The well permit database was exported from AREV to ASCII text
files by county. This text file included permit no., permit holder,
location in Township/Range/Section/QQ, facility name, permit status,
use code, priority date, yield, well depth, static water depth, main
water bearing zone (mwbz) top, mwbz bottom, well diameter, county,
and water district. These text files were sorted and imported into
an INFO database. Any records where field types didn't match (ie
a character entry such as a person's name into the integer well depth
field) were discarded initially. After importing each county 84,700
unique records existed. Next the record number, township, range,
section, and qq were exported to a text file and run through a
fortran program called TRLL (Township Range to Lat-Long) which was
created by Charles Morgan and Jesse McNellis of USGS in 1965 and
updated by Dennis Grasso in 1988. This routine generates latitude
and longitude values for the points from township/range descriptions.
Using the GENERATE utility within ARC/INFO these records were
converted to a point coverage and the original INFO file was joined
using the common record number. An initial QA/QC process was then
performed that deleted wells that didn't exist spatially within
the county that they were permitted, wells that listed static
water depth as less than 0, wells that listed a use-code as
OIL, and wells with mwbz listed as greater than 3000 feet. With
these routines completed, 74,370 wells remained. The next procedure
removed deep wells that existed in the same quarter-quarter section
as shallower wells. An automated script was used to check through each
qq section marking for deletion all wells except for the shallowest
within that qq leaving 30,987 wells.
In order to not interpolate
across various aquifer boundaries the wells were divided up into
three categories based on geologic age of the aquifer. The three
categories were Tertiary, Quaternary, and Older.
Finally, a spherical Kriging procedure was performed to generate an
estimated surface from the points in each of the three layers
and they were then combined using the geologic age boundaries. The
cell size used was 100 meters.
Other methods of Kriging were evaluated, but spherical Kriging was
chosen based on aesthetic and "believability" reasons by the reviewer.
The cell size of 100 meters was chosen based on the intended scale of
1:100,000. This scale was chosen for the Wyoming Ground-Water
Vulnerability Project of which this layer was a part.