Assessment of local thermal heterogeneity of flux sites during the LITFASS-2003-experiment from airborne infrared imagery

The LITFASS-2003 experiment was aimed at determining area-averaged turbulent fluxes of sensible and latent heat over a heterogeneous landscape at the meso-gamma scale. Techniques employed included long-path scintillometry, airborne measurements and the land-use weighted averaging of fluxes measured with eddy-covariance systems over all relevant types of land use. These local flux measurements at field scale were performed at 14 different sites over patches of homogeneous land use, the site selection was based on visual inspection of the sites with respect to surface characteristics, obstacles, fetch conditions and terrain levelness. Surface temperature mapping across the study area was performed on selected days of the experiment using an infrared thermometer aboard the Helipod, a sensor package carried by a helicopter, and from infrared imagery by a Tornado reconnaissance aircraft of the German air force.

The infrared images from the Tornado flights were calibrated with the help of synchronous Helipod flight tracks. The images, available with a spatial resolution of about 1-2 m, were then analysed with respect to the local surface temperature distribution in the footprint area around the different surface flux stations. Mean surface temperatures differed up to about 8 K for different footprint sectors. Within a certain footprint sector, we found a standard deviation of surface temperature up to about 6 K due to differences in local soil moisture and vegetation density. These temperature anomalies could be identified consistently in the images from three flights performed on different days and at different times of the day (with correlation coefficients between the temperature patterns of the different days up to r2 = 0.92).

Non-closure of the local surface energy balance was related to the surface temperature heterogeneity in the footprint area of the flux sites. It could be shown, that the size of the energy balance residuum is increased for sectors with a large standard deviation of surface temperature. We should remark that most of the obvious structures in the surface temperature field could not or just hardly be recognised in the corresponding visible site and area photographs. Surface temperature mapping should therefore be an advisable step of the site selection procedure for micrometeorological flux measurements.