Validating climate models for computing evapotranspiration in hydrologic studies: how relevant are climate model simulations over Florida?
- First Online:
- Cite this article as:
- Obeysekera, J. Reg Environ Change (2013) 13(Suppl 1): 81. doi:10.1007/s10113-013-0411-0
- 250 Downloads
A non-stationary climate requires a new paradigm in the current practice of using past observations for future planning. Next to precipitation, evapotranspiration is the most important variable in water budgets of regions such as Florida, the location of this study. Dynamical simulations using Regional Climate Models (RCMs) provide all the variables necessary to compute potential or reference crop evapotranspiration (RET) for hydrologic modeling under climate change scenarios. Data sets for the variables necessary to compute RET using the Penman–Monteith model were obtained from RCMs provided by the North American Regional Climate Change Assessment Program to evaluate their skills in estimating RET during the late twentieth century simulations. Comparison of RET values simulated from RCMs with those computed from a high-resolution weather-monitoring network and satellite data show significant biases in spatial patterns and seasonality. In particular, seasonal patterns of RET computed from RCMs show a phase shift with peak values occurring during mid-summer months, whereas observed data show earlier peaks in May and June. The phase shift reflects similar behavior in model predictions of incoming solar radiation. In addition, relative humidity values estimated from RCMs are significantly higher than observed values and the wind speed estimates from many RCMs have a significant bias during dry season months. Further improvements to RCMs may be needed before they can be used effectively for hydrologic modeling under future climate change scenarios.