SENSITIVITY OF ERS-1 AND JERS-1 RADAR DATA TO BIOMASS AND STAND STRUCTURE IN ALASKAN BOREAL FOREST

Thirty-two boreal forest sites were identified and sampled in the cent ral region of Alaska to evaluate the sensitivity of the C-band ERS-1 a nd the L-band JERS-1 radar platforms to site biophysical properties. A growing body of research has shown a significant radar backscatter re sponse to biomass in a variety of forest systems. Alaskan boreal fores ts may be well suited to radar remote sensing. The sites selected repr esent black spruce (Picea mariana) and white spruce (Picea glauca) sta nds in a post-fire chronosequence. Black spruce biomass ranged from le ss than 1 kg/m(2) to 5.6 kg/m(2) and white spruce from 8.8 to 21.5 kg/ m(2). Results indicate both ERS-1 and JERS-1 backscatter is responsive to biomass, density, and height, though other factors, principally su rface moisture conditions, are often a stronger influence. Sensitivity to forest biomass and structure appears greatest when surface moistur e conditions are minimized as a factor. Biomass correlations with the radar backscatter were strongest in the late winter Imagery when all s ites had a snow cover, and late summer when the surface is most dry. E RS-1 data may be more sensitive to surface moisture conditions than th e JERS-1 data due to the shorter wavelength of the C-band sensor, thou gh this is inconclusive because of limited JERS-1 L-band data for comp arison. Also, though the ERS-1 platform has proved to provide a very s table signal, results must be interpreted with caution as the dynamic range for our study sites is often less than 4 dB, and the uncertainty of the backscatter estimate is +/- 1.5 dB.