COVAR is a program for generating two-dimensional fields of autocorrelated parameters which are normally or log-normally distributed (e.g., hydraulic conductivity). The program uses a technique based on (Choleski) matrix decomposition. The generated parameter field represents a major requirement in the stochastic analysis via Monte-Carlo techniques. The program requires as input a list of x,y coordinates for which the parameters need to be generated, and statistical parameters describing the distribution of the hydrologic parameter of interest. The program facilitates a maximum of 20x20 nodes. For larger problems, the program can be easily modified and recompiled. The program can also be modified easily to be used as a subroutine in a Monte Carlo type two-dimensional stochastic flow or mass transport model.

Program data input is interactive. Output is a text file with tabular results. COVAR, a public domain research code, is distributed on a DOS-formatted disk containing source code, executable image, and an example data set. The user's manual consists of installation instructions, a description of the theory and computer code, and an example application.

SYSTEM REQUIREMENTS

Intel 80i86 based computer, 640 Kb RAM, math co-processor, about 500 Kb free disk space, and CGA graphics board/monitor.

Authors: S.A. Williams and A.I. El-Kadi (IGWMC).