| Hydrologic systems have been interfered by human activities directly or indirectly at various scales and human increasingly becomes a significant component of the hydrologic cycle. It no longer makes sense to study only natural hydrological cycles in river basins where human impacts on natural processes are intensive and extensive. Understanding and predicting the effects of indirect human interferences, such as land use change and urbanization or climate change, have been and will still be a topic of hydrologic research. Recently, researchers attempt to incorporate direct human interferences such as irrigation, groundwater pumping, return flow, and reservoir operation into hydrologic simulation models. Conventionally, direct human interferences are added or subtracted from water balance outside of the modeling systems by assuming a linear relationship between the human and natural systems. However, nonlinear impacts of human interferences and nonlinear response of water storage and flows may exist because of the temporal change of water uses, the dynamics of natural processes, and the interaction between the two. Thus direct human interferences should be modeled as endogenous variables in order to understand the interaction within the coupled human and natural systems.;However, current data availability of human activities is limited and water use data involve considerable uncertainties. Corresponding to this issue, the goal of this thesis research is to improve the understanding of the coupled human and hydrological systems by detecting, estimating and modeling both indirect and direct human interferences. This thesis research is expected to advance more accurate understanding of human interferences to hydrologic systems, as well as providing new methodology to study heavily managed watersheds as coupled human-natural systems. The broader impact of the scientific exploration includes: (1) water use estimation -- an extension of the irrigation scheduling approach; (2) farmers' behavior of water use, which, if coupled with economic analysis, may be used to explain water use decisions and derive implications for water saving; (3) methodology in dealing with comprehensive input errors in hydrologic modeling, which attempts to bring human input data errors to the attention of hydrologic modeling community; and (4) new sequential data assimilation approaches, which are expected to be used for other problems. |
