README  -- CONTEMPORARY CROP TYPE MAPS

In response to requests from the scientific community we have produced contemporary maps of four major crop types that are consistent with the fractions of cropland area in each gridcell in our 2005 maps. We have named this release "croptypes.rc1" and it can be downloaded from our website at: http://luh.unh.edu. After unpacking the file you should find four grids of data (C3_annuals.txt, C3_N_fixing.txt, C3_perennials.txt, and C4_annuals.txt), each giving the fraction of the 2005 cropland area in each half-degree gridcell occupied by each of the four crop types in 2005), along with some figures and related information.

Our crop type maps are derived from the maps of harvested area of 175 different crops in: Monfreda et al. (2008), "Farming the planet: 2. Geographic distribution of crop areas, yields, physiological types, and net primary production in the year 2000", Global Biogeochemical Cycles, Vol.22, GB1022, doi:10.1029/2007GB002947. We have modified this data in the following ways, in order to make it consistent with our 2005 cropland map:

1) We aggregated the harvested fraction of each gridcell for these 175 crop types into harvested fraction of each gridcell for four major types: C3-annuals, C3-N-fixing, C3-perennials, and C4-annuals. Of the C3-perennials, many would be considered "woody". The Excel spreadsheet shows which of the original 175 croptypes were mapped to each of the four new croptypes. In addition to aggregating the various crop-types into four major types, the data was also aggregated from the original 5 minute resolution to the half-degree resolution used in our products.
2) After aggregation the 4 classes were normalised so that the sum of their fractions added to 1 (which had not previously been the case due to some multiple cropping). 
3) For gridcells that had non-zero cropland area in our 2005 map, but no corresponding crop-type fractions, we used the crop-type fraction information from the closest possible gridcell. A map showing the gridcells that required using this "nearest neighbor" data is included. The total area of cropland that required using nearest neighbor crop-type information is approximately 240,000 km^2 which is around 1.5% of the global cropland area (and impacts around 15% of cropland-containing gridcells).

Included in our process was the inherent assumption that the fraction of a gridcell that was harvsted for a crop type (i.e. the Mondreda et al. data) was roughly correlated with the fraction of the total cropland area that was occupied by that crop type. Hence the fractional information contained within the four crop-type grids can be interpreted as the fraction of each gridcell's cropland area occupied by each of the four major crop-types. To compute the fraction of each *gridcell* occupied by each of the four major crop-types (rather than the fraction of the cropland area), multiply the crop-type fractions by the 2005 cropland fraction map from our products. 

For further questions about the crop type maps, please contact us at:
Louise Parsons Chini: lchini@umd.edu, +1-301-405-4050
George Hurtt: gchurtt@umd.edu, +1-301-405-8541
Steve Frolking: steve.frolking@unh.edu, +1-603-862-0244