Introduction to GRASS

Credit:

These materials are adapted from Professor Larry Huggins ASM 521 materials.

Goal:

To understand the fundamental character of the GRASS GIS and its user interface.

Scope:

This material is intended as background for hands-on use of GRASS.

Terms:

New technical terminology: geo-referenced data, grid-cell data (also known as raster data), spatial data, tabular data, and vector data.

GRASS, A Specific GIS:

The advantages of GRASS are its rich array of data analysis functions, its low cost (public domain software), and the growing libraries of available data sets. This lesson is not a tutorial on using GRASS. Instead, it provides a conceptual overview of a user's interaction with GRASS. The hypothesis is that remembering commands will be easier if you first have a general sense of what's happening.

For many people, the concept of "a database" connotes a single large file with a highly structured internal organization. That is not an appropriate metaphor for GRASS. Rather, a GRASS database is a hierarchal set of subdirectories, each with numerous files that are created automatically as one requests each new analyses.

Typically, a GRASS user will make personal analyses of a region(s) for which basic data already exists. GRASS facilitates this operational style by saving within your home directory personal variations of basic data. Thus, you may make modifications to the GIS without corrupting the public library files used only to initialize your personal setup. (By the way, the rumor that GIS developers are really disk salespersons in disguise is untrue, despite contrary evidence of a GIS's impact on disk space needs!)

GRASS Infrastructure:

GRASS stores spatial data in three formats: raster data, tabular and vector data. While it has the capability to analyze only raster data, routines are included to convert stored vector data into a raster format for analyses. Retrieval/reporting capabilities include:

At start-up GRASS accesses a three-tiered directory structure referred to in its internal documentation as:

database:: The name for the sub-directory in which all personal GRASS database sub-directories and files will be stored. The default is a sub-directory named "data" within your $HOME directory.
location:: You may create and use several independent GIS databases. Each must be identified by a unique "location name" which will cause a separate sub-directory by that name to be created within your "database" directory. If you wish to start from existing library files (you do!), names will be available from the GIS site administrator and/or by substituting the word "list" to display local libraries.
mapset:: Multiple-scenarios of a location are easily maintained. This is accomplished with parallel GIS files under a separate directory hierarchy for each uniquely named mapset, e.g. conventional-till and no-till might be two GIS databases (mapsets) that differ only by the tillage management used on targeted fields.

GRASS attempts to make its detailed hierarchal directory and file structure transparent to the user. While definitely efficient for the experienced user, it can make learning the system seem counter-intuitive.

The GRASS Fundamentals document shows how to initiate a GRASS session.

Credits: *GRASS is a creation of the U.S. Army Construction Engineering Research Laboratory (CERL) at Champaign, IL. It is public domain software running under a Unix(Tm) operating system with an active enhancement program. Many graphics in this module are from CERL publications or have been created by Jean Messersmith and Bernie Engel.

How can this presentation be improved?