Characterization of soil/vegtation on flood irrigated hayfields in Grand Teton National Park, Wyoming : a predictive evaluation tool for agricultural wetlands

Abstract

The Elk Ranch hayfield in Grand Teton National Park (GTNP) has been historically flood-irrigated since the early 20th Century. The park service is now considering closing irrigation to restore native plant communities and enhance Spread Creek fisheries and will need information on the extent of irrigation-created wetlands and how irrigation cessation would change the vegetative component of the ranch. The main objective of this study was to assess the relation between soil and vegetation characteristics of wetland community types at the ranch and to determine if any of the relationships could be used to differentiate between naturally occurring and irrigation created wetlands. Vegetation data were collected from transects centered on a soil pit at 28 randomly located sample points throughout the hayfield. Twenty-six of the 28 sample plots were classified as wetland based on criteria listed by the US Fish and Wildlife Service. Bray-Curtis dissimilarity and nonmetric multi-dimensional scaling were used to analyze percent foliar cover, wetland index value (WIV), soil texture, percent organic matter, redox contrast and abundance, and depth to groundwater and soil saturation for each of the sampled points. The WIV and redox contrast had the greatest dissimilarity (D²), 0.90, and 0.71 respectively across the hayfield. The other measured characteristics had D² values ranging from 0.23 to 0.49 and were strongly correlated with the WIV and redox contrast measures. However, inclusion of these measures contributed little to the differences already identified. Categorical organization of WIV and redox measures indicated that naturally occurring wetlands could be differentiated from wetlands created by flood irrigation in former upland vegetation communities. Combining wetland index value and soil redox contrast suggests park managers could identify wetland community types likely to remain or transition following cessation of flood irrigation at the Elk Ranch. Additional testing at other GTNP sites will be necessary to test the broad application of this approach and refine the assessment categories.