Evapotranspiration BEE

Evapotranspiration at EFForTS-BEE: We recorded land and canopy surface temperatures at the 56 BEE plots in August and September 2017 using a radiometric thermal camera (FLIR Tau 2 640, FLIR Systems Inc., Wilsonville, OR, USA) attached to a TeAx ThermalCapture module (TeAx Technology GmbH, Wilnsdorf, Germany) mounted on a multicopter drone (MK EASY Okto V3; HiSystems, Moormerland, Germany). The thermal images were recorded around noon, covering each plot once over a nine day period encompassing varying weather conditions. The land and canopy surface temperatures were the main input for modeling latent heat flux (in W m-2) and deriving evapotranspiration (ET, mm h-1) using the QGIS3 Plugin 'QWaterModel' (Ellsäßer et al., 2020), which bases on the DATTUTDUT energy balance model. Measured short-wave radiation and relative humidity were used as further input variables to support the prediction of evapotranspiration with QWaterModel. The dataset consists of the following files: 1.) Results table of evapotranspiration (ET, mm h-1) at the 56 BEE plots (Evapotranspiration_BEE.xlmx), including auxiliary variables needed for the calculation of ET with QWaterModel: plot ID, input and output filenames, number of pixels per plot, longitude and latitude, date and time of drone flight, the minimum and maximum temperature thresholds in QWaterModel (tmin_threshold, max_threshold, in °C), observed minimum and maximum temperature (Tmin, Tmax, in degrees K), the QWaterModel settings atm_transmissivity, atm_emissivity and surface_emissivity, the measured short wave irradiance (sw_irradiance_measured, in W m-2), the measured relative humidity (RH, in %), along with the mean QWaterModel outputs per plot, i.e. mean latent heat flux (W m-2), mean evaporative fraction and mean evapotranspiration (ET, mm h-1). 2.) The land surface temperature maps of the 56 plots as key input files for QWaterModel. For each plot, there is a .tif image file along with a .xml info file. The files are labeled as 'Plot_[plot nr.].tif'. 3.) The evapotranspiration (ET) maps of the 56 plots as key output files of QWaterModel. For each plot, there is a unix executable file along with a .xml info file. The files are labeled as 'Plot_[plot nr.]_DATTUTDUT'. From these ET maps, the mean ET was calculated for each plot, as presented in the Evapotranspiration_BEE.xlmx results file. ---------- More details on the applied methods can be found in the following publications: 1.) Ellsäßer, F., A. Röll, C. Stiegler, Hendrayanto & D. Hölscher, 2020. Introducing QWaterModel, a QGIS plugin for predicting evapotranspiration from land surface temperatures. Environmental Modelling & Software, 104739. 2.) Ellsäßer, F., A. Röll, J. Ahongshangbam, P.-A. Waite, Hendrayanto, B. Schuldt & D. Hölscher, 2020. Predicting tree sap flux and stomatal conductance from drone-recorded surface temperatures in a mixed agroforestry system – a machine learning approach. Remote Sensing 12, 4070. 3.) Ellsäßer, F., C. Stiegler, A. Röll, T. June, Hendrayanto, A. Knohl & D. Hölscher, 2021. Predicting evapotranspiration from drone-based thermography - a method comparison in a tropical oil palm plantation. Biogeosciences 18, 861-872.