Research On Stochastic Fuzzy Method And Application In Watershed Runoff Simulation And Water Mmanagement

Under the coupling impacts of climate change and anthropogenic activities, watershed systems can no longer cope with changing environmental conditions in time, a lot of serious environmental problems and natural hazards have occurred in the watersheds around the world. However, complex watershed systems and changing environmental conditions make uncertainties be incorporated within the watershed modeling studies, and as such, these uncertainties should be carefully considered, reflected and analyzed in the watershed models. However, many theoretical and field studies have recognized that runoff simulation is significantly influenced by uncertainties inherent in hydrological flow processes, and thus may affect model predictions accuracy. These uncertainties generally emerge from heterogeneity of hydrogeological environment and scarcity of related data, and they can further be related to aquifer characteristics, and/or physical properties of the watershed. Consequently, This dissertation take the hydrological modeling and water resources management as the research point, select the typical river basin:Xiangxi River Watershed, Zhangweinan River Basin and Kaidu Watershed, China as the study area. The research includes hydrometeorological, uncertainty analysis, water resources management and optimization, hydrological modeling and other field. Main research works and innovative achievements were introduced as follows:(1) Model effects of DEM resolution on topography-based watershed runoff simulation under uncertainty for the Xiangxi River watershed, China. The TOMODEL will be used for simulating water movement in the entire land phase of the hydrological cycle. Four experimental catchments with hydrological and meteorological data will be selected to carry out the research, and in each catchment hydrological station has been built to get the detailed hydrological and meteorological data. Different levels of DEM grid sizes between 30 m and 200 m are used, and parameter calibrations are conducted at each level. Fuzzy analysis technique will be used for analyzing the relationships between the DEM resolution and TOPMODEL efficiency in considering Nash, SSE and SAE values. The results obtained will be useful for helping planers to establish effective water exploitation and allocation policies and thus improve the local ecosystem sustainability.(2) Develop a hybrid stochastic-fuzzy-analysis (HSFA) approach for rain runoff simulation, through integrating a semi-distributed hydrological model with stochastic-fuzzy analysis for water resources management in the Kaidu Watershed, northwest China. The HSFA will be used for simulating water movement in the entire land phase of the hydrological cycle. Besides, the uncertainties in hydrological modeling associated with fuzzy and stochastic features are analyzed. The results reveal that uncertainties in parameters M, ln(TO), and ChVel have significant impacts on predicted flow in flood season (June to August). The percentage of the observations lay within the 95% confidence intervals increased with the a-cut levels, and the ranges of the residuals were decreased with the increase of a-cut levels and typically lower than that in the conventional TOPMODEL.(3) A double-side fuzzy chance-constrained fractional programming (DFCFP) method is developed for planning water resources management under uncertainty. In DFCFP the system marginal benefit with per unit of input under uncertainty can also be balanced. The DFCFP is applied to a real case of water resources management in Zhangweinan River Basin, China. Results show that the amounts of water allocated to the two cities (Anyang and Handan) would be both different under minimum and maximum reliability degrees. Results discover that the marginal benefit of the system solved by DFCFP is bigger than the system benefit under the minimum and maximum reliabilities degrees, which not only improve economic efficiency in the mass, but also remedy water deficiency. Compared with the traditional DFCP method, the solutions obtained from DFCFP lead to significant higher than the DFCP model, and the DFCFP has advantages in water conservation.(4) Develop a linear fractional two stage programming (FTSP) model for agricultural water management and planning in the Zhangweinan River Irrigation Area, China. The developed model incorporate approaches of two-stage stochastic programming (TSP) and linear fractional programming (LFP) within a general optimization framework, such that uncertainties expressed as both fractional and probability distributions can be addressed. Moreover, it can reflect tradeoffs between conflicting economic benefits and the associated penalties attributed to the violation of irrigation targets. The results demonstrated that different pre-regulating irrigation target policies lead to different system benefit and target-violation risk and scenario 1 balanced the system benefit and irrigation target violating risk; water allocation rate of the two channels are different under different scenarios, and under high inflow level, all the fifteen subareas have allocated water flow.