Page 153 - Geochemical Anomaly and Mineral Prospectivity Mapping in GIS
P. 153

154                                                             Chapter 6



















             Fig. 6-4. Log-log plots of density (d) of points representing occurrences of epithermal Au deposits
             in the Aroroy district (Philippines) as a function of distance or radius  r from  each point. The
             individual straight lines fitted through the linear portions of plots satisfy the power-law relation in
             equation (6.3).


             where d is the density of points in polygonal patterns defined by circles of radius r from
             such points, C is a constant and D r is the radial-density fractal dimension of the pattern
             defined by the points. Feder (1988) calls D r the cluster dimension of fractal point pattern.
             The radial-density relation has been applied by Carlson (1991), Agterberg (1993) and
             Wei and  Pengda (2002) to characterise spatial distributions  of mineral  deposit
             occurrences. For the application of the radial-density relation in a GIS, a small pixel size
             is used to represent each point (i.e., mineral deposit occurrence) as just one pixel. The
             radial-density is then derived as the ratio of the number of points to the total number of
             pixels common to overlapping circles of radius r from certain points plus the number of
             pixels  in non-overlapping circles  of radius  r from certain  points. If the spatial
             distribution of the points is fractal, then the radial-density of the points should decrease,
             following a power-law function, with increasing radius from the points.
                The point pattern of the occurrences of epithermal Au deposits in the Aroroy district
             has two  radial-density fractal dimensions (Fig.  6-4). The straight line fit through the
             plots when r ≤ 2.8 km has a slope of -1.1141, whereas the straight line fit through the
             plots when  r  > 2.8 km has  a slope  of -0.4099.  These results indicate that the spatial
             distribution of the occurrences of epithermal Au deposits in the Aroroy district is non-
             random but fractal. A plausible interpretation of the results shown in Fig. 6-4, together
             with the results shown in Tables 6-I and 6-II, is that, in the Aroroy district, there are
             fractal fracture systems that controlled the epithermal  mineralisations over scales
             (lengths  or widths) ranging  from about  0.5  km to at  most 2.8 km and that there are
             fractal hydrothermal systems that controlled the epithermal mineralisations over scales
             (lengths, widths or diameters) ranging from about 2.8 km to about 7 km.
                The estimated fractal  dimensions of the  pattern  of epithermal Au deposits in the
             Aroroy district are somewhat inconsistent [0.2891 (D b) versus 1.1141 (D r)] at scales less
             2.8 km and [1.0905 (D b) versus 0.4099 (D r)] at scales greater than 2.8 km, although the
   148   149   150   151   152   153   154   155   156   157   158