If you encounter problems with the file provided on this page, please contact Soil Survey at 406-587-6818.
The information on this page is also available in Acrobat Reader format. Once you have saved the file to a local drive, open Acrobat Reader then open the saved file on your local drive.
Introduction (PDF; 525 KB)
ArcGIS 9.0 Introduction to Spatial Analyst for Initial and Update/Maintenance Soil Surveys
Updated November 2005
- To be able to use datasets derived from elevation data for soil survey applications.
- To provide soil scientists with tools to help view, describe, and analyze the relationship between landform/landform positions and the components found on those positions.
Overview of Spatial Analyst
For soil survey activities, Spatial Analyst provides a mechanism to analyze the relationship between DEMs (rasters) and Soil Polygons (vectors).
Rasters are cells that are arranged in rows and columns.
- DEMs are rasters that have a specific elevation value in each cell (cell sizes are typically 10 or 30 meters).
- DOQs are rasters that have a specific pixel value in each cell (cell sizes are typically 1 or 2 meters).
Vectors represent geographic data, such as points, lines, and polygons.
The differences between rasters and vectors:
- Rasters have continuous data based on cells without defined boundaries. Each cell can have a different value.
- Vectors have defined boundaries and are uniform within the boundary.
How are DEMs created?
- Hypsography (developed from contour lines); most DEMs are created by this method.
- (Note: Hypsography method has a contour line bias that can create unintended results.)
- Survey points from survey-grade GPS
- Satellite Imagery
DEMs are a useful and powerful tool, but you need to understand their limitations.
- The “M” in DEM stands for Model.
- DEM Limitations
- Random errors (from misplaced contour lines)
- Contour line bias (limitation of Hypsography)
- Scale error (landform features are finer than the resolution of a DEM; e.g., short steep slopes
Know the limits of your data.
Check metadata about scale and resolution.
Make sure 10-meter DEM data is not resampled 30-meter data.