Before any information was obtained, however, a model had to be chosen. After examining the two major pieces of software available for modeling the Green River basin by the interested parties (State Engineers' Office and Wyoming Water Development Office), the decision was made to use the Wyoming State Engineers' Office's version of WIRSOS. The choice of this model was based on its ability to handle a large amount of data in the form of water right diversion inputs as well as its more precise handling of them. Unfortunately, the treatment of reservoirs is rudimentary in WIRSOS compared to the Bureau of Reclamation's model, HYDROSS.
After the decision to use WIRSOS, the development of the necessary data for the model occurred. Throughout the data input stage, assumptions based upon the opinions of experienced individuals and actual studies involving parts of the system assisted in the resolution of several key problems. An initial model was then developed that, permitted further refinement. Starting with this basic model, parts of the individual model were altered to permit the calibration of the system. The changes in the model varied from simply altering derived runoff data to adding efficiency diversions and deleting non-contributing runoff areas from the system. One of the largest changes, however, occurred in the reservoir operations. Since the percentages of water released and stored differed throughout the entire run-time of the model, few of the values could act as constants. This resulted in several different input files that forced the necessity of separate runs and then compilation of their output. Even with these measures, one year of data required deletion from the model due to inaccurate yields based on the eight percent error rule.
Although this model does provide for an accurate interpretation of the river system, years that have a Flaming Gorge Reservoir inlet runoff below 800,000 acre-feet tend to be overestimated. This results from large conveyance losses in the river system that do not follow the general trends used in the model for other years without adversely affecting these years with flows greater than 800,1000 acre-feet. For this reason, year one of the model should not be used to predict or test any results for the entire basin. In contrast, most individual areas of the model accurately simulate low flow years for their separate regions. These localities can be determined by looking at the check stations in Appendix A and determining when the percentage error for year one meet the calibration requirements established.
The only other major issue in this model pertains to the description of the reservoir parameters. Even though the basic physical data represents those of record, the operating methods forced the making of some extreme assumptions involving release schedules and related diversions since no data was available to properly determine these quantities. With no exact solutions to these parameters, they became tools to assist in the calibration of the stream segments that they affected. This resulted in the parameters differing in respect to which check stations that relied upon them.
As for the software itself, the way that reservoirs are handled in WIRSOS must be altered. More of the processes that occur need to be represented to eliminate some of the assumptions that were made to allow for calibration. other than this fact, WIRSOS provides an excellent foundation for water management models to be established in Wyoming.
Continued construction of this model should result from the specific uses that can be incorporated into it. The specific topics that need to be addressed include the addition of more runoff data, more diversion permits, and further definition of the reservoirs. As the range of these inputs increases, the more appropriately it should define the actual status of the Green River.
Meena, 1993 Table of Contents
Water Resources Data System Library | Water Resources Data System Homepage