Figure 1 - Looking south to the site of Hecla
James B. McGuire
GEO 386G – Final Project
Introduction
The summer of 2001 was spent mapping the
The most outstanding feature when
mapping Hecla is the presence of large dome, just
south of the abandoned buildings seen in Figure 1. On the east slope of
Figure 2 -
GIS
Starting Points
The purpose of this project is to integrate data
collected in the field with readily available digital data for the state of
Figure 3 - DOQs
and topo map with a square drawn around Hecla area
Production
of Digital Data
A digitized geologic map by Ruppel et al. (1993) is available
for the Dillon Quadrangle from the USGS (http://greenwood.cr.usgs.gov/pub/open-file-reports/ofr-97-0738/dillon.tar.Z). This is a 1:250,000 map
that is in a custom projection and does not provide the necessary accuracy for
examining features on a meter-scale.
This map was in ARC/INFO format, and was changed to a shapefile and reprojected into
MSPC. A 1:24000 map by E-an Zen (1988) provides a much better view. A digital image of this map was created with
a scanner, and the Hecla area was cropped out using
Adobe Photoshop. The scanned map was
neither in the right projection, nor was true north known. Using the topo map
as a projected base, “rubber sheeting,” a process whereby an image can be
rotated and stretched to fit points, was
attempted. An extension called ImageWarp 2.0 was downloaded from the ESRI site.
When activated, the extension calls for the user to select at least 4
ground control points (GCPs) to be selected on the
projected image, and the image to be projected.
The confluence of streams, a stream running into a lake, and other
defining geographic features were used to rubber sheet the Zen map TIFF image to
the correct location (Figure 4).
Figure 4 - Zen's map of the Hecla area rubber sheeted to fit Mt. Tahepia DOQs in MSPC
Once the map was rubber sheeted so that it registered
with the rest of the data in MSPC, the problem became digitizing the geologic
data on it. After considering turning
the map TIFF into a grid to digitize the mapped units, it was decided that
digitizing the map by hand in ArcView would be most
effective. A new polygon theme was
created to put digitized Zen geologic units in.
Putting all of the different colored units shown in Figure 4 was fairly
difficult. In ArcView,
polygons can be added to the theme which is being edited by: 1) drawing a whole
polygon, 2) appending a polygon to an existing polygon, or 3) drawing a line
through an existing polygon. When a
complex series of polygons such as those seen above are needed, careful planning
should be done before drawing them, because you can only append a new polygon
to a single polygon that already exists.
If one were to draw the green units at the bottom of Figure 4, and then
the units at the top right and top left, the purple DJ (Devonian
Jefferson Dolomite) couldn’t be drawn in because it would need to be appended
to three other polygons.
Figure 5 - Digitized Zen data
Once
all of the polygons were added and island polygons of intrusive dikes were
subtracted, a polygon theme that showed the extent of the Paleozoic sediments
and Cretaceous igneous rocks existed.
Figure 5 shows digitized Zen data with some of the original TIFF image
still visible behind it.
During the
field season, a series of samples from across the field area were
collected. Using a Magellan handheld GPS
receiver, sample locations were taken in UTM coordinates (NAD27, UTM zone 12)
and were recorded in a field book. The
formation/intrusion unit was also recorded with a description of the sample. The data was entered into a Microsoft Excel
spreadsheet, and save as a .dbf (level IV) file. The samples.dbf
file was added as an Event Theme (with Easting as the x-value and Northing as
the y-value), and projected way off of the map because it was not correctly
projected. This data was converted to a shapefile. With the ArcView Projection Utility (under the File menu), the UTM
NAD27 point data were reprojected to MSPC NAD83. A visual inspection and using the ArcView measurement tool revealed that the point data were
within 10 m (approximate) error of GPS device of where they were supposed to
be. A custom legend was contructed which displayed each sample location
color-referenced to the unit from which it was collected (an attribute field
called “unit”). 3D Analyst was loaded as
an extension, and the Hecla DEM was used as the
surface. The digitized Zen Map was
placed on it, as were the point data.
Production of a 3-D map confirms that units that should be ridge-forming
actually are, and also helps in referencing sample location. Figure 6 shows this 3-D view.
Figure 6 - Zen map with sample point data, looking NNW
Using the Data
Though a map was created in the
field, the AutoCAD digitizing equipment was not available for creation of .dxf files to import into ArcView. A digital version of the Zen map, which has
known inaccuracies, is nonetheless going to be important for later comparisons
with a digitized map in the future. With
my point data, that presumably has correct unit types associated with it, a
first-order test of the accuracy of the Zen map can be done. The goal is to test for each point whether or
not it is contained within the polygon which it should be. In simple terms, the unit indicated for each
point should match the unit of the Zen map polygon that it overlays. To accomplish this, the Spatial Analyst and
Spatial Tools extensions were loaded.
First the attribute table of the Projected Sample Locations theme was
loaded. The “Point” field is selected so
that it is shaded. Next, the attribute
table for the digitized Zen map is selected, and then its “Polygon” field is
selected so that it is shaded. Then
selecting a “Spatial Join” will append all relevant Zen attributes to the
Projected Sample Locations theme attribute table. By then forming a query that asks if the
“Unit” field equally the joined Zen map unit field, a list of samples that were
collected within the unit polygon that they were supposed to be in is generated
(Figure 7).
Figure 7 - Selected points that
fulfill query
First-order
accuracy of the map is done by finding the percentages of these points that
fulfill the above query. Only 22 of 75
samples (29.3%) fall in the units defined by Zen for their geographic
coordinates.
These sample collection data can
also be used in a simpler manner to help define areas which provided especially
fruitful for sample collection. Though
much of the Hecla area is exposed rock, some parts of
the
To
make this data useful in the field, it needs to be superimposed on a map. The Zen Hecla map
is a good first choice. First, the Zen
map is converted to a grid using the “Convert to Grid” command. Then, the Location Density theme is added as
a Brightness Theme in the “Advanced” portion of the Legend Editor. The result is shown in the transition between
Figures 8 and 9.
Figure 8 - Sample Location Density grid
Figure 9 - Zen's map with lighter
areas showing places to collect samples
Conclusion
With adequate data and time, an ArcView GIS can provide a myriad of useful overlays for the
geologist. When my final map is in
digital form, its addition to the GIS will allow for further analysis of
discrepancies between the Zen map and the Dillon Quadrangle map.
References
Ruppel, E.T., O’Niell,
J.M., Lopez, D.A. 1993. Geologic Map of the Dillon 1° X 2°
quadrangle,