Creating an Environmental Geographic Information System for the City of Kumasi, Ghana

Abstract: The city of Kumasi in Ghana struggles with a number of environmental issues, including excessive road traffic, air and water pollution, flooding, and inadequate solid waste management. If there is a group that is directly affected by these issues, it certainly is the city’s population. At the same time, the people of Kumasi—who are the constituents of the local administration, called KMA—have few to none means of obtaining objective information about the state of the urban environment, and therefore no way of holding the city administration accountable concerning environmentally relevant decisions. This case study aims to explore the possibility of alleviating this transparency issue by creating an ‘environmental information system’ (EIS) for the city. The term EIS in this context denotes an information system which can be used to publish environmental information on the web, to be utilised by students, professionals, NGOs, and the general public. The case study seeks to provide answers to two research questions: What are the software requirements for an EIS for Kumasi? And: How can free software be used to satisfy these requirements? The case study takes an approach based on Soft Systems Methodology and agile software development techniques to explore the software requirements. As part of the study, a prototype of the EIS was developed in order to explore the requirements even more, and in order to determine the applicability of currently available free software. The results of the requirements analysis include the following observations: geographical information is essential in presenting the city’s environmental issues, therefore the EIS is based on geographic information system (GIS) software and techniques; the information should be presented to the public in an easy-to-use and easy-to-understand way in order to reach the largest possible percentage of the target group; the environmental information that is available at local institutions (such as the largest local university KNUST, the city administration, and the Environmental Protection Agency) is scarce and semantically and syntactically heterogeneous—therefore, the EIS must be able to consolidate such information in order to present it in an easy-to-understand way; many of the involved actors have no or little knowledge in GIS techniques, therefore the EIS must be usable without such knowledge. In the implementation process, heavy use was made of free software components: GeoServer for publishing geographical data using WMS and WFS; PostgreSQL with the PostGIS extension for data storage; JPA/Hibernate for storing metadata in PostgreSQL; Spring MVC, jQuery UI and many other libraries for creating a user-friendly web application; OpenLayers for displaying and editing geographical data in the web application; GeoTools for handling geographical data on the server-side. During implementation, actual environmental information was entered into the EIS in order to provide a realistic semantic environment for the agile development process. The study concludes that—while the implemented prototype does not include all of the features which were identified as required, and while a ‘full’ soft systems analysis (as opposed to the ‘soft systems perspective’ which was applied) would have led to a more complete picture of the software’s organisational environment—the implementation of an environmental information system for Kumasi, based solely on free software, is viable in the current technical and organisational environment. KNUST is foreseen to be an adequate organisation to manage the development and operation of the system, since the necessary technical knowledge is available. The successful operation of the EIS relies on environmental information being provided by data producers such as the KMA, the EPA, the Ghana Statistical Service, and various departments at KNUST.