Drought impacts on ecosystem functions of the U.S. National Forests and Grasslands: Part I evaluation of a water and carbon balance model
Understanding and quantitatively evaluating the regional impacts of climate change and variability (e.g., droughts) on forest ecosystem functions (i.e., water yield, evapotranspiration, and productivity) and services (e.g., fresh water supply and carbon sequestration) is of great importance for developing climate change adaptation strategies for National Forests and Grasslands (NFs) in the United States. However, few reliable continental-scale modeling tools are available to account for both water and carbon dynamics. The objective of this study was to test a monthly water and carbon balance model, the Water Supply Stress Index (WaSSI) model, for potential application in addressing the influences of drought on NFs ecosystem services across the conterminous United States (CONUS). The performance of the WaSSI model was comprehensively assessed with measured streamflow (Q) at 72 U.S. Geological Survey (USGS) gauging stations, and satellite-based estimates of watershed evapotranspiration (ET) and gross primary productivity (GPP) for 170 National Forest and Grassland (NFs). Across the 72 USGS watersheds, the WaSSI model generally captured the spatial variability of multi-year mean annual and monthly Q and annual ET as evaluated by Correlation Coefficient (R = 0.71–1.0), Nash–Sutcliffe Efficiency (NS = 0.31–1.00), and normalized Root Mean Squared Error (0.06–0.48). The modeled ET and GPP by WaSSI agreed well with the remote sensing-based estimates for multi-year annual and monthly means for all the NFs. However, there were systemic discrepancies in GPP between our simulations and the satellite-based estimates on a yearly and monthly scale, suggesting uncertainties in GPP estimates in all methods (i.e., remote sensing and modeling). Overall, our assessments suggested that the WaSSI model had the capability to reconstruct the long-term forest watershed water and carbon balances at a broad scale. This model evaluation study provides a foundation for model applications in understanding the impacts of climate change and variability (e.g., droughts) on NFs ecosystem service functions.