MapRios

<b>Raster dataset of the following manuscript:</b><b>Looking upstream: </b><b>analyzing</b><b> the protection of </b><b>the drainage area of </b><b>Amazon river</b><b>s</b>.<b>Aim:</b><br>We provide explicit spatial information on accumulated deforestation, mining, and protection across the river network.<b>Input datasets:</b><br>We analyze land use and land cover along Amazonian rivers using several datasets. The adopted Amazon Basin limit is provided by the HydroSHEDS level 2 basin product (Lehner et al., 2008), including the Amazon and Tocantins-Araguaia basins. We used the global HydroSHEDS products at 15 arcsecs spatial resolution, based on elevation data obtained in 2000 by NASA's Shuttle Radar Topography Mission (SRTM). HydroSHEDS provides georeferenced hydrographic information at various scales, including river networks, watershed boundaries, drainage directions, and flow accumulations. The deforestation and mining areas in 2020 were obtained from MapBiomas Amazon Project Collection, which contains annual land use and land cover maps for the region based on automatic classification of satellite imagery. The protected areas were obtained from World Database on Protected Areas (WDPA), updated monthly and managed by the United Nations Environment Programme's World Conservation Monitoring Centre.<br>All non-natural land use and land cover classes were reclassified as deforestation areas, reprojected, and downgraded to the Hydrosheds pixel resolution. For this process, we calculated the fraction of each Hydrosheds pixel covered by the 30 m deforested pixels and multiplied the results by the Hydrosheds pixel areas. The MapBiomas project considers mining as all areas of extraction of minerals with soil exposure without differentiating the type of mining (industrial, artisanal, or illegal). Many overlapping PAs in the WDPA with different categories and designations (national, regional, and international PAs) were maintained in this study.<b>Output dataset:</b><br>From the 15 arc-second (~500 m) flow direction matrix, we calculated the upstream area for each pixel (flow accumulation). The next step was to identify the river network, which is considered the channelized river. For this step, a threshold of 20 km² on the flow accumulation matrix was applied to determine the beginning of the river network. A similar method was used to compute the accumulated area occupied by each land use type upstream of the drainage pixel. The value of each feature (deforested/mining/protected areas) accumulated in a pixel equals the sum of the values of the feature areas in all pixels that drain to it, based on their flow direction information. The final step was determining the percentage between each land use accumulated and the contributing area for each river network pixel.<br>