The Response of Wetlands Benthic Macroinvertebrates to Short-term Drawdown


Material Information

The Response of Wetlands Benthic Macroinvertebrates to Short-term Drawdown
Physical Description:
Non-Thesis Project
Hayworth, Jennifer
University of Florida
Publication Date:


Subjects / Keywords:
Spatial Coverage:
United States -- Florida -- Collier -- Immokalee -- Corkscrew Swamp Sanctuary
26.39 x -81.62


General Note:
42 Pages

Record Information

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

This item is only available as the following downloads:

Full Text


/" I'The ResponseofWetlandBenthic Macroinvertebrates to Short-term DrawdownJennifer Hayworth DepartmentofEnvironmental Engineering Sciences UniversityofFloridaA Non-Thesis Project Presentedtothe UniversityofFlorida in Partial Fulfillmentofthe Requirements for the DegreeofMasterofScience April 2000


" Abstract:Short-term effectsofwater drawdown on benthicmacroinvertebrate communitiesofSouth Florida cypress systems were investigated. The time period andinsitumechanismofbenthic responsetodrawdown conditions wereofparticular interest duetolackofdata, as well as the increasing stress placedonwetland systems in proximitytohigh groundwater extraction areas. Microcosm experiments were conductedbysubjecting soil cores collected from the field to varying time periods (3,5,and 10 days)ofdrawdown. After each drawdown treatment, cores were sectioned, preserved, and processedtodetermine vertical distributionofmacroinvertebrate populations. Cores were also sub sampled for moisture and organic matter content, bulk density, and porosity. The prominent macroinvertebrate communities, annelids and dipterans, migrated down through the soil column over time reaching depths of20-30cm. Chironomidae and Ceratopogonidae were the dominant dipteran families found. Macroinvertebrate response was evident between the third and fifth dayofdrawdown treatment. Vertical migration seemstobe a viable drought response mechanism for benthos. Benthic migration, however, was not consistently correlated with organic matter content or total percent moisture content.


IntroductionHydrologyisa major environmental determinateofecosystem character, influencing floral and faunal community structure by the frequency and durationofhydrological alterations. Wetland systems, by their very nature, experience dynamic hydrofluxes.Inparticular, cypress swamps and domes are naturally characterized by extreme hydrological fluctuations where severe drawdownofthe system is likely to occur. Hydroperiods that drop the water table to greater than 150embelow the soil surface for several months are not unusual (Duever 1975). The hydroperiodicityofcypress systems that are somewhat hydrologically isolated from the surrounding water table are regulated primarily by the factorsofevaporation, precipitation, and groundwater recharge (Cutright 1974). Ifgroundwater sources are depleted by withdrawal, in conjunction with minimal rainfall, normally unpredictable hydroperiods are further altered with deleterious effects to wetland ecosystems. Intense hydrological change is occurring in South Florida, where exploding population growth and development has lead to increased consumptive water use withdrawals. Active pumpingofgroundwater from municipal well fields is a prominent method for satisfying consumptive needs. Pumping has intensified the hydrological flux. within cypress systems in close proximity to extraction areas through rapid drawdownofthe surrounding water table and underlying aquifer. Long-term mechanical pumpingofcypress swamps in and around Tampa, Florida has led to severe impacts including sinkhole development in susceptible areas and saltwater intrusion into underlying groundwater (Bradbury&Courser 1977, Fretwell 1988, Rochow 1985). Adverse changes in vegetation, wildlife, and soil composition have also been noted with rapid, extended drawdown (Biological Research Associates, Inc. 1988, Reddy&Patrick 1998,


Weller&Voigts 1983). Lowered water tables favor the developmentofwoody versus herbaceous vegetation, decrease wetland species diversity, and encourage invasionofterrestrial and exotic species into wetland habitats, favoring successiontoanupland system over time (Biological Research Associates, Inc. 1988, Taber 1982, Weller & Voigts 1983). In cypress systems, cypress tree growth decreases and mortality increases with prolonged water drawdown (Keeland et al.1997, Taber 1982). Wetland wildlife populations fluctuate with hydrology, particularly those depending solelyonwetland habitats for survival,Le.aquatic invertebrates and fish. With declines in these populations duetodrawdown, higher vertebrates such as waterfowl, muskrats, bears, and alligators are deprivedofvaluable food sources, prompting species migrationorextinction (parker 1960, Weller&Voigts 1983).Inaddition to the biotic changes, prolonged dry conditions alter soil character in wetlands by promotingtheoxidation, shrinkage and compactionofthe usually well-developed organic layer (parker 1960, Reddy & Patrick 1998). Dry wetlands are more susceptible to fire, which is another routeoforganic matter loss and vegetation change (Bradbury & Courser 1977, Parker 1960). Wetland impacts have been documented in ecosystems within three hundred metersofactive wells (Rochow 1985). Current regulations governing water withdrawals from wetland systems in Florida allow for a one foot or 0.30 meter drop in the surficial aquifer at the wetland edge over a ninety day periodofno recharge before adverse impactstowetland hydrology and functioning are expected (F.S. 1996). An attempt to evaluate this regulation scientifically through quantificationofspatial and temporal impacts is needed.


Itiscrucial to detennine the adaptations, both long and short-term,ofaquatic species within wetlands undergoing ever increasing hydrological stress through pumping regimes. Specifically, the responseofbenthic invertebrates to drawdowniscritical, as benthos are fundamental food web components and nutrient cyclers in cypress swamps and other wetlands (Mitsch&Gosselink 1986, Murkin&Kadlec 1986). Macroinvertebrates have several mechanisms for surviving dessication. Species can emigrate from the system as adults by timing their life cycle to avoid seasonal dry events (Batzer&Wissinger 1996, Mackay 1985, Tauber etal.1998). This strategy is only viable under a fairly predictable wet-dry seasonality. Survival strategiesin situinvolve either the productionofdesiccation resistant structuresormaterials,ormigration down through the soil towards moisture (Jackson&Mclachlan 1991, Wiggins etal.1980, Mackay 1985).Insitusurvival mechanisms would be employed under unexpected and prolonged drought conditions, such as would occur with extreme consumptive pumping. Previous studies have dealt with the long-term effectsofnatural droughtonbenthic turnover rates, biomass, and diversity during and after a drought event.Mosthave noted a high faunal turnover rate with drawdown, wet taxa being replaced by terrestrial taxa, and a rapid recoveryoftaxa richness and diversity following rewetting (Bataille&Baldassare 1993, Driver 1977, Jefferies 1994). While most researchoninvertebrates has been conducted in prairie potholes and reservoirs, Leslie etal.(1997) specifically looked at benthic community structure in Florida pondcypress swamps. The study found no significant change in wet taxa representation with drought, suggesting survivalofbenthosin situ.


Two studies have addressed the vertical migration through soil by benthos as a survival strategytodrought in a reservoir. Over a 100-day drawdown period, Paterson&Fernando (1969) found that oligochaete species burrowed downward (>20 cm) to avoid desiccation and winter freezing. Chironomid numbers remained stable at 3 to 6 cm below the surface, indicating that downward migration was not pronounced.Incontrast, Kaster & Jacobi (1978) found that chironomid abundance was altered within21daysofdroughtatdepths from 2to8 cm below surface, and also discovered burrows deeper than20cm for both oligochaetes and chironomids. This study clearly demonstrated benthic migrationtodeeper soil depthstoavoid the effectsofdrawdown. Although both studies looked at benthic migration depth, they samplednofurther than 20 cm becauseofimpenetrable, frozen ground. Despite limited and conflicting data, little research has focusedonthe short-term responseofwetland benthic invertebratestorapid drawdown, particularly in cypress swamps (Bataille&Baldassare 1993, Driver 1977, Jefferies 1994, Leslie etal.1997, Riley&Bookhout 1990). Examinationofboth benthic response timetodrawdown and deep soil migration as a short-term drought escape mechanism is crucialtofully comprehend the temporal impactsofextended drawdown. Determining benthic drought' tolerance is vitaltounderstandinghowbenthos adapt to rapid, severe drawdown conditions initiated by unpredictable natural drought, as well as human-induced hydrologic alteration.Inordertoinvestigate the vertical migrationofbenthic macro invertebrates in responsetowater drawdown, this study employed drought simulation experiments


conductedonsoil cores from a South Florida cypress ecosystem. These experiments assessed both response time and response mechanisms for benthic dessication survival.MethodsStudy Sites The primary study sites were in cypress sloughs located within the Corkscrew Swamp Sanctuary in Collier County. The sanctuary comprises 4,275 hectares, with 300 hectaresofvirgin cypress.The sanctuary is adjacent to agriculture and residential development. The hydrologyofthe sanctuary is quite diverse, but hydroperiodsofthe sampling sites remain fairly constant between 150 and 300 days (Duever 1975). Sites were located 30 to 70 m into the swamp from the fringing marshland areas. Prior to the study, these sites went through prolonged natural drought, but sampling was confined to areas where water level never went below the soil surface. Duringthe low water period (May 1999), samples were collected randomly from an open water slough area surrounded by Taxodium distichum. The macrophyte community included Thalia geniculata, Pontederia cordata, Pistia stratiotes, C/adium jamaicense, and Cephalanthus occidentalis. Sampling during flooded conditions (July and August 1999) occurred within 30-40 mofthe swamp fringe, within the closed canopy area. This area was dominated by Taxodium ascend ens, with an understoryofAcer rubrum, Limnobium spongia, Pontederia cordata, Annona glabra, Sabal minor, and numerous fern species. Samples were also collected once from an inundated cypress dome (FP5) located in the FlintPinStrand Management Area inLeeCounty, approximately thirty minutes from Corkscrew Swamp. The dome was approximately 60 m in diameter with an open canopied water center comprisedoffloating macrophytes and peripherally ringed by


cypress. The original intentofthe project was to sample exclusively from this dome system, but duetounforeseen logistical problems, only one sampling event took place and Corkscrew Swamp became the sampling focus. A3 a resultofdifferences in the two systems, the FP5 sampling event will only be considered for gross comparisonsofbenthic communities between pre and post drought seasons.Sampling MethodsTostudy the responsesofbenthic macroinvertebrates within cypress domes to rapid drawdown, three microcosm experiments were conducted. The microcosms were designed to simulate rapid drawdown conditions within wetland soils by subjecting soil cores to varying regimesofwater loss. For each sampling event, one40_m2area within the cypress canopy was delineated. Cores were randomly collected from the field down to a soil depthof30cmwithin this area basedonwater depth (at surface, but no greater than one meter deep) and organic matter depth (>15 cm). Cores were collected using an aluminum corer(1m length) designed to reduce compression forces. The corer had an inside polycarbonate liner (9.3 cmill)that remained stationary to hold roots, while the outside sleeve cut cleanly through the soil profile with sharp cutting teeth (patent pending). The corer was designed only for organic substrate. Each core was sealedonthe bottom with a rubber stopper, left open on top for ventilation, and shielded from light with aluminum foiltodecrease soil temperature fluctuation. All cores were allowed to dry out while in transport back to the lab, except those designated as controls, which were continuously maintained with water collected from the field.


Inthe field, water column DO, water temperature, andpHwere measured from the middleofthe water column and water depth was measuredateach sampling site. These measurements were not collected in May, as there was insufficient surface water.Inthelaboratory, soil cores were kept upright indoors under stable, ambient temperatures. Triplicate cores were subjected to oneoffour treatments: continuously saturated (control), or a drawdown periodofthree, five or ten days (time zero drought beganondayoffieldcollection). To simulate rapid drawdown, soil water was allowedtopercolate down through the soil layerstodrop the ''water table" within the cores.Atthe endofthe experimental drawdown period, soil cores were extracted using a metal rod tipped with a rubber stopper, and divided into 3 cm increments for the top 9 cm, and into 10cmincrements below. All soil increments were preserved with ethanol, stained with rose bengal, and passed through au.s.Standard No. 30 sieve. Macroinvertebrates were hand picked from the retained fraction, identifiedtothe lowest taxonomic level practical, and total numbers tallied. Representative invertebrate biomass values for dipterans and oligochaetes were based on published values for Florida (Brightman 1975). Soil cores from the cypress dome in Flint Pin Strand were randomly collectedbythe same corer device along a transect line from dome centertoedge, sectioned downto.30cmin the same increments as described above, sieved in the field, and the preserved samples were returned to the lab for analysisofmacroinvertebrates. During core processing, the 3 cm and10cm increment samples were subsampled for moisture content. 2 cm3and 6.5 cm3subsamples from the 3 cm and 10 cm increments, respectively, were placed into pre-weighed bottles. Air and oven dry weight(lOOCfor at least 24 hours) for the sub samples were obtainedtodetermine mass


moisture content. Organic matter was determined through loss on ignition (500Cfor 1 hour)(APHA1992).Dryweight bulk density and porosity also were calculated for the subsamples.Inordertocalculate porosity, particle density was estimated fromtheorganic matter content accordingtothe formula: particle density=2.65 0.02 % organic matter (Bonneau&Souchier 1982).Statistical AnalysisTocompare total macroinvertebrate density, richness, and abundance relativetoboth lengthofdrawdown and depth in the soil profile, data were pooled from each depth increment for the triplicatesofeach treatment. Density was reported as numbers per3cm3interval.Inorder to normalize the intervals for comparison, the10cmincrements were transformedbydividing the numbers by3.3.A square root transformationofthe data was performedtoreduce high variances characteristicoflowsample size. A one wayANOVAtest (SPSS version9.0)compared macroinvertebrate biomass and density for each depth increment within a treatmenttopercent moisture, percent organic matter, porosity, and bulk density. P values less than 0.05 were considered significant.Twopost hoc tests, Duncan and Tamhane, were appliedtodetermine significant relationships. Pearson correlations were run between all parameters and compared at the levelofthefive soil depth intervals. ResultsWater QualityWaterdepthatthe FP5 cypress dome during the single February sampling event was approximately74cmin the center, and11to23cm at the edge. Dissolved oxygen


fluctuated between 0.05to0.65 mgIL along the transect from center to edge, respectively, with water temperatures ranging from 20.0to21.6 C.pHwas between 5.76 and 6.36. Corkscrew Swamp had similarlylow oxygen conditions, averaging 0.71 in July and 0.88 mgIL in August. Water temperatures in July and August ranged from 25.0 to 28.6 C, respectively.pHremained fairly constant for both July and August (6.85 6.91). Water depth in July varied between15to25cm, and 28 to31cm at the August sampling. Water measurements were not taken for theMaysampling period as there was less than 2cmsurface water. Soil Parameters Within Microcosms Percent organic matter decreased significantly with increasing core depth, in both the July and August microcosms (Table1).Organic matter was fairly uniform in the top 9 cmofall soil cores at approximately 80% in the 0-3 cm increment and 40-60% by 7-9cm(Figures 1-3). The July and August cores had sand lens starting at the10cm increment and organic content dropped to less than 30%at21-30cm. Bulk densityofthe soil generally increased with depth, particularly for the July and August cores containing the sand lens at 10 cm (Figures 4-6). Bulk density wasnotstatistically affected by the lengthofdrawdown (Table1).Porosity did not alter over the courseofthe drawdown treatments, but was reduced significantly by the sand lensat10cmfor July and August (Table1,Figures 4-6). Porosity and bulk density were negatively correlated in July and August (p<0.01) (Table 2). Porosity was positively correlated with percent organic matter and moisture for all three setsofexperiments, while bulk density was negatively correlatedwith these parameters only during July and August.


Despite allowing corestodrain freely, percent moisture did not differ statisticalIy relativetocore depth or the temporal extentofdraining during the May experiment (TableI,Figure I). Moisture differed significantly by depth in the July and August cores, where there was a significant decrease at the10cm level probably due to the presenceofthe sand lens (Figures 2-3). Moisture was positively correlated with percent organic matter inalIexperimental periods (Table 2).Benthic MacroinvertebratesMacroinvertebrate density was low for all sampling periods, ranging from73total individuals in the July experiment to 136 in August (Table 4), and decreased from the surface downward (Figures 7-10). Intra-treatment variance was high in alI experiments.Mostmacroinvertebrates were found in the 0-3 cm and 4-6 cm depth intervals.Forthe February and May periods, 0-3 cm abundance was significantly different from alI other depthlevels (p=0.05&p=0.004, respectively).InJuly, the 0-3cm was different from the 4-6cm and 21-30cm levels (p=0.01). The August experiment showed significant differences between 0-3 and 21-30 cm (p=0.01). Only the July experiment had significant differences among drawdown treatments for macroinvertebrate density (Table 3). The five-day drawdown treatment was different fromalIthe other treatments and the control. Although August had the highest total abundance, the February experiment had the highest mean macroinvertebrate density and July had the lowest total abundance and mean density (Table 4, FigureII).


drawdown within threetofive days, whether due to increased mortalityorvertical migration. Although inter-treatment differences were significant only in July, the othertwoexperiments exhibited the same trendofdeclining numbers over the lengthofthe drawdown period. While this couldbea resultofthe core microcosm itself, control cores maintained macroinvertebrates (July&August only) and were not significantly different from the treatments. A precipitous drop in densities istobeexpected with drawdown (Jefferies 1994, Kaster & Jacobi 1978, Paterson & Fernando 1969), and indeed, the February sampling event, which experienced no drawdown, exhibited higher mean densities than the microcosms. Vertical migration within the soil seems to be a viable drought response for macroinvertebrates, as suggested by the current experiments. Macroinvertebrates from the February event were present predominantly in the top soil interval (0-3 cm).Inthe microcosms, most organismswerefound in the upper soil layers (0-6 cm), but they also migrated downward in the cores as demonstrated by the lackofsignificant difference between the 0-3 cm layer and the next three consecutive soil layers (4-20 cm). Very few organisms were found at the 21-30 cm depth, suggesting a possible lower limit for vertical migration. The specific mechanism prompting vertical migration is unclear. Macroinvertebrate densities were correlated with percent organic matter, percent moisture, and porosity. Organic matter content decreased throughout the depth profile in all three experiments, but was only correlated with densities during theMayexperiment. Macroinvertebrates exhibit food source preferences, requiring appropriate ingestible particle size and nutritional quality from organic matter (Gardiner 1972). However,


Meanmacroinvertebrate biomass was highest during February (Figure 12).Itdecreased with soil depth, following the density trend, and displayed a significant depth relationship for theMayand August experiments (Table 3). Diptera and Oligochaeta were the dominant taxa in all experiments. Within Diptera, Chironomidae, Tipulidae, and Ceratopogonidae were the major genera. The Chironomidae were mostly composedofChironomusspp. andPo/ypedi/umspp..InFebruary, July, and August, Chironomidae and Oligochaeta accounted for 85% (February and July) and 82% (August)oftotal macroinvertebrate abundance (Table 4). Tipulidae and Ceratopogonidae dominated inMay(47% and 30%). All were represented throughout the soil depths, with none showing preference in burrow depth (Figures 13 14). All genera decreased at the 4-6cm increments, with the exceptionofoligochaetes in the July microcosm (Figure 14). Dipteran species dominated in February and May, while Oligochaeta dominated July and August microcosms. Total species richness was highest for May, while the February and July experiments exhibited the lowest richness (Figure15). Macroinvertebrate densities were positively correlated with decreasing organic matter content for May (p=0.02) and August (p=0.05), and moisture content (p=0.02) and porosity (p=O.02) for August (Table 2). There was no significant relationship between macroinvertebrate density and soil bulk density (Table 2).DiscussionMacroinvertebrate densities in the study were very low overall, a productofboth the limited numberofcore replicates, as well as the characterofSouth Florida cypress


systems. Duever (1975) found low numbersofchironomid larvae and oligochaetesatCorkscrew Swamp, and sampling in recent years by Corkscrew biologists has yielded similarly low densities. The rangeofannual mean densities for cypress systems varies from 307 numberslm2inanEverglades water conservation area (Rader 1994) to 4229 numberslm2in Florida pondcypress swamps (Leslie etal.1997). The mean densities for all experiments (February August) ranged from 2-4 #/cm3orapproximately 0.02 0.04 #/m3 .Lowdensities made statistical comparisons difficult, but still some trendsofresponses to drawdownwereevident. The macro invertebrate communitiesofthis study, dipterans and oligochaetes, were very indicativeofdrought conditions (Bataille & Baldassare 1993, Kaster & Jacobi 1978, Leslie etal1997,Paterson&Fernando 1969). There were obvious seasonal differences, as the most abundant taxa in the spring months (February and May) were dipterans, while oligochaetes were dominant during summer (July and August). Within the Diptera, Ceratopogonidae and Tipulidae were most abundant in February and May, being replaced by Chironomidae in the latter months. The difference in species richness between May and the other experiments canbeattributed to the onsetofdrought inMayand its abatement in the July and August months. Higher diversity is expected withtheinitial daysofdrought, then decreasing during a prolonged drought period (Jefferies 1994). Many studies have shown that macro invertebrates, especially dipterans and oligochaetes, can survivedrought periods, but none have looked specifically at the response time within the soil. Only Kaster & Jacobi (1978) noted an initial responsetodrawdown onset for benthos (>7 days). Macroinvertebrates in our study respondedto


organismsinthis study appear to be migrating downward independentoforganic matter content. Moisture contentofthe soil is undoubtedly critical to the organisms, as they require water to maintain osmotic balance and to facilitate oxygen adsorption through the integument (Gardiner 1972). Percent moisture decreased with soil depth, but did not change over the courseofthe drawdown time period. Although densities were correlated with moisture change down the core profile in August, the relationship was not apparent in eitherofthe other microcosmsoracross drawdown treatments. Soil cores were analyzed for total moisture content, and no differentiation was made between pore water and water held in the interstitial spacesofthe organic matter. These interstitial spaces may allow the persistenceofhumidity levels sufficient enough to supply adequate moisture to benthos. This subtle difference could influence benthic behavior, specifically migration patterns down a desiccating soil core. The physical structureofthe soil in termsofcompaction and resistance may be important to organism movement. In July and August, porosity decreased drastically down the soil profile with an increase in bulk density and decrease in moisture content. The sand lens present in the July and August cores likely account for this shift. When soil columns lose water, they compact from the surface down within a few days, and porosity would be expected to be higher in the lower versus the upper sediments (EPA 1977). The well-compacted sand lens present in July and August experiments confounds the patternofporosity change. The porosity decrease in August was slightly correlated with decreasing macroinvertebrate numbers, so organisms may be respondingtothe degreeofsoil compaction; however, this was not apparent in the other experiments.


The extentofvertical migration may be a constraintofburrow depth preference by different macroinvertebrate communities. Charbonneau and Hare (1998) found burrow depthtobe very dependent upon species composition, particularlyofChironomidae assemblages. Three speciesofChironomusburrowed to different depths and at different rates. The mean burrow depthoftheseChironomusspp. was approximately 6 cm. A strong seasonal variability was also noted, as larvae migrated much further into the soil with a drop in temperature. Perhaps the lackofprofound vertical migration within the current cores can be attributed to species variability and an under representationoftaxa that migrate deeper. The species present might also need a more drastic change in water content to stimulate the vertical migration impulse. Restrictionsofthe microcosm design should be taken into account when assessing study results. Container size and organism residence times have been shown to affect results in limnological microcosms (Stephensoneta1.1984). The core environment itself could have been a factor by inadvertently altering invertebrate behavior or mortality rates. Unfortunately, the amountofcontrols within the microcosms were inadequate to statistically determine this affect. Processing a replicate core before each drawdown treatment and prior to the adventofthe experiments would have been optimal. Core effects on species assemblages would also have been better served by improvementofthe controls. Large interand intra-variability was exhibited within the cores, which was unavoidable. Macroinvertebrate communities are naturally characterized by patchy distribution and as such, core to core variability is expected to be high (Leslieeta1.1997, Turner and Trexler 1997). Although the amountofcores required to significantly reduce


this variability is impractically large (Streever and Portier 1994), the study would be improved with more sample replicates. Despite the limitationsofthe design, the microcosms were effective research tools as they allowed isolationofphysical factors difficult to achieve in the field, and they have been widely used in studying microfaunal response to aquatic perturbations (Schratzberger&Warwick 1998, Warwick 1993).Ofcourse, the resultsofthese microcosms should be verified by field research. In addition to the above mentioned experimental design flaws, future research on macro invertebrate drought response using microcosms should include several improvements. Further removalofconfounding soil factors is necessary, perhaps by maintaining constant organic matter content and fluctuating the water table only. The additionofknown macroinvertebrate assemblages and numbers to the cores would enable illuminationofspecies soil migration preferences. Extending drawdown periods past ten days is crucial to determine the maximum dessication tolerance limit for benthos. Core oxygen profile data, potentially ascertained from redox potential measurements, are needed as well, as oxygen limitations certainly affect benthic behavior. Acknowledgements I would like to thank the membersofmy committee, Dr. Thomas Crisman, Dr. Joseph Prenger, and Dr. William Wise for their guidance and support. I also thank the South Florida Water Management District for fundingofthis project. The National Audubon Society permitted access to sites within the Corkscrew Swamp Sanctuary.I am grateful to the following for their assistance in the field: Shannon Ludwig, Shanna


Ratnesar, Scott Lynch, Martin Horwitz, and David Skelton. Mark Fowlkes and Joe Prenger helped with taxonomic identification. Mark Clark assisted with experimental design. Cara Stallman and Andrew Muss provided critical commentary. Dr. Lauren Chapman provided statistical advice.


References American Public Health Association (APHA). 1992. Standard Methods for the ExaminationofWater and Wastewater.18thEdition. A.P.H.A., Washington, D.C. Bataille, K.J. and G.A. Baldassarre. 1993. Distribution and abundanceofaquatic macroinvertebrates following drought in three prairie pothole wetlands. Wetlands 13: 260-269. Batzer, D. P. and S.A. Wissinger. 1996. Ecologyofinsect communities in nontidal wetlands. Annu. Rev. Entomol. 41: 75-100. Biological Research Associates, Inc. 1988. Ecological monitoringofthe Morris Bridge wellfield. Annual report submitted to CityofTampa Water Department. Tampa, FL, USA. Bonneau, M. and B. Souchier. 1982. Constituents and PropertiesofSoils. Academic Press, London, England. Bradbury, K.R. and W.D. Courser. 1977. Fourth annual reportofthe St.PetersburgSouth Pasco well field study. Technical Report 1977-4, Southwest Florida Water Management District. Brooksville, FL, USA. Brightman, R.S. 1975. Distributionofbenthic macroinvertebrates in the cypress domes, p.401-429. Annual Report on Cypress Wetlands for Water Management, Recycling and Conservation. Center for Wetlands, UniversityofFlorida, Gainesville, FL. Charbonneau, P. andL.Hare. 1998. Burrowing behavior and biogenic structuresofmud-dwelling insects.J.N.Am.Benthol. Soc. 17(2): 239-249. Courser, W.D. 1972. Investigationsofthe effectofPinellas County Eldridge-Wilde well field's aquifer coneofdepression on cypress head water levels and associated vegetation. Memorandum submitted to Southwest Florida Water Management District, Ft. Myers, FL, USA. 1973. Investigationsofthe effectofPinellas County Eldridge-Wilde well field's aquifer coneofdepressionofcypress pond water levels and associated vegetation. Memorandum submitted to Southwest Florida Water Management


District, Ft. Myers, FL,USACutright, B.L.1974.Hydrogeologyofacypress swamp north-central Alachua County, Florida. Master's Thesis. UniversityofFlorida, Gainesville, FL,USADriver,E.A 1977. Chironomid communities in small prairie ponds: some characteristics and controls. Freshwat. BioI.7:121-133.EPA1977.Lake drawdown as a methodofimproving water quality. Technical report600/3-77-005,OfficeofResearch and Development. Corvallis, OR,USAFlorida Statutes (F.S.).-1996.Section373.042,Florida Statutes. Tallahassee, FL,USAFretwell,lD.1988.Water resources and effectsofground-water development in Pasco County, Florida. Water-resources investigations report87-4188,U.S. Geological Survey. Tallahassee, FL,USAGardiner, M.S.1972.The Biology ofInvertebrates. McGraw-Hill Book Company,NewYork, NY,USAJackson, J.M. andAJ.Mclachlan.1991.Rainpools on peat moorland as island habitats for midge larvae. Hydrobio.209: 59-65.Jefferies,M.1994.Invertebrate communities and turnover in wetland ponds affectedbydrought. Freshwat. BioI.32: 603-612.Kaster, J.L. and G.Z. Jacobi.1978.Benthic macroinvertebratesofa fluctuating reservoir. Freshwat. BioI.8:283-290.Keeland, B.D., W.H. Conner, andRRSharitz.1997.A comparisonofwetland tree growth response to hydrologic regime in Louisiana and South Carolina. For. Ecol. Manag.90: 237-250.Leslie,AJ.,Crisman, T.L., Prenger,J.P., and K.C. Ewel.1997.Benthic macroinvertebratesofsmall Florida pondcypress swamps and the influenceofdry periods. Wetlands17: 447-455.Mackay,Rl1984/85.Survival strategiesofinvertebrates in disturbed aquatic habitats. l Minn. Acad. Sci.50: 28-30.Mitsch, w.J. andlG.Gosselink.1986.Wetlands. Van Nostrand Reinhold Company, New York, NY,USAMurkin, H.R. andlAKadlec.1986.Relationships between waterfowl and macroinvertebrate densities in a northern prairie marsh.J.Wild. Man.50: 212-


217. Parker, G.G. 1960. Groundwater in the central and southern Florida Flood Control District. Proc. Soil Crop Scien .. Soc. Fl. 20: 211-231. Paterson, C.G. and C.H. Fernando. 1969. The effectofwinter drainage on reservoir benthic fauna. Can. J. Zool. 47: 589-595. Rader, R.B. 1994. Macroinvertebratesofthe Northern Everglades: species composition and trophic structure. Florida Scient. 57(1,2): 22-33. Reddy, K.R. and W.H. Patrick. 1998. BiogeochemistryofWetlands. UniversityofFlorida, Gainesville, FL,USARiley, T.Z. andT.ABookhout. 1990. Responseofaquatic macroinvertebrates to early spring drawdown in nodding smartweed marshes. Wetlands10:173-185. Rochow, T.F. 1985. Vegetational monitoring at the Cypress Creek well field in Pasco County, Florida. Technical Report 1985-5, Southwest Florida Water Management District. Brooksville, FL,USASchratzberger,M.andR.M. Warwick. 1998. Effectsofphysical disturbance on nematode communities in sand and mud: a microcosm experiment. Mar. Bio. 130: 643-650. SPSS Inc. 1999. Base 10.0 User's Guide, Version9.SPSS Inc., Chicago, ILL,USAStephenson, G.L.,P.Hamilton, N.K. Kaushik, J.B. Robinson, and K.R. Solomon. 1984. Spatial distributionofplankton in enclosuresofthree sizes. Can.J.Fish.Aq.Sci. 41: 1048-1054. Streever, W.J. and K.M. Portier. 1994. A computer program to assist with sampling design in the comparisonofnatural and constructed wetlands. Wetlands14:199 205. Taber, R.G. 1982. Historic impact on wetlands within the Eldridge-Wilde well field. Memorandum submitted to Southwest Florida Water Management District, Ft. Myers, FL,USAWork order number 238. Tauber,MJ.,Tauber,CA.,Nyrop, J.P., and M.G. Villani. 1998. Moisture, a vital but neglected factor in the seasonal ecologyofinsects: hypotheses and testsofmechanisms. Ent. Soc.Am.27: 523-530. Turner,AM.and J.C. Trexler. 1997. Aquatic invertebrate assemblagesofmarshes: a


characterization and comparisonofsampling methods. Southeast Florida Environmental Research Program andDept.ofBiological Sciences. Florida International University, Miami, FL. Warwick, R.M.1993.Environmental impact studiesonmarine communities: pragmatical considerations. Aust.J.Ecol.18:63-80.Weller, M.W. and D.K. Voigts.1983.Changes in the vegetation and wildlifeuseofasmall prairie wetland following a drought. Proc. Iowa Acad. Sci.90(2): 50-54.Wiggins, G.B., Mackay, R.J., and1M.Smith.1980.Evolutionary and ecological strategiesofanimals in annual temporary pools. Arch. Hydrobiol. Suppl.58:97206.


Table1.SummaryofANOVAtestsfordifferences between drawdown treatment(T)and soil depth (0) for physical soil parameters within each microcosm. Runformicrocosms only.Microcosm May July August Parameter F Value F Value F ValueDTDTDTPercent Organic Matter 1.69 2.902 78.482" 0.424 65.45" 0.413 Percent Moisture 1.086 1.913 55.581" 0.131 64.068" 0.473 Porosity 0.591 1.619 50.4" 0.152 60.31" 0.184 Soil bulk density 0.287 0.874 89.522" 0.239 89.628" 0.221 Significant at p<0.05.


Table 2. SummaryofPearson correlations run between all parameters. Comparisons were made at the level of soil depth interval. Invertebrate numbers were analyzed using square root transformed data. Runformicrocosms only. Pearson coefficients given.Microcosm ParameterMayJuly August DOM M PS8DOM M PS8DOMMPS8Macroinvertebrate Density (D) 0.370* 0.294*0.302* Percent Organic Matter (OM) 0.611* -0.946* 0.946* 0.847* -0.938* Percent Moisture (M) 0.846* 0.902* 0.922* 0.929* -0.955* 0.921* -0.966* Porosity (P) 0.855* -0.964* Soilbulkdensity(S8)-0.964**Significantatp<0.05.


Table3.SummaryofANOVA testsfordifferences between drawdown treatment (T) and soildepth(D)forbiological parameterswithineachsampling event. Invertebrate numbers were analyzed using squareroottransformed data. SamplingEventFebruaryMayJulyAugustParameterF Value FValue F Value F ValueDTDTDTDTMacroinvertebrate Density6.454--4.746" 1.682 3.3222.658" 3.2490.48Biomass5.258--3.015"-1.415-2.178-"Significantatp<0.05.(-)Denotesnostatisticalanalysis.


Table4.Total and relative abundanceofdominant taxaforeach sample period. Total abundancepooledforall cores collectedforeach sampling event. Sampling Event Taxa Total Total Relative Abundance Abundance1%)February Oliaochaeta320.32Amohiooda1 0.01Chiranamidae530.52Ceratooogonidae9 0.09Caleoptera6 0.06Total101May Oliaochaeta2 0.02Amohiooda2 0.02Chironomidae100.10Ceratooogonidae290.30Coliembola6 0.06Coleootera2 0.02Tioulidae460.47Total97July Oligochaeta53 0.73Amohiooda8 0.11Chiranomidae9 0.12Ceratooogonidae2 0.03Coleoptera1 0.01Total73August Oligochaeta720.53Amohiooda17 0.13Chironomidae390.29Ceratooogonidae3 0.02Coleoptera5 0.04Total136


DAY 321--30 113% OM 0% MOISTUREI120 100 +---..,.,.--------------------------------1 % l!t:O-+--"'''''''''=-_...L----,-_.1:0:.:.::.:.:.:.1_--'-_.,....... 0-34-67-910-20Depth Intervals(em)DAY 50-3120 100 +----T------T-------:;:-------------------1 % 4 2 :+_==f .. :l ... ::): .. :r ... :1 .. :1;m:': ... :i .. ... :i .. .. :: .. .... .. 1------II:;:;IiO+---""="----J'---r---l.:.:.:.:.:.:.L--'--,----"'="'--'--,---I:O=:i:L..-.l..--,--.....L:=:::L...-.l..--l 4-67--9 1 0--20 21--30DepthIntervals(em)DAY 10--10-207-9DepthIntervals(em)4-60-3120 .,.-----------------------------------, 100 %EI f----II:I 1----111----1::11/1111110+---"'.:.:.:.:..:.1-_-'--..,..---1.:.:.:.:.:.:.1.._-'----._-'=='__--'-_.,..-......=.:1-_.1...-...,.-..1.=:.:.1-_-'---\ 21--30CONTROL4-60-3120r--------------------------------,100 80% __ l'.....L.t 7-910-2021--30DEPTH INTERVALS (em) Figure1.Mean percent organic matter and moisture for each drawdown treatment and control during the May 1999microcosm e>q>eriment.


DAY321--300-3 113 %OM [] %MOISTUREI100 % __ I_LI-J 4-67-910-20Depth Intervals(em)DAYS21--30 120 .,---T--------------------------------. 100 %+::;:;;:::::a:_--l_,---"'.... 0-34-67-910--20DepthIntervals(em)DAY 100--34-6CONTROL 4-6 0-3,120 100+-----------------,-----------------180 % .......l-__..J:-........l-.-I_=.---J 7-910-2021--30DEPTH INTERVALS (em) Figure2.Mean percent organic matter and moisture for each drawdown treatmentandcontrol during the July 1999microcosmexperiment.


DAY 3 113% OM[] % MOISTUREI21-307-9Depth Intervals (em) 4-6 0-3120-,--------------------------------,100 % _-_T"__:III;:;,';:;,IIII<;l11L----l_...--_..Ii::::;{L_L-r----'.....==II __ Ll--l 10-20DAY 50-3120 100 %E ii'1----11 1-----1:I!I!ll:ll!I!1 I 1-----ITf1 ; r 4-67-910--20 21--30Depth Intervals (em)DAY 1021--3010-207-9Depth Intervals (em)120 -.--------------------------------, 100 -r----'T:-------------------I% __ r---.JI-I 0-34-6CONTROL21--30 4-6 0-3120,-------------------------------,100 %!I 7-910--20DEPTHINTERVALS(em)Figure3.Meanpercentorganic matterandmoisture foreachdrawdowntreatmentandcontrolduringtheAugust 1999microcosm experiment.


DAY 3 11;1 Bulk Density[]PorosityI1.5 -r------------------------------,C') 1+---------------,------+------------l 0.5 -f-mm..----i:::::::r-eO+......"""':.:L.._"'-...,..---'="""-_....L.---,_-==L----L_,-....L:O="'-_J..........,...---"==-_-'---! 0-34-67-9Depth Intervals (em)10-2021-30DAY51.5 C') 1 5r+rE-.,.,..1-:-:I-, -0.5, f:!::: Clmtr1$1[:@t 0............0-34-67-910-2021--30Depth Intervals (em)DAY101.5 C') 1E0-0.5 Cl 00-34--67-910--20 21--30DEPTHINTERVALS(em)CONTROL ()...3 7-9DEPTHINTERVALS(em)21-30Fi9ure4.Meen bulk density and porosiiyforeach drewdown treatment and control during the May 1999 microcosm eJ

DAY 3I III Bulk Density C PorosityI r r:::.r::m 1.5 C'l 1 5r--.... 0.5 Cl 0 =. 0-3 4-6 7-9Depth Intervals(em)10-2021-30DAYS1-__---l-%-I---__---lr+-I--_--f'*'"'"'C-+-,...I---_--1d:I:.:: .. ::.:: .. :I.:: .. :i.:: .. :I.:I .. :1.:: .. ::r--=-",-'1-----l::::U:::: r1.5 C'l 1E -r0 .... 0.5 Cl 0 ""'*'"" 0-34--67-9Depth Intervals(em)10--20 21--30DAY 10 1----_ .. -j-/----.....:JT,-, .. ---l .. 1.5 C'l 1 5.... 0.5 Cl 0 0-34-67-910--20DEPTH INTERVALS21--30CONTROL21-3010-207-9DEPTH INTERVALS0-31.5 -r-----------------------:::::b-----.!0.:-+-1-__ 1'1T11-+-""-1---14-8 Figure5.Meanbulk density and porosity foreachdrawdown treatment and control during the July 1999 microcosm e)(Jleriment. Althoughporosityisscaledthesameasbulkdensity on the graphs, is


DAY 3 )1;1 Bulk Density[]PorosityI 1---=-.-J .. -J ..: ""'--t-----tW:jr+-t-----tI:I--...------l1 1.5 C'l 1 6 -0.5 Cl 0 .,..,.,..,., 0-34-67-9Depth Intervals (em)10-2021-30DAYS21-3010-207-9Depth Intervals (em)4-61.5 -,-------------------------Jb-----,!0.: _-_-_-_-+ -;r+------'""'"1f1[4\ I O+-"""""=L.-......L.---,_""""="-_-'--,----Io.:.:.llo.:.:L.._"----.--"':.:.:.:.:oL---L_,.-""""'='--....L..----j 0-3DAY 1021-307-910-20OEPTH INTERVALS(em)1.5 -,-------------------------------,L: .. = .. -l. r+----1__-.-.. -.:-l.:-. ---1--------..........-,...,---I. 0-34-6CONTROL 21-30 1.5 I0.:-l------------T-r-T-----""'-t-,.,.,.-T-----/ ..:I...:'..:I...:'..:i... :1 .. :1.:=:: ..:1...:1..:1.... :11-----.---1.,.,.., II O+-...IiZ!:!i:i1._.l...-...,..-....L:O:i:o:.:.1._.l...-...,..-....o;.:;:;:;:;;;:L_..L.--...,........,.I;,;,;,;,;,:,L_.l...-...,..---L:o:o:.:.:.:L_...L...-lD-3 7-9 lG-20 OEPTH INTERVALS(em)Figure 6. Mean bulk density and porosity foreachdrawdown treatment and control during the August 1999 microcosm e>q:>eriment. Althoughporosityisscaled thesameasbulk denstty on the graphs, tt is unttless.


Sel1--.-SeI3 -,6.'Se14-..Sel2\\\2\ \ \. \ \\.\\\\ \.\\\ \. \\ \. \\r \. 1L------,r \ loo. ....I """d II.. ....--'I' ......"! __ __ -i __ o .. fif---l'It'''' -I 15 14 13 1211109 ;:;; 8Eu-7 'It --2 -3-sl-_-+.....J SoilDepthIntervals (em) Figure 7. Mean macroinvertebrate density (+-5E)forFebruary sampling. Data pooledforeachsetoftriplicate eores collected.


23222I2019 18 1716IS... 14e"-13 :!. 12..c .!: II Gli! 10 .a! 9 c8 'f "7.. :::ii 6S43 2I,, ,, ,, ,,,, ,, ,,, 3 DayOrawdown-..SOayOrawdown--. 10 DayOrawdown _Control-2 -3-5.1.....JSoilDepthIntervals(em) Figure8.Meanmacroinvertebrate density (+-8E) comparedtosoil depth incrementforMay.


543 M Eu_2-1-2-3 \.'\' I .'\'I,I,, 3Day /....I, Drawdown\ 1. .I, -..5Day I... / 'J Orawdown. "-, \.../ I. --.'-10Day. ,,."Drawdown I ./.. ....,, .I _-Control. ...../. ., ,.. I. .',. ,... .. / .----....., .. ...W /fI:J?l"lj/t>(:ltV'" Soil Depth Intervals (em)Figure9.Mean macro invertebrate density (+-5E) comparedtosoil depth incrementforJuly.


18171615 14 131211109 ;;; E8"'II: -7 IIIC 6.. C 5 ...!4c 3 'e or 2 ::E 10-1-2 -3-'I-5-6 -7 -a'l \ \ \ \ I\ .... "" \.. ..."" ......... ._-.... I..... --"-..-............... --... :':"'-.;--Jl -.-.Ii -r::f"J?r::f..,('"'\SoilDepth Intervals (em) Figure 10.Meanmacroinvertebrate density (+-SE) compared to soil depth incrementforAugust. 3DayDrawdown.. 5DayDrawdown--.'-10DayDrawdown_-Control


1098 ... 7Eu'It 6 ..ccP c5 cP f!.c'cP1:: 4 cP > ce u3 co:E 2 1 0February May July AugustFigure 11. Mean macrolnvertebratedensity (+-SE) compared across all experiments. Pooledfromall coreaforeach experiment.


8-,-----"k---------------------------, 76 I:: 5 iii i:2February'May-.a-July _AugustT.. .; -..o-I-__ ",:-=II=..........""' ........-lI-...........,.c::..i':"Oioolil..;.:...-C" SoilDepthIntervals (em)Figure 12. Mean macroinvertebrate biomass (+oSE) pooledateach depth intervalforeach experimenL


Dominant Taxa: February35 30105o '\ \\__ Ollgochaeta" \__Chlronomldae -..Ceratopogonldae '\\'\\'\\ vr -7-9Soil Depth Intervals(eml 1Cl-20 21-3035305oDominant Taxa: May )(. \\ __Ollgochaeta__Chlronomldae\\ -..Ceratopogonldae '\\---*Tlpulldae... \k..4-6 7-9 1Cl-20 Soil Depth Intervals(eml 21-30Figure 13. Dominant taxa at each soil depth increment for February and May.


Dominant Taxa: July3Or----------------------, 25t------------------,f-\------1 20+---------------+-----\------1 1r-.=--::O::-lIg-oc...,ha-e-:-ta-----,;z: Chironomidae ... 15 +--------------+------+-------j--.Ceratopogonklae J -M-Amphlpoda10 21-3010-207-9 SoilDepthInlerval. (eml Dominant Taxa: August4035305o \\ \\ -+-Oligochaeta \\__Chlronomldae ......Ceratopogonldae\\.-M-Amphipoda ""'-.\..............""'-.\ ... -.",.4-67..Jil 10-205011Depth Inlerval (eml 21-30Figure 14. Dominant taxa at each soil depth increment for July and August.


76 5 eu 4>< =It:: 321o ....//"\ FebruaryMayJuly AugustFigure 15. Mean total species richness for individual experiments.