ETL 1110-1-175
30 Jun 97
Figure 5-3. (Sheet 2 of 4)
example, such a remedy may be entirely appro-
approach may be necessary to avoid an undue
priate; however, some investigators may support
amount of compromise.
the idea that the situation is actually dealing with
e. The restriction of measured data, set B, is
two unique and homogeneous domains. Therefore,
the second alternative considered, distributing the
shown in Figure 5-4b and the basic univariate sta-
kriging process so that each homogeneous domain
tistics are listed in Table 4-1 (bedrock B). The
is addressed independently, becomes more attrac-
restriction exercise resulted in removing 17 meas-
tive. In more complicated applications where a
ured locations and in the truncation of the north-
large number of domains are present, a distributed
eastern part of the area so that the area became
5-10