<%BANNER%>

Organic Soil Subsidence and Nutrient Leaching in an Impacted Florida Wetland

Permanent Link: http://ufdc.ufl.edu/UFE0044282/00001

Material Information

Title: Organic Soil Subsidence and Nutrient Leaching in an Impacted Florida Wetland
Physical Description: 1 online resource (86 p.)
Language: english
Creator: Van Dam, Bryce R
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2012

Subjects

Subjects / Keywords: ch4 -- co2 -- eutrophication -- hydrology -- oxidation -- subsidence -- wetland
Soil and Water Science -- Dissertations, Academic -- UF
Genre: Soil and Water Science thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Current water management practices in the Upper St. Johns River Basin have resulted in significant losses of wetland soils in the St. Johns Marsh Conservation Area to subsidence (SJMCA). Because organic soils accrete at a very slow rate and oxidize rapidly under drained conditions, this soil subsidence is of major concern to water managers. With this permanent loss of soil material to oxidation comes the mineralization of other nutrients that contribute to downstream water quality degradation. Managers have relied on historic soil subsidence rates established as early as 1956 in the Everglades Agriculture Area which have been used to model soil loss in the SJMCA. While these well documented subsidence rates are certainly valid for the environments in which they were determined, their applicability to other systems is questionable. Therefore, we seek to model the effect of hydrologic status on soil subsidence for soils in the SJMCA. Field and laboratory studies were conducted in which CO2 and CH4 fluxes were monitored over changing hydrologic conditions. A separate lab study determined a subsidence rate that can be modeled along with gaseous carbon losses.
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 Bryce R Van Dam.
Thesis: Thesis (M.S.)--University of Florida, 2012.
Local: Adviser: Osborne, Todd Z.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2012-11-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2012
System ID: UFE0044282:00001

Permanent Link: http://ufdc.ufl.edu/UFE0044282/00001

Material Information

Title: Organic Soil Subsidence and Nutrient Leaching in an Impacted Florida Wetland
Physical Description: 1 online resource (86 p.)
Language: english
Creator: Van Dam, Bryce R
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2012

Subjects

Subjects / Keywords: ch4 -- co2 -- eutrophication -- hydrology -- oxidation -- subsidence -- wetland
Soil and Water Science -- Dissertations, Academic -- UF
Genre: Soil and Water Science thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Current water management practices in the Upper St. Johns River Basin have resulted in significant losses of wetland soils in the St. Johns Marsh Conservation Area to subsidence (SJMCA). Because organic soils accrete at a very slow rate and oxidize rapidly under drained conditions, this soil subsidence is of major concern to water managers. With this permanent loss of soil material to oxidation comes the mineralization of other nutrients that contribute to downstream water quality degradation. Managers have relied on historic soil subsidence rates established as early as 1956 in the Everglades Agriculture Area which have been used to model soil loss in the SJMCA. While these well documented subsidence rates are certainly valid for the environments in which they were determined, their applicability to other systems is questionable. Therefore, we seek to model the effect of hydrologic status on soil subsidence for soils in the SJMCA. Field and laboratory studies were conducted in which CO2 and CH4 fluxes were monitored over changing hydrologic conditions. A separate lab study determined a subsidence rate that can be modeled along with gaseous carbon losses.
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 Bryce R Van Dam.
Thesis: Thesis (M.S.)--University of Florida, 2012.
Local: Adviser: Osborne, Todd Z.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2012-11-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2012
System ID: UFE0044282:00001


This item has the following downloads:


Full Text

PAGE 1

1 ORGANIC SOIL SUBSIDE NCE AND NUTRIENT LEA CHING IN AN IMPACTED FLORIDA WETLAND By BRYCE VAN DAM A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2012

PAGE 2

2 2012 Bryce Van Dam

PAGE 3

3 ACKNOWLEDGMENTS through every step in this process My committee chair, Dr. Todd Z. Osborne lended assistance with every aspect of this work, from experimental design to critical reviews. My advisory committee: Dr. Kanika S. Inglett, Dr. Angelique K. Bochnak, and Dr. Mark W. Clark also provided a great deal of advice and guidance with my research. I also thank Ken Snyder of the SJRWMD for his prowess at the helm of the airboat.

PAGE 4

4 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 3 LIST OF TABLES ................................ ................................ ................................ ............ 6 LIST OF FIGURES ................................ ................................ ................................ .......... 7 LIST OF ABBREVIATIONS ................................ ................................ ............................. 9 ABSTRACT ................................ ................................ ................................ ................... 10 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 11 Environmental/Social Significance ................................ ................................ .......... 17 St Johns Marsh Conservation Area: ................................ ................................ ....... 19 Objectives ................................ ................................ ................................ ............... 21 2 SOIL SUBSIDENCE IN THE SJMCA ................................ ................................ ...... 22 Ba ckground Statement ................................ ................................ ........................... 22 Field Methods ................................ ................................ ................................ ... 23 Lab Methods ................................ ................................ ................................ ..... 27 Subsidence Study ................................ ................................ .......................... 27 Water Table Study ................................ ................................ ............................ 28 Drain Study ................................ ................................ ................................ ....... 29 Results ................................ ................................ ................................ .................... 30 Insitu Study ................................ ................................ ................................ ....... 30 Laboratory ................................ ................................ ................................ ........ 33 Subsidence ................................ ................................ ................................ 33 Water table study ................................ ................................ ....................... 35 Drain study ................................ ................................ ................................ 37 Discussion ................................ ................................ ................................ .............. 39 Insitu Study ................................ ................................ ................................ ....... 39 Laboratory ................................ ................................ ................................ ........ 42 Subsidence study ................................ ................................ ....................... 42 Water table stud y ................................ ................................ ....................... 43 Drain study ................................ ................................ ................................ 44 Concluding Remarks ................................ ................................ ............................... 45 3 NUTRIENT EXCHANGE IN THE SJMCA ................................ ............................... 47 Opening Statement ................................ ................................ ................................ 47 Methods ................................ ................................ ................................ .................. 50

PAGE 5

5 Site Description ................................ ................................ ................................ 50 Soil Chemistry ................................ ................................ ................................ .. 55 Laboratory Nutrien t Flux ................................ ................................ ................... 55 Field Nutrient Flux ................................ ................................ ............................ 58 Results ................................ ................................ ................................ .................... 59 Soil Chemistry ................................ ................................ ................................ .. 59 Lab Nutrient Flux ................................ ................................ .............................. 60 Field Nutrient Flux ................................ ................................ ............................ 62 Discussion ................................ ................................ ................................ .............. 66 Soil Chemistry ................................ ................................ ................................ .. 66 Lab Nutrient Flux ................................ ................................ .............................. 68 Field Nutrient Flux ................................ ................................ ............................ 69 Concluding Remarks ................................ ................................ ............................... 70 4 SUMMARY AND CONCLUSIONS ................................ ................................ .......... 7 2 APPENDIX ADDITIONAL FIGURE S ................................ ................................ ......................... 77 LIST OF REFERENCES ................................ ................................ ............................... 81 BIOGRAPHICAL S KETCH ................................ ................................ ............................ 86

PAGE 6

6 LIST OF TABLES Table page 1 1 Selected literature subsidence rates ................................ ................................ ... 13 2 1 Li 8100 parameters used for field and lab ................................ .......................... 26 2 2 Estimates of subsidence fractions (percent of total in parenthesis). n=37 .......... 34 2 3 Least squares models for CO 2 (g C m 2 yr 1 ) and CH 4 (g C m 2 yr 1 ) Fluxes. Adj R 2 of 0.77 and 0.13 respectively. ................................ ................................ .. 37 3 1 Soil chemica l characteristics (Average Std Error) ................................ ............ 59 3 2 Nutrient fluxes (g m 2 ) given as mean SE (n=37). Asterisked p values are ................................ ................................ ............. 60

PAGE 7

7 LIST OF FIGURES Figure page 1 1 Location of study site ................................ ................................ .......................... 20 2 1 Map of field locations.. ................................ ................................ ........................ 24 2 2 Diagram and picture of study site layout. ................................ ............................ 25 2 3 Picture of water table experiment. Cores were randomly assigned to each of 6 water table treatments. Photo courtesy of Bryce Van Dam. ............................ 29 2 4 CO 2 and CH 4 fluxes with water depth. ................................ ................................ 31 2 5 Log CO 2 :CH 4 flux and Water Level (cm). n=22, R 2 Adj=0.72 ............................. 32 2 6 Total respiration (g C m 2 yr 1 ) and water level (cm) ................................ ........... 33 2 7 Average soil elevation (cm) and days drained (n=37) ................................ ........ 34 2 8 Laboratory CO 2 and CH 4 fluxes with water depth. ................................ .............. 36 2 9 Average Q 10 for CO 2 flux with varying water depths (cm). Smoothing Spline Fit, lambda=100, R 2 = 0.42 ................................ ................................ .................. 37 2 10 CO 2 flux after draining, and with air tempe rature ................................ ................ 38 2 11 Soil moisture (%) after draining. P value = 0.7128, n=3 ................................ ..... 39 3 1 Location of study site. ................................ ................................ ......................... 52 3 2 Map of field sites. ................................ ................................ ................................ 53 3 3 Diagram and picture of field site layout. ................................ .............................. 54 3 4 Soil cores in laboratory. Container is insulated fiberglass; a plywood lid was placed over the bath when not being sampled to prevent light from entering. Soil cores were placed randomly into each of 13 5 g al buckets. Cores were flooded by pouring DI water into buckets outside of the cores. .......................... 57 3 5 C:N and depth by field site. ................................ ................................ ................. 60 3 6 Nutrient flux % change by vegetative community (n=9). ................................ ..... 61 3 7 Average TP (mg kg 1 ) and TDP% change. Reciprocal fit. R 2 Adj = 0.998, n=4 ... 62 3 8 Field nutrient fluxes by site. ................................ ................................ ................ 63

PAGE 8

8 3 9 DOC and TP fluxes regressed against TKN. ................................ ...................... 64 3 10 Field nutrient fluxes compared with laboratory nutrient fluxes ............................ 65 4 1 Sketch of current and proposed water table regime for SJMCA ......................... 75 A 1 Field DO (mg L 1 ), DOC (mg L 1 ), NH 4 (mg L 1 ), TDKN (mg L 1 ), TP (mg L 1 ), and pH. n=3 ................................ ................................ ................................ ........ 78 A 2 Depth profiles for TC (%), TN (%), TP (ug kg 1 ), and TS (ug kg 1 ) by site. n=1 ... 79 A 3 Major nutrient fluxes (mg/L, n=3) and pH (n=1) for all field sites ........................ 80

PAGE 9

9 LIST OF ABBREVIATION S C Carbon CO 2 Carbon Dioxide DOC Dissolved Organic Carbon FID Flame Ionization Detector GC Gas Chromatograph ICP Inductuvely Coupled Plasma N Nitrogen OM Organic Matter P Phosphorus PVC Polyvinyl Chloride SJMCA St Johns Marsh Conservation Area TC Total Carbon TDKN To tal Dissolved Kjeldahl Nitrogen TDP Total Dissolved Phosphorus TEA Terminal Electron Acceptor TN Total Nitrogen TP Total Phosphorus TS Total Sulphur USJRB Upper St Johns River Basin

PAGE 10

10 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science ORGANIC SOIL SUBSIDE NCE AND NUTRIENT LEA CHING IN AN IMPACTED FLORIDA WETLAND By Bryce Van Dam May 2012 Chair: Todd Z Osborne Maj or: Soil and Water Science Current water management practices in the Upper St. Johns River Basin have resulted in significant losses of wetland soils in the St. Johns Marsh Conservation Area (SJMCA) to subsidence Because organic soils accrete at a very slow rate and oxidize rapidly under drained conditions, this soil subsidence is of major concern to water managers. With this permanent loss of soil material to oxidation comes the mineralization of other nutrien ts that contribute to downstream water quality degradation. Managers have relied on historic s oil subsidence rates established as early as 1956 in the Eve rglades Agriculture Area which have been used to model soil loss in the SJMCA. While these well docum ented subsidence rates are certainly valid for the environments in which they were determined, their applicability to other systems is questionable. Therefore, we seek to model the effect of hydrologic status on soil subsidence for soils in the SJMCA. Fiel d and laboratory studies were conducted in which CO 2 and CH 4 fluxes were monitored over changing hydrologic conditions. A separate lab study determined a subsidence rate that can be modeled along with gaseous carbon losses.

PAGE 11

11 CHAPTER 1 INTRODUCTION Wetlands, being ecosystems of significant biological productivity, are major contributors to the global carbon cycle. In response to changes in hydrologic, biologic, climatologic, or chemical conditions, wetlands can serve as either a source or a sink for carbon (Schedlbauer 2010). It is well established that when wetlands are flooded, the ability of microorganisms to use oxygen in their metabolic processes is greatly reduced (Schedlbauer et al. 2010; Freeman et al. 1996; Lafleur et al 2005). While some mi crobes can use alternate oxidants like nitrate, Fe 3+ or even sulfate, the rate of decomposition using these oxidants is much slower than for these microbes that use O 2 Thus, when a wetland is flooded, soil organic matter (SOM) can build up in the soil (due to slower oxidation rates) causing an overall increase in soil volume (accretion). Conversely, when floodwater is removed, organic matter can be rapidly oxidized and released to the atmosphere as CO 2. This loss of carbon to the atmosphere is one of t he major causes of soil subsidence in wetlands, which can be defined as the lowering of surface elevation after drainage due to causes other than erosion ( Parent et al. 1977) Total sub sidence rates in excess of 2 cm 1 yr have been shown to occur in drained Everglades soils (Reddy and Delaune 2008). In a previous study in the Everglades, Stephens (1956) found a slightly highe r rate of approximately 1.25 in yr 1 (3.175 cm yr 1 ). Subsidence rates from 3.4 (Sch ipper et al. 20 02) to 0.6 1 cm yr 1 (Schothorst 1977) have been shown to exist as well in New Zealand pasture and a Netherland moor respectively. According to Vepraskas and Ewing (2006), total subsidence is a function of two processes, termed primary and secondary subsi dence. Primary subsidence results from

PAGE 12

12 the loss of the buoyant force provided by water; once pore water leaves the soil system, the support it provides to the overlying particles is lost. When air fills the pore spaces once occupied by water, the soil may compact under its own weight. The process of primary subsidence was estimated by Vepraskas and Ewing (2006) to account for 4 cm yr 1 or 69% of total subsidence in a North Carolina agricultural field. Secondary subsidence is caused by oxidation of the organ ic carbon to inorganic carbon that can be lost to the atmosphere as CO 2 This process can occur through biotic means like micro bial decomposition or through a biotic processes like fire. Vepraskas and Ewing (2006) estimate the soil lost to secondary subside nce to account for the remaining 1.9 cm yr 1 or 31% of total subsidence for the same North Carolina histosol mentioned earlier. It appears that complete consensus has not been reached on the magnitude of post drainage soil loss that can be expected. Much v ariability exists between subsidence rate estimates based on experimental methods, geography, pedology, and many other environmental conditions. Therefore, any models attempting to incorporate soil subsidence for environmental management purposes must take into account all of the previously mentioned factors, as small differences in subsidence rates could result in misleading results Once an organic soil is drained, both primary and secondary subsidence may act to lower the soil surface elevation Literature rates for subsidence are listed in Table 1 1. As time progresses after drainage, the relative role of each fraction of subsidence will change. According to Wsten et al. (1997) and Vepraskas and Ewing (2006) the process of soil subsidence can b e broken down into two main factors: compaction and shrinkage (primary subsidence), and oxidation (secondary subsidence). During the first

PAGE 13

13 5 10 years of water t able drawdown, compaction and shrinkage dominate relative to oxidation. In a study by Schothorst (1977) primary subsidence accounted for 1 4 cm of the 4 10 cm total subsidence over 6 years for peat in the Netherlands. One of the primary factors in physical subsidence is the destruction of macropore structure. Kechavarzi et al (2010) demonstrated tha t peat soils that have retained natural hydrology contain more macropores than peat that has been drained. As water in flooded soil exerts a buoyant force on soil particles, upon draining, this force is removed, allowing the soil particles to settle and co nsolidate (Drexler et al. 2009). This physical compaction of drained portions in peat not only results in a lowering of the surface, but also causes an increase in bulk density (Drexler et al. 2009) and subsequent decrease in saturated hydraulic conductivi ty. As both density and conductivity influence plant and microbial activity in the rhizosphere, and these factors also influence subsidence, many feedback systems may exist, making the prediction of subsidence rates very difficult. Table 1 1. Selected li terature subsidence rates Source Region Soil Type Subsidence rate (cm yr 1 ) Stephens 1956 Everglades Peat 3.18 Ewing and Vepraskas 2006 North Carolina Organic a gricultural 4 Schipper et al. 2002 New Zealand Peat b og 3.4 Rojstaczer and Deverel 1996 Northern California Alluvial sediment 2.3 1.5 Wosten et al. 1997 Malaysia Peat 2 Parent et al. 1977 Quebec Muck 0.99 4.53 Reddy et al. 2006 Florida (Blue Cypress Marsh) Peat 0.25 0.5 (only secondary subsidence) Drexler et al. 2009 Northern California Peat 3.5 8.2 Average literature subsidence 2.9 This Study Florida Peat 4.7

PAGE 14

14 Oxidation, or mineralization of organic carbon to CO 2 is the primary cause of secondary subsidence. Under aerobic conditions (as when the water t able is below the soil surface), this process can be accomplished by microorganisms using O 2 as an electron acceptor. The production of CO 2 during OM oxidation results in the permanent loss of carbon from the soil system, and contributes to overall soil subsid ence (Reddy et al. 2006; Berglund and Berglund 2011; Vepraskas and Ewing 2006; Morris et al. 2004). While flooded, carbon mineralization is primarily a function of the oxygen diffusion rate. Once drained, the rate limiting factor shifts from terminal elec tron acceptor availability to temperature (Kechavarzi et al. 2010). At temperatures over 20 o C, the rate of OM decomposition roughly doubles for every 10 o C increase in temperature (Q 10 ~2). As temperature drops below 20 o C, Q 10 increases exponentially, with a theoretical limit of Q 10 at 0 o C (Kirschbaum 1995) While the actual rate of decomposition decreases with decreasing temperature, the temperature sensitivity of decomposition i ncrease s with decreasing temperature. For example, if soil tempera tures increase from 10 o C to 20 o C, the rate of S OM oxidation will increase by a larger factor than if the temperature increases from 20 o C to 30 o C (Inglett et al 2011) In addition to loss of soil carbon, aerobic decomposition also can lead to a release of inorganic nutrients that might otherwise remain in the soil under flooded conditions (Reddy and DeLaune 2008; Osborne et al 2011) Wetlands play a large role in this nitrogen cycle; anaerobic conditions allow denitrifying microbes to compete fo r organic matter which facilitates the reduction of NO 3 to N 2 gas (process of denitrification) which can be lost to the atmosphere (Bowden 1987; Knowles 1982) This is the primary means of N loss from the soil system, so without anaerobic soils

PAGE 15

1 5 inorganic Reddy (1994) found the rates of denitrification in Lake Apopka Marsh (a hypereutrophic constructed wetland) to be between 20 and 50 mg N kg 1 day 1 According to Reddy and DeLaune (2 008) the rate of organic nitrogen (ON) mineralization when O 2 is present is roughly 2.6 times greater than when the wetland is anaerobic. A previous study showed that for Everglades surface soil (0 10 cm), ON mineralization when O 2 was present was 8.4 time s as high as when NO 3 was used as a terminal electron acceptor ( TEA ) (White and Reddy 1997). The inorganic nitrogen (NH 4 + ) released by the mineralization of organic nitrogen can be converted to NO 3 by various aerobic soil microbes (White and Reddy 1997) This nitrate released can be lost from the wetland and transferred to N limited ecosystems like most lakes, rivers, and estuaries, ultimately leading to eutrophication and fish kills. Fish kills and harmful algal blooms on the St Johns River over the sum mer season have become a common yet environmentally and economically destructive force (Chapman et al. 2001; Paerl 2007; Steidinger et al. 1998; Maher 1997) Phosphorus (P) is another element that plays a large role in shaping the chemical environment of wetlands. P is often a limiting nutrient in freshwater wetlands, meaning that increases in its concentration will lead to proportionate increases in biologic activity. As many Florida wetlands are already low P systems, any additional P may result in a change in trophic status (Reddy et al. 1999). This P induced eutrophication of previously oligotrophic systems is epitomized in the recent history of the Florida Everglades. Here, P loading has caused a significant change in microbial

PAGE 16

16 ecology (White and Re ddy 2001) resulting in altered sediment dynamics (Reddy and (Lorenzen et al. 2001). Unlike nitrogen, there is no significant gaseous form of P, so any P added to a wetland will remain in the system unless it is removed through some exogenous process. Its availability in wetla nds is gover ned mainly by adsorption desorption and redox reactions occurring on soil particles ( Twinch 1986; Cadenhead 1997; White and Reddy 2001 ). The degree to which P will adsorb or desorb in the soil is primarily a function of the difference between P concentratio ns in solution and in the soil, although other factors like Fe and Ca concentration also play a large role. If the total dissolved phosphorus ( TDP ) concentration is greater than the equilibrium P concentration (EPC=[P] at which net adsorption equals 0), then P will be taken up by the soil. On the other hand, when TDP is less than EPC, the soil will release P to the water column. Other mechanisms influencing the mobility of p hosphorus in soil include pH and Eh; both of these factors demonstrate a strong negative relationship with P availability (Reddy and DeLaune 2008). During periods of water t able drawdown, aerobic conditions can extend into wetland soils, causing microbiall y associated P to be rapidly mineralized and lost to the water column. A study by Olila et al. (1997) showed that a wetland drained for 6 weeks released roughly 10 times as much dissolved r eactive p hosphate (DRP also known as soluble reactive phosohorus ( SRP) ) as did a wetland that was drained for just 3 weeks. This study also showed that the availability of phosphorus within the soil intensified with increased drainage time. The phosphorus released during the drained period was highly labile and could pos sibly contribute to downstream eutrophication. A 1986 stu dy by Twinch showed a slightly more complex

PAGE 17

17 effect of drying on P exchange characteristics. In this study, sediment samples from a hypereutrophic lake were taken, with half being allowed to air dry, and the other half were stored wet. Although drying decreased the amount of bioavailable P in the sediment, the most no t able effect of drying was to: increase labile P by 127%, increase EPC by 199%, and decrease the P buffer capacity by 20%. Overall, the biogeochemistry of nutrients (primarily N and P) is strongly linked with water t able dynamics. When flooded, wetlands can effectively remove nitrogen from the water column, and have the ability to sequester phosphorus (depending on various soil physical an d chemical characteristics). Once drained, the process of denitrification halts, and bound phosphorus can be released (Olila et al. 1997; Twinch 1987; Reddy and DeLaune 2008). The biogeochemistry of carbon is also control l ed strongly by hydrologic factors; when flooded, carbon generally remains in the soil system, but can be lost through oxidation when water is removed. Environmental/Social Significance draining and reclamation of w etland ecosystems. The Upper St Johns River Basin (USJRB) is no exception; altered hydrology has caused severe environmental degradation in the years since its reclamation in the early 20 th century. Historically, wetlands covered almost 1800 km 2 of the USJRB. As a result of decades of draining and reclamation, much of the original USJRB wetland area was lost. After reaching a low of < 300 km 2 in 1985, the SJRWMD now manages roughly 590 km 2 of wetlands that are in various states of degradation (Br enner et al 2001). The primary services of these wetlands are to ensure: flood protection, environmental enhancement, water supply, and recreation. The effect of draining status on soil physical properties is well

PAGE 18

18 documented, with the overall end result b 2010). Drexler et al (2009) noted that when wetlands in the San Joaquin Delta were drained for farming, bulk density in creased significantly in the surface layer. Often, changes in the physical properties (often bulk density and hydraulic conductivity) of peat soils after draining are permanent, and cannot be significantly restored even if the historic hydrologic condition s are reestablished (Holden et al. 2006). While Price (2003) noted a slight increase in hydraulic conductivity after peat soil was rewetted, this followed a decrease in conductivity of up to two orders of magnitude after draining. In wetlands like the SJMC A that are managed for flood control, the ability for peat to quickly assimilate and store water is vital. Holden et al. (2006) show that hydrographs for drained wetlands demonstrate quicker runoff responses to rainfall. While this response may have been b eneficial when the wetlands were first drained for agricultural purposes, it now means that the wetland may be less effective at preventing flooding downstream. Over the last few decades, fish kills and harmful algal blooms have become a common occurrenc e on the St Johns River (Steidinger et al. 1998; Chapman et al. 2001). Both of these events have a similar cause: excess nutrients. According to Chapman et al. (2001) cyanobacteria are the primary culprits of harmful algal blooms (HABs) in Floridian freshw ater systems, and their occurrence is often predicated by changes in N or P limitations within the river. These Cyanobacterial blooms are not only unsightly, but they also may produce toxins that can attack the human liver, skin, and nervous system. Prolon ged exposure to these HABs can lead to conditions ranging

PAGE 19

19 limited to humans; frequently the occurrence of a HAB is followed by hypoxia and associated fish/ invertebrate kills. In August of 1999, an Anabaena bloom was implicated in causing a fish kill where over 10,000 menhaden, Brevoortia Tyrannus, and B. Smithi died (Chapman et al. 2001). As St Johns River HABs are often linked with nutrient leaching from upstream wetla nds, it is important to determine how hydrologic management effects nutrient movement. St Johns Marsh Conservation Area : The SJMCA is a region that contains primarily freshwater marshes, hardwood swamps, and small lakes. Hydrologic inputs to the SJMCA in clude precipitation, surface flow from the west, and channelized flow from the east and south. As seen in Figure 1 1, pasture land constitutes the western boundary of the wetland, while the eastern boundary consists of a water management and a water conser vation area.

PAGE 20

20 Figure 1 1 Location of study site The C 40 canal was dug during a period in history when many Florida wetlands were being reclaimed for agriculture or development. The result of this canal has been a severe over draining of the SJMCA To combat the environmental degradation that has resulted from this over draining a se ries of 8 earthen dams were placed in the C 40 canal although some have since been removed The objective of this construction was to facilitate the re flooding of the SJMCA. The average maximum and minimum annual air temperatures in the SJMCA are 27.5 and 17.3 o C respectively. The average annual precipitation 50.06 inches (Overpeck et al. 2010).

PAGE 21

21 The St Johns Marsh Conservation Area (SJMCA) covers roughly 13,880 ha of marshes and lakes in Brevard County, Southeast Florida. This area has historically been subject to periods of severe draining and hydrolologic isolation/alteration. Some of the primary results of this draining have been wetland system degradation, dimi nished water quality, and vegetation community shifts (commonly an invasion of high nutrient species like Typha sp ) (Miller et al. 2009). Previous studies have shown that soil subsidence is occurring in the upper St Johns River Basin, although a rate for s ubsidence has not yet been established for this region (Reddy et al. 2006). Objectives The primary research objective of this study is to determine: 1) what are the major causes of soil subsidence, and 2) how can the current hydrologic management scheme be st be modified to reduce or eliminate organic soil subsidence and allow the SJMCA to meet its requirements of: flood protection, environmental enhancement, water supply, and recreation. In this study, both field and laboratory experiments will be undertak en in order to build a subsidence model for soils in the SJMCA. As subsidence is related to carbon cycling, selected biogeochemical fluxes of carbon from the wetland to the atmosphere and hydrosphere will be determined.

PAGE 22

22 CHAPTER 2 SOIL SUBSIDENCE I N THE SJMCA Background Statement When flooded, carbon fixed by wetland vegetation can be stored as organic matter in the soil. This process is termed accretion, and describes the buildup of organic matte r that makes organic soils (muck peat, and histosols ). If flooding conditions are removed (either seasonally or anthropogenically), primary and secondary soil subsidence may occur, leading to a lowering of the soil surface elevation According to Vepraskas and Ewing (2006), the significant factors influenci ng primary and secondary soil subsidence are: original soil thickness, depth to water t able mineral content, temperature, precipitation, and management practices. Other factors like wild fire can also cause rapid loss of soil carbon (Stephens 1956). As s econdary soil subsidence (loss of soil organic carbon) through aerobic decomposition) results in the permanent depletion of soil material, this process is of great concern for wetlands that rely on peat soils to aid in flood control. Holden et al. (2006) f ound that when large wetlands are drained, rainfall runoff became much more flashy and was transferred downstream with great efficiency The SJMCA is located in a subtropical region where intense rainfall is not uncommon One of the primary purposes of th e SJMCA is to catch runoff from high precipitation events like hurricanes, and prevent it from flooding nearby urban areas. Because previous management practices have altered the effectiveness of the wetland in flood prevention, it is vital to understand h ow these practices influence soil subsidence. When the factors leading to soil subsidence are better understood, the SJMCA can be managed to more effectively prevent floods and maintain water quality.

PAGE 23

23 The overall objective of this research is to quantify the effect of current hydrologic management schemes on both primary and secondary subsidence. As many researchers claim that microbial oxidation of SOM is a major cause of subsidence (Morris et al. 2004; Reddy et al. 2006), a primary goal of this study wil l be to understand the relationship between hydrology and CO 2 flux (a measure of aerobic respiration). To accomplish this task, both field and laborat o ry studies were conducted in order to determine how carbon is cycled through the SJMCA, and what effect h ydrology has on this cycling. Methods Field Methods All of the field sites for this study are located within the SJMCA ( Figure 1 1) The location of these sites were chosen to maximize elevation and vegetation gradients. Two transects of approximately 0.5 km each have been established along the main C 40 canal Thes e sites can be seen in Figure 2 1. Each Transect consists of three sites: one close to the canal, one at a moderate distance, and one far away from the canal, with the objective of creating a pr ofile of wet to dry across the wetland. Four more sites were set up on the north side of the marsh where multiple vegetative communities existed in close proximity. The 4 vegetation communities chosen were: Willow ( Salix caroliniana), Sawgrass ( Cladium jam aicense ), Cattail ( Typha spp. ), and open water. One last site (SJM3 CTR) was created on the western side of the SJMCA where the soil was characteristically dry. Thus, a total of 11 sites were established in the SJMCA. At each site, 20 cm diameter PVC collar installed in triplicate, as well as a single shallow (~1.5m) monitoring well. The 20cm diameter collars were 30cm in height, with the bottom half inserted into the soil; the top

PAGE 24

24 2 half was left protruding from the soil to allow coupling of the Li 8100. These collars were left in place throughout the study so as to not disturb the soil when repeated measurements were taken. Each well was made by drilling a series of holes into a 5cm (2in) diameter, 1.2 1.5m long PVC pipe. Figure 2 1. Map of field locations A) Overall f ield sites in the SJMCA B) Vegetation communities. C) Transect 2. D) Transect 1. A B A C D A 1 2

PAGE 25

25 At the center of each field site, an auger just under 5cm (2in) in diameter was used to excavate soil to allow installation of the wells. Depth to water in the wells was measured using a tape measurer equipped with a water depth sounder. An image and diagram of an individu al site can be seen in Figure 2 2. Figure 2 2. Diagram and picture of study site layout. A) Individual site diagram B) P hotograph of one site (SJM2 T3). Photo courtesy of Bryce Van Dam A B

PAGE 26

26 Monthly field measurements were conducted in SJMCA from March 2011 to September 2011. A Li 8100 soil respiration monitor equipped with a 20cm chamber was used to determine CO 2 flux. Table 2 1 shows the parameters that were used to run the Li 8100. CO 2 flux was measured at each site during each visit, along with water depth and soil temperatur e ( 11.7 cm Taylor d igital temperature probe) For the first three months, three replicate CO 2 measurements were taken at each site, but in subsequent months, only 2 replicates were used because of time constraints. Every final CO 2 flux value is composed of two separate measurements, which were averaged together. Table 2 1 Li 8100 parameters used for field and l ab Obs Length (min) Pre Purge (min) Post Purge (min) IRGA Av eraging (sec) Dead Band (min) Chamber Diameter (cm) Obs # Field 1.5 1.75 0.5 0.5 4 0.5 10 2 Lab 1.5 2.0 0.5 0.5 4 0.5 20 2 As CH 4 sampling is a much more time consuming process, monitoring frequency was constrained to field access availability. To determine methane flux, as gas septum was installed on the Li 8100 (in line with tubing connecting the chamber and the analyzer) that allowed periodic gas sampling. A 20 cm Li cor chamber was placed over the same PVC collars used for CO 2 measurement, and allowed to equilibrate for 30 minutes to 1 hr. During sampling in Januar y, methane samples were collected using a chamber made from 20cm diameter PVC collars and end caps. Septa were installed to allow sampling, and small CPU cooling fans (the type commonly found in desktop computers) were used for air circulation. These fans were turned on for 5 10 seconds before each sample was taken to ensure complete mixing of the gas. Gas samples

PAGE 27

27 were taken at pre determined time intervals and transferred into evacuated gas vials. Analysis was completed on a GC equipped with an FID detecto r, and methane flux was calculated using a linear regression of concentration against time. For this and all other studies, statistics were analyzed using JMP 8.0.2 All comparisons were made using one way ANOVA or student ized t test s at Lab Metho ds To relate soil subsidence with corresponding rates of OM oxidation, our field work was supplemented with laboratory studies. The goal of these lab studies wa s to model how soils react to changing hydrology on shorter temporal scales. The data collected in these lab experiments were used along with the results from the field in order to create a more accurate model of soil subsidence. There were three laboratory studies study t able r espectively. The subsidence study involved a complete draining of soils in order to measure the potential rate of soil subsidence. In the drain experiment, soil cores were drained completely as in the subsidence study, but were monitored for CO 2 flux and s oil moisture instead of subsidence. In the water t able experiment, the goal was to determine the influence of water depth on carbon loss. Subsidence Study To determi ne total subsidence rates, 37 soil cores ( 10cm diameter ) were taken from the SJMCA, and br ought back to the laboratory. These cores were taken on March 16, 2011 when the SJMCA was dry at all field sites. 9 cores were taken from each of the 4 vegetation communities, and one from SJM3 CTR. These cores were all cut down to a common length of rough ly 15 cm. A cap with three small holes was placed on the bottom of each core to facilitate wetting and draining. Flooding from bottom to top was

PAGE 28

28 desired because we wanted to avoid moving nutrients down the soil profile, as well as prevent gas bubbles from being trapped in the soil matrix. The cores were then flooded with DI water for 9 days as part of a concurrent experiment, and subsequently were allowed to drain under gravity. Following draining, the depth of soil was measured 15 times over 148 days total The ambient air temperature was regulated by an air conditioning unit, and was generally between 21 24 o C. The location of these soi l cores can be seen in Figure 3 1. A simple linear regression of soil depth and time was used to determine the total subsid ence rate. Water Table Study To understand the link between hydrology and gaseous C losses, laboratory experiments were conducted in which the water t able was manipulated to varying levels, and CO 2 /CH 4 fluxes were measured. Additionally, an experiment was conducted wherein flooded cores were allowed to drain under gravity, and CO 2 flux was measured at pre determined time intervals. For both experiments, 32 cores (10cm wide x 30cm deep ) were taken from the SJMCA near site SJM1 T2. Caps with one small hole were placed on the bottom of each core, and the cores were put in 5 gallon buckets and moved to a South facing bench in a greenhouse. The cores were always covered with a black plastic tarp when not being sampled. 18 of the original 32 cores were randomly selected using a random number generator found at < http://www.random.org >. for each of 6 water level treatments (+1, 2.5, 7.5, 10, 15, and 20 cm). Each treatment contained three replicates, and the cores were all fl ooded wi th DI water as seen in Figure 2 3 Water was kept at the appropriate level by topping off the buckets containing the cores with DI water. The cores remained flooded for 22 days before measurement began to ensure that equilibrium was reached. To mea sure CO 2 flux, a Li 8100 with

PAGE 29

29 10cm collar was used. As with the field measurements, CH 4 flux was determined by placing a septum in between the chamber and the analyzer to allow gas extraction. Gas samples were taken every 5 10 minutes for a total of 30 min utes and transferred into evacuated glass vials. Methane analysis was completed on a Shimadzu GC FID, and flux was calculated using a linear regression. Soil temperature was also monitored in each core during flux measurement using a temperature probe. Figure 2 3. Picture of water table experiment Cores were randomly assigned to each of 6 water t able treatments. Photo courtesy of Bryce Van Dam. Drain Study The other 18 cores were flooded to +2 cm with DI for 28 days (also placed in 5 gallon buckets on a South facing bench in a greenhouse ) and were subsequently drained. CO 2 flux was measured 12 times over the following 90 days. CO 2 flux was measured using a Li 8100 equipped with a 10 cm collar. Measurements took place at varying times throughout the da y in order to account for different t emperatures that exist in the environment. Average daily air temperature was 31.5 0 C and ranged from 27.4 to 35.65 0 C, and the t emperature during individual measurements ranged from 24.66 to

PAGE 30

30 38.19 0 C. Cores were periodically removed in triplicate sectioned into 2cm samples, and analyzed for water content by drying at 105 o C for 24 hrs. Results Insitu Study Over the course of the study, CO 2 flux ranged from 106 to 7602 gC m 2 yr 1 and was 3036 171 2 gC m 2 yr 1 on average for all field sites The water level was quite v ariable, ranging from a low of 116 cm below the soil surface during the summer to a high of 29 cm over the surface in the winter. Average surface soil temperature ranged from 14.2 to 27.7 0 C (using an 11.7 cm probe) mirroring average air temperature for the same area. While CO 2 flux did not exhibit a strong linear relationship with water level (R 2 =0.28), flux measurements when flooded were significantly lower than when drained test, A lpha=0.05), as seen in Figure 2 4 When drained, CO 2 flux was 2595 387 gC m 2 yr 1 (n=57), and decreased to 417 78 gC m 2 yr 1 (n=12) when the water level rose above the soil surface. These standard errors for drained and flooded CO 2 flux of 387 and 78 gC m 2 yr 1 respectively indicate that aerobic decomposition is quite spatially and temporally variable.

PAGE 31

31 A B C D Figure 2 4. CO 2 and CH 4 fluxes with water depth. A) CO 2 flux (gC m 2 yr 1 ). B ) CH 4 flux (gC m 2 yr 1 ),and flooding status. C ) CO 2 Flux (gC m 2 yr 1 ) and depth to water (R 2 Adj = 0.27, n=69). D ) CH 4 Flux (gC m 2 yr 1 ) and depth to water (n=33). Methane flux contrasted with CO 2 flux in its relationship with water level. When drained, average CH 4 flux was 7.2 6 gC m 2 yr 1 (n=22), and increased to 30 14 gC m 2 yr 1 (n=11) when the water level rose above the soil surface. While flooding did result in a higher (and much more variable) CH 4 that any differe nce was not significant.

PAGE 32

32 Figure 2 5. Log CO 2 :CH 4 flux and Water Level (cm). n=22, R 2 Adj=0.72 Figure 2 5 shows that as the water level dropped, the ratio of CO 2 to CH 4 production rates increased logarithmically. While this relationship appears to be li near, extrapolation beyond water levels of 100 to +20 cm would likely not be accurate. As the water t able rises over +20 cm, methane oxidation within the water column would cause CO 2 :CH 4 to increase. Total carbon respiration (CO 2 + CH 4 ) was 447 79 g C m 2 yr 1 when flooded, and increased to 2603 388 g C m 2 yr 1 when drained ( Figure 2 6 ). According t

PAGE 33

33 Figure 2 6 Total respiration ( g C m 2 yr 1 ) and water level (cm) Laboratory Subsidence Using a linear regression, total soil subsidence was found to occur at a rate of 4.7 cm yr 1 for the first 148 days after drainage, as shown in Figure 2 7 This rate accounts for a decrease in soil volume of at least 31% (each core was 15 cm long). Both primary and secondary subsidence influenced this rate.

PAGE 34

34 Figure 2 7 Average soil elevation (cm) and days drained ( n=37 ) Given field CO 2 fluxes of between 2066.2 and 3493.6 gC m 2 yr 1 (25 and 75% quartiles respectively), average bulk density of 0.173 g cm 3 and assuming the peat soil is 58% carbon, the relative contributions of primary and secondary subsid ence can be calculated. Table 2 2 shows these findings. Secondary subsidence (oxidative losses) was found to account for between 44 and 83% of total subsidence. Table 2 2 Estimates of subsidence fractions (percent of total in parenthesis) n=37 Factor Low Estimate (cm yr 1) High Estim ate (cm yr 1) Average (cm yr 1) Primary Subsidence 0.78 (17%) 2.65 (56%) 1.72 (36.5%) Secondary Subsidence 2.05 (44%) 3.92 (83%) 2.99 (63.5%)

PAGE 35

35 It was noted that the peat pulled back from the sides of the cores over the study period, mirroring the peat cracking during the dry season in the field. In most cores, peat was observed floating upon flooding (for a concurrent study). Additionally, peat w as seen floating in the SJMCA when it became flooding, indicating that similar processes are happening in the field as in the laboratory. Water table study In this study, CO 2 flux increased linearly from 441 to 923 g C m 2 yr 1 as the water t able decreased from +1 to 20 cm ( Figure 2 8 ) All CH 4 fluxes recorded when the water t able was beneath the surface were essentially negligible but increased to an average of 30 g C m 2 yr 1 when flooded.

PAGE 36

36 Figure 2 8 Laboratory CO 2 and CH 4 fluxes with water depth. A ) CO 2 (g C m 2 yr 1 ) flux. B ) CH 4 (g C m 2 yr 1 ) Fluxes at different water t able treatments (cm). A least squares method was used to fit models for the cumulative effect of soil temperature and water depth on both CO 2 and CH 4 fl uxes. Th ese models are shown in Table 2 3. While all soil cores showed an increase in CO 2 flux with increasing temperatures, treatments with lower water levels (drier surface soil) were more sensitive to temperature changes. Using the definition of Q 10 fr om Inglett et al. (2011) Q 10 =(k 2 /k 1 ) 10/(T2 T1) ( where k 1 and k 2 are the CO 2 fluxes at lower at lower and higher A B

PAGE 37

37 temperatures respectively, and T 1 and T 2 are the low and high tem peratures respectively), i t was found that Q 10 showed a negative relations hip with water level ( Figure 2 9 ). Q 10 increased from 1.48 when the soil was flooded to 4.37 when drained to 20cm. Table 2 3 Least squares models for CO 2 (g C m 2 yr 1 ) and CH 4 (g C m 2 yr 1 ) Fluxes Adj R 2 of 0.77 and 0.13 respectively. CO2 Parameter Est imate Std Error P > |t| CH4 Parameter Estimate Std Error P > |t| Intercept 649 194 0.0043 Intercept 28.3 19.6 0.1834 Soil Temp (0C) 53.2 9.34 < 0.0001 Soil Temp (0C) 0.806 1.01 0.4452 Water Depth (cm) 22.8 4.92 0.0003 Water Depth (cm) 0.926 0.553 0.1823 Figure 2 9 Average Q 10 for CO 2 flux with varying water depths (cm). Smoothing Spline Fit, lambda=100, R 2 = 0.42 Drain study After an unexpected initial CO2 Flux as high as 50757 g C m 2 yr 1 during the first day after draining, a steady rate of between 600 and 1200 g C m 2 yr 1 was observed

PAGE 38

38 for the remainder of the study. Results of this experiment are shown graphically in Figure 2 10 Figure 2 10 CO 2 flux after draining, and with air temperature. A ) CO 2 flux (g C m 2 yr 1 ) after draining B ) Air temperature ( 0 C) and CO 2 Flux (g C m 2 yr 1 ), Adj R 2 =0.32 n=18 Because these cores were incubated in a greenhouse, ambient air temperature was quite variable (ranging from 21.11 to 35.65 0 C). When potential outlier CO 2 fluxe s A B

PAGE 39

39 were removed, air temperature accounted for 32% of the variability between measurements (using a linear regression). While this experiment was designed to study the effect of changing soil moisture on CO 2 flux, average soil moisture within the cores did not change between these sampling dates. The same held true when only surface soils (0 8cm) were considered. Soil cores were sampled at 22, 59, and 90 days for soil moist ure ( Figure 2 11 ). Figure 2 11 Soil moisture (%) after draining. P value = 0.7128 n=3 Discussion Insitu Study When the water t able was below the surface of the marsh, CO 2 flux became high and quite variable. Some of this variability in drained CO 2 flux could be explained by the fluctu ating nature of soil temperature during the summer months. A study by Inglett et al. (2011) showed that temperature, vegetation, and their interaction was a significant factor in determining greenhouse gas flux (aerobic and anaerobic) for soils in the same general region in Florida. The current study also demonstrates the importance of

PAGE 40

40 temperature on aerobic respiration; Q 10 values of over 4 were found for potential CO 2 flux. It is also interesting to note that according to a studentized t erage water level did not differ significantly between any pair of field sites. Because the soil elevation does differ between sites, this lack of difference in water levels indicates either inaccuracies in measurement, or that the potentiometric surfaces at these sites mirrors their respective soil elevations. This lack of difference in average water levels was seen despite the fact that the transe cts were set up for the express purpose of studying the effect of differing water levels on CO 2 flux. Wetland s are generally very productive ecosystems and can have significant impacts on the global carbon cycle While under normal hydrologic conditions, wetlands are considered to be net carbon sinks, however, they can become sources when drained (Schedlbauer et al. 2010; Kechavarzi et al. 2010; Jauhiainen et al. 2008; Armentano et al. 1986). Primary productivity in wetlands has been estimated to be between 1300 3200 g C m 2 yr 1 (Miller and Fujii 2009). Therefore, using the CO 2 and CH 4 fluxes found in this stu dy, the SJMCA would undergo rapid oxidative soil loss (roughly 6 8 kg C m 2 yr 1 ) if drained, but would be in approximate equilibrium while flooded. This relates to an average net flux of carbon from the SJMCA to the atmosphere of 0.97 Tg C yr 1 under dra ined conditions during the study period ( not a full year). Over the course of the study, methane contributed a relatively small amount to overall carbon flux (7.2 30 gC m 2 yr 1 ). Lowest methane fluxes were seen when the water t able was below the surface, while highest fluxes occurred when th e wetland was flooded ( Figure 2 4). While drained CH 4 fluxes were consistently low, flooded CH 4 flux

PAGE 41

41 was quite variable. Although factors like temperature and nutrient status certainly contribute d to some of this variability, measurement methods are also of concern. When CO 2 flux is measured, the observer only needs to approach the chamber once to initiate the experiment, then does not return until the measurement is completed. As methane flux was measured using gas samples taken at 5 10 minute intervals, the observer must approach the chamber a number of times, pos sibly disturbing the soil, rele asing excess methane to the chamber. Care was taken to ensure this sour c e of error was kept to a minimum but some level of disturbance was inevi t able Given that the SJMCA was only flooded for about 25% of the year during this study, CO 2 flux should be considered the main gaseous loss of carbon from this wetland. Methane fluxes found in this study represen t the difference between the processes of methanogenesis (production) and methane oxidation within the soil plant air continuum. Methanogenesis is the result of a series of microbially mediated reactions of which the final step is the actual production of meth a ne. Methane oxidation is also a microbial process that occurs at the aerobic anaerobic interface within the soil, often oxidized portions of the rhizosphere (Segers 1998; Altor and Mitsch 2008). Thus, the gross methane flux in the SJMCA is almost cer tainly much higher than the net fluxes shown here. A 1990 study of Everglades peat soils showed that 91% of the methane produced in soils is oxidized before it can be lost to the atmosphere. This oxidation occurs in the aerobic portion of the soil/floodwat er system (King et al. 1990). As seen in Figure 2 4 (c ,d ), the inflection point for both CO 2 and CH 4 fluxes did not line up exactly with a water depth of 0 cm as expected. CO 2 flux began to decrease and CH 4 flux began to increase as the water t able reache d about 10 cm Therefore, even

PAGE 42

42 though the potentiometric water level was well below the soil surface, capillarity action was sufficient to generate anaerobic conditions at the soil surface. This information could be very us eful in managing the SJMCA, it indicates that soil loss could be kept to a minimum while still dr awing the water t able to 10 cm below the soil surface. This water t able drawdown would allow the wetland to approximately store an additional 1.4 x10 7 m 3 of wa ter which is vital given the importance of the flood control mission of the wetland. Laboratory Subsidence study Total subsidence was found to occur at a potential rate of 4.7 cm yr 1 This rate includes primary and secondary subsidence, which were calculated to account for 36.5 and 63.5% of total subsidence respectively. A study by Vepraskas and Ewing (2006) found primary and secondary subsidence to account for 69 and 31% of total sub sidence respectively. Their study compared subsidence rates between soils drained for 15, 20, and 30 years. As time drained increased, primary subsidence became more influential relative to secondary subsidence. Therefore, as the soils in the current study were drained for less than a year, it would follow that secondary subsidence would be the main cause of soil elevation change which was the case Nevertheless, the SJMCA has been chronically overdrained since the mid 20 th century. If this drainage time is taken into account, it would then be expected that primary subsidence would dominate relative to secondary subsidence based on previous studies The findings in this study could thus be interpreted as b eing somewhat unconventional in that secondary subsidence was higher than primary subsidence for soil that has been hydrologically altered for a significant period of time.

PAGE 43

43 As was seen in the water t able study (and in many previous studies (Inglett et al. 2011; Kirschbaum 2006; Chapm an et al. 1996)), temperature is an important factor in determining the rate of organic matter oxidation. Thus any differences in temperature could introduce a large source of error into this subsidence model. The average air temperature in the study area is 22.4 0 C, and the air temperature in the laboratory was between 21 and 24 0 C. This is a maximum difference in temperature between field and lab of 1.6 0 C. A study of BCM soils (Inglett et al. 2011) found Q 10 values to be in the range of 1.4 to 2.1. If Q 10 is assumed to be the mean of these values, 1.8, CO 2 flux would only be a maximum of 11% different between the field and lab. While temperature differences are only a source of minimal error, there are certainly other factors that could cause the potent ial subsidence rate to differ from the actual rate, including vegetation, fauna, aeolian soil loss, fire, etc. Although these factors cannot be accurately accounted for using this subsidence model, it should still be relatively accurate in predicting soil elevation change over time. Water t able study Figure 2 9 shows CO 2 flux increased linearly from 441 to 923 g C m 2 yr 1 as the water t able decreased from +1 to 20 cm. All CH 4 fluxes while drained were essentially negligible but increased to 30 g C m 2 yr 1 when flooded. A least squares model showed that water depth and temperature were both significant predictors of CO 2 flux, but not of CH 4 flux. The inclusion of water temperature increased the model reliability markedly for CO 2 flux, but decreased reli ability slightly for CH 4 flux (Adj R 2 increase of + 148% and 19% respectively). This is consistent with literature reports that aerobic decomposition is more temperature sensitive than anaerobic decomposition (Chapman et al. 1996).

PAGE 44

44 An interesting finding of this study was that the temperature sensitivity ( Q 10 ) was greater when the water t able was low, and decreased when flooded. Q 10 values were between 1.48 and 4.37. Values in the range of 1.5 to 3.5 are typical for soils in this region according to previ ous studies by Inglett et al. (2011). Although determining temperature sensitivity was not part of this study, this finding offers valuable insight into the factors effecting soil oxidation in the SJMCA. This finding indicates that the season in which the water t able is drawn down is very important. During winter months, the depth of water in the wetland should have a small influence on the CO 2 flux, while during summer months, the flooding status will be critical. Another possible explanation for this obse rved relationship is that increas ing recalcitrance of OM is related to greater temperature sensitivities (Conant et al. 2008) Thus, as time progresses after draining, the peat becomes more recalcitrant, and possibly more sensitive to temperature changes. Drain study After draining, CO 2 flux was initially very high, but quickly decreased to a steady rate of between 600 and 1200 g C m 2 yr 1 within the first 10 days. These anomalously high initial CO 2 fluxes are most likely a result of ebullition of gas bubbles from the recently drained soil. While the soil cores were flooded, gas accumulated in pore spaces, so when the cores were drained, this gas was released. CO 2 flux was expected to increase to a maximum while soils drained to optimal levels, then to decrease as the soil became too dry for respiration. This trend was not seen, possibly because soils never dried out fully. Figure 2 12 shows that the water content of the soil cores did not change significantly between days 22, 59, and 90 after d raining. As described in the methods, a black plastic tarp was draped over the cores

PAGE 45

45 during this experiment to inhibit the growth of plants. This tarp may have had the unintended effect of also limiting water vapor transport from the cores to the atmospher e. Conclu ding Remarks The objective of this research was to quantify the effect of current hydrologic management schemes on both primary and secondary subsidence. This study indicated that under current management practices, soil subsidence will continue to be a factor in the SJMCA. While drained, the potential subsidence rate was 4.7 cm yr 1 When the soils are allowed to drain for approximately 1 month per year or more, total subsidence (primary and secondary) will exceed sediment accumulation, causing n et soil loss. Furthermore, the finding that secondary subsidence exceeded primary subsidence is somewhat novel and indicates that the legacy of drainage in the SJMCA might be different than in other wetlands with similar hydrology. While vegetative type may not be particularly critical in determing CO 2 flux rates, temperature and water depth are key to predicting carbon mobility. During both field and laboratory studies, hydrology exerted a strong influence on CO 2 flux; rates were low when the soil was fl ooded, and high and variable when the water t able was below the surface. CO 2 Flux when drained was roughly 6 times higher than when flooded. Similarly, CH 4 flux showed a strong trend with water level. When drained, flux was near zero (7.2 6 g C m 2 yr 1 ), but became high and variable when flooded (30 14 g C m 2 yr 1 ). Following draining, CO 2 flux was initially high and variable, but quickly (within 10 days) approached a steady rate of between 600 and 1200 g C m 2 yr 1 Overall, hydrologic management a ppears to have a strong influence on carbon dynamics in the SJMCA. Because wetlands are characterized by intermediate water

PAGE 46

46 t able conditions (water level at or near the surface), small changes in hydrology can have significant effects on carbon cycling. As current hydrologic management results in a water t able dra w down during the summer months, when temperature is high, oxidative carbon loss will continue to be a problem in the SJMCA.

PAGE 47

47 CHAPTER 3 NUTRIENT EXCHANGE IN THE SJMCA Opening Statement Wetla nds can effectively function as sinks or sources of nutrients to downstream water bodies. The nutrients of most concern in Florida are carbon (C), nitrogen (N), and phosphorus (P). When wetlands are flooded, the diffusion of oxygen into the soil is greatly diminished, restricting the ability of aerobic microbes to compete for organic matter. When oxygen is not available, other electron acceptors (NO 3 Fe 3+ SO 4 2 etc) are used by anaerobic microbes to break down organic carbon (Mclatchey and Reddy 1998; Wh ite and Reddy 1997). In the case of nitrogen, when NO 3 (aq) is reduced during the oxidation of organic matter, gaseous species are formed (N 2 0 and N 2 ). This process is termed denitrification, and can result in a permanent removal of nitrogen from the wetla nd. In freshwater systems, denitrification can lead to a nitrogen flux from the wetland to the atmosphere of up to 0.5g N m 2 day 1 (Knowles 1982 ; Bowden 1987; Hansen et al. 1993 ). If not denitrified, t his nitrogen would be transferred downstream where it could contribute to water quality degradation. In addition to providing this biogeochemical sink for nitrogen, wetlands may also retain phosphorus. Under aerobic conditions plant litter can be fully de composed, allowing all organic phosphorus to be mineralized to PO 4 3 Wetlands, on the other hand, are characterized by anaerobic conditions where S OM is not fully mineralized. In this case, phosphorus within the S OM is protected to a greater extent from m icrobial activity. Therefore, wetlands can serve as a sink for phosphorus, sequestering this nutrient along with carbon in thick layers of peat (Mclatchey and Reddy 1998, Olila et al. 1997).

PAGE 48

48 While under flooded conditions, wetlands can restrict the flow o f nutrients from upstream lands to downstream water bodies, upon draining, this mechanism is lost. When wetlands are drained, oxygen can penetrate into soil organic matter that is highly reduced. This has profound effects on the fate of soil C, N, and P. W hile flooded, the major end products of decomposition are CH 4 and DOC. When drained, porewater DOC that accumulated during anaerobosis is rapidly mineralized to CO 2 and lost to the atmosphere (Ellis et al. 2009; Freeman et al. 1996; Leiros et al. 1999). E llis et al (2009) found that a water t able drawdown of 12 weeks resulted in a 12.6% decrease in porewater DOC. This aerobic DOC mineralization brings a concomitant flux of P and N. A 1996 study by Freeman et al. demonstrated that water t able drawdown incr eased phosphatase (enzyme responsible for hydrolysing P containing substrates) activity by 31 67%. This increase in phosphatase activity can result in a large flux of PO 4 2 into the water column upon re flooding. As no common gaseous form of phosphorus exi sts, the P mineralized during draining remains in place within the OM until reflooding. Olila et al. (1997) found that draining lake Apopka sediment for 6 weeks resulted in 334 mg P m 2 d 1 being released to the water column. Similar to phosphorus, NH 4 + ( ammonium) that was accumulated in the soil when flooded can be nitrified (microbially oxidized to NO 3 ) when drained. A study by Malecki et al. in 2004 showed that sediments in the Lower St Johns River (LSJR) fluxed significantly more ammonium into the wat er column when the soil was anaerobic than when aerobic. This ammonium is oxidized to nitrate through the process of nitrification. Under flooded conditions, this nitrate would be rapidly used by denitrifying microbes,

PAGE 49

49 and released as N 2 gas. Nitrate remai ns in the soil when drained, and is lost to the water column when re flooded (Bowden 1987). While nitrogen and phosphorus are necessary nutrients in all biological systems, they can lead to imbalances in water chemistry when in excess concentrations (eutr ophication) This nutrient imbalance can cause the proliferation of nuisance species of algae and macrophytes (Maher 1997; Malecki et al. 2004 Lorenzen 2001 ). These nutrients can come from point and non point sources, both of which are of significant conc ern in the St Johns River watershed. According to a report by Maher in 1997, the loads of nitrogen and phosphorus in the lower St Johns River are dominated by contributions from upstream flow. As this upstream component of nutrients originates in the marsh es in and around the SJMCA, it is vital to understand how these nutrients move through the wetland and how water management may be utilized to reduce nutrient loading (deterring subsidence of organic soils) In many oligotrophic aquatic systems phosphorus is a limiting nutrient, and if added in excess of a critical amount can lead to the development of HABs or crashes in dissolved oxygen While phosphorus can be assimilated into wetland soils effectively when flooded, it is rapidly mineralized and can be lost to the water column when the soil is drained and subsequently flooded (Olila et al. 1997). As the current management practices result in severe water t able fluctuations (flooded in the winter and drained in the summer), yearly phosphorus fluxes during the summer months are frequent. Interestingly, these P fluxes from the USJRB can initiate a positive feedback cycle in the lower SJR. Acc ording to a study by Malecki et al. (2004), algal blooms caused by nutrient loading can cause decreases in porewater O 2 and subsequent anoxia This

PAGE 50

50 anoxia increased sediment to water P flux by a factor of 40, further enhancing eutrophication. According to a 210 Pb nutrient accumulation study by Brenner et al. (2001), the USJRB has experienced increased P loading over the last 70 years. This phosphorus loading in conjunction with seasonal water t able fluctuations has created a situation where DO crashes, fish kills, and harmful algal blooms ( HABs ) are a common occurrence on the lower St Johns River, a water body economic and environmental well being. Clearly, more needs to be known about the effect of water t able dyna mics on nutrient biogeochemistry so the wetlands of the USJRB can be best managed to prevent floods without causing water quality degradation. Nutrient mineralization and subsequent eutrophication of downstream water bodies is a serious problem in the SJ R watershed. As such, it is vital to know how major nutrient biogeochemistry (cycling of nitrogen, phosphorus, and sulphur) is affected by soil drying and rewetting under alt ered hydrologic conditions. The objective of this research is to determine how var ious water management schemes affect nutrient losses from the SJMCA. Potential nutrient mineralization rates will be determined using both laboratory and field studies. Methods Site Description All of the field sites for this study are loc ated within the SJMCA ( Figure 3 1). The location of these sites were chosen to maximize elevation and vegetation gradients. Two transects of approximately 0.5 km each have been established along the main C 40 canal. Each Transect consists of three sites: one close to the canal, one at a

PAGE 51

51 moderate distance, and one far away from the canal, with the objective of creating a profile of wet to dry across the wetland. Four more sites were set up on the north side of the marsh where multiple vegetative communities existed in close proximity. The 4 vegetation communities chosen were: Willow ( Salix caroliniana ), Sawgrass ( Cladium jamaicense ), Cattail ( Typha spp. ), and slough One last site (SJM3 CTR) was created on the w estern side of the SJMCA where soil was characteristically dry. Thes e sites can be seen in Figure 3 2.

PAGE 52

52 Figure 3 1 Location of study site

PAGE 53

53 Figure 3 2 Map of field sites. A) Overall f ield sites in the SJMCA. B) Vegetation communities. C) Transect 2. D) Transect 1. Thus, a total of 11 sites were established in the SJMCA. At each site, 20 cm diameter PVC collars and 10cm diameter PVC pipes ( 1 1.5 m tall) were installed in triplicate, as well as a single shallow (~1.5m) monitoring wel l. Each well was made by dril ling a series of holes into a 5 cm diameter, 1 1.5 m long PVC pipe. At the center of each fi eld site, an auger just under 5 cm in diameter was used to excavate soil to allow 1 A B C D

PAGE 54

54 installation of the wells. Depth to water in the wells was measured using a tape measurer equipped with a water depth sounder A diagram and a picture of a fie ld site can be seen in Figure 3 3 Figure 3 3 Diagram and picture of field site layout. A) Individual site diagram B) P hotograph of one site (SJM2 T3). Central well, nutrient flux pip es, and gas flux collars can be seen. Photo courtesy of Bryce Van Dam. A B

PAGE 55

55 Soil Chemistry Soil cores taken from the SJMCA were sectioned by depth and analyzed for total C, N, P, and S in this study One core (5 cm diameter by 20 30 cm in length) was taken from each of the 11 field site s described in the site description section Due to soil compaction, it was not possible to fill some cores to the top. These cores were extruded by hand and cut into 2 cm segments. This soil was then dried at 105 o C for 24 hrs, s ieved then ball milled for 15 minutes prior to analysis. Total carbon and nitrogen were determined using a Leco Tru Spec combustion analyzer. Total sulfur was dete rmined by HCl extraction/ICP (Thermo ICP6500). TP was determined by atomic emission on a Per kinElmer 5300DV (EPA 200.7). For this and all other studies, statistics were analyzed using JMP 8.0.2. All comparisons were made using one way ANOVA or studentized t test at Laboratory Nutrient Flux To reduce variability often found in field flux studies and look at more controlled water level effects, 37 soil cores were taken from SJMCA on 16 March 2011 (9 cores from each of 4 vegetation communities) and brought back to the laboratory where they were placed into buckets and flooded to 10cm over the soil surface with DI water. The soil at this region had been drained for approximat ely 3 months prior to removing these cores. All of these cores were cut down to 15 cm in length to ensure the soil volume f or each sample was the same. While site water could have been used to flood the cores, the use of DI water offered 2 distinct advantages for this study. DI water is more representative of rain water, and reduces a significant source of excess nutrients tha t could cause error in nutrient flux calculations. The 4 vegetation communities sampled were: Willow ( Salix caroliniana), Sawgrass ( Cladium jamaicense ), Cattail ( Typha spp. ),

PAGE 56

56 and open water. One last site (SJM3 CTR) on the western side of the SJMCA was use d where the soil was chronically overdrained. Before flooding, 3 small holes were drilled into PVC caps placed on the bottom of the cores. These holes allowed water to enter the soil from the bottom, so as to not move nutrients down the soil profile. These cores were kept in a dark tub in an indoor lab oratory (seen in Figure 3 4 ). A plywood cover was placed over the cores to prevent light entry. The ambient air temperature was regulated by an air conditioning unit, and was generally between 21 24 o C. After 9 days, water standing above the soil was sampled, filtered (0.45 um), acidified (to a pH of 2.0 with H 2 SO 4 ), and analyzed for DOC, TDP, and TDKN. Sampling was accomplished by lowering a small tube half way between the water surface and the soil, and pullin g water into a 60uL syringe. Standing w ater was not mixed prior to sampling as to prevent soil material from contaminating water samples. DOC was found using the high temperature combustion method with a Shimadzu TOC5050A CE ( SM5310B ).

PAGE 57

57 Figure 3 4 Soil c ores in laboratory. Container is insulated fiberglass; a plywood lid was placed over the bath when not being sampled to prevent light from entering. Soil cores were placed randomly into each of 13 5 gal buckets. Cores were flooded by pouring DI wate r into buckets outside of the cores. TDP was determined spectrophotometrically (Bran and Lubbe AA3 auto analyzer) using the ascorbic acid method ( SM4500P F ). TDKN was determined using a sul furic acid digestion, with analysi s on a Technicon AA2 auto analy EPA351.2 ) After sampling, the cores were allowed to drain under gravity for 215 days. Following this period, the cores were again flooded with DI water to an average of 13cm above the

PAGE 58

58 soil surface (total subsidence during the 215 days w as 3 cm on average ). After 9 days, surface water samples were again taken, filtered (0.45um), acidified (H 2 SO 4 ), and analyzed for DOC, TDP, and TDKN as before (TDKN was determined on an Astoria Pacific A2 segmented flow analyzer) Nutrient flux was calcula ted as in the field methods. Field Nutrient Flux PVC pipes were installed in triplicate at each of the 11 field sites. These pipes were 10 cm in diameter and were sunk about 30 cm into the soil ( Figure 3 3) The 10 cm diameter PVC pipes had one small hole drilled in the side to allow pressure equilibration upon flooding. Additionally, a loosely fitted PVC cap was placed on the top to prevent contamination by rainwater, insects, and birds from entering the nutrient flux tube. The purpose of these pipes was to allow sampling of soil pore water without contamination from surface or rain water. Once the water t able reached the surface, pore water fluxed from the soil into the inside the columns was removed from the tubes using a peristaltic pump. It could not b e determined exactly when the field sites became flooded, but effort was made to sample as soon as possible after flooding. Water was not mix ed within the tube in order to prevent soil from contaminating water samples. Water samples were analyzed by the St Johns River Water Manag ement District Laboratory for DO DOC, NH 4 TP, TDKN Turbidity and pH (along with other analytes that are not immediately relevant to this study). Nutrient flux was calculated as g m 2 using the pipe diameter and d epth of water at each location.

PAGE 59

59 Results Soil Chemistry Results from soil chemical analysis are shown in full in Figure A 2 in the Appendix and are summarized in Table 3 1. Soil t otal carbon increased with depth from 40.0 1.8 to 51.9 1.0 %. Total phosphorus decreased fro m a high of 1240.5 61.9 mg kg 1 to a low of 216.6 29.6 mg kg 1 Total nitrogen remained relatively constant with depth (average of 3.2 0.05%) but was highest at 18 20 cm. Total sulphur was also constant with depth (average of 0.29 0.01mg kg 1 ). C: N ratios ( Figure 3 3) were quite variable from site to site, but averaged 14.4 0.2 across depths and locations, typical for soils in this region. In general, C:N increased with depth, although site SJM2 T3 was an exception to this trend. Table 3 1 Soil chemical characteristics (Average Std Error) Site Average Bulk Density (g cm 3) TC (%) TN (%) TP ( g kg 1) TS (mg kg 1) SJM1 T1 0.16 0.01 43.1 2.8 3.18 0.17 0.62 0.08 160.6 0.03 SJM1 T2 0.14 0.01 48.1 4.9 3.42 0.24 0.64 0.39 130.4 0.06 SJM1 T3 0.12 0.01 48.8 1.1 3.57 0.06 0.55 0.08 161.7 0.04 SJM2 T1 0.29 0.04 36.6 3.6 2.76 0.25 0.52 0.05 399.0 0.02 SJM2 T2 0.23 0.03 42.8 2.3 3.26 0.14 0.62 0.09 368.2 0.03 SJM2 T3 0.16 0.007 45.7 1.2 3.47 0.18 0.82 0.11 315.6 0.02 SJM3 SAW 0.17 0.08 47.8 1.0 3.13 0.08 0.76 0.14 364.4 0.02 SJM3 TYP 0.15 0.004 45.5 1.1 3.20 0.04 1.26 0.10 355.6 0.02 SJM3 OW 0.17 0.01 49.5 5.5 3.26 0.33 0.56 0.11 317.4 0.04 SJM3 WIL 0.13 0.004 50.1 1.3 3.19 0.06 0.63 0.13 393.8 0.02 SJM3 CTR 0.17 0.02 52.4 1.0 3.00 0.08 0.49 0.08 241.1 0.03

PAGE 60

60 Figure 3 5 C:N and depth by field site. Lab Nutrient Flux As seen in Table 3 2, both TDP and TDKN fluxes were significantly h igher (students t slightly lower after drai ning, but not significantly so. Table 3 2 Nutrient fluxes (g m 2 ) given as mean S E (n=37) A sterisked p values are significantly different ( ) Post f ield Post drain (215 days) p value DOC 0.42 0.020 0.37 0.03 0.24 TDP 0.021 0.002 0.030 0.004 0.028* TDKN 0.062 0.004 0.20 0.020 <0.0001*

PAGE 61

61 Differences in nutrient fluxes were also calculated as % change from post field flux, and were compared between vegetation communities. These results are shown in Figure 3 6 ; values for % change greater than 0 indicate that nutrient fluxes were higher afte r draining than before. For example, TDP % change for the Slough vegetation community was approximately 100%, reflecting a two fold increase in phosphorus flux following draining. Figure 3 6 Nutrient flux % change by vegetative community (n=9) Interestingly, the site that had the highest soil TP values (Typha or SJM3 TYP) exhibited the lowest change in P flux following draining, and the site with the second lowest TP concentrations (Slough or SJM3 OW) showed the highest change in P flux.

PAGE 62

62 Figure 3 7 shows that soil TP was very highly correlated (R 2 Adj = 0.998) with TDP % change. It is important to note that soil cores measured for TP were not the same ones used for the nutrient flux experiment, although they did come from the same field site (les s than 10m apart) Figure 3 7 Average TP (m g kg 1 ) and TDP % change. Reciprocal fit. R 2 Adj = 0.998, n=4 Field Nutrient Flux Overall data from the field nutrient fl ux stud y are shown in Figure A 1 and A 3 in the Appendix As seen in Figure 3 8 nutrient fluxes in the field ranged from 25 50 g m 1 for DOC, from near 0 to 1.75 g m 1 for TP, and from 2 12 g m 1 for TDKN Notably, between vegetation communities, TP flux was highest in the willow ( Salix caroliniana ) dominated site ( Figure 3 8 ), and lowest in the open water (or slough) site. DOC and TDKN fluxes were also highest at the willow dominated site. The Cladium dominated site released the least DOC and TP, but had the second highest TDKN flux (one anomalously high value raised the mean ).

PAGE 63

63 Figure 3 8 Field nutrient flux es by site. A ) Field TDKN fluxes (g/m 2 ). B ) DOC fluxes (g/m 2 ). C ) TP fluxes (g/m 2 ). (n =3 ). A C B

PAGE 64

64 Least squares regression analysis suggests, DOC, NH 4 and D OC concentrations were the only significant predictors of TDKN flux. For DOC flux, TP, NH 4 D OC, and TDKN were all significant predictors. Only PO 4 and DOC were significant predictors of TP flux ( Figure 3 9 ). Figure 3 9 DOC and TP fluxes regressed against TKN A ) T KN (mg L 1 ) and TP(mg L 1 ) (R 2 Adj = 0.27, n=31 ). B) TDKN (mg L 1 ) and DOC (mg L 1 ) (R 2 Adj = 0.34, n=31 ) TDKN outliers of 40.2 and 50.6 mg L 1 were removed. During the laboratory nutrient flux experiment, it was possible to impose an equal level of flooding for all soil cores. This way, all samples would be diluted to the same B A

PAGE 65

65 level and any differences in nutrient flux would be a function of soil processes, no t experimental error. In this field experiment it was not feasible to sample with all soils flooded to the same level. Consequently, all field water samples were diluted to different levels based on their relative elevations. Therefore, in order to compare nutrient fluxes between the laboratory and field studies, fluxes were normalized by depth of flooding and area sampled to create Figure 3 10 The units for this Figure are grams per m 2 of soil per cm of floodwater (g m 2 cm 1 ). In general, nutrient fluxes in the field were approximately 10 times greater than in the laboratory (both initial and post drain). A B C Figure 3 10 Field nutrient fluxes compared with laboratory nutrient fluxes. A ) DOC fluxes (g m 2 cm 1 ). B ) TP fluxes (g m 2 cm 1 ). C ) TDKN fluxes (g m 2 cm 1 ) for field and lab studies

PAGE 66

66 Discussion Soil Chemistry Soil TP was relatively consistent between the transect sites, but was more variable between the different vegetation communi ties. Soil TP at the T ypha dominated site (1.26 mg kg 1 ) was significantly higher (students t test, than at all other sites. This finding was expected as typha is known to become established where phosphorus is found in excess (Lorenzen et al. 2001; Maher 1997). Both TC and TN were much more co nsistent from site to site, ranging from 36.59 52.75% and 2.76 3.57% respectively. Because these soils are highly organic, high TC and TN values were expected. Only the slough site (SJM3 OW) showed significantly higher TN and TC than any other vegetation c ommunities. In soil profiles from all sites, TP was highest in the surface and decreased in concentration with increasing depth. This is characteristic for soils that have experienced nutrient loading in their recent history. Another possible explanation for the high TP seen in surface soils could be the presence of benthic algae. As TP analysis does not exclude phosphorus associated with microbial biomass, any benthic algae in the soil will contribute to a higher final TP value. As opposed to larger plant s that need a high carbon structure for support, algae have no such need, therefore have much lower C:P ratios (Brenner et al. 2001 ; Reddy et al 1999 ) Nevertheless, the fact that surface soils at all sites were P enriched indicates to me that allochthono us sources of P are the primary culprit. At the C:N ratio s seen in this study ( 14.4 0.2 ) ammonification (mineralization of organic nitrogen to NH 4 + ) and subsequent N loss to pore/flood water can be expected during both flooded and drained conditions. Ad ditionally, this low C:N ratio indicates that

PAGE 67

67 upon draining, initial organic N mineralization will occur relatively rapidly. This is because microbial N demands (for biomass buildup) are quickly satisfied, allowing subsequent N to be released to the soil s olution (Reddy and DeLaune 2008). Interestingly, the one site that highest surface C:N ratios (and a trend with depth opposite of the other sites) also had by far the lowest pH when flooded. Both TN and TC exhibited a negative relationship with bulk densi ty, with higher nutrient levels corresponding with lower bulk densities. As soils with more organic carbon generally have lower bulk densities relative to more mineral soils, this relationship between TC and density is consistent. It was expected that soil bulk density would exhibit a strong relationship with depth, but interestingly any association was weak (Adj R 2 = 0.23, n=136). TC and TN were good predictors of bulk density (Adj R 2 of 0.29 and 0.20 respectively) while TP and TS were not. Average N:P rat ios (top 10cm) were not significantly different between the various vegetation communities in this study altho u gh TP was significantly higher in the typha site than in the other vegetation communities This could very well be due to the fact that only one core was taken from each site and sampling was only conducted once. It is generally accepted in the literature that invasive species like cattails ( Typha spp.) gain dominance in wetlands when phospho rus limitations are decreased (Lorenzen et al 2001; Wright and Reddy 2001; Reddy et al 1993 ; Liu et al. 2010 ) As the sites chosen for this study were intended to examine the chemical differences between soil with invasive and native vegetation, we expecte d to see more P limitation at the native sites (Cladium and slough/open water) compared to the sites with invasive species. Nevertheless, average N:P ratios (top 10cm) were not significantly different

PAGE 68

68 between the various vegetation communities in this stud y indicating that all surface soils were similarly P limited. While this seems contradictory, it should be noted that vegetation observed during sampling is only representative of conditions during recent years. As the SJMCA has undergone severe changes i n hydrology over the last few decades, soil deeper in the profile was most likely produced by vegetation different from that observed during this study. Thus, it is not surprising that soil chemical profiles generally appear to converge as depth increases. For comparison, the high and low TP values found in this study (1240 to 216 mg kg 1 ) are similar to impacted and normal Everglades soils respectively (Osborne et al. 2011). Lab Nutrient Flux Both TDP and TDKN fluxes increased after soils were drained, whi le DOC flux decreased. This difference can be explained in part by the C:N ratios found previously. As soil C:N was 14.4, and microbial biomass is gener ally between 8 and 12 (Cleveland and Liptzin 2007) it would be expected that upon draining, soil carbon would be lost to the atmosphere as CO 2 and nitrogen would be microbially cycled within the soil, causing the C:N ratio to decrease, approaching 10. A previous study by Olila et al. (1997) found that floc sediments drained for 6 weeks exhibited a 10 fold higher phosphorus flux to the water column after flooding relative to cores drained for only 3 weeks. Th is study demonstrates that for peat soils, a similar relationship exists; increased drainage time results in higher P fluxes to the water column. Using average nutrient flux values (after draining), nutrient loads for the entire SJMCA are 51.3, 4.2, and 27.8 metric t ons for DOC, TDP, and TDKN respectively.

PAGE 69

69 Field Nutrient Flux The willow dominated site exhibited the highest TP flux upon flooding. This is i nteresting because the willow site also had relatively low soil TP values. A similar trend was seen during the laboratory study where TDP flux increased more after draining for low TP soils than for high TP soils (seen in Figure 3 7). Higher (not significa ntly) DOC and TDKN fluxes were also seen for the willow site. These high nutrient fluxes from the willow site were not expected, but could be due to a few different factors. First, the rooting structure of the willow community in the field site was very th ick; adventitious roots were common and very dense. These root structures are produced by the plant for the purpose of oxygen introduction to the rhizosphere, which can then be used to mineralize phosphorus by microbes (Lorenzen et al. 2001) Additionally, prescribed burns are commonly used by the SJRWMD to manage the spread of willow with mixed success (Lee et al. 2005). As a result of fire, phosphorus that is normally unavailable is mineralized and can be leached into the water column. In 1991, Spencer an d Hauer observed that soluble reactive phosphorus (SRP) in streams increased between 5 and 60 times during wildfire occurrence. While this 1991 study shows that fire can mineralize P, the effects on P flux may only be relevant for short periods of time aft er the fire. Liu et al (2010) demonstrated that the first 24 hours after burning were most crucial to the fate of mineralized phosphorus. Therefore, if fire (wild or prescribed) has occurred in the salix community, it is reasonable that P flux would be hi gh. I t was expected that field sites with the highest CO 2 fluxes would produce the largest nutrient fluxes when soils flooded however, this trend was not observed. In all cases, nutrient fluxes in the field were roughly 10 times as high as those in the l ab. While it was expected that soils in the field would release more nutrients than in the lab,

PAGE 70

70 this magnitude of difference was somewhat surprising. Many factors may have led to the relatively high flux of nutrients from the field sites. First, there is s ignificantly more soil in the field; using a metal rod, the peat was found to be at least 3 4m deep at most sites (soil cores were only 15cm long). While this difference in total soil volume is certainly a concern, a study of USJRB soils by Reddy et al. (2 006) showed that carbon mineralization is highest in the top 20cm; beneath this depth, mineralization decreases significantly. Second, flora and fauna may contribute to nutrient cycling in the field, while they are excluded from the lab study. Additionally temperatures in the field are much more variable than in the laboratory. A study by Inglett et al. (2011) shows that both temperature and vegetation are significant predictors of aerobic/anaerobic respiration. Conclu ding Remarks The objective of this stu dy was to determine how water management affect s nutrient movement within the SJMCA. Hydrology does appear to have an influence on nutrient movement through and from the wetland. When soils were drained and re flooded fluxes of dissolved nitrogen and phos phorus increased significantly Additionally, s oils that were populated with invasive plants like cattails and willow generally released more nutrients (not always significantly) than soil colonized by native species like sawgrass. While both the insitu a nd laboratory studies measure similar factors, their respective utilities to management in the SJMCA are different. Figure 3 10 shows that the two studies produced somewhat disparate results. As the laboratory nutrient flux study occurred in a much more co ntrolled way (cores flooded and sampled at same time), its results should be more useful in predicting the rate of nutrient loss after water t able drawdown as a function of vegetation type The primary utility of t he insitu study is

PAGE 71

71 to show the influence of environmental factors on nutrient flux. The finding that insitu nutrient fluxes were much greater than those seen in the laboratory indicates that factors like vegetation, temperature, and macrofauna have a significant influence on th e loss of nutrients from the SJMCA. Given that nutrients fluxed from the SJMCA will be translocated downstream to the SJR, where they can contribute to HABs and fish kills, care must be taken when the water level is allowed to decrease. This study has shown that phosphorus in particular responds very strongly to water t able drawdowns, increasing from 0.021 to 0.030 g m 2 after drainage. As phosphorus is the nutrient most often linked with the occurrence of HABs and DO crashes (Chapman et al. 2001; Paerl 2007; Steidinger et al. 1998), and summer storms can rapidly flush this phosphorus downstream where it can cause ecological harm; water t able s during the summer months particularly should be kept at a level high enough to maintain anaerobi c conditions in the surface soils. This management scheme would allow for much of the flood control properties of the wetland to be maintained, while still limiting the loss of nutrients from the SJMCA.

PAGE 72

72 CHAPTER 4 SUMMARY AND CONCLUSI ONS The primary research objective s of this study were to answer the question s: 1) what are the major causes of soil subsidence, and 2) how can the current hydrologic management scheme best be modified to reduce or eliminate organic soil subsidence and allow the SJMCA to meet its requirements of: flood protection, environmental enhancement, water supply, and recreation. Multiple laboratory and insitu studies were undertaken in order to answer to these questions. This study indicates that under current management practice s, soil subsidence will continue to be a factor in the SJMCA. When the soils are allowed to drain for approximately 1 month per year or more, total subsidence (primary and secondary) will exceed sediment accumulation, causing net soil loss. Although some p revious studies have suggested that soils drained for multiple years experience more primary subsidence than secondary, this study provides data to the contrary. Here, it was shown that secondary subsidence exceeded primary subsidence for the SJMCA, a wetl and that has been chronically overdrained. While vegetative type may not be particularly critical in determing CO 2 flux rates, temperature and water depth are key to predicting carbon mobility. Both CO 2 and CH 4 fluxes demonstrated a strong relationship wi th hydrology. CO 2 flux in the field was only 417 g C m 2 yr 1 when flooded, but increased to 2595 g C m 2 yr 1 when the water t able fell below the surface. CH 4 flux was 30 gC m 2 yr 1 when flooded, but decreased to 7.2 g C m 2 yr 1 when the soil was draine d below the soil surface. During the course of the study, the water t able was beneath the soil surface for roughly 8 months out of the year. Thus, aerobic respiration of soil OC is the most important factor regulating carbon

PAGE 73

73 cycling in the wetland and sho uld be considered when manipulating the water t able in the SJMCA in order to avoid excessive soil loss. The C:N ratios found in this study indicate that the carbon quality of surface soils in the SJMCA is relatively high This implies that upon water t able drawdown, significant mineralization of SOC should be expected with concomitant oxidative carbon losses The high carbon quality of surface soils in the SJMCA could als o partially explain the unexpectedly high fraction of total subsidence caused by s econdary factors. High levels of TP in surface soils could also theoretically help explain the high level of secondary subsidence, although no relationship was found in this study between phosphorus and CO 2 flux. Conversely, the high C:N ratios seen lower in soil profiles indicate lower qualit y carbon deeper in the soil. This refractory carbon source generally yield s a lower average respiration rate relative to more labile substrates (Conant et al. 2008; Kechavrzi et al 2010) although a 2011 study by Inglett et al. indicates that both anaerobic and aerobic CO 2 fluxes are not correlated with carbon quality As seen in the insitu subsidence study, once the water t able dropped below approximately 20 cm, the rate of CO 2 flux did not change significantly This suggests that allowing the deeper soils in the SJMCA to become exposed to oxygen does not have a significant effect on secondary subsidence. The implication for management here is that the negative effect of wetland drainage is limited to the first 20 cm of water t able drainage; beneath this level, additional drainage has little to no effect on gaseous carbon loss While the gaseous loss of carbon is clearly a significant cause of soil subsidence, soil material can be lost through processes besides oxi dation. Aqueous carbon lost from

PAGE 74

74 the soil while flooded could also be considered a c ontributing factor to subsidence. This study showed that roughly 1 g C m 2 was lost from the soil for every centimeter of floodwater Although this loss of carbon is small relative to secondary subsidence, it is still a loss that is not currently assessed. This study also shows that hydrology also has an influence on how nutrients move through the wetland. When soils were drained and re flooded, fluxes of dissolved nitrogen and phosphorus increased significantly Soils that were populated with invasive plants like cattails and willow generally released more nutrients (not always significantly) than soil colonized by native species like sawgrass. Areas c olonized by cattails also had higher levels of soil TP, possibly allowing for higher phosphorus fluxes when flooded. Given the current pattern in the SJMCA of drainage during the warm summer months and flooding during the winter, persistent soil loss will continue. As drained CO 2 flu x was found to be 6 times higher than flooded flux, and thes e rates are temperature dependant, draining of the SJMCA during the summer is the worst possible case for carbon stability within the wetland. Additionally, e xtreme se asonal nutrient fluxes can occur, leading to the algal blooms and fish kills that bli ght the St Johns River annually. In light of both of these factors, soil carbon loss concomitant with nutrient flux, the soils in the SJMCA must be better managed to prote citizens and environment. It would seem that sites with high nutrient loss upon flooding would also exhibit high CO 2 fluxes because microbial decomposition rates influence both factors. Nevertheless, this relation ship between nutrient flux and CO 2 flux was not seen. In both laboratory an insitu studies, TDP, TDKN, and DOC fluxes predicted less than 5% of the

PAGE 75

75 variability in average field CO 2 fluxes (R 2 Adj < 0.05 for all comparisons ) T his indicates that the factors controlling nutrient losses in the SJMCA are different than those governing CO 2 flux. While water t able drawdowns during the summer may be necessary to prevent flooding in the surrounding urban areas, the unintended consequences of oxygenati ng the soil are severe. Over time, the soil that retards the flow of water during high rainfall events will be lost due to primary and secondary subsidence. Figure 4 1 Sketch of current and proposed water t able regime for SJMCA Therefore, if t he water t able can be kept high enough to maintain anaerobic conditions during the warm summer months, oxidative carbon loss (as well as subsequent nutrient mineralization) could be diminished substantially. While keeping the water level at the soil surfac e would be ideal, an additional drawdown of 10 cm would only have a small impact on the loss of soil carbon through oxidation. Thus, the SJMCA

PAGE 76

76 should not be drained below 10 cm. Average yearly water t able s would be kept the same, but would be lower in the winter and higher in the summer ( Figure 4 1). This would not only help to reduce soil and nutrient losses, but would also go some way to restore natural wetland hydrology to the SJMCA.

PAGE 77

77 APPENDIX ADDITIONAL FIGURE S

PAGE 78

78 Figure A 1 Field DO (mg L 1 ), DOC (mg L 1 ), NH 4 (mg L 1 ), TDKN (mg L 1 ), TP (mg L 1 ) and pH. n=3

PAGE 79

79 Figure A 2 Depth profiles for TC (%), TN (%), TP (ug kg 1 ), and TS (ug kg 1 ) by site. n=1

PAGE 80

80 Figure A 3 Major nutrient fluxes (mg/L n=3 ) and pH (n=1) for all field sites

PAGE 81

81 LIST OF REFERENCES Altor AE, Mitsch WJ (2008) Pulsing hydrology, Methane emissions and Carbon Dioxide fluxes in created marshes: A 2 year ecosystem study. Society 28:423 438. Armentano ATV, Menges ES, Jo urnal S, Sep N (1986) Patterns of Change in the Carbon Balance of Organic Soil Wetlands of the Temperate Zone Journal of Ecology 74:755 774 Berglund Berglund K (2011) Influence of water t able level and soil properties on emissions of greenhouse gases f rom cultivated peat soil. Soil Biology and Biochemistry 43:92 A 3 31 Bowden WB (1987) The biogeochemistry of nitrogen in freshwater wetlands. Biogeochemistry 4:313 348 Brenner M, Schelske CL, Keenan LW (2001) Historical rates of sediment and nutrient accumul ation in marshes of the Upper S t. Johns River Basin, Florida USA. Jour nal of Paleolimnology 2:241 257 Cadenhead M (1997) Eutrophicati on: Causes and Effects. In J.R. Maher (ed.) Health of the river: A survey of the factors affecting water quality in the Lo wer St. Johns River basin. 10 14 Chapman AD, Williams CD, Burns JW, et al. (2001) Assessm ent of Cyanotoxins in Reservoirs and Rivers. 1 57 the micr obial biomass? Biogeochemistry 85:235 252 Conant RT, Drijber R a., Haddix ML, et al. (2008) Sensitivity of organic matter decomposition to warming varies with its quality. Global Change Biology 14:868 877 atter in a Wetland Receiving Hypereutrophic Lake Water: H. Role of Inorganic Electron Acceptors in Nutrient Release. Journal of En vironmental Quality 943:937 943 Drexler JZ, de Fontaine CS, Deverel SJ (2009) The Legacy of Wetland Drainage on the Remaining Peat in the Sacramento San Joaquin Delta, California, USA. Wetlands 29:372 386 Ellis T, Hill PW, Fenner N, et al. (2009) The interactive effects of elevated carbon dioxide and water t able draw down on carbon cycling in a Welsh ombrotrophic bog. Ecologica l Engineering 35:978 986

PAGE 82

82 Freeman C, Liska G, Ostle NJ, et al. (1996) Microbial activity and enzymic decomposition processes following peatland water t able drawdown. Plant and Soil 180:121 127 Hanson G, Groffman P, Gold A (1993) Denitrification in Riparian Wetlands receiving High and Low Groundwater Nitrate Inputs. Journal of Environmental Quality 23:917 922 Holden J, Wallage ZE, Lane SN, McDonald AT (2011) Water Table dynamics in undisturbed, drained and restored blanket peat. Journal of Hydrology 402:10 3 9 4 Inglett KS, Inglett PW, Reddy KR, Osborne TZ (2011) Temperature s ensitivity of greenhouse gas production in wetland soils of different vegetation. Biogeochemistry Jauhiainen J, Limin S, Silvennoinen H, Vasander H (2008) Carbon dioxide and methane fluxes in drained tropical peat before and after hydrological restoration. E cology 89:3503 14 Ktterer T, Reichstein M, Andrn O, Lomander A (1998) Temperature dependence of organic matter decomposition: a critical review using literature data analyzed with different models. Biology and Fertility of Soils 27:258 262 Kechavarzi C, Dawson Q, Bartlett M, Leeds Harrison PB (2010) The role of soil moisture, temperature and nutrient amendment on CO2 efflux from agricultural peat soil microcosms. Geoderma 154:203 21 0 King GM, Roslev P, Skovgaard H (1990) Distribution and rate of methane oxidation in sediments of the F lorida everglades. Applied and E nviro nmental Microbiology 56:2902 11 Kirschbaum M (1995) The temperature dependence of soil organic matter decomposition and the effect of global warming on soil organic C storage. Soil Biology and Biochemistry 27:753 760 Lafleur PM, Moore TR, Roulet NT, Frolking S (2005) Ecosystem Respiration in a Cool Temperate Bog Depends on Peat Temperature But Not Water Table Ecosyst ems 8:619 629 Lee MAB, Snyder KL, Valentine Darby P, et al. (2005) Dormant Season Prescribed Fire as a Management Tool for the Control of Salix caroliniana Michx. in a Floodplain Marsh. Wetlands Ecology and Management 13:479 487 Leirs M, Trasar Cepeda C, Seoane S, Gil Sotres F (1999) Dependence of mineralization of soil organic matter on temperature and moisture. Soil Biology and Biochemistry 31:327 335

PAGE 83

83 Liu GD, Gu B, Miao SL, et al. (2010) Phosphorus Release from Ash and Remaining Tissues of Two Wetland Sp ecies after a Prescribed Fire. Journal of Environment Quality 39:1585 Lorenzen B, Brix H, Mend elssohn I, et al. (2001) Growth biomass allocation and nutrient use efficiency in Cladium jamaicense and Typha domingensis as affected by phosphorus and oxygen a vailability. Aquatic Botany 70:117 133 Malecki LM, White JR, Reddy KR (2004) Nitrogen and Phosphorus Flux rates from Sediment in the Lower St. Johns River Estuary. Journal of Environmental Quality 1545 1555. Maher, J. Water Quality & the St. Johns River. D epartment of Environmental Protection. 1 9 Mclatchey GP, Reddy KR (1998) Regulation of Organic Matter Decomposition and Nutrient Release in a Wetland Soil. Journal of Env ironmental Quality 27:1268 1274 Miller RL, Fujii R (2009) Plant community, primary pro ductivity, and environmental conditions following wetland re establishment in the Sacramento San Joaquin Delta, California. Wetlands Ecology and Management 18:1 16 Morris DR, Glaz B, Daroub SH (2004) Organic Soil Oxidation Potential Due To Periodic Flood a nd Drainage Depth Under Sugarcane. Soil Science 169:600 608 Automated Colorimetry. US Environmental Protection Agency 1 15 Olila O., Reddy K., Stites D. (1997) Influence of draining on soil p hosphorus forms and distribution in a constructed wetland. Ecological Engineering 9:157 169 Osborne TZ, Newman S, Scheldt DJ, Kalla PI, Bruland GL, Cohen MJ, Scinto LJ, Ellis JR. Landscape Patterns of Significant Soil Nutrients and Contaminants in the Grea ter Everglades Ecosystem: Past, Present, and Future. Critical Reviews in Environmental Science and Technology 41:121 148 Overpeck JT, Meehl G a, Bony S, Easterling DR (2011) Climate data challenges in the 21st century. Science (New York, NY) 331:700 2 Paer l H (2007) Nutrient and other environmental controls of harmful cyanobacterial blooms along the freshwa ter marine continuum. 217 233 Parent LE, Millette JA, Mehuys GR (1977) Subsidence and Erosion of a Histosol. Soil Science Societ y of America Journal 46 :404 408 Price JS (2003) Role and character of seasonal peat soil deformation on the hydrology of undisturbed and cutover peatlands. Water Resources Research 39:1 10

PAGE 84

84 Reddy KR, DeLaune RD (2008) Biogeochemistry of Wetlands. CRC Press Reddy KR, Delaune RD, K och MS, Debusk WF (1993) Long Term Nutrient Accumulation Rates in the Everglades. Soil Science Socie ty of America Journal 1147 1155 istry in Subtropical Ecosystems Reddy KR, Osborne T, Inglett K, Corstanje R (2006) Influence of Water Levels on Subsidence of Organic Soils in the Up per St. Johns River Basi. 1 119 Rojstaczer S, Deverel SJ (1996) Land subsidence in drained histosols and highly organic mineral soils of Califor nia. Soil Science Socie ty of A merica Journal, 59: 1162 1167 Schedlbauer JL, Oberbauer SF, Starr G, Jimenez KL (2010) Seasonal differences in the CO2 exchange of a short hydroperiod Florida Everglades marsh. Agricultural and Forest Meteorology 150:994 1006 Schipper LA, Mcleod M, S cott N, et al. (2002) Subsidence rates and carbon loss in peat soils following conversion to pasture in the Waikato Region New Zealand. Soil Science 91 93 Schothorst C (1977) Subsidence of Low Moor Peat Soils in the Western Netherlands. Geoderma 17: 265 291 underlying wetland methane fluxes. Biogeochemistry 23 51 Spencer CN, and Hauer FR. Phosphorus and nitrogen dynamics in streams during a wildfire. Journal of the Nort h American Benthological Society 10:24 30 Stephens J (1956) Subsidence of Organic Soils in the Florida Everglades. Soil Science Society of America Journal 2 5 Steidinger KA, Landsberg JH, Truby EW, Roberts BS (1998) First report of Gymnodinium Pulchellum ( Dinophyceae) in North America and associated fish kills in the Indian River Florida. J Phycol 437:431 437 Twinch AJ (1987) Phosphate exchange characteristics of wet and dried sediment samples from a hypertrophic reservoir: Implications for the measurement s of sediment phosphorus status. Water Research 21:1225 1230 Vepraskas JM, Ewing MJ (2006) Estimating Primary and Secondary Subsidence in an Organic Soil 15, 20, and 30 Years After Drainage. Wetlands 26:119 130

PAGE 85

85 White JR, Reddy KR (1997) Influence of Select ed Inorganic Electron Acceptors on Organic Nitrogen Mineralization in Everglades Soils. Soil Science Soc iety of America Journal 558 563 Wright AL, Reddy KR (2001) Phosphorus Loading Effects on Extracellular Enzyme Activity in Everglades Wetland Soils. Soil Science Society of America Journal 65:588 595 Wsten JH., Ismail A., van Wijk AL. (1997) Peat subsidence and its practical implications: a case study in Malaysia. Geoderma 78:25 36

PAGE 86

86 BIOGRAPHICAL SKETCH Bryce Van Dam was born in Tallahassee, FL in 1987. After finishing high school Bryce enrolled at the University of West Florida in 2005 I n 2010, he graduated with a Bachelor of Science in environmental science/natural s cie nce, and a certificate in g eograp hic information s cience. In 2012 Bryce completed a Master of Science degree under Dr Todd Osborne in the Soil and Water Science Department with a focus on wetland sciences at the University of Florida