APPLICATION OF GEOCHEMICAL END-MEMBER MIXING ANALYSIS TO DELINEATE WATER SOURCES TO A FORESTED RIPARIAN WETLAND

Water levels in coastal wetlands greatly influence carbon release into or capture from the environment and hence are critical to global climate change. These levels are influenced by streams, groundwater, soil type, topography, precipitation, evaporation, etc. However, the relative influence of each of these factors is usually hard to ascertain based on hydrology alone. We used an end-member mixing technique involving geochemical data to delineate the influence of various sources of water to a small forested riparian wetland in the 7000-hectare Turkey Creek Watershed present in the Francis Marion National Forest near Charleston, SC. This watershed is characterized by very flat slopes and overall elevation of 4-14 m above sea level. The soils are primarily poorly-drained and influenced by seasonally high water tables. We installed a series of water table wells and drive-point piezometers to a depth of 10 ft below ground surface in the stream bed and at the margin areas adjacent to the riparian zone. Over a span of two months in July and August 2007, water samples were collected on a weekly basis. Water levels were also monitored during this period. Major cation and anion concentrations were analyzed in these water samples. Due to exceptional drought conditions during that period, we were unable to sample precipitation. Preliminary results indicated that the water tables levels were very low in most wells, with some wells not yielding any water during this period. Chemical analysis indicated that that some well water contained higher levels of sulfate and chloride compared with others possibly due to influence of a deeper groundwater system of higher salinity. End-member mixing analysis performed on the data to delineate specific sources of water to the wetland was inconclusive. Additional data are required to establish a more definitive model of surface and subsurface contributions to the groundwater in the region. One small benefit from this study is that the data collected are most likely representative of a “worst-case scenario” condition of extreme drought. In short, we have the “dry end” of the climate spectrum identified with regard to how a riparian wetland system can respond hydrologically and chemically as one component of the water budget within this Coastal Plain watershed.