A digital land cover map of South America has been produced using remotely sensed satellite data acquired between 1995 and the year 2000. The mapping scale is defined by the 1 km spatial resolution of the map grid‐cell. In order to realize the product, different sources of satellite data were used, each source providing either a particular parameter of land cover characteristic required by the legend, or mapping a particular land cover class. The map legend is designed both to fit requirements for regional climate modelling and for studies on land cover change. The legend is also compatible with a wider, global, land cover mapping exercise, which seeks to characterize the world's land surface for the year 2000. As a first step, the humid forest domain has been validated using a sample of high‐resolution satellite images. The map demonstrates both the major incursions of agriculture into the remaining forest domains and the extensive areas of agriculture, which now dominate South America's grasslands.

No abstract

Mapping and monitoring forest extent is a common requirement of regional forest inventories and public land natural resource management, including in Australia. The state of Victoria, Australia, has approximately 7.2 million hectares of mostly forested public land, comprising ecosystems that present a diverse range of forest structures, composition and condition. In this paper, we evaluate the performance of the Random Forest (RF) classifier, an ensemble learning algorithm that has recently shown promise using multi-spectral satellite sensor imagery for large area feature classification. The RF algorithm was applied using selected Landsat Thematic Mapper (TM) imagery metrics and auxiliary terrain and climatic variables, while the reference data was manually extracted from systematically distributed plots of sample aerial photography and used for training (75%) and accuracy (25%) assessment. The RF algorithm yielded an overall accuracy of 96% and a Kappa statistic of 0.91 (confidence interval (CI) 0.909-0.919) for the forest/non-forest classification model, given a Kappa maximised binary threshold value of 0.5. The area under the receiver operating characteristic plot produced a score of 0.91, also indicating high model performance. The framework described in this study contributes to the operational deployment of a robust, but affordable, program, able to collate and process OPEN ACCESSRemote Sens. 2013, 5 2839 large volumes of multi-sourced data using open-source software for the production of consistent and accurate forest cover maps across the full spectrum of Victorian sclerophyll forest types.

No abstract

No abstract