Authors: Anthony Preucil – RTI International
Title: Dynamic Parameterization of the Priestley–Taylor Coefficient for Improved PET Estimation Using FluxNet Observations
Presentation Type: Poster Presentation
Abstract: Potential evapotranspiration (PET) represents the rate of evapotranspiration (ET) that would occur under prevailing atmospheric conditions if water were not limiting. It is a fundamental variable in hydrologic and land surface modeling, often serving as a basis for estimating actual ET. The Priestley-Taylor (PT) method is widely used to estimate PET because of its simplicity, typically employing a constant coefficient (αₚₜ = 1.26). However, this assumption overlooks well-established variations in αₚₜ driven by differences in hydroclimate and vegetation physiology, which can reduce model accuracy across diverse ecosystems. In this study, we analyze observations from 246 FluxNet eddy covariance (EC) sites to calibrate αₚₜ across a wide range of biomes. The results reveal pronounced spatial and seasonal variability in αₚₜ, highlighting the limitations of a fixed coefficient and the need for dynamic parameterization. We further develop empirical relationships that relate αₚₜ to environmental conditions, enabling improved PET estimation. When incorporated into the Evaporative Stress Index (ESI), this optimized framework reveals systematic biases in stress estimates arising from the conventional PT formulation. Overall, the findings demonstrate the value of a more flexible and ecologically grounded representation of the PT coefficient for improving PET estimates in land surface models and drought monitoring applications.