OCR Text |
Show planning for the development and management of Great Salt Lake and its basin. This approach recognizes the difficulties in commensurating environmental quality, for example, with economic terms ( costs or benefits). Accordingly, the various societal and economic objectives and goals discussed in the previous sections are each modeled in its own unit and terms. Consequently, several non- commensurable objective functions are produced. These objectives are then evaluated at the highest level in the hierarchy by means of the surrogate worth trade- off method •( SWT) ( Haimes and Hall, 1974). A major feature of the SWT method is its capability to quantitatively and systematically evaluate non- commensurable multiobjective functions in terms understood and acceptable to the decision- maker. The application of the SWT method is discussed in Chapter V of this report. The models associated with each of the subsystems at each layer in the hierarchy are likely to be different from each other - both in structure, scope, and complexity. The models associated with the first layer in the hierarchy, for all three different decomposition schemes discussed ( hydrological- geographical, temporal, and functional decomposition), are information oriented models - to distinguish from optimization oriented models. They are aimed at providing information and future prediction related to the specific aspect of the system that they ( model) represent. The models associated with the first level- second layer, in the hierarchy may be classified as both information oriented- predictive models and optimization oriented ones. The optimization procedure itself, by manipulating the control measures ( both the technical and nontechnical ones), is carried on at the second level- second layer of the hierarchy. The utility and objective functions and the system's constraints are all constructed at the first level- second layer of the hierarchy. For example, consider the recreation- tourism subsystem. A utility function or functions should be constructed which relates the desirable goals associated with this subsystem such as stable lake water level, low health hazard, and easy access to the control measures such as construction of dikes and tributary storage reservoirs, adequate sewage treatment, mosquito control, and the development of parks, beaches, and associated features. Under the SWT approach utility functions are not necessarily expressed in monetary terms. They may be in units of level of the water in the lake, number of users of the recreation facilities, level of health hazard, sensitivity of water level in the lake to other control measures, such as flow of tributaries to the lake and reservoir operations. The construction of the utility functions in non- commensurable terms to each other is made viable by the SWT method ( Haimes and Hall, 1974; Haimes, Hall and Freedman, 1975). Thus, while these optimization oriented objective functions are constructed at the first level- second layer of the hierarchy, they will be all together ( all from first level subsystems) analyzed and their trade- offs evaluated at the second level in the hierarchy in order to achieve a solution which is acceptable to the decision- makers. As suggested by Figure 7, a comprehensive management of the water resources system of Great Salt Lake is necessarily based on a realistic and adequate representation of the physical aspects of the system ( first layer). For this reason, the component subsystems at this level will be emphasized during the early stages of the study to develop a comprehensive model of the entire system depicted by Figure 7. As was indicated earlier, a close examination of any of the component subsystems shown by this figure would reveal the major internal processes. For example, with reference to the box of Figure 7 which indicates the " lake watershed," the hydrologic processes within this box logically could be represented by the typical block flow diagram of Figure 10. In this diagram the blocks represent storage locations within the subsystem and the lines represent various processes by means of which water is transferred from one storage location to another. Thus, the subsystem which represents the lake watershed is identified and the modeling process is able to continue. This same procedure will be followed in identifying the subsystems of the near shore and the lake itself and eventually for the entire system as shown by Figure 7. As the real world system is better understood, the conceptual model is adjusted to coincide more closely with the system of the real world. In this case, the filtering loss is lessened between the real world and the conceptual model, as indicated by Figure 3. Watershed submodels As indicated earlier, initial emphasis in model development under this project will be placed on the physical aspects of the lake system. The three major space units which are delineated for this portion of the total system are indicated by the three boxes at the bottom of the diagram of Figure 7. The lake watershed is composed primarily of the drainage basins of the Jordan, Weber, and Bear Rivers. Under previous projects at the Utah Water Research Laboratory ( UWRL), hydrologic models of these three basins already have been developed ( Israelsen et al., 1968; Hill et al., 1970; Wanget al., |