Spatial data infrastructures for the Amazon: a first step towards a global forest information system

The preservation of the world’s rain forests is one of the major environmental challenges of our generation. Rain forests are home to a large portion of the world’s biodiversity, and play a major role in climate regulation and in the hydrological cycle of the Earth. Despite their biological richness and ecological importance, most of the world’s rain forests are under increasing pressure of deforestation caused by human actions. Deforestation is particularly intense in countries with large-scale agricultural production, such as Brazil and Indonesia. In the Brazilian Amazonia, which covers an area of 4 million km, 17% of the original forest has been cleared, according to the Brazilian deforestation monitoring system, developed by the Brazilian National Institute for Space Research (INPE). Science Magazine called the Brazilian system “the envy of the world”, but warned that, in spite of recent improvements in forest monitoring, uncertainties are still large (Kintisch 2007). Because of data and capacity gaps, current information on global deforestation is extremely patchy. There is much disagreement in the scientific community about the magnitude and extent of deforestation worldwide. As an example, the FAO 2005 Forest Resource Assessment (FAO 2006) points out that deforestation in Brazilian Amazonia from 2001-2005 equals 31,000 km/yr. Using data imagery from a different satellite, Hansen et al. (2008) estimated that forest clearings for Brazil in the same period were 26,000 km/yr. INPE ́s detailed estimates are 22,000 km/yr for the same period (Valeriano, Monteiro et al. 2008). If we compare them to INPE’s data, the FAO 2005 report and Hansen et al. overestimated deforestation in Amazonia by 63% and 18% respectively. Furthermore, making such estimates is a large effort, and comprehensive results take long to be obtained and validated. Information science and technology has a key role in the improvement of the monitoring efforts, but the efficient collection, organization, and distribution of large volumes of data still represent a great challenge for information science and technology.