The spatial resolution of precipitation mapping is limited by the spacing of the point data. This is particularly severe for global mapping which includes large data sparse areas. Global land precipitation maps have been produced using NDVI and elevation information to add spatial detail to point precipitation data. The extreme, 90th, 50th, and 10th percentile values of 14 monthly precipitation statistics have been mapped for all global land areas. In addition, errors for each map grid cell were modeled as a function of a grid cell's distance from the nearest station. The error model was used to compare errors produced by augmenting the station data with NDVI and elevation data verses errors produced using only station data. Augmenting the station data reduced the errors for amount type statistics (e.g., extreme 5-day total) for growing season months and introduced additional errors for number of day type statistics (e.g., 50th percentile of the number of days with precipitation). Consequently the final mapping program augments station data with NDVI and elevation information for amount type statistics and does not augment the station data for number of day type statistics.