Effects of environmental conditions on boreal forest classification and biomass estimates with SAR

During both SIR-C/X-SAR missions (April and October of 1991), periods of freezing and above freezing temperatures were experienced within the southern study area (SSA) of the boreal ecosystem-atmosphere study (BOREAS). This paper discusses the observed effects of different environmental conditions on forest classification and biomass estimates derived from the SAR data of a portion of this Canadian boreal forest test site. The SIR-C backscatter was found to be lowest when the trees and background were frozen, in agreement with other studies. Simple linear regression models relating measured above-ground woody biomass to SIR-C backscatter were not the same for each date, but statistical tests showed that the slopes were not significantly different, suggesting that a simple correction by an offset is possible. The classification results indicated that forest type classifications are more variable than the biomass estimates due to temperature conditions.