What raster-based GIS systems really work?

Question

Raster GISes like GRASS, ArcGIS/Spatial Analyst, and Idrisi can perform a rich set of data processing and analytical procedures loosely known as "map algebra." In today's computing environment it is becoming commonplace to maintain rasters of 100,000,000 cells or more in many different formats and to demand relatively complex calculations like viewsheds, watersheds, and terrain identification, as well as image processing capabilities.

It seems that many open source, free, and inexpensive solutions are out there. But which ones really hold up in practice? That is, which ones can handle large grids efficiently, can easily get data in and out, are reasonably bug free, and offer a full complement of analytical procedures? What are the pitfalls or hidden limitations that you don't find out until you have invested lots of time in learning these systems? (This last question is one that is not easily answered with Web searching and where I am hopeful respondents can offer valuable advice.)

I am especially interested in solutions that can both integrate well with and compete with the popular (but expensive) commercial systems (which means Windows compatibility is important).

Answer 1

I can't speak to SAGA or some of the other systems, but I have used GRASS extensively, including for an global-scale analysis of ~720M cells which required robust implementations of raster algebra and complex terrain operations. (As an aside, with the discontinuation of ArcInfo, GRASS is arguably the longest continuously developed GIS).

GRASS data and tools can be readily accessed through QGIS, which provides a nice ArcView GUI analog. QGIS itself is gaining nice raster analysis capabilities, such as the GDALTools plugin, but these are quite new and lack the maturity and depth of GRASS itself.

Another prospect is the raster package for R: R has a large userbase, the source of methods are easy to access, and it includes the cutting edge of many statistical techniques. However, it lacks in image processing tools and may not be sufficient for the kinds of tasks you're interested in.

Lastly, GDAL forms a solid basis of many, if not most modern GIS systems, and has very fast implementations of many common map algebra operations. It can be used through its Python interface or through direct C/C++ in times when the 'layer' abstraction proves insufficient.

Answer 2

We use a mixture - from Spatial Analyst, SAGA, Ermapper, a bit of GRASS, but in the end we tend to go to Geosoft - though that is because we do a lot of geophysical enhancement processing. Spatial Analyst/ArcGIS is good because you can easily extend functionality through the toolboxes/geoprocessing but we've found the actual Spatial Analyst processing routines are often not the best. Of late we've built toolboxes to access SAGA modules from within ArcGIS so we can continue to use the functionality without having to import/export - the toolbox looks after all that as necessary. We'll probably look at doing a similar thing to access GRASS functionality as well

Answer 3

You can now work and do map algebra with rasters of almost unlimited size in a spatial database with PostGIS. I personnaly work with SRTM and climate data at the scale of Canada. I can do intersection between raster and vector layers in a very fast and transparent way. I can also use a whole set of map algebra functions.

Answer 4

Manifold with Surface Tools is very good in terms of importing formats and handling large rasters, analysis can be done directly between matching rasters or with implicit reprojection. There's GPU support for a number of raster functions, and there's strong support for automation with a variety of scripting languages and SQL. The price is good at a few hundred US.

General doc for Surface Tools:

http://www.georeference.org/doc/surface_tools.htm

Here's the current list of functions available to the Surface Transform dialog, which accepts custom expressions to perform calculations between multiple rasters:

http://www.georeference.org/doc/transform_dialog_functions_and_operators.htm

One pitfall is that exports of "surfaces" (rasters) cannot be done to GeoTIFF (images can). I usually export to SDTS and convert that to GeoTIFF with GDAL. The mapping of coordinate systems from Manifold's (own) support and other systems like GDAL's family is not perfect, but problems are pretty rare.

Answer 5

I have heard of quiet a few people using SAGA. But I personally have very little experience with it.

http://www.saga-gis.org/en/index.html

Answer 6

For this article "Estimating daily Land Surface Temperatures in mountainous environments by reconstructed MODIS LST data (full text PDF) I have processed 11,000 MODIS LST images easily in GRASS GIS, in a parallelized manner on our cluster. Great fun since it just works.

Answer 7

we use SAGA for monitoring data from dose rate and gamma spectrometry measurements (airborne or ground, natural background, old mine dumps etc.) processing. I has many useful modules for us and we enjoy it much.

PS: as the SAGA map output has its limitations, for more advanced maps we combine it with Quantum GIS.

Answer 8

Speak for myself, I am biased in this case. But I mostly use IDRISI for raster GIS. Largely because IDRISI offers most comprehensive tools for raster analysis if you compare it with other GIS software. From various classification and prediction statistical models to watershed and cost distance analysis, it has pretty much everything we need for daily raster analysis. It also has an extension for ArcGIS. It has improved its ability to handle large data. However, none of the GIS software really can calculate a 1000000 by 1000000 cost distance in a minute yet.