The most widely available small scale soil data for the U.S. is the Soil Survey Geographic (SSURGO) database, maintained by the NRCS. SSURGO data are available in Arc Interchange format or as Digital Line Graph (DLG-3) optional file format. Some state or regional NRCS offices may distribute the data in additional formats.

To use SSURGO data in the models described in this tutorial, the vector datasets must be imported into the GIS and converted to raster using the appropriate polygon attribute. Each polygon in the vector data refers to a particular soil classification, which is typically made up of several soil horizons with varying K-factors. For our purposes, it is most appropriate to use the K-factor for the surface horizon.

To generate the raster in ArcView or ArcGIS:

1. Generate a table containing K-factor for only the topmost layer.
   1. Open the table for the layer attributes.
   2. Sort descending by layernum.
   3. Select all layer 1 records and export selection as new "layer1" table.
2. Open the attribute table for the soil coverage and the new layer1 table, select the "musym" field in both and join the layer1 table to the coverage. Note: musym field data in the layer table may contain extra characters identifying the county, in which case a new field would need to be added using string functions to make sure the musym (map unit symbols) match for joining.
3. Select the soil coverage and perform "convert to grid" at a resolution and extent to match the best resolution of your other data inputs. Select the Kfact field to use as the value for the output grid.

To generate the raster in GRASS:

1. m.in.e00 lw_soil.e00
2. v.to.rast lw_soil out=lw_soil.2
3. ls $LOCATION/dig_cats/lw_soil.*

   lw_soil.awch      lw_soil.laydepl   lw_soil.phh
   lw_soil.awcl      lw_soil.layernum  lw_soil.phl
   lw_soil.bdh       lw_soil.musym     lw_soil.poly#
   lw_soil.bdl       lw_soil.omh       lw_soil.surftex
   lw_soil.compname  lw_soil.oml       lw_soil.texture
   lw_soil.kfact     lw_soil.permh     lw_soil.wtdeph
   lw_soil.laydeph   lw_soil.perml     lw_soil.wtdepl
   

4. Create a raster map layer with the names of the soils using the category file lw_soil.musym and a raster map layer for K factor from the lw_soil.kfact file as follows:
   1. g.copy rast=lw_soil.2,lw_soil.musym
   2. cp $LOCATION/dig_cats/lw_soil.musym $LOCATION/cats/
   3. r.mapcalc "lw.kfe00=1.*@lw_soil.kfe00"

Soil data on Internet:

• http://www.nhq.nrcs.usda.gov/land/index/intro.html
• ftp://ftp.ftw.nrcs.usda.gov/pub/ssurgo

R-factor for a watershed or military installation will usually not be a spatial dataset, but rather a constant, as is used in the tutorial examples. When modeling very large basins or regions that contain mountainous areas, as well as plains, it may be beneficial to create a map to represent R and use that in the calculations instead of the constant. R values may be obtained from RUSLE databases, which provide values for major cities such as those listed here, or from the U.S. Department of Agriculture's Agriculture Handbook 703. 

Up
Rainfall