Quick Look
An easy-to-use tool that delineates valley bottoms based on the topographic position index (TPI).
- Difficulty: 1
- Technical level: 1
- Expense: 1
- Scale: Typically 10 m or 30 m
- Accuracy: 2
(Ratings are given on a 1-5 scale. Click on any rating for an explanation)
Method Overview
The Land Facet Corridor Designer (LFCD) (Jenness, Brost, and Beier, 2011) is an extension for ArcGIS, developed at Jenness Enterprises, to design wildlife corridors by identifying continuous landscape units as defined by topographic and soil characteristics. The tool has been repurposed as an easy-to-use tool for delineating valley bottoms, as an aid to riparian mapping. The LFCD is integrated into ESRI’s ArcDesktop as a toolbar.
Figure 1: TPI (right) is calculated from a DEM (left) (from Jenness, Brost, and Beier, 2011)
The LFCD performs topographic analysis using the topographic position index (TPI), which is the difference between each pixel’s elevation and the average elevation of the surrounding pixels. TPI is calculated from a DEM producing a new raster of TPI values (Figure 1). A negative TPI value indicates that the pixel is at a lower elevation than the surrounding cells (e.g., in a valley), while a positive TPI indicates that the surrounding cells are lower (e.g., on a ridge). A TPI value near zero indicates either a hillside slope or a flat area, and these can be distinguished using the slope at that point (i.e., a slope near zero suggests a flat area, while a slope above some threshold indicates a hillside slope).
By varying the TPI neighborhood size (the number of surrounding cells) the user can control the scale-or resolution-of the modeled topographic position. A large TPI neighborhood tends to minimize smaller features and capture larger ones. A small TPI neighborhood emphasizes finer-scale features. This allows the user to tailor the products to the scale of the investigation.
Similar Methods
Workflow
Determine TPI Neighborhood Size
The first step is for the user to determine the appropriate TPI neighborhood size for the study area. This decision depends on the scale of the project. The TPI should be calculated with several neighborhood sizes and compared to a hillshade to determine the best neighborhood size for the project.
Slope Position
Once the appropriate TPI neighborhood size has been selected, the TPI can be thresholded into classes. The user examines the TPI relative to a hillshade to determine the appropriate break points between classes. The software divides the landscape into a minimum of three classes using the user-defiended thresholds: canyon (valley), slopes, and ridges. The addition of a slope-angle criterion allows the “slopes” class to be subdivided into mid-slope versus flat areas. Additional TPI thresholds can distinguish between canyon versus lower slopes, or between ridges versus upper slopes.
The valley bottoms can be extracted from the classified TPI product and incorporated into a riparian mapping project.
The tool produces a raster dataset delineating valley bottoms, slopes, and ridges.
Riparian Example
None published
References
Jenness, J.; Brost, B.; Beier, P. 2011. Land Facet Corridor Designer: extension for ArcGIS. Jenness Enterprises. Available at: http://www.jennessent.com/arcgis/land_facets.htm.
Limitations
It can be difficult to establish an ideal TPI neighborhood size that captures all of the valley bottoms.
Data Inputs
- Digital elevation model (DEM)
Software/Hardware Requirements
This method requires the Land Facet Corridor Designer, ArcDesktop, and the Spatial Analyst extension. The method algorithm is reasonably efficient; however, processing time will vary with study area and TPI neighborhood size. The extension can be downloaded from: http://www.jennessent.com/arcgis/land_facets.htm
Additional Resources
http://www.jennessent.com/arcgis/land_facets.htm
http://www.jennessent.com/downloads/Land_Facet_Tools.pdf
http://www.jennessent.com/downloads/TPI-poster-TNC_18x22.pdf