This data set contains images of fractional cover estimates of photosynthetic vegetation (PV) canopy, nonphotosynthetic vegetation (NPV), and exposed soils (S) derived from Landsat images (30-m resolution) obtained for two ranches in the Brazilian Amazon from 1996 to 2002. The Fazenda Vitoria ranch is located in eastern Para near the city of Paragominas and is a mosaic of primary forest, logged forest, secondary forest, and pasture with moderately dissected topography. The Fazenda Nova Vida ranch is located in the state of Rondonia in western Amazonia and is a mosaic of primary forest, logged forest, and pastures. For Fazenda Vitoria, two dry-season Landsat images were obtained, subset, and analyzed. For Nova Vida three dry-season images and one end-of-wet-season image were obtained, subset, and analyzed. Spectral mixture analysis, which decomposes individual satellite pixels into constituent cover fractions of surface materials, was used with a general probabilistic modeling approach to derive subpixel cover fractions of PV, NPV, and S. There are six GeoTIFF (.tif) files with this data set.

This data set provides active fire locations and estimates of annual fire frequencies for South America from 2000-2007. Data from the Moderate Resolution Imaging Spectroradiometer (MODIS) sensors aboard the Terra (2000-2007) and Aqua (2003-2007) satellite platforms were analyzed to determine spatial and temporal patterns in satellite fire detections. The analysis considered a high-confidence subset of all MODIS fire detections to reduce the influence of false fire detections over small forest clearings in Amazonia (Schroeder et al., 2008). The number of unique days on which the active fire detections were recorded within a 1 km radius was estimated from the subset of active fire detections and the ArcGIS neighborhood variety algorithm. There are 14 data files with this data set: 7 GeoTIFF (.tif) files of fire frequency at MODIS 250 m resolution, where each grid cell value represents the number of days in that year on which active fires were detected, and 7 shape files of active fire locations for the years 2001-2007.

This data set is a five-class land cover for Southern Para for the years 1984 (Landsat MSS), 1988 (Landsat TM), 1996, and 2003 (Landsat ETM+). The final classification shows five classes derived using visual comparison (Water, Clouds/Shadow, Forest, Not Forest, Background). These data were used in 2007 to illustrate the nature of deforestation in Southern Para, Brazil over the past twenty years (Simmons et al. 2007). There are four annual GeoTIFF files distributed with this data set. Each GeoTIFF file and accompanying *.tfw file have been compressed into a single *.zip file.

This data set provides (1) areal estimates of deforestation events (>25 ha) that were identified from 2001-2004 in Mato Grosso by the Brazilian Institute for Space Research (INPE) as part of the Program for the Estimation of Deforestation in the Brazilian Amazon (PRODES) and (2) the classification of the post-deforestation land use as either cropland, cattle pasture, or not in production (deforested areas that were never fully cleared or returned immediately to secondary forest) in the years after the large deforestation events from 2002-2005. Data are provided in ESRI shapefile format. There are five compressed (*.zip) data files with this data set. Each shapefile represents one year of post-deforestation land use. Land use in the years following deforestation was estimated using annual time series of MODIS NDVI (normalized difference vegetation index) and EVI (enhanced vegetation index). Metrics of vegetation phenology derived annual time series of MODIS NDVI and EVI data were analyzed using a decision-tree classifier to characterize the major cover type in each area of new deforestation. Post-deforestation land use for each large deforestation event was classified based on the classification of MODIS phenology metrics for all years following deforestation during 2002-2005.

This data set provides the results of analyses of Landsat Enhanced Thematic Mapper Plus (ETM+) images for selective logging activity in the Brazilian states of Para, Mato Gross, Rondonia, Roraima, and Acre over the years 1999 through 2001. Images were analyzed using the Carnegie Landsat Analysis System (CLAS) to detect and to quantify the amount of damage due to selective logging in the major timber-production states of the Brazilian Amazon. This approach provided automated image analysis using atmospheric modeling for detection of forest canopy openings, surface debris, and bare soil exposed by forest disturbances; and pattern-recognition techniques. CLAS provides detailed measurements of forest-canopy damage at a spatial resolution of 30 x 30m. Fifteen GeoTiff format files are included -- one for each of the three years from 1999-2001 for each of the five states. Each GeoTiff is a single band image where each pixel represents if logging activity was or was not detected. A zero (0) value indicates that no logging was detected, while a value of one (1) indicates that damage from logging was detected. The 15 GeoTiff (*.tif) files have been compressed into one *.zip file.

This data set provides the results of the two modeled scenarios for future patterns of deforestation across the Amazon Basin from 2002 to 2050. This larger defined Amazon Basin (PanAmazon area) includes the Amazon River watershed, the Legal Amazon in Brazil, and the Guiana region. The model SimAmazonia was used to simulate monthly deforestation in the Amazon Basin from 2002 to 2050 for two scenarios: (1) a "Business-as-Usual" scenario, which considered the deforestation trends across the basin and projected the rates by using historical images and their variations from 1997 to 2002 and then added to that the effect of paving a set of major roads, and (2) a "Governance" scenario, that also considered the current deforestation trends, but assumed a 50% limit imposed for deforested land within each basin's subregion, and that existing and proposed Protected Areas (PAs), play a decisive role in limiting deforestation as well (Soares et al., 2006).The provided data products include one GeoTiff (*.tif) for each year (2002 to 2050) for both model scenarios for a total of 98 files. The files have been compressed in two *.zip files, one for each model scenario. There is also one comma-delimited file that contains the model input data derived from satellite deforestation maps.

This data set includes classified land cover transition maps at 30-m resolution derived from Landsat TM, MSS, ETM+ imagery and aerial photos of Altamira, Santarem, and Ponta de Pedras, in the state of Para, Brazil. The Landsat images were classified into several types of land use (e.g., forest, secondary succession, pasture, annual crops, perennial crops, and water) and subjected to change detection analysis to create transition matrices of land cover change. Dates of acquired images represent the most cloud-free image retrievals from 1970-2001 for each site and are therefore not continuous. There are 3 GeoTIFF files (.tif) with this data set.

This data set reports observations of fires in the vicinity of Maraba, Para, Brazil, from November 3-5th, 2001, and in Mato Grosso, Brazil, between Cuiaba and Alta Floresta, for July 12-15th, 2002. These ground-based data were collected by visual inspection from roads primarily during daylight hours. Data include fire position and time, estimates of fire size, and type of vegetation burned. There is one comma-delimited ASCII file with this data set.

This data set provides hot pixel data, as an indicator of fires that were detected by the GOES-8 satellite for the state of Acre, Brazil. Image data were collected for extended periods over the course of 3 years (1998, 2000 and 2001). Data were filtered to select only pixels identified and processed by the GOES-8 Automated Biomass Burning Algorithm (ABBA), where estimates of sub-pixel fire characteristics including size and temperature were able to be determined. There are three comma-delimited ASCII data files with this data set.

This data set provides the results of controlled burns conducted to assess the flammability of mature forests on the Catuaba Experimental Farm of the Federal University of Acre - Rio Branco, Acre, Brazil. Controlled burns were conducted in 1998, and the rate of fire spread was calculated based on the duration of the fire and the measured extent of the burned area. Environmental variables measured included type of forest, canopy openness, leaf area index, number of days without rainfall, precipitation, height of litter, litter humidity, brushwood humidity, amount of water in the ground, air temperature, and relative humidity. Results from 50 fires set in 1998 are reported. There is one comma-delimited data file with this data set.These data are part of a larger study reported in the thesis by Elsa Renee Huamon Mendoza, Susceptibility of primary forest to fire in 1998 and 1999: A case study in Acre, south-eastern Amazonia, Brazil. The thesis, in Portuguese, is included as a companion file with this data set.