Estimating Soil Evaporative Fluxes from Soil Surface Temperature Data and a Model of Soil Heat Flow

Water vapor fluxes (ET) are a combination of soil evaporative fluxes (E) and plant transpiration fluxes (T). One common method of estimating E and T separately employs a soil-plant-atmosphere model, which would be verified by comparing the model’s ET predictions against observed ET measurements. Such a model may yield very good estimates of ET but may still need other observational data or some other external information (about E and/or T separately) to determine how well the model simulates E and T. Here we propose a novel method of estimating E from a numerical solution to the soil heat flow equation. The conceptual basis of this method follows from a reinterpretation of the soil heat flux estimated using the apparent thermal conductivity (which includes the effects of soil evaporation/condensation), rather than the soil heat flux alone (which does not include the effects of E or condensation). The soil heating model is driven by observed soil surface temperature measured every 5 minutes and is evaluated against the observed profile of soil temperature. Other model inputs include soil profile data from single-needle thermal conductivity probes and soil moisture data from TDR probes. At this stage the model estimates of E cannot be independently verified, but they are tested for reasonableness by comparing with contemporaneous eddy covariance flux observations.