Independent verification and benchmark testing of the UNSAT-H computer code, Version 2. 0

Independent testing of the UNSAT-H computer code, Version 2.0, was conducted to establish confidence that the code is ready for general use in performance assessment applications. Verification and benchmark test problems were used to check the correctness of the FORTRAN coding, computational efficiency and accuracy of the numerical algorithm, and code, capability to simulate diverse hydrologic conditions. This testing was performed using a structured and quantitative evaluation protocol. The protocol consisted of: blind testing, independent applications, maintaining test equivalence and use of graduated test cases. Graphical comparisons and calculation of the relative root mean square (RRMS) values were used as indicators of accuracy and consistency levels. Four specific ranges of RRMS values were chosen for in judging the quality of the comparison. Four verification test problems were used to check the computational accuracy of UNSAT-H in solving the uncoupled fluid flow and heat transport equations. Five benchmark test problems, ranging in complexity, were used to check the code`s simulation capability. Some of the benchmark test cases include comparisons with laboratory and field data. The primary findings of this independent testing is that the UNSAT-H is fully operationaL In general, the test results showed that computer code produced unsaturated flow simulationsmore » with excellent stability, reasonable accuracy, and acceptable speed. This report describes the technical basis, approach, and results of the independent testing. A number of future refinements to the UNSAT-H code are recommended that would improve: computational speed and accuracy, code usability and code portability. Aspects of the code that warrant further testing are outlined.« less