Environmental data management: The identification of outliers

Increased emphasis on pollution control and abatement has often necessitated development of large scale data bases. While sophisticated techniques have been developed and employed for data storage and manipulation, parallel developments in analyzing the accuracy and reliability of the data have been absent (see, for instance, the broad spectrum of requirements outlined in the Federal Water Pollution Control Act Amendments of 1972).This paper centers on the latter and sets out a procedure for data editing and “outlier” identification based on an application of discriminant analysis. A hypothetical example is included along with some suggested applications.