LAUSR

ABSTRACT Some key geologic data can only be obtained from wells. This article provides original technology concerning the geological mapping validation of complex anomalous zones based on information obtained from irregular or sparse drilling. Only complex non‐unique cases of geologic characteristics delineation (i.e. characterized by multiple solutions) are considered here. Representative groups of multiple mapping solutions (i.e. essential variations in anomaly zones delineation in terms of size and location) are analysed in order to optimize and verify the mapping parameters. Minor interpolation discrepancies not influencing the geologic interpretation, for example, resulting from specificities of different interpolation techniques or insignificant mapping procedure corrections, are neglected. The mapping error can be quantitatively assessed by the cross validation error analysis. Maps can be optimized by the proper choice of the interpolation algorithm and data filtering. Using trial‐and‐error schema, interpolation and filtering options are tested and mapping errors are compared. Single assessments or selective tests of the mapping accuracy often cannot lead to the best solution. The optimization procedure can be used for mapping verification, accuracy assessment and mapping parameter optimization. It may also be used to find an optimal balance between the mapping accuracy and the resolution by interpolants (i.e. initial well data) test processing. The interpolation algorithms used here are most technologically fit for a check comparison – not as a substitution of the existing mapping technologies. After positive mapping validation, a common procedure of interpolation can be started. Franke's theoretical work on scattered data interpolation and Krilov's smart averaging interpolation tests were used here to substantiate the well data mapping validation technology. The authors tried not to delve deep into the theory and interpolation testing details using due references and concentrating on the technology procedure and algorithm description.