Quantifying Potential Evapotranspiration in a Changing Climate (Core Science Lecture)

Potential evapotranspiration (PET) is one of the two crucial atmospheric inputs to hydrologic water-balance models, the other being precipitation. Climate models provide precipitation as an output, but not PET. To project potential impact of various scenarios of climate change on water balance, therefore, hydrologists often resort to some model of PET (e.g., Penman-Monteith, Thornthwaite) to convert projected atmospheric variables to PET for input to hydrologic models. Questions arise: How reliable are the various PET methods for this purpose? Indeed, how can we evaluate their reliabilities without waiting for the future to play out? How sensitive are hydrologic projections to PET models? Does the scatter in change of actual evapotranspiration and runoff obtained by using multiple PET methods truly represent uncertainty? What changes in PET might be expected under a given scenario of climate change and what controls those changes? An attempt to begin to address these questions will be made in this lecture.