Your objective in this lab will be to calculate the RKLS factor (part of the RUSLE, see your readings) for an area in Oregon. The dataset available is the Powder River Basin - unfortunately, due to LS processing limitations (time), you will not run the entire watershed.
Copy them to the D:\ drive USING ARC/CATALOG- work here. Remember, ALWAYS use arcCatalog to move, copy, etc GIS files.
Warning - as you work farther along, keep your filenames and folder names to 8 or less characters with NO spaces.
Because processing the DEM - to get the LS factor - will take days, you will first need to clip it to a manageable size (at most one tenth the current size). There are two methods. The first is to build a shapefile, draw a box, convert it to a coverage, and then head into workstation and clip it. I can't get this to work - something is truly goofy. However, method two will work. Bring pow_dem up in an arcmap window. By moving the cursor around, get the coordinates of two opposite corners of the area you want. You will then clip the DEM based upon the min/max x,y values with Workstation - grid.
NOTES on Arc Workstation: This is good ole fashioned command line programming. If you don't know what to type, you're hosed. Some of the basic commands:
Now, open up Arc Workstation, set your workspace to wherever your DEM and new coverage are, go into GRID, and use the gridclip command to reduce your DEM to manageable size. The syntax is: gridclip <ingrid> <outgrid> box <xmin> <ymin> <xmax> <ymax>
Where gridclip is the command, ingrid is your DEM, outgrid is a filename for your new clipped DEM, box is a command, and the numbers are the x,y min/max coordinates of the area you are interested in.
There is no need to clip the R-factor grid - there are no processing limitations with these guys.
Next, we have to rasterize the soils layer. Do this in ArcMap, Spatial Analyst. In the pulldown menu, goto Convert - features to raster. Fill in the appropriate boxes (KFACT, 30m resolution, filepath, etc). Before running this, go into the spatial analyst options and set the extent and mask to your new, small DEM. This will ensure the kfact layer is the same size as the DEM you will be using.
Now, you should have a a small DEM, a small rasterized K-factor layer, and the R-factor layer. For file management reasons, nuke everything else. Using arcCatalog. If you screwed up somewhere, you can always copy the original files and re-clip.
We now need to calculate the LS factor. To do this, head off to my slope page(http://www.onlinegeographer.com/slope/slope.html) and download version 4 for windows of the RUSLE code. Copy this file into your working subdirectory.
Now, go into Arc Workstation and set your workspace. Don't go into GRID! type "&r rusleV4pc.aml" (no quotes - this should be the file you just downloaded) and type in the answers to the questions. Remember, the watershed grid is the same as your DEM. Let it run for as long as it takes!
For more info about the code, read the info on the website. That page will also tell you what all the final output files represent.
AMLs are just text files - as such you can open them in Word, etc to see what they're all about.
You now have R, K, and LS factor raster files. The last step is to multiply them together. Do this in Spatial Analyst. Make sure you set your analysis extent (in Spatial Analyst) to either your K or LS factor file (or small DEM) and your output resolution to 30m. Otherwise, you'll crunch for the entire US at a very low res. Your final output represents physical erosion potential (it excludes the vegetation component). Refer to my paper (one of readings 7) - I did this whole mess for the entire State of Oregon.
As to grading.... Hmmm.... Howsabout:
3 points. Due Friday, 17 April.