Focus on deforestation: zooming in on hot spots in highly fragmented ecosystems in Costa Rica

Most estimates of deforestation are derived for areas that are not intrinsically related to the causes of deforestation, making it hard to link such estimates with strategies for improved land management. In this paper a method is presented to identify local areas of high rates of deforestation within larger areas, using remote sensing and geographic information systems (GIS). These deforestation hot spots are identified using a local analysis approach, considering for every location only land cover changes taking place in the immediate vicinity, and they thus reflect local deforestation. This method has potential application in improved land policy development, land management and it could supplement land use change modeling at the landscape level. Specifically, the method could be used to undertake a risk assessment of the effectiveness of ecological corridors. The method was applied to the province of Limón in Costa Rica, where commercial agriculture has caused large scale deforesta- tion. The deforestation hot spots in the period 1986–1997 that were identified with this method represent a large proportion of the total deforestation having taken place in the province, and for each of the hot spots specific drivers of deforestation could be identified.