Preliminary assessment of ALOS/PALSAR images for the international roughness index estimation

This article evaluates the relationship between SAR backscattering coefficients and the International Roughness Index (IRI). We used full-polarized ALOS/PALSAR amplitude images and field data gathered using laser sensors on-board of vehicles. We also evaluated LANDSAT-5/TM images for comparison. The study area is located in Brazil´s Federal District. Simple regression was used for quantitative analysis employing Pearson´s correlation coefficients (r) and determination coefficients (r²). The HH polarized image yielded an r value of 0.5040 and a r² value of 0.2541. The HV polarized image obtained an r value of 0.4824 and a r² value of 0.2327 while the VV polarized image reached an r value of 0.4382 and a r² value of 0.1920. All r² values were considered significant. Those results show that there is an important relationship between SAR backscattering and IRI. Considering that, new researches will be conducted using high spatial resolution SAR sensors and different wavelengths.