T. Hengl1, C. H. Grohmann2, R. S. Bivand3, O. Conrad4, and A. Lobo5
1 IBED University of Amsterdam, UvA B2.34, Nieuwe Achtergracht 166, 1018 WV Amsterdam, The Netherlands
2 Institute of Geosciences, University of São Paulo, Rua do Lago, 562, Cidade Universitária 05508-080 São Paulo – SP, Brazil
3 Norwegian School of Economics and Business Administration, Helleveien 30, 5045 Bergen, Norway
4 University of Hamburg, Institute for Geography, Bundesstraße 55, Raum 810, 20146 Hamburg, Germany
5 Institut de Ciències de la Terra “Jaume Almera” (CSIC), Solé Sabarís s/n E–08028 Barcelona, Spain
Two of the most used open source desktop GIS software for the analysis of DEMs are SAGA (System for Automated Geoscientific Analyses) GIS and GRASS (Geographic Resources Analysis Support System) GIS (Wood, 2008; Steiniger and Bocher, 2009). SAGA has been under development since 2001 at the University of Göttingen (the SAGA development team, has since moved to University of Hamburg), Germany, with aim of simplifying the implementation of new algorithms for spatial data analysis. In 2004, most of SAGA’s source code was published using an Open Source Software license. The functionality of SAGA is described in Böhner et al. (2002) and Böhner et al. (2008); the software design, methods, and usage are explained in detail in Conrad (2007). GRASS GIS, now one of the eight initial Software Projects of the Open Source Geospatial Foundation (OSGeo), is probably the most known open source GIS software in the world. Its functionality and usage are described in detail in Neteler and Mitasova (2008). GRASS itself is a collection of modules (they vary from version to version). Although originally a Linux-based project, the most recent version of GRASS (6.3; development version) is now also available for MS Windows machines. GRASS is a much larger project than SAGA considering the number of developers/institutions involved, although their functionality considering the DEM analysis is about similar. Both SAGA and GRASS are increasingly rich considering the functionality they offer: the latest version of SAGA (2.0.3.) contains 48 libraries with 300 modules; GRASS 6.3 contains over 350 routines. Both in fact provide more functionality for the analysis of DEMs than proprietary low-end products such as the basic installation of ArcGIS 9.2. By linking SAGA/GRASS with R environment for statistical computing, a powerful combination is created that allows fusion of GIS and statistical functionality in the same code (Grohmann, 2004; Brenning, 2008).
In this article we present the results of a comparative analysis of performances of the two GIS software for the analysis of elevation data. We focus on DEM generation, extraction of hydrological features (stream networks), and extraction of gridded DEM derivatives. We will base our comparison on objective and subjective criteria: measures of accuracy, processing speed, but then also on the user’s satisfaction following questionnaires. Our intention is not really to name the winner, but to see what the basic differences are, and to suggest ways to combine the strengths of the two packages.