Geoscience Reference
In-Depth Information
Fig. 4.7 (continued)
step, crosses a predetermined level Q . Setting Q
0.5 for the present example
resulted in the selection of 24 explanatory variables. Q -values are small (
<
0.05) for
the first nine variables that were selected.
Finally, Fig. 4.9 shows how well smaller sets of known copper deposits in parts
of the study area perform in predicting copper potential in the entire study area.
Nineteen of the 27 (10 km
10 km) cells with known copper deposits occurred in
western Quebec, mainly in the Noranda and Val d'Or mining districts. The contour
map of Fig. 4.9b is based on these 19 control cells only. There is strong similarity
between Fig. 4.9a, b that was based on all 27 control cells. Using the eight control
cells in Ontario (western part of study area) produced the pattern of Fig. 4.9c , which
underestimates Fig. 4.9a .
Search WWH ::




Custom Search