OCR Text |
Show 3.3 Applications of model to Salt Lake County Parametric applications of the model developed were performed with Salt Lake County data for a number of conditions. As was discussed in Section 3.1, there are 10 major water supply agencies that extensively use groundwater in Salt Lake County. The objectives of the applications presented in this section were ( 1) to test and refine the model formulated, ( 2) determine minimum cost annual pumping schedules for the ten water agencies for operation of the system over a specified time period, ( 3) examine the effects of demand variability on optimal groundwater use patterns in the County, and ( 4) develop recommendations for least cost use of groundwater during a drought period for the County. The general steps for each application are briefly enumerated below: ( 1) Analyze historical pumping record for each water supply agency. The average annual groundwater use for each agency in the county was analyzed to compute a number of summary statistics. These statistics included the average annual pumping, the ratio of the maximum annual pumping to the average annual pumping, and the ratio of the minimum annual pumping to the average annual pumping. These statistics were computed for each well node as well as each agency. The intent of such an analysis is to assess the temporal variability in groundwater use, to determine appropriate ranges for groundwater demand, and to aid the specification of lower and upper limits of pumping. Summary statistics obtained for the 1973 to 1982 historical record are presented in Table 3.1. ( 2) Select candidate well nodes for optimization. Typically the number of individual wells operated by water supply agencies is quite large. The finite difference formulation used for simulating the aquifer system aggregates all wells in a finite difference cell at the node associated with the cell. Not all the nodes where pumping activities are concentrated are candidates for optimization. The wells located at some of these nodes may belong to users or agencies that are not considered in the management plan. Further to reduce the size of the management problems to be solved it may be appropriate to restrict the number of candidate wells where decisions on annual pumpage are considered. Such a selection can come from several considerations. The foremost of these is a restriction of candidate well nodes to locales where there is a significant potential for improved aquifer yield. Other locations may be dictated by proximity to concentrations of demand or from geographical or political considerations. Likewise some locations may be infeasible due to the lack of suitable sites for pumping stations or from political considerations. For the Salt Lake County applications, the main criterion considered was to restrict candidate well sites for optimization to areas of high aquifer productivity. These were identified as areas ( cells of the finite difference grid) where existing pumping exceeded 0.6 cubic feet/ second ( cfs). There are 65 wells corresponding to 39 nodes. The 39 nodes correspond to 47 well node locations. This happens because in some cases more than one agency has wells assigned to the same node. Waddell et al. ( 19 87) used pumping rates of 0.3 cfs as a criteria to identify areas of high productivity. They identified 122 well nodes as exceeding this threshold. The higher threshold of 0.6 cfs was used in this study to ( 1) reduce the computer memory requirements for optimization, ( 2) increase the groundwater use in zones of higher productivity in order to reduce the concentration of areas of high drawdown reported by Waddell et al. ( 1987). All other wells are considered to pump at historical pumping rates 54 |