Cerrado vegetation study using optical and radar remote sensing: two Brazilian case studies

Abstract

The amount of phytomass is a crucial parameter in ecology as a whole. Conventional methods to estimate phytomass parameters are often prohibitive in terms of time, environment, and manpower. Seasonality in Brazilian savanna physiognomies is marked by green leaves lost during the dry season and regrowth during the wet season. Branches and trunks remain the same through the seasons. The amounts of the phytomass components can be estimated by nondestructive methods using remote sensing. This paper presents case studies where the abilities to predict vegetation variation are tested using optical and radar images. The foliar component leaf area index (LAI) obtained in the field is related to the normalized difference vegetation index (NDVI) obtained from satellite images, and both methods present a strong relationship with green leaves biomass. Japanese Earth Resources Satellite (JERS-1) images are used to estimate the aboveground woody biomass. Campo cerrado physiognomy showed the highest seasonal variation in NDVI, and cerradão the lowest variation in NDVI. During August, the lower the NDVI values, the lower the solar albedo and the higher the photosynthetically active radiation (PAR) albedo. During November, the higher the NDVI values, the lower the PAR and the higher the solar albedo. The same proportion in seasonal variation was observed with LAI values. A significant equation is proposed to estimate trunk and branch biomass using allometric parameters and JERS-1 backscattering.