Comparison of Two Temperature-Based Methods of Estimating Potential Evapotranspiration (PET) in Texas

Assessment of future drought characteristics requires knowledge of potential evapotranspiration (PET) which is commonly estimated using the Penman-Monteith (PM) method. This method requires knowledge of climatic variables such as relative humidity and wind speed which are not always readily available, particularly for regional-scale studies. Alternate methods such as the Thornthwaite (Th) and Hargreaves (Hg) have been widely-used as they only require temperature input. In this study, we have applied both the Th and Hg methods to a suite of 22 global climate models (GCMs) available from the recently released Intergovernmental Panel for Climate Change Coupled Model Intercomparison Project Phase 5 (IPCC CMIP5) under the Representative Concentration Pathway 8.5 (RCP 8.5) to evaluate the 21st century evolution of PET in different climate divisions of Texas. While both methods produce statistically significant (p<0.01) increasing PETs as we move into the 21st century, the slope of these trends are two to three times greater for the Th compared with the Hg. The Th method appears to be very sensitive to temperature and overestimates PET in response to the increasing temperatures projected by GCMs the under the RCP 8.5, particularly during the summer (June-July-August) months. Studies that use the Th method may thus overestimate future drying due to the pronounced increase in PETs. As such, the Th method appears to be unsuitable for drought assessment studies in Texas and where only temperature data may be available, we recommend the use of the Hg method.