Impacts of surface bounadry layer turbulence on surface temperature measurements at different spatial resolutions

Surface temperature displays temporal fluctuations driven by turbulence intensity of both surface (SBL) and planetary (PBL) boundary layers. An experiment conceived to evaluate the error induced by these fluctuations on instantaneous satellite measurements at different spatial resolutions is described. It is based on the reconstruction of surface temperature time series from sequences of thermal images acquired using a TIR camera aboard an helicopter in stationary flight over 5 different surface types. It is shown that the spatial averaging performed over the pixel allows one to reduce the contribution of the high frequency SBL turbulence to the error on Ts measurements. This error rapidly decreases with pixel size, and is found to be around ±0.5°C at 50m resolution for the cases we studied. Unfortunately the experimental set up revealed not as well suited to evaluate the contribution of low frequency PBL turbulence to Ts error, and our recommendations are given for future experiments.