Relations Between Temperature and Evapotranspiration: How Complex are They?

David I. Stannard

Abstract

Evapotranspiration (ET), the flux of water from the land surface to the atmosphere, ultimately is driven by a vapor pressure gradient at the land-surface atmosphere interface. Many variables in the soil-plant-atmosphere system, including temperature, affect this gradient. For instance, evaporation from a moist surface increases with surface temperature because saturation vapor pressure increases with temperature. This fact has led to several successful potential ET equations based primarily on air temperature (which is highly correlated with surface temperature). However, because of the interrelations between many variables, it is difficult to generalize about the interdependence of ET and temperature. In temperate climates, soil moisture and soil temperature tend to be negatively correlated, which often reduces ET at warmer temperatures. Vegetative stomatal controls on transpiration also are affected by temperature, usually reducing ET at extremely warm or cool temperatures. Although ET cools a surface in the short term, continued ET dries a surface that is not replenished (by precipitation or shallow groundwater), leading to greater surface warming under a given solar input. Additionally, sustained regional ET increases atmospheric vapor pressure and can lead to cloud formation, both of which feed back on surface temperature.

At the field scale, relatively simple analytical models are available to predict ET from state variables such as temperature. At this scale, interactions between the surface of interest and the upper planetary boundary layer (PBL) are unimportant. At the regional, continental and global scales, however, there is constant feedback between surface processes and the upper PBL. Numerical atmospheric circulation models, coupled with realistic surface process models, probably are the only hope of accurately predicting the effects of such changes as global warming on ET and other hydrologic fluxes. For example, if warming is the result of increased CO₂, the effects of increased CO₂ on stomates, plant growth, and atmospheric chemistry also need to be evaluated, to accurately predict the hydrology. To achieve greater realism, models are becoming increasingly interdisciplinary; however a large amount of uncertainty still exists in the prediction of future climate.