| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,*Soil and Water Science Dep., Univ. of Florida, Gainesville, FL 32611-0510
* Corresponding author (wdebusk{at}ene.com)
Received for publication October 17, 2001.
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: WCA, Water Conservation Area
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Decomposition of organic matter is governed by the chemical composition of the substrate and by environmental factors. Among the more important environmental factors are temperature, moisture, nutrients, and availability of electron acceptors (especially O2) (Swift et al., 1979; Heal et al., 1981; Reddy and D'Angelo, 1994). In wetlands, the presence of floodwater or saturated conditions limits the availability of O2 in the soil profile, therefore decomposition frequently proceeds at a greatly reduced rate with less favorable electron acceptors. Hydroperiod, which encompasses frequency, duration, and depth of flooding, is thus a primary factor governing the degree of organic matter accumulation in wetlands.
Nutrient availability can affect decomposition rate by limiting the growth rate of microbial decomposers. Although nutrient loading is typically greater in wetlands than in uplands due to location within the landscape, nutrient availability may be low relative to the pool of available organic C in wetlands (Reddy and D'Angelo, 1994). Nitrogen (N) and phosphorus (P) both have been identified as microbial growth-limiting nutrients in wetlands (Westermann, 1993).
Over the past several years, widespread attention has been focused on ecological effects of nutrient loading from agricultural lands into the historically P-limited Everglades of southern Florida. Phosphorus enrichment in water and soil has been associated with encroachment of cattail and other rapidly colonizing vegetation into the native sawgrass marsh (Davis, 1991; Steward and Ornes, 1983; Toth, 1987, 1988). Soil P enrichment has been well documented in the Water Conservation Areas, most notably WCA-2A (South Florida Water Management District, 1992; DeBusk et al., 1994, 2001), and has been linked to an increase in peat accretion rate (Craft and Richardson, 1993; Reddy et al., 1993), while also being associated with increased rates of organic carbon turnover (DeBusk and Reddy, 1998) and N cycling (White and Reddy, 1999).
The overall objective of this study was to evaluate the response of soil organic C mineralization (measured as soil respiration) to soil phosphorus enrichment and changes in water level in a nutrient-impacted Everglades marsh. The Everglades encompasses a variety of wetland ecosystems that are historically adapted to low nutrient availability, seasonal dry-down, and periodic droughts. The extensive network of canals constructed across the Everglades, and the Everglades Agricultural Area to the north, has caused significant alteration of both nutrient and hydrologic regimes in the marsh ecosystem. In this study we investigated the interactive effects of varying nutrient enrichment and water levels on the rate of organic C turnover in soil–water microcosms.
|
MATERIALS AND METHODS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
A north–south transect approximately 10 km long was established for soil, water, and vegetation sampling along the gradient of nutrient enrichment in WCA-2A. Sampling sites were located along the transect at seven locations, starting near the S-10C inflow structure at the Hillsboro Canal and ending in the interior marsh region (Fig. 1b). A total of 10 sampling sites were established among the seven locations on the transect, as follows. One site each was situated at 0.75 and 2.2 km from the inflow (Sites T1 and T2), within the highly nutrient-impacted cattail-dominated area. Two sites each were located at 3.1, 4.0, and 5.0 km from the inflow, in the moderately impacted transitional vegetation zone characterized by patches of cattail and sawgrass and mixed stands of cattail and sawgrass. At each transect point, one sampling site was located within a cattail stand (Sites T3, T4, and T5, respectively) and one within a sawgrass stand (Sites C1, C2, and C3, respectively). Sites C4 and C5 were located within the low-nutrient interior marsh (sawgrass), at distances of 6.9 and 10.2 km from the inflow.
Sampling Methodology
Soil cores were taken on 11 July 1994 at each site along the WCA-2A transect. Two cores were taken from each site, approximately 2 m apart. Soil cores were collected intact in 50-cm-long sections of rigid, clear acrylic tubing with an inside diameter of 14.6 cm. Coring was accomplished by pushing the acrylic tubes through the peat by hand while cutting around the perimeter of the core tubes with a serrated knife to sever plant litter and the root mat on the surface. Peat was cored to a depth of 30 cm, including an overlying detritus and periphyton layer of variable thickness. Intact cores were excavated with a shovel, and the bottom openings were plugged with a polypropylene disks (2.54 cm thick) machined to a diameter slightly smaller than the core inside diameter and fitted with dual rubber O-rings for a water-tight seal. Core tubes were capped for transport from the field, such that air headspace was reduced or eliminated, to prevent turbulent mixing and possible disruption of the core stratigraphy.
Soil–Water Microcosms
Intact soil cores (soil–water microcosms) were held upright in aluminum racks, and placed in an ambient-temperature water bath to minimize temperature variation during the study. Water temperature in the microcosms was continuously monitored with thermocouples connected to a data logger (Model CR10; Campbell Scientific, Logan, UT). The soil–water microcosms were housed in an open-air shade house. Partial shading of direct sunlight, to simulate field conditions, was provided by draping shade cloth over the water bath containing the microcosms. Measured levels of photosynthetically active radiation (PAR) beneath the shade cloth were in the neighborhood of 25% of full sunlight at midday, which was similar to PAR measured at ground level in cattail and sawgrass marshes.
Surface water depth was adjusted in each microcosm to 10 cm by siphoning off excess water. During the experimental period, distilled water was added as needed to microcosms to compensate for evaporation of surface water. Microcosms were allowed to stabilize for two weeks before initiation of experiments.
Dissolved Oxygen Profiles
Vertical profiles of dissolved O2 were measured in the microcosms during midday hours with a needle-type dissolved oxygen microelectrode (Model 757; Diamond General, Ann Arbor, MI) mounted on a motorized screw-drive profiling apparatus. The microelectrode was slowly (typically 3 mm min-1) driven downward through the water column and litter layer to the peat surface. In addition, diel measurements of dissolved O2 were taken within the litter layer of microcosms collected from high- and low-impact areas (Sites T1 and C5). Profiles of dissolved O2 were recorded, as described above, and repeated every 6 h for a period of 24 h.
Carbon Dioxide and Methane Flux Measurements
To avoid interferences in the measurement of microcosm CO2 flux related to algal photosynthesis and respiration, a major portion of the periphyton mat was removed from the water column and litter layer of the microcosms for the soil respiration study. The remaining periphyton biomass consisted of residual epiphytic algae on the plant detritus. Autotrophic activity within the residual periphyton was inhibited by placing an opaque cover over the microcosms to exclude sunlight. To compensate for the loss of daytime production of O2 by the periphyton, artificial aeration of the upper region of the water column was provided, as described below.
Flux of CO2 and CH4 from the microcosms was measured according to the following procedure. Core tubes were fitted with polyethylene caps, which were attached with silicone sealant to provide an airtight seal. A gas manifold was used to deliver air from a compressed air cylinder to each microcosm through inlet and outlet ports in the cap. Air was passed through a CO2 trap (2 M NaOH) before delivery to the microcosms. The air stream was bubbled through the microcosm water column at a rate of approximately 30 mL min-1 to about a 5-cm depth via polyethylene tubing connected to the inlet port. Air was allowed to flow through the outlet ports into the atmosphere through a short length of silicone rubber tubing. Aliquots of microcosm headspace gas were obtained by sampling through the wall of the outflow tubing with a 1-mL syringe fitted with a 25-gauge hypodermic needle. Duplicate samples were taken from each of the microcosms during a sampling event. In addition, air flow rate was measured for each microcosm with a Manostat Calcuflow flowmeter (Barnant Company, Barrington, IL). The syringes containing headspace gas samples were immediately inserted into a large rubber stopper to restrict loss of sample through the needle, and taken to the lab for analysis of CO2 and CH4.
Mass of CO2 and CH4 in the headspace gas samples was determined by direct sample injection into a dual-detector gas chromatograph (Model 5840A; HP, Palo Alto, CA). Separate determinations of CO2 and CH4 were made from the duplicate samples taken from each microcosm, as a check of analytical precision. The resulting values were then averaged to obtain a single reading for each microcosm per sampling event. Analysis of CO2 was performed with a thermal conductivity detector (TCD) and Poropak N (Supelco, Bellefonte, PA) column with He as a carrier gas. Oven, injector, and detector temperatures were set to 60, 140, and 200°C, respectively. For CH4 analysis, a flame ionization detector (FID) and Carboxen 1000 (Supelco) column with N2 carrier gas were used. Oven, injector, and detector temperatures were set to 120, 120, and 200°C, respectively, for this analysis. Mass flux of CO2 and CH4 from each microcosm was calculated as follows:
 |
Measurements of CO2 and CH4 flux were repeated for successively lower water levels in the microcosms. Water level was sequentially lowered from the initial depth of 10 cm to a level flush with the peat surface (0-cm depth), then 5, 10, and 15 cm below the peat surface. This was accomplished by draining water through a port near the bottom of each acrylic core tube. Water level was maintained at each depth for five days. Headspace samples were taken from each microcosm on the fifth day, at three separate times during the day.
|
RESULTS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
Dissolved O2 profiles in Microcosms T4, T5, C4, and C5, representing the transitional and low-nutrient regions of the WCA-2A marsh, reflected the reduced size and depth of litter and the presence of a benthic periphyton layer above the fine sediment and peat (Fig. 2). Distinct peaks of dissolved O2 occurred near the surface of the benthic periphyton mat, in the region of maximum photosynthetic activity.
Diel measurements of dissolved O2 in microcosms representing high-nutrient (T1) and low-nutrient (C5) sites revealed fundamental differences in dissolved O2 dynamics within the soil–water profiles at the two sites (Fig. 3)
. The midday (1200 h) dissolved O2 profile for both T1 and C5 were indicative of high primary productivity within the periphyton mats. Between 1200 and 1800 h, O2 concentration in the periphyton zone decreased considerably; however, a redistribution of O2 took place in both T1 and C5 during this time frame, such that the lower litter layer and flocculent sediment became oxygenated in T1, and O2 concentration in the water column of C5 increased. However, high O2 demand in the litter and sediment caused rapid depletion of dissolved O2 between 1800 h and midnight. Overnight (0000–0600 h), dissolved O2 levels in the litter layer and water column had decreased to 
20% saturation in C5 and close to zero (essentially anaerobic) in T1.
|
Summary data for CO2, CH4, and total gaseous C flux from the soil–water microcosms are presented in Table 1. Overall, flux of CO2 due to microbial respiration in the soil and litter varied significantly (analysis of variance [ANOVA], 
= 0.05) with water level in the microcosms, within the experimental range of +10 to -15 cm, relative to the peat surface. For all microcosms, representing a wide range of soil P enrichment, the lowest CO2 flux was recorded at a water level of 0 cm (saturated but not flooded), with significantly increasing flux in response to decreasing water level between 0 and -15 cm ( = 0.05). The CO2 flux measured under flooded conditions (10-cm water depth, artificially aerated) was also significantly higher than for the 0-cm water depth, similar in magnitude to rates measured for both -5 or -10 cm water depths. Thus, the lowest CO2 flux occurred under saturated conditions where the water level was poised at the peat surface. Under these conditions, plant litter that was suspended in the water column under flooded conditions was instead vertically compressed and confined to a thin saturated layer overlying the peat. Rapid vertical depletion of O2 in the compressed litter layer was detected within a few millimeters of the litter surface by measurement of dissolved O2 with a microelectrode (data not shown).
|
= 0.05) under saturated, nonflooded conditions (water depth = 0 cm) than for either drained or flooded conditions. Otherwise, CH4 flux showed no consistent trend with water depth, and there was no significant difference in CH4 flux among the flooded or drained conditions (+10, -5, -10, and -15 cm water levels) imposed on the microcosms. It should also be noted that, for the most part, CH4 flux was approximately one to two orders of magnitude lower than CO2 flux and, consequently, was a relatively minor contributor to total gaseous C flux from the microcosms.
Soil P enrichment also exerted a significant effect on CO2 flux. Overall, flux of CO2 decreased significantly with increasing distance of the sampling site from the WCA-2A inflow (i.e., with decreasing soil P concentration), according to linear regression analysis ( = 0.05). This trend, however, was not significant at water levels of 0 and -5 cm. On the other hand, flux of CH4 varied significantly with increasing distance from the WCA-2A inflow only when the water level was reduced to -15 cm. In this case, unlike the response of CO2 flux, CH4 flux increased with distance from the inflow.
Response of the sum total of CO2 and CH4 flux as a function of water level and distance from inflow was summarized by a response surface of total gaseous C flux (Fig. 4) . Flux of CO2 represented roughly 90 to 99% of the total C flux, thus the trends in CO2 flux discussed earlier also apply to flux of total gaseous C. Most prominent in this summary figure is the significant response of gaseous C flux to water level, including the decrease in C flux between flooded and drained/saturated conditions, and the sharp increase in C flux as water table was lowered from the peat surface (saturated conditions) to 15 cm below the peat surface. Superimposed on this response is the decrease in C flux with increasing distance of the sampling sites from the WCA-2A surface inflow structure, corresponding to decreasing P enrichment of soil, water, and vegetation.
|
|
DISCUSSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Of particular interest in this study was the combined effect of flooding–draining and P enrichment. The interaction of these factors was graphically summarized by the response surface in Fig. 4, of total gaseous C (CO2 + CH4) flux, as a function of distance from the WCA-2A inflow and water level. Interactions among C flux, water level, and P enrichment (as distance from inflow) are evidenced by increased 
flux/water level toward the inflow, and decreased flux/distance with water level. This relationship carries potentially significant implications with respect to nutrient cycling and energy flow in nutrient-impacted wetlands that may also be affected by hydrologic alteration (such as the Everglades).
Differences in biogeochemical characteristics of the soil profile (litter and peat) between cattail and sawgrass stands did not significantly contribute to overall variability in CO2 flux. Although the central tendency of CO2 flux was slightly higher for cattail-dominated sites than for corresponding sawgrass-dominated sites within the transitional zone (i.e., Sites T3, T4, and T5 vs. C3, C4, and C5), the differences were not significant for = 0.05. In contrast, CH4 flux was significantly higher in cattail sites than in corresponding sawgrass sites.
Although relatively small in magnitude, methane flux exhibited greater variability than CO2 flux from the microcosms. Because of the absence of emergent macrophytes in the microcosms, an important conduit for CH4 emissions from wetlands (Schutz et al., 1991), and the low solubility of CH4 in water, ebullition probably was the primary mechanism for efflux of CH4. Given that ebullition is an intermittent process, CH4 measurements taken over short time intervals would presumably yield highly variable results. Another source of uncertainty typically exists in the estimation of CH4 production in flooded soils, namely CH4 oxidation to CO2 at the anaerobic–aerobic interface. Oxidation of methane by methanotrophic bacteria is potentially a major sink for methane in wetlands (Kelly and Chynoweth, 1979; Schipper and Reddy, 1996). Thus, the amount of CH4 collected above the microcosm soil or floodwater surface may substantially underestimate the actual CH4 production in the soil. Happell and Chanton (1993) estimated that an average of 46% of CH4 produced in a northern Florida swamp forest soil was oxidized to CO2 near the sediment surface.
In comparison with C flux measured during the present study, the average CO2 flux from intact (80 cm deep) peat columns from three Canadian wetlands ranged from 1.9 to 11.7 µg C cm-2 h-1 under saturated conditions and 9.0 to 15.8 µg C cm-2 h-1 for a water table depth of 40 cm (Moore and Dalva, 1993). Total C flux from undisturbed cores collected from the Everglades Agricultural Area (Montverde muck, a sawgrass peat) averaged 8.4 µg C cm-2 h-1 for a water table depth of 15 cm (Volk, 1973). Using in situ measurement of CO2 and CH4 flux from northern Florida swamp forests, Happell and Chanton (1993) estimated the average flux of CO2 under flooded conditions to be approximately three to five times less than under drained conditions. After adjusting for estimated root respiration, the average rate of C mineralization under flooded conditions based on combined CO2 and CH4 fluxes was 1.6 µg C cm-2 h-1. In situ measurement of CH4 flux from a WCA-2A slough community, using a static chamber method, yielded an average value of 0.28 µg C cm-2 h-1 (Schipper and Reddy, 1994)
Aeration of the water column presumably accelerated decomposition of the extensive, and relatively labile, plant litter and algal detritus. The potential contribution of the litter layer to total gaseous C efflux in the Everglades marsh is considerable, based on results of related studies in WCA-2A (DeBusk and Reddy, 1998). The CO2 flux from flooded microcosms was most representative of maximum daytime levels under field conditions. Taking into account diurnal variation in dissolved O2 concentration in the water column and litter, the mean daily rate of respiration in the field was probably considerably lower than indicated by microcosm measurements. Therefore, flux measurements made during aeration of the microcosm water column should be considered maximum, or potential, rates sustainable in the field for short periods of time and are useful primarily for comparison among sites along the nutrient gradient and for scaling rate parameters in ecosystem models.
Light availability in the water column under field conditions is dependent on density of the plant canopy (including floating vegetation) and of plant litter within the water column. The oligotrophic sawgrass marsh, as well as the aquatic sloughs, in the interior portions of WCA-2A generally support a lower standing crop of live macrophytes and plant litter (Toth, 1987; Davis, 1991). Light availability in the water column may be relatively high, giving rise to extensive periphyton growth (South Florida Water Management District, 1992). Based on microcosm O2 profiles, it is reasonable to expect that the water column, and portions of the litter layer, might be substantially oxygenated during daytime hours, not only in the low-nutrient sawgrass marsh, but also in the more eutrophic areas of WCA-2A. Belanger et al. (1989) observed that the high-nutrient cattail marsh in WCA-2A was frequently anoxic based on measurements using conventional dissolved O2 probes. However, results of this study suggest that spot measurements of dissolved O2 in the field without serious consideration of spatial or temporal variability could give very misleading information on O2 availability in the water column and litter layer. Systematic field measurements are a prerequisite to characterization of spatial and temporal variability of dissolved O2 in the litter layer. As yet, it is unclear whether or not the O2 supplied by algal photosynthesis is significant in terms of supporting aerobic microbial heterotrophs, including lignin-decomposing fungi.
|
CONCLUSIONS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
flux/water level increased toward the inflow. Overall, water level exerted the greater effect on microbial respiration in the soil and litter layer. Dissolved O2 profiles in Everglades microcosms provided evidence that algal photosynthesis can maintain aerobic conditions in a significant portion of the litter layer during daylight hours. The magnitude of C flux from flooded, partially aerated microcosms demonstrated a substantial potential for C mineralization in the relatively labile litter layer. The use of soil–water microcosms devoid of the native emergent vegetation and its associated microflora and fauna resulted in a much-simplified representation of the Everglades marsh ecosystem, but provided a relatively controlled environment for isolating the effects of water level and P enrichment on soil respiration. Thus, despite obvious limitations imposed by the use of microcosms, results of this study are useful in demonstrating trends in organic C turnover in response to environmental variables, especially those that relate to ecosystem management.
|
ACKNOWLEDGMENTS |
|---|
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
W.F. DeBusk, current address: Ecology & Environment, Inc., 220 W. Garden St., Suite 404, Pensacola, FL 32501. 
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Related articles in JEQ:
This article has been cited by other articles:
|
|
 |
|
  J. R. White, K. R. Reddy, and J. Majer-Newman Hydrologic and Vegetation Effects on Water Column Phosphorus in Wetland Mesocosms Soil Sci. Soc. Am. J., May 23, 2006; 70(4): 1242 - 1251. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Corstanje, S. Grunwald, K. R. Reddy, T. Z. Osborne, and S. Newman Assessment of the spatial distribution of soil properties in a northern everglades marsh. J. Environ. Qual., May 1, 2006; 35(3): 938 - 949. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. R. White and K. R. Reddy Nitrification and Denitrification Rates of Everglades Wetland Soils along a Phosphorus-Impacted Gradient J. Environ. Qual., November 1, 2003; 32(6): 2436 - 2443. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Journal of Natural Resources and Life Sciences Education |
Vadose Zone Journal | ||||
| Soil Science Society of America Journal | Journal of Plant Registrations | The Plant Genome | |||
