Quantifying drought in eastern Texas floodplain forests using remote sensing and water-energy balance models

Abstract:

Floodplain hydrology is a key driver of floodplain forest ecosystems. Changing climate and intensifying water resource management can increase drought, yet processes controlling plant water availability and use are poorly understood. Variation of water and energy availability within floodplain forests is significant, depending not only on seasonality and hydrologic connectivity to rivers but also on heterogeneities in soil, topography, and vegetation communities. We examined tree responses to water deficits in three floodplains in eastern Texas (Angelina, Sabine, and Trinity rivers) using remotely sensed, gridded estimates of gross primary productivity and transpiration. We compared these responses to gridded datasets of precipitation, evaporative demand, and soil water potential using energy-balance models in the coupled water and energy balance framework postulated by Budyko, but at the resolution of a grid instead of a watershed. Water and energy balances during 2003-2017 at monthly and annual scales within the three floodplains revealed spatial variation in apparent soil moisture subsidies that affects drought responses. Distributed tree-ring data provided ground-based validation of remotely sensed forest responses to water deficit. Results also suggest a potential dryness index for characterizing hydrological drought specific to floodplains, using a combination of remotely sensed data and a water-energy balance model.