Page 97 - Geochemical Anomaly and Mineral Prospectivity Mapping in GIS
P. 97

96                                                              Chapter 4























             Fig. 4-8. Spatial distributions of the classified interpolated soil Fe data (Fig. 4-2, uppermost left)
             based on three threshold values (1.6% Fe, 7.2% Fe, 8.6%  Fe) defined in  the log-log plot
             representing relationships between areas and soil Fe concentrations (Fig. 4-7).



             Attribute table operation for area calculations

                For every raster map of geo-objects (see Chapter 2), such as a map of a discretised
             geochemical surface, there is an associated table. The areas (usually expressed in number
             of pixels or boxes) of polygonal geo-objects, such as areas enclosed within two contours
             (i.e., class limits) of uni-element concentrations, are  recorded automatically in the
             associated attribute table during the discretisation process. For example in Fig. 4-10, the
             second column (NPIX_CL) contains the number of pixels (or boxes) representing each
             class of uni-element concentrations. The  other columns in the table are created  or
             calculated.  The third column  (cl  min), containing the minimum values  (or lower
             limits) of each class, is created based on the first column. The fourth to last columns are
             calculated via arithmetic operations  using the values in the  NPIX_CL column. The
             values in the fourth column (npix cum) are derived by calculating the cumulative sum
             of the number of pixels in the NPIX_CL column successively from the lowest to the
             highest class. The values in fifth column (npix tot) are derived by aggregating the
             total number of pixels in the NPIX_CL column. The total number of pixels is recorded
             repeatedly in all cells of the npix_tot column because arithmetic operations on values
             in at least two columns  are executed  by rows. The values in the sixth column
             (npix_above_cl), which refer to cumulative sums of pixels in classes higher than a
             certain class,  are  derived  by subtracting  the values in  the fourth column from the
             corresponding values in  the fifth column [i.e., (npix_above_cl)=(npix_tot)–
             (npix cum)]. The values in the last column (npix_equal_above_cl_min), which
             refer to cumulative sums of pixels of uni-element concentrations equal to or greater than
   92   93   94   95   96   97   98   99   100   101   102