A preliminary study on the feasibility of continuously estimating evapotranspiration from vegetated surfaces using the three-temperatures method.

Although Evapotranspiration (ET) has long been recognised as a key process for the redistribution of water and energy at a global scale, there remains uncertainty in the actual ET rates at a local scale from vegetated Sustainable Drainage Systems (SuDS). ET has been seen to account for between 61 % and 21 % of the water balance in these systems, demonstrating its significance in the overall system performance. There is a requirement to improve our understanding of the variance of ET rates from SuDS and similar systems. By extension, there is a need for robust ET estimation methods which can be readily applied to a variety of SuDS at different spatial and temporal scales.

The Three-Temperatures (3-T) method is one such approach, which only requires net radiation and surface temperatures from the vegetated surface and a corresponding imitation surface, alongside the overlying air temperature. This method has been previously applied to a variety of different surface types, spatial scales and environments. However, it has been met with a varying degree of success and often only produced spot ET estimates. Furthermore, its limitations are not fully understood and producing a continuous record of ET estimates allows us to see when and under what conditions spot estimates of 3-T ET may be considered credible.

This preliminary study aimed to determine if reasonable continuous ET estimates could be achieved from the 3-T method for a small vegetated surface analogous to SuDS and or green infrastructure (GI). This included the establishment of an experimental setup, which captured the relevant 3-T parameters and those required to calculate hourly reference ET rates as determined by the FAO 56 Penman–Monteith (P-M) method, to use for comparison purposes. Practical considerations (e.g. building shadowing) and sensitivity analysis of 3-T ET estimates to changes in the 3-T parameters were also explored, to provide a deeper understanding of the method’s robustness.

Initial results indicated that the 3-T method can produce periods of ‘reasonable’ continuous hourly ET values, between 0.0 mm.hr^-1 to 0.5 mm.hr^-1 under preferred conditions. Following a period (up to 3 days) of dry weather conditions, the cumulative reference ET was 2.3 mm and the corresponding 3-T ET was 2.9 mm, showing a total difference of 26% at the end of 3 days. The tendency of the 3-T method to produce higher ET estimates during the day compared to the reference ET values, was attributed to instances where the surface temperatures approach that of the air temperature. The preliminary findings show promise for the 3-T method to produce continuous records of ET, but have also highlighted the need for further research on the method’s application to vegetated SuDS and or GI.