Making GIS and R play nicely - some tips?
Is the state of the art approach to use GRASS a la Bivand and Neteler's paper (http://agec144.agecon.uiuc.edu/csiss/Rgeo/)? Is it better to dump the grids to an imagine file (or bil) and read them with rgdal? I work mostly in Windows (b/c of ESRI - damn them!) but switch gears onto Linux when the task requires it. I have never used GRASS. Show me the way!
Hello Disclaimer - I've never used GRASS, and I haven't used ESRI much. You might consider using Manifold GIS: http://www.manifold.net/ Much more Windows (and user) friendly IMO, and has many options for input and output for most formats, as well as generic text and binary. (Give it a few years and Manifold will have taken over much of the GIS market.) You could output your grids to binary files (or ASCII), and with some modification of my simple script write the projection metadata to XML - then script-import these to Manifold. This scripting could be done all within R if that was desirable using RDCOMClient I have some R routines that export with Manifold-friendly metadata in XML, very rudimentary but might give you some ideas. Passing grid (matrix) data from R to Manifold: http://www.georeference.org/Forums/forums/thread-view.asp?tid=789&posts=9 Running R from Manifold: http://www.georeference.org/Forums/forums/thread-view.asp?tid=650&posts=8 Running Manifold from R: http://www.georeference.org/Forums/forums/thread-view.asp?tid=876&posts=9 Cheers, Mike. ############################################### Michael Sumner - PhD. candidate Maths and Physics (ACE CRC & IASOS) and Zoology (AWRU) University of Tasmania Private Bag 77, Hobart, Tas 7001, Australia Phone: (03) 6226 1752 http://www.acecrc.org.au/