Take the 2-minute tour ×
Geographic Information Systems Stack Exchange is a question and answer site for cartographers, geographers and GIS professionals. It's 100% free, no registration required.

I'm working with NOAA's PRECipitation REConstruction over Land dataset, and am looking for an efficient way to sample the data by pixels on NASA's MODIS grid, using open source tools. The PRECL data uses 0.5 degree blocks, so I'll have quite a few repeated values.

I'm currently using a mapping generated by ArcGIS that specifies the array index within the PRECL data for every MODIS pixel. This works (and it's fast, as I'm just doing a series of array lookups), but it'd be nice to get away from that extra data set, especially if there were some way to do the transformation in code, without Arc.

Thanks in advance, all!

share|improve this question
did you notice this similar question? Does it help your situation? gis.stackexchange.com/questions/5041/… –  Brad Nesom Jan 29 '11 at 3:37
IF not. I use auotdesk raster design which does contain a function to resample/transform by pixel. It has several methods. Arcgis as well has an extension called spatial analyst. If you have access to that it will resample to a new dataset and you can specify from several methods. –  Brad Nesom Jan 29 '11 at 3:41
add comment

1 Answer 1

up vote 3 down vote accepted

R is an excellent programming language designed for data manipulation and statistical computation. There is a very nice package called sp that adds good support for handling spatial data. The spsample function can pull a statistical sample from spatial data using 7 different methods. The sampling area may be controlled by specifying a bounding box and/or a grid cell size.

Combine with the rgdal package for geospatial IO and you should be ready to rock.

share|improve this answer
I find the maptools R package very useful (it depends on the sp package). You can read a raster directly in using it (I haven't used sp so perhaps you can also do it there also) –  djq Feb 2 '11 at 19:46
Hey, I'll take it. R is great, though I'm working on moving to Incanter, using Clojure. –  Sam Ritchie Feb 4 '11 at 0:54
add comment

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.