Prediction of Occurrence of Discrete Events - Geomathematics: Theoretical Foundations, Applications and Future Developments - page 148

Granite Contact Binary Pattern
Mapid : Mgr8

Legend

less favourable
more favourable (1.00 KM)

20 KM

Weight for feature present w+ = 0.3150
Weight for feature absent w− = −0.0562

Fig. 5.6 Binary pattern for proximity to Devonian granite contact within Goldenville Formation
(Source: Bonham-Carter et al. 1988 , Fig. 3d)

Table 5.1 Weights, contrasts and their standard deviations for predictor maps
W +

(W ) W (W ) C

(C) C/ (C)

σ σ σ σ

Goldenville Fm 0.3085 0.1280 1.4689 0.4484 1.7774 0.4663 3.8117

Anticline axes 0.5452 0.1443 0.7735 0.2370 1.3187 0.2775 4.7521

Au, biogeochem. 0.9045 0.2100 0.2812 0.1521 1.1856 0.2593 4.5725

Lake sed. signature 1.0047 0.3263 0.1037 0.1327 1.1084 0.3523 3.1462

Golden-Hal contact 0.3683 0.1744 0.2685 0.1730 0.6368 0.2457 2.5918

Granite contact 0.3419 0.2932 0.0562 0.1351 0.3981 0.3228 1.2332

0.0185 0.0247

NW lineaments 0.2453 0.0062 0.1417 0.2833 0.0872

1.2406 1.4610

Halifax Fm. 0.5793 0.1204 0.1257 0.5928 2.4646

Devonian granite 1.7360 0.7086 0.1528 0.1248 1.8888 0.7195 2.6253

Source: Bonham-Carter et al. ( 1990 , Table 1)
The last column is the “studentized” value of C , for testing the hypothesis that C ¼ 0. Values
greater than 1.96 indicate that the hypothesis can be rejected at α ¼ 0.05. Note that the hypothesis
cannot be rejected for the granite contact and NW lineaments

is approximately satisfied because nowhere does the observed curve break the 95 %
confidence envelope surrounding the predicted curve. An interesting result is that if
the gold occurrences with known production are plotted on a graph of posterior
probability versus cumulative area, there appears to be a positive correlation