<%BANNER%>

Coupling Microbial Activity and Nutrient Cycling in Soils of the Everglades Agricultural Area

Permanent Link: http://ufdc.ufl.edu/UFE0041061/00001

Material Information

Title: Coupling Microbial Activity and Nutrient Cycling in Soils of the Everglades Agricultural Area
Physical Description: 1 online resource (179 p.)
Language: english
Creator: Ye, Rongzhong
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: agricultural, biogeochemical, evergaldes, histosol, microbial, nutrient
Soil and Water Science -- Dissertations, Academic -- UF
Genre: Soil and Water Science thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: COUPLING MICROBIAL ACTIVITY AND NUTRIENT CYCLING IN SOILS OF THE EVERGLADES AGRICULTURAL AREA The Everglades Agricultural Area (EAA) is an important agricultural region, but has been implicated in contributing to the deterioration of water quality in the Everglades wetlands primarily through runoff of nutrients. The objective of this research was to investigate impacts of selected agricultural management practices on soil biochemical processes as related to microbial activities, nutrient cycling, and water quality. Long-term cultivation and nutrient and water management of EAA increased organic matter decomposition rates and altered supply of bioavailable nutrients. Evaluation of the major land uses in the EAA indicated that labile phosphorus (P) was the main soil chemical property affected by agricultural management practices. Phosphorus fertilization resulted in 465% higher concentrations of labile P in sugarcane soils compared to uncultivated soils. The uncultivated sites were the least disturbed, but the most P-limited and therefore had the smallest size of microbial population and lowest efficiency of C and N utilization. Statistical analysis revealed that P availability was the primary factor regulating the soil microbial community structure and function for land uses. Increased P availability promoted significant accumulation of P in microbial biomass, which indicates that P fertilization is likely to enhance microbial activity and potentially increase organic matter decomposition and soil subsidence. Future management practices should consider the role of labile P on the function of microbial communities and their control of nutrient cycling. Elemental sulfur (S) application to EAA soils is used as one of management practices to decrease pH and therefore increase the bioavailability of nutrients to crops. Sulfur application at rates up to 448 kg ha-1 did not significantly reduce soil pH due to the high buffering capacity against acidification. However, the highest S rate did increase P concentrations in the Fe-Al bound P fraction by 55% compared to unamended soils at 2 months. The stimulatory effects were limited and did not last beyond 2 months. Higher S application rates may be needed to overcome the soil buffering capacity and increase nutrient availability. However, the risk of P export as runoff should be well recognized since the Ca-bound P pool, which comprised of 32% of the total soil P, is not stable under acidic conditions. Similar to labile P, water-extractable potassium (K) and acetic acid extractable Zinc (Zn) increased at 2 months after S application at the highest S rates by 71% and 134% respectively, but the stimulatory effects did not extend beyond 2 months. Correspondingly, S application at current recommended rates did not increase sugar yield. The failure of S application to enhance nutrient availability throughout the growing season indicates the limited benefit of applying elemental S to reduce pH and increase nutrient availability to sugarcane. Alternatives, such as different P and micronutrient fertilizer application methods, timings, and sources, may be better for increasing nutrient availability for these changing soils. Sulfur did promote short-term changes in soil microbial activities. The activities of phosphatase and glucosidase in soils receiving 448 kg S ha-1 were 115% and 560% higher, respectively, than unamended soils at 2 months. Yet, microbial respiration and potential N and P mineralization rates were not affected by S amendment, suggesting that S application under the current recommendations would not enhance soil subsidence and regeneration rates of N and P. However, mineralized S increased with increasing S application rates and the effects continued throughout the growing season. Averaged across time, mineralizable S was 2, 6, 9, and 27 mg kg-1 d-1 for the increasing S application rates, indicating that S fertilization may pose an environmental hazard to the nearby aquatic ecosystem through potential export of SO4. Therefore, using higher S application rates to produce favorable responses in terms of nutrient availability and sugarcane yield should be well evaluated in terms of both effects on sugarcane yield and adverse environmental effects. Agricultural management practices have significantly altered nutrient distribution and regeneration in EAA soils, especially the P cycling. Phosphorus fertilization is to stimulate microbial activities and soil oxidation rates. Current land use as sugarcane cropping continues to promote soil subsidence. Elemental S application under current recommendation rates introduced temporary and limited effects on increasing nutrient availability to sugarcane and posed minimal impacts on microbial activity and function. However, S application at a large scale may increase the risk of SO4 export from the fields and stimulate nutrient regeneration rates and microbial activity.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Rongzhong Ye.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Wright, Alan.
Local: Co-adviser: Reddy, Konda R.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-04-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041061:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041061/00001

Material Information

Title: Coupling Microbial Activity and Nutrient Cycling in Soils of the Everglades Agricultural Area
Physical Description: 1 online resource (179 p.)
Language: english
Creator: Ye, Rongzhong
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: agricultural, biogeochemical, evergaldes, histosol, microbial, nutrient
Soil and Water Science -- Dissertations, Academic -- UF
Genre: Soil and Water Science thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: COUPLING MICROBIAL ACTIVITY AND NUTRIENT CYCLING IN SOILS OF THE EVERGLADES AGRICULTURAL AREA The Everglades Agricultural Area (EAA) is an important agricultural region, but has been implicated in contributing to the deterioration of water quality in the Everglades wetlands primarily through runoff of nutrients. The objective of this research was to investigate impacts of selected agricultural management practices on soil biochemical processes as related to microbial activities, nutrient cycling, and water quality. Long-term cultivation and nutrient and water management of EAA increased organic matter decomposition rates and altered supply of bioavailable nutrients. Evaluation of the major land uses in the EAA indicated that labile phosphorus (P) was the main soil chemical property affected by agricultural management practices. Phosphorus fertilization resulted in 465% higher concentrations of labile P in sugarcane soils compared to uncultivated soils. The uncultivated sites were the least disturbed, but the most P-limited and therefore had the smallest size of microbial population and lowest efficiency of C and N utilization. Statistical analysis revealed that P availability was the primary factor regulating the soil microbial community structure and function for land uses. Increased P availability promoted significant accumulation of P in microbial biomass, which indicates that P fertilization is likely to enhance microbial activity and potentially increase organic matter decomposition and soil subsidence. Future management practices should consider the role of labile P on the function of microbial communities and their control of nutrient cycling. Elemental sulfur (S) application to EAA soils is used as one of management practices to decrease pH and therefore increase the bioavailability of nutrients to crops. Sulfur application at rates up to 448 kg ha-1 did not significantly reduce soil pH due to the high buffering capacity against acidification. However, the highest S rate did increase P concentrations in the Fe-Al bound P fraction by 55% compared to unamended soils at 2 months. The stimulatory effects were limited and did not last beyond 2 months. Higher S application rates may be needed to overcome the soil buffering capacity and increase nutrient availability. However, the risk of P export as runoff should be well recognized since the Ca-bound P pool, which comprised of 32% of the total soil P, is not stable under acidic conditions. Similar to labile P, water-extractable potassium (K) and acetic acid extractable Zinc (Zn) increased at 2 months after S application at the highest S rates by 71% and 134% respectively, but the stimulatory effects did not extend beyond 2 months. Correspondingly, S application at current recommended rates did not increase sugar yield. The failure of S application to enhance nutrient availability throughout the growing season indicates the limited benefit of applying elemental S to reduce pH and increase nutrient availability to sugarcane. Alternatives, such as different P and micronutrient fertilizer application methods, timings, and sources, may be better for increasing nutrient availability for these changing soils. Sulfur did promote short-term changes in soil microbial activities. The activities of phosphatase and glucosidase in soils receiving 448 kg S ha-1 were 115% and 560% higher, respectively, than unamended soils at 2 months. Yet, microbial respiration and potential N and P mineralization rates were not affected by S amendment, suggesting that S application under the current recommendations would not enhance soil subsidence and regeneration rates of N and P. However, mineralized S increased with increasing S application rates and the effects continued throughout the growing season. Averaged across time, mineralizable S was 2, 6, 9, and 27 mg kg-1 d-1 for the increasing S application rates, indicating that S fertilization may pose an environmental hazard to the nearby aquatic ecosystem through potential export of SO4. Therefore, using higher S application rates to produce favorable responses in terms of nutrient availability and sugarcane yield should be well evaluated in terms of both effects on sugarcane yield and adverse environmental effects. Agricultural management practices have significantly altered nutrient distribution and regeneration in EAA soils, especially the P cycling. Phosphorus fertilization is to stimulate microbial activities and soil oxidation rates. Current land use as sugarcane cropping continues to promote soil subsidence. Elemental S application under current recommendation rates introduced temporary and limited effects on increasing nutrient availability to sugarcane and posed minimal impacts on microbial activity and function. However, S application at a large scale may increase the risk of SO4 export from the fields and stimulate nutrient regeneration rates and microbial activity.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Rongzhong Ye.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Wright, Alan.
Local: Co-adviser: Reddy, Konda R.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-04-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041061:00001


This item has the following downloads:


Full Text

PAGE 1

1 COUPLING MICROBIAL ACTIVITY AND NUTRIENT CYCLING IN SOILS OF THE EVERGLADES AGRICULTURAL AREA By RONGZHONG YE A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE RE QUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 20 1 0

PAGE 2

2 20 1 0 Rongzhong Ye

PAGE 3

3 To those who have faith and love for me

PAGE 4

4 ACKNOWLEDGMENTS To my major advisor, Dr. Alan L. Wright, I would like to express my deepest thanks f or his trust, endless support, and guidance throughout my graduate study. I would also like to express my appreciation to my major co advisor, Dr. K. R amesh Reddy, for his teachings and guidance. I thank my committee members, Dr. Andrew Ogram, Dr. Linda Young, and Dr. Susan Newman for their comments and suggestions regarding the dissertation. I thank Dr. Mabry McCray for his kind help with the field work and review of manuscripts. I thank Dr. Yigang Luo, Ms. Yu Wang, Dr. Kanika S. Inglett, Mr. Gavin Wilson, Ms. Yubao Cao, Mrs. Dawn Lucas, Dr. Abid Al -Agely, and my colleagues in the Wetland Biogeochemistry Laboratory for their assistance and support. My special thanks to my wife, my parents, and my family for their faith, support and love for me.

PAGE 5

5 TABL E OF CONTENTS page ACKNOWLEDGMENTS ...................................................................................................... 4 LIST OF TABLES ................................................................................................................ 9 LIST OF FIGURES ............................................................................................................ 11 ABSTRACT ........................................................................................................................ 13 CHAPTERS 1 INTRODUCTION ........................................................................................................ 16 Everglades Agricultural Area ...................................................................................... 17 Land Use Change ....................................................................................................... 20 Sulfur in the Everglades ............................................................................................. 21 Objectives and Hypothesis ......................................................................................... 24 2 LAND USE EFFECTS ON SOIL NUTRIENT CYCLING AND MICROBIAL COMMUNITY DYNAMICS IN THE EVERGLADES AGRICULTURAL AREA, FLORIDA ..................................................................................................................... 28 Int roduction ................................................................................................................. 28 Materials and Methods ............................................................................................... 30 Site Description .................................................................................................... 30 Soil S ampling and Physical -Chemical Analysis .................................................. 31 Microbial Biomass ................................................................................................ 32 Potentially Mineralizable N and P ........................................................................ 32 Enzyme Activity Assay ......................................................................................... 33 Community -Level Physiological Profile by BIOLOG Assay ................................ 34 Statistica l Analysis ............................................................................................... 35 Results ........................................................................................................................ 35 Soil Physical and Chemical Properties ............................................................... 35 Microbial Biomass ................................................................................................ 36 Extracellular Enzyme Activity .............................................................................. 36 Microbial Community Composition and Function ............................................... 36 Discussion ................................................................................................................... 37 Nutrient Distribution and Cycling ......................................................................... 37 Microbial Community Dynamics .......................................................................... 39 Indicators of Land Use Changes ......................................................................... 44 Conclusions ................................................................................................................ 44

PAGE 6

6 3 MULTIVARIATE ANA LYSIS OF CHEMICAL AND MICROBIAL PROPERTIES IN HISTOSOLS AS INFLUENCED BY LAND -USE TYPES ..................................... 49 Introduction ................................................................................................................. 49 Materials and Methods ............................................................................................... 51 Site Description .................................................................................................... 51 Soil Chemical Properties ..................................................................................... 52 Soil Microbi al Parameters .................................................................................... 53 Microbial biomass .......................................................................................... 53 Potentially mineralizable N and P ................................................................. 54 Enzyme activity assay ................................................................................... 54 Community -level physiological profile by BIOLOG assay ............................ 55 Data Analysis ....................................................................................................... 56 Results ........................................................................................................................ 56 Land-Use Effects on Integrated Soil Chemical Properties ................................. 56 Land-Use Eff ects on Integrated Microbial Properties ......................................... 57 Dependent Relationship between Chemical Properties and Microbial Parameters ........................................................................................................ 58 Discussion ................................................................................................................... 59 Land-Use Effects on Soil Chemical Properties ................................................... 59 Land-Use Effects on Microbial Parameters ........................................................ 61 Variable Reduction ............................................................................................... 62 Dependent Relationship between Chemical Properties and Microbial Parameters ........................................................................................................ 63 Conclusions ................................................................................................................ 64 4 SULFURINDUCED CHANGES IN PHOSPHORUS DISTRIBUTION IN EVERGLADES AGRICULTURAL AREA SOILS ....................................................... 73 Introduction ................................................................................................................. 73 Materials and Methods ............................................................................................... 76 Site Description .................................................................................................... 76 Soil Sampling and Analysis ................................................................................. 76 Statistical Analysis ............................................................................................... 78 Results and Discussion .............................................................................................. 78 Soil pH .................................................................................................................. 78 Phosphorus Distribution ....................................................................................... 79 Labile P .......................................................................................................... 79 Fe -Al bound P ................................................................................................ 80 Ca -bound P .................................................................................................... 82 Humic -fulvic acid P ........................................................................................ 83 Residual P ..................................................................................................... 83 Phosphorus Distribution and Availability in S -Amended Soils ........................... 84 Conclusions ................................................................................................................ 86

PAGE 7

7 5 MICROBIAL ECO -PHYSIOLOGICAL RESPONSE OF A CALCAREOUS HISTOSOL TO SULFUR AMENDMENT ................................................................... 93 Introduction ................................................................................................................. 93 Material and Methods ................................................................................................. 95 Site Description .................................................................................................... 95 Soil Sampling and Analysis ................................................................................. 96 Statisti cal Analysis ............................................................................................... 98 Results ........................................................................................................................ 98 Soil Physical and Chemical Properties ............................................................... 98 Extracellular Enzyme Activities ............................................................................ 99 Microbial Biomass .............................................................................................. 100 Microbial Mediated Mineralization ..................................................................... 100 Discussion ................................................................................................................. 101 Enzymatic Activities ........................................................................................... 101 Microbial Biomass .............................................................................................. 103 Microbial Mediated Organic Matter Mineralization ........................................... 104 Seasonal Fluctuations in Microbial Indices ....................................................... 106 Conclusions .............................................................................................................. 107 6 SEASONAL CHANGES IN NUTRIENT AVAILABILITY IN SULFUR -AMENDED EVERGLADES SOILS UNDER SUGARCANE ....................................................... 116 Introduc tion ............................................................................................................... 116 Material and Methods ............................................................................................... 118 Site Description .................................................................................................. 1 18 Soil Samp ling and Analysis ............................................................................... 119 Sugar Yield ......................................................................................................... 120 Statistical Analysis ............................................................................................. 120 R esults and Discussion ............................................................................................ 121 Soil pH ................................................................................................................ 121 Dissolved Organic C and Extractable N ............................................................ 122 Phosphorus ........................................................................................................ 123 Potassium ........................................................................................................... 124 Calcium ............................................................................................................... 125 Magnesi um ......................................................................................................... 125 Sulfur .................................................................................................................. 126 Copper ................................................................................................................ 126 Iron ...................................................................................................................... 127 Manganese ......................................................................................................... 127 Zinc ..................................................................................................................... 127 Soil Properties and Micronutrient Availability .................................................... 128 Comparison of Soil Extractants ......................................................................... 130 Nutrient Availability and Sugar Yield ................................................................. 131 Con clusions .............................................................................................................. 132

PAGE 8

8 7 SULFUR POOLS, TRANSFORMATIONS, AND MINERALIZATION IN EVERGLADES AGRICULTURAL AREA SOILS ..................................................... 142 Introduction ............................................................................................................... 142 Materi a ls and Methods ............................................................................................. 144 Site Description .................................................................................................. 144 Soil Sampling and Lab oratory Analysis ............................................................. 145 Statistical Analysis ............................................................................................. 147 Results and Discussion ............................................................................................ 147 Soil pH ................................................................................................................ 147 Extractable SO4-S .............................................................................................. 148 Extractable Organic S ........................................................................................ 149 Elemental S ........................................................................................................ 149 Sulfatase Activity ................................................................................................ 150 Organic S Mineralization .................................................................................... 151 Conclusions .............................................................................................................. 151 8 SYNTHESIS .............................................................................................................. 159 Land Use Effects on Soil Nutrient Cycling and Microbial Community Dynamics in the Everglades Agricultural Area, Florida ......................................................... 159 Multivariate Analysis of Chemical and Microbial Properties in Histosols as Influenced by Land-Use Types ............................................................................. 160 Sulfur -Induced Changes in Phosphorus Distribution in Everglades Agricultural Area Soils .............................................................................................................. 160 Microbial Eco -Physiological Response of a Calcareous Histosol to Sulfur Amendment ........................................................................................................... 161 Seasonal Changes in Nutrient Availability in Sulfur -Amended Everglades Soils under Sugarcane ................................................................................................... 162 Overall Conclusions .................................................................................................. 163 LIST OF REFERENCES ................................................................................................. 167 BIOGRAPHICAL SKETCH .............................................................................................. 179

PAGE 9

9 LIST OF TABLES Table page 2 -1 Properties of forest, sugarcane, turf and uncultivated soils .................................. 46 2 -2 Microbial biomass, enzymatic activities, potentially mineralizable N (P MN) and potentially mineralizable P (PMP) of forest, sugarcane, turf and uncultivated soils. ................................................................................................... 46 2 -3 Significant correlations of soil and microbial properties ........................................ 47 3 -1 Chemical properties of cypress, sugarcane, and uncultivated soils. .................... 66 3 -2 The ratio of between -cluster sum of square to total sum of square (R2) and the ra tio of between -cluster sum of square to within-cluster sum of square (R2/(1 -R2)) for chemical variables in defining differences among clusters ........... 66 3 -3 Microbial parameters in cypress, sugarcane, and uncultivated soils ................... 67 3 -4 Stepwise discriminant model for differentiating cypress, sugarcane, and uncultivated soils based on microbial parameters ................................................ 67 4 -1 Chemical properties of the Dania soil in the Everglades Agricultural Area before fertilizer application ..................................................................................... 87 4 -2 Two way ANOVA on different pools of P in soils am ended with variable S application rates. .................................................................................................... 87 4 -3 Pearson correlations coefficients for P fractions and soil properties. ................... 88 5 -1 Ex tractable nutrients (mg kg1) and pH in soil amended with elemental S during the sugarcane growing season ................................................................ 108 5 -2 Potentially mineralizable C (Cmin), N (Nmin), P (Pmin), and microbial biomass C (MBC), N (MBN), P (MBP) in soil amended with elemental S during the sugarcane growing season. ............................................................... 109 5 -3 Significant correlation coefficients between selected chemical properties and microbial functional activities. .............................................................................. 110 5 -4 Microbial metabolic coefficient (qCO2), microbial biomass C to organic matter content ratio (MBC/OM), and potential N and P mineralization quotient (qPMN and qPM P) in soils amended with elemental S duri ng the sugarcane growing season .................................................................................................... 111 6 -1 Comparisons of soil test extractants on concentrations of available nutrients .. 134

PAGE 10

10 6 -2 Pearson correlation coefficients (r) between pH, organic matter content, and concentrations of acetic acidextractable nutrients ............................................. 134 6 -3 Pearson correlat ion coefficients (r) between soil pH, organic matter content, and concentrations of Mehlich-3 extractable nutrients ....................................... 135 6 -4 Pearson correlation coefficients (r) between soil pH, organic matter content, and concentrations of water extractable nutrients .............................................. 135 6 -5 Multiple regression models relating soil nutrient concentrations (mg kg1) with sugar yield (Mg ha1) at different times. ............................................................... 135 7 -1 Chemical properties of the Histosols in the Everglades Agricultural Area before fertilizer application. .................................................................................. 153

PAGE 11

11 LIST OF FIGURES Figure page 1 -1 Map of the Everglades Agri cultural Area in south Florida ................................... 26 1 -2 Conceptual scheme showing selected land managements that potenti ally affect biogeochemical processes in EAA soils. ..................................................... 27 2 -1 Microbial community metabolic diversities of forest, sugarcane, turf and uncultivated soils. ................................................................................................... 48 2 -2 Principal component analysis of community level physiological profiles from forest, sugarcane, turf and uncultivated soils. ....................................................... 48 3 -1 Dendrogram from K means cluster method applied to soil chemical data. ......... 68 3 -2 Cluster means across chemical variables ............................................................. 69 3 -3 Score plots of principal components analysis on soil chemical properties and microbial parameters. ............................................................................................. 70 3 -4 Loading plots of principal components analysis on soil chemical properties and microbial parameters ...................................................................................... 71 3 -5 Canonical plots of discriminant analysis on soil microbial parameters. ............... 72 4 -1 Soil pH changes in response to different S application rates thr oughout the sugarcane growing season .................................................................................... 89 4 -2 The percentage of labile P and Fe -Al bound P of soil total P throughout the sugarcane growing season .................................................................................... 89 4 -3 Concentrations of labile P and Fe-Al bound P in soils after S application for different rates .......................................................................................................... 90 4 -4 The percentage of three P fractions to total P ....................................................... 91 4 -5 Concentrations of P in Ca-bound, humic -fulvic acid, and residual fractions of soils at various times after S application ............................................................... 91 4 -6 Distribution of P among inorganic and organic pools after S application ............. 92 5 -1 Concentrations of labile P in soils after elemental S application at different rates ...................................................................................................................... 112 5 -2 Activities of phosphatase (a) and glucosidase (b) in response to different elemental S application throughout the sugarcane growing season .................. 113

PAGE 12

12 5 -3 Activities of leucine aminopeptidase (a) and sulfatase (b) in response to different elemental S application rates throughout the sugarcane growing season .................................................................................................................. 114 5 -4 Microbial biomass P in soils after elemental S application at different rates ..... 115 5 -5 Mineralized S in soils after S application at different rates ................................. 115 6 -1 Soil pH changes in response to different S application rates throughout the sugarcane growing season .................................................................................. 136 6 -2 Seasonal dynamics of dissolved organic C, extractable NH4-N, NO3N, and SO4-S after S application ..................................................................................... 137 6 -3 Seasonal dynamics of acetic acid, Mehlich3, and water extractable P and K after S amendment ............................................................................................... 138 6 -4 Seasonal dynamics of acetic acid, Mehlich 3, and water extractable Ca and Mg after S amendment ......................................................................................... 139 6 -5 Seasonal dynamics of acetic acid, Mehlich3, and water extractable Cu and Fe after S amendment ......................................................................................... 140 6 -6 Seasonal dynamics of acetic acid, Mehlich3, and water extractable Mn and Zn after S amendment ......................................................................................... 141 7 -1 Soil pH changes in response to different S application rates throughout the sugarcane growing season .................................................................................. 154 7 -2 Seasonal dynamics of extractable SO4-S after S application ............................. 155 7 -3 Seasonal dynamics of extractable organic S after S application ........................ 155 7 -4 Seasonal dynamics of elemental S after S application ....................................... 156 7 -5 Sulfatase activities in response to different elemental S application rates throughout the sugarcane growing season ......................................................... 157 7 -6 Potential S mineralization in soils after S app lication at different rates. ............. 158 8 -1 Conceptual model of the microbial response to P fertilization in EAA soils under sugarcane .................................................................................................. 165 8 -2 Summary of biogeochemical processes in EAA soils after S applications ........ 166

PAGE 13

13 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Require ments for the Degree of Doctor of Philosophy COUPLING MICROBIAL A CTIVITY AND NUTRIENT CYCLING IN SOILS OF THE EVERGLADES AGRICULTU RAL AREA By Rongzhong Ye May 20 1 0 Chair: Alan L. Wright Co -chair: K. R amesh Reddy Major: Soil and Water Science The E verg lades Agricultural Area (EAA) is an important agricultural region, but has been implicated in contributing to the deterioration of water quality in the Everglades wetlands primarily through runoff of nutrients The objective of this research was to invest igate impacts of selected agricultural management practices on soil biochemical processes as related to microbial activities, nutrient cycling, and water quality. Longterm cultivation and nutrient and water management of EAA increased organic matter decomposition rates and altered supply of bioavailable nutrients Evaluation of the major land uses in the EAA indicated that labile phosphorus (P) was the main soil chemical property affected by agricultural management practices. Phosphorus fertilization re sulted in 465% higher concentrations of labile P in sugarcane soils compared to uncultivated soils. The uncultivated sites were the least disturbed, but the most P -limited and therefore had the smallest size of microbial population and lowest efficiency of C and N utilization. Statistical analysis revealed that P availability was the primary factor regulating the soil microbial community structure and function for land uses. Increased P availability promoted significant accumulation of P in m icrobial

PAGE 14

14 biomass which indicates that P fertilization is likely to enhance microbial activity and potentially increase organic matter decomposition and soil subsidence. Future management practices should consider the role of labile P on the function of microbial com munities and their control of nutrient cycling Elemental sulfur ( S) application to EAA soils is used as one of management practices to decrease pH and therefore increase the bioavailability of nutrients to crops. Sulfur application at rates up to 448 kg ha1 did not significantly reduce soil pH due to the high buffering capacity against acidification. However, the highest S rate did increase P concentrations in the Fe-Al bound P fraction by 55% compared to unamended soils at 2 months. The stimulatory e ffects were limited and did not last beyond 2 months. Higher S application rates may be needed to overcome the soil buffering capacity and increase nutrient availability. However, the risk of P export as runoff should be well recognized since the Ca-boun d P pool, which comprised of 32% of the total soil P, is not stable under acidic conditions. Similar to labile P, water extractable potassium (K) and acetic acid extractable Z i n c (Zn) increased at 2 months after S application at the highest S rates by 71% and 134% respectively, but the stimulatory effects did not extend beyond 2 months. Correspondingly, S application at current recommended rates did not increase sugar yield. The failure of S application to enhance nutrient availability throughout the gr owing season indicates the limited benefit of applying elemental S to reduce pH and increase nutrient availability to sugarcane. Alternatives, such as different P and micronutrient fertilizer application methods, timings, and sources, may be better for increasing nutrient availability for these changing soils.

PAGE 15

15 Sulfur did promote short -term changes in soil microbial activities. The activities of phosphatase and glucosidase in soils receiving 448 kg S ha1 were 115% and 560% higher, respectively, than uname nded soils at 2 months. Yet, microbial respiration and potential N and P mineralization rates were not affected by S amendment, suggesting that S application under the current recommendations would not enhance soil subsidence and regeneration rates of N and P. However, mineralized S increased with increasing S application rates and the effects continued throughout the growing season. Averaged across time, mineralizable S was 2, 6, 9, and 27 mg kg1 d1 for the increasing S application rates, indicating t hat S fertilization may pose an environmental hazard to the nearby aquatic ecosystem through potential export of SO4. Therefore, using higher S application rates to produce favorable responses in terms of nutrient availability and sugarcane yield should b e well evaluated in terms of both effects on sugarcane yield and adverse environmental effects. Agricultural management practices have significantly altered nutrient distribution and regeneration in EAA soils, especially the P cycling. P hosphorus fertili zation is to stimulate microbial activities and soil oxidation rates Current land use as sugarcane cropping continues to promote soil subsidence. Elemental S application under current recommendation rates introduced temporary and limited effects on incr easing nutrient availability to sugarcane and posed minimal impacts on microbial activity and function. However, S application at a large scale may increase the risk of SO4 export from the fields and stimulate nutrient regeneration rates and microbial act ivity.

PAGE 16

16 CHAPTER 1 INTRODUCTION S oil support s diverse microbial communities that play important roles in ecosystem level processes such as decomposition of organic matter and nutrient cycling (Wright and Reddy, 2001) In natural systems, the soil microbi al community composition and act ivity are related to the efficiency of nutrient cycling and ecosystem function. However, the richnes s, abundance and activity are vulnerable to influence by biogeochemical properties such as pH, moisture, organic matter co ntent, and nutrient s (Carreira et al., 2000; Allison et al., 2007; Castillo and Wright, 2008; Morra et al., 2009) Alterations in the physical and chemical nature of soil s may lead to shifts in microbial community composition and changes in microbial func tion. Decomposition of organic matter is the major nutrient source in soil s and is mediated by heterotrophic microorganisms. M icroorganisms play a critical role in regulating nutrient cycling as s oil microbial biomass is considered both a source and sink for nutrients (Sicardi et al., 2004; Zhang et al., 2006). Plants exclusively utilize inorganic nutrients and therefore their growth depends on decomposition of organic matter, particularly in natural and agricultural ecosystems having low nutrient inputs (Yao et al., 2000; Chen et al., 2003). Microbial activity, such as extracellular enzyme activities and respiratory rates, reflects the potential of nutrient immobilization and mineralization, and hence are widely utilized to indicate changes in soil qual ity,

PAGE 17

17 organic matter status, and nutrient cycling processes (Stevenson et al., 2004; Monkiedje et al., 2006; Nogueira et al., 2006). Nutrient limitation is a key factor commonly constraining agricultural production and soil microbial populations A gricult ural management practices, such as organic amendments and fertiliz er use, are commonly applied to enhance crop yields. Th e se practices result in alterations of nutrient distribution in the soil profile and shifts in microbial community structure and funct ion. For instance, t illage enhances org anic matter turnover by increasing the presence of ox ygen throughout the soil profile and consequently enhances microbial respiration (Gesch et al., 2007). Fertilization overcomes potential nutrient limitations and stimulates microbial activity and crop growth (Castillo and Wright, 2008 a ). Organic amendments often improve soil characteristics and enhance microbial activity (Tejada et al., 2006). However, e xcessive nutrients from fertilization or decomposition can b e immobilized in soil organic matter, or lost as leaching or runoff from fields, which may cause serious environmental problems to the adjacent ecosyst ems. Thus, n utrient management in agricultural fields is especially important for critical or sensitiv e ecosystems, such as those in s outh Florida near the Everglades. Everglades Agricultural Area The Everglades Agricultural Area (EAA) is located south of Lake Okeechobee and north of the Water Conservation Areas (WCAs) in south Flori d a (Fig 1 1 ). I t cons ists of an area of approximately 283,300 ha which was drained in the early

PAGE 18

18 1900s for agricultural production (Chen et al., 2006). Currently, the land use is predominantly sugarcane with a smaller portion dedicated to rice, vegetable s, and turf sod product ion. The soils are primarily H istosols with an organic matter content as high as 8090 % (Snyder et al. 2005). D ue to the drain ing and establishment of the EAA, soil subsidence has increasingly become a crit ical problem Subsidence resulted from the oxi dation of organic matter upon initiation of drainage in the early 1900s (Gesch et al., 2007 ). Along with subsidence, exces sive N and P are released during organic matter mineralization which has altered nutrient cycling and soil processes (Morris and Gilb ert, 2005). The current longterm estimat e of soil subsidence is approximately 1.5 cm yr1 ( Morris and Gilbert, 2005 ). At this rate of loss, soils may become too shallow for many agricultural use and sugarcane production in the near future (Anderson and Flaig, 1995; Morris and Gilbert, 2005). Longterm cultivation of these soils specifically tillage has resulted in the incorporation of bedrock CaCO3 into surface soil and has gradually increas ed the pH from the historic 5.0 to 5.5 to approximately 7.0 to 7.5 today (Snyder, 2005). As a result, P and micronutrient availability to crops has decreased and necessitated new fertilizer management practices. Elemental S is occasionally applied for the purpose of reducing pH and therefore improving the availab ility of micronutrients and P to sugarcane (Schueneman, 2001). The microbial oxi dation of elemental S to SO4 produces acidity which reacts with the soil and reduces pH, which in turn releases P and micronutrients from adsorption sites into soil solution. However, the buffering

PAGE 19

19 capacity of these calcareous Histosols is strong and can counteract the acidifying effects of elemental S oxidation, t hus effects of amendments may only be temporary and may need to be repeated each growing season (Beverly and Ander son, 1986). Sugarcane is the dominant crop grown in the EAA, and requi r es approximately 3 0 kg P ha1 year1 and extensive tillage for pre-plant preparation and weed control (Rice et al. 2006). Elements that are of nutritional concern for sugarcane product ion include N, P, K, Mg, B, Cu, Fe, Mn, Si, and Zn (Rice et al., 2006). To prevent nutrient deficiency and maximize sugar production, fertilizer recommendations were introduced by the University of Floridas Institute of Food and Agricultural Science (IFA S) (Anderson, 1985; Rice et al., 2006). Longterm P application has resulted in P accumulation in the soil profile as well as export into Everglades wetlands through canal systems, which was a major factor contributing to the deterioration of water qualit y and alterations of the Everglades wetland ecosystem (Childers et al. 2003). Studies have demonstrated that most of applied P fertilizers enter the Fe -Al bound and Ca-Mg bound P fractions, which together account for 60% of the total P in EAA soils (Wrigh t, 2009). Phosphorus retained in Fe-Al and Ca Mg bound pools of these calcareous soils are considered relatively stable under current drained conditions (Wright, 2009). Several factors are capable of influencing P stability and mobility in the soil profi le including pH, microbial activity, and soil amendments (Arai et al. 2005; Jaggi et al. 2005). Application of elemental S at a rate of 500 g g1 significantly enhanced available P concentrations in alkaline soils having a pH of 10.2 (Jaggi et

PAGE 20

20 al., 2005) Jaggi et al. suggested that increased P concentrations resulted from lowered pH and replacement of SO4 with PO4 from soil adsorption sites. However, such stimulatory effects were not observed for other studies of calcareous soils with a pH of 8.1 (Hass an and Olson, 1966). The ecosystem of south Florida has been greatly modified as a result of agriculture and urbanization Driven factors include deterioration of water quality and alterations in ecosystem -level processes (Chimney and Goforth, 2006), t hus i mprovement s in water qualit y is a major goal of Everglades rehabilitation projects (Gabriel, 2008) Reducing nutrient exports from the EAA to the remnant Everglade s is an ongoing strategy for improving water quality (Chimney and Goforth, 2006; Gabriel, 2008) Therefore, a better understanding o f nutrient dynamics and land management within the EAA is critical. Land Use Change Acknowledging the fact that subsidence clouds the future of agriculture in the EAA, strategies have been implemented to increa se sustainability of these soils (Grigg et al., 2002; Morris et al., 2004). Major l and use changes are considered inevitable in the near future, likely i n the order of decades (Anderson and Rosendahl, 1998; Snyder 2005 ). An emerging interest is to convert current land uses back to prior uses as wetland prairie. Under flooded conditions degradation of organic matter and the subsidence rate can be reduced or eliminated (Grigg et al., 2002). Snyder (2005) proposed possible land use s in the EAA over the n ext 50 years

PAGE 21

21 including growing pasture grasses and utilizing forestry in areas with shallow organic soils. Additionally, in consideration of the continuous growing population and urbanization, more agricultural lands are expected to change to home develop ment sites with turfgrass coverage (Anderson and Rosendahl, 1998). Clear ly, many land use options can be explored to serve as alternatives to traditional agriculture. However, far less is known about the influence of land use changes on the EAA ecosystem with regard to nutrient cycling and microbial community dynamics. Sulfur in the Everglades Sulfur is required by all biological materials as an essential macronutrient (Wang et al., 2006) and it is present in soils in various forms, each of which play imp ortant biological and chemical roles. Sulfate is the most abundant form of inorganic S found in most soil s and the main form available to plants although reduced forms, such as elemental S, thiosulfate, and sulfide, can be found under anaerobic conditions (Zhou et al., 2005). However, the majority of soil S is in organic form which consists of two components, ester S and C bound S, serving as the main source for inorganic S (Solomon et al., 2001). Mineralization of organic S compounds is mediated by heterotrophic microbial activity and closely associated with C and N mineralization (Gharmakher et al., 2009) McGill and Cole (1981) proposed a conceptual model for cycling of organic C, N, S and P through soil organic matter, in which S mineralization inv olves biological and biochemical processes. In biological processes, SO4 is released as a by product of C oxidation,

PAGE 22

22 while SO4 can also be a direct product of enzymatic hydrolysis (McGill and Cole, 1981). Potential S sources to the EAA include soil amen dments, fungicides, and fertilizers (Orem, 2007). Since the 1920s, S has been applied to soil as CuSO4 to enhance crop yields via increasing the availability of Cu to these micronutrient poor soils (Allison et al., 1927). In later years, elemental S was recommended to lower soil pH when it exceeded 6.6 for the purpose of improving the availability of micronutrients needed for sugarcane growth (Anderson, 1985). The IFAS recommended rate is 560 kg S ha1, but Schueneman (2001) indicated that actual S appli cation rates in the EAA are lower than the IFAS recommendations. Everglades Agricultural Area growers tend to use micronutrient sprays to alleviate nutrient deficiency caused by elevated pH since it is more cost effective, so S application may not be cons idered necessary at this stage. However, due to the increasing soil pH and decreasing soil depth to bedrock since 1985, revision of this r ecommendation may be required. There is a need to determine the level of S application producing favorable responses in terms of nutrient availability and sugarcane yield. There is widespread S contamination of surface water and sediments in the northern Everglades (Bates et al., 1998; Bates et al., 2002). Potential sources contributing to the S enrichment include agricultural S, groundwater, rainwater, seawater aerosol, S flux from sediments, and influx of water from Lake Okeechobee

PAGE 23

23 (Bates et al., 2002). However, it has been suggested that S from fertilizers and soil amendments are the primary contributors to SO4 enrichment of the Everglades (Bates et al., 2002; Orem, 2007). The environmental and ecological significance of S in Everglades wetlands is principally the stimulation of MeHg formation, which is catalyzed by SO4-reducing bacteria under anaerobic conditions (Bates et al., 2002). The MeHg is a neurotoxin that is bioaccumulative and found in high concentrations in fish and other wildlife in the Everglades (Orem, 2007). During the process of SO4 reduction, P bound to the organic matter c an be released As such, the presence of high SO4 concentrations can lead to internal eutrophication (Gilmour et al., 2007). Two primary mechanisms are possibly responsible for the internal eutrophication. First, production of sulfide result s in the release of nutrient s p articularly NH4-N and PO4. Second, NH4-N and PO4 are mobilized through the generation of alkalinity (Gilmour et al., 2007 ; Gabriel et al., 2008). Furthermore, SO4 reduction produces sulfide, which can be toxic to the aquatic plants and animals (Orem, 200 4). In consideration of the adverse impacts that S poses to the Everglades wetlands, reducing potential S exports from the EAA is beneficial for protecting water quality and ecosystem health (Gabriel at al., 2008). Explicit quantification of S budgets and transformations within EAA soils is limited, so there is a need to study the seasonal dynamics of S as related to S application practices and the risk of S exportation.

PAGE 24

24 Objectives and Hypothesis The present study wa s designed to experimentally determine the influence of select land uses (sugarcane, forest, and uncultivated) and management on the microbial activity and nutrient cycling in EAA soils (Fig. 1 -2) The specific objectives and hypotheses were to: Investigate impacts of land uses (sugarcane, fore st, and uncultivated) on microbial activities and cycling of C, N, and P (Chapter 2). Hypothesis: Long -term land management (P fertilization) has altered C, N and P cycling as well as microbial activity and community composition Identify linkages between m icrobial activities and soil biogeochemical properties using multivariate methods (Chapter 3). Hypothesis: Differences in microbial function under various landuse types can be explained by variations in soil biogeochemical properties Determine effects of S amendment on P cycling and related processes during the sugarcane growing season (Chapter 4). Hypothesis: Sulfur application will reduce soil pH and consequently influence soil P forms and availability Evaluate the microbial ecophysiological response to S amendment (Chapter 5). Hypothesis: Sulfur application will increase nutrient availability and stimulate microbial activity Determine S addition effects on nutrient availability to sugarcane (Chapter 6). Hypothesis: Micronutrient availability to sugar cane will be enhanced due to S application Quantify S forms and transformations in soils amended with elemental S (Chapter 7) Hypothesis: Sulfur application will modify S forms and transformations in EAA soils

PAGE 25

25 The completion of this study is expected to provide information to better manage nutrient and lands within the EAA while minimizing adverse environmental impacts. More specifically, completion of this project would provide information to a) predict effects of future land use changes on nutrient cycl ing; b) reduce nutrient export from fields; c) improve water quality; d) satisfy nutrient requirements for crops; e) optimize fertilizer use efficiency; and f) contribute to more efficient and effective management of south Florida ecosystems.

PAGE 26

26 Fig 1 1 Map of the Everglades Agricultural Area in south Florida. Source: South Florida Water Management District

PAGE 27

27 EAA Soil SOM Microbial Biomass Bioavailable Nutrients EA Land Use Change Water Table Management Phosphorus Fertilization Sulfur Application BMP STAs Crops EAA Soil SOM Microbial Biomass Bioavailable Nutrients EA Land Use Change Water Table Management Phosphorus Fertilization Sulfur Application BMP STAs Crops Fig. 1 2. Conceptual scheme showing selected land managements that potentially affect biogeochemical processes in EAA soils. BMP, best management practices; SOM, soil organic matter; EA, enzymatic activity; STAs, storm water treatment areas.

PAGE 28

28 CHAPTER 2 LAND USE EFFECTS ON SOIL NUTRIENT CYCLING AND MICROBIAL COMM UNITY DYNAMICS IN THE EVER GLADES AGRICULTURAL AREA, FLORIDA Introduction S oil suppo rt s divers e microbial communities that play important roles in ecosystem level processes such as decomposition of organic matter and nutrient cycling (Wright and Reddy, 2001a ). In natural systems, soil microbial community composition and act ivity are relat ed to the efficiency of nutrient cycling and ecosystem function (Yao et al. 2000). However, the richnes s, abundance and activity of the microbial community is vulnerable to influence by soil physical and chemical properties such as pH, moisture, organic matter content, and nutrient availability Alterations in the physical and chemical nature of the soil may lead to shifts in microbial community composition and changes in microbial function. Agricultur al practices such as fertilization and tillage influ ence soil chemical properties and nutrient dynamics throughout the soil profile (Gesch et al., 2007; Wright et al., 2007). Therefore, changes in land use s with different intensity or history of agricultural practices consequently result s in distinctive changes in microbial community composition and function (Grayston et al., 2004). Lan d use changes can also disrupt carbon (C) and nitrogen (N) dynamic s and organic matter storage in soils across a range of habitats (Garcia-Oliva et al, 2006; Monkiedje et al., 2006), which are commonly viewed as major factors caus ing shifts in microbial community composition (Schimel and Bennett, 2004; Cookson et al., 2007). The Everglades Agricultural Area (EAA) is located south of Lake Okeechobee and north of the Water Cons ervation Areas (WCAs) in south Florida. It consists of an area of approximately 283,300 ha which was artificially drained in the early 1900s for agricultural production (Chen et al., 2006). Currently, the land use is primarily

PAGE 29

29 sugarcane with a smaller port ion dedicated to vegetable s The soils in the EAA are primarily H istosols with organic matter content s as high as 8090% (Snyder et al. 1978). Due to the drain ing and establishment of the EAA, soil subsidence has become a critical problem, which resulted from the oxidation of organic matter (Gesch et al., 2007) Along with subsidence, excessive N and phosphorus (P) were released from organic matter mineralization, which altered nutrient cycling and soil processes (Morris and Gilbert, 2005). The current l ongterm estimate of soil subsidence is approximately 1.5 cm yr1 (Morris and Gilbert, 2005). At this rate of loss, soils will become too shallow for agricultural use and sugarcane production in the future (Anderson and Flaig, 1995; Morris and Gilbert, 200 5). Acknowledging the fact that the subsidence clouds the future of agriculture in EAA, strategies have been implemented to increase sustainability of agriculture (Grigg et al., 2002; Morris et al., 2004). However, land use changes in EAA are considered in evitable in the near future, likely on the order of decades (Anderson and Rosendahl, 1998; Snyder 2005 ). An emerging interest is to convert current land uses back to prior uses as prairie. Under flooded conditions, degradation of organic matter and the s ubsidence rate can be reduced or eliminated (Grigg et al., 2002). Snyder (2005) further proposed the possible land use s in the EAA over the next 50 years, which included growing pasture grasses and planting cypress trees in areas with shallow organic soils Additionally, in consideration of the continuous growing population and urbanization, more agricultural lands are expected to change to home development sites with turfgrass coverage (Anderson and Rosendahl, 1998). Obviously, many land use options can be explored to serve as alternatives to traditional agriculture. However, far less is known about the

PAGE 30

30 influence of land use change on EAA ecosystems with regard to nutrient cycling and microbial community dynamics. The objective s of the present study w ere t o determine and characterize the impacts of land use on the soil microbial community composition and activity and to investigate the effects of land use on nutrient cycling. Materials and Methods Site Description The study sites are located in the northern EAA at the Everglades Research and Education Center near Belle Glade, FL. The long-term average annual rainfall is 133 cm and temperature is 24C. All soils are Dania muck (euic, hyperthermic, shallow Lithic Medisaprists) with a depth to the bedrock of ap proximately 45 cm. These organic soils developed under seasonal flooding and low nutrient status and supported vegetation adapted to these conditions, primarily sawgrass ( Cladium jamaicense Crantz). Due to conversion to agricultural use by drainage, the dominant vegetation shifted to annual crops of vegetable and sugarcane ( Saccharum sp.) in the early 1900s. Four land uses were selected for this study to mirror possible land uses in the future: soils under forest for 19 years, fields under sugarcane product ion for approximately 50 years, turfgrass lawns for 60 years, and fields under perennial pasture for approximately 100 years. Four field sites were randomly sampled for each land use. The forest soils were previously cropped to sugarcane but planted to bal d cypress ( Taxodium distichum ) and pond cypress ( Taxodium ascendens ) in 1988. T hese fields did not receive any fertilization after land use change, but were extensively tilled prior to seedling establishment, and no further management has been applied. The sugarcane fields were managed for vegetable production from the early 1900s to the 1950s, but mainly

PAGE 31

31 for sugarcane since the 1950s. Fertilization was applied at a rate of 40 kg P ha1 yr1 (Gilbert and Rice, 2006) prior to planting. Sugarcane is planted f rom August through January and harvested from October through April. Tillage operations included several disking (to 15 cm depth) after crop harvest, subsoil chiseling (to 30 cm depth) to improve drainage, and frequent in-season tine cultivations (to 4 cm depth) for weed control (Morris et al., 2004). The turf fields were vegetated by St. Augustinegrass [ Stenotaphrum secundatum (Walt) Kuntze] turf since the mid 1940s. The uncultivated field was primarily occupied by paragrass [ Panicum purpurascens (L.) Radd i] and bermudagrass [ Cynodon dactylon (L.) Pers] Turf and uncultivated fields were periodically mowed with residues returned to soil, but received no fertilization and tillage since establishment. Soil Sampling and PhysicalChemical Analysis Surface soi l (0 15 cm) samples were collected from 4 replicate fields of each land use on March 2007. The soils were homogenized after the removal of large plant particles and stored at 4C until use. Moisture content was measured as the mass loss after drying at 70 C for 5 days. Soil organic matter content was estimated by the loss onignition method after ashing at 550C for 4 hours (Anderson, 1976). Total organic C was then calculated from the organic content by using a factor of 0.51 (Anderson, 1976). Total C, tot al N, and total P were determined using the oven dried (70C) and ground soil. Total C and N were measured with a Carlo Erba NA 1500 CNS Analyzer (Haak -Buchler Instruments, Saddlebrook NJ), while total P was evaluated after ashing (Bremner, 1996) using th e ascorbic acid molybdenum blue method (Kuo, 1996) with an AQ2+ discrete analyzer (Seal Analytical Inc., Mequon, WI).

PAGE 32

32 Microbial Biomass Microbial biomass C (MBC), N (MBN) and P (MBP) were measured by the fumigationextraction method (White and Reddy, 2001). The amount of K2SO4extracted C was determined with a total organic carbon analyzer TOC -5050A (Shimadzu, Norcross, GA). The microbial biomass C was calculated from the difference in extractable C between fumigated and unfumigated samples using a conver sion factor of 0.37. After digestion, the K2SO4extracts were measured for total Kjeldahl N using an AQ2+ discrete analyzer (Seal Analytical Inc., Mequon, WI). The microbial biomass N was calculated from the difference in total Kjeldahl N between fumigated and unfumigated samples using KEC = 0.54. The total P content of the NaHCO3 extracts used for microbial biomass P analysis was measured for P as previously described (Kuo, 1996). The microbial biomass P was determined as the difference in total P of NaHC O3 extracts from fumigated and unfumigated samples. Labile inorganic P (NaHCO3-Pi) was measured for unfumigated soil extracts and analyzed for P as previously described (Kuo, 1996). Potentially Mineralizable N and P Potentially mineralizable N (PMN) was d etermined according to methods of White and Reddy (2000) based on a 10day incubation followed by extraction with 2M KCl. Extracts were analyzed for NH4-N (White and Reddy, 2000). Potentially mineralizable P (PMP) was measured using the method of Corstanje et al. (2007) with slight modifications. 0.5 g dry soil were placed in 30 -ml serum bottles and mixed with 5 ml of water. The bottles were then capped incubated in the dark at 40C for 10 days. At 10 d, 20 ml of 1M HCl was injected and soil shaken for 3 hours. Extracts were filtered through

PAGE 33

33 were directly extracted with 25 ml of 1M HCl and extracts were analyzed for total P. The potentially mineralizable P was determined as the difference in total P of extracts between incubated and non -incubated soil. Enzyme Activity Assay Approximately 1 g moist soil was placed in polypropylene centrifuge tubes, mixed with 30 ml of water, and shaken for 25 minutes. Homogenized samples were further diluted 5 times for enzyme assays. Enzyme assays were conducted in three or four replicates with controls to offset non enzymatic production. For cellobiohydrolase assay, the substrate used was 2 mM pnitrophenol -cellobioside (ACROS Organics, Geel, Belgium). 0.75 ml of diluted samples and 0.75 ml of substrates were mixed in 2ml micro centrifuge tubes and incubated at 20C for 20 hours with gentle shaking. At the end of the incubation, the mixtures were centrifuged at 10,000 rpm for 1 minute, 0 .75 ml supernatants were transferred to new tubes, followed by the addition of 0.075 ml of 1 N NaOH to stop the reaction and develop the color. The mixtures were then analyzed for absorbance with a spectrophotometer UV 160 (Shimadzu, Norcross, GA) at 420 n m. The enzymatic activity was expressed as mg pnitrophenol released per gram dried soil per hour. Leucine aminopeptidase assay was conducted in 96 well microtiter plates (Prenger and Reddy, 2004). 200 l of samples was incubated with 50 l substrates, 5 m M L Leucine 7 -Amino -4 methylcoumarin (Biosynth, Naperville, IL), at 20C for 8 hours. The florescence readings were collected at 1 hour intervals using a fluorescence plate reader, Bio-TEK FL600 (Bio -TEK Instruments Inc., Winooski, VT), at a setting of 365 nm excitation and 450 nm emission. Enzyme activity was determined by calculating

PAGE 34

34 the mean florescent reading changes over time with a standard curve and expressed as mg 7 amino4 -methylcoumarin released per gram dried soil per hour. Alkaline phosphatase and sulfatase assays were conducted according to the methods of Wright and Reddy (2001a). Community -Level Physiological Profile by BIOLOG Assay Community -Level physiological profiles (CLPPs) were determined by direct incubation of fresh soil extracts in BIOLOG Eco -Plates (31 substrates) (BIOLOG Inc.). Approximately 1 g moist soil samples was mixed with 20 ml of water and gently shaken for 20 minutes. The homogenized samples were then diluted 400 times and soil particles allowed to settle for 15 minutes a t 4C. 150 l of the supernatants was subsequently dispensed into each well of Eco-Plates and incubated at 20C for 7 days. Optical densities were measured every 6 or 12 hours using Bio -TEK FL600 (Bio -TEK Instruments Inc., Winooski, VT) at 590 nm. Absorbance values of each well with C sources were blanked against control wells before analysis. Negative values were considered as 0. Community metabolic diversity (CMD) was calculated by summing up the numbers of positive responses in each plate. Positive responses were defined as any absorbance values greater than 0.25 (Garland, 1997). Average well color development (AWCD) was determined as described by Garland (1996, 1997). To overcome possible interference by inoculum density on color development, absorbance values for various C sources were standardized by dividing the blanked value of each well by the AWCD of the plate and were subsequently used for principal component analysis.

PAGE 35

35 Statistic al Analysis Significant differences among land uses for all the variabl es were determined by oneprofiles was carried out by principal component analysis (Preston -Mafham et al., 2002; Campbell et al., 2003). All computations were conducted on computing software JMP V4 (SAS Institute Inc., NC). Results Soil Physical and Chemical Properties Soil organic matter content was highest in uncultivated soil and decreased in the order of forest, sugarcane and turf soil, respectively (Table 2 1). Total C and N were also highest in uncultivated, averaging 461 g C kg1and 32 g N kg1, respectively. However, in contrast to total C and N, total P was lowest in uncultivated soils averaging 0.78 g P kg1. Extractable C was significantly higher in sugarcane (2.34 g C kg1) than other land uses (Table 2 -1). However, no difference in extractable C was found among forest, turf and uncultivated soil. Extractable NH4-N in forest and sugarcane soils was lower than turf and uncultivated soils (Table 2 -1). Labile inorganic P was higher in sugarcane soil (96 mg P kg1) than forest (43 mg P kg1) and uncultivated (17 mg P kg1). The highest inorganic P was found in turf soil (173 mg P kg1); however, it was not statistically different from other land uses. Potentially mineralizable N was four times lower in sugarcane (2.66 mg N kg1 d1) than other land uses (Table 2 -2). Potentially mineralizable P varied widely in the turf soil, averaging 33.1 mg P kg1 d1, which was about 8 times higher than other land uses (Table 2 -2), but th e difference was not significant.

PAGE 36

36 Microbial Biomass Soil microbial biomass C, ranging from 9.3 to 13.5 g C kg1, was lower in uncultivated soil than forest and sugarcane (Table 2 2). The microbial biomass C to organic matter ratio in the turf soil (23374 mg kg1) was two times higher than uncultivated soil which was also higher than the forest (15183 mg kg1) and sugarcane (16857 mg kg1) soils. As well, the microbial biomass C to organic matter ratio was lower in uncultivated soil than forest and sugarca ne. Additionally, microbial biomass N was lowest in uncultivated soil (0.12 g N kg1) and highest in turf soil (0.32 g N kg1). Soil microbial biomass P was not significantly different among land uses. Extracellular Enzyme Activity Cellobiohydrolase activ ity was highly varied in uncultivated soil, averaging 271 mg g1 h1, which was not significantly different from other land uses (Table 2 2). However, activities in forest, sugarcane and turf soils varied considerably, averaging 140, 51, and 191 mg g1 h1, respectively. Uncultivated soil had higher alkaline phosphatase (1.21 mg g1 h1) than forest (0.70 mg g1 h1) and sugarcane (0.94 mg g1 h1). Sulfatase activities were higher in uncultivated soil (0.42 mg g1 h1) than sugarcane soil (0.24 mg g1 h1), but were not different from turf soil (0.44 mg g1 h1) and forest soil (0.32 mg g1 h1). Microbial Community Composition and Function We used the CMD and AWCD to describe the average numbers of substrates potentially utilized and respiration rates from the C -sources by microbial communities. Both CMD (Fig. 2 -1) and AWCD (data not shown) followed a sigmoidal curve over time. On day 1, no color development was observed; nevertheless on day 3.25 both CMD and AWCD of forest and turf reached their midpo int. Since the actual rates of color development changed with time, the comparison among land uses in relative rates was

PAGE 37

37 made in the interval of day 1 to day 3.25. The CMD of the microbial communities was higher in forest and turf than sugarcane and uncult ivated (Fig 2 -1). On day 3.25, more than 55% of the wells were positive in plates inoculated with turf or forest soil, yet only 15% of wells were found positive in plates from sugarcane soils, and 6% of wells from uncultivated soils show ed positive respon ses (Fig. 2 1 ). The rates of color development (AWCD) followed the same pattern as CMD, being highest in turf and forest and lower for sugarcane and uncultivated (data not shown). Principal component analysis on the utilization patterns of all of the 31 substrates on day 3.25 revealed clear differentiation among land uses in the composition of active members of the soil microbial community (Fig. 2 -2). In particular, ordination axis 1 demonstrated obvious separation between turf and either the uncultivated o r sugarcane. Furthermore, ordination axis 1 separated the soil microbial community in forest fro m that of uncultivated soil No apparent separations were observed on axis 2. However, microbial community composition in forest soil s was distinct from sugarcane soils on ordination axis 3. Discussion Nutrient Distribution and Cycling To investigate land use effect s on nutrient distribution and cycling, samples from the 0 15 cm depth were utilized since surface soils are most susceptible to change s in chemical and physical properties (Garcia-Oliva et al., 2006). Long term cultivation and fertilization in the EAA greatly altered nutrient distribution and increased the organic matter turnover rates (Tab le 2 -1) Uncultivated soils had higher organic matter content than other land uses, and were the only land use never subjected to tillage, while various tillage practices were often applied to sugarcane fields to maintain drainage and

PAGE 38

38 for weed control (Morris et al., 2004). It has been reported that t illage practices alter the below ground ecosystem and expose subsurface organic matter to the aerobic environment which increases organic matter decomposition rates and may lead to soil subsidence (Reicosky and Lindstrom, 1993; Gesch et al., 2007). Thus, t illage operations in sugarcane likely contributed to its lower organic matter content than other land uses Lohila et al. (2003) proposed that tillage-induced soil C loss was likely greater for H istosols than mineral soils. Since the dominant soil of EAA is H istosols and the subsidence problem is considered a result of aerobic oxidation of organic C (Shih et al., 1998), it is reasonable to postulate that frequent tillage may worsen the subsidence problems in EAA (Gesch et al 2007). Labile inorganic P is a readily available fraction of P that remains soluble until either absorbed or precipitated to Fe, Al, Ca, and Mg (Anderson et al., 1994). Therefore, it is mobile with drainage, runoff or shallow groundwater. Sugarcane soil had about 6 times more labile inorganic P than uncultivated soil and 2 times more than forest due to its intensive fertilization history. However, we did not find any significant difference in labile organic P and PMP among soils of those land use s In addition, labile inorganic P of the sugarcane soil accounted for 10% of its total P stock, which was higher than other land uses, especially uncultivated soil (2%). In contrast, uncultivated soil had 10% of its total P as labile organic P, which was significantly greater than other land uses. Thus, intens ive fertilization and management increased soil P retention in inorganic forms rather than organic and conversely, land uses with minimal cultivation had a greater P sequestration in organic fractions (Graham et al., 2005). Furthermore, practices under sug arcane cropping releases more dissolved inorganic P to drainage water than both

PAGE 39

39 uncultivated and forest, which poses a greater threat to the downstream ecosystems. The Florida Everglades is highly sensitive to small increases in P concentrations (Noe et al ., 2001). Phosphorus in the drainage water from the EAA is considered to be the key contributor to the eutrophication of Lake Okeechobee and the Everglades (Childers et al., 2003). Our results support the idea that reduced management practices and P fertil ization intensity potentially decrease the P content in the EAA drainage water (Izuno et al 1991), thus land uses that minimize soil disturbance decrease potential for eutrophication of downgradient aquatic systems. The C: N: P molar ratios of forest, s ugarcane, turf and uncultivated soil were 939:54:1, 1178:67:1, 399:23:1, and 1528:93:1, respectively, suggesting that the uncultivated soil was the most P limited land use, followed by the sugarcane and forest. There were no obvious differences in C: N rat ios among land uses. However, the difference in both the C: P and N: P ratios were significant, indicating that agricultural practices greatly influenced the P sequestration in the soil. Alkaline phosphatase activity plays an important role in the P -limite d ecosystem with respect to the regeneration of P from organic forms (Wright and Reddy, 2001a). Hence, it was not surprising to find that the uncultivated soil exhibited the highest alkaline phosphatase activity, followed by the sugarcane and forest soils. No significant correlation between alkaline phosphatase activity and soil P parameters implied other environmental factors likely contributed to this observation Microbial Community Dynamics Microbial communities are in close contact with soil microenvir onments, and therefore are easily subjected to change following alteration of soil properties (Corstanje

PAGE 40

40 et al., 2007). Hence, changes in land use may cause a shift in the composition of active fractions of microbial communities, which can be explained by the changes to indices of microbial activity such as respiratory capacities and extracellular enzymatic activities (Wright and Reddy, 2001b; Corstanje et al., 2007). Significant correlations between microbial biomass C and either organic matter or dissol ved organic C has been frequently observed (Yao et al., 2000; Cookson et al., 2007). However, in the present study we did not find any of such correlation. The microbial biomass C to organic matter ratio is thought to be indicative of the organic matter qu ality and availability (Monkiedje et al., 2006). Our results showed that the ratio was highest in turf and lowest in uncultivated soil (Table 2 2), which suggested that turf had the highest chemical diversity in organic matter sources and efficiency of C u tilization and conversely, the uncultivated soil had the lowest. This statement was further supported by the findings that microbial biomass C to organic matter ratio was significant ly correlated with AWCD (R2 = 0.81) and CMD (R2 = 0.65) Uncultivated soil has lower overall plant coverage than turf, forest, and sugarcane (Shih et al., 1982), thus it is reasonable to expect lower chemical diversity of organic matter sources. Analysis of PMN to microbial biomass N ratio also revealed that uncultivated soil ha d a significantly higher ratio than other land uses, indicating that its microbial communities had the lowest efficiency in utilizing N resources. Regarding the fact that uncultivated soil had lower microbial biomass C and N, it was possible that uncultiva ted soil had the lowest abundance of microbial populations, which may explain its low efficiency in C and N utilization. No statistical difference was found in PMP to total P ratio and to microbial biomass P ratio, suggesting less effects of land use on th e overall potential P

PAGE 41

41 turnover rates and utilization efficiency. In consideration of its lower abundance of microbial populations and poorer efficiency of C and N utilization, uncultivated soil may experience lower rates of organic matter degradation, whic h was also evidenced by its higher organic matter content (Table 2 1). Shih et al. (1982) reported that oxidation rates were higher in uncultivated than sugarcane and forest due to its higher soil temperature, mainly resulting from lower vegetative cover and greater exposed area. Uncultivated fields were periodically mowed with residues returned back to soil, and soil was periodically covered with layers of plant residues, which provided inputs of organic matter and prevented the temperature increases. Thus results of temperature effects on soil oxidation may be confounded by differences in sampling time and sites. Interestingly we found significant correlations between microbial biomass C to organic matter ratio and total P (R2 = 0.70), labile inorganic P (R2 = 0.72), total labile P (R2 = 0.73), and microbial biomass P (R2 = 0.58) implying that P plays the critical role in determining the efficiency of C utilization and degradation of organic matter by soil microorganisms. It is well known that nutrient av ailability such as P, greatly influences soil microbial activity and function (Wright and Reddy, 2001a, b; Corstanje et al., 2007). As demonstrated previously, uncultivated soil was the most P limited system, followed by the sugarcane and forest soils. Pr obably, P deficiency deeply confined the microbial population limited the microbial activity, and subsequently inhibited the efficiency of C utilization and reduced oxidation rates of organic matter for uncultivated soil. Extracellular enzymes are excret ed by the microorganisms to the soil for sequestering nutrients. The activities of those enzymes are critical for the degradation of soil organic matter and plant detritus (Wright and Reddy, 2001a) and regulated by the

PAGE 42

42 availability of the substrates and ot her environmental factors (White and Reddy, 1999; Corstanje et al., 2007). To characterize the effects of land uses on microbial activity, we conducted the enzymatic assays of cellobiohydrolase in the C cycle, leucine aminopeptidase in the N cycle, alkaline phosphatase in the P cycle and arylsulfatase in the S cycle. Most of the enzymes were sensitive to the changes in land use, except leucine aminopeptidase, indicating possible differences in the composition of active fractions of soil microbial populations and biochemical processes among different land uses (Monkiedje et al., 2006). Application of land management and plant coverage greatly affected the distribution and availability of substrates and the quantity and quality of organic matter sources, which primarily contributed to differences in enzyme activities. Significantly negative correlations were found between the leucine aminopeptidase activities and moisture content (R2 = -0.55), loss on-ignition (R2 = 0.80), labile organic P (R2 = 0.70), and mi crobial biomass C (R2 = 0.58), respectively. It was possible that leucine aminopeptidase was sensitive to factors other than the quality and quantity of organic matter sources. Sole C -source utilization patterns are commonly used to evaluate changes in m icrobial community composition and functional diversity and have been successfully applied to ranges of soil habitats undergoing changes in land use (Garland, 1996; Campbell et al., 2003). BIOLOG EcoPlates were employed to investigate the utilization patte rns of 31 different substrates by soil microbes. Our results indicated clearly that the microbial community composition and functional diversity differed across land uses. The CMD was highest in turf and forest soils and lowest in uncultivated soil (Fig. 2 -1). Since the difference in the CMD profile reflects the variation in substrate diversity and

PAGE 43

43 availability (Grayston et al., 2004), the C sources in turf and forest soils appeared more divisive than those of the sugarcane and uncultivated fields, which further affirmed our aforementioned statements on the results of microbial biomass C to organic matter ratios. Average well color development describes the average respiration of the C sources by the microbial community (Garland, 1997). Results demonstrated that microbial communities of turf and forest soils were more functionally adapted to use those C resources than sugarcane and uncultivated soil which was consistent with our previous postulation that turf had the highest efficiency of C utilization and u ncultivated soil the lowest. R esults also indicated that microbial communities of turf and forest had higher capacities to acclimatize to alterations in land use (PrestonMafham et al., 2002). Principal component analysis on the color development profiles revealed no similarity in the microbial community composition between any two of those four land uses, except that of the forest and turf (Fig. 2 -2). It has been proposed recently that both dissolved organic C and dissolved organic N indeed greatly regulat e the composition of the active fractions of the soil microbial community (Schimel and Bennett, 2004; Cookson et al., 2007). In the present study, we did not find any such effect. On the contrary, we found remarkable correlations between AWCD and total P ( R2 = 0.70), labile inorganic P (R2 = 0.64), total labile P (R2 = 0.62), soil C to P ratio (R2 = 0.92), and soil N to P ratio (R2 = 0.92), respectively. Therefore, concentrations of labile inorganic P in soil profiles play a critical role in regulating mi crobial community composition. Moreover, labile inorganic P is of importance in determining the C utilization efficiency and organic matter decomposition by microbes, which was evidenced by the fact that it was highly negatively correlated with organic mat ter content (Table 2 -3). Apparently, good

PAGE 44

44 agreement between results of microbial biomass C to organic matter ratio and BIOLOG profiles analysis firmly supported the statement that long -term intensive P fertilization in EAA might stimulate microbial communi ties with higher efficiency of C utilization, which in turn will enhance the organic matter decomposition rates, and, consequently, result in increasing soil subsidence and nutrient regeneration. Indicators of Land Use Changes Indicators have been widely recommended to specify early effects of land use change or nutrient enrichment (Corstanje et al., 2007; Monkiedje et al., 2006). Soil organic matter is one of the key indicators for its role as nutrient sources and impacts on soil physical structure. Organic matter content of soils in the EAA approximates 8090%. Hence, alteration of the organic matter content in the short term is probably not sensitive enough to indicate changes in land use. Instead, our results show that microbial biomass C to organic ma tter ratio was highly distinct across land uses, which indicates substrate availability to soil microorganisms, and can be used as an indicator of prospective alterations in organic matter status along with land use changes in the EAA. Soil enzyme activiti es are generally the most sensitive indicators of changes of belowground microbial communities (Sicardi et al., 2004). In our study, cellobiohydrolase activity exhibited the most variation among land uses, suggesting that this enzyme can be considered as a sensitive indicator of land use changes. Conclusions Long term cultivation and fertilization in the EAA greatly altered soil nutrient distribution and increased organic matter decomposition rates. Sugarcane cropping sequestered more P in inorganic fracti ons and may pose threats to the downstream Everglades ecosystems. Uncultivated soils retained more P in organic fractions.

PAGE 45

45 Uncultivated soil was the most P -limited system and thus had the lowest efficiency of C and N utilization. Soil microbial community s tructure and metabolic diversity significantly changed after variable long-term land management. Turf and forest soils had the highest diversity of C sources and utilization rates of C resources, while the uncultivated soil had the lowest. Soils of forest and turf were close to each other in terms of microbial community composition, but were significantly different from sugarcane or uncultivated. Labile inorganic P played important roles in regulating organic matter decomposition and microbial community com position and function. Our results also support the notion that changes in microbial activity represent a shift in microbial community composition. Microbial biomass C to organic matter ratio and cellobiohydrolase activity was sensitive indicators of alter ations in land uses. Turf soils potentially had a higher rate of soil subsidence and uncultivated soils had the lowest. Land use change from sugarcane cropping to turf grass in EAA is likely to enhance soil subsidence. Nonetheless, land use change from sug arcane cropping to uncultivated tends to slow down the oxidation rates of organic matter and subsequently may minimize soil subsidence.

PAGE 46

46 Table 2 1. Properties of forest, sugarcane, turf and uncultivated soils (0 -15 cm). Forest Sugarcane Turf Uncultivat ed Moisture (%) 49 a 54 a 46 ab 37 b LOI (%) 83 ab 81 a 57 a 85 b TC (g kg 1 ) 449 a 445 a 319 a 461 b TN (g kg 1 ) 30 a 29 a 21 ab 32 b TP (g kg 1 ) 1.25 a 0.98 b 2.71 ab 0.78 c TOC (g kg 1 ) 424 ab 411 a 293 a 433 b DOC (g kg 1 ) 1.13 a 2.34 b 1.14 a 1 .14 a DON (mg kg 1 ) 141 a 197 b 148 ab 132 a NH 4 N (mg kg 1 ) 12 a 10 a 21 b 42 c LIP (mg kg 1 ) 43 a 96 b 173 abc 17 c TLP (mg kg 1 ) 122 a 165 b 236 ab 97 a Notes: LOI, Loss -On -Ignition; TOC, Total Organic C; DOC, Dissolved Organic C; DON, Dissolved Or ganic N; NH4N, Extractable NH4 +; LIP, Labile Inorganic P (NaHCO3extractable P); TLP, Total Labile P; Different letters following numbers indicate significant differences among land uses (p < 0.05) Table 2 2. Microbial biomass, enzymatic activities, po tentially mineralizable N (PMN) and potentially mineralizable P (PMP) of forest, sugarcane, turf and uncultivated soils (0 -15 cm). Forest Sugarcane Turf Uncultivated MBC (g kg 1 ) 12.7 a 13.5 a 13.3 ab 9.3 b MBN (g kg 1 ) 0.24 a 0.16 a 0.32 ab 0.12 b MBP (mg kg 1 ) 52 a 48 a 99 a 41 a MBC/OM (mg kg 1 ) 15183 a 16857 a 23374 b 10918 c PMN (mg kg 1 d 1 ) 10.14 a 2.66 b 10.42 a 12.79 a PMP (mg kg 1 d 1 ) 4.75 a 4.18 a 33.10 a 3.92 a CBHase (mg g 1 h 1 ) 140 a 51 b 191 c 271 abc LAPase (mg g 1 h 1 ) 1.42 a 1.9 6 a 3.64 a 2.58 abc APase (mg g 1 h 1 ) 0.70 a 0.94 a 1.29 ab 1.21 b Sulfatase (mg g 1 h 1 ) 0.32 ab 0.24 a 0.44 ab 0.42 b Notes: MBC, Microbial Biomass C; MBN, Microbial Biomass N; MBP, Microbial Biomass P; OM, Organic matter content; CBHase, Cellobiohydrolase; LAPase, Leucine Aminopeptidase; APase, Alkaline Phosphatase; Different letters following numbers indicate significant differences among land uses (p < 0.05)

PAGE 47

47 Table 2 3. Significant correlations of soil and microbial properties at p < 0.05, n = 1 6. OM Total P LIP MBC/OM MBC MBN MBP CMD AWCD OM 1 Total P 0.61 1 LIP 0.67 0.92 1 MBC/OM -0.72 0.70 0.7 2 1 MBC NS NS NS 0.48 1 MBN NS NS NS NS 0.76 1 MBP NS 0.56 NS 0.58 0.62 0.62 1 CMD 0.57 0.56 NS 0.65 NS NS NS 1 AWCD -0.78 0.70 0.6 4 0.81 NS NS NS 0.88 1 Notes: OM, Organic Matter Content; LIP, Labile Inorganic P; MBC/OM, Microbial Biomass C to Organic Matter Content ratio; MBC, Microbial Biomass C; MBN, Microbial Biomass N; MBP, Microbial Biomass P; CMD, C ommunity Metabolic Diversity; AWCD, Average Well Color Development; NS, not significant at p < 0.05.

PAGE 48

48 Fig 2 1. Microbial community metabolic diversities of forest, sugarcane, turf and uncultivated soils Different letters indicate significant diffe rences between land uses ( p < 0.05). Prin 1Prin 3 Forest Turfgrass Pasture Sugarcane Prin 1Prin 3 Forest Turfgrass Pasture Sugarcane Forest Turfgrass Pasture Sugarcane Fig 2 2. Principal component analysis of community -level physiological profiles from forest, sugarcane, turf and uncultivated soils.

PAGE 49

49 CHAPTER 3 MULTIVARIATE ANALYSI S OF CHEMICAL AND MI CROBIAL PROPERTIES I N HISTOSO LS AS INFLUENCED BY LAND -USE TYPES Introduction Univariate data analysis is essential in many experiments, but it is considered appropriate when just one variable is measured for several samples ( Sena et al., 2002). To better understand whole soil ecosyst em processes, various properties are systematically collected and multivariate analytical methods are employed, which allows analysis of multiple variables simultaneously while interpreting results with better summarized information (Sena et al., 2002). A lthough considered underutilized (Sena et al., 2000), multivariate methods have been well recognized and commonly applied in soil research. For instance, principal component analysis (PCA) has been performed to investigate management impacts on soil quali ty (Wander and Bollero, 1999) and microbial community structure and function (Grayston et al., 2004; Bossio et al., 2005; Allison et al., 2007; Cookson et al., 2007). Recently, application of canonical correlation analysis (CCA) and discriminant analysis (DA) has also been reported for soils research (Zhang et al., 2006; Banning and Murphy, 2008; Sanchez Moreno et al., 2008). Canonical correlation analysis is a method used to assess the dependent relationships between two data sets. The method is design ed to find linear combinations of variables in one data set that account for the most variation in a linear combination of variables for the other data set (Lattin et al., 2003). In this way, much of the relationship between two data sets is detected and visualized. A potential

PAGE 50

50 application for CCA in soil research is the identification of the relationship between soil chemical properties and microbial community structure and function. S oil support s diverse microbial populations and, i n natural systems, m icrobial act ivity is essentially related to the efficiency of nutrient cycling and organic matter turnover (Wright and Reddy, 2001). Yet the richnes s, abundance, and activity of microbial communit ies are influenced by chemical properties such as organic matter content and nutrient availability ( Rutigliano et al., 2004; Ye et al., 2009) Alterations in the chemical conditions of soil s may lead to shifts in microbial community composition and changes in microbial function which is frequently observed upon change in land uses (Nogueira et al., 2006; Cookson et al., 2007; Castillo and Wright, 2008a) The Everglades Agricultural Area (EAA) in south Florida was drained in the early 1900s and converted from wetlands to sugarcane and vegetable cropping. Soils of the EAA are primarily H istosols with high organic matter content, and contain high N yet low P and micronutrient concentrations that require supplemental fertilization (Snyder, 2005; Ye et al., 2009). These soils have undergone subsidence since they w ere drained at rates currently at 1.5 cm/yr (Shih et al., 1998). This has decreased the soil depth to bedrock to the point where major changes in soil chemical properties, such as pH, are becoming problematic and increasingly prohibitive for agricultural use. Sugarcane is the major land use in the EAA and requi r es approximately 3 0 kg P ha1 yr1 and extensive tillage ( Rice et al., 2006). Its l ong-term cultivation typically changes soil chemical properties and microbial community structure and function (G rayston et al., 2004;

PAGE 51

51 Cookson et al., 2007; Castillo and Wright, 2008a, 2008b). Due to economic factors, soil subsidence, and water management concerns, current land use patterns indicate a shift from sugarcane cropping back to the historic seasonally -flo oded wetland prairie ecosystem, tree islands, or pastures (Snyder, 2005). The purpose of this study was to evaluate land use effects on soil biochemical processes using multivariate analytical methods. Land uses under sugarcane cropping and cypress were compared with uncultivated land. Application of multivariate methods was performed to determine (1) whether land uses distinguished by integrated soil chemical properties and microbial parameters (2) which variables contributed most for those differenti ations (3) and whether discriminations in microbial parameters were depend ent on soil chemical properties. Materials and Methods Site Description The study sites are located in the northern EAA near Belle Glade, FL (2639N, 8038W). All soils are Dania muck (euic, hyperthermic, shallow Lithic Medisaprists) with depths to the bedrock of approximately 45 cm. Three land uses with different management history wer e selected for this study: soils under forest for 21 years soils under sugarcane production for approximately 50 years, and reference soils that have never been cultivated since drainage. The forest soils were previously cropped to sugarcane but planted to bald cypress ( Taxodium distichum ) and pond cypress (Tax odium ascendens ) in 1988. T hese did not receive any fertilization after land use

PAGE 52

52 change, but were tilled prior to seedling establishment with no further management applied. The sugarcane soils were managed for vegetable production from the early 1900s to the 1950s, but mainly for sugarcane since the 19 6 0s. Phosphorus fertilization is commonly appl ied at 30 kg ha1 yr1 prior to planting (Rice et al. 2006). Tillage operations included several diskings (to 15 cm depth) after crop harvest, subsoil chiseli ng (to 30 cm depth) to improve drainage, and frequent in-season tine cultivations (to 4 cm depth) for weed control (Morris et al., 2004). The uncultivated soils were primarily occupied by paragrass [ Panicum purpurascens (L.) Raddi ] and bermudagrass [ Cynodon dactylon (L.) Pers] and mowed periodically with residues returned to soil, and received no fertilization and tillage Soil Chemical Properties Surface soil (0 15 cm) samples were collected from four r eplicate sites of each land use in March 2007. The s oils were homogenized after the removal of visible plant particles and stored at 4C Soil organic matter content was measured by the loss onignition method after ashing at 550C for 4 hr (Wright et al., 2008). Total C, total N, and total P were determi ned using the oven dried (70C) soil. Total C and N were measured with a Carlo -Erba NA 1500 CNS Analyzer (Haak -Buchler Instruments, Saddlebrook NJ) while total P was measured after ashing (Bremner, 1996) using the ascorbic acidmolybdenum blue method (Kuo, 1996) with an AQ2+ discrete analyzer (Seal Analytical Inc., Mequon, WI). Labile inorganic N (LIN) was determined by extraction with 0.5 M K2SO4 followed by colorimetric analysis of NH4 (Castillo and Wright, 2008b). Labile

PAGE 53

53 organic N (LON) was calculate d from the difference between total N of the K2SO4 extracts and LIN (Castillo and Wright, 2008b). Labile inorganic P (LIP) was analyzed as described previously after NaHCO3 extraction (Kuo, 1996). The NaHCO3 extracts were measured for total P after Kjeld ahl digestion (Castillo and Wright, 2008b), with labile organic P (LOP) being difference between total extractable P and LIP. Soil Microbial Parameters Microbial biomass Microbial biomass C (MBC), biomass N (MBN) and biomass P (MBP) were measured by the f umigationextraction method (White and Reddy, 2001). The amount of K2SO4extracted C was determined with a Shimadzu TOC -5050A total organic carbon analyzer. The MBC was calculated as the difference in extractable C between fumigated and unfumigated sample s using a conversion factor of 0.37. Dissolved organic C (DOC) was referred to as the total C contained in the unfumigated K2SO4 extract After digestion, the K2SO4 extracts were measured for total Kjeldahl N using an AQ2+ discrete analyzer (Seal Analyti cal Inc., Mequon, WI). The MBN was calculated as the difference in total N between fumigated and unfumigated samples using k= 0.54. The MBP was determined as the difference in total P of NaHCO3 extracts between fumigated and unfumigated soil The P content of the NaHCO3 extracts was measured as pr eviously described (Kuo, 1996).

PAGE 54

54 Potentially mineralizable N and P Potentially mineralizable N (PMN) was determined according to methods of White and Reddy (2000) based on a 10d incubation followed by extraction with 2 M KCl and analysis of NH4. Potentially mineralizable P (PMP) was measured using the method of Corstanje et al. (2007) with slight modifications. Soil (0.5 g) w as placed in 30 -ml serum bottles and mixed with 5 ml double distilled water. The bottle s were then capped and incubated in the dark at 40C for 10 d. At 10 d, 20 m L of 1 M HCl was injected and samples were shaken for 3 h r. Extracts were then filters. Another set of equivalent weight samples, without incubation, were directly extracted with 25 m L of 1 M HCl and extracts analyzed for P. The PMP was determined as the difference in P concentration between extrac ts from incubated and non-incubated soil Enzyme activity assay Four enzyme activities were measured in this study, including glucosidase in the C cycle, leucine aminopeptidase in the N cycle, alkaline phosphatase in the P cycle and arylsulfatase in the S cycle. Approximately 1 g moist soil was placed in polypropylene centrifuge tubes, mixed with 30 m L of water, and shaken for 25 min. Homogenized samples were further diluted 5 times for enzyme assays which were conducted in triplicate with controls to a ssess non enzymatic production. Leucine aminopeptidase assay was conducted in 96well microtiter pl ates (Prenger and Reddy, 2004) with 200 L samples incubated with 50 L of 5 mM L l eucine 7a mino -4 methylcoumarin

PAGE 55

55 (Biosynth Naperville, IL) at 20C for 8 h r. The florescence readings were collected at 1 h r intervals using a Bio -TEK FL600 fluorescence plate reader (Bio -TEK Instruments Inc., Winooski, VT) a t a setting of 365 n m excitation and 450 n m emission. Enzyme activity was determined by calculating th e mean fl u orescent reading changes over time with a standard curve and expressed as mg 7 amino 4 methylcoumarin released per g soil per h r Alkaline phosphatase and sulfatase assays were conducted according to Wright and Reddy, 2001. Community -level physiological profile by BIOLOG assay Community -l evel physiological profiles (CLPPs) were determined by direct incubation of fresh soil extracts in BIOLOG Eco-Plates (31 substrates) (BIOLOG Inc.). Approximately 1 g moist soil was mixed with 20 m L of water and gently shaken for 20 min. The homogenized samples were then diluted 400 times and soil particles allowed settling for 15 min at 4C. 150 L of the supernatant was subsequently dispensed into each well of Eco-Plates and incubated at 20C for 7 d. Optical densities were measured every 6 or 12 hr using Bio -TEK FL600 (Bio -TEK Instruments Inc., Winooski, VT) at 590 n m. Absorbance values of each well with C sources were blanked against control wells before analysis. Negative values were considered as 0. Ave rage well color development (AWCD) was determined as described by Garland (1996; 1997) To overcome possible interference by inoculum density on color development, absorbance values for various C sources were standardized by dividing the blanked value of each well by the AWCD of the plate and were subsequently used for PCA.

PAGE 56

56 Data Analysis T o classify soils by integrated chemical properties, data were standardized to zero mean and unit variance and subject to cluster analysis (CA) according to K means partit ioning method. The importance of a given variable X in defining the differences among clusters was justified by the ratio of between-cluster sum of squares of X to the total sum of squares of X (R2) and the ratio of between-cluster sum of squares of X to within -cluster sum of squares of X [R2/(1 -R2)]. High ratios indicate high importance. Microbial data were analyzed with DA to determine differences among land uses, and stepwise variable selection was conducted to identify the important variables in disc riminating land uses. Due to the existence of missing data, MBP and PMP were not eligible for DA and hence excluded from the analysis. Principal components analysis was used to reduce the numbers of chemical and microbial variables by extracting the most important principal components separately. Canonical correlation analysis was then carried out to investigate the dependent relationship between extracted chemical and microbial principal components The s ignificant level was set at = 0.05. Application of PCA and DA was conducted with JMP 7 (SAS Institute Inc., NC), while CA and CCA were performed with SAS 9.1 (SAS Institute Inc., NC). Results Land Use Effects on Integrated Soil Chemical Properties Soil chemical properties are list ed in Table 3 1. The K means procedure suggested a three-group clustering as the best clustering scheme for these land uses

PAGE 57

57 (Fig. 3 -1). The dendrogram clearly displayed that cluster 1 contained only soils from cypress, cluster 2 included soils solely und er sugarcane production, and cluster 3 exclusively was comprised of uncultivated soils. The cluster of sugarcane soil was closer to cypress than to uncultivated soil. Analysis o f the variances revealed that LIP, LIN, DOC, and total P were the most import ant parameters defining the differences among clusters (Table 3 -2). Additionally, the plot of cluster means across all chemical variables demonstrated that sugarcane, cypress, and uncultivated soils were highly distinguished by total P and LIP (Fig. 3 -2). Variable reduction was made by PCA while retaining as much as possible the original variances. The analysis extracted two principal components, C -Prin1 and C Prin2, representing 54% and 19% of the original variances, respectively. Score plots showed that sugarcane, cypress, and uncultivated soils were distributed separately along ordination axis Prin1, while on axis Prin2 cypress was separated from sugarcane and uncultivated (Fig. 3 -3). Variables with significant loadings on Prin1 were total C (R2 = 0.8 0), total N (R2 = 0.84), DOC (R2 = -0.81), L I N (R2 = 0.81), LO N (R2 = -0.75), and LIP (R2 = -0.95), while total P had high negative loadings on Prin2 (R2 = 0.74) (Fig. 3 4). Land Use Effects on Integrated Microbial Properties Considering the high numbers of variables (31) for BIOLOG data, PCA was first conducted to created new variables representing the carbon utilization patterns. Two variables were extracted (CLPP1 and CLPP2), and each explained 20% of the total

PAGE 58

58 variance in the BIOLOG data set. The new variables were then combined with other microbial parameters and used for DA (Table 3 -3). Canonical plots showed that cypress, sugarcane, and uncultivated soils were significantly different on the first discriminant function (Canonical 1), and the secon d discriminant function (Canonical 2) de monstrated the differences between uncultivated soils and those under sugarcane and cypress (Fig. 3 -5). To identify variables that significantly contributed to the discriminations, stepwise variable selection was conducted. The CLPP2 was the first significant variable selected into the model with a partial R2 of 0.78, and was followed by MBC (R2 = 0.79), MBN (R2 = 0.60) and PMN (R2 = 0.84) (Table 3 -4). Application of PCA extracted two principal components, M -Prin1 and M -Prin2, from the original microbial data, which together explained 68% of the total variance (Fig. 3 -3). Score plots indicated that uncultivated soils were distinct from sugarcane and cypress soils on the ordination axis M -Prin1, while sugarcane and c ypress were separated on axis M -Prin2. Glucosidase exhibited the highest loading (R2 = 0.81) on M Prin1 followed by sulfatase (R2 = 0.80), phosphatase (R2 = 0.77), MBC (R2 = -0.77), PMN (R2 = 0.70), leucine aminopeptidase (R2 = 0.67) and MBN (R2 = -0.65), while CLPP1 exhibited the highest loading on M -Prin2 (R2 = 0.81) (Fig. 3 4). Dependent Relationship between Chemical Properties and Microbial Parameters Canonical correlation analysis was performed with chemical (C -Prin1, 2) and microbial (M -Prin1, 2) pri ncipal components, and two pairs of canonical variates (CVs) were extracted. Canonical correlation between the first chemical canonical variate (C -

PAGE 59

59 CV1) and the first microbial canonical variate (M -CV1) was significant (R = 0.91), with a goodness of fit of p = 0.0006. Approximately 83% of the variance in M -CV1 was explained by C -CV1. The importance of a given principal component for obtaining the maximum correlation between C -CVs and M -CVs was expressed as a standardized canonical coefficient. The coeffi cient of M -Prin1 for M -CV1 was 0.99 and of M -Prin2 was -0.17. Therefore, M -Prin1 gave a greater contribution increasing the M -CV1, while M -Prin 2 contributed less in an opposite way. The coefficients of C Prin1 and C -Prin2 for C -CV1 were 0.81 and 0.58, respectively. The second pair of canonical variates explained 43% of the variance shared by M -CV2 and C -CV2 (R = 0.65, p = 0.03). The canonical coefficients of M -Prin1 and M -Prin2 for M -CV2 were 0.17 and 0.99, respectively, while those of C -Prin1 and C -Pri n2 for C -CV2 were 0.58 and 0.81, respectively. Further redundancy analysis revealed that 42% of total variance in M Prin1 and M -Prin2 was explained by C -CV1 and another 21% by C -CV2. Discussion Land Use Effects on Soil Chemical Properties Cluster analysi s is a method that involves classifying objects into groups so that objects within each group are relatively similar, while objects in different groups are relatively dissimilar (Lattin et al., 2003). This analytical method has been used to identify discr iminations of management effects on soil properties (Sena et al., 2002; Gila et al., 2008; Mic et al., 2008). In the present study, a K means partitioning method was utilized to describe the heterogeneity of EAA soils under different land uses by

PAGE 60

60 integrated analysis using chemical parameters. Soils were clustered into three groups in corresponding to the three land uses (Fig. 3 1), indicating that cypress, sugarcane, and uncultivated soils were highly different in the sense of soil chemistry. Apparently long -term cultivation of the EAA soils resulted in such discrimination. It has been reported that agricultural management poses remarkable impacts on the distribution of C (Wu et al., 2004; Zhang et al., 2006), N (Cookson et al., 2007) and P (Castillo and Wright, 2008b; Wright, 2009). Soils of the EAA are primarily organic and contain high N and low P and micronutrient concentrations Therefore, s ugarcane production requi r es supplemental fertilization and as well as extensive tillage (Rice et al. 2006). Fertilization is likely to promote nutrient accumulation, especially P, in the soil profile (Wright, 2009), while tillage disrupt organic matter and leads to nutrient stratification (Gesch et al., 2007). Analysis of cluster means across all variables further revealed that cypress, sugarcane, and uncultivated soils were highly distinguished by total P and LIP (Table 3 1, Fig. 3 -2), indicating a significant difference in P availability between land uses. It was reasonable since sugarcane soils received long -term P fertilization of approximately 30 kg ha1 y r1, and those of uncultivated did not, whereas the cypress soils received P application up until their establishment. This may also explain why LIP (R2 = 0.91; R2/(1 -R2) = 10.4) was the most importan t variable in defining clusters (Table 3 -2), and why the cluster of cypress was closer to sugarcane rather than to uncultivated soil (Fig. 3 -1).

PAGE 61

61 Land Use Effects on Microbial Parameters Discriminant analysis utilizes the information from a set of variable s to achieve the clearest possible discrimination between or among groups (Lattin et al., 2003). In the present study, DA was applied to differentiate soils under multiple land-uses as well as to determine whether the difference was actually significant. Analysis of discriminant functions clearly showed that sugarcane, cypress, and uncultivated soils had distinctive microbial characteristics (Fig. 3 5). Similar results have also been observed for other ecosystems, in which nutrient availability caused shifts in microbial community structure and function (Allison et al., 2007; Cookson et al., 2007; Matsushita et al., 2007). The stepwise variable selection procedure initially suggested that phosphatase (R2 = 0.63), sulfatase ( R2 = 0.51), glucosidase ( R2 = 0.62), PMN ( R2 = 0.67), MBC ( R2 = 0.68), MBN (R2 = 0.57), and CLPP2 ( R2 = 0.78) were all significant and eligible to enter the discriminant model. However, only CLPP2, MBC, MBN, and PMN were finally included, which indicated their great significance to th e discrimination (Table 3 4). In other words, cypress, sugarcane, and uncultivated soils were highly distinguished by those four variables. Considering the fact that CLPP2, MBC, and MBN characterized the microbial community composition and population siz e, it may be true that microbial community structure, rather than function, was more sensitive to agricultural management. This concept was also supported by other studies (Bossio et al., 2005; Zhang et al., 2006)

PAGE 62

62 Variable Reduction Principal component analysis is a use ful method for dimension reduction (Lattin et al., 2003) as it allows for re express ion of data with fewer variables while capturing as much of the available variance as possible. The objective of variable reduction is to make data analys is more manageable and straightforward. Application of PCA in the present study successfully reduced nine chemical variables to two principal components and was able to capture 73% of the original variance, while nine microbial parameters were reduced to two principal components explaining 68% of the total variance. The reduction increased the sample size to variable ratio and made subsequent CCA analysis and interpretation easier. In addition, score plots showed that cypress, sugarcane, and uncultivated soils were dispersed differently along ordination axes (Fig. 3 -3), indicating their significant difference in both soil chemistry and microbial community structure and function, which further supports the aforementioned results of CA and DA. Principal component loadings determine the amount of variance of a given variable accounted by principal components (Lattin et al., 2003) Our results demonstrated that LIP had the highest loadings on C Prin1 (R2 = -0.95), while total P was the only variable that had significant loadings on C -Prin2 (R2 = 0.74), which is in agreement with the previous statement that LIP and total P significantly contributed to the discrimination in soil chemistry (Figs. 3 2 and 3 4).

PAGE 63

63 Dependent Relationship between Chemical Properties and Microbial Parameters Microbial communities are i n close contact with soil micro environments, and therefore are easily subj ected to change following alter ation of soil chemical properties (Corstanje et al. 200 7 ). Thus land use patterns are likely to affect the microbial communit y structure and function, which can be described by the changes in microbial parameters such as respiratory capacities microbial biomass and extracellular enzymatic activities (Wright and Reddy, 2001; Castillo and Wright, 2008 a ). In the present study, multivariate data analysis demonstrated the contrasting soil chemistry and distinctive microbial community composition and function across soils under different land uses (Figs. 3 -1 and 3 5). It was likely that discriminations i n microbial parameters were directly connected to variations in soil chemical properties, and CCA suggested that the dependent relationship was indeed significant. Changes in soil chemical properties resulted in alterations in microbial community composit ion and function. Nutrient availability often plays a critical role in regulat ing the microbial community structure and function (Cookson et al., 2007; Corstanje et al. 2007 ; Wright and Reddy, 2008 ). Cluster analysis suggested that LIP and total P were the most important variables discriminating the integrated chemistry of cypress, sugarcane, and uncultivated soils (Table 3 2, Fig. 3 -2), which was further supported by the results of PCA (Fig. 3 4). In consideration of the fact that Everglades soils are historically P limited (Wright and Reddy, 2008), it was likely that P availability was the major factor

PAGE 64

64 that controlled the microbial communit ies of these land uses, which indeed are P limited (Table 3 1). S oil chemical propert ies and the micro bial commu nity are two key components in ecosystem function (Finlay et al., 1997 ). Net results of chemical microbial interactions define the way ecosystem s function. Factors capable of interrupting or modifying the interactions are likely to cause significant ecol ogical impacts. Our results suggested that a gricultural cultivation has drastically altered soil chemical properties with in the EAA, especially P availability Intensive application of P fertilizers is likely to stimulate microbial communities, which in turn would enhance organic matter decomposition rates and contribute to greater rates of soil subsidence. Thus, future P fertilization should be evaluated well for its potential long -term impact on microbial activity, which in turn may affect the sustainab ility of different land uses of these subsiding soils. Conclusions Longterm cultivation in the EAA significantly altered nutrient distribution and availability for different land uses, especially for P cycling. Correspondingly, soil microbial community c omposition and function was modified, and applications of CA and DA successfully described the alterations. Canonical correlation analysis clearly demonstrated a significant dependence relationship between chemical properties and microbial community composition and function. Phosphorus availability was one of the major factors regulating the soil microbial activity and function for the land uses. Intensive application of P fertilizer is likely to stimulate microbial community and

PAGE 65

65 subsequently increase soil oxidation and subsidence. Future land use changes in the EAA should consider effects of P on the functioning of microbial communities and their control of nutrient cycling.

PAGE 66

66 Table 3 1. Chemical properties of cypress, sugarcane, and uncultivated soils (0 15 cm) with standard error values, n = 4. Unit Cypress Sugarcane Uncultivated Loss On Ignition % 83 (3) 81 (2) 85 (1) Total C g kg 1 449 (3) 445 (2) 461 (2) Total N g kg 1 30 (0.5) 29 (0.3) 32 (0.3) Total P g kg 1 1.3 (0.07) 1.0 (0.01) 0.8 (0.01) Dissolved organic C g kg 1 1.1 (0.1) 2.3 (0.1) 1.1 (0.2) Labile organic N mg kg 1 141 (11) 197 (8) 132 (16) Labile inorganic N mg kg 1 12 (1) 10 (1) 42 (5) Labile organic P mg kg 1 79 (5) 69 (4) 80 (4) Labile inorganic P mg kg 1 43 (9) 96 (4) 17 (3) Table 3 2. The ratio of between-cluster sum of square to total sum of square (R2) and the ratio of between -cluster sum of square to within -cluster sum of square (R2/(1 -R2)) for chemical variables in defining differences among clusters, n = 12. R 2 R 2 /(1 R 2 ) Loss On Ignition 0.21 0.26 Total C 0.72 2.60 Total N 0.63 1.72 Total P 0.86 6.10 Dissolved organic C 0.86 6.30 Labile organic N 0.65 1.83 Labile inorganic N 0.87 6.49 Labile organic P 0.30 0.43 Labile inorganic P 0.91 10.43

PAGE 67

67 Table 3 3. Micr obial parameters in cypress, sugarcane, and uncultivated soils (015 cm) with standard error values, n = 4. Unit Cypress Sugarcane Uncultivated Microbial biomass C g kg 1 13 (1.0) 14 (0.5) 9 (0.6) Microbial biomass N g kg 1 0.2 (0.04) 0.2 (0.01) 0.1 (0. 01) Potentially mineralizable N mg kg 1 d 1 10 (2.5) 3 (0.3) 13 (1.6) Leucine aminopeptidase mg kg 1 d 1 1.4 (0.4) 2.0 (0.2) 2.6 (0.7) Phosphatase mg g 1 h 1 0.7 (0.12) 1 (0.08) 1 (0.08) Sulfatase mg g 1 h 1 0.3 (0.04) 0.2 (0.04) 0.4 (0.05) Glucosidas e mg g 1 h 1 0.2 (0.01) 0.1 (0.02) 0.3 (0.04) CLPP1 None 1.2 (1.0) 1.6 (0.4) 0.4 (1.8) CLPP2 None 2 (0.7) 1 (0.2) 3 (0.9) CLPP1 and 2, first and second principal components extracted from datasets of community level physiology profiles. Table 3 4. S tepwise discriminant model for differentiating cypress, sugarcane, and uncultivated soils based on microbial parameters, n = 12. Variable Partial R 2 F value P value CLPP2 0.78 15.9 0.001 Microbial biomass C 0.79 15.3 0.002 Microbial biomass N 0.60 5.3 0 .039 Potentially mineralizable N 0.84 16.0 0.004 CLPP2, second principal component extracted from datasets of community level physiology profiles.

PAGE 68

68 1 2 3 1 2 3 Fig 3 1. Dendrogram from K means cluster method applied to soil chemical data.

PAGE 69

69 2 1 0 1 2 LOI TC TN TP DOC NH4LON LIP LOPCluster MeansCypress Sugarcane Uncultivated 2 1 0 1 2 LOI TC TN TP DOC NH4LON LIP LOPCluster MeansCypress Sugarcane Uncultivated Fig 3 2. Cluster m eans across chemical variables. LOI, loss on-ignition; TC, total C; TN, total N; TP, total P; DOC, dissolved organic C; NH4, labile inorganic N; LON, labile organic N; LIP, labile inorganic P; LOP, labile organic P.

PAGE 70

70 Chemical Properties Microbial Properties 4 3 2 1 0 1 2 3 4Prin 2 19% 4 3 2 1 0 1 2 3 4 Prin 1 54% 4 3 2 1 0 1 2 3 4Prin 2 18% 4 3 2 1 0 1 2 3 4 Prin 1 50% Cypress Sugarcane Uncultivated Cypress Sugarcane Uncultivated Chemical Properties Microbial Properties 4 3 2 1 0 1 2 3 4Prin 2 19% 4 3 2 1 0 1 2 3 4 Prin 1 54% 4 3 2 1 0 1 2 3 4Prin 2 18% 4 3 2 1 0 1 2 3 4 Prin 1 50% Cypress Sugarcane Uncultivated Cypress Sugarcane Uncultivated Fig 3 3. Score plots of principal components analysis on soil chemical properties and microbial parameters.

PAGE 71

71 0.5 0.0 0.5 Component 2 LOI TC TN TP DOC NH4 LON LIP LOP 0.5 0 0.5 Component 1 1.0 1.0 1.0 1.0 Chemical Properties 0.5 0.0 0.5 Component 2 LOI TC TN TP DOC NH4 LON LIP LOP 0.5 0 0.5 Component 1 1.0 1.0 1.0 1.0 Chemical Properties 1.0 1.0 1.0 0.5 0.0 0.5 Component 2 LAPase APA Sulfatase Glucosidase CLPP1 CLPP2 MBC MBN PMN 0.5 0 0.5 Component 1 1.0 Microbial Properties 1.0 1.0 1.0 0.5 0.0 0.5 Component 2 LAPase APA Sulfatase Glucosidase CLPP1 CLPP2 MBC MBN PMN 0.5 0 0.5 Component 1 1.0 Microbial Properties Fig 3 4. Loading plots of principal components analysis on soil chemical properties and microbial parameters. LOI, loss on-ignition; TC, total C; TN, total N; TP, total P; DOC dissolved organic C; NH4, labile inorganic N; LON, labile organic N; LIP, labile inorganic P; LOP, labile organic P ; MBC, microbial biomass C; MBN, microbial biomass N; MB P microbial biomass P; LAPase, leucine aminopeptidase; APA, alkaline phosphatase; CLPP1, 2, first and second principal components extracted from datasets of community -level physiology profiles.

PAGE 72

72 0 5 10 15 20Canonical 2 Cypress Sugarcane Uncultivated 0 5 10 15 20 25 30 35 40 45 Canonical 1 0 5 10 15 20Canonical 2 Cypress Sugarcane Uncultivated 0 5 10 15 20 25 30 35 40 45 Canonical 1 Fig 3 5. Canonical plots of discriminant analysis on soil microbial parameters.

PAGE 73

73 CHAPTER 4 SULFURINDUCED CHANGES IN P HOSPHORUS DISTRIBUTI O N IN EVERGLADES AGRICULTU RAL AREA SOILS Introduction The Everglades Agricultural Area (EAA) in south Florida was drained in the early 1900s and converted to sugarcane and vegetable cropping. The EAA primarily consists of Histosols with high organic matter content, approximately 85% by weight, which contain high N yet low P and micronutrient concentrations that require supplemental fertilization (Snyder 2005; Castillo and Wright 2008). Due to the conversion of land use from seasonally -flooded wetlands to agricultural use, oxidation or subsidence of the drained peatlands has occurred at a rate currently approximating 1.5 cm yr1 (Shih et al. 1998). Consequently, the depth of the soil has declined considerably to the point of causing significant interactio n with the underlying bedrock. Cultivation of these drained peatlands, specifically the use of tillage, has resulted in incorporation of bedrock CaCO3 into soil, which has gradually increased the pH since drainage from the historic 5.05.5 to approximatel y 7.0 7.5 today (Snyder 2005; Gabriel et al. 2008). Subsequently, these soil pH increases have decreased P and micronutrient availability to crops and necessitated new fertilizer management practices. Sugarcane is the dominant crop grown in the EAA, but it requires approximately 30 kg P ha1 year1 and extensive tillage for pre plant preparation and weed control (Rice et al. 2006). Long -term P application has resulted in P accumulation in soil profile and as well as export into Everglades wetlands thr ough canal systems, which was a major factor contributing to the

PAGE 74

74 deterioration of water quality and alterations of the Everglades wetland ecosystem (Childers et al. 2003). An understanding of P transformations and distribution in soil is necessary to ma ximize efficient P use by sugarcane, while minimizing potential export from fields into adjacent wetlands. Fractionation schemes have been developed to determine the P distribution and allocation in different pools related to their degree of recalcitrance (Reddy et al. 1998; Harrell and Wang, 2007). Such schemes assume that different extractants selectively extract discrete P chemical forms sequentially (Adhami et al. 2007). Though the assumption and procedure is operationally defined, the methods prov ide a convenient way to characterize the availability and mobility of P in soils and to assess their impacts on the environment (Maguire et al. 2000). Soils with high labile P content indicate high P availability to plants and also potential export by le aching or runoff. Inorganic P associated with CaCO3 or bound to Fe and Al oxides are considered relatively stable, but may be susceptible to dissolution and regeneration upon change in environmental conditions (Castillo and Wright 2008). Phosphorus asso ciated with organic pools is unstable in these drained peatlands due to organic matter oxidation and subsequent P mineralization (Wright 2009). Several factors are capable of influencing P stability and mobility in the soil profile including pH, microbial activity, and soil amendments (Arai et al. 2005; Jaggi et al. 2005). Elemental S is occasionally applied in the EAA as soil amendment for the purpose of reducing pH and therefore increasing P availability to crops (Gabriel,

PAGE 75

75 et al. 2008). The microbia l oxidation of elemental S to SO4 produces acidity which reacts with the soil and reduces pH, which in turn releases P bound to Ca and Fe minerals into soil solution. However, the buffering capacity of these calcareous histosols is strong and can countera ct the acidifying effects of S oxidation, thus effects of amendments are temporary and may need to be repeated each growing season (Beverly and Anderson, 1986). Problems with large scale S application include a potential pulsed P release from soils that m ay pose runoff or leaching hazards into proximal aquatic ecosystems (Santoso et al. 1995). Additionally, increased nutrient availability resulting from pH reduction can potentially stimulate the microbial population to decompose organic matter and increa se soil oxidation rates. Everglades wetlands of south Florida are traditionally P limited and sensitive to small increases in P loading (Noe et al. 2001). Reducing P export from the EAA is critical to fulfilling the emerging interests of protecting wat er quality and restoring south Florida ecosystems. Due to the increases in pH and the decreasing depth to bedrock of soils in the EAA, use of S application to counteract the rising pH may increase in the future. Therefore, a better understanding of how S influences pH and P distribution and availability within various pools during the sugarcane growing season is essential and the objective of this study.

PAGE 76

76 Materials and Methods Site Description The experimental site is located in the central EAA on Dania muck (euic, hyperthermic, shallow Lithic Haplosaprist) with a depth to bedrock of approximately 45 cm. The experimental design was a randomized complete block with four S application rates and four field replications. Each field plot measured 9 m x 13 m an d consisted of 6 rows of sugarcane ( Saccharum officinarum ) Elemental granular S (90%) was applied at rates of 0, 112, 224, and 448 kg S ha1 to the furrow and covered after planting sugarcane in the furrow. Other fertilization was provided using typical guidelines for this region (Rice et al. 2006). All fertilizers were soil applied just prior to planting and included 17 kg N ha1 and 37 kg P ha1 as monoammonium phosphate, 228 kg K ha1 as KCl, 8.5 kg Mn ha1, 4.5 kg Cu ha1, 5.6 kg Fe ha1, 2.8 kg Zn ha1, and 1.1 kg B ha1. All plots received typical cultural practices including cultivation and herbicide application. Water was applied via seepage irrigation in field ditches approximately 182 m apart. Sugarcane cultivar CP 89-2143 was planted in November 2007 and harvested in February 2009. Soil Sampling and Analysis Soil samples were collected before planting and fertilizer application and then in January 2008, May 2008, August 2008, and December 2008, corresponding to approximately 0, 2, 6, 9, a nd 13 months after planting, respectively. Twelve soil (0 15 cm) cores (2.54 cm diameter) were randomly collected from rows within each field

PAGE 77

77 plot and composited. Soils were homogenized after the removal of visible plant residues and stored at 4 C. Soi l pH was measured with a soil to water ratio of 1:3 after equilibration for 30 min. Total organic C was measured by loss on-ignition at 550C for 4 hr after conversion to organic C with a coefficient factor of 0.51 (Wright et al. 2008). Total organic N was measured by Kjeldahl digestion followed by NH4 analysis (Bremner 1996). Extractable NH4-N and NO3-N were determined by extraction with 2 N KCl followed by colorimetric analysis (Castillo and Wright 2008). Acetic acid extractable nutrients were measured according to guidelines for muck soils (Sanchez 1990) by extracting 4 g soil with 25 mL 0.5 N acetic acid for 1 hr, then filtering through Whatman #42 filters. Extracts were analyzed for Ca, Mg, Fe, and Al concentrations by ICP (Perkin -Elmer, Waltham MA) using EPA method 200.7. Select soil nutrient concentrations and properties before S application are listed on Table 4 1. Soils underwent a sequential chemical P fractionation procedure (Reddy et al. 1998; Wright 2009 ). Approximately 1 g soil was extracted with 25 mL water for 1 hr, passed through 0.45 m membrane filters, and analyzed for P (labile P). The remaining samples were extracted with 25 mL of 0.1 N NaOH for 17 hr, filtered and analyzed for Fe-Al bound P (Fe -Al -P), followed by the extract ion of remaining samples with 25 ml of 0.5 N HCl for 24 hr and analysis of Cabound P (Ca P). The remaining samples were further digested with 6 N HCl for 1 hr at 150C and analyzed for residual P. Three mL of NaOH extracts was digested with 11 N H2SO4 f or 4 hr at 350C and analyzed for NaOH -TP. The humic -fulvic acid fraction was

PAGE 78

78 calculated by subtraction of NaOH -P i from NaOH -TP. Phosphorus concentrations of extracts were measured using the ascorbic acidmolybdenum blue method (Kuo 1996) with an AQ2+ d iscrete analyzer (Seal Analytical Inc., Mequon, WI). Statistical Analysis A mixed model was fit using restricted maximum likelihood in the MIXED procedure of SAS (Littell et al., 2006). The fixed effects were S application rate, time and their interaction. Block was a random effect. Degrees of freedom were adjusted using the Kenward-Roger adjustment. An exponential covariance structure was used to model the correlation among observations taken from the same plot over time. Significant differences amon g individual treatments and time intervals were regression analysis was employed to determine relationships among P fractions and soil properties. All statistical analysis was carried out with SAS 9.1 (SAS Institute). Results and Discussion Soil pH Sulfur application within the range of 0 to 448 kg S ha1 did not significantly reduce soil pH ( p = 0.14) (Fig. 4 -1). In the present study, the background soil pH prior to S application was 6.2, which was not significantly different from soils collected at any time of the growing season. However, soil pH dropped slightl y at 2 months after S addition in the highest S application rate and increased thereafter from 6.0 to 6.4 at the end of growing season, which indicated a lagged effect of soil

PAGE 79

79 buffering. Additionally, no interaction effect of S application rate and sampling time (p = 0.58) was found. The limited effect of acidification may result from the relatively low rates of S application and from the high buffering capacity of this calcareous organic soil (Bloom 2000; Jaggi et al ., 2005; Deubel et al. 2007). Soils with high concentrations of carbonates and bicarbonates are highly buffered against acidification (Bloom 2 000; Rogovska et al. 2007). The buffering effects often take place more slowly than the formation of sulfuric acid, therefore, a re-increase of soil pH is possible (Bloom 2000; Deubel et al. 2007). A limited reduction in soil pH after S application wa s observed in other studies of calcareous soils (Hassan and Olson 1966). Phosphorus Distribution Two way ANOVA demonstrated that the time effect was significant across all P fractions indicating significant seasonal changes in P distribution (Table 4 2) The main effect of S application rate was solely found significant on fractions of labile P and Fe-Al -P. No interaction effect of S application rate and time was observed for any of the P fractions. Labile P Labile P is commonly considered the most bi ological available form of P and consistently represented the smallest fraction of total P throughout the growing season, decreasing from 1.1% to 0.3% from 2 to 13 months (Fig. 4 -2). Similar results were also observed in other studies (Maguire et al. 200 0; Castillo and Wright 2008). The concentration of labile P decreased from 15 to 3 mg P kg1 from

PAGE 80

80 2 to 13 months after S application (Fig. 4 3). Phosphorus in labile form is highly mobile, unstable, and prone to loss either by leaching, runoff, or plant uptake, which explains the seasonal decrease observed in this study. An increase in available P as a result of decreased soil pH has been documented in other studies (Deluca et al. 1989; Codling, 2008). In the present study, labile P was significantly higher in soils receiving 448 kg S ha1 (13 mg P kg1) when compared to soils receiving 112 ( 6 mg P kg1) and 224 kg S ha1 (6 mg P kg1) (Fig. 4 -3). There are two primary mechanisms by which S influences P availability: lowering of soil pH (Gabriel et al. 2008) and replacement of PO4 with SO4 and release of PO4 from association with Fe, Al, and Ca (Jaggi et al. 2005). The significant positive correlations between extractable SO4 and labile P (Table 4 3) were likely indicative of a partial stimulatory effect of SO4 on the release of labile P. Statistical analysis also revealed significant correlations of labile P to Fe -Al -P (R2 = 0.87), Ca -P (R2 = 0.49), humic -fulvic acid P (R2 = 0.36), and soil pH (R2 = -0.33), suggesting possible P replenishment from other pools to the labile fraction (Table 4 3). Yet, Fe-Al -P (R2 = 0.77) was the major component in explaining the variance in labile P, w hile Ca -P and humic -fulvic acid P contributed to a better prediction. Labile P = 0.51 + 0.35 (Fe Al -P) + 0.01 ( Ca P) 0.07 (Humic -fulvic acid P) (R2 = 0.84, Cp = 3, p <0.0001) Fe Al bound P The Fe-Al fraction contains P associated with amorphous and crystalline Fe and Al oxides (Arai et al. 2005). Therefore, it was not surprising to find that Fe -Al -P

PAGE 81

81 was strongly c orrelated to Fe and Al content (Table 4 -3), a finding that was also observed in other studies (Ryan et al. 1984; Maguire et al. 2000). Phosphorus concentrations in the Fe-Al bound fraction displayed a clear declining pattern during the growing season (F ig. 4 -3). Concentrations decreased steadily from the beginning to the end of the season, averaging 62, 48, 41 and 34 mg P kg1, respectively, for 2, 6, 9, and 13 months after S application. This observed decrease indicates that the Fe -Al bound P was a major source of P for sugarcane. Throughout the season, this fraction had the second lowest contribution to total P, averaging 4% (Fig. 4 -2). In acidic soils, the Fe -Al -P is frequently the dominant fraction involved in P retention (Mozaffari and Sims 1996 ). However, the calcareous nature of this organic soil is likely to encourage more P sequestration in the Ca-bound rather than the Fe-Al bound fraction. Nonetheless, Fe and Al oxides play important roles in controlling P chemistry in soils with high CaCO3 content (Halajnia et al. 2009). Other researchers also suggested that Fe and Al help to control P loss by leaching and runoff (Arai et al. 2005; Harrell and Wang, 2007). The pools of Fe-Al P were not affected by S application rates from 0 to 224 k g S ha1 (Fig. 4 3). However, P concentrations in this fraction were significantly higher in soils amended with 448 kg S ha1, averaging 59 mg P kg1, than those receiving lower S rates. The Pearson correlation coefficient between Fe-Al -P and soil pH was significantly negative (R2 = 0.54) (Table 4 3), indicating that a small decrease in soil pH is likely to encourage P retention in this fraction. Inversely, an increase in soil pH may potentially promote P desorption from fixation sites (Gessa et al. 20 05), which

PAGE 82

82 may explain the continuous losses of Fe-Al -P over time along with increasing soil pH (Figs. 4 2 and 4 -3). Multiple regression analysis, in addition to correlation analysis, showed that 88% of the variance in Fe-Al -P could be explained by labile P, humic fulvic acid P, and extractable Fe concentrations. Fe -Al -P = 25.4 + 2.0 (labile P) + 0.2 (humic -fulvic acid P) + 2.2 (Fe) (R2 = 0.88, Cp = 4, p < 0.0001) Ca -bound P Phosphorus stocks in the Ca bound fraction were much higher than those of labile and Fe-Al fractions, contributing 2835% of the total P (Fig. 4 -4) as a result of high Ca concentrations in the soil. Sugarcane cropping in the EAA requires multiple tillage applications prior to and during the growing season. This consequently results in the incorporation of the bedrock CaCO3 into soil and promotes P retention in Ca-bound fractions (Castillo and Wright 2008 b ). Correlation analysis revealed that Ca -P concentration was significantly correlated to soil Ca (R2 = 0.56), Mg (R2 = 0.40), and Mn (R2 = 0.47) concentrations indicating that some portion of P in this soil also exists as Mg -P and MnP. Association of P compounds with Mn in highly calcareous soils has recently been reported (Adhami et al. 2007), as such P compounds may originate f rom hureaulite [Mn5H2(PO4)4.4H2O] and reddingite [Mn3(PO4)2.2H2O]. The size of the Ca -P fraction declined gradually during the growing season from 454 mg P kg1 to 301 mg P kg1 by 13 months (Fig. 4 5). Previous studies suggested that under cultivated conditions Ca P, rather than

PAGE 83

83 organic fractions, represents a stable P pool (Zhang and MacKenzie 1997; Castillo and Wright 2008). The Ca P fraction is considered relatively stable under alkaline condition, but unstable under acidic conditions. In the present study, S application did not impact the size of the Ca P fraction as a result of the limited reduction in soil pH, which implied that Ca P in this calcareous soil was not sensitive to slight changes in pH as was the Fe -Al -P and labile P fractions (Hassan and Olson, 1966). Humic -fulvic acid P The organic pools are primarily comprised of humic -fulvic acids and the more recalcitrant residual fraction (Turner et al. 2005). Sulfur application did not influence P concentrations in this fraction at any sampling time or alter its proportion to total P. Phosphorus concentrations in the humic -fulvic acid fraction fluctuated during the growing season (Fig. 4 5). The percentage of humic -fulvic acid P to total P followed the same pattern, varying from 12% to 17% (Fig. 4 -4). Phosphorus distribution in this fraction was negatively correlated to labile P, Ca-P and residual P fractions (Table 4 3). Residual P Residual P was the most abundant P fraction for all sampling times, accounting for 47 51% of the total P ( Fig. 4 4). This fraction is considered unavailable to crops since the P is in organic form that must be decomposed before becoming available. However, due to subsidence, P contained in this residual fraction will eventually be made available to crops. S ulfur application did not alter P concentrations in this

PAGE 84

84 fraction or the contribution of this fraction to total P. Phosphorus concentrations in this fraction fluctuated during the growing season in a contrasting pattern to the humic -fulvic acid fraction, averaging 565, 501, 598 and 547 mg P kg1, respectively, for 2, 6, 9 and 13 months after S application (Fig. 4 5). Phosphorus Distribution and Availability in S Amended Soils Phosphorus concentrations in the top 15 cm of soil averaged 1244 mg P kg1 ini tially and decreased significantly to an average of 1066 mg P kg1 at the end of the sugarcane growing season. Soils are capable of releasing P constantly into solution over a long period of time (Arai et al. 2005) due to release of mineral -bound P and d ecomposition of organically -bound P. However, P release varies with soil properties (Zhou and Li 2001). Active P fractions, such as labile P, Fe -Al -P, and Ca -P, comprise of the primary pools of desorbable P in weekly acidic and calcareous soils (Maguire et al. 2000). In the present study, the size of the inorganic P pools decreased by 13 months and consequently contributed to the decline in total P concentrations and the percentage of inorganic P to total P (Fig. 4 -6). The P concentrations in this soi l were a net result of the long -term balance between inputs (fertilization, rainfall and soil oxidation) and export (leaching, runoff, and sugarcane uptake). At the current rate of soil oxidation, approximately 6090 kg P ha1 is generated annually (Wrigh t 2009), which is greater than typical P fertilization rates to sugarcane (Rice et al. 2006) and P removed as harvested biomass (23 kg P ha1) (Coale et al. 1993). The significant reduction in total P indicates subsidence is a major source of P in runo ff from EAA fields, which contributes to deterioration of

PAGE 85

85 water quality and modifications to proximal aquatic ecosystems (Childers et al. 2003). Therefore, land management that minimizes soil subsidence is likely to reduce the potential for P export. Th e majority of P in this soil was retained in organic forms (Figs. 4 2 and 4 -4), indicating the susceptibility of this soil to oxidation under drained conditions typical of sugarcane production in this region (Castillo and Wright 2008; Wright 2009). Sulf ur application at 448 kg S ha1 promoted P accumulation in labile P and Fe Al -P fractions (Fig. 4 3) suggesting increased P availability to crops and as well as potential increased risk of P export (Codling, 2008). However, no significant effects of S add ition were found on total P and total inorganic and organic P concentrations. It was likely that the effects of S amendment on labile P and Fe-Al -P were confounded by the noninfluential effects of sulfur on Ca-P, and considering the fact that Ca -P was th e dominant inorganic fraction. A declining trend in pH for soils receiving 112 to 448 kg S ha1 corresponded to a declining trend in the size of Ca-P pool (Figs. 4 4 and 4 -5). It is reasonable to postulate that application rates beyond 448 kg S ha1 woul d continue to reduce soil pH and reach a point causing significant releases of P from the Ca bound fraction (Gessa et al. 2005), which may be a cause of concern since Ca-P comprised 32% of total P and more than 80% of total inorganic P in this soil (Fig. 4 4). Nonetheless, using the current recommended S application guidelines and rates, the risk of P export from the Ca-bound P would be minimal. However, due to the increasing pH problem for EAA soils, there may be a need for greater S application rates i n the future, which may overcome the soils

PAGE 86

86 buffering capacity and in fact release large amounts of P from the Ca-bound pool and therefore pose an environmental hazard to nearby aquatic ecosystems. Conclusions Organic P was the major form of P in this soil averaging 63% of total P, while the Ca P fraction dominated the inorganic pools, contributing 32% of total P. Total P concentrations in the surface soil decreased significantly at the end of growing season as a result of considerable reduction in inorganic P, especially labile P and Fe -Al -P, which comprised of the majority of available P for crops. Under current sugarcane production, organic P in this soil is susceptible to oxidation and a potential source for P loss. Application of S at rates up to 448 kg S ha1 introduced limited effects on reduction in soil pH, yet a small decrease in soil pH promoted P accumulation in labile and Fe -Al bound fractions, which increased P availability and as well as the risk of P export from these two fractions. The p ool of Ca-P was relatively stable under current S application guideline and rates. Higher S rates than currently recommended may overcome the soils buffering capacity and consequently release large amounts of P from the Ca bound pool and pose an environm ental hazard.

PAGE 87

87 Table 4 1. Chemical properties of the Dania soil in the Everglades Agricultural Area before fertilizer application. Numbers in parenthesis is Soil Property Unit Concentration Total organic C g kg 1 416 (3) Total N g kg 1 38 (0.6) Total P mg kg 1 850 (9) Extractable NO 3 N mg kg 1 290 (39) Extractable NH 4 N mg kg 1 16 (3) Extractable Ca mg kg 1 720 (49) Extractable Mg mg kg 1 105 (5) Extractable Fe mg kg 1 13 (1) Extractable Al mg kg 1 1.1 (0.2) Table 4 2. Two way ANOVA on different pools of P in soils amended with variable S application rates. Variable P Value Treatment Time Interaction Labile P i 0.0 1 4 0.00 4 0.2 34 Fe Al bound P 0.0 0 4 < 0.0001 0.069 Ca bound P 0.272 0.0 22 0. 8 5 8 Humic fulvic acid P 0. 350 0.514 0. 397 Residual P 0.1 45 <0.0001 0. 275 Total P 0. 766 0.01 1 0. 8 6 8

PAGE 88

88 Table 4 Labile P i Fe -Al -P Ca -P Humic fulvic acid P Residual P Ca Mg Al Fe Mn pH Labile Pi 1 Fe Al P 0.87 1 Ca P 0.49 0.34 1 Humic fulvic acid P 0.36 NS 0.39 1 Residual P NS NS 0.60 0.58 1 Ca 0.26 NS 0.56 0.53 0.65 1 Mg 0.40 0.37 0.40 0.44 0.25 0.66 1 Al 0.62 0.68 NS 0.29 NS NS 0.46 1 Fe 0.36 0.54 NS 0.27 NS NS 0.78 0.57 1 Mn 0.38 0.36 0.47 0.49 0.26 0.64 0.88 0.38 0.78 1 pH 0.33 0.54 NS NS 0.52 0.30 0.3 0.51 0.50 NS 1 Ca, Mg, Al, Fe, and Mn = acetic acidextractable concentrations; NS = not significant.

PAGE 89

89 5.0 5.5 6.0 6.5 7.0 2 6 9 13 Months after S Application 0 112 224 448 5.0 5.5 6.0 6.5 7.0 2 6 9 13 Months after S Application 0 112 224 448 0 112 224 448 Fig. 4 1. Soil pH changes in response to different S application rates (0, 112, 224, and 448 kg S ha1) throughout the sugarcane growing season. Error bars represent the standard error of the mean. 0 1 2 3 4 5 6 7 Labile P Fe Al PPercentage of Total P2 6 13 9 0 1 2 3 4 5 6 7 Labile P Fe Al PPercentage of Total P2 6 13 9 Fig 4 2. The percentage of labile P and Fe-Al bound P of soil total P throughout t he sugarcane growing season. Error bars represent the standard error of the mean.

PAGE 90

90 0 5 10 15 20 25 30 35 40 Months after S ApplicationLabile P (mg kg1)0 112 224 448 0 20 40 60 80 100 120 2 6 913Fe Al bound P (mg kg1) 0 5 10 15 20 25 30 35 40 Months after S ApplicationLabile P (mg kg1)0 112 224 448 0 112 224 448 0 20 40 60 80 100 120 2 6 913Fe Al bound P (mg kg1) Fig 4 3. Concentrations of labile P and Fe Al bound P in soils 2, 6, 9, and 13 months after S application for different rates (0, 112, 224, and 448 S kg ha1). Error bars represent the standard error of the mean.

PAGE 91

91 0 10 20 30 40 50 60 Ca bound P Humic fulvic acid P Residual PPercentage of Total P2 6 9 13 0 10 20 30 40 50 60 Ca bound P Humic fulvic acid P Residual PPercentage of Total P2 6 9 13 Fig 4 4. The percentage of three P fractions to total P. Due to lack of significant effects of S, data for S application rates were averaged for presentation. Error bars represent the standard error of the mean. 0 100 200 300 400 500 600 700 0 2 4 6 8 10 12 14 Months after S ApplicationPhosphorus Concentration (mg kg1) Ca bound P Humic fulvic acid P Residual P 0 100 200 300 400 500 600 700 0 2 4 6 8 10 12 14 Months after S ApplicationPhosphorus Concentration (mg kg1) Ca bound P Humic fulvic acid P Residual P Fig 4 5. Concentrations of P in Ca bound, humic -fulvic acid, and residual fractions of soils at various times after S application. Due to lack of significant effects of S, data for S application rates were averaged for presentation. Error bars represent the standard error of the mean.

PAGE 92

92 250350 450 550 650 750 0 2 4 6 8 10 12 14 Months after S ApplicationPhosphorus Concentration (mg kg1) Total inorganic P Total organic P 250350 450 550 650 750 0 2 4 6 8 10 12 14 Months after S ApplicationPhosphorus Concentration (mg kg1) Total inorganic P Total organic P Fig 4 6. Distribution of P among inorganic and organic pools after S application. Error bars represent the standard error of the mean.

PAGE 93

93 CHAPTER 5 MICROBIAL ECO -PHYSIOLOGICAL RESPON SE OF A CALCAREOUS H ISTOS OL TO SULFUR AMENDMENT Introduction Microbial communities play important roles in organic matter degradation and nutrient regeneration in soils. Likewise, soil physical and chemical properties and environmental factors greatly influence microbial activit ies and community composition (Allison et al., 2007). Microbial functional activities, such as extracellular enzyme activities (Allison et al., 2007), microbial respiration (Castillo and Wright, 2008; Iovieno et al., 2009), and nutrient mineralization rat es (Corstanje et al., 2007; Wright et al., 2009) have been widely used as indicators to assess soil disturbance. Monitoring the microbial ecophysiological response to soil disturbance provides insights in understanding their effects and extent on nutrient cycling and organic matter turnover (Corstanje et al., 2007). The EAA in south Florida was historically a seasonally -flooded prairie ecosystem but was converted to agricultural use by drainage in the early 1900s. Longterm drainage has resulted in oxid ation of these Histosols resulting in a decreased depth to bedrock. The current estimate of soil loss is 1.5 cm yr1 and many soils are less than 51 cm in depth, such as those classified as the Dania series (Shih et al., 1998; Snyder, 2005). Land use con version and soil oxidation has contributed to nutrient export from agricultural fields into adjacent wetlands, resulting in SO4 and P enrichment. Sulfate export from the EAA into Everglades wetlands has been implicated in causing stimulation of Hg methyla tion (Gabriel et al., 2008).

PAGE 94

94 Longterm cultivation of these drained peatlands, specifically the use of tillage, coupled with soil oxidation, has also resulted in incorporation of bedrock CaCO3 into surface soil and has gradually increased the pH from the historic 5.0 to 5.5 to approximately 7.0 to 7.5 today (Snyder, 2005; Gabriel et al., 2008). As a result, P and micronutrient availability to crops have decreased and necessitated new fertilizer management practices. Elemental S is occasionally applied as soil amendment for the purpose of reducing pH and therefore increasing P and micronutrient availability (Gabriel, et al., 2008). The microbial oxidation of elemental S to SO4 produces acidity which in turn releases P bound to Ca minerals, which may resul t in a pulsed flux of P to soil solution. Also, the reduction in pH creates a more favorable environment for the soil microbial community and may enhance nutrient availability through increases in organic matter mineralization rates. Thus, there is concern that widespread S application may stimulate nutrient export from these soils which can harm down gradient wetlands. However, the buffering capacity of these calcareous Histosols is strong and may counteract the acidifying effects of S oxidation, thus e ffects of amendments may only be temporary and minimally effective (Beverly and Anderson, 1986). Increased nutrient availability resulting from S amendment is likely to stimulate microbial activity and subsequently alter nutrient cycling and organic mate r turnover (Wright and Reddy, 2001; Castillo and Wright, 2008). Due to the increases in pH and the decreasing depth to bedrock of soils in the EAA, the need for S amendments may increase in the future. Investigations of the responses of microbial functional

PAGE 95

95 activities to variable S application rates are necessary and would provide insight in understanding the response of this system to S application. These results can then be used to help to formulate fertilizer and nutrient management solutions for bett er soil management in the EAA. Material and Methods Site Description The experimental site is located in the central EAA on Dania muck (euic, hyperthermic, shallow Lithic Haplosaprist) with a depth to bedrock of approximately 50 cm. The experimental des ign was a randomized complete block with four S application rates and four field replications. Each field plot measured 9 m x 13 m and consisted of 6 rows of sugarcane ( Saccharum spp. ) Sugarcane cultivar CP 89 2143 was planted in November 2007 and harvested in February 2009. Elemental granular S (90%) was applied at rates of 0, 112, 224, and 448 kg S ha1 to the furrow and covered after planting. Other fertilization was provided using typical recommendations and guidelines for this region and soil type (Rice et al., 2006). All fertilizers were soil applied prior to planting and included 17 kg N ha1 and 37 kg P ha1 as monoammonium phosphate, 228 kg K ha1 as KCl, 8.5 kg Mn ha1, 4.5 kg Cu ha1, 5.6 kg Fe ha1, 2.8 kg Zn ha1, and 1.1 kg B ha1. All p lots received common cultural practices including tillage and herbicide application. Water was applied as needed via seepage irrigation in field ditches approximately 182 m apart.

PAGE 96

96 Soil Sampling and Analysis Soil samples were collected before planting an d fertilizer application and then in January 2008, May 2008, August 2008, and December 2008, corresponding to approximately 0, 2, 6, 9, and 13 months after planting, respectively. Twelve soil (0 15 cm) cores (2.54 cm diameter) were randomly collected from rows within each field plot and composited. Samples were homogenized after the removal of visible plant residues and stored at 4 C. The pH was measured using a soil to water ratio of 1:3 after equilibration for 30 min. Total organic C was measured by loss on-ignition at 550C for 4 hr after conversion to organic C with a coefficient factor of 0.51 (Wright et al., 2008). Dissolved organic C was measured by extraction with 0.5 M K2SO4 and analyzed with a TOC -5050A total organic C analyzer (Shimadzu, Nor cross, GA). Total N was measured by Kjeldahl digestion followed by NH4 analysis (Bremner, 1996). Extractable NH4-N and NO3-N were determined by extraction with 2 M KCl followed by colorimetric analysis (Castillo and Wright, 2008). Total P was determined using the ascorbic acid molybdenum blue method after Kjeldahl digestion, and labile inorganic P measured after Mehlich1 extraction (Kuo, 1996). Microbial biomass C and N were measured by the fumigationextraction method using a conversion factor of 0.37 and 0.54, respectively (Wright et al., 2009). The microbial biomass P was determined as the difference in total P of NaHCO3 extracts between fumigated and unfumigated samples (Wright et al., 2009). Potentially mineralizable N and P were determined based on a 10d incubation

PAGE 97

97 followed by extraction with 2 M KCl (for N) and 1 M HCl (for P) and subtraction of initial N and P concentrations (Wright et al., 2009). Mineralized S was calculated as the difference in water extractable SO4 between residual soil and after a 10 d incubation, with SO4 being analyzed by ion chromatography (Gharmakher et al., 2009). Microbial respiratory activities were measured as CO2 production after 5 and 10 d incubation (Corstanje et al., 2007) and expressed as the slope of the regression of cumulative CO2 production over the 10d incubation period (Wright and Reddy, 2008). In consideration of the potential acidification effect induced by S oxidation and in order to resemble real soil pH variations between treatments, all enzyme activities were assayed with nonbuffered solutions. Approximately, 1 g moist soil was placed in polypropylene centrifuge tubes, mixed with 30 ml of sterile water, and shaken for 25 min. Homogenized samples were further diluted 5 times for enzyme assays. Enzyme assays were conducted using 6 replicates with controls to offset nonenzymatic production. Phosphatase, glucosidase, and leucine aminopeptidase assays were conducted in 96well microtiter plates. The enzyme substrates were 1 mM 4 MUF -phosphate ( Sigma, St. Louis, MO), 1 mM 4 MUF -D glucopyranoside (Sigma, St. Louis, MO), and 2.5 mM L Leucine 7 -Amino -4 methylcoumarin (Biosynth, Naperville, IL), respectively. 200 L of sample was incubated with 50 l substrates at 20C for 3 hr. The florescence readings were collected at 0.5 hr i ntervals using a Bio -TEK FL600 fluorescence plate reader (Bio-TEK Instruments Inc., Winooski, VT) at a setting of 365 m excitation and 450 n m emission. Enzyme

PAGE 98

98 activity was determined by calculating the mean florescent reading changes over time. Sulfatas e activity was determined colorimetrically according to the methods of Wright and Reddy (2001). Statistical Analysis A mixed model was fit using restricted maximum likelihood in the MIXED procedure of SAS (Littell et al., 2006). The fixed effects were S application rate, time and their interaction. Block was a random effect. Degrees of freedom were adjusted using the Kenward-Roger adjustment. An exponential covariance structure was used to model the correlation among observations taken from the same plot over time. Significant differences among individual treatments and time intervals were performed to assess relationships between variables. All statistical analyses were carried out with SAS 9.1 (SAS Institute). Results Soil Physical and Chemical Properties Soil chemical properties were listed on Table 5 1. Application of S at a range from 0 to 448 kg S ha1 did not significantly affect soil pH, extractable NH4-N, NO3-N, and SO4, and dis solved organic C. Dissolved organic C increased significantly from 2 (1346 mg kg1) to 6 months (1572 mg kg1), and then decreased toward the end of the season. Inversely, extractable NO3N decreased from 2 (383 mg kg1) to 6 months (16 mg kg1), but then increased at 13 months (49 mg kg1). Extractable NH4-N and SO4 both decreased gradually throughout the growing season. The

PAGE 99

99 interaction of S application and time was significant for labile P. Further analysis revealed that labile P was substantially hi gher in soils amended with 448 kg S ha1 (118 mg kg1), than soils receiving lower S rates at 2 months after application (49 mg kg1) (Fig. 5 -1). However, the higher P concentrations were not observed at later months. Extracellular Enzyme Activities Th e S application effect on phosphatase activity was significant, yet the effect was only observed at 2 months after S application (Fig. 5 2a). Phosphatase activity at 2 months was considerably higher for soils receiving 448 kg S ha1 (131 mg MUF kg1 h1) than soils receiving 0, 112, and 224 kg S ha1, which averaged 61, 76, and 81 mg MUF kg1 h1, respectively. Phosphatase activity also fluctuated seasonally, with the lowest activity observed at 6 months (52 mg MUF kg1 h1) and the highest activity at 9 months (102 mg MUF kg1 h1). Glucosidase activity significantly increased at 2 months, averaging 15, 56, 58, and 99 mg MUF kg1 h1 for the increasing S application rates (Fig. 5 2b). Glucosidase was also significantly higher at 2 and 9 months than at 6 and 13 months. Analysis of variance showed that leucine aminopeptidase activity was not affected by any S rate, averaging 68 mg MUF kg1 h1 (Fig. 5 -3a). However, leucine aminopeptidase decreased significantly toward the end of the sugarcane growing season. Sulfatase activity did not respond to S application at any rate (Fig. 5 3b) and was largely unaffected by seasonality.

PAGE 100

100 Microbial Biomass Microbial biomass C was not altered as a result of S amendment, but did fluctuate during the growing season (Table 5 -2). The highest concentration occurred at 9 months (17 g kg1), followed by 2 months (13 g kg1) and 6 and 13 months (11 g kg1). Similarly, the size of the microbial biomass N pool did not change after S amendment and was stable throughout the growing season. Microbial biomass P increased significantly at 2 months at the highest S rate (177 mg kg1), which was about 3 times higher than for lower S application rates (Fig. 5 4). However, the stimulating effect did not extend beyond 2 months. Micro bial-Mediated Mineralization Aerobic CO2 production rates did not differ between soils receiving variable S application rates (Table 5 2). The production was the highest at 9 months after S application (44 mg CO2-C kg1 d1) and lowest at 2 months (26 mg CO2-C kg1 d1), corresponding to temperature patterns. Similarly, N and P mineralization rates were not influenced by S amendment (Table 5 2). The highest overall rates of N mineralization were found at 9 months (10 mg kg1 d1) and the lowest rates at 13 months (2 mg kg1 d1). Sulfur mineralization rates significantly increased as a result of S application (Fig. 5 -5). Overall, mineralized S was 1421% greater for soils receiving 448 kg S ha1 than unamended soil, 375% greater than soils receiving 112 kg S ha1, and 219% greater for soils receiving 224 kg S ha1. Nonetheless, mineralized S rates significantly decreased throughout the growing season.

PAGE 101

101 Discussion Labile P is considered the most bioavailable form of P in soils. Sulfur application at 448 kg S ha1 significantly increased concentrations of labile P at 2 months (Fig. 5 1), suggesting increased P availability to sugarcane as well as soil microorganisms (Codling, 2008). There are two primary mechanisms by which S influences P availability: lowering of soil pH (Gabriel, et al., 2008) and replacement of PO4 with SO4 and the release of P from association with Fe, Al, and Ca (Jaggi et al., 2005). These results are supported by the fact that labile P was significantly correlated with pH (R2 = 0 .35, p = 0.005) and extractable SO4 (R2 = 0.29, p = 0.019). However, increased P availability was not observed at later months indicating limited long -term effects of S on the reduction in soil pH due to the high buffering capacity of this calcareous organic soil (Jaggi et al, 2005; Snyder, 2005). Enzymatic Activities Extracellular enzymes are excreted by the microorganisms to the soil for the purpose of sequestering nutrients. The enzyme activities, especially hydrolases, are known to be involved in or ganic matter turnover and nutrient cycling in terrestrial systems (Wright and Reddy, 2001; Corstanje et al., 2007). Phosphatase catalyzes the hydrolysis of organic P ester resulting in the release of P, and thus plays an important role in P regeneration f rom soils (Wright and Reddy, 2001). Glucosidase catalyzes the hydrolysis of glycosides and its activity reflects the state of organic matter and processes occurring therein (Tejada et al., 2006). Our results showed that phosphatase activities were 115% h igher for soils receiving the highest S rates

PAGE 102

102 than unamended soils at 2 months after S application, while glucosidase activities were 573% higher (Fig. 5 2). High enzyme activity may indicate nutrient limitation (Sinsabaugh et al., 1993; Allison et al., 2007), and the highest activity occurred during the winter sampling when soil oxidation rates are typically lowest (Snyder, 2005). Thus, the nutrient -supplying capacity of this soil was low at this time. In fact, soil oxidation typically provides a major portion of the sugarcane nutrient requirements for this soil (Rice et al., 2006). Negative correlations between nutrient availability and related enzyme activities have been demonstrated (Wright and Reddy, 2001; Allison and Vitousek, 2005). However, no s uch correlations were observed in the present study, but instead, both phosphatase and glucosidase activities were positively correlated to the concentrations of labile P (Table 5 3), which together may suggest the P limitation for organic matter turnover in these high C soils. (Allison et al., 2007). The EAA soils are primarily Histosols with high organic matter content, approximately 85% by weight, which contain high N yet low P and micronutrient concentrations that require supplemental fertilization (Sn yder, 2005; Castillo and Wright, 2008). The background C: N: P molar ratio at this site was 1274:98:1, indicating this soil was indeed P limited, which may constrain the activities and growth rates of microorganisms (Fontaine et al., 2003). As demonstrat ed previously, application of S increased the concentration of labile P at 2 months (Fig. 5 1) and both activities of phosphatase and glucosidase simultaneously increased (Fig. 5 2). It is plausible that the application of S had a transitory effect on soi l pH which stimulated P release to the soil environment and hence enhanced the

PAGE 103

103 microbial enzymatic activities. Thus, application of S at the current recommended rate had a short -lived effect on microbial activity. The combination of soil buffering capacit y and sugarcane nutrient uptake likely minimized P release and accumulation in soil that was initiated by S application. These results indicate that higher S rates may be needed to prolong the response of the soil microbial community and maintain nutrient availability throughout the growing season. Leucine aminopeptidase and sulfatase were also assayed in this study to represent N and S cycles. Leucine aminopeptidase is involved in the degradation of proteins (Larson et al., 2002), while sulfatase hydroly zes aromatic ester sulfates (Knauff et al., 2003). Both enzyme activities were not influenced by any rate of S application (Fig. 5 3). Since enzymes are biological catalysts capable of catalyzing specific chemical reactions, their responses may vary based on substrate quality and real microbial community composition (Corstanje et al., 2007). Soil enzyme activities may also depend on other parameters, such as seasonal variation in soil moisture, pH and temperature (Knauff et al., 2003; Wallenstein et al., 2009), which may explain the lack of response to S amendment in this soil. Microbial Biomass Microbial biomass P was more sensitive to S addition than biomass C and N at 2 months, as biomass P for soils receiving 448 kg S ha1 was 314% higher than soils receiving the lower rates (Fig. 5 4). Correlation analysis revealed that microbial biomass P was significantly correlated to labile P, total P, extractable SO4, pH, NH4N, and NO3-N (Table 5 -3). Considering the fact that this Histosol is P limited, i t was

PAGE 104

104 likely that increased labile P at 2 months after S application caused P immobilization into microbial biomass, which may explain the lower labile P concentrations at subsequent sampling times during the growing season. Increasing biomass P as a resu lt of increased P availability in Everglades soils has been well documented (Corstanje et al., 2007; Castillo and Wright, 2008; Wright et al., 2009). Microbial-M ediated Organic Matter Mineralization Agricultural practices in EAA soils, such as tillage and P fertilization, showed variable effects on organic C mineralization (Morris et al., 2004). The present study indicated that S amendment did not appear to influence microbial aerobic respiration rates indicating that S application will not further sti mulate soil oxidation. Correspondingly, the metabolic coefficient (qCO2), the proportion of aerobic respiration to microbial biomass C, did not change across S rates (Table 5 4), suggesting that application at rates up to 448 kg S ha1 did not alter organic matter turnover. Glucosidase activity is known to reflect the state of organic matter cycling (Tejada et al., 2006). However, glucosidase activity was not correlated with C mineralization rates (Table 5 -3). Glucosidase is one of several soil enzymes involved in C mineralization process and hence its activities alone may not be able to reflect the overall real C mineralization, which may explain why no increased CO2 production was observed at 2 months at 448 kg S ha1 when glucosidase activity was sign ificantly enhanced. Potentially mineralizable N and the quotient (mineralized N as a function of biomass N) characterize the potential N turnover in the system (Corstanje et al.,

PAGE 105

105 2007; Castillo and Wright, 2008). No clear effects of S amendment on N turn over rates in this Histosol were observed (Table 5 -2 and 5 4). Likewise, net P regeneration did not appear to be impacted by S, even after phosphatase activities were significantly increased at 2 months, suggesting that P mineralization was not limited by phosphatase activity (Carreira et al., 2000). Studies have shown no correlation between phosphatase activity and gross P mineralization, whereas others have shown positive relationships (Carreira et al., 2000). Agriculture in the EAA has been identified as a major source of P enrichment in Everglades wetlands, which damaged the ecosystem and impaired water quality (Wright et al., 2009). Our results indicated that S application under current guidelines may not enhance the microbial mediated P regeneration in EAA soils and thus minimize the risk of P exports from agricultural fields. Potential S mineralization was the only mineralization process showing a clear response to S amendment. It has been proposed that S mineralization involved two processes, biological and biochemical mineralization, in which SO4 is released as a by product of C oxidation and as a product of enzymatic hydrolysis (McGill and Cole, 1981; Chen et al., 2001). Nonetheless, in the present study, mineralized S was not correlated wi th CO2 production or sulfatase activity. Instead, S mineralization was highly correlated with glucosidase activity (Table 5 3), which may suggest that net mineralization of S was not simply dependent on either biological or biochemical processes (Eriksen et al., 1998). Oxidation of elemental S may also contribute to the observed increased mineralized S rates with increasing S application rate.

PAGE 106

106 Elemental S is oxidized by soil microorganisms producing SO4, which can accumulate in soil solution. The elemen tal S oxidation rate may be slow (Deluca et al., 1989) and SO4 accumulation in this study is likely a result of both organic S mineralization and elemental S oxidation to SO4. Our results demonstrated that mineralized S increased concurrently with higher S application rates and the effects continued throughout the growing season, although stimulatory effects diminished with subsequent sampling times (Fig. 5 5). Seasonal Fluctuations in Microbial Indices Seasonal variations in soil microbial activities ar e common (Wallenstein et al., 2009) and dependent on environmental factors such as rainfall and temperature. Soil disturbance resulting from agricultural practices are also known to pose impacts on microbial community composition and activity (Knauff et al., 2003; Morris et al., 2004; Wright and Reddy, 2008). During the growing season, tillage was applied to improve drainage and weed control, which has been found to affect the size of the microbial biomass pool and organic matter mineralization rates (Mor ris et al., 2004; Castillo and Wright, 2008 a ). Our results showed clear seasonal fluctuations for most of the microbial parameters, suggesting the net results of interactions between environmental factors, soil management, and the sugarcane growing season Most effects occurred within 2 months of S application, suggesting a strong effect of fertilization for this site. Nutrient concentrations decreased during the growing season due to sugarcane uptake, thus nutrient limitations may have minimized microbi al responses to S application at subsequent sampling times. In fact, these

PAGE 107

107 soils are often low in plant available nutrients with the exception of N (Rice et al., 2006). Conclusions Application of elemental S at 448 kg S ha1 increased P availability at 2 months, which subsequently stimulated some enzyme activities and simultaneously promoted labile P to be immobilized in microbial biomass. However, these effects were temporary and not observed beyond 2 months. There was limited effect of S application on increasing the P availability due to the high buffering capacity of this organic soil against pH reduction. Overall, S amendment at rates up to 448 kg S ha1 did not appear to pose significant impacts on organic matter turnover and N and P regeneration rates, suggesting that S application will not stimulate soil oxidation and result in large -scale nutrient flux from soil. Using the current recommended S application guidelines and rates, impacts on microbial activities and functions should be minimal. However, due to the increasing pH trend for these soils, there may be a need for higher S application rates in the future. These higher S rates may overcome the soils buffering capacity and release large amounts of P, potentially stimulating microbial f unctional activities and altering organic matter dynamics. Additionally, oxidation of elemental S would produce large amounts of SO4 and therefore may pose an environmental hazard to the nearby aquatic ecosystem.

PAGE 108

108 Table 5 1. Extractable nutrients (mg kg1) and pH in soil amended with elemental S during the sugarcane growing season. Values denote means and the standard error is in parentheses. Dissolved organic C Extractable NH 4 N Extractable NO 3 N Extractable SO 4 Labile P i pH Treatment 0 kg S ha 1 1315 (58) 12 (2) 110 (41) 107 (40) 47 (4) 6.2 (0.1) 112 kg S ha 1 1583 (93) 10 (1) 115 (39) 145 (40) 46 (4) 6.5 (0.1) 224 kg S ha 1 1419 (56) 10 (1) 120 (42) 141 (37) 56 (8) 6.3 (0.1) 448 kg S ha 1 1381 (45) 12 (3) 121 (42) 179 (42) 66 (11) 6.1 (0.1) Time 2 months 1346 (47) 19 (3) 383 (19) 376 (32) 66 (12) 6.0 (0.1) 6 months 1572 (67) 10 (0) 16 (1) 129 (13) 48 (1) 6.3 (0.0) 9 months 1436 (80) 9 (1) 19 (1) 21 (3) 53 (4) 6.4 (0.1) 13 months 1344 (66) 6 (0) 49 (2) 47 (7) 46 (8) 6.4 (0 .1)

PAGE 109

109 Table 5 2. Potentially mineralizable C (Cmin), N (Nmin), P (Pmin), and microbial biomass C (MBC), N (MBN), P (MBP) in soil amended with elemental S during the sugarcane growing season. Values denote means and the standard error is in parentheses. Cmin (mg kg 1 d 1 ) Nmin (mg kg 1 d 1 ) Pmin (mg kg 1 d 1 ) MBC (g kg 1 ) MBN (g kg 1 ) MBP (mg kg 1 ) Treatment 0 kg S ha 1 34 (1.9) 6 (0.8) 6 (3.6) 12 (0.9) 0.28 (0.03) 33 (4.5) 112 kg S ha 1 39 (2.7) 8 (1.1) 4 (1.5) 13 (0.8) 0.27 (0.03) 31 (2.8) 2 24 kg S ha 1 38 (3.4) 7 (1.0) 5 (2.3) 13 (0.6) 0.27 (0.03) 29 (3.9) 448 kg S ha 1 32 (2.0) 6 (0.9) 2 (1.2) 13 (0.8) 0.29 (0.03) 57 (13.2) Time 2 months 26 (1.3) 8 (1.1) 6 (2.8) 13 (0.4) 0.27 (0.01) 64 (13.2) 6 months 36 (1.8) 6 (0.3) 6 ( 2.7) 11 (0.9) 0.27 (0.02) 33 (4.0) 9 months 44 (3.3) 10 (0.3) 3 (2.3) 17 (0.2) 0.35 (0.04) 28 (4.1) 13 months 37 (1.2) 2 (0.2) 1 (0.6) 11 (0.2) 0.23 (0.01) 27 (1.1)

PAGE 110

110 Table 5 3. Significant correlation coefficients ( p < 0.05) between selected chemical pr operties and microbial functional activities (n = 64). pH DOC NH 4 NO 3 P i SO 4 LAP PHO GLU SUL Cmin Nmin Pmin Smin MBC MBN MBP pH 1 DOC 0.63 1 NH 4 0.42 NS 1 NO 3 0.51 NS 0.61 1 P i 0.35 NS 0.6 0 0.28 1 SO 4 0.46 NS 0.60 0.88 0.29 1 LAP 0.42 0.57 NS NS NS NS 1 PHO NS NS NS NS 0.28 NS NS 1 GLU 0.28 NS 0.30 0.41 0.35 0.35 NS 0.64 1 SUL NS 0.29 NS 0.41 NS NS NS NS NS 1 Cmin 0.52 0.50 0.33 0.51 NS 0.45 NS NS NS NS 1 Nmin NS NS NS NS NS NS 0.44 NS NS 0.29 NS 1 Pmin NS NS NS NS NS NS NS NS NS NS NS NS 1 Smin 0.38 NS 0.31 NS 0.49 0.39 NS 0.28 0.41 NS NS NS NS 1 MBC NS NS NS NS NS NS NS 0.34 NS NS NS 0.41 NS NS 1 MBN NS NS NS NS NS NS NS NS NS NS NS 0.36 NS NS NS 1 MBP 0.38 NS 0.57 0.49 0.63 0.47 NS NS 0.35 NS 0.37 NS NS 0.61 NS NS 1 *, significant at p = 0.05; NS, not significant; DOC, dissolved organic C; NH4, extractable NH4-N; NO3, extractable NO3-N; P i extractable labile P; SO4, extractable SO4-S; MBC, microbial biomass C; MBN, microbial biomass N; MBP, microbial biomass P; LAP, leucine aminopeptidase; PHO, phosphatase; GLU, glucosidase; SUL, sulfatase; Cmin, potentially mineralizable C; Nmin, potentiall y mineralizable N; Pmin, potentially mineralizable P; Smin, potentially mineralizable S;

PAGE 111

111 Table 5 4. Microbial metabolic coefficient (qCO2), microbial biomass C to organic matter content ratio (MBC/OM), and potential N and P mineralization quotient (qPMN and qPMP) in soils amended with elemental S during the sugarcane growing season. Values denote means and the standard error is in parentheses. qCO2 (100) MBC/OM (%) qPMN (mg N g 1 MBN) qPMP (mg P g 1 MBP) Treatment 0 kg S ha 1 0.31 (0.04) 1.5 (0.1) 22 (3.3) 192 (130) 112 kg S ha 1 0.31 (0.04) 1.7 (0.1) 28 (4.1) 136 (54) 224 kg S ha 1 0.29 (0.02) 1.6 (0.1) 26 (4.4) 153 (71) 448 kg S ha 1 0.28 (0.04) 1.6 (0.1) 23 (5.0) 42 (22) Time 2 months 0.20 (0.01) 1.6 (0.1) 31 (4.0) 160 (59) 6 months 0.39 (0.05) 1.3 (0.1) 24 (2.3) 212 (121) 9 months 0.27 (0.02) 2.1 (0.0) 36 (4.8) 109 (72) 13 months 0.33 (0.01) 1.4 (0.0) 8 (0.9) 45 (24)

PAGE 112

112 Months after S Application 0 40 80 120 160 2 6 9 13 0 112 224 448Labile P (mg kg1)Months after S Application 0 40 80 120 160 2 6 9 13 0 112 224 448 0 112 224 448Labile P (mg kg1) Fig 5 1. Concentrations of labile P in soils at 2, 6, 9, and 13 months after elemental S applicati on at different rates (0, 112, 224, and 448 S kg ha1). Error bars represent the standard error of the mean.

PAGE 113

113 Months after S Application 0 50 100 150Phosphatase Activity mg MUF kg1h1Glucosidase Activity mg MUF kg1h1a 0 20 40 60 80 100 120 2 6 9 13 0 112 224 448 b Months after S Application 0 50 100 150Phosphatase Activity mg MUF kg1h1Glucosidase Activity mg MUF kg1h1a 0 20 40 60 80 100 120 2 6 9 13 0 112 224 448 b 0 20 40 60 80 100 120 2 6 9 13 0 112 224 448 0 112 224 448 b Fig 5 2. Activities of phosphatase (a) and glucosidase (b) in response to different elemental S application (0, 112, 224, and 448 kg S ha1) throughout the sugarcane growing season. Error bars represent the standard error of the mean.

PAGE 114

114 0 50 100 150 200 250300 2 6 9 13 Months after S ApplicationSulfatase Activity mg pnitrophenol kg1h1 0 20 40 60 80 100120Leucine Aminopeptidase Activity mg MUF kg1h10 112 224 448 a b 0 50 100 150 200 250300 2 6 9 13 Months after S ApplicationSulfatase Activity mg pnitrophenol kg1h1 0 20 40 60 80 100120Leucine Aminopeptidase Activity mg MUF kg1h10 112 224 448 0 112 224 448 a b Fig 5 3. Activities of leucine aminopeptidase (a) and sulfatase (b) in response to different elemental S application rates (0, 112, 224, and 448 kg S ha1) throughout the sugarcane growing season. Error bars represent the standard error of the mean.

PAGE 115

115 0 50 100 150 200 2 69 13 Months after S ApplicationMicrobial Biomass P (mg kg1)0 112 224 448 0 50 100 150 200 2 69 13 Months after S ApplicationMicrobial Biomass P (mg kg1)0 112 224 448 0 112 224 448 Fig 5 4. Microbial biomass P in soils at 2, 6, 9, and 13 months after elemental S application at different rates (0, 112, 224, and 448 S kg ha1). Error ba rs represent the standard error of the mean. 0 10 2030 40 50 2 6 9 13 Months after S applicationPotentially Mineralizable S (mg kg1d1)0 112 224 448 0 10 2030 40 50 2 6 9 13 Months after S applicationPotentially Mineralizable S (mg kg1d1)0 112 224 448 0 112224 448 Fig 5 5. Mineralized S in soils at 2, 6, 9, and 13 months after S application at different rates (0, 112, 224, and 448 S kg ha1). Error bars represent the standard error of the mean.

PAGE 116

116 CHAPTER 6 SEASONAL CH ANGES IN NUTRIENT AVAILABI LI TY IN SULFUR-AMENDED EVERGLADES SOILS UND ER SUGARCANE Introduction The availability of essential elements is known to affect the yield and quality of crops (Heitholt et al., 2002; Parsons et al., 2007). Soils are the major sour ces for plant nutrients; however, their nutrient availability varies during the growing season depending on characteristics such as soil organic matter content, pH, and cationexchange capacity (Cancela et al., 2002; Strahm and Harrison, 2007). Soil manag ement practices, such as fertilization and amendments, are commonly employed to enhance the nutrient supply and increase crop yields. Nonetheless, the status and behavior of nutrients in soil are difficult to predict (Moral et al., 2002; Moreno -Caselles e t al., 2005; Herencia et al., 2008). Interactions among nutrients affect their availability to crops (Rice et al., 2006) as an over abundance of one nutrient may lead to a deficiency of another. For instance, excessive P fertilization can decrease Zn availability due to precipitation of Zn3(PO4)2 (Li et al., 2007). Thus, applying proper amounts of fertilizers or amendments minimizes nutrient imbalances, maximizes crop yields, and improves fertilizer use efficiency. Soil testing is widely used for evalu ating nutrient availability and justifying fertilizer application rates to maximize crop production while minimizing adverse environmental impacts, including the runoff or leaching of excess nutrients (Rice et al., 2006; Slaton et al., 2009). Various test ing methods have been introduced, including acids, salts and chelates to assess nutrient availability in soils (Cancela et

PAGE 117

117 al., 2002). However, no universal standards have been achieved (Wang et al., 2004). Disagreements on nutrient availability and fert ilizer recommendations with similar soils and crops as a result of different extraction methods have been documented (Cancela et al., 2002; Wang et al., 2004). Extractants vary in their extracting capabilities and therefore dissimilar extraction methods account for different degrees of nutrient availability. A sound and acceptable soil testing should be correlated with crop yield (Korndrfer et al., 1995). The Everglades Agricultural Area (EAA) in south Florida was historically a seasonally -flooded prair ie ecosystem, but was converted to agricultural use by drainage in the early 1900s. The soils are primarily Histosols with high organic matter content, approximately 85% by weight, which contain high N yet low P and micronutrient concentrations that requi re supplemental fertilization (Snyder, 2005). Upon drainage and land use conversion, high decomposition rates of these drained Histosols resulted in subsidence and a decreased depth to the underlying bedrock. The current estimate of soil loss is 1.5 cm y r1 and many soils are less than 51 cm in depth, such as those classified as the Dania series (Shih et al., 1998; Snyder, 2005). Longterm cultivation of these drained soils, specifically the use of tillage coupled with soil oxidation, has resulted in inc orporation of bedrock CaCO3 into surface soil and has gradually increased the pH from the historic 5.05.5 to approximately 7.0-7.5 today (Snyder, 2005). As a result, P and micronutrient availability to crops has decreased and necessitated new fertilizer management practices to maintain nutrients at concentrations sufficient for optimal crop growth. Application of

PAGE 118

118 elemental S is recommended in the EAA when soil pH exceeds 6.6 for the purpose of reducing pH and therefore increasing P and micronutrient avai lability (Schueneman, 2001). The recommendation rate of 448 kg S ha1 was initially established in 1985 (Anderson, 1985), but due to the changes in soil conditions since 1985, revision of this recommendation may be required. Strong buffering capacity of these calcareous Histosols is likely to counteract the acidifying effects of elemental S oxidation, and thus effects of amendments may only be temporary and minimally effective (Beverly and Anderson, 1986). There is a need to determine the level of S appl ication producing favorable responses in terms of nutrient availability and sugarcane yield. These results can then be used to help formulate fertilizer and nutrient management solutions for better sugarcane management in the EAA. Thus, the objective of t his research was to evaluate various S application rates for their effects on nutrient availability during the sugarcane growing season and to assess the effectiveness of three soil test extractants in predicting sugarcane yield. Material and Methods Site Description The experimental field is located in the central EAA on Dania muck (euic, hyperthermic, shallow Lithic Haplosaprist) with a depth to bedrock of approximately 50 cm. The experimental design was a randomized complete block with four S application rates and four field replications. Each field plot measured 9 m x 13 m and consisted of 6 rows of sugarcane ( Saccharum spp. ) Sugarcane cultivar CP 89 2143 was planted in November 2007 and harvested in February 2009. Elemental

PAGE 119

119 granular S (90%) was app lied at rates of 0, 112, 224, and 448 kg S ha1 to the furrow and covered after planting. Other fertilization was provided using the typical recommendations and guidelines for this region and soil type (Gilbert and Rice, 2006). All fertilizers were soil applied prior to planting and all field plots received 17 kg N ha1 and 37 kg P ha1 as monoammonium phosphate, 228 kg K ha1 as KCl, 8.5 kg Mn ha1, 4.5 kg Cu ha1, 5.6 kg Fe ha1, 2.8 kg Zn ha1, and 1.1 kg B ha1. All plots received common cultural pra ctices including tillage and herbicide application. Water was applied as needed via seepage irrigation in field ditches approximately 182 m apart. Soil Sampling and Analysis Soil samples were collected before planting and fertilizer application and then in January, May, August, and December 2008, corresponding to approximately 0, 2, 6, 9, and 13 months after planting, respectively. Twelve soil (0 -15 cm) cores (2.54 cm diameter) were randomly collected within each field plot and composited. Samples were homogenized after the removal of visible plant residues and stored at 4C. Soil pH was measured using a soil to water ratio of 1:3 after equilibration for 30 min. Organic matter content was determined by loss onignition at 550C for 4 hr (Wright et al. 2008). Dissolved organic C (DOC) was measured by extraction with 0.5 M K2SO4 and analyzed with a TOC -5050A total organic C analyzer (Shimadzu, Norcross, GA). Extractable NH4N and NO3-N were determined by extraction with 2 M KCl followed by colorimetri c analysis (Castillo and Wright, 2008). Water extractable SO4-S was analyzed by ionic chromatography after shaking 3 g soil with

PAGE 120

120 25 mL water for 0.5 hr, followed by filtering through Whatman No. 41 filter paper (Gharmakher et al., 2009). Three different extractants were tested in this study: water, 0.5 N acetic acid, and Mehlich-3. Water and acetic acid are the soil test extractants for use on muck soils, and Mehlich3 is the soil test extractant for sandy soils in Florida (Morgan et al., 2009; Mylavarapu, 2009). Phosphorus concentrations for different extracts were determined using the ascorbic acidmolybdenum blue method (Kuo, 1996) after shaking 4 mL air -dry soil with 50 mL of extractant for 50 min, followed by filtering through Whatman No. 2 (acetic a cid extraction) and No. 5 (water and Mehlich3 extraction) filter paper, respectively. Extractable macro and micronutrients in different extractants were then analyzed by ICP. Sugar Yield Harvestable stalks were counted in 2 of the 4 middle rows of each plot in August. Stalk weights were determined by cutting and weighing 20 stalks from 2 of the middle 4 rows of each plot (40 stalks total) in February 2009. Sugarcane yield was calculated by multiplying stalk number by stalk weight, and dividing by unit area. Sugar yield (Mg ha1) was determined according to the theoretical recoverable sugar method utilizing sugar content of harvested cane and sugarcane yield (Glaz et al., 2002). Statistical Analysis A mixed model was fit using restricted maximum likelihood in the MIXED procedure of SAS (Littell et al., 2006). The fixed effects were S application rate, time

PAGE 121

121 and their interaction. Block was a random effect. Degrees of freedom were adjusted using the Kenward-Roger adjustment. An exponential covariance s tructure was used to model the correlation among observations taken from the same plot over time. Significant differences among individual treatments and time intervals were perf ormed to assess relationships between variables. Stepwise multiple regression was conducted to evaluate the relative importance of extractable nutrients in predicting sugar yield. A p value of 0.1 and 0.05 was used as the entry and staying values, respec tively, in the stepwise selection method (Majchrzak et al., 2001). All statistical analyses were carried out with SAS 9.1 (SAS Institute). Results and Discussion Soil pH Soil pH was not affected by S application (Fig. 6 -1). The background pH prior to S application was 6.2, which did not differ from soils collected during the growing season. The limited effect of acidification may result from a S application rate too low to cause a change in pH and from the high buffering capacity of this calcareous or ganic soil (Jaggi et al, 2005; Deubel et al., 2007). When the original S recommendation for sugarcane of 448 kg S ha1 was established years ago, soil pH was considerably lower. However, the rise in pH and decrease in soil depth to bedrock likely increas ed the capacity of these soils to resist changes in pH. Thus, higher S application rates may be necessary to produce the same response as 448 kg ha1 did in the 1980s (Anderson, 1985). Soils with high concentrations of

PAGE 122

122 carbonates and bicarbonates are hig hly buffered against acidification (Rogovska et al, 2007). The buffering effects often take place more slowly than the formation of sulfuric acid from elemental S (Deubel et al., 2007). A limited reduction in soil pH after S application was also observed in other studies of calcareous soils (Hassan and Olson, 1966). Dissolved Organic C and Extractable N Application of S at a range from 0 to 448 kg S ha1 did not affect DOC, but the concentrations varied seasonally (Fig. 6 2). Averaged across treatments DOC significantly increased from 2 (1346 mg kg1) to 6 months (1572 mg kg1), but then decreased toward the end of the growing season. Extractable NH4-N and NO3-N were not influenced by S application, but concentrations fluctuated during the growing sea son (Fig. 6 2). Extractable NH4-N significantly decreased from 2 (19 mg kg1) to 6 months (10 mg kg1), while extractable NO3-N exhibited the same trend from 2 (383 mg kg1) to 6 months (16 mg kg1). Oxidation of the muck soil provides most of the N requ irement for sugarcane grown in the EAA (Gilbert and Rice, 2006). However, the soil N pool can change rapidly under the impacts of environmental (precipitation, temperature) and management (tillage, irrigation) factors (Rice et al., 2006). Considering the fact that sugarcane biomass accumulation is the greatest during the summer months of the rainy season (Rice et al., 2006), it was likely that plant uptake and leaching losses contributed to the low NO3-N concentrations as the growing season progressed.

PAGE 123

123 Phosphorus Sulfur application at 448 kg S ha1 significantly increased concentrations of acetic acid, Mehlich-3, and water extractable P at 2 months (Fig. 6 3), suggesting increased P availability to sugarcane caused by S application (Codling, 2008). Th ere are two primary mechanisms by which S application could influence P availability: lowering of soil pH and replacement of PO4 with SO4 (Gabriel et al., 2008), or the release of P from association with Fe, Al, and Ca caused by pH reduction (Jaggi et al., 2005). However, increased P availability was not observed at later months indicating limited longterm effects of S on the reduction in soil pH due to the high buffering capacity of this calcareous organic soil (Snyder, 2005). Acetic acid extractable P decreased progressively during the season (Fig. 6 -3). The concentrations at 13 months were 214% lower than those at 2 months, 107% lower than those at 6 months, and 75% lower than those at 9 months. Similarly, water extractable P decreased gradually from 2 months (15 mg kg1) to 13 months (3 mg kg1). Mehlich 3 extractable P did not change much during the same period. The EAA soils are traditionally P limited and sugarcane production requires supplemental P fertilization (Morgan et al., 2009). Therefo re, the reduction in acetic acid and water extractable P during the season was likely a result of sugarcane uptake. Seasonal trends in P concentrations would be expected to show declines from planting to harvest, corresponding to uptake of extractable P b y sugarcane. However, P mineralized from soil organic matter also contributes to the available P pool.

PAGE 124

124 Across treatments and sampling times, acetic acid extracted 834% more P than water, while Mehlich3 extracted 559% more P than water (Table 6 -1). No significant difference was found between the amounts of P extracted by acetic acid and Mehlich-3. Water primarily extracts P in soil solution, while Mehlich3 and acetic acid also extract P adsorbed or complexed with Ca, Mg, Fe, and Al, in addition to sol uble P (Wright et al., 2007; Wright, 2009). Thus, it was not surprising that Mehlich-3 and acetic acid extracts contained more P than water. Potassium Sulfur application at the highest rates significantly increased K availability at 2 months (Fig. 6 3), but the treatment effect was not observed at subsequent sampling times. At 2 months, acetic acid extractable K for soils receiving 448 kg S ha1 (1214 mg kg1) was significantly higher than for soils receiving 112 kg S ha1 (647 mg kg1) and unamended soil (708 mg kg1). Likewise, Mehlich3 extractable K for soils amended with the highest rates (949 mg K kg1) was significantly higher than soils receiving 112 kg S ha1 (493 mg K kg1) and unamended soil (549 mg kg1). Water extractable K was only differ ent between soils receiving 112 (395 mg kg1) and 448 kg S ha1 (713 mg kg1). Acetic acid extractable K decreased 440% from 2 months to 13 months, while during the same period Mehlich3 and water extractable K decreased 448% and 891%, respectively (Fig. 6 -3). Despite the fact that EAA soils have high cationexchange capacities, K is usually weakly held on the exchange sites (Gilbert and Rice, 2006). Therefore, K movement out of the soil profile occurs readily depending on precipitation patterns. The st imulatory effect of S on

PAGE 125

125 extractable K concentrations was likely attributed to the replacement of K+ by H+ at adsorbing sites, while plant uptake and leaching were responsible for seasonal decreases in K availability. Averaged across treatments and sampl ing times, acetic acid extracted the same amounts of K as Mehlich-3 (Table 6 -1). However, acetic acid and Mehlich3 extracted 106% and 66% more K than water. Correlation analysis revealed that water extractable K was highly correlated to acetic acid (R2 = 0.98) and Mehlich3 extractable K (R2 = 0.96), while the later two were also strongly correlated (R2 = 0.99), indicating that the three extractants may indeed extract the same pools of K. Calcium None of the extractable Ca concentrations exhibited S ef fects, but all displayed seasonal fluctuation (Fig. 6 4). Both acetic acid and water extractable Ca decreased significantly from 2 to 9 months and then increased to 13 months. Mehlich3 extractable Ca was highest at 9 months (19516 mg kg1), followed by 13 (17359 mg kg1), 2 (14386 mg kg1), and 6 months (15224 mg kg1). Acetic acid extractable Ca was 3457% higher than water extractable Ca, while Mehlich -3 extracts had 3022% more Ca than water extracts. Magnesium Sulfur application did not affect Mg av ailability (Fig. 6 4). Acetic acid extractable Mg decreased significantly from 2 (1735 mg kg1) to 6 (665 mg kg1) months, but remained constant to 9 months (671 mg kg1). Mehlich -3 extractable Mg increased from 6 (580 mg kg1) to 9 months (774 mg kg1) and then decreased to 13

PAGE 126

126 months (711 mg kg1). Water extractable Mg decreased gradually during the season, being 307% lower at 13 than 2 months. Acetic acid extracted 44% more Mg than Mehlich3, while 1155% more than water (Table 6 -1). Sulfur Sulfur application significantly increased SO4 concentrations as a result of S oxidation (Jaggi et al., 2005) (Fig. 6 -2). Extractable SO4-S in soils receiving 448 kg S ha1 was 131%, 201%, and 270% higher than unamended soils at 6, 9, and 13 months, respectively. Similar to extractable K, SO4 concentrations decreased significantly from 2 (376 mg kg1) to 6 months (129 mg kg1) and continued to decrease from 6 to 9 months (21 mg kg1). In the EAA, soil oxidation generally supplies sufficient S to satisfy sugarcan e nutrient requirements (Gilbert and Rice, 2006). Therefore, there is potential for S application at high rates to increase the risk of SO4 export from fields. Lower SO4-S concentrations at 6 and 9 months were likely due to SO4 uptake by sugarcane and lo sses as runoff or leaching during precipitation events. Copper Extractable copper was not affected by S application at any time during the growing season (Fig. 6 5). Acetic acid extractable Cu decreased significantly from 2 (0.3 mg kg1) to 9 months ( 0.1 mg kg1), while water extractable Cu remained unchanged during the season. Mehlich-3 extractable Cu increased from 6 (0.1 mg kg1) to 9 months (0.2 mg kg1) and then to 13 months (1.6 mg kg1). Across

PAGE 127

127 treatments and sampling times, acetic acid and water extracted similar amounts of Cu, but both extracted less than Mehlich-3 (Table 6 -1). Iron Similar to Cu, extractable Fe did not respond to any rate of S application, but did fluctuate seasonally (Fig. 6 5). Acetic acid extractable Fe decreased signif icantly from 2 (13.6 mg kg1) to 6 months (9.4 mg kg1) and toward the end of the season (7.2 mg kg1). Inversely, Mehlich-3 extractable Fe increased from 2 (5.8 mg kg1) to 13 months (8.0 mg kg1). Water extracts contained similar amounts of Fe as aceti c acid, but extracted 61% more Fe than Mehlich3 (Table 6 1). Manganese The availability of Mn was not affect by S application, but varied during the season (Fig. 6 6). Similar to K and Fe, acetic acid extractable Mn decreased incrementally from 2 (23.9 mg kg1) to 13 months (9.4 mg kg1). Mehlich -3 extractable Mn decreased from 2 (4.2 mg kg1) to 6 months (3.7 mg kg1), and then increased form 6 to 9 months (5.3 mg kg1). Acetic acid extractable Mn was 7900% higher than water extractable Mn, while Mehlich 3 extracts had 2414% more Mn than water extracts. Zinc Acetic acid extractable Zn in soils receiving 448 kg S ha1 (7.2 mg kg1) was significantly higher than for unamended soils (2.5 mg kg1) at 2 months (Fig. 6 6). However, the stimulating effec ts were not observed beyond 2 months. Mehlich-3 extractable Zn did not exhibit any treatment effects during the growing season.

PAGE 128

128 Interestingly, at 6 months after S application, water extractable Zn for soils receiving 448 kg S ha1 (0.3 mg kg1) was signi ficantly higher than for soils receiving 112 kg S ha1 (0.2 mg kg1) and unamended soils (0.2 mg kg1). Similar to K, Fe, and Mn, acetic acid extractable Zn decreased gradually from 2 to 13 months. The decreases in nutrient availability can be attributed to losses of nutrient as leaching and sugarcane uptake. Meanwhile, fixation and chelating of Zn to organic matter, clay minerals, and carbonates may also remove it from the available pools. Zinc concentrations in acetic acid and Mehlich-3 extracts did n ot differ, but both extracted 943% and 817% more Zn than water extracts, respectively (Table 6 1). Soil Properties and Micronutrient Availability Organic matter and soil pH are two major properties that influence nutrient availability and mobility (Her encia et al., 2008; Provin et al., 2008). Organic matter provides ligands that chelate the micronutrients and promotes the formation of soluble micronutrient organic matter complexes and therefore increases nutrient availability (Herencia et al., 2008). However, organic matter can also immobilize nutrients through the same complexation mechanism (Wei et al., 2006). In the present study, organic matter content was only significantly correlated with Mehlich3 extractable Fe and acetic acid extractable Mn ( Tables 6 -2, 6 -3, and 6 -4), indicating that organic matter was unlikely the dominant factor influencing nutrient availability during the sugarcane growing season. Changes in soil pH can mobilize nutrients from unavailable phases to available pools. Studies have shown that the availability of nutrients to crops depends on soil

PAGE 129

129 pH (Wei et al., 2006). In the EAA, elemental S is introduced as soil amendment for the purpose of reducing pH and therefore increasing nutrient availability (Rice et al., 2006). Our results suggested that application of S up to 448 kg ha1 introduced limited effects on soil pH and therefore had little influence on enhancing nutrient availability. Statistical analysis revealed that pH was significantly correlated with acetic acid ext ractable Fe and Zn and Mehlich3 extractable Cu and Mn (Table 6 -2 and 6 -3). Nonetheless, no correlations were found between soil pH and any of water extractable nutrients (Table 6 4), which suggests that soluble nutrients were not as sensitive as adsorbed or complexed nutrients to small changes in soil pH. Phosphate can affect micronutrient availability by direct precipitation of nutrient cations. However, the effect varies among micronutrients and depends on other soil properties, such as water content (Li et al., 2007), pH (Wei et al., 2006) and metal solubility (Shuman, 1988), which helps to explain the varied relationships between P and micronutrient concentrations (Table 6 2, 6 -3, and 6 4) in this study. Correlation analysis also revealed significantly negative correlations between water extractable Ca and Fe (R2 = 0.86) and Mn (R2 = -0.29) indicating that increasing CaCO3 content in these soils was likely to decrease micronutrient availability. Calcium carbonate is able to precipitate micronutrient ions in soil solution during the formation of carbonate depending on desorption characteristics of micronutrients and the solubility of carbonate (Wei et al., 2006). Mehlich3 extractable Ca was also significantly correlated to Mehlich3 extractable Fe and Mn, which may further

PAGE 130

130 suggest that Fe and Mn availability in these calcareous soils was affected by the CaCO3 content. Comparison of Soil Extractants Nutrients exist in soils as water soluble, exchangeable, and non exchangeable forms. The contribution of these pools towards nutrient availability to plants depends on the dynamic equilibrium among different fractions. Therefore, different extraction methods reflect the degree of nutrient availability (Wright et al., 2007). Water extracts represent the readily available chemical forms, whereas acetic acid and Mehlich-3, as acid solutions, extract the pool consisting of soluble, exchangeable, and some of nonexchangeable fractions (Cancela et al., 2002; Wright et al., 2007). Our results indicated th at, as expected, acetic acid and Mehlich-3 extracted more P, K, Ca, M g, Mn, and Zn than water (Table 6 -1). However, acetic acid and Mehlich-3 did not extract more Fe than water, but in fact water extracted more Fe than Mehlich3. Acetic acid and Mehlich3 have been deemed satisfactory extractants for soil testing on EAA soils (Korndrfer et al., 1995; Hochmuth et al., 1996). Mehlich3 solution contains large amounts of salts, strong acids, and EDTA. Salts are present mainly for extracting major cation such as P, K, Ca, and Mg, while micronutrient extraction is accomplished by metal -EDTA complexation (Wang et al., 2004). Nonetheless, compared to 0.5 N acetic acid, Mehlich3 extracted equal amounts of P, K, and Zn, but less Ca, Mg, Mn, and Fe (Table 6 1) indicating that acidity may in fact control the Ca, Mg, Mn, and Fe availability in this calcareous organic soil.

PAGE 131

131 Strong acidity can help to dissolve Ca, Mg, Mn, and Fe from precipitates in soils. Acetic acid is more acidic than Mehlich 3 and appeared l ess affected by soil buffering capacity and the presence of free CaCO3 (Korndrfer et al., 1995) and therefore extracted more Ca, Mg, Mn, and Fe. In other words, acetic acid method tends to extract relatively high amounts of nonexchangeable nutrients and thus may overestimate the concentrations of available nutrients. Mehlich3 extracted more Cu than acetic acid and water, suggesting that extractable Cu was likely present in a complex with organic matter rather than as an insoluble precipitate. In fact, copper is often associated with dissolved organic matter (Wright et al., 2007). Nutrient Availability and Sugar Yield Sulfur amendment did increase the availability of P, K, and Zn at 2 months after application. Nonetheless, S application did not increase sugar yield. The yields for soils amended with 0, 112, 224, and 448 kg S ha1 averaged 16, 17, 16, and 17 Mg sugar ha1, respectively. Results suggest that current recommended S application guidelines and rates in the EAA may not be high enough to ac hieve the projected goals. Higher application rates may be required to overcome the soils buffering capacity and significantly increase nutrient availability and sugar yield. Our results also indicated that S application is likely to increase the risk o f SO4 export from fields. Large scale S application should be well evaluated, since SO4 export from the EAA into Everglades wetlands has been implicated in causing stimulation of Hg methylation (Gabriel et al., 2008). Nonetheless, it has been reported th at actual grower S application rates in the EAA are lower than the current recommended rates

PAGE 132

132 (Schueneman, 2001). Everglades Agricultural Area growers tend to use micronutrient sprays to alleviate nutrient deficiency caused by elevated pH since it is more cost effective, so S application may not be considered necessary at this stage. Significant linear regression equations for prediction of sugar yields with extractable nutrients are listed in Table 6 5. Stepwise multiple regressions identified the most significant model considering Mehlich3 extractable P prior to planting as the main predictor, which explains 93% of the variation in sugar yield. Nonetheless, the correlation between Mehlich-3 extractable P and sugar yield was negative. As describe d previously, Mehlich 3 may not accurately reflect the P availability to sugarcane during the growing season. In fact, P measured as Mehlich-3 extractable may overestimate available P since it extracts P found in Ca and FeAl fractions which are considered unavailable to crops (Wright, 2009), which may explain the negative correlation between Mehlich3 extractable P and sugar yield. Regression equations for 2, 6, and 9 months had low coefficients of determinant ranging from 0.30 to 0.60, suggesting import ant factors influencing sugar yield were not quantified (Anderson et al., 1999). Meanwhile, the response of sugar yield to a specific factor may not be linear. General linear models only offered rough approximations of the relationships and therefore may not be adequate in this case (Korndrfer et al., 1995; Anderson et al., 1999). Conclusions Sulfur application at rates up to 448 kg ha1 had limited effects on the reduction in soil pH due to the high soil buffering capacities and generally failed to en hance the

PAGE 133

133 nutrient availability. Correspondingly, S application at current recommendation rates did not increase sugar yield. Considering the increasing pH and the decreasing depth to bedrock of soils in the EAA, new S application guidelines with higher amendment rates may be needed. Sulfur application increased SO4 concentrations in soils and also the risk for export from fields. Therefore, large scale of S application should be evaluated for their potential to adversely affect proximal sensitive wetland ecosystems. However, it may not be economically viable to increase S fertilizer recommendations because of the high cost of elemental S. An alternative, such as different P and micronutrient fertilizer application methods, timings, and sources, may be a better alternative to increase nutrient availability in these changing soils. Multiple regression analysis suggested that the parameter most influencing sugar yield was available P.

PAGE 134

134 Table 6 1. Comparisons of soil test extractants on concentrations of available nutrients (mg kg1). Values denote the mean across S rates and time with standard error values in parenthesis Acetic acid Mehlich 3 Water P 76 (11) 53 (4) 8 (1) K 387 (41) 313 (32) 188 (27) Ca 18936 (600) 16621 (287) 532 (54) Mg 957 (5 9) 664 (12) 76 (8) Cu 0.19 (0.02) 0.51 (0.08) 0.20 (0.01) Mn 15 (0.8) 4.7 (0.1) 0.2 (0) Fe 10 (0.4) 7 (0.2) 11 (0.6) Zn 3.2 (0.3) 2.9 (0.2) 0.3 (0) Table 6 2. Pearson correlation coefficients (r) between pH, organic matter content, and concentrations of acetic acid extractable nutrients (n= 64). pH OM Ca Cu Fe K Mg Mn P Zn pH 1 OM 0.45 1 Ca 0.29 0.71 1 Cu NS NS NS 1 Fe 0.51 NS NS 0.41 1 K 0.61 NS 0.32 0.25 0.80 1 Mg 0.31 NS 0.66 0.38 0.78 0.79 1 Mn NS 0.28 0.64 0.47 0.78 0.73 0.88 1 P 0.31 NS NS NS 0.37 0.60 NS 0.26 1 Zn 0.29 NS NS 0.55 0.41 0.49 NS 0.38 0.66 1 Notes: OM, organic matter content; *, significant a

PAGE 135

135 Table 6 3. Pearson correlation coefficie nts (r) between soil pH, organic matter content, and concentrations of Mehlich-3 extractable nutrients (n=64). pH OM Ca Cu Fe K Mg Mn P Zn pH 1 OM 0.45 1 Ca 0.35 0.29 1 Cu 0.28 NS NS 1 Fe NS 0.37 0.60 0.38 1 K 0.61 NS 0.58 0.44 0.44 1 Mg 0.38 NS 0.91 0.32 0.66 0.54 1 Mn 0.36 NS 0.67 0.57 0.71 0.45 0.77 1 P 0.34 NS NS NS NS 0.52 NS NS 1 Zn NS NS NS NS 0.46 NS NS 0.26 0.36 1 Notes: OM, organic matter content; *, significant at Table 6 4. Pearson correlation coefficients (r) between soil pH, organic matter content, and concentrations of water extractable nutrients (n = 64). pH OM Ca Cu Fe K Mg Mn P Zn pH 1 OM 0.45 1 Ca 0.3 9 NS 1 Cu NS NS NS 1 Fe NS NS 0.86 NS 1 K 0.56 NS 0.90 NS 0.74 1 Mg 0.42 NS 0.99 NS 0.86 0.90 1 Mn NS NS 0.29 0.26 NS 0.27 0.28 1 P 0.32 NS 0.42 NS 0.26 0.63 0.39 NS 1 Zn NS NS NS 0.54 NS NS NS 0.42 NS 1 Notes: OM, organic matter content; *, significant at Table 6 5. Multiple regression models relating soil nutrient concentrations (mg kg1) with sugar yield (Mg ha1) at different times. Time Equation R 2 0 months Y = 57 1.42 (M P) 0.93 2 months Y = 20 0.04 (W Mg) + 0.01 (W SO 4 ) 0.52 6 months Y = 21 54.06 (W Mn) 0.30 9 months Y = 14 0.04 (A K) + 0.17 (W K) 0.60 13 months Y = 10 + 0.13 (NO 3 N) + 15.09 (W Mn) 1.06 (W P) 0.81 Notes: Y, sugar yield; M, Mehlich3 extractable nutrient; W, water extractable nutrient; A acetic acid extractable nutrient.

PAGE 136

136 5.0 5.5 6.0 6.5 7.0 26 9 13 Months after S ApplicationSoil pH0 112 224 448 5.0 5.5 6.0 6.5 7.0 26 9 13 Months after S ApplicationSoil pH0 112 224 448 0 112 224 448 Fig 6 1. Soil pH changes in response to different S application rates (0, 112, 224, and 448 kg S ha1) throughout the sugarcane growing season. Error bars represent the standard error of the mean.

PAGE 137

137 Concentration (mg kg1) 0 100 200 300 400 500 Nitrate 0 100 200 300 400 500 0 2 4 6 8 10 12 14 Months after S Application 0 112 224 448 Sulfate 4 12 20 28 36 1000 1500 2000 2500 Dissolved Organic Carbon AmmoniumConcentration (mg kg1)0 2 4 6 8 10 12 14 Months after S Application Concentration (mg kg1) 0 100 200 300 400 500 Nitrate 0 100 200 300 400 500 0 2 4 6 8 10 12 14 Months after S Application 0 112 224 448 Sulfate 4 12 20 28 36 1000 1500 2000 2500 Dissolved Organic Carbon AmmoniumConcentration (mg kg1)0 2 4 6 8 10 12 14 Months after S Application Fig 6 2. S easonal dynamics of dissolved organic C, extractable NH4N, NO3-N, and SO4-S after S application at 0, 112, 224, and 448 kg S ha1. Error bars represent the standard error of the mean.

PAGE 138

138 0 112 224 448 0 100 200 300 400 500Acetic acid Extractable (mg kg1)Potassium Phosphorus 0 40 80 120 160Mehlich 3 Extractable (mg kg1)Months after S Application 0 200 400 600 800 1000Water Extractable (mg kg1) 0 10 20 30 40 0 2 4 6 8 10 12 14 Months after S Application 0 250 500 750 1000 1250 1500 0 250 500 750 1000 1250 1500 0 2 4 6 8 10 12 14 0 112 224 448 0 100 200 300 400 500Acetic acid Extractable (mg kg1)Potassium Phosphorus 0 40 80 120 160 0 40 80 120 160Mehlich 3 Extractable (mg kg1)Months after S Application 0 200 400 600 800 1000Water Extractable (mg kg1) 0 10 20 30 40 0 2 4 6 8 10 12 14 Months after S Application 0 250 500 750 1000 1250 1500 0 250 500 750 1000 1250 1500 0 2 4 6 8 10 12 14 Fig 6 3. Seasonal dynamics of acetic acid, Mehlich-3, and water e xtractable P and K after S amendment at 0, 112, 224, and 448 kg S ha1. Error bars represent the standard error of the mean.

PAGE 139

139 Acetic acid Extractable (mg kg1) Mehlich 3 Extractable (mg kg1) Water Extractable (mg kg1) 12500 17500 22500 27500 32500 Magnesium Calcium 500 1000 1500 0 2000 2500 12000 16000 20000 24000 400 600 800 1000 0 500 1000 1500 0 50 100 150 200 250 0 112 224 448 Months after S Application 0 2 4 6 8 10 12 14 Months after S Application0 2 4 6 8 10 12 14Acetic acid Extractable (mg kg1) Mehlich 3 Extractable (mg kg1) Water Extractable (mg kg1) 12500 17500 22500 27500 32500 Magnesium Calcium 500 1000 1500 0 2000 2500 12000 16000 20000 24000 400 600 800 1000 0 500 1000 1500 0 50 100 150 200 250 0 112 224 448 Months after S Application 0 2 4 6 8 10 12 14 Months after S Application0 2 4 6 8 10 12 14 Fig 6 4. Seasonal dynamics of acetic acid, Mehlich-3, and water extractable Ca and Mg after S amendment at 0, 112, 224, and 448 kg S ha1. Error bars represent the standard error of the mean.

PAGE 140

140 5 7 9 11 13 15 IronCopper 0 0.2 0.4 0.6 0.8 0 0.5 1.0 1.5 2.0 2 4 6 8 10 0.1 0.2 0.3 0.4 0.5 0 4 8 12 16 20 0 112 224 448 Months after S Application 2 4 6 8 1012 14 Months after S Application 2 4 6 8 10 12 14Acetic acid Extractable (mg kg1) Mehlich 3 Extractable (mg kg1) Water Extractable (mg kg1)0 0 5 7 9 11 13 15 IronCopper 0 0.2 0.4 0.6 0.8 0 0.5 1.0 1.5 2.0 2 4 6 8 10 0.1 0.2 0.3 0.4 0.5 0 4 8 12 16 20 0 112 224 448 Months after S Application 2 4 6 8 1012 14 Months after S Application 2 4 6 8 10 12 14Acetic acid Extractable (mg kg1) Mehlich 3 Extractable (mg kg1) Water Extractable (mg kg1)0 0 Fig 6 5. Seasonal dynamics of acetic acid, Mehlich-3, and water extractable Cu and Fe after S amendment at 0, 112, 224, and 448 kg S ha1. Error bars represent the standard error o f the mean.

PAGE 141

141 0 5 10 15 20 25 30 Manganese 0 2 4 6 8 10 12 Zinc 2 4 6 8 0 2 4 6 8 10 0 0.1 0.2 0.3 0.4 0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14 Months after S Application Months after S Application 0.1 0.2 0.3 0.4 0.5 0.6 0 112 224 448Acetic acid Extractable (mg kg1) Mehlich 3 Extractable (mg kg1) Water Extractable (mg kg1) 0 5 10 15 20 25 30 Manganese 0 2 4 6 8 10 12 Zinc 2 4 6 8 0 2 4 6 8 10 0 0.1 0.2 0.3 0.4 0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14 Months after S Application Months after S Application 0.1 0.2 0.3 0.4 0.5 0.6 0 112 224 448Acetic acid Extractable (mg kg1) Mehlich 3 Extractable (mg kg1) Water Extractable (mg kg1) Fig 6 6. Seasonal dynamics of acetic acid, Mehlich-3, and water extractable Mn and Zn after S amendment at 0, 112, 224, and 448 kg S ha1. Error bars represent the standard error of the mean.

PAGE 142

142 CHAPTER 7 SULFUR POOLS, TRANSF ORMATIONS, AND MINERALIZATION IN EVERGLADES AGRICULTU RAL AREA SOILS Introduction Sulfur is an element required by all living organisms as an essential macronutrient (Wang et al., 2006) Sulfur exists in soils in various forms, each of which play important biological and chemical functions Sulfate is the most abundant form of inorganic S found in most soil s as well as the main form available to plants although reduced forms, such as elemental S, thiosulfate, and sulfide, are important for anaerobic soils (Zhou et al., 2005). However, the bulk of soil S in natural and managed ecosystems is in organic form which is directly impacted by microbial activity through decomposition processes (Solomon et al., 2001) Sulfur dynamics is widely varia ble among soils and often closely associ ated with other nutrient cycles It is widely believed that S mineralization in soils involves both biological and biochemical processes (McGill and Cole, 1981) As in biological process, SO4 is released as a by -product during organic matt er decomposition while biochemical mineralization releases SO4 through hydrolysis of ester SO4, which is catalyzed by extracellular enzyme activity, primarily sulfatase (McGill and Cole, 1981; Wright and Reddy, 2001). Water quality is a critical issue fa cing rehabilitation of the Florida Everglades (Gabriel at al., 2008), as evidenced by SO4 contamination of the northern Everglades, which has been implicated in the stimulation of MeHg formation in soils and water ( Bates et al., 2002). Methylmercury is a neurotoxin that is bioaccumulated

PAGE 143

143 in higher organisms and found at high concentrations in fish and other wildlife in the Everglades (Orem, 2007). Potential sources contributing to the SO4 enrichment in Everglades wetlands include ground water, rainwater, s ea aerosol, internal S flux from sediments, and surface water inputs from Lake Okeechobee and the Everglades Agricultural Area (EAA) (Schueneman, 2002). I t has been reported that S from the EAA is the likely key contributor (Schueneman, 2001; Bates et al. 2002; Orem 2007). The E AA is located south of Lake Okeechobee and north of the Water Conservation Areas (WCA) of south Florida Historically, it was a seasonally -flooded prairie ecosystem, but was converted to agricultural use by drainage in the earl y 1900s The soils of the EAA are predominately H istosols w ith high organic matter content and but low P and micronutrient concentrations that require supplemental fertiliza tion (Snyder, 2005; Ye et al., 2009). Since this area has changed from the wetlan d to agricultural ecosystem in the 1920s, several nutrient deficiencies were evident, mainly P but also micronutrients, particularly Cu. Thus, CuSO4 was used to alleviate Cu deficiency to crops (Allison et al., 1927). Sulfur has also been applied to soil s as part of pesticides that were commonly used to support sugarcane and vegetable production. In recent years, formation of shallow soils resulting from soil subsidence (Snyder, 2005; Wright, 2009) and increasing pH due to incorporation of limestone bed rock into soils has increased the need for amendments to decrease soil pH. Elemental S is recommended to reduce soil pH when it exceeds 6.6 for the purpose of improving the availability of P and micronutrients to sugarcane

PAGE 144

144 (Anderson, 1985; Schueneman, 2001). The microbial oxi dation of elemental S to SO4 produces acidity which reacts with the soil and reduces pH, which in turn increases P and micronutrient availability However, SO4 is soluble in water and su sceptible to export from the field as runoff d uring precipitation events to down stream Everglades wetlands. In consideration of the adverse impacts that S may pose to Everglades wetlands, reducing potential S export from the EAA is beneficial for protecting water quality and ecosystem health (Gabri el at al., 2008). N onetheless, explicit quantification of S budgets and transformations within EAA soils is rare. Due to the increasing pH trend for EAA soils, demand for S application may continue to exist or increase in the future. Minimal research is available for S cycling in Everglades soils, such that there is a strong need to determine the influence of elemental S application on soil pH, S distribution, and transformation in soils Materi a ls and Methods Site Description The experimental field is located in the central EAA on Dania muck (euic, hyperthermic, shallow Lithic Haplosaprist) with a depth to bedrock of approximately 50 cm. The experimental design was a randomized complete block with four S application rates and four field replications w ith four sampling times encompassing the entire growing season. Each field plot measured 9 m x 13 m and consisted of 6 rows of sugarcane (Saccharum spp. ) cultivar CP 892143 planted in November 2007 and harvested in February 2009. Elemental granular S (9 0%) was applied at rates of

PAGE 145

145 0, 112, 224, and 448 kg S ha1 to the furrow and covered after planting. Other fertilization was provided using typical recommendations and guidelines for this region and soil type (Gilbert and Rice 2006). F e rtilizers were soilapplied prior to planting and all field plots received 17 kg N ha1 and 37 kg P ha1 as monoammonium phosphate, 228 kg K ha1 as KCl, 8.5 kg Mn ha1, 4.5 kg Cu ha1, 5.6 kg Fe ha1, 2.8 kg Zn ha1, and 1.1 kg B ha1. All plots received common cultural practices including tillage and herbicide application. Water was applied as needed via seepage irrigation in field ditches approximately 182 m apart. Soil Sampling and Laboratory Analysis Soil samples were collected before planting and fertilizer applic ation and then in January 2008, May 2008, August 2008, and December 2008, corresponding to approximately 2 6 9 and 13 months after planting, respectively. Twelve soil (015 cm) cores (2.54 cm diameter) were randomly collected from rows within each fiel d plot and composited to yield one sample per plot Soils were homogenized after the removal of visible plant residues and stored at 4C until analysis Soil pH was measured using a soil to water ratio of 1:3 after equilibration for 30 min. Total organi c C was measured by loss on-ignition at 550C for 4 h r after conversion to organic C with a coefficient factor of 0. 51 (Wright et al. 2008). Total N was measured by Kjeldahl digestion followed by NH4 analysis (Bremner, 1996). Extractable NH4-N and NO3-N were determined by extraction with 2 M KCl followed by colorimetric analysis ( Ye et al., 2009). T otal P was measured using the ascorbic acid molybdenum blue method after Kjeldahl digestion, and l abile inorganic P was

PAGE 146

146 measured after Mehlich 3 extraction ( Ye et al., 2009). Acetic acid extractable nutrients were measured according to guidelines for muck soils (Sanchez 1990) by extracting 4 g soil with 25 mL of 0.5 N acetic acid for 1 hr, then filtering through Whatman #42 filters. Extracts were analyzed for Ca, Mg, Fe, and Al concentrations by ICP using EPA method 200.7. Select soil nutrient concentrations and properties before S application are listed on Table 7 -1. Water extractable SO4-S was analyzed by ionic chromatography (Perkin -Elmer, Waltham, MA) aft er shaking 2 g field soil with 25 mL water for 0.5 hr, followed by filtering through Whatman No. 42 filter p aper. Aliquots of extracts were analyzed for total extractable S by ICP. Extractable organic S was calculated by subtracting extractable SO4-S fro m total extractable S. Elemental S was determined as described by Pansu and Gautheyrou (2006), with slight modification. Approximately 5 g soil was extracted with 10 mL acetone for 30 min and centrifuged at 5,000 g for 15 min. Extracts were then analyzed for S content colorimetrically at 420 nm. To avoid interference by organic matter, unamended soils were used as controls to calibrate all readings. Potentially mineralizable S was measured based on the methods of White and Reddy (2000) using a 10 day i ncubation followed by extraction with water. Extracts were analyzed for SO4-S as previously described. The sulfatase activity in soils was assayed using the colorimetric methods described by Wright and Reddy (2001).

PAGE 147

147 Statistical Analysis A mixed model wa s fit using restricted maximum likelihood in the MIXED procedure of SAS (Littell et al., 2006). The fixed effects were S application rate, time and their interaction. Block was a random effect. Degrees of freedom were adjusted using the Kenward-Roger ad justment. An exponential covariance structure was used to model the correlation among observations taken from the same plot over time. Significant differences among individual treatments and time intervals were Pearson correlation was employed to determine relationships between variables All statistical analysis was carried out with SAS 9.1 (SAS Institute). Results and Discussion Soil pH Elemental S is commonly considered beneficial in alkaline soils to lower soil pH, supply SO4 to plants, and increase P and micronutrient availability (Lindemann et al., 1991; Schueneman, 2001; Yang et al., 200 8). However, the effectiveness of elemental S is n ot observed until elemental sulfur is oxidized, which depends on factors such as application rates, soil p roperties, and the activity of S oxidizing microorganisms (Yang et al., 2008). In the present stud y, S application did not significantly reduce soil pH during the growing season (Fig. 7 -1). A limited reduction in soil pH after S application was also observed in other studies of calcareous soils (Hassan and Olson, 1966). The limited effect of acidific ation may result from a S application rate too low to cause a change in pH and from the high buffering capacity

PAGE 148

148 of this calcareous organic soil. S oils with high concentrations of carbonates and bicarbonates are highly buffered against acidification (Rogov ska et al, 2007). When the original S recommendation for sugarcane of 448 kg S ha1 was established years ago (Anderson, 1985), soil pH in the EAA was considerably lower. However, the rise in pH and decrease in soil depth to bedrock since that time (Snyd er, 2005) increased the capacity of these soils to resist changes in pH. Thus, higher S application rates may be necessary to produce the same response as 448 kg ha1 did in the 1980s (Anderson, 1985). Extractable SO4-S Elemental S application significa ntly increased SO4S concentrations in soils throughout the growing season (Fig. 7 2 ). E xtractable SO4S in soils receiving 448 kg S ha1 was 36%, 131% 201%, and 270% higher than unamended soils at 2, 6, 9, and 13 months, respectively. Sulfate -S concent rations significantly decreased from 2 (376 mg kg1) to 6 months (129 mg kg1), and continuing to 9 months (21 mg kg1). The declining trend in SO4-S concentrations was likely due to SO4 uptake by sugarcane and loss as runoff or leaching during precipitat ion events. Extractable SO4-S was significantly correlated to extractable organic S ( R2 = 0.89), elemental S (R2 = 0.52), and potentially mineralizable S ( R2 = 0.39), which may suggest that microbial oxidation of elemental S and mineralization of organic S were two major sources of soil SO4-S (Jaggi et al., 2005 ; Zhou et al., 2005). In the EAA, soil oxidation generally supplies sufficient SO4 to satisfy sugarcane nutrient requirements (Gilbert and Rice, 2006), therefore S application at high rates is likely to increase

PAGE 149

149 SO4 concentrations in the soil and thus the risk of export from fields and into sensitive Everglades wetlands. Extractable Organic S Extrac table organic S contains S associated with particular organic matter and generally accounts for only a small proportion of the total soil S (Dias et al., 2003). However, it is likely to be readily mineralized to SO4 and as such this pool is considered an important source of available S to crops, especially in soils containing low inorganic SO4 (Dias et al., 2003; Kaiser and Guggenberger, 2005). Extractable organic S was not affected by S application during the growing season, but exhibited a similar decreasing pattern as extractable SO4-S, indicating that extractable organic S was as mobile as SO4 in th ese organic soils (Fig. 7 3). Extractable organic S averaged 58, 22, 4, and 16 mg S kg1 at 2, 6, 9, and 12 months, respectively, and was significantly correlated to elemental S ( R2 = 0.44) and potentially mineralizable S (R2 = 0.26), and extractable SO4-S (R2 = 0.89), suggesting that production of extractable organic S and S mineralization were controlled by the same factors (Valeur et al., 2000). Elemental S Elemental S was not detected in unamended soils during the growing season, but it was significantly higher in soils receiving 448 kg S ha1 (215 mg kg1) than soils receiving 112 (44 mg kg1) and 224 kg S ha1 (20 mg kg1) (Fig. 7 4). Elemental S contents in soils receiving the highest S rate was 771% and 334% higher than soils amended with 112 and 224 kg S ha1 at 2 months, respectively. However, at 13

PAGE 150

150 months after S application, elemental S was only detected in soils receiving 448 (9 mg kg1) and 224 kg S ha1 (0.5 mg kg1). The decreasing patterns of elemental S in soils throughout the season w ere due to the oxidation of elemental S to SO4. It has been reported that oxidation of elemental S in some calcareous soils is slow and may take several years (Lindemann et al., 1991; Cifuentes and Lindermann, 1993). Our results showed that a relatively high concentration of elemental S persisted in soils receiving 448 kg S ha1 by 13 months after application, suggesting that higher S application in these calcareous organic soils is li kely to maintain high levels of elemental S and SO4 in soils for long p eriods of time. Sulfatase Activity Sulfatase is an en zyme that hydrolyzes ester S and releases SO4, and hence plays an important role in organic S min eralization (Chen et al., 2001). Sulfur application at a range from 0 to 448 kg ha1 had minimal effect s on sulfatase activity during the growing season (Fig. 7 -5). Sulfatase activity can be influenced by several soil properties, such as SO4 concentration, organic matter, pH, and seasonal variations in soil moisture (Knauff et al., 2003) Decomposition of organic matter in this Histosol typically supplies enough S needed for crop growth (Snyder, 2005), thus sulfate is probably at a high enough concentration to minimize sulfatase activity (Wright and Reddy, 2001). Sulfatase activity averaged 240, 157, 223, and 216 mg p -nitrophenol kg1 h1 at 2, 6, 9, and 13 months after S application.

PAGE 151

151 Organic S Mineralization Potential S mineralization rates increased concurrently with increasing S application rates, and the effects continued throughout the growing seas on (Fig. 7 6). Overall, mineralized S was 1421% greater for soils receiving 448 kg S ha1 than unamended soil, 375% greater than soils receiving 112 kg S ha-1 and 219% greater for soils receiving 224 kg S ha1. However, mineralized S rates significantly decreased from 2 (16 mg kg1 d1) to 6 months (12 mg kg1 d1), and continued to decrease from 9 (11 mg kg1 d1) to 13 months (3 mg kg1 d1). It has been well recognized that organic S is mineralized to SO4 by hydrolysis of ester SO4 catalyzed by sulfa tase or by mineralization of C -bound S due to microbiological activity ( McGill and Cole, 1981; Chen et al., 2001; Gharmakher et al., 2009). In the present study, no significant correlation between mineralized S and both sulfatase activity and C mineraliza tion rates was found (data not shown). Instead, S mineralization rates were significantly correlated to S application rate ( R2 = 0.60), elemental S concentrations ( R2 = 0.85), extractable SO 4 -S ( R2 = 0.39), and extractable organic S (R2 = 0.26). It was l ikely that oxidation of elemental S to SO4 was primarily responsible for the increased mineralized S rates rather than organic S mineralization (Eriksen et al., 1998). Conclusions Sulfur application under current recommendations and guidelines for sugarcan e had limited effects on the reduction of soil pH, thus its use for enhancing soil nutrient availability appears limited. Higher rates than currently recommended

PAGE 152

152 may be needed to affect a change in soil pH, which would then have the effect of increasing nutrient concentrations in soil for longer duration during the growing season. However, S application at 448 kg ha1 significantly increased elemental S and SO4 concentrations in soil solution. Sulfur application did not stimulate the sulfatase activities during the growing season, whereas it significantly enhanced potential S mineralization rates, which was largely attributed to the oxidation of elemental S. Both extractable SO4 and dissolved organic S decreased significantly throughout the growing season likely due to uptake by sugarcane, but also potentially by runoff or leaching through the shallow soils. Large -scale S amendment of EAA soils, or an increase in S application rates, is likely to increase SO4-S concentrations in soil, which may enhance t he potential for S export to sensitive Everglades wetlands during field drainage or precipitation events, leading to S enrichment of downgradient wetlands and contributing to the stimulation of MeHg.

PAGE 153

153 Table 7 1 C hemical properties of the Histosols in t he Everglades Agricultural Area before fertilizer application. Soil P roperty Unit Concentration Total o rganic C g kg 1 416 Total N g kg 1 38 Total P mg kg 1 850 Extractable NO 3 N mg kg 1 290 Extractable NH4-N mg kg 1 16 Extractable P mg kg 1 48 Extr actable Ca mg kg 1 720 Extractable Mg mg kg 1 105 Extractable Fe mg kg 1 13 Extractable Al mg kg 1 1.1

PAGE 154

154 5.0 5.5 6.0 6.5 7.0 2 6 9 13 Months after S ApplicationSoil pH0 112 224 448 5.0 5.5 6.0 6.5 7.0 2 6 9 13 Months after S ApplicationSoil pH0 112 224 448 0 112 224 448 Fig 7 1. Soil pH changes in response to different S application rates (0, 112, 224, and 448 kg S ha1) throughout the sugarcane growing season. Error bars represent the standard error of the mean.

PAGE 155

155 0 100 200 300 400 500 0 2 4 6 8 10 12 14 Month after S ApplicationExtractable SO4S (mg kg1) 0 112 224 448 0 100 200 300 400 500 0 2 4 6 8 10 12 14 Month after S ApplicationExtractable SO4S (mg kg1) 0 112 224 448 Fig 7 2. Seasonal dynamics of extractable SO4-S after S application at 0, 112, 224, and 448 kg S ha1. Error bars represent the standard error of the mean. 0 20 40 60 80 0 2 4 6 8 10 12 14 Months after S Application Extractable Organic S (mg kg1) 0 112 224 448 0 20 40 60 80 0 2 4 6 8 10 12 14 Months after S Application Extractable Organic S (mg kg1) 0 112 224 448 Fig 7 3. Seasonal dynamics of extrac table organic S after S application at 0, 112, 224, and 448 kg S ha1. Error bars represent the standard error of the mean.

PAGE 156

156 0 100 200 300 400 500 600 0 2 4 6 8 10 12 14 Months after S ApplicationElemental S (mg kg1) 0 112 224 448 0 100 200 300 400 500 600 0 2 4 6 8 10 12 14 Months after S ApplicationElemental S (mg kg1) 0 112 224 448 Fig 7 4. Seasonal dynamics of elemental S after S application at 0, 112, 224, and 448 kg S ha1. Error bars represent the standard error of the mean.

PAGE 157

157 0 50 100 150 200 250 300 2 6 9 13 Months after S ApplicationSulfatase Activity (mg p nitrophenol kg1h1)0 112 224 448 0 50 100 150 200 250 300 2 6 9 13 Months after S ApplicationSulfatase Activity (mg p nitrophenol kg1h1)0 112 224 448 0 112 224 448 Fig 7 5 Sulfatase activities in response to different elemental S application rates (0, 112, 224, and 448 kg S ha1) throughout the sugarcane growing season. Error bars represent the standard error of the mean.

PAGE 158

158 0 10 2030 40 50 2 6 9 13 Months after S application 0 112 224 448Potentially Mineralizable S (mg kg1d1) 0 10 2030 40 50 2 6 9 13 Months after S application 0 112 224 448 0 112224 448Potentially Mineralizable S (mg kg1d1) Fig 7 6 P otential S m ineraliz ation in soils at 2, 6, 9, an d 13 months after S application at different rates (0, 112, 224, and 448 S kg ha1). Error bars represent the standard error of the mean.

PAGE 159

159 CHAPTER 8 SYNTHESIS Laboratory and field studies were conducted to characterize chemical properties and microbial activities of EAA soils with different management history (Chapter 2). Multivariate analytical methods were utilized to assess differences between land uses and soils (Chapter 3). The impacts of S amendment on P distribution, availability, and stability were investigated for soils under sugarcane production (Chapter 4). The effects of S on soil microbial eco -physiological response, as measured using extracellular enzyme activities and potential mineraliza tion rates, were determined for sugarcane soils (Chapter 5). The effectiveness of S amendment on enhancing macroand micronutrient availability during the sugarcane growing season was evaluated (Chapter 6). Sulfur distribution and transformations were q uantified for S amended soils under sugarcane production (Chapter 7). A summary and implications of the experimental results related to objectives is present below. Land Use Effects on Soil Nutrient Cycling and Microbial Community Dynamics in the Everglad es Agricultural Area, Florida Four land uses (sugarcane, turfgrass, forest and uncultivated soil) were characterized for nutrient cycling and microbial activity. Longterm cultivation and management significantly altered nutrient distribution and cycling in soil profiles, as well as the microbial community composition and activity Turf soils, followed by forest and sugarcane, had the highest diversity of C sources and utilization rates of C resources, while the uncultivated soils had the lowest diversity The uncultivated soils were the least managed, but the most P limited and therefore had the smallest size of microbial population and the lowest C and N utilization efficiency. Labile inorganic P played an important role in regulating organic matter de composition and microbial community

PAGE 160

160 structure and function. Land use changes from sugarcane cropping to turf increased microbial activity and organic matter decomposition rates, indicating that changes from agricultural to urban land uses may further cont ribute to soil subsidence. Nonetheless, land use change from sugarcane cropping to uncultivated sites tends to slow down the oxidation rates of organic matter and subsequently may minimize soil subsidence. Multivariate Analysis of Chemical and Microbial P roperties in Histosols as Influenced by LandUse Types Soils from three land uses with different management intensity and history were analyzed for chemical and microbial properties. Cluster analysis on chemical properties demonstrated that soils from dif ferent land uses were perfectly clustered into their own groups, which were distinguished by labile inorganic P and total P. Likewise, integrated soil microbial characteristics were distinctive between land uses. Microbial biomass C and N, community -level physiological profile components, and potentially mineralizable N contributed most to such discriminations. C anonical correlation analysis suggested that variations in microbial properties between land uses were largely explained by difference in soil c hemical properties, especially P availability. Longterm agricultural m anagement especially P fertilization, altered soil nutrient availability and consequently modified the microbial community structure and function. Intensive P fertilization is likely to stimulate microbial community and alter microbial processes so future land use changes should consider the role of labile P on microbial community function and its control of nutrient cycling. Sulfur I nduced Changes in Phosphorus Distribution in Ever glades Agricultural Area Soils Phosphorus fractionation procedures were performed to quantify P distribution in soil during the sugarcane growing season. The majority of P in this soil was retained in

PAGE 161

161 organic forms (63%) followed by Ca bound (32%), Fe-Al bound (4%) and labile fractions (1%). Labile and Fe Al bound P comprised of the majority of available P for crops, and the size of these pools decreased throughout the growing season. Under current sugarcane production, P storages in the organic pools are susceptible to oxidation and a potential source for P loss from fields. Sulfur application within a range from 0 to 448 kg S ha1 did not significantly decrease soil pH due the high buffering capacity against acidification. As a result, the stimulato ry effects of S on increasing labile and Fe-Al bound P were limited and temporary, whereas S addition did not impact the sizes of Ca-bound and organic P fractions. Therefore, S application under current recommendation guidelines and rates has minimal impa ct on increasing P availability beyond a short -term response ( 2 months) and is unlikely to enhance the potential for P export from agricultural fields into wetlands. Higher S application rates may overcome the soils buffering capacity and consequently r elease large amounts of P from the Ca bound fractions and pose an environmental hazard. Microbial EcoPhysiological Response of a Calcareous Histosol to Sulfur Amendment Application of elemental S at 448 kg S ha1 stimulated the activities of phosphatase and glucosidase and simultaneously promoted labile P to be immobilized into microbial biomass. However, these effects were temporary and not observed beyond 2 months due to the high buffering capacity of this organic soil against pH reduction. Microbial biomass C and N were not affected by S amendment The C, N, and P mineralization rates were independent of S addition, though all rates varied seasonally, suggesting that S application did not stimulate soil oxidation. U sing the

PAGE 162

162 current recommended S app lication guidelines, impacts on microbial activities and functions sh ould be minimal. However, due to the increasing pH trend for these soils there may be a need for higher S application rates in the future. These higher S rates may overcome the soils buffering capacity and release large amounts of P potentially stimulating microbial functional activities and altering organic matter dynamics. Seasonal Changes in Nutrient Availability in Sulfur Amended Everglades Soils under Sugarcane Soils under sugarc ane cultivation were amended with elemental S at four rates up to 448 kg S ha1 to decrease pH and enhance nutrient availability. Water extractable P and K for soils receiving the highest S rate were significantly higher than for unamended soils only at 2 months after application, indicating a short -term enhancement of macronutrient availability. Similarly, soils amended with 448 kg S ha1 contained more acetic acid extractable Zn than unamended soil, but the stimulatory effects did not extend beyond 2 months. The failure of S to enhance nutrient availability throughout the growing season indicates the limited benefit of applying elemental S to reduce pH and increase nutrient availability to sugarcane. As a result, S application did not increase sugar yi eld. Considering the trends of increas ing pH and the decreasing depth to bedrock of soils in the EAA, new S application guidelines with higher amendment rates may be needed. Sulfur application increased SO4 concentrations in soils and also its risk of ex port from fields. Therefore, large scale of S application should be evaluated for its potential to adversely affect proximal sensitive wetland ecosystems. It may not be economically viable to increase S fertilizer recommendations because of the high cost of elemental S. An alternative, such as different P and micronutrient fertilizer

PAGE 163

163 application methods, timings, and sources, may be a better alternative to increase nutrient availability for these changing soils. Three extractants, water, Mehlich-3, and acetic acid, were evaluated for their potential to extract plant available nutrients and their relationship with sugarcane yield. Generally, acetic acid and Mehlich-3 extracted more nutrients from these calcareous organic soils than water, whereas acetic acid extracted more nonexchangeable nutrients and may not actually reflect nutrient availability to sugarcane. Mehlich3 extractable P was identified as the single parameter most significantly correlated with sugar yield, so this extractant should be further evaluated for its potential to replace acetic acid as the extractant used for nutrient assessment for sugarcane grown on muck soils. Overall Conclusions Longterm cultivation and management has significantly altered the soil chemical properties, espe cially P availability, and microbial community composition and function in EAA soils. Long-term P fertilization has resulted in accumulation of P in soil profile and enhanced P availability, which consequently stimulated the microbial activity and function and organic matter turnover rates (Fig. 8 -1). Current land use as sugarcane cropping requires P fertilization (Rice et al., 2006) and therefore would continue to promote organic matter mineralization. Future land use and management should consider the impacts of P on microbial communities and their control of nutrient cycling Elemental S application under current recommendation rates introduced temporary and limited effects on increasing nutrient availability to sugarcane and posed minimal impacts on m icrobial activity and function. Therefore, elemental S application is not beneficial at this stage under the current recommendations. Large scale application

PAGE 164

164 is likely needed to overcome the high soil buffering capacity against acidification and produce desirable responses in terms of micronutrient availability and crop yield. However, S application at a large scale may stimulate nutrient regeneration rates and microbial activity and increase the risk of SO4 export from the fields into adjacent wetlands (Fig. 82).

PAGE 165

165 Soils under Sugarcane Fertilization (30 40 kg P ha1 yr1)Microbial Community P limitation Nutrients Organic Matter C:P = 1178:1 N:P = 67:1 Enzymatic Activity Labile P (96 mg1 kg1) Runoff and Leaching Soils under Sugarcane Fertilization (30 40 kg P ha1 yr1)Microbial Community P limitation Nutrients Organic Matter C:P = 1178:1 N:P = 67:1 Enzymatic Activity Labile P (96 mg1 kg1) Runoff and Leaching Fig. 8 1. Conceptual model of the microbial response to P fertilization in EAA soils under sugarcane. C:P, molar ratios of total carbon to phosphorus; N:P, molar ratios of total nitrogen to phosphorus.

PAGE 166

166 Runoff and Leaching Soils under Sugarcane Elemental S (448 kg S ha1) Microbial Biomass P limitation Nutrients Organic Matter pH = 6.2 OM = 81% Enzymatic Activity pH (5.8) + SO4S (441 mg kg1) Buffering C:P = 1263:1 N:P = 98:1 Runoff and Leaching Micronutrients Zn (7 mg kg1) Labile P (265 mg kg-1) Sugarcane Runoff and Leaching Soils under Sugarcane Elemental S (448 kg S ha1) Microbial Biomass P limitation Nutrients Organic Matter pH = 6.2 OM = 81% Enzymatic Activity pH (5.8) + SO4S (441 mg kg1) Buffering C:P = 1263:1 N:P = 98:1 Runoff and Leaching Micronutrients Zn (7 mg kg1) Labile P (265 mg kg-1) Sugarcane Fig. 8 2. Summary of biogeochem ical processes in EAA soils after S applications at 448 kg S ha1 at 2 months. C:P, molar ratios of total carbon to phosphorus; N:P, molar ratios of total nitrogen to phosphorus; OM, organic matter content; SO4S, water extractable sulfate; Labile P, acet ic acid extractable phosphorus.

PAGE 167

167 LIST OF REFERENCES Adhami E. Memarian, H.R Rassael F Mahdavi E. Maftoun M. Ronaghi A. Fasaei R G ., 2007 Relationship between phosphorus fractions and properties of highly calcareous so ils. Australian Journal of Soil Research 45, 255261. Allison, R.V., Byyan, O.C., Hunter, J.H., 1927. The simulation of plant response on the raw soils of the Florida Everglades through the use of copper sulphate and other chemicals. Florida Agricultural Experiment Station B ulletin 190, University of Florida, Gainesville, FL. Allison, S.D., Vitousek, P.M., 2005. Responses of extracellular enzymes to simple and complex nutrient inputs. Soil Biology & Biochemistry 37, 937 944. Allison, V.J., Condron, L.M., Peltzer, D.A., Richa rdson, S.J., Turner, B.L., 2007. Changes in enzyme activities and soil microbial community composition along carbon and nutrient gradients at the Franz Josef chronosequence, New Zealand. Soil Biology & Biochemistry 39, 1170 -1781. Anderson, D.L. 1985 Crop soil fertility recommendations of the Everglades soil te sting laboratory. EREC Belle Glade Report EV -1985 10. Univ. of Florida, Belle Glade FL. Anderson, D.L. Flaig, E.G. 1995 Agricultural best management practices and surface water improvement and m anagemen t. Water Science and Technology 31, 109121. Anderson, D.L., Hanlon, E.A., Bottcher, A.B., Ceric, A. 1994 Phosphorus in drainage waters from organic soils (Histosols) of Florida. SS LAH 1, Univ ersity of Florida, Gainesville FL. Anderson, D.L., Portier, K.N., Obreza, T.A.. Collins, M.E., Pitts, D.J., 1999. Tree regression analysis to determine effects of soil variability on sugarcane yields. Soil Science Society of America Journal 63, 592-600. Anderson, D.L. Rosendahl, P.C. 1998. Development an d management of land/water resources: the Everglades, agriculture, and south Florida. Journal of Ameri can Water Resources Association 34, 235-249. Anderson, J.M. 1976 An ignition method for determination of total phosphorus in lake sediments. Water Research 10, 329-331. Arai Y ., Livi K.J T Sparks D L ., 2005 Phosphate reactivity in longterm poultry litter amended southern Delaware sandy soils. Soil Sci ence Soc iety of America Journal 69, 616629.

PAGE 168

168 Banning, N.C., Murphy, D.V., 2008. Effect of heat induced disturbance on microbial biomass and activity in forest soil and the relationship between disturbance effects and microbial community structure. Applied Soil Ecology 40, 109119. Bates, A.L., Orem, W.H., Harvey, J.W., Spiker, E.C., 2002. Tracing sourc es of sulfur in the Florida Everglades. Journal of Environmental Quality 31, 287299. Beverly R.B., Anderson, D., 1986. Effects of acid source on soil pH. Soil Sci ence 143, 301-303. Bloom, P. R ., 2000 Soil pH and pH buffering. In: Summer M., (ed ), Handbook of soil science. CRC Press, Boca Raton, FL p B333-352. Bossio, D.A., Girvan, M.S., Verchot, L., Bullimore, J., Borelli, T., Albrecht, A., Scow K.M., 2005. Soil microbial community response to land use change in an agricultural landscape of western Ken ya. Microbial Ecol ogy 49 50 -62. Bremner, J.M., 1996. Nitrogen -total. In: Bartels J M ., (ed) Methods of soil analysis. Part 3. ASA and SSSA, Madison, WI, p 10851122. Campbell, C.D., Chapman, S.J., Cameron, C.M., Davidson, M.S., and Potts, J.M., 2003 A rapid microtiter plate method to measure carbon dioxide evolved from carbon substrate amendments so as to determine the physiological profiles of soil microbial communities by using whole soil. Appl ied and Environ mental Microbiol ogy 69, 3593 3599. Cancela, R.C., de Abreu, C.A., Paz -Gonzlez, A., 2002. DTPA and Mehlich-3 micronutrient extractability in natural soils. Communications in Soil Science and Plant Analysis 33 28792893. Carreira, J.A., Garca -Ruiz, R., Li tor J., Harrison A.F., 2000. Changes in soil phosphatase activity and P transformation rates induced by application of N and S-containing acidmist to a forest canopy. Soil Biology & Biochemistry 32, 1857 1865. Castillo, M.S., Wright, A.L., 2008a. Microbial activity and phosphorus availability in a subtropical soil under different land use. World Journal of Agricultural Science 4, 314-320. Cas tillo, M.S., Wright, A.L., 2008b. Soil phosphorus pools for Histosols under sugarcane and pasture in the Everglades, USA Geoderma 145, 130-135. Chen, C.R. Condron, L.M., Davis, M.R., Scherlock, R.R., 2001. Effects of landuse change from grassland to forest on soil sulfur and arylsulfatase activity in New Zealand. Australian Journal of Soil Research 39, 749757.

PAGE 169

169 Chen, M., Daroub, S.H. Lang T.A., Diaz O. A. 2006 Specific conductance and ionic characteristics of farm canal in the Everglades Agricultural Area. J ournal of Environmental Quality 35 141 -150. Childers D L ., Doren, R F Jone s, R., Noe, G.B., Rugge, M., Scinto L J ., 2003 Decadal change in vegetation and soil phosphorus patterns across the Everglades landscape. Journal of Environmental Quality 32, 344-362. Chimney, M.J., Goforth, G., 2006. History and description of he Everglades nutrient removal project, a subtropical constructed wetland in s outh Florida (USA). Ecological Engineering 27, 268-278. Cifuentes, F.R., Lindemann, W.C., 1993. Organic matter stimulation of elemental sulfur oxidation in a calcareous soil. Soil Science Society of America Journal 57, 727731. Coale, F J Sanchez C A Izuno, F T Bottcher A. B ., 1993 Nutrient accumulation and removal by sugarcane grown on Everglades histosols. Agronomy Journal 85, 310315. Codling E. E ., 2008. Effects of soil acidity and cropping on solubility of by -product immobilized phosphorus and extractable aluminum, calcium, and iron from two highphosphorus soils. Soil Science 173, 552559. Cookson, W.R., Osman, M., Marschner, P., Abaye, D.A., Clark, I., Murphy, D.V., Stochdale, E.A., Watson, C.A., 2007. Controls on soil nitrogen cycling and mic robial community composition across land use and incubation temperature. Soil Biology & Biochem istry 39, 744756. Corstanje, R., Reddy, K.R., Prenger, J.P., Newman, S., Ogram, A.V. 2007 Soil microbial eco physiological response to nutrient enrichment in a sub tropical wetland. Ecol ogical Indicat ors 7, 277289. DeLuca, T.T., Skogley, E.O., Engel, R.E., 1989. Bandapplied elemental sulfur to enhance the phytoavailability of phosphorus in alkaline calcareous soils. Biology & Fertility of Soils 7, 346 -350. D eubel, A., Braune, H., Tanne berg, H., Merbach, W., 2007. Conversion and acidifying effect of elemental sulphur in an alkaline loess soil. Archives of Agronomy and S oil Science 53, 161-171. Dias, L.E., Ribeiro, E.S., Alvarez, V.H., Mello, J., 2003. Organic and mineral sulfur fractions in Brazilian soils submitted to consecutive harvests. Commun ications in Soil Science and Plant Anal ysis 34, 357373. Eriksen, J., Murphy, M.D., Schnug, E., 1998. The soil sulphur cycle. In: Schnug, E., (Ed.), Kluwer Academic Pu blishers, The Netherlands, pp. 3974.

PAGE 170

170 Finlay, B.J., Maberly, S.C., Copper, J.I., 1997. Microbial diversity and ecosystem function. Oikos 80, 209213. Fontaine, S., Mariotti, A., Abbadie, L., 2003. The priming effect of organic matter: a question of microb ial competition? Soil Biology & Biochemistry 35, 837 -843. Gabriel M ., Redfield, G., Rumbold D., 2008 Sulfur as a regional water quality concern in south Florida. South Florida Environmental Report, Appendix 3B -2. South Florida Water Management District, West Palm Beach, FL. Garcia -Oliva, F., Lancho, J.F.G., Montano, N.M. 2006. Soil carbon and nitrogen dynamics followed by a forest to pasture conversion in western Mexico. Agroforest ry Syst ems 66, 93 -100. Garland, J.L. 1996 Analytical approaches to th e characterization of samples of microbial communities using patterns of potential C source utilization. Soil Biol ogy & Bioche mistry 28, 213-221. Garland, J.L. 1997 Analysis and interpretation of communi ty level physiological profiles in microbial ecolog y. FEMS Microb iology Ecol ogy 24 289 -300. Gesch, R.W., Reicosky, D.C., Gilbert, R.A., Morris, D.R., 2007 Influence of tillage and plant residue management on respiration of a Florida Everglades H istosol Soil & Tillage Res earch 92 156 -166. Gessa, C E M immo T Deiana, S Marzadori C ., 2005 Effect of aluminum and pH on the mobility of phosphate through a soil -root interface model. Plant and Soil 272, 301-311. Gharmakher, H.N., Machet, J.M., Beaudoin, N., Recous, S., 2009. Estimation of sulfur mineralization and relationships with nitrogen and carbon in soils. Biology & Fertility of Soils 45, 297 304. Gilbert, R.A., Rice, R.W., 2006. Nutrient requirements for sugarcane production on Florida muck soils. UF -IFAS SS-AGR -226, Gainesville, FL. Gilmour, C.C. Orem, W., Krabbenhoft, D., Mendelssohn, I.A., 2007. Preliminary assessment of sulfur sources, trends and effects in the Everglades. In: 2007 South Florida Environmental Report, Appendix 3B 3, South Florida Water Management District, West Palm Beach, FL. Graham, S.A., Craft, C.B., McCormick, P.V., Aldous, A., 2005 Forms and accumulation of soil P in natural and recently restored peatlands upper Klamath Lake, Oregon, USA. Wetlands 25, 594 -606.

PAGE 171

171 Grayston, S.J., Campbell, C.D., Bardgett, R.D., Mawdsley, J.L., Clegg, C.D., Ritz, K., Griffiths, B.S. 2004 Assessing shifts in microbial community structure across a range of grasslands of different management intensity using CLPP, PLFA and community DNA techniques. Appl ied Soil Ecology 25, 63 -84. Grigg, B.C., Zinder, G.H., Millar, J.D., 2002 Seasonally maintained shallow water tables improve sustainability of histosols planted to sugarcane. Journal of the American Soc iety of Sugarcane Technologists 22, 62 -72. Halajnia, A. Haghnia G.H., Fotovat A. Khorasani R ., 2009 Phosphorus fractions in calcareous soils amended with P fertilizer and catt le manure. Geoderma 150, 209213. Harrell, D.L., Wang J J ., 2007 Evaluation of threeand five-step inorganic phosphorus chemical fractionation procedures along with inductively coupled plasma determination for calca reous soils. Soil Science 172, 55-67. Hassan, N., Olson, R A., 1966 Influence of applied sulfur on availability of soil nutrients for corn ( Zea mays L. ) nutrition. Soil Science Society of America Journal 30,284286. Heitholt, J.J., Sloan, J.J., MacKown, C.T., 2002. Copper, manganese, and zinc fertilization effects on growth of soybean on a calcareous soil. Journal of Plant Nutrition 25, 1727 1740. Herencia, J.F., Ruiz, J.C., Morillo, E., Melero, S., Villaver de, J., Maqueda, C., 2008. The Effect of organic and mineral fertilization on micronutrient availability in soil. Soil Science 173, 69 80. Hochmuth, G. Hanlon, E.A. Snyder, G.H. Nagata, R.T ., Schueneman, T. 1996. UF IFAS BUL 313, Gainesville, FL. Iovi eno, P., Morra, L., Leone, A., Pagano, L., Alfani, A., 2009. Effect of organic and mineral fertilizers on soil respiration and enzyme activities of two Mediterranean horticultural soils. Biology & Fertility of Soils 45, 555561. Izuno, F.T., Sanchez, C.A., Coale, F.J., Bottcher, A.B., Jones D.B. 1991. Phosphorus concentrations in drainage water in the Everglades Agriculture Area. Journal of Environmental Quality 20, 608 -619. Jaggi, R C Aulakh, M.S., Sharma, A.R., 2005 Impacts of elemental S applied un der various temperature and moisture regimes on pH and available P in acidic, neutral and alkaline soils. Biology & Fertility of Soils 41, 5258. Kaiser, K., Guggenberger, G., 2005. Dissolved organic sulphur in soil water under Pinus sylvestris L. and Fagus sylvatica L. stands in northeastern Bavaria, Germany variations with seasons and soil depth. Biogeochemistry 72, 337-364.

PAGE 172

172 Knauff, U., Schulz, M., Scherer, H.W., 2003. Arylsufatase activity in the rhizosphere and roots of different crop species. European Journal of Agronomy 19, 215-223. Korndrfer, G.H., Anderson, D.L., Portier, K.M., Hanlon, E.A., 1995. Phosphorus soil test correlation to sugarcane grown on Histosols in the Everglades. Soil Science Society of America Journal 59,16551661. Kuo S., 1996 Phosphorus. In: B ridgham J M ( e d) Methods of soil analysi s. Part 3. SSSA, Madison, WI, p p 869919. Larson, J.L., Donald, R.Z., Sinsabaugh, R.L., 2002. Extracellular enzyme activity beneath temperate trees growing under elevated carbon dioxide and ozon e. Soil Science Society of America Journal 66, 18481856. Lattin, J., Carroll, J.D., Green P.E., 2003. Analyzing multivariate data. Thomson Learning C A. Li, B.Y., Zhou, D.M., Cang, L., Zhang, H.L., Fan, X.H., Qin S.W., 2007. Soil micronutrient availabili ty to crops as affected by long -term inorganic and organic fertilizer applications. Soil and Tillage Research 96,166 -173. Lindemann, W.C., Aburto, J.J., Haffner, W.M., Bono, A.A., 1991. Effect of sulfur source on sulfur oxidation. Soil Science Society of A merica Journal 55, 8590. Little, R.C., Milliken, G.A., Stroup, W.W., Wolfinger, R.D., Schabenberber, O., 2006. SAS for Mixed Models, Second Edition. SAS Institute Inc., Cary, NC. Lohila, A., Aurela, M., Regina, K., Laurila, T. 2003 Soil and total ecosys tem respiration in agricultural fields: effects of soi l and crop type. Plant and Soil 251, 303 -317. Maguire, R O Sims J T C oale, F J ., 2000 Phosphorus fractionation in biosolids amended soils: relationship to soluble and desorbable phosphorus. Soil Sci ence Soc iety of America J ournal 6 4, 20182024. Majchrzak, R.N., Olson, K.R., Bollero, G., Nafziger, E.D., 2001. Using soil properties to predict wheat yields on Illinois soils. Soil Science 166, 267-280. Matsushita, M., Ito, S., Meguro, S., Kawachi, S., 2007. Structure of soil microbial communities in sugi plantations and seminatural broad leaved forests with different land-use history. Canadian Journal of Forest Research 37, 236246. McGill, W.B., Cole, C.V., 1981. Comparative aspects of cycling of organic C, N, S and P through soil organic matter. Geoderma 26, 267286. Monkiedje, A., Spiteller, M., Fotio, D., Sukul, P. 2006 The effect of land use on soil health indicator in Peri -Urban agriculture in the humid forest zone of southern Cameroon. J ournal of Environ mental Qual ity 35, 2402-2409.

PAGE 173

1 73 Moral, R., Moreno-Caselles, J., Perez Murcia, M., Perez -Espinosa, A., 2002. Improving the micronutrient availability in calcareous soils by sewage sludge amendment. Communications in Soil Science and Plant Analysis 3 3, 3015-3022. Moreno -Caselles, J., Moral, R., Perez Murcia, M.D., Perez Espinosa, A., Paredes, C., Agull, E., 2005. Fe, Cu, Mn, and Zn input and availability in calcareous soils amended with the solid phase of pig slurry. Communications in Soil Science and Plant Analysis 36 525-534. Morgan, K.T., McCray, J.M. Rice, R.W., Gilbert, R.A., Baucum L.E. 2009. Review of c urrent s ugarcane f ertilizer r ecommendations: A r eport from the UF/IFAS s ugarcane f ertilizer s tandards t ask f orce. UF EDIS SL 295, Gainesville, FL. Morris, D.R., Gilbert, R.A., 2005 Inventory, crop use and soil subsidence of histosols in Florida. Journal of Food, Agriculture & Environment 3, 190 -193. Morris, D.R., Gilbert, R.A., Reicosky, D.C., Gesch, R.W., 2004. Oxidation potentials of soil organic matter in histosols under different tillage methods. Soil Science Society of America Journal 68, 817-826. Mozaffari, M., Sims J T ., 1996 Phosphorus transformations in poultry litter amended soils of the Atlantic Coastal Plain. J ournal of Environ mental Quality 25,1375 1365. Mylavarapu, R.S. 2009. UF/IFAS extension soil testing laboratory (ESTL) analytical procedures and training manual. UF Circular 1248, Gainesville, FL. Noe, G.B., Childers, D.L., Jones R D ., 2001 Phosphorus biogeochemistry and the impact of phosphorus enrichment: why is the Everglades so unique? Ecosystems 4, 603-624 Nogueira, M.A., Albino, U.B., Brandon-Junior, O., Braun, G., Cruz, M.F., Dias, B.A., Duarte, R.T.D., 2006. Promising indicators for assessment of agroecosystems al teration among natural, reforested and agricultural land use in southern Brazil. Agriculture, Ecosystems and Environment 115, 237-247. Orem, W.H., 2007. Sulfur contamination in the Florida Everglades: initial examination of mitigation strategies. U.S. Geol ogical Survey Open File Report 2007 1374. Pansu, M., Gautheyrou, J 2006. Titration of elementary sulphur. In : Handbook of soil analysis. Pansu, M., Gautheyrou, J. (ed). Springer, NY. pp. 867-869 Parsons, K.J., Zheljazkov V.D. MacLeod, J., Caldwell, C. D., 2007. Soil and tissue phosphorus, potassium, calcium, and sulfur as affected by dairy manure application in a no-till corn, wheat, and soybean rotation. Agronomy Journal 99, 13061316.

PAGE 174

174 Prenger, J.P., Reddy, K.R., 2004 Extracellular enzyme activity lev els in a freshwater marsh after cessation of nutrient loading. Soil Science Society of America Journal 68, 1796-1804. Preston Mafham, J., Boddy, L., Randerson, P.F. 2002 Analysis of microbial community functional diversity using sole-carbon -source utiliz ation profiles a critique. FEMS M icrobiology Eco logy 42, 114. Provin, T.L., Wright, A.L., Hons, F.M., Zuberer, D.A., White, R.H., 2008. Seasonal dynamics of soil micronutrients in compost amended bermudagrass turf. Bioresource Technology 99, 26722679. Reddy K. R Wang Y ., DeBusk, W. F Fisher M M Newman, S., 1998 Forms of soil phosphorus in selected hydrologic units of t he Florida Everglades. Soil Science Society of America Journal 62, 11341147. Reicosky, D.C. Lindstrom, M.J. 1993 Fall tillage method: effect on short -term carbon dioxide f lux from soil. Agronomy Journal 85, 12371243. Rice, R. W Gilbert R.A. Lentini R S., 2006 Nutrient requirements for Florida sugarcane. UF -IFAS SS -AGR -228, Gainesville FL. Rogovska, N P Blackmer A. M Mallarino A P., 2007 Relationships between soybean yield, soil pH, and soil carbonate concentration. Soil Science Society of America Journal 71, 12511256. Rutigliano, F.A., DAscoli, R., Santo A. V.D., 2004. Soil microbial metabolism and nutrient status in a Mediterranean area as affected by plant cover. Soil Biology & Biochemistry 36, 1719-1729. Ryan, J Curtin D Cheema, M.A., 1984 Significance of iron oxides and calcium carbonate particle size in phosphate sorption by calcareous soils. Soil Scienc e Society of America Journal 48, 7476. Sanchez C.A., 1990 Soil -testing and fertilization recommendations for crop production on organic soils in Florida. Univ ersity of Fl orida Agric ulture Exp erimental Sta tion Bul letin 876. Univ ersity of Florida, Gainesv ille FL. Sanchez Moreno, S., Smukler, S., Ferris, H., OGeen, A.T., Jackson, L.E., 2008. Nematode diversity, food web condition, and chemical and physical properties in different soil habitats of an organic farm. Biology & Fertility of Soils 44, 727-744. Santoso, D Lefroy R D B. Blair G J ., 1995 A comparison of sulfur extractions for weathered acid soi ls. Australian Journal of Soil Research 33, 125-133. Schimel, J.P. Bennett, J. 2004. Nitrogen mineralization: challenges of a changing paradigm. Ecology 85, 591 -602.

PAGE 175

175 Schueneman, T.J., 2001. Characterization of sulfur sources in the EAA. Soil and Crop Science Society of Florida Proceedings 60, 49 -52. Sena, M.M., Frighetto, R.T.S., Valarini, P.J., Tokeshi, H., Poppi, R.J., 2002. Discrimination of manage ment effects on soil parameters by using principal component analysis: a multivariate analysis case study. Soil & Tillage Research 67, 171181. Sena, M.M., Poppi, R.J., da Costa, L.M., Martinez, M.A., 2000. Avaliaco do uso de mtodos quimiomtricos em an lise de solos (Evaluation of the use of chemometric methods in soil analysis) (in Portuguese, with English abstract). Qumica Nova 23, 547-556. Shih, S.F., Glaz, B., Barnes Jr., R.E., 1998 Subsidence of organic soils in the Everglades Agricultural Area during the past 19 years. Soil and Crop Sci ence Soc iety of Fl orida Proc eedings 57 20 29. Shih, S.F., Rahi, G.S., Ozaki, H.Y., Smajstrla, A.G., 1982 Effect of water table and crops on soil temperature. Soil and Crop Science Society of Florida Proceedings 41, 47-54. Shuman, L.M., 1988. Effect of phosphorus level on extractable micronutrients and their distribution among soil fractions. Soil Science Society of America Journal 52, 1 36 141. Sicardi, M., Garcia -Prechac, F., Frioni, L., 2004 Soil microbial indicators sensitive to land use conversion from pasture to commercial Eucalyptus grandis (Hill ex Maiden) plantations in Uruguay Applied Soil Ecology 27, 125133. Sinsabaugh, R.L., Antibus, R.K., Linkins, A.E., McClaugherty, C.A., Rayburn, L., Repert, D., Weiland, T., 1993. Wood decomposition: nitrogen and phosphorus dynamics in relation to extracellular enzyme activity. Ecology 74, 1586 -1593. Slaton, N.A., Golden, B.R., Norman, R.J., Wilson, C.E., DeLong, R.E., 2009. Correlation and calibration of soil pot assium availability with rice yield and nutrimental status. Soil Science Society of America Journal 73, 1192 1201. Snyder, G.H., 2005 Everglades Agricultural Area soil subsidence and land use projections. Soil and Crop Science Society of Florida Proceedings 64 44 -51. Snyder, G.H., Burdine, H.W., Crockett, J.R., Gascho, G.J., Harrison, D.S., Kidder, G., Mishoe, J.W. 1978. Water table management for organic soil conservation and crop production in the Florida Everglades. Florida Agricultural Experiment Stations Bulletin, 801, Univ. of Florida, Gainesville, FL. Solomon, D., Lehmann, J., Tekalign, M., Fritzsche, F., Zech, W., 2001. Sulfur fractions in particle -size separates of the subhumid Ethiopian highlands as influenced by land use changes. Geoderma 102, 41 -59.

PAGE 176

176 Strahm, B.D., Harrison, R.B., 2007. Mineral and organic matter controls on the sorption of macronutrient anions in variable -charge soils. Soil Science Society of America Journal 71, 19261933. Tejada, M., Hernandez, M.T., Garcia, C., 2006. Applica tion of two organic amendments on soil restoration: effects on the soil biological properties. Journal of Environmental Quality 35, 10101017. Turner B. L ., Cade Menum, B. J ., Condron, L .M., Newman, S., 2005 Extraction of soil organic phosphorus. Talanta 6 6, 294306. Valeur, I., Andersson, S., Nilsson, S.I., 2000. Calcium content of liming material and its effect on sulphur release in a coniferous forest soil. Biogeochemistry 50, 120. Wallenstein, M.D., McMahon, S.K., Schimel, J.P., 2009. Seasonal variation in enzyme activities and temperature sensitivities in Arctic tundra soils. Global Change Biology 15, 16311639. Wander, M.M., Bollero, G.A., 1999. Soil quality assessment of tillage impacts in Illinois. Soil Science Society of America Journal 63, 961-971. Wang, J., Solomon, D., Lehmann, J., Zhang, X., Amelung, W., 2006. Soil organic sulfur forms and dynamics in the Greta Plains of north America as influenced by longterm cultivation and climate. Geoderma 133, 160-172. Wang, J.J, Harrell, D.L., Henderson, R.E., Bell, P.F., 2004. Comparison of soil test extractants for phosphorus, potassium, calcium, magnesium, sodium, zinc, copper, manganese, and iron in Louisiana soils. Communications in Soil Science and Plant Analysis 35, 145-160. Wei, X., Hao, M., Shao, M., Gale W.J., 2006. Changes in soil properties and the availability of soil micronutrients after 18 years of cropping and fertilization. Soil and Tillage Research 91, 120130. White, J.R., Reddy, K.R. 1999 Influence of nitrate and phosphorus loading on denitrifying enzyme activity in Everglades wetland soils. Soil Science Society of America Journal 63, 1945-1954. White, J.R., Reddy, K.R. 2000 Influence of phosphorus loading on organic nitrogen mineralization of Everglades soils. Soil Science Society of America Journal 64, 15251534. White, J.R., Reddy, K.R. 2001 Influence of selected inorganic electron acceptors on organic nitrogen mineralization in everglades soils. Soil Science Society of America Journal 65, 941948.

PAGE 177

177 Wright, A.L. 2009. Soil phosphorus stocks and distribution in chemical fractions for long -term sugarcane, pasture, turfgrass, and forest systems in Florida. Nutrient Cycling in Agroecosystem s 83:223231. Wright, A.L., Dou, F., Hons, F.M. 2007. Crop species and tillag e effects on carbon sequestration in subsurface soil. Soil Science 172, 124-131. Wright, A.L., Provin, T.L., Hons, F.M., Zuberer, D.A., White, R.H., 2007. Soil micronutrient availability after compost addition to St. Augustinegrass. Compost Science & Utili zation 15, 127134. Wright, A.L. Reddy, K.R. 2001a. Phosphorus loading effects on extracellular enzyme activity in Everglades wetland soil. Soil Science Society of America Journal 65, 588-595 Wright, A.L. Reddy, K.R. 2001 b Heterotrophic microbial act ivity in northern Everglades wetland soils. Soil Science Society of America Journal 65, 1856 -1864. Wright, A.L., Reddy, K.R., 2008. Catabolic diversity of periphyton and detritus microbial communities in a subtropical wetland. Biogeochemistry 89, 199207. Wright A.L., Reddy, K.R., Corstanje, R., 2009 Patterns of heterotrophic microbial activity in eutrophic and oligotrophic peatlands. European Journal of Soil Biology 45, 131137. Wright, A.L., Reddy, K.R., Newman, S., 2009. Microbial indicators of eutrophication in Everglades wetlands. Soil Science Society of America Journal 73, 15971603. Wright, A. L Wang, Y. Reddy, K.R., 2008. L oss on-ignition method to assess soil organic carbon in calcareous Everglades wetlands. Communications in Soil Science and Pl ant Anal ysis 39, 3074 3083. Wu, T., Schoenau, J.J., Li, F., Qian, P., Malhi, S.S., Shi, Y., Xu, F., 2004. Influence of cultivation and fertilization on total organic carbon and carbon fractions in soils from the Loess Plateau of China. Soil and Tillage Res earch 77, 5968. Yang, Z., Haneklaus, S., Singh, B.R., Schnug, E., 2008. Effect of repeated applications of elemental sulfur on microbial population, sulfate concentration, and pH in soils. Communications in Soil Science and Plant Anal ysis 39, 124140. Ya o, H., He, Z., Wilson, M.J., Campbell, C.D., 2000 Microbial biomass and community structure in a sequence of soil with increasing fertility and changing land use. Microb ial Ecol ogy 40, 223-237. Ye, R., Wright, A.L., Inglett, K. Wang, Y. Ogram, A.V. Reddy K.R., 2009. Land use effects on soil nutrient cycling and microbial community dynamics in the Everglades Agricultural Area, Florida. Communications in Soil Science and Plant Anal ysis 40, 2725 2742.

PAGE 178

178 Zhang, C., Huang, L., Luan, T., Jin, J., Lan, C., 2006 Structure and function of microbial communities during the early stages of revegetation of barren soils in the vicinity of a Pb/Zn Smelter. Geoderma 136, 555-565. Zhang, T.Q., Mackenzie A.F ., 1997 Changes of soil phosphorous fractions under long term c orn monoculture. Soil Science Society of America Journal 61, 485-493. Zhou, M, Li Y., 2001 Phosphorus -sorption characteristics of calcareous soils and limestone from the southern Everglades and adjacent farmlands. Soil Science Society of America Journal 65, 14041412. Zhou, W., He, P., Li, S., Lin, B., 2005. Mineralization of organic sulfur in paddy soils under flooded conditions and its availability to plants. Geoderma 125, 8593.

PAGE 179

179 BIOGRAPHICAL SKETCH Rongzhong Ye was born in Huian County, Fujian Provi nce, P.R.China. In 1996, he graduated from the 1st Middle School of Huian County, Fujian Province, P.R.China. He then attended the Longyan Teachers College, Fujian Province with the intention of becoming a high school teacher. In 1998, he went to the F ujian Normal University for his bachelors degree in b iological e ducation. After completed his undergraduate study, he continued to pursue his first masters degree in 2000. In 2003, he worked for the 3rd Institute of Oceanography, the State Oceanic Admi nistration of China with a Master of Science degree in zoology (developmental biology). After received an offer from Nicholls State University, USA, he resigned the job and pursued his 2nd Master of Science degree in marine and environmental biology at th e Department of Biological Sciences of NSU. In 2006, he was enrolled in the graduate program of the Soil and Water Science Department at the University of Florida. He graduate d in May 20 1 0 with a Ph.D. in soil and water science and a minor in statistics.