In-Situ Quantification and Prediction of Water Yield From Southern US Pine Forests

Forest management can play an important role in landscape-scale water balances and thus regional water supply planning, necessitating improved quantification and prediction of forest water yield (i.e., rainfall minus evapotranspiration (ET)). We used high frequency soil moisture data to quantify soil ET and interception in 30 pine stands capturing regional variation in aridity, hydrogeology, and forest management. We evaluated typical forest rotation stages (i.e., clear cuts through mature stands), as well as stands restored to historical, lower biomass conditions. Our results supported the expectation that forest management can strongly influence local water yield. A simple model using leaf area index (LAI), hydrogeologic setting, and climate aridity (P:ET) explained nearly 80% of observed water yield variation. LAI emerged as the dominant forest structural control, influencing both soil ET and interception rates, with each unit decrease in LAI increasing water yield by nearly 10 cm. While other forest attributes (e.g., basal area, groundcover, species) were less important for predicting stand-level water balances, aridity and hydrogeologic setting emerged as highly significant predictors of water yield. We further observed small and short-lived effects of low-intensity prescribed fires on soil ET and no discernible effect of pine species, suggesting that maintaining low density pine forests-regardless of species-is a viable management strategy for increasing water yield. Overall, our results illustrate the utility of soil moisture-based methods for stand-level water balances and provide useful models for predicting landscape water yield under a range of forest management and hydroclimatic settings now and in the future.