Search

Abundant vegetation species and associated complex forest stand structures in moist tropical regions often create difficulties in accurately classifying land-use and land-cover (LULC) features. This paper examines the value of spectral mixture analysis (SMA) using Landsat Thematic Mapper (TM) data for improving LULC classification accuracy in a moist tropical area in Rondônia, Brazil. Different routines, such as constrained and unconstrained least-squares solutions, different numbers of endmembers, and minimum noise fraction transformation, were examined while implementing the SMA approach. A maximum likelihood classifier was also used to classify fraction images into seven LULC classes: mature forest, intermediate secondary succession, initial secondary succession, pasture, agricultural land, water, and bare land. The results of this study indicate that reducing correlation between image bands and using four endmembers improve classification accuracy. The overall classification accuracy was 86.6% for the seven LULC classes using the best SMA processing routine, which represents very good results for such a complex environment. The overall classification accuracy using a maximum likelihood approach was 81.4%. Another finding is that use of constrained or unconstrained solutions for unmixing the atmospherically corrected or raw Landsat TM images does not have significant influence on LULC classification performances when image endmembers are used in a SMA approach. [ABSTRACT FROM AUTHOR]

High deforestation rates in Amazonia have motivated considerable efforts to monitor forest changes with satellite images, but mapping forest distribution and monitoring change at a regional scale remain a challenge. This article proposes a new approach based on the integrated use of Moderate Resolution Imaging Spectroradiometer (MODIS) and Landsat Thematic Mapper (TM) images to rapidly map forest distribution in Rondônia, Brazil. The TM images are used to differentiate forest and non-forest areas and the MODIS images are used to extract three fraction images (vegetation, shade and soil) with linear spectral mixture analysis (LSMA). A regression model is built to calibrate the MODIS-derived forest results. This approach is applied to the MODIS image in 2004 and is then transferred to other MODIS images. Compared to INPE PRODES (Brazil's Instituto Nacional de Pesquisas Espaciais – Programme for the Estimation of Deforestation in the Brazilian Amazon) data, the errors for total forest area estimates in 2000, 2004 and 2006 are −0.97%, 0.81% and −1.92%, respectively. This research provides a promising approach for mapping fractional forest (proportion of forest cover area in a pixel) distribution at a regional scale. The major advantage is that this procedure can rapidly provide the spatial and temporal patterns of fractional forest cover distribution at a regional scale by the integrated use of MODIS images and a limited number of Landsat images. [ABSTRACT FROM AUTHOR]