| This research deals with numerical watershed simulation as a basis for management of multi-scale, multi-process environmental systems. The research focuses on finding a parsimonious strategy for incorporating spatially distributed process understanding, the representation of coupled processes in watershed models, and numerical strategies for solving multi-state, distributed environmental models. The dissertation has four main themes: (1) Linking geospatial data from national data sources in support of distributed watershed simulation to develop model-data coupling strategy; (2) Implementing surface, sub-surface process coupling between a coastal watershed and estuary; (3) Investigate the role of process coupling in simulating space-time dynamics of wetland landscapes in an integrated watershed simulation; (4) The distributed “age” of water in watershed flow systems.;In a watershed, hydrology is the primary driver of various physical, chemical, and biological processes. A fundamental understanding of the geospatial and temporal nature of catchment hydrology is important for transport of nutrients and sediments. Numerical watershed models are increasingly being used by scientists for hypothesis testing, designing of observation networks and by agency professionals for watershed management and related decision-making. Central to the next generation of models is the incorporation of geospatial properties of the watershed (for e.g. topography, vegetation, soil, geology, regolith, etc.) in addition to the national observing system for precipitation, meteorological parameters, streamflow, groundwater, and soil-moisture, etc. Empirical and conceptual models face challenges in applicability at different scales and transferability of calibrated parameters due to lack of relation between the model parameters and the geospatial watershed properties. More importantly, such models may not provide sufficient insight regarding the internal hydrologic processes of the catchment. Physics-based distributed modeling frameworks that utilize geospatial information hold the potential to improve the performance of the watershed predictions, while providing a more comprehensive tool for assessing impacts, advantages, and disadvantages of management scenarios and decision making.;Implementations of Penn State Integrated Hydrologic Model (PIHM) for watersheds of multiple scales in the Chesapeake Bay watershed were studied where new generation of high-resolution accessible national geospatial data serve as a practical basis for numerical watershed simulations. PIHMgis, a tightly coupled interface to PIHM, was developed to efficiently and accurately prototype, execute, and analyze distributed model simulations of a watershed in an open source GIS using national geospatial data products. New process couplings were implemented in the model and examined at watersheds of multiple scale: (a) to estimate groundwater discharge to estuary across the shoreline in a coastal watershed (b) to simulate the location, function, and dynamics of wetlands based on the simulated near surface availability of water. A coupled flow-transport-age model was developed to simulate the transport of the environmental isotopic tracers, estimate the spatial distribution of age in multi-state system, and seasonality of the watershed’s residence time. |
