Using Canopy Temperature to Infer Hydrologic Processes in Floodplain Forests

Abstract:

Decreased water availability due to hydrologic modifications, groundwater withdrawal, and climate change threaten the hydrological architecture of floodplain forests globally. The relative contributions of different sources of water (e.g., precipitation, surface flooding, and groundwater) to soil moisture on floodplains is poorly constrained, so identification of areas of water stress within a floodplain can provide valuable information about floodplain hydrology. Canopy temperature is a useful indicator of moisture stress and has long been used in agricultural and natural landscapes. Accordingly, thermal infrared (TIR) remote sensing data (spatial resolution of 1 km) from NASA’s MODIS sensor was used to examine patterns of spatiotemporal variation in water stress in two floodplain forests over 12 growing seasons. On the upper Sabine River floodplain, Texas, increasing rainfall-derived soil moisture corresponded with increased heterogeneity of LST but there was weak association between river stage and heterogeneity. On the lower White River floodplain, Arkansas, distinct differences in LST between two reaches were observed during low flow years, while little relationship was observed between LST spatial variability and rainfall-derived soil moisture on either reach. The differences in hydrological control on these floodplain ecosystems have important ramifications for varying resilience to climate change and water resource management.