Towards Remote Sensing for Hydrologic Prediction in Ungauged Basins
Jeffrey P. Walker
Earth observing satellites have the potential to revolutionise our understanding and prediction of ungauged basins, but historically remote sensing data has not been widely used in hydrology. Moreover, during the last decade there has been a dramatic increase in the number of hydrologically relevant remote sensing missions. There are four main ways that such remote sensing observations may be used in hydrologic models: (i) as parametric input data, including soil and land cover properties; (ii) as initial condition data, such as initial snow water storage; (iii) as time-varying hydrological flux data to propagate model predictions, such as precipitation and radiation meteorologic forcing data; and (iv) as time-varying hydrologic state or flux data to constrain model predictions, such as soil moisture content or evapotranspiration. This paper focuses on the latter and demonstrates how: (i) rootzone soil moisture may be retrieved from AMSR-E near-surface soil moisture; (ii) soil moisture and groundwater may be retrieved from GRACE temporal gravity; (iii) soil moisture and runoff may be retrieved from TOPEX/POSEIDON water level measurements; (iv) snow depth and temperature may be retrieved from MODIS snow cover and AMSR-E snow water equivalent; and v) soil moisture, soil temperature and evapotranspiration may be retrieved from Landsat, MODIS or GOES skin temperature or inferred latent and sensible heat flux. Coordinated field campaigns within gauged catchments have, and will continue to, contribute significantly to advances in use of remote sensing for hydrologic prediction in ungauged basins.