COS 121-3 - Quantifying terrestrial and marine ecosystem connectivity with stable isotopes at the watershed scale

Thursday, August 10, 2017: 8:40 AM
E142, Oregon Convention Center
Eric J. Chapman, Marine Science Center, University of New England, Biddeford, ME, Carrie J. Byron, Marine Science Department, University of New England, Biddeford, ME, Theodore V. Willis, Aquatic Systems Group, University of Southern Maine, Portland, ME and Karen A. Wilson, Environmental Science and Policy, University of Southern Maine, Gorham, ME

Terrestrial, freshwater, and marine ecosystems are tightly linked through material exchange and energy transfer. This link is especially clear in the oft-used example of the Gulf of Mexico hypoxic zone that is generated from excessive nutrients from agrarian land use in the Midwestern United States. In this study, we used stable isotope analysis (SIA), existing streamflow data, and geospatial datasets to quantify linkages between terrestrial and coastal ecosystems at the watershed scale along the Maine coastline. We analyzed stable isotopes of carbon (δ13C) and nitrogen (δ15N) of marine fish tissue from marine systems adjacent to three terrestrial watersheds that represent increasing levels of development and decreasing levels of freshwater input: Saco Bay, St. George, and Cobscook Bay. Using this coastline gradient we investigated the influence of terrestrial land cover, land use, and various terrestrial catchment characteristics on nearshore bay food web structure.


The Saco Bay catchment had the highest mean annual riverine input volume by an order of magnitude (1.24 x 109 m3) and was followed by St. George (2.32 x 108 m3) and Cobscook Bay (1.76 x 108 m3). Along our coastline watershed gradient, we found distinct δ13C and δ15N signatures in fish tissues. We observed higher 15N values in samples from St. George compared to Saco Bay and Cobscook Bay; this finding was explained by differences in fish lengths (R2 = 0.694, p < 0.001). With respect to δ13C values, St. George and Cobscook Bay had less variable and higher δ13C values, indicating that St. George and Cobscook Bay were more prominently influenced by marine derived carbon than Saco Bay. Regression analysis showed that as the annual input volume of freshwater increased, δ13C values decreased significantly (R2 = 0.286, p < 0.001). Our results suggest that near ocean (< 200 m from the coast) terrestrial catchment characteristics and land use have profound influence on nearshore ecosystem structure, highlighting the need to explicitly consider the impacts of coastal land use on near shore ecosystem structure by municipal zoning and planning boards.