Forest discrimination with multipolarization imaging radar

Abstract

The relations between polarization signatures and biophysical characteristics through a range of different forest environments were investigated using airborne synthetic-aperture (SAR) images acquired at L-band on March 1, 1984 in South Carolina. SAR data acquired in four linear polarization states with 10-m spatial resolution were encoded as color composite images and compared to US Forest Service forest stand data. The most useful correlative forest data were stand basal area, forest age, site condition index, and forest management type. It is found that the multipolarization images discriminate variation in tree density or difference in the amount of understory, but no evidence has been found for discrimination between evergreen and deciduous forest types.