Downscaled land cover for SSP IAM "marker" scenarios, 2010-2100

This dataset is comprised of data from the Shared Socio-economic Pathways (SSP) database, downscaled using the Downscalr package (available here). The data comprises of global projections for land use modelled by GLOBIOM, covering the period from 2000 to 2100. The projections are based on combinations of three different scenario models – Representative Concentration Pathways (RCP), Shared climate Policy Assumptions (SPA), and Shared Socio-economic Pathways (SSP). The dataset consists of 18 NetCDF files, each representing a different combination of RCP, SPA, and SSP pathways. Each file contains projected data for the entire 100-year period. Each file has the following structure:1. One raster file containing data on land use types (GlOBIOM land use projections)2. One raster file containing information on pixel size (redundant) (pixel_area). This raster file is redundant for analysis purposes. Within the first layer (GLOBIOM land use projections), each raster has 80 bands. Each of these bands represents one type of land use in a given decade. The key for both of these is listed below. Land Classes lc_class ID Land Use Class name Definition Dynamics Initialisation 1 cropland_other cropland area, excluding 2nd. generation bioenergy plantations (but includes 1st generation bioenergy crops); both n-fixing and not; both perennial (e.g., oil palm) and annual can increase from deforestation or conversion of grassland or other natural land; can decrease if not used anymore initialized with GLC2000, further harmonized with SPAM and FAO (crop specific areas); 2 cropland_other cropland dedicated to 2nd generation bioenergy short rotation plantations; perennial cropland can increase from deforestation or conversion of other natural land or grassland; can decrease if not used anymore initialized with GLC2000, further harmonized with SPAM and FAO (crop specific areas) 3 grassland grassland used for feeding livestock, can be both rangeland or pasture, both temporary or permanent grassland can increase from deforestation or conversion of other natural land; can decresae if not used anymore or converted to cropland initialized in year 2000 with GLC2000, but a large share of grassland is not used (and rebalanced to 'other') 4 forest_unmanaged forests areas not managed, can be both primary or secondary (e.g., if used before year 2000), was present in year 2000 and excludes new forest (afforestation) can decrease from conversion to managed forest (as simulated by the G4M model), cropland or grassland initialized in year 2000 with GLC2000, further hamonized with SPAM, FAO and information from G4M model 5 forest_managed forests areas managed (for extractive use or carbon sequestration), includes both forest present in year 2000 and new forest (e.g., afforestation) can decrease from conversion to cropland or grassland, can increase as a result of afforestation of other natural land (only in suitable pixels, as simulated by the G4M model) initialized in year 2000 with GLC2000, further hamonized with SPAM, FAO and information from G4M model 6 restored land that was used as grassland or cropland and set aside for restoration (only from 2020 onwards) only in BIOD scenarios, and cannot decrease (i.e., in each pixel, any increase at any time step is new restoration land) none 7 other other vegetated (primary or secondary non-forest and non-agricultural vegetation, including schrubland, tundra, wetlands), and non-vegetated (bare land, deserts, water, ice or permanent snow) areas can increase as a result cropland or grassland abandonment (in all time steps for NOBIOD scenarios, before 2020 in BIOD scenarios) initialized with GLC2000, further harmonized with SPAM and FAO (crop specific areas); 8 built_up_areas built-up areas static value (does not change over time or scenarios) initialized with GLC2000, further harmonized with SPAM and FAO (crop specific areas); Years Time value Year 0 2010 10 2020 20 2030 30 2040 40 2050 50 2060 60 2070 70 2080 80 2090 90 2100 The spatial dimension is divided into pixels, according to the GLOBIOM's SimU classification system. Within each pixel, a share of each land cover class is given. These pixels are sometimes accumulated into larger combined pixels, where data for land cover is homogenous across a wider area. Some pixels are not simulated as they contain no crop land cover (e.g., glaciers, lakes). The data is spatially explicit at SimU Level, with cell size between 5 arc minutes and 0.5 degrees. The dataset was created as a Fast Track Analysis for the Bending the Curve of Biodiversity trends exercise, on the 11th of February 2018. An overview of the Bending the curve exercises is linked here. The data is intended for research and analysis purposes, under a Creative Commons license.