Temporal and spatial patterns of wetland extent influence variability of surface water connectivity in the Prairie Pothole Region, United States

Abstract

Quantifying variability in landscape-scale surface water connectivity can help improve our understanding of the multiple effects of wetlands on downstream waterways. We examined how wetland merging and the coalescence of wetlands with streams varied both spatially (among ecoregions) and interannually (from drought to deluge) across parts of the Prairie Pothole Region. Wetland extent was derived over a time series (1990–2011) using Landsat imagery. Changes in landscape-scale connectivity, generated by the physical coalescence of wetlands with other surface water features, were quantified by fusing static wetland and stream datasets with Landsat-derived wetland extent maps, and related to multiple wetness indices. The usage of Landsat allows for decadal-scale analysis, but limits the types of surface water connections that can be detected. Wetland extent correlated positively with the merging of wetlands and wetlands with streams. Wetness conditions, as defined by drought indices and runoff, were positively correlated with wetland extent, but less consistently correlated with measures of surface water connectivity. The degree of wetland–wetland merging was found to depend less on total wetland area or density, and more on climate conditions, as well as the threshold for how wetland/upland was defined. In contrast, the merging of wetlands with streams was positively correlated with stream density, and inversely related to wetland density. Characterizing the degree of surface water connectivity within the Prairie Pothole Region in North America requires consideration of (1) climate-driven variation in wetness conditions and (2) within-region variation in wetland and stream spatial arrangements.