Normalized back scattering radar cross section of tropical rain forest in Rondonia, northern Brazil

Abstract

Taking an advantage of Synthetic Aperture Radar (SAR) on board Japanese Earth Resources Satellite-1 (JERS-1) in capability of observing land surfaces through clouds, an attempt is made to extract deforested areas of tropical rain forests in Rondonia, Brazil. During the study the features of the normalized back scattering radar cross section σ ° are clarified. It was found that dense forests are characterized with the high values while sparse forests, bushes, range lands are characterized with the low values of σ ° depending on the density and physical characteristics of scatterers. In average following values are found: primary forests −6 ∼ −7 dB, secondary forests −7.5 ∼ −8.5 dB, pastures −10 ∼ −13 dB, rivers −15 ∼ −17 dB. It was also found that there is a seasonal effect on the values of σ ° reflecting the variation of the surface condition during the rainy and dry seasons.