You are here:
Projecting the impacts of climate change on wetland dynamics across the central U.S. using Sentinel-1 and Sentinel-2 based time series (2017-2021)
Citation:
Vanderhoof, M., J. Christensen, L. Alexander, C. Lane, AND H. Golden. Projecting the impacts of climate change on wetland dynamics across the central U.S. using Sentinel-1 and Sentinel-2 based time series (2017-2021). AGU 2023 Fall Meeting, San Francisco, CA, December 10 - 15, 2023.
Impact/Purpose:
Using remote sensing technology, we project a decline in surface water extent, with drying most consistent across the Southeast, Southwest, and Midwest. Projected declines under the driest 15% of conditions at these sites suggest impacts of drought are likely to be amplified. Projected changes were seasonally variable, with the greatest decline in surface water extent expected in summer and fall seasons. In contrast, some northern sites showed a projected increase in surface water in most seasons, attributable to increases in winter and spring precipitation exceeding increases in temperature.
Description:
Frequent remotely sensed observations of surface waters at fine spatial scales will provide critical data to support the management of aquatic habitat, including projecting the impacts of climate change on freshwater ecosystems. We developed surface inundation algorithms for Sentinel-1 and Sentinel-2, respectively, at 12 sites (>536,000 km2) across the conterminous U.S. (2017-2021). Each image was classified into open water (e.g., lakes, rivers, ponds, reservoirs), vegetated water (e.g., wetlands, riparian corridors), and non-water at 20 m resolution. We found that the two open water time series were highly correlated over time, but sensor-specific differences, such as sensitivity to vegetation structure versus pixel color, complicated the data integration for mixed-pixel, vegetated water. We then propose an approach to integrate the two algorithms using an accuracy-based voting scheme. The integrated Sentinel-1 and Sentinel-2 data was used to track monthly surface water extent for 32 sites across the central United States. Time series of surface water extent was modeled as a function of weather using site-specific statistical models to account for variability (e.g., topography, water storage capacity, land use) in the response of surface water extents to precipitation. Climate change adjusted variables were based on MACA-CMIP5 (MACAv2-METDATA) models, as the change between current, 2006-2025 and future, 2061-2080. Annually, 19 of the 32 sites at RCP4.5, and 22 of the 32 sites at RCP8.5 were projected to show a decline in surface water extent, with future drying most consistent across the Southeast, Southwest, and Midwest. Projected declines under the driest 15% of conditions at these sites suggest the impacts of drought are likely to be amplified. Projected changes were seasonally variable, with the greatest decline in surface water extent expected in summer and fall seasons. In contrast, some northern sites showed a projected increase in surface water extents in most seasons, attributable to increases in winter and spring precipitation exceeding increases in temperature.