Creation of a hydrological modelling environment to assist in the decision making of water-related activities

Abstract

In South Africa, water is a scarce resource and it has become very important to manage this resource effectively. The State developed a regulating framework, under the hospice of the Minister of Water Affairs and Forestry, which protects the country‟s water resources from over-exploitation by ensuring that it is protected, used, developed, conserved, and managed, in a sustainable and equitable manner. The laws and policies governing the use of water resources are contained in the National Water Act (South Africa, 1998), the National Water Policy (South Africa, 1997a), the National Water Resource Strategy, and the Water Services Act (South Africa, 1997b). In addition some water-related functions were transferred to Catchment Management Agencies and Water Users‟ Associations, and it is their task to ensure that the strategies, laws and policies are implemented.

Effective water management can only be performed by making use of hydroinformatics which assists with simulations and estimations. As a result input data will be collected, added to a Relational Database Management System and output results generated. A Geographic Information System with the support of a geodatabase will allow users to store spatial and temporal data.

The research project investigated different water-related data models (ArcHydro, Hydstra, GML, HYMOS, and WinHSPF), as well as hydrological modelling frameworks (BASINS, OMS, OpenMI, SPATSIM, and TIME) to determine whether they were adequate to assist with the decision making of water-related activities.

It was found that these data models and hydrological modelling frameworks did not allow users to add new datasets to their existing data structures and in many cases only had a limited set of functions. For these reasons it was decided to develop a comprehensive, modifiable, geodatabase that will function in a modelling environment which will allow users to save their data in a centralised database. Additionally the functionality provided by other data models and modelling frameworks may be linked and used in the new modelling environment.

A methodology that has been followed was to first establish the objectives of the research project, gather the necessary data, investigate various data models and hydrological modelling frameworks, determine the requirements for the modelling environment, design and create the modelling environment, design and create the geodatabase, and finally selecting the study area which will provide the research project with the necessary data.

The following findings were made concerning the research project: firstly, that ArcHydro will be used as example data model to assist in designing the geodatabase. Secondly, that UML will be used as a development tool to assist with the development of the geodatabase. Thirdly, that the geodatabase will be generated from the XML schema and be made available to ArcCatalog. Fourthly, that data from different users/providers (Hydstra, Stats SA, Weather Bureau, Department of Water Affairs and Forestry, etc.) be inserted into the geodatabase. Fifthly, that any other hydrological modelling framework may make use of the data stored in the geodatabase. Finally, ArcGIS was selected as GIS application and Microsoft Access as a storage area.