Site Productivity Grid

Site Grid Update 2016/05/11

We have results and now have to figure out distribution…

Ready for some teaser pictures.  Here you go the long awaited site grid for the western United States.

So let’s define the project that has been undertaken.  The following are excerpts from the returned metadata of the contractor that provided the GIS work.  (Data Directions, LLC.  in Eugene, Oregon, USA)

5 Acre Point Grid – Site Index Data w/FIPS

Creator:
   Data Directions LLC
   www.datadirections.biz

Date: June 23, 2014, revised July 29, 2014, August 30, 2014, October 31, 2014, November 21, 2014, December 27, 2014, January 19, 2015, August 17, 2015.

Data Description:

The data is provided as an ArcGIS Shapefile of points spaced on a 5 acre grid. The layer is populated with attributes (see below) useful in predicting forest productivity. The layer is trimmed to remove data points which are not within forested regions. Data is provided by State/UTM Zone. In some cases data is further divided into North/Central/South regions to prevent exceeding Shape file limitations. Data is not contiguous where data sets meet and may contain some overlap. Note: Hawaii was not trimmed to forested regions but provided in its entirety.

Development Procedures:

Beginning with 30 Meter USGS Digital Elevation Models (DEMs), the DEM files are mosaicked to create a single model within a State/UTM region. This model is then Clipped to a 1:100,000 scale State layer to reduce excess processing. Slope and Aspect models are generated from the DEM and the 30 Meter DEM is then resampled to a 5 acre DEM.

The 5 acre DEM is next clipped to a layer which represents “US Forested Lands”. The forested lands layer was derived from a US Forest Service 1000 Meter Raster Layer representing US Forested Lands. This layer was buffered by 20,000 meters. The resulting clipped DEM is converted to points that include the elevation attribute in meters.

This 5-acre point grid layer is next processed and evaluated against the Slope and Aspect models as well as precipitation data by month. The precipitation data was obtained from the 800 Meter Raster data detailing normal monthly totals from 1981-2010 available through the PRISM climate group, at Oregon State University (www.prism.oregonstate.edu). The precipitation data for Hawaii was obtained from a 250 Meter Raster data set detailing normal monthly totals from 1978-2007 available through the University of Hawaii (http://rainfall.geography.hawaii.edu). The precipitation data for Alaska** was a 2Km Raster dataset acquired through the NOAA National Centers for Environmental Information, Gridded Monthly Temperature and Precipitation Data for Alaska, British Columbia and Yukon Study. The individual,monthly data files for the most recent 30 years (1980 – 2009) were averaged. Attribute fields are added for slope, aspect and monthly precipitation where a point overlays a specific grid cell.

The resulting point layer is next re-projected to the Geographic Coordinate system Longitude/Latitude using the WGS84 datum. Attributes are added representing Longitude and Latitude. Next using a nationwide county layer the county Federal Information Processing Standard code (FIPS) is added based on point location. Finally, the completed point layer is clipped to a 1:100,000 scale State layer.

Attributes (Name, Type, Description):

  • GRID_CODE; Double; Elevation in Meters from DEM.
  • Slope; Float; % Slope at point.
  • Aspect; Float; Aspect in degrees at point.
  • Precip_Jan; Long Integer; Jan. Normal Precipitation (1981-2010) millimeters X 100*
  • Precip_Feb; Long Integer; Feb. Normal Precipitation (1981-2010) millimeters X 100*
  • Precip_Mar; Long Integer; Mar. Normal Precipitation (1981-2010) millimeters X 100*
  • Precip_Apr; Long Integer; Apr. Normal Precipitation (1981-2010) millimeters X 100*
  • Precip_May; Long Integer; May Normal Precipitation (1981-2010) millimeters X 100*
  • Precip_Jun; Long Integer; June Normal Precipitation (1981-2010) millimeters X 100*
  • Precip_Jul; Long Integer; July Normal Precipitation (1981-2010) millimeters X 100*
  • Precip_Aug; Long Integer; Aug. Normal Precipitation (1981-2010) millimeters X 100*
  • Precip_Sep; Long Integer; Sep. Normal Precipitation (1981-2010) millimeters X 100*
  • Precip_Oct; Long Integer; Oct. Normal Precipitation (1981-2010) millimeters X 100*
  • Precip_Nov; Long Integer; Nov. Normal Precipitation (1981-2010) millimeters X 100*
  • Precip_Dec; Long Integer; Dec. Normal Precipitation (1981-2010) millimeters X 100*
  • POINT_X; Double; Longitude, Decimal Degrees, WGS84 datum.
  • POINT_Y; Double; Latitude, Decimal Degrees, WGS84 datum.
  • FIPS_Code; Character(5); FIPS code representing US county containing point.

Note: if both Slope and Aspect values are zero or less the point location is on a water surface.

* Precipitation data for Hawaii is provided in millimeters rounded to the nearest millimeter.

** Due to the lack of a complete coverage of USGS 30M DEM data for the State of Alaska, the coverage of the dataset is not complete and contains gaps in nearly every Alaska UTM Zone.

After the completion of this gargantuan task and error checking was performed it was then necessary to process the data to develop a level of site productivity.  This first round of processing listed here is using the Forest Projection and Planning System (FPS) developed by The Forest Biometrics Research Institute (FBRI) based in Portland, Oregon, USA.  The data from the above GIS was fed into FPS and the Site Grid calculations were then performed using the “Regional model” option in the Update Physical Site routine.  The output images above are in a white to green scale.  Where pure white are areas that were not deemed to be forest land and the darkest green are the most productive using the FPS 10 Meter site method.  no direct scale is provided in the images because no field checking has been done.  But the range is between 0 and 3.1 meters per decade of growth for the second log (second 33 foot log) of the tree.

Google Earth overlays are available here: