Porwal, Alok
The fuzzy gamma operator function (Equation 1) was then used for combining fuzzy membership values and
generating final favourability maps. The algebraic product term in the right hand side of the equation, viz., IT Lx).
i=l
returns a value of 0 for most of the area, because the fold axes map has a value of “no data” for all grids except for the
grids which fall on the buffers of fold axes. This results in a combined fuzzy membership value of zero on
0.20 multiplication for all these areas. This map was therefore omitted, and the algebraic product term was computed using
0.12 four maps only. Several values of y were tested, and it was found that large values of y give excessive areas of high
0.16 favourability (eg. Fig. 3C, y = 1). Figs. 3A and 3B show the favourability maps generated using y = 0 and y = 0.500,
0.08 respectively.
0.12
0.20 A B : C
0.20
912 0.00 - 0.10 3 50 000-019 2 33 000-020 0 0.0
0.08 0.10-0.19 Levent 0192037 0 0 030-040 0 0.0
on 0.19 - 0.29 10 167 0.37-0.56 0 0 0.40 - 0.60 0 0.0
0 0.29 - 0.38 3 50 0.56-0.74 1 1.7 0.60 - 0.80 0 0.0
0.12 0.38- 0.48 43 717 0.74 -0.93 57 93.0 0.80 - 1.00 60 100.0
Table 2 Distribution of deposits with respect to the combined fuzzy membership classes derived using different values
of the fuzzy gamma operator (A. y — 0, B. y 2 0.500, and C. y 1)
0.12
0.06
0.24
0.24 B C
0.36
0.39
0.24
0.18
0.18
0.36
0.30
0.30
0.18
024 [[] Study Area [7] Study Area
0.48 Basemetals Basemetals [Shy tres
0.48 Fuzzy Membership | uzzy Membership - Basemetals
F1] 0-0.1 [7] 0 - 0.19 Fuzzy Membership
048 m. E: es
0.48 038.048 | 074 - 0.53 : | 05-08
e Fig. 3 Favourability maps generated using different values of fuzzy gamma operator (A. 'y 2 0, B. y= 0.500, C. y=1)
0.48
0.15 4 DISCUSSION
0.27
0.39 The fuzzy model at y= 0.500 predicts 95% of the known mineral deposits in high favourability areas (Table 2B,
0.21 Fig. 3B). At y= 1 ( the fuzzy algebraic sum operator), the model predicts all mineral deposits in high favourability area,
0.42 but it also predicts unrealistically large areas of high favourability (Fig. 2C). On the other hand, the fuzzy algebraic
0.45 product operator (y= 0), predicts most of the known mineral deposits in low to moderate favourability areas (Fig. 2A).
0.51 In addition, several areas with high favourability, which contain no known basemetal occurrence, are mapped.
0.48 Especially notable is a moderately-high favourability belt with combined fuzzy membership value of 0.56-0.74 in the
0.54 north-western part of the study area, which is predicted by the model when a Y value of 0.5 or larger is applied. This belt
0.24 is occupied by meta-carbonates and meta-basites of the Delhi supergroup and contains no significant mineral
033 occurrence. Similar rocks host important copper and polymetallic mineralisation outside the study area. Work is
continuing to verify whether these areas genuinely represent favourable areas warranting further exploration, or whether
the modelling techniques used need further refinement.
International Archives of Photogrammetry and Remote Sensing. Vol. XXXIII, Part B7. Amsterdam 2000. 1183