Abstract
Palavras-chave: deteccao de incendios florestais, relatorio de ocorrencia de incendios, sensoriamento remoto. AbstRAct - The remote sensing detection of hotspots is an important part of the monitoring system of forest fires for two major Brazilian federal agencies elated to preventing and combating these occurrences, IBAMA and ICMBio. To evaluate the effectiveness and accuracy of data from the Vegetation Fires Database of the National Institute of Space Research, we used the following two data sources: data recorded by the Center for the Prevention and Combat of Fire in Itatiaia National Park from 2008 to 2010 and data about areas of large fires that occurred in the Park in 2001, 2007 and 2010, obtained by interpretation of satellite images from LANDSAT 5 and CBERS 2. We analyzed information from the AQUA, TERRA, ERS-2, MGS- 02 and NOAA and GOES satellites series for the aforementioned area and period to study a total of 92 hotspots, whose data were compared to those from 101 records of fire occurrence. Ninety-six percent of hotspot detections were omitted in relation to fires reported. The TERRA and AQUA satellites, which carry MODIS sensors, showed the best results in the detection of fires and in the accuracy of the hotspot's position in relation to the burned area.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have