Using the R statistical data analysis language on GRASS 5.0 GIS database files

Many researchers wish to explore and analyse spatial data, but typical software does not readily permit such integration. This paper presents a simple interface between two open-source software systems, the GRASS geographical information system, and the R statistical data analysis language. The platform used here is GNU/Linux, because both systems compile and install cleanly; R runs cleanly in Windows environments as well. The interface allows floating point and category data to be passed both ways for raster map layers and sites files; NULL raster cells in GRASS interchange with NA (not available) values in R. Because both systems are developing rapidly and GRASS database internals change often, the interface uses ASCII transfer via temporary files generated by standard programs. The interface operates by running R from within the GRASS environment, and issues commands to GRASS programs through the R system() function. The accompanying code is constructed as an R package, and contains wrapper functions for R plotting, and for R analytical functions returning gridded output, such as trend surface and kriging prediction, kernel density estimation of point patterns, and bicubic spline interpolation. These typical spatial analytical techniques, also often available in some form in GIS, are amply buttressed in R by a large range of other statistical and graphical functions, giving substantial insight into the data or results being handled. The interface will be extended to vector data, and will be coordinated with other database integration packages in R and GRASS. (C) 2000 Elsevier Science Ltd. All rights reserved.