Multiscale environmental and management effects on forest water use and productivity in contrasting U.S. regions

Forest carbon and water budgets, and their coupling through water-use efficiency (WUE; the ratio of carbon uptake to water loss), are influenced by multiple factors, including forest structure, climate, topography, soil, and management strategies. Drivers of forest productivity and water use are generally well understood at scales of individual trees to stands. However, as that scale extends to landscapes and biomes, the relative importance of these drivers may change, challenging our ability to predict changes in regional forest health. Here, we applied multiscale geographically weighted regression to quantify the influence of environmental factors and forest management types on net primary production (NPP), evapotranspiration (ET), and WUE across two ecologically contrasting U.S. forest regions: the Southeast (SEUS) and the Pacific Northwest (PNW). Results revealed that the relationships between NPP, ET, and WUE and predictors were relatively uniform across the SEUS, likely due to the region's relatively homogeneous humid climate and management approaches. In contrast, the effects of predictors in the PNW were less uniform due to its complex topography, climatic gradients, and diverse management types. Results also indicated clear regional contrasts in the effects of forest management on carbon-water dynamics. In the SEUS, ecological management maximized WUE in most locations, whereas in the PNW, shifting to preservation management produced the greatest increase in WUE. This study presents a spatially explicit framework for enhancing regional carbon and water budget predictions and informing forest management under changing environmental conditions.