ADDRESSING UNCERTAINTY IN GEOLOGIC DATA AND MAPS

Uncertainty within geologic data and geologic maps is rarely analyzed or discussed. This is undoubtedly due, in part, to the difficulty of analyzing large sets of paper files. The increasing use of computers for storing and serving large geologic data sets has made it easier to systematically analyze these data and any geologic maps and models created from them.

General error theory can be a useful method for characterizing the uncertainty within geologic data. For most geologic data, errors can be due to inaccuracies in: record keeping, lithologic description, identification of lithologic contact, lithologic generalization, identification of location, evaluation of elevation, and correlation. Error theory was used for evaluating the total uncertainty of lithologic data from water well records stored in the central database at the Illinois State Geological Survey.

While analysis by error theory can be used to evaluate the uncertainty within the data used in mapping, other methods can be used to more fully evaluate the uncertainty within specific maps. The area of influence method was developed to use three map characteristics for estimating uncertainty within maps of ore deposits. This method uses the spatial distribution of data points, the probability of misidentification of the targeted unit, and the size of the targeted geologic feature to calculate the probability that additional, unidentified targets can exist. I will use the area of influence method, with examples from an area in Illinois, to quantify the uncertainty surrounding both the identification of channelized glacio-fluvial sand deposits, and the identification of geomorphic features within the bedrock surface.

Insight gained from the use of error theory and the area of influence method will be used to describe the uncertainty included in geologic maps. By considering the mapping objectives with these two uncertainty assessments, I will address appropriate uses, use limitations, and several scale dependent issues of various geologic maps and models.