<%BANNER%>

Litter Decomposition and Phosphorus Release in Okeechobee Isolated Wetlands


PAGE 1

LITTER DECOMPOSITION AND PHOSPH ORUS RELEASE IN OKEECHOBEE ISOLATED WETLANDS By NATALIE BALCER A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2006

PAGE 2

Copyright 2006 by Natalie Balcer

PAGE 3

To my parents who have supported me throughout this entire journey.

PAGE 4

iv ACKNOWLEDGMENTS I would like to thank Dr. Mark Clark for his insight, guidance, and support. I thank him for inspiring me with his wealth of knowledge on scientific issues. Thanks go to my other committee members (Dr. Ramesh Reddy and Dr. Patrick Bohlen) for their efforts and advice. Special thanks go to Dr. Ed Dunne for his advice and help in the field. Many other students and staff in the Wetland Biogeochemistry Laboratory have provided additional insight and assistance during the past 2 years. Funding was provided by the South Florida Water Management Distri ct, Florida Department of Agriculture and Consumer Services, and the Florida Departme nt of Environmental Protection. Lastly, thanks go to Darren Cole for continuously s upporting my decision to come to Gainesville to further my education.

PAGE 5

v TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES............................................................................................................vii LIST OF FIGURES...........................................................................................................ix ABSTRACT......................................................................................................................x ii CHAPTER 1 INTRODUCTION AND SITE DESCRIPTION..........................................................1 Introduction................................................................................................................... 1 Rationale for Research..................................................................................................5 Site Description............................................................................................................8 2 PLANT TISSUE CHARACTERIZATION...............................................................14 Introduction.................................................................................................................14 Materials and Methods...............................................................................................16 Results........................................................................................................................ .18 Discussion...................................................................................................................24 3 SHORT TERM PHOS PHORUS LEACHING...........................................................27 Introduction.................................................................................................................27 Materials and Methods...............................................................................................29 Results........................................................................................................................ .32 Discussion...................................................................................................................45 4 LITTER DECOMPOSITION AND LONG TERM PHOSPHORUS RELEASE.....52 Introduction.................................................................................................................52 Materials and Methods...............................................................................................56 Field Methods......................................................................................................56 Laboratory Methods............................................................................................61 Results........................................................................................................................ .62 Discussion...................................................................................................................73 5 SYNTHESIS AND CONCLUSIONS........................................................................81

PAGE 6

vi LIST OF REFERENCES...................................................................................................87 BIOGRAPHICAL SKETCH.............................................................................................93

PAGE 7

vii LIST OF TABLES Table page 2-1. Neutral detergent fibe r and acid detergent fiber fractionation methods (Rowland and Roberts, 1999)...................................................................................................18 2-2. The C/N and C/P ratios for live and sene sced tissue. Values represent mean ( 1 standard deviation)...................................................................................................22 3-1. Cumulative P release (mg P/g litter) and percent tissue P released from four species over a 17 day period, under aer obic and anaerobic conditions. Values represent mean ( 1 standard deviation). Lowercase letters indicate significant differences over the entire 17 day study between species, water treatment, and redox conditions with a p-value of .05.....................................................................40 3-2. Mean concentrations ( 1 standard deviation) of water column P for various species under aerobic and anaerobic conditions after 17 days. Negative DOP values are due to high standard errors and represent virt ually no measurable DOP in the water column.........................................................................................41 3-3. Mean concentrations ( 1 standard de viation) of water co lumn TKN for various species under aerobic and anaerobic conditions after 17 days. Letters indicate significant differences between species and redox condition..................................41 3-4. The P-Values associated with initial nutrient parameters of senesced tissue to estimate the best predictor of P flux fo r site water under aerobic conditions on Day 17......................................................................................................................43 3-5. The R-Square values from %P and cu mulative P release correlation in site water under aerobic and an aerobic conditions...................................................................44 3-6. The P-Values associated with initial nut rient parameters of live tissue to estimate the best predictor of P flux for site water under aerobic conditions on Day 17.......44 4-1. Species decomposition in each zone of litterbag deployment after 12 months. Values represent mean ( 1 standard deviation ). .....................................................66 4-2. Significance and R2 values of the relationship betw een the change in P content and initial senesced substrate quality characteristics for each sampling period.......67 4-3. Significance and R2 values of the relationship betw een the change in P content and initial live substrate quality ch aracteristics for each sampling period...............69

PAGE 8

viii 5-1. Variation in short and moderate-t erm P assimilation or release from 1m 2 of litter over 2 months for the 4 domina nt species investigated............................................83

PAGE 9

ix LIST OF FIGURES Figure page 1-1. Hydrological modifica tions of Lake Okeechobee and its watershed have resulted in increased channelization, transportation of contaminants to the lake and loss of wetlands (FDEP, 2001)..........................................................................................3 1-2. The P retention mechanism in a wetland, the potential fate of P initially assimilated in macrophyte plant tissue and different P fractions such as Dissolved Organic Phosphorus (DOP), Pa rticulate Organic Phosphorus (POP), and Dissolved Inorganic Phosphorus (DIP)...............................................................6 1-3. Overland view of the study site including th ree historically is olated wetlands on Larson Dixie Ranch....................................................................................................9 1-5. Average hydroperiod for center, edge, and upland zones of seasonally isolated wetlands on Larson-Dixie Ranch.............................................................................11 1-6. Species occurrence in seasonally is olated wetlands located in the Okeechobee Basin. A) Wetland Centers, B) We tland Edge, C) Surrounding Uplands...............12 1-7. Soil total phosphorus storage in wetland zones and surrounding upland..................13 2-1. Extraction sequence used to dete rmine carbon quality of vegetation.......................17 2-2. The C/N and N/P ratios of live tissu e for species surveyed. Values represent mean ( 1 standard deviation)..................................................................................19 2-3. Mean C/P ratios in live tissue of speci es surveyed. Values represent mean ( 1 standard deviation)...................................................................................................20 2-4. Fiber quality (NDF, ADF, SADF a nd Residual Fiber) percentages for species surveyed...................................................................................................................20 2-5. Residual fiber fractions for live tissu e. Values represent mean ( 1 standard deviation)..................................................................................................................21 2-6. Nutrient ratios of C:N in senesced tissue of dominant species. Values represent mean ( 1 standard deviation)..................................................................................21 2-7. Comparison of phosphorus values fo r live and senesced vegetation. Values represent mean ( 1 standard deviation)..................................................................22

PAGE 10

x 2-8. Neutral detergent fibe r content contained in the se nesced tissue of 4 dominant species. Values represent mean ( 1 standard deviation)........................................23 2-9. Residual fiber content contained in th e senesced tissue of 4 dominant species. Values represent mean ( 1 standard deviation)......................................................23 3-1. The P leaching study. A) Overview of the experimental setup and aerobic and anaerobic treatments B) Individual fluxing container and tubing bubbling ambient air in the water column through the hypodermic needle............................31 3-2. Phosphorus leaching rates of 4 sene sced species averaged across all 3 water treatments A) aerobic B) anaerobic c onditions. Values represent mean ( standard deviation)...................................................................................................34 3-3. Litter phosphorus release rate for P. hemitomon litter under (a) aerobic and (b) anaerobic conditions with 3 different wate r treatments. Values represent mean ( 1 standard deviation)............................................................................................36 3-4. Litter phosphorus release rate for P. hydropiperoides litter under a) aerobic and b) anaerobic conditions with 3 water treatments. Values represent mean ( 1 standard deviation)...................................................................................................37 3-5. Phosphorus release rate for P. notatum litter under a) aerobic and b) anaerobic conditions with 3 water treatments. Va lues represent mean ( 1 standard deviation)..................................................................................................................38 3-6. Phosphorus release rate for J. effusus litter under a) aerobic and b) anaerobic conditions with 3 water treatments. Va lues represent mean ( 1 standard deviation)..................................................................................................................39 3-7. Percent mass loss from the vegetati on over a 17 day period. Values represent mean 1 standard deviation.....................................................................................42 3-8. Bivariate fit of cumulative P flux by initial senesced tissue P content. Correlation is for site water treatm ent under aerobic conditions on day 17.............43 4-1. Litterbag distribution and deployment locations in 4 hydrological zones within the wetland...............................................................................................................58 4-2. Hydrological information for the s easonally isolated wetlands on Larson Dixie Ranch. A) The yearly stage information in meters from April 1, 2004 – March 10, 2006 B) the average hydroperiod for the center, edge, and upland zones during November, March, and July..........................................................................59 4-3. Aerial view of the three historic ally isolated wetlands on Larson Dixie Ranch where the litterbags were deployed..........................................................................59

PAGE 11

xi 4-4. Litterbag experiment s howing A) litterbags and netti ng perpendicular to transect in upland, B) litterbags attached to soil and covered by nylon netting as a precaution against cattle, and C) a cl ose up of litterbag filled with P. notatum. Pictures D, E, and F illustrate the litter bag collection after 8 months of exposure showing D) P. notatum growing in and through the li tterbag and E) the large amount of vegetation covering the bags in the upland and F) in the transitional zone..........................................................................................................................6 0 4-5. Litter decomposition of all species in all hydrologic zone among the 3 different wetlands, Larson East (LE), Larson West (LW), and Lars on South (LS). Values represent mean 1 standard deviation.....................................................................63 4-6. Decomposition in 4 wetland zones ove r a 12 month period. Values represent mean 1 standard deviation.....................................................................................64 4-7. Average litter decomposition of each species over a 12 month period. Values represent mean 1 standard deviation.....................................................................65 4-8. Change in % P in the 4 hydrologi cal zones over a 12 month period. Values represent mean 1 standard deviation.....................................................................66 4-9. Change in % P in the 4 dominant species over a 12 month period. Values represent mean 1 standard deviation.....................................................................67 4-10. Correlation between initial NDF fracti on in the senesced tissue and P loss or gain after 12 months.................................................................................................68 4-11. Correlation between initial C:P in th e live tissue and P loss or gain after 12 months......................................................................................................................70 4-12. Change in litter %N (all species combined) among different wetland hydrologic zones over time. Values represen t mean 1 standard deviation.............................70 4-13. Change in litter %N among species over time. Values represent mean 1 standard deviation....................................................................................................71 4-14. Comparison of residual fiber conten t of initial and 12 month exposed litter among four species tested. Values represent mean ( 1 standard deviation)..........72 4-15. Residual fiber content of species in each hydrologic zone after 12 months. Values represent mean ( 1 standard deviation)......................................................73 4-16. Litterbags collected after 12 mont hs from A) wetland center, which were approximately 20-30 cm underneath the soil surface and B) wetland edge, which were on top of the soil surface.................................................................................77

PAGE 12

xii Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science LITTER DECOMPOSITION AND PHOSPH ORUS RELEASE IN OKEECHOBEE ISOLATED WETLANDS By Natalie Balcer December 2006 Chair: Mark Clark Major Department: Soil and Water Science Phosphorous (P) is the leading nutrient that contributes to degraded water quality and the eutrophication of Lake Okeechobee. Isolated wetlands within the Lake’s watershed are potentially an important phosphor us sink reducing P con centrations in the water column before surface runoff reaches the lake. Hydrologic restoration of ditched and partially drained wetlands is hypothesi zed to enhance P assimilative capacity and therefore further reduce P contaminant load s. Phosphorus dynamics in decomposing litter contributes significantly to overall P dynamics in these heavily vegetated systems. Our study focused on shortand long-t erm P release from aboveground biomass after vegetative senescence. Our first objective of this proj ect was to characterize fiber quality and nutrient content ch aracterization of live and se nesced tissue of different wetland species. By quantifying the amount of TN TC, and TP in the litter, as well as running fiber quality characteristics, some pred ictions were made regarding the relative lability and recalcitrance of the detrital material. Ou r second objective involved a

PAGE 13

xiii laboratory experiment to investigate the shortterm release of P in response to inundation under aerobic and anaerobic condi tions from the senesced tissu e of 4 species common to these wetlands: Paspalum notatum, Polygonum hydropiperoides, Juncus effusus and Panicum hemitomon. Our third objective was to invest igate the influence of hydrological zones on the decomposition rate and long-term P release of senesced vegetation. This was accomplished using litterbags deployed in the field. Fiber quality and nutrient content varied significantly among the four plant species and played a significant role in P release and decomposition rate. The P leaching study showed that senesced, air-dried vegetati on had the potential to release a substantial amount of P within 72 hours. There were statisticaly significant differences in the leaching rates among species, with th e greatest P leaching occurring in P. hydropiperiodes and the least amount of leaching in P. hemitomon In addition, short term P leaching was strongly correlated with th e initial P content in the senesced tissue. Species type also insignificantly influenced P release and decomposition in the field over a 12-month period. Paspalum notatum had a significantly higher decomposition rate after 12 months compared to the other three species. The C:P in the senesced tissue predicted phosphorus release up to a 2-month time period. Our results provide further insight into th e potential implications of vegetative species changes that occur in response to hydrological restorati on, or other wetlandrelated Best Management Practices (BMPs) to reduce P loading into the waterways of the Okeechobee Basin.

PAGE 14

1 CHAPTER 1 INTRODUCTION AND SITE DESCRIPTION Introduction Wetlands have the ability to store an d transform nutrients from surrounding uplands (Richardson, 1985; Reddy et al., 1995; Jordan, 2003). Wetlands presently make up 17% of the Okeechobee Basin; however this percentage has decreased in recent times due to ditching and drainage in an effort to improve pastures for dairy and beef cattle operations (Haan, 1995; Tiner, 2003). Curre ntly, greater than 50% of the wetlands located in the basin have been at least pa rtially ditched and drai ned to increase the acreage of improved pastures (McKee, 2005). Wetlands already provide a valuable role in improving the water quality of Lake Okeech obee by storing water in the watershed and sequestering phosphorous (P), the leading nut rient causing eutrophication of the lake (Davis and Marshall, 1975; Fede rico et al., 1981). However, it is thought that hydrologic restoration of those wetlands that have been ditched and drained may provide a significant increase in P storage and will assist in efforts to address P TMDL targets set for the basin. Lake Okeechobee is the 2nd largest lake contained entirely within the United States, covering an area of 1890 km2 (Reddy et al., 1995). It was formed by a buildup of peat near the southern edge of the lake th at restricted outflow (Brezonik and Engstrom, 1998). Lake Okeechobee provides many values including flood control; water supply; a habitat for migratory birds, fish, and other species; as well as a multimillion dollar recreational and commercial fish ing industry. It also serves as the head waters for the

PAGE 15

2 Everglades. Consequently, the water quality of the lake has the potential to impact numerous downstream communities (Steinman, 2003). Over the past 100 years, Lake Okeec hobee has had numerous anthropogenic alterations and impacts. After several deva stating hurricanes when the lake overflowed its natural boundaries, a dyke was built around the perimeter, establishing total control of water entering and leaving the la ke. Water-level regulations have been set to maintain the integrity of the dyke surrounding the lake which provides flood protection but has altered the wetland communities that lie with in the dyke perimeter. Rivers and many tributaries were channelized, allowing adjacent wetland areas to be drained and stormwater to be moved rapidly off the la ndscape. The loss of historic sinuous flow along many of the rivers and st reams, like the Kissimmee Rive r, has resulted in a loss of their ability to treat water; and instead, the de graded water is discharged directly into the lake (Figure 1-1). Presently agriculture is the primary land use in the Okeechobee basin. The watershed is dominated by beef ranches and dairy operations (Flaig et al., 1995). Land use has impacted the water quality in the Okeechobee Basin, causing an increased nutrient load to Lake Okeechobee. The incr eased contaminant load has resulted in a decline in species diversity, lower amounts of available oxygen, and reduced clarity in the Lake. These changes have caused a sh ift from aquatic vegetation to phytoplankton, which has reduced the assimilative capacity of the lake and made the sediments less stable, and caused an increase in the occu rrence of cyanobacteria blooms (South Florida Water Management District (SFWMD), 1993; Brezonik and Engstrom, 1998; Carpenter et al., 1998; Havens and Shelske, 2001).

PAGE 16

3 Phosphorous is the nutrient that contribut es most significantly to degraded water quality and eutrophication (Davis and Marshal, 1975; Federico et al., 1981; Carpenter et al., 1998). Most of the phosphorus enters the lake from non-poi nt sources in agricultural areas (Anderson and Flaig, 1995). During heavy rainfall, P from fertilizers and wastes are incorporated into runoff from the surr ounding agricultural areas, which then drains into Lake Okeechobee. Dairies have been f ound to produce the most P load per unit area. They make up 27% of the land in the Okeec hobee Basin, yet are res ponsible for 49% of the P load to the lake (Flaig et al., 1995; Hiscoc k et al., 2003). Figure 1-1. Hydrological m odifications of Lake Okeechobee and its watershed have resulted in increased channelization, transportation of contaminants to the lake and loss of wetlands (FDEP, 2001).

PAGE 17

4 The second highest P load comes from im proved pastures used for beef cattle production. Cattle ranches make up 33% of th e landscape and are responsible for 51% of the P entering from north of the lake (Flaig et al., 1995; Hi scock et al., 2003). Decreasing water quality in Lake Okeechobee has become a big concern in Florida primarily because it is a drinking-wa ter source for communities in the area, and those populations are increasing (Anderson a nd Flaig, 1995). Best Management Practices (BMPs) such as the use of wetlands to store and retain P, planting vegetative species that increase the P uptake, fenc ing cattle from water bodies, and capturing and recycling manure have prevented an increase in the P load in the last decade (B ottcher et al., 1995; Flaig et al., 1995). Even with the use of BMPs, the P content in the lake sediments is still high from historic agriculture runoff and back-pumping of drainage water in the Everglades Agricultural Area (EAA). Thes e conditions result in a high internal P loading, regardless of new external loading of P from the watershed. In 1987, the Lake Okeechobee Surface Wa ter Improvement and Management Plan (SWIM) was initiated (SFWMD, 1993). The goal was to improve water quality of the lake using a watershed management approa ch. The SWIM Plan identified four of the 41 basins around the lake as “priority basins,” because they were contributing 35% of the P load while occupying only 12% of the la nd area (FDEP, 2001). Lake Okeechobee has also been listed as an impaired water body by the Clean Water Act (CWA) section 303 (d) and therefore a Total Maximum Daily Load or TMDL that limits the amount of P the lake can receive has been established. Between 1995 and 2000, the lake received an average of 640 tons of P per year (FDEP, 2001). In 2002, the P load meas ured into Lake Okeechobee was 543 tons. The

PAGE 18

5 TMDL target P load is 140 tons per year with a concentration of 40ppb P in the pelagic areas of the lake by 2015. To reach this goal an increased effort in use of BMPs and Best Available Technology (BATs) in the Okeechobee Basin (SFWMD et al., 2004a) is underway. Rationale for Research BMPs are needed to reduce the use of P and increase P retention on agricultural landscapes. Wetlands have the ability to store and transform nutrients from the surrounding upland due to high organic matter co ntent, the presence of vegetation, and low oxygen availability, but they may also f unction as nutrient sources depending on the physical and chemical characteristics of the water column and the soil (Richardson, 1985; Reddy et al., 1995; Jordan, 2003). Nutrients a nd contaminants can become sorbed to the soil or organic matter, immobilized by plants and microbes, and mineralized by microbial processes. This degree to which P can be reduced depends on the land use, soil type, retention time, slope, hydraulic connectivity, an d vegetation type of the area (Raisin and Mitchell 1995). The large number of wetlands within the f our priority basins may have a potential to significantly reduce P load if drainage pathways from ditched wetlands are blocked and hydrologically restored to th eir historically isolated state. Restoration would increase the hydroperiod, thereby increasi ng the wetland area and the poten tial to retain P by plant immobilization, soil adsorption and reduced organic matter decomposition rates. Macrophytes have the ability to immobili ze large amounts of nutrients from the water column; however this should be t hought of as only a temporary storage for nutrients (Fig. 1-2). It is es timated that 35-80% of the total P contained in plant biomass is eventually lost to the water column, or transported out of the system after senescing,

PAGE 19

6 and the processes influencing nutrient rel ease rates are poorly unde rstood (Richardson, 1989; Reddy et al., 1995). Because a high perc entage of phosphorus is assimilated and potentially released during the decompositi on process, it is important to understand factors that might regulate the rate and total amount of P re leased (Jordan and Whigham, 1989). Figure 1-2. The P retention mechanism in a wetland, the potential fate of P initially assimilated in macrophyte plant tissue and different P fractions such as Dissolved Organic Phosphorus (DOP), Particulate Organic Phosphorus (POP), and Dissolved Inorganic Phosphorus (DIP). Wetlands are carbon based systems in whic h most of the energy driving the food web is derived from litter turnover (Turner, 1993; Corstanje, et al., 2006). Because the rate of organic matter turnover drives the system, it is importa nt to understand the fate of detrital matter in the wetland, as well as how much P is released from litter during short and long term decomposition. Numerous st udies have shown that carbon quality, primarily the fraction of recalcit rant portions of the plant, as well as the ratio of available Litterfall POP POP DOP DIP DOP DIP A dsorbed P OUTFLOW INFLOW Peat Accretion PIP Fe, Al, Ca bound P P leaching / mineralization

PAGE 20

7 N to carbon determine the rate of litter breakdown (Melill o, 1982; Berg, 1998; Villar, 2001). Nutrient loss due to leaching is not as wi dely investigated but has been shown to be a significant nutri ent and carbon source (Turner, 1993). Studies have determined that up to 80% of N and P contained in the litter as well as up to 80% of the litter biomass can be lost within 2 months after senescence due to short term leaching and mineralization in a tidal freshwater marsh (Simpson et al., 1978; Qiu, 2005). Another study preformed by C. Boyd (1971) concluded that the rate of nutrient loss was greatest during the first 4 months after senescence. Th e study reported that 50% of N and P were released and 60% of the biomass ( Juncus effusus) decomposed during the first 4 months in the field while only an additional 10% of the remaining mass wa s lost during the remaining 8 months. It is apparent however, that the species t ype or environmental conditions under which leaching and decomposition occur has a large influence on decomposition rates (Villar et al., 2001). There is a wide range of P thought to be released through leaching and mineralization processes. The rate of P loss is determined by a range of factors including; the decomposition rate of the litter, the species of vegetation, as well as factors such as temperature, moisture, nutrien ts, pH, carbon quality, and the microbial community which influence decomposition (M elillo, 1982; Benner, 1985). It has been determined that there is a significant di fference between the amount of phosphorus assimilation in different species of vegetati on (McJanet et al., 1995), but the amount and rate of P release of different wetland species after vegetative senescence remains unknown. If a species can be found that has si milar uptake rates of P, yet releases a

PAGE 21

8 significantly lower amount of P from its biom ass, that species coul d be recommended to ranchers for the purpose of P retention in orde r to decrease the amount of P release after senescence. Site Description This study was conducted on three historica lly isolated wetlands located within the same pasture on the Larson Dixie Ranch (N 0270 20.966’, W 0800 56.465’) (Fig. 1-3). Larson Dixie Ranch is a cow-calf operation within one of the f our priority basins (S-154) of the Lake Okeechobee Watershed (Fig. 14). Cow-calf operations make up roughly 48% of the agricultural land within the Ok eechobee Basin. Isolated wetlands cover approximately 12,000 ha in the four priority sub-basins and 400,000 ha within the entire Okeechobee watershed (Reddy et al., 1995). Pr esently about 50% of the historically isolated wetlands within the basin are ditche d to enhance drainage and increase the area of improved pastures. Hydrological conn ections between wetl ands on the landscape allows surface water to be transported to the lake, and results in a large P load and reduced water quality to Lake Okeechob ee during the wet season. Often, wetlands located within the Okeechobee Basin are seas onally flooded from early June to January, and remain relatively dry during the late winter and spring months. The average hydroperiod for the wetland center, edge, a nd upland zones ranged from 100-200 days / y, 5-55 days / y, and 0-15 days / y respectively (Fig. 1-5). The average area of each wetland in this study was 2.64 ha. Every wetland was drained by one ditch that drained into a larg er drainage swale and transported the water offsite. The wetlands were completely surrounded by grazing pastures that were dominated by Paspalum notatum Flugge (Bogdan) (Bahia grass), while the wetlands

PAGE 22

9 themselves were dominated by emergent marsh vegetation consisting of Juncus effusus L. Panicum spp., Polygonum hydropiperoides Michx, and Pontedaria cordata var. lancifolia (Muhl.) Torr (Fig. 1-6). Figure 1-3. Overland view of the study site in cluding three historically isolated wetlands on Larson Dixie Ranch The soils within the Okeechobee Basin are primarily poorly drained, sandy, Spodosols that have limited P retention w ithin the upper horizons and a water table generally less than 20 cm from the soil surface (Anderson and Flaig, 1995; Haan, 1995). The P content was the greatest in center soils compared to the edge and upland soils of the seasonally isolated wetlands on Larson Dixi e Ranch (Fig. 1-7). The isolated wetlands Larson South (LS) Larson East (LE) Larson West (LW)

PAGE 23

10 on the study site are flooded throughout the ma jority of the year with an occasional dry period from late March to early June. Soil on the Larson Dixie ranch is classified as Siliceous, hyperthermic Spodic, Psammaquents (B asinger series), which is a deep, poorly drained, rapidly permeable soil formed from sandy marine sediments (Anderson and Flaig, 1995). Figure 1-4. Location of Larson Dixie Ranch wi thin the 4 priority basins (McKee, K.A. 2005. Predicting Phosphorus Storage in Historically Isolated Wetlands within the Lake Okeechobee Priority Basins. Masters Thesis. University of Florida. Gainesville, FL.) The soils within the Okeechobee Basin are primarily poorly drained, sandy, Spodosols that have limited P retention w ithin the upper horizons and a water table generally less than 20 cm from the soil surface (Anderson and Flaig, 1995; Haan, 1995). The P content was the greatest in center soils compared to the edge and upland soils of Larson Dixie Ranch

PAGE 24

11 the seasonally isolated wetlands on Larson Dixi e Ranch (Fig. 1-7). The isolated wetlands on the study site are flooded throughout the ma jority of the year with an occasional dry period from late March to early June. Soil on the Larson Dixie ranch is classified as Siliceous, hyperthermic Spodic, Psammaquents (B asinger series), which is a deep, poorly drained, rapidly permeable soil formed from sandy marine sediments (Anderson and Flaig, 1995). Objective 1: Characterize plant tissue from live and recently senesced vegetation of dominant species found in the thre e isolated wetlands on the Larson Dixie Ranch. Objective 2: Determine the short term P release from senesced plant material under aerobic and an aerobic conditions. Objective 3: Measure the litter decomposition and l ong term P release rate of four dominant species in situ at four different hydrologic regimes. The objectives mentioned above and th eir related hypotheses, methods, and results are discussed further in Chapters 2, 3, and 4 respectively. Chapter 5 summarizes results from all three chapters and eval uates the potential effects of hydrological restoration of isolated wetlands on decompos ition rates and P release as a BMP in the Okeechobee Basin. -10 40 90 140 190 240 CenterEdgeUplandHydroperiod (days / year) Figure 1-5. Average hydroperiod for center, edge, and upland zones of seasonally isolated wetlands on Larson-Dixie Ranch

PAGE 25

12 Figure 1-6. Species occurrence in seasona lly isolated wetlands located in the Okeechobee Basin. A) Wetland Centers, B) Wetland Edge, C) Surrounding Uplands. Wetland Centers 0 5 10 15 20 25Lud w i g ia Luz + Pasp. Other Panicum Polygonum Po n t ed aria R e s idualFrequency Wetland edges 0 5 10 15 20 25 30A ltern a nt h era Andropogon Bahia Ele o cha r is Juncus L ud wig ia Luz.+ Pasp. Mixed Oth e r Pani cu m P o lygo n um Ponted a ria R e sidualFrequency Surronding uplands 0 20 40 60 80 100A n d r op o go n B a h i a J u ncu s Ot he r Residua lFrequency C B A

PAGE 26

13 0 50 100 150 200 250 300 350 400 centeredgeuplandTotal P (mg/m2) Figure 1-7. Soil total phosphorus storag e in wetland zones and surrounding upland. Objective 1: Characterize plant tissue from live and recently senesced vegetation of dominant species found in the thre e isolated wetlands on the Larson Dixie Ranch. Objective 2: Determine the short term P release from senesced plant material under aerobic and an aerobic conditions. Objective 3: Measure the litter decomposition and l ong term P release rate of four dominant species in situ at four different hydrologic regimes. The objectives mentioned above and th eir related hypotheses, methods, and results are discussed further in Chapters 2, 3, and 4 respectively. Chapter 5 summarizes results from all three chapters and eval uates the potential effects of hydrological restoration of isolated wetlands on decompos ition rates and P release as a BMP in the Okeechobee Basin.

PAGE 27

14 CHAPTER 2 PLANT TISSUE CHARACTERIZATION Introduction Wetland vegetation is able to assimilate nutrients into its biomass from the soil and surrounding water column, thereby potentially reducing nutrient levels discharged from the wetland through ditches. Macrophytes onl y serve as a temporar y sink for nutrients however, and tend to undergo leaching and mine ralization of biomass after senescence. Observations from several studies suggest that the degree of d ecomposition and nutrient leaching is greatest during the first 4 months after senescence when up to 50% of N, P, and biomass can be lost in marsh comm unities (Boyd, 1971; Richardson, 1989; Flaig, 1995). Numerous physical and environmental factors such as temperature, moisture content, nutrients, pH, carbon quality, and the microbial community can influence litter decomposition and leaching rates (Melillo, 1982; Benner, 1985). Nutrient ratios such as C:N and C:P, as well as the quantity of residual fiber in the plant tissue is also important and often us ed to determine the physical and chemical composition of vegetation (Day, 1982; DeBusk and Reddy 1998; Lewis, 2005). Numerous studies have shown that nitroge n and lignin content (one of the most recalcitrant organic st ructural compounds) are the 2 most important factors that determine the rate of decomposition (Melillo, 1982; Berg, 1998; Carreiro, 2000; Villar, 2001; Corstanje, 2006). During the decomposition process, carbon compounds are lost, and in the short-term nutrients are either gain ed or lost depending on the surrounding environment and the nutrient ratios of the litter.

PAGE 28

15 Due to the role of C:N and lignin conten t play in decomposition, it may be possible to predict P release by assessing these two parameters in conjunction with wetlands where P assimilation and retention is being c onsidered. It is nece ssary to perform an initial characterization on live and recently senesced vegetation in order to quantify nutrient concentrations and fibe r quality of the dominant ve getative species on the site. Lignin is the organic com pound contained in plant biomass that is the most recalcitrant or resistant to decomposition. It is thought to be the limiting factor of longterm decomposition because it can only be degraded by a few organisms. Microbes are able to degrade lignin with the help of various enzymes (Benner, 1984); however it is extremely resistant to decomposition in anaerobic environments even by microbes (Criquet et al., 2001). Lignin helps provide stru ctural support to the plant in order to remain upright, as well as pr ovide a protective layer against microbial attack; therefore it is often abundant in the bark of woody trees (Hammel, 1997). Due to a greater assimilative capacity of wetland vegetation compared to edge and upland vegetation, there may be significant differences in the nutrient content of the plant species found in these 3 respec tive zones (McJanet et al., 1995) Differences in nutrient content may cause some species to have a lower or higher C/N ratio or residual fiber content compared to others, and may ultimatel y influence the quantity of P released from senesced vegetation through short term leach ing and long term decomposition. It is important to characterize th e nutrient and residual fiber to assess whether any tissue parameters are strongly correlated to P retenti on so that the amount of P released from a species after senescing may be predicted with some certainty. This information could provide further insight as to how species shif t as a result of hydrol ogical restoration, or

PAGE 29

16 other practices may impact the amount of P released into the water column after vegetative senescence. Hypothesis 1: The species of vegetation surveyed will have different amounts of carbon, nitrogen, phosphorus, and residual fiber. Materials and Methods Live and recently senesced (standing d ead) vegetation were harvested from Larson Dixie ranch in November 2004. Species found in the wetland center, edge, and upland were collected, placed in paper bags according to species. After the material was brought back to the laboratory, all the live species co llected were analyzed for TN, TC, TP, and tissue fiber analysis, while only the senesced ti ssue of 4 of the live species collected were analyzed for fiber and nutrient content due to their dominance on the site and presence in different hydrological zones. The 4 domi nant species were all perennial species including: Paspalum notatum Flugge (Bogdan) (Bahia grass) in the upland, and Juncus effusus L. Panicum spp., Polygonum hydropiperoides Michx in the edge and interior of the wetland. Vegetation was dried at 60 C for at least 72 hrs. The plants were then ground to pass a # 40 mesh sieve using a Wiley Mill. Analysis of total carbon and total nitrogen was done using a Thermo Electron Flash EA1112 Nitrogen/Carbon analyzer. Tissue P was analyzed using acid digestion of as hed tissue (Anderson 1976) and P content was quantified using an auto analyzer (Met hod 365.4; USEPA, 1993). An Ankom 200 Fiber Analyzer was used to quantify the NDF, ADF, and residual fiber fractions of tissue; this method is further described below (Ankom Technology Corp., 1998a). The means of all species were compared using Tukey-Kramer HSD (honestly signifi cant differences) to determine significant differences.

PAGE 30

17 Fiber Analysis Method Residual fiber content in tissue has been quantified in the past using the Klasson method by the hydrolysis of cellu lose with the use of 72% H2SO4 (Rowland and Roberts, 1999). This technique alone has proven to be time consuming due to lengthy pretreatments and is not easily reproducible in the laboratory (R owland and Roberts, 1999). The neutral detergent fiber (NDF ) and acid detergent fiber (ADF) method provides a simple technique for quantifying am ounts of labile components as well as hemicellulose contained in the plant biomass, and is the method used in this study (Fig. 2-1). Sequential NDF and ADF extractions were recommended to improve the digestion of cell wall proteins and minimize the influen ce of tannins on the recovery of residual fiber (Terrill et al., 1994). Figure 2-1. Extraction sequence used to determine carbon quality of vegetation. In this study “residual fiber” is referre d to as lignin. This material is highly recalcitrant, however also contains non-carbon compounds, th erefore it is functionally defined as any compound that has not been previously leached or degraded during the fractionation scheme. The NDF, ADF, and H2SO4 detergent fiber (strong acid detergent fiber or SADF) extractions sequentially re move sugars, hemicellu lose, and cellulose fractions of the plant material. The NDF ex traction removes the labi le components of the Dried, ground vegetation sample Neutral detergent fiber digestion 72% H2SO4 di g estion Residual Fiber Acid detergent fiber digestion Sugars Hemicellulose Cellulose “Lignin” Increasing Recalcitrance

PAGE 31

18 vegetation, such as sugars and starch. Th e ADF extraction remove s hemicellulose, while the SADF extraction strips away any remaining hemicellulose as well as cellulose, leaving behind lignin and ash (Table 2-1). Table 2-1. Neutral detergent fiber and acid detergent fiber fractionation methods (Rowland and Roberts, 1999) Results There was a wide range of C:N and N:P in the live tissue of the species surveyed (Fig. 2-2). Panicum hemitomon had the highest C:N of all the live species characterized, while L. fluitans had the lowest C/N ratio compared to other live species. However, Paspalum notatum had the highest N/P ratio out of all the species surveyed while B. caroliniana had the lowest N/P ratio. There were also significant differences in C:P in the live species surveyed (Fig. 23) In addition to having the highest C:N, P. hemitomon also had the highest C:P, followed by P. notatum, and J. effusus respectively, while P. cordata had the highest C:P associated with the live tissue.

PAGE 32

19 0 5 10 15 20 25 30 35 40Panicum hemitomon Juncus effusus Paspalum notatum Baccopa caroliniana Ludwigia repens Polygonum hydropiperoides Paspalum accumnatum Sagittaria lancifolia Hydrocotyle umbellata Pontederia cordata Alternanthera philoxeroides Luziola fluitans C:N N:P Figure 2-2. The C/N and N/P ratios of live ti ssue for species surveyed. Values represent mean ( 1 standard deviation). In addition, a wide range of NDF (neutral detergent fiber) and residual fiber was found in the different species surveyed (Fig. 2-4 & 2-5). Hydrocotyle umbellata, had a lower C/N and C/P ratio and had the highest NDF fraction in th e live tissue, resulting in lower percentages of ADF and SADF. Panicum hemitomon had the lowest NDF fraction associated with its biomass, and theref ore higher ADF and SADF percentages. Paspalum notatum, the dominant upland species, had the lowest percentage of residual fiber compared to all other species, and A. philoxeroides had the highest residual fiber content. Each of the 4 dominant species had a significantly lower P c ontent in the senesced tissue compared to the live tissue (Fig. 2-7). Polygonum hydropiperoides contained a significantly higher live and senesced P tissu e content compared to other 3 dominant species of vegetation, wh ile the P content in P. hemitomon was significantly lower in the

PAGE 33

20 senesced tissue compared to the other 3 species of senesced vegetation. In addition to a higher %P in the live tissue of the 4 domina nt species, the C/P ratio in senesced tissue was also significantly lower than the C/P valu es observed in the live tissue (Table 2-2). 0 50 100 150 200 250 300 350Panicum hemitomon Juncus effusus Paspalum notatum Baccopa caroliniana Ludwigia repens Polygonum hydropiperoides Paspalum accumnatum Sagittaria lancifolia Hydrocotyle umbellata Pontederia cordata Alternanthera philoxeroides Luziola fluitans C:P Figure 2-3. Mean C/P ratios in live tissue of species survey ed. Values represent mean ( 1 standard deviation). 0% 20% 40% 60% 80% 100%Panicum hemitomon Juncus effusus Paspalum notatum Paspalum accumnatum Polygonum hydropiperoides Pontederia cordata Luziola fluitans Baccopa caroliniana Alternanthera philoxeroides Ludwigia repens Sagittaria lancifolia Hydrocotyle umbellata % of Plant Tissue NDF ADF SADF Residual Fiber Figure 2-4. Fiber quality (NDF, ADF, SA DF and Residual Fiber) percentages for species surveyed.

PAGE 34

21 0 5 10 15 20 25 30Panicum hemitomon Juncus effusus Paspalum notatum Baccopa caroliniana Ludwigia repens Polygonum hydropiperoides Paspalum accumnatum Sagittaria lancifolia Hydrocotyle umbellata Pontederia cordata Alternanthera philoxeroides Luziola fluitans % Residual Fiber Figure 2-5. Residual fiber fractions for live ti ssue. Values represent mean ( 1 standard deviation). 0 10 20 30 40 50 60 Panicum hemitomon Paspalum notatum Polygonum hydropiperoides Juncus effusus C/N Ratio Live C:N Senesced C:N Figure 2-6. Nutrient ratios of C:N in senesced tissue of dominant species. Values represent mean ( 1 standard deviation).

PAGE 35

22 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 Panicum hemitomon Paspalum notatum Polygonum hydropiperoides Juncus effusus%P Live %P Senesced %P Figure 2-7. Comparison of phos phorus values for live and senesced vegetation. Values represent mean ( 1 standard deviation). Table 2-2. The C/N and C/P ratios for live a nd senesced tissue. Values represent mean ( 1 standard deviation). Species Live C:N Senesced C:N Live C:P Senesced C:P Panicum hemitomon 34.7 1.6 51.8 2.4 289.3 7.4 2165.1 129.8 Paspalum notatum 24.4 0.5 50.7 4.4 274.9 9.0 520.4 56.3 Polygonum hydropiperoides 20.5 0.7 40.4 0.5 160.0 19.7 350.5 33.7 Juncus effusus 25.1 0.9 40.3 1.4 257.9 12.4 557.2 15.0 Additionally, the NDF fraction in the sene sced tissue was lower compared to the live tissue (Fig. 2-8). Differences in NDF c ontent between live and senesced tissue were significant in P. hydropiperoides and P. notatum The highest NDF fraction was seen in P. hydropiperoides There were also differences in th e residual fiber content in the live and senesced tissue. Residual fiber was slight ly lower in the live tissue of the 4 dominant species compared to the senesced tissue (Fi g. 2-9). The only significant difference in the residual fiber content between the live and senesced tissue was seen in P.

PAGE 36

23 hydropiperoides. In addition to having the highest NDF fraction, this species also had significantly higher residual fi ber content in live and senes ced tissue compared to the 3 other species. 0 5 10 15 20 25 30 35 40 45 50 Panicum hemitomonPaspalum notatumPolygonum hydropiperoides Juncus effusus% NDF Live Tissue Senesced Tissue Figure 2-8. Neutral detergen t fiber content contained in the senesced tissue of 4 dominant species. Values represen t mean ( 1 standard deviation). 0 2 4 6 8 10 12 14 16 18 Panicum hemitomon Paspalum notatum Polygonum hydropiperoides Juncus effusus % Residual Fiber Live Residual Fiber Senesced Residual Fiber Figure 2-9. Residual fiber c ontent contained in the sene sced tissue of 4 dominant species. Values represent mean ( 1 standard deviation).

PAGE 37

24 Discussion Different qualities of vegetative tissue were observed in this study depending on the species, the landscap e position which the species was found, and whether the vegetation was live or senesce d. A range of chemical and physical characteristics in live and senesced tissue of different vegetative species may influence P release rates through short term leaching and long term d ecomposition. Species included in the characterization with a low N and P content, a hi gh percentage of residual fiber, or a high C/N or C/P ratio may have slower decom position or leaching rates which may also decrease the rate of P release. High nutrient ratios may limit the amount of microbial degradation as well as the decomposition rate. Senesced vegetation with a high N and P content as well as a low residual fiber percentage, may have a greater rate of decomposition because it will serve as a nutrient rich substrate with a higher likelihood of microbial colonization. Various studies have reported increased decomposition of tissues with a hi gher N content (Berg, 1998; Carreiro, 2000; Corstanje, 2006), and an N limitation may suppress decomposition rates due to a lack of microbial colonization of the substrate. In general, a C:N > 30 or a C:P > 200 in the senesced tissue may resu lt in an N or P limitation and decrease microbial decomposition under aerobi c conditions(Fenchel et al., 1998). Based on values in Table 2-2, vegetation collected from the Larson Dixie site suggests that all species of senesced tissue may have an N and P limitation, due to ratios greater than 30:1 and 200:1 re spectively, and may not result in a net mineralization of N and P under aerobic conditions. Species with a nutrient limitation and restricted microbial decomposition may release a lower amount of P compared to species with a high nutrient content. Furthermor e, the nutrient ratios are likel y to be higher in the live

PAGE 38

25 tissue compared to the senesced tissue due to a leaching event which may have removed some of the labile N and P or a resorbtion of nutrients back into the living plant tissue. Observations have shown that some sp ecies of emergent marsh vegetation, like P. hemitomon, can remain as standing dead biomass for up to 2 months or more after senescence (Lewis, 2005), which may allow multiple leaching events to take place before the senesced vegetation enters the water column. Whether the species is an annual or pere nnial may also influence the amount of nutrient uptake, resorbtion, and P re lease rate. It is believed th at the species used in this study were all perennial, and may therefore ha ve a higher degree of nutrient resorbtion back into the live tissu e of the plant, especially into the below ground tissues. Perennial species may also have a lower rate of bi omass accumulation, and a lower P release rate compared to annual species (Tobe, 1998; Fraser and Karnezis, 2005). On highly disturbed landscapes there may be a higher pr oportion of annual plan t species which may further contribute to eutrophication of nearby water bodies due to a potentially greater rate of decomposition and P release, possibly as a re sult of higher nutrient concentrations, NDF, and a lower residual fibe r content compared to perennial species. Different amounts of NDF, ADF, SADF, and residual fiber were seen in the vegetative species surveyed. It is likely that species with high residual fiber content will have a slower decomposition rate and possibl y a slower P release rate compared to species with low residual fiber content and a higher NDF fraction. The NDF content was higher for each species in the live tissue compared to the senesced tissue which may indicate the NDF or labile car bon fraction in the vegetation is rapidly lost in some species after senescence. Significant differences between the NDF fractions in the live and

PAGE 39

26 senesced vegetation of P. notatum and P. hydropiperoides, suggest that soluble sugars and carbohydrates may be lost the fastest in th ese species or they had been senesced for a longer period of time before collection. A leaching event could have occurred after senescence, removing some of the labile su gars and carbohydrates associated with the standing dead vegetation. Residual fiber also made up a higher percentage of biomass in the senesced vegetation due to the reduction of NDF in the litter biomass. In summary this chapter addressed th e hypothesis that different species of vegetation found on isolated wetlands in the Okeechobee Basin contained different nutrient levels quantified by the C/N ratio and %P in live and senesced tissues, as well as various fiber qualities. Findings suggest that the N and P content and fiber quality varied significantly among the different species collected, which suppor ts the original hypothesis. Results also suggest some sene sced species may provide a better substrate for the colonization of microorganisms, and possibly a rapid rate of decomposition and P release due to a lower C/N ratio and residua l fiber fraction, a higher P content and NDF fraction. Lastly, there were significant diffe rences in the substrate quality between the live and senesced tissue in terms of C:N, C:P, NDF, and residual fibe r content, indicating that leaching, and a rapid breakdown of s ugar occurs shortly after senescence.

PAGE 40

27 CHAPTER 3 SHORT TERM PHOSPHORUS LEACHING Introduction The term leaching, as used in this thesis, is a natural process where mass is lost from senesced vegetation due to the release of soluble or ganic and inorganic compounds from the plant biomass (Robertson, 1988). Th e leaching process is the first of three phases of decomposition, and is subsequently followed by microbial mineralization and physical and biological fragmentation (Va liela, 1985; Webster and Benfield, 1986). There is evidence that leaching is correlated with rainfall when standing dead vegetation is attached to the plant (Taylor et al, 1989; Qiu, 2005), and may continue after the litter is incorporated into the detrital layer and immersed in the water column. The most rapid period of leaching usually lasts from a few days to a few weeks (Davis et al., 2003) depending on water availability, and is not mediated by microbial processes. The greatest leaching rates usually occur during the first rain fall or the first hours after emersion in the water column (Tope, 2003; Qiu, 2005). Leaching processes can result in a large nutrient flux of P, N, and C from plant biomass into the water column during fall and spring as a result of vegetative senescence (Mitsch et al., 1989). This is the primary reason why leaching has recently become one of the primary concerns relating to water qu ality in agricultural areas (Kuusemets and Mander, 2002). The processes influenc ing P leaching rates are poorly understood, however temperature, moisture, pH, Eh, P co ncentration, and land us e, are the primary variables influencing the P leaching rate of vegetation (Melillo, 1982; Benner, 1985;

PAGE 41

28 Richardson, 1989; Moore and Reddy, 1994; Findla y et al, 2003; Flaig, 1995; Lewis, 2005). Short-term P leaching from senesced tissue can lead to a significant release of P into the environment. Anywhere from a 2050% loss of the total P in the plant biomass can be released from the vegetation in a fe w hours and upwards of 80% of the total P can be released during the first 2 months of mineralization (Simpson et al., 1978; Webster and Benfield, 1986; Tope, 2003; Qiu, 2005; Reddy et al., 2005). Leaching is an important nutrient cycli ng process to consider when understanding the fate of P within a wetland ecosyst em. A significant por tion (30-80%) of the bioavailable P in the water column can be immobilized by macrophytes (Dolan et al., 1981), depending on the plant species, the grow th rate of the plant, plant density, harvesting frequency, climate, and the oxygen ava ilability in the sediment (Reddy et al., 1995). Vegetation only serves as a temporary si nk for nutrients however. It is estimated that 35-80% of the tota l P contained in the biomass is lost to the water column, to the soil or transported out of the sy stem after plant senescence (R ichardson, 1989; Reddy et al., 1995). A portion of the total P obtained in the standing dead tissue is resorbed back into the living part of the plant, preventing all the P contained within the senesced biomass from leaching out. Nutrient resorbtion in a hardwood forest was estimated to be approximately 30-34% of the total N and P c ontained within the senesced tissue (Ryan, 1982). The main objective of this study is to qua ntify the amount and rate of P loss from senesced vegetation into the water column under aerobic and anaerobic conditions, and in response to N enrichment. This study provi ded a more suitable sampling frequency to

PAGE 42

29 assess short-term P release associated with different species of litter in the wetland that would have been difficult to quantify in the field. The litterbag st udy in the subsequent chapter, evaluated decomposition over a longer period of time. It is likely that different species will ha ve various leaching rates after vegetative senescence, due to the fact that some speci es have higher tissue P concentrations than others (Hobbie, 1992; Knops et al., 2002). The sp ecies that releases th e lowest amount of P in this study relative to its P assimilation rate could be used to increase P storage on historically isolated wetlands. In addition, knowing species specific P leaching characteristics provides managers and la ndowners with guidance on beneficial and problematic plants when trying to re tain P on pastures and in wetlands. Hypotheses A greater P flux will be observed unde r anaerobic conditions than aerobic conditions. Elevated nitrogen levels will increase P fl ux rates when compared to ambient site water and site water diluted with deionized water. Short term P release will be positively related to the concentration of P in the senesced tissue. Materials and Methods Recently senesced (standing dead) vegetation was harvested from Larson Dixie ranch in November 2004, this was the same ma terial used in the characterization study described in the previous ch apter as well as the decom position study explained in the following chapter. Dominant species in the wetland center, edge, and upland were collected then placed in paper bags according to species. Species that were adapted to longer hydroperiods included Panicum hemitomon, which had a P concentration of 0.209 0.01 mg P/g tissue, and Polygonum hydropiperoides with 1.278 0.11 mg P/g tissue.

PAGE 43

30 The dominant edge species was Juncus effusus, which had 0.804 0.02 mg P/g tissue. Lastly, the dominant upland species was Paspalum notatum which had 0.830 0.08 mg P/g tissue. The senesced vegetation was air dried due to evidence that oven drying can artificially alter chemical composition as well as leaching rates (Tope, 2003). Approximately 1.5 grams of air dried litter material was placed in a 250 ml covered plastic container and filled with 200 ml of treatment water. The litter added to the containers was scaled down from the mass of litter naturally present within a 1m x 1m quadrat. Vegetation was misted with de ionized (DI) water approximately 12 hours before the treatment water was added in order to reconstitute the litt er so it wouldn’t float up to the top of the container and to bring the moisture content of the air dried litter up to field conditions. Empty fluxing containers, wi th no vegetation added, were filled with 200 ml of treatment water and used as the co ntrol. A hypodermic needle was inserted into the head space of the containe r through a rubber septum in the top of the container. This needle functioned as a pressure relief vent. A longer needle was pushed through the same rubber septum and below the water surface to bubble atmospheric gas (aerobic conditions) or nitrogen gas (anaer obic conditions) (Fig. 3-1) in to the container and gently mix the water column. In addition to species and oxygen availabi lity treatments, three water quality conditions were tested. Water treatments c onsisted of low P water collected from a cypress slough near the research site, which is defined as “Site Water” in this chapter (SRP concentration of 0.09 0.00 mg/L, a DOP concentration of 0.18 0.00 mg/L, and an NO3 concentration of 0.22 0.00 mg/L). Other water treatments included site water

PAGE 44

31 diluted 50% with DI water (SRP concentr ation of 0.04 0.00, a DOP concentration of 0.12 0.01, and an NO3 concentration of 0.15 0.00), and site water spiked with 3 mL of 1000 ppm NO3 to double the N concentration to 0 .41 ppm (SRP concentration of 0.09 0.00, a DOP concentration of 0.18 0.00, and an NO3 concentration of 0.41 0.00). A B Figure 3-1. The P leaching st udy. A) Overview of the expe rimental setup and aerobic and anaerobic treatments B) Indivi dual fluxing container and tubing bubbling ambient air in the water column through the hypodermic needle. This experimental design resulted in a to tal of 90 flux containers. Flux containers were covered and kept in darkness to prev ent algae growth. The water column was sampled a total of 5 times. The timescale of sampling was after 2 hrs, 24 hrs, 3 days, 7 days, and 17 days, with an initial characteriz ation of site water before the experiment began. At each sampling period 20 ml of wa ter was removed from the flux container and analyzed for SRP. At time zero and after 17 days, an additional 40 ml of water was removed and analyzed for Dissolved Organic Phosphorus (DOP) and Total Kjeldahl Nitrogen (TKN). The DOP samples were filtered and digested in the autoclave (Method 365.1; USEPA, 1993), while TKN samples were unfiltered and digested according to EPA Method 351.2 (USEPA, 1993).

PAGE 45

32 After each sampling period, an equivalent volume of treatment water was added to each container to maintain a constant volume of 200 mL. The dilution factor and P mass loss in sample water were taken into account when calculating cumulative flux of tissue P over the course of the experiment. At the end of the experiment, litter and any newly formed particulates within the water column were filtered using a glass fiber filter, oven dried at 70 C for at least 24 hours, and we ighed to determine the mass loss from the litter. Samples and glass fiber filters were subsequently ground to pass a # 40 mesh sieve using a Wiley Mill and analyzed for total carbon and total nitrogen using a Thermo Electron Flash EA1112 Nitrogen/Carbon analyzer. The mean P flux rates of each species we re compared using Tukey-Kramer HSD (honestly significant difference) to determ ine significant differences between the cumulative P flux among species. Regression an alyses and one way analysis of variance (ANOVA) were used to determine which s ubstrate quality parame ter, quantified in Chapter 2, was the best predictor of short term P release using the R2 and p-values. Results Cumulative P flux was averaged, combin ing three water treatments and ranged from an average of 0.01 0.01 mg P/ g tissue over a period of 17 days( P. hemitomon ) to 0.96 0.17 mg P /g tissue ( P. hydropiperoides ) under aerobic conditions (Figure 3-2a) Cumulative P flux from P. notatum and J. effusus fell between this range with values of 0.59.15 and 0.35.08 mg P /g tissue, respectiv ely, over the same 17 day period under aerobic conditions. Leaching was greater under aerobic conditions in each species except J. effusus, which had the greatest P flux under anaerobic conditions. The water treatment and redox condition had little effect on the P re leased from senesced vegetation over a 17 days period, and an obvious trend between N enrichment or redox and P release was not

PAGE 46

33 seen. The only statistically significan t differences observed between aerobic and anaerobic conditions over 17 days was in P. notatum, where P flux was significantly higher under aerobic conditions and J. effusus where P fluxes were greater under anaerobic conditions (Table 3-1). At the end of the study, on Day 17, the only significant difference between aerobic and anaerobic conditions was seen in P. hemitomon. Significant differences in flux rate among individual spec ies under aerobic conditions (Fig. 3-2a) occurred within the fi rst 2 hours. At this sampling period, P flux from P. notatum and P. hydropiperoides were significantly higher than the other 2 species, and P. hemitomon had a significantly lower P release compared to all other species. After 1 day of incubation and for th e remainder of the study, each species each species had a significant ly different flux rate. Flux containers incubated under anaerobi c conditions also began to show significant differences in P fl ux rates only 2 hrs after water was added (Fig. 3-2b). At this sampling period P. hydropiperoides again had a significantly greater P flux, while P. hemitomon still had the smallest. This trend co ntinued throughout the course of the study. At day 17, there were significan t differences seen among each species. After a period of 17 days, P. hemitomon, had the least amount of P leaching from the litter. The P released from P. hemitomon was significantly lower at every sampling period compared to other species of ve getation under both aerobic and anaerobic conditions (Fig. 3-2). The cumulative P flux for P. hemitomon peaked after just 2 hours, then declined nearly to 0 mg P/g tissue. The minimum 2 hr flux value for this species was 0.085 0.009 mg P/g tissue observed in th e site water + DI under aerobic conditions, while the maximum flux value was 0.101 0.031 mg P/g tissue observed in the site water

PAGE 47

34 treatment under anaerobic conditions (Fig. 3-3a ). There was not a significant difference between redox conditions or among wa ter treatments (Fig. 3-3) in P. hemitomon over the course of 17 days (Table 3-1), however ther e was a significant difference between redox condition for this species on Day 17 only.. 0 0.2 0.4 0.6 0.8 1 1.2 03691215 Time (days)Cumulative P flux (mg P/g tissue) P. hemitomon P. notatum P. hydropiperoides J. effusus Control (a) 0 0.2 0.4 0.6 0.8 1 1.2 03691215 Time (days)Cumulative P flux (mg P/g plant tissue) P. hemitomon P. notatum P. hydropiperoides J. effusus Control (b) Figure 3-2. Phosphorus leaching rates of 4 se nesced species averag ed across all 3 water treatments A) aerobic B) anaerobic c onditions. Values represent mean ( standard deviation).

PAGE 48

35 The species with the highest P flux unde r aerobic and anaerobic conditions was P. hydropiperoides (Fig. 3-2). The P flux of this sp ecies continuously increased throughout the experiment, with the highest cumula tive flux of 1.00 .215 mg P/g plant tissue observed under aerobic conditions in the site water + N treatment, and 0.976 0.223 under anaerobic conditions in the site water + DI treatment after 17 days (Fig. 3-4). The lowest 17 day P flux for this species was s een in the site water + DI treatment under aerobic conditions (0.94 0.15) and site wate r + N treatment under anaerobic conditions (0.74 0.08). -0.03 -0.01 0.01 0.03 0.05 0.07 0.09 0.11 0.13 0.15 051015 Time (days)Cumulative P flux (mg P/g tissue) Site Water Site Water + N Site Water + DI (a)

PAGE 49

36 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 051015 Time (days)Cumulative P flux (mg P/g tissue) Site Water Site Water + N Site Water + DI (b) Figure 3-3. Litter phosphorus release rate for P. hemitomon litter under (a) aerobic and (b) anaerobic conditions with 3 different water treatments. Values represent mean ( 1 standard deviation). The remaining 2 species, P. notatum and J. effusus had a moderate amount of P released from the litter compared to P. hydropiperoides There were not any significant differences between water treatments in P. notatum under aerobic conditions, however the site water + N treatment had a signif icantly lower P flux for this species under anaerobic conditions compared to the other 2 treatments over the course of 17 days (Fig. 3-5). Juncus effusus however, had no significant diffe rences in P flux between water treatments under aerobic or anaerobic c onditions (Fig. 3-6), however P flux was significantly higher under anaerobic conditions compared to aerobic conditions for this species over the course of the entire study (Table 3-1).

PAGE 50

37 0 0.2 0.4 0.6 0.8 1 1.2 1.4 051015 Time (days)Cumulative P flux (mg P/g tissue) Site Water Site Water + N Site Water + DI (a) 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 051015 Time (days)Cumulative P flux (mg P/g plant tissue) Site Water Site Water + N Site Water + DI (b) Figure 3-4. Litter phosphorus release rate for P. hydropiperoides litter under a) aerobic and b) anaerobic conditions with 3 water treatments. Values represent mean ( 1 standard deviation)

PAGE 51

38 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 051015 Time (days)Cumulative P flux (mg P/g tissue) Site Water Site Water + N Site Water + DI (a) 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 051015 Time (days)Cumulative P flux (mg P/g tissue) Site Water Site Water + N Site Water + DI (b) Figure 3-5. Phosphorus release rate for P. notatum litter under a) aerobic and b) anaerobic conditions with 3 water treatments. Values represent mean ( 1 standard deviation).

PAGE 52

39 0.00 0.10 0.20 0.30 0.40 0.50 0.60 051015 Time (days)Cumulative P flux (mg P/g tissue) Site Water Site Water + N Site Water + DI (a) 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 051015 Time (days)Cumulative P flux (mg P/g tissue) Site Water Site Water + N Site Water + DI (b) Figure 3-6. Phosphorus release rate for J. effusus litter under a) aerobi c and b) anaerobic conditions with 3 water treatments. Va lues represent mean ( 1 standard deviation).

PAGE 53

40 Table 3-1. Cumulative P release (mg P/g litte r) and percent tissue P released from four species over a 17 day period, under aer obic and anaerobic conditions. Values represent mean ( 1 standard devi ation). Lowercase letters indicate significant differences over the entire 17 day study between species, water treatment, and redox conditions with a p-value of .05. 17 Day Total Cumulative Flux (mg P/g tissue) Species Treatment Aerobic % P Released Anaerobic %P Released Site Water 0.002 (a) 0.003 0.80 0.007 (a) 0.001 3.55 Site Water + N -0.002 (a) 0.001 -1.08 0.003 (a) 0.004 1.28 Panicum hemitomon Site Water + DI 0.024 (a) 0.026 11.61 0.011 (a) 0.006 5.45 Site Water 0.551 (b) 0.126 66.37 0.414 (c) 0.203 49.86 Site Water + N 0.694 (b) 0.170 83.58 0.197 (d) 0.034 23.78 Paspalum notatum Site Water + DI 0.539 (b) 0.144 64.94 0.377 (c) 0.022 45.41 Site Water 0.925 (e) 0.149 72.34 0.911 (e) 0.066 71.22 Site Water + N 1.004 (e) 0.215 78.55 0.738 (f) 0.082 57.74 Polygonum hydropiperoides Site Water + DI 0.937 (e) 0.149 73.32 0.976 (e) 0.223 76.32 Site Water 0.355 (g) 0.112 44.16 0.461 (h) 0.030 57.33 Site Water + N 0.342 (g) 0.079 42.52 0.636 (h) 0.275 79.09 Juncus effusus Site Water + DI 0.351 (g) 0.055 43.72 0.517 (h) 0.074 64.32 Over the span of 17 days, SRP concentrati on significantly increased from the initial 0.09 mg/L in the site water flux co ntainers of every species except P. hemitomon Approximately 96% of the P in the water co lumn at the end of the study was SRP, a readily labile form, while the small re maining P fraction was DOP (Table 3-2). In general TKN concentrations were hi gher under anaerobic conditions compared to aerobic conditions, however few significan t differences were seen (Table 3-3). Polygonum hydropiperoides was the only species where TKN concentrations were higher under aerobic conditions (in site water and site water + N tr eatment). In addition there were not any significant diffe rences observed in the TKN concentrations among species or water treatments.

PAGE 54

41 Table 3-2. Mean concentrations ( 1 standard deviation) of water column P for various species under aerobic and anaerobic conditions after 17 days. Negative DOP values are due to high standard errors and represent virt ually no measurable DOP in the water column. SRP (mg/L) DOP (mg/L) Species Treatment Aerobic Anaerobic Aerobic Anaerobic P. hemitomon Site Water 0.03 0.01 0.03 0.01 0.06 0.02 0.09 0.02 P. notatum 2.72 0.67 2.07 1.09 0.22 0.06 0.15 1.07 P. hydropiperoides 4.99 0.84 4.84 0.37 0.18 1.00 -0.03 0.33 J. effusus 1.87 0.65 2.37 0.25 0.05 0.53 0.13 0.33 P. hemitomon Site Water + N 0.02 0.00 0.03 0.00 0.06 0.03 0.07 0.01 P. notatum 3.57 0.88 0.92 0.16 0.23 1.91 0.18 0.13 P. hydropiperoides 5.51 1.07 3.97 0.34 0.05 1.11 0.04 0.64 J. effusus 1.77 0.34 3.61 1.65 0.06 0.31 0.14 1.56 P. hemitomon Site Water + DI 0.11 0.00 0.02 0.00 0.09 0.16 0.06 0.01 P. notatum 2.68 0.74 1.74 0.17 0.12 0.70 0.23 0.13 P. hydropiperoides 5.13 0.73 5.17 1.27 0.23 0.57 0.11 1.29 J. effusus 1.77 0.27 2.67 0.43 0.09 0.28 0.24 0.20 Table 3-3. Mean concentrati ons ( 1 standard deviation) of water column TKN for various species under aerobic and anaerobi c conditions after 17 days. Letters indicate significant differences between species and redox condition. TKN (mg/L) Species Treatment Aerobic Anaerobic P. hemitomon Site Water 3.61 (a) 2.42 6.21 (a) 0.24 P. notatum 5.53 (a) 0.32 6.31 (a) 0.68 P. hydropiperoides 7.34 (a) 3.54 6.08 (a) 1.69 J. effusus 3.46 (a) 0.58 6.47 (a) 2.13 P. hemitomon Site Water + N 4.41 (A) 1.19 6.62 (A) 1.54 P. notatum 5.15 (A) 1.08 6.49 (A) 0.60 P. hydropiperoides 6.02 (A) 1.28 5.4 (A) 1.55 J. effusus 3.81 (A) 0.18 6.31 (B) 1.50 P. hemitomon Site Water + DI 7.07 (a’) 3.48 6.47 (a’) 1.54 P. notatum 6.02 (a’) 1.94 9.47 (a’) 5.17 P. hydropiperoides 3.53 (a’) 0.38 5.13 (a’) 2.28 J. effusus 2.97 (b’) 0.73 6.78 (b’) 1.04

PAGE 55

42 After 17 days of incubation, there were no significant differen ces in the mass loss among species, although a substa ntial decline in mass occu rred. The largest mass loss was observed in P. hydropiperoides while P. hemitomon had the smallest. 0 5 10 15 20 25 J. effususP. hydropiperoidesP. notatumP. hemitomon% Mass Loss Figure 3-7. Percent mass loss from the vegetation over a 17 day period. Values represent mean 1 standard deviation. Multiple ANOVAS were run in order to determine which parameter provided the best predictor of the amount of P released fr om vegetation (Table 34). Regressions were made between the cumulative P flux on day 17 and initial tissue concentration of N, P, and C and various ratios of these parameters. It was determined with 99.99% confidence that a relationship exists between P contained in the tissue and the amount of P that tissue will release to the water column. Tissue P c ontent explained 89% of the variability in P flux in site water under aerobic conditions on Day 17, while %NDF explained 76% of the variability in P flux (Fig. 3-8).

PAGE 56

43 Table 3-4. The P-Values associated with initia l nutrient parameters of senesced tissue to estimate the best predictor of P flux fo r site water under aerobic conditions on Day 17. Substrate Quality Parameter P-value R 2 % C .314 .101 % SADF .252 .129 % N .110 .235 % Residual Fiber .104 .242 C:N .062 .306 % ADF .030 .390 C:P .001 .670 N :P .001 .687 % NDF .0002 .757 % P < .0001 .887 -0.25 0 0.25 0.5 0.75 1 1.25cumulative flux (mgP/g tissue) 0 .025 .05 .075 .1 .125 .15 %P senesced Figure 3-8. Bivariate fit of cumulative P flux by initial senesced tissue P content. Correlation is for site water treatm ent under aerobic conditions on day 17. Table 3-5 shows R-square values for the correlations between cumulative P release and tissue P concentration for each sampling period under aerobic and anaerobic conditions. The correlation improves throughout the course of the study under aerobic P. hemitomon P. notatum P. hydropiperoides J. effusus Cumulative Flux= -0.21 + 8.58x R 2 = .887

PAGE 57

44 conditions with the highest R value o ccurring on day 17, while under anaerobic conditions the R value is the highest on day 1 and 7. Phosphorus concentration in the litter is a slightly be tter indicator under anaerobic conditi ons as to the amount of P that will be released from that species. Other initial substrate quality parameters such as %NDF, N:P, and C:P had significant p-values however the relationship between these parameters and the amount of P released from the senesced vegetation was not as strong compared to the relationship between the initi al %P in the senesced tissue and cumulative P release. Table 3-5. The R-Square values from %P and cumulative P release correlation in site water under aerobic and anaerobic conditions R-Square values Time (days) 0.08 1 3 7 17 Aerobic 0.732 0.861 0.853 0.864 0.887 Anaerobic 0.885 0.914 0.911 0.914 0.908 Table 3-6. The P-Values associated with in itial nutrient parameters of live tissue to estimate the best predictor of P flux fo r site water under aerobic conditions on Day 17. Substrate Quality Parameter P-value R2 N:P 0.7696 0.009 % Residual Fiber 0.0967 0.251 % ADF 0.0129 0.477 % C 0.0076 0.526 % P 0.0014 0.658 C:P 0.0007 0.696 % N <.0001 0.829 C:N <.0001 0.830 % SADF <.0001 0.862 % NDF <.0001 0.883 Initial substrate quality parameters of the live tissue characterized in Chapter 2, were also correlated with the short term cumu lative P release from each senesced species of vegetation. The strongest correlation was between short term P flux and % NDF or the labile fiber fraction associated with the live tissue with an R2 of .883 (Table 3-6).

PAGE 58

45 Discussion Findings suggest the potentia l for vary rapid and signi ficant amounts of tissue P release from plant litter upon exposure to rainfall or inundation, and the amount of P released from the senesced tissue was pr imarily dependant on the species type. Phosphorus release from the 4 litter species tended to peak during the first 2 hours after the water treatment was added to the vegeta tion through the first 3 days of the study. After day 7, bacterial growth began to appear in the flux containers likely resulting in water column P uptake and a possible explana tion for the slight decline in cumulative flux values. Approximately 70% of the P in the tissue of P. hydropiperoides was released on average after 17 days, while P. notatum and J. effusus both released 55% of the P that was initially associated with the biomass, and P. hemitomon released only 4% of the P associated with its biomass. The P flux values closely follow the amount of P each species originally contained in the bioma ss, which helps explain why the cumulative P flux from P. hydropiperoides was greater than the cumulative flux for other species because this species had the highest percentage of P initially, while P. notatum and J. effusus initially had similar P concentrations, and P. hemitomon had the least amount of P. The cumulative amount of P released over the 17 day period is just below the 80% value that was lost during biomass decom position in the study preformed by Reddy et al. (1995) and below the 35% value reported by Richardson (1989). Panicum hemitomon contained only 25% of the P in the senesced tissue of P. notatum and J. effusus, and 6.2% of the P contained in P. hydropiperoides. In addition to having the lowest P flux, P. hemitomon also had the highest initi al C:N and the lowest NDF fraction out of all the live species survey ed. Neutral detergent fiber content was the

PAGE 59

46 parameter that was the best predictor of short term P release in the live tissue. The small amount of P within the tissue of P. hemitomon was held onto tightly throughout the study. A large quantity of P was released from th e senesced tissue within a relatively short period of time. The rate at which P was released from the senesced vegetation quickly increased between 2 hours and 3 days, after whic h the release rate slow ed and leveled off, then stopped or reversed. This information suggests that a potentially large amount of P can be released from the senesced tissue s hortly after vegetative senescence due to rainfall or upon entering the water column. Results from this study are similar to a leaching experiment preformed by Davis et al. (2003) which reported the most rapid leaching from Rhizophora mangle L. leaf litter after 2 da ys of incubation. High quantities of P fluxing into the water column after senescence could have the potential to increase the eutrophication of seasonally isolated wetlands and downstream waterways. Hydrological restoration could also cause spec ies shifts which may significantly impact the amount and rate of P leaching. In each species with the exception of P. hydropiperoides, there was a negative P flux observed in either aerobic or anaerobic co nditions during the last 7-14 days of the study. A decrease in the cumulative P flux can only be explained by microbial uptake of P from the water column or P uptake by the li tter. Near day 7, a bacterial growth is thought to have occurred around the needles bubbling gases into the fluxing containers since light was available during incubation. It is likely that the negative slopes depicted in the graphics above were due to microbial nut rient uptake. In future studies, it may be

PAGE 60

47 interesting to have a light vs. dark treat ment to quantify the P uptake by algae and bacteria. Looking at the cumulative flux curves for each species, it is estimated that P leaching took place for approximately the firs t 7 days of this study. Around day 7, the P flux slowed, leveled off, or became nega tive, suggesting microbial mineralization processes began to take over. It was suspected that the water spiked with N may increase P release during mineralization because all sp ecies initially had a high C:N, and would have been N limited under aerobic conditions. A nitrogen addition would increase the N concentration as well as the amount of nutri ents available to microbes, and therefore stimulating mineralization, C breakdown and P release beyond a 7 day period. The low NDF fraction and initial P concentrations in P. hemitomon is likely the reason why this species had the lowest change in mass after a 17 days, while P. hydropiperoides underwent the highest mass change b ecause it’s initial NDF fraction was higher than the 3 other species. It is likely that the NDF fr action or the soluble sugars and starches associated with the litter is what mostly contributed to the mass loss of the litter, along with the N and P a ssociated with this fiber fr action. A even higher mass loss was reported in a 21 day leaching study prefor med by Davis et al. (2003), where leaching accounted for 33% of the dry mass from R. mangle leaves. Physical factors such as the fiber quality of the litter are likely to influence short term leaching and mass loss more than the nutrient content of the litter or th e C/N ratio, which may in fluence P release over a longer period of time. This may be a possi ble reason why the %NDF in the live tissue had the strongest correlation with short term P release. It is thought that the majority of

PAGE 61

48 the mass differences resulted from nutrient leaching through day 7, and perhaps microbial mediated decomposition processes for the remainder of the study. The empty flux containers used as the c ontrol had an extremely low amount of measurable P flux. This could be attributed to a small amount of sediment that was suspended in the water column when the treat ment water was added to the containers. The control did not have any vegetation, and therefore had a post experiment C:N of 10.0 6.2 as a result of bacterial gr owth which appeared in all of the containers around day 7. Overall, there did not appear to be a la rge difference in P release under aerobic or anaerobic conditions, an d there wasn’t a clear trend as to which condition caused the most P to flux from the vegetation. Even though there were no si gnificant differences between treatments on Day 17, N enrichment seemed to enhance P release under aerobic conditions, but inhibite d P flux, or had no effect on P re lease under anaer obic conditions. Anaerobic microbes often require a lower amount of available N to carry out metabolic activities since NO3 is often not available in reduced e nvironments. It is possible that the site water + N treatment had no effect under anaerobic conditions because the C:N was initially lower than 80:1; therefore the species used in this study w ould not be considered N limited under anaerobic conditions. Because enough N was already present to carry out metabolic activities unde r anaerobic condition, additiona l N may not have made a difference in increasing the P mineralization after Day 7. The large proportion of SRP in the water co lumn is significant because SRP is an extremely labile form of P that is readily bioavailable. After rainstorms in the Okeechobee Basin, there is a high possibility that substantial amounts of labile P can easily be transported out of ditched we tlands and increase P loading downstream

PAGE 62

49 waterways and possibly even to Lake Okeechobee. Table 3-3 illustrates that TKN concentrations were slightly higher under anaerobic condi tions, although few significant differences were observed. It is likely that there would be a higher concentration of reduced forms of N in the anaerobic treatment due to the utilization of N as an electron acceptor for microbial respiration. Results from this experiment show that approximately 90% of the P releases from litter after vegetative senescen ce can be predicted using the initial P tissue concentration of the senesced tissue (Figure 3-8), and ther efore P fluxes are likely to vary depending on the species type. Live tissue parameters such as the % NDF can also be used to predict short term P flux in the event senesced tissue is not available. Th e relationship between short term P release and the % NDF contained in the living tissue appears to be just as strong as the relationship between short term P release and the initial % P contained in the senesced tissue. This could indicate th at a substantial amount of P contained in the senesced vegetation is contained within the labile NDF fraction of the plant, and is released along with the sol uble sugars and carbohydrates. In addition, there were a greater number of significant relationships betw een short term P release and an initial live substrate quality characteristic such as the %SADF, C:N and %N. The relationship between short term P release and the initial P content in the live tissue may have not been as strong due to relatively high amount of P resorbtion from senesced tissue into live plant tissue. If changes in the P mass can be predicted with confidence using a live tissue parameter, the application of the data may be more useful since live vegetation is more abundant than senesced tissue throughout most of the year.

PAGE 63

50 The results from this short term leachi ng study are important because it is likely that short term P release from senesced ti ssue can be predicted with a high level of confidence for many different species found on historically isolat ed wetlands in the Okeechobee Basin, not just the four species used in this study. Therefore the release and transport of P may be managed using short te rm P release rates predicted by an initial substrate quality parameter in live or senes ced tissue; species that may release a high degree of P after senescence can be avoi ded. Future studies may indicate that P. hemitomon should be promoted along with hydrologi cal restoration because this species may reduce P transport and improve the water quality in the Okeechobee watershed compared to other species within th e wetland. It is likely however, that P. hemitomon will only persist on these landscapes if grazing is limited. If overgrazed, there may be a shift to an alternate species with a different P release rate. Relative to the original h ypotheses there were no strong trends to show that oxygen availability and N enrichment enhance P leaching, however P. hemitomon, the species with the lowest initial P concentration, did release the lowest amount of P released compared to other species of vegetation. Also the majority of P loss was attributed to leaching rather than microbial mineralization pr ocesses that seemed to take effect after 7 days of incubation. This study suggests th at leaching is an extremely important component of P cycling and nutrient movement especially in the fall when many species begin to senesce and there is a high proportion of standing d ead vegetation. This study not only provides an increased understanding of nutrient losses from senesced vegetation, but also provides some degree of species sp ecific leaching potential based on initial tissue

PAGE 64

51 P concentration. This information should pr ovide managers with better insight into wetland related Best Management Practices and reduction of nutrien t discharge offsite.

PAGE 65

52 CHAPTER 4 LITTER DECOMPOSITION AND LONG TERM PHOSPHORUS RELEASE Introduction Organic matter is made up primarily of carbon, and is a driver of many ecosystem heterotrophic processes. Orga nic carbon can function as a nu trient source for microbes, can adsorb toxic compounds or nutrients, a nd can provide an exchange capacity for cations (Cotrufo, 2006). The breakdown of or ganic matter or microbi al mineralization of nutrients is known as decomposition. Decom position occurs in all ecosystems and it is often accompanied by a net release of nutrients from senesced vegetation (Mitsch et al., 1989). The environmental factors influenc ing organic matter decomposition include moisture, temperature, electron acceptor av ailability, and pH (Mel illo, 1982; Benner, 1985; Qualls and Richardson, 2000). Plant tissue substrate quality partly dete rmines decomposition rates in addition to environmental factors listed above (Melillo, 1982; Berg, 1998; Vill ar, 2001). Substrate quality is determined partly by the recalcitrance or lability of the fiber fractions within the plant tissue. In addition, the amount of available nutrients su ch as N and P relative to the available carbon content is also a component of substrate quality. Substrate quality is an important parameter to consider when estimating decomposition rates of different plant species because it can either enhance or inhib it microbial colonization of the litter. The more labile organic polymers pr esent in the litter material such as starches and sugars, the easier microbes can utilize the litte r as a substrate for nutrients.

PAGE 66

53 In most cases nutrient enrichment has been shown to increase decomposition because N and P are often limiting factors co ntrolling microbial growth (Qualls, 1984; Taylor et al., 1989; Corstanje et al., 2006). Others have f ound that nutrient enrichment has no effect or can cause a lower rate of decomposition due to a possible C limitation (Berg et al., 1998; Carreiro et al., 2000; Villar et al., 2001). It may also be that the influence of nutrient content is variable dur ing different stages of decomposition. Lewis (2005) suggests that lit ter nutrient content may be the cont rolling factor during the initial phase of decomposition, or before 30% of the in itial mass is lost but thereafter available carbon may become the primary factor limiting decomposition rate. Lignin is the organic compound that is th e most recalcitrant or resistant to decomposition. It is the limiting factor of long-term decomposition because it can only be digested by a few organisms. Microbes are able to degrade lignin with the help of various enzymes (Benner, 1984); however it is extremely resistant to decomposition under anaerobic environments (Criquet et al ., 2001). Lignin helps provide structural support to plants so they can remain upright as well as provide a protective barrier against microbial attack (Hammel, 1997). Li gnin content or the li gnin/N ratio may be a better predictor of deco mposition rates compared to the P concentration or the C/N ratio (Melillo et al., 1982; Sinsabaugh et al ., 1993; DeBusk and Reddy, 1998; Rowland and Roberts, 1999; Carreiro et al., 2000). For instance litter with a high N concentration and low lignin content is likely to be a relatively labile subs tance with a rapid decomposition rate. Lignocellulose is another reca lcitrant C fraction. It is comprised of approximately 25% lignin and it forms a protective layer ar ound the cellulose tissu e, preventing rapid

PAGE 67

54 breakdown (Donnelly et al., 1990). Cellulose ge nerally decomposes faster than lignin, but slower than more labile fiber frac tions such as sugars, carbohydrates, and hemicellulose. Lignocellulose is formed as labile components of the vegetation decompose and lignin and cellulose decompos ition rates slow down (M elillo et al., 1982, 1989). Although lignocellulose formation ma y increase the litte r recalcitrance, decomposition will continue with the addition of nutrients or new substrates to the litter (Donnelly et al., 1990), if optimum conditi ons for decomposition such as warm temperatures, moisture, and oxygen are provided. When living plants are present in a we tland, they are likely accumulating P and incorporating it into their biom ass, however when the plant se nesces and litter enters the water column, the litter may become a P source (Moore and Reddy, 1994). During different phases of decomposition, various am ounts of P could potentially be released from litter depending on how much P is stored in the fiber fractions of the litter. The species type is likely to be an important factor when esti mating long-term P release since some species accumulate more nutrients than others (Hobbie, 1992; Knops, et al., 2002). McJanet et al. (1995) report ed significant differences in the N and P content of 41 different wetland plant species with N c oncentrations ranging from 0.25-2.1% and P concentrations of 0.13-1.1%. The age of the sene sced material may also play a role in the rate of decomposition. Younger tissues genera lly have less recalcitrant fractions and a higher proportion of N and P associated with their biomass compared to older plant tissues. The emergent macrophyte species used in th is study retain st anding dead biomass for a longer period of time (up to one year ) compared to submerged aquatic vegetation,

PAGE 68

55 which have virtually a nonexistent standing dead phase During the standing dead phase, biomass is likely exposed to leaching from rainfall as well as colonization and initial breakdown by aerobic microbes. Therefor e, standing dead litter may go through a significant degree of nutrient leaching before th e litter enters the water column. Leaching occurring before the litter enters the wa ter column could aff ect nutrient and fiber character of substrate quality and thereby in fluence decomposition ra te as well as longterm P release. The fiber quality, hydroperiod, and depth of inundation are likely to influence the decomposition rate of various species which may significantly affect the quantity of P that is released during decomposition. A higher degree of decomposition is likely to occur under aerobic conditions; however moistu re may be a limiting factor during the standing dead phase. Recalcitrant organic matter may be resistant to breakdown under anaerobic conditions, and may retain P in the recalcitrant portions of the litter, therefore litter may serve as a storage mechanism for P as organic matter continues to accumulate in the wetland and eventually form new soil. This may help decrease the P concentration in the surrounding water column as well as reduce P export from the wetland and nutrient loading downstream. Although the P assimilation rate by plants is a critical factor in assessing potential efficacy of wetland P storage, if plants are allowed to senesce in-situ the overall effectiveness of a wetland to immobilize and st ore P is also dependant on the release rate of P during litter decomposition. The rela tionship between decomposition rates and substrate quality, as well as substrate quality and plant species suggest that the type of plant species present in a wetland is also a cr itical factor regulati ng litter mineralization

PAGE 69

56 rates and long-term P storage. Therefor e, this chapter investigates long-term decomposition rates and compares these ra tes to various subs trate qualities and environmental parameters. A better understa nding of the relationship between tissue substrate quality and P litter mineralization wi ll assist in evaluating the implications of vegetative community change in response to hydrologic restoration on downstream P load to Lake Okeechobee. Hypotheses Decomposition rates will be lowest in wetland centers due to longer hydroperiod and anaerobic conditions. Tissue with high C:N ratios and or high residual fiber content will have slower decomposition rates than tissue with low C: N ratios or low residual fiber content. Materials and Methods Field Methods Litter decomposition rates were determined by measuring percent mass loss of standing dead vegetation collected from wetland center, wetland edge, and upland communities. Recently senesced sta nding dead biomass was collected from P. hemitomon, P. hydropiperoides (representative of the dominant vegetation in the wetland center), J. effusus (representative of the dominant edge species), and P. notatum (representative of the dominant upland sp ecies) during November 2004. The biomass was air dried in the laboratory for 10 days resulting in less variability in the litter moisture content without altering the chemi cal composition by oven drying. The litter was then manually broken into smaller segments approximately 7-9 cm in length. Six grams of air dried vegetation was placed in 15 cm x 15 cm litterbags made of 1mm screen mesh, and each litterbag was individually labeled with plastic labels. The initial

PAGE 70

57 amount of litter in each bag was based field litter biomass estimates and normalized to 15 x 15cm area. The litterbags were deployed on April 1st, 2005 in 3 isolated wetlands located within the same pasture. Litterb ags were deployed along a hydrological gradient from the center of the wetland to the upland in four different we tland zones defined as center, edge, transitional zone, and upla nd (Fig. 4-1), which had a hydroperiod of 100200 days / yr, 5-60 days / yr, and 0-15 days / yr respectively (Fig. 4-2). Five sets (one set for each time collection) of 8 litterbags (2 bags for of each of four species) were deployed in each zone within th e three wetlands to determine the influence of different environmental factors and substr ate quality on decomposition rates (Fig. 4-2). Each set of litterbags was tied together with monofilament fishing line, spaced evenly across the soil surface. Monofilament fishing line was att ached to the soil with a metal nail on the corner of each bag, and then cove red by 3 x 3 cm plastic nylon netting to prevent the cattle from eating or relocating the litterbags (Fig. 4-3). Litterbag coordinates were determined using GPS, and mesh coveri ngs were marked with one inch diameter PVC poles to aid in relocation. One set of litterbags from each zone in each wetland was harvested during each sampling period. Litterbags se ts were harvested June 1st, 2005, July 27th, 2005, December 8th, 2005, and April 5th, 2005, representing 2, 4, 8, and 12 months of exposure. The location of each set of litte rbags was found using GPS and a metal detector. Once a set of litterbags was located, they were placed in plastic Ziploc ba gs and stored in a cooler and transporte d to the laboratory.

PAGE 71

58 Figure 4-1. Litterbag distribution and de ployment locations in 4 hydrological zones within the wetland. -1.500 -1.000 -0.500 0.000 0.500 1.0004/1/04 7/10 / 04 10/18/04 1/26/0 5 5/6/05 8/ 1 4 / 05 11/22/ 0 5 3/2/06Stage (m) A = site of litterbag deployment Upland Edge Cente r Transitional

PAGE 72

59 0 20 40 60 80 100 120 140 160 180 200 CenterEdgeUplandHydrperiod (days/yr) 11/1/04 3/1/05 7/1/05 B Figure 4-2. Hydrological information for th e seasonally isolated wetlands on Larson Dixie Ranch. A) The yearly stage information in meters from April 1, 2004–March 10, 2006, B) the average hydrope riod for the center, edge, and upland zones during November, March, and July. Figure 4-3. Aerial view of the three historically isol ated wetlands on Larson Dixie Ranch where the litterbags were deployed. Larson West Larson East ( LE ) Larson South ( LS ) Average Area = 2.5 ha

PAGE 73

60 A B C D E F Figure 4-4. Litterbag experiment showi ng A) litterbags and netting in upland, B) litterbags attached to soil and covered by nylon netting as a precaution against cattle, C) a close up of litterbag filled with P. notatum, D) P. notatum growing in and through the litterbag after 8 months of exposure, E) the large amount of vegetation covering the bags in the upland after 8 months, F) vegetation covering netting in the transitional zone after 8 months.

PAGE 74

61 Laboratory Methods After litterbags were collected from the field and brought back to the laboratory, vegetation growing on the outside of the litt erbag was removed and sediment was rinsed out of the litterbags using tap water. Th e procedure for sediment removal was to dunk each litter bag into its own individual containe r of tap water, gently agitate the bag while submersed and let drain. This procedure was repeated three times. Litterbags were not rinsed in the field because site water was not always present at the time of collection. After rinsing, litterbags were cut open and remaining green vegetation and roots were removed. Litterbag contents were then placed in a labeled paper ba g and dried at 60 C for at least 72 hrs. After drying, litter was weighed to determine mass loss after each exposure time. In an effort to further reduce mass change errors associated with inclusion of field sediments in to the litter, litterbags depl oyed in the center were also sieved using a 250 m or #60 mesh to remove small particulate matter mostly representing mineral soil. After drying and weighing all samples were ground using a Wiley Mill and passed through a # 40 mesh. An alysis of total carbon and total nitrogen was determined using a Thermo Electron Flash EA1112 Nitrogen/Carbon analyzer. Tissue phosphorus content was determined us ing acid digestion of ashed tissue and analyzed using colorimetric procedures (Method 365.4; USEPA, 1993). An Ankom 200 Fiber Analyzer was used to quantify the ne utral detergent fiber (NDF) which is the starches, sugars and labile components associat ed with the plant tissu e, the acid detergent fiber (ADF) which is the hemicellulose fibe r fraction, the strong acid detergent fiber (SADF) which is the cellulose fiber fract ion, and the residual fiber or “lignin” percentages of the litter (Ankom Technology Corp., 1998a). Fiber analysis was only conducted on litter collected 12 months after deployment.

PAGE 75

62 The decomposition rates and P content mean s of all species were compared using Tukey-Kramer HSD (honestly significant diffe rences) method to determine significant differences between species decomposition and species P mass. Regression analyses and one way analysis of variance (ANOVA) were us ed to determine which substrate quality parameter, quantified in Chapter 2, was the be st predictor of % mass loss and long term P release after 12 months using the R2 and p-values. Results There were no significant differences in decomposition rates among the three wetlands when all species were combined. Ov erall, results indicate a slightly slower decomposition rate in Larson South (LS) comp ared to Larson East (LE) which had the highest percent mass loss of the three wetlands (Fig. 4-5). Litter decomposition in the four hydrologic zones (center, edge, transitional zone, and upland) did not vary significantly over the co urse of this study. The only significant difference between litter mass loss and hydrologic zone occurre d after 2 months when the center litterbags had a higher percent mass loss compared to the other three zones (Fig. 46). Decomposition rates over time were be st modeled by an exponential decay curve, with similar rates between zones. Although there were not any additional si gnificant differences in decomposition rate between the four hydrologic zones after the first 2 months, there were significant differences in decomposition rates among the four different plant species. Paspalum notatum had a significantly higher decomposition ra te over 12 months compared to the 3 other species of senesced vegetation During the first 2 months, there were not any significant differences among the decomposition rates of the four species. However, P.

PAGE 76

63 notatum had a significantly greater percent mass lo ss compared to the three other species during the 4 and 12 month sa mpling periods (Fig. 4-7). 40 50 60 70 80 90 100 024681012 Time (months)% Mass Remaining LE LW LS Figure 4-5. Litter decomposition of all sp ecies in all hydrologic zone among the 3 different wetlands, Larson East (LE), Larson West (LW), and Larson South (LS). Values represent mean 1 standard deviation. The greatest amount of decomposition occurred in P. notatum litter in the upland with roughly a 62% mass loss after 12 mont hs. The least amount of decomposition occurred in P. hydropiperoides in the edge (a 32 % mass lo ss after 12 months) (Table 41). An average of 43 % of the original l itter mass from each species within the four hydrological zones had decomposed after 12 mo nths. In addition, the highest rate of decomposition in every species was observe d during the first four months where approximately 34% of the original bioma ss had decomposed, while roughly only an additional 9% of the remaining biomass was lost during the next 8 months (Fig. 4-7).

PAGE 77

64 40 50 60 70 80 90 100 024681012 Time (months) Mass Remaining (%) center edge transitional upland Figure 4-6. Decomposition in 4 wetland zones over a 12 month period. Values represent mean 1 standard deviation. None of the initial substrate quality paramete rs characterized in Chapter 2, in either the live or senesced tissue, was highly corr elated with the mass loss among the 4 different species after 12 months in the field. The in itial substrate quality parameter in senesced tissue that most closely associated with the mass remaining after 12 months was carbon content, this relationship grew progressively stronger and wa s not significant after only 2 months however. Although the relationship wa s significant, it only explained 37% of the variability in the decomposition rate (R2 = .368 and a p-value < .0001). The substrate quality parameter in the live tissue which had the best association with mass loss after 12 months was the initial N/P ratio (R2 = .262 and a p-value < .0001), however this relationship was not consistent th roughout the course of this study. y = 84.815e-0.0408xR2 = 0.77 y = 88.61e-0.036x R2 = 0.69 y = 92.16e-0.046xR2 = 0.90 y = 91.37e-0.049xR2 = 0.89

PAGE 78

65 30 40 50 60 70 80 90 100 024681012 Time (months) Mass Remaining (%) P. hemitomon P. notatum P. hydropiperoides J. effusus Figure 4-7. Average litter decomposition of each species over a 12 month period. Values represent mean 1 standard deviation. There were not any significant differences in the % P remaining in any of the four hydrological zones throughout the entire 12 month study. The % P in the center remained relatively constant during 12 months of deployment. After 12 months, the % P contained in the litter was highe r in the edge and transitional zone compared to the center and the upland, although the differences in % P were not significant (Fig. 4-8). During the first 2 months, P. hydropiperoides released approximately 50% of the initial P contained in the biomass, while J. effusus released approximately 30% of its original P mass during the firs t 2 months in the field. Panicum hemitomon had a significantly greater % P remaining compared to the other 3 species surveyed during each sampling period over 12 months, while P. hydropiperoides had the significantly lowest % P remaining during every sampling period except after 4 months. In addition, P. y = 91.41e-0.039x R2 = 0.87 y = 89.51e-0.066x R2 = 0.85 y = 86.51e-0.035x R2 = 0.73 y = 90.74e-0.042x R2 = 0.85

PAGE 79

66 hemitomon accumulated approximately 50% more P than was originally contained in the biomass of the litter during the first 4 months of deployment (Fig. 4-9). Table 4-1. Species decomposition in each zone of litterbag deployment after 12 months. Values represent mean ( 1 standard deviation ). Species Zone Mass Loss (%) P. hemitomon center 42 7 edge 36 15 transitional 41 7 upland 36 8 P. notatum center 58 4 edge 48 9 transitional 58 5 upland 62 13 P. hydropiperoides center 36 16 edge 32 13 transitional 43 6 upland 39 4 J. effuses center 42 19 edge 36 10 transitional 42 10 upland 42 5 0 50 100 150 200 250 300 024681012 Time% P Remaining Center Edge Transitional Upland Figure 4-8. Change in % P in the 4 hydrol ogical zones over a 12 month period. Values represent mean 1 standard deviation.

PAGE 80

67 0 50 100 150 200 250 300 024681012 Time% P Remaining P. hemitomon P. notatum P. hydropiperoides J. effusus Figure 4-9. Change in % P in the 4 dominant species over a 12 month period. Values represent mean 1 standard deviation. Table 4-2. Significance and R2 values of the relationship betw een the change in P content and initial senesced substrate quality characteristics for each sampling period. Time (months) 2 4 8 12 Substrate Quality Parameter p-value R2 p-value R2 p-value R2 p-value R2 % SADF <.0001 .283 <.0001 .263 <.0001 .271 <.0001 .307 % C <.0001 .384 <.0001 .165 <.0001 .330 <.0001 .250 ln C:N <.0001 .479 <.0001 .520 <.0001 .483 <.0001 .544 ln N:P <.0001 .573 <.0001 .521 <.0001 .573 <.0001 .530 % N <.0001 .586 <.0001 585 <.0001 581 <.0001 .626 % ADF <.0001 .638 <.0001 .550 <.0001 .614 <.0001 .626 % Residual Fiber <.0001 .694 <.0001 .581 <.0001 .665 <.0001 .664 ln C:P <.0001 .719 <.0001 .685 <.0001 .720 <.0001 .701 % P <.0001 .818 <.0001 .738 <.0001 .806 <.0001 .787 % NDF <.0001 .813 <.0001 .767 <.0001 .802 <.0001 .796 When many initial substrate quality parameters in the senesced tissue were correlated with the change in P over time the strongest association was between the initial NDF fraction contained in the biomass and th e loss or gain of P mass through time (Table

PAGE 81

68 4-2) with an R2 of .796 after 12 months and a p-value <.0001 (Fig. 4-10). In addition, the correlation between these two parameters rema ined strong regardless of hydrologic zone, and the initial P content explaine d at least 77% of the variabilit y in the loss or gain in P mass regardless of sampling time. Figure 4-10. Correlation between initial NDF fr action in the senesced tissue and P loss or gain after 12 months. Initial substrate quality parameters of the live tissue characterized in Chapter 2, were also correlated with the long term changes in P mass in the senesced litter. The initial C:P ratio in the live tissue was the best parameter to predict P loss or gain after 12 months (R2 = 0.790 and p-value < .0001) (Table 4-3). The ability to predict the change in P using the initial C:P in the live tissue remained consistent throughout each sampling period. The second best parameter to predic t the change in P after 12 months was the -0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5 0.6 P loss / gain (g) 21 22 23 24 25 26 27 28 29 30 % NDF P. hemitomon P. notatum P. hydropiperoides J. effusus P loss/gain (g) = -1.77 + 0.07 % NDF R2 = .796

PAGE 82

69 initial %P contained in the live tissue. Figure 4-11 shows the strong correlation between the change in P mass and the initial C:P in the live tissue of the four dominant species. Table 4-3. Significance and R2 values of the relationship betw een the change in P content and initial live substrate quality ch aracteristics for each sampling period. Time (months) 2 4 8 12 Substrate Quality Parameter p-value R2 p-value R2 p-value R2 p-value R2 N:P <.0001 .280 .0001 .143 <.0001 .242 <.0001 .217 C:N <.0001 .437 <.0001 .501 <.0001 .460 <.0001 .465 % N <.0001 .480 <.0001 .524 <.0001 .499 <.0001 .496 % C <.0001 .498 <.0001 .546 <.0001 .503 <.0001 .586 % ADF <.0001 .548 <.0001 .553 <.0001 .544 <.0001 .592 % Residual Fiber <.0001 .593 <.0001 .455 <.0001 .557 <.0001 .547 % NDF <.0001 .580 <.0001 .675 <.0001 .602 <.0001 .655 % SADF <.0001 .719 <.0001 .741 <.0001 .726 <.0001 .753 % P <.0001 .834 <.0001 .715 <.0001 .810 <.0001 .786 C:P <.0001 .833 <.0001 .726 <.0001 .813 <.0001 .790 The %N remaining in the li tter doubled on average and only a slight decrease in N was seen during the first 2 months in every zone excluding the center. The N content in the litter increased from 2-8 months in every zone, while a decrease in the %N was observed from 8-12 months in the upland and transitional zones (Fig. 4-12). There were not any significant differences seen between hydrological zones over the 12 month study. The only significant difference between hydrol ogical zones was seen after 2 months when the center had a significantly higher % N remaining.

PAGE 83

70 Figure 4-11. Correlation between initial C:P in the live tissue and P loss or gain after 12 months. 0 50 100 150 200 250 300 024681012% N Remaining Center Edge Transitional Upland Figure 4-12. Change in litter %N (all species combined) among different wetland hydrologic zones over time. Values repr esent mean 1 standard deviation. -0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5 0.6 P loss/gain (g) 140 160 180 200 220 240 260 280 300 C:P live P loss/gain (g) = 1.23 0.0051 C:P live R 2 = 0.790 P. hemitomon P. notatum P. h y dro p i p eroides J e ff usus

PAGE 84

71 The %N remaining in P. notatum and J. effusus was significantly higher than the %N in P. hemitomon and P. hydropiperiodes over the course of 12 months (Fig. 4-13). The C:N for each species was roughly the invers e of the %N remaining in the litter, and after 8 months exposure, changes in C: N in each species appeared to stabilize, and converged around 15-25 after 12 months Polygonum hydropiperoides had the lowest C:N after2 and 4 months, while P. notatum had the lowest C:N afte r 8 and 12 months. In addition, the average %N in the litter doubled after 1 year in the field. 0 50 100 150 200 250 300 024681012% N Remaining P. hemitomon P. notatum P. hydropiperoides J. effusus Figure 4-13. Change in litter %N among speci es over time. Values represent mean 1 standard deviation. The residual fiber content of the litter after 12 months was significantly higher than the initial fiber content for every sp ecies (Fig. 4-14). After 12 months, P. hydropiperoides had the highest residual fiber fraction among the 3 species surveyed in the upland, edge, and transitional hydrologic zones. In addition, P. hydropiperoides had

PAGE 85

72 a significantly higher residual fiber fraction compared to J. effusus and P. hemitomon in the center zone. There were also significant differences in residual fiber content within each species among the 4 hydrologic zones. Re sidual fiber content in bags deployed in the upland and transitional hydrologic zones were genera lly significantly lower in residual fiber content than litte rbags located in the center a nd edge hydrologic zone (Fig. 4-15). 0 5 10 15 20 25 30 35 40 45 50 P. hemitomonP. notatumP. hydropiperoides J. effusus% Residual Fiber 0 months 12 months Figure 4-14. Comparison of residual fiber cont ent of initial and 12 month exposed litter among four species tested. Values repr esent mean ( 1 standard deviation).

PAGE 86

73 0 10 20 30 40 50 60 P. hemitomonP. notatumP. hydropiperoides J. effusus Species% Residual Fiber Center Edge Transitional Zone Upland Figure 4-15. Residual fiber content of speci es in each hydrologic zone after 12 months. Values represent mean ( 1 standard deviation). Discussion An average of 43% of the original litter had decomposed after 12 months exposure. The rate of decomposition in this study was lower than the 50% loss in litter mass observed within 4 months of a litterbag study of J. effusus in a freshwater marsh (Boyd, 1971), but greater than the 23% mass loss in J. effusus observed over 268 days in an additional freshwater marsh decom position study (Kuehn et al., 2000). Lower decomposition rates in our study may be attributed to the initial low N content of the litter or perhaps a lower degree of wetting and dr ying compared to Boyd’s study. All species in this study were N limited initially under aerobi c conditions for the first 4-8 months, and higher N content could have possibly enhanced decompositi on rates (Qualls, 1984; Taylor et al., 1989; Corstanje et al., 2006) Most decomposition took place during the first 4 months of the study. Other studies ha ve reported a leveling o ff or a lower rate of

PAGE 87

74 decomposition after the first 4 months, which may indicate that most of the labile components associated with the biomass of th e senesced tissue have been lost by this time (Boyd, 1971; Berg et al., 1998; Villa r et al., 2001; Davis et al., 2003). Regardless of zone (center, edge, transitional, or upland), P. notatum decomposed faster than P. hemitomon, P. hydropiperoides, and J. effusus, and lost an average of 56% of the biomass over the 12 month period (Fig. 4-7). Results from initial characterization of litter quality indicate that P. notatum also had the lowest initial amount of residual fiber. A low percentage of re sidual fiber in the biomass of P. notatum could have been responsible for a higher degree of decomposition compared to other species. Lower litter recalcitrance can increase the likelihood of microbial colonization since carbon compounds are more bioavailable and ma terial can be broken down more easily. Results from Chapter 2 indicate that P. hydropiperoides had the greatest initial NDF (non-detergent fiber) fracti on as well as the most residu al fiber compared to the 3 other species of vegetation. S ugars, starches and other com ponents that are easily broken down make up the NDF fraction. It is likely that the leaves of this species are relatively labile and contain the majori ty of the NDF fraction, while the woody stems of the plant contain most of the residual fiber. The high NDF fraction as we ll as the significantly high N and P content of this species could explain why P. hydropiperoides had the highest degree of decomposition for the first 2 months (Fig. 4-7). After this fraction was consumed by microbes, the decomposition rate decreased because the remaining biomass was primarily composed of residual fi ber. After 12 months exposure, P. hydropiperoides had the highest amount of residual fiber compar ed to the other species in each zone of litterbag deployment.

PAGE 88

75 Panicum hemitomon had a relatively high percentage of residual fiber initially in the senesced tissue and the lowest N and P c ontent, which may explain why this species had the lowest amount of decomposition prim arily during the first 2 months. Slow decomposition of species with a low nutrien t content could be supporting evidence that nutrients control decomposition un til more than 30% of the original mass is lost, similar to findings by Lewis (2005). Although litter quality appears to have some effect on decomposition rates, a strong relationship between a live or senesced initial substrate quali ty parameter and the % mass loss after 12 months was not seen. Wh en initial substrate quality parameters were correlated with % mass loss after 2 months, a strong association was still not observed. It was surprising that decomposition rate was not more significantly influenced by hydrological zones of the wetland. It was originally thought that increased oxygen availability in the upland would promote decomposition since oxygen is the preferred electron acceptor and yields the most energy during microbial catabo lism. Results from this experiment suggest that species type is what primarily determines short-term P release, while the location may influence d ecomposition and P release over a time period greater than one year. Presence of cattle is an environmental factor on the research site that may significantly impact decomposition rate of litter, especially in the center where soil is primarily organic and more influenced by trampling. Cattle may actually encourage decomposition in the center by stepping on li tterbags, pushing them underneath the soil surface, and breaking the litter into smaller fragments. After the litterbags had been

PAGE 89

76 deployed in the field for 12 months the bags located in the cent er were buried up to 30 cm beneath the soil surface. While litterbags located in the 3 other hydrological zones were covered with a mat of vegetation, they we re not pushed underneath the soil surface by cattle, indicating that the bags in the cent er may have had environmental factors which could influence mass loss that were not an issu e in other zones (Fig. 4-16). It is likely that mechanical fragmentation was not seen in the edge because the soil was much firmer in this zone compared to the center. Fiber analysis results from the litter collected after 12 m onths of exposure indicate that the residual fiber conten t after 12 months was greates t in the center compared to other hydrologic zones. Results support the id ea that residual fiber or “lignin” is not broken down under anaerobic conditions. Th is may be the strongest evidence for potentially less decomposition in the center compared to the upland, and may suggest that greater mass was lost in the center due to mech anical fragmentation of the as a result of cattle trampling. Lower % mass loss of litte r in the wetland edge compared to the wetland center may indicate that cattle have a greater effect on litter decomposition this zone (Table 4-1). In the future it may be beneficial to repeat this study with cattle exclosures in order to quantify the impact that cattle may have on decomposition and P release in isolated wetlands Many different confounding physical and environmental factors present in the field may have been the reason why a clear relationship between initial substrate quality parameters and th e percent mass remaining the litter after 12 months was not seen.

PAGE 90

77 A B Figure 4-16. Litterbags collected after 12 months from A) wetland center, which were approximately 20-30 cm underneath the soil surface and B) wetland edge, which were on top of the soil surface. Litter often alternates between releasing and absorbing nutrients as it decomposes (Jordan et al., 1989). The N content in each species was higher after 12 months compared to the initial %N contained in th e litter at the beginni ng of the study (Fig. 413). The %N remaining in the tissue dec lined in some species during the 12 month period. The decrease in the %N re maining is likely due to microbial mineralization/immobilization processes. It is possible that a leaching event could have decreased the %N contained in the tissue during the first 2 months of exposure. It is thought that N was assimilated by microbes fr om the water column, or sorbed to the surface of the litter as particles from manur e or urine of cattle in the upland and transitional zone where standi ng water was rarely present, causing and increase in %N remaining. An increase in N, or a decrease in the C/N ratio has been reported in other decomposition studies (Brinson, 1977; Kuehn et al., 2000; Davis et al ., 2003). Villar et al. (2001) reported that the N c ontent of litter increased 7 times after 2 years in the field while the N concentration doubled over 1 ye ar for each of the species in this study.

PAGE 91

78 A slight decrease in the %N remaining th e edge (0-2 month period) and in the upland and transitional (0-2 and 8-12 month periods) may suggest leaching or microbial mineralization during the dry months of the year (Dec. – June) instead of N accumulation relative to C losses seen in the wetland edge and center. Panicum hemitomon acted as a P sink throughout the course of this study, while other species alternated between a P source a nd a P sink. Some species of litter in this study, P. hemitomon and P. notatum, had a greater amount of %P remaining after 12 months compared to the %P in itially in the litter, while P. hydropiperoides and J. effusus had lower or roughly the same %P remaining in the litter after 12 mont hs (Fig. 4-9). It was surprising not to see a decline in the %P remaining in the litter in every species due to mineralization processes; however similar results of an increas e in the P mass have been reported from past decomposition studi es (Kuehn et al, 2000; Villar et al., 2001; Davis et al., 2003). It is likely that the presence of uncontroll ed environmental factors such as cattle, cycles of wetting and drying, changing moistu re content, and various temperatures in different hydrological zones c ould have had a significant imp act on the %P remaining in the litter after 12 months. The increase in P content throughout the study was likely caused by microbial assemblages adhe ring to the litter or manure pa rticles. It is not likely that soil particles adhering to the litter woul d have caused an increase in P. Figure 1-7 reports the mg P/m2 stored in the center, edge, a nd upland soil which is less than the5200mg P/m2 stored in P. hemitomon, the species with the lowest P content. Findings from the laboratory leaching study provided an estimate of short-term P release (described in Chapter 3) and P rel ease rates were well correlated with initial P

PAGE 92

79 content in the senesced tissue. When P rel ease was evaluated at the field scale and over longer periods, the initial %NDF contained in the senesced tissue had the strongest relationship with long term change in P ove r 12 months. It is possible that the NDF fraction of the litter has a high P content and as labile sugars and starches are consumed or leached much of the P associated with th at fiber fraction is re leased, hence the strong association between these 2 parameters. The relationship between the initial C/P ratio in the live tissue and the %P remaining after 12 mo nths was just as str ong as the relationship between the initial %NDF in the senesced tissue and %P remaining. This information could be beneficial in managing seasonally isolated wetlands in the Okeechobee Basin because possible to predict the %P remain ing after 12 months using tissue quality information from live or senesced vegetation. Differences in P release among species during this study suggest that species composition of a wetland may significantly in fluence P storage capacity in litter and soils. A marsh dominated by P. hemitomon would have significantly %P remaining in the litter compared to a marsh dominated by P. hydropiperoides. The large amount of P released by P. hydropiperoides after senescing could indicate th at much of the P in this species was associated with the NDF fiber fraction and therefore relatively labile compared to the P contained in P. hemitomon which may be associated with residual fiber fraction and remain incorporated in the litter for a longer period of time. Because wetland species often have different hydrologic tolerances, changes in hydrologic regime of a wetland may result in changes in species dominance within the wetland The relationship between species specific P release rates and factors that may result in shifts in species composition could have significant effect on water quality of the isolated

PAGE 93

80 wetland and potentially influence efforts to address P loading to Lake Okeechobee. Therefore, the resulting effects of hydrologi c restoration on vegeta tive community shifts should be considered to determine implicati ons of this management action on P storage capacity of wetlands. In summary, relative to the original hypotheses, there was no significant difference between the decomposition rates obse rved in the 4 wetland hydrologic zones. Although the results from this study suggest th at wetland centers have relatively the same rate of decomposition compared to uplands, these results may have been due to many confounding environmental factors including th e presence of cattle which may cause mechanical fragmentation of the litter. If th ese factors had not been present it is likely that wetland centers would have lower amounts of decomposition compared to upland hydrologic zones. Alternatively, if findings of this study were not confounded by other factors, it is possible that expected differences in deco mposition rate among hydrologic regimes require longer time periods (greater than one year) to become evident. In addition, we can conclude that both P. hemitomon and P. hydropiperoides had the lowest decomposition rates compared to P. notatum and J. effusus It is believed that high residual fiber content limited P. hydropiperoides decomposition, while P. hemitomon decomposition may have been limited by relativ e high residual fiber content and a low N and P concentration which possibly decr eased the rate of litter breakdown.

PAGE 94

81 CHAPTER 5 SYNTHESIS AND CONCLUSIONS Four plant species dominant on the site, bu t located in different hydrological zones within the wetlands and adjacent pasture were evaluated to determine the rate of P release during the first and second phases of decom position. From these results, correlations were made between a substrate quality char acteristic and P release from litter over time due to decomposition and leaching. Findings suggest a strong relati onship between some initial substrate quality parameters (%NDF, C:P, or P content) and short and long term P release that can potentia lly be used as a predictor of P stability and storage under various restoration techniques a nd management practices. Results pertinent to plant tissue characteri zation (Chapter 2) i ndicated that live and senesced tissue of species surveyed contained different nutrient and fiber contents. Live tissue had a significantly higher N and P content, but had a lo wer residual fiber content compared to senesced tissue of the same speci es. These results may indicate that there was either a leaching event that took place th at removed some N, P, and labile fiber fractions associated with se nesced tissue or some nutrien t resorbtion took place before senesced tissue was collected. In subseque nt chapters, it was re vealed that nutrient content and labile fiber of an individual species can signifi cantly influence the amount of P released within a 12 month period. Info rmation provided by characterizing live and senesced plant tissue from different vegeta tive species was important to assess which tissue parameters are strongly correlated to P retention in an attempt to predict the amount of P released from dominant species present.

PAGE 95

82 Species type was the major factor that infl uenced short-term P release (Chapter 3). Despite the original hypotheses, redox condition and nitrogen enrichment did not seem to have a significant effect on P leaching over the 17 day period, however there were significant differences in P leaching among the four dominant species investigated. Panicum hemitomon the species that contained the lowe st amount of P initially, released a much lower amount of P compared to P. hydropiperoides, the species that had the greatest initial P content. Thes e results suggest that a shift in dominant species as a result of hydrological restoration or other mechanism could significantly a ffect the amount of P leached from senesced vegetation and coul d consequently impact water quality of wetlands as well as Lake Okeechobee. Below is a hypothetical situat ion comparing short and moderate-term P release from four dominan t species present in different hydrological zones on the site (Table 5-1). Panicum hemitomon was the only species that continually increased the %P remaining in the bioma ss throughout 12 months. The amount of P remaining in the other 3 species fluctuated c onsiderably over time and therefore the net release or assimilation of P for the 3 remain ing species would be different depending on the time period. In the leaching study, the greatest amount of P was released from each species of senesced vegetation within 2-72 hrs after treat ment water was added to containers, and P leaching decreased considerably or ended after 72 hours of incubation. Air dried, senesced vegetation had a high leaching pot ential upon initial cont act with treatment water. In a field situation, a large quantity of P could potentially be released into the water column during a heavy rainstorm afte r senescence or the first time the wetland becomes inundated af ter litterfall.

PAGE 96

83 Table 5-1. Variation in shor t and moderate-term P assimilation or release from 1m 2 of litter over 2 months for the 4 dominant species investigated. Species Initially 250 g litter / m2 P. hemitomon P. notatum P. hydropiperoides J. effusus Initial P (g) 5.2 20.7 32.0 20.1 P remaining after leaching (g) 5.0 9.2 9.1 9.0 Litter mass remaining after 2 months (g) 215.3 208.1 198.8 206.0 P remaining after 2 months decomposition (g) 5.7 20.9 13.8 14.1 Net release (-) or assimilation (+) of P in litter (g) + 0.5 + 0.2 18.2 16.0 There was a strong relationship between ini tial P content of sene sced tissue and the amount of P released over a 17 day pe riod. These relationships had high R2 values, indicating that initial tissue P content in se nesced tissue could possibly be used as a predictor of short-term P rele ase from leaching. From this finding it may be possible to predict short-term P release across a variety of species present on isolated wetlands within the Okeechobee Basin if the P c ontent of senesced species is known. Because senesced tissue is often hard to find during the summer months in isolated wetlands in the Okeechobee Basin, it may be easier to characterize live tissue of different species. The initial P content of live tissue was not the best substrate quality parameter to predict P leaching rates (as compared to se nesced tissue), however there was a strong relationship between short term P release a nd the initial % non-detergent fiber (%NDF) in the live tissue. The litter decomposition study (Chapter 4) at tempted to predict long-term P release after litter had been exposed to field condi tions for 1 year. A significant relationship between the %P remaining in the litter after 12 months and the initial NDF fraction in the

PAGE 97

84 senesced tissue was observed. In addition, phosphorus cont ent remaining in the litter after 12 months was also predictable using the C:P in the live tissue. Phosphorus content among the four species seemed to fluctuate considerably over 12 months; resulting in an increase in %P re maining in some species while others had a lower %P remaining compared to their initial condition. It is thought that an increase in the microbial biomass or manure particles adhe ring to the litter even after washing may have caused an increase in P content of litter tissue. Due to many environmental factors introduced in the field that were not present in the lab, a strong relationship be tween an initial substrate qua lity parameter in either the live or senesced tissue and the % mass loss of the litter could not be drawn, although it appeared that litter quality had some eff ect on decomposition since one of the four dominant species did have a si gnificantly greater rate of decomposition. It is possible that cattle present on the research site had a significant influence on the litter quality or decomposition processes even after 2 months in wetland centers. In rejection of the origin al decomposition hypotheses li tterbags deployed in the wetland center did not have the slowest d ecomposition rate among the four hydrologic zones. Landscape position did not seem to significantly influence decomposition rates as over the 12 month study period as originally thought. It is be lieved that cattle may have a higher impact than biogeochemical processes, or that the role of hydrology may not be significant until later phases in the decomposition process. The presence of cattle is thought to have enhanced decomposition in th e center due to mechanical breakdown of the litter within this zone. There is little previous research on the effect that cattle may have on ecosystem processes and P release. Future studies may want to consider cattle

PAGE 98

85 exclosures in each hydrologic zone to bett er compare litter decomposition and nutrient changes in areas where cattle are presen t compared to areas where they are not. Plant species type appeared to be the major factor influencing decomposition and nutrient release rates in isolat ed wetlands during the time pe riod investigated. Parameters such as the NDF fraction, as well as the quantity of P in the litter are unique to various species and were significant factors determin ing the rate of decomposition and nutrient release. Of the four species evaluated, P. notatum had a significantly higher rate of decomposition compared to the other three species. This species also had the lowest initial residual fiber fraction. There were no significant diffe rences identified in decomposition rate of the three remaining species, however there were signif icantly higher percent mass losses observed at certain sampling periods in some species compared to others. Polygonum hydropiperoides had the lowest decomposition rate co mpared to other species. This species had a significantly gr eater residual fiber content in itially which may have limited decomposition. Panicum hemitomon had a slightly higher mass loss over the course of 12 months compared to P. hydropiperoides. In addition, P. hemitomon had the lowest initial N and P content compared to the thre e other species. In regards to the second hypothesis in Chapter 4, it seems that fiber qua lity and nutrient content may play a role in decomposition since species with the lowest initial residual fiber content ( P. notatum) had the fastest decomposition rate, while speci es with the highest residual fiber content (P. hydropiperoides) and the lowest initial nutrient content ( P. hemitomon) had the lowest mass losses.

PAGE 99

86 Interactions between plant s ubstrate quality, environmenta l conditions and effect of cattle disturbance on wetland P storage are not well understood and this study provides additional insight into this dynamic topic. Important findi ngs in this study include a surprisingly high amount of P flux from litter sh ortly after water column exposure. On average, approximately 46% of the P contained in the litter was released within the first 2 – 72 hrs. In addition, P. hemitomon released a fraction of the P that was released by the three other species. Further studies may indicat e that this species s hould be promoted to optimize wetland efficacy as a BMP while acco mpanied with hydrological restoration to increase P storage, however in order to promote P. hemitomon cattle grazing must be limited.. Panicum hemitomon may have the ability to reduce P transport out of the wetland which will subsequently reduce nutri ent loading and eutrophication in Lake Okeechobee compared to other species such as P. hydropiperoides The results presented in this study also pr ovide insight on the effect cattle ma y have in isolated wetlands and P storage in senesced vegetation, as well as valuable information on improving the water quality of the Okeechobee Basin and increasing P storage in isolated wetlands with the use of a particular wetland species.

PAGE 100

87 LIST OF REFERENCES Anderson, D.L. and E.G. Flaig. 1995. Agricu ltural best management practices and surface water improvement and mana gement. Water Sci. Tech. 31, 109-121. Ankom Technology Corporation. 1998a. Method for determining Acid Detergent Fiber, Neutral Detergent Fiber and Crude Fibe r, using the Ankom Fiber Analyzer. Ankom Technology Corporation, 14 Turk Hill Park, Fairport New York 14450, USA. Benner R., M.A. Moran, R.E. Hodson. 1985. Effects of pH and plant source lignocellulose biodegr adation rates in two wetland ecosystems, the Okefenokee Swamp and a Georgia Salt Marsh. Limnology and Oceanography 30, 489-499. Benner, R., S.Y. Newell, A.E. Maccubin, a nd R.E. Hodson. 1984. Relative contributions of bacteria and fungi to rates of degr adation of lignocellulo sic detritus in saltmarsh sediments. Appl. Environ. Microbiol. 48, 36-40. Berg, B., Jansson, P-E., Meentenmeyer, V ., and Krantz, W. 1998a. Decomposition of tree root litter in a climatic transect of coniferous forests in northern Europe: A synthesis. Scand. J. For. Res. 13, 402-412. Berg, M.P., Kniese, J.P., Zoomer, R., and Verhoef, H.A. 1998. Long-term decomposition of successive organic strata in a nitrogen saturated Scots Pine forest soil. For. Ecol. Mgt. 107, 159-172. Bottcher, A.B., T.K. Tremwel, and K.L. Campbell. 1995. Best management practices for water quality improvement in the La ke Okeechobee watershed. Ecological Engineering 5, 341-356. Boyd, C.E. 1971. The dynamics of dry matter and chemical substances in a Juncus effusus population. American Midland Naturalist. 86, 28-45. Brezonik, P.L., and D.R. Engstrom. 1998. Mode rn and historic accumulation rates of phosphorus in Lake Okeechobee, Flor ida. Journal of Paleolimnology 20, 31–46. Brinson, M.M. 1977. Decomposition and nutrient ex change of litter in an alluvial swamp forest. Ecology 58, 601-609. Carpenter, S.R., N.F. Caraco, D.L. Correll, R.W. Howarth, A.N. Sharpley, V.H. Smith. 1998. Nonpoint pollution of surface wate rs with nitrogen and phosphorus. Ecological Applications 8, 559-568.

PAGE 101

88 Carreiro, M.M., Sinsabaugh, R.L., Repert D.A., Parkhurst, D.F. 2000. Microbial enzyme shifts explain litter decay re sponses to simulated nitrogen deposition. Ecology 81, 2359-2365. Corstanje, R., K.R. Reddy, K.M. Portier. 2006. Typha latifolia and Cladium jamaicense litter decay in response to exogenous nut rient enrichment. Aquatic Botany 84, 7078. Cotrufo, M.F. 2006. Quantity of standing litter: A driving factor of root dynamics. Plant and Soil 281, 1-3. Criquet, S., Farnet, A.M., Tagger, and S., LePetit, J. 2001. Annual variations of phenoloxidase activity in evergreen oak li tter: Influence of certain biotic and abiotic factors. Soil Biochemistry 32, 1505-1513. Davis, F.E. and M.M. Marshall. 1975. Chemical and biological inve stigations of Lake Okeechobee January 1973 June 1974. Tec hnical Publication 75-1. South Florida Water Management District, West Palm Beach, Florida. Davis, S.E., C. Corronado-Molina, D.L. Childers, and J.W. Day. 2003. Temporally dependent C, N, and P dynamics associated with the decay of Rhizophora mangle L., leaf litter in oligotrophic mangrove wetlands of the Southern Everglades. Aquat. Bot. 75, 199-215. Day, F.P. 1982. Litter decomposition rates in the seasonally flooded Great Dismal Swamp. Ecology 63, 670-678. DeBusk, W.F. and K.R. Reddy. 1998. Turnover of detrital organic carbon in a nutrientimpacted Everglades marsh. Soil Sci. Soc. Am. J. 62, 1460–1468. DeBusk, W.F. and K.R. Reddy. 2005. Litter dynamics and decomposition in a phosphorus enriched everglades marsh. Biogeochemistry 75, 217-240. Dolan, T.J., S.E. Bayley, J. Zoltek, A.J. Hermann. 1981. Phosphorus dynamics of a Florida freshwater marsh receiving treated wastewater. Journal of Applied Ecology 18, 205-219. Donnelly, P.K., Entry, J.A., Crawford, D.L., and Cromack, K.C., Jr. 1990. Cellulose and lignin degradation in forest soils: Response to moisture, temperature, and acidity. Microb. Ecology 20, 289-295. Federico, A.C., K.G. Dickson, C.R. Krat zer, and F.E. Davis. 1981. Lake Okeechobee water quality studies and eu trophication assessment. Re port, South Florida Water Management District, West Palm Beach, Florida.

PAGE 102

89 Fenchel, T., G.M. King, and T.H. Blac kburn. 1998. Bacterial biogeochemistry: The ecophysiology of mi neral cycling, 2nd ed. Academic Press, San Diego, CA, USA. Findlay, S., P. Groffman, a nd S. Dye. 2003. Effects of Phragmites australis removal on marsh nutrient cycling. Wetlands Ecology and Management 11, 157-165. Findlay, S., P. Groffman, a nd S. Dye. 2003. Effects of Phragmites australis removal on marsh nutrient cycling. Wetlands Ecology and Management 11, 157-165. Flaig, E.G. and K.R. Reddy. 1995. Fate of phosphorus in the Lake Okeechobee watershed, Florida, USA: overview and recommendations. Environmental Engineering 5, 127-142. Florida Department of Environmental Prot ection. 2001. Total Maximum Daily Load for Total Phosphorus – Lake Okeec hobee. [Online] Available at: http://www.dep.state.fl.us/water/wqssp/lakeo_tmdl.htm (modified 01 Jul 2004; accessed 21 Aug 2004; verified 20 D ec 2004). Florida Department of Environmental Protection, Tallahassee, Florida. Fraser, L.H. and J.P. Karnezis. 2005. A compar ative assessment of seedling survival and biomass accumulation for fourteen we tland plant species grown under minor water-depth differences. Wetlands 5, 520-530. Haan, C.T. 1995. Fate and transport of phosphorus in the Lake Okeechobee Basin, Florida. Ecological Engineering 5, 331-339 Hammel, K.A. 1997. Fungal degradation of lig nin. In Cadisch, G. and Giller, K.E. (eds.) Driven by Nature: Plant Litter Quality and Decomposition. University Press, Cambridge, UK, pp. 33-45. Havens, K.E and C.L. Schelske. 2001. Th e importance of considering biological processes when considering total maximum daily loads (TMDL) for phosphorus in shallow lakes and reservoirs Environmental Pollution 113, 1-9 Hiscock, J.G., C.S. Thourot, and J. Za ng. 2003. Phosphorus budget-land use relationships for the northern Lake Okeechobee watershe d, Florida. Ecological Engineering 21, 63-74. Hobbie, S. E. 1992. Effects of plant species on nutrient cycling. Trends in Ecology & Evolution 7:336–339. Jordan, T.E., D.F. Whighamm, D.L. Correll. 198 9. The role of litter in nutrient cycling in a brackish tidal marsh. Ecology 70, 1906-1915. Jordan, T.E., D.F. Whighamm, K.H. Hofm ockel, M.A. Pittek. 2003. Nutrient and sediment removal by a restored wetland r eceiving agricu ltural runoff. J. Environ. Qual. 32, 1534–1547.

PAGE 103

90 Knops, J. M. H., K. L. Bradley, and D. A. Wedin. 2002. Mechanisms of plant species impacts on ecosystem nitrogen cycling. Ecology Letters 5:454–466. Kuehn K.A., M.J. Lemke, K. Suberkropp, R.G. Wetzel. 2000. Microbial biomass and production associated with decaying leaf litter of the emergent macrophyte Juncus effusus. Limnology and Oceanography 45, 862-870. Kuusemets, V. and U. Mander. 2002. Nutr ient flows and management of a small watershed. Landscape Ecology 17, 59–68. Lewis, C.G. 2005. Linkages Among Vegetative Substrate Quality, Biomass Production, and Decomposition in Maintaining Ever glades Ridge and Slough Vegetative Communities. Masters Thesis. University of Florida. Gainesville, FL. McJanet, C.L., P.A. Keddy, F.R. Pi ck. 1995. Nitrogen and Phosphorus tissue concentration in 41 wetland plants: A co mparison across habitats and functional groups. Functional Ecology 9, 231-238. McKee, K.A. 2005. Predicting Phosphorus Stor age in Historically Isolated Wetlands within the Lake Okeechobee Priority Basins. Masters Thesis. University of Florida. Gainesville, FL. Melillo, J.M., J.D. Aber, A.E. Linkins, A. Ricca, B. Fry, and K.J. Nadelhoffer. 1989. Carbon and nitrogen dynamics along the d ecay continuum: Plant litter to soil organic matter. Plant and Soil 115, 189-198. Melillo J.M., J.D. Aber, J.F. Muratore. 1982. Nitrogen and lignin control of hardwood leaf litter decomposition dynamics. Ecology 63, 621-626. Mitsch, W. J., B. C. Reeder, and D. M. Kalr er. 1989. The role of we tlands in the control of nutrients with a case st udy of western Lake Erie. p.129–157. In W. J. Mitsch and S. E. Jorgensen (eds.) Ecological Engi neering: an Introduction to Ecotechnology. John Wiley & Sons, Inc., New York, NY, USA. Moore, P.A. and K.R. Reddy. 1994. Role of Eh and pH on phosphorus geochemistry in sediments of Lake Okeechobee, Florida. Journal of Environmental Quality 23, 955964. Qiu S., A. McComb., R. Bell., an d J. Davis. 2005. Leaf-litte r application to a sandy soil modifies phosphorus leaching over the wet season of southwestern Australia. Hydrobiologia 545, 33–43. Qualls, R.G. 1984. The role of leaf litter nitrogen immobilization in the nitrogen budget of a swamp stream. J. Environ. Qual. 13, 640–644.

PAGE 104

91 Qualls, R.G. and C.J. Richardson. 2000. Phosphorus Enrichment Affects Litter Decomposition, Immobilization, and So il Microbial Phosphorus in Wetland Mesocosms. Soil Sci. Soc. Am. J. 64, 799–808. Raisin, G.W. and D.S. Mitchell. 1995. The us e of wetlands for the control of non-point source pollution. Wat. Sci. Tech. 32, 177-186. Reddy, K.R., O.A. Diaz, L.J. Scinto, M. Ag ami. 1995. Phosphorus dynamics in selected wetlands and streams of the Lake Okeec hobee Basin. Ecological Engineering 5, 183-207. Reddy, K.R, R.G. Wetzel, and R.H. Kadlec. 2005. Biogeochemistry of phosphorus in wetlands. In Sims, J.T. and Sharpley, A. N. (eds) Phosphorus: Agriculture and the Environment. American Society of Agronomy Inc., Madison, WI, p. 263-316. Redfield, A.C. 1958. The biochemical control of chemical factors in the environment. Am Sci. 46, 205-221. Richardson, C.J. 1985. Mechanisms cont rolling phosphorus retention capacity in freshwater wetlands. Science 228, 1424-1427. Robertson, A. 1988. Decomposition of mangrove l eaf litter in Australia. Journal of Experimental Marine Biology and Ecology 116, 235-247. Rowland, A.P., and J.D. Roberts. 1999. Ev aluation of Lignin and Lignin Nitrogen Fractionation Following Alternative De tergent Fiber Pre-treatment Methods. Commun. Soil Sci. Plant Anal. 30 (1&2), 279-292. Ryan, D.F., and F.H. Bormann. 1982. Nutrient resorbtion in northern hardwood forests. Bioscience 32, 29-32. Simpson, R.L., D.F. Whigham, and R. Walk er. 1978. Seasonal patterns of nutrient movement in a freshwater tidal marsh. In R.E. Good et al, eds. Freshwater Wetlands: Ecological Processes and Mana gement Potential. New York (NY). Academic Press. p 243-257. Sinsabaugh, R.L., Antibus, R.K., and Linkins, A.E. 1991. An enzymatic approach to the analysis of microbial activity during plant litter decomposition. Agriculture, Ecosystems, and Environment 34, 43-54. Sinsabaugh, R.L., Antibus, R.K., Linkins A.E., McClaugherty, C.A., Rayburn, L., Repert, D., and Weiland, T. 1993. Wood decomposition: Nitrogen and phosphorus dynamics in relation to extr acellular enzyme activity. Ecology 74,1586-1593. South Florida Water Management Distri ct. 1993. Surface Water Improvement and Management (SWIM) Plan Update for Lake Okeechobee, Volume 1, Planning

PAGE 105

92 Document. South Florida Water Management District, West Palm Beach, Florida. South Florida Water Management District Florida Department of Environmental Protection, and Florida Department of Ag riculture and Consumer Services. 2004a. Lake Okeechobee Protection Plan. South Florida Water Management District, West Palm Beach, Florida. Steinman, A.D., J. Concklin, P.J. Bohlen, and D.G. Uzarski. 2003. Influence of cattle grazing and pasture land use on macroinvert ebrate communities in freshwater wetlands. Wetlands 23, 877–889. Taylor, B., Parkinson, D., and Parsons, W. F.J. 1989. Nitrogen a nd lignin content as predictors of litter decay rates: A microcosm test. Ecology 70, 97-104. Terrill, T.H., W.R. Windham, J.J. Evans, and C.S. Hoveland. 1994. Effect of drying method and condensed tannin on detergent fiber analysis of Sericea lespedeza J. Sci. Food Agric. 66, 337-343. Tiner, R.W. 2003. Gegraphically isolated wetlands of the United States. Wetlands 23, 494-516. Tobe J.D. (ed) 1998. Florida wetland plan ts: An identification manual. Florida Department of Environmental Protection. Tallahassee, FL. Tope, B. 2003. White Paper, A study of the l eaching processes and characteristics for red alder (Alnus rubra) and Himalayan blackberry (Rubus sp.) leaf litter. Turner, R.E. 1993. Carbon, nitrogen, and phos phorus leaching rates from Spartina alterniflora salt marshes. Mari ne Ecology Progress Series 92,135-140. Twilley, R.R., Lugo, A.E., Patterson-Zucca, C ., 1986a. Litter production and turnover in basin mangrove forests in s outhwest Florida. Ecology 67, 670–683. Valiela, I., Teal, J.M., Allen, S.D., Van Etten, R., Goehringer, D., Volkmann, S., 1985. Decomposition in salt marsh ecosystems: the phases and major factors affecting disappearance of above-ground orga nic matter.Biol. Ecol. 89, 29–54 Villar.C.A, L. de Cabo, P. Vaithiyanatha n and C. Bonetto. 2001. Litter decomposition of emergent macrophytes in a floodplain marsh of the Lower Paran River. Aquatic Botany 70, 105–116 Webster J.R. and Benfield E.F. 1986. Vasc ular plant breakdown in freshwater ecosystems. Ann.Rev. Ecol. Syst. 17, 567–594.

PAGE 106

93 BIOGRAPHICAL SKETCH Natalie was born in Ft. Pierce, FL in 1982 where she developed a love for the outdoors at an early age. After completing high school, she attended Virginia Tech in Blacksburg where she became interested in studying wetlands during her last semester after taking a forested wetland course. After graduating from Virginia Tech in 2003, she worked for the Indian River Research and Educ ation Center in Ft. Pi erce as a Biological Control Technician in the Entomology Depa rtment. In 2004, she began her master’s degree at the University of Florida. In December 2006, she completed her master’s degree and will be joining a cons ulting firm in Sarasota, FL.


Permanent Link: http://ufdc.ufl.edu/UFE0016520/00001

Material Information

Title: Litter Decomposition and Phosphorus Release in Okeechobee Isolated Wetlands
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0016520:00001

Permanent Link: http://ufdc.ufl.edu/UFE0016520/00001

Material Information

Title: Litter Decomposition and Phosphorus Release in Okeechobee Isolated Wetlands
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0016520:00001


This item has the following downloads:


Full Text












LITTER DECOMPOSITION AND PHOSPHORUS RELEASE IN OKEECHOBEE
ISOLATED WETLANDS














By

NATALIE BALCER


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


2006

































Copyright 2006

by

Natalie Balcer

































To my parents who have supported me throughout this entire journey.















ACKNOWLEDGMENTS

I would like to thank Dr. Mark Clark for his insight, guidance, and support. I

thank him for inspiring me with his wealth of knowledge on scientific issues. Thanks go

to my other committee members (Dr. Ramesh Reddy and Dr. Patrick Bohlen) for their

efforts and advice. Special thanks go to Dr. Ed Dunne for his advice and help in the

field. Many other students and staff in the Wetland Biogeochemistry Laboratory have

provided additional insight and assistance during the past 2 years. Funding was provided

by the South Florida Water Management District, Florida Department of Agriculture and

Consumer Services, and the Florida Department of Environmental Protection. Lastly,

thanks go to Darren Cole for continuously supporting my decision to come to Gainesville

to further my education.















TABLE OF CONTENTS



A C K N O W L E D G M E N T S ................................................................................................. iv

LIST OF TABLES ................................................... vii

LIST OF FIGURES .................................. .. ......... ............................ ix

ABSTRACT ................................................... .................... xii

CHAPTER

1 INTRODUCTION AND SITE DESCRIPTION......................................................1..

In tro d u c tio n ................................................................................................................... 1
Rationale for Research.................................................................... .. 5
S ite D description ..................................................... ............................................ . 8

2 PLANT TISSUE CHARACTERIZATION .............. ....................................14

In tro d u c tio n ................................................................................................................. 1 4
M materials and M ethods .................... ............................................................... 16
R e su lts.. . ........ ........................................................... ......................................1 8
D iscu ssion ................................................................................... ...................... 24

3 SHORT TERM PHOSPHORUS LEACHING......................................................... 27

In tro d u ctio n ................................................................................................................. 2 7
M materials and M ethods .. ...................................................................... ................ 29
R e su lts.. . ........ ........................................................... ......................................3 2
D iscu ssion ................................................................................... ...................... 45

4 LITTER DECOMPOSITION AND LONG TERM PHOSPHORUS RELEASE ..... 52

In tro d u ctio n ................................................................................................................. 5 2
M materials and M ethods .. ...................................................................... ................ 56
F ie ld M eth o d s ......................................................................................................5 6
L laboratory M ethods .............. ................ ..................................... 61
R e su lts .. . ........ ........................................................... ...................................... 6 2
Discussion ........................ ................................................. 73

5 SYNTHESIS AND CONCLUSIONS......................................................... .............81









L IST O F R EFE R E N C E S .... ........................................................................ ................ 87

BIOGRAPHICAL SKETCH ......................................................................................93














LIST OF TABLES


Table page

2-1. Neutral detergent fiber and acid detergent fiber fractionation methods (Rowland
and Roberts, 1999) .......................... .......... ........................ 18

2-2. The C/N and C/P ratios for live and senesced tissue. Values represent mean ( 1
standard deviation). ........... ... .............. .... ........ ...... ............... 22

3-1. Cumulative P release (mg P/g litter) and percent tissue P released from four
species over a 17 day period, under aerobic and anaerobic conditions. Values
represent mean ( 1 standard deviation). Lowercase letters indicate significant
differences over the entire 17 day study between species, water treatment, and
redox conditions w ith a p-value of .05 ............................................... ................ 40

3-2. Mean concentrations ( 1 standard deviation) of water column P for various
species under aerobic and anaerobic conditions after 17 days. Negative DOP
values are due to high standard errors and represent virtually no measurable
D O P in the w ater colum n ........................................ ........................ ................ 4 1

3-3. Mean concentrations ( 1 standard deviation) of water column TKN for various
species under aerobic and anaerobic conditions after 17 days. Letters indicate
significant differences between species and redox condition. ................................41

3-4. The P-Values associated with initial nutrient parameters of senesced tissue to
estimate the best predictor of P flux for site water under aerobic conditions on
D a y 1 7 ................................................................. ................................................ .... 4 3

3-5. The R-Square values from %P and cumulative P release correlation in site water
under aerobic and anaerobic conditions.............................................. ................ 44

3-6. The P-Values associated with initial nutrient parameters of live tissue to estimate
the best predictor of P flux for site water under aerobic conditions on Day 17.......44

4-1. Species decomposition in each zone of litterbag deployment after 12 months.
Values represent mean ( 1 standard deviation). ................................ ................ 66

4-2. Significance and R2 values of the relationship between the change in P content
and initial senesced substrate quality characteristics for each sampling period.......67

4-3. Significance and R2 values of the relationship between the change in P content
and initial live substrate quality characteristics for each sampling period ............69









5-1. Variation in short and moderate-term P assimilation or release from Im 2 of litter
over 2 months for the 4 dominant species investigated.......................................83














LIST OF FIGURES
Figure page

1-1. Hydrological modifications of Lake Okeechobee and its watershed have resulted
in increased channelization, transportation of contaminants to the lake and loss
of w wetlands (F D E P 200 1) ........................................ ......................... .............. .3...

1-2. The P retention mechanism in a wetland, the potential fate of P initially
assimilated in macrophyte plant tissue and different P fractions such as
Dissolved Organic Phosphorus (DOP), Particulate Organic Phosphorus (POP),
and D issolved Inorganic Phosphorus (D IP) .................................... ..................... 6

1-3. Overland view of the study site including three historically isolated wetlands on
Larson Dixie Ranch.................. ............. .............................. 9

1-5. Average hydroperiod for center, edge, and upland zones of seasonally isolated
w wetlands on Larson-D ixie R anch ........................................................ ............... 11

1-6. Species occurrence in seasonally isolated wetlands located in the Okeechobee
Basin. A) Wetland Centers, B) Wetland Edge, C) Surrounding Uplands ............ 12

1-7. Soil total phosphorus storage in wetland zones and surrounding upland............... 13

2-1. Extraction sequence used to determine carbon quality of vegetation. .................... 17

2-2. The C/N and N/P ratios of live tissue for species surveyed. Values represent
m ean ( 1 standard deviation) ............................................................ ............... 19

2-3. Mean C/P ratios in live tissue of species surveyed. Values represent mean ( 1
standard deviation). ........... ... .............. .... ........ ...... ............... 20

2-4. Fiber quality (NDF, ADF, SADF and Residual Fiber) percentages for species
su rv ey e d ............................................................................................................. .. 2 0

2-5. Residual fiber fractions for live tissue. Values represent mean ( 1 standard
d ev iatio n ).............................................................................................................. .. 2 1

2-6. Nutrient ratios of C:N in senesced tissue of dominant species. Values represent
m ean ( 1 standard deviation) ............................................................ ................ 2 1

2-7. Comparison of phosphorus values for live and senesced vegetation. Values
represent m ean ( 1 standard deviation) ............................................. ................ 22









2-8. Neutral detergent fiber content contained in the senesced tissue of 4 dominant
species. Values represent mean ( 1 standard deviation)...................................23

2-9. Residual fiber content contained in the senesced tissue of 4 dominant species.
Values represent mean ( 1 standard deviation). ................................ ................ 23

3-1. The P leaching study. A) Overview of the experimental setup and aerobic and
anaerobic treatments B) Individual fluxing container and tubing bubbling
ambient air in the water column through the hypodermic needle.........................31

3-2. Phosphorus leaching rates of 4 senesced species averaged across all 3 water
treatments A) aerobic B) anaerobic conditions. Values represent mean (
standard deviation). ........... ... .............. .... ........ ...... ............... 34

3-3. Litter phosphorus release rate for P. hemitomon litter under (a) aerobic and (b)
anaerobic conditions with 3 different water treatments. Values represent mean
( 1 standard deviation)........................................... ......................... ................ 36

3-4. Litter phosphorus release rate for P. hydropiperoides litter under a) aerobic and
b) anaerobic conditions with 3 water treatments. Values represent mean ( 1
standard deviation) ............................... ............................................. 37

3-5. Phosphorus release rate for P. notatum litter under a) aerobic and b) anaerobic
conditions with 3 water treatments. Values represent mean ( 1 standard
d ev iatio n ).............................................................................................................. .. 3 8

3-6. Phosphorus release rate for J. effusus litter under a) aerobic and b) anaerobic
conditions with 3 water treatments. Values represent mean ( 1 standard
d ev iatio n ).............................................................................................................. .. 3 9

3-7. Percent mass loss from the vegetation over a 17 day period. Values represent
m ean 1 standard deviation....................................... ...................... ................ 42

3-8. Bivariate fit of cumulative P flux by initial senesced tissue P content.
Correlation is for site water treatment under aerobic conditions on day 17 ..........43

4-1. Litterbag distribution and deployment locations in 4 hydrological zones within
th e w etlan d .............................................................................................................. 5 8

4-2. Hydrological information for the seasonally isolated wetlands on Larson Dixie
Ranch. A) The yearly stage information in meters from April 1, 2004 March
10, 2006 B) the average hydroperiod for the center, edge, and upland zones
during N ovem ber, M arch, and July..................................................... ................ 59

4-3. Aerial view of the three historically isolated wetlands on Larson Dixie Ranch
w here the litterbags w ere deployed .................................................... ................ 59









4-4. Litterbag experiment showing A) litterbags and netting perpendicular to transect
in upland, B) litterbags attached to soil and covered by nylon netting as a
precaution against cattle, and C) a close up of litterbag filled with P. notatum.
Pictures D, E, and F illustrate the litterbag collection after 8 months of exposure
showing D) P. notatum growing in and through the litterbag and E) the large
amount of vegetation covering the bags in the upland and F) in the transitional
z o n e ....................................................................................................... ........ .. 6 0

4-5. Litter decomposition of all species in all hydrologic zone among the 3 different
wetlands, Larson East (LE), Larson West (LW), and Larson South (LS). Values
represent m ean 1 standard deviation ............................................... ................ 63

4-6. Decomposition in 4 wetland zones over a 12 month period. Values represent
m ean 1 standard deviation....................................... ...................... ................ 64

4-7. Average litter decomposition of each species over a 12 month period. Values
represent m ean 1 standard deviation ............................................... ................ 65

4-8. Change in % P in the 4 hydrological zones over a 12 month period. Values
represent m ean 1 standard deviation ............................................... ................ 66

4-9. Change in % P in the 4 dominant species over a 12 month period. Values
represent m ean 1 standard deviation ............................................... ................ 67

4-10. Correlation between initial NDF fraction in the senesced tissue and P loss or
gain after 12 m months .......................................................................................... 68

4-11. Correlation between initial C:P in the live tissue and P loss or gain after 12
m o n th s .................................................................. ............................................... ... 7 0

4-12. Change in litter /oN (all species combined) among different wetland hydrologic
zones over time. Values represent mean 1 standard deviation..........................70

4-13. Change in litter %N among species over time. Values represent mean 1
standard deviation. ... ............................ ........ .................... 71

4-14. Comparison of residual fiber content of initial and 12 month exposed litter
among four species tested. Values represent mean ( 1 standard deviation) ..........72

4-15. Residual fiber content of species in each hydrologic zone after 12 months.
Values represent mean ( 1 standard deviation). ................................ ................ 73

4-16. Litterbags collected after 12 months from A) wetland center, which were
approximately 20-30 cm underneath the soil surface and B) wetland edge, which
w ere on top of the soil surface .................................... ..................... ................ 77















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

LITTER DECOMPOSITION AND PHOSPHORUS RELEASE IN OKEECHOBEE
ISOLATED WETLANDS

By

Natalie Balcer

December 2006

Chair: Mark Clark
Major Department: Soil and Water Science

Phosphorous (P) is the leading nutrient that contributes to degraded water quality

and the eutrophication of Lake Okeechobee. Isolated wetlands within the Lake's

watershed are potentially an important phosphorus sink reducing P concentrations in the

water column before surface runoff reaches the lake. Hydrologic restoration of ditched

and partially drained wetlands is hypothesized to enhance P assimilative capacity and

therefore further reduce P contaminant loads. Phosphorus dynamics in decomposing

litter contributes significantly to overall P dynamics in these heavily vegetated systems.

Our study focused on short- and long-term P release from aboveground biomass

after vegetative senescence. Our first objective of this project was to characterize fiber

quality and nutrient content characterization of live and senesced tissue of different

wetland species. By quantifying the amount of TN, TC, and TP in the litter, as well as

running fiber quality characteristics, some predictions were made regarding the relative

liability and recalcitrance of the detrital material. Our second objective involved a









laboratory experiment to investigate the short-term release of P in response to inundation

under aerobic and anaerobic conditions from the senesced tissue of 4 species common to

these wetlands: Paspalum notatum, Polygonum hydropiperoides, Juncus effusus, and

Panicum hemitomon. Our third objective was to investigate the influence of hydrological

zones on the decomposition rate and long-term P release of senesced vegetation. This

was accomplished using litterbags deployed in the field.

Fiber quality and nutrient content varied significantly among the four plant

species and played a significant role in P release and decomposition rate. The P leaching

study showed that senesced, air-dried vegetation had the potential to release a substantial

amount of P within 72 hours. There were statistical significant differences in the

leaching rates among species, with the greatest P leaching occurring in P.

hydropiperiodes and the least amount of leaching in P. hemitomon. In addition, short

term P leaching was strongly correlated with the initial P content in the senesced tissue.

Species type also insignificantly influenced P release and decomposition in the field over

a 12-month period. Paspalum notatum had a significantly higher decomposition rate

after 12 months compared to the other three species. The C:P in the senesced tissue

predicted phosphorus release up to a 2-month time period.

Our results provide further insight into the potential implications of vegetative

species changes that occur in response to hydrological restoration, or other wetland-

related Best Management Practices (BMPs) to reduce P loading into the waterways of the

Okeechobee Basin.














CHAPTER 1
INTRODUCTION AND SITE DESCRIPTION

Introduction

Wetlands have the ability to store and transform nutrients from surrounding

uplands (Richardson, 1985; Reddy et al., 1995; Jordan, 2003). Wetlands presently make

up 17% of the Okeechobee Basin; however this percentage has decreased in recent times

due to ditching and drainage in an effort to improve pastures for dairy and beef cattle

operations (Haan, 1995; Tiner, 2003). Currently, greater than 50% of the wetlands

located in the basin have been at least partially ditched and drained to increase the

acreage of improved pastures (McKee, 2005). Wetlands already provide a valuable role

in improving the water quality of Lake Okeechobee by storing water in the watershed and

sequestering phosphorous (P), the leading nutrient causing eutrophication of the lake

(Davis and Marshall, 1975; Federico et al., 1981). However, it is thought that hydrologic

restoration of those wetlands that have been ditched and drained may provide a

significant increase in P storage and will assist in efforts to address P TMDL targets set

for the basin.

Lake Okeechobee is the 2nd largest lake contained entirely within the United

States, covering an area of 1890 km2 (Reddy et al., 1995). It was formed by a buildup of

peat near the southern edge of the lake that restricted outflow (Brezonik and Engstrom,

1998). Lake Okeechobee provides many values including flood control; water supply; a

habitat for migratory birds, fish, and other species; as well as a multimillion dollar

recreational and commercial fishing industry. It also serves as the head waters for the









Everglades. Consequently, the water quality of the lake has the potential to impact

numerous downstream communities (Steinman, 2003).

Over the past 100 years, Lake Okeechobee has had numerous anthropogenic

alterations and impacts. After several devastating hurricanes when the lake overflowed

its natural boundaries, a dyke was built around the perimeter, establishing total control of

water entering and leaving the lake. Water-level regulations have been set to maintain

the integrity of the dyke surrounding the lake, which provides flood protection but has

altered the wetland communities that lie within the dyke perimeter. Rivers and many

tributaries were channelized, allowing adjacent wetland areas to be drained and

stormwater to be moved rapidly off the landscape. The loss of historic sinuous flow

along many of the rivers and streams, like the Kissimmee River, has resulted in a loss of

their ability to treat water; and instead, the degraded water is discharged directly into the

lake (Figure 1-1).

Presently agriculture is the primary land use in the Okeechobee basin. The

watershed is dominated by beef ranches and dairy operations (Flaig et al., 1995). Land

use has impacted the water quality in the Okeechobee Basin, causing an increased

nutrient load to Lake Okeechobee. The increased contaminant load has resulted in a

decline in species diversity, lower amounts of available oxygen, and reduced clarity in

the Lake. These changes have caused a shift from aquatic vegetation to phytoplankton,

which has reduced the assimilative capacity of the lake and made the sediments less

stable, and caused an increase in the occurrence of cyanobacteria blooms (South Florida

Water Management District (SFWMD), 1993; Brezonik and Engstrom, 1998; Carpenter

et al., 1998; Havens and Shelske, 2001).









Phosphorous is the nutrient that contributes most significantly to degraded water

quality and eutrophication (Davis and Marshal, 1975; Federico et al., 1981; Carpenter et

al., 1998). Most of the phosphorus enters the lake from non-point sources in agricultural

areas (Anderson and Flaig, 1995). During heavy rainfall, P from fertilizers and wastes

are incorporated into runoff from the surrounding agricultural areas, which then drains

into Lake Okeechobee. Dairies have been found to produce the most P load per unit area.

They make up 27% of the land in the Okeechobee Basin, yet are responsible for 49% of

the P load to the lake (Flaig et al., 1995; Hiscock et al., 2003).


Early 1800s Present














A-O



EVEROLADES
THE EVERGLADES AGRICULTURAL
AREA



Figure 1-1. Hydrological modifications of Lake Okeechobee and its watershed have
resulted in increased channelization, transportation of contaminants to the
lake and loss of wetlands (FDEP, 2001).









The second highest P load comes from improved pastures used for beef cattle

production. Cattle ranches make up 33% of the landscape and are responsible for 51% of

the P entering from north of the lake (Flaig et al., 1995; Hiscock et al., 2003).

Decreasing water quality in Lake Okeechobee has become a big concern in

Florida primarily because it is a drinking-water source for communities in the area, and

those populations are increasing (Anderson and Flaig, 1995). Best Management Practices

(BMPs) such as the use of wetlands to store and retain P, planting vegetative species that

increase the P uptake, fencing cattle from water bodies, and capturing and recycling

manure have prevented an increase in the P load in the last decade (Bottcher et al., 1995;

Flaig et al., 1995). Even with the use of BMPs, the P content in the lake sediments is still

high from historic agriculture runoff and back-pumping of drainage water in the

Everglades Agricultural Area (EAA). These conditions result in a high internal P

loading, regardless of new external loading of P from the watershed.

In 1987, the Lake Okeechobee Surface Water Improvement and Management

Plan (SWIM) was initiated (SFWMD, 1993). The goal was to improve water quality of

the lake using a watershed management approach. The SWIM Plan identified four of the

41 basins around the lake as "priority basins," because they were contributing 35% of the

P load while occupying only 12% of the land area (FDEP, 2001). Lake Okeechobee has

also been listed as an impaired water body by the Clean Water Act (CWA) section 303

(d) and therefore a Total Maximum Daily Load or TMDL that limits the amount of P the

lake can receive has been established.

Between 1995 and 2000, the lake received an average of 640 tons of P per year

(FDEP, 2001). In 2002, the P load measured into Lake Okeechobee was 543 tons. The









TMDL target P load is 140 tons per year with a concentration of 40ppb P in the pelagic

areas of the lake by 2015. To reach this goal an increased effort in use of BMPs and Best

Available Technology (BATs) in the Okeechobee Basin (SFWMD et al., 2004a) is

underway.

Rationale for Research

BMPs are needed to reduce the use of P and increase P retention on agricultural

landscapes. Wetlands have the ability to store and transform nutrients from the

surrounding upland due to high organic matter content, the presence of vegetation, and

low oxygen availability, but they may also function as nutrient sources depending on the

physical and chemical characteristics of the water column and the soil (Richardson, 1985;

Reddy et al., 1995; Jordan, 2003). Nutrients and contaminants can become sorbed to the

soil or organic matter, immobilized by plants and microbes, and mineralized by microbial

processes. This degree to which P can be reduced depends on the land use, soil type,

retention time, slope, hydraulic connectivity, and vegetation type of the area (Raisin and

Mitchell 1995).

The large number of wetlands within the four priority basins may have a potential

to significantly reduce P load if drainage pathways from ditched wetlands are blocked

and hydrologically restored to their historically isolated state. Restoration would increase

the hydroperiod, thereby increasing the wetland area and the potential to retain P by plant

immobilization, soil adsorption and reduced organic matter decomposition rates.

Macrophytes have the ability to immobilize large amounts of nutrients from the

water column; however this should be thought of as only a temporary storage for

nutrients (Fig. 1-2). It is estimated that 35-80% of the total P contained in plant biomass

is eventually lost to the water column, or transported out of the system after senescing,









and the processes influencing nutrient release rates are poorly understood (Richardson,

1989; Reddy et al., 1995). Because a high percentage of phosphorus is assimilated and

potentially released during the decomposition process, it is important to understand

factors that might regulate the rate and total amount of P released (Jordan and Whigham,

1989).





INFLOW





POP mineralization



+ PIP Fe,

DIP DOp POP Adsorbed P*- Al, Ca
DIP DOP bound P

Figure 1-2. The P retention mechanism in a wetland, the potential fate of P initially
assimilated in macrophyte plant tissue and different P fractions such as
Dissolved Organic Phosphorus (DOP), Particulate Organic Phosphorus
(POP), and Dissolved Inorganic Phosphorus (DIP).

Wetlands are carbon based systems in which most of the energy driving the food

web is derived from litter turnover (Turner, 1993; Corstanje, et al., 2006). Because the

rate of organic matter turnover drives the system, it is important to understand the fate of

detrital matter in the wetland, as well as how much P is released from litter during short

and long term decomposition. Numerous studies have shown that carbon quality,

primarily the fraction of recalcitrant portions of the plant, as well as the ratio of available









N to carbon determine the rate of litter breakdown (Melillo, 1982; Berg, 1998; Villar,

2001).

Nutrient loss due to leaching is not as widely investigated but has been shown to

be a significant nutrient and carbon source (Turner, 1993). Studies have determined that

up to 80% of N and P contained in the litter as well as up to 80% of the litter biomass can

be lost within 2 months after senescence due to short term leaching and mineralization in

a tidal freshwater marsh (Simpson et al., 1978; Qiu, 2005). Another study preformed by

C. Boyd (1971) concluded that the rate of nutrient loss was greatest during the first 4

months after senescence. The study reported that 50% of N and P were released and 60%

of the biomass (Juncus effusus) decomposed during the first 4 months in the field while

only an additional 10% of the remaining mass was lost during the remaining 8 months. It

is apparent however, that the species type or environmental conditions under which

leaching and decomposition occur has a large influence on decomposition rates (Villar et

al., 2001).

There is a wide range of P thought to be released through leaching and

mineralization processes. The rate of P loss is determined by a range of factors

including; the decomposition rate of the litter, the species of vegetation, as well as factors

such as temperature, moisture, nutrients, pH, carbon quality, and the microbial

community which influence decomposition (Melillo, 1982; Benner, 1985). It has been

determined that there is a significant difference between the amount of phosphorus

assimilation in different species of vegetation (McJanet et al., 1995), but the amount and

rate of P release of different wetland species after vegetative senescence remains

unknown. If a species can be found that has similar uptake rates of P, yet releases a









significantly lower amount of P from its biomass, that species could be recommended to

ranchers for the purpose of P retention in order to decrease the amount of P release after

senescence.

Site Description

This study was conducted on three historically isolated wetlands located within

the same pasture on the Larson Dixie Ranch (N 0270 20.966', W 0800 56.465') (Fig. 1-3).

Larson Dixie Ranch is a cow-calf operation within one of the four priority basins (S-154)

of the Lake Okeechobee Watershed (Fig. 1-4). Cow-calf operations make up roughly

48% of the agricultural land within the Okeechobee Basin. Isolated wetlands cover

approximately 12,000 ha in the four priority sub-basins and 400,000 ha within the entire

Okeechobee watershed (Reddy et al., 1995). Presently about 50% of the historically

isolated wetlands within the basin are ditched to enhance drainage and increase the area

of improved pastures. Hydrological connections between wetlands on the landscape

allows surface water to be transported to the lake, and results in a large P load and

reduced water quality to Lake Okeechobee during the wet season. Often, wetlands

located within the Okeechobee Basin are seasonally flooded from early June to January,

and remain relatively dry during the late winter and spring months. The average

hydroperiod for the wetland center, edge, and upland zones ranged from 100-200 days /

y, 5-55 days / y, and 0-15 days / y respectively (Fig. 1-5).

The average area of each wetland in this study was 2.64 ha. Every wetland was

drained by one ditch that drained into a larger drainage swale and transported the water

offsite. The wetlands were completely surrounded by grazing pastures that were

dominated by Paspalum notatum Flugge (Bogdan) (Bahia grass), while the wetlands









themselves were dominated by emergent marsh vegetation consisting of Juncus effusus

L., Panicum spp., Polygonum hydropiperoides Michx, and Pontedaria cordata var.

lancifolia (Muhl.) Torr (Fig. 1-6).


Figure 1-3. Overland view of the study site including three historically isolated wetlands
on Larson Dixie Ranch

The soils within the Okeechobee Basin are primarily poorly drained, sandy,

Spodosols that have limited P retention within the upper horizons and a water table

generally less than 20 cm from the soil surface (Anderson and Flaig, 1995; Haan, 1995).

The P content was the greatest in center soils compared to the edge and upland soils of

the seasonally isolated wetlands on Larson Dixie Ranch (Fig. 1-7). The isolated wetlands









on the study site are flooded throughout the majority of the year with an occasional dry

period from late March to early June. Soil on the Larson Dixie ranch is classified as

Siliceous, hyperthermic Spodic, Psammaquents (Basinger series), which is a deep, poorly

drained, rapidly permeable soil formed from sandy marine sediments (Anderson and

Flaig, 1995).


Figure 1-4. Location of Larson Dixie Ranch within the 4 priority basins (McKee, K.A.
2005. Predicting Phosphorus Storage in Historically Isolated Wetlands
within the Lake Okeechobee Priority Basins. Masters Thesis. University of
Florida. Gainesville, FL.)

The soils within the Okeechobee Basin are primarily poorly drained, sandy,

Spodosols that have limited P retention within the upper horizons and a water table

generally less than 20 cm from the soil surface (Anderson and Flaig, 1995; Haan, 1995).

The P content was the greatest in center soils compared to the edge and upland soils of










the seasonally isolated wetlands on Larson Dixie Ranch (Fig. 1-7). The isolated wetlands

on the study site are flooded throughout the majority of the year with an occasional dry

period from late March to early June. Soil on the Larson Dixie ranch is classified as

Siliceous, hyperthermic Spodic, Psammaquents (Basinger series), which is a deep, poorly

drained, rapidly permeable soil formed from sandy marine sediments (Anderson and

Flaig, 1995).


* Objective 1: Characterize plant tissue from live and recently senesced vegetation
of dominant species found in the three isolated wetlands on the Larson Dixie
Ranch.

* Objective 2: Determine the short term P release from senesced plant material
under aerobic and anaerobic conditions.

* Objective 3: Measure the litter decomposition and long term P release rate of four
dominant species in situ at four different hydrologic regimes.

The objectives mentioned above and their related hypotheses, methods, and

results are discussed further in Chapters 2, 3, and 4 respectively. Chapter 5 summarizes

results from all three chapters and evaluates the potential effects of hydrological

restoration of isolated wetlands on decomposition rates and P release as a BMP in the

Okeechobee Basin.

240

190 -

140 -

90 -



-10 s
Center Edge Upland

Figure 1-5. Average hydroperiod for center, edge, and upland zones of seasonally
isolated wetlands on Larson-Dixie Ranch














25 -
20 -20
S, 15
D- 10

0


4~.

'V


o ~*9 ~


0


100
> 80
, 60
I- 40
L 20
0


0


C)
'S


Figure 1-6. Species occurrence in seasonally isolated wetlands located in the
Okeechobee Basin. A) Wetland Centers, B) Wetland Edge, C) Surrounding
Uplands.










400

350

300

E 250

200

150
i-
100

50


center edge upland

Figure 1-7. Soil total phosphorus storage in wetland zones and surrounding upland.

* Objective 1: Characterize plant tissue from live and recently senesced vegetation
of dominant species found in the three isolated wetlands on the Larson Dixie
Ranch.

* Objective 2: Determine the short term P release from senesced plant material
under aerobic and anaerobic conditions.

* Objective 3: Measure the litter decomposition and long term P release rate of four
dominant species in situ at four different hydrologic regimes.

The objectives mentioned above and their related hypotheses, methods, and

results are discussed further in Chapters 2, 3, and 4 respectively. Chapter 5 summarizes

results from all three chapters and evaluates the potential effects of hydrological

restoration of isolated wetlands on decomposition rates and P release as a BMP in the

Okeechobee Basin.














CHAPTER 2
PLANT TISSUE CHARACTERIZATION

Introduction

Wetland vegetation is able to assimilate nutrients into its biomass from the soil and

surrounding water column, thereby potentially reducing nutrient levels discharged from

the wetland through ditches. Macrophytes only serve as a temporary sink for nutrients

however, and tend to undergo leaching and mineralization of biomass after senescence.

Observations from several studies suggest that the degree of decomposition and nutrient

leaching is greatest during the first 4 months after senescence when up to 50% of N, P,

and biomass can be lost in marsh communities (Boyd, 1971; Richardson, 1989; Flaig,

1995). Numerous physical and environmental factors such as temperature, moisture

content, nutrients, pH, carbon quality, and the microbial community can influence litter

decomposition and leaching rates (Melillo, 1982; Benner, 1985).

Nutrient ratios such as C:N and C:P, as well as the quantity of residual fiber in the

plant tissue is also important and often used to determine the physical and chemical

composition of vegetation (Day, 1982; DeBusk and Reddy 1998; Lewis, 2005).

Numerous studies have shown that nitrogen and lignin content (one of the most

recalcitrant organic structural compounds) are the 2 most important factors that determine

the rate of decomposition (Melillo, 1982; Berg, 1998; Carreiro, 2000; Villar, 2001;

Corstanje, 2006). During the decomposition process, carbon compounds are lost, and in

the short-term nutrients are either gained or lost depending on the surrounding

environment and the nutrient ratios of the litter.









Due to the role of C:N and lignin content play in decomposition, it may be possible

to predict P release by assessing these two parameters in conjunction with wetlands

where P assimilation and retention is being considered. It is necessary to perform an

initial characterization on live and recently senesced vegetation in order to quantify

nutrient concentrations and fiber quality of the dominant vegetative species on the site.

Lignin is the organic compound contained in plant biomass that is the most

recalcitrant or resistant to decomposition. It is thought to be the limiting factor of long-

term decomposition because it can only be degraded by a few organisms. Microbes are

able to degrade lignin with the help of various enzymes (Benner, 1984); however it is

extremely resistant to decomposition in anaerobic environments even by microbes

(Criquet et al., 2001). Lignin helps provide structural support to the plant in order to

remain upright, as well as provide a protective layer against microbial attack; therefore it

is often abundant in the bark of woody trees (Hammel, 1997).

Due to a greater assimilative capacity of wetland vegetation compared to edge and

upland vegetation, there may be significant differences in the nutrient content of the plant

species found in these 3 respective zones (McJanet et al., 1995). Differences in nutrient

content may cause some species to have a lower or higher C/N ratio or residual fiber

content compared to others, and may ultimately influence the quantity of P released from

senesced vegetation through short term leaching and long term decomposition. It is

important to characterize the nutrient and residual fiber to assess whether any tissue

parameters are strongly correlated to P retention so that the amount of P released from a

species after senescing may be predicted with some certainty. This information could

provide further insight as to how species shift as a result of hydrological restoration, or









other practices may impact the amount of P released into the water column after

vegetative senescence.

Hypothesis 1: The species of vegetation surveyed will have different amounts of
carbon, nitrogen, phosphorus, and residual fiber.

Materials and Methods

Live and recently senesced (standing dead) vegetation were harvested from Larson

Dixie ranch in November 2004. Species found in the wetland center, edge, and upland

were collected, placed in paper bags according to species. After the material was brought

back to the laboratory, all the live species collected were analyzed for TN, TC, TP, and

tissue fiber analysis, while only the senesced tissue of 4 of the live species collected were

analyzed for fiber and nutrient content due to their dominance on the site and presence in

different hydrological zones. The 4 dominant species were all perennial species

including: Paspalum notatum Flugge (Bogdan) (Bahia grass) in the upland, and Juncus

effusus L., Panicum spp., Polygonum hydropiperoides Michx in the edge and interior of

the wetland.

Vegetation was dried at 600 C for at least 72 hrs. The plants were then ground to

pass a # 40 mesh sieve using a Wiley Mill. Analysis of total carbon and total nitrogen

was done using a Thermo Electron Flash EA 1112 Nitrogen/Carbon analyzer. Tissue P

was analyzed using acid digestion of ashed tissue (Anderson 1976) and P content was

quantified using an auto analyzer (Method 365.4; USEPA, 1993). An Ankom 200 Fiber

Analyzer was used to quantify the NDF, ADF, and residual fiber fractions of tissue; this

method is further described below (Ankom Technology Corp., 1998a). The means of all

species were compared using Tukey-Kramer HSD (honestly significant differences) to

determine significant differences.









Fiber Analysis Method

Residual fiber content in tissue has been quantified in the past using the Klasson

method by the hydrolysis of cellulose with the use of 72% H2SO4 (Rowland and Roberts,

1999). This technique alone has proven to be time consuming due to lengthy

pretreatments and is not easily reproducible in the laboratory (Rowland and Roberts,

1999). The neutral detergent fiber (NDF) and acid detergent fiber (ADF) method

provides a simple technique for quantifying amounts of labile components as well as

hemicellulose contained in the plant biomass, and is the method used in this study (Fig.

2-1). Sequential NDF and ADF extractions were recommended to improve the digestion

of cell wall proteins and minimize the influence of tannins on the recovery of residual

fiber (Terrill et al., 1994).













Figure 2-1. Extraction sequence used to determine carbon quality of vegetation.

In this study "residual fiber" is referred to as lignin. This material is highly

recalcitrant, however also contains non-carbon compounds, therefore it is functionally

defined as any compound that has not been previously leached or degraded during the

fractionation scheme. The NDF, ADF, and H2SO4 detergent fiber (strong acid detergent

fiber or SADF) extractions sequentially remove sugars, hemicellulose, and cellulose

fractions of the plant material. The NDF extraction removes the labile components of the









vegetation, such as sugars and starch. The ADF extraction removes hemicellulose, while

the SADF extraction strips away any remaining hemicellulose as well as cellulose,

leaving behind lignin and ash (Table 2-1).

Table 2-1. Neutral detergent fiber and acid detergent fiber fractionation methods
(Rowland and Roberts, 1999)
NDF Plant Fraction ADF

Soluble Protein
Neutral Detergent Soluble Lipids
Fraction Soluble Carbohydrates Acid Detergent Soluble
Pectin Fraction
Insoluble Protein

Hemicellulose

Neutral Detergent Fiber Cellulose
Fraction Lignin
Ligafied Acid Detergent Fiber
Lignified Nitrogen
Ash



Results

There was a wide range of C:N and N:P in the live tissue of the species surveyed

(Fig. 2-2). Panicum hemitomon had the highest C:N of all the live species characterized,

while L. fluitans had the lowest C/N ratio compared to other live species. However,

Paspalum notatum had the highest N/P ratio out of all the species surveyed while B.

caroliniana had the lowest N/P ratio.

There were also significant differences in C:P in the live species surveyed (Fig. 2-

3). In addition to having the highest C:N, P. hemitomon also had the highest C:P,

followed by P. notatum, and J. effusus respectively, while P. cordata had the highest C:P

associated with the live tissue.










40
mC:N
35 ON:P

30

25 -

20 -

15 -
15













Figure 2-2. The C/N and N/P ratios of live tissue for species surveyed. Values represent
mean ( 1 standard deviation).

In addition, a wide range of NDF (neutral detergent fiber) and residual fiber was

found in the different species surveyed (Fig. 2-4 & 2-5). Hydrocotyle umbellata, had a

lower C/N and C/P ratio and had the highest NDF fraction in the live tissue, resulting in

lower percentages of ADF and SADF. Panicum hemitomon had the lowest NDF fraction

associated with its biomass, and therefore higher ADF and SADF percentages. Paspalum

notatum, the dominant upland species, had the lowest percentage of residual fiber

compared to all other species, and A. philoxeroides had the highest residual fiber content.

Each of the 4 dominant species had a significantly lower P content in the senesced tissue

compared to the live tissue (Fig. 2-7). Polygonum hydropiperoides contained a

significantly higher live and senesced P tissue content compared to other 3 dominant

species of vegetation, while the P content in P. hemitomon was significantly lower in the












senesced tissue compared to the other 3 species of senesced vegetation. In addition to a


higher %P in the live tissue of the 4 dominant species, the C/P ratio in senesced tissue


was also significantly lower than the C/P values observed in the live tissue (Table 2-2).


350

300

250

200
a.
150

100

50

0


Figure:


E am Sa

c a 0 a) m -i '
cL C#L b



2-3. Mean C/P ratios in live tissue of species surveyed. Values represent mean
( 1 standard deviation).


100%


80%
U)
. 60%


- 40%


20%


0 % .

0) r M 0
o (U -o
C 0. 0 r 0 L.

M 0 6
ND AR

ENDF EADF ESADF B Residual Fiber


Figure 2-4. Fiber quality (NDF, ADF,
species surveyed.


SADF and Residual Fiber) percentages for









30

25

20

,. 15

c 10

5
0


H~


i


Figure 2-5. Residual fiber fractions for live tissue. Values represent mean ( 1 standard
deviation).


60

50

40
-* -
0
S30O
z

20

10


M Live C:N 0 Senesced C:N


0 ---


Panicum Paspalum
hemitomon notatum


Polygonum Juncus effusus
hydropiperoides


Figure 2-6. Nutrient ratios of C:N in senesced tissue of dominant species. Values
represent mean ( 1 standard deviation).













M Live %P IE Senesced %P

T


Panicum Paspalum
hemitomon notatum


0.35

0.3

0.25

0.2
-0
0.15

0.1

0.05

0


Figure 2-7. Comparison of phosphorus values for live and senesced vegetation. Values
represent mean ( 1 standard deviation).

Table 2-2. The C/N and C/P ratios for live and senesced tissue. Values represent mean
(+ 1 standard deviation).
Senesced
Species Live C:N C:N Live C:P Senesced C:P
Panicum
hemitomon 34.7 1.6 51.8 2.4 289.3 7.4 2165.1 129.8
Paspalum notatum 24.4 0.5 50.7 4.4 274.9 9.0 520.4 56.3
Polygonum
hydropiperoides 20.5 0.7 40.4 0.5 160.0 19.7 350.5 33.7
Juncus effusus 25.1 0.9 40.3 1.4 257.9 12.4 557.2 15.0

Additionally, the NDF fraction in the senesced tissue was lower compared to the

live tissue (Fig. 2-8). Differences in NDF content between live and senesced tissue were

significant in P. hydropiperoides and P. notatum. The highest NDF fraction was seen in

P. hydropiperoides. There were also differences in the residual fiber content in the live

and senesced tissue. Residual fiber was slightly lower in the live tissue of the 4 dominant

species compared to the senesced tissue (Fig. 2-9). The only significant difference in the

residual fiber content between the live and senesced tissue was seen in P.


Polygonum Juncus effusus
hydropiperoides











hydropiperoides. In addition to having the highest NDF fraction, this species also had


significantly higher residual fiber content in live and senesced tissue compared to the 3


other species.


* Live Tissue
] Senesced Tissue


50

45

40

35

30
LL
z 25

20

15

10

5

0


Neutral detergent fiber content contained in the senesced tissue of 4
dominant species. Values represent mean ( 1 standard deviation).


E Live Residual
Fiber

0 Senesced
Residual Fiber









T


Paspalum Polygonum Juncus effusus
notatum hydropiperoides


Panicum
hemitomon


Figure 2-9. Residual fiber content contained in the senesced tissue of 4 dominant
species. Values represent mean ( 1 standard deviation).


Panicum hemitomon Paspalum notatum Polygonum Juncus effusus
hydropiperoides


Figure 2-8.




18 T


16

14

12
I.Q
S10


"2 8
ry
6

4

2









Discussion

Different qualities of vegetative tissue were observed in this study depending on

the species, the landscape position which the species was found, and whether the

vegetation was live or senesced. A range of chemical and physical characteristics in live

and senesced tissue of different vegetative species may influence P release rates through

short term leaching and long term decomposition. Species included in the

characterization with a low N and P content, a high percentage of residual fiber, or a high

C/N or C/P ratio may have slower decomposition or leaching rates which may also

decrease the rate of P release.

High nutrient ratios may limit the amount of microbial degradation as well as the

decomposition rate. Senesced vegetation with a high N and P content as well as a low

residual fiber percentage, may have a greater rate of decomposition because it will serve

as a nutrient rich substrate with a higher likelihood of microbial colonization. Various

studies have reported increased decomposition of tissues with a higher N content (Berg,

1998; Carreiro, 2000; Corstanje, 2006), and an N limitation may suppress decomposition

rates due to a lack of microbial colonization of the substrate. In general, a C:N > 30 or a

C:P > 200 in the senesced tissue may result in an N or P limitation and decrease

microbial decomposition under aerobic conditions(Fenchel et al., 1998).

Based on values in Table 2-2, vegetation collected from the Larson Dixie site

suggests that all species of senesced tissue may have an N and P limitation, due to ratios

greater than 30:1 and 200:1 respectively, and may not result in a net mineralization of N

and P under aerobic conditions. Species with a nutrient limitation and restricted

microbial decomposition may release a lower amount of P compared to species with a

high nutrient content. Furthermore, the nutrient ratios are likely to be higher in the live









tissue compared to the senesced tissue due to a leaching event which may have removed

some of the labile N and P or a resorbtion of nutrients back into the living plant tissue.

Observations have shown that some species of emergent marsh vegetation, like P.

hemitomon, can remain as standing dead biomass for up to 2 months or more after

senescence (Lewis, 2005), which may allow multiple leaching events to take place before

the senesced vegetation enters the water column.

Whether the species is an annual or perennial may also influence the amount of

nutrient uptake, resorbtion, and P release rate. It is believed that the species used in this

study were all perennial, and may therefore have a higher degree of nutrient resorbtion

back into the live tissue of the plant, especially into the below ground tissues. Perennial

species may also have a lower rate of biomass accumulation, and a lower P release rate

compared to annual species (Tobe, 1998; Fraser and Karnezis, 2005). On highly

disturbed landscapes there may be a higher proportion of annual plant species which may

further contribute to eutrophication of nearby water bodies due to a potentially greater

rate of decomposition and P release, possibly as a result of higher nutrient

concentrations, NDF, and a lower residual fiber content compared to perennial species.

Different amounts of NDF, ADF, SADF, and residual fiber were seen in the

vegetative species surveyed. It is likely that species with high residual fiber content will

have a slower decomposition rate and possibly a slower P release rate compared to

species with low residual fiber content and a higher NDF fraction. The NDF content was

higher for each species in the live tissue compared to the senesced tissue which may

indicate the NDF or labile carbon fraction in the vegetation is rapidly lost in some species

after senescence. Significant differences between the NDF fractions in the live and









senesced vegetation of P. notatum and P. hydropiperoides, suggest that soluble sugars

and carbohydrates may be lost the fastest in these species or they had been senesced for a

longer period of time before collection. A leaching event could have occurred after

senescence, removing some of the labile sugars and carbohydrates associated with the

standing dead vegetation. Residual fiber also made up a higher percentage of biomass in

the senesced vegetation due to the reduction of NDF in the litter biomass.

In summary this chapter addressed the hypothesis that different species of

vegetation found on isolated wetlands in the Okeechobee Basin contained different

nutrient levels quantified by the C/N ratio and %P in live and senesced tissues, as well as

various fiber qualities. Findings suggest that the N and P content and fiber quality varied

significantly among the different species collected, which supports the original

hypothesis. Results also suggest some senesced species may provide a better substrate

for the colonization of microorganisms, and possibly a rapid rate of decomposition and P

release due to a lower C/N ratio and residual fiber fraction, a higher P content and NDF

fraction. Lastly, there were significant differences in the substrate quality between the

live and senesced tissue in terms of C:N, C:P, NDF, and residual fiber content, indicating

that leaching, and a rapid breakdown of sugar occurs shortly after senescence.














CHAPTER 3
SHORT TERM PHOSPHORUS LEACHING

Introduction

The term leaching, as used in this thesis, is a natural process where mass is lost

from senesced vegetation due to the release of soluble organic and inorganic compounds

from the plant biomass (Robertson, 1988). The leaching process is the first of three

phases of decomposition, and is subsequently followed by microbial mineralization and

physical and biological fragmentation (Valiela, 1985; Webster and Benfield, 1986).

There is evidence that leaching is correlated with rainfall when standing dead vegetation

is attached to the plant (Taylor et al, 1989; Qiu, 2005), and may continue after the litter is

incorporated into the detrital layer and immersed in the water column. The most rapid

period of leaching usually lasts from a few days to a few weeks (Davis et al., 2003)

depending on water availability, and is not mediated by microbial processes. The

greatest leaching rates usually occur during the first rainfall or the first hours after

emersion in the water column (Tope, 2003; Qiu, 2005).

Leaching processes can result in a large nutrient flux of P, N, and C from plant

biomass into the water column during fall and spring as a result of vegetative senescence

(Mitsch et al., 1989). This is the primary reason why leaching has recently become one

of the primary concerns relating to water quality in agricultural areas (Kuusemets and

Mander, 2002). The processes influencing P leaching rates are poorly understood,

however temperature, moisture, pH, Eh, P concentration, and land use, are the primary

variables influencing the P leaching rate of vegetation (Melillo, 1982; Benner, 1985;









Richardson, 1989; Moore and Reddy, 1994; Findlay et al, 2003; Flaig, 1995; Lewis,

2005).

Short-term P leaching from senesced tissue can lead to a significant release of P

into the environment. Anywhere from a 20-50% loss of the total P in the plant biomass

can be released from the vegetation in a few hours and upwards of 80% of the total P can

be released during the first 2 months of mineralization (Simpson et al., 1978; Webster

and Benfield, 1986; Tope, 2003; Qiu, 2005; Reddy et al., 2005).

Leaching is an important nutrient cycling process to consider when understanding

the fate of P within a wetland ecosystem. A significant portion (30-80%) of the

bioavailable P in the water column can be immobilized by macrophytes (Dolan et al.,

1981), depending on the plant species, the growth rate of the plant, plant density,

harvesting frequency, climate, and the oxygen availability in the sediment (Reddy et al.,

1995). Vegetation only serves as a temporary sink for nutrients however. It is estimated

that 35-80% of the total P contained in the biomass is lost to the water column, to the soil

or transported out of the system after plant senescence (Richardson, 1989; Reddy et al.,

1995). A portion of the total P obtained in the standing dead tissue is resorbed back into

the living part of the plant, preventing all the P contained within the senesced biomass

from leaching out. Nutrient resorbtion in a hardwood forest was estimated to be

approximately 30-34% of the total N and P contained within the senesced tissue (Ryan,

1982).

The main objective of this study is to quantify the amount and rate of P loss from

senesced vegetation into the water column under aerobic and anaerobic conditions, and in

response to N enrichment. This study provided a more suitable sampling frequency to









assess short-term P release associated with different species of litter in the wetland that

would have been difficult to quantify in the field. The litterbag study in the subsequent

chapter, evaluated decomposition over a longer period of time.

It is likely that different species will have various leaching rates after vegetative

senescence, due to the fact that some species have higher tissue P concentrations than

others (Hobbie, 1992; Knops et al., 2002). The species that releases the lowest amount of

P in this study relative to its P assimilation rate could be used to increase P storage on

historically isolated wetlands. In addition, knowing species specific P leaching

characteristics provides managers and landowners with guidance on beneficial and

problematic plants when trying to retain P on pastures and in wetlands.

Hypotheses

* A greater P flux will be observed under anaerobic conditions than aerobic
conditions.

* Elevated nitrogen levels will increase P flux rates when compared to ambient site
water and site water diluted with deionized water.

* Short term P release will be positively related to the concentration of P in the
senesced tissue.

Materials and Methods

Recently senesced (standing dead) vegetation was harvested from Larson Dixie

ranch in November 2004, this was the same material used in the characterization study

described in the previous chapter as well as the decomposition study explained in the

following chapter. Dominant species in the wetland center, edge, and upland were

collected then placed in paper bags according to species. Species that were adapted to

longer hydroperiods included Panicum hemitomon, which had a P concentration of 0.209

0.01 mg P/g tissue, and Polygonum hydropiperoides with 1.278 + 0.11 mg P/g tissue.









The dominant edge species was Juncus effusus, which had 0.804 0.02 mg P/g tissue.

Lastly, the dominant upland species was Paspalum notatum which had 0.830 + 0.08 mg

P/g tissue.

The senesced vegetation was air dried due to evidence that oven drying can

artificially alter chemical composition as well as leaching rates (Tope, 2003).

Approximately 1.5 grams of air dried litter material was placed in a 250 ml covered

plastic container and filled with 200 ml of treatment water. The litter added to the

containers was scaled down from the mass of litter naturally present within a Im x Im

quadrat. Vegetation was misted with deionized (DI) water approximately 12 hours

before the treatment water was added in order to reconstitute the litter so it wouldn't float

up to the top of the container and to bring the moisture content of the air dried litter up to

field conditions. Empty fluxing containers, with no vegetation added, were filled with

200 ml of treatment water and used as the control. A hypodermic needle was inserted into

the head space of the container through a rubber septum in the top of the container. This

needle functioned as a pressure relief vent. A longer needle was pushed through the same

rubber septum and below the water surface to bubble atmospheric gas (aerobic

conditions) or nitrogen gas (anaerobic conditions) (Fig. 3-1) into the container and gently

mix the water column.

In addition to species and oxygen availability treatments, three water quality

conditions were tested. Water treatments consisted of low P water collected from a

cypress slough near the research site, which is defined as "Site Water" in this chapter

(SRP concentration of 0.09 0.00 mg/L, a DOP concentration of 0.18 0.00 mg/L, and

an NO3 concentration of 0.22 0.00 mg/L). Other water treatments included site water









diluted 50% with DI water (SRP concentration of 0.04 0.00, a DOP concentration of

0.12 0.01, and an NO3 concentration of 0.15 0.00), and site water spiked with 3 mL of

1000 ppm NO3 to double the N concentration to 0.41 ppm (SRP concentration of 0.09 +

0.00, a DOP concentration of 0.18 + 0.00, and an NO3 concentration of 0.41 + 0.00).













A B

Figure 3-1. The P leaching study. A) Overview of the experimental setup and aerobic
and anaerobic treatments B) Individual fluxing container and tubing
bubbling ambient air in the water column through the hypodermic needle.

This experimental design resulted in a total of 90 flux containers. Flux containers

were covered and kept in darkness to prevent algae growth. The water column was

sampled a total of 5 times. The timescale of sampling was after 2 hrs, 24 hrs, 3 days, 7

days, and 17 days, with an initial characterization of site water before the experiment

began. At each sampling period 20 ml of water was removed from the flux container and

analyzed for SRP. At time zero and after 17 days, an additional 40 ml of water was

removed and analyzed for Dissolved Organic Phosphorus (DOP) and Total Kjeldahl

Nitrogen (TKN). The DOP samples were filtered and digested in the autoclave (Method

365.1; USEPA, 1993), while TKN samples were unfiltered and digested according to

EPA Method 351.2 (USEPA, 1993).









After each sampling period, an equivalent volume of treatment water was added to

each container to maintain a constant volume of 200 mL. The dilution factor and P mass

loss in sample water were taken into account when calculating cumulative flux of tissue P

over the course of the experiment. At the end of the experiment, litter and any newly

formed particulates within the water column were filtered using a glass fiber filter, oven

dried at 700 C for at least 24 hours, and weighed to determine the mass loss from the

litter. Samples and glass fiber filters were subsequently ground to pass a # 40 mesh sieve

using a Wiley Mill and analyzed for total carbon and total nitrogen using a Thermo

Electron Flash EA 112 Nitrogen/Carbon analyzer.

The mean P flux rates of each species were compared using Tukey-Kramer HSD

(honestly significant difference) to determine significant differences between the

cumulative P flux among species. Regression analyses and one way analysis of variance

(ANOVA) were used to determine which substrate quality parameter, quantified in

Chapter 2, was the best predictor of short term P release using the R2 and p-values.

Results

Cumulative P flux was averaged, combining three water treatments and ranged

from an average of 0.01 0.01 mg P/g tissue over a period of 17 days(P. hemitomon) to

0.96 0.17 mg P /g tissue (P. hydropiperoides) under aerobic conditions (Figure 3-2a).

Cumulative P flux from P. notatum and J. effusus fell between this range with values of

0.59.15 and 0.35+.08 mg P /g tissue, respectively, over the same 17 day period under

aerobic conditions. Leaching was greater under aerobic conditions in each species except

J. effusus, which had the greatest P flux under anaerobic conditions. The water treatment

and redox condition had little effect on the P released from senesced vegetation over a 17

days period, and an obvious trend between N enrichment or redox and P release was not









seen. The only statistically significant differences observed between aerobic and

anaerobic conditions over 17 days was in P. notatum, where P flux was significantly

higher under aerobic conditions and J. effusus where P fluxes were greater under

anaerobic conditions (Table 3-1). At the end of the study, on Day 17, the only significant

difference between aerobic and anaerobic conditions was seen in P. hemitomon.

Significant differences in flux rate among individual species under aerobic

conditions (Fig. 3-2a) occurred within the first 2 hours. At this sampling period, P flux

from P. notatum and P. hydropiperoides were significantly higher than the other 2

species, and P. hemitomon had a significantly lower P release compared to all other

species. After 1 day of incubation and for the remainder of the study, each species each

species had a significantly different flux rate.

Flux containers incubated under anaerobic conditions also began to show

significant differences in P flux rates only 2 hrs after water was added (Fig. 3-2b). At

this sampling period P. hydropiperoides again had a significantly greater P flux, while P.

hemitomon still had the smallest. This trend continued throughout the course of the

study. At day 17, there were significant differences seen among each species.

After a period of 17 days, P. hemitomon, had the least amount of P leaching from

the litter. The P released from P. hemitomon was significantly lower at every sampling

period compared to other species of vegetation under both aerobic and anaerobic

conditions (Fig. 3-2). The cumulative P flux for P. hemitomon peaked after just 2 hours,

then declined nearly to 0 mg P/g tissue. The minimum 2 hr flux value for this species

was 0.085 0.009 mg P/g tissue observed in the site water + DI under aerobic conditions,

while the maximum flux value was 0.101 0.031 mg P/g tissue observed in the site water











treatment under anaerobic conditions (Fig. 3-3a). There was not a significant difference

between redox conditions or among water treatments (Fig. 3-3) in P. hemitomon over the

course of 17 days (Table 3-1), however there was a significant difference between redox

condition for this species on Day 17 only..


1.2 -


0 3 6 9 12 15
Time (days)
- P. hemitomon ---P. notatum -A P. hydropiperoides
- J. effusus -- Control


-


0 3 6 9 12 15
Time (days)
P. hemitomon ---P. notatum -A P. hydropiperoides
--)-J. effusus -- Control


(b)

Figure 3-2. Phosphorus leaching rates of 4 senesced species averaged across all 3 water
treatments A) aerobic B) anaerobic conditions. Values represent mean (+
standard deviation).


- --* - - -- *









The species with the highest P flux under aerobic and anaerobic conditions was P.

hydropiperoides (Fig. 3-2). The P flux of this species continuously increased throughout

the experiment, with the highest cumulative flux of 1.00 + .215 mg P/g plant tissue

observed under aerobic conditions in the site water + N treatment, and 0.976 0.223

under anaerobic conditions in the site water + DI treatment after 17 days (Fig. 3-4). The

lowest 17 day P flux for this species was seen in the site water + DI treatment under

aerobic conditions (0.94 + 0.15) and site water + N treatment under anaerobic conditions

(0.74 0.08).





0.15

0.13 -4 -Site Water

0.11 --D--Site Water + N
U)
0.09 -k Site Water + DI
a-
I 0.07

0.05

a- 01
0.03 U--- ---,, ,,




-0.01 5 10 15

-0.03
Time (days)

(a)











0.8

0.7

0.6

U-M -Site Water
6. 0.5
M l -C- Site Water + N

x 0.4 A Site Water + DI

0.3
0.2


0.1

0
0 5 10 15
Time (days)

(b)

Figure 3-3. Litter phosphorus release rate for P. hemitomon litter under (a) aerobic and
(b) anaerobic conditions with 3 different water treatments. Values represent
mean ( 1 standard deviation).

The remaining 2 species, P. notatum and J. effusus had a moderate amount of P

released from the litter compared to P. hydropiperoides. There were not any significant

differences between water treatments in P. notatum under aerobic conditions, however

the site water + N treatment had a significantly lower P flux for this species under

anaerobic conditions compared to the other 2 treatments over the course of 17 days (Fig.

3-5). Juncus effusus however, had no significant differences in P flux between water

treatments under aerobic or anaerobic conditions (Fig. 3-6), however P flux was

significantly higher under anaerobic conditions compared to aerobic conditions for this

species over the course of the entire study (Table 3-1).























- -__-% %-M


-4* Site Water
-- Site Water + N
- A Site Water + DI


0 5 10 15
Time (days)


-* -Site Water

-0-Site Water + N

- k Site Water + DI


0 5 10 15
Time (days)


Figure 3-4. Litter phosphorus release rate for P. hydropiperoides litter under a) aerobic
and b) anaerobic conditions with 3 water treatments. Values represent mean
(+ 1 standard deviation)


1.2





S0.8
x

o. 0.6


0.4
E

0.2


0


1.2


1 1.0
0.

- 0.8


. 0.6
0.

. 0.4


L 0.2


0.0



















- *-Site Water
-0- Site Water + N
- A Site Water + DI


0 5 10 15
Time (days)


- - -


- *-Site Water

-C-Site Water + N
- A Site Water + DI


5 10 15
Time (days)


Figure 3-5. Phosphorus release rate for P. notatum litter under a) aerobic and b)
anaerobic conditions with 3 water treatments. Values represent mean (+ 1
standard deviation).


0.7

S0.6

C- 0.5

x 0.4
a.
w 0.3

E 0.2

0.1

0.0
















0.50



0.40


x 0.30



" 0.20
E


0.10



0.00


5 10 15
Time (days)


-4* Site Water
-- Site Water + N
- k Site Water + DI


0 5 10 15
Time (days)


Figure 3-6. Phosphorus release rate for J. effusus litter under a) aerobic and b) anaerobic
conditions with 3 water treatments. Values represent mean ( 1 standard
deviation).


-* -Site Water
-- Site Water + N
- A Site Water + DI


0.9

6 0.8

= 0.7

0.6

x 0.5

0.4

| 0.3
E
5 0.2

0.1

0.0










Table 3-1. Cumulative P release (mg P/g litter) and percent tissue P released from four
species over a 17 day period, under aerobic and anaerobic conditions. Values
represent mean ( 1 standard deviation). Lowercase letters indicate
significant differences over the entire 17 day study between species, water
treatment, and redox conditions with a p-value of .05.
17 Day Total Cumulative Flux (mg P/g tissue)
% P %P
Species Treatment Aerobic Released Anaerobic Released
0.002 0.007
Site Water (a) 0.003 0.80 (a) 0.001 3.55
Panicum -0.002 0.003
hemitomon Site Water + N (a) 0.001 -1.08 (a) 0.004 1.28
0.024 0.011
Site Water + DI (a) 0.026 11.61 (a) 0.006 5.45
0.551 0.414
Site Water (b) 0.126 66.37 (c) 0.203 49.86
Paspalum 0.694 0.197
notatum Site Water + N (b) 0.170 83.58 (d) 0.034 23.78
0.539 0.377
Site Water + DI (b) 0.144 64.94 (c) 0.022 45.41
0.925 0.911
Site Water (e) 0.149 72.34 (e) 0.066 71.22
Polygonum 1.004 0.738
hydropiperoides Site Water + N (e) 0.215 78.55 (f) 0.082 57.74
0.937 0.976
Site Water + DI (e) 0.149 73.32 (e) 0.223 76.32
0.355 0.461
Site Water (g) 0.112 44.16 (h) 0.030 57.33
0.342 0.636
Juncus effusus 0.342 0.636
Site Water + N (g) 0.079 42.52 (h) 0.275 79.09
0.351 0.517
Site Water + DI (g) 0.055 43.72 (h) 0.074 64.32

Over the span of 17 days, SRP concentration significantly increased from the initial

0.09 mg/L in the site water flux containers of every species except P. hemitomon.

Approximately 96% of the P in the water column at the end of the study was SRP, a

readily labile form, while the small remaining P fraction was DOP (Table 3-2).

In general TKN concentrations were higher under anaerobic conditions compared

to aerobic conditions, however few significant differences were seen (Table 3-3).

Polygonum hydropiperoides was the only species where TKN concentrations were higher

under aerobic conditions (in site water and site water + N treatment). In addition there

were not any significant differences observed in the TKN concentrations among species

or water treatments.










Table 3-2. Mean concentrations ( 1 standard deviation) of water column P for various
species under aerobic and anaerobic conditions after 17 days. Negative DOP
values are due to high standard errors and represent virtually no measurable
DOP in the water column.
SRP (mg/L) DOP (mg/L)
Species Treatment Aerobic Anaerobic Aerobic Anaerobic
P. hemitomon Site Water 0.03 0.01 0.03 0.01 0.06 0.02 0.09 0.02

P. notatum 2.72 0.67 2.07 1.09 0.22 0.06 0.15 1.07
P.
hydropiperoides 4.99 0.84 4.84 0.37 0.18 1.00 -0.03 0.33
J. effusus 1.87 0.65 2.37 0.25 0.05 0.53 0.13 0.33
Site Water +
P. hemitomon N 0.02 0.00 0.03 0.00 0.06 0.03 0.07 0.01
P. notatum 3.57 0.88 0.92 0.16 0.23 1.91 0.18 0.13
P.
hydropiperoides 5.51 1.07 3.97 0.34 0.05 1.11 0.04 0.64
J. effusus 1.77 0.34 3.61 + 1.65 0.06 0.31 0.14 1.56
Site Water +
P. hemitomon DI 0.11 0.00 0.02 0.00 0.09 0.16 0.06 0.01
P. notatum 2.68 0.74 1.74 0.17 0.12 0.70 0.23 0.13
P.
hydropiperoides 5.13 0.73 5.17 1.27 0.23 0.57 0.11 1.29
J. effusus 1.77 0.27 2.67 0.43 0.09 0.28 0.24 0.20


Mean concentrations (a 1 standard deviation) of water column TKN for
various species under aerobic and anaerobic conditions after 17 days. Letters
ni dictate significant differences between species and n


Table 3-3.


TKN (mg/L)
Species Treatment Aerobic Anaerobic
3.61 6.21
P. hemitomon Site Water (a) 2.42 (a) 0.24
5.53 6.31
P. notatum (a) 0.32 (a) 0.68
7.34 6.08
P. hydropiperoides (a) 3.54 (a) 1.69
3.46 6.47
J. effusus (a) 0.58 (a) 2.13
4.41 6.62
P. hemitomon Site Water + N (A) 1.19 (A) 1.54
5.15 6.49
P. notatum (A) 1.08 (A) 0.60
6.02 5.4
P. hydropiperoides (A) 1.28 (A) 1.55
3.81 6.31
J. effusus (A) 0.18 (B) 1.50
7.07 6.47
P. hemitomon Site Water + DI (a') 3.48 (a') 1.54
6.02 9.47
P. notatum (a') 1.94 (a') 5.17
3.53 5.13
P. hydropiperoides (a') 0.38 (a') 2.28
2.97 6.78
J. effusus (b') 0.73 (b') 1.04


.










After 17 days of incubation, there were no significant differences in the mass loss

among species, although a substantial decline in mass occurred. The largest mass loss

was observed in P. hydropiperoides while P. hemitomon had the smallest.


25 n


u 15
In
0
-j


5 10


J. effusus


P. hydropiperoides P. notatum P. hemitomon


Figure 3-7. Percent mass loss from the vegetation over a 17 day period. Values
represent mean + 1 standard deviation.

Multiple ANOVAS were run in order to determine which parameter provided the

best predictor of the amount of P released from vegetation (Table 3-4). Regressions were

made between the cumulative P flux on day 17 and initial tissue concentration of N, P,

and C and various ratios of these parameters. It was determined with 99.99% confidence

that a relationship exists between P contained in the tissue and the amount of P that tissue

will release to the water column. Tissue P content explained 89% of the variability in P

flux in site water under aerobic conditions on Day 17, while %NDF explained 76% of the

variability in P flux (Fig. 3-8).









Table 3-4. The P-Values associated with initial nutrient parameters of senesced tissue to
estimate the best predictor of P flux for site water under aerobic conditions on
Day 17.
Substrate Quality Parameter P-value R 2
%C .314 .101
% SADF .252 .129
%N .110 .235
% Residual Fiber .104 .242
C:N .062 .306
% ADF .030 .390
C:P .001 .670
N:P .001 .687
% NDF .0002 .757
%P <.0001 .887


1.25


U 1

0.5
o 0.75
E
x 0.5

,U
= 0.25


00


-0.25 -- I|'---
0 .025 .05 .075 .1 .125 .15
%P senesced

Figure 3-8. Bivariate fit of cumulative P flux by initial senesced tissue P content.
Correlation is for site water treatment under aerobic conditions on day 17.

Table 3-5 shows R-square values for the correlations between cumulative P

release and tissue P concentration for each sampling period under aerobic and anaerobic

conditions. The correlation improves throughout the course of the study under aerobic









conditions with the highest R value occurring on day 17, while under anaerobic

conditions the R value is the highest on day 1 and 7. Phosphorus concentration in the

litter is a slightly better indicator under anaerobic conditions as to the amount of P that

will be released from that species. Other initial substrate quality parameters such as

%NDF, N:P, and C:P had significant p-values, however the relationship between these

parameters and the amount of P released from the senesced vegetation was not as strong

compared to the relationship between the initial %P in the senesced tissue and cumulative

P release.

Table 3-5. The R-Square values from %P and cumulative P release correlation in site
water under aerobic and anaerobic conditions
R-Square values
Time (days) 0.08 1 3 7 17
Aerobic 0.732 0.861 0.853 0.864 0.887
Anaerobic 0.885 0.914 0.911 0.914 0.908

Table 3-6. The P-Values associated with initial nutrient parameters of live tissue to
estimate the best predictor of P flux for site water under aerobic conditions on
Day 17.
Substrate Quality Parameter P-value R2
N:P 0.7696 0.009
% Residual Fiber 0.0967 0.251
% ADF 0.0129 0.477
% C 0.0076 0.526
% P 0.0014 0.658
C:P 0.0007 0.696
% N <.0001 0.829
C:N <.0001 0.830
% SADF <.0001 0.862
% NDF <.0001 0.883


Initial substrate quality parameters of the live tissue characterized in Chapter 2,

were also correlated with the short term cumulative P release from each senesced species

of vegetation. The strongest correlation was between short term P flux and % NDF or the

labile fiber fraction associated with the live tissue with an R2 of .883 (Table 3-6).









Discussion

Findings suggest the potential for vary rapid and significant amounts of tissue P

release from plant litter upon exposure to rainfall or inundation, and the amount of P

released from the senesced tissue was primarily dependant on the species type.

Phosphorus release from the 4 litter species tended to peak during the first 2 hours after

the water treatment was added to the vegetation through the first 3 days of the study.

After day 7, bacterial growth began to appear in the flux containers likely resulting in

water column P uptake and a possible explanation for the slight decline in cumulative

flux values.

Approximately 70% of the P in the tissue of P. hydropiperoides was released on

average after 17 days, while P. notatum and J. effusus both released 55% of the P that

was initially associated with the biomass, and P. hemitomon released only 4% of the P

associated with its biomass. The P flux values closely follow the amount of P each

species originally contained in the biomass, which helps explain why the cumulative P

flux from P. hydropiperoides was greater than the cumulative flux for other species

because this species had the highest percentage of P initially, while P. notatum and J.

effusus initially had similar P concentrations, and P. hemitomon had the least amount of

P. The cumulative amount of P released over the 17 day period is just below the 80%

value that was lost during biomass decomposition in the study preformed by Reddy et al.

(1995) and below the 35% value reported by Richardson (1989).

Panicum hemitomon contained only 25% of the P in the senesced tissue of P.

notatum and J. effusus, and 6.2% of the P contained in P. hydropiperoides. In addition to

having the lowest P flux, P. hemitomon also had the highest initial C:N and the lowest

NDF fraction out of all the live species surveyed. Neutral detergent fiber content was the









parameter that was the best predictor of short term P release in the live tissue. The small

amount of P within the tissue of P. hemitomon was held onto tightly throughout the

study.

A large quantity of P was released from the senesced tissue within a relatively short

period of time. The rate at which P was released from the senesced vegetation quickly

increased between 2 hours and 3 days, after which the release rate slowed and leveled off,

then stopped or reversed. This information suggests that a potentially large amount of P

can be released from the senesced tissue shortly after vegetative senescence due to

rainfall or upon entering the water column. Results from this study are similar to a

leaching experiment preformed by Davis et al. (2003) which reported the most rapid

leaching from Rhizophora mangle L. leaf litter after 2 days of incubation. High

quantities of P fluxing into the water column after senescence could have the potential to

increase the eutrophication of seasonally isolated wetlands and downstream waterways.

Hydrological restoration could also cause species shifts which may significantly impact

the amount and rate of P leaching.

In each species with the exception of P. hydropiperoides, there was a negative P

flux observed in either aerobic or anaerobic conditions during the last 7-14 days of the

study. A decrease in the cumulative P flux can only be explained by microbial uptake of

P from the water column or P uptake by the litter. Near day 7, a bacterial growth is

thought to have occurred around the needles bubbling gases into the fluxing containers

since light was available during incubation. It is likely that the negative slopes depicted

in the graphics above were due to microbial nutrient uptake. In future studies, it may be









interesting to have a light vs. dark treatment to quantify the P uptake by algae and

bacteria.

Looking at the cumulative flux curves for each species, it is estimated that P

leaching took place for approximately the first 7 days of this study. Around day 7, the P

flux slowed, leveled off, or became negative, suggesting microbial mineralization

processes began to take over. It was suspected that the water spiked with N may increase

P release during mineralization because all species initially had a high C:N, and would

have been N limited under aerobic conditions. A nitrogen addition would increase the N

concentration as well as the amount of nutrients available to microbes, and therefore

stimulating mineralization, C breakdown and P release beyond a 7 day period.

The low NDF fraction and initial P concentrations in P. hemitomon is likely the

reason why this species had the lowest change in mass after a 17 days, while P.

hydropiperoides underwent the highest mass change because it's initial NDF fraction was

higher than the 3 other species. It is likely that the NDF fraction or the soluble sugars

and starches associated with the litter is what mostly contributed to the mass loss of the

litter, along with the N and P associated with this fiber fraction. A even higher mass loss

was reported in a 21 day leaching study preformed by Davis et al. (2003), where leaching

accounted for 33% of the dry mass from R. mangle leaves. Physical factors such as the

fiber quality of the litter are likely to influence short term leaching and mass loss more

than the nutrient content of the litter or the C/N ratio, which may influence P release over

a longer period of time. This may be a possible reason why the %NDF in the live tissue

had the strongest correlation with short term P release. It is thought that the majority of









the mass differences resulted from nutrient leaching through day 7, and perhaps microbial

mediated decomposition processes for the remainder of the study.

The empty flux containers used as the control had an extremely low amount of

measurable P flux. This could be attributed to a small amount of sediment that was

suspended in the water column when the treatment water was added to the containers.

The control did not have any vegetation, and therefore had a post experiment C:N of 10.0

+ 6.2 as a result of bacterial growth which appeared in all of the containers around day 7.

Overall, there did not appear to be a large difference in P release under aerobic or

anaerobic conditions, and there wasn't a clear trend as to which condition caused the

most P to flux from the vegetation. Even though there were no significant differences

between treatments on Day 17, N enrichment seemed to enhance P release under aerobic

conditions, but inhibited P flux, or had no effect on P release under anaerobic conditions.

Anaerobic microbes often require a lower amount of available N to carry out metabolic

activities since NO3 is often not available in reduced environments. It is possible that the

site water + N treatment had no effect under anaerobic conditions because the C:N was

initially lower than 80:1; therefore the species used in this study would not be considered

N limited under anaerobic conditions. Because enough N was already present to carry

out metabolic activities under anaerobic condition, additional N may not have made a

difference in increasing the P mineralization after Day 7.

The large proportion of SRP in the water column is significant because SRP is an

extremely labile form of P that is readily bioavailable. After rainstorms in the

Okeechobee Basin, there is a high possibility that substantial amounts of labile P can

easily be transported out of ditched wetlands and increase P loading downstream









waterways and possibly even to Lake Okeechobee. Table 3-3 illustrates that TKN

concentrations were slightly higher under anaerobic conditions, although few significant

differences were observed. It is likely that there would be a higher concentration of

reduced forms of N in the anaerobic treatment due to the utilization of N as an electron

acceptor for microbial respiration.

Results from this experiment show that approximately 90% of the P releases from

litter after vegetative senescence can be predicted using the initial P tissue concentration

of the senesced tissue (Figure 3-8), and therefore P fluxes are likely to vary depending on

the species type. Live tissue parameters such as the % NDF can also be used to predict

short term P flux in the event senesced tissue is not available. The relationship between

short term P release and the % NDF contained in the living tissue appears to be just as

strong as the relationship between short term P release and the initial % P contained in

the senesced tissue. This could indicate that a substantial amount of P contained in the

senesced vegetation is contained within the labile NDF fraction of the plant, and is

released along with the soluble sugars and carbohydrates. In addition, there were a

greater number of significant relationships between short term P release and an initial live

substrate quality characteristic such as the %SADF, C:N and %N. The relationship

between short term P release and the initial P content in the live tissue may have not been

as strong due to relatively high amount of P resorbtion from senesced tissue into live

plant tissue. If changes in the P mass can be predicted with confidence using a live tissue

parameter, the application of the data may be more useful since live vegetation is more

abundant than senesced tissue throughout most of the year.









The results from this short term leaching study are important because it is likely

that short term P release from senesced tissue can be predicted with a high level of

confidence for many different species found on historically isolated wetlands in the

Okeechobee Basin, not just the four species used in this study. Therefore the release and

transport of P may be managed using short term P release rates predicted by an initial

substrate quality parameter in live or senesced tissue; species that may release a high

degree of P after senescence can be avoided. Future studies may indicate that P.

hemitomon should be promoted along with hydrological restoration because this species

may reduce P transport and improve the water quality in the Okeechobee watershed

compared to other species within the wetland. It is likely however, that P. hemitomon

will only persist on these landscapes if grazing is limited. If overgrazed, there may be a

shift to an alternate species with a different P release rate.

Relative to the original hypotheses there were no strong trends to show that oxygen

availability and N enrichment enhance P leaching, however P. hemitomon, the species

with the lowest initial P concentration, did release the lowest amount of P released

compared to other species of vegetation. Also the majority of P loss was attributed to

leaching rather than microbial mineralization processes that seemed to take effect after 7

days of incubation. This study suggests that leaching is an extremely important

component of P cycling and nutrient movement especially in the fall when many species

begin to senesce and there is a high proportion of standing dead vegetation. This study

not only provides an increased understanding of nutrient losses from senesced vegetation,

but also provides some degree of species specific leaching potential based on initial tissue









P concentration. This information should provide managers with better insight into

wetland related Best Management Practices and reduction of nutrient discharge offsite.














CHAPTER 4
LITTER DECOMPOSITION AND LONG TERM PHOSPHORUS RELEASE

Introduction

Organic matter is made up primarily of carbon, and is a driver of many ecosystem

heterotrophic processes. Organic carbon can function as a nutrient source for microbes,

can adsorb toxic compounds or nutrients, and can provide an exchange capacity for

cations (Cotrufo, 2006). The breakdown of organic matter or microbial mineralization of

nutrients is known as decomposition. Decomposition occurs in all ecosystems and it is

often accompanied by a net release of nutrients from senesced vegetation (Mitsch et al.,

1989). The environmental factors influencing organic matter decomposition include

moisture, temperature, electron acceptor availability, and pH (Melillo, 1982; Benner,

1985; Qualls and Richardson, 2000).

Plant tissue substrate quality partly determines decomposition rates in addition to

environmental factors listed above (Melillo, 1982; Berg, 1998; Villar, 2001). Substrate

quality is determined partly by the recalcitrance or liability of the fiber fractions within the

plant tissue. In addition, the amount of available nutrients such as N and P relative to the

available carbon content is also a component of substrate quality. Substrate quality is an

important parameter to consider when estimating decomposition rates of different plant

species because it can either enhance or inhibit microbial colonization of the litter. The

more labile organic polymers present in the litter material such as starches and sugars, the

easier microbes can utilize the litter as a substrate for nutrients.









In most cases nutrient enrichment has been shown to increase decomposition

because N and P are often limiting factors controlling microbial growth (Qualls, 1984;

Taylor et al., 1989; Corstanje et al., 2006). Others have found that nutrient enrichment

has no effect or can cause a lower rate of decomposition due to a possible C limitation

(Berg et al., 1998; Carreiro et al., 2000; Villar et al., 2001). It may also be that the

influence of nutrient content is variable during different stages of decomposition. Lewis

(2005) suggests that litter nutrient content may be the controlling factor during the initial

phase of decomposition, or before 30% of the initial mass is lost but thereafter available

carbon may become the primary factor limiting decomposition rate.

Lignin is the organic compound that is the most recalcitrant or resistant to

decomposition. It is the limiting factor of long-term decomposition because it can only

be digested by a few organisms. Microbes are able to degrade lignin with the help of

various enzymes (Benner, 1984); however it is extremely resistant to decomposition

under anaerobic environments (Criquet et al., 2001). Lignin helps provide structural

support to plants so they can remain upright, as well as provide a protective barrier

against microbial attack (Hammel, 1997). Lignin content or the lignin/N ratio may be a

better predictor of decomposition rates compared to the P concentration or the C/N ratio

(Melillo et al., 1982; Sinsabaugh et al., 1993; DeBusk and Reddy, 1998; Rowland and

Roberts, 1999; Carreiro et al., 2000). For instance litter with a high N concentration and

low lignin content is likely to be a relatively labile substance with a rapid decomposition

rate.

Lignocellulose is another recalcitrant C fraction. It is comprised of approximately

25% lignin and it forms a protective layer around the cellulose tissue, preventing rapid









breakdown (Donnelly et al., 1990). Cellulose generally decomposes faster than lignin,

but slower than more labile fiber fractions such as sugars, carbohydrates, and

hemicellulose. Lignocellulose is formed as labile components of the vegetation

decompose and lignin and cellulose decomposition rates slow down (Melillo et al., 1982,

1989). Although lignocellulose formation may increase the litter recalcitrance,

decomposition will continue with the addition of nutrients or new substrates to the litter

(Donnelly et al., 1990), if optimum conditions for decomposition such as warm

temperatures, moisture, and oxygen are provided.

When living plants are present in a wetland, they are likely accumulating P and

incorporating it into their biomass, however when the plant senesces and litter enters the

water column, the litter may become a P source (Moore and Reddy, 1994). During

different phases of decomposition, various amounts of P could potentially be released

from litter depending on how much P is stored in the fiber fractions of the litter. The

species type is likely to be an important factor when estimating long-term P release since

some species accumulate more nutrients than others (Hobbie, 1992; Knops, et al., 2002).

McJanet et al. (1995) reported significant differences in the N and P content of 41

different wetland plant species with N concentrations ranging from 0.25-2.1% and P

concentrations of 0.13-1.1%. The age of the senesced material may also play a role in the

rate of decomposition. Younger tissues generally have less recalcitrant fractions and a

higher proportion of N and P associated with their biomass compared to older plant

tissues.

The emergent macrophyte species used in this study retain standing dead biomass

for a longer period of time (up to one year) compared to submerged aquatic vegetation,









which have virtually a non-existent standing dead phase. During the standing dead

phase, biomass is likely exposed to leaching from rainfall as well as colonization and

initial breakdown by aerobic microbes. Therefore, standing dead litter may go through a

significant degree of nutrient leaching before the litter enters the water column. Leaching

occurring before the litter enters the water column could affect nutrient and fiber

character of substrate quality and thereby influence decomposition rate as well as long-

term P release.

The fiber quality, hydroperiod, and depth of inundation are likely to influence the

decomposition rate of various species which may significantly affect the quantity of P

that is released during decomposition. A higher degree of decomposition is likely to

occur under aerobic conditions; however moisture may be a limiting factor during the

standing dead phase. Recalcitrant organic matter may be resistant to breakdown under

anaerobic conditions, and may retain P in the recalcitrant portions of the litter, therefore

litter may serve as a storage mechanism for P as organic matter continues to accumulate

in the wetland and eventually form new soil. This may help decrease the P concentration

in the surrounding water column as well as reduce P export from the wetland and nutrient

loading downstream.

Although the P assimilation rate by plants is a critical factor in assessing potential

efficacy of wetland P storage, if plants are allowed to senesce in-situ, the overall

effectiveness of a wetland to immobilize and store P is also dependant on the release rate

of P during litter decomposition. The relationship between decomposition rates and

substrate quality, as well as substrate quality and plant species suggest that the type of

plant species present in a wetland is also a critical factor regulating litter mineralization









rates and long-term P storage. Therefore, this chapter investigates long-term

decomposition rates and compares these rates to various substrate qualities and

environmental parameters. A better understanding of the relationship between tissue

substrate quality and P litter mineralization will assist in evaluating the implications of

vegetative community change in response to hydrologic restoration on downstream P

load to Lake Okeechobee.

Hypotheses

* Decomposition rates will be lowest in wetland centers due to longer hydroperiod
and anaerobic conditions.

* Tissue with high C:N ratios and or high residual fiber content will have slower
decomposition rates than tissue with low C:N ratios or low residual fiber content.


Materials and Methods

Field Methods

Litter decomposition rates were determined by measuring percent mass loss of

standing dead vegetation collected from wetland center, wetland edge, and upland

communities. Recently senesced standing dead biomass was collected from P.

hemitomon, P. hydropiperoides (representative of the dominant vegetation in the wetland

center), J. effusus (representative of the dominant edge species), and P. notatum

(representative of the dominant upland species) during November 2004. The biomass

was air dried in the laboratory for 10 days resulting in less variability in the litter

moisture content without altering the chemical composition by oven drying. The litter

was then manually broken into smaller segments approximately 7-9 cm in length. Six

grams of air dried vegetation was placed in 15 cm x 15 cm litterbags made of 1mm

screen mesh, and each litterbag was individually labeled with plastic labels. The initial









amount of litter in each bag was based field litter biomass estimates and normalized to 15

x 15cm area. The litterbags were deployed on April 1st, 2005 in 3 isolated wetlands

located within the same pasture. Litterbags were deployed along a hydrological gradient

from the center of the wetland to the upland in four different wetland zones defined as

center, edge, transitional zone, and upland (Fig. 4-1), which had a hydroperiod of 100-

200 days / yr, 5-60 days / yr, and 0-15 days / yr respectively (Fig. 4-2).

Five sets (one set for each time collection) of 8 litterbags (2 bags for of each of four

species) were deployed in each zone within the three wetlands to determine the influence

of different environmental factors and substrate quality on decomposition rates (Fig. 4-2).

Each set of litterbags was tied together with monofilament fishing line, spaced evenly

across the soil surface. Monofilament fishing line was attached to the soil with a metal

nail on the corner of each bag, and then covered by 3 x 3 cm plastic nylon netting to

prevent the cattle from eating or relocating the litterbags (Fig. 4-3). Litterbag coordinates

were determined using GPS, and mesh coverings were marked with one inch diameter

PVC poles to aid in relocation.

One set of litterbags from each zone in each wetland was harvested during each

sampling period. Litterbags sets were harvested June 1st, 2005, July 27th, 2005,

December 8th, 2005, and April 5th, 2005, representing 2, 4, 8, and 12 months of exposure.

The location of each set of litterbags was found using GPS and a metal detector. Once a

set of litterbags was located, they were placed in plastic Ziploc bags and stored in a

cooler and transported to the laboratory.















Transitional '"



Edge


Center




a = site of litterbc deployment





Figure 4-1. Litterbag distribution and deployment locations in 4 hydrological zones
within the wetland.


1.000



0.500 -



0.000


O -0.500 "



-1.000 -



-1.500

61 "6 7-. 77
w 6


















120

100

7'


0 ---


Center Edge


Figure 4-2.


H 11/1/04
H 3/1/05
07/1/05


Upland


Hydrological information for the seasonally isolated wetlands on Larson
Dixie Ranch. A) The yearly stage information in meters from April 1,
2004-March 10, 2006, B) the average hydroperiod for the center, edge, and
upland zones during November, March, and July.


Figure 4-3. Aerial view of the three historically isolated wetlands on Larson Dixie
Ranch where the litterbags were deployed.






60















A B













C D













E F

Figure 4-4. Litterbag experiment showing A) litterbags and netting in upland, B)
litterbags attached to soil and covered by nylon netting as a precaution
against cattle, C) a close up of litterbag filled with P. notatum, D) P.
notatum growing in and through the litterbag after 8 months of exposure, E)
the large amount of vegetation covering the bags in the upland after 8
months, F) vegetation covering netting in the transitional zone after 8
months.









Laboratory Methods

After litterbags were collected from the field and brought back to the laboratory,

vegetation growing on the outside of the litterbag was removed and sediment was rinsed

out of the litterbags using tap water. The procedure for sediment removal was to dunk

each litter bag into its own individual container of tap water, gently agitate the bag while

submersed and let drain. This procedure was repeated three times. Litterbags were not

rinsed in the field because site water was not always present at the time of collection.

After rinsing, litterbags were cut open and remaining green vegetation and roots were

removed. Litterbag contents were then placed in a labeled paper bag and dried at 600 C

for at least 72 hrs. After drying, litter was weighed to determine mass loss after each

exposure time. In an effort to further reduce mass change errors associated with

inclusion of field sediments into the litter, litterbags deployed in the center were also

sieved using a 250 [m or #60 mesh to remove small particulate matter mostly

representing mineral soil. After drying and weighing all samples were ground using a

Wiley Mill and passed through a # 40 mesh. Analysis of total carbon and total nitrogen

was determined using a Thermo Electron Flash EA 112 Nitrogen/Carbon analyzer.

Tissue phosphorus content was determined using acid digestion of ashed tissue and

analyzed using colorimetric procedures (Method 365.4; USEPA, 1993). An Ankom 200

Fiber Analyzer was used to quantify the neutral detergent fiber (NDF) which is the

starches, sugars and labile components associated with the plant tissue, the acid detergent

fiber (ADF) which is the hemicellulose fiber fraction, the strong acid detergent fiber

(SADF) which is the cellulose fiber fraction, and the residual fiber or "lignin"

percentages of the litter (Ankom Technology Corp., 1998a). Fiber analysis was only

conducted on litter collected 12 months after deployment.









The decomposition rates and P content means of all species were compared using

Tukey-Kramer HSD (honestly significant differences) method to determine significant

differences between species decomposition and species P mass. Regression analyses and

one way analysis of variance (ANOVA) were used to determine which substrate quality

parameter, quantified in Chapter 2, was the best predictor of % mass loss and long term P

release after 12 months using the R2 and p-values.

Results

There were no significant differences in decomposition rates among the three

wetlands when all species were combined. Overall, results indicate a slightly slower

decomposition rate in Larson South (LS) compared to Larson East (LE) which had the

highest percent mass loss of the three wetlands (Fig. 4-5).

Litter decomposition in the four hydrologic zones (center, edge, transitional zone,

and upland) did not vary significantly over the course of this study. The only significant

difference between litter mass loss and hydrologic zone occurred after 2 months when the

center litterbags had a higher percent mass loss compared to the other three zones (Fig. 4-

6). Decomposition rates over time were best modeled by an exponential decay curve,

with similar rates between zones.

Although there were not any additional significant differences in decomposition

rate between the four hydrologic zones after the first 2 months, there were significant

differences in decomposition rates among the four different plant species. Paspalum

notatum had a significantly higher decomposition rate over 12 months compared to the 3

other species of senesced vegetation. During the first 2 months, there were not any

significant differences among the decomposition rates of the four species. However, P.










notatum had a significantly greater percent mass loss compared to the three other species

during the 4 and 12 month sampling periods (Fig. 4-7).



100 m
-<-LE
--- LW
90 i- LS


S0 .


70 \ -


60 --


50 -


40
0 2 4 6 8 10 12
Time (months)

Figure 4-5. Litter decomposition of all species in all hydrologic zone among the 3
different wetlands, Larson East (LE), Larson West (LW), and Larson South
(LS). Values represent mean + 1 standard deviation.

The greatest amount of decomposition occurred in P. notatum litter in the upland

with roughly a 62% mass loss after 12 months. The least amount of decomposition

occurred in P. hydropiperoides in the edge (a 32 % mass loss after 12 months) (Table 4-

1). An average of 43 % of the original litter mass from each species within the four

hydrological zones had decomposed after 12 months. In addition, the highest rate of

decomposition in every species was observed during the first four months where

approximately 34% of the original biomass had decomposed, while roughly only an

additional 9% of the remaining biomass was lost during the next 8 months (Fig. 4-7).










100
< y y 84.815e-0 0408x
SR2 0.77
90 \ 8860
y = 88.61e-36x
SR = 0.69
S 80 T y 92.16e-0 046x
00 \ 2 = 0.90
*y 91.37e-0 049x
70 "X
E.., R2=0.89

CO ------------^


50 -

40
0 2 4 6 8 10 12
Time (months)

-0--center -4-edge A- transitional x upland


Figure 4-6. Decomposition in 4 wetland zones over a 12 month period. Values
represent mean + 1 standard deviation.

None of the initial substrate quality parameters characterized in Chapter 2, in either

the live or senesced tissue, was highly correlated with the mass loss among the 4 different

species after 12 months in the field. The initial substrate quality parameter in senesced

tissue that most closely associated with the mass remaining after 12 months was carbon

content, this relationship grew progressively stronger and was not significant after only 2

months however. Although the relationship was significant, it only explained 37% of the

variability in the decomposition rate (R2 = .368 and a p-value < .0001). The substrate

quality parameter in the live tissue which had the best association with mass loss after 12

months was the initial N/P ratio (R2 = .262 and a p-value < .0001), however this

relationship was not consistent throughout the course of this study.










100

90

80 \

"o 70 ---

W 60

S50

40 y 91.41e-0039 y 89.51e- 066x y 86.5 e-0 035x X y 90.74e-0 042x
R2 0.87 R2 = 0.85 R2 = 0.73 R = 0.85
30
0 2 4 6 8 10 12
Time (months)

-0- P. hemitomon -U- P. notatum A- P. hydropiperoides x J. effusus


Figure 4-7. Average litter decomposition of each species over a 12 month period.
Values represent mean + 1 standard deviation.

There were not any significant differences in the % P remaining in any of the four

hydrological zones throughout the entire 12 month study. The % P in the center

remained relatively constant during 12 months of deployment. After 12 months, the % P

contained in the litter was higher in the edge and transitional zone compared to the center

and the upland, although the differences in % P were not significant (Fig. 4-8).

During the first 2 months, P. hydropiperoides released approximately 50% of the

initial P contained in the biomass, while J. effusus released approximately 30% of its

original P mass during the first 2 months in the field. Panicum hemitomon had a

significantly greater % P remaining compared to the other 3 species surveyed during each

sampling period over 12 months, while P. hydropiperoides had the significantly lowest %

P remaining during every sampling period except after 4 months. In addition, P.










hemitomon accumulated approximately 50% more P than was originally contained in the

biomass of the litter during the first 4 months of deployment (Fig. 4-9).

Table 4-1. Species decomposition in each zone of litterbag deployment after 12 months.
Values represent mean ( 1 standard deviation).
Species Zone Mass Loss (%)
P. hemitomon center 42 7
edge 36 15
transitional 41 7
upland 36 8
P. notatum center 58 4
edge 48 9
transitional 58 5
upland 62 13
P. hydropiperoides center 36 16
edge 32 13
transitional 43 6
upland 39 4
J. effuses center 42 19
edge 36 10
transitional 42 10
upland 42 5


300


250


C 200


E 150
0a
100


50


0


0 2 4 6 8 10 12
Time


-*-- Center -U- Edge A- Transitional


Change in % P in the 4 hydrological zones over a
represent mean + 1 standard deviation.


X Upland


12 month period. Values


Figure 4-8.










300


250

>---------------<
200 -"


150 .-


100 ----

50 -


0-
0 2 4 6 8 10 12
Time

-- P. hemitomon --- P. notatum A- P. hydropiperoides X J. effusus

Figure 4-9. Change in % P in the 4 dominant species over a 12 month period. Values
represent mean + 1 standard deviation.

Table 4-2. Significance and R2 values of the relationship between the change in P content
and initial senesced substrate quality characteristics for each sampling period.
Substrate Time (months)
Quality 2 4 8 12
Parameter p-value R2 p-value R2 p-value R2 p-value R2
% SADF <.0001 .283 <.0001 .263 <.0001 .271 <.0001 .307
% C <.0001 .384 <.0001 .165 <.0001 .330 <.0001 .250
In C:N <.0001 .479 <.0001 .520 <.0001 .483 <.0001 .544
In N:P <.0001 .573 <.0001 .521 <.0001 .573 <.0001 .530
% N <.0001 .586 <.0001 585 <.0001 581 <.0001 .626
% ADF <.0001 .638 <.0001 .550 <.0001 .614 <.0001 .626
% Residual <.0001 .694 <.0001 .581 <.0001 .665 <.0001 .664
Fiber
In C:P <.0001 .719 <.0001 .685 <.0001 .720 <.0001 .701
% P <.0001 .818 <.0001 .738 <.0001 .806 <.0001 .787
% NDF <.0001 .813 <.0001 .767 <.0001 .802 <.0001 .796

When many initial substrate quality parameters in the senesced tissue were

correlated with the change in P over time the strongest association was between the initial

NDF fraction contained in the biomass and the loss or gain of P mass through time (Table









4-2) with an R2 of .796 after 12 months and a p-value <.0001 (Fig. 4-10). In addition, the

correlation between these two parameters remained strong regardless of hydrologic zone,

and the initial P content explained at least 77% of the variability in the loss or gain in P

mass regardless of sampling time.

0.6-
0.5 P loss/gain (g) = -1.77 + 0.07 % NDF

0.4 R .796
0.3 -
S0.2 -

0.1 -
0
-0.1 P. hemitomon

-0.2 P. notatum
-0.3 hydropiperoides

-0.4 --
-0.5 -

21 22 23 24 25 26 27 28 29 30

% NDF


Figure 4-10. Correlation between initial NDF fraction in the senesced tissue and P loss or
gain after 12 months.

Initial substrate quality parameters of the live tissue characterized in Chapter 2,

were also correlated with the long term changes in P mass in the senesced litter. The

initial C:P ratio in the live tissue was the best parameter to predict P loss or gain after 12

months (R2= 0.790 and p-value < .0001) (Table 4-3). The ability to predict the change in

P using the initial C:P in the live tissue remained consistent throughout each sampling

period. The second best parameter to predict the change in P after 12 months was the









initial %P contained in the live tissue. Figure 4-11 shows the strong correlation between

the change in P mass and the initial C:P in the live tissue of the four dominant species.

Table 4-3. Significance and R2 values of the relationship between the change in P content
and initial live substrate quality characteristics for each sampling period.
Substrate Time (months)
Quality 2 4 8 12
Parameter p-value R2 p-value R2 p-value R2 p-value R2
N:P <.0001 .280 .0001 .143 <.0001 .242 <.0001 .217
C:N <.0001 .437 <.0001 .501 <.0001 .460 <.0001 .465
% N <.0001 .480 <.0001 .524 <.0001 .499 <.0001 .496
% C <.0001 .498 <.0001 .546 <.0001 .503 <.0001 .586
% ADF <.0001 .548 <.0001 .553 <.0001 .544 <.0001 .592
% Residual <.0001 .593 <.0001 .455 <.0001 .557 <.0001 .547
Fiber
% NDF <.0001 .580 <.0001 .675 <.0001 .602 <.0001 .655
% SADF <.0001 .719 <.0001 .741 <.0001 .726 <.0001 .753
% P <.0001 .834 <.0001 .715 <.0001 .810 <.0001 .786
C:P <.0001 .833 <.0001 .726 <.0001 .813 <.0001 .790

The %N remaining in the litter doubled on average and only a slight decrease in N

was seen during the first 2 months in every zone excluding the center. The N content in

the litter increased from 2-8 months in every zone, while a decrease in the %N was

observed from 8-12 months in the upland and transitional zones (Fig. 4-12). There were

not any significant differences seen between hydrological zones over the 12 month study.

The only significant difference between hydrological zones was seen after 2 months

when the center had a significantly higher % N remaining.










0.6

0.5

0.4-

, 0.3-

a 0.2 -

-C 0.1

Q, 0 -+ P. hemitomon
-0.1 : P. notatum

-0.2 K hdropperoides

-0.3 -

-0.4 P loss/gain (g) = 1.23 0.0051 C:P live
R = 0.790
-0.5 16 1 2 2 2 2 2
140 160 180 200 220 240 260 280


300


C:P live



Figure 4-11. Correlation between initial C:P in the live tissue and P loss or gain after 12
months.


E 150


100


50


0


4 6 8 10 12


- -- Center


-*- Edge


- A- Transitional


X Upland


Figure 4-12. Change in litter %N (all species combined) among different wetland
hydrologic zones over time. Values represent mean + 1 standard deviation.







71


The %N remaining in P. notatum and J. effusus was significantly higher than the

%N in P. hemitomon and P. hydropiperiodes over the course of 12 months (Fig. 4-13).

The C:N for each species was roughly the inverse of the %N remaining in the litter, and

after 8 months exposure, changes in C: N in each species appeared to stabilize, and

converged around 15-25 after 12 months. Polygonum hydropiperoides had the lowest

C:N after and 4 months, while P. notatum had the lowest C:N after 8 and 12 months. In

addition, the average %N in the litter doubled after 1 year in the field.


300


250 -


200 -

100 -- -------






50


0
0 2 4 6 8 10 12

*- P. hemitomon --- P. notatum n,-. P. hydropiperoides X J. effusus


Figure 4-13. Change in litter %N among species over time. Values represent mean + 1
standard deviation.

The residual fiber content of the litter after 12 months was significantly higher than

the initial fiber content for every species (Fig. 4-14). After 12 months, P.

hydropiperoides had the highest residual fiber fraction among the 3 species surveyed in

the upland, edge, and transitional hydrologic zones. In addition, P. hydropiperoides had







72


a significantly higher residual fiber fraction compared to J. effusus and P. hemitomon in

the center zone. There were also significant differences in residual fiber content within

each species among the 4 hydrologic zones. Residual fiber content in bags deployed in

the upland and transitional hydrologic zones were generally significantly lower in

residual fiber content than litterbags located in the center and edge hydrologic zone (Fig.

4-15).


50 -

45 -

40 -

35 -

S30 -


n, 20
15 -





P. hemitomon P. notatum P. J. effusus
hydropiperoides

l 0 months E 12 months


Figure 4-14. Comparison of residual fiber content of initial and 12 month exposed litter
among four species tested. Values represent mean (+ 1 standard deviation).










60 -

50 -

S40 -



30





P. hemitomon P. notatum P. J. effusus
hydropiperoides
Species
2 Center 0 Edge E Transitional Zone Upland

Figure 4-15. Residual fiber content of species in each hydrologic zone after 12 months.
Values represent mean (+ 1 standard deviation).

Discussion

An average of 43% of the original litter had decomposed after 12 months

exposure. The rate of decomposition in this study was lower than the 50% loss in litter

mass observed within 4 months of a litterbag study of J. effusus in a freshwater marsh

(Boyd, 1971), but greater than the 23% mass loss in J. effusus observed over 268 days in

an additional freshwater marsh decomposition study (Kuehn et al., 2000). Lower

decomposition rates in our study may be attributed to the initial low N content of the litter

or perhaps a lower degree of wetting and drying compared to Boyd's study. All species

in this study were N limited initially under aerobic conditions for the first 4-8 months,

and higher N content could have possibly enhanced decomposition rates (Qualls, 1984;

Taylor et al., 1989; Corstanje et al., 2006). Most decomposition took place during the

first 4 months of the study. Other studies have reported a leveling off or a lower rate of









decomposition after the first 4 months, which may indicate that most of the labile

components associated with the biomass of the senesced tissue have been lost by this

time (Boyd, 1971; Berg et al., 1998; Villar et al., 2001; Davis et al., 2003).

Regardless of zone (center, edge, transitional, or upland), P. notatum decomposed

faster than P. hemitomon, P. hydropiperoides, and J. effusus, and lost an average of 56%

of the biomass over the 12 month period (Fig. 4-7). Results from initial characterization

of litter quality indicate that P. notatum also had the lowest initial amount of residual

fiber. A low percentage of residual fiber in the biomass of P. notatum could have been

responsible for a higher degree of decomposition compared to other species. Lower litter

recalcitrance can increase the likelihood of microbial colonization since carbon

compounds are more bioavailable and material can be broken down more easily.

Results from Chapter 2 indicate that P. hydropiperoides had the greatest initial

NDF (non-detergent fiber) fraction as well as the most residual fiber compared to the 3

other species of vegetation. Sugars, starches and other components that are easily broken

down make up the NDF fraction. It is likely that the leaves of this species are relatively

labile and contain the majority of the NDF fraction, while the woody stems of the plant

contain most of the residual fiber. The high NDF fraction as well as the significantly

high N and P content of this species could explain why P. hydropiperoides had the

highest degree of decomposition for the first 2 months (Fig. 4-7). After this fraction was

consumed by microbes, the decomposition rate decreased because the remaining biomass

was primarily composed of residual fiber. After 12 months exposure, P. hydropiperoides

had the highest amount of residual fiber compared to the other species in each zone of

litterbag deployment.









Panicum hemitomon had a relatively high percentage of residual fiber initially in

the senesced tissue and the lowest N and P content, which may explain why this species

had the lowest amount of decomposition primarily during the first 2 months. Slow

decomposition of species with a low nutrient content could be supporting evidence that

nutrients control decomposition until more than 30% of the original mass is lost, similar

to findings by Lewis (2005).

Although litter quality appears to have some effect on decomposition rates, a

strong relationship between a live or senesced initial substrate quality parameter and the

% mass loss after 12 months was not seen. When initial substrate quality parameters

were correlated with % mass loss after 2 months, a strong association was still not

observed.

It was surprising that decomposition rate was not more significantly influenced by

hydrological zones of the wetland. It was originally thought that increased oxygen

availability in the upland would promote decomposition since oxygen is the preferred

electron acceptor and yields the most energy during microbial catabolism. Results from

this experiment suggest that species type is what primarily determines short-term P

release, while the location may influence decomposition and P release over a time period

greater than one year.

Presence of cattle is an environmental factor on the research site that may

significantly impact decomposition rate of litter, especially in the center where soil is

primarily organic and more influenced by trampling. Cattle may actually encourage

decomposition in the center by stepping on litterbags, pushing them underneath the soil

surface, and breaking the litter into smaller fragments. After the litterbags had been









deployed in the field for 12 months the bags located in the center were buried up to 30 cm

beneath the soil surface. While litterbags located in the 3 other hydrological zones were

covered with a mat of vegetation, they were not pushed underneath the soil surface by

cattle, indicating that the bags in the center may have had environmental factors which

could influence mass loss that were not an issue in other zones (Fig. 4-16). It is likely

that mechanical fragmentation was not seen in the edge because the soil was much firmer

in this zone compared to the center.

Fiber analysis results from the litter collected after 12 months of exposure indicate

that the residual fiber content after 12 months was greatest in the center compared to

other hydrologic zones. Results support the idea that residual fiber or "lignin" is not

broken down under anaerobic conditions. This may be the strongest evidence for

potentially less decomposition in the center compared to the upland, and may suggest that

greater mass was lost in the center due to mechanical fragmentation of the as a result of

cattle trampling. Lower % mass loss of litter in the wetland edge compared to the

wetland center may indicate that cattle have a greater effect on litter decomposition this

zone (Table 4-1). In the future it may be beneficial to repeat this study with cattle

exclosures in order to quantify the impact that cattle may have on decomposition and P

release in isolated wetlands. Many different confounding physical and environmental

factors present in the field may have been the reason why a clear relationship between

initial substrate quality parameters and the percent mass remaining the litter after 12

months was not seen.






















Figure 4-16. Litterbags collected after 12 months from A) wetland center, which were
approximately 20-30 cm underneath the soil surface and B) wetland edge,
which were on top of the soil surface.

Litter often alternates between releasing and absorbing nutrients as it decomposes

(Jordan et al., 1989). The N content in each species was higher after 12 months

compared to the initial %N contained in the litter at the beginning of the study (Fig. 4-

13). The %N remaining in the tissue declined in some species during the 12 month

period. The decrease in the %N remaining is likely due to microbial

mineralization/immobilization processes. It is possible that a leaching event could have

decreased the %N contained in the tissue during the first 2 months of exposure. It is

thought that N was assimilated by microbes from the water column, or sorbed to the

surface of the litter as particles from manure or urine of cattle in the upland and

transitional zone where standing water was rarely present, causing and increase in %N

remaining. An increase in N, or a decrease in the C/N ratio has been reported in other

decomposition studies (Brinson, 1977; Kuehn et al., 2000; Davis et al., 2003). Villar et

al. (2001) reported that the N content of litter increased 7 times after 2 years in the field

while the N concentration doubled over 1 year for each of the species in this study.









A slight decrease in the %N remaining the edge (0-2 month period) and in the

upland and transitional (0-2 and 8-12 month periods) may suggest leaching or microbial

mineralization during the dry months of the year (Dec. June) instead of N accumulation

relative to C losses seen in the wetland edge and center.

Panicum hemitomon acted as a P sink throughout the course of this study, while

other species alternated between a P source and a P sink. Some species of litter in this

study, P. hemitomon and P. notatum, had a greater amount of %P remaining after 12

months compared to the %P initially in the litter, while P. hydropiperoides and J. effusus

had lower or roughly the same %P remaining in the litter after 12 months (Fig. 4-9). It

was surprising not to see a decline in the %P remaining in the litter in every species due

to mineralization processes; however similar results of an increase in the P mass have

been reported from past decomposition studies (Kuehn et al, 2000; Villar et al., 2001;

Davis et al., 2003).

It is likely that the presence of uncontrolled environmental factors such as cattle,

cycles of wetting and drying, changing moisture content, and various temperatures in

different hydrological zones could have had a significant impact on the %P remaining in

the litter after 12 months. The increase in P content throughout the study was likely

caused by microbial assemblages adhering to the litter or manure particles. It is not likely

that soil particles adhering to the litter would have caused an increase in P. Figure 1-7

reports the mg P/m2 stored in the center, edge, and upland soil which is less than

the5200mg P/m2 stored in P. hemitomon, the species with the lowest P content.

Findings from the laboratory leaching study provided an estimate of short-term P

release (described in Chapter 3) and P release rates were well correlated with initial P









content in the senesced tissue. When P release was evaluated at the field scale and over

longer periods, the initial %NDF contained in the senesced tissue had the strongest

relationship with long term change in P over 12 months. It is possible that the NDF

fraction of the litter has a high P content and as labile sugars and starches are consumed

or leached much of the P associated with that fiber fraction is released, hence the strong

association between these 2 parameters. The relationship between the initial C/P ratio in

the live tissue and the %P remaining after 12 months was just as strong as the relationship

between the initial %NDF in the senesced tissue and %P remaining. This information

could be beneficial in managing seasonally isolated wetlands in the Okeechobee Basin

because possible to predict the %P remaining after 12 months using tissue quality

information from live or senesced vegetation.

Differences in P release among species during this study suggest that species

composition of a wetland may significantly influence P storage capacity in litter and

soils. A marsh dominated by P. hemitomon would have significantly %P remaining in the

litter compared to a marsh dominated by P. hydropiperoides. The large amount of P

released by P. hydropiperoides after senescing could indicate that much of the P in this

species was associated with the NDF fiber fraction and therefore relatively labile

compared to the P contained in P. hemitomon which may be associated with residual

fiber fraction and remain incorporated in the litter for a longer period of time. Because

wetland species often have different hydrologic tolerances, changes in hydrologic regime

of a wetland may result in changes in species dominance within the wetland. The

relationship between species specific P release rates and factors that may result in shifts

in species composition could have significant effect on water quality of the isolated









wetland and potentially influence efforts to address P loading to Lake Okeechobee.

Therefore, the resulting effects of hydrologic restoration on vegetative community shifts

should be considered to determine implications of this management action on P storage

capacity of wetlands.

In summary, relative to the original hypotheses, there was no significant

difference between the decomposition rates observed in the 4 wetland hydrologic zones.

Although the results from this study suggest that wetland centers have relatively the same

rate of decomposition compared to uplands, these results may have been due to many

confounding environmental factors including the presence of cattle which may cause

mechanical fragmentation of the litter. If these factors had not been present it is likely

that wetland centers would have lower amounts of decomposition compared to upland

hydrologic zones. Alternatively, if findings of this study were not confounded by other

factors, it is possible that expected differences in decomposition rate among hydrologic

regimes require longer time periods (greater than one year) to become evident. In

addition, we can conclude that both P. hemitomon and P. hydropiperoides had the lowest

decomposition rates compared to P. notatum and J. effusus. It is believed that high

residual fiber content limited P. hydropiperoides decomposition, while P. hemitomon

decomposition may have been limited by relative high residual fiber content and a low N

and P concentration which possibly decreased the rate of litter breakdown.














CHAPTER 5
SYNTHESIS AND CONCLUSIONS

Four plant species dominant on the site, but located in different hydrological zones

within the wetlands and adjacent pasture were evaluated to determine the rate of P release

during the first and second phases of decomposition. From these results, correlations

were made between a substrate quality characteristic and P release from litter over time

due to decomposition and leaching. Findings suggest a strong relationship between some

initial substrate quality parameters (%NDF, C:P, or P content) and short and long term P

release that can potentially be used as a predictor of P stability and storage under various

restoration techniques and management practices.

Results pertinent to plant tissue characterization (Chapter 2) indicated that live and

senesced tissue of species surveyed contained different nutrient and fiber contents. Live

tissue had a significantly higher N and P content, but had a lower residual fiber content

compared to senesced tissue of the same species. These results may indicate that there

was either a leaching event that took place that removed some N, P, and labile fiber

fractions associated with senesced tissue or some nutrient resorbtion took place before

senesced tissue was collected. In subsequent chapters, it was revealed that nutrient

content and labile fiber of an individual species can significantly influence the amount of

P released within a 12 month period. Information provided by characterizing live and

senesced plant tissue from different vegetative species was important to assess which

tissue parameters are strongly correlated to P retention in an attempt to predict the

amount of P released from dominant species present.









Species type was the major factor that influenced short-term P release (Chapter 3).

Despite the original hypotheses, redox condition and nitrogen enrichment did not seem to

have a significant effect on P leaching over the 17 day period, however there were

significant differences in P leaching among the four dominant species investigated.

Panicum hemitomon, the species that contained the lowest amount of P initially, released

a much lower amount of P compared to P. hydropiperoides, the species that had the

greatest initial P content. These results suggest that a shift in dominant species as a result

of hydrological restoration or other mechanism could significantly affect the amount of P

leached from senesced vegetation and could consequently impact water quality of

wetlands as well as Lake Okeechobee. Below is a hypothetical situation comparing short

and moderate-term P release from four dominant species present in different hydrological

zones on the site (Table 5-1). Panicum hemitomon was the only species that continually

increased the %P remaining in the biomass throughout 12 months. The amount of P

remaining in the other 3 species fluctuated considerably over time and therefore the net

release or assimilation of P for the 3 remaining species would be different depending on

the time period.

In the leaching study, the greatest amount of P was released from each species of

senesced vegetation within 2-72 hrs after treatment water was added to containers, and P

leaching decreased considerably or ended after 72 hours of incubation. Air dried,

senesced vegetation had a high leaching potential upon initial contact with treatment

water. In a field situation, a large quantity of P could potentially be released into the

water column during a heavy rainstorm after senescence or the first time the wetland

becomes inundated after litterfall.









Table 5-1. Variation in short and moderate-term P assimilation or release from lm 2 of
litter over 2 months for the 4 dominant species investigated.
Initially 2 Species
250 g litter / m2 P. hemitomon P. notatum P. hydropiperoides J. effusus
Initial P (g) 5.2 20.7 32.0 20.1

P remaining after
leaching (g) 5.0 9.2 9.1 9.0

Litter mass remaining
after 2 months (g)
215.3 208.1 198.8 206.0
P remaining after 2
months
decomposition (g)
5.7 20.9 13.8 14.1
Net release (-) or
assimilation (+) of P
in litter (g) +0.5 +0.2 18.2 -16.0

There was a strong relationship between initial P content of senesced tissue and the

amount of P released over a 17 day period. These relationships had high R2 values,

indicating that initial tissue P content in senesced tissue could possibly be used as a

predictor of short-term P release from leaching. From this finding it may be possible to

predict short-term P release across a variety of species present on isolated wetlands

within the Okeechobee Basin if the P content of senesced species is known.

Because senesced tissue is often hard to find during the summer months in isolated

wetlands in the Okeechobee Basin, it may be easier to characterize live tissue of different

species. The initial P content of live tissue was not the best substrate quality parameter to

predict P leaching rates (as compared to senesced tissue), however there was a strong

relationship between short term P release and the initial % non-detergent fiber (%NDF)

in the live tissue.

The litter decomposition study (Chapter 4) attempted to predict long-term P release

after litter had been exposed to field conditions for 1 year. A significant relationship

between the %P remaining in the litter after 12 months and the initial NDF fraction in the









senesced tissue was observed. In addition, phosphorus content remaining in the litter

after 12 months was also predictable using the C:P in the live tissue.

Phosphorus content among the four species seemed to fluctuate considerably over

12 months; resulting in an increase in %P remaining in some species while others had a

lower %P remaining compared to their initial condition. It is thought that an increase in

the microbial biomass or manure particles adhering to the litter even after washing may

have caused an increase in P content of litter tissue.

Due to many environmental factors introduced in the field that were not present in

the lab, a strong relationship between an initial substrate quality parameter in either the

live or senesced tissue and the % mass loss of the litter could not be drawn, although it

appeared that litter quality had some effect on decomposition since one of the four

dominant species did have a significantly greater rate of decomposition. It is possible

that cattle present on the research site had a significant influence on the litter quality or

decomposition processes even after 2 months in wetland centers.

In rejection of the original decomposition hypotheses litterbags deployed in the

wetland center did not have the slowest decomposition rate among the four hydrologic

zones. Landscape position did not seem to significantly influence decomposition rates as

over the 12 month study period as originally thought. It is believed that cattle may have a

higher impact than biogeochemical processes, or that the role of hydrology may not be

significant until later phases in the decomposition process. The presence of cattle is

thought to have enhanced decomposition in the center due to mechanical breakdown of

the litter within this zone. There is little previous research on the effect that cattle may

have on ecosystem processes and P release. Future studies may want to consider cattle









exclosures in each hydrologic zone to better compare litter decomposition and nutrient

changes in areas where cattle are present compared to areas where they are not.

Plant species type appeared to be the major factor influencing decomposition and

nutrient release rates in isolated wetlands during the time period investigated. Parameters

such as the NDF fraction, as well as the quantity of P in the litter are unique to various

species and were significant factors determining the rate of decomposition and nutrient

release. Of the four species evaluated, P. notatum had a significantly higher rate of

decomposition compared to the other three species. This species also had the lowest

initial residual fiber fraction.

There were no significant differences identified in decomposition rate of the three

remaining species, however there were significantly higher percent mass losses observed

at certain sampling periods in some species compared to others. Polygonum

hydropiperoides had the lowest decomposition rate compared to other species. This

species had a significantly greater residual fiber content initially which may have limited

decomposition. Panicum hemitomon had a slightly higher mass loss over the course of

12 months compared to P. hydropiperoides. In addition, P. hemitomon had the lowest

initial N and P content compared to the three other species. In regards to the second

hypothesis in Chapter 4, it seems that fiber quality and nutrient content may play a role in

decomposition since species with the lowest initial residual fiber content (P. notatum)

had the fastest decomposition rate, while species with the highest residual fiber content

(P. hydropiperoides) and the lowest initial nutrient content (P. hemitomon) had the lowest

mass losses.









Interactions between plant substrate quality, environmental conditions and effect of

cattle disturbance on wetland P storage are not well understood and this study provides

additional insight into this dynamic topic. Important findings in this study include a

surprisingly high amount of P flux from litter shortly after water column exposure. On

average, approximately 46% of the P contained in the litter was released within the first 2

- 72 hrs. In addition, P. hemitomon released a fraction of the P that was released by the

three other species. Further studies may indicate that this species should be promoted to

optimize wetland efficacy as a BMP while accompanied with hydrological restoration to

increase P storage, however in order to promote P. hemitomon, cattle grazing must be

limited.. Panicum hemitomon may have the ability to reduce P transport out of the

wetland which will subsequently reduce nutrient loading and eutrophication in Lake

Okeechobee compared to other species such as P. hydropiperoides. The results presented

in this study also provide insight on the effect cattle may have in isolated wetlands and P

storage in senesced vegetation, as well as valuable information on improving the water

quality of the Okeechobee Basin and increasing P storage in isolated wetlands with the

use of a particular wetland species.













LIST OF REFERENCES

Anderson, D.L. and E.G. Flaig. 1995. Agricultural best management practices and
surface water improvement and management. Water Sci. Tech. 31, 109-121.

Ankom Technology Corporation. 1998a. Method for determining Acid Detergent Fiber,
Neutral Detergent Fiber and Crude Fiber, using the Ankom Fiber Analyzer.
Ankom Technology Corporation, 14 Turk Hill Park, Fairport New York 14450,
USA.

Benner R., M.A. Moran, R.E. Hodson. 1985. Effects of pH and plant source
lignocellulose biodegradation rates in two wetland ecosystems, the Okefenokee
Swamp and a Georgia Salt Marsh. Limnology and Oceanography 30, 489-499.

Benner, R., S.Y. Newell, A.E. Maccubin, and R.E. Hodson. 1984. Relative contributions
of bacteria and fungi to rates of degradation of lignocellulosic detritus in salt-
marsh sediments. Appl. Environ. Microbiol. 48, 36-40.

Berg, B., Jansson, P-E., Meentenmeyer, V., and Krantz, W. 1998a. Decomposition of
tree root litter in a climatic transect of coniferous forests in northern Europe: A
synthesis. Scand. J. For. Res. 13, 402-412.

Berg, M.P., Kniese, J.P., Zoomer, R., and Verhoef, H.A. 1998. Long-term
decomposition of successive organic strata in a nitrogen saturated Scots Pine
forest soil. For. Ecol. Mgt. 107, 159-172.

Bottcher, A.B., T.K. Tremwel, and K.L. Campbell. 1995. Best management practices for
water quality improvement in the Lake Okeechobee watershed. Ecological
Engineering 5, 341-356.

Boyd, C.E. 1971. The dynamics of dry matter and chemical substances in a Juncus
effusus population. American Midland Naturalist. 86, 28-45.

Brezonik, P.L., and D.R. Engstrom. 1998. Modern and historic accumulation rates of
phosphorus in Lake Okeechobee, Florida. Journal of Paleolimnology 20, 31-46.

Brinson, M.M. 1977. Decomposition and nutrient exchange of litter in an alluvial swamp
forest. Ecology 58, 601-609.

Carpenter, S.R., N.F. Caraco, D.L. Correll, R.W. Howarth, A.N. Sharpley, V.H. Smith.
1998. Nonpoint pollution of surface waters with nitrogen and phosphorus.
Ecological Applications 8, 559-568.