Microwave observations of boreal forests in the NOPEX area of Sweden and a comparison with observations of a temperate plantation in the United Kingdom

Abstract

Within the NOPEX programme, fully polarimetric dual frequency Synthetic Aperture Radar (SAR) data were acquired over the Siggefora and Norunda forest test sites in Sweden during 1994 and 1995. These data were taken by the SAR system (EMISAR) of the Technical University of Denmark as part of an aircraft programme sponsored by the European Space Agency. Simultaneously with these overflights, ground campaigns were conducted in 64 sample plots to measure the standing biomass density of 20 forest compartments. The microwave backscatter observed by EMISAR was found to decrease from a constant saturation value only at the very lowest biomass densities-such behaviour was expected at C and but not at L-band where a more gradual decrease was anticipated starting at medium biomass density levels. A much larger ground dataset was therefore taken from estimates of forest biomass density made by the Swedish landowners, and another microwave dataset was taken from the SAR system on board the Japanese Earth Resources Satellite (JERS-1), both of which confirmed these unexpected findings. The EMISAR results in Sweden have been compared with earlier data acquired by the SAR system (AIRSAR) of the Jet Propulsion Laboratory over the temperate plantation around Thetford in the United Kingdom. This is an area having some similarity to the Swedish sites in its biophysical characteristics but showing a quite different relationship between microwave response and biomass density. The capability of the fully polarimetric EMISAR and AIRSAR data to provide information on the dominant, scattering mechanism was employed to help investigate these contrasting results. The different behaviour of the Swedish sites may arise from the greater surface roughness or the greater moisture content of the boreal forest in this part of Scandinavia. The accuracy with which biophysical parameters can be retrieved from SAR measurements of forests depends considerably upon vegetation structure and ground conditions. The implications of these results for global inventory and for monitoring of forests using satellite microwave remote sensing are discussed briefly.