Multivariate Statistical Techniques for Assessing Regional Ground-Water Quality and Quantity

Regional assessment of the chemical character of ground water requires that unbiased estimates of the chemical species in ground water be obtained. With recent regulations regarding ground-water contamination and disposal of toxic wastes, it is necessary that the methods for assessment of man-induced versus natural chemical character are not biased by experimental design, and that estimates of chemical species present in ground water are not influenced to a large degree by temporal variation and sample collection strategies. It is a requirement as well that populations which are compared are homogeneous in terms of statistical parameters. Implementing statistical design techniques early in the water-quality assessments can lead to maximizing information while minimizing the number of samples needed. The ultimate objective of this project is to apply and extend statistical design theory to better understand and define ground-water quality parameters. It is the objective also to understand the relations between chemical variables in ground- water systems and to decompose the chemical variability within the system according to contributing sources. The understanding of chemical variability within the ground-water system will provide a basis for application of methodology to other areas with similar hydrogeology.

For information on additional projects in the National Research Program, see Indexes to NRP projects and bibliographies