Linking geographic information systems and runoff modelling for environmental planning

One of the important consequences of development plans is their effect on the runoff characteristics of a particular catchment. Environmental planners in the early stages of a planning process may wish to use computer-based hydrologic models to predict this effect and arrive at planning or management decisions in the least possible time. However, hydrologic models are seldom accessible to planners or managers because they are designed for engineers or modellers. Using them requires considerable expertise and ? great amount of time, and their data requirements and operating rules are intimidating not only to planners but to many engineers as well. Moreover, the visual display of development plans in a catchment is conventionally based on paper, and is rudimentary and fixed. The aim of this work is to develop an approach to making a conventional runoff model accessible to environmental planners by using a geographic information system (GIS). Criteria were developed for selecting from a plethora of runoff models. Using these criteria, a simple, existing runoff-routing model called RORB was selected. When using RORB, new map-based data will have to be prepared for each alternative plan to be investigated. As production of these map-based data is time-consuming and requires considerable expertise, automating it through a linkage with a GIS is a major key to making the RORB model more accessible. Because of the specific procedural requirements of RORB, criteria for selecting an appropriate GIS to use had to be developed. Using these criteria, it was established that a new GIS did not have to be developed, and an existing one called ARC/INFO was suitable. The GIS and the model were then linked using a loose-coupling design, which involved a peer-to-peer linkage architecture. The coupled system used the GIS as an automated, map-based data provider for RORB. This virtually eliminated the need for expertise in producing map-based data and reduced the production time by three quarters. The visual display of catchment models and plans was excellent. The hydrologic results obtained using the RORB model in conjunction with the automated GIS-based approach were not significantly different from those obtained by conventional use of RORB by engineers