Despite a long history of synergy, current techniques for integrating Geographic Information System (GIS) software with hydrologic simulation models do not fully utilize the potential of GIS for modeling hydrologic systems. Part of the reason for this is a lack of GIS data models appropriate for representing fluid flow in space and time. Here we address this challenge by proposing a spatiotemporal data model designed specifically for large‐scale river basin systems. The data model builds from core concepts in geographic information science and extends these concepts to accommodate mathematical representations of fluid flow at a regional scale. Space–time is abstracted into three basic objects relevant to hydrologic systems: a control volume, a flux and a flux coupler. A control volume is capable of storing mass, energy or momentum through time, a flux represents the movement of these quantities within space–time and a flux coupler insures conservation of the quantities within an overall system. To demonstrate the data model, a simple case study is presented to show how the data model could be applied to digitally represent a river basin system.

Mitigating water contamination, improving water security, and increasing sustainability involve environmental awareness and conscientious decision-making by denizens and stakeholders. Achieving such awareness requires visually compelling geospatial decision-making tools that take into account the probabilistic and spatially distributed nature of water contamination. Inspired by the success of weather maps, this paper presents a novel STochastic Reliability-based Risk Evaluation And Mapping for watershed Systems and Sustainability (STREAMS) tool that produces and effectively communicates the risk of water contamination as maps. STREAMS is integrated with ArcGIS geoprocessing tools and uses physics-based reliability theory to compute the spatial distribution of risk, which is defined as the probability of exceeding a safety threshold of water contamination within a watershed. A quantitative analysis of the efficacy of mitigation strategies is conducted by estimating risk reduction from best management practices throughout the entire watershed. Two case studies at different spatial scales are presented, demonstrating STREAMS application to watersheds with varied properties.

New developments in geographic information systems are providing a stronger framework for synthesizing geospatial and temporal water resources data into a single information system. Arc Hydro is a data model and toolset for water resources application, built within the ArcGIS software system. Arc Hydro includes a network of rivers and streams, a means for defining drainage areas and linking them to the network, connections to time series recorded at gaging sites, and a three-dimensional model for channel shape. Application of Arc Hydro is facilitated in the United States by the existence of national datasets for digital elevation, watershed boundaries, hydrographic networks, and stream gauging sites. An example application of Arc Hydro to water availability modelling in Texas is presented in which it is shown how raster delineation of drainage areas in subbasins is combined with a river network representation of the whole basin to allow water supply reliability to be evaluated throughout large river basins. Some perspectives on the possible future evolution of this technology are provided.

Tillamook Bay, on the coast of the Oregon, has a history of bacterial pollution problems. A method to model non-point source pollution and the benefits of best management practices (BMPs) was developed using a Geographic Information System (GIS) platform, specifically Arc/Info and ArcView. A 100-foot grid digital elevation model was analyzed to estimate storm water discharge using spatially distributed annual precipitation data and calibrated runoff coefficients. Non-point source loads of bacteria were determined for each grid cell based on expected mean concentration (EMC) values that were assigned to runoff from each land use in the watershed. These non-point loads were accumulated down streams to the receiving estuary. The potential load reduction resulting from BMP implementation was calculated using data related to BMP effectiveness and level of implementation. Predicted concentration grids also were calculated along streams based on accumulated loads and flows. The model indicates that most of the bacteria load is coming from the dairy lands, but that implementation of BMPs can effectively reduce loads to the rivers and, ultimately, the bay. Results also show that the model can reasonably reproduce flow and load conditions.