Page 45 - Geochemical Anomaly and Mineral Prospectivity Mapping in GIS
P. 45

Spatial Data Models, Management and Operations                        41































           Fig. 2-14. Transformation of data points into a surface grid. (A) Control data points; each point is
           characterised by its  x-coordinate (east-west or across page width),  y-coordinate (north-south or
           down page length), and z-coordinate (value beside a point). Point data are identified by numbering
           them sequentially, as read by software, from 1 to i. Thus, a control point i has coordinates x i  and y i
           and a value z i . (B) A regular grid of nodes is superimposed on the map. These grid nodes are also
           numbered sequentially from 1 to k. Each grid node has coordinates x k  and y k , and a value to be
           estimated z k . (C) The value z k  at a grid node k is estimated from n control points found within a
           search neighbourhood, of specified area of  influence, centred  at  k. (D) Completed grid with
           estimated values of z k .


           ‘weighted moving average’  methods, inverse distance method and  kriging  have  been
           usually applied to model geochemical surfaces. Point-to-surface transformations of
           geochemical  data by spatial interpolation  are applicable in fractal analysis of
           geochemical anomalies (see Chapter 4).
              Methods of point-to-surface  transformations are also applicable to line-to-surface
           transformations of linear geo-objects representing continuous variables (e.g., isolines of
           elevation). That means all points or samples of points along linear geo-objects are used
           in triangulation  or gridding methods.  In contrast,  methods of point-to-area
           transformations are not readily applicable to line-to-area transformations, particularly if
           linear  geo-objects represent qualitative variables. For  linear geo-objects  representing
           qualitative variables (e.g., curvi-linear structures, lithologic contacts, etc.), the idea of
           line-to-area transformation is to generate zones of proximity to linear features through an
           operation known as buffering or dilation, which depends on distance calculations. Points
           or polygons can also be buffered (i.e., point-to-area or area-to-area transformations) if
   40   41   42   43   44   45   46   47   48   49   50