welch (1999) procedure for use in Corps watereways (James and Barko, 2003). Ann (1995) tested seve ral different alum doses (1.4-21.8 g kg-1) on 5 g of air-dried wetland soils incubated on a mechanical shaker with 25 mL of distilled, deionized water ( DDI) for 3 days and analyzed for soluble reactive P (SRP). In a similar study Ann et al . (2000) tested several different alum doses (3.7-23 g kg-1) on air-dried wetland soils, put the mi xtures in columns and flooded them sampling weekly for SRP over a period of 12 weeks. The effectiveness of an alum treatment is often determined by improved transparency based on secchi depth and decrea sed whole-lake TP concentrations (Welch et al., 1982; Connor and Mart in, 1989; Jacoby et al., 1994; Welch and Cooke, 1999). The efficiency of alum to reduce internal loadi ng in lakes can be determined by estimating the amount of P removed from the P cycle. Assu ming all Al added remains in the sediment and the observed Al:Al-P ratio is valid, the P inactivated can be estimated (Welch et al., 1982; Rydin et al., 2000).

PAGE 37

18 Hypotheses and Objectives Based upon the reviewed literature, se veral hypotheses can be generated regarding th e use of alum in wetlands. Fi rst, it appears that alum will effectively sequester P in a municipal wastewater treatme nt wetland. However, this alum treatment will decrease the growth and nutrient uptake of the aquatic macrophytes, as well as result in a decline in biomass and activity of the microbial community. Fi nally, changes in the soil mineralogy will be evident due to the addition of alum. The overall purpose of this research, therefor e, is to determine the effectiveness of alum treatment in the OEW, a wastewater treatment wetland for the city of Orlando. To determine this, both laboratory scale and fiel d scale experiments were utilized. At the laboratory scale, an intact core incubation st udy was used to determine the most effective dosage and effectiveness of alum and its a lternatives (alum residual, PAC, PNAS) in immobilizing P under anaerobic conditions typical of wetlands. At the field scale, intact cores were collected periodically over a one -year period (0 mo., 4 mo., 8 mo., 12 mo.) from both an alum-treated cell and a c ontrol cell for soil characterization, Al characterization, P cycling, and mineralogical composition analysis. Furthermore, microbial parameters such as microbial bioma ss P, potentially mineralizable P, and soil oxygen demand were measured to determine th e direct effect of alum on microbial biomass and activity. In addi tion, the effect of the alum application on the growth and nutrient uptake of the dominant wetland plants, Typha spp., Scirpus spp., and submerged aquatic vegetation (SAV) growing in the OEW was addressed. Site Description The Orlando Easterly Wetland s Reclama tion Project (OEW) located in Orange County, Florida is one of the oldest and largest constructed treatment wetlands in the

PAGE 38

19 United States, located east of Orlando in Christmas, FL. The wetland was built in 1986, designed by Post, Buckley, Schuh & Jernigan, Inc. (PBS&J) for the City of Orlando’s Iron Bridge Regional Water Pollution Cont rol Facility (WPCF) which needed an alternative discharge point for its wastewater effluent (Burney et al., 1989). The main goal in designing the system was to use m acrophytes to facilitate additional nutrient removal for an average daily flow of up to 132 thousand m3d-1 of effluent from the Iron Bridge WPCF servicing Orlando, Winter Pa rk, Maitland, and Cassselberry (2002). before discharging into the St. Johns Rive r (Black and Wise, 2003). The main goal in designing the system was to use macrophytes to facilitate additional nutrient removal for an average daily flow of up to 35 mgd of ef fluent from the Iron Bridge WPCF before discharging it to the St. Johns River (PBS&J, 1997). The site selected for construction was a 664 ha piece of land located 3.2 km west of the main channel of the St. Johns River (SJR ). Historically, the land had been part of the riparian wetland adjacent to the SJR, but was drained for cattle pasture around the turn of the last century (Burney et al, 1989). The site had a natura l topographic gradient of 0.2% downward from west to east so PBS&J used that gradient to divide the land into eighteen cells, with an average elevation drop across each cell of approximately 1 m (Martinez and Wise, 2003) (Figure 1-1). The resulting residence time has varied from 21 days during the dry season up to 65 days duri ng the rainy season. To control water depth in each cell 32 water control structures were included in the design (Martinez and Wise, 2003). Water exits the wetla nd through a weir control structure and flows into a receiving ditch. From there water can fl ow directly to the SJR or by sheet flow

PAGE 39

20 Figure 1-1. The Orlando Easterly Wetla nd 1995 digital ortho quad (Orange County, Florida). through Seminole Ranch, a natural marsh adj acent to the OEW owned by the St. Johns River Water Management Dist rict (SJRWMD) (PBS&J, 1997). Construction of the 494 ha wetland began in 1986. Approximately 29 km of berm were constructed by February 1987, using fill material from a 36 ha lake (Lake Searcy) constructed in the hardwood swamp (Burney et al, 1989; PB S&J, 1997). In July 1987 the OEW went online, receiving 30,280 m3d-1 from the Iron Bridge WPCF as permitted by the Florida Department of Environmenta l Protection (FDEP). From 1988 to 1996 the annual average flow remained slightly over 49,200 m3d-1 (PBS&J, 1997). The overall average influent total phosphorus (TP) c oncentration from 1988 to 2005 was 0.22 mg L-1, however, annual TP concentratio ns ranged from 0.02 – 3.30 mg L-1. Since its start, the OEW has exceeded performance expectations. The N and P permit limits established by the FDEP were 2.31 mg L-1 and 0.2 mg L-1, respectively (PBS&J, 1997). From 1988 to 1995 the average total nitrogen (T N) discharged was 0.81 mg L-1 and the average TP

PAGE 40

21 discharged was 0.07 mg L-1 (PBS&J, 1997). Miner (2001) estimated the OEW stores 73% of the P input to the system, 66% of whic h is stored in the soil. The rate of P accumulation in the OEW over its first 14 years of operation was estimated to be approximately 0.51 g m-2 yr-1 while the organic matter acc reted at a rate of 260 g m-2 yr-1 (Miner, 2001). A secondary objective of the OEW design was to provide wildlife habitat. To achieve this, three different vegetative communities were established using over 2.2 million plants, including cattails, bulrushes , duckweed, and water lilies (Reagin, 2002). Cells 1 through 12 make up a 166-ha, deep mars h designed primarily for nutrient removal (PBS&J, 1997). Cells were plan ted with either cattails ( Typha spp.), giant bulrush ( Scirpus californicus), or a combination of the two as a large-scale experiment to test their treatment abilities and competitive effects between the two species. Cells 13 through 15 consist of a 154-ha, mixed marsh that was planted with submergent and emergent macrophytes including Pontederia cordata, Sagitt aria lancifolia, Najas guadalupensis, Lemna spp., Salvinia minima, Pi stia stratiotes, and Limnobium spongia . These cells serve as a transitional zone, provi ding nutrient removal but also serving as a diverse wildlife habitat. Finally, cells 16 and 17 composed a 162-ha hardwood swamp, designed primarily as a wildlife habitat (Burney, 1989; PBS&J, 1997). Approximately 160,000 wetland trees were planted throughout the cells with an understory of emergent herbaceous species (PBS&J, 1997). Even Lake Chipster located in the hardwood swamp was designed to maximize wildlife habitat by us ing an irregular shor eline, varied slopes and water depths, as well as placing debris fr om construction within the lake to serve as fisheries habitat.

PAGE 41

22 The wetland has consistently reduced nutrient concentrations to meet the discharge permit requirements. Recently, however, P c oncentrations have fluctuated during the winter months resulting in concern over th e P binding capacity of the soil. Several management strategies are now being im plemented to ensure permit requirements continue to be met.

PAGE 42

23 CHAPTER 2 RESTORATION OF PHOSPHORUS SEQUESTRATION IN TREATMENT WETLAND SOIL USING AL-CONTAINING AMENDM ENTS Introduction There are two prim ary mechanisms of phosphorus (P) removal in freshwater wetlands, burial and sorption (Reddy et al., 1999b). Organic matte r accretion over time provides long-term P storage through humification and peat formation (Craft and Richardson, 1993; Pant and Reddy, 2001; Craf t and Chiang, 2002) while adsorption and retention of P in wetland soils can fluctuate, controlled by in teractions of redox potential, pH, iron (Fe), aluminum (Al), magnesium (Mg) , and calcium (Ca) minerals. Phosphorus removal under oxic conditions is usually at tributed to binding with ferric iron (Fe3+), forming insoluble complexes (Upchurch et al., 1974). Under anoxic conditions typical of wetland soils, Fe3+ is reduced to soluble ferrous iron (Fe2+) leading to liberation of P to the overlying water column (Lee et al., 1977). Aluminum, Ca, and Mg can form insoluble compounds, retaining P under both aerobic and anaerobic conditions, with binding capacity affected by the pH of the system. Aluminum tends to bind P most efficiently at pH 6-8 (Cooke et al., 1993), while Ca has been found to bind P at pHs above 8 (Diaz et al., 1994; Gomez et al., 1999). Over time, the P removal capacity of treatment wetlands may decline as soil exchange sites become saturated. Little resear ch has been done on methods to restore the treatment capacity of older constructed wetla nds since most treatment wetlands in use today are relatively young, and treatment eff ectiveness has not declined until recently.

PAGE 43

24 Extensive research has been done on lake rehabilitation, however, using a variety of techniques and their applicabil ity for treatment wetlands are now being tested (Ann et al., 2000; Hunter et al., 2001; Steven s et al., 2002; Jamieson et al., 2003; Wang et al., 2006). Several restoration options in clude dredging, nutrie nt inactivation/prec ipitation, biotic harvesting, and sediment exposure and desiccation to restore ecological function. Chemical amendments for nutrient inactiva tion have several advantages over other restoration methods including the ease of app lication and relatively low cost. However, their effectiveness in wetlands to inactivate P, their longevity, and e ffect on the flora and fauna are issues that need to be investigated. The chemical amendment used most often for P inactivation in lakes and coagulation in the wastewat er treatment industry is Al2(SO4)3H2O (alum). When added to the water column alum dissociates, forming aluminum ions (Al3+) that are immediately hydrated. Through several rapi d hydrolytic reactions an insoluble, gelatinous, poorly crystallin e aluminum hydroxide (Al(OH)3) floc is formed (Ebeling et al., 2003). This floc has high P adsorption prop erties with a surface area greater than 600 m2 g-1 (Huang et al., 2002). The size of the floc formed is directly related to the alum dose (Chakraborti et al., 2003) and can rem ove both soluble and particulate P both by adsorption and physical entrapment (Galarneau and Gehr, 1997). Immo bilization of 1 mg of phosphate (PO4 3-) theoretically requires 0.28 mg of Al3+, however, the alum floc also binds with organic matter whic h is typically abundant in treatment wetlands, reducing its P treatment efficiency and requiring an in creased alum dosage (Van Hullebusch et al., 2002).

PAGE 44

25 The controlling factor in the effectiveness and toxicity of alum is the pH of the system. Alum solution itself has a pH of approximately 2.4 (Beecroft et al., 1995; Lind, 2003), and therefore tends to decrease the pH of the system to which it is added. As long as the pH of the system remains betw een 6 and 8, insoluble polymeric Al(OH)3 will dominate (May et al., 1979) and P inactivation results. However, if the pH decreases to between 4 and 6, soluble intermediates wi ll occur, releasing bound P (Cooke et al., 1993a). Below pH 4 soluble Al3+ dominates which may result in Al toxicity (Cooke et al., 1993b), and at pH 8 or grea ter the aluminate ion (Al(OH)4 -) dominates due to its amphoteric nature, releasing bound P and increasing soluble Al (Cooke et al, 1993a). Aluminate, similar to Al3+, is associated with Al toxicity in plants (Kinraide, 1990, Eleftheriou et al., 1993; Ma et al., 2003). There are also several Al-containi ng alternatives to alum, including polyaluminum chloride (PAC), which wa s developed for water treatment use (Viraraghavan and Wimmer, 1988). Polyaluminum chloride (Aln(OH)mCl(3n-m)) is a partially hydrolyzed aluminum chloride so lution which has recently been found to provide stronger, faster settli ng flocs than alum. In the PA C manufacturing process, the Al becomes further polymerized, therefore partially eliminating the polymerization process that occurs when alum is added to water (Boisvert and Jolicoeur, 1999). Polyaluminum chloride contains preformed, highly charged polymers such as the dimer Al2(OH)2 4+ and the Al13 species AlO4Al12(OH)24(H2O)12 7+ inducing rapid coagulation (Viraraghavan and Wimmer, 1988). This results in higher cation charge, thus increasing its P binding capacity so that lower doses can achieve equivalent treatment efficiency (Jiang and Graham, 1998). Another important bene fit of PAC is that it does not decrease

PAGE 45

26 the pH as much as alum (Lind, 2003) and is effective over a wide r pH range (Jiang and Graham, 1998). Another alum alternative is partially-neu tralized aluminum sulfate (PNAS) also known as polyaluminum hydroxysulfate which was developed by Koether et al. (1993; 1997). Partially-neutralized aluminum sulf ate is a solution formed by adding powdered calcium carbonate to concentrated alum (Beecroft et al., 1995; Koether et al., 1997). It is similar to PAC in that it contains seve ral preformed aluminum polymers including monomers [Al(OH)]2+, small octahedrally coordinate d Al species, and larger Al13 polymers (Koether et al., 1993). The pH of PNAS is 3.3-3.4 depending on whether a concentrated or dilute solution is used (B eecroft et al., 1995) which is less acidic than alum, similar to PAC, and thus preferre d when adding Al to a natural system. A final alternative to alum is alum residual, a solid formed in large quantities as a byproduct in many potable water tr eatment plants that use alum in their treatment process (Butkus et al., 1998). The dried sludge is an Al-based water tr eatment plant residual (WTR) that was originally researched as a soil amendment in agricultural uplands with the negative effect that it was found to adso rb plant available soil P (Bugbee and Frink, 1985). Alum residual has a high P sorption capacity due to the oxides which make up a significant fraction of WTRs (Dayton and Bast a, 2005). Therefore, WTRs are now being applied to soils and poultry litter in an effort to reduce P in surface water runoff (Staats et al., 2004; Silveira et al. 2006). The applica tion of alum residual has minimal risk of environmental pH reduction as compared to the direct addition of sulf ate ions associated with alum (Dao et al., 2001). However, a much larger quantity of Al in the form of Al-

PAGE 46

27 WTR is required to reach the sorption capaci ty of alum (Zvomuya et al., 2006) since intraparticle diffusion is required for P sorption (Makris et al., 2004). There is no published research on the uti lization of Al-containing amendments in a constructed wastewater treatment wetland for reducing P concentrations. Additionally, little research has been done on the effect of alum on th e microbial populations in the sediment of lakes or wetland soils. The rate of microbial activity and structure of the microbial community is largely dependent on environmental factors. Both the size and activity of the microbial pool influences the ability of a wetland to remove nutrients (White and Reddy, 1999; White and Reddy 2003) and other contaminants (White et al., 2006a). Microbes are generally sensitive to soil acidity (Degens et al., 2001) and soluble Al (Robert, 1995). The microbial biomass, therefore, has the potential of being a sensitive indicator of impact to soil nut rient dynamics (Powlson and Jenkinson, 1981) due to alum application. This is due to th e close relationship betw een microbial biomass nutrients and levels of mineralizable nutrien ts available in the soil (Jenkison and Ladd, 1981). An alum study by Connor and Martin (1989) on a shallow New Hampshire lake suggested that the dissolved oxygen concentr ations within the lake may have been affected by suppression in activ ity or reduction in the populati on of microbial organisms, but neither was measured. Bacterial cells ar e colloidal in nature and can therefore be aggregated by the alum floc (Eriksson and Axberg, 1981). This bacterial removal may delay the reestablishment of e ssential nutrient cycling (Buls on et al., 1984). Therefore, the size and activity of the microbial pool needs to be assessed with respect to Al amendments to fully understand effects on nut rient cycling within a treatment wetland.

PAGE 47

28 The hypotheses of this study were that al l Al containing amendments would reduce water column P, with alum alternatives aff ecting the pH less. Secondly, the pH reduction attributed to the different amendments will decrease the microbial biomass pool size and activity, and increase Al availability. The specific objectives of this study were to determine: (i) an effective dose of the Al containing amendments (alum, PAC, alum residual, and PNAS) as determined by the P flux at the soil-water interf ace, (ii) the effect of each Al amendment on soluble Al and microbial biomass and activity of the amended soils and (iii) the ability of each amendment to remove P from the water column. Materials and Methods Site Description The Orlando Easter ly Wetlands (OEW) R eclamation Project located in Orange County, Florida is one of the oldest and largest constructed treatment wetlands in the United States, located east of Orlando in Christmas, FL. The wetland was built in 1986, designed by Post, Buckley, Schuh & Jernigan, Inc. (PBS&J) for the City of Orlando’s Iron Bridge Regional Water Pollution Cont rol Facility (WPCF) which needed an alternative discharge point for its wastewater effluent (Burney et al., 1989). The main goal in designing the system was to use macr ophytes to facilitate nutrient removal for an average daily flow of up to 132,489 m3d-1 of effluent from the Iron Bridge WPCF before discharging into the St. Johns River (Black and Wise, 2003). The 494 ha wetland rests on a 664 ha piece of land located 3.2 km west of the main channel of the St. Johns River (SJR). Histori cally, the land had been part of the riparian wetland adjacent to the SJR, but was drained fo r cattle pasture around th e turn of the last century (Burney et al, 1989). The site ha s a natural topographic gradient of 4.6 m downward from west to east allowing water to flow by gravity through a series of cells

PAGE 48

29 with an average elevation drop across each cell of approximately 1 m (Martinez and Wise, 2003) (Figure 2-1). Water exits the we tland through a weir c ontrol structure and flows into a receiving ditch. From there wa ter can flow directly to the SJR or by sheet flow through Seminole Ranch, a natural marsh adjacent to the OEW owned by the St. Johns River Water Management District (SJR WMD). Cells 1 through 12 and 15 are deep marsh, designed primarily for nutrient rem oval, planted with either cattails (Typha spp.), giant bulrush ( Scirpus californicus ), or a combination of the two. Cells 13, 14, and 16 through 18 consist of a mixed marsh do minated by submergent and emergent macrophytes including Ceratophyllum demersum , Limnobium spongia, Myriophyllum spicatum, Najas guadalupensis, Nuphar luteum , Nymphaea odorata, Pontederia cordata, Figure 2-1. Site map of Orlando Easterly Wetland, Christmas, Florida. Soils collected in cell 10 (star), dominated by Typha spp. Arrows indicate water flow paths. Project Location Created by: Lynette Malecki Dated: 12/30/04

PAGE 49

30 Sagittaria lancifolia, and Sagittaria latifolia . These cells serve as a diverse wildlife habitat while continuing to provide nu trient removal (Martinez and Wise, 2003). The overall average influe nt total phosphorus (TP) concentration from 1988 to 2005 was 0.22 mg L-1, however, annual inflow TP concentrations ranged from 0.02 – 3.30 mg L-1 during the same time period. Since its inception, the OEW has exceeded performance expectations. The TP discharg e permit limit established by the FDEP is 0.2 mg L-1 (Wang et al., 2006). From 1988 to 1995 the average TP discharged was 0.07 mg L-1 (Sees and Turner, 1997), however TP values are considerably higher from December to February in recent years (Wang et al., 2006). Field Sampling, Laboratory Set-Up and Analysis Seventy-eight push cores (7 cm i.d.) were collected within a 10 m2 area from a Typha spp. dominated cell (cell 10) within the OE W (Figure 2-1). Six replicates were collected for each chemical amendment (alu m, PAC, alum residual, PNAS) at three dosage rates (9, 18, and 36 g Al m-2), and six control cores. These dosage rates were selected based on previous alum research performed on OEW soil by Simon (2003) and DB Environmental, Inc. (2004) as well as an extensive review of lake alum application rates (Appendix A). The alum and PAC we re obtained from General Chemical Corporation, the alum residual originated from the Me lbourne, FL Potable Water Treatment Plant, and the PNAS was synthesized in the laboratory according to Koether et al. (1997). Phosphorus flux rates were determined by measuring changes in water column concentrations of intact sediment cores over time (Fisher and Reddy, 2001; Malecki et al., 2004; Steinman et al., 2004). Floodw ater was replaced with filtered (0.45m) site water to maintain a 20-cm water column to init iate the intact core flux study. Cores were

PAGE 50

31 sealed and allowed to equilibrate overnight while water columns were purged with N2 gas (with 300 mg L-1 CO2). Time zero samples were colle cted from the equilibrated cores followed by the addition of treatments to the water column. Within twenty-four hours a visible floc had formed at the soil-water interfac e of those cores treated with alum and PNAS. Cores were purged daily and redox probes were installed in the water column at mid-depth and at 5-cm depth in the soil of two cores selected randomly from each treatment to confirm anaerobic c onditions throughout the study. Cores were incubated in the dark in a wa ter bath maintained at 20 C. Water samples were taken at designated time intervals (0, 1, 3, 5, 7, 10, 14 d) over a two week period, filtered through 0.45m syringe filters, and analyzed for soluble Al, and soluble reactive P (SRP). The water co lumn pH of the system was measured in triplicate cores for each treatment throughout the study. Flux calculations were based on the immediate change in water column con centrations of SRP over time . Flux rates were calculated using the linear portion of the c oncentration versus time curves of SRP over the first day as this was the rate of maximum P sorption. Three soil cores of each treatment dosage ( 39 cores total) were then sectioned into depth increments of 0-5 and 5-10 cm afte r the two-week incubation to determine variations in microbial bioma ss and activity as well as the P content and physicochemical characteristics of the soil. Prior to sectio ning, any visible floc was removed from the soil surface so as to not directly influence th e characterization results. The remaining triplicate cores were spiked once per week with 0.180 mg L-1 P as KH2PO4 (ACS Certified, Fisher Scientific, Fair Lawn, NJ), a concentration equivale nt to that entering the OEW, for three weeks to analyze the P uptake capacity of the floc formed.

PAGE 51

32 The following physicochemical parameters were measured on the sectioned soil samples: pH, bulk density (Blake and Ha rtge, 1986), mass loss on ignition (LOI), microbial biomass P (MBP), soil oxygen demand (SOD), potentially mineralizeable P (PMP), total P (TP), inorganic P fractionati on (Reddy et al., 1998), 1N HCl – extractable metals, and oxalate-extractable Al (McKea gue and Day, 1966). Microbial biomass P was determined by a 24 h chloroform fumigationextraction (CFE) technique (Brookes et al., 1982; Hedley and Stewart, 1992; Ivano ff et al., 1998). Soil oxygen demand was determined by placing 10 g (wet weight) of soil into 250 mL biochemical oxygen demand (BOD) dark bottles and the bottles filled with oxygen saturated distilled, deionized water (APHA, 1992). The initial D.O. concentra tions were recorded using a YSI model 58 dissolved oxygen meter (Yellow Springs Instrument Company, Yellow Springs, OH) equipped with a YSI model 5905 BOD stirring probe. The BOD bottles were incubated in the dark for 24 h at 21 C. Following in cubation, the final D.O. concentrations were measured (Fisher and Reddy, 2001, Malecki et al., 2004). The potentially mineralizeable P (PMP) rate was determined using an anaerobic, waterlogged incubation at 40 C (Chua, 2000). Glass se rum bottles (50 mL) were prepared by weighing out field moist soil (e quivalent of 0.5 g dry weight) and adding 5 mL of distilled, deionized (DDI) water. Bo ttles were capped with butyl rubber stoppers and sealed with aluminum crimps. The h eadspace was evacuated and replaced with O2free N2 gas for incubation in the dark at 40 C for 10 d. A duplicate set of subsamples was weighed into 50 mL centrifuge tubes for the controls. Sample s were shaken and extracted with 25 mL of 1.0 M HCl for 3 h on a reciprocal shaker and supernatant filtered through a 0.45m membrane filter. The supernatant was analyzed for total inorganic P

PAGE 52

33 (TPi) using automated, colorimetric analysis (Method 365.1, USEPA, 1993). Potentially mineralizable P (mg kg-1 d-1) was calculated as the difference in the incubated and Time 0 extractable TPi on a sediment mass basis, divided by the incubation time. Total P analysis involved combustion of 0.5 g oven-dried subsamples at 550 C for 4 h in a muffle furnace followed by dissoluti on of the ash in 6 M HCl on a hot plate (Andersen, 1976). Total P was analyzed using an automated ascorbic acid method (Method 365.4, USEPA, 1993). Ash content wa s calculated to determine mass loss on ignition (LOI), indicating the organic ma tter content in the wetland soil (Lim and Jackson, 1982). Calcium, Mg, Fe, and Al concentrations were determined from ovendried soil treated with 25 mL of 1.0 M HCl and placed on a reciprocal shaker for 3 h. The supernatant was filtered through 0.45m membrane filters and analyzed for Ca, Mg, Al, and Fe (DeBusk et al., 1994; Reddy et al., 1998). Metal analyses were determined by inductively coupled arg on plasma spectrometry ( Vista MPX CCD simultaneous ICP-OES manufactured by Varian , Inc., Walnut Creek, CA; Method 200.7, USEPA, 1993). Statistical Analysis Paired t-tests were used to determin e significant differences (p<0.05) among soil properties in the 0-5 cm and 5-10 cm sect ioned intervals (Microsoft Excel, 2000). Additionally, Pearson product-moment corr elation coefficients (p<0.05) between parameters were calculated (Microsoft Ex cel, 2000). Data normality was determined using the Kolmogorov-Smirnov test (Minitab 13.32, 2000) and data were transformed to fit a normal distribution (Microsoft Ex cel, 2000). One-way ANOVAs and multiple comparisons by Tukey’s W were used on so il variables, while repeated measure ANOVAs followed by multiple comparison were used on water column data to determine significant differences (p<0.05) among tr eatments and dosage rates (Minitab 13.32,

PAGE 53

34 2000). Parameters that could not be transf ormed to follow a normal distribution were analyzed non-parametrically using the Kruskal-Wallis ANOVA on ranks multiple comparisons with Dunn’s test (Minitab 13.32, 2000) . Linear regression analysis was also used (Microsoft Excel, 2000). Results and Discussion Soil Physicochemical Characteristics There were no significant di fferences in bulk density or organic ma tter content among cores. The bulk density of all cores wa s greater in the subsurface 5-10 cm layer (Table 2-2) than in the surface 0-5 cm (Table 2-1) while the opposite was true of the organic matter content as indicated by LOI. Detrital deposition and decomposition at the soil surface resulted in the higher LOI. The increased organic matter in the surface layer allows the soil to remain porous, thereby decreasing the bulk dens ity (Brady and Weil, 1999). There was a significant negative corr elation between bulk density and LOI in the surface layer of cores treated with alum, PAC, and PNAS. The total P concentration was greater in the surface layer (Table 2-1) than subsurface layer (Table 2-2) for all cores, and significantly greater in the control cores, all cores dosed with alum residual, as well as the low dosage alum cores and high dosage PNAS cores. However, there were no signifi cant differences in the total P concentration among treatments or dosage in either layer. Linear regression analysis found TP was directly related to the percent LOI with an R2 of 0.83. Nutrient content typically increases with an increase in organic matter (Farnham and Finney, 1965). Similar to the LOI and TP, the amorphous oxa late-extractable Al was also greater at the surface (Table 2-1) th an in the subsurface layer (Tab le 2-2). Expected in the treated cores due to the surface application of Al, this was also found in the control cores

PAGE 54

35 Table 2-1. Mean soil physicochemical characterization data for the 0-5 cm depth of cores taken from the Orlando Easterly We tland, (n=3) 1 standard deviation. Treatment Dosage Bulk density pH LOI Total P Oxalate Al g Al m-2 g cm-3 pH units % mg kg-1 mg kg-1 Alum 36.0 0.05 0.04 5.41 0.32 71.3 10.91235 177 5344 1779 18.0 0.05 0.04 5.37 0.30 70.0 17.9 1024 378 3188 1296 9.0 0.03 0.01 5.59 0.04 75.8 6.73 1171 131 2514 710.1 PAC 36.0 0.09 0.13 5.37 0.10 50.3 27.5 699 465 2707 2553 18.0 0.03 0.01 5.71 0.17 77.1 6.01 1034 182 2106 183.8 9.0 0.02 0.00 5.69 0.08 75.8 2.09 973 182 1797 187.6 Alum residual 36.0 0.03 0.02 5.93 0.20* 80.5 1.99 1006 77.5 5375 994.6 18.0 0.06 0.02 5.80 0.09 55.9 10.8686 152 3079 1347 9.0 0.02 0.00 5.82 0.04 67.7 8.07 921 216 2416 1108 PNAS 36.0 0.05 0.03 5.27 0.19 68.5 4.47 872 99.9 7733 2926 18.0 0.02 0.01 5.39 0.07 79.8 4.26 1148 119 7116 619.3 9.0 0.03 0.01 5.77 0.11 74.0 5.73 1004 261 3529 769.1 Control 0.0 0.06 0.01 5.75 0.04 66.3 4.04 720 24.5 890.4 322.2* LOI = loss on ignition; PAC = polyaluminum chloride; PNAS = partially-neutralized aluminum sulfate. * Significantly greater than alum, PAC, and PNAS at same dosage at the 0.05 probability level. Table 2-2. Mean soil physicochemical char acterization data for the 5-10 cm depth of cores taken from the Orlando Easterly Wetland, (n=3) 1 standard deviation. Treatment Dosage Bulk density pH LOI Total P Oxalate Al g Al m-2 g cm-3 pH units % mg kg-1 mg kg-1 Alum 36.0 0.27 0.17 5.42 0. 3041.6 16.9485 187 542 198 18.0 0.29 0.14 5.05 0.27* 34.3 15.9333 235 451 208 9.0 0.30 0.19 5.34 0.03 34.5 24.3244 87.9 582 348 PAC 36.0 0.38 0.57 5.48 0.0543. 4 42.1512 559 1496 1837 18.0 0.32 0.21 5.49 0.10 39.4 30.1339 311 493 278 9.0 0.22 0.03 5.33 0.07 62.4 11.5660 156 798 59.0 Alum residual 36.0 0.32 0.09 5.45 0.1437.7 21.3314 239 653 137 18.0 0.43 0.29 5.54 0.16 29.6 21.2288 194 591 468 9.0 0.13 0.02 5.58 0.14 53.6 16.1615 202 952 216 PNAS 36.0 0.30 0.14 5.43 0.18 37.3 0.90367 31 765 73.0 18.0 0.23 0.12 5.56 0.15 58.5 23.4580 276 824 169 9.0 0.19 0.10 5.74 0.12 48.1 23.9484 238 679 233 Control 0.0 0.54 0.34 5.66 0.0916.6 5.68119 71.6 291 230 LOI = loss on ignition; PAC = polyaluminum chloride; PNAS = partially-neutralized aluminum sulfate. * Significantly less than all other treatments at same dosage and control at the 0.05 probability level.

PAGE 55

36 due to the natural mineralogy of the wetland soil and Al associated with organic matter. The surface layer of the cont rol cores (890.4 322.2 mg kg-1) did have significantly less oxalate-extractable Al than the core s treated with alum (3682 1725 mg kg-1), alum residual (3623 1678 mg kg-1), and PNAS (6126 2500 mg kg-1), specifically at the highest Al dosage rate. There was a signi ficant positive correlation between the Al dosage rate and oxalate Al in the surface of both the alum and alum residual treated cores. There were no significant differences in oxalate-extractable Al in the 5-10 cm layer among treatments or dosage indicating the Al hydroxide floc remained at the soil surface. Generally, the soil pH was greater in the surface layer (Table 2-1) than in the subsurface layer (Table 2-2) however, there we re few significant differences. There were several significant differences in soil pH among treatments and dosages. In the surface layer, the alum residual treate d cores (5.9 0.1) had significa ntly greater soil pH values than both the alum (5.5 0.2) and PNAS (5.5 0.3) treated cores. At the highest dosage specifically, the alum residual treated cores ha d significantly greater pH values than the alum, PAC, and PNAS treated co res. Within treatment compar isons found that the pH of soils treated with 36 g Al m-2 PNAS were significantly lower than the pH values of cores that received only 9 g Al m-2 PNAS, while in all other treatments there were no significant differences. There was however a significant negative correlation between pH and dosage rate found for both the PNAS and PA C treated cores. As expected, the higher Al loading rates yielded lower soil pH values. In the 5-10 cm layer (Table 2-2) the alum treated cores (5.3 0.3) had significantly lower pH values than the alum residual tr eated cores (5.5 0.1) , PNAS (5.6 0.2), and

PAGE 56

37 control (5.7 0.1) cores. At the 18 g Al m-2 dosage specifically, the alum cores had significantly lower pH values than the alum residual, PNAS, and control cores of the same dosage. There were, however, no signifi cant differences in soil pH among dosage rates within any treatment. Soil Microbial Characteristics Microbial biomass P (MBP) tended to decrea se with depth (Table 2-3, T able 2-4). The MBP was significantly greater in the su rface than in the subs urface layer of the control, 18 g Al m-2 PAC and PNAS, as well as the 9 g Al m-2 alum residual cores where recently deposited bioavailable nutrients are more easily accessible. There were no significant differences in MBP among treatment types, nor within treatment by dosage. Grouping all treatments, the surface soil of 9 g Al m-2 dosed cores had significantly greater MBP concentrations th an both the 18 and 36 g Al m-2. Pearson product correlation analysis found a significant nega tive correlation between Al dosage and MBP for the alum residual and PNAS treated co res. No trend was found in the 5-10 cm subsurface layer suggesting the effect of Al am endments in the short term only negatively impacts the microbial biomass in the surface soil. Microbial activity, as indicated by PMP and SOD rates, was alsosignificantly greater in the surface than subsurface layer (Table 2-3, Table 2-4). Potentially mineralizable P rates were signif icantly greater in the 18 g Al m-2 PNAS treated cores corresponding to the significantly higher microbi al biomass in the surface layer. Similar to MBP, there were no significant differences in PMP among treatment types, nor within treatment by dosage, but when all treatments were grouped, the surface soil of 9 g Al m-2 dosed cores had significantly great er PMP rates than the 36 g Al m-2. Pearson product

PAGE 57

38 Table 2-3. Mean soil microbial characterization data for the 0-5 cm depth of cores taken from the Orlando Easterly Wetland, (n=3) 1 standard deviation. Treatment Dosage MBP, NS PMP, NS SOD g Al m-2 mg kg-1 mg kg-1 d-1 mg kg-1 h-1 Alum 36.0 79.8 50.5 7.48 4.71 66.6 24.3 18.0 93.8 56.7 11.9 22.4 42.7 25.3 9.0 139 57.3 37.1 38.4 74.5 20.2 PAC 36.0 115 89.4 12.2 9.91 66.4 52.0 18.0 128 39.3 8.44 7.38 77.1 13.1 9.0 143 27.7 11.1 21.5 98.7 9.60 Alum residual 36.0 67.5 38.6 8.65 10.4 155 31.8 18.0 63.9 18.5 12.3 10.5 92.1 51.2 9.0 178 27.8 21.5 3.08 175 49.4 PNAS 36.0 42.9 63.1 1.87 8.21 99.0 46.5 18.0 88.5 19.5 16.6 3.61 184 62.4 9.0 137 64.7 18.8 10.1 164 32.6 Control 0.0 70.0 8.28 6.18 4.48 72.9 18.7 NS = No significant differences among mean values by dosage or treatment. PAC = polyaluminum chloride; PNAS = partially-neutralized aluminum sulfate. MBP = microbial biomass P; PMP = potentially mineralizable phosphorus; SOD = sediment oxygen demand. Table 2-4. Mean soil microbial characterization data for the 5-10 cm depth of cores taken from the Orlando Easterly Wetland, (n=3) 1 standard deviation. Treatment Dosage MBP, NS PMP, NS SOD, NS g Al m-2 mg kg-1 mg kg-1 d-1 mg kg-1 h-1 Alum 36.0 43.0 10.9 1.79 1.53 20.2 17.6 18.0 103 53.6 0.65 2.32 12.8 4.22 9.0 47.3 35.1 3.13 2.54 18.0 16.2 PAC 36.0 79.4 62.7 9.76 7.98 33.3 31.1 18.0 30.2 33.1 16.7 25.3 13.7 9.23 9.0 83.8 65.5 5.13 2.87 16.1 3.98 Alum residual 36.0 36.0 11.5 2.63 1.52 8.70 4.42 18.0 35.0 21.9 0.94 0.45 12.2 8.79 9.0 52.4 20.3 12.6 7.90 13.2 3.80 PNAS 36.0 40.8 2.96 5.43 5.42 10.0 0.25 18.0 34.9 16.6 3.36 2.89 20.7 13.6 9.0 62.2 33.8 18.0 24.8 8.49 1.26 Control 0.0 23.9 8.33 2.15 2.45 3.85 1.07 NS = No significant differences among mean values within treatment by dosage, or among treatments. PAC = polyaluminum chloride; PNAS = partially-neutralized aluminum sulfate. MBP = microbial biomass P; PMP = potentially mineralizable phosphorus; SOD = sediment oxygen demand.

PAGE 58

39 correlation analysis indicated a strong negative correlation between the Al dosage rate and surface layer PMP for the alum, alum residual, and PNAS treated cores. This corresponds to the same trend found in the size of the microbial pool. Soil oxygen demand rates were significantly higher in the surface than subsurface (Table 2-3, Table 2-4). There were no significant differences within treatment by dosage, but there were differences among treatments. In the surface layer, the alum and PAC treated cores had significantly lower SOD rate s than the alum residual and PNAS treated cores, while in the subsurface layer there we re no significant differences similar to the MBP and PMP results. Pearson product correlation analysis indica ted a strong negative correlation between Al dosage rate and surface layer SOD rates for the PAC and PNAS treated cores. This trend once again did not pe rsist in the subsurface layer indicating both microbial biomass and activity in the 5-10 cm layer remained unaffected by the Al amendments with impacts only apparent in the 0-5 cm layer for this short term experiment. Soil Phosphorus Forms Organic P was dominant in all co res, averaging 70% organic P (Po) (Table 2-5) and 30% inorganic P (Pi) (Table 2-6) in the surface layer of all treated cores, and 78% Po, 22% Pi in the surface layer of control cores. All P fractions, with the exception of HClextractable P, were significantly greater in the surface layer than in the subsurface, similar to the TP values previously discussed. The overall decrease in P storage capacity with depth may be attributed to the significan t decrease in all metal concentrations (Al, Ca, Fe, Mg) with depth (Table 2-7, Table 2-10). In the surface laye r, the KCl–extractable P consisting of labile, readily bioavailable P made up the smallest portion (0.3 4%) of the total P pool, generally incr easing with a decrease in Al dosage rate although there

PAGE 59

40 Table 2-5. Mean organic phosphorus derive d from inorganic phos phorus fractionation data for the 0-5 cm depth interval of cores taken from the Orlando Easterly Wetland, (n=3) 1 standard deviation. Treatment Dosage NaOH Po Residue Po Total Po g Al m-2 mg kg-1 mg kg-1 mg kg-1 Alum 36.0 322 257 414 179 737 434 18.0 207 152 270 41.3 477 166 9.0 283 51.0 349 40.1 632 10.9 PAC 36.0 293 200 334 110 628 305 18.0 271 68.3 299 31.5 570 66.8 9.0 247 20.5 342 30.6 590 50.9 Alum residual 36.0 284 41.4 418 195 702 227 18.0 169 40.3 247 69.5 416 109 9.0 373 35.4 434 100 807 132 PNAS 36.0 295 127 346 117 640 145 18.0 378 52.5 396 153 774 199 9.0 299 83.7 374 90.3 673 168 Control 0 150 18.6 333 33.9 517 96.1 PAC = polyaluminum chloride; P NAS = partially-neutralized aluminum sulfate. Table 2-6. Mean inorganic phos phorus fractionation data for th e 0-5 cm depth interval of cores taken from the Orlando Easterly Wetland, (n=3) 1 standard deviation. Treatment Dosage KCl Pi* NaOH Pi HCl Pi Total Pi g Al m-2 mg kg-1 mg kg-1 mg kg-1 mg kg-1 Alum 36.0 3.31 1.75 259 206 249 136 511 77.3 18.0 4.73 6.97 138 137 75.2 25.5 218 152 9.0 17.1 12.6 135 26.8 83.7 24.1 235 34.4 PAC 36.0 14.5 12.5 144 106 53.2 27.1 211 146 18.0 15.3 7.97 140 14.3 76.2 19.5 231 27.5 9.0 27.6 15.1 97.7 11.3 63.1 15.5 188 39.3 Alum residual 36.0 10.3 5.77 145 15.1 69.4 13.4 225 26.6 18.0 10.3 8.60 88.3 11.6 67.2 3.61 166 16.3 9.0 35.8 19.0 221 96.0 90.5 14.0 348 64.0 PNAS 36.0 2.31 1.34 195 97.6 57.4 21.6 254 98.7 18.0 3.16 0.95 262 60.7 193 199 458 228 9.0 7.55 3.85 185 68.0 104 35.7 297 107 Control 0 11.6 3.47 57.3 10.9 77.2 24.2 146 35.2 PAC = polyaluminum chloride; P NAS = partially-neutralized aluminum sulfate. * Significantly less than all other P fractions at the 0.05 probability level.

PAGE 60

41 were no significant differences. The HClextractable Ca and Mg bound P comprised 620% of the total P, while as expected, the NaOH-extractable reactive Al and Fe bound P was the dominant Pi fraction in the Al treated cores, making up 13-22% of the total P pool while only accounting for 9% of the total P in the control cores. Table 2-7 shows the 1 M HCl-extractable metals for the surface (0-5 cm) soils. There were no significant differences within treatment by dosage. In all treated cores there were significantly greater Ca concentrations than Mg concentr ations, indicating most of the P from the P fractionation was associated with Ca from the buffered wetland soil, rather than Mg. In the PAC and PNAS treated cores Ca concentrat ions were also signi ficantly greater than Fe concentrations. Miner (2001) concl uded there was no significant P retention associated with Fe in the OEW, instead pre dominantly affected by Al concentrations. There were no significant diffe rences in metals in the control cores, however, the control cores did have significantly lower Al concentrations than all other treated cores Table 2-7. Mean 1M HCl-extractable metals fo r the 0-5 cm depth interval of cores taken from the Orlando Easterly Wetland, (n=3) 1 standard deviation. Treatment Dosage Al Ca * Fe Mg g Al m-2 mg kg-1 mg kg-1 mg kg-1 mg kg-1 Alum 36.0 4960 2828 21368 1181 1694 153.9 651.7 96.12 18.0 4401 2803 21326 5402 1704 504.0 657.3 251.9 9.0 3005 581.3 23479 481.3 1766 152.6 839.6 36.15 PAC 36.0 2759 1741 18319 8468 1440 730.1 799.3 94.53 18.0 3062 680.6 24586 1629 1786 19.65 876.2 152.3 9.0 1684 400.2 21264 1381 1618 165.7 822.2 71.72 Alum residual 36.0 5239 2119 22847 3577 1889 366.7 918.9 181.6 18.0 2727 227.9 20923 5954 1662 449.6 721.6 249.4 9.0 2161 850.1 25143 710.6 2072 696.7 964.1 121.7 PNAS 36.0 11261 8328 20656 5075 1779 178.0 583.8 166.9 18.0 11267 1400 23949 1466 1739 303.9 722.1 18.08 9.0 4804 864.2 25120 600.4 1800 241.2 789.6 30.71 Control 0.0 693.0 233.1* 17953 3990 1577 129.4 640.3 146.5 PAC = polyaluminum chloride; PNAS = partially-neutralized aluminum sulfate. * Significantly greater than Al, Fe, and Mg c oncentrations at the 0.05 probability level.

PAGE 61

42 Table 2-8. Mean organic phosphorus derive d from inorganic phos phorus fractionation data for the 5-10 cm depth interval of cores taken from the Orlando Easterly Wetland, (n=3) 1 standard deviation. Treatment Dosage NaOH Po Residue Po Total Po g Al m-2 mg kg-1 mg kg-1 mg kg-1 Alum 36.0 52.3 32.6 254 111 211 184 18.0 46.6 14.2 166 57.6 213 54.7 9.0 77.2 59.3 274 313 351 371 PAC 36.0 139 137 252 234 391 371 18.0 58.0 37.0 209 186 267 221 9.0 77.4 7.53 271 114 348 110 Alum residual 36.0 68.6 21.1 198 101 266 120 18.0 45.8 37.5 161 223 207 259 9.0 92.8 26.7 234 79.7 327 81.7 PNAS 36.0 52.6 13.3 215 76.6 268 89.9 18.0 74.5 36.5 313 333 387 369 9.0 84.0 45.1 151 81.7 235 38.5 Control 0 32.4 19.5 102 77.8 134 96.9 PAC = polyaluminum chloride; PNAS = partially-neutralized aluminum sulfate. Table 2-9. Mean inorganic phos phorus fractionation data for the 5-10 cm depth interval of cores taken from the Orlando Easterly Wetland, (n=3) 1 standard deviation. Treatment Dosage KCl Pi NaOH Pi HCl Pi Total Pi g Al m-2 mg kg-1 mg kg-1 mg kg-1 mg kg-1 Alum 36.0 2.36 0.82 19.6 11.5 96.3 65.1 394 436 18.0 7.33 10.1 19.5 11.2 75.6 61.3 102 82.4 9.0 2.98 2.34 27.8 27.7 157 138 188 140 PAC 36.0 9.92 8.50 74.7 89.4 63.3 65.2 148 159 18.0 2.88 2.21 22.3 17.1 44.2 30.3 69.4 41.7 9.0 2.94 1.96 34.2 11.0 186 223 223 232 Alum residual 36.0 1.05 0.33 26.7 7.09 29.3 9.30 57.0 14.5 18.0 2.05 1.52 17.5 19.7 60.3 39.6 81.7 54.6 9.0 3.74 0.68 42.1 9.15 96.9 21.7 143 28.4 PNAS 36.0 1.54 1.43 17.8 5.36 39.6 25.2 58.9 31.3 18.0 2.61 2.42 29.6 17.7 51.2 30.8 83.4 50.5 9.0 9.67 7.33 36.5 27.2 109 69.9 155 93.6 Control 0 1.44 0.99 10.4 6.62 7.58 4.16 19.4 11.7 PAC = polyaluminum chloride; PNAS = partially-neutralized aluminum sulfate.

PAGE 62

43 corresponding to the low NaOH Pi. Generally Mg made up 3% of the HCl-extractable metals in the surface of Al treated cores, Fe 6%, Al 9-26% and Ca dominated making up 67-82% of the extracted metals. The organic P fractions consisted of Na OH-extractable non-reactive P associated with humic and fulvic acids as well as b acteria incorporated P which accounted for 2435% of the total P in the surface laye r of all cores, while the residue Po representing the refractory organic P and any othe r inert mineral P fractions no t extracted with salt, acid, or base composed 32-45% of the total P in tr eated cores and 53% in the control cores. Both of these Po fractions were significantly grea ter than the Ca and Mg bound fraction of all four Al treatment types, while the residue Po was significantly greater than the NaOH Pi in the PAC, alum resi dual, and control cores. The subsurface layer was once again dominated by organic P averaging 69% Po (Table 2-8) and 31% Pi (Table 2-9) for treated cores and 87% Po and 13% Pi in the control cores. Labile P repr esented 0.3-2% of the total P, followed by the Al and Fe bound P which only made up 5-14% of the subs urface P pool indicating once again that the Al treatments remained at the surface. The dominant inorganic P form in the subsurface layer was the Ca and Mg bound P which ranged from 5-33% of the total P pool. Table 2-10 shows the 1M HCl-extractabl e metal concentrations in the subsurface layer. The Ca concentrations were significantly higher than the Al, Fe, and Mg concentrations for all treated and control co res. Generally Ca made up 88% of the HClextractable metals within the 5-10 cm layer of all cores, Al 5%, Fe 4%, and Mg 3%. There were no significant diffe rences in extractable metals within treatment by dosage.

PAGE 63

44 The organic acid and bacterial P made up 1226% of the total P while the residue Po was the dominant subsurface P fraction in all cores composing 39-66% of the total P. Table 2-10. Mean 1M HCl-extractable metals for the 5-10 cm depth interval of cores taken from the Orlando Easterly We tland, (n=3) 1 standard deviation. Treatment Dosage Al Ca* Fe Mg g Al m-2 mg kg-1 mg kg-1 mg kg-1 mg kg-1 Alum 36.0 473.1 156.2 11379 3914 896.2 431.2 353.8 163.3 18.0 446.4 158.6 10611 3927 593.0 82.32 296.2 68.89 9.0 508.4 251.1 9319 5120 736.9 383.8 313.4 153.0 PAC 36.0 1234 1482 11040 9848 870.8 798.3 446.9 332.2 18.0 455.3 250.2 9276 7789 240.7 167.5 349.7 267.1 9.0 688.7 24.20 15752 2374 497.3 179.0 512.4 65.02 Alum residual 36.0 579.5 120.5 9385 6418 315.4 140.8 356.6 170.9 18.0 495.7 405.2 8244 6440 309.7 244.6 274.2 203.4 9.0 755.7 159.5 14951 5212 503.5 232.1 474.3 83.91 PNAS 36.0 589.2 43.64 10298 1072 312.4 97.05 358.4 38.02 18.0 640.7 154.2 14603 5534 420.7 98.63 515.3 221.5 9.0 672.4 419.4 12675 4617 565.7 459.0 498.8 194.1 Control 0.0 266.7 205.0 3571 2485 122.1 33.74 174.4 90.51 PAC = polyaluminum chloride; PNAS = partially-neutralized aluminum sulfate. * Significant at the 0.05 probability level. Water Column Results Soluble Reactive Phosphorus – Incubation Study The initial Al application rapidly reduced the soluble reactive P (SRP) concentrations in the anaerobic water colu mn of cores treated with alum, PAC, and PNAS, stabilizing by day three (Figure 2-2). Th e SRP concentration in cores treated with alum residual gradually decreased over the two week time period, while concentrations gradually increased over time in the water colu mn of control cores. Repeated measures analysis of log transformed data determined that all four treatments resulted in significantly lower water column SRP concentrations than the control at all three

PAGE 64

45 0.00 0.15 0.30 0.45 02468101214 Time (d)SRP (mg L-1) Control Alum residual PAC Alum PNAS (a) 0.00 0.15 0.30 0.45 02468101214 Time (d)SRP (mg L-1) Control Alum residual PAC Alum PNAS (b) 0.00 0.15 0.30 0.45 02468101214 Time (d)SRP (mg L-1) Control Alum residual PAC Alum PNAS (c) Figure 2-2. Changes in soluble reactive phos phorus concentration in the water column under anaerobic conditions at treatm ent dosage (a) 36.0, (b) 18.0, and (c) 9.0 mg Al m-2 for soil cores from the Orlando Easterly Wetland (n=6).

PAGE 65

46 dosages. Of the treatments, the alum and PNAS were significantly more effective at sequestering P than the PAC, and the alum residual was least effective. The alum residual was a solid amendment, however, wh ile the alum, PNAS, and PAC were liquid and thus able to sorb P more readily in the short term than sorption to a solid phase with reduced surface area. There is an apparent diffusion limitation on the P adsorption rate because some of the Al in alum residual is already occluded with respect to P access. Nevertheless, the solid may prove more effec tive at a longer time s cale as the residual breaks down, increasing the surface area availa ble to react and in turn increasing P sorption (Dayton and Basta, 2005). Additionally, there was no significant difference in water column SRP with dosage rate in the alum residual treat ed cores, averaging 0.17 0.06 mg L-1 over the two week time period. For the cores treated with alum, PNAS, and PAC, however, the SRP concentrations were significantl y lower in the 36 and 18 mg Al m-2 treated cores than in the 9 mg Al m-2. These differences are apparent wh en data are log transformed, thus from a management and cost benefit perspe ctive, the lowest dosage rate would be appropriate. The Kruskal-Wallis test was used to determine effects of treatments and dosages on the SRP flux rates (Table 2-11). The SRP uptake rates of all treatments (-78.9 mg m-2 d-1 to -2.75 mg m-2 d-1) were equal or greater than the mean release rate of the controls (2.96 mg m-2 d-1) suggesting once again that any applic ation rate might provide effective treatment of soil-released P in the short term. The net negative flux rates in the treated cores indicate that the Al hydroxide floc not only removed P from the water column but also intercepted any P fluxing out of the soil, as indicated by the control cores. There

PAGE 66

47 were no significant differences in P flux rates among dosage levels for any of the treatments. At all dosage rates, however, th e alum, PNAS, and PAC were more effective at sequestering P than the alum residual. Table 2-11. Mean soluble reactive phosphorus flux from constructed wetland soil under anaerobic water column conditions , (n=6) 1 standard deviation. Treatment Dosage g Al m-2 Average P Flux mg m-2 d-1 Alum 36.0 -78.9 33.0 18.0 -69.0 30.6 9.0 -69.6 19.0 PAC 36.0 -68.5 19.1 18.0 -75.7 22.2 9.0 -64.2 14.0 Alum residual 36.0 -2.75 0.93 18.0 -3.42 0.54 9.0 -3.61 4.04 PNAS 36.0 -71.8 3.47 18.0 -62.0 14.7 9.0 -66.0 10.4 Control* 0.00 2.96 1.00 PAC = polyaluminum chloride; PNAS = partially-neutralized aluminum sulfate. * Significantly greater than all other flux rates at the 0.05 probability level. Soluble Reactive Phosphorus – Re-Spiking Study After the addition of P at time 0 (day 15 of core incubation study), 8, and 16 days soluble reactive P concen trati ons rapidly decreased in the wa ter column of cores treated with alum, PNAS, and PAC, stabilizing by the second day at all dosage rates. In the alum residual treated cores, the SRP concen tration continued to decrease over the seven day time period, never fully binding the added P (Figure 2-3). Interestingly the control cores also showed some capacity for initial P retention although continuing to release P during the prior two-week inc ubation study. This indicates that when P levels remain high in the water column, the soil does still have the capacity to act as a sink for P.

PAGE 67

48 0.00 0.15 0.30 0.45 0.60 081624 Time (d)SRP (mg L-1) Control Alum residual PAC Alum PNAS (a) 0.00 0.15 0.30 0.45 0.60 081624 Time (d)SRP (mg L-1) Control Alum residual PAC Alum PNAS (b) 0.00 0.15 0.30 0.45 0.60 081624 Time (d)SRP (mg L-1) Control Alum residual PAC Alum PNAS (c) Figure 2-3. Changes in soluble reactive phos phorus concentration in the water column during weekly P spiking at treatment dosage (a) 36.0, (b) 18.0, and (c) 9.0 mg Al m-2 for soil cores from the Orlando Easterly Wetland (n=3).

PAGE 68

49 However, as treatment efficiency of the ve getation and algae increases in the summer maintaining a low water column SRP concentrati on, the soil may then act as a P source. Figure 2-3 also shows that cores treated with alum residual did have a significantly greater sorption capacity than control cores over time. Additionally, the sorption capacity of alum, PNAS, and PAC were significantly greater than that of alum residual at all dosage levels over time, wh ile alum and PNAS had greater sorption capacities than PAC at the 36 and 18 mg Al m-2 dosage level, similar to the results of the incubation study. At the lowest Al dosa ge, there was a clear separation among all treatment types with PNAS having a superior P uptake capacity and nearly full recovery to baseline SRP concentrations ev en after three spiking events. Water Column pH The pH tended to remain rela tively stable in the anaerobic water columns over the two week time period (Figure 2-4). The alum treated cores always maintained the lowest water column pH values which were significan tly lower than all other treatments at the 36 and 18 g Al m-2 dosage rate averaging 2.6 0.2 and 3.2 0.3 respectively (Table 212). The pH of alum treated cores was also significantly lower than all treatments but PNAS in the 9 g Al m-2 dosage rate, averaging 5.1 0.3. The PNAS treated cores had the next most acidic water column with signifi cantly lower values than that of the PAC, alum residual, and control cores at all three Al dosage levels averaging 4.1 0.9. Finally, the PAC treated cores had significantly lower values than the alum residual and control cores at all three Al dosage levels, averag ing 4.9 1.0. The alum residual and control cores maintained the highest pH values at or above 6.0 for the study duration. The pH values of the water column of cores treated with alum, PNAS, and PAC all fell below 6.0 regardless of Al dosage rate. Additionally, for all treatments th e cores receiving the

PAGE 69

50 0.00 2.00 4.00 6.00 8.00 02468101214 Time (d)pH Control Alum residual PAC Alum PNAS (a) 0.00 2.00 4.00 6.00 8.00 02468101214 Time (d)pH Control Alum residual PAC Alum PNAS (b) 0.00 2.00 4.00 6.00 8.00 02468101214 Time (d)pH Control Alum residual PAC Alum PNAS (c) Figure 2-4. Changes in water column pH unde r anaerobic conditions at treatment dosage (a) 36.0, (b) 18.0, and (c) 9.0 mg Al m-2 for soil cores from the Orlando Easterly Wetland (n=3).

PAGE 70

51 Table 2-12. Mean water column data collected over 14 days in soil cores incubated under anaerobic conditions, (n=18) 1 standard deviation. Treatment Dosage Average pH Average soluble Al g Al m-2 pH unit mg L-1 Alum 36.0 2.63 0.19 31.4 4.36 18.0 3.19 0.29 6.12 2.77 9.0 5.14 0.31 0.15 0.01 PAC 36.0 3.73 0.71 2.55 1.22 18.0 5.26 0.34 0.17 0.02 9.0 5.57 0.42 0.16 0.02 Alum residual 36.0 5. 96 0.15 0.13 0.01 18.0 6.26 0.13 0.12 0.00 9.0 6.15 0.15 0.12 0.00 PNAS 36.0 3.14 0.13 25.9 4.79 18.0 4.30 0.61 1.88 1.91 9.0 5.20 0.25 0.16 0.03 Control 0 6.09 0.25 0.10 0.01 PAC = polyaluminum chloride; PNAS = partially-neutralized aluminum sulfate. highest Al dosage had significantly lower wate r column pH values than the mid and lowlevel Al dosages. The pH values during this study were extr emely low which presents two potential problems. First, there is th e possibility of aluminum toxic ity when pH values are below 4.0 and second is the remobiliz ation of the Ca-bound soil P. However, this study was done in intact cores, while in a flow thr ough system as in a wetland, the pH would not have likely remained low for a long period. This research also corroborates the assertion that the primary benefit of utilizing alterna tives to alum such as PNAS and PAC, as indicated by Beecroft et al. (1995) and Jia ng and Graham (1998), is that they do not decrease the water column pH as severely due to their polymerized composition and therefore may cause less environmental impact.

PAGE 71

52 Water Column Soluble Aluminum Soluble or dissolved Al concentrations ra pidl y peaked in the water column after initially dosing the cores, and then slowly d ecreased over the next te n days (Figure 2-5). Due to the close relationship between pH a nd Al speciation, the low pH of the water column in cores treated with alum, PNAS, and PAC resulted in a re latively high release of dissolved Al into the wa ter column initially. Overall, alum (12.6 2.38 mg Al L-1) and PNAS (9.32 2.24 mg Al L-1) treated cores had significantly higher water column soluble Al concentrations th an the PAC (0.96 0.42 mg Al L-1), alum residual (0.12 0.00 mg Al L-1), and control cores (0.10 0.01 mg Al L-1) over the course of the experiment. The Al concentrations in th e alum, PNAS, and PAC treated cores were significantly higher at the highest Al dosage rate than at the lower dosage rates, as would be expected. Regression analysis determined a strong linear relationship between the average soluble Al concentration and dosage for both alum (R2 = 0.98) and alum residual (R2 = 0.99) treated cores and strong logari thmic relationship for the PAC (R2 = 0.92) and PNAS (R2 = 0.97) treated cores. Due to the low pH values (< 6) and high Al concentrations, there is potential for aluminum toxicity to occur within the wetland water column affecting both fish and benthos. However, the pH effect does not appear to be of concern in mobilizing the Ca-bound P pool since the pH in the soil was not affected as dramatically as the water column.

PAGE 72

53 0.0 20.0 40.0 60.0 80.0 0 51015 Time (d)Al (mg L-1) Alum PNAS PAC Alum residual Control (a) 0.0 5.0 10.0 15.0 20.0 02468101214 Time (d)Al (mg L-1) Alum PNAS PAC Alum residual Control (b) 0.0 0.5 1.0 1.5 2.0 051 01 5 Time (d)Al (mg L-1) Alum PNAS PAC Alum residual Control (c) Figure 2-5. Changes in water column solubl e aluminum concentra tion under anaerobic conditions at treatment dosage (a ) 36.0, (b) 18.0, and (c) 9.0 mg Al m-2 for soil cores from the Orlando Easterly Wetland (n=6).

PAGE 73

54 Conclusions Alum , as well as the three tested alterna tives to alum (PAC and PNAS to a greater extent than the alum residual) are all effective at short te rm P sequestration in wetland soils. However, both the surface soil and the water column characteristics show reason for concern in utilizing such treatments. Du e to decreased soil pH, as well as low water column pH values and high soluble Al concen trations, the minimum dosage necessary for effective treatment should be used in treatm ent wetlands. Additionally, the spiking of P into the water column to mimic loading to the wetland implied that while the treatments remained effective at sequestering P in the short term, the long term efficacy of a one time treatment may be short lived when utilizi ng alum or its alternatives in a treatment wetland. Instead, a continuous slow-drip injection system at low dose may prove to be a more effective management strategy. More research is needed to determine the long term impacts to the microbial community as well as potential population sh ifts. In the short term, Al amendments negatively impact both microbial biomass a nd activity in the surface soil, however no trends were found in the subsurface soil layer. As the floc layer of Al becomes buried through detrital deposition and sedimentation it may in turn influence the subsurface microbial pool. Changes in soil and water colu mn characteristics may also result in Al toxicity for aquatic macrophytes present in treatment wetlands. Finally, the long term efficacy of P sequestration by Al amendments in treatment wetlands needs to be determined. Treatment wetland management generally re quires monitoring the outflow of the wetland to meet specific discharge criteria. During the winter months, wetlands in the southern United States may become less eff ective at treating P as plants senesce and

PAGE 74

55 microbial activity slows. Application of alum or Al-c ontaining amendments to soil proximal to the outflow regions of the we tland may provide an effective management tool to maintain discharge concentrations w ithin permitted values during these inefficient wetland treatment times.

PAGE 75

56 CHAPTER 3 INFLUENC E OF ALUM ON WATER QUALITY AND MACROPHYTE GROWTH: A MESOCOSM STUDY Introductio n Alum (Al2(SO4)3H2O) is the chemical amendment used most often for phosphorus (P) inactivation in lakes and coagulation in th e water treatment industry. When added to water, alum dissociates, forming aluminum ions (Al3+) that are immediately hydrated. An insoluble, gela tinous, poorly crystalline aluminum hydroxide (Al(OH)3) floc is formed through several rapid hydr olytic reactions (Ebeling et al., 2003). This floc has high P adsorption properties with a surface area greater than 600 m2 g-1 (Huang et al., 2002). The size of the floc form ed is directly rela ted to the alum dose (Chakraborti et al., 2003) and can remove both soluble and particulate P both by adsorption and physical entrapment (Galarneau and Gehr, 1997). The controlling factor for effectiveness a nd toxicity of alum is the pH of the system. Alum solution itself has a pH of approximately 2.4 (Beecroft et al., 1995; Lind, 2003) and therefore tends to decrease the pH of the system to which it is added. As long as the pH of the system remains betw een 6 and 8, insoluble polymeric Al(OH)3 will dominate (May et al., 1979) and P inactivati on results. However, if pH decreases to between 4 and 6, soluble intermediates wi ll occur, releasing bound P (Cooke et al., 1993a). Below pH 4 soluble Al3+ dominates which may result in Al toxicity (Cooke et al., 1993b), and at pH 8 or grea ter the aluminate ion (Al(OH)4 -) dominates due to its amphoteric nature, releasing bound P and increasing soluble Al (Cooke et al, 1993a).

PAGE 76

57 Aluminate, similar to Al3+, is associated with Al toxicity in plants (Kinraide, 1990, Eleftheriou et al., 1993; Ma et al., 2003). While alum has been used for P inact ivation in eutrophic lakes since 1968 (Blomsquist et al., 1971) there has been little research done on its potential effectiveness in aging treatment wetlands with reduced P sorption capacities (Simon, 2003; DB Environmental, Inc., 2004; Malecki-Brown a nd White, 2007a). Additi onally, there is not a clear comprehension of the impact of incr eased aluminum (Al) concentrations on the aquatic macrophytes in alum-treated ecosystems. The speciation of Al determines its mobility, bioavailability, and toxicity in aquatic ecosystems (Bertsch, 1990). In the case of plants, Al(H2O)6 3+ (or simply referred to as Al3+), Al(OH)2+, and some complex polymers such as Al13O4(OH)24(H2O)12 7+ are the most toxic (Kinraide, 1991; Kochian, 1995), wh ile Al that complexes with organic anions is relatively nonphytotoxic (Wickstrm et al ., 2000; Jansen et al., 2002). Flow paths, residence times, temperature, and organic complexations can all affect aluminum solubility, thus seasonal va riations in Al speciation may also be observed (Van Benschoten and Edzwald, 1990; Easthouse, 1993). The uptake and retention of bioavailable metals in aquatic macrophytes is largely determined by plant physiology and genotypic differences among plants which control their ability to accumulate th e available forms (Guilizzoni, 1991; Outridge and Noller, 1991; Jackson, 1998). Submerged aquatic vege tation (SAV) has a reduced root system, absorbing a large portio n of soluble nutrients and metals directly from the water column into shoots, while rooted floating-leaved and emergent plants tend to obtain most metals from the soil (Crowder, 1991; Rai et al ., 1995; Sparling and Lowe , 1998; Thiebaut and

PAGE 77

58 Muller, 2000). Submerged plants may uptake more metals than emergents due to a higher surface/biomass ratio (Baudo et al., 1981; Guilizzoni, 1991; Albers and Camardese, 1993; Rai et al., 1995; Cardwell et al., 2002). Additionally, emergent species with shallow root systems tend to absorb higher metal concentrations than deeply rooted plants due to greater exposu re at the sediment-water in terface (Mudroch and Capobianco, 1978; Guilizzoni, 1991). The uptake of bioavailable metals gene rally occurs via surface adsorption or absorption, followed by passive and activ e transport across cell membranes, incorporating metals into pl ant biochemical functions or storing metals in a bound form (Outridge and Noller, 1991; Ra i et al., 1995). The pH of wetland soils is near neutral (6.5-7.5) typically favoring metal immobiliza tion (Gambrell, 1994). However, several studies have shown that wetland acidification can increase the bioavailability of metals, resulting in elevated concentrations in aquatic plants (Lehtonen, 1989; Albers and Camardese, 1993; Jackson et al., 1993; V azques et al., 2000; Cardwell et al., 2002; Gallon et al., 2004). It is important to determine how alum treatment will affect the bioavailability of metals and nutrients for aquatic plants. Ava ilable P is an essential nutrient for aquatic vegetation growth and the alum floc whic h settles to the soil surface may not only deprive plants of P, but Al may have adve rse effects on plants as well, because Al concentrations are typically very low (0.02% of total weight) (Hutchinson, 1945). Aluminum toxicity in plants is related to the activity of the Al3+ ion which often inhibits root elongation and respiration in upland plan ts resulting in roots that are thickened, stubby, brittle, and often ineffi cient in nutrient absorption (Schier, 1985; Jarvis and

PAGE 78

59 Hatch, 1986; Jansen et al. 2002). However, few studies have looked at the possibility of Al toxicity in aquatic macrophytes (Genseme r and Playle, 1999), and even fewer studies have specifically looked at the e ffect of alum on aquatic macrophytes. Sparling and Lowe (1998) applied aluminum sulfate (Al2SO4) and or sulfuric acid to several constructed wetland ponds, comparing metal concentrations in four plants as influenced by soil type, acidification, and life form (rooted emergent or non-rooted submerged). Their results were similar to t hose of previous studies. Submerged plants tended to have higher metal concentrations than the emergents and emergents growing in acidified ponds had higher Al concentrations than those in non-acidi fied ponds (Sprenger and McIntosh, 1989; Jackson et al., 1993). However, they did not find any significant differences in tissue Al concentration between the control and aluminum sulfate treated wetlands, nor any increase in aqueous Al in the ponds treated with aluminum sulfate. Results were attributed to high dissolved organic carbon (DOC) concentrations which may have bound the aqueous Al. A wetland mesocosm experiment by Collins et al. (2005) looked at the use of two rooted emergents, a rooted floating-leaved, and SAV to treat metal-laden (84 mg Al L-1) effluent runoff from coal stockp iles. Higher concentrations of Al were found in the roots than shoots, and metal biomass concentrations increased as the concentration in the soil (rooted) or water column (non-rooted) increas ed. The shoots of SAV and floating plants accumulated higher Al, Fe, and Mn concentra tions than those of the emergents which was attributed to greater surface area and foliar uptake of these metals. An acid intolerance study of the submersed macrophyte Vallisneria americana (Gris et al., 1986) found a significant decrease in biomass at pH 5 or 6 compared to the

PAGE 79

60 controls at pH 7.5. They found tissue Al c oncentrations to be inversely related to biomass production with Al concentrations 2-3 times higher in plants grown at pH 5 versus those grown at higher pH values. Th is suggested Al toxici ty as the cause of reduced growth, with symptoms of Al toxicity such as th e premature browning of leaf tips as well as smaller and fewer buds produced. Kaggwa et al. (2001) looked at the effect s of alum sludge discharge from a water treatment plant into a natural swamp in Uganda. Phragmites, Cladium , and Cyperus biomass, shoot productivity and re-growth we re monitored and plant tissue analyzed for total nitrogen (TN), total P (TP), Al, calcium (Ca), and magnesium (Mg). There was no effect on plant biomass but productivity was hindered, suggesting the alum sludge may have affected the root system, reducing nutrien t availability to the plants. Additionally, the plant tissue TP did not increase with bi omass, indicating the P supply to the plants may have been inhibited by the Al (Kaggw a et al., 2001). Conversely, it has been suggested that alum should not interfere in the growth of rooted aquatic macrophytes because they are able to access the high P content in the subsurface sediment (Carignan and Kalff, 1979; Welch and Schrieve, 1994). There are several methods to quantify eff ects of management techniques on aquatic vegetation. Measurement of biomass is the most sensitive measure of plant abundance and therefore is the preferred method in eval uating the effect of management practices, such as the application of alum, on target species (Madsen and Bloomfield, 1993). Biomass is the mass per unit area of living plan t material reported as grams of dry plant material per square meter (Madsen, 1993; Brower et al., 1998). Both the biomass and

PAGE 80

61 productivity of aquatic macrophytes may be affe cted by increased meta l concentrations in the soil or water column. Productivity measurements are another me thod of quantifying effects on aquatic vegetation and is a measure of the net annual growth of vegetative biomass, estimated by the rate of increase of bi omass over time (Boyd, 1970; Elzinga et al., 2001) or by measuring shoot or leaf increments over ti me (Dickerman and Stewart, 1986; Kirkman, 1996). The latter provides a non-destructive, in situ growth rate with low variance such that small differences can be detected (Kirkman, 1996). Williams and Murdock (1972) developed a method to estimate plant producti on in subtropical or tropical regions by multiplying the average standing crop by the rati o of the stem growth to the average stem biomass. This was adapted for cattails in a study by Davis (1984) and was used in this experiment for both cattails and bulrush. The overall goal of this study was to de termine the effects of continuous lowdosage alum treatment on P sequestration, overa ll water quality, and emergent and SAV biomass production in flow through mesocosm s receiving advanced secondary treated wastewater. The hypotheses were that alum application w ould improve P sequestration and water quality within all mesocosms, regardless of plant type or continued external loading. Secondly, alum would impact the biomass of both SAV and emergents due to nutrient limitations or Al toxicity. Materials and Methods Site Description The Orlando Easter ly Wetlands (OEW) R eclamation Project located in Orange County, is one of the oldest and largest c onstructed treatment wetlands in the United States, located east of Orlando in Christ mas, FL. The wetland was built in 1986,

PAGE 81

62 designed by Post, Buckley, Schuh & Jernigan, In c. for the City of Orlando’s Iron Bridge Regional Water Pollution Control Facility (WPCF) which needed an alternative discharge point for its wastewater effluent (Burney et al., 1989). The main goal in designing the system was to use macrophytes to facilitate nut rient removal for an average daily flow of up to 132,489 m3d-1 of effluent from the Iron Bridge WPCF before discharging into the St. Johns River (Black and Wise, 2003). The 494 ha wetland lies on a 664 ha piece of land located 3.2 km west of the main channel of the St. Johns River (SJR). Hist orically, the land was part of the riparian wetland adjacent to the SJR, but was drained fo r cattle pasture around th e turn of the last century (Burney et al., 1989). The site has a natural topographi c gradient of 4.6 m downward from west to east allowing water to flow by gravity through a series of cells with an average elevation drop across each cell of approximately 1 m (Martinez and Wise, 2003) (Figure 3-1). Water exits the we tland through a weir c ontrol structure and flows into a receiving ditch. From there wa ter can flow directly to the SJR or by sheet flow through Seminole Ranch, a natural marsh adjacent to the OEW owned by the St. Johns River Water Management District. Cells 1 through 12 and 15 are deep marsh, designed primarily for nutrient remova l, planted with either cattails (Typha spp.), giant bulrush ( Scirpus californicus), or a combination of the two. Cells 13, 14, and 16 through 18 consist of a mixed marsh dominated by submergent and emergent macrophytes including Ceratophyllum demersum , Limnobium spongia, Myriophyllum spicatum, Najas guadalupensis, Nuphar luteum, Nymphaea odorata, Pontederia cordata, Sagittaria lancifolia, and Sagittaria latifolia . These cells serve as a diverse wildlife habitat while continuing to provide nutrient removal (Martinez and Wise, 2003).

PAGE 82

63 Figure 3-1. Site map of Orlando Easterly Wetland, Christmas, Florida. Plants collected in cell 1 (Scirpus californicus), cell 10 ( Typha spp.), and cell 13 (Najas guadalupensis ). Mesocosm location designated by star. The overall average influent TP co ncentration from 1988 to 2005 was 0.22 mg L-1, however, annual inflow TP concentr ations ranged from 0.02 – 3.30 mg L-1 during the same time period. Since its inception, the OE W has exceeded performance expectations. The TP discharge permit limit established by the Florida Department of Environmental Protection is 0.2 mg L-1 (Wang et al., 2006). From 1988 to 1995 the average TP discharged was 0.07 mg L-1 (Sees and Turner, 1997), however TP values have been considerably higher from December to February in recent years (Wang et al., 2006). Project Location Created by: Lynette Malecki Dated: 12/30/04 2 1 3 4 5 6 7 8 9 10 11 12 15 13 14 16 17 18 Lake

PAGE 83

64 Mesocosm Establishment Eighteen circular me socosms, each with a surfa ce area of 1.86 m2 and depth of 0.88 m, were established in June 2004. They in cluded triplicate expe rimental and control mesocosms planted with either Typha spp., Scirpus californicus , or SAV ( Najas guadalupensis dominated) utilizing a randomized bl ock design. Approximately 0.3 m of homogenized soil was added to each mesocosm. The soil used to initiate the mesocosms originated from a dredged spoil pile with a mixture of organi c material from cell 1, 3, 4, 7 and 8 of the OEW (Figure 3-1). Soil used had a TP content of 111 13.4 mg kg-1, 285 21.2 mg kg-1 oxalate-extractable Al, 19.9% 0.02 moisture content and soil pH of 5.2 0.3. Each mesocosm contained a polyvinyl chloride (PVC) drain which permitted control of water leve ls to within 3 cm. The wa ter flowing through the mesocosms originated from the outflow of cell 15 and was pumped to a head tank and distributed (on a one hour timer) via gravity to each mesocosm at a rate of 360 L d-1. This provided a hydraulic loading rate of 9.6 cm d-1, and retention time of approximately 9.5 days, similar to the retention time of the SAV cells within the OEW. On June 30, 2004, the mesocosms were planted to begin the five month grow-in / stabilization period. The six SAV mesocosm s were established with 4.05 0.06 kg wet weight (WW)/ meso cosm, or 2.18 kg WW m-2 Najas guadalupensis harvested onsite. The stocking density for the Typha spp. and Scirpus californicus mesocosms was 15 plants / mesocosm, averaging 1.88 kg WW m-2 Typha spp., and 1.04 kg WW m-2 Scirpus californicus both harvested onsite (Figure 3-1). The leaves of the Typha and Scirpus plants were cut to a length of 78.7 cm, just above the surface of the mesocosms in order to begin with a more uniform biomass. A 53 cm water column was maintained in the

PAGE 84

65 SAV mesocosms throughout the entire grow-i n, while in the emergent mesocosms a 28 cm water column was maintained for the firs t month to allow the plants to establish before raising the water column to 53 cm for the remainder of the study. Water quality parameters such as pH, dissolved oxygen (D.O.), temperature, and conductivity were monitored weekly using a handheld YSI 85 during the grow-in. Additionally, two months prio r to the start of the experiment, water samples were collected for analyses from the inflow and each mesocosm outflow using a peristaltic pump. These included: unfiltered acidified for TP and total kjeldahl N (TKN) analysis; filtered (0.45m Whatman polydiscTM AS 50 mm inline filters) acidified for total dissolved P (TDP), total dissolved N (TDN), DOC, ammonium-N (NH4-N), and nitrate-N (NO3-N); and filtered for soluble reactive P (SRP) and dissolved Al. Acidified samples were refrigerated at 4 C until digested and or analyzed while the non-acidified samples were frozen until analyzed. Water TP a nd TDP were digested by autoclaving and analyzed using automated, colorimetric analysis (Method 365.1, USEPA, 1993) while TKN and TDN were digested by the Kjeldahl procedur e (Method 351.2, USEPA, 1993). Experiment Initiation On Decem ber 1, 2004, liquid alum was pumped from a head tank via peristaltic pumps on an ART-3 repeat cycle timer ( 120v, 15 amp) to the designated mesocosms through black polyethylene tubi ng at a rate of 0.81 g Al m-2 d-1 for three months resulting in a total addition of 68.2 g Al m-2. Green stem counts were taken six times over the course of the experiment. Plant bioma ss and tissue nutrient concentrations were determined from one 25 cm x 25 cm quadrat harvested from each mesocosm upon initiation and completion of the study. Emergent species were clipped at the soil surface while all submerged vegetation falling within the quadrat of the water column to the soil

PAGE 85

66 surface was collected. All plants were rins ed thoroughly with tap and then distilled water, and all visible algae or epiphytes wiped off. Emergent plants were then cut into roughly 10-cm pieces, and all plant material was placed in paper bags, dried at 40 C at least 48 hours or until constant weight was reached (APHA, 1998), and ground using a Wiley mill. Plant tissue was analyzed for total C (TC) and total N (TN) using an Elemental Combustion ECS 4010 CHNS-O Analyzer (Cos tech Analytical Technologies, Inc., Valencia, CA). Total P and metal analysis involved combustion of 0.2 g oven-dried subsamples of plant tissue at 550 C for 4 h in a muffle furnace followed by dissolution of the ash in 6 M HCl on a hot plate (Anders en, 1976). Total P was analyzed using an automated ascorbic acid method (Method 365.4, US EPA, 1993). Total Al, Ca, iron (Fe), and Mg were analyzed by inductively coupled argon plasma spectrometry ( Vista MPX CCD simultaneous ICP-OES manufactured by Varian , Inc., Walnut Creek, CA; Method 200.7, USEPA, 1993). Water sample s from each mesocosm were collected twice weekly for the first month, and weekly for the remainder of the experiment. All water quality and nutrient variables were coll ected in the same manner as the grow-in. The rate of net annual aboveground producti on was required to calculate the uptake of nutrients by aboveground Typha spp. and Scirpus californicus biomass over the study period (Dickerman and Stewart, 1986). In each Typha and Scirpus mesocosm, triplicate newly emerged shoots were tagged loosely around the base. On each subsequent sampling date the individual Typha leaves were tagged and labeled as they emerged. Typha leaf and Scirpus shoot lengths were measured from the soil surface to the most distal portion of the leaf / s hoot weekly during the three m onth experiment and biweekly

PAGE 86

67 thereafter until all tagged plants were dead (within 46 weeks). A nonlinear regression curve of leaf length to dry weight was established for both Typha spp. and Scirpus californicus by collecting thirty Scirpus shoots of various sizes from cells 1 (R2 = 0.95) and twelve Typha plants (69 leaves) from cell 10 within the OEW (R2 = 0.94), to convert the lengths measured during the experime nt into weights (Appendix B). Annual production was then calculated by multiplying the mean annual standing crop by the ratio of annual leaf growth to mean leaf weight (Davis, 1984). Annual P and N uptake associated with growth were calculated by multiplying the growth to weight ratio by the mean P and N storages (g m-2) for each emergent mesocosm (Davis, 1984). Additionally, the relative growth rate (RGR) was cal culated for all mesocosms as: RGR = ln(W2)– ln(W1)/(t2-t1) where W1 and W2 are the initial and final pl ant dry weights at time 0 (t1) and time 84 day (t2) (Forchhammer, 1999; Brix et al ., 2002; Hadad et al., 2006). Statistical Analysis Data norma lity was determined using the Kolmogorov-Smirnov test (Minitab 13.32, 2000) and data were transformed to fit a normal distributi on (Microsoft Excel, 2000). One-way ANOVAs and multiple comparisons by Tukey’s W were used on plant tissue and growth variables. Repeated meas ure analysis of variance (RMANOVA) using a general linear model followed by Tukey’s W multiple comparison was used on water column data to determine significant di fferences (p<0.05) between plant type and treatment (Gurevitch and Chester, 19 86; Minitab 13.32, 2000). Linear regression analysis and Pearson product correlation coefficients were also used to determine significant (p<0.05) relationships (Microsoft Excel, 2000).

PAGE 87

68 Results and Discussion Water Quality Characteristics pH The pH of all me socosms remained above 6.5 and below 10.0 throughout the experiment (Figure 3-2). The SAV mesocosms did have significantly higher pH values than those of the emergent mesocosms. Measurements were taken in each mesocosm at approximately the same time of day (mid-m orning). The underwater photosynthesis of 6.0 7.0 8.0 9.0 10.0 0142842567084Time (d)pH SAV control Scirpus control Typha control SAV alum Scirpus alum Typha alum Figure 3-2. Water column pH in Orla ndo Easterly Wetland mesocosms (n=3). SAV during the day removes dissolved inor ganic carbon from the water column which can result in an elevated pH (Dennison et al., 1993). This pH change can affect the toxicity and solubility of metals including Al. The leaves of emergent vegetation are primarily above the water surface so the shift in pH would not be expected (Farve et al., 2004). Overall, the mesocosm water columns treated with alum had significantly lower pH values than the control mesocosms, averaging 8.1 0.6 and 8.8 0.5, respectively. Specifically, the SAV and Typha tanks treated with alum had significantly lower pHs than their control counterparts. At a pH value greater than 8, the dominant Al species is

PAGE 88

69 Al(OH)which can result in Al toxicity as well as P release (Entranc o, 1986; Cooke et al., 1993b). However, when water samples were an alyzed for soluble Al , all concentrations throughout the experiment were below the pr actical quantification limit of 20 mg Al L-1. These results are similar to the findings of Sparling and Lowe (1998) which were attributed to high DOC concentrations binding the available soluble Al. Dissolved Oxygen Dissolved oxygen levels ra nged between 3.4 and 8.8 mg L-1 throughout the experiment (Figure 3-3) with an overa ll average water temperature of 18.8C 4.0 during the study. Similar to pH, the SAV mesocosms had significantly higher D.O. values than those of the emergent mesocosm s due to the underwater photosynthesis of the SAV oxygenating the water column. Overall, the alum-treated mesocosms (5.74 1.41 mg L-1) had significantly greater D.O. concentr ations than the c ontrols (5.08 1.25 mg L1). The alum-treated SAV tanks had significan tly higher D.O. levels than the control 2.0 4.0 6.0 8.0 10.0 0142842567084 Time (d) Dissolved oxygen (mg L-1) SAV control EAV control SAV alum EAV alum Figure 3-3. Water column dissolved oxyge n content in Orlando Easterly Wetland submerged aquatic vegetation (SAV) a nd emergent aquatic vegetation (EAV) mesocosms (SAV n=3, EAV n=6).

PAGE 89

70 SAV tanks, as well as all other Scirpus californicus and Typha spp. alum-treated and control tanks. Alum is not only used for P removal, but al so to clarify turbid lakes by acting as a source of positive electrolytes neutralizing negatively-charged soil particles, which causes them to settle out of the water column (Davis and Gloor, 1981; Berg and Berns, 1985). The increased water clarity may stim ulate rapid plant growth and in turn increased photosynthesis (Ent ranco, 1986) which may explain why the SAV tanks treated with alum had such high D.O. concentrations. Conductivity The average conductivity within the mesocos ms ranged between 300 and 630 S cm-1. There were no significant differences between plant type but overall, the alumtreated mesocosms did have significantly greater conductivities than the controls, averaging 505 76.1 and 463 67.4 S cm-1, respectively, over the three month study. The increased conductivity can be attributed to the increased concentrat ion of sulfate ions added to the alum-treated tanks. Dissolved Organic Carbon Concentrations of DOC remained relatively stable in the control mesocosms over time while in the alum -treated mesocosms there wa s a gradual decreasing trend (Figure 3-4). There were no significant diffe rences in DOC due to plant type, however, as Figure 3-4 shows, the alum-treated tanks did have signi ficantly lower DOC concentrations than the controls from December to February as determined by RMANOVA (Table 3-1). For all three plant types, the alum-treated mesocosms were significantly lower than all other controls with Typha spp. tanks showing the most removal (23.8%) on average, followed by the SAV (18.0%) and Scirpus californicus (14.0%). As mentioned previously, the

PAGE 90

71 positively charged Al3+ and resulting alum floc bi nd organic matter by neutralizing negatively-charged particles (Sparling and Lowe, 1998; Van Hullebusch et al., 2002). Table 3-1. Water quality characteristics of inflow water and treated outflow leaving mesocosms. Parameter Control Treatment Alum Treatment Inflow Outflow Reduction Outflow Reduction mg L-1 mg L-1 % mg L-1 % Soluble reactive P 0.25 0.09 0.10 0.06 44.5 31.8 0.03 0.02 83.0 12.5 Total dissolved P 0.26 0.09 0.13 0.07 41.8 21.8 0.05 0.03 58.7 16.8 Total P 0.26 0.10 0.14 0.06 41.8 22.4 0.10 0.04 59.1 17.3 Particulate P 0.01 0.01 0.02 0.01 -2.14 2.47 0.05 0.04 -7.74 6.12 Ammonium-N 0.03 0.01 0.01 0.01 63.2 20.3 0.01 0.01 62.9 23.7 Total dissolved N 0.59 0.08 0.58 0.05 1.40 11.5 0.52 0.08 11.3 15.6 Total kjeldahl N 0.63 0.08 0.64 0.08 16.0 10.7 0.57 0.09 33.6 27.2 Dissolved organic C 8.71 0.51 8.67 0.51 -0.11 8.10 7.07 1.25 18.6 13.8 Outflow values are means of 18 samplings on 9 control and alum-treated mesocosms, from Dec. 2004 to Feb. 2005. All mesoco sms received the same inflow water, therefore one sample was taken on each sampling trip. Negative percent reduction = net increase. 0.0 2.0 4.0 6.0 8.0 10.0 12.0 0142842567084 Time (d)Dissolved organic C (mg L-1) Control Alum Inflow Figure 3-4. Water column dissolved oxygen concentration in Orlando Easterly Wetland mesocosms (n=9) and inflow water.

PAGE 91

72 In the control mesocosms the SAV were most efficient at removing DOC from the inflow water averaging 2.6% removal. The emergent Scirpus and Typha tanks averaged net increases of 0.6 and 2.4% respectively. The SAV had lower DOC concentrations perhaps due to the physical filtration of the wa ter by the dense SAV throughout the water column. Dissolved Inorganic Nitrogen There was no detectable NO3-N entering or leaving the mesocosm water columns throughout the study either indicating an inhi bition of nitrification due to low D.O. concentrations (Malecki et al., 2004) or rapid denitr ification conve rting the NO3-N to NH4-N, nitrous oxide (N2O) or dinitrogen gas (N2) (Christian, 1989). Denitrification rates are directly controlled at the microbial level by NO3-N availability, oxygen inhibition, organic matter content and temper ature (Seitzinger, 1988; Rysgaard et al., 1994). There were very low levels of NH4-N entering the mesocosms and even lower concentrations being exported from the meso cosms (Figure 3-5, Table 3-1). There were 0.00 0.01 0.02 0.03 0.04 0.05 01428425670 Time (d)Ammonium-N (mg L-1) Control Alum Inflow Figure 3-5. Water column ammonium-nitrogen concentrations in Orlando Easterly Wetland mesocosms (n=9) and inflow water.

PAGE 92

73 no significant differences based on plant type or treatment wi th both the control and alum tanks averaging 0.01 0.01 mg NH4-N L-1 throughout the study. On average the mesocosms were able to rem ove 63.0% of the inflow NH4-N. Total Dissolved Nitrogen The total dissolved N includes both the dissolved inorganic N (DIN) discussed previously as well as the dissolved organic N (DON). There was a clear treatme nt effect beginning after eighteen days of slow drip alum application (Figure 3-6). There were no significant differences in TDN concentrations based on plant type, but the alum-treated tanks did have significantly lower TDN con centrations as determined by RMANOVA. The emergent Scirpus and Typha control mesocosms had significantly more TDN than the alum-treated emergent tanks averagi ng 11.9 and 12.8% removal in the alum-treated mesocosms as compared to 0.1 and 1.6% removal in the controls, respectively. There was no significant difference between SAV mesoco sms, but those treated with alum did 0.30 0.45 0.60 0.75 0.90 0142842567084 Time (d)Total dissolved N (mg L-1) SAV control EAV control SAV alum EAV alum Inflow Figure 3-6. Total dissolved nitrogen concen trations in the water column of Orlando Easterly Wetland submerged aquatic vegetation (SAV), emergent aquatic vegetation (EAV) mesocosms (SAV n= 3, EAV n=6), and inflow water.

PAGE 93

74 have greater removal efficiency, averaging 9.2% removal as compared to the controls which only averaged 2.5% removal. While al um did not have any effect on the DIN it was able to reduce water column TDN concentr ations due to its ab ility to bind organic matter and thus the organically-bound N (S parling and Lowe, 1998; Van Hullebusch et al., 2002). Total Kjeldahl Nitrogen The TKN includes all dissolved and partic ulate organically-bound N and ammonia. Similar to the TDN results, the re is an eleven day lag time before there is a sufficient amount of alum floc to bind the organic matte r associated with the TKN (Figure 3-7). There was no significant difference in TKN c oncentrations based on plant type but the alum-treated tanks did have significantly lowe r TKN concentrations than the controls. Specifically, the alum-treated Typha tanks had significantly lower TKN concentrations than their control counterpart s averaging 24.1% removal. 0.40 0.50 0.60 0.70 0.80 0.90 0142842567084 Time (d)Total kjeldahl N (mg L-1) Control Alum Inflow Figure 3-7. Total kjeldahl nitrogen concen trations in the wate r column of Orlando Easterly Wetland mesocosms (n=9) and inflow water.

PAGE 94

75 Similar to the DOC results, the SAV control mesocosms were most efficient at removing TKN from the inflow water av eraging 2.1% removal. The emergent Typha and Scirpus tanks actually averaged ne t increases of 2.9 and 0.1% respectively most likely due to the decomposition of detrital materi al in the water column. The SAV control mesocosms may have been able to remove TKN due to the physical filtration of the particulate organic matter by the de nse SAV throughout the water column. Soluble Reactive Phosphorus The mesocosm s treated with alum were much more effective at SRP removal. The alum-treated mesocosms were able to mainta in low water column concentrations ranging between 0.088 0.004 mg L-1 throughout the experiment whil e SRP concentrations in the control mesocosms were highl y variable (0.278-0.039 mg L-1), influenced by fluctuations in the inflow SRP (Figure 3-8). Overall, the Scirpus mesocosms did have significantly greater water column SRP concentrations than the Typha and SAV mesocosms. One possible reason for this difference could be attr ibuted to the significantly smaller initial 0.00 0.10 0.20 0.30 0.40 0.50 0142842567084 Time (d)Soluble reactive P (mg L-1) SAV control Scirpus control Typha control SAV alum Scirpus alum Typha alum Inflow Figure 3-8. Water column soluble react ive phosphorus concentrations in Orlando Easterly Wetland mesocosms (n=3) and inflow water.

PAGE 95

76 biomass, growth, and productivity of the Scirpus plants as compared to the other two vegetation types, which will be discussed later in the text. The water column of the alumtreated mesocosms had significantly less SRP than the controls as determined by RMANOVA. For all three plant types, the alum-treated mesocosms had significantly lower SRP than their co ntrol counterparts with Typha tanks showing the most removal (86.4%) on average, followed closely by the SAV (85.2%) and Scirpus (77.3%). Of the controls, the SAV were once again the most efficient at SRP removal averaging 51.0% removal while the emergent Typha and Scirpus tanks averaged removals of 48.1 and 34.3%, respectively. Total Dissolved Phosphorus The TDP, similar to the TDN, include s both the SRP and dissolved organic P (DOP). Figure 3-8 and 3-9 look very sim ilar due to the dominance of the SRP fraction in the water column averaging 73% of the TD P for all mesocosms. There were no 0.00 0.10 0.20 0.30 0.40 0.50 0142842567084 Time (d)Total dissolved P (mg L-1) Control Alum Inflow Figure 3-9. Water column total dissolved phosphorus concentrations in Orlando Easterly Wetland mesocosms (n=9) and inflow water.

PAGE 96

77 significant differences in TDP concentrations attributed to plant type. However, the water column of alum-treated mesocosms had significantly less TDP than the controls, similar to the SRP results. For all three plant types the alum-treated mesocosms had significantly lower TDP concentrations than all controls, with Typha tanks once again showing the most removal (44.7%) on average, followed closely by the SAV (38.1%) and Scirpus (37.9%). Of the controls, the SAV was once again the most efficient at TDP removal, averaging 27.8% removal, while the emergent Typha and Scirpus tanks averaged removals of 20.7 and 10.0%, respectively. Total Phosphorus The water column TP includes particulate P (PP) in addition to SRP and DOP. The average inflow TP to all mesocosms was co mposed of 94% SRP, 4% DOP and 2% PP. There were no significant di fferences in TP treatment based on plant type, but the mesocosms treated with alum had significantl y lower water column TP concentrations than those of the controls (Figure 3-10). In particular, the emergent Typha and Scirpus mesocosms treated with alum had significantly lower TP concentrations than all control mesocosms, averaging 64.3% and 57.3% removal, respectively. The average composition of the outflow TP from emergent tanks treated with alum was 60% SRP, 29% PP, and 11% DOP while in the SAV mesocosms outflow TP consisted of 46% PP, 39% SRP, and 15% DOP. In the control mesocosms, SAV was most efficient at TP removal, averaging 45.9% removal, while the Typha and Scirpus tanks averaged removals of 42.5 and 37.0%, respectively. The composition of the

PAGE 97

78 0.00 0.10 0.20 0.30 0.40 0.50 0142842567084 Time (d)Total P (mg L-1) SAV control EAV control SAV alum EAV alum Inflow Figure 3-10. Total phosphorus concentrations in the water column of Orlando Easterly Wetland submerged aquatic vegetation (SAV) and emergent aquatic vegetation (EAV) mesocosms (SAV n= 3, EAV n=6), and inflow water. outflow TP from all control mesocosms was simi lar to that of the inflow, dominated by 56% SRP, with increased (28%) DOP and (16%) PP. Particulate Phosphorus The calculated PP was the only water quality param eter for which the alum-treated mesocosms had significantly greater concentr ations than the control mesocosms (Figure 3-11, Table 3-1). This sugge sts that some of the AlOH3 floc remained suspended in the water column and was being exported from the mesocosms in the less bioavailable, particulate form. As indicated by the com position of the outflow TP in the previous section, this was particularly true in the dense vegetati on of the alum-treated SAV mesocosms where PP was, on average, the dominant form exported. While total suspended solids (TSS) were not measured in this study, the export of PP may have resulted in increased TSS concentr ations (Bostan et al., 2000) as well. This may be critical to note in treatment wetlands with permitted PP or TSS discharges. Both

PAGE 98

79 0.00 0.05 0.10 0.15 0.20 0142842567084 Time (d)Particulate P (mg L-1) SAV control EAV control SAV alum EAV alum Inflow Figure 3-11. Particulate phos phorus concentration in the water column of Orlando Easterly Wetland submerged aquatic vegetation (SAV), emergent aquatic vegetation (EAV) mesocosms (SAV n= 3, EAV n=6), and inflow water. the control and alum-treated mesocosms served as net sources of PP as indicated by negative percent removal values (Table 3-1) . In addition to the floc-bound P, the increased PP can also be attributed to the decomposition of plant material within both the control and treated mesocosms. Macrophyte Characteristics Emergent Growth Study There were no significant diffe rences in winter growth ch aracteristics between those emergent macrophytes growing in the control and alum-treated m esocosms (Table 3-2). The Typha spp. did have significantly greater tota l growth and a greater annual growth rate than the Scirpus californicus. This led Typha to have a significantly greater growth / weight ratio, annual biomass, and annual production rate than the Scirpus plants. This may explain why, with the exception of DOC and TKN, both the alum and control mesocosms planted with Typha always performed better than the S cirpus mesocosms. The Typha plants were able to uptake more P and N since the plants grew larger, quicker,

PAGE 99

80 and produced more biomass (61190 14949 g m-2 yr-1) than the Scirpus plants (3489 1638 g m-2 yr-1). Table 3-2. Emergent macrophyte growth ch aracteristics within mesocosms at the Orlando Easterly Wetland. Values are means 1 standard deviation (n = 9). Plant Type Treatment Total growth Annual growth Growth / Weight ratio Annual biomass Annual production g plant-1 g yr-1 g m-2 g m-2 yr-1 Scirpus Alum 5.42 0.41 8.69 2.73 2.25 0.39 1496 574.9 3470 1798 Control 5.18 2.33 7.89 3.26 2.15 0.32 1600 585.1 3508 1478 Typha* Alum 32.7 7.26 40.7 8.05 4.34 0.79 15637 2171 68636 21274 Control 25.4 7.66 34.2 9.01 3.77 0.89 14468 1385 53744 8625 * All Typha parameters significantly greater than Scirpus at the 0.05 probability level. Each Typha plant on average had at least four living leaves, producing approximately seventeen leaves per year (Table 3-3). Leaf growth and mortality occurred on a continuum throughout the lifespan of the Typha plants with young leaves emerging and increasing weight while older leaves decreased in weight and died, exhibiting a unimodal pattern of growth (F igure 3-12). This resulted in annual production far exceeding annual biomass (Table 3-2) similar to the findings of Davis (1984). The average Typha leaf weighed over twice as much as the average Scirpus Table 3-3. Scirpus californicus shoot and Typha spp. leaf characteristics within mesocosms at the Orlando Easterly We tland. Values are means 1 standard deviation (n = 9 plants). Plant Type Treatment Live Leaves Live leaf production Leaf turnover Shoot/Leaf weight Shoot/Leaf longevity leaves plant-1 leaves yr-1 leaves leaf-1 yr-1 g plant-1 weeks Scirpus Alum na na na 3.77 0.56b 34.7 9.07 Control na na na 3.70 1.42b 33.9 5.70 Typha Alum 4.44 0.57 16.2 0.48 3.70 0.54 9.13 0.39a 38.8 6.48 Control 4.18 0.28 17.0 0.40 4.08 0.35 8.88 0.39a 37.6 7.53 a,b Significantly different at the 0.05 probability level.

PAGE 100

81 300 500 700 900 051015202530354045 Time (weeks)Biomass (g m-2) Alum Control Figure 3-12. Unimodal grow th pattern exhibited in Typha spp. leaves in Orlando Easterly Wetland mesocosms (n=9). shoot with both averaging 84% moisture, how ever, there was no signi ficant difference in Typha leaf and Scirpus shoot longevity, on the whole, aver aging slightly over 36 weeks. There was a significantly greater number of living Scirpus plants per mesocosm than there were Typha throughout the study (Figure 3-13). Additionally, despite the variability in Scirpus control mesocosms, repeated meas ure analysis found there to be a significantly greater number of live plants in the control mesocosms than in the Scirpus mesocosms treated with alum while there were no significant differences in the number of Typha plants per mesocosm. This suggests that while alum application may not have 0 50 100 150 200 0142842567084 Time (d)Green Stem Count Typha control Typha alum Scirpus control Scirpus alum Figure 3-13. The number of live plants in each Orlando Easterly Wetland mesocosm throughout the experiment (n=3 standard error).

PAGE 101

82 .affected the growth characteristics of the Scirpus californicus that were able to get established, it may have hindered the overall quantity of emerging shoots. Destructive Biomass and Growth At the initiation of the expe riment, there were no signifi cant differences in biomass between alum and control mesocosms, however, there were differences among macrophyte type (Table 3-4). The SAV me socosms had significantly greater biomass than both emergents, and the Typha had significantly more biomass than the Scirpus mesocosms at the start of the experiment. By the end of the experiment, there were no significant differences in biomass based on plant type, however, the biomass in the SAV tanks treated with alum was significantly le ss than the biomass of the SAV growing in the control mesocosms. This suggests that alum may have negatively impacted the SAV due to its life form, while the emergent plants remained relatively unaffected. Submerged aquatic vegetation absorbs a larg e portion of its nutrients and metals directly from the water column into the shoot s, while emergent plants tend to obtain them from the soil (Crowder, 1991; Rai et al., 1995; Sparling and Lowe, 1998; Madsen and Cedergreen, 2002). The reduced biomass in the alum-treated SAV mesocosms may be attributed to the significantly lower P concentration in the water column, or possibly to Al toxicity because the SAV mesocosms treated with alum did have significantly lower pH values than their control c ounterparts (Gris et al., 1986). To compare the SAV growth with the emer gents, the RGR was calculated for each mesocosm (Table 3-4). Alum treatment did no t significantly affect th e RGR of any of the aquatic macrophytes, however the SAV and Scirpus mesocosms treated with alum did have lower RGRs than the controls. The Scirpus mesocosms had significantly greater RGRs than both the SAV and Typha planted mesocosms as indicated by the

PAGE 102

83 Table 3-4. Biomass and relative growth rates (RGR) based on total dry weight of plants harvested from the Orlando Easterly We tland mesocosms. Values are means 1 standard deviation (n=3). Plant Biomass (g m-2) RGR (g g-1 d-1) Time 0 d Time 84 d Alum Control Alum Control Alum Control SAV 706 12.7 898 71.1 209 56.7 537 312 -0.006 0.002a -0.003 0.002a Scirpus 85.3 38.6 91.9 31.7 367 245 748 625 0.016 0.004b 0.023 0.014b Typha 349 161 436 353 392 55.3 322 211 0.002 0.004a -0.004 0.008a SAV = submerged aquatic vegetation a,b Significantly different at the 0.05 probability level. significantly greater Scirpus biomass on day 84 than day 0. In the SAV mesocosms as well as the Typha control mesocosms growth actually declined from December to February corresponding to the decrease in biomass. Total Carbon, Nitrogen, Phosphorus Concentrations and Storage Throughout the study, the SAV had significantly less C than the emergent macrophytes (Table 3-5). This can be attri buted to the im portance of struct ural C relative to plant life form development (Duarte, 1992; Pompo et al., 1999). Emergents must use C toward structural cell wall carbohydrates as well as support structures to maintain the shape and elevation of aerial stems and leaves as co mpared to the SAV which are supported by the water and thus do not require as much C for structural components (Sterner and Elser, 2002). There were no significant differences in C tissue content due to alum application, nor were there differences over time. While the SAV C tissue concentrations were low, their C storage capacity was equivalent to that of Typha at experiment initiation due to the high biomass of the SAV while the C storage in the Scirpus plants was significantly lower than the Typha and SAV due to the low biomass (Table 3-6). By experiment completion there were no significant differences in C storage among plants, however, the

PAGE 103

84 SAV growing in mesocosms treated with alum did have a significantly lower amount of C stored than the controls corresponding to the decrease in biomass. Table 3-5. Carbon, nitrogen, and phosphorus tissue concentrations in aquatic macrophytes harvested from the Orlando Easterly Wetland mesocosms. Values are means 1 standard deviation (n=3). Time 0 d Time 84 d Treatment Nutrient SAV Scirpus Typha SAV Scirpus Typha mg g-1 DW mg g-1 DW mg g-1 DW mg g-1 DW mg g-1 DW mg g-1 DW Alum Total C 316 43.3 381 1.52 417 10.3 341 12.3 404 17.4 421 4.43 Total N 18.3 4.08a 13.8 5.21ab 9.60 2.35b 16.2 0.38a 9.32 2.52b 7.60 0.86b Total P 14.2 8.21 16.9 4.25 8.16 2.27 27.3 7.98 15.5 1.91 14.4 1.61 TC:TN 15 5.8 31 14 33 23 21 1.2 46 17 56 6.5 TC:TP 27 13 23 5.3 54 15 13 3.7 26 3.6 30 3.7 TN:TP 1.6 0.7 0.8 0.3 1.3 0.5 0.8 0.2 0.8 0.1 0.7 0.2 Control Total C 310 30.1 375 30.9 398 25.8 330 3.08 399 10.9 406 12.0 Total N 16.2 3.35 12.3 2.26 8.11 5.56 15.4 3.27 7.43 0.46 9.00 1.99 Total P 13.5 6.41 17.5 4.72 11.4 4.73 23.8 6.67 20.7 9.68 15.8 1.55 TC:TN 16 4.3 31 4.6 64 34.1 22 4.3 54 4.9 47 11 TC:TP 26 10 22 4.7 39 16 15 3.5 22 7.9 26 1.9 TN:TP 1.3 0.4 0.7 0.3 0.7 0.2 0.7 0.1 0.8 0.2 0.7 0.2 SAV = submerged aquatic vegetation. a,b Significantly different at the 0.05 probability level. Table 3-6. Carbon, nitrogen, and phosphorus st orage in aquatic macrophytes harvested from the Orlando Easterly Wetland mesocosms. Values are means 1 standard deviation (n=3). Time Nutrient Treatment SAV Scirpus Typha g m-2 g m-2 g m-2 0 days Total C Alum 223 32.5 32.4 14.6 146 67.9 Control 280 49.5 35.0 13.9 172 140 Total N Alum 12.9 2.78 1.05 0.15 3.16 0.89 Control 14.7 4.18 1.17 0.57 4.75 6.33 Total P Alum 10.0 5.72 1.37 0.39 3.09 2.24 Control 12.4 6.85 1.63 0.76 5.78 6.60 84 days Total C Alum 145 31.2 151 107 165 23.3 Control 230 26.2 297 247 132 86.8 Total N Alum 6.84 1.1 3.05 1.24 2.99 0.56 Control 10.8 2.45 5.67 4.87 2.67 1.61 Total P Alum 12.4 5.45 5.66 3.93 5.58 0.48 Control 16.5 4.44 13.5 9.48 5.12 3.38 SAV = submerged aquatic vegetation.

PAGE 104

85 Nitrogen tissue concentrations were significantly higher in the SAV than Typha at time 0 and higher than both emergents by the en d of the experiment (Table 3-5). This may reflect the ability of SAV to efficiently uptake available N from the water column or a lack of bioavailable N in the soil for th e emergents to utilize (Madsen and Cedergreen, 2002). There were no significant differences in tissue N concentration due to alum application, nor were there differences between start and end concentrations. The SAV also had significantly greater N storage (10.06 3.32 g m-2) than the Typha (3.11 1.14 g m-2) and Scirpus (3.55 2.36 g m-2) throughout the experiment related to both the high N concentration and biomass (Table 3-6) . Additionally, the N storage in the Scirpus plants significantly increase d over time (1.16 4.36 g m-2) which can be attributed to the significant increase in Scirpus biomass over time. Alum application did not affect the N tissue storage. At the start of the experiment there we re no significant differences in tissue P concentration among plants, however due to the high initial biomass, the SAV did store significantly more P (11.2 5.79 g m-2) than both the Scirpus (1.50 0.56 g m-2) and Typha (4.43 4.65 g m-2) plants (Table 3-5, Table 3-6). By then end of the experiment the SAV tissue had a significantl y higher P concentration than Typha as well as storage. Overall, both P storage and tissue P concen trations were significantly higher at experiment completion than at its initia tion. However, there were no significant differences in P storage or tissue concentratio ns of plants growing in the alum-treated mesocosms as compared to the controls. This is a surprising result considering the significant sequestration of P from the water column by the alum floc which should have limited P availability to the SAV. In the s hort-term, however, it confirms the findings of

PAGE 105

86 Welch and Schrieve (1994) as well as the assertion by Carignan and Kalff (1979) that alum would not interfere with rooted aquati c macrophytes because they are able to access P from the soil. Using the annual growth to mean wei ght ratios from the emergent growth experiment N and P uptake rates were calculat ed for the emergent plants. There were no significant differences with treatment or plant type. The average Typha and Scirpus N uptake rates were 14.4 11.9 and 5.87 3.70 g m-2 yr-1 respectively while the P uptake rates were slightly higher at 20.2 14.3 and 11.5 8.07 g m-2 yr-1. Overall, TC:TN ratios within the plant tissue significantly in creased with time suggesting a reduction in N availability, while the TC:TP ratios signi ficantly decreased with time indicating an increase in P availabil ity (Table 3-5) (Stern er and Elser, 2002). This corresponds to the significant increase in pl ant tissue P over the course of the experiment while there was no corresponding increase in tissue N. The TN:TP ratios also significantly decreased with time overall, once again suggesting system nutrient enrichment (Downing, 1997) as well as a tre nd in N limitation with time rather than P (Guildford and Hecky, 2000). When analyzing the data by plant type, the SAV tissue had significantly lower TC:TN ratios than both the Typha and Scirpus tissue, influenced by the reduced need for C-rich structural biomass in the SAV tissue. Throughout the study, the SAV had significantly less TC than the emergent macrophyt es (Table 3-7). This can be attributed to the importance of structural C relative to plant life form devel opment (Duarte, 1992; Pompo et al., 1999). Emergents must use C toward structural cell wall carbohydrates as well as support structures to maintain the shape and elevation of aer ial stems and leaves

PAGE 106

87 as compared to the SAV which are supported by the water and thus do not require as much C for structural components (Sterner and Elser, 2002). Additionally, the TC:TP ratio significantly increased between the st art and end of the e xperiment in the SAV tissue which suggests P availability was limited, possibly due to Al toxicity, resulting in the decreased biomass. However, there we re no significant differences in any of the ratios specifically due to alum application. Other Macronutrient and Micronutrient Concentrations Upon starting the experime nt there were no significant differences in tissue Al concentrations between the alum and control mesocosms for each plant, however, the SAV plant tissue had a significantly greater Al concentration than both emergent plants (Table 3-7). As previously mentioned, s ubmerged plants are t hought to take up higher metal concentrations generally due to a higher surface/biomass ratio and foliar uptake (Baudo et al., 1981; Guilizzoni, 1991; Albers and Camardese, 1993; Rai et al., 1995; Sparling and Lowe, 1998; Cardwell et al., 2002; Collins et al., 2005). By the end of the experiment there was significantly more Al in the plants growing in the alum-treated tanks than in the controls. Specifically, the SAV growing in the alum-treated mesocosms had significantly higher Al con centrations than the SAV grow ing in the control tanks as well as all emergent plant tissue sampled from both alum and control tanks. This hyperaccumulation of Al suggests that the si gnificant decrease in SAV biomass can be attributed to Al toxicity rather than nutrient deficiencies, similar to the findings of Gris et al. (1986). There were no significant di fferences in Ca concentrations between the alum and control mesocosms for each plant, however, the SAV plant tissue had a significantly greater Ca concentration than both emerge nts, similar to the Al concentration. Typha had

PAGE 107

88 Table 3-7. Mean tissue concen trations of metals in aquatic macrophytes harvested from the Orlando Easterly Wetland mesocosms. Values are means 1 standard deviation (n=3). Time 0 d Time 84 d Treatment Element SAV Scirpus Typha SAV Scirpus Typha mg kg-1 mg kg-1 mg kg-1 mg kg-1 mg kg-1 mg kg-1 Alum Al 60.8 4.92 29.7 16.3 41.0 7.26 3315 1823 59.2 33.2 176 52.9 Ca 37989 28847 1438 153 6591 859 13372 6390 1516 154 5337 604 Cu 1.11 0.20 1.68 1.00 1.07 0.24 1.14 0.08 1.09 0.05 1.19 0.03 Fe 308 196 24.1 8.86 70.7 49.0 532 289 69.8 79.7 85.4 54.7 Mg 2222 112 509 99.5 466 30.1 2949 749.7 412 75.1 485 49.5 Mn 38.6 1.88 35.5 10.5 20.4 5.70 43.1 5.34 55.1 16.5 27.9 9.45 Mo 2.00 1.06 2.61 2.66 1.70 0.59 1.79 0.93 2.31 1.02 2.15 0.61 Ni 0.45 0.04 0.81 0.42 1.35 1.39 0.69 0.06 1.28 1.30 0.52 0.18 Zn 124 146 18.1 6.06 60.8 66.2 55.5 26.1 21.2 8.31 39.5 22.1 Control Al 65.8 56.2 38.9 24.0 41.9 14.8 66.3 32.6 25.6 13.7 44.6 19.4 Ca 30108 14407 2071 1322 6840 802 28989 6771 1474 486 4466 804 Cu 0.96 0.10 1.16 0.21 1.65 0.86 0.95 0.12 1.23 0.14 1.68 0.33 Fe 290 129 117 49.2 132 125 371 134 75.8 44.6 155 117 Mg 1757 351 484 97.6 603 81.8 1853 167.3 373 32.8 559 67.5 Mn 53.1 15.4 27.0 2.91 18.4 4.44 46.2 6.92 42.6 17.0 25.7 1.31 Mo 1.90 1.17 1.63 0.95 1.20 0.10 0.88 0.01 2.20 0.58 2.59 1.47 Ni 0.90 0.80 0.73 0.40 1.33 1.23 0.80 0.15 1.32 1.12 0.38 0.16 Zn 94.2 73.9 69.6 40.9 73.8 97.9 58.7 18.1 33.4 18.2 31.1 12.2 SAV = submerged aquatic vegetation. significantly greater Ca concentrations than Scirpus throughout the study as well (Table 3-7). At study completion, the SAV in th e alum-treated mesocosms had significantly lower Ca concentrations than the SAV growi ng in the control tanks. Aluminum has been shown to interfere with Ca absorption and tr ansport in terrestrial plants, resulting in reduced Ca concentrations in both the roots and shoots of Al stressed plants (Bennet et al., 1987; Thornton et al., 1987). The relationshi p between Ca availability and Al toxicity is complex, involving a variety of physiologi cal mechanisms which continue to be debated (Matsumoto, 2000; Barcelo and Po schenrieder, 2002; Rengel and Zhang, 2003). Similar to Al and Ca tissue concentrati ons, Fe and Mg concentrations were also significantly higher in the SAV tissue than the Typha and Scirpus throughout the study (Table 3-7). There were no differences in Fe concentrations based on alum application,

PAGE 108

89 however by the end of the experiment cont rol SAV tissue had significantly lower Mg concentrations than those grown in the alum-tre ated tanks. The addi tion of either Ca or Mg to upland plants has been shown to allevi ate symptoms of Al t oxicity (Silva et al., 2001). Perhaps the increased SAV Mg uptake is a defensive physiological response to the elevated Al concentration and Ca defi ciency, since Ca and Mg are the two major divalent cations able to co mpete with Al at exchange sites (Rufyikiri et al., 2002). Overall, the emergent tissues as well as the control SAV contained significantly more Ca followed by Mg than all other metals. In the SAV tissue from the alum-treated mesocosms there was no significant differen ce between the Ca and Mg concentrations and while the Ca concentrations were significa ntly greater than the Al concentrations, the Mg concentrations were not, once again indicat ive of the Al toxicity within the SAV. Additionally, the Al concentrati ons were significantly higher than the Fe concentrations in the alum-treated SAV tissue. Alum may have also influenced Scirpus plant nutrients with Fe concentrations being significantly grea ter than Al concentrations in the controls while not in the alum-treated tissue due to the elevated Al levels. Conclusion Alum was eff ective at P sequestration within the water column of treatment wetland mesocosms. The ability of alum to bind not only P, but organic molecules as well, resulted in improved SRP, TDP, TP, TDN, TKN, and DOC wetland treatment efficiency. However, there was an overal l increase in PP being exported from the wetlands which can be attributed to P bound to the settling alum floc, thus remaining unavailable upon settling downstream. During the winter months, wetlands in th e southern United States may become less effective at treating P as plants senes ce and microbial activity slows. Constructed

PAGE 109

90 treatment wetland permits require outflow effluent to meet specific discharge criteria regardless of season. This study showed that application of alum pr oximal to the outflow regions of the wetland may provide an effectiv e management tool to maintain discharge concentrations within permitted values during the inefficient winter treatment times as long as total suspended solids are monitored due to the increase in flocculent particulates. This is one of the first studies to specif ically investigate th e effect of alum on wetland macrophytes which play a critical functi on in nutrient removal within treatment wetlands. The research showed that Typha spp. remained relatively unimpacted by the use of alum while Scirpus californicus shoot emergence was reduced in the short term. It has been noted that over time the alum floc migrates downward in th e soil profile due to sediment and detrital deposition. This may result in even greater adverse impacts on the plants due to reduced nutrient availability and increased Al concentrations in the root zone. Submerged aquatic vegetation were immediately impacted by alum application resulting in decreased biomass attributed to Al toxicity al though the P concentration of the tissue was no affected. Therefore the use of alum should be restricted to treatment wetland areas dominated by emergent vegetati on. The long term efficacy of alum P sequestration as well as its long term im pact on aquatic macrophytes in treatment wetlands needs further investigation.

PAGE 110

91 CHAPTER 4 INFLUENCE OF ALUM ON TREATMENT WETLAND SOIL PHYSICOCHEMICAL AND MICR OBIAL CHARACTERISTICS Introduction Alum (Al2(SO4)3H2O) is the chemical amendment used most often for phosphorus (P) inactivation in lakes and coagulation in the wastewater treatment industry. When added to the water column alum dissociates, forming aluminum ions (Al3+) that are immediately hydrated. Through se veral rapid hydrolytic reactions an insoluble, gelatinous, poorly crystalline aluminum hydroxide (Al(OH)3) floc is formed (Ebeling et al., 2003). This floc has high P adsorption properties and can remove both soluble and particulate P both by adsorption and physical entrapment (Galarneau and Gehr, 1997). The controlling factor in the effectiveness and toxicity of alum is the pH of the system. Alum itself has a pH of appr oximately 2.4 (Beecroft et al., 1995; Lind, 2003) and therefore tends to decrease th e pH of the system to which it is added. If the pH of the system remains between 6 and 8, insoluble polymeric Al(OH)3 will dominate (May et al., 1979) and P inactivation results. However, if the pH decreases to between 4 and 6 soluble intermediates will occur, releasing bound P(Cooke et al., 1993a). Below pH 4 soluble Al3+ dominates which may result in Al toxi city (Cooke et al., 1993b), and at pH 8 or greater the aluminate ion (Al(OH)4 -) dominates due to its amphoteric nature, releasing bound P and increasing soluble Al (Cooke et al, 1993a). Aluminate, similar to Al3+, is

PAGE 111

92 associated with Al toxicity in plants (Kinra ide, 1990, Eleftheriou et al., 1993; Ma et al., 2003, Malecki-Brown and White, 2007b). The rate of microbial activity and structur e of the microbial community is largely dependent on environmental factors. Bo th size and activity of the microbial pool influences the nutrient removal of a wetla nd (White and Reddy, 1999; White and Reddy 2003) as well as the removal of other emerging contaminants (White et al., 2006a). Microbes are generally sensitive to both soil acidity (Degens et al ., 2001) and soluble Al (Illmer et al., 1995; Robert, 1995; Pina and Cervantes, 1996). The microbial biomass, therefore, has the potential of being a sensitive indicator of impact to soil nutrient dynamics (Powlson and Jenkinson, 1981) due to alum application. This is due to the close relationship between microbial bioma ss nutrients and levels of mineralizable nutrients available in the so il (Jenkison and Ladd, 1981; Illmer et al., 1995; Gutknecht et al., 2006). An alum study by Connor and Martin ( 1989) on a shallow New Hampshire lake suggested that the dissolved oxygen (D.O.) con centrations within the lake may have been affected by suppression in activity or re duction in the populati on of BOD-producing organisms, however neither was measured. Malecki-Brown and White (2007) carried out a laboratory core study utilizing various Al amendments and determined an inverse relationship existed between Al dose and micr obial biomass and activity. Bacterial cells are colloidal particulates, and may therefore be aggregated by the alum floc (Eriksson and Axberg, 1981). This bacterial removal may delay the reestablishment of essential nutrient cycling (Bulson et al ., 1984). Therefore, the size and activity of the microbial

PAGE 112

93 pool needs to be assessed with respect to alum application to fully understand effects on P cycling within a treatment wetland. Additionally, the characterization of soil P is necessary to determine shifts in exchangeable, metal oxide, hydroxide bound, a nd organically bound P pools due to alum application which can not only influence soil microbial mineralization, but also nutrient availability to the aquatic macrophytes. Th e aquatic macrophytes selected for use in treatment wetlands can significantly impact the P treatment efficiency and overall nutrient budget (Tanner, 1996; Reckhow & Chapra 1999; Thiebaut and Muller, 2000; Allen et al., 2002). Macrophyte sp ecies composition can alter the soil biogeochemistry as well (Wigand et al. 1997; White et al., 2004; Ne ubauer et al. 2005; White et al., 2006b). While alum has been used for P inact ivation in eutrophic lakes since 1968 (Blomquist et al., 1971) there has been little research done on its potential effectiveness in aging treatment wetlands with reduced P sorption capacities (Simon, 2003; DB Environmental, Inc., 2004; Malecki-Brown a nd White, 2007). Additionally, there is not a clear comprehension of the impact of incr eased Al concentrations on the biomass and activity of the microbial co mmunity, and therefore nutrient cycling of alum-treated ecosystems. The goal of this study was to in vestigate the effects of alum application on the physicochemical soil characteristics, P forms available, microbial biomass pool size, and microbial activity of treatment wetland s dominated by differing aquatic macrophytes. Materials and Methods Site Description The Orlando Easter ly Wetlands (OEW) R eclamation Project located in Orange County, Florida is one of the oldest and largest constructed treatment wetlands in the United States, located east of Orlando in Christmas, FL. The wetland was built in 1986,

PAGE 113

94 designed by Post, Buckley, Schuh & Jernigan, Inc. (PBS&J) for the City of Orlando’s Iron Bridge Regional Water Pollution Cont rol Facility (WPCF) which needed an alternative discharge point for its wastewater effluent (Burney et al., 1989). The main goal in designing the system was to use m acrophytes to facilitate additional nutrient removal for an average daily flow of up to 132,489 m3d-1 of effluent from the Iron Bridge WPCF before discharging into the St. Johns River (Black and Wise, 2003). The 494 ha wetland rests on a 664 ha piece of land located 3.2 km west of the main channel of the St. Johns River (SJR). Histori cally, the land had been part of the riparian wetland adjacent to the SJR, but was drained fo r cattle pasture around th e turn of the last century (Burney et al, 1989). The site ha s a natural topographic gradient of 4.6 m downward from west to east allowing water to flow by gravity through a series of cells with an average elevation drop across each cell of approximately 1 m (Martinez and Wise, 2003) (Figure 4-1). Water exits the we tland through a weir c ontrol structure and flows into a receiving ditch. From there wa ter can flow directly to the SJR or by sheet flow through Seminole Ranch, a natural marsh adjacent to the OEW owned by the St. Johns River Water Management District (SJR WMD). Cells 1 through 12 and 15 are deep marsh, designed primarily for nutrient remova l, planted with either cattails ( Typha spp.), giant bulrush ( Scirpus californicus ), or a combination of the two. Cells 13, 14, and 16 through 18 consist of a mixed marsh do minated by submergent and emergent macrophytes including Ceratophyllum demersum , Limnobium spongia, Myriophyllum spicatum, Najas guadalupensis, Nuphar luteum , Nymphaea odorata, Pontederia cordata, Sagittaria lancifolia, and Sagittaria latifolia . These cells serve as a diverse wildlife habitat while continuing to provide nu trient removal (Martinez and Wise, 2003).

PAGE 114

95 Figure 4-1. Site map of Orlando Easterly Wetland, Christmas, Florida. Plants collected in cell 1 (Scirpus californicus), cell 10 ( Typha spp.), and cell 13 (Najas guadalupensis ). Mesocosm location designated by star. The overall average influent total P (TP) concentration from 1988 to 2005 was 0.22 mg L-1, however, annual inflow TP concen trations range from 0.02 – 3.30 mg L-1 during the same time period. Since its inception, the OEW has exceeded performance expectations. The TP discharge permit limit established by the FDEP is 0.2 mg L-1 (Wang et al., 2006). From 1988 to 1995 the average TP discharged was 0.07 mg L-1 (Sees and Turner, 1997), however TP values ar e considerably higher from December to February in recent years (Wang et al., 2006). Mesocosm Establishment Eighteen circular me socosms, each with a surfa ce area of 1.86 m2 and depth of Project Location Created by: Lynette Malecki Dated: 12/30/04 2 1 3 4 5 6 7 8 9 10 11 12 15 13 14 16 17 18 Lake

PAGE 115

96 0.88 m were established in June 2004. They in cluded triplicate experimental and control units for Typha spp., Scirpus californicus, and SAV ( Najas guadalupensis dominated) utilizing a randomized bloc k design. Approximately 0.3 m of homogenized soil was added to each mesocosm. The soil used to initiate the mesocosms originated from a dredged spoil pile with a mixture of organi c material from cell 1, 3, 4, 7 and 8 of the OEW (Figure 4-1). Initial so il characteristics included a TP content of 111 13.4 mg kg1, 285 21.2 mg kg-1 oxalate-extractable Al, 19.9% 0.02 moisture content and soil pH of 5.2 0.3. Each mesocosm contained a polyvinyl chloride (PVC) drain which permitted control of water leve ls to within 3 cm. The wa ter flowing through the mesocosms originated from the outflow of cell 15 and was pumped to a head tank and distributed (on a one hour timer) via gravity to each mesocosm at a rate of 360 L d-1 with a hydraulic loading rate of 9.6 cm d-1, providing a retention time of approximately 9.5 days, similar to the retention time of the SAV cells within the OEW. On June 30, 2004, the mesocosms were planted to begin the five month grow-in / stabilization period. The six SAV mesocosm s were established with 4.05 0.06 kg (wet weight) WW/ mesocosm, or 2.18 kg WW m-2 Najas guadalupensis harvested onsite. The stocking density for the Typha spp. and Scirpus californicus mesocosms was 15 plants / mesocosm, averaging 1.88 kg WW m-2 Typha spp., and 1.04 kg WW m-2 Scirpus californicus both harvested onsite (Figure 4-1). A 53 cm water column was maintained in the SAV mesocosms throughout the entire grow-in, while in the emergent mesocosms a 28 cm water column was maintained for the first month to allow the plants to establish before raising the water column to 53 cm for the remainder of the study.

PAGE 116

97 Experiment Initiation On Decem ber 1, 2004, liquid alum was pumped from a head tank via peristaltic pumps on ART-3 repeat cycle timer (120v, 15 amp) to the designated mesocosms through black polyethylene tubi ng at a rate of 0.81 g Al m-2 d-1 for three months resulting in a total addition of 68.2 g Al m-2. Soil physicochemical and microbial characteristics were determined from a core collected from each mesocosm upon initiation and completion of the study. Cores were sectione d into 0-5 cm and 5-10 cm intervals in the field and placed in labeled Ziploc bags in coolers of ice for tran sportation back to the laboratory. Samples were then transferred to polyethylene cont ainers and stored at 4 C for analysis. Laboratory Analysis The following physicochemi cal variables were measured on the sectioned soil samples: pH, bulk density (Blake and Hartge, 1986), mass loss on ignition (LOI), TP and total Al (AlT), oxalate-extractable Al (Alox) (McKeague and Day, 1966), microbial biomass P (MBP), soil oxygen demand (SOD) (APHA, 1992; Fisher and Reddy, 2001; Malecki et al., 2004), potentially minerali zeable P (PMP), inorganic P fractionation (Reddy et al., 1998), 1N HCl – extractable meta ls. Microbial biomass P was determined by a 24 h chloroform fumigation-extracti on (CFE) technique (Brookes et al., 1982; Hedley and Stewart, 1992; Ivanoff et al., 1998) . The PMP rate was determined using an anaerobic, waterlogged incubation at 40 C (Chua, 2000; Maleck i-Brown and White, 2007). Total P analysis involved combustion of 0.5 g oven-dried subsamples at 550 C for 4 h in a muffle furnace followed by dissoluti on of the ash in 6 M HCl on a hot plate (Andersen, 1976). Total P was analyzed using an automated ascorbic acid method

PAGE 117

98 (Method 365.4, USEPA, 1993) while AlT was determined by inductively coupled argon plasma spectrometry (model Vista MPX CCD simultaneous ICP-OES manufactured by Varian, Inc., Walnut Creek, CA) at = 396.152 nm (Barnes, 1975; Campbell et al., 1983; Easthouse et al., 1993; Rydin and Welch, 1999; Rydin et al., 2000; Method 200.7, USEPA, 1993). Ash content was calculated to determine mass loss on ignition (LOI), indicating the organic matter content in the wetland soil (Lim and Jackson, 1982). The amount of Al present in crystalline form was calculated as the difference between the AlT and Alox (Dolui and Chakraborty, 1998). Ac id-extractable Ca, Mg, Fe, and Al concentrations were determined from ovendried soil treated with 25 mL of 1.0 M HCl and placed on a reciprocal shaker for 3 h. The supernatant was filtered through 0.45m membrane filters and analyzed by inductively coupled argon plasma spectrometry (DeBusk et al., 1994; Reddy et al., 1998; Malecki et al., 2004). Statistical Analysis Paired t-tests were used to determin e significant differences (p<0.05) among soil properties in the 0-5 cm and 5-10 cm sect ioned intervals (Microsoft Excel, 2000). Additionally, Pearson product-moment correlation coefficients among variables were calculated (Microsoft Excel, 2000) to determine significant relationships (p<0.05). Data normality was determined using the Kolm ogorov-Smirnov test (Minitab 13.32, 2000) and data were transformed to fit a normal dist ribution (Microsoft Excel, 2000). One-way ANOVAs and multiple comparisons by Tukey’ s W were used on soil variables to determine significant differences (p<0.05) am ong mesocosm plant types and treatment (Minitab 13.32, 2000).

PAGE 118

99 Results and Discussion Soil Physicochemical Characteristics There were no significant di fferences in bulk density or organic ma tter content among mesocosms at the initiation of the st udy. The bulk density was greater in the subsurface 5-10 cm layer (Table 4-2) than in the surface 0-5 cm (T able 4-1) while the opposite was true of the organic matter cont ent as indicated by LOI. Organic matter deposition and decomposition at the soil surf ace resulted in the higher LOI. The increased organic matter in the surface layer allows the soil to remain porous, thereby decreasing the bulk density (Brady and Weil, 1999). Overall, the organic matter content remained constant over the course of the e xperiment which may be attributed to slower plant growth and less organic depos ition during the winter months. Similar to the organic matter, TP concentrations were generally greater in the surface layer (Table 4-1) than subsurface laye r (Table 4-2). Nutrient content typically increased with an increase in organic matte r (Farnham and Finney, 1965). Total soil Al on the other hand was generally greater in the subsurface layer at the start of the experiment, however by the conclusion of the experiment AlT tended to be greater in the surface layer. There were no signif icant differences in the TP or AlT concentrations among mesocosm plant types in either layer throughout the experiment. Furthermore, soil TP remained unaffected by alum application while AlT concentrations significantly increased in the surface layer of alum-tre ated mesocosms by the end of the study as would be expected. The ammonium oxalate extraction was used to quantify the poorly crystalline amorphous and organically bound Al oxides, h ydroxides, and oxy-hydroxides in the soil (McKeague and Day, 1966; McKeague et al., 1971; Kodama and Ross, 1991), excluding

PAGE 119

100Table 4-1. Mean soil physicochemical char acterization data for the 0-5 cm depth of Orlando Easterly Wetland mesocosms. Values are means 1 standard deviation (n=3). Treatment Parameter Units Time 0d Time 84d SAV Scirpus Typha SAV Scirpus Typha Alum Bulk Density g cm-3 0.89 0.16 0.89 0.15 0.89 0.09 0.94 0.26 0.91 0.31 0.83 0.07 pH pH units 5.95 0.43 5.73 0.76 6.17 0.02 5.56 1.09 5.61 0.58 3.83 0.63 LOI % 8.24 0.77 12.0 3.45 10.6 3.20 5.77 2.34 4.56 1.73 9.20 3.94 Total P mg kg-1 196 155 241 133 175 9.44 161 42.9 173 79.1 201 44.0 Total Al mg kg-1 1871 731.8 2356 879.0 2839 370.8 3544 2290 1989 247.3 4214 1126 Oxalate Al mg kg-1 418 16.6 610 252 688 126 1148 477 916 355 2157 546 HCl Ca mg kg-1 1468 328.6 4156 2500 3220 871.6 3383 766.1 8395 6379 10620 6486 HCl Mg mg kg-1 26.0 13.6 46.1 19.8 57.5 36.6 64.9 54.6 69.5 12.7 104 55.7 HCl Fe mg kg-1 587 314 791 112 793 218 706 430 972 402 826 244 HCl Al mg kg-1 431 114 588 32.3 528 102 566 335 672 127 531 154 Control Bulk Density g cm-3 0.92 0.08 0.94 0.19 0.94 0.11 0.98 0.1 0.81 0.31 0.99 0.03 pH pH units 6.06 0.42 6.01 0.51 6.07 0.30 5.79 0.05 5.72 0.28 5.87 0.36 LOI % 9.71 3.55 9.26 1.81 12.0 4.81 4.17 2.19 7.01 3.08 4.14 1.42 Total P mg kg-1 118 25.1 128 28.5 220 91.6 155 31.3 235 77.8 146 39.6 Total Al mg kg-1 2058 539.3 1579 494.9 2299 500.2 1823 498.7 2455 1986 2210 884.8 Oxalate Al mg kg-1 580 156 536 162 588 127 611 62.9 612 360 751 207 HCl Ca mg kg-1 3967 3309 6335 3419 3092 1202 2432 788.7 4205 4179 1778 332.1 HCl Mg mg kg-1 36.5 17.2 133 149 38.7 41.5 68.6 82.2 63.8 58.3 10.6 16.0 HCl Fe mg kg-1 934 86.1 561 320 843 191 891 324 676 82.9 718 301 HCl Al mg kg-1 559 78.5 394 116 564 86.5 1158 830 612 688 1002 438 SAV = submerged aquatic vegetation; LOI = loss on ignition.

PAGE 120

101Table 4-2. Mean soil physicochemical char acterization data for the 5-10 cm depth of Orlando Easterly Wetland mesocosms. Value s are means 1 standard deviation (n=3). Treatment Parameter Units Time 0d Time 84d SAV Scirpus Typha SAV Scirpus Typha Alum Bulk Density g cm-3 1.06 0.07 1.01 0.13 1.04 0.04 1.08 0.08 1.02 0.17 1.07 0.30 pH pH units 6.06 0.63 5.67 0.66 5.90 0.18 6.87 0.34 7.10 0.07 5.15 0.99 LOI % 11.5 6.24 10.5 1.54 7.17 3.49 5.72 4.36 3.91 1.18 4.51 3.47 Total P mg kg-1 227 134 190 21.8 123 16.4 129 65.6 161 23.9 112 38.1 Total Al mg kg-1 2001 1235 2555 266 2071 626 1506 285.4 1673 668.1 1705 516.6 Oxalate Al mg kg-1 672 90.2 747 96.4 567 252 635 107 567 202 690 243 HCl Ca mg kg-1 1467 318.6 2724 164.9 2434 245.4 1710 462.1 2192 1739 2210 603.3 HCl Mg mg kg-1 23.1 7.78 65.9 43.2 47.2 20.8 53.8 30.5 31.4 46.0 41.7 29.2 HCl Fe mg kg-1 444 179 864 332 899 166 546 269 641 392 761 261 HCl Al mg kg-1 345 73.2 519 153 607 35.7 380 143 1173 1102 574 118 Control Bulk Density g cm-3 1.09 0.04 1.06 0.05 1.16 0.05 0.97 0.27 1.21 0.21 1.19 0.20 pH pH units 6.10 0.28 5.75 0.30 6.12 0.47 6.76 0.16 6.76 0.46 6.39 0.59 LOI % 8.18 3.46 9.12 0.88 11.0 3.19 6.64 2.30 6.48 2.45 2.68 1.77 Total P mg kg-1 124 26.1 100 15.3 231 135 194 162 209 110 114 39.9 Total Al mg kg-1 1699 766.5 1556 587.8 2382 291.9 2699 2165 1706 421.7 1568 109.4 Oxalate Al mg kg-1 631 241 527 181 588 269 613 279 501 174 576 177 HCl Ca mg kg-1 3221 1357 3998 2772 4569 2196 3051 2931 6305 5199 10193 7784 HCl Mg mg kg-1 72.1 54.4 58.9 65.1 38.0 12.8 43.2 44.6 41.7 25.7 71.4 59.7 HCl Fe mg kg-1 869 344 764 457 742 196 923 345 655 404 774 130 HCl Al mg kg-1 581 208 566 328 520 105 873 531 472 153 744 297 SAV = submerged aquatic vegetation; LOI = loss on ignition.

PAGE 121

102 all crystalline forms (Parfitt and Childs, 1988). At the initiation of the experiment the Alox concentrations tended to vary between th e surface (Table 4-1) and subsurface layer (Table 4-2) with no significant difference between layers, averaging 595 167 mg kg-1 overall. However, by the end of the expe riment amorphous Al concentrations were significantly higher, with n early twice as much Al in the 0-5 (1032 634 mg kg-1) than 510 cm layer (593 200 mg kg-1). This was expected in the mesocosms receiving alum due to the surface application of Al, but it was also true of the control mesocosms due to the natural Al associated with organic matter. There were no differences in Alox due to plant type upon st udy completion. Similar to the AlT, however, the alum-treated mesoco sms did have significantly higher Alox concentrations in the surface layer, containi ng more than double the amorphous Al as the controls. Specifically, the alum-treated Typha tanks had significantly higher surface Alox concentrations than the controls of all three plant type. Oxalate-extractable Al concentrations did not differ between experi mental and control mesocosms in the 5-10 cm layer similar to the AlT, indicating the Al(OH)3 floc remained at the soil surface. Similar to the Alox, soil pH was variable between the surface and subsurface layer (Table 4-1, Table 4-2) at the start of the experiment with no signi ficant differences in either layer nor among mesocosms. By the end of the experiment, the pH was significantly lower in the surface soil, averaging 5.9 0.2 in the controls and 5.0 1.1 in the alum-treated mesocosms. In contrast, th e subsurface soil pH was significantly higher than at the start of the experiment. Corresponding to the high oxalate-extractable Al in the surfac e soil of the alumtreated Typha tanks, the pH was signifi cantly lower than in the Typha control

PAGE 122

103 mesocosms. Unlike the trend in Al, however, the subsurface soil pH of the alum-treated Typha tanks was also significantly lower than th e controls suggesting that while the floc itself may not have penetrated down through th e soil profile, the effects of alum may have. The pH was negativ ely correlated to the Alox in both the surface (p<0.01) and subsurface (p<0.05) of the emergent mesocosm s while this relationship was not evident in the SAV mesocosms. The thick stands of SAV may have intercepted the settling alum floc resulting in irregular alum coverage on th e soil surface. This could in turn lead to reduced alum effectiveness (Welch and Schrieve, 1994; Welch and Cooke, 1999). There was no initial trend in crystalline Al with depth, however by the end of the study there was generally more in the surface layer than subsurface. Crystalline Al concentrations within the soil did not vary wi th plant type nor treatment in either layer throughout the experiment, averaging 1483 813 mg Al kg-1 overall. Using the change in ratio of Alox / AlT over time however, the crystallizati on of soil Al was detected in the surface layer of the control mesocosms as i ndicated by a significant decrease in the Alox / AlT ratio (Mahaney et al., 1991; Bera et al., 2005). This suggests that the Al(OH)3 floc at the surface of the alum-treated mesocosm s may hinder natural soil development due to an increasing amount of am orphous material which inhi bits crystallization. All HCl-extractable metals were significantly higher in the surface than subsurface, similar to the TP and LOI, and there were no differences due to plant type throughout the study. Calcium was the dominant HCl-extractable metal at time 0 in both the soil surface and subsurface for all plant types making up 72% of the metals extracted on average, which was significantly higher than the Al, Fe, and Mg extracted. There was significantly more Fe (17%) a nd Al (10%) than Mg (1%) pr esent in the mesocosms as

PAGE 123

104 well. By the completion of the study there wa s significantly more Ca in the surface layer of the alum-treated mesocosms while in th e subsurface layer Ca was significantly lower than in the controls. Calcium remained the dominant acid-extractable metal composing 73% of the total metals extracted comp ared to 15% Fe, 11% Al, and 1% Mg. Soil Microbial Characteristics There were no significant differences in MBP with depth throughout the study (Table 4-3, Table 4-4). Howe ver general trends at the st art indicated the mi crobial biomass was greater in the s ubsurface of the SAV mesocosm s and the surface layer in the emergent mesocosms. By the end of th e study there was a significant decrease in microbial biomass in all mesocosms which is attributed to the cold winter season during which the experiment was carried out. The mesocosms treated with alum had significantly lower MBP concentrations than th e controls in both the surface (66% lower) and subsurface (50% lower) soil layers. These results are similar to the pH findings, indicating once again that while the floc itself remained at th e surface, the effects of the alum penetrated down through the soil profile. Additionally, simila r to the soil pH, MBP was inversely related to the Alox present in both soil layers (p<0.05). While there were no Table 4-3. Mean soil microbial characteriza tion data for the 0-5 cm depth in Orlando Easterly Wetland mesocosms. Values ar e means 1 standard deviation (n=3). Treatment Plant Time 0d Time 84d MBP PMP SOD MBP PMP SOD mg kg-1 mg kg-1 d-1 mg kg-1h-1 mg kg-1 mg kg-1 d-1 mg kg-1h-1 Alum SAV 25.7 15.0 0.60 0.29 6.33 1.03 1.62 0.93 1.67 1.41 7.36 1.64 Scirpus 30.8 12.3 0.62 0.35 3.43 2.03 1.51 1.05 0.32 0.19 5.25 2.13 Typha 25.2 3.10 0.72 0.08 10.1 4.450.38 0.44 0.60 0.21 6.21 0.69 Control SAV 24.6 4.91 0.52 0.32 5.15 1.79 1.99 0.87 1.18 0.21 10.8 2.84 Scirpus 37.1 8.58 0.93 0.53 6.71 2.71 5.73 1.15 1.46 0.10 8.95 1.68 Typha 32.9 9.95 0.64 0.08 5.16 3.72 2.86 1.53 1.88 1.30 7.51 4.39 MBP = microbial biomass P; PMP = potentially mineralizable phosphorus; SOD = soil oxygen demand; SAV = submerged aquatic vegetation.

PAGE 124

105 Table 4-4. Mean soil microbial characteriz ation data for the 5-10 cm depth of Orlando Easterly Wetland mesocosms. Values are means 1 standard deviation (n=3). Treatment Plant Time 0d Time 84d MBP PMP SOD MBP PMP SOD mg kg-1 mg kg-1 d-1 mg kg-1h-1 mg kg-1 mg kg-1 d-1 mg kg-1h-1 Alum SAV 50.7 37.5 0.85 0.34 3.66 0.90 2.28 1.64 2.46 3.35 5.26 1.15 Scirpus 33.5 8.70 0.38 0.17 3.41 0.81 2.20 0.20 1.65 2.36 3.70 1.36 Typha 20.2 1.77 1.98 1.52 3.66 1.06 0.89 0.85 0.61 0.02 4.42 1.22 Control SAV 34.4 14.3 0.46 0.27 2.80 0.95 2.93 2.07 1.09 0.68 5.13 1.74 Scirpus 20.8 6.69 1.13 1.02 3.91 1.15 5.45 3.33 0.96 0.92 3.79 2.42 Typha 35.8 13.3 0.72 0.43 3.50 1.82 2.48 0.43 1.61 0.55 4.52 1.42 MBP = microbial biomass P; PMP = potentially mineralizable phosphorus; SOD = soil oxygen demand; SAV = submerged aquatic vegetation. significant differences with depth, there was a clear trend of higher MBP in the 0-5 cm layer of the controls while in the alum-treat ed mesocosms the 5-10 cm layer had a greater MBP. This suggests the microbes either migrated downward in an attempt to avoid the effects of alum such as the low pH, high Al, or nutrient limitations (Table 4-1) (Koepple et al., 1997; Zhou, 2002), which would explain the increased MBP in the subsurface of the alum-treated mesocosms as compared to their surface soil; or alum reduced the microbial population present in the surface soil. Microbial activity, as indicated by SOD ra tes was generally greater in the surface than subsurface layer at the start of the study while the PMP rates were variable (Table 43, Table 4-4). There were no significant di fferences in SOD or PMP among mesocosms at the start of the experiment in either soil la yer. At the end of the experiment there were no significant differences ba sed on plant type for either SOD or PMP, however both parameters indicated significantly greater mi crobial activity in the 0-5 cm layer of the control mesocosms than in the alum-treated mesocosms. This suppression in activity corresponds to the significant decrease in micr obial biomass due to alum application.

PAGE 125

106 There were no significant di fferences between control and alum-treated mesocosms in microbial activity in the subsurface laye r, nor were there significant differences between depths. Similar to the trend in MB P however, the PMP rates were greater in the surface of the controls while in the alum-t reated mesocosms the subsurface layer had higher rates. The surface soil PMP rates were positively correlated (p<0.05) with soil pH similar to the MBP concentrations. This sugge sts that alum indirec tly impacted both the microbial biomass and activity due to the sens itivity of microbes to the increased soil acidity (Degens et al., 2001). Soil Phosphorus Forms Organic and inorganic P were essentially equivalent in all me socosms throughout the study averaging 55% organic P (Po) (Table 4-6) and 45% inorganic P (Pi) (Table 4-5) in the 0-5 cm layer. There were no significan t differences with treatment or plant type. All P fractions were greater in the surface la yer than in the subsurface (Table 4-7, Table 4-8) similar to the TP values previously di scussed. The overall decr ease in P storage Table 4-5. Mean inorganic phosphorus fracti onation data from the 0-5 cm layer of Orlando Easterly Wetland mesocosms. Values are means 1 standard deviation (n=3). Treatment Fraction Time 0d Time 84d mg kg-1 SAV Scirpus Typha SAV Scirpus Typha Alum KCl Pi 0.43 0.07 0.41 0.07 0.38 0.04 0.23 0.11 0.16 0.08 0.26 0.16 NaOH Pi 27.3 6.39 28.4 12.6 41.8 7.03 32.8 5.30 35.4 2.25 66.7 6.56 HCl Pi 31.6 39.1 64.1 60.9 21.6 1.85 29.4 33.9 53.2 55.1 6.42 2.50 Total Pi 59.3 43.6 92.9 49.7 63.8 6.81 62.4 30.0 88.8 57.1 73.4 7.95 Control KCl Pi 0.42 0.09 0.49 0.13 0.45 0.13 0.26 0.20 0.31 0.20 0.20 0.03 NaOH Pi 27.9 6.96 29.3 11.0 25.0 7.39 31.5 9.90 45.8 27.3 41.3 16.5 HCl Pi 26.0 10.8 33.2 25.2 38.5 26.4 53.2 56.0 71.4 82.8 21.2 0.97 Total Pi 54.3 5.05 63.0 17.9 64.0 23.6 85.0 47.7 117 62.9 62.7 16.0 SAV = submerged aquatic vegetation.

PAGE 126

107 Table 4-6. Mean organic phosphorus derive d from inorganic phos phorus fractionation data from the 0-5 cm layer of Orlando Easterly Wetland mesocosms. Values are means 1 standard deviation (n=3). Treatment Fraction Time 0d Time 84d mg kg-1 SAV Scirpus Typha SAV Scirpus Typha Alum NaOH Po 33.5 5.24 29.6 12.3 44.7 7.73 37.5 8.37 37.9 3.03 69.3 5.12 Residue Po 34.1 20.4 48.0 14.3 59.5 4.02 53.7 19.8 44.0 14.5 48.9 4.61 Total Po 67.6 23.2 77.6 26.5 104 11.5 91.2 11.5 82.0 16.9 118 6.58 Control NaOH Po 32.7 8.23 45.8 12.5 78.5 20.7 34.9 10.3 49.4 28.6 45.2 15.4 Residue Po 35.1 12.3 39.9 4.74 75.0 16.9 47.8 8.21 66.4 38.9 50.1 18.7 Total Po 36.6 11.4 46.4 9.06 83.0 20.5 82.7 17.7 116 66.6 95.3 33.9 SAV = submerged aquatic vegetation. Table 4-7. Mean inorganic phosphorus fracti onation data from the 5-10 cm layer of Orlando Easterly Wetland mesocosms. Values are means 1 standard deviation (n=3). Treatment Parameter Time 0d Time 84d mg kg-1 SAV Scirpus Typha SAV Scirpus Typha Alum KCl Pi 0.40 0.06 0.39 0.10 0.31 0.10 0.21 0.03 0.22 0.10 0.20 0.05 NaOH Pi 20.0 5.87 33.6 8.57 33.8 6.87 23.1 2.36 26.6 4.18 36.1 12.7 HCl Pi 86.4 72.2 24.0 19.5 11.1 3.68 66.0 81.5 61.1 26.0 11.5 0.38 Total Pi 107 73.3 58.0 12.5 45.1 6.67 89.2 81.5 87.9 21.9 47.8 12.4 Control KCl Pi 0.43 0.14 0.38 0.10 0.41 0.16 0.20 0.09 0.18 0.07 0.19 0.07 NaOH Pi 25.9 6.72 32.0 7.13 24.2 8.81 36.9 21.0 30.7 17.9 30.7 12.2 HCl Pi 25.0 18.6 16.0 9.61 90.3 97.0 20.0 1.49 93.8 138 19.5 13.6 Total Pi 51.3 12.1 48.4 3.04 115 88.3 57.2 20.6 125 120 50.4 16.2 SAV = submerged aquatic vegetation. Table 4-8. Mean organic phosphorus derive d from inorganic phos phorus fractionation data from the 5-10 cm layer of Orlando Easterly Wetland mesocosms. Values are means 1 standard deviation (n=3). Treatment Fraction Time 0d Time 84d mg kg-1 SAV Scirpus Typha SAV Scirpus Typha Alum NaOH Po 26.3 8.21 37.9 7.38 37.9 5.49 27.4 4.33 28.4 4.65 38.9 14.1 Residue Po 42.4 20.4 48.0 2.85 35.9 7.31 37.0 8.52 41.9 9.37 36.3 7.27 Total Po 68.7 24.0 85.8 4.88 73.8 12.7 64.4 10.7 70.3 14.0 75.1 20.2 Control NaOH Po 29.6 7.39 35.6 5.76 26.1 9.39 42.1 22.8 32.0 17.7 32.9 10.3 Residue Po 39.8 12.1 37.4 4.16 41.5 17.2 56.2 30.4 43.8 10.8 38.8 0.91 Total Po 69.4 16.4 72.9 6.90 67.6 25.3 98.3 51.9 75.8 26.7 71.8 9.44 SAV = submerged aquatic vegetation.

PAGE 127

108 capacity with depth may be attributed to the significant decrease in metal concentrations (Al, Ca, Fe, Mg) with depth (Table 4-1, Table 4-2). In the surface layer, the KCl–extractable Pi, consisting of labile, readily bioavailable P, made up the significantly sma llest portion (0.1 – 0.4%) of the total P pool (Figure 4-2). The NaOH-extract able reactive Al and Fe bound Pi comprised 17-25 % of the total P in the mesocosms initially. Due to alum application this percentage increased to 21-35% in the treated mesocosms while in the controls fractions remained similar to the initial values. The HClextractable Ca and Mg bound Pi made up 12-38% of the total P pool initially and remained re latively stable averaging 3-32 % at the end of the study. There was very little HCl-extractable Mg in the soil (Table 4-1) indicating most of the P in the HCl Pi fraction was associated with Ca from the buffered wetland soil, rather than Mg. Additionally, because this Ca bound P frac tion did not increase with time it suggests that although the water column pH was relatively high and Ca concentrations increased over time, photosynthetically driven Ca-P pr ecipitation (Dierberg et al., 2002) did not occur. The organic P fractions consisted of NaOH-extractable non-reactive Po associated with humic and fulvic acids as well as bact eria incorporated P which accounted for 1727% of the total P in the surf ace layer of all mesocosms at the start of the experiment. Alum application resulted in an increase in this percentage to 22-36% in the treated mesocosms (Figure 4-2), nearly e quivalent to the amount of NaOH Pi, while the control mesocosms once again remained similar to the initial values. The residue Po representing the refractory organic P and any other inert mineral P fractions not extracted with salt,

PAGE 128

109 acid, or base composed 26-35% of the total P, remaining relatively constant throughout the study. TIME 0 d TIME 84 d 0-5 cm KCl Pi 0.41 0.06 (0.0%) KCl Pi 0.22 0.11 (0.0%) NaOH Pi 32.2 11.3 (21%) NaOH Pi 39.0 13.6 (23%) HCl Pi 41.1 43.4 (26%) HCl Pi 44.9 58.7 (26%) NaOH Po 36.0 11.0 (23%) NaOH Po 43.0 13.6 (25%) Residue Po 47.2 17.6 (30%) Residue Po 44.8 8.49 (26%) Total P 155 42.8 (100%) Total P 172 45.8 (100%) 5-10 cm KCl Pi 0.37 0.09 (0.0%) KCl Pi 0.23 0.12 (0.0%) NaOH Pi 29.1 9.23 (20%) NaOH Pi 28.3 8.97 (20%) HCl Pi 40.5 51.2 (28%) HCl Pi 42.4 38.6 (30%) NaOH Po 34.0 8.45 (23%) NaOH Po 31.1 9.14 (22%) Residue Po 42.1 12.1 (29%) Residue Po 38.4 8.41 (27%) Total P 146 49.9 (100%) Total P 145 49.3 (100%) Figure 4-2. Inorganic and organic phosphorus forms in the surface and subsurface soil of alum-treated Orlando Easterly Wetland me socosms at the start and end of the study. Values are means (mg kg-1) 1 standard deviation (n=9) with percent of total P pool in parenthesis. The initial subsurface 5-10 cm soil had nearly equal amounts of organic and inorganic P similar to the surface layer averaging 52% Po (Table 4-7) and 48% Pi (Table 4-8) while it was dominated by or ganic P by the end averaging 64% Po versus 36% Pi. Labile P once again represente d 0.1-0.4% of the total P, id entical to the surface soil (Figure 4-2). The Al and Fe bound P made up 11-28% of the subsurface P pool initially, and only increased slightly in the alum-treat ed mesocosms (15-29%). This indicates once again that the alum floc remained primarily at the surface of the soil where substantial shifts in the NaOH Pi and Po pools were evident. The dominant inorganic P form in the subsurface layer of most mesocosms was the Ca and Mg bound P which ranged from 950% of the total P pool which, similar to the surface soil, can be primarily attributed to

PAGE 129

110 the significantly high Ca concentration (Table 4-2). The organic acid and bacterial P fraction made up 15-32% of the total P thr oughout the study in the subsurface, while the recalcitrant Po composed 22-36% of the total P. Mass Balance of Phosphorus Utilizing the soils data presented here as well as the water quality and aquatic macrophyte data presented in the previous chapter (Malecki-Brown and White, 2007b) a P mass balance can be calculated. The seve n components of the mass balance calculated included the inflow TP, 0-5 cm surface soil , 5-10 cm subsurface soil, microbial-bound P in surface soil, microbial-bound P in subsurface soil, aboveground plant biomass, and outflow TP (Figure 4-3, Figure 4-4). All me socosms received the same inflow water but the mesocosms treated with alum removed on average 0.058 g P d-1 from the water column while the control mesocosms averaged approximately 0.043 g P d-1 removal. Both water column and microbial biomass P storage were insignificant, comprising less than 1% of the total P within both th e alum-treated and control mesocosms. The largest component within the SAV mesocosms was the aboveground biomass averaging 44% of the TP in the alum-treated SAV tanks and 52% in the controls (Figure 4-3). This is nearly double the am ount of biomass P present in the Scirpus and Typha emergent aquatic vegetation (EAV) and may result in a reduction in P treatment with time. As the SAV detritus accumulates and undergoes mineralization a substantial amount of P will be released back into the water column which may decrease treatment efficiency. In the EAV mesocosms, on the ot her hand, P was primarily present in the soil components averaging approximately 35% of th e total P in each layer for both the alum and control mesocosms (Figure 4-4).

PAGE 130

111 Figure 4-3. Phosphorus mass balance in th e submerged aquatic vegetation mesocosms after three months. Values are percent of total mesocosm P, with mean value (g) 1 standard deviation in parenthesi s (n=3). Inflow and outflow P values are in g d-1. Figure 4-4. Phosphorus mass balance in em ergent aquatic vegetation mesocosms ( Typha and Scirpus ) after three months. Values are perc ent of total mesocosm P, with mean value (g) 1 standard deviation in parenthesis (n=6). Inflow and outflow P values are in g d-1. Alum: 0.1% (0.04.01) Control: 0.1% (0.05.03) Alum: 29% (14.5.32) Control: 23% (14.0.20) 5-10 cm Alum: 27% (13.3.81) Control: 25% (14.9.13) 0-5 cm Alum: 0.2% (0.09.03) Control: 0.2% (0.09.03) Alum: 44% (22.1.3)Control: 52% (30.8.26) Alum: 0.3% (0.09.03) Control: 0.2% (0.09.03) 5-10 cm Alum: 35% (13.3.49) Control: 35% (18.03.1) 0-5 cm Alum: 0.1% (0.03.01) Control: 0.1% (0.05.02) Alum: 38% (14.4.22) Control: 31% (15.8.63) Alum: 27% (10.5.65)Control: 34% (17.34.6)

PAGE 131

112 Interestingly, there was approximately 8% less P stored in the alum-treated SAV biomass than in the controls which corresponded to the alum-treated tanks containing approximately 6% more P in the surf ace soil as a result of the Al(OH)3 floc and 2% more in the subsurface soil than the control mesocosms. Similarly, there was approximately 7% less P stored in the alum-t reated EAV biomass than in the controls wh ich directly corresponded to 7% more P present in the surf ace soil as a result of the surface floc in the alum-treated EAV mesocosms. Therefore, while there was little evidence of alum affecting the emergent macrophytes in Malecki-Brown and White (2007b) on a mass basis it appears that there may have been a re duction in P availability to the plants from the soil. The reduction in alum-treated SAV biomass P resulted from the Al toxicity that occurred in the SAV (Malecki-Brown and White, 2007b). Conclusions The use of alum is prolific in m any areas of the United States to manage eutrophic water bodies. The findings of this research suggest a closer look is needed at the resulting changes that may be occurring in th e biogeochemistry of the soil or sediment treated with alum. The P mass balance s uggests ecosystems dominated by SAV will primarily see alum impacts associated with the vegetation while in systems dominated with emergent macrophytes, the soil and asso ciated microbes will be most impacted. This directly corresponds to the largest P compartments in each system, thus the use of alum may completely alter the P nutrient dynamics of a system in the short term and it is critical that treatment wetland managers understand the short and possibly long term implications when using alum in wetland systems.

PAGE 132

113 The increase in amorphous Alox within the surface layer of the soil was directly associated with a decrease in soil pH th roughout the upper 10 cm of soil. The increased Alox concentration and reduced pH both in turn negatively impacted the microbial biomass as well as activity. Furthermore, the addition of alum seemed to hinder Al crystallization processes with in the surface soil, leading to reduced development in the alum-treated mesocosms when compared to th e controls. More research is needed to determine the long-term impacts of lowdose alum application on the microbial community as well as overall nutrien t cycling within treatment wetlands.

PAGE 133

114 CHAPTER 5 SPATIAL AND TEMPORAL EF FECTS OF CONTINUOUS LOW -DOSAGE ALUM ON FIELD SOIL CHARACTERISTICS Introduction Alum (Al2(SO4)3H2O) is the chemical amendment used most often for phosphorus (P) inactivation in lakes and coagulation in the wastewater treatment industry. When added to the water column alum dissociates, forming aluminum ions (Al3+) that are immediately hydrated. Through se veral rapid hydrolytic reactions an insoluble, gelatinous, poorly crystalline aluminum hydroxide (Al(OH)3) floc is formed (Ebeling et al., 2003). This floc has high P ad sorption properties, able to remove both soluble and particulate P both by adsorption and physical entrapment (Galarneau and Gehr, 1997). The controlling factor in the effectiveness and toxicity of alum is the pH of the system. Alum itself has a pH of appr oximately 2.4 (Beecroft et al., 1995; Lind, 2003) and therefore tends to decrease the pH of the system to which it is added. As long as the pH of the system remains between 6 and 8, insoluble polymeric aluminum hydroxide (Al(OH)3) will dominate (May et al., 1979) and P inactivation results. If the pH decreases to between 4 and 6 soluble interm ediates will occur, releasing bound P(Cooke et al., 1993a). Below pH 4 soluble Al3+ dominates which may result in Al toxicity (Cooke et al., 1993b), and at pH 8 or greater the aluminate ion (Al(OH)4 -) dominates due to its amphoteric nature, re leasing bound P and increasing soluble Al (Cooke et al,

PAGE 134

115 1993a). Aluminate, similar to Al3+, is associated with Al toxi city in plants (Kinraide, 1990, Eleftheriou et al., 1993; Ma et al., 2003; MaleckiBrown and White, 2007b). While alum has been used for P inact ivation in eutrophic lakes since 1968 (Blomquist et al., 1971) there has been little research done on its potential effectiveness in aging treatment wetlands with reduced P sorption capacities (Simon, 2003; DB Environmental, Inc., 2004; Malecki-Br own and White, 2007a; Malecki-Brown and White, 2007b; Malecki-Brown et al., 2007a). Additionally, there is not a clear comprehension of the impact of increased Al concentrations on the biomass and activity of the microbial community, and therefore nutri ent cycling of alum-treated ecosystems. The rate of microbial activity and structur e of the microbial community is largely dependent on environmental factors. Bo th size and activity of the microbial pool influences the nutrient removal of a wetla nd (White and Reddy, 1999; White and Reddy 2003) as well as the removal of other emerging contaminants (White et al., 2006a). Microbes are generally sensitive to both soil acidity (Degens et al ., 2001) and soluble Al (Illmer et al., 1995; Robert, 1995; Pina and Cervantes, 1996). The microbial biomass, therefore, has the potential of being a sensitive indicator of impact to soil nutrient dynamics (Powlson and Jenkinson, 1981) due to alum application. This is due to the close relationship between microbial bioma ss nutrients and levels of mineralizable nutrients available in the so il (Jenkison and Ladd, 1981; Illmer et al., 1995; Gutknecht et al., 2006). An alum study by Connor and Martin ( 1989) on a shallow New Hampshire lake suggested that the dissolved oxygen (D.O.) con centrations within the lake may have been affected by suppression in activity or reduc tion in the population of microorganisms,

PAGE 135

116 however neither was measured. Malecki-Brown and White (2007a) carried out a laboratory core study utilizing various Al amendments and determined an inverse relationship existed between Al dose and micr obial biomass and activity. In contrast, several studies of phytoplankton in acidic lakes have shown increased phosphatase production to compete for phosphate in the pres ence of high Al concentrations (Jansson, 1981; Bittl et al., 2001). It is possible that alum application may cause similar results within the water column or surface soil, stimulating microbial activity or enzyme production in order to compete for the organically-bound P. Therefore, the objective of this study was to assess the size and activity of the microbial pool, with respect to alum application, to better understand effects of alum on P cycling both spatially and temporally within a treatm ent wetland. Additionally the characterization of soil P was n ecessary to determine shifts in exchangeable, metal oxide, hydroxide-bound, and organically-bound P pools due to alum application which may not only influence soil microbial mineralization, but also nutrient avai lability to aquatic macrophytes. The hypothesis was that long-t erm application of low-dose alum would affect the bioavailabil ity of metals and nutrients within the soil, and in turn impact microbial biomass and activity within a c onstructed wastewater treatment wetland. Materials and Methods Site Description The Orlando Easter ly Wetlands (OEW) R eclamation Project located in Orange County, is one of the oldest and largest c onstructed treatment wetlands in the United States, located east of Orlando in Christ mas, FL. The wetland was built in 1986, designed by Post, Buckley, Schuh & Jernigan, In c. for the City of Orlando’s Iron Bridge Regional Water Pollution Control Facility (WPCF) which needed an alternative discharge

PAGE 136

117 point for its wastewater effluent (Burney et al., 1989). The main goal in designing the system was to use macrophytes to facilitate additional nutrient removal for an average daily flow of up to 132,489 m3d-1 of effluent from the Iron Bridge WPCF before discharging into the St. Johns River (Black and Wise, 2003). The 494 ha wetland rests on a 664 ha piece of land located 3.2 km west of the main channel of the St. Johns River (SJR). Histori cally, the land had been part of the riparian wetland adjacent to the SJR, but was drained for pasture by a cattle ranch around the turn of the last century (Burney et al., 1989). The site has a natural topographic gradient of 4.6 m downward from west to eas t allowing water to flow by gravity through a series of cells with an average elevation drop across e ach cell of approximately 1 m (Martinez and Wise, 2003) (Figure 5-1). Water exits the we tland through a weir c ontrol structure and flows into a receiving ditch. From there wa ter can flow directly to the SJR or by sheet flow through Seminole Ranch, a natural marsh adjacent to the OEW owned by the St. Johns River Water Management District. The overall average influent TP co ncentration from 1988 to 2005 was 0.22 mg L-1, however, annual inflow TP concentr ations ranged from 0.02 – 3.30 mg L-1 during the same time period. Since its inception, the OEW has exceeded performance expectations. The TP discharge permit limit established by th e Florida Department of Environmental Protection is 0.2 mg L-1 (Wang et al., 2006). From 1988 to 1995 the average TP discharged was 0.07 mg L-1 (Sees and Turner, 1997), however TP values have been considerably higher from December to February in recent years (Wang et al., 2006).

PAGE 137

118 Figure 5-1. Figure 5-1. Site map of Orlando East erly Wetland, Christmas, Florida. Typha spp. dominated cell 9 served as the control, and ce ll 10 was treated with alum for one year. Transect Establishment May 23, 2005 duplicate belt transects 2.5 m wide were established in each cell stretching from the inflow weir on the southw est of each cell to th e two outflow weirs on the northeast end. Each transect was divide d into four plots spaced to capture any resultant gradients formed from alum applicati on. The plots were established at 0-10 m, 10-35 m, 35-80 m, and 80-150 m from the inflow and marked using 2.5 m polyvinyl chloride poles that were pounded into the so il using dead-blow mallets and remained throughout the study (Figure 5-2). Project Location Created by: Lynette Malecki Dated: 12/30/04 2 1 3 4 5 6 7 8 9 10 11 12 15 13 14 16 17 18 Lake

PAGE 138

119 Figure 5-2. Map of study transe cts established in cell 9 and 10 of the Orlando Easterly Wetland, Christmas, Florida. Experiment Initiation Beginning August 3, 2005, commercial grade liquid alum (General Chemical Corp.) was pumped via a solar-charged chemi cal injection pump to the inflow of cell 10 through volume regulated black polyethylene tu bing at an average rate of 0.24 g Al m-2 d1 for one year resulting in a total addition of 88.6 g Al m-2. Water samples were collected weekly for the first three months, then every other week for the remainder of the study. Created by: Lynette Malecki Brown Dated: 12/11/06 9 10

PAGE 139

120 In addition, plant and soil samples were collected at time 0 (Aug. 2005), 4 mo. (Dec. 2005), 8 mo. (April 2006), and 12 mo. (Aug. 2007). Triplicate soil cores were collected from each transect plot and sectioned into 0-5 cm and 5-10 cm intervals for a total of 96 samples. Soil was placed in labeled Ziploc bags within ice-filled coolers for transpor t to the laboratory. Samples were then transferred to polyethylene containers and refr igerated at 4 C for characterization. The following physicochemical variables were meas ured on the sectioned soil samples: pH, bulk density (Blake and Hartge, 1986), mass loss on ignition (LOI), TP and total Al (AlT), oxalate-extractable Al (Alox) (McKeague and Day, 1966), microbial biomass P (MBP), soil oxygen demand (SOD) (APHA, 1992; Fisher and Reddy, 2001; Malecki et al., 2004), potentially mineralizeable P (PMP), inorganic P fractionation (Reddy et al., 1998; Reddy et al., 1995), and 1N HCl – extrac table metals. Microbial biomass P was determined by a 24 h chloroform fumigationextraction (CFE) technique (Brookes et al., 1982; Hedley and Stewart, 1992; Ivanoff et al., 1998). The PMP rate was determined using an anaerobic, waterlogged incubati on at 40 C (Chua, 2000; Malecki-Brown and White, 2007a). Total P analysis involved combustion of 0.5 g oven-dried subsamples at 550 C for 4 h in a muffle furnace followed by dissoluti on of the ash in 6 M HCl on a hot plate (Andersen, 1976). Total P was analyzed using an automated ascorbic acid method (Method 365.4, USEPA, 1993) while AlT was determined by inductively coupled argon plasma spectrometry (model Vista MPX CCD simultaneous ICP-OES manufactured by Varian, Inc., Walnut Creek, CA) at = 396.152 nm (Barnes, 1975; Campbell et al., 1983; Easthouse et al., 1993; Rydin and Welch, 1999; Rydin et al., 2000; Method 200.7,

PAGE 140

121 USEPA, 1993). Ash content was calculated to determine mass loss on ignition (LOI), indicating the organic matter content in the wetland soil (Lim and Jackson, 1982). The amount of Al present in crystalline form was calculated as the difference between the AlT and Alox (Dolui and Chakraborty, 1998). Ac id-extractable Ca, Mg, Fe, and Al concentrations were determined from ovendried soil treated with 25 mL of 1.0 M HCl and placed on a reciprocal shaker for 3 h. The supernatant was filtered through 0.45m membrane filters and analyzed by inductively coupled argon plasma spectrometry (DeBusk et al., 1994; Reddy et al., 1998; Malecki et al., 2004). X-ray Diffraction Analysis Soil fr actionation is often required to simplify the complexity of mineralogical data so that isolation and identif ication of specific phases can be determined (Amonette, 2002). It is often used to isolate the silt a nd clay minerals for use in x-ray diffraction (XRD). Triplicate soil samples originating fr om the first plot of each transect (~10 m from inflow) in the control and alum-treated cell after 12 mo. of alum application were selected for analysis. The first step in the process was to remove all excess porewater and associated salts from soil samples by centrifuging soil fo r 10 minute at 6000 rpm and removing the supernatant via syringe. Next, the sand fracti on was separated from the silt and clay by wet sieving the soil through a 270 mesh sieve us ing distilled water. The silt+clay fraction was collected and used for XRD to identify th e minerals present (Whitting and Allardice, 1986), and in particular any crys talline sulfate or Al forms present in the alum-treated soil. Suspension mounts were dewatered by suction onto ceramic tiles that were then stored in desiccators to prevent cracki ng or contamination until run on the x-ray diffractometer.

PAGE 141

122 Statistical Analysis Data norma lity was determined using the Kolmogorov-Smirnov test (Minitab 13.32, 2000) and data were transformed to fit a normal distributi on (Microsoft Excel, 2000). One-way ANOVAs and multiple comparisons by Tukey’s W were used on all soil characterization variables to determine significant differences (p<0.05) between the control and treated cell (Minitab 13.32, 2000). Linear regression an alysis and Pearson product correlation coefficients were used to determine si gnificant (p<0.05) relationships (Microsoft Excel, 2000). Results and Discussion Soil Physicochemical Characteristics Bulk Density and LOI There were no significant di fferences in bulk density or organic ma tter content between cells at the initiation of the study at either soils de pth (Table 5-1, Table 5-2). During the winter sampling the subsurface soil in the control cell had significantly greater organic matter content as indicated by LOI and significantly (p<0.005) lower bulk density than the subsurface soil in the alum-tr eated cell. By the end of the experiment the surface soil of the alum-treat ed cell had significantly (p< 0.005) more organic matter and lower bulk density than the control cell. Alum is not only used for P removal, but also to clarify turbid lakes by acting as a source of positive electrolytes, causing them to settle out of the water column (Davis and Gloor, 1981; Berg and Berns, 1985). Thus the ability of the positively charged alum floc to bi nd organic matter by neut ralizing the negativelycharged particles (Sparling and Lowe, 1998; Van Hullebusch et al., 2002) may have resulted in the higher soil LOI.

PAGE 142

123Table 5-1. Soil physicochemical characteriza tion data for the 0-5 cm depth in the Or lando Easterly Wetland. Values are means 1 standard deviation (n=3). Parameter Distance from Alum Control inflow (m) 0 mo 4 mo 8 mo 12 mo 0 mo 4 mo 8 mo 12 mo Bulk Density 0-10 1.22 0.49 0.95 0.29 1.00 0.45 0.03 0.03 0.11 0.14 0.77 0.56 0.70 0.46 0.11 0.09 g cm-3 10-35 0.13 0.17 0.15 0.10 0.08 0.06 0.07 0.04 0.06 0.04 0.06 0.07 0.13 0.11 0.22 0.26 35-80 0.04 0.01 0.13 0.12 0.12 0.15 0.05 0.05 0.05 0.03 0.04 0.05 0.04 0.02 0.06 0.02 80-150 0.02 0.01 0.11 0.17 0.02 0.00 0.01 0.01 0.10 0.15 0.16 0.33 0.04 0.01 0.06 0.05 Soil pH 0-10 6.5 0.5 5.1 0.4 4.8 0.1 5.8 0.1 5.9 0.2 6.0 0.9 6.7 0.3 6.8 0.1 10-35 7.1 0.6 6.8 0.1 6.8 0.1 6.8 0.1 6.2 0.6 6.2 0.5 6.6 0.3 6.7 0.1 35-80 7.0 0.5 6.7 0.2 6.8 0.1 6.9 0.1 6.2 0.4 6.8 0.1 6.7 0.1 6.8 0.1 80-150 6.8 0.5 6.8 0.1 7.0 0.1 7.0 0.0 6.4 0.4 6.8 0.1 6.9 0.1 6.9 0.1 LOI 0-10 12.5 13.1 3.18 3.23 4.3 3.64 72.4 22.2 62.3 20.5 16.7 25.6 12.9 19.2 39.3 21.2 % 10-35 69.9 21.3 57.1 23.1 65.3 10.1 68.1 8.35 54.3 28.8 63.4 19.2 47.5 27.2 45.9 24.0 35-80 85.1 1.52 66.7 21.8 68.5 21.2 72.5 12.2 76.4 2.90 67.9 21.2 77.5 4.34 60.6 11.5 80-150 87.2 2.41 74.6 26.2 89.5 2.63 64.5 22.9 71.5 27.1 67.9 30.2 70.0 8.18 57.9 25.5 LOI = loss on ignition.

PAGE 143

124Table 5-2. Soil physicochemical characteriza tion data for the 5-10 cm depth in the Or lando Easterly Wetland. Values are means 1 standard deviation (n=3). Parameter Distance from Alum Control inflow (m) 0 mo 4 mo 8 mo 12 mo 0 mo 4 mo 8 mo 12 mo Bulk Density 0-10 1.44 0.24 1.40 0.31 1.40 0.30 0.56 0.58 0.32 0.21 0.60 0.55 0.60 0.34 0.39 0.31 g cm-3 10-35 0.59 0.34 0.65 0.21 0.52 0.27 0.39 0.29 0.35 0.18 0.47 0.45 0.38 0.19 0.44 0.33 35-80 0.18 0.12 0.45 0.28 0.41 0.32 0.32 0.15 0.05 0.01 0.12 0.15 0.06 0.04 0.23 0.27 80-150 0.10 0.05 0.48 0.44 0.14 0.12 0.62 0.61 0.41 0.52 0.17 0.22 0.30 0.31 0.41 0.35 Soil pH 0-10 5.7 0.3 4.9 0.5 4.8 0.1 5.7 0.1 5.4 0.6 5.9 0.8 6.6 0.3 6.8 0.2 10-35 6.2 0.6 6.7 0.1 6.8 0.2 6.9 0.1 5.4 0.4 6.3 0.4 6.3 0.6 6.7 0.1 35-80 6.5 0.5 6.6 0.2 6.7 0.2 6.8 0.2 6.5 0.6 6.9 0.3 6.8 0.0 6.8 0.0 80-150 6.0 0.1 6.7 0.2 6.9 0.1 7.0 0.1 6.1 0.2 6.9 0.2 6.9 0.1 6.9 0.1 LOI 0-10 2.62 3.21 1.53 1.86 2.16 3.62 39.8 40.1 34.2 28.5 20.2 15.4 13.4 6.99 21.4 0.13 % 10-35 40.2 22.7 13.8 5.75 28.9 15.7 38.1 27.2 38.5 15.4 35.3 30.4 22.3 14.9 18.5 0.13 35-80 82.4 3.55 35.9 27.0 51.1 32.3 38.4 22.1 73.6 11.5 54.6 28.5 62.3 16.4 44.5 0.26 80-150 79.6 3.46 35.1 25.3 68.9 21.3 34.3 40.6 35.0 28.5 53.0 31.6 32.2 30.8 31.4 0.33 LOI = loss on ignition.

PAGE 144

125 Overall, bulk density was significantly greater (p<0.05 all year) in the subsurface 5-10 cm layer (Table 5-2) than in the surf ace 0-5 cm (Table 5-1) while the opposite was true of the organic matter content (p<0.01 all year) throughout the entire study. Organic matter deposition and decomposition at the soil su rface most likely resulted in the higher surface LOI. The increased organic matter in the surface layer allows the soil to remain porous, thereby decreasing the bulk density (Brady and Weil, 1999) as indicated by a significant inverse correlation be tween the two variables for all four sampling events. During the first three sampling events there were also significant differences in bulk density and LOI spatially within the cells. The bulk densities in the surface soil of plots closest to the inflow were significantly higher than all those further from the inflow while the opposite was true of LOI at the 4 and 8 mo. sampling events, and at time 0 the 10 m plot had significantl y less organic matter than the 80 and 150 m plots, corresponding to the inverse relationship betw een the two variables. This result was expected because the inflow culverts of both cells have a very high rate of hydraulic loading which only allows the heavy sand part icles to settle out in close proximity. Vegetation is primarily dominated by seasonal SAV within a 1-2 m radius of inflows, with detrital accumulation typically being ca rried downstream for deposition, while dense Typha stands dominate the remainder of the cells, explaining the higher LOI and lower bulk densities. Soil pH As discussed previously, one concern when dosing an aquatic system with alum is the possible reduction in water column or soil pH . At experiment initiation, the alum cell (10) had significantly greater surface soil pH values (Table 5-1) than the control cell while in the following three samplings ther e were no significant differences in pH

PAGE 145

126 between the cells in either th e surface or subsurface (Table 5-2). This would suggest alum application did not impact the soil pH, however, average surface soil pH values did drop below 6 in the plots closest to th e alum inflow during the 4, 8, and 12 mo. samplings. At a pH between 4 and 6, soluble Al intermediates may occur releasing bound P, altering the toxicity and solubili ty of metals (Cooke et al., 1993b). Overall, soil pH was significantly greater (p<0.001) in the surface than subsurface at time 0, specifically in the plots closest to the inflow in both cells and the farthest plot in the experimental cell. There were no si gnificant differences w ith depth throughout the rest of the study, however the general trend was 0-5 cm 5-10 cm. This was significant in the 35 m plot of alum cells during the 4 mo. winter sampling and 150 m plots of both cells during the 8 mo. spring sampling. There were no significant spatial differen ces in the surface soil pH at experiment initiation, while subsurface values in the farthe st two plots were significantly greater than the plots closest to inflow in the control cell. This was also the case for both cells in the 0-5 and 5-10 cm layer during the 4 mo. samp ling, and again for both layers in the alum cell during the last sampling event (Figure 5-3). Soil pH was directly correlated (p<0.01) to organic matter content and inversely rela ted to bulk density during the 4 and 8 mo. sampling events. Soil Total Phosphorus There were no significant di fferences in TP between cells with respect to both surface and subsurface so il layers, at initiation of the experiment (Table 5-3, Table 5-4). After four months of alum application, the subsurface soil in the alum-treated cell had significantly (p<0.005) lower P concentrations th an the subsurface of the control cell. At 8 and 12 months after alum initiation the surface soil of the alum-treated cell had

PAGE 146

127 Table 5-3a. Soil nutrient data for the 0-5 cm depth in alum-treated cell 10 of the Orlando Easterly Wetland. Values are means 1 standard deviation (n=3). Parameter Distance from Time after alum initiation (mo) inflow (m) 0 4 8 12 Total P 0-10 188.5 181.1 100.8 76.57 648.5 703.3 19658 12781 mg kg-1 10-35 4790 2364 3369 2025 7985 6827 17046 13265 35-80 9685 1335 5075 4944 10749 10986 13202 8634 80-150 13050 3028 8968 5922 15879 9267 11149 9576 Total Ca 0-10 16587 15896 1514 1398 1476.9 1201 36247 8465 mg kg-1 10-35 95593 27598 42893 29858 45674 17707 38800 10028 35-80 113773 20432 71651 61527 58476 37843 40397 6662 80-150 89817 4545 84749 54151 132154 37665 42861 8280 Total Fe 0-10 7610 4363 504.1 712 222.2 138 3152 937 mg kg-1 10-35 15835 5662 7308 4961 7366 1762 5905 1994 35-80 14552 8983 9175 8951 7758 6134 4028 1806 80-150 7517 2902 7359 3325 10601 4303 3974 436 Total Mg 0-10 1725 1583 145.1 127 116.6 97.1 3247 910 mg kg-1 10-35 7577 2429 3398 2233 3836 1505 3165 863 35-80 8324 793.6 5486 4666 4956 3276 3219 447 80-150 8039 680.3 6674.6 4249 12171 3928 3935 827 Table 5-3b. Soil nutrient data for the 0-5 cm depth in control cell 9 of the Orlando Easterly Wetland. Values are means 1 standard deviation (n=3). Parameter Distance from Time after alum initiation (mo) inflow (m) 0 mo 4 mo 8 mo 12 mo Total P 0-10 5467 3364 622.9 949.4397 680.3 2446 1562 mg kg-1 10-35 4324 3801 4789 3730 2584 2736 2008 1607 35-80 6322 1119 4549 2211 5194 1493 2717 1128 80-150 8399 4078 6722 3373 3713 418 4331 3711 Total Ca 0-10 69435 33267 10114 17091 5219 8537 27922 9493 mg kg-1 10-35 70309 21370 51668 36877 28024 28762 23020 11010 35-80 118696 42119 65585 31791 72115 19312 33159 7981 80-150 88951 33616 72169 35248 54841 21992 33508 18950 Total Fe 0-10 19401 6968 2452 4501 2001 2858 6747 1413 mg kg-1 10-35 24451 17212 13583 8789 6593 4948 7690 4658 35-80 21330 7260 12830 6482 17401 4440 7235 1322 80-150 21915 9150 10356 4822 9190 2605 5254 2485 Total Mg 0-10 6038 2920 794.2 1255 449.8 698 2550 807 mg kg-1 10-35 5960 2326 4115 3210 2535 2694 1969 1028 35-80 8445 1216 5505 2679 6132 1500 2826 724 80-150 7321 2879 5660 2771 4515 1695 2600 1209

PAGE 147

128 Table 5-4a. Soil nutrient data for the 5-10 cm depth in alum-treated cell 10 of the Orlando Easterly Wetland. Values are m eans 1 standard deviation (n=3). Parameter Distance from Time after alum initiation (mo) inflow (m) 0 4 8 12 Total P 0-10 35.72 25.9 43.45 39.13 128.4 176.0 2756 2911 mg kg-1 10-35 922.7 741.3 160.9 65.36 895.5 683.9 5280 10663 35-80 7854 1585 2233 3885 2778 2792 571.7 289.2 80-150 6113 1032 1001 843.6 4318 2340 1819 3404 Total Ca 0-10 3939 1797 3790 6802 20020 16938 18689 16240 mg kg-1 10-35 50459 23500 4276 1448 37159 44534 19883 13101 35-80 77144 10559 23494 43944 37061 47337 35467 50298 80-150 79076 3442 14784 12989 48417 34845 17687 19250 Total Fe 0-10 1849 1015 569.6 734.1 3053 2471 2215 1958 mg kg-1 10-35 12502 10454 663 269.4 5835 6701 3146 2150 35-80 13663 17062 3069 5984 6377 7700 2847 2395 80-150 10274 1679 2192 1568 6839 3955 2417 1264 Total Mg 0-10 408.4 244.3 136.4 119.3 1541 1242 1412 1333 mg kg-1 10-35 4009 2209 353.2 119.2 3248 3862 1714 1097 35-80 5811 1472 1989 3659 3251 4066 2823 3646 80-150 6399 492 1122 910.2 3900 2758 1573 1500 Table 5-4b. Soil nutrient data for the 5-10 cm depth in control cell 9 of the Orlando Easterly Wetland. Values are means 1 standard deviation (n=3). Parameter Distance from Time after alum initiation (mo) inflow (m) 0 mo 4 mo 8 mo 12 mo Total P 0-10 1392 1528 458.6 378.7 242.6 136.6 589.0 404.6 mg kg-1 10-35 935.9 539.3 2814 3954 461.8 384.3 415.5 319.8 35-80 6179 2747 3233 2685 3270 1810 1737 1411 80-150 1524 1707 3901 3230 1172 2030 1891 2873 Total Ca 0-10 38635 20350 5907 34800 3237 2313 10541 5760 mg kg-1 10-35 39918 19628 27209 36130 5815 5810 9130 6336 35-80 86492 11178 45968 39577 53988 29881 24544 14569 80-150 45664 32839 49263 40065 33637 27367 20121 19455 Total Fe 0-10 50310 53546 4062 6596 2016 2142 4436 3323 mg kg-1 10-35 17528 6722 8013 10451 2079 929 4218 1966 35-80 21232 16328 9982 8375 12425 6849 5551 3077 80-150 15215 5434 7600 5352 5639 3954 4031 2780 Total Mg 0-10 3934 2075 557.8 2896 284.1 159.5 1127 671.9 mg kg-1 10-35 3089 1250 2228 3042 552.8 554.9 802.7 576.7 35-80 7039 815.3 3859 3348 4514 2524 2118 1262 80-150 3761 2670 3804 3046 2813 2313 1533 1390

PAGE 148

129 significantly more P than that of the control, while there we re no significant differences in the 5-10 cm layer. An increase in soil TP was expected in the alum-treated cell since the alum floc should continually bind P from the water column, eventu ally settling to the soil surface. The Penriched alum floc then ac ts to retard P release from the sediment to the water column (Peterson et al., 1973; Connor and Ma rtin, 1989), not only binding mobile P but also P bound to Fe (Rydin and Welch, 1998). This may explain why after 4 months of alum application the subsurface P was significantly lower, as the mobile and Fe-bound P migrated upward due to the high P adsorption properties (Huang et al., 2002) of the alum floc. However, once all Al ad sorption sites were occupied on the initial settled floc layer, P diffusing upward replenis hed the available sites on the Fe (Rydin et al, 2000) resulting in no significant differences between cell subsurface layers in the last two sampling events. Total P concentrations were significantly greater (p<0.005) in the surface soil (Table 5-3) than subsurface (Table 5-4) thr oughout the entire study, si milar to the organic matter content. Loss on ignition and TP were significantly positively correlated with each other at time 0, 4 and 8 months. Nutrie nt content typically increases with an increase in organic matter (Farnham and Finney, 1965). Additionally, TP concentrations were significantly greater (p<0.001) during both August samplings (time 0 and 12 mo.) than during the December and April samplings which may be attributed to seasonal uptake and release by Typha spp. During maximum growth in the spring Typha spp. uptakes nutrients primarily from the soil (Crowder, 1991; Rai et al., 1995; Sparling and Lowe, 1998; Thiebaut and Muller, 2000), however during senescence in late July through

PAGE 149

130 August P is translocated to the rhizomes or released by decomposition (Davis, 1984; Emery and Perry, 1995) increasing soil TP concentrations. Similar to the spatial pattern in LOI and soil pH there were significantly lower (p<0.001) TP concentrations in the first two pl ots closest to the inflow than in the back two plots for both cells at both depths at the start of the experiment. This was also the case in the surface soil at the 0-10 m plot dur ing the 4 and 8 mo. samplings while in the final sampling TP values were so variable th ere were no significant spatial differences at either depth. Total Calcium and Magnesium There were no significant differences in surface soil Ca or Mg concentrations throughout the first three sampling events (T able 5-3). During the 12 mo. sampling however, the alum-treated cell had significantly more Ca and Mg than the control cell. During the winter, 4 mo. sampling subsurface Ca and Mg concentrations were both significantly greater in the control cell th an in the alum cell, while there were no significant differences in the other three samplings (Table 5-4). Throughout the entire study both Ca and Mg concentrations were significantly greater (p<0.005) in the surface soil (Table 5-3) than subsurface (Table 5-4) similar to the organic matter and TP content of the soil. Surface soil Ca and Mg concentrations were consistently correlated (p<0.05) with each other throughout the study. Both variables were also directly correlated with LOI a nd TP (p<0.01) throughout. Additionally, in both cells Ca and Mg concentrations were significan tly greater at the star t of the study than in the latter three samplings while the 12 m o. summer sampling had significantly higher concentrations than during the 4 and 8 mo., wi nter and spring samplings. This seasonal

PAGE 150

131 pattern may be attributed to the growth and decomposition of Typha since the same trend was also seen in soil TP. Spatially, Ca and Mg concentrations were significantly lower (p<0.001) in the first plot closest to the inflow than in the back two plots for both cells at both depths at the start of the experiment, as well as the subs urface soil in the 10-35 m plot having lower (p<0.001) quantities than the 35-80 m plot. During the 4 and 8 mo. samplings both variables were significantly lower in the plot closest to the inflow than at all other plots along the transect in both the surface and subsur face soil. In the final sampling, similar to TP values, Ca and Mg concentrations were so variable there were no significant spatial differences at either depth. Total Iron Total Fe concentration s were significantly greater (p<0.05) in both the surface and subsurface soil layers of the control cell th roughout the entire study as compared to concentrations in the alum-treated cell (Tab le 5-3, Table 5-4), possibly due to natural mineralogical variations. Ther e was no significant difference in Fe concentrations with depth initially, but Fe was significantly grea ter (p<0.05) in the surface 0-5 cm soil than subsurface 5-10 cm layer in the remaining th ree samplings. In both cells initial Fe concentrations were significan tly greater than during all ot her samplings, similar to the trend in Ca and Mg concentrations as well. Surface soil Fe concentrations were directly correlated to Ca and Mg concentrations during the winter and spring samplings. Spatial differences in Fe concentrations were not as pronounced as those of other variables with no differences in either soil layer at the initiation or completion of the study. At the 4 and 8 mo. samplings surface so il Fe concentrations were significantly

PAGE 151

132 lower in the 0-10 m plot, closes t to the inflow, than in all other plots along the transect. The same was true of the subsur face soil during the 8 mo. sampling. Soil Aluminum Characterization Total Aluminum Surface soil AlT concentrations increased five tim es or more in the alum-treated cell from experiment start to fi nish while in the control cell AlT decreased to 33% of the original concentration (Table 55). Thus at time 0 the control cell had significantly more Al in both the 0-5 and 5-10 cm soil layers th an the designated cell to be treated with alum. After four months of al um application (~ 29 g Al m-2) there was no longer a difference in surface soil AlT concentrations but the contro l cell still had significantly more Al in the subsurface (Table 5-6). By the 8 and 12 mo. samplings the alum-treated cell had significantly higher surface soil AlT concentrations than the control and there were no significant differences in subsurface concentrations. The surface soil AlT concentrations were directly corre lated (p<0.01) to both the amorphous Alox and calculated crystalline Al thr oughout the study with all Al forms showing good agreement. Additionally AlT was positively correlated to soil TP in all but the first sampling suggesting a direct relationship between the amount of Al present and TP bound. There were no significant differences in AlT concentration with depth at the start of the study but from the 4 mo. sampling on, con centrations were significantly greater (p<0.005) in the surface than subsurface. As mentioned, the control cell AlT concentrations were significan tly greater at the start of th e experiment than during all other samplings while the opposite was true in the alum-treated cell where the 12 mo. AlT concentrations were significantly greater than the previous three.

PAGE 152

133 Spatial differences in AlT were highly variable re sulting in no significant differences in the first or last sampling. At the 4 and 8 mo. samplings surface soil AlT concentrations were significantly lower in the 0-10 m plot, closest to the inflow, than in all other plots along the transect. The same was true of the subsurface soil during the 8 mo. sampling similar to the spatial pattern in total Fe. Amorphous Aluminum Initially and in the 4 mo sampling there was significantly more Alox in both the surface and subsurface soil of the control cell as co mpared to the alum-treated cell (Table 5-5, Table 5-6). However, in the last two sa mplings the alum-treated surface soil had significantly higher amorphous Al than the cont rol with no significant differences in the subsurface similar to the AlT results. The increased amorpho us Al can be attributed to the addition of alum in the treated cell wh ich took over 4 months of continuous low dosage application to exceed the natural mineralogi cal source of Alox in the control cell. Throughout the entire study Alox concentrations were signi ficantly greater (p<0.005) in the surface soil (Table 5-5) than subsurface (Table 5-6) similar to the LOI and TP concentrations. This was expected in the ce ll receiving alum due to the settling floc, but it was also true of the control cell due to the natural Al associated with organic matter. Interestingly, surface soil TP concentrations were not directly correlated (p<0.01) to Alox concentrations until the 8 and 12 mo. samplings , similar to the time period it took for the addition of Al in the form of alum to exceed the natural Al concentration present in the control cell, providing further evidence that the alum floc was effectively binding P. Additionally, when comparing samplings the initial samples from the alum-treated cell had significantly less (p<0.001) Alox than concentrations during the remaining

PAGE 153

134 Table 5-5a. Aluminum characterization for the 0-5 cm depth interval in alum-treated cell 10 of the Orlando Easterly Wetland. Valu es are means 1 standard deviation (n=3). Parameter Distance from Time after alum initiation (mo) inflow (m) 0 4 8 12 Total Al 0-10 7962 8412 1273 1926 2261 2122 57706 62904 mg kg-1 10-35 19497 8520 14736 7984 64270 71435 90036 77158 35-80 11845 3498 15067 15218 47071 56929 58500 61525 80-150 4915 742.6 11478 5408 30321 9558 89262 88539 Oxalate Al 0-10 38.81 54.29 409.7 332.5 1441 1875 26796 32545 mg kg-1 10-35 444.5 191.8 2807 1970 14892 12706 39489 35905 35-80 414.0 183.9 1412 1256 10089 12368 25848 28398 80-150 234.3 90.39 1089 515.9 2955.8 1079.6 37635 36995 Crystalline Al 0-10 5974 7725 5229 7950 93.21 98.23 31010 30491 mg kg-1 10-35 19052 8357 2720 4644 49378 59032 50548 42692 35-80 11432 3494 6718 7926 36982 44683 32652 33203 80-150 4680 692.3 1407 1451 27992 10331 51627 52891 Total Al extracted via acid digestion (Andersen, 1976); oxalate Al extracted using acid oxalate (McKeague and Day, 1966); crystalline Al calcula ted as total – oxalate (Dolui and Chakraborty, 1998). Table 5-5b. Aluminum characterization for the 0-5 cm depth interval in control cell 9 of the Orlando Easterly Wetland. Values are means 1 standard deviation (n=3). Parameter Distance from Time after alum initiation (mo) inflow (m) 0 4 8 12 Total Al 0-10 30033 14861 2980 3898 2405 3459 9213 920.8 mg kg-1 10-35 27027 11784 20213 15353 9843 10165 8584 3156 35-80 24209 3687 16416 7931 18825 3875 8935 2354 80-150 14754 4285 11257 5511 9455 4763 5469 2878 Oxalate Al 0-10 3501 2109 337.5 258.9 623.8 734.2 1528 186.3 mg kg-1 10-35 2955 1276 2672 1081 1647 1022 1760 716.7 35-80 2011 156.8 1757 576.4 1931 145.5 1560 503.7 80-150 1331 574.7 1334 601.8 1203 364.7 977.9 386.0 Crystalline Al 0-10 26532 12904 5596 4595 1780.8 2727 7685 786.8 mg kg-1 10-35 24072 10631 14956 12642 8196 9184 6823 2471 35-80 22198 3700 15674 7433 16711 4443 7375 1861 80-150 13423 3823 5952 7350 8252 4429.5 4491 2550

PAGE 154

135 Table 5-6a. Aluminum characterization for the 5-10 cm depth interval in the alumtreated cell 10 of the Orlando Easter ly Wetland. Values are means 1 standard deviation (n=3). Parameter Distance from Time after alum initiation (mo) inflow (m) 0 4 8 12 Total Al 0-10 1520 1985 1241 2300 2507 2686 5367 4030 mg kg-1 10-35 10360 7443 842.3 404.8 2563 2633 26523 51213 35-80 11323 6313 4170 8362 11006 14215 6040 5050 80-150 12625 2092 2748 1846 8152 3795 8398 4766 Oxalate Al 0-10 48.45 42.77 129.1 70.60 347.8 633.5 1416 2005 mg kg-1 10-35 166.4 121.7 204.9 70.38 800.9 856.8 12487 27669 35-80 389.6 303.0 318.0 293.2 10010 1043 1267 1427 80-150 595.6 112.2 457.6 166.2 785.3 263.4 2478 2448 Crystalline Al 0-10 1471 1950 2540 3097 200.1 150.2 3951 2486 mg kg-1 10-35 10194 7330 11704 13486 1762 1838 14036 23608 35-80 12086 6529 10109 9330 10005 13225 4773 4087 80-150 12029 2045 3171 5451 7367 3597 6264 2454 Table 5-6b. Aluminum characterization for the 5-10 cm depth interval in control cell 9 of the Orlando Easterly Wetland. Values are means 1 standard deviation (n=3). Parameter Distance from Time after alum initiation (mo) inflow (m) 0 4 8 12 Total Al 0-10 26863 15340 4003 8661 2097 1174 7121 4251 mg kg-1 10-35 16232 6133 12303 16175 2434 1645 4174 1556 35-80 18288 3167 11956 10254 13649 8489 6628 3778 80-150 11008 7735 8193 6867 6215 5465 5185 4875 Oxalate Al 0-10 1183 683.8 899.4 579.3 580.5 278.9 944.0 559.6 mg kg-1 10-35 650.7 250.7 2049 1865 661.8 402.5 750.9 313.3 35-80 1559 205.0 1339 861.0 1536 612.4 1117 705.3 80-150 677.3 674.0 971.9 640.2 812.4 673.0 673.1 617.1 Crystalline Al 0-10 24380 16079 5458 11919 1517 929.9 6176 3721 mg kg-1 10-35 15581 5922 11383 14214 1772 1252 3423 1245 35-80 16729 3005 11328 8481 10443 6121 5510 3081 80-150 10330 7075 6470 5738 5731 5689 4512 4300

PAGE 155

136 three sampling events. At 4 months, the alum-treated soil Alox was significantly less than at 8 and 12 months, and in the final sampli ng the alum-treated soil had a significantly greater Alox than at all other sampling times. Thus while control surface Alox remained stable throughout the study averaging 1.70 0.87 g kg-1 the amorphous Al continued to increase in the surface soil of the experimental cell due to the continuous alum application. There were no significant diffe rences in either the surf ace or subsurface soil at the initial, 4, or 12 mo. sampling due to high spatial variability be tween transects. Only in the April, 8 mo. sampling event was the Alox in the 0-10 m plot si gnificantly less than the 35 and 80 m plots in the surface so il, and less than that in the 80 m plot in the subsurface soil. This, suggests that the alum floc did not readily settle out until flow velocity decreased away from the inflow, carrying the alum through the cell. Crystalline Aluminum Crystalline Al concentrations followed the same temporal trend as that of AlT. At time 0 the control cell had significantly more cr ystalline Al in both soil layers than cell 10. After four months of alum application there was no longer a difference in surface soil crystalline Al concentrations but the control cell still had significantly more Al in the subsurface (Table 5-6). By the 8 and 12 mo. samplings the alum-treated cell had significantly higher surface soil crystalline Al concentrations than the control and there were no significant differences in subsurface concentrations. Similar to Alox concentrations, surface soil TP concentrations were not directly correlated (p<0.01) to the crystalline Al concentrations until the 8 and 12 mo. samplings, suggesting not only that the Al hydroxide floc was binding P but also may have been transforming into a more crystalline phase able to incor porate P into the structure.

PAGE 156

137 However, when the change in ratio of Alox / AlT over time was calculated, crystallization, or the si gnificant decrease in the Alox / AlT ratio (Mahaney et al., 1991; Bera et al., 2005), did not occur. Rather, th e ratio increased in both cells with the time 0 ratio significantly less than at all other sa mpling periods. Additionally, the change in Alox / AlT ratio was significantly greater in the alum-t reated cell than the control as would be expected due to the increase in amorphous Al. Similar to the AlT, there were no significant differences in crystalline Al concentration with depth at the start of the study but from the 4 mo. sampling on, concentrations were significantly greater (p< 0.005) in the surface than subsurface. Also, the control cell crystalline Al concentrations were significantly greater at the start of the experiment than during all other samplings while the opposite was true in the alumtreated cell where the 12 mo. Al concentrations were significantly greater than the 4 and 8 mo. sample concentrations. Spatial differences in crystalline Al we re highly variable with no significant differences in either soil la yer during the first or last sa mpling. At the 4 and 8 mo. samplings surface soil Al concentrations were significantly lower in the 0-10 m plot, closest to the inflow, than in all other plots along the transect. Soil Microbial Characteristics Microbial Biomass Phosphorus Measurements of MBP are used to assess the importance of the m icrobial pool in P cycling, measuring the P immobilized in microbes, thus serving as an indicator of microbial pool size in the system as well as indicating organic P removal efficiency (Kunc, 1994; Powlson, 1994). At the start of the experiment microbial biomass was significantly greater in the surface layer of th e control cell than in the experiment cell

PAGE 157

138 while there were no significant differences in the subsurface soil (Table 5-7, Table 5-8). During the winter and spring samplings there were no significant differences between cells in the surface soil, however the contro l had significantly greater microbial biomass in the subsurface layer than the alum-treated cell. Additionally, microbial biomass levels were significantly lower in December and April than they were during the two summer August sampling events for both cells which can be attributed to the cold seasonal temperatures which reduce biological processes (McDonnell and Hall, 1969). Surprisingly, by the final sampling there wa s significantly more microbial biomass in the surface soil of the alum -treated cell than the control with no significant differences in the subsurface. Previous short-term st udies had indicated decreased soil pH and increased Al due to alum application ha d significantly reduced MBP (Malecki-Brown and White, 2007a; Malecki-Brown et al., 2007a). Recall, by the final sampling event the surface soil of the alum-treated cell had significantly greater TP (bound in the floc layer) than the control cell which may have stimulat ed the increase in biomass if the microbes were able to access the Al-bound nutrients (J ansson, 1981; Bittl et al., 2001). Increased microbial biomass with increased nutrient c oncentrations has been documented in other systems (Reddy et al., 1999b; White and Reddy, 2000; Malecki et al., 2004) however those systems were not alum-treated. There were no significant di fferences in MBP with depth at experiment initiation, however in all three remaining sampling even ts MBP was significantl y greater (p<0.05) in the surface than subsurface (Table 5-7, Table 5-8). This may be explained by the decrease in TP and LOI with depth as we ll, decreasing nutrient availability to the microbial pool. There were also no spatial differences in MBP at time 0 however during

PAGE 158

139Table 5-7. Microbial characteri zation of the 0-5 cm surface so il in cell 9 and 10 of the Orlando Easterly Wetland. Values are means 1 standard deviation (n=3). Parameter Distance from Alum Control inflow (m) 0 mo 4 mo 8 mo 12 mo 0 mo 4 mo 8 mo 12 mo MBP 0-10 6.16 8.62 1.25 2.63 0.46 0.92 150 92.0 72.4 77.9 10.8 11.7 3.46 2.46 79.7 57.3 mg kg-1 10-35 32.5 49.5 26.5 32.7 57.6 33.7 53.8 31.8 52.9 50.9 83.0 77.4 36.7 40.5 69.2 47.8 35-80 68.9 46.1 78.6 71.5 60.7 42.1 82.3 58.5 146 35.3 80.2 53.6 73.7 29.9 57.6 29.5 80-150 28.1 26.0 70.0 48.5 177 89.5 227 103 82.0 67.9 133 68.9 131 87.1 82.7 50.9 SOD 0-10 6.71 4.11 3.59 2.41 5.04 9.16 76.1 108 58.0 28.3 5.62 6.84 8.24 9.49 38.2 30.1 mg kg-1hr-1 10-35 34.7 24.0 34.4 28.3 50.6 27.1 21.6 21.0 60.7 38.1 33.6 10.3 37.6 20.1 40.1 34.2 35-80 72.5 15.8 36.7 16.5 44.7 18.2 23.1 14.7 124 40.5 32.3 18.4 45.6 11.6 25.0 9.06 80-150 119 51.0 63.1 54.3 87.1 27.4 29.4 43.0 102 81.0 55.4 29.1 65.0 35.6 21.5 14.6 PMP 0-10 1.10 0.74 0.44 0.32 0.91 1.16 41.4 35.7 19.3 20.1 1.56 2.79 2.02 2.10 39.6 33.6 mg kg-1d-1 10-35 16.4 11.8 8.02 9.24 29.5 26.1 62.0 87.1 12.2 9.12 11.1 6.89 21.3 16.4 22.7 11.8 35-80 33.2 17.4 10.9 9.58 14.5 10.3 40.3 28.4 15.1 10.2 10.9 6.80 16.7 4.72 14.4 10.3 80-150 24.9 12.5 8.22 11.2 36.8 14.1 35.6 23.0 27.0 14.9 28.5 18.3 22.2 10.3 13.1 11.0 MBP = microbial biomass phosphorus; PMP = potentially mineralizeable phosphorus; SOD = soil oxygen demand.

PAGE 159

140Table 5-8. Microbial characterization of the 5-10 cm subsurface soil in cell 9 and 10 of the Orlando Easterly Wetland. Values are means 1 standard deviation (n=3). Parameter Distance from Alum Control inflow (m) 0 mo 4 mo 8 mo 12 mo 0 mo 4 mo 8 mo 12 mo MBP 0-10 6.70 11.1 0.68 1.28 0.13 0.21 16.5 11.3 50.7 61.9 2.59 2.93 3.29 1.80 13.8 6.56 mg kg-1 10-35 67.5 45.9 3.20 1.98 5.50 5.66 13.4 12.2 36.4 47.2 44.2 69.2 8.81 11.8 12.1 12.8 35-80 203 161 15.1 29.7 9.74 17.6 19.4 17.7 149 111 28.3 26.5 32.0 17.1 38.0 28.3 80-150 5.67 2.83 18.0 13.9 36.3 29.6 74.1 81.7 13.6 12.8 76.7 81.5 47.4 59.8 18.3 29.5 SOD 0-10 3.21 0.77 3.61 5.60 2.19 3.44 57.3 60.8 24.1 16.7 9.17 7.64 5.66 4.89 11.3 5.42 mg kg-1hr-1 10-35 6.90 5.10 5.77 2.84 6.24 3.26 24.8 23.7 18.3 7.29 16.7 23.6 12.8 9.87 11.2 8.37 35-80 19.1 11.7 11.8 12.4 13.1 9.32 35.5 42.4 103 15.1 24.2 14.4 29.3 11.7 13.1 8.31 80-150 31.6 14.6 9.07 4.80 25.0 15.3 102 129 18.4 16.4 32.9 25.8 24.2 18.5 13.8 15.5 PMP 0-10 0.10 0.06 0.12 0.15 0.41 0.67 2.88 3.93 1.30 1.19 0.99 0.65 1.01 0.94 6.80 4.94 mg kg-1d-1 10-35 1.82 1.55 0.87 0.99 0.85 0.87 4.22 4.31 1.15 0.89 5.98 9.06 2.06 3.48 4.16 4.61 35-80 5.00 5.33 3.06 5.79 4.64 9.14 4.45 5.92 16.9 11.9 15.8 14.4 8.07 4.01 7.73 6.29 80-150 10.8 4.77 1.95 3.26 7.64 8.30 6.01 8.24 5.27 7.87 20.3 16.1 6.70 6.33 4.35 7.59 MBP = microbial biomass phosphorus; PMP = potentially mineralizeable phosphorus; SOD = soil oxygen demand.

PAGE 160

141 the 4 and 8 mo. samplings MBP 0-10 m from the inflow was significantly lower than MBP at plots farthest from the inflow in both cells at both depths corresponding to the spatial differences in LOI and TP during the same time periods. Both LOI and TP were directly correlated (p<0.05) to MBP in the surface layer at ti me 4 and 8 mo. as well as in the subsurface during the first three samplings . During the last sampling only LOI was correlated to subsurface MBP. Soil Oxygen Demand Soil oxygen dema nd is used as an indicator of the activity of the microbial population. Soil oxygen demand wa s significantly greater in the control cell than experimental cell at the start of the experiment in both the surface a nd subsurface soil and continued to be the case in the subsurface soil after 4 and 8 months (Table 5-7, Table 58). This corresponds to the greater microbia l biomass as indicated by MBP in the control cell. The SOD was significantly (p<0.05) corr elated to MBP in the surface soil during the first three samplings and in the subsur face during the 4 and 8 mo. samplings. Soil oxygen demand was also correlated to TP and LOI during the first three sampling events in both soil layers indicating not only populati on growth but also microbial activity is dependant upon available nutrients. In the final sampling however, the 5-10 cm soil layer in the alum-treated cell had significantly great er SOD than the control cell and was not correlated with any of the parameters. Throughout the first 8 months of the e xperiment SOD rates were significantly greater (p<0.005) in the surface soil (Table 5-7) than subsurface (Table 5-8) similar to the MBP concentrations during the winter and sp ring samplings. Generally microbes in the surface layer are able to use the available la bile substrates while more recalcitrant material is left for microbes in the subsurface layer resulting in lower levels of activity

PAGE 161

142 (Henrichs and Reeburgh, 1987). Soil oxygen de mand rates were also significantly lower in the 4 mo., December sampling than during the two summer August sampling events for both cells which can be attributed to the cold seasonal temperatures which reduce biological activity (McDonnell and Hall, 1969). Similar to the spatial pattern in MBP, LOI, and TP, SOD rates closest to the infl ow were significantly lower (p<0.001) than SOD rates farthest from the inflow in surf ace soil at time 0, 4 and 8 mo. and in the subsurface at 4 and 8 months. Potentially Mineralizable Phosphorus Potentially mi neralizable P rates serve as another indicator of microbial activity although in this experiment results were hi ghly variable. There were no significant differences in PMP rates between cells at either the beginning or end of the experiment in either soil layer (Table 5-7, Table 5-8). During the December sampling surface soil in the control cell had greater PMP rates than in the alum-treated cell, and the same was true of the subsurface soil during the April sampling. There were significant (p<0.01) positive correlations between PMP rates, LOI, and TP in both soil layers at the start of the experiment. During the December sampling PMP rates were directly correlated (p<0.01) with LOI and MBP in both the surface and subs urface, while subsurface rates were also correlated with TP concentrations. April PMP rates were significantly correla ted to LOI, TP, soil pH, and MBP in both layers, excluding soil pH in the subsurface. In the fi nal summer sampling PMP rates were significantly correlated to surface soil TP and subsurface LOI, TP, and MBP. Additionally there was signifi cant correlations (p<0.01) be tween PMP and SOD rates in the surface soil at time 4 and 8 mo. as well as in the subsurface during the first three

PAGE 162

143 samplings indicating good agreement between the two measures of microbial activity used in this experiment. Throughout the entire study PMP rates were significantly greate r (p<0.05) in the surface than subsurface, similar to trends in MBP, SOD, TP, and LOI. As previously mentioned, microbes in the surface layer are able to use the available labile substrates while more recalcitrant material is left for microbes in the subsurface layer resulting in lower levels of activity (Henrichs and Ree burgh, 1987). Similar to the SOD rates, PMP rates were also significantly lower in the December sampling than during the two summer sampling events which can be attributed to the cold seasonal temperatures which reduce biological activity (McDonnell and Hall , 1969). Spatially similar to most other soil variables, PMP rates in the sandy area closest to the cell inflow were significantly lower than PMP rates farthest from the inflow in both cells at both depths during the first three samplings (Figure 5-8). There were no significant differences with distance for the 12 mo. sampling due to increa sed transect variability. Soil Phosphorus Forms Overall averages of organic P (Po) and inorganic P (Pi) were essentially equivalent in the cell treated with alum averaging 46% Pi and 54% Po in the 0-5 cm surface soil while in the control cell there was substantially less Pi averaging 24% and substantially more Po averaging 76%. Of even greater importance is the change in Pi versus Po in the surface soil over the course of the experiment. At time 0 both cells had the majority of P bound organically, both averaging 32% Pi and 68% Po. However, after receiving twelve months of alum, 60% of the P in cell 10 wa s inorganically bound and only 40% remained in the organic form while Pi in the control cell remained constant at 32% and 68% Po.

PAGE 163

144 Surprisingly, alum effects were also observed in the subsurface 5-10 cm layer despite both physical and microbi al parameters suggesting th at the Al hydroxide floc remained at the surface. At the st art of the experiment subsurface Pi averaged 28% and Po 72% in both cells, very similar to the surface soil. By the 12 mo. sampling subsurface Pi was nearly equal to Po averaging 46% in the alum-treated cell while in the control cell Pi remained constant at 28% similar to the surf ace soil. All P fractions were significantly greater (p<0.05) in the surface layer than subsurface (Table 5-9 through Table 5-12) similar to the TP values previously discussed. The overall decrease in P storage capacity with depth can be attributed to the decreas e in metal concentrations (Al, Ca, Fe, Mg) available for P sorption with dept h (Table 5-3 to Table 5-5). Soluble Phosphorus In the surf ace layer, KCl–extractable Pi consisting of labile, readily bioavailable P consistently made up the smallest fracti on (0.02 – 12%) of the total P pool, remaining significantly lower than all other P fractions throughout the year (Table 5-9). Soluble P concentrations were significantl y greater in the surface soil of cell 10 at the start of the study, however there were no signifi cant differences in surface KCl Pi concentrations between cells for the remainder of the study. During the 4 mo. winter sampling, subsurface soil in the control cel l had significantly greater KCl Pi concentrations than in the alum-treated cell but once again, there we re no significant diffe rences between cells for the remainder of the study (T able 5-10). Overall, KCl Pi concentrations were significantly higher during the fi rst sampling than the last in both cells, and the 4 mo. sampling was greater than both the 8 and 12 mo. samplings in the control cell. In both surface and subsurface soil a spatial gradient in soil soluble P concentrations was apparent in both the alum-treated and contro l cell. At the start of the

PAGE 164

145Table 5-9. Mean inorganic phosphorus fracti onation data from the 0-5 cm soil layer in cell 9 and 10 of the Orlando Easterly Wetland. Values are means 1 standard deviation (n=3). Parameter Distance from Alum Control inflow (m) 0 mo 4 mo 8 mo 12 mo 0 mo 4 mo 8 mo 12 mo KCl Pi 0-10 3.18 4.04 1.88 2.75 0.15 0.15 29.0 33.5 11.0 7.03 3.75 3.58 2.94 4.56 9.14 6.27 mg kg-1 10-35 12.0 10.5 6.02 4.44 1.73 1.51 1.64 1.93 9.19 6.89 19.3 19.5 6.31 6.90 5.48 5.01 35-80 31.0 18.5 25.1 15.6 7.56 12.9 2.67 2.17 16.1 8.81 22.9 17.1 13.2 8.02 6.50 4.13 80-150 121 53.1 65.0 40.6 80.0 35.3 25.9 23.1 30.9 15.5 40.0 30.4 18.0 7.86 27.7 39.5 NaOH Pi 0-10 8.673 7.271 35.65 52.58 185.5 262.4 3047 3909 368.5 268.8 51.28 59.32 39.17 46.76 116.8 18.80 mg kg-1 10-35 75.16 35.93 381.7 378.9 2215 2208 4818 5719 262.0 172.3 266.7 135.4 134.6 110.2 150.0 133.0 35-80 126.4 14.17 242.3 176.6 1261 1684 2439 2686 152.2 31.46 133.5 55.70 117.2 21.68 92.57 37.96 80-150 118.4 24.05 153.0 88.32 537.6 278.3 3841 3444 158.1 77.92 144.6 67.19 146.5 42.18 103.6 49.69 HCl Pi 0-10 10.15 5.995 17.82 35.74 82.30 168.1 126.0 68.78 85.42 49.99 26.75 35.22 20.79 16.50 273.2 231.5 mg kg-1 10-35 85.00 61.55 96.46 50.56 304.5 239.2 257.7 77.38 60.30 38.29 93.07 41.94 61.50 55.07 66.46 47.18 35-80 113.7 45.02 187.4 197.7 95.33 84.74 119.0 67.34 105.4 86.31 149.0 204.1 63.52 18.07 85.14 30.55 80-150 197.7 87.82 173.1 169.0 297.6 112.4 161.9 36.35 135.6 79.46 157.9 126.0 207.7 192.1 196.5 158.0 TPi 0-10 22.01 12.07 55.35 90.93 268.0 348.3 3202 3944 464.9 319.2 81.78 97.14 62.90 66.85 399.2 224.7 mg kg-1 10-35 172.2 88.17 484.1 408.6 2521 2267 5077 5741 331.5 204.9 379.0 187.2 202.4 170.7 221.9 169.1 35-80 271.1 32.77 454.8 305.8 1363 1764 2561 2727 273.7 118.5 305.3 250.3 193.9 40.20 184.2 55.43 80-150 437.4 138.3 391.1 262.1 915.3 229.6 4028 3448 321.2 141.5 342.5 202.9 372.1 233.9 327.9 190.2

PAGE 165

146Table 5-10. Mean inorganic phosphorus fracti onation data from the 5-10 cm soil layer in cell 9 and 10 of the Orlando Easterly Wetland. Values are means 1 standard deviation (n=3). Parameter Distance from Alum Control inflow (m) 0 mo 4 mo 8 mo 12 mo 0 mo 4 mo 8 mo 12 mo KCl Pi 0-10 0.22 0.30 1.30 2.01 0.36 0.25 3.10 2.11 4.09 6.16 1.99 1.69 1.47 0.83 2.04 1.02 mg kg-1 10-35 4.40 4.60 1.17 1.02 1.78 1.40 0.57 0.21 0.91 0.47 12.1 18.8 1.09 1.27 1.04 0.78 35-80 17.1 11.0 9.82 14.5 3.31 3.29 1.12 1.18 15.6 6.44 22.5 13.9 8.89 4.12 2.90 1.51 80-150 20.4 13.1 8.10 8.22 17.6 15.7 4.26 6.13 18.2 22.8 18.5 15.2 7.42 7.18 2.80 2.90 NaOH Pi 0-10 3.649 1.603 30.41 39.00 44.81 93.10 179.5 291.3 89.92 55.87 56.84 36.78 35.42 10.08 50.95 25.80 mg kg-1 10-35 24.66 16.80 28.30 18.97 77.26 103.8 1234 2959 47.22 23.90 199.6 257.6 35.55 24.27 41.35 20.06 35-80 80.12 53.28 50.78 86.78 115.4 121.6 134.3 125.7 138.2 22.18 101.5 52.58 94.17 32.28 72.87 44.98 80-150 100.0 42.99 48.51 28.65 94.73 49.34 264.5 279.2 59.71 46.61 108.7 71.81 72.00 60.05 41.76 47.16 HCl Pi 0-10 1.706 3.447 6.876 11.95 29.62 43.07 66.04 64.44 15.84 14.23 11.69 10.90 16.67 10.19 48.54 34.43 mg kg-1 10-35 18.73 19.08 11.14 13.01 21.35 18.64 70.70 73.44 14.64 23.68 46.31 58.41 19.23 36.97 15.22 16.65 35-80 169.6 287.8 39.12 58.87 79.11 77.16 41.47 31.42 107.3 62.50 74.03 49.72 51.77 11.19 49.91 19.62 80-150 251.6 247.9 65.65 56.03 191.8 122.7 73.76 52.91 17.93 17.42 128.8 17.76 64.04 63.47 94.27 119.3 TPi 0-10 5.578 4.023 38.59 49.33 74.79 115.4 248.6 285.8 109.8 73.83 70.52 42.56 53.56 9.873 101.5 47.24 mg kg-1 10-35 47.80 36.22 40.61 28.74 100.4 120.8 1305 3020 62.77 46.34 257.9 330.7 55.87 61.07 57.62 35.83 35-80 266.8 296.9 99.71 158.5 197.8 183.3 176.9 152.1 261.1 84.97 198.0 93.61 154.8 26.35 125.7 63.46 80-150 372.1 232.5 122.3 89.01 304.1 160.9 342.5 326.5 95.86 78.87 256.0 89.83 143.5 130.3 138.8 169.0

PAGE 166

147 experiment the first two plots closest to th e inflow in cell 10 and the second, 10-35 m plot in the control cell had significantly lower KCl Pi concentrations than the furthest two plots in both soil layers. By the 4 mo. samp ling the 0-10 m plot closest to the inflow still had significantly lower KCl Pi concentrations than the fu rthest two plots in both soil layers and the 10-35 m plot had significantly less P than the farthest plot of the alumtreated cell. The 0-10 m plot also had a lowe r soluble P concentrati on than the farthest (80-150 m) plot in the surface soil of the control cell. There were no significant differences in the control cell during the last two samplings while in the alum-treated cell the first three plots stretching 0-80 m all had significantly less KCl Pi than the farthest plot from the alum source in surface soil at th e 8 mo sampling. The same was true of the 10-35 m and 35-80 m plots in the final sampling. Calcium and Magnesium-bound Phosphorus The HCl-extractable Ca and Mg (HCl Pi ) bound in the control cell gradually increased (on a % basis) throughout the year, averaging 9-17% of the total P pool, while the percent bound to Ca and Mg in the alum-tre ated cell decreased fr om 14% initially to 3% by the end of the study. There was substantially less Mg in the soil than Ca (Table 53, Table 5-4) suggesting the majority of P in the HCl Pi fraction was associated with Ca. Additionally, total Ca and Mg c oncentrations in the soil did not follow similar trends to the percent P-bound suggesting other influenc es. In both cells, water column pH averaged 6.9 0.4 throughout the study while so il pH values were slightly more acidic, in both cases favoring the binding of P to Fe and Al. In the alum-treated cell, excess Al hydroxide sorption capacity may have been av ailable to not only preferentially bind P naturally associated with Al and Fe with in the soil but also P bound to Ca and Mg.

PAGE 167

148 Temporal trends were variable. In the first two samplings there were no significant differences in P concentrations bound to Ca and Mg between cells in the surface soil while during the winter concentrations were si gnificantly greater in the subsurface soil of the control cell. In the 8 mo. sampling th e surface soil of the alum-treated cell had significantly more HCl Pi than the control on a concentra tion basis, while there were no significant differences between cells in the fi nal sampling due once again to variability in the data. Overall, HCl Pi concentrations were significantl y higher in the alum-treated cell at the 12 mo. sampling than during the first two samplings. Spatial trends in HCl Pi were apparent with the last plot having significantly greater HCl Pi concentrations than all other plots in the surface and subsurface of cell 10 while in the control cell the 10-35 m had significantly lower concentrations than the 35-80 m plot in the subsurface soil. At the 4 and 8 mo. sa mplings the first plot closest to the inflow had significantly less HCl Pi bound than the last plot in both layers of the control cell. There were no spatial differences in the alum-treated cell during the 4 mo. sampling while surface soil the first two plots had significantly less HCl Pi bound than the last plot in the 8 mo. sampling. Similar to most variables, there were no si gnificant spatial trends in the 12 mo. sampling due to high spa tial variability between transects. Aluminum and Iron–bound Phosphorus The reactive NaOH-extractable P asso ciated with both amorphous oxyhydroxide and crystalline Al and Fe oxides was the dom in ant fraction of inorganic P in both cells. In the control cell the percent P bound in the NaOH Pi fraction decreased from 22% initially to 14% of the tota l P by the end of the study corre sponding to the 8% increase in HCl Pi mentioned previously. In contrast, NaOH Pi in the alum-treated cell significantly increased from 12% at time 0 up to 57% on average by the 12 mo. sampling.

PAGE 168

149 The NaOH Pi followed a nearly identical temporal trend to that of the AlT suggesting the majority of P in this fraction was bound to Al instead of Fe. At time 0 the control cell had significantly greater NaOH Pi concentrations in bot h the 0-5 and 5-10 cm soil layers than the designated cell to be trea ted with alum (Table 59, Table 5-10). After four months of alum a pplication (~ 29 g Al m-2) there was no longer a difference in surface soil NaOH Pi but the control cell still had signi ficantly more Aland Fe-P in the subsurface. By the 8 and 12 mo. samplings the alum-t reated cell had significantly higher NaOH Pi concentrations in both the surface and s ubsurface compared to the control which can be attributed to the addition of alum. Furt hermore, the time 0 and 4 mo. soil samples had significantly lower NaOH Pi concentrations than the 8 and 12 mo. samplings in the alumtreated cell and even th e 8 mo. sampling yielded significantly lower NaOH Pi concentrations than the 12 mo. alum-treated samples while there were no significant differences in NaOH Pi concentrations among sampling events in the control cell. Spatially, there was no significant difference in surface soil NaOH Pi at experiment initiation. In the subsurface, plots closest to the inflow had significantly lower values than the 80 and 150 m plots in cell 10 and the 35 m plot was less than the 80 m plot in the control cell. During the 4 mo. winter sampling both cells had significantly lower NaOH Pi concentrations in the surface soil of plots closest to the inflow compared to the 10-35 m plots, continuing in the alum-treated cel l through the 8 mo. sampling. Data were highly variable due to the hete rogeneous nature of the alum floc deposition resulting in no spatial differences in either layer during the final sampling.

PAGE 169

150 Organically-bound Phosphorus The NaOH-extractable non-reactive P (NaOH Po) in the fractionation represents the P associated with humic and fulvic acids as well as bacteria incorporated P. Values remained relatively constant in the control cell composing 33-38% of the total P while in the alum-treated cell the frac tion increased from 27% initia lly to 35% by the end of the study. The NaOH Po concentrations followed a nearly id entical temporal trend to that of the Alox. Initially and in the 4 mo sampli ng there was significantly more NaOH Po in both the surface and subsurface soil of the control cell as compared to the alum-treated cell (Table 5-11, Table 5-12). However, in the last two samplings the alum-treated surface soil had significantly higher organicbound P than the control with no significant differences in the subsurface. Similar to the trend in NaOH Pi, the time 0 and 4 mo. soil samples had significantly lower NaOH Po concentrations than the 8 and 12 mo. samplings in the alum-treated cell and even the 8 m o. sampling yielded signi ficantly lower NaOH Po concentrations than the 12 mo. alum-treated samples while there were no significant differences in NaOH Po concentrations among sampling ev ents in the control cell. Spatially, there was no significant difference in surface soil NaOH Po at experiment initiation while in the subsurface the control cell had significantly greater concentrations at the 35-80 m plot than a ll others along the transect s uggesting an area of organic deposition within the cell whil e in cell 10 the first two plots had significantly lower contents than the back two plots. Duri ng the 4 mo. sampling th ere were no spatial differences in the alum-treated cell wh ile in the control, surface soil NaOH Po in the first plot was significantly lower than in the last plot. By the 8 mo. sampling surface soil of the alum-treated soil in the plot closest to the inflow was significantly lower than all

PAGE 170

151Table 5-11. Mean organic phosphorus derived from the inorganic phosphorus fractionation data from the 0-5 cm soil layer in cel l 9 and 10 of the Orlando Easterly Wetland. Values are means 1 standard deviation (n=3). Parameter Distance from Alum Control inflow (m) 0 mo 4 mo 8 mo 12 mo 0 mo 4 mo 8 mo 12 mo NaOH Po 0-10 13.69 12.22 39.49 72.40 105.2 160.4 2194 2526 427.5 281.6 74.63 92.91 73.03 109.1 282.1 120.0 mg kg-1 10-35 174.9 91.68 351.1 228.1 1267 1051 2825 2859 383.2 213.2 482.9 256.8 234.9 189.8 263.2 155.4 35-80 310.0 50.21 358.4 167.6 783.8 1119 1361 1296 399.9 46.72 353.4 132.6 300.4 40.72 240.8 81.79 80-150 319.1 56.98 236.3 136.8 645.5 186.3 2458 1706 395.3 220.8 427.3 200.6 390.2 134.5 300.5 164.2 Residue Po 0-10 30.26 29.60 36.78 76.79 11.60 14.25 260.4 102.3 254.5 120.2 92.96 129.0 80.47 127.8 280.7 118.6 mg kg-1 10-35 346.3 192.8 236.0 112.3 324.4 115.6 336.0 111.3 231.8 124.2 369.8 149.6 197.7 134.5 219.4 124.7 35-80 409.0 71.09 416.8 187.5 310.7 160.9 277.9 79.62 438.3 66.35 388.1 137.2 321.9 70.47 326.7 113.8 80-150 288.6 87.68 280.9 140.6 419.2 156.4 265.7 103.6 375.1 226.6 425.5 197.3 444.8 122.7 350.8 151.7 Total Po 0-10 43.95 41.72 76.27 149.2 116.8 174.2 2455 2561 681.9 393.7 167.6 220.3 153.5 236.8 562.8 235.9 mg kg-1 10-35 521.2 260.3 587.2 331.2 1592 990.8 3161 2830 615.0 326.1 852.7 402.8 432.5 321.5 482.6 275.9 35-80 718.9 101.6 775.2 351.4 1095 1217 1639 1293 838.3 83.13 741.5 260.5 622.4 80.85 567.5 194.9 80-150 607.6 132.2 517.1 273.5 1065 246.2 2724 1625 770.4 439.9 852.8 393.8 835.0 249.7 651.3 314.2

PAGE 171

152Table 5-12. Mean organic phosphoru s derived from the inorganic phosphorus fractionation data from the 5-10 cm soil layer in ce ll 9 and 10 of the Orlando Easterly Wetland. Values are means 1 standard deviation (n=3). Parameter Distance from Alum Control inflow (m) 0 mo 4 mo 8 mo 12 mo 0 mo 4 mo 8 mo 12 mo NaOH Po 0-10 7.774 4.413 51.39 72.75 34.20 67.72 145.3 151.2 160.4 91.73 98.87 73.01 56.50 28.19 99.54 46.90 mg kg-1 10-35 47.26 32.65 52.06 35.14 70.64 60.06 716.1 1644 85.36 26.98 299.5 363.8 74.68 41.42 88.51 60.52 35-80 172.5 124.8 84.20 129.7 117.4 113.8 112.8 77.79 357.3 65.00 259.5 166.6 231.0 106.8 194.7 124.8 80-150 167.4 67.42 75.52 42.06 172.8 90.61 264.2 295.8 116.3 102.0 299.8 213.8 180.5 157.1 93.79 111.1 Residue Po 0-10 5.680 2.057 64.61 96.81 8.595 8.593 254.6 269.4 169.4 115.5 124.2 94.88 75.48 37.54 113.3 61.55 mg kg-1 10-35 97.23 85.33 65.49 30.86 102.5 53.32 144.1 100.4 102.1 48.79 212.9 240.8 96.64 74.23 94.77 72.69 35-80 455.0 283.8 167.4 198.0 198.6 139.6 178.6 132.0 406.2 84.21 322.1 160.7 307.8 110.4 240.5 136.7 80-150 510.7 241.3 220.0 165.9 369.6 111.6 125.2 124.5 163.1 145.3 316.4 206.7 199.2 148.7 168.7 186.1 Total Po 0-10 13.45 3.175 116.0 169.5 42.79 75.79 399.9 302.7 329.9 205.7 223.1 167.8 132.0 65.42 212.8 105.7 mg kg-1 10-35 144.5 106.1 117.5 59.90 173.1 100.7 860.2 1668 187.4 73.55 512.4 602.3 171.3 115.3 183.3 132.3 35-80 627.5 391.1 251.6 326.8 316.0 251.0 291.4 201.7 763.5 140.4 581.6 322.2 538.8 213.7 435.1 260.3 80-150 678.2 303.2 295.6 207.3 542.4 183.4 389.4 412.8 279.4 247.3 616.2 419.2 379.7 305.5 262.4 294.3

PAGE 172

153 others along the transect while there were no significant differences in the subsurface or 12 mo. sampling for either cell. Residual Phosphorus The residue Po represents the highly resistant refractory organic P and any other inert mineral P fractions not extracted with salt, acid, or base. It composed 30-35% of the total P in the control cell and 30-42% of the total P in the alum-treated cell, remaining relatively constant throughout the study with no significant differences between cell concentrations or temporal differences in either layer, however, there were apparent spatial differences. At time 0, the plot closest to the in flow had significantly less residue Po than all others in cell 10 surface soil wh ile in the subsurface, the first two plots had significantly lower values than the last tw o plots. In the control cell, the 10-35 and 80-150 m plots had significantly lower residue Po concentrations than the 35-80 m plot subsurface soil similar to the NaOH Po concentrations, suggesting once agai n a high depositional area. After four months, the plot closest to the inflow had significantly less residue Po than the plot farthest from the inflow in the surface of the control cell and subs urface of the alumtreated cell. Alum-treated surface soil in the first plot had less recalcitrant Po than the 35 and 80 m plots during the 4 mo. sampling. In surface soil during the spring 8 mo. sampling, 0-10 m plots in the alum cell had significantly less residue Po than all others and less than the 150 m plot in the control cell. In the subs urface of the alum-treated the 0-10 m plots continued to have significantly less residue Po than all others while in the control cell the first two plot s contained less than the 35-80 m plot. Spatial variability resulted in no significant differences in th e last sampling similar to most variables.

PAGE 173

154 X-Ray Diffraction X-ray diffraction analysis of the combined silt an d clay fraction within both cells revealed the predominance of quart z and presence of kaolinite (Al2Si2O5(OH)4) as well. The presence of organic matter in both cel ls was also indicated by a broad amorphous hump on the spectrum. Surface soil samples from the 12 mo. sampling were used to detect any crystallization of the Al-bound P over the course of a year. Over time this floc should slowly consolidate and crystalli ze as larger units of polymeric Al(OH)3 are formed (Burrows, 1977). The exact transformation path of poorly crystallin e to crystalline Al hydroxide is unclear (Bertsch and Parker, 1996) and, as these results suggest, may not occur in some wetlands due to the high concen tration of organic acids present. Humic and fulvic acids are high molecular weight acids that tend to prevent Al(OH)3 crystallization because functional groups on the acids form stable complexes with the Al (Kodama and Schnitzer, 1980; Singer and Huang, 1990). Conclusions The long-term application of low-dosage alum had variable results. Soil pH rem ained relatively constant throughout the experiment averaging 6.5, optimal for Al-P binding. There was, however, a significant incr ease in surface soil Al associated with the hydrolyzed alum floc in the alum-treated cell. Additionally there was a significant increase in surface soil TP bound by the alum floc over time which corresponded to eight times more NaOH Pi at the end of the study than at the beginning in the alum-treated cell. The increased soil TP can be attributed to the reduction in water column SRP, TDP, and TP concentrations by the hydrolyzed floc wh ich then settles to the soil surface. Despite these significant alterations in so il characteristics there appeared to be no discernable impact on the micr obial population or activity throughout the year. Nor did

PAGE 174

155 alum application impact wetland soil mineralogy. Therefore, alum may be considered as an effective, long-term management tool in aging treatment wetlands. Phosphorus should remain bound within the floc la yer and move downward in th e soil profile as sediment and organic matter are deposited at the soil su rface. Caution should be taken, however, in implementing alum applications as both short-term core studies and a winter mesocosm experiment have shown negative impacts, including decreased microbial biomass and activity, Al toxicity in submerged aquatic vegetation, and increased particulate P export.

PAGE 175

156 CHAPTER 6 SPATIAL AND TEMPORAL EF FECTS OF CONTINUOUS LOW-DOSAGE ALUM ON FIELD WATER QUALITY AND AQUATIC MACROPHYTE CHARACTERISTICS Introduction Alum (Al2(SO4)3H2O) is the chemical amendment used most often for phosphorus (P) inactivation in lakes and coagulation in the wastewater treatment industry. When added to the water column alum dissociates, forming aluminum ions (Al3+) that are immediately hydrated. Through se veral rapid hydrolytic reactions an insoluble, gelatinous, poorly crystalline aluminum hydroxide (Al(OH)3) floc is formed (Ebeling et al., 2003). This floc has high P adsorption properties, removing both soluble and particulate P both by adsorption and physical entrapment (Galarneau and Gehr, 1997). The controlling factor in the effectiveness and toxicity of alum is the pH of the system. Alum solution itself has a pH of approximately 2.4 (Beecroft et al., 1995; Lind, 2003) and therefore tends to decrease the pH of the system to which it is added. As long as the pH of the system remains betw een 6 and 8, insoluble polymeric Al(OH)3 will dominate and P inactivation results (May et al ., 1979). If the pH d ecreases to between 4 and 6 soluble intermediates will occur, re leasing bound P (Cooke et al., 1993a). Below pH 4 soluble Al3+ dominates which may result in alum inum (Al) toxicity (Cooke et al., 1993b), and at pH 8 or greater the aluminate ion (Al(OH)4 -) dominates, due to its amphoteric nature, releasing bound P and increasing soluble Al (Cooke et al, 1993a).

PAGE 176

157 Aluminate, similar to Al3+, is associated with Al toxicity in plants (Kinraide, 1990, Eleftheriou et al., 1993; Ma et al., 2003; Malecki-Brown and White, 2007b). The speciation of Al determines its mobility, bioavailability, and toxicity in aquatic ecosystems (Bertsch, 1990). In the case of plants, Al(H2O)6 3+ (or simply referred to as Al3+), Al(OH)2+,and some complex polymers such as Al13O4(OH)24(H2O)12 7+ are the most toxic (Kinraide, 1991; Kochian, 1995), while the Al that comp lexes with organic anions is relatively harmless (Wickstrom et al., 2000) . Therefore, a clear comprehension of the quantity and Al form present is needed to assess the impact of increased concentrations in alum treated ecosystems. The uptake of bioavailable metals in a quatic macrophytes generally occurs via surface adsorption or absorption, followed by pa ssive and active transport across cell membranes, incorporating meta ls into biochemical functions or storing them in a bound form (Outridge and Noller, 1991; Rai et al., 1995). The pH of wetland soils is near neutral (6.5-7.5) typically favoring metal immobilization (Gambrell, 1994). However, several studies have shown that wetland acidi fication can increase the bioavailability of metals resulting in elevated concentrations in aquatic plants (Leht onen, 1989; Albers and Camardese, 1993; Jackson et al., 1993; V azques et al., 2000; Cardwell et al., 2002; Gallon et al., 2004). While alum has been used for P inact ivation in eutrophic lakes since 1968 (Blomquist et al., 1971) there has been little research done on its potential effectiveness in aging treatment wetlands with reduced P sorption capacities (Simon, 2003; DB Environmental, Inc., 2004; Malecki-Br own and White, 2007a; Malecki-Brown and White, 2007b; Malecki-Brown et al., 2007a; Malecki-Brown et al ., 2007b). It is

PAGE 177

158 important to determine how the long-term a pplication of low-dose alum will affect the bioavailability of metals and nutrients within the water, soil, and in turn the macrophytes within constructed wastewater treatment wetla nds. All three compartments are critical in the nutrient cycling and treatmen t efficiency of P. The hypothesis of this research was that alum application would improve P re moval and overall water quality within a constructed wastewater treatment wetland while in no way impacting the vegetation within a Typha spp. dominated system. Materials and Methods Site Description The Orlando Easter ly Wetlands (OEW) R eclamation Project located in Orange County, is one of the oldest and largest c onstructed treatment wetlands in the United States, located east of Orlando in Christ mas, FL. The wetland was built in 1986, designed by Post, Buckley, Schuh & Jernigan, In c. for the City of Orlando’s Iron Bridge Regional Water Pollution Control Facility (WPCF) which needed an alternative discharge point for its wastewater effluent (Burney et al., 1989). The main goal in designing the system was to use macrophytes to facilitate additional nutrient removal for an average daily flow of up to 132,489 m3d-1 of effluent from the Iron Bridge WPCF before discharging into the St. Johns River (Black and Wise, 2003). The 494 ha wetland rests on a 664 ha piece of land located 3.2 km west of the main channel of the St. Johns River (SJR). Histori cally, the land had been part of the riparian wetland adjacent to the SJR, but was drained for pasture by a cattle ranch around the turn of the last century (Burney et al., 1989). The site has a natural topographic gradient of 4.6 m downward from west to eas t allowing water to flow by gravity through a series of

PAGE 178

159 cells with an average elevation drop across e ach cell of approximately 1 m (Martinez and Wise, 2003) (Figure 6-1). Water exits the we tland through a weir c ontrol structure and Figure 6-1. Site map of Orlando East erly Wetland, Christmas, Florida. Typha spp. dominated cell 9 served as the control, and ce ll 10 was treated with alum for one year. flows into a receiving ditch. From there wa ter can flow directly to the SJR or by sheet flow through Seminole Ranch, a natural marsh adjacent to the OEW owned by the St. Johns River Water Management District. Cells 1 through 12 and 15 are deep marsh, designed primarily for nutrient remova l, planted with either cattails (Typha spp.), giant bulrush ( Scirpus californicus), or a combination of the two. Cells 13, 14, and 16 through 18 consist of a mixed marsh dominated by submergent and emergent macrophytes including Ceratophyllum demersum , Limnobium spongia, Myriophyllum spicatum, Najas Project Location Created by: Lynette Malecki Dated: 12/30/04 2 1 3 4 5 6 7 8 9 10 11 12 15 13 14 16 17 18 Lake

PAGE 179

160 guadalupensis, Nuphar luteum, Nymphaea odorata, Pontederia cordata, Sagittaria lancifolia, and Sagittaria latifolia . These cells serve as a diverse wildlife habitat while continuing to provide nutrient removal (Martinez and Wise, 2003). The overall average influent total P (TP) concentration from 1988 to 2005 was 0.22 mg L-1, however, annual inflow TP concentr ations ranged from 0.02 – 3.30 mg L-1 during the same time period. Since its incep tion, the OEW has exceeded performance expectations. The TP discharge permit limit established by the Florida Department of Environmental Protection is 0.2 mg L-1 (Wang et al., 2006). From 1988 to 1995 the average TP discharged was 0.07 mg L-1 (Sees and Turner, 1997), however, TP values have been considerably higher from December to February in recent years (Wang et al., 2006). Pre-experiment Water Quality Monitoring Water quality parame ters including pH, dissolved oxygen (D.O.), temperature, and conductivity were monitored at cell inflow and outflow weirs weekly for six months prior to experiment initiation (Feb. 3, 2005 – July 28, 2005) using a handheld YSI 85 (YSI Inc., Yellow Springs, OH). Additionally, three types of water samples were collected from the inflow and outflows of each cell (Fi gure 6-2) using an ISCO portable peristaltic pump. Samples included: unfiltered acidified for TP analysis; filtered (0.45m Whatman polydiscTM AS 50 mm inline filters) acidified for total dissolved P (TDP) and DOC; and filtered for soluble reactive P (SRP) and dissolved Al. Acidified samples were collected in wide-mouth high density polyethylene (HDPE) 125 or 60 mL bottles and refrigerated at 4 C unt il digested and or analyzed. N on-acidified samples were collected in 20 mL HDPE scintillation vials and frozen until analyzed via automated, colorimetric analysis for SRP (Method 365.1, USEPA, 1993). Dissolved Al was determined by

PAGE 180

161 inductively coupled arg on plasma spectrometry ( Vista MPX CCD simultaneous ICP-OES manufactured by Varian , Inc., Walnut Creek, CA) (Method 200.7, USEPA, 1993). Water TP and TDP were digested by autoclaving a nd then analyzed colorimetrically (Method 365.4, USEPA, 1993). Transect Establishment On May 23, 2005 duplicate belt transects 2.5 m wide were established in each cell stretching from the inflow weir on the southw est of each cell to th e two outflow weirs on the northeast end. Each transect was divide d into four plots spaced to capture any resultant gradients formed from alum applicati on. The plots were es tablished at distances of 0-10 m, 10-35 m, 35-80 m, and 80-150 m from the inflow and marked using 2.5 m polyvinyl chloride poles that were pounded into the soil using dead-blow mallets and remained throughout the study (Figure 5-2). To prevent constant disturbance of the cells by wading in to take weekly water samples along transects, black vi nyl tubing (0.95 cm i.d.) was installed from the berms to the far interior of each plot. Due to the unforeseen preference for chewing on vinyl by otters and muskrats over th e next two months (Kadlec et al., 2007), all tubing was replaced with flexible black polyethylene tubing. Pool filter pump baskets fitted with black poly hardware mesh (0.64 x 0.95 cm) fit ove r the uptake end of each tube to filter out soil and debris as well as prevent clogging within the tubing when using the peristaltic pump to retrieve water samples. Black tubing was used to prevent algae gr owth, and lines remained water-filled at all times such that the weight would keep them in place under the water surface or dense vegetation to prevent tube heating, as well as to maintain suction for easier sampling.

PAGE 181

162 Depending on the transect sampled the pump wa s allowed to run from 20 sec to 4 min prior to sampling to assure a ne w, clean sample was collected. ` Figure 6-2. Map of study transe cts established in cell 9 and 10 of the Orlando Easterly Wetland, Christmas, Florida. Experiment Initiation Beginning August 3, 2005, commercial grade liquid alum (General Chemical Corp.) was pumped via solar-charged pump s to the inflow of cell 10 through volume Created by: Lynette Malecki Brown Dated: 12/11/06 9 10

PAGE 182

163 regulated black polyethylene tubing at an average rate of 158 mL min-1 (60 gal d-1) for one year resulting in a to tal addition of 88.6 g Al m-2. Water samples were collected weekly for the first three months, then every other week for the remainder of the study. In addition, plant and soil samples were collected at time 0 (Aug. 2005), 4 mo (Dec. 2005), 8 mo (April 2006), and 12 mo (Aug. 2007). Plant biomass and tissue nutrient concentra tion were determined from triplicate 0.1 m2 quadrats randomly selected and harvested from each field plot. All above-ground plant material from living Typha plants within the quadrat was removed and placed in labeled Ziploc bags within icefilled coolers for transport to the laboratory. Plants were counted and rinsed thoroughly w ith tap and then distilled wate r, and all visible algae or epiphytes wiped off. Plant tissue was then separated into live (green) and standing dead (brown) leaves which were counted, cut into roughly 10 cm pieces, placed in paper bags and weighed. Bags were dried at 40 C until a constant weight was reached (APHA, 1998) to determine biomass (g m-2) as well as the percent allocation of biomass to live and dead aboveground components (Smith and Newman, 2001). The relative growth rate (RGR) of Typha was calculated as: RGR = ln(W2)–ln(W1)/(t2-t1) where W1 and W2 are the initial and final plant dr y weights at time 0, 4, 8 (t1) and time 4, 8, and 12 mo. (t2) (Forchhammer, 1999; Brix et al., 2002; Hada d et al., 2006). Tissue was then ground using a Wiley mill and analyzed for total C (TC) and total N (TN) using an Elemental Combustion ECS 4010 CHNS-O Analyzer (Cos tech Analytical Technologies, Inc., Valencia, CA). Total P and total metal analysis involved dry ashing 0.2 g oven-dried subsamples of plant tissue at 550 C for 4 h in a muffle fu rnace followed by dissolution of the ash in 6

PAGE 183

164 M HCl on a hot plate (Gorsuch, 1970; Anders en, 1976; Jones et al., 1991). Total P was analyzed using the automated ascorbic acid-molybdenum blue colorimetric method (Method 365.4, USEPA, 1993). Initial and final plant samples were analyzed for total Al, calcium (Ca), potassium (K ), iron (Fe), and magnesium (Mg) by inductively coupled argon plasma spectrometry (Vista MPX CCD simultaneous ICP-OES manufactured by Varian, Inc., Walnut Creek, CA) (Barnes, 1975; Campbell et al., 1983; Easthouse et al., 1993; Rydin and Welch, 1999; Rydin et al., 2000; Method 200.7, USEPA, 1993). Total P and N concentrations were used to calculate the nutrient resorption efficiency (RE) (Killingbeck, 1996; Aerts et al., 1999) and nutr ient use efficiency (NUE) (Vitousek, 1982). The RE serves as a measur e of nutrient conservation since in nutrientpoor wetlands, plants tend to resorb and tr anslocate the majority of nutrients from senescing tissue to belowground tissue for storage, therefore having a high RE. Nutrient resorption efficiency was calculated as: (N or P (g m-2) in green biomass – N or P (g m-2) in standing dead biomass) N or P (g m-2) in green biomass where biomass is aboveground harvested mate rial (Killingbeck, 1996). Nutrient use efficiency measures the eff ectiveness of macrophytes to utilize nutrients to produce biomass (Vitousek, 1982). Nutrient-poor wetlands generally have a high NUE, producing large amounts of biomass with little loss of N or P in litterfall corresponding to the high rate of nutrient reso rption occurring. Nutrient use efficiency was calculated as: aboveground standing dead biomass (g m-2) / N or P (g m-2) in standing dead biomass (Vitousek, 1982). x 100%

PAGE 184

165 Statistical Analysis Data norma lity was determined using the Kolmogorov-Smirnov test (Minitab 13.32, 2000) and data was transformed to fit a normal distribution (Microsoft Excel, 2000). One-way ANOVAs and multiple comparisons by Tukey’s W were used on plant biomass and nutrient parameters while rep eated measure ANOVAs using a general linear model followed by Tukey’s W multiple comparison were used on water column data to determine significant differences (p<0.05) betw een the control and treated cell (Minitab 13.32, 2000). Linear regression analysis and Pearson product correlation coefficients were also used to determine significant (p <0.05) relationships (Mic rosoft Excel, 2000). Results and Discussion Water Quality Characterization pH The pH of both cells rema ined above 5.5 and below 8.5 throughout the experiment (Figure 6-3). The overall average water column pH in both cells was 6.9 0.4 which is ideal for the alum floc to bind P and remain insoluble. On only one measured occasion did the inflow pH, where alum entered the cell, drop below 6.0 where soluble Al intermediates may have been present, possibly re-releasing previously bound P (Cooke et al., 1993b). However, all water samples were an alyzed for soluble Al and concentrations throughout the experiment were below the pr actical quantification limit of 20 mg Al L-1. These results are similar to the findings of Sparling and Lowe (1998) which were attributed to high DOC concentrations bi nding with any available soluble Al.

PAGE 185

166 5.0 6.0 7.0 8.0 9.0 04896144192240288336384 Time (d)pH Inflow Outflow (a) 5.0 6.0 7.0 8.0 9.0 04896144192240288336384 Time (d)pH Inflow Outflow (b) Figure 6-3. Water column pH of (a) alum-t reated cell 10, and (b) control cell 9 in the Orlando Easterly Wetland (inflow n=1, outflow n=2). Dissolved Oxygen Dissolved oxygen levels ra nged between 0.4 and 6.1 mg L-1 throughout the experiment (Figure 6-4) with an overall av erage water temperature of 21.3C 4.7 in the control cell and 19.9C 5.4 in the alum-treated cell throughout the study. Inflow D.O. concentrations were greater th an outflow concentrations due to weir design and water velocity. Prior to entering each cell, water flows over a weir waterfall from the previous

PAGE 186

167 0.0 2.0 4.0 6.0 8.0 050100150200250300350400 Time (d)Dissolved Oxygen (mg L-1) Control Inflow Control Outflow Alum Inflow Alum Outflow Figure 6-4. Water column dissolved oxygen content in cell 9 and 10 of the Orlando Easterly Wetland (inflow n=1, outflow n=2). cell and is then channeled th rough a large culvert which rapidly flows into the succeeding cell. Therefore water is oxygena ted as it enters each cell but as it slows in velocity for treatment, passing through the densely vegetated Typha-dominated cells. Oxygen is consumed until reaching the outflow weir where the process begins again. Dissolved Organic Carbon Concentrations of DOC were highly variable along transect s during the first 100 days of the study in both cells , averaging 20.5 34.1mg L-1 in the control cell and 15.5 27.8 mg L-1 in the alum-treated cell (Figure 6-5). The alum-treated cell did have lower DOC concentrations than the control in the remainder of the study averaging 7.4 3.1 and 9.8 10.2 mg L-1 respectively. Both cells displayed net increases in DOC from the inflow to outflow, averaging a 4.8% increa se in DOC in the control cell and 1.2% increase in the alum-treated cell. The 0-10 cm plot in the alum-treated cell, closest to the inflowing alum was the only plot in both cel ls with an average net removal of DOC (4.6%) while the same plot in the control cell was the greatest source of DOC averaging a

PAGE 187

168 0 25 50 75 100 125 04896144192240288336384 Time (d)Dissolved organic C (mg L-1) 0-10 m 10-35 m 35-80 m 80-150 m (a) 0 25 50 75 100 125 04896144192240288336384 Time (d)Dissolved organic C (mg L-1) 0-10 m 10-35 m 35-80 m 80-150 m (b) 4 6 8 10 12 14 04896144192240288336384 Time (d)Dissolved organic C (mg L-1) Control Inflow Control Outflow Alum Inflow Alum Outflow (c) Figure 6-5. Water column dissolved organic C concentrations in (a) control cell 9 transects, (b) alum-treat ed cell 10 transects, (c) in flow and outflow weirs of the Orlando Easterly Wetland (n=2).

PAGE 188

169 222% increase in conc entration compared to the DOC of inflow water entering the cell (Figure 6-5 a, b). Alum is not only used for P removal, but also to clarify turbid lakes by acting as a source of positive electrolytes ne utralizing negatively-charged soil particles which causes them to settle out of the wa ter column (Davis and Gloor, 1981; Berg and Berns, 1985; Sparling and Lowe, 1998; Van Hullebusch et al., 2002). Wetlands, however, have a shallower water column and de nse stands of vegetation that can result in a continuous deposition of carbon to the wa ter column resulting in decreased alum effectiveness in DOC removal. Soluble Reactive Phosphorus The cell treated with alum was more e ffective at SRP removal averaging 36.4% with water column concentrati ons ranging between 0.00 1.03 mg L-1 throughout the experiment (Figure 6-6 a). Initially, SRP c oncentrations were highl y variable throughout the cell, however, at day 216 (approximatel y 7 months) alum had migrated throughout the entire cell with outflow SRP concentra tions decreasing steadily until remaining below 0.05 mg L-1 from day 261 onward, no longer in fluenced by peaks in inflow concentrations. All internal sampling point s along the transect also maintained SRP concentrations below the concentration of inflow SRP after day 216 once again suggesting alum had penetrated thro ughout the cell (Figure 6-6 a). There was a spatial gradient in P removal efficiency of the cell with plots closest to the alum inflow averaging 46.8% removal, the 35 m plots averag ing 45.6%, and the 80 m plots averaging 22.7% removal. The plot fart hest from the inflow (150 m) served as a net source of P up until day 160 when alum proba bly reached that area of the cell. At the 150 m plot P increased on average 65.2% over the course of the study (-149% from day 0-160 and +52.7% thereafter) suggesting P was fluxing out of the soil into the water

PAGE 189

170 0.0 0.4 0.8 1.2 050100150200250300350400 Time (d)Soluble reactive P (mg L-1) Inflow Outflow 10 m 35 m 80 m 150 m (a) 0 0.4 0.8 1.2 050100150200250300350400 Time (d)Soluble reactive P (mg L-1) Inflow Outflow 10 m 35 m 80 m 150 m (b) Figure 6-6. Water column soluble reactive phos phorus concentrations in (a) alum-treated cell 10, and (b) control cell 9 of th e Orlando Easterly Wetland (n=2). column. In particular, the 80-150 m plot in the west transect corre sponds to an area of high soil TP averaging 11,366 2538 mg kg-1 at the start of the study. This directly shows the ability of alum application to transform wetland soils high in P from sources to sinks. Soluble reactive P concentrations throughout the control cel l were in close agreement, influenced directly by fluctu ations in the inflow SRP (0.01 0.68 mg L-1),

PAGE 190

171 with an overall removal efficiency of 8.0% (Figure 6-6 b). The second plot from the inflow (80 m) within the control cell wa s a relatively constant P source averaging a removal efficiency of -1.7% throughout the stud y suggesting release of P from the soil to the water column in that area of the cell. The other plots within the control cell averaged 8.9-14.5% SRP removal efficiencies. Total Dissolved Phosphorus The TDP includes both the SRP and dissolved organic P (DOP). Figure 6-6 and 6-7 look very similar due to the dom inance of the SRP fraction with in the water column and the results are very similar as well. To tal dissolved P concentrations were directly correlated (p<0.01) to those of SRP in both cells throughout the study. The alum-treated cell averaged 39.1% TDP removal thr oughout the study, with water column concentrations ranging between 0.004 1.10 mg L-1 (Figure 6-7 a). Initially, TDP concentrations were highly variable th roughout the cell similar to SRP, however approximately seven months after alum a pplication began, the hydrolyzed floc had migrated throughout the entire cell with outflow TDP concentrations decreasing steadily until remaining below 0.10 mg L-1 from day 244 onward, no longer influenced by peaks in inflow concentratio ns (Figure 6-7 a). Also similar to SRP removal, there was a spatial gradient in TDP removal within the alum-treated cell. Plots closest to the alum inflow averaged 49.1% removal, the 3580 m plots averaged 46.2%, and the 35-80 m plots averaged 24.7% removal. Once again, the plot farthest from the inflow at 150 m served as a net source of P up until day 160 after which the alum floc extended into that area of the cell, averaging 55.6% P addition over the course of the study (-147% from day 0-160 and +41.3% thereafter).

PAGE 191

172 0.0 0.4 0.8 1.2 04896144192240288336384 Time (d)Total dissolved P (mg L-1) Inflow Outflow 10 m 35 m 80 m 150 m (a) 0.0 0.4 0.8 1.2 04896144192240288336384 Time (d)Total dissolved P (mg L-1) Inflow Outflow 10 m 35 m 80 m 150 m (b) Figure 6-7. Water column dissolved reactiv e phosphorus concentrations in (a) alumtreated cell 10, and (b) cont rol cell 9 of the Orlando Easterly Wetland (n=2). Total dissolved P concentrations thr oughout the control cel l were in close agreement, influenced directly by fluc tuations in inflow TDP (0.10 0.71 mg L-1), with an overall removal efficiency of 4.9%, slightly lower than th at of SRP (Figure 6-7 b). Similar to the findings in spatial SRP, the s econd plot from the inflow (80 m) within the control cell was also a relative ly constant TDP source averagi ng a removal efficiency of -3.7% throughout the study while the other plots averaged 1.1 6.2% TDP removal.

PAGE 192

173 Total Phosphorus The water column TP includes particulat e P (PP) in addition to SRP and DOP. The average inflow TP to both cells was com posed of 50% SRP, 26% DOP and 24% PP. The average alum-treated cell outflow water ( 0.16 0.11 mg L-1) had a much lower TP concentration than the av erage of the control cell outflow (0.32 0.15 mg L-1) which is surprising when looking at Figure 6-8. The grap h of cell 10 shows that by the time water flowed through the first plot the TP con centration actually incr eased, continuing to decrease as the water moved through the cell to the outflow (Figur e 6-8 a), averaging a 30.4% removal. Similar to SRP and TDP the control cell TP concen trations are very consistent throughout th e cell (0.05-0.74 mg L-1) averaging a net P removal of 4.0% (Figure 6-8 b). While the inflow and outflow TP concentra tions in the alum-treated cell resemble the prior graphs of SRP and TDP, the graphs of internal TP concentrations along the transects are considerably diffe rent which can be attributed to the PP component of the TP concentration. The average composition of the outflow TP from both cells was very similar to the alum-treated cell averaging 77% SRP, 13% PP, and 10% DOP while in the control, outflow TP consisted of 89% SRP, 2% PP, and 9% DOP. The primary difference being a 12% increase in particulate-bound P and corresponding 12% decrease in SRP in the alum-treated cell. Similar re sults were found in Mal ecki-Brown and White (2007b). Total P concentrations in the water column were dir ectly correlated (p<0.01) to PP concentrations in both cells during the firs t four months of the experiment. From winter until study completion, however, TP was onl y correlated to PP in the alum-treated cell, while in the control cell TP was direc tly related (p<0.01) to TDP concentrations.

PAGE 193

174 0.0 0.7 1.4 2.1 2.8 04896144192240288336384 Time (d)Total P (mg L-1) Inflow Outflow 10 m 35 m 80 m 150 m (a) 0.0 0.7 1.4 2.1 2.8 04896144192240288336384 Time (d)Total P (mg L-1) Inflow Outflow 10 m 35 m 80 m 150 m (b) Figure 6-8. Water column total phosphorus co ncentrations in (a) alum-treated cell 10, and (b) control cell 9 of the Orlando Easterly Wetland (n=2). Particulate Phosphorus The calculated PP was the only water quality param eter for which the alum-treated cell had greater concentrations than the contro l (Figure 6-9) as indi cated in the previous discussion of TP. Cell 10 averaged 0.021 0.025 mg P L-1 at the outflow and -171% PP removal, acting as a net source while th e control cell averaged 0.006 0.008 mg P L-1 and 5.8% PP removal. The increased PP in the alum-treated cell suggests that some of

PAGE 194

175 0.0 0.5 1.0 1.5 2.0 2.5 04896144192240288336384 Time (d)Particulate P (mg L-1) Inflow Outflow 10 m 35 m 80 m 150 m (a) 0.0 0.5 1.0 1.5 2.0 2.5 04896144192240288336384 Time (d)Particulate P (mg L-1) Inflow Outflow 10 m 35 m 80 m 150 m (b) Figure 6-9. Water column particulate phosphorus concentrations in (a) alum-treated cell 10, and (b) control cell 9 of the Orlando Easterly Wetland (n=2). the AlOH3 floc remained suspended or was re suspended from areas of deposition, flowing through the cell to be exported from the wetland, in th is case into the receiving cell (cell 14), in the less bioava ilable, particulate form. This is particularly evident in Figure 6-9 a, where PP concentrations are on av erage 73% greater in the 0-10 m plot than in the inflowing water prior to alum additi on. As water flows through tortuous paths of

PAGE 195

176 dense Typha , the AlOH3 floc slowly settles out of the water column to the soil surface averaging 45% more PP than the inflow concentration after 35 m, 23% after 80 m and only 6.4% after 150 m. Particulate P concentrations in the co ntrol cell also varied throughout the study which can most likely be attributed to decomposition and deposition of macrophyte detritus within the wetland wa ter column (Reddy et al., 1999a). While total suspended solids (TSS) were beyond the scope of this study, the increased concentration of PP may have resulted in increased TSS concentrations (Bostan et al., 2000) as well and should be investigated. Typha spp. Characterization Plant and Leaf Densities There were no significant di fferences in plant densities between cells 9 and 10 at the start or end of the study. Th is suggests no negative impa ct on the Typha plants due to one year of alum application (Table 6-1). In fact, during the winter and spring samplings the alum-treated cell had significantly more plants per square meter than the control cell. There was also no significant difference in the number of standing dead leaves per plant in the cells throughout th e study averaging 2.0 1.7 overall (Table 6-1). However, there was a decline in the num ber of live leaves produced pe r plant in the alum-treated cell. At the start of the study there were no significant differences between cells with Typha averaging 8 live leaves per plant. For the remainder of the study, however, Typha plants in the control cell had significantly more live leaves than those plants growing in the alum-treated cell, averaging two to three more leaves per plant during the 4, 8, and 12 mo. samplings. This may indicate that although the alum-treated cell was producing more plants, they were not growing as rapidl y as those of the cont rol cell. Nutrient

PAGE 196

177 limitations may have resulted in slower pl ant growth, impacted by P and possibly N inactivation (Malecki-Brown and White, 2007b) due to alum addition. Biomass During all sampling events both the per cen t allocated to living biomass and quantity harvested was always significantly gr eater than the allocation and quantity of dead biomass (Table 6-1, Table 6-2). There were no significant differences among biomass in the cells throughout the study. Live biomass ranged from 19-1343 g m-2 or 43-100% of the total biomass while standing dead biomass ranged from 0-536 g m-2 or 061% of the total biomass throughout the year. In both cells the quantity of live and dead biomass were significantly greater during th e 12 mo. summer sampling than during the 4 mo. winter or 8 mo. spring samplings. There was a significantly greater allocati on of biomass to liv ing tissue at the 8 mo. sampling (90%) than 12 mo. (81%), and a significantly greater allocation to standing dead tissue during the 12 mo. summer sampling as compared to the 8 mo. spring sampling. It was during this time that st anding dead biomass was inversely related (p<0.01) to the dead tissue TP content. Du e to variability in tr ansects there were no significant spatial differences in biomass within either cell. Both figur es indicate an area of increased live and standing d ead biomass in the farthest plot from the inflow along the eastern transect of cell 10, corresponding to an area of high soil P, averaging 13,850 592.7 mg kg-1 in the surface soil (0-5 cm) at the 8 mo. sampling. Live tissue biomass was directly correlated (p<0.05) to surface soil TP during th e 4 and 8 mo. samplings when biomass quantities were reduced. These results indicate that continuous low-dose alum application for one year did not impact Typha biomass similar to the findings of a mesocosm study by Malecki

PAGE 197

178Table 6-1. Plant density and biomass allocation of Typha spp. plants harvested from cell 9 and 10 in the Orlando Easterly Wetland. Values are means 1 standard deviation (n=6). Parameter Distance from Alum Control inflow (m) 0 mo 4 mo 8 mo 12 mo 0 mo 4 mo 8 mo 12 mo Plant Density 0-10 20 8.0 22 7.5 20 6.3 18 4.1 10 8.0 15 12 18 7.5 25 5.5 plants m-2 10-35 20 8.4 15 5.5 27 8.2 25 8.4 25 8.4 17 8.2 25 5.5 25 5.5 35-80 25 8.4 17 4.1 25 8.4 23 5.5 25 8.4 17 8.2 25 8.4 23 8.2 80-150 10 9.0 18 4.1 23 8.2 20 9.0 15 9.0 10 0.0 13 5.2 13 5.2 Live Leaves 0-10 5.5 4.0 4.4 1.4 6.9 1.2 6.3 1.2 8.5 4.0 7.5 4.0 9.0 3.1 7.3 2.3 plant -1 10-35 6.0 2.0 6.5 1.2 7.1 1.2 4.9 1.4 7.8 1.5 6.2 1.2 6.9 1.5 5.6 1.1 35-80 8.8 4.0 5.3 1.8 6.2 0.9 5.4 1.4 12 4.0 9.3 3.9 10 2.2 7.8 2.8 80-150 9.5 5.6 5.2 1.4 6.4 0.9 6.6 1.0 9.0 5.6 7.5 2.7 11 5.6 10 0.5 Dead Leaves 0-10 4.7 1.3 0.9 0.5 1.1 0.7 1.2 0.7 2.0 1.3 1.9 0.9 1.1 0.8 2.9 1.3 plant -1 10-35 3.3 2.7 2.3 1.5 1.6 0.9 3.5 2.7 2.8 2.7 1.3 1.6 1.5 0.9 2.5 1.5 35-80 2.2 2.6 1.8 1.3 1.3 0.6 3.4 2.6 3.8 2.6 2.2 1.9 2.7 2.0 2.3 2.3 80-150 2.0 3.0 1.3 0.4 2.2 2.0 3.1 3.0 0.5 1.1 2.0 2.0 1.9 1.3 1.0 1.1 Live Biomass 0-10 83 19 77 19 92 4.3 84 11 99 4.0 89 13 94 6.0 78 14 % 10-35 90 20 90 10 91 9.3 74 22 90 20 88 15 90 6.2 72 8.4 35-80 93 10 75 18 91 5.2 72 19 86 17 79 19 90 9.1 86 17 80-150 74 20 87 6.8 83 20 84 14 98 6.0 82 18 89 10.3 96 6.5 Dead Biomass 0-10 17 19 23 19 8.0 4.3 16 11 1.2 4.0 11 13 5.5 6.0 22 14 % 10-35 10 20 10 10 8.8 9.3 26 22 10 20 12 15 10 6.2 28 8.4 35-80 6.7 10 25 18 8.9 5.2 28 19 14 17 21 19 10 9.1 14 17 80-150 26 20 13 6.8 17 20 16 14 2.3 6.0 18 18 11 10.3 4.0 6.5 Dead and live tissue were not separated for analysis, biomass back -calculated for time 0. Biomass calculations based on dry weight.

PAGE 198

179Table 6-2. Biomass of Typha spp. plants on a dry weight basis, harvested from cell 9 and 10 in the Orlando Easterly Wetland. Values are means 1 standard deviation (n=6). Parameter Distance from Alum Control inflow (m) 0 mo 4 mo 8 mo 12 mo 0 mo 4 mo 8 mo 12 mo Live Biomass 0-10 307 348 237 184 279 144 534 348 767 348 273 147 512 324 613 279 g m-2 10-35 632 295 247 96.7 421 123 380 175 486 295 149 73.5 369 146 545 295 35-80 533 333 284 157 333 241 459 333 547 333 195 47.1 411 200 289 227 80-150 788 492 513 243 449 137 584 324 396 492 166 93.6 470 492 488 221 Dead Biomass 0-10 60.4 62.0 108 113 9.04 3.72 83.8 62.1 3.95 3.70 103 139 19.6 25.4 211 152 g m-2 10-35 67.3 56.0 52.7 44.3 6.82 5.99 146 142 51.9 44.0 62.9 55.9 27.3 17.0 219 163 35-80 40.6 36.0 98.1 97.4 8.83 6.48 142 70.5 66.4 69.0 92.2 83.4 37.2 35.7 55.9 69 80-150 125 200 85.5 74.1 41.0 43.6 163 205 11.0 7.90 44.5 62.9 22.7 7.87 42.6 58 Dead and live tissue were not separated for analysis, biomass back -calculated for time 0.

PAGE 199

180 Brown and White (2007b). Emergent plants tend to obtain a large portion of their nutrients and metals directly from the soil root zone (Crowder, 1991; Rai et al., 1995; Sparling and Lowe, 1998; Madsen and Cede rgreen, 2002). It suggests that alum therefore does not impact growth of rooted aquatic macrophytes like Typha because they are able to access the P content in the subsurface soil (Carignan and Kalff, 1979; Welch and Schrieve, 1994) beneath the floc layer. Relative Growth Rate To comp are the growth of Typha within the control and alum cells, the RGR was calculated for each (Table 6-3). There were no significant differences in plant growth rates throughout the study either by cell or di stance. There were seasonal differences however, with growth declining from August to December resulting in significantly Table 6-3. Relative growth rate s based on total dry weight of Typha spp. harvested from cell 9 and 10 of the Orla ndo Easterly Wetland. Values are means 1 standard deviation (n=6). Distance from Time RGR (g g-1 d-1) inflow (m) (mo.) Alum Control 10 0-4 0.002 0.011 -0.014 0.008 35 -0.012 0.003 -0.015 0.005 80 -0.010 0.008 -0.011 0.008 150 -0.003 0.018 -0.012 0.010 10 4-8 0.003 0.016 0.006 0.012 35 0.007 0.004 0.011 0.007 80 0.002 0.009 0.008 0.008 150 0.000 0.009 0.009 0.014 10 8-12 0.001 0.008 0.009 0.013 35 0.005 0.005 0.015 0.004 80 0.005 0.011 0.002 0.007 150 0.002 0.013 0.014 0.011 10 0-12 -0.001 0.004 -0.004 0.008 35 -0.007 0.005 0.000 0.005 80 -0.004 0.008 -0.010 0.015 150 -0.001 0.007 0.002 0.007

PAGE 200

181 lower RGRs than the December to April or April to August time periods. The calculated annual RGRs were also si gnificantly lower than the 4-8 and 8-12 mo. rates but significantly greater than the in itial 0-4 month growth rates. Total Carbon and Nitrogen Concentrations Throughout the entire study, live Typha tissue fr om cell 10 had a significantly greater total C (TC) content than the plants growing in the cont rol cell (Table 6-4). It was also true of the standing dead collected in the winter sampling. Because the live tissue in the alum-treated cell had a significantly greater TC concentration even at time 0, before alum application began, it can be attributed to an inter-cell difference rather than an effect of alum application. In both cells live tissue TC content was signi ficantly lower during the winter than during the 8 and 12 mo. samp lings while there were very few spatial differences throughout the cells. Unlike TC, there were no significant differe nces in total N (TN) concentrations prior to experiment initiation (Table 64). However, TN concentrations were significantly higher in live a nd dead tissue harvested from the control cell during the 4 mo. winter and 12 mo. summer samplings while there were no significant differences in live or dead tissue TN content during the 8 mo. spring sampling when nutrients were translocated for flowering (Davis, 1984). Malecki-Brown and White (2007b) found alum was not only efficient at bindi ng P but also significantly decreased the total dissolved N and total kjeldahl N concentrations within the treated water column. Therefore, alum application may have resulted in a lack of bioavailable N in the soil for the emergents to utilize (Madsen and Cedergreen, 2002). Additionally, in both cells live tissue TN concentrations were greater in the winter than in the 8 or 12 mo. samplings while the

PAGE 201

182Table 6-4. Carbon, nitrogen, and phosphorus tissue concentrations in Typha spp. ha rvested from cell 9 and 10 of the Orlando Ea sterly Wetland. Values are means 1 standard deviation (n=6). Tissue Parameter Distance from Alum Control status inflow (m) 0 mo 4 mo 8 mo 12 mo 0 mo 4 mo 8 mo 12 mo Live Total C 0-10 435 10.4 415 4.21 424 15.3 444 5.61 427 1.06 418 20.1 430 15.2 435 12.9 mg g-1 DW 10-35 436 4.38 419 5.58 447 23.7 448 8.46 436 6.46 411 12.2 431 13.5 449 5.36 35-80 431 2.93 423 9.03 457 49.1 444 4.91 414 3.42 391 4.50 413 16.6 418 7.83 80-150 429 3.57 415 10.1 436 2.20 440 7.05 423 1.54 404 15.2 406 20.0 423 8.48 Total N 0-10 7.99 3.23 11.3 2.04 10.2 1.50 7.46 1.16 9.03 1.41 13.1 3.92 8.30 1.09 6.89 1.90 mg g-1 DW 10-35 7.72 0.93 9.10 2.40 9.10 1.36 6.71 1.06 7.27 0.91 15.6 3.44 9.35 0.88 7.46 0.81 35-80 8.30 0.29 8.20 1.23 9.60 1.45 6.62 1.23 9.41 1.92 17.2 2.17 9.92 1.90 11.2 2.26 80-150 6.80 1.5410.1 0.82 9.12 1.02 6.79 1.42 9.30 0.32 14.0 4.73 11.9 2.85 10.5 1.77 Total P 0-10 11.4 3.11 16.9 2.62 12.8 1.04 11.6 2.23 13.8 1.74 17.6 2.77 14.8 2.73 14.2 5.23 mg g-1 DW 10-35 12.0 0.42 14.1 2.59 13.4 1.23 12.1 3.68 13.5 1.99 21.7 4.86 15.4 2.16 13.7 3.04 35-80 12.6 1.62 12.2 2.29 14.3 2.07 10.7 3.27 15.6 2.37 23.6 4.63 16.9 3.12 17.5 4.00 80-150 13.9 8.03 18.1 3.92 15.1 2.60 12.0 3.65 12.1 7.29 22.6 4.90 15.2 4.26 18.5 4.10 Dead Total C 0-10 na 437 17.1 416 10.8 445 9.11 na 432 21.0 435 12.2 447 25.8 mg g-1 DW 10-35 na 432 17.3 440 4.60 450 4.62 na 414 39.5 426 23.5 432 10.3 35-80 na 443 7.42 427 18.1 432 10.3 na 399 16.2 419 18.9 432 10.3 80-150 na 425 18.6 440 13.5 446 23.9 na 404 26.7 422 7.15 438 12.6 Total N 0-10 na 0.47 0.42 6.00 0.68 4.74 0.78 na 0.66 0.74 5.21 0.74 4.00 0.31 mg g-1 DW 10-35 na 0.23 0.17 6.33 1.69 4.07 0.93 na 0.36 0.32 5.95 0.79 7.13 2.29 35-80 na 0.39 0.39 4.76 0.40 7.13 2.29 na 0.90 0.74 6.61 0.95 7.13 2.29 80-150 na 0.40 0.30 5.13 0.71 4.49 0.90 na 0.27 0.35 5.85 0.90 6.21 0.89 Total P 0-10 na 11.6 6.78 6.01 1.00 5.28 0.85 na 9.35 5.05 5.64 2.38 7.50 3.22 mg g-1 DW 10-35 na 6.02 3.42 7.19 2.68 5.49 2.10 na 11.4 4.75 9.32 4.22 7.23 2.76 35-80 na 6.20 2.54 5.30 1.89 13.5 3.08 na 18.3 3.85 9.90 4.74 13.5 3.08 80-150 na 6.43 1.34 6.51 1.94 6.18 3.19 na 13.9 4.15 6.02 1.21 10.7 1.34 Dead and live tissue were not separated for analysis.

PAGE 202

183 opposite was true of the TC cont ent of the tissue, w ith lowest concentrations occurring in the winter suggesting an inve rse relationship. The same re lationship was found in the TC and TN of the dead tissue with TC concentrations significantly gr eater during the 12 mo. sampling than the 8 mo., with the op posite true of TN concentrations. Overall, there were no significant differen ces in TC between live and dead tissue while TN concentrations were twice as high in live tissue as compared to standing dead suggesting either leaching or tr anslocation of N. There were no significant differences in TC:TN at the start of the study (Table 6-5) . However, during the 4 and 8 mo. samplings TC:TN ratios were significantly greater in the live tissue from the alum-treated cell than the control, as well as in th e dead tissue at 4 and 12 months . This suggests a reduction in N availability to the Typha growing in the alum-treated cel l, corresponding to the low TN tissue concentrations (Sterner and Elser, 2002). Both the nitrogen RE and NUE were calcu lated to get a bette r understanding of the N cycling dynamics within the cells. Ther e were no significant differences in the RE between cells (Table 6-6). Nitrogen RE av eraged 77.0% 32.4 in the control cell and 81.0% 31.1 in the alum-treated cell over the c ourse of the study. Resorpton efficiencies were significantly greater during the spring 8 mo. sampling than during the winter (4 mo.) or summer (12 mo.). Maximum N c onservation during the spring corresponds to maximum growth of Typha spp. typical in the spring (Davis, 1984; Emery and Perry, 1995). The NUE was significantly greater in the Typha from the alum-treated cell than the control cell. The increasingly higher NUE is indicative of N-poor wetland, producing large amounts of biomass with little loss of N in the standing dead and litterfall

PAGE 203

184Table 6-5. Total phosphorus, car bon, and nitrogen ratios in Typha spp. harvested from cell 9 and 10 of the Orlando Easterly Wetland. Values are means 1 standard deviation (n=6). Tissue Parameter Distance from Alum Control type inflow (m) 0 mo 4 mo 8 mo 12 mo 0 mo 4 mo 8 mo 12 mo Live TC:TN 0-10 59.6 25.4 39.5 6.43 47.2 5.03 177 95.6 48.2 3.87 33.3 12.4 49.7 6.96 139 95.9 10-35 56.8 6.26 47.4 8.45 47.1 5.37 228 118 60.5 5.30 24.6 6.97 46.6 6.40 192 126 35-80 52.9 0.65 50.3 11.5 45.3 3.93 200 171 44.9 3.85 24.5 5.35 37.9 9.71 152 95.8 80-150 65.8 3.46 50.9 22.3 57.0 15.3 147 86.3 45.5 25.6 33.6 9.61 39.1 7.46 83.5 26.4 TC:TP 0-10 39.6 11.7 25.1 3.93 33.4 2.45 39.6 7.93 31.1 3.87 24.4 4.87 29.8 6.18 35.9 15.7 10-35 36.3 0.91 30.0 3.77 33.5 3.90 40.1 12.3 32.7 5.30 19.8 4.60 28.2 4.87 33.8 9.67 35-80 34.6 4.68 35.9 9.10 32.2 2.57 44.8 14.2 26.8 3.85 17.1 2.91 25.3 5.96 25.1 6.84 80-150 37.4 22.1 23.8 4.63 29.4 4.54 40.3 12.7 42.7 25.6 18.7 4.52 29.1 9.47 23.5 5.69 TN:TP 0-10 0.69 0.10 0.68 0.18 0.71 0.10 0.31 0.23 0.65 0.09 0.79 0.26 0.60 0.11 0.28 0.07 10-35 0.64 0.05 0.64 0.14 0.72 0.11 0.21 0.09 0.54 0.01 0.83 0.23 0.62 0.15 0.22 0.11 35-80 0.65 0.10 0.76 0.34 0.71 0.06 0.20 0.06 0.60 0.03 0.77 0.19 0.70 0.21 0.19 0.09 80-150 0.54 0.20 0.50 0.11 0.54 0.13 0.23 0.13 0.95 0.60 0.57 0.08 0.74 0.22 0.33 0.13 Dead TC:TN 0-10 na 92.5 18.3 75.9 10.6 157 95.5 na 73.8 26.5 81.5 14.8 141 94.2 10-35 na 97.9 12.5 73.9 16.7 211 97.8 na 65.4 33.5 73.1 7.05 130 46.4 35-80 na 105 20.0 91.4 13.6 233 177 na 41.6 11.1 66.3 12.8 210 141.1 80-150 na 87.8 10.2 71.2 25.6 126 36.2 na 68.2 24.1 72.5 11.0 90.8 32.9 TC:TP 0-10 na 44.7 15.5 71.9 11.8 39.6 7.93 na 55.0 21.5 89.5 48.3 35.9 15.7 10-35 na 63.6 15.6 66.6 18.6 40.1 12.3 na 21.9 21.7 55.8 25.3 33.8 9.67 35-80 na 78.6 22.4 90.9 35.0 44.8 14.2 na 22.6 5.05 48.8 25.7 25.1 6.84 80-150 na 68.5 15.0 74.8 26.5 40.3 12.7 na 31.6 10.2 71.4 14.4 23.5 5.69 TN:TP 0-10 na 0.48 0.21 0.97 0.22 0.31 0.23 na 0.75 0.16 1.19 0.58 0.28 0.07 10-35 na 0.67 0.17 0.95 0.35 0.21 0.09 na 0.34 0.29 0.78 0.39 0.22 0.11 35-80 na 0.72 0.26 1.01 0.47 0.2 0.06 na 0.58 0.11 0.81 0.51 0.19 0.09 80-150 na 0.79 0.18 1.29 0.90 0.23 0.13 na 0.62 0.21 0.97 0.32 0.33 0.13 Dead and live tissue were not separated for analysis. TC = total carbon; TN = tota l nitrogen; TP = total phosphorus.

PAGE 204

185Table 6-6. Nitrogen and phosphorus resorption efficiency and nutrient use efficiency of Typha spp. harvested from cell 9 and 10 of the Orlando Easterly Wetland. Values are means 1 standard deviation (n=6). Parameter Distance from Alum Control inflow (m) 4 mo 8 mo 12 mo 4 mo 8 mo 12 mo N RE 0-10 81.8 18.6 95.9 3.18 88.0 9.48 87.5 11.4 94.1 8.55 80.3 13.3 % 10-35 89.5 17.8 94.0 6.41 72.4 28.1 79.9 0.34 93.2 4.74 76.9 11.2 35-80 74.0 25.8 94.0 4.68 70.8 25.8 65.6 35.7 94.0 7.31 81.5 28.0 80-150 92.5 4.57 86.4 15.5 85.7 14.7 85.4 21.3 91.5 11.8 96.6 5.40 N NUE 0-10 207 29.8 185 20.7 230 25.1 188 63.6 152 31.6 251 20.2 ratio 10-35 218 31.1 179 21.8 247 46.9 157 62.6 168 7.25 243 30.2 35-80 255 31.8 168 37.2 247 46.4 98.0 10.7 172 24.6 151 41.1 80-150 199 18.4 218 24.2 230 44.4 132 36.3 150 27.0 164 25.3 P RE 0-10 78.6 32.4 96.5 2.18 90.6 7.89 83.1 11.9 95.3 5.39 77.2 25.5 % 10-35 68.0 34.0 94.2 7.49 86.7 17.2 90.3 13.4 94.3 7.15 78.4 13.7 35-80 63.0 34.6 96.3 3.86 65.8 20.4 69.2 46.6 96.5 6.03 85.5 27.2 80-150 88.9 8.20 86 19.3 90.5 7.88 93.3 8.20 93.6 7.04 96.9 5.45 P NUE 0-10 103 49.9 174 25.2 194 37.6 111 54.7 169 32.7 153 58.0 ratio 10-35 147 39.7 154 56.7 222 83.1 98.0 37.7 180 127 168 100 35-80 171 44.6 214 90.5 154 80.4 57.0 14.4 114 76.9 16.7 80-150 162 36.3 163 54.6 215 130 77.9 23.6 171 31.3 94.7 13.1 RE = resorption efficiency; NUE = nutrient use efficiency. n=2.

PAGE 205

186 (Vitousek, 1982). This explains the high TC:TN ratios which correspond to the high rate of nutrient resorption occurring in the alum-treated cell. Nitrogen use efficiency was significantly greater in the alum-treated cel l during the winter (4 mo.) and summer (12 mo.) than during the spring when Typha generally flowers (Davis, 1984). In the control cell NUE was significantly greater at 12 months than at 4 or 8 months. Total Phosphorus Concentrations Similar to biom ass, during all sampling events living tissue TP was significantly greater (nearly double) the standi ng dead TP concentrations in both cells. There were no significant differences in TP tissue concentrations between cells at the start of the experiment averaging 13,120 3200 mg P kg-1 throughout both cells (Table 6-4), directly correlated (p<0.01) to TN tissue concentra tions. Interestingly, there was one flower stalk / spike removed from a plant harvested from cell 10 that was also analyzed for TP. The flower contained nearly four times (50,500 mg P kg-1) as much P as the average shoot material, evidence of the i nordinate amount of P required by Typha biomass, particularly when flowering. In the 4 mo. winter sampling, TP concentr ations of both live and dead plant tissue in the control cell were significantly greater than in the alum-treated cell suggesting alum may have impacted Typha, possibly translocating nutrients to the rhizomes more rapidly than would typically occur due to nutrient limitation (Aerts et al., 1999), reflecting the scarcity of bioavailable P due to the P seque stration of the alum floc in the surface soil (Malecki-Brown et al., 2007b). There was a direct correlation (p<0.01) between the crystalline Al concentration in the surface 0-5 cm soil layer and both live and dead tissue TP concentrations during this sampling.

PAGE 206

187 However, by the 8 mo. spring sampling there were once again no significant differences in live or dead TP concentrati ons among plants similar to the start of the experiment (Table 6-4). Plants growing in the control cell av eraged 15,190 3230 mg P kg-1 and plants in the alum-treated cell av eraged slightly less at 13,870 1890 mg P kg-1 (Figure 6-12). Dead tissue TP concentrations along the western tran sect in cell 10 clearly contained less P than the remainder of th e cell (Figure 6-13), perhaps indicating a preferential path and impact of alum (Mal ecki-Brown et al., 2007b). In addition, similar to the time 0 sampling, there was one flower stalk/spike removed fr om a plant harvested in cell 9. This flower containe d nearly five times (71,110 mg P kg-1) as much TP as the average shoot material. The 12 mo. summer sampling TP concentratio ns of both live and dead plant tissue in the control cell were significantly greater th an in the alum-treated cell similar to the winter TP findings (Table 6-4) as well as the same temporal trend occurring in TN concentrations. The variable temporal result s in tissue TN and TP concentrations warrant further investigation, possibly for multiple ye ars at shorter time intervals to discern whether differences can be attrib uted to alum or are due to other factors associated with inter-cell variability. During both the 4 and 12 mo. samplings live tissue TP was directly correlated (p<0.05) to dead tissu e TP in the control cell while there was no such relation in plants within the alum-treated cell at the 4 mo. sampling and during the 12 mo. sampling there was an inverse relationship (p<0.01) between the two suggesting a high rate of resorption was necessary as in a P-limited system (Vitousek, 1982). Overall, there were no significant differences in the TN:TP ratios in live Typha between cells throughout the study (Table 6-5). In both cells liv e tissue TN:TP ratios

PAGE 207

188 were significantly lower during the last samplin g than in the previous three, suggesting system nutrient enrichment (Downing, 1997) as well as a trend in N limitation with time rather than P (Guildford and Hecky, 2000). C ontrastingly, dead tissue in the alum-treated cell, had significantly greater TN:TP ratios th an dead tissue in the control cell during the last two samplings implying increased P limitation over time. Typha plants may have been translocating P from standing dead to be instead utilized by the live tissue. The TC:TP ratios were also significantly greater in the alumtreated cell during all samplings in both live and dead tissue corresponding to the significantly higher C content of plants growing in the alum-treated cell as well as a de crease in P availability (Sterner and Elser, 2002). Similar to the differences between cells, intra-cell spatial di fferences were not significant at the start or 8 mo. spring sampling in either live or dead tissue. During the winter, 4 mo. sampling live tissue TP concentr ations were significantly greater (p<0.04) in the 35-80 m plot than in the last plot of the alum treated cell while dead tissue TP concentrations in the control 35-80 m plot we re significantly greater than those in the 010 m plot (Table 6-3). Additionally during the final summer sampling, live tissue TP in the control 35-80 m plot was significan tly greater than in the 10-35 m plot. Both the phosphorus RE and NUE were calculated to better understand P cycling dynamics within the cells. Due to the variability in TP and biomass concentrations, there were no significant differences in RE or NUE between cells (Table 6-6). Phosphorus RE averaged 84-95% in the cont rol cell and 77-94% in the alum -treated cell. The RE of Typha was significantly greater (p<0.005) during the 8 mo. spring sampling than during the 12 mo. summer sampling corresponding to the maximum growth of Typha spp.

PAGE 208

189 typical in the spring and sene scence in late July through A ugust (Davis, 1984; Emery and Perry, 1995). The average P NUE was generally greater in the alum-treated cell averaging 149206 compared to the control which averaged an NUE of 87-155 throughout the study. A higher NUE would be expected in the alum-treated cell because P bound to the alum floc is not bioavailable to the Typha, thus they are restricted to subsurface P availability. Both the P RE and NUE may increase in the alum-treated Typha over time as the floc layer moves downward through the soil profile, due to accretion of partic ulates, into direct contact with the root zone. Metal Concentrations There were no significant diffe rences in total Al, Ca, Fe, K, or Mg concentrations between live Typha tissue from the alum-treated and control cell, during the initial or final summer samplings (Table 6-7). This indicates that while N and P limitations were present in Typha growing in the alum-treated cell, they were not associated with Al toxicity or micronutrient defici encies. There were al so no spatial differe nces in nutrient concentrations within the cells. Calcium and K were the dominant me tals present in both live and dead tissue, significantly greater than concentrations of Al, Fe, and Mg (Table 67, Table 6-8). During the final sampling, standing dead Typha tissue in the alum-treated cell had significantly greater Al concentrations than ti ssue from the control cell (Table 6-8). In addition, Ca concentrations were significantly lower in standing dead of the alum-treated cell. The ability of Al to reduce Ca uptake and transloca tion in plants has been well documented (Huang et al., 1992; Rengel, 1992). These results imply that leaves may have been dying as an internal mechanism of detoxifying the high Al concentration

PAGE 209

190 Table 6-7. Live tissue Typha spp. metal concentrations, on a dry weight basis, in cell 9 and 10 of the Orlando Easterly Wetland. Parameter Distance from Alum Control inflow (m) 0 mo 12 mo 0 mo 12 mo Total Al 0-10 721.9 196 103 940.4 336 92.9 mg kg-1 10-35 596.1 342 129 729.4 343 171 35-80 389.4 335 147 557.9 256 198 80-150 1321 314 136 429.8 187 93.1 Total Ca* 0-10 96200 53499 14821 65409 47733 16596 mg kg-1 10-35 75342 43468 12312 80387 44813 8819 35-80 84188 47980 26633 102564 53230 13140 80-150 180906 50932 25758 59394 55996 22766 Total Fe 0-10 2225 246 169 1589 569 512 mg kg-1 10-35 668.6 259 170 969.8 417 80.3 35-80 983.5 568 295 1363 382 279 80-150 1283 454 309 431.0 296 81.1 Total K* 0-10 68359 91527 27074 89621 99504 32140 mg kg-1 10-35 58224 105808 31074 90274 89086 24956 35-80 71266 95181 27296 77547 87451 26227 80-150 98219 75001 43218 32346 86535 26129 Total Mg 0-10 5736 6834 1696 7620 5329 3666 mg kg-1 10-35 6073 6532 2764 8466 7031 1742 35-80 9270 7082 1766 10657 8560 5427 80-150 9532 7135 1816 4979 7790 2224 Dead and live tissue were not separated for analysis; n=1. * Significantly greater than Al, Fe, and Mg concentrations at the 0.05 probability level.

PAGE 210

191 Table 6-8. Dead tissue Typha spp. metal concentrations, on a dry weight basis, in cell 9 and 10 of the Orlando Easterly Wetland during the 12 month sampling. Distance from Parameter inflow (m) Alum Control Total Al* 0-10 1018 785.5 494 282 mg kg-1 10-35 2495 3162 484 168 35-80 2905 1721 400 254 80-150 2954 1761 454 74.2 Total Ca* 0-10 67875 49393 135577 101933 mg kg-1 10-35 50022 19317 88843 56659 35-80 31353 9273 107340 36994 80-150 23671 8777 107717 42348 Total Fe 0-10 595.7 308.3 3056 2259 mg kg-1 10-35 2127 2275 2090 1205 35-80 1160 654.6 2132 2411 80-150 3917 1783 3225 3480 Total K 0-10 31840 39000 45721 34986 mg kg-1 10-35 25123 24173 31177 22499 35-80 30736 23382 44200 26884 80-150 31713 27373 26421 31792 Total Mg 0-10 9328 3241 10935 3350 mg kg-1 10-35 8603 1623 7786 4163 35-80 13419 9095 14742 6871 80-150 10354 5052 11125 1793 * Alum tissue concentration significantly different from control at the 0.05 probability level. within the plants (Delhaize and Ryan, 1995). Malecki-Brown and White (2007b) found similar results in SAV tissue growing in alum-treated mesocosms. Conclusions The long-term application of low-dosage al um in a m unicipal wastewater treatment wetland did not appear to impact the water column pH, which remained circumneutral throughout the year. Alum application did not improve DOC removal despite being used in lakes for this purpose. Highly productiv e wetland systems are often found to be net sources of DOC, perhaps reducing the overall pe rformance of alum. Regardless, alum application was extremely effective in incr easing both SRP and TDP removal efficiencies resulting in improved TP removal as well, on average sequestering over 25% more P than

PAGE 211

192 the control cell over the course of a year. However, alum application did substantially increase PP concentrations, particularly in areas near the inflowing Al(OH)3 floc which readily adsorbed bioavailable SRP transforming it into particulate form. The increased PP concentrations in the water column may have been due to a rapid flow velocity at the area of alum inflow or reduced residence tim e within the wetland, either of which could prevent settling of the flocculent particulat e material out of the water column. This finding may be critical in the effective management of treatment wetlands with permitted PP or TSS discharges. Alum should be introduced in areas of low flow velocity, at an adequate distance from the discha rge point to allow for settling. Results were variable when looking at the impact of alum application on the emergent macrophyte Typha spp. While Typha growing in the alum-treated cell did produce a greater number of shoots per square meter, their development seemed impaired, producing fewer leaves than their control counterparts as well as containing significantly lower tissue TP during two of the four samplings. In addition, standing dead tissue from the alum-treated cell during the la st sampling had symptoms associated with Al toxicity. However, there were no significant differences in the biomass, RGR, RE, or NUE between cells suggesting alum appli cation in emergent-dominated treatment wetlands did not impact plant nutrient cycli ng overall. I recommend further investigation into the possible impact the Al(OH)3 floc layer may have on the plants as it moves downward through the soil profile into the root zone where the majority of metals and nutrients are obtained.

PAGE 212

193 CHAPTER 7 EFFECTS OF ALUM ON A MUNICIPAL WASTEWATER TREATMENT WETLAND: A SUMMARY The effectiveness of alum in immo bilizing phosphorus (P) from m unicipal wastewater treatment wetlands and it s impacts on macrophytes, microbes, soil characteristics, and water quality were inves tigated using short-term core and mesocosm studies as well as a long-t erm field study (Table 7-1). The anaerobic core study investigated not only the use of alum at various dosage ra tes but also three aluminum (Al) containing alternatives: polya luminum chloride (PAC), al um residual, and partiallyneutralized aluminum sulfat e (PNAS). There is no previous documentation in the literature regarding the usage of any of these chemical amendments for P sequestration in treatment wetlands. Nor has there ever been direct measurement of the impact of alum application on the biomass or activity of the microbi al community in aquatic environments. Alum, as well as the three tested alternatives to alum were all effective at short term P sequestration. However, applications of alum, PAC, and PNAS resulted in decreased water column pH values and high soluble Al concentrations. Ther e was also a direct relationship between the dose of Al used a nd resulting impact on the microbial biomass and activity within the soil. Mesocosm results also indicated a significant decrease in microbial biomass and activity attributed to in creased soil acidity and Al concentrations in the alum-treated mesocosms (Table 7-1). Therefore, the minimum dosage necessary

PAGE 213

194 for effective treatment should be used by wetland managers. Additionally, it was determined that the efficacy of a one-time alum treatment may be short lived in treatment wetlands due to continuous P loading. Inst ead, a continuous low-dosage alum injection system would be a more effective management strategy. Both the mesocosm and field study demons trated that a continuous low-dosage application of alum proximal to the outfl ow regions of the wetland can provide an effective management tool to maintain discharge concentrations within permitted values, particularly during inefficient winter treatmen t times, when plants senesce and microbial activity slows. The ability of alum to bind not only P but organic molecules as well resulted in improved soluble react ive P, total dissolved P, total P, total dissolved N, total kjeldahl N, and dissolved orga nic C wetland treatment efficien cy. However, in both the mesocosm and field study there was an overall increase in particulate P, particularly in areas near the inflowing Al(OH)3 floc which readily adsorbed bioavailable SRP transforming it into a particulate form. Th e resulting export of PP from alum-treated wetlands may be extremely important in treatment wetlands with permitted PP or total suspended solid discharges. Therefore, it is recommended that alum be introduced in areas of low flow velocity an adequate distance from discharge points to allow for settling of particulates. Aquatic macrophytes are critical to the pr ocess of nutrient removal and cycling in constructed wastewater treatment wetlands. The mesocosm and field study demonstrated that the uptake and retention of bioavailabl e metals in aquatic macrophytes are largely dependent on life form. Submerged aquatic vegetation (SAV) absorb a large portion of soluble nutrients and metals directly from the water column and were therefore

PAGE 214

195 Table 7-1. Summary of vari ables impacted by alum app lication in a constructed wastewater treatment wetland duri ng a tri-scale study from 2004-2006. Variable Unit Core Study (Ch2) Mesocosm Study (Ch3-4) Field Study (Ch 5-6) Soil oxalate Al mg kg-1 + + + Soil pH pH units nsd nsd Soil total P mg kg-1 nsd nsd + Microbial biomass mg kg-1 nsd -, nsd, + Microbial activity mg kg-1 hr-1 (d-1) nsd , nsd Water column Al mg L-1 + nsd nsd Water column DOC mg L-1 na Water column pH pH units nsd Water column SRP mg L-1 Water column total P mg L-1 na Water column particulate P mg L-1 na + + Live plant biomass g m-2 na SAV = EAV = nsd nsd Live plant Al mg kg-1 na SAV = + EAV = nsd nsd Live plant Ca mg kg-1 na SAV = EAV = nsd nsd Dead plant biomass g m-2 na na nsd Dead plant Al mg kg-1 na na + Dead plant Ca mg kg-1 na na = Alum significantly decreased variable compared to control. + = Alum significantly increased variable compared to control. na = Variable not measured during study. nsd = No significant difference between alum and control. susceptible to Al toxicity as a result of alum application. Emergent plants such as Scirpus californicus and Typha spp. tend to obtain most metals and nutrients from the subsurface soil, and therefore remained relatively unaffected by alum application. Further investigation is needed howev er to determine the possible impact of the Al hydroxide floc layer on the emergent macrophytes as it migrates downward through the soil profile into the root zone, where the majority of metals and nutrients are obtained. The effectiveness of alum in P sequestration and relative ease of application make it a viable management alternative in agi ng constructed wetlands. However there are many ecological impacts associated with alum application that may completely alter the P

PAGE 215

196 nutrient dynamics of a wetland system. The impacts shown in my research include a reduction in the microbial pool and activ ity which can in turn slow phosphorus mineralization in the short-term. The reduc tion in bioavailable inorganic P bound in the Al floc, and reduced mineralization will result in nutrient limitations to both wetland microbes and macrophytes. If diurnal pH cha nges result in a wate r column pH greater than 8 or less than 4, SAV biomass may signifi cantly decline due to Al toxicity. This can not only result in a decreased nutrient remova l efficiency of a treatment wetland due to less biomass for nutrient uptake, but it can also result in th e re-release of a substantial portion of P stored in the SAV detrital materi al. Both the mesocosm and field study also implicated potential long-term impacts of alum on emergent macrophytes. These may include a reduction in shoot emer gence as was demonstrated by Scirpus , or growth limitations such as fewer leaves produced as was the case in Typha growing in the alumtreated field cell. It is cr itical that treatment wetland managers understand all of the shortand possibly long-term im plications of using alum in treatment wetland systems.

PAGE 216

197 APPENDIX A CORE STUDY RAW DATA Lake Alum Application Summary Table A-1. Literature review of previous alum applications to aquatic eco systems. Study Area Alum Dosage Reference Amisk Lake,Baptise Lake, and Crooked Lake, Canada 13 g Al4+ m-2 Burley et al., 2001 Annabessacook Lake, Maine 25 g Al m-2 Connor and Smith, 1986 Bass Bay, Wisconsin 90 g Al m-2 Rydin and Welch, 1999 Braidwood Lagoon, Australia 1 lb 1000 gal-1 = 1 kg 10 kl-1 = 1067 kg (lake volume 10910 kl) May, 1974 Bullhead Lake, Wisconsin 13 mg Al3+ L-1 hypolimnetic water = 32,500 L or 42,500 kg Narf, 1985 Campbell Lake, Washington 10.9 g m-2 Rydin et al., 2000 Courtille Lake, France 1.5 mg Al3+ L-1 = 2.65 g m-2 Hullebusch et al., 2003 Cuyahoga River, Ohio 17.0 m3 alum d-1 3.0 m3 alum d-1 Barbiero et al., 1988 Eau Galle Reservoir, Wisconsin 11.3 g Al m-2 = 4.5 mg Al L-1 = 2653 kg Barko et al., 1990 Green Lake, Washington 8.6 mg Al L-1 Jacoby et al., 1994 Horseshoe Lake, Wisconsin 18 mg Al L-1 (200 mg L-1 alum) Peterson et al, 1973 Lake Apopka Basin, Florida 3.7 23 g kg-1 soil (1 g kg-1 = 56 g m-2) Ann et al., 2000 Lake Ballinger, Washington 5 g m-2 Rydin et al., 2000 Lake Delavan, Wisconsin 12 g Al m-2 Rydin and Welch, 1999 Lake Erie, Washington 10.9 g m-2 Rydin et al., 2000 Lake Hollingsworth, Florida 85,500 gal alum (355 acre lake) Harper and Medley, 2004 Lake Leba, Nebraska 10 mg L-1 (34,065 L) Holz and Hoagland, 1999 Lake Morey, Vermont 44 g Al m-2 Smeltzer, 1990 Lake S usser See, Germany 2 mg Al L-1 yr-1 Lewandowski et al., 2003 Liberty Lake, Washington 10 mg L-1 Bulson et al., 1984 Long Lake, Wisconsin 14 mg Al3+ L-1 Narf, 1990 Medical Lake, Washington 12.2 g m-2 Rydin et al., 2000 Mirror Lake, Wisconsin 6.6 mg Al L-1 Garrison and Knauer, 1984 Newman Lake, Washington 31 mg L-1 (2.8 mg L-1 as Al) Doke et al., 1995 Phantom Lake, Washington 4.2 g m-2 Rydin et al., 2000 Pickerel Lake, Wisconsin 7 mg Al3+ L-1 = 61,000 L Narf, 1990 Shadow Lake, Wisconsin 5.7 mg Al L-1 Garrison and Knauer, 1984 Snake Lake, Wisconsin 12 mg Al3+ L-1 Narf, 1990 West Twin Lake, Ohio 26 mg Al L-1 Cook et al., 1982 Wind Lake, Wisconsin 80 g Al m-2 20 g Al m-2 Rydin and Welch, 1999 Winnipeg Reservoir, Canada 5 mg alum L-1 Klassen et al., 1970

PAGE 217

198 Water Column Analyses Table A-2. Raw soluble reac tive phosphorus data for anaerobic core incubation study. Treatment & Dosage Core # T0 T1 T3 T5 T7 T10 T14 alum 36 1 0.205 < det limit 0.003 < det limit 0.007 0.006 0.007 alum 36 2 0.233 < det limit 0.004 < det limit 0.002 0.006 0.014 alum 36 3 0.214 < det limit 0.003 < det limit 0.003 0.006 0.007 alum 36 4 0.325 0.0013 0.001 < det limit 0.003 0.007 0.007 alum 36 5 0.339 < det limit 0.001 < det limit 0.007 0.007 0.007 alum 36 6 0.219 < det limit 0.002 0.0016 0.006 0.007 0.007 alum 18 7 0.208 0.001 0.002 0.005 0.003 0.007 0.004 alum 18 8 0.188 0.001 0.001 0.003 0.002 0.006 0.001 alum 18 9 0.372 0.001 0.002 0.002 0.002 0.006 0.002 alum 18 10 0.141 0.000 0.001 0.002 0.002 0.006 0.003 alum 18 11 0.191 0.000 0.001 0.001 0.001 0.005 0.002 alum 18 12 0.116 0.001 0.001 0.001 0.001 0.007 0.008 alum 9 13 0.229 0.003 0.007 0.011 0.006 0.022 0.008 alum 9 14 0.235 0.003 0.003 0.005 0.005 0.018 0.007 alum 9 15 0.138 0.002 0.004 0.004 0.004 0.015 0.005 alum 9 16 0.252 0.002 0.004 0.004 0.005 0.017 0.009 alum 9 17 0.132 0.002 0.004 0.003 0.003 0.010 0.005 alum 9 18 0.250 0.001 0.003 0.004 0.007 0.011 0.010 PAC 36 19 0.157 0.001 0.006 < det limit 0.006 0.006 0.001 PAC 36 20 0.206 0.007 0.002 < det limit < det limit 0.005 < det limit PAC 36 21 0.296 0.003 0.005 0.0584 0.007 0.006 0.006 PAC 36 22 0.234 0.021 0.002 0.0004 0.006 0.008 0.006 PAC 36 23 0.133 0.002 0.000 0.0042 0.024 0.018 0.014 PAC 36 24 0.218 0.007 0.010 0.0560 0.007 0.007 0.008 PAC 18 25 0.323 0.000 0.002 0.002 0.003 0.010 0.018 PAC 18 26 0.200 0.001 0.003 0.004 0.003 0.010 0.005 PAC 18 27 0.210 0.000 0.003 0.001 0.006 0.013 0.006 PAC 18 28 0.265 0.001 0.004 0.002 0.004 0.009 0.004 PAC 18 29 0.134 0.002 0.005 0.007 0.005 0.011 0.007 PAC 18 30 0.203 0.001 0.005 0.005 0.004 0.009 0.004 PAC 9 31 0.231 0.004 0.006 0.004 0.007 0.012 0.007

PAGE 218

199 Table A-2. Continued. Treatment & Dosage Core # T0 T1 T3 T5 T7 T10 T14 PAC 9 32 0.212 0.006 0.011 0.009 0.015 0.028 0.024 PAC 9 33 0.203 0.005 0.006 0.005 0.006 0.020 0.008 PAC 9 34 0.112 0.002 0.006 0.014 0.015 0.048 0.022 PAC 9 35 0.182 0.002 0.017 0.017 0.123 0.130 0.093 PAC 9 36 0.208 0.002 0.009 0.007 0.012 0.030 0.015 alum res 36 37 0.213 0.181 0.138 0.156 0.172 0.147 0.126 alum res 36 38 0.234 0.183 0.142 0.152 0.131 0.153 0.117 alum res 36 39 0.260 0.231 0.198 0.201 0.204 0.177 0.089 alum res 36 40 0.132 0.124 0.119 0.138 0.153 0.169 0.145 alum res 36 41 0.236 0.227 0.197 0.196 0.170 0.161 0.130 alum res 36 42 0.216 0.165 0.135 0.120 0.121 0.105 0.081 alum res 18 43 0.219 0.212 0.225 0.172 0.136 0.318 0.197 alum res 18 44 0.160 0.131 0.122 0.089 0.069 0.068 0.025 alum res 18 45 0.212 0.191 0.179 0.114 0.112 0.158 0.107 alum res 18 46 0.326 0.298 0.305 0.214 0.202 0.270 0.164 alum res 18 47 0.154 0.173 0.165 0.143 0.088 0.046 0.128 alum res 18 48 0.266 0.245 0.264 0.236 0.196 0.220 0.166 alum res 9 49 0.222 0.177 0.243 0.219 0.196 0.321 0.256 alum res 9 50 0.206 0.200 0.198 0.202 0.119 0.190 0.112 alum res 9 51 0.220 0.228 0.213 0.152 0.185 0.236 0.262 alum res 9 52 0.220 0.212 0.233 0.223 0.210 0.262 0.219 alum res 9 53 0.113 0.121 0.132 0.107 0.117 0.163 0.142 alum res 9 54 0.109 0.076 0.122 0.086 0.102 0.144 0.137 PNAS 36 55 0.211 < det limit 0.004 0.006 0.004 0.007 0.006 PNAS 36 56 0.202 < det limit 0.002 0.005 0.004 0.013 0.015 PNAS 36 57 0.226 < det limit < det limit 0.004 0.003 0.006 0.010 PNAS 36 58 0.210 < det limit < det limit 0.004 0.003 0.007 0.006 PNAS 36 59 0.219 < det limit < det limit 0.004 0.002 0.006 0.007 PNAS 36 60 0.199 < det limit < det limit 0.003 0.002 0.005 0.008 PNAS 18 61 0.196 0.009 0.006 0.005 0.006 0.010 0.002 PNAS 18 62 0.311 0.094 0.002 0.003 0.004 0.009 0.002 PNAS 18 63 0.120 0.003 0.001 0.002 0.003 0.010 0.003 PNAS 18 64 0.219 0.001 0.001 0.001 0.002 0.006 0.001 PNAS 18 65 0.211 0.001 0.004 0.001 0.002 0.007 0.002 PNAS 18 66 0.141 0.000 0.000 0.001 0.002 0.005 0.001 PNAS 9 67 0.155 0.004 0.004 0.007 0.006 0.013 0.003 PNAS 9 68 0.229 0.003 0.003 0.002 0.005 0.012 0.005 PNAS 9 69 0.214 0.002 0.003 0.016 0.005 0.009 0.008 PNAS 9 70 0.163 0.003 0.001 0.004 0.005 0.010 0.005

PAGE 219

200 Table A-2. Continued. Treatment & Dosage Core # T0 T1 T3 T5 T7 T10 T14 PNAS 9 71 0.210 0.002 0.001 0.002 0.004 0.008 0.004 PNAS 9 72 0.205 0.003 0.001 0.003 0.003 0.010 0.004 control 73 0.225 0.252 0.274 0.291 0.290 0.299 0.310 control 74 0.186 0.216 0.251 0.274 0.238 0.269 0.252 control 75 0.262 0.284 0.302 0.320 0.330 0.367 0.305 control 76 0.227 0.264 0.285 0.325 0.296 0.354 0.267 control 77 0.215 0.257 0.274 0.261 0.295 0.341 0.350 control 78 0.188 0.219 0.230 0.258 0.247 0.272 0.246 PAC = polyaluminum chloride; alum res = alum residual; PNAS = partially-neutralized aluminum sulfate Dosage = 36, 18, or 9 g Al m-2

PAGE 220

201 Table A-3. Raw soluble reac tive phosphorus data for anaerobic core spiking study. Treatment T0 T1 T2 T4 T7 T0 T1 T2 T4 T7 T0 T1 T2 T4 T7 alum 36 0.0939 0.0170 0.0305 0.0114 0.0125 0.0743 0.0190 0.0055 0.0043 0.0018 0.0830 0.0076 0.0059 0.006 0.009 alum 36 0.0861 0.0484 0.0320 0.0141 0.0096 0.0988 0.0325 0.0209 0.0116 0.0107 0.0638 0.0126 0.0036 0.011 0.011 alum 36 0.0612 0.0116 0.0073 0.0034 0.0036 0.0898 0.0410 0.0037 0.0033 0.0026 0.0521 0.0072 0.004 0.006 0.006 alum 18 0.036 0.005 0.001 0.001 0.0027 0.0751 0.0346 0.0235 0.026 0.019 0.111 0.071 0.064 0.035 0.024 alum 18 0.025 0.003 0.000 0.002 0.0062 0.0673 0.0285 0.0192 0.021 0.012 0.102 0.066 0.051 0.037 0.025 alum 18 0.030 0.013 0.007 0.004 0.005 0.0547 0.0274 0.017 0.024 0.011 0.111 0.04 0.012 0.011 0.011 alum 9 0.096 0.072 0.086 0.064 0.0645 0.1697 0.1446 0.1402 0.129 0.114 0.207 0.211 0.155 0.18 0.165 alum 9 0.157 0.063 0.035 0.037 0.0347 0.1391 0.1051 0.0842 0.082 0.064 0.175 0.138 0.109 0.094 0.057 alum 9 0.077 0.057 0.040 0.037 0.0291 0.1339 0.1107 0.0773 0.086 0.068 0.169 0.146 0.11 0.095 0.069 PAC 36 0.0314 0.0096 0.0057 0.0038 0.0036 0.0733 0.0124 0.0087 0.0060 0.0037 0.0861 0.0464 0.029 0.020 0.019 PAC 36 0.0120 0.0032 0.0032 0.0034 0.0026 0.0338 0.0132 0.0112 0.0124 0.0080 0.0393 0.0675 0.023 0.023 0.011 PAC 36 0.0739 0.0529 0.0443 0.0486 0.0285 0.1391 0.0917 0.0882 0.0896 0.0068 0.1921 0.1613 0.107 0.089 0.063 PAC 18 0.166 0.127 0.119 0.095 0.0563 0.2111 0.1568 0.1307 0.138 0.083 0.171 0.21 0.186 0.167 0.131 PAC 18 0.107 0.042 0.038 0.016 0.0215 0.1477 0.0858 0.0758 0.061 0.13 0.149 0.116 0.089 0.077 0.057 PAC 18 0.144 0.059 0.046 0.043 0.0207 0.1523 0.1021 0.1155 0.076 0.044 0.164 0.141 0.137 0.108 0.089 PAC 9 0.159 0.163 0.116 0.116 0.0982 0.2763 0.2427 0.2561 0.212 0.237 0.282 0.345 0.305 0.288 0.273 PAC 9 0.223 0.242 0.136 0.190 0.1855 0.3386 0.2742 0.2466 0.2 0.208 0.286 0.346 0.296 0.283 0.305 PAC 9 0.111 0.105 0.102 0.112 0.0761 0.219 0.1607 0.1813 0.147 0.151 0.234 0.238 0.203 0.205 0.187 alum res 36 0.234 0.298 0.277 0.298 0.2548 0.4177 0.356 0.3496 0.327 0.343 0.438 0.425 0.408 0.403 0.379 alum res 36 0.236 0.200 0.183 0.211 0.1534 0.5767 0.3653 0.3712 0.275 0.339 0.412 0.388 0.394 0.394 0.333 alum res 36 0.170 0.162 0.125 0.135 0.1223 0.2643 0.2392 0.238 0.18 0.177 0.256 0.264 0.225 0.207 0.196 alum res 18 0.262 0.323 0.278 0.268 0.2245 0.4053 0.3241 0.3633 0.319 0.316 0.434 0.475 0.393 0.382 0.367 alum res 18 0.241 0.552 0.461 0.512 0.4097 0.6402 0.6185 0.5781 0.456 0.512 0.523 0.564 0.513 0.513 0.439 alum res 18 0.231 0.174 0.232 0.186 0.2688 0.4047 0.38 0.4161 0.382 0.383 0.378 0.408 0.397 0.402 0.345 alum res 9 0.238 0.278 0.174 0.327 0.2836 0.4488 0.4219 0.4022 0.303 0.389 0.451 0.481 0.457 0.438 0.368 alum res 9 0.159 0.271 0.121 0.200 0.1962 0.3563 0.3002 0.3231 0.299 0.295 0.344 0.366 0.35 0.345 0.343 alum res 9 0.299 0.289 0.260 0.249 0.221 0.362 0.3205 0.3385 0.293 0.311 0.338 0.389 0.361 0.4 0.383 PNAS 36 0.0542 0.0018 0.0180 0.0075 0.0028 0.1062 0.0053 0.0049 0.0043 0.0045 0.0393 0.0034 0.002 0.002 0.005 PNAS 36 0.0813 0.0423 0.0079 0.0161 0.0077 0.1027 0.0190 0.0060 0.0033 0.0047 0.1068 0.0080 0.009 0.004 0.007 PNAS 36 0.0698 0.0443 0.0106 0.0075 0.0040 0.0743 0.0010 0.0020 0.0028 0.0016 0.0224 0.0028 0.004 0.004 0.004 PNAS 18 0.017 0.006 0.003 0.002 0.005 0.0444 0.0199 0.0167 0.01 0.008 0.127 0.07 0.041 0.022 0.012 PNAS 18 0.022 0.016 0.004 0.002 0.0029 0.0577 0.0272 0.0131 0.006 0.007 0.114 0.058 0.025 0.016 0.009 PNAS 18 0.022 0.004 0.001 0.001 0.0034 0.0431 0.0059 0.0052 0.009 0.007 0.045 0.018 0.012 0.009 0.011 PNAS 9 0.057 0.032 0.014 0.010 0.0045 0.1087 0.0176 0.0119 0.007 0.006 0.17 0.045 0.017 0.007 0.007 PNAS 9 0.078 0.049 0.032 0.020 0.0096 0.1307 0.0662 0.0356 0.019 0.014 0.14 0.056 0.02 0.015 0.006 PNAS 9 0.054 0.059 0.027 0.019 0.0172 0.1335 0.0864 0.0702 0.046 0.042 0.153 0.106 0.082 0.056 0.031 control 0.238 0.455 0.346 0.404 0.3428 0.5311 0.4908 0.468 0.358 0.411 0.558 0.548 0.444 0.513 0.486 control 0.401 0.489 0.428 0.469 0.3681 0.6424 0.5785 0.6200 0.498 0.513 0.493 0.680 0.688 0.676 0.696 control 0.296 0.364 0.255 0.345 0.2989 0.4125 0.3774 0.4027 0.362 0.374 0.414 0.524 0.428 0.46 0.497

PAGE 221

202 APPENDIX B MESOCOSM STUDY RAW DATA Grow-in Water Qua lity Table B-1a. Raw mesocosm water column pH data for 20 weeks in 2004 after planting, prior to alum addition. Plant Mesocosm 7/8 7/15 7/22 7/298/68/128/198/269/169/30 10/1410/2811/1111/24 S 1 7.9 8.5 7.6 8.3 8.18.2 8.2 7.8 7.5 7.8 7.5 7.6 7.8 7.6 S 2 8.7 7.7 7.6 8.3 8.78.3 8.3 8.0 7.7 7.9 7.6 7.6 7.8 7.8 N 3 s 9.3 9.3 8.9 9.1 8.68.8 9.4 9.4 9.3 9.0 9.7 8.9 9.3 9.3 N 3 b 8.7 7.7 7.9 8.7 8.48.4 9.3 8.4 8.3 8.2 8.7 8.1 8.1 8.0 T 4 7.9 7.7 7.7 8.3 8.38.3 8.0 8.2 8.4 8.7 8.2 8.0 7.9 7.8 N 5 s 8.7 8.0 8.4 9.1 8.58.9 8.3 9.1 9.3 8.7 8.7 9.1 9.4 8.6 N 5 b 7.8 7.6 7.6 8.1 8.48.6 8.1 8.3 8.4 8.5 8.3 7.9 8.1 8.2 T 6 7.9 7.8 7.7 8.3 8.58.4 8.3 8.2 8.4 8.7 8.1 8.4 8.0 8.1 N 7 s 9.6 9.0 8.9 9.0 9.49.4 9.6 9.1 8.9 9.0 9.3 8.9 9.3 9.3 N 7 b 8.3 6.7 7.7 8.6 8.68.4 9.5 8.6 8.6 8.2 8.9 8.1 8.2 8.6 S 8 8.1 8.0 7.8 8.6 9.09.5 9.3 8.6 8.1 8.4 8.3 7.9 8.2 8.2 N 9 s 9.7 9.4 8.1 8.8 9.59.3 9.8 9.8 9.1 9.6 7.9 9.1 9.0 9.1 N 9 b 7.3 7.7 7.6 8.0 9.07.9 9.1 8.7 8.4 8.4 7.8 8.2 8.2 8.6 T 10 7.9 7.6 7.6 8.4 8.88.5 8.7 8.6 8.3 8.4 8.1 7.9 8.0 7.9 S 11 7.9 8.0 7.5 8.2 8.48.6 9.0 8.3 8.2 8.1 7.8 7.8 7.9 7.7 T 12 7.9 7.6 7.5 8.3 8.48.8 9.6 8.5 7.9 8.1 7.8 7.8 8.2 8.3 S 13 7.9 7.7 7.7 8.4 8.48.6 9.0 8.4 8.1 8.2 7.9 8.8 8.3 8.4 T 14 7.7 7.6 7.6 8.3 8.69.0 9.0 8.4 8.0 8.1 7.9 7.8 8.1 9.2 S 15 7.6 7.3 7.6 8.2 8.38.3 8.3 8.2 8.0 8.0 7.8 8.1 7.9 7.6 N 16 s 9.7 9.5 8.0 8.5 8.79.4 10.08.9 9.0 9.4 9.6 8.6 8.6 9.8 N 16 b 8.0 8.1 7.6 8.2 8.28.7 9.9 8.2 8.4 8.3 8.7 8.1 7.9 8.9 N 17 s 8.9 8.7 8.1 8.7 8.79.6 10.29.5 8.8 9.3 9.4 8.7 9.1 9.3 N 17 b 7.9 7.7 7.6 8.5 8.68.5 10.08.8 8.3 8.7 8.1 7.9 8.1 7.7 T 18 7.9 7.7 7.6 8.3 8.59.1 9.2 8.4 8.2 8.7 8.0 8.0 8.2 8.3 s = surface water column pH in SAV mesocosm, b = pH approximately 1” from soil surface in SAV mesocosm S = Scirpus californicus ; N = submerged aquatic vegetation; T = Typha spp.

PAGE 222

203 Table B-2. Raw outflow sol uble reactive phosphorus (mg L-1) data for mesocosms nine weeks prior to alum addition. Plant Mesocosm 10/1/0410/14/ 0410/28/0411/11/0411/18/04 11/24/04 S 1 0.069 0.017 0.006 0.025 0.01 0.028 S 2 0.075 0.016 0.011 0.025 0.004 0.024 N 3 0.021 0.005 0.006 0.016 0.015 0.013 T 4 0.044 0.022 0.010 0.023 0.022 0.07 N 5 0.029 0.007 0.007 0.022 0.009 0.025 T 6 0.012 0.011 0.006 0.027 0.007 0.049 N 7 0.034 0.002 0.002 0.003 0.005 0.012 S 8 0.072 0.015 0.013 0.029 0.010 0.077 N 9 0.021 0.019 0.010 0.003 0.007 0.034 T 10 0.019 0.011 0.007 0.018 0.008 0.061 S 11 0.045 0.016 0.014 0.019 0.003 0.054 T 12 0.014 0.023 0.011 0.020 0.006 0.054 S 13 0.065 0.025 0.012 0.023 0.026 0.048 T 14 0.032 0.016 0.011 0.019 0.007 0.016 S 15 0.073 0.032 0.021 0.033 0.036 0.006 N 16 0.046 0.014 0.008 0.016 0.012 0.004 N 17 0.018 0.012 0.003 0.010 0.006 0.003 T 18 0.038 0.027 0.013 0.018 0.013 0.005 Inflow 0.108 0.080 0.052 0.085 0.108 0.127 DI Blank 0.000 0.000 0.000 0.000 0.001 0.001 S = Scirpus californicus ; N = submerged aquatic vegetation; T = Typha spp.

PAGE 223

204 Table B-3. Raw outflow total kjeldahl nitrogen data (mg L-1) for mesocosms nine weeks prior to alum addition. Plant Mesocosm 10/1/04 10/14/ 04 10/28/0411/11/0411/18/04 11/24/04 S 1 0.666 0.600 0.718 0.586 0.466 0.536 S 2 0.607 0.482 0.630 0.675 0.795 0.615 N 3 0.841 0.600 0.748 0.705 0.526 0.496 T 4 0.666 0.698 0.689 0.586 0.556 0.496 N 5 0.812 0.836 0.600 0.645 0.466 0.466 T 6 0.607 0.718 0.777 0.615 0.526 0.526 N 7 0.548 0.718 0.836 0.675 0.466 0.496 S 8 0.724 0.659 0.718 0.615 0.496 0.526 N 9 0.783 0.836 0.895 0.675 0.496 0.496 T 10 0.724 0.954 0.895 0.675 0.795 0.526 S 11 0.636 0.718 0.659 0.675 0.586 0.466 T 12 0.695 0.659 0.689 0.566 0.466 0.466 S 13 0.959 0.718 0.954 0.675 0.675 0.406 T 14 0.724 0.748 0.748 0.675 0.496 0.466 S 15 0.724 0.836 0.659 0.795 0.466 0.556 N 16 0.724 0.659 0.777 0.705 0.466 0.526 N 17 0.666 0.659 0.807 0.675 0.586 0.556 T 18 0.724 0.659 0.630 0.645 0.436 0.615 Inflow 0.607 0.541 0.659 0.795 0.496 0.550 DI Blank 0.000 0.000 0.026 0.048 0.000 0.000 S = Scirpus californicus ; N = submerged aquatic vegetation; T = Typha spp.

PAGE 224

205 Table B-4. Raw outflow disso lved organic carbon (mg L-1) data for mesocosms 9 weeks prior to alum addition. Plant Mesocosm 10/1/0410/14/ 0410/28/0411/11/0411/18/04 11/24/04 S 1 8.209 10.340 8.884 7.970 8.893 8.941 S 2 8.252 9.938 8.084 8.290 9.598 8.732 N 3 7.952 8.695 8.310 8.384 8.005 8.156 T 4 8.215 9.883 8.691 9.295 9.869 8.507 N 5 7.994 9.792 8.333 8.199 8.47 8.312 T 6 7.926 11.14 8.984 8.748 8.926 8.896 N 7 8.139 9.204 7.921 8.065 7.864 8.158 S 8 8.506 10.030 8.200 8.597 8.818 8.396 N 9 8.138 10.380 8.587 8.819 9.188 8.418 T 10 8.253 11.230 9.002 9.405 9.279 8.354 S 11 7.992 9.920 8.543 9.005 9.385 8.402 T 12 8.296 10.300 8.426 8.931 9.168 8.638 S 13 7.671 10.310 8.331 8.722 8.43 8.816 T 14 8.148 10.720 8.812 8.931 8.917 9.363 S 15 8.347 9.836 8.470 8.922 8.518 9.7 N 16 7.857 10.390 8.168 8.489 8.129 7.997 N 17 7.995 10.310 8.223 8.211 7.874 8.052 T 18 8.604 9.862 8.549 8.792 8.868 9.069 Inflow 7.855 9.706 8.012 8.390 8.593 8.233 DI Blank 1.204 0.696 0.465 0.315 0.467 S = Scirpus californicus ; N = submerged aquatic vegetation; T = Typha spp.

PAGE 225

206 Table B-5. Initial characterization of soil used in mesocosm establishment (n=5). Parameter Units Moisture % 19.9 0.02 pH pH units 5.21 0.28 LOI % 8.24 1.21 Total P mg kg-1 111 13.4 HCl Ca mg kg-1 1418 235.0 HCl Mg mg kg-1 16.8 3.79 HCl Fe mg kg-1 600 57.2 HCl Al mg kg-1 455 35.3 Oxalate Al mg kg-1 285 21.2 PMP mg kg-1 d-1 0.60 0.18 KCl Pi mg kg-1 0.21 0.15 NaOH Pi mg kg-1 24.9 2.11 HCl Pi mg kg-1 5.67 3.30 Sum TPi mg kg-1 30.7 4.66 NaOH Po mg kg-1 36.3 3.12 Residue Po mg kg-1 34.4 16.7 Sum TPo mg kg-1 70.7 18.7 LOI = loss on ignition; PMP = pot entially mineralizeable phosphorus.

PAGE 226

207 y = 0.0002x2 0.0065x + 0.170 R2 = 0.9520 2 4 6 8 10 12 14 16 050100150200250 Length (cm)Dry Weight (g) Figure B-1. Scirpus californicus leaf length to dry weight regression harvested from cell 2 in the Orlando East erly Wetland (n=30). y = 0.0002x2 0.0173x + 7.617 R2 = 0.9380 2 4 6 8 10 12 14 16 050100150200250 Length (cm)Dry Weight (g) Figure B-2. Typha spp. leaf length to dry weight regr ession harvested from cell 10 in the Orlando Easterly Wetland (n=12 plants, 69 leaves).

PAGE 227

APPENDIX C FIELD STUDY SPATIAL DATA Table C-1. Spatial coordinate s f or a ll samples collected within cell 9 and 10 of the Orlando Easterly Wetland (UTM Zone 17, NAD83). Distance from inflow (m) Site Northing Easting 10 9A1 3160707 499587.1 35 9A2 3160735 499581.6 80 9A3 3160784 499600.7 150 9A4 3160892 499611.5 10 9B5 3160670 499907.6 35 9B6 3160673 499910.4 80 9B7 3160762 499644.1 150 9B8 3160799 499706.6 10 10C9 3160528 499850.6 35 10C10 3160556 499861.5 80 10C11 3160593 499875.0 150 10C12 3160670 499904.9 10 10D13 3160528 499891.3 35 10D14 3160538 499915.8 80 10D15 3160538 499918.5 150 10D16 3160556 499983.7 208

PAGE 228

209 Figure C-1. Field study tran sects in control cell 9 and alum-treated cell 10 of the Orlando Easterly Wetland. C C 9 9 C C 1 1 0 0 C C 1 1 1 1 C C 1 1 2 2 D D 1 1 3 3 D D 1 1 4 4 D D 1 1 5 5 D D 1 1 6 6 A A 1 1 A A 2 2 A A 3 3 A A 4 4 B B 5 5 B B 6 6 B7 B B 8 8 9 10

PAGE 229

210Table C-2. Soluble reactive phosphorus coll ected in 2005 from replicate transects in cell 9 and 10 of the Orlando Easterly Wet land. Weir or Plot 8/4 8/11 8/18 8/23 9/7 9/14 9/21 9/28 10/13 10/19 10/26 11/9 11/23 12/12 12/27 5X 0.42 0.242 0.162 0.172 0.472 0.171 0.52 0.353 0.447 0.45 0.08 0.394 0.235 0.552 0.329 9X 0.406 0.237 0.176 0.165 0.338 0.182 0.433 0.323 0.296 0.464 0.117 0.427 0.201 0.603 0.343 9Y 0.396 0.203 0.146 0.095 0.271 0.174 0.434 0.258 0.227 0.368 0.099 0.391 0.245 0.553 0.324 9A1 0.4 0.218 0.154 0.141 0.454 0.148 0.508 0.287 0.435 0.465 0.095 0.377 0.227 0.602 0.314 9A2 0.389 0.277 0.249 0.161 0.413 0.175 0.509 0.332 0.432 0.416 0.458 0.377 0.231 0.622 0.316 9A3 0.378 0.256 0.047 0.007 0.427 0.142 0.446 0.466 0.03 0.065 0.373 0.519 0.333 9A4 0.387 0.19 0.115 0.056 0.39 0.118 0.413 0.15 0.349 0.366 0.122 0.363 0.038 0.45 0.336 9B5 0.26 0.205 0.062 0.125 0.429 0.153 0.107 0.339 0.445 0.45 0.098 0.386 0.223 0.616 0.313 9B6 0.357 0.201 0.056 0.066 0.444 0.036 0.499 0.321 0.436 0.451 0.01 0.356 0.224 0.522 0.271 9B7 0.194 0.181 0.021 0.011 0.448 0.116 0.529 0.406 0.377 0.097 0.383 0.242 0.672 0.338 9B8 0.445 0.572 0.457 0.187 0.165 0.234 0.59 0.318 6X 0.354 0.024 0.18 0.128 0.158 0.010 0.192 0.102 0.264 0.253 0.16 0.053 0.32 0.54 0.223 10X 0.174 0.177 0.118 0.096 0.08 0.164 0.052 0.147 0.15 0.149 0.246 0.179 0.272 0.391 0.14 10Y 0.236 0.117 0.108 0.093 0.115 0.162 0.062 0.118 0.149 0.143 0.217 0.159 0.395 0.34 0.099 10C9 0.318 0.018 0.171 0.082 0.007 0.007 0.164 0.074 0.204 0.173 0.179 0.109 0.143 0.536 0.168 10C10 0.29 0.021 0.15 0.039 0.013 0.021 0.071 0.084 0.177 0.158 0.167 0.094 0.164 0.527 0.153 10C11 0.097 0.595 0.473 0.374 0.265 0.008 0.175 0.082 0.167 0.116 0.174 0.134 0.25 0.504 0.295 10C12 0.992 1.033 0.809 0.642 0.676 0.170 0.457 0.674 0.511 0.406 0.579 0.325 0.23 0.431 0.321 10D13 0.336 0.024 0.174 0.054 0.017 0.008 0.147 0.086 0.186 0.212 0.179 0.085 0.111 0.524 0.122 10D14 0.287 0.134 0.166 0.019 0.033 0.346 0.018 0.059 0.18 0.187 0.168 0.198 0.258 0.498 0.183 10D15 0.519 0.043 0.187 0.07 0.043 0.165 0.112 0.017 0.257 0.218 0.224 0.244 0.488 10D16 0.243 0.274 0.276 0.212 0.289 0.32 0.395 0.426

PAGE 230

211Table C-3. Soluble reactive phosphorus coll ected in 2006 from replicate transects in cell 9 and 10 of the Orlando Easterly Wet land. Weir or Plot 1/10 1/24 2/7 2/21 3/7 3/21 4/4 4/21 5/11 5/25 6/16 6/27 7/11 7/27 8/10 8/17 5X 0.175 0.427 0.252 0.423 0.296 0.445 0.559 0.660 0.291 0.141 0.367 0.186 0.106 0.178 0.229 0.243 9X 0.168 0.534 0.216 0.493 0.305 0.425 0.517 0.573 0.239 0.154 0.330 0.145 0.116 0.185 0.198 0.190 9Y 0.16 0.515 0.207 0.458 0.273 0.432 0.499 0.471 0.201 0.137 0.293 0.122 0.105 0.134 0.177 0.192 9A1 0.016 0.422 0.225 0.381 0.264 0.347 0.543 0.656 0.287 0.083 0.368 0.205 0.109 0.177 0.238 0.256 9A2 0.168 0.424 0.247 0.425 0.315 0.484 0.564 0.623 0.325 0.422 0.180 0.098 0.179 0.112 0.214 9A3 0.181 0.413 0.257 0.234 0.293 0.396 0.339 0.395 0.186 0.111 0.178 0.227 0.247 9A4 0.139 0.546 0.184 0.514 0.277 0.411 0.630 0.516 0.272 0.103 0.353 0.115 0.191 0.169 0.158 0.184 9B5 0.102 0.342 0.154 0.39 0.283 0.446 0.474 0.679 0.295 0.180 0.352 0.185 0.107 0.178 0.227 0.256 9B6 0.178 0.414 0.247 0.429 0.283 0.452 0.564 0.673 0.361 0.199 0.104 0.192 0.237 0.258 9B7 0.112 0.43 0.252 0.417 0.295 0.438 0.563 0.651 0.163 0.194 0.315 0.136 0.125 0.104 0.192 0.192 9B8 0.166 0.441 0.24 0.416 0.287 0.410 0.409 0.381 0.203 0.039 0.017 0.028 0.376 6X 0.028 0.51 0.02 0.021 0.009 0.001 0.004 0.318 0.105 0.089 0.244 0.000 0.005 0.116 0.125 0.042 10X 0.09 0.248 0.163 0.028 0.207 0.200 0.110 0.066 0.032 0.018 0.028 0.025 0.037 0.022 0.035 0.034 10Y 0.062 0.275 0.224 0.035 0.221 0.070 0.045 0.024 0.029 0.007 0.015 0.018 0.033 0.021 0.021 0.021 10C9 0.022 0.484 0.016 0.016 0.008 0.001 0.002 0.057 0.073 0.056 0.209 0.000 0.002 0.101 0.113 0.038 10C10 0.019 0.461 0.017 0.012 0.007 0.000 0.006 0.031 0.039 0.048 0.170 0.000 0.003 0.023 0.039 0.049 10C11 0.023 0.368 0.033 0.035 0.049 0.002 0.004 0.003 0.009 0.069 0.061 0.034 10C12 0.146 0.384 0.053 0.016 0.122 0.098 0.042 0.024 0.018 0.125 0.104 0.136 0.175 0.312 0.309 10D13 0.019 0.35 0.02 0.013 0.009 0.000 0.002 0.016 0.040 0.014 0.174 0.000 0.011 0.069 0.063 0.059 10D14 0.015 0.344 0.048 0.026 0.121 0.034 0.025 0.070 0.017 0.070 0.041 0.011 0.020 0.029 0.046 0.071 10D15 0.033 0.365 0.108 0.026 0.123 0.016 0.039 0.011 0.037 0.000 0.025 0.021 0.041 0.048 10D16 0.426 0.019 0.371 0.038 0.023 0.161 0.039 0.018 0.016 0.019 0.066 0.058 0.048 0.022 0.035 0.038

PAGE 231

212 LIST OF REFERENCES 2002. Iron Bridge Regional Wa ter Reclama tion Facility. Water Environment & Technology 14: 32-33. Aerts, R., J. T. A. Verhoven, and D. F. Whigham. 1999. Plant-mediated controls on nutrient cycling in temperate fens and bogs. Ecology 80: 2170-2181. Albers, P. H. and M. B. Camardese. 1993. Effects of acidification on metal accumulation by aquatic plants and invertebrates. 1. Constructed Wetlands. Environmental Toxicology and Chemistry 12: 959-967. Allen, W. C., P. B. Hook, J. A. Biederman, and O. R. Stein. 2002. Temperature and wetland plant species effects on wastewater treatment and root zone oxidation. Journal of Environmental Quality 31: 1010-1016. Amonette, J. E. 2002. Methods for determ ination of mineralogy and environmental availability. In Dixon, J. B. and D. G. Schulze (eds.) Soil Mineralogy with Environmental Applications. Soil Science Society of America, Inc.: Madison, WI. pp. 153-197. Andersen, J. M. 1976. An ignition method for determination of total phosphorus in lake sediments. Water Research 10: 329-331. Ann, Y. K. 1995. Phosphorus immobilization by chemical amendments in a constructed wetland. Ph.D. Dissertation. University of Florida, Gainesville, FL. Ann, Y., K. R. Reddy, and J. J. Delfino. 2000. Influence of chemical amendments on phosphorus immobilization in soils from a constructed wetland. Ecological Engineering 14: 157-167. APHA. 1998. Standard methods for the ex amination of water and wastewater. 20th ed. American Public Health Association: Baltimore, MD. Auter, T., T. A. Debusk, F. E. Dierberg, M. Sees. 2003. An evaluation of techniques for rejuvenating the nutrient removal capacity of treatment wetland communities. In Proceedings of the 12th Annual North American Lake Management Society Southeastern Lakes Management Confer ence. June 2-5, 2003: Orlando, FL. Bachand, P. A. M. and A. J. Horne. 2000. Denitrification in c onstructed wetland freewater surface wetlands: II. Effects of ve getation and temperature. Ecological Engineering 14: 17-32.

PAGE 232

213 Baker, J. P. 1982. Effects on fish and metals associated with acidification. In Johnson, R. E. (ed.) Acid Rain/Fisheries. American Fisheries Society. pp. 165-176. Barbiero, R., R. E. Carlson, G. D. Cooke, A. W. Beals. 1988. The effects of a continuous application of aluminum sulfat e on lotic benthic inve rtebrates. Lake and Reservoir Management 4: 63-72. Barcelo, J. and C. Poschenrieder. 2002. Fast root growth responses, root exudates, and internal detoxification as clues to the mechanisms of aluminium toxicity and resistance: A review. Environmenta l and Experimental Botany 48:75-92. Barko, J. W., D. Gunnison, and S. R. Carpen ter. 1991. Sediment interactions with submersed macrophytes growth and comm unity dynamics. Aquatic Botany 41: 4165. Barko, J. W., W. F. James, W. D. Taylor, and D. G. McFarland. 1990. Effects of alum treatment on phosphorus and phytoplankton dynamics in Eau Gale Reservoir: a synopsis. Lake and Reservoir Management 6: 1-8. Barnes, R. B. 1975. The determination of speci fic forms of aluminum in natural water. Chemical Geology 15: 177-191. Baudo, R., G. Galanti, P. Guilizzoni, and P. Varini. 1981. Relationship between heavy metals and aquatic organisms in Lake Mezzola hydrographic system (Northern Italy). 4. Metal concentration in six submerged aquatic macrophytes. Memorie dell'Istituto Italiano di Idrobiologia. 39: 203-225. Beecroft, J. R., M. C. Koether, and G. W. vanLoon. 1995. The chemical nature of precipitates formed in soluti ons of partially neutralized aluminum sulphate. Water Research 29: 1461-1464. Bennet, R. J., C. M. Breen and M. V. Fey. 1987. The effects of aluminum on root cap function and root development in Zea mays L. Environmental and Experimental Botany 27: 91-104. Bera, R. R., A. Seal, M. Banerjee, and A.K. Dolu. 2005. Nature and profile distribution of iron and aluminum in relation to pedoge nic processes in some soils developed under tropical environmen t in India. Environm ental Geology 47: 241-245. Berg, D. J. and T. A. Burns. 1985. Distributi on of aluminum in the tissues of three fish species. Journal of Fres hwater Ecology 3: 113-120. Bertsch, P. M. 1990. Aluminum speci ation: Methodology a nd applications. In Norton, S. A., S. E. Lindberg, and A. L. Page (eds .) Advances in Environmental Science. Acidic Precipitation Vol. 4. Spri nger-Verlag: New York, NY. pp. 63-105. Bertsch, P. M. and P. R. Bloom. 1996. Aluminum. In D. L. Sparks (ed.) Methods of soil analysis. Part 3. SSSA Book Series 5. SSSA: Madison, WI. pp.517-550.

PAGE 233

214 Bertsch, P. M. and D. R. Parker. 1996. Aqueous polynuclear aluminum species. In G. Sposito (ed.) The Environmental Chemistry of Aluminum, 2nd ed. CRC Press: Boca Raton, FL. pp 117-168. Bittl,T., J. Vrba, J. Nedoma, and J. Kopacek. 2001. Impact of ionic aluminium on extracellular phosphatases in acidified lakes. Environm ental Microbiology 3: 578587. Black, C. A. and W. R. Wise. 2003. Evalua tion of past and potenti al phosphorus uptake at the Orlando Easterly Wetland. Ecological Engineering 21: 277-290. Blake, G. R. and K. H. Hartge. 1986. Bulk Density. In Klute, A. (ed.) Methods of Soil Analysis. Part 1, Agronomy No. 9. American Society of Agronomy: Madison, WI. pp. 363-375. Blomquist, K., M. Cronholm, and A. Jernelv. 1971. Frsk med fosfatreduktion genom kemisk fllning i Lngsjn [In Swedis h, English summary]. Vatten 27(2). Bodelier, P. L. E., J. A. Libochant, C. W. P. M. Blom, and H. J. Laanbroek. 1996. Dynamics of nitrification and denitrif ication in root-oxygenated sediments and adaptation of ammonia-oxidizing bacteria to low-oxygen or anoxic habitats. Applied Environmental Microbiology 62: 4100-4107. Boisvert, J. P. and C. Jolicoeur. 1999. Infl uences of sulfate and/or silicate present in partially prehydrolyzed Al(III) flocculants on Al(III) speciation in diluted solutions. Colloid and Surfaces A: Physicochemi cal and Engineering Aspects 155: 161-170. Boisvert, J., T. C. To, A. Berrak, and C. Jolicoeur. 1997. Phosphate adsorption in flocculation processes of aluminum sulpha te and poly-aluminum-silicate-sulphate. Water Research 31: 1939-1946. Bostan, V., J. Dominik, M. Bostina, and M. Pardos. 2000. Forms of particulate phosphorus in suspension and in bottom se diment in the Danube Delta. Lakes & Reservoirs: Research an d Management 5: 105-110. Bottero, J. Y., J. M. Case, F. Fiessinger, and J. E. Poivier. 1980. Studies of hydrolyzedaluminum chloride solutions. I. Nature of aluminum species and composition of aqueous solutions. Jour nal of Physical Chemistry 84: 2933-2939. Boyd, C. E. 1970. Production, mineral accumu lation and pigment concentrations in Typha latifolia and Scirpus americanus. Ecology 51: 285-290. Brady, N. C. and R. R. Weil. 1999. Soil organic matter. In The Nature and Properties of Soil. Prentice-Hall, Inc.: Upper Sa ddle River, New Jersey. pp. 446-490. Brix, H. 1994. Functions of macrophytes in constructed wetlands . Water Science & Technology 29: 71.

PAGE 234

215 Brix, H. 1997. Do macrophytes play a role in constructed treatment wetlands? Water Science & Technology 35: 11-17. Brix, H., K. Dyhr-Jensen, and B. Lorenzen. 2002. Root-zone acidity and nitrogen source affects Typha latifolia L. growth and uptake kinetics of ammonium and nitrate. Journal of Experime ntal Botany 53: 2441-2450. Brookes, P. C., A. Landman, G. Pruden, and D. S. Jenkinson. 1985. Chloroform fumigation and the release of soil nitrogen: a rapid direct extraction method to measure microbial biomass nitrogen in so ils. Soil Biology and Biochemistry 17: 83-84. Brookes, P. C., D. S. Powlson, and D. S. Jenkinson. 1982. Measurement of microbial biomass phosphorus in soil. Soil Biology and Biochemistry 14: 319-329. Brower, J. E., J. H. Zar, and C. N. von Ende. 1998. Biomass Measurements. In Field and Laboratory Methods for General Ecology. WCB/McGraw-Hill: Boston, MA. pp. 200-205. Bulc, T., D. Vrhovek, and V. Kukanja. 1997. The use of constructed wetland for landfill leachate treatment. Wa ter Science & Technology 35: 301-306. Bulson, P. C., D. L. Johnstone, H. L. Gi bbons, and W. H. Funk. 1984. Removal and inactivation of bacteria during alum treatment of a lake. Applied and Environmental Microbiology 48: 425-430. Burney, J. L., S. T. Baccus, and J. B. Lee. 1989. An evaluation of wildlife utilization in a man-made freshwater wetland system in central Florida, U. S. A. In Webb, F. J., Jr. (ed.) Proceedings of the Sixteenth Annual Conference on Wetlands Restoration and Creation. Hillsborough Community College, Plant City, FL. pp. 24-48. Burrows, H. D. 1977. Aquatic aluminum chemistry, toxicology, and environmental prevalence. Critical Reviews in Environmental Control 7: 167-216. Caffrey, J. M. and W. M. Kemp. 1991. Seasonal and spatial patterns of oxygen production, respiration and r oot-rhizome release in Potamogeton perfoliatus L. and Zostera marina L. Aquatic Botany 40: 109-128. Campbell, P. G. C., M. Bisson, R. Bougie, A. Tessier, and J. P. Villeneuve. 1983. Speciation of aluminum in acidic freshw aters. Analytical Chemistry 55: 22462252. Cardwell, A. J., D. W. Hawker, and M. Greenway. 2002. Metal accumulation in aquatic macrophytes from southeast Queensland, Australia. Chemosphere 48: 653-663. Carignan, R. and J. Kalff. 1979. Quantifica tion of the sediment phosphorus available to aquatic macrophytes. Journal of the Fish eries Research Board of Canada 36: 10021008.

PAGE 235

216 Chakraborti, R. K., K. H. Ga rdner, J. F. Atkinson, and J. E. Van Benschoten. 2003. Changes in fractal dimension during aggr egation. Water Re search 37: 873-883. Chao, G. Y., J. Baker, A. P Sabina, and A. C. Roberts. 1985. Doyleite, a new polymorph of Al(OH)3, and its relationship to bayerite, gi bbsite, and nordstra ndite. Canadian Mineralogy 23: 21-28. Christian, R. R. 1989. Microbial ecology and organic detritus in estuaries. In Day, J. W., Jr., C. A. S. Hall, W. M. Kemp, and A. Y ez-Arancibia (eds.) Estuarine Ecology. John Wiley & Sons: New York. pp. 257-308. Chua, T., 2000. Mineralization of organic phosphorus in a subtropical freshwater wetland. Ph.D. dissertation, University of Florida, Gainesville, FL. Clarke, E. and A. H. Baldwin. 2002. Respons es of wetland plants to ammonia and water level. Ecological Engineering 18: 257-264. Collins, B. S., R. R. Sharitz, and D. P. C oughlin. 2005. Elemental composition of native wetland plants in constructed mesoco sm treatment wetlands. Bioresource Technology 96: 937-948. Connor, J. N. and M. R. Martin. 1989. An assessment of sediment phosphorus inactivation, Kezar Lake, New Hampshir e. Water Resources Bulletin 25: 845-853. Connor, J. N. and G. N. Smith. 1986. An e fficient method of applying aluminum salts for sediment phosphorus inactivation in la kes. Water Resources Bulletin 22: 661664. Cooke, G. D., E. B. Welch, A. B. Martin, D. G. Fulmer, J. B. Hyde, and G. D. Schrieve. 1993a. Effectiveness of Al, Ca, and Fe salts for control of internal phosphorus loading in shallow and deep la kes. Hydrobiol ogia 253: 323-335. Cooke, G. D., E. B. Welch, S. A. Pete rson, and P. R. Newroth. 1993b. Phosphorus activation and sediment oxidation. In Restoration and Management of Lakes and Reservoirs. Lewis Publishe rs: Boca Raton, FL. pp. 161-209. Cross, A. F. and W. H. Schleslinger. 1995. A literature review of the Hedley fractionation: Applications to the biog eochemical cycle of soil phosphorus in natural ecosystems. Geoderma 64: 197-214. Crowder, A. 1991. Acidification, metals, and macrophytes. Envi ronmental Pollution 71: 171-203. Culley, J. L. 1993. Density and compressibility. In Carter, M. R. (ed.) Soil Sampling and Methods of Analysis. Lewi s Publishers: Boca Raton. pp. 529-539.

PAGE 236

217 Dayton E. A. and N. T. Basta. 2005. A method for determining the phosphorus sorption capacity and amorphous aluminum of alum inum-based drinking water treatment residuals. Journal of Environmental Quality 34: 1112-1118. Dao, T.H., L.J. Sikora, A. Hamasaki, and R.L. Chaney. 2001. Manure phosphorus extractability as affected by alum inumand iron by-products and aerobic composting. J. Environ.l Qual. 30:1693-1698. Davis, J. A. and R. Gloor. 1981. Adsorption of dissolved organics in lake water by aluminum oxide. Effect of molecula r weight. Environmental Science and Technology 16: 1223-1230. Davis, S. M. 1984. Cattail leaf produc tion, mortality, and nutrient flux in Water Conservation Area 2A. Technical Pub lication #84-8. South Florida Water Management District: West Palm Beach, FL. 40 p. DB Environmental, Inc. 2004. An evaluation of the use of alum to enhance phosphorus removal performance by the Orlando Easterly Wetlands. Report prepared for Post, Buckley Schuh & Jernigan, Inc. DB E nvironmental, Inc.: Rockledge, FL. 118 p. DeBusk, W. F., K. R. Reddy, M. S. Koch, and Y. Wang. 1994. Spa tial distribution of soil nutrients in a northern Everglades ma rsh: Water Conservation Area 2A. Soil Science Society of America Journal 58: 543-552. Degens, B. P., L. A. Schipper, G. P. Sparli ng, and L. C. Duncan. 2001. Is the microbial community in a soil with reduced catabolic diversity less resistant to stress or disturbance? Soil Biology & Biochemistry 33: 1143-1153. Delhaize, E., and P. R. Ryan. 1995. Aluminum toxicity and toleran ce in plants. Plant Physiology 107: 315-321. Dennison, W. C., R. J. Orth, K. A. Moore, J. C. Stevenson, V. Carter, S. Kollar, P. Bergstrom and R. A. Batiuk. 1993. Assessing water quality with submersed aquatic vegetation. BioScience 43: 86-94. Denny, P. 1997. Implementation of construc ted wetlands in developing countries. Water Science & Technology 35: 27-34. Diamadopoulos, E. and A. Benedek. 1984. Aluminum hydrolysis effects on phosphorus removal from wastewaters. Wa ter Pollution Control 56: 1165-1172. Diaz, O. A., K. R. Reddy, and P. A. Moore. 1994. Solubility of inorganic P in stream water as influenced by pH and Ca con centration. Water Research 28: 1755-1763. Dickerman, J. A. and A. J. Stewart. 1986. Estimates of net annual aboveground production: Sensitivity to sampli ng frequency. Ecology 67: 650-659.

PAGE 237

218 Dierberg, F. E., T. A. Debusk, J. Potts, and B. Gu. 2002. Biological uptake vs. coprecipitation of soluble reactive phosphorus by “P-enriched” and “P-deficient” Najas guadalupensis in hard and soft waters. Verhandlungen der Internationalen Vereinigung fr theoretische und angewandte Limnologie 28: 1865-1870. Dolui, A. K. and N. Chakraborty. 1998. Differentiation of forms of extractable aluminum in relation to pedogenic processe s in two alfisols and an entisol. International Journal of Tr opical Agriculture 16: 257-266. Driscoll, C. T. and W. D. Schecher. 1990. The chemistry of aluminum in the environment. Environmental geochemistry and health 12: 28-49. Duarte, C. M. 1992. Nutrient concentration of aquatic plants: Patterns across species. Limnology and Oceanography 37: 882-889. Dunst, R. C., S. M. Born, P. D. Uttormark, S. A. Smith, S. A. Nichols, J. O. Peterson, et al. 1974. Survey of Lake Rehabilitation Techniques and Experiences. Technical Bulletin No. 75. Department of Natural Resources, Madison, WI. 179 p. Easthouse, K. B., D. E. Spyridakis, and E. B. Welch. 1993. Alumin um chemistry in two pristine alpine lakes of the Central Ca scades, Washington. Water, Air, and Soil Pollution 71: 377-390. Ebeling, J. M., Sibrell, P. L., Ogden, S. R., Summerfelt, S. T., 2003. Evaluation of chemical coagulation-flocculation aids for the removal of suspended solids and phosphorus from intensive recirculating aquaculture effluent discharge. Aquacultural Engineering 29: 23-42. Eleftheriou, E. P., M. Moustakas and N. Frag iskos. 1993. Aluminate-induced changes in morphology and ultrastructure of Thinopyrum roots. Journal of Environmental Botany 44: 427-536. Elzinga, C. L., D. W. Salzer, J. W. Will oughby, and J. P. Gibbs. 2001. Field techniques for measuring vegetation. In Monitoring Plant and Animal Populations. Blackwell Science, Inc.: Malden, MA. pp. 205-230. Emery, S. L. and J. A. Perry. 1995. Aboveground biomass and phosphorus concentrations of Lythrum salicaria (Purple Loosestrife) and Typha spp. (Cattail) in 12 Minnesota wetlands. Entranco. 1986. Wapato Lake Restorati on: A discussion of de sign considerations, construction techniques a nd performance evaluation. Final Report, Entranco Engineering: Bellevue, WA. Entranco. 1987. Pattison and Long Lakes restoration project. Final Report, Entranco Engineering, Inc.: Bellevue, WA.

PAGE 238

219 Eriksson, L. and C. Axberg. 1981. Direct influence of wastewater pollutants on flocculation and sedimentation behavior in biological wastewater treatment-I model system E. coli. Water Research 15: 421-432. Exall, K. N. and G. W. Vanloon. 2000. Usi ng coagulants to remove organic matter. Journal of American Water Work s Association 92(11): 93-102. Farnham, R. S. and M. R. Finney. 1965. Cla ssification and propertie s of organic soils. Advanced Agronomy 17: 115-162. Farve, M., W. Harris, F. Dierberg, and K. Portier. 2004. Association between phosphorus and suspended solids in an Ev erglades treatment wetland dominated by submersed aquatic vegetation. Wetlands Ecology and Management 12: 365-375. Faulkner, S. P. and C. J. Richardson. 1989. Physical and chemical characteristics of freshwater wetland soils. In D. A. Hammer (ed.) Constructed Wetlands for Wastewater Treatment. Munici pal, Industrial and Agricultural. Lewis Publishers: Chelsea, MI. pp. 41. Fisher, M. M. and K. R. Reddy. 2001. Phos phorus flux from wetland soils affected by long-term nutrient loading. Journal of Environmental Quality 30: 261-271. Forchhammer, N. C. 1999. Production potentia l of aquatic plants in systems mixing floating and submerged macrophytes. Freshwater Biology 41: 183-191. Foy, R. H. 1985. Phosphorus inactivation in an eutrophic lake by direct addition of ferric aluminum sulfate: Impact on iron and phosphorus. Freshwater Biology 15: 613-630. Freeman, R. A. and W. H. Ev erhart. 1971. Toxicity of aluminum hydroxide complexes in neutral and basic media to rainbow trout. Transactions of the American Fisheries Society 100: 644-658. Furumai, H. and S. Ohgaki. 1982. Fract ional composition of phosphorus forms in sediments related to release. Water Science and Technology 14: 215-226. Fytianos, K. and A. Kotzakiota. 2005. Se quential fractionation of phosphorus in lake sediments of northern Greece. Environmental Monitoring and Assessment 100: 191-200. Galarneau, E. and R. Gehr. 1997. Phosphorus removal from wastewaters: Experimental and theoretical support for alternative mechanisms. Water Research 31: 328-338. Gallon C., C. Munger, S. Prmont and P. G. C. Campbell. 2004. Hydroponic study of aluminum accumulation by aquatic plants: E ffects of fluoride and pH. Water, Air & Soil Pollution 153: 135-155.

PAGE 239

220 Gambrell, R. P. 1994. Trace and toxic metals in wetlands. Journal of Environmental Quality 23: 883-891. Gensemer, R. W. and R. C. Playle. 1999. Th e bioavailability and toxicity of aluminum in aquatic environments. Critical Re views in Environmental Science and Technology 29:315. Goldberg, S., J. A. Davis, and J. D. Hem. 1996. The surface chemistry of aluminum oxides and hydroxides. In Sposito, G. (ed.) The environmental chemistry of aluminum. CRC Press/Lewis Publ ishing: Boca Raton, FL. pp. 271-331. Gomez, E., C. Durillon, G. Rofes, and B. Pi cot. 1999. Phosphate adsorption and release from sediments of brackish lagoons: pH, O2 and loading influence. Water Research 33: 2437-2447. Goransson, A. and T. D. Eldhuset. 1991. E ffects of aluminum on growth and nutrient uptake of small Picea abies and Pinus sylvetris plants. Trees, structure and function. Tresey 5: 136-142. Gorsuch, T. T. 1970. Destruction of organic matter. In International Series of Monographs in Analytical Chemistry, Vo lume 39. Pergannon Press: New York, NY. Gris, D., J. E. Titus, and D. J. Wagner. 1986. Environmental pH influences growth and tissue chemistry of the submersed macrophytes Vallisneria americana. Canadian Journal of Botany 64: 306-309. Guildford, S. J. and R. E. Hecky. 2000. Total nitrogen, total p hosphorus, and nutrient limitation in lakes and oceans: Is there a common relationship? Limnology and Oceanography 45: 1213-1223. Guilizzoni, P. 1991. The role of heavy meta ls and toxic materials in the physiological ecology of submersed macrophytes . Aquatic Botany 41: 87-109. Gumbricht, T. 1993. Nutrient removal proc esses in freshwater submersed macrophytes systems. Ecological Engineering 2: 1-30. Gundersen, D. T., S. Bustaman, W. K. Seim and L. R. Curtis. 1994. pH, hardness, and humic acid influence aluminum toxicity to rainbow trout ( Oncorhynchus mykiss) in weakly alkaline waters. Canadian Jour nal of Fisheries and Aquatic Science 51: 1345-1355. Gurevitch, J. and S. T. Chester, Jr. 1986. An alysis of repeated measures experiments. Ecology 67: 251-255. Gutknecht JLM, Goodman RM, Balser TC. 2006. Linking soil process and microbial ecology in freshwater wetland ecosystems. Plant and Soil 289: 17-34.

PAGE 240

221 Hadad, H. R., M. A. Maine, and C. A. Bonetto. 2006. Macrophyte growth in a pilotscale constructed wetland for industrial wastewater treatment. Chemosphere 63: 1744-1753. Hanson, A. T. and J. L. Cleasby. 1990. The effects of temperature on turbulent flocculation: fluid dynamics and chemistr y. Journal of American Water Works Association 82(11): 56-73. Hatcher, D., J. Eaton, M. Gibson, and R. L eah. 1999. Methodologies for surveying plant communities in artificial channels. Hydrobiologia 415: 87-91. Hedley, M. J. and J. W. B. Stewart. 1982. Method to measure microbial phosphorus in soils. Soil Biology and Biochemistry 14: 319-329. Hieltjes, A. H. M. and L. Lijklema. 1980. Fractionation of inorganic phosphates in calcareous sediments. Journal of Environmental Quality 9: 405-407. Henrichs, S. M. and W. S. Reeburgh. 1987. Anaerobic mineralization of marine sediment organic matter: Rates and the ro le of anaerobic proce sses in the oceanic carbon economy. Geomicrobi ology Journal 5: 191-237. Hoge, V. 2003. The use of liquid alum and alum residual for nutrient control within wetland restoration projects. In Proceedings of the 12th Annual North American Lake Management Society Southeastern La kes Management Conference. June 2-5, 2003: Orlando, FL. Hsu, P. H. 1989. Aluminum hydroxides and oxyhydroxides. In Dixon, J. B. and S. B. Weed (eds.) Minerals in soil environments. 2nd ed. SSSA Book Series 1. SSSA: Madison, WI. pp. 331-378. Hsu, P. H. 1975. Precipitation of phosphate fr om solution using aluminum salt. Water Research 9: 1155-1161. Huang, J. W., D. L. Grunes, and L. V. Kochian, 1992. Aluminum effects on the kinetics of calcium uptake into cells of th e root wheat apex. Planta 188: 414-421. Huang, P. M., M. K. Wang, N. Kmpf, and D. G. Schulze. 2002. Aluminum hydroxides. In Dixon, J. B. and D. G. Schulze (eds.) Soil Mineralogy with Environmental Applications. Soil Science Society of America, Inc.: Madison, WI. pp. 261-289. Hunter, R. G., D. L. Combs, and D. B. George. 2001. Nitrogen, phosphorus, and organic carbon removal in simulated wetland treatment systems. Archives of Environmental Contamination and Toxicology 41: 274-281. Hupfer, M., R. Gchter, and R. Giovanoli. 1995. Transformation of phosphorus species in settling seston and during early sediment diagenesis. Aquatic Sciences 57: 305324.

PAGE 241

222 Hutchinson, G. E. 1945. Aluminum in soils, plants, and animals. Soil Science 60: 2940. Illmer, P., K. Marschall, and F. Schinner. 1995. Influence of available aluminum on soil-microorganisms. Letters in Applied Microbiology 21: 393-397. Ivanoff, D. B., K. R. Reddy, and S. Robins on. 1998. Chemical fractionation of organic phosphorus in selected Histos ols. Soil Science 163: 36-45. Jackson, L. J. 1998. Paradigms of metal accumu lation in rooted aquatic vascular plants. Science of the Total Environment 219: 223-231. Jackson, L.J., J. Kalff, and J.B. Rasmu ssen. 1993. Sediment pH and redox potential affect the bioavailability of Al, Cu, Fe, Mn, and Zn to rooted aquatic macrophytes. Canadian Journal of Fisheries and Aquatic Sciences. 50: 143-148. Jacoby, J. M., H. L. Gibbons, K. B. Stoops, and D. D. Bouchard. 1994. Response of a shallow, polymictic lake to buffered alum treatment. Lake and Reservoir Management 10: 103-112. James, W. F. and J. W. Barko. 2003. Alum dosage determination based on redoxsensitive sediment phosphorus concentra tions. Water Quality Technical Notes Collection (ERDC WQTN-PD-13). U.S. Army Engineer Research and Development Center: Vicksburg, MS. 9 p. Jamieson, T. S., G. W. Stratton, R. Gordon, and A. Madani. 2003. The use of aeration to enhance ammonia nitrogen removal in cons tructed wetlands. Ca nadian Biosystems Engineering 45: 1.9-1.14. Jansen, S, M. R. Broadley, E. Robbrecht, and E. Smets. 2002. Aluminum hyperaccumulation in angiosperms: A review of its phylogenetic significance. The Botanical Review 68: 235-269. Jansson, M. 1981. Induction of high phosphatase activity by aluminum in acid lakes. Archiv fur Hydrobiologie 93: 32-44. Jarvis, S. C. and D. J. Hatch. 1986. The eff ects of low concentrations of aluminum on the growth and uptake of nitrate-N by Wh ite Clover. Plant and Soil 95: 43-55. Jenkinson, D. S. and J. N. Ladd. 1981. Micr obial biomass in soil: Measurement and turnover. In E. A. Paul and J. N. Ladd (eds.) Soil Biochemistry, Volume 5. Marcel Dekker: New York. pp. 415-471. Jiang, J. Q. and N. J. D. Graham. 1998. Pre-polymerised inor ganic coagulants and phosphorus removal by coagulation – A review. Water SA 24: 237-244. Jones, Jr., J. B., B. Wolf, and H. A. Mills. 1991. Preparation and Analysis. In Plant Analysis Handbook. Micro-Macro Publishing, Inc.: Athens, GA. pp. 23-26.

PAGE 242

223 Kadlec, R. H. and R. L. Knight. 1996. Tr eatment Wetlands. Lewis Publishers: Boca Raton. 893 p. Kadlec, R. H., J. Pries, and H. Mustard. 2007. Muskrats ( Ondatra zibethicus) in treatment wetlands. Ecologi cal Engineering 29: 143-153. Kaggwa, R. C., C. I. Mulalelo, P. Denny, a nd T. O. Okurut. 2001. The impact of alum discharges on a natural tropical wetland in Uganda. Water Research 35: 795-807. Kmpf, N., A. C. Scheinost, and D. G. Schulze. 2000. Oxide minerals. In Sumner, M. E. (ed.) Handbook of soil science. CRC Press: Boca Raton, FL. pp. F125-F168. Kane, D. A. and C. F. Rabeni. 1987. Effects of aluminum and pH on the early life stages of smallmouth bass ( Micropterus dolomieui ). Water Research 21: 633-639. Karathanasis, A. D. and W. G. Harris. 1994. Quantitative thermal analysis of soil materials. In Quantitative Methods in Soil Mine ralogy. Soil Science Society of America: Madison, WI. pp. 360-411. Kennedy, R. H. 1978. Nutrient inactivation with aluminum sulfate as a lake reclamation technique. PhD Dissertation, Ke nt State University, Kent, OH. Kennedy, R. H. and G. D. Cooke. 1982. C ontrol of lake phosphorus with aluminum sulfate: Dose determination and applicat ion techniques. Water Resources Bulletin 18: 389-395. Kennedy, R. H., J. W. Barko, W. F. James, W. D. Taylor, and G. L. Godshalk. 1987. Aluminum sulphate treatment of a eutr ophic reservoir: Rationale, application methods, and preliminary results. La ke and Reservoir Management 3: 85-90. Killingbeck, K. T. 1996. Nutrients in senesced leaves: Keys to the search for potential resorption and resorption prof iciency. Ecology 77: 1716-1727. Kinraide, T. B. 1990. Assessing the rh izotoxicity of the aluminate ion, AlOH4 -. Plant Physiology 93:1620-1625. Kinraide, T. B. 1991. Identity of the rhiz otoxic aluminum species. Plant and Soil 134: 167-178. Kirkman, H. 1996. Baseline and monitoring me thods for seagrass meadows. Journal of Environmental Management 47: 191-201. Kunc, F. 1994. Methods for the analysis of soil microbial communities. In Ritz, K., J. Dighton, and K. E. Giller (eds.) Beyond the Biomass. John Wiley & Sons: Chichester, United Ki ngdom. pp. 21-28. Knight, R. L. 1997. Wildlife habitat and public use benefits of treatment wetlands. Water Science & Technology 35: 35-43.

PAGE 243

224 Knight, R. L., R. W. Ruble, R. H. Kadlec, and S. Reed. 1993. Wetlands for wastewater treatment performance data. In Moshri, G. A. (ed.) Cons tructed wetlands for water quality improvement. Lewis Publishing: New York. pp. 35-58. Knocke, W. R., S. West, and R. C. Hoe hn. 1986. Effects of low temperature on the removal of trihalomethane precursors by co agulation. Journal of American Water Works Association 78(4): 189-195. Kochian, L. V. 1995. Cellular mechanisms of aluminum toxicity and resistance in plants. Annual Review of Plant Phys iology and Plant Molecular Biology 46: 237260. Kodama, H. and G. J. Ross. 1991. Iron dissolution methods used to remove and characterize inorganic compounds in soils . Soil Science Society of America Journal 55: 1180-1187. Kodama, H. and M. Schnitzer. 1980. Effect of fulvic acid on th e crystallization of aluminum hydroxides. Geoderma 24: 195-205. Koepple, C., Popovic, M., and R. K. Bajpai . 1997. Microbial migration in soil. Influence of sodiumhexametaphosphate. Annals of the New York Academy of Sciences 829: 250-262. Koether, M. C., J. E. Deutschman, and G. W. vanLoon. 1993a. A bench-scale evaluation of the performance of a polynuc lear aluminum coagulent. In Robertson, W., R. Tobin, and K. Kjartanson (eds.) Disi nfection Dilemma: Microbiological Control Versus By-products. American Wwater Works Association: Denver. pp. 303-319. Koether, M. C., J. E. Deutschman, and G. W. vanLoon. 1993b. Chemical characterization of a polynuclear aluminum coagulent. Inte rnational Symposium on Chemistry and Biology of Municipal Water Treatment: Current Status and Future Directions. Burlington, ON. Lakatos, G., M. K. Kiss, M. Kiss and P. Juhsz. 1997. Application of constructed wetlands for wastewater treatment in H ungary. Water Science & Technology 35: 331-336. Lamb, D. S. and G. C. Bailey. 1981. Acute and chronic effects of alum to midge larva (Diptera: Chironomidae). Bulletin of Environmental Contamination and Toxicology 27: 59-67. Lee, G. F., W. C. Sonzogni, and R. D. Spear. 1977. Significance of oxic vs anoxic conditions for Lake Mendota sediment phosphorus release. In Golterman, H. L. (ed.) Interactions Between Sediments a nd Fresh Water. Proceedings of an International Symposium held at Amsterda m, Netherlands, Sept. 6-10. Dr. W. Junk B.V. Publishers: Hague. pp. 294-306.

PAGE 244

225 Lehtonen, J. 1989. Effects of acidification on the metal levels in aquatic macrophytes in Espoo, S Finland. Annales Botanici Fennici 26: 39-50. Lewandowski, J., I. Schauser, and M. Hupf er. 2003. Long term effects of phosphorus precipitations with alum in hypereutrophic Lake Ssser See (Germany). Water Research 37: 3194-3204. Lijklema, L. 1980. Interaction of or thophosphate with ion (III) and aluminum hydroxides. Environmental Sc ience and Technology 14: 537-541. Lim, C.H., and M.L. Jackson. 1982. Disso lution for total elemental analysis. In Page, A. L. et al. (eds.) Methods of soil analysis. Part 2. 2nd ed. Agron. Monogr. 9, ASA and SSSA: Madison, WI. pp. 1. Lind, C. B. 2003. Alum chemistry, storage an d handling in lake trea tment applications. In Proceedings of the 12th Annual North American Lake Management Society Southeastern Lakes Management Confer ence. June 2-5, 2003: Orlando, FL. Luderitz, V. and F. Gerlach. 2002. Phosphorus removal in different constructed wetlands. Acta Biotechnologica 22: 91-99. Ma, G., P. Rengasamy, and A. J. Rathjen. 2003. Phytotoxicity of aluminium to wheat plants in high-pH solutions. Australian Journal of Experimental Agriculture 43: 497-501. Madsen, J. D. 1993. Biomass techniques for monitoring and assessi ng control of aquatic vegetation. Lake and Rese rvoir Management 7: 141-154. Madsen, J. D. and J. A. Bloomfield. 1993. Aquatic vegetation quantification symposium: An overview. Lake and Reservoir Management 7: 137-140. Madsen, J. D. and M. S. Adams. 1989. The distribution of submerged aquatic macrophyte biomass in a eutrophic stream, Badfish Creek: the effect of environment. Hydrobiology 171: 111-119. Madsen, J. D., M. S. Adams, and P. Ruffier. 1988. Harvest as a control for sago pondweed (Potamogeton pectinatus L.) in Badfish Creek, Wisconsin: frequency, efficiency and its impact on the stream community oxygen metabolism. Journal of Aquatic Plant Management 26: 20-25. Madsen, T. V. and N. Cedergreen. 2002. S ources of nutrients to rooted submerged macrophytes growing in a nutrient-rich stream. Freshwat er Biology 47: 283-291. Mahaney WC, Hancock, RGV, Sanmugadas K (1991) Extractable Fe-Al and geochemistry of late Pleistocene Paleos ol in the Dalijia Shan, Western China. Journal of Southeast Asia n Earth Science 6: 75-82.

PAGE 245

226 Makris, K.C, W.G. Harris, G.A. O’ Connor, and T.A. Obreza. 2004. Phosphorus immobilization in micropores of drinkingwater treatment residuals: Implications for long-term stability. Envi ron. Sci. Tech. 38:6590-6596. Malecki, L. M, J. R. White, and K. R. Reddy. 2004. Nitrogen and phosphorus flux rates from sediment in the Lower St. Johns Ri ver Estuary. Journal of Environmental Quality 33:1545-1555. Malecki-Brown, L. M. and J. R. White. 2007a. Phosphorus sequest ration in aluminum amended soils from a municipal wast ewater treatment wetland. Soil Science Society of America Journal (in review). Malecki-Brown, L. M., and J. R. White. 2007b. Alum application: effects on water quality and aquatic macrophytes in a m unicipal wastewater treatment wetland. Science of the Total Environment (in prep). Malecki-Brown, L. M., J. R. White, and K. R. Reddy. 2007a. Soil biogeochemical characteristics influenced by alum applica tion in a municipal wastewater treatment wetland. Journal of Environmental Quality (in review). Malecki-Brown, L. M., J. R. White, and M. S ees. 2007b. Spatial and temporal effects of continuous low-dosage alum on soil charac teristics in a municipal wastewater treatment wetland. Environmental Science and Technology (in prep). Mander, . and T. Mauring. 1997. Construc ted wetlands for wastewater treatment in Estonia. Water Scien ce & Technology 35: 323-330. Marschner, H. 1995. Functions of Mineral Nutrients: Micronutrients. In Mineral Nutrition of Higher Plants. 2nd Edition. Academic Press: London. pp. 313-404. Martinez, C. J. and W. R. Wise. 2003. Hydraulic analysis of Orlando Easterly Wetland. Journal of Environmental Engineering 129: 553-560. Matsui, Y., A. Yuasa, Y. Furuya, T. Kamei. 1998. Dynamic analysis of coagulation with alum and PACl. Journal of American Water Works Association 98(10): 96-106. Matsumoto, H. 2000. Cell biology of aluminum toxicity and tolerance in higher plants. International Review of Cytology 200: 1-46. May, H. M., P. A. Helmke, and M. L. Jackson. 1979. Gibbsite solubility and thermodynamic properties of hydroxyaluminum i ons in aqueous solutions at 25 C. Geochimica Cosmochimica Acta 43: 861-868. McDonnell, A. J. and S. D. Hall. 1969. E ffects of environmental factors on benthal oxygen uptake. Journal of Wate r Pollution Control 41: 353-363.

PAGE 246

227 McKeague, J. A., J. E. Brydon, and N. M. Miles. 1971. Differentiation of forms of extractable iron and aluminum in soils. Soil Science Society of America Proceedings 35: 33-38. McKeague, J. A. and J. H. Day. 1966. Dithi oniteand oxalate-extract able Fe and Al as aids in differentiating various classes of so ils. Canadian Journal of Soil Science 46: 13-22. Microsoft. 2000. Microsoft Excel for Window s. Microsoft Corporation, Redmond WA. Miner, C. L. 2001. Storage and partitioning of soil phosphorus in the Orlando Easterly Wetland treatment system. MS Thesis, University of Florida, Gainesville, FL. Minitab. 2000. Minitab statis tical software. Version 13.32. Minitab, State College, PA. Moffett, M. R. 1979. Changes in the micr ocrustacean communities of East and West Twin Lakes, Ohio, following lake restora tion. MS Thesis, Kent State University, Kent, OH. Morris, J. K. and W. R. Knocke. 1984. Te mperature effects on the use of metal-ion coagulants for water treatment. Journa l of American Water Works Association 76(3): 74-79. Mudroch, A. and J. Capobianco. 1978. Study of selected metals in marshes on Lake St. Clair, Ontario. Archiv fr Hydrobiologie 84: 87-108. Narf, R. P. 1990. Interactions of Ch ironomidae and Chaoboridae (Diptera) with aluminum sulfate treated lake sediment. Lake and Reservoir Management 6: 3342. Neubauer, S. C., K. Givler, S. K. Valen tine, and J. P. Megonigal. 2005. Seasonal patterns and plant-mediated controls of subsurface wetland biogeochemistry. Ecology 86: 3334-3344. Neville, C. M. 1985. Physiological response of juvenile rainbow trout, Salmo gairdneri , to acid and aluminum – prediction of fi eld responses from laboratory data. Canadian Journal of Fisherie s and Aquatic Sciences 42:2004-2019. Newman, S. and K. Pietro. 2001. Phosphor us storage and release in response to flooding: implications for Everglades st ormwater treatment areas. Ecological Engineering 18: 23-28. Nguyen, L. M. 2000. Phosphate incorporati on and transformation in surface sediments of a sewage-impacted wetland as influenced by sediment sites, sediment pH and added phosphate concentrations. Ecological Engineering 14: 139-155. Olila, O. G. and K. R. Reddy. 1997. Influence of redox potential on phosphate uptake by sediments in two sub-tropical eutrop hic lakes. Hydrobiologia 345: 45-57.

PAGE 247

228 Omoike, A. I. And G. W. Vanloon. 1999. Removal of phosphorus and organic matter removal by alum during wastewater treatment. Water Research 33: 3617-3627. Outridge, P. M. and B. N. Noller. 19 91. Accumulation of toxic trace elements by freshwater vascular plants. Reviews of Environmental Contamination and Toxicology. 121: 1-63. Pant, H. K. and K. R. Reddy. 2001. Hydrol ogic influence on stability of organic phosphorus in wetland detritus. Journal of Environmental Quality 30: 668-674. Parfitt, R. L. and C. W. Childs. 1988. Estimation of forms of Fe and Al: A review and analysis of contrasting so ils by dissolution and Moessbauer methods. Australian Journal of Soil Research 26: 121-144. PBS&J (Post, Buckley, Schuh & Jernigan, Inc.). 1997. 1996 Monitoring Report. PBS&J, Winter Park, FL. Peltier, W. H. and E. B. Welch. 1970. Factor s affecting growth of rooted plants in a reservoir. Weed Science 18: 7-9. Peltier, W. H. and E. B. Welch. 1969. Factors affecting growth of rooted aquatics in a river. Weed Science 17: 412-416. Pettecrew, E. L. and J. Kalff. 1992. Wate r flow and clay retention in submerged macrophytes beds. Canadian Journal of Fisheries and Aquatic Science 49: 24832489. Pina, R. G. and C. Cervantes. 1996. Microbia l interactions with aluminium. Biometals 9: 311-316. Pompo, M. L., R. Henry, and V. Moschini -Carlos. 1999. Chemical composition of tropical macrophyte Echinochloa polystachya (H.B.K.) Hitchcock in Jurumirim Reservoir (So Paulo, Brazil). Hydrobiologia 411: 1-11. Powlson, D. S. 1994. Methods for the analysis of soil microbial communities. In Ritz, K., J. Dighton, and K. E. Giller (eds.) Beyond the Biomass. John Wiley & Sons: Chichester, United Kingdom. pp. 3-20. Powlson, D. S. and D. S. Jenkinson. 1981. A comparison of the organic matter, biomass, adenosine triphosphate and mineralizable nitrogen contents of ploughed and directdrilled soils. Journal of Ag ricultural Science 97: 713-721. Psenner, R., B. Bostrom, M. Dinka, K. Pettersson, R. Puckso, and M. Sager. 1988. Fractionation of phosphorus in suspended matter and sediment. Archiv fur Hydrobiologie Beiheft 30: 98-111.

PAGE 248

229 Psenner, R. and R. Pucsko. 1988. Phosphorus fractionation: advantages and limits of the method for the study of sediment p origins and interactions. Archiv fur Hydrobiologie Beiheft 30: 43-59. Psenner, R., R. Pucsko, and M. Sager. 1984. Fractionation of or ganic and inorganic phosphorus compounds in lake sediment. An attempt to characterize ecologically important fractions. Archiv fur Hydrobiologie Supplement 70: 111-155. Pyrzy ska, K., S. Guer, and E. Bulska. 2000. Flow-injection speciation of aluminium. Water Research 34: 359-365. Rai, U. N., S. Sinha, R. D. Tripathi, and P. Chandra. 1995. Wast ewater treatability potential of some aquatic macrophytes: Removal of heavy metals. Ecological Engineering. 5: 5-12. Rajan, S. S. S., K. W. Perrott, and W. H. M. Saunders. 1974. Identification of phosphatereactive sites of hydrous alumina from proton consumption during phosphate adsorption at constant pH values. The Journal of Soil Science 38: 211-217. Ramamoorthy, S. 1988. Effect of pH on speciat ion and toxicity of aluminum to rainbow trout ( Salmo gairdneri ). Canadian Journal of Fi sheries and Aquatic Sciences 46:634-642. Ratnaweera, H., J. Fetting, and H. Odegoard. 1992. Particle and phosphate removal mechanisms with prepolymerized coagulants. In Proceedings of the 5th International Gothenberg Symposium on Ch emical Treatment. Springer-Verlag: New York. pp. 3-18. Reagin, M. 2002. Nature’s filter: construc ted wetlands. American City & County 117: 20-26. Reckhow, K. H. and S. C. Chapra. 1999. Modeling excessive nutrient loading in the environment. Environmental Pollution 100: 197-207. Reddy, K. R., E. M. D’Angelo, and T. A. Debusk. 1989. Oxygen transport through aquatic macrophytes: The role in wastewater treatment. Journal of Environmental Quality 19: 261-267. Reddy, K. R., O. A. Diaz, L. J. Scinto, and M. Agami. 1995. Phosphorus dynamics in selected wetlands and streams of the lake Okeechobee Basin. Ecological Engineering 5: 183-207. Reddy, K. R., Y. Wang, W. F. Debusk, M. M. Fisher, and S. Newman. 1998. Forms of soil phosphorus in selected hydrologic un its of the Florida Everglades. Soil Science Society of America Journal 62: 1134-1147.

PAGE 249

230 Reddy, K. R., R. H. Kadlec, E. Flaig, and P. M. Gale. 1999a. Phosphorus retention in streams and wetlands: A review. Critical Reviews in Environmental Science and Technology 29: 83-146. Reddy, K. R., J. R. White, A. Wright, a nd T. Chua. 1999b. Influence of phosphorus loading on microbial processes in so il and water column of wetlands. In Reddy, K. R., G.A. O’Connor, and C.L. Schelske (eds.) Phosphorus Biogeochemistry in Subtropical Ecosystems: Florida as a Case Example. CRC/Lewis Publishing: Boca Raton, FL. pp. 249-273. Rengel, Z. and W. H. Zhang. 2003. Role of dynamics of intracellular calcium in aluminium-toxicity syndrome. New Phytologi st 159: 295-314. Rengel, Z. 1992. Role of calcium in al uminum toxicity. New Phytology 121: 499-513. Robert, M. 1995. Aluminum toxicity: A majo r stress for microbes in the environment. In P. M. Huang et al. (eds.) Environmental impact of soil component interactions. Vol. II. Metals, other inorganics, and microbial activities. CRC Press/Lewis Publishing: Boca Raton, FL. pp.227-247. Rufyikiri, G., J. E. Dufey, R. Achard, and B. Delvaux. 2002. Cation exchange capacity and aluminum–calcium–magnesium binding in roots of bananas cultivated in soils and in nutrient solutions. Communications in Soil Science and Plant Analysis 33: 991-1009. Rydin, E. and E. B. Welch. 1999. Dosing alum to Wisconsin Lake sediments based on in vitro formation of aluminum bound phosphate . Journal of Lake and Reservoir Management 15: 324-331. Rydin, E. and E. B. Welch. 1998. Aluminum dose required to inactivate phosphate in lake sediments. Water Research 32: 2969-2976. Rysgaard, S. N., N. P. Sloth, K. Jensen, and L. P. Nielsen. 1994. Oxygen regulation of nitrification and denitrif ication in sediments. Limnology and Oceanography 39: 1643-1652. Sand-Jensen, K., C. Prahl, and H. Stokholm. 1982. Oxygen release from roots of submerged aquatic macrophytes. Oikos 38: 349-354. Schier, G. A. 1985. Response of red spruce and balsam fir seedlings to aluminum toxicity in nutrient solutions. Canadi an Journal of Forest Research 15:29-33. Schoenau, J. J. and R. E. Karamanos. 1993. Sodium bicarbonate-extractable P, K, and In (Carter, M. R., Ed.) Soil Sampling and Methods of Analysis. Lewis Publishers: Boca Raton. pp. 51-58.

PAGE 250

231 Schreijer, M., R. Kampf, S. Toet, and J. Verhoeven. 1997. The use of constructed wetlands to upgrade treated sewage efflue nts before discharge to natural surface water in Texel Island, the Netherlands – pilot study. Water Science & Technology 35: 231-237. Sees, M., Turner, D., 1997. Orlando Easterly Wetland; Management of a constructed wetland treatment system. In Proceedings of the 6th Annual North American Lake Management Society Southeastern Lake s Management Conference. May 7-9, 1997. Palm Beach Shores, FL. Seitzinger, S. P. 1988. Denitrification in freshwater and coastal marine ecosystems: Ecological and geochemical significance. Limnol ogy and Oceanography 33:702724. Shen, Y. H. and B. A. Dempsey. 1998. Synthesis and speciation of polyaluminum chloride for water treatment. Environment International 24: 899-910. Shutes, R. B. E., D. M. Revitt, A. S. Mungur, and L. N. L. Scholes. 1997. The design of wetland systems for the treatment of urba n run off. Water Science & Technology 35: 19-25. Silva, I. R., T. J. Smyth, D. W. Israel, C. D. Raper, and T. W. Rufty. 2001. Magnesium ameliorates aluminum rhizotoxicity in s oybean by increasing citric acid production and exudation by roots. Plan t Cell Physiology 42: 546-554. Silveira, M. L., M. K. Miyittah, and G. A. O'Connor. 2006. Phos phorus Release from a Manure-Impacted Spodosol: E ffects of a Water Treatment Residual. Journal of Environmental Quality 35: 529-541. Simon, S. M. 2003. Phosphorus retention and release of soils in a constructed wetland for wastewater treatment. M.S. thesis, University of Florid a, Gainesville. 107 p. Singer, A. and Huang, P. M. 1990. Effects of humic acid on the crystallization of aluminum hydroxides. Clays and Clay Mineralogy 38: 47-52. Small, M. and C. Wurm. 1977. Data Report. Meadow/Marsh/Pond System. Brookhaven National Laboratory, BNL 50675. Smeltzer, E., and R. A. Kirn, and S. Fiske. 1999. Long – term water quality and biological effects of alum treatment of La ke Morey, Vermont. Journal of Lake and Reservoir Management 15: 173-184. Smeltzer, E. 1990. A successful alum/alumi nate treatment of Lake Morey, Vermont. Lake and Reservoir Management 6: 9-19. Smith, S. M. and S. Newman. 2001. Growth of southern cattail ( Typha domingensis pers.) seedlings in response to fire-rela ted soil transformations in the northern Florida Everglades. Wetlands 21: 363-369.

PAGE 251

232 Smith, J. S. and E. A. Paul. 1990. The significance of soil microbial biomass estimations. In Bollag, J. M. and G. Stotzky (eds.) Soil Biochemistry. Vol. 6. Marcel Dekker: New York. Sparling, D. W. and T. P. Lowe. 1998. Meta l concentrations in aquatic macrophytes as influenced by soil and acidification. Wa ter, Air, and Soil Pollution 108: 203-221. Sprenger, M. and A. McIntosh. 1989. Relationship between concentrations of aluminum, cadmium, lead, and zinc in water, sediments, and aquatic macrophytes in six acidic lakes. Archives of Environmental C ontamination and Toxicology 18: 225-231. Srinivasan, P. T. and T. Viraraghavan. 2002. Characterisation and concentration profile of aluminum during drinking-water treatment. Water SA 28: 99-106. Steinman, A., R. Rediske, and K. R. Reddy. 2004. The reduction of internal phosphorus loading using alum in Spring Lake, Michig an. Journal of Environmental Quality 33: 2040-2048. Sterner, R. W. and J. J. Elser. 2002. The stoichiometry of autotroph growth. In Ecological Stoichiometry. Princeton Univ ersity Press: Princeton, NJ. pp. 80-134. Stevens, C. E., A. W. Diamond, and T. S. Gabor. 2002. Anuran call surveys on small wetlands in Prince Edward Island, Canada restored by dredging of sediments. Wetlands 22: 90-99. Tangen, G., T. Wickstrom, S. Lierhagen, R. Vogt, and W. Lund. 2002. Fractionation and determination of aluminum and iron in soil water samples using SPE cartridges and ICP-AES. Environmental Science and Technology 36: 5421-5425. Tanner, C. C. 1996. Plants for constructed wetland treatment systems – A comparison of the growth and nutrient uptake of eight emergent species. Ecological Engineering 7: 59-83. Taylor, G. J. 1989. Aluminum toxi city and tolerance in plants. In Adriano, D. C. and A. H. Johnson (eds.) Advances in Environmenta l Science. Acidic Precipitation Vol. 2. Springer-Verlag: New York. pp. 327-361. Taylor, G. J. and A. A. Crowder. 1984. Copper and nickel tolerance in Typha latifolia clones from contaminated and uncontaminated environments. Canadian Journal of Botany 62:1304-1308. Templeton, D. M., F. Ariese, R. Cornelis, L. Danielsson, H. Muntau, H. P. Van Leeuwen, and R. Lobinski. 2000. Guidelines for terms related to chemical speciation and fractionation of elements – Definitions, structural aspects, and methodological approaches (IUPAC recommendations 2000) . Pure and Applied Chemistry 72: 1453-1470.

PAGE 252

233 Tettenhorst, R. and A. Hoffmann. 1980. Crys tal chemistry of boehmite. Clays and Clay Minerals 28: 373-380. Thiebaut, G. and S. Muller. 2000. Linking phosphorus pools of water, sediment and macrophytes in running waters. Annales de Limnologie – Intern ational Journal of Limnology 39: 307-316. Thornton, F.C., M. Schaedle, and D. J. Raynal. 1987. Effects of aluminum on red spruce seedlings in solution culture. Environmental and Experimental Botany 27: 489– 498. Topp, G. C. 1993. Soil water content. In Carter, M. R. (ed.) Soil Sampling and Methods of Analysis. Lewis Publishers: Boca Raton. pp. 541-557. Upchurch, J. B., J. K. Edzwald, and C. R. O’Melia. 1974. Phosphates in sediments of Pamlico Estuary. Environmental Science and Technology 8: 56-58. USEPA 600/R-93/100 (U.S. Environmental Pr otection Agency). 1993. Methods of Chemical Analysis of Water and Wast es. Environmental Monitoring Support Laboratory: Cincinnati, OH. Van Benschoten, J. E. and J. K. Edzwald. 1990a. Chemical aspects of coagulation using aluminum salts – I. Hydrolytic reactions of alum and polyaluminum chloride. Water Research 24: 1519-1526. Van Benschoten, J. E. and J. K. Edzwald. 1990b. Measuring Al dur ing water treatment: Methodology and application. Journal of American Water Works Association 82: 71-78. Van Hullebusch, E. V., F. Auvray, V. Deluchat, P. M. Chazal, and M. Baudu. 2003. Phosphorus fractionation and short-term m obility in the surface sediment of a polymictic shallow lake treated with a low dose of alum (Courtille Lake, France). Water, Air, and Soil Pollution 146: 75-91. Van Hullebusch, E. V., V. Deluchat, P. M. Chazal, and M. Baudu. 2002. Environmental impact of two successive chemical treatme nts in a small shallow eutrophied lake: Part I. Case of aluminum sulphate . Environmental Po llution 120: 617-626. Vazques, M. D., J. Lopez, S. Diaz, and A. Carballeira. 2000. The effect of pH on the kinetics of intraand extracellular uptake of Al in Fontinalis antipyretica: Changes in the cellular contents of K, Mg, and Ca. Archives of Environmental Contamination and Toxicology 39: 74-85. Verhoeven, J. T. A. 1986. Nutrient dynami cs in minerotrophic peat mires. Aquatic Botany 25: 117.

PAGE 253

234 Viola, R. E., J. F. Morrison, and W. W. Cleland. 1980. Interactions of metal (III)adenosine 5’-triposphate complexes with yeast hexokinase. Biochemistry 19: 3131-3137. Viraraghavan, T. and C. H. Wimmer. 1988. Po lyaluminum chloride as an alternative to alum coagulation: a case study. Proceedi ngs of the Canadian Society for Civil Engineering Annual Conference, Calgaray, Alberta. pp.480-498. Vitousek, P. M. 1982. Nutrient cycling and nu trient use efficiency. American Naturalist 119: 553-572. Vymazal, J. 2001. Types of constructed we tlands for wastewat er treatment: Their potential for nutrient removal. In Vymazal, J. (ed.) Transfor mations of Nutrients in Natural and Constructed Wetlands. Backhuys Publishers: Leiden, Netherlands. pp. 1. Wang, H., J. W. Jawitz, J. R. White, C. J. Martinez, and M. Sees. 2006. Rejuvenating the largest municipal treatment wetland in Florida. Ecological Engineering 26: 132-146. Welch, E. B. and G. D. Cooke. 1999. Effectiveness and longevity of phosphorus inactivation with alum. Journal of Lake and Reservoir Management 15: 5-27. Welch, E. B. and G. D. Schrieve. 1994. Al um treatment effectiven ess and longevity in shallow lakes. Hydrobiologia 275/276: 423-431. Welch, E. B., J. P. Michaud, and M. A. Pe rkins. 1982. Alum control of internal phosphorus loading in a shallow lake. Water Research Bulletin 18: 929-936. White, J.R., M.A. Belmont, and C.D. Metcalfe. 2006a. Pharmaceutical compounds in wastewater: wetland treatment as a potenti al solution. Scientific World J. 6:17311736. White, J. R., and K. R. Reddy. 1999. The influence of nitrate and phosphorus loading on denitrification enzymes activity in Evergl ades wetland soils. Soil Science Society of America Journal 63: 1945-1954. White, J. R. and K. R. Reddy. 2000. Influence of phosphorus loading on organic nitrogen mineralization of Everglades soils. Soil Science Society of America Journal 64: 1525-1534. White, J. R. and K. R. Reddy. 2003. Nitrificatio n and denitrification rates of everglades wetland soil along a phosphorus-impacted grad ient. Journal of Environmental Quality 32: 2436-2443. White, J. R., K. R. Reddy, and J. Maje r-Newman. 2006b. Hydrologic and vegetation effects on water column phosphorus in we tland mesocosms. Soil Science Society of America Journal 70: 1242-1251.

PAGE 254

235 White, J. R., K. R. Reddy, and M. Z. Moustafa. 2004. Influence of hydrologic regimeand vegetation on phosphorus retenti on in Everglades stormwater treatment area wetlands. Hydrological Processes 18: 343-355. White, J. S., S. E. Bayley, and P. J. Curtis. 2000. Sediment storage of phosphorus in a northern prairie wetland receiving munici pal and agro-industrial wastewater. Ecological Engineering 14: 127-138. Whitting, L. D. and W. R. Allardice. 1986. X-ray diffraction techniques. In Klute, A. (ed.) Methods of Soil Analysis. Part 1. 2nd ed. Agronomy 9. American Society of Agronomy: Madison, WI. Wickstrm, T., N. Clarke, K. Derome, J. Derome, and E. Rgeberg. 2000. Comparison study of five analytical methods for the fractionation and subsequent determination of aluminium in natural water samples. Journal of Environmental Monitoring 2: 171-181. Wigand C., Stevenson J.C. & Cornwell J.C. 1997. — Effects of different submersed macrophytes on sediment biogeochemistry. Aquatic Botany 56: 233-244. Williams, R. B. and M. B. Murdoch. 1972. Co mpartmental analysis of the production of Juncus roemerianus in a North Carolina salt marsh. Chesapeake Science 13: 6979. Ye, Z. H., A. J. M. Baker, M. H. Wong, and A. J. Willis. 1997. Zinc, lead and cadmium tolerance, uptake and accumulation by Typha latifolia . New Phytologist 136:469480. Zhang, G. and G. J. Taylor. 1988. Effect of aluminum on gr owth and distribution of aluminum in tolerant and sensitive cultivars of Triticum aestivum L. Communications in Soil Science and Plant Analysis 19: 1195-1205. Zhou, J., B. Xia, D. S. Treves, L.-Y. Wu, T. L. Marsh, R. V. O’neill, A. V. Palumbo, and J. M. Tiedje. 2002. Spatial and reso urce factors influenc ing high microbial diversity in soil. Applied and E nvironmental Microbiology 68: 326-334. Zvomuya, F., C.J. Rosen, and S.C. Gupta. 2006. Phosphorus sequestration by chemical amendments to reduce leaching from wastew ater applications. J. Environ. Qual. 35:201-215.

PAGE 255

236 BIOGRAPHICAL SKETCH Lynette Marie Malecki was born in South Be nd, Indiana. She graduated from Saint Joseph’s High School in South Bend with a 4.2 GPA as a Saint Joseph’s Scholar. She attended Saint Mary’s College in Notre Da m e, Indiana for her undergraduate degree. Lynette graduated suma cum laude in May of 1999 with a Bachel or of Science degree in biology, with chemistry and mathematic minors. She also was the recipient of the Sigma Xi Award for Outstanding Research in 1999 for her senior comprehensive research project investigating the effect of acidification on plant ti ssue somatic embryogenesis. In the fall of 1999 she entered the Envir onmental Engineering Department at the University of Florida in the Center for Wetla nds to pursue her Master of Science degree. Here she studied wetland community structur e changes along a nutrient gradient in the Orlando Easterly Wetland, one of the largest and oldest treatment wetlands in the United States. In 2001 she transferre d to the Soil and Water Scien ce Department in the Wetlands Biogeochemistry Laboratory at the University of Florida to continue her pursuit of an M.S. degree. Here she studi ed the spatial and seasonal variability of phosphorus and nitrogen present in sediment throughout the Lower St. Johns River (LSJR) as well as studying the nutrient fluxes from these sedime nts. She graduated with her M.S. in December 2002 and received the IFAS Thesis of the Year award for her research. She received an IFAS Alumni Fellowship to remain at the University of Florida to pursue her doctorate. She returned to the Orlando Easterly Wetland to assess the use of alum as a possible management strategy in sequestering phosphorus and to investigate any impacts

PAGE 256

237 on the vegetation, microbial community, or mineralogy. In fall 2005 she passed her qualifying exams becoming a Ph.D. candidate, a nd a few weeks later married Eric Ahrens Brown, changing her name to Lynette Malecki Br own. She finished her field research in October 2006 and laboratory research in February 2007 while diligently writing, and submitting two manuscripts prior to her defens e. She hopes to obtain an Environmental Scientist position at one of the Water Manageme nt Districts within th e state of Florida.