A Research Work Plan and Scope of Work

Submitted to:

Sherry Mitchell-Bruker
South Florida Natural Resources Center
Everglades National Park
Homestead, FL

By

Daniel L. Childers
Michael Ross
Southeast Environmental Research Center &
Department of Biological Sciences
Florida International University
Miami, FL 33199

And

Lynn Leonard
Department of Earth Sciences
Center for Marine Science
University of North Carolina at Wilmington
Wilmington, NC

November, 2001

In accordance with Cooperative Agreement #1443CA 5280-01-021 between the U.S. Department of the Interior, Everglades National Park (ENP) and Florida International University (FIU), this Work Plan and Scope of Work describes specific activities to be undertaken by FIU and by the University of North Carolina - Wilmington (UNC-W) as a subcontractor, during the two year contract period of this project. This Work Plan represents a large portion of a greater research program that also includes Dr. Helena Solo-Gabriele at the University of Miami, FL (UM) and Dr. Sherry Mitchell-Bruker at the SFNRC-ENP. This Work Plan touches on all aspects of this greater research program, but where tasks are the responsibility of either Dr. Solo-Gabriele or Dr. Mitchell-Bruker, we refer the reader to the UM contract or to Dr. Mitchell-Bruker for details. Close contact will be maintained with Sherry Mitchell-Bruker, Co-Investigator and Project Manager at the South Florida Natural Resources Center (SFNRC-ENP) throughout this time.

Introduction and Background

The Everglades wetland landscape is as distinctive as it is heterogeneous. This landscape is characterized by teardrop-shaped tree islands in a wetland mosaic of strand-shaped sawgrass marsh and deeper slough systems. Increasingly, evidence shows that this landscape structure is a result of the once historic, regular flow of water downslope from Lake Okeechobee to the Gulf of Mexico. The last 100 years of water management in this system has compartmentalized the landscape and impounded water, though. In most of the remaining Everglades, flow is either greatly reduced and sporadic or is non-existent. Analyses of historical data support the hypothesis that the Everglades wetland landscape was not only structured by water flow but that this structure was also historically maintained by regular water flow. Furthermore, McVoy and Crisfield (2001) hypothesized that this flow-maintained landscape was actually a dysequilibrium condition. They argued that deep sloughs would naturally tend to fill in, probably with organic sediments, while the elevations of the higher sawgrass strands would be mediated by mean hydroperiod. Flow thus kept the sloughs scoured, maintaining pathways for downstream movement of large volumes of water in a non-equilibral way (McVoy and Crisfield, 2001). The loss of both the "corrugated" ridge and slough topographic features and of many tree islands in much of the remaining Everglades point to the validity of this hypothesis. In this research plan, we propose to quantify flow regimes around tree islands and in open marsh in Shark River Slough, ENP. We also propose to experimentally manipulate flow rates in SRS marshes in order to quantify the importance of water flow to inorganic sediment and organic matter transport through this landscape. Both aspects of our research will help parameterize hydrodynamic and ecological models that will be used to test the implications of different restoration options to maintaining and enhancing water flow in the Everglades landscape.
Flows across vegetated marsh surfaces are inherently complex. Flow energies may be dissipated by friction with the marsh surface and by wake formation around closely spaced plant stems (Nepf et al. 1997). The resulting fine scale flow dynamics and turbulent exchanges are known to affect: 1) oxygen and nutrient exchanges at the soil-water interface (Escartin and Aubrey, 1995); 2) microclimate regulation between plant stems (Dade, 1993); 3) the nutritional environment for suspension-feeding animals (Irlandi and Peterson, 1991), larval recruitment (Eckman, 1983), and; 4) sediment deposition and retention processes (Leonard and Luther, 1995). Hydrodynamic exchanges may also influence: 1) substrate selection by invertebrate larvae (Butman, 1987); 2) the distribution of meiofauna (Palmer, 1986); 3) the export of primary (detritus) and secondary production from the marsh surface to adjacent flowing-water systems (Peterson and Turner, 1994), and; 4) the trophic relationship between protozoan suspension feeders and their prey (Shimeta et al., 1995). Although flow regime may impact many different wetland processes, little field data exist which describe detailed flow behavior through the emergent vegetated ridge and slough system of the Everglades.
Previous (e.g. Leonard et al., 1995, Leonard 1997) and on-going (Leonard et al., in review) work in other marsh systems has shown that fine scale flow structure controls suspended particle distribution over both vertical and horizontal scales. Flow patterns resulting from the synergistic interactions of channel morphology, marsh microtopography, and length of inundation exert significant control over the spatial distributions of suspended and deposited particulate matter in many vegetated systems (Leonard, 1997, Christiansen et al., 2000). Similar interactions may be responsible for shaping and maintaining the geomorphic framework of the tree island/ridge and slough system in the Everglades.
Another complexity in understanding particulate dynamics in vegetated wetland systems arises from the fact that the vertical structure of marsh flows may be strongly related to canopy architecture. Flow profiles collected in Spartina alterniflora and Juncus roemerianus canopies in Gulf of Mexico marshes and Atriplex portulacoides marshes in the North Sea indicate that vertical flow profiles are not uniform and that they deviate from the logarithmic shaped profile encountered in areas lacking vegetation. Leonard and Luther (1995) demonstrated that profile shape was affected by both inter-profile and intra-profile biomass distribution. Intra-profile baffling potential was maximum at heights where plant material was most abundant, and inter-profile baffling potential was greatest at maximum plant densities. Even for very low velocities (<5 cm sec-1)--such as those expected for the ridge and slough system-plant stem densities over about 1.25 cm may cause transitional to fully developed turbulence to occur (based on measured Reynolds Numbers). In many cases, flow conditions in wetland canopies are neither fully turbulent nor fully laminar because of plant/flow interactions, (Kadlec, 1990; Leonard and Luther, 1995). Consequently, microturbulence and flow energy in a vegetated canopy may be most strongly affected by basal stem diameter and plant canopy architecture. This is of particular interest to flow through Everglades marshes because most sloughs that were once either unvegetated or sparsely vegetated (in the pre-drainage system) are now vegetated by Eleocharis and Panicum. This change in both plant density and morphology may thus exert strong control on maximum possible flow rates as well as whether those flows are laminar or turbulent. And both flow rate and flow character directly control rates of both sediment entrainment and transport..
Tree islands are as fundamental to the Everglades landscape, history, and ecosystem function as the broad marshes that run among them. Restoration of the Everglades will not be successful without protecting their place and character. Currently, however, our knowledge of tree islands is insufficient to predict how they will respond to changing water management and restoration. We have ideas about how tree islands form, but don't know how they develop over time from these beginnings, and what conditions will enable them to persist. We don't know where they need more water, where they need less. From the uniformity of their orientation and shape from place to place in the Everglades, we know that surface water flow plays a vital role, but at present we lack functional understanding of how water flow controls the sedimentation processes that build or maintain tree islands.
The tree island profiled in Figures 1 and 2 is one of thousands in ENP. The teardrop-shaped landform narrows and extends to the south-southwest, as do most of its neighbors. The island has formed above and downstream of a large limestone outcropping. In the tree island's "head", i.e., above the outcrop proper, tall trees common to dry or mesic tropical uplands are rooted in several feet of mineral soil. In its "tail", temperate trees of shorter stature grow under much more hydric conditions. The organic soils of the tail are elevated by several dm above the adjacent marshes, but there is little evidence of bedrock control. One scenario suggested by this profile is that differential sedimentation in the lee of the tree island head has, over many centuries, raised the surface of the tail, allowing woody plants to survive, albeit in smaller form. Other alternatives are possible. For instance, several mechanisms may cause the initial forest in the tree island head to concentrate nutrients, which may be transported downstream to create a tail of relatively high productivity in comparison to adjacent wetlands. Either scenario implicates water flow and material transport processes as fundamental to landscape character. In this proposal, we propose to initiate a modeling effort, supported by data from natural and experimentally manipulated landscapes, that will allow us to simulate these process under different flow regimes.
In May 1999, a large group of research scientists from universities, federal agencies (including ENP, through the active involvement of S. Mitchell-Bruker), state agencies (including SFWMD), and non-governmental organizations (including National Audubon Society) were awarded funding by the National Science Foundation to establish a Long Term Ecological Research (LTER) site in Everglades National Park. For detailed information about the NSF LTER Network, see www.lternet.edu. This new LTER is known as the Florida Coastal Everglades LTER Program (FCE LTER; for more information on the FCE program, see http://fcelter.fiu.edu). Some of the research being conducted by the FCE LTER program is basic science. However, much of the LTER research is quite similar to the work we are proposing here. Water flow is central to the focus of FCE LTER research, particularly as it relates to the oligohaline regions of ENP estuaries. Because of this overlap, we propose that this CESI proposal be viewed as a valuable partnership between the FCE LTER Program and ENP.

General Goals and Objectives

We will quantify the flow regime within and immediately adjacent to tree islands. Additionally, we will measure water flow rates in open marshes between tree islands, and both upstream and downstream of tree islands, focusing on how the islands themselves modify flow regimes in the landscape. We hypothesize a positive feedback mechanism whereby tree islands accelerate local water flow by channeling water around their higher elevations (the Bernoulli Effect), which leads to maximal flows along the upstream island flanks that scour sediments here but deposit them immediately downstream of the islands (in the "tail" region). In this way, tree islands actually maintain their structure, and those not "anchored" in place by geologic features may actually migrate downstream very gradually. More tree islands in a landscape should thus relate to higher peak flow rates for a given mass flux of water and thus a greater potential for maintenance of the islands in the mosaic.
Water flow is important to the open marsh as well. In addition to the well-substantiated hypothesis that flow is critical to maintenance of the [sawgrass] ridge and [deeper water] slough topography of ENP marshes, we suggest that flow is also the critical force responsible for particle transport. Everglades wetlands contain very little inorganic sediment, and most particles found in these marshes are organic in nature. Furthermore, Everglades wetlands are characterized by very clear water. The bulk of particulate organic matter in these marshes takes the form of flocculent detritus ("floc") found in an unconsolidated layer just above the soil surface. We hypothesize that 1) water flow is critical to moving this floc as bedload through Everglades marshes; 2) the minimum flow necessary to mobilize and move floc controls the importance of that floc to local ecological processes, including soil dynamics and plant productivity, and; 3) the downstream transport of floc is the most important organic matter and nutrient input to Everglades estuaries by the upstream freshwater landscape.
In a separate project directed by S. Mitchell-Bruker, a model of deposition, scouring, and transport of sediment through vegetated shallow waters typical of the Florida Everglades is already being developed. The simulator considers horizontal flow of water assuming that water is essentially hydrostatic in the vertical. A vertically averaged diffusion equation is used for simultaneous overland and porous-medium flow. The simulator can be run in either steady, quasi-steady, or fully transient modes. Sediment is characterized by an arbitrary number of discrete particle sizes, each with a characteristic particle density and diameter. Each sediment class is considered independently. Transport of a sediment class is performed using the advection-diffusion equation, with settling and scour providing sinks and sources for each cell. Settling rates are based on the fall velocity of a spherical particle, while several formulations are provided to model scour. The initial distribution of suspended sediment within the column will be based on field measurements. Thus, a final objective of the work proposed here is to provide flow measurements and suspended sediment characterization to support this modeling and to run model simulations.
We have structured both our research and this Work Plan around research questions. In the case of each question, we discuss our rationale for this question and briefly present a methodology or experimental design for answering the question. Before presenting our specific research questions, however, we discuss our general field sampling approach (below). For each question, the investigators responsible for this research objective will be noted. Where Dr. Helena Solo-Gabriele is the investigator responsible for a task, we refer the reader to the UM contract or to other documents. The UM contract supports the participation of Dr. Solo-Gabriele and a graduate student. Supplies and equipment for all aspects of the study, including the UM portion, are included in this contract.

General Field Sampling Approach

Baseline data collection will focus on the wetland mosaic surrounding three major Shark Slough tree islands: Black Hammock, Gumbo Limbo Hammock, and an unnamed island referred to as Satin Leaf Hammock (Figures 3 and 4). All three are subjects of a current study of forest structure and function ("Everglades National Park tree islands: interaction of hydrology, vegetation, and soils": Cooperative Agreement CA 5280-00-017). In that study, M. Ross (with collaborators Jayachandran and Oberbauer) has established permanent plots in which plant and soil processes are monitored in hardwood hammock, bayhead, and bayhead swamp environments, in relation to background variation in water level (from nearby longterm recorders), nutrient availability, and other physical factors. The study also includes a paleoecological component which attempts to reconstruct the history of the different tree island sub-environments, and the development of the landform as a whole. A missing element in the ongoing research is the integration of the study tree island into the surrounding landscape via sheetflow pathways and sediment transport processes.
We addressed a number of questions of scale when setting up our sampling scheme. A key objective of this research is that our empirical approaches be designed to enhance our modeling efforts. Our experimental/sampling design centers around transects around the 3 tree islands discussed above: One transect through the marsh upstream of each tree island (600 - 900m), one transect along the island periphery, and one lateral transect from either side of the island (200 - 300m each). To the extent possible, we will link the lateral transects to 1 of M. Ross' 3 existing island transects. The spot measurements of flow discussed below will be made along transects from tree islands into the marsh. Flow sampling of the lateral transects will follow an exponential design, with the shortest intervals between sampling points at community boundaries (i.e. the tree island-marsh boundary or the slough-sawgrass boundary). We will establish initial transect locations and dimensions using field reconnaissance combined with aerial photography. After the transects have been established, we will conduct a flow regime reconnaissance along each transect. From this information, we will determine the location of the routine flow monitoring stations (discussed below) on each transect.
With all field work, we will employ the minimum tool approach. In early phases of the project we will experiment with the feasibility of accessing sites by airboat, kayak, boardwalks, and/or by foot to ensure minimum impact. During data collection, sampling will always begin at the most downstream sites and progress upstream. Elevation benchmarks will be established using differential GPS or optical surveying techniques. Installed equipment, wells, and sampling sites will be surveyed and tied to the benchmark. Removable PVC poles will be used to mark sampling sites. In all cases, we will minimize short and long-term impacts on the system we are studying.
The hypotheses described in the previous section will be tested with long-term field manipulative experiments in which we actually modify the flow regime around relatively large areas of marsh, both reducing/eliminating flow and increasing flow. At each experimental site, we will have marsh sampling locations both within the manipulated area and outside of it (as controls). We will quantify water flow and calculate water flux at both places. We will quantify floc transport, floc production and accretion, floc nutrient content, and floc organic content. We will calculate floc transport, as bulk flux, for each experimental marsh and relate these flux values to the flow regime. The key to this type of experimental manipulation is the physical modification of flow regimes. We will accomplish this using 75 cm high walls of 12 mil plastic sheeting established as "flow fences" in the marsh. In the no flow treatments, we will construct walls that block and divert flow from the downstream experimental marsh. In the flow enhancement treatments we will construct walls that concentrate and direct accelerated flow through the downstream experimental marsh (the Venturi Effect). We have a great deal of experience with these kinds of flow manipulations. For example, a graduate student working under D. Childers has used this method to experimentally reduce water flow around experimental tree islands in Southern Everglades marshes south of the C-111 canal for 2 years now. These walls are inexpensive, require minimal maintenance for long-term deployment, and effectively reduce flows by 70% or more (see Figure 5). We are thus confident of the efficacy of this simple yet elegant design.
The model simulations will be based on data from water samples collected at the tree island and marsh sites, and from characterizations of suspended sediments in the water column. Particle density and diameter distribution will be determined through sedimentation tests, sieving and filtering techniques. In addition, Doppler flow meters will be used to further characterized the particle density distribution in the water column, under various flow regimes. The results of these suspended sediment analyses will be used to provide input to the sediment transport model. In addition, measured flow rates will be used to calibrate the model. Finally, we will model a range of scenarios representing typical flow regimes in the Everglades, examining model results to ascertain the role of flow and sediment transport in the formation and maintenance of tree islands and the ridge and slough topography.

Specific Research Questions and Methodologies

Primary Research Question 1: What is the role of ecological pattern and process in controlling flow in the Ridge and Slough Landscape?

Question 1.1: How does plant community structure and/or primary productivity control flow regimes? We will quantify vegetative resistance measuring plant biomass and biovolume in both slough and sawgrass marsh communities. These measurements will be made in 1m X 1m plots in both slough and sawgrass marsh using nondestructive allometric models that already exist (Daoust and Childers, 1999; L.Leonard and D.Childers). We will also use intensive flow measurements to identify transition zones in flow between the slough and sawgrass marsh communities. We expect these physical/hydrologic interfaces to be points of minimum sediment/floc transport and maximum deposition in the sloughs (L.Leonard and H.Solo-Gabriele, see UM contract). Wind is likely to play an important role in this interaction between vegetation and water flow, particularly in broad, relatively open sloughs. We will obtain local wind conditions with a hand-held anemometer when flow data are collected (H.Solo-Gabriele). Finally, we will explore the interactions between vegetation and water flow in more detail using ecosystem manipulations in which we modify the properties of the plant canopy through clipping to reduce stem density. If possible, we will carry out these experiments in conjunction with the flow enhancement manipulation discussed above. In addition to carefully measuring vertical and horizontal profiles in water flow [in these canopy manipulations], we will also measure floc transport, short-term materials deposition, and long-term changes in soil elevation (L.Leonard and D. Childers).

Question 1.2: How does plant community structure control materials deposition? We will quantify the relationship between plant community structure and materials deposition in the marsh by quantifying surficial sediment accumulation in the vegetation plots (see Question 1.1; L.Leonard and D.Childers). In tree islands, we will examine this relationship by quantifying primary production as well as sediment accumulation, using asediment trap or marker horizon method (M.Ross).

Question 1.3: How does plant community structure control topography and soil elevation dynamics? Because changes in soil elevation are very slow, we view this as primarily a long-term question. In the marsh, we will quantify long-term changes in net soil elevation by measuring depth to bedrock twice each year (wet season and dry season) in the vegetation monitoring plots. We will also quantify long-term changes in net soil elevation using the stainless steel pin method, and we will measure surficial materials accumulation using a marker horizon approach (D.Childers). In the tree islands, we will also measure change in net soil elevation (pin method) and surficial materials accumulation data (marker horizon method; M.Ross).

Question 1.4: What are sources of sediments and materials in different Ridge and Slough communities? Questions of sources of organic matter are central to the research being conducted in the FCE LTER Program. As part of this research, Dr. Rudolf Jaffé (FIU) has demonstrated the efficacy of several sophisticated organic geochemical techniques to differentiate broad patterns of sources of estuarine organic matter. We will work with Dr. Jaffé to use one of these techniques--molecular markers (biomarkers)-to characterize the organic geochemical signatures of materials produced by saw grass marsh, slough marsh, and tree islands. We will examine both soil cores and water samples. If we are able to differentiate organic matter sources among these 3 closely located communities, we will apply these techniques to floc and sediment samples (M.Ross and D.Childers). M. Ross will be examining materials deposited on tree island soil surfaces as part of his ongoing study (working with Pete Stone, SCDEP), with the same goal in mind (M.Ross). We will assess the physical characteristics of the suspended materials using standard qualitative or gross particles composition assessment techniques (L.Leonard and H.Solo-Gabriele, see UM contract). Finally, we will attempt to address the importance of periphyton mat movement (particularly during large wind events) by physically tagging a minimum of 3 periphyton mats per study slough and tracking their positions in time and space (D.Childers, H.Solo-Gabriele, and S.Mitchell-Bruker).

Primary Research Question 2: What is the role of intra-annual and interannual variability in the hydrologic regime on the following processes/features?

Question 2.1: On short-term (years) changes in soil elevation & topography? We will answer this question with the soil elevation change and sedimentation data being collected in the marsh (D.Childers) and tree island (M.Ross) communities. Additionally, if a dry-down occurs, we will quantify soil microtopography within the vegetation plots (L.Leonard).

Question 2.2: On rates of material transport and flow? We will address this question through our measurements of temporal changes in flow and material load (including particle characteristics and nutrient concentration of suspended materials) over study period (H.Solo-Gabriele, see UM contract). We will also monitor surface water and ground water at 4 locations per tree island site. We will use these data to identify regional hydrologic gradients, and they will be incorporated into the modeling efforts (M.Ross, H.Solo-Gabriele, and S.Mitchell-Bruker).

Primary Research Question 3: What is the role of the flow in maintaining a Ridge and Slough Landscape?

Question 3.1: What is role of material transport in tree island development and maintenance? We will use our measurements of change in net soil elevation (pin method) and surficial sediment accumulation (trap) to address the maintenance aspect of this. We will use compositional and textural analysis of the surficial soils to ascertain the recent depositional/erosional environment, particularly downstream of tree islands in tails (L.Leonard and M.Ross). The quality of materials being transported is also important to this question, so we will determine the characteristics of particulates being transported via bedload at the tree island-marsh interface (L.Leonard and H.Solo-Gabriele, see UM contract). We also hypothesize that major wind events may be an important mechanism of materials transport when they force "rafts" of periphyton mat onto tree islands. This episodic deposition of periphyton wrack may be an important source of material to build tree island soil elevation. We will address this phenomenon through a combination of the long-term soil elevation measurements and our periphyton tagging and tracking experiments (D.Childers, H.Solo-Gabriele, and S.Mitchell-Bruker).

Question 3.2: What is role of material transport in marsh [sawgrass ridge and slough] development and maintenance? As with Question 3.1, we will use compositional and textural analysis of the surficial soils at sawgrass-slough interfaces, periphyton mat tagging, and determination of the characteristics of particulates being transported via bedload at the slough-sawgrass interface to answer this question (same responsibilities as Question 3.1). Additionally, our flow manipulation experiments will allow us to directly address this question as we use ecosystem manipulations to increase and decrease ambient flow rates (D.Childers and L.Leonard).

Question 3.3: What is the role of flow in mediating energy and nutrient exchange among Ridge and Slough communities? Flow and hydrology will affect energy and nutrient exchanges both above and below the soil surface. We will quantify temporal change in nutrient concentrations in ground water on an intermittent basis along our sampling transects (S.Mitchell-Bruker). We will attempt to integrate surface and groundwater measurements by quantifying nutrient concentrations in pore waters at the tree island-marsh and the sawgrass-slough interfaces (M.Ross). We will also periodically sample surface water at 4 transect perimeter stations to quantify any nutrient gradients that may be a result of island-flow interactions (D.Childers). Finally, our flow manipulation experiments will allow us to address this question on a local scale as we use ecosystem manipulations to increase and decrease ambient flow rates (D.Childers and L.Leonard).

References Cited

Butman, CA 1987. Larval settlement of soft-sediment invertebrates: the spatial scales of pattern explained by active habitat selection and the emerging role of hydrodynamic processes. Oceanography and Marine Biology Annual Review 25: 113-165.

Christiansen, T., P.L. Wiberg, and T.G. Milligan. 2000. Flow and sediment transport on a tidal salt marsh surface. Estuarine, Coastal and Shelf Science 50: 315-331.

Dade WB 1993. Near-bed turbulence and hydrodynamic control of diffusional mass transfer at the sea floor. Limnology and Oceanography 38:52-69.

Daoust, R. and D.L.Childers, 1998. Quantifying aboveground biomass and estimating productivity in nine Everglades wetland macrophytes using a non-destructive allometric approach. Aquatic Botany. 62:115-133.

Ekman, JE 1983. Hydrodynamic processes affecting benthic recruitment. Limnology and Oceanography 28:241-257.

Escartin, J and DG Aubrey 1995. Flow structure and dispersion within algal mats. Estuarine Coastal and Shelf Science 40:451-472.

Irlandi, EA and CH Peterson 1991. Modification of animal habitat by large plants. Oecologia 87:307-318.

Kadlec, RH 1990. Overland flow in wetlands: vegetation resistance. Journal of Hydraulic Engineering 116: 691-706.

Leonard, LA, Beavers R.L, and PA Wren. (in review). Flow Dynamics and Sedimentation in Spartina alterniflora and Phragmites australis Marshes of the Chesapeake Bay. Submitted to Wetlands.

Leonard, LA 1997. Controls of sediment transport and deposition in an incised mainland marsh basin, southeastern North Carolina. Wetlands, 17(2): 263-275.

Leonard, LA & ME Luther 1995. Flow hydrodynamics in tidal marsh canopies. Limnology and Oceanography 40:1474-1484.

Leonard, LA, AC Hine, and ME Luther 1995. Suspended sediment transport and deposition in a J. roemerianus marsh, west-central Florida. Journal of Coastal Research 11(2): 322-336.

McVoy, C. and E. Crisfield, 2001. The role of water and sediment flows in the ridge and slough landscape. SFWMD White Paper.

Nepf, H.M., J.A. Sullivan, and R.A. Zavistoski. 1997. A model for diffusion within emergent vegetation. Limnology and Oceanography 42: 1735-1745.

Palmer, MA 1986. Hydrodynamics and structure: Interactive effects on meiofauna dispersal. Journal of Experimental Marine Biology and Ecology 104:53-68.

Peterson, GW and RE Turner 1994. The value of salt marsh edge vs. interior as a habitat for fish and decapod crustaceans in a Louisiana tidal marsh. Estuaries 17: 235-262.

Reed, DJ 1992. Effect of weirs on sediment deposition in Louisiana coastal marshes. Environmental Management 16(1): 55-65.

Shimeta, J, PA Jumars, and EJ Lessard 1995. Influences of turbulence on suspension feeding by planktonic protozoa; experiments in laminar shear fields. Limnology and Oceanography 40: 845-589.