Recipient Organization
WEST VIRGINIA UNIVERSITY
886 CHESTNUT RIDGE RD RM 202
MORGANTOWN,WV 26505-2742
Performing Department
Forestry
Non Technical Summary
The Clean Water Act regulates pollution discharge into waters of the US and requires states to adopt water quality standards for waters to protect public health or welfare and enhance quality of water for its citizens. Additionally, states are required to establish total maximum daily loads (TMDLs) for impaired waters. If a state fails to regulate pollution discharge, the US EPA is authorized to directly regulate the state's waters. Most states have water quality protections in place for streams and lakes, but relatively few states have water quality standards for wetlands. In order to set wetland water quality standards we need quality data on water quality and the potential impacts on biological communities such as invertebrates, turtles, and small mammals. These types of data are important because wetland organisms serve as important biological indicators of ecosystem health and function. We will sample 200 wetlands cross West Virginia for a variety of water quality variables to represent a gradient of pristine to highly impacted sites. A smaller sample of wetlands will be studied to determine how wetland quality affects wildlife communities (invertebrates, small mammals, turtles). Water quality will be measured in the lab and the field. Turtles will be captured and heavy metal levels from blood and toe nails evaluated. Invertebrate communities will be related to water quality and habitat structure and small mammals will be evaluated in response to surrounding landuse and habitat structure. These data will help to fill in existing data gaps on West Virginia's wetlands and contribute to development of appropriate management actions and guidelines for improved environmental quality and human health.
Animal Health Component
70%
Research Effort Categories
Basic
20%
Applied
70%
Developmental
10%
Goals / Objectives
This study builds on the existing progress of the EPA-approved WV Wetland Program Plan and focuses on wetland water quality criteria and biotic communities for WV. Our overarching goal is to advance WV's wetland program by incorporating water quality, small mammal communities, and turtle communities and contaminant levels into the existing Level 1 to 3 assessments. Specifically, we will:characterize water quality conditions (and biological indicators: vegetation and invertebrates) for a statewide wetland sample stratified by ecoregion, WVWRAM water quality scores, HGM, and NWI system class (palustrine class);evaluate water quality conditions and turtle heavy metal levels in relation to surrounding landuse, and biotic and abiotic parameters; andevaluate small mammal and turtle community composition, abundance, and diversity in relation to wetland types and abiotic and biotic factors.
Project Methods
We will sample 200 wetlands quarterly over a 1- or 2-year period for total AL, alkalinity, total chloride, dissolved oxygen, total Fe, fecal coliform and/or E. coli, hardness, hot acidity, total K, total Mg, total Mn, total Na, NO2-NO3-N, total P, dissolved Pb, pH, total Se, specific conductance, sulfate, total dissolved solids, total suspended solids, dissolved Zn, and total Zn. Physicochemical variables such as pH, specific conductance, and dissolved oxygen saturation will be measured in situ at the time of sampling. Elemental analyses will be conducted using an inductively coupled plasma optical emission spectrometer capable of elemental detection at sub-parts-per-billion (ppb) scale. Concentrations of dissolved organic carbon and total dissolved nitrogen will be determined using a Shimadzu total organic carbon and total nitrogen analyzer. Concentrations of individual nitrogen species will be analyzed using a Hach DR 3800 spectrophotometer. Fecal coliform and E. coli will be quantified using the IDEXX system. Statistical comparison of water quality results among wetland types and ecoregions will be performed using established parametric and non-parametric methods.We will sample floodplain wetlands for macroinvertebrates once during April to June and again from October to December to facilitate wetland water quality index of biotic integrity development. We will collect ten 5-cm diameter (15-cm deep) soil core samples with a PVC corer and four 50-cm wide, 5-m long D-frame sweep net samples at each wetland to ensure coverage of the entire breadth of the wetland. Samples will be processed and the number of individuals by family (and functional feeding groups) will be tallied, and biomass obtained by oven-drying samples between 50-60 °C for at least 48 hours to a constant mass (0.0001 g).Our previously developed disturbance gradient will be refined and scores above and below the 75th and 25th percentile will be used to categorize reference and stressed conditions. Nektonic and benthic macroinvertebrate data will be used to derive candidate metric values that are evaluated for their capacity to discriminate between reference and stressed sites. Responsive metrics will be identified using box-and-whisker plots, screened for redundancy using Spearman's R correlation (>0.80), and analyzed with multivariate analysis of variance (MANOVA).We will target 30 sites for turtle abundance, movement, and contaminants. Turtles will be trapped using aquatic hoop nets baited with sardines. For each individual captured, 1 mL of blood will be drawn from the dorsal or ventral caudal vein, and 1-2 mm trimmings from 3-4 hind toenails will be collected for analysis. Both blood and nail samples will be collected in the field, where blood samples will be taken with 1 mL non-heparinized syringes, and nail samples will be taken with human or dog nail clippers (depending on the size of the turtle). Turtles will be shell notched and some will have transmitters applied to them. Samples will be stored in 1.5 mL Eppendorf tubes and stored in a cooler followed by transfer to a freezer at -20°C immediately upon return from the field until analysis.Later we will track turtle locations via telemetry and points will be plotted on ArcGIS. Radiotagged turtles will be found every 2-5 days between April and October. Relocations will be made once monthly from November to March to confirm hibernation locations, take note of new hibernacula if applicable, and observe when turtles leave hibernacula. Telemetry for each day will begin at the most recent location. Locations of individuals that are tracked in inaccessible locations (e.g. deeper water, private land) will be estimated via triangulation and GPS point projection. For each turtle relocation, the landscape and distance from nearest wetland will be determined on ArcGIS.The effects of surrounding landscape types on turtle abundance and movement will be quantified through digitized turtle radiolocations within each buffer zone to identify average distances traveled and determine preferences of landscape type. Soil composition within 30 m of each site will also be quantified by calculating average percent sand in soil using the Soil Survey Geographic Database and associated raster data.Blood, nail, and soil samples will be tested for cadmium (Cd), chromium (Cr), lead (Pb), total mercury (THg), selenium (Se), and zinc (Zn). Samples will be analyzed according to the EPA Method 3050B with a Perkin Elmer NexIon 2000 inductively-coupled plasma mass spectrometer.N-mixture models will be used to quantify the influence of habitat quality on abundance of turtles. The adequacy of models for air temperature, water temperature, pH, dissolved oxygen, water depth at trap site, and average depth of wetland will be calculated by determining goodness-of-fit with Akaike Information Criteria (AIC) values. Landscape analyses will be completed through skewness indices to determine land use preferences in turtles within the 1 km buffers. Modal centers of activity (MCA) will be determined by identifying areas with relatively discrete clusters of radiolocations, and the habitat classification will be determined for each individual from the center point in each MCA segment. Favored landscape types will be determined for each sex class, if possible. Logistic regressions will be fit to determine correlations for turtle abundance in wetlands based on average distance from other wetlands, average distance from roads, and wetland size. The best model to explain turtle abundance in wetlands will be determined based on AIC values, with the best-fit model having the lowest value.We will trap small mammal communities at mitigated and natural wetlands (paired study) and also at high-elevation wetland sites with various surrounding landuse practices across West Virginia using 5.08 x 6.35 x 16.51 cm (2 x 2.5 x 6.5") folding Sherman Live Traps. We will place traps along two parallel transects with traps spaced 10 m apart and transects spaced 50 m apart. Upon capturing a small mammals, we will record species, weight, and length and we will mark or tag individuals. After each transect has been set up, an assessment of ground cover percentage will be conducted. This assessment will be made every 10m, at every trap along each transect. The trap will be centered in a 1 m2 quadrat. The percentage of ground cover type will be recorded for the following types: herbs, shrubs, trees, bare soil, leaf litter, rock, and water. These categories will be assessed using a Daubenmire scale of six cover classes. Depth to the nearest 0.5 cm will be obtained for leaf litter and water. Dominant species ( >50%) will be recorded for any vegetative cover class (herb, shrub, tree) that is present. In addition to a percentage of ground cover, We will also record the percent canopy cover at every trap along each transect using a spherical densiometer. This survey will be taken at the end of every trapping period to reduce disturbance prior to animal capture.Historic anthropogenic land uses will be put into GIS as layers and used to determine where specific historic land uses have occurred. Specifically, we will use the NCLD (National Land Cover Database) dataset to help determine past land use. By using these data in conjunction with historical records that are obtained for the sites, we will be able to determine which historical land uses have been performed at or around high elevation wetland sites. Cluster analysis, which works by grouping sites that are most similar in terms of any metric that is being considered, will be used for determining sites that have similar community composition. Abundance of small mammals will be estimated using the package RMark for analysis of capture-recapture data. Species abundance, richness, and diversity will be analyzed using N-mixture models.