Wetlands are among the most sensitive ecosystems to climate change as their hydrology, which is directly influenced by climate, is an important driver of the establishment and maintenance of wetland habitat types and processes. Semi-arid wetlands are believed to be especially sensitive as small changes in temperature and precipitation can have large impacts in the timing and duration of flooding. However, tools to forecast climate impacts on wetlands are severely limited relative to other ecosystem types. Landscape-level hydrologic data for wetlands is scarce because tracking changes in wetland water levels over weeks and months requires the installation of expensive monitoring equipment or visiting sites many times a year for several years. We derived hydrographs of individual wetlands from a time series of 581 Landsat satellite images detailing the seasonal flooding and drying for thousands of wetlands for almost three decades (1983 – 2011). Our method using a sub-pixel technique called spectral mixture analysis to detect fine scale changes (<30 m) in surface water changes. We combined this dataset with soil moisture and groundwater variables simulated by the physically based Variable Infiltration Capacity (VIC) hydrologic model to develop site specific regression models for wetlands across the Columbia Plateau (CP) ecoregion. We then used these models to project the impacts of climate change using global climate model scenarios for the 2080s (A1B emissions scenario).
Results/Conclusions
Our remote sensing dataset allowed us to examine historic patterns of flooding and drying, understand how climate variables relate to current hydrology as well as project future hydrologic response under climate change, across a broad spatial extent. We found that wetlands within a similar geographic location may have drastically different hydrologic responses under climate change, not only in magnitude, but also in directionality, with some wetlands getting wetter and others getting drier. While our results may seem counterintuitive this bi-directional response is driven by the underlying hydrologic inputs for wetlands. In general, groundwater driven wetlands were more sensitive to increases in winter precipitation levels and had increases in max./min. water levels, dried less frequently, and dried later in the season, because of their ability to store water in winter months. On the other end of the spectrum, surface water driven wetlands, were more sensitive to increases in temperature, and decreases or no change in max./min. water levels, dried more frequently, and dried earlier.