|
Craig Schwabe 1. Introduction The need and access to spatial information in South Africa for use in decision making and development planning has been a long-standing issue. With the implementation of the Reconstruction and Development Programme (RDP) in 1994 a strategy was implemented which provided the necessary policies and social incentives for comprehensive spatial data to be made available and developed for use in transforming South Africa into a new state under the guidance of the Government of National Unity (GNU). The RDP essentially commits the government to improve the quality of life and standard of living of all South Africans and in order for this to be effectively achieved requires comprehensive sets of spatial information (RDP White Paper, 1994). The RDP is an integrated growth and development policy that aims to not only address poverty, inequality and sustainable economic growth but to transform South Africa into a democratic, non-racial and non-sexist state. The government feels that this can best be achieved by following six basic principles, namely, that there is a need for an integrated and sustainable programme, the RDP must be people driven, that there is peace and security for all so that a new nation can be built through the process of reconstruction and development and underlying all of these is the need for a truly democratic South Africa. There is no doubt that for the aims of the RDP to be achieved requires information for decision making whether it be in the prioritising of regions for funding or in the monitoring of the development process or in addressing the "skewed distribution of resources" (Harris et al, 1995). A model which clearly sets out the strategies for developing the necessary information in the decision making process is required. The model is further necessitated by the fact that more and more government officials and planners are seeing the need for spatial information in their decision making (de Man, 1988; Harris et al, 1995). Presently, there is still an urgent need for baseline information especially data of the right sort (Chorley, 1988). For decision making on the allocation of funding for any particular development activity requires an understanding of the needs of people in relation to their access to services and infrastructure at a particular spatial level. What is seen to be the case is that there is a great deal of information within government departments and service providers but it is not very accessible. This may be a result of it not being in a spatial format or it being held within the Organisation because of it being confidential or the Organisation seeing that its power base, through its information, may be lost. Accessibility to information is often a problem because the dissemination of information, especially in an easy to use format, is not seen as a priority by an Organisation (Chorley, 1988). Communication amongst institutions that are involved in the gathering of baseline spatial information is seen as a major hindrance in the use of information for processes such as the RDP. What is needed in this regard is the co-ordination of efforts to clearly document what spatial information is available and which organisations are involved in the gathering of such information. Mechanisms need to be examined which will facilitate communication amongst organisations developing GIS databases. If this is not undertaken, it is envisaged that millions of rand will be continually spent on duplicating the gathering of baseline information. A visionary approach needs to be implemented whereby effective communication is achieved within the information and GIS industries in South Africa. A good start in this regard would be for an Organisation to champion the cause and drive the process of informing people about what information is available and where is it being housed (Campbell, 1994). More partnerships also needs to be formed to be able to address the large and diverse spatial information needs of government and to be able to put together the teams with the necessary skill to undertake the analyses of the information. GIS and spatial information is a powerful tool for decision making but policy makers, government officials and planners need to be made more aware of its potential application in their specific areas of interest (Chorley, 1988). A holistic model for developing spatial information and providing it for decision making will address many of the existing problems being faced within South Africa. Not only would such a model improve decision making but it would result in the more effective administration and monitoring of funds allocated by government to the RDP and other programmes for improving people's access to basic services and infrastructure. The model would facilitate a more equitable and efficient distribution of services and infrastructure and it would ensure more balanced regional development (Al-Ankary, 1991). An important component of such a model would be to focus on the needs of government (Campbell, 1994). In doing so, it is imperative that spatial information providers take a proactive role in the initial development of baseline information. This information can then be presented to government and its applications to their decision making illustrated. By following such approach decision-makers and planners will have been presented with a picture which they can relate to and support. The spatial information model should also ensure that once the baseline data sets have been captured that there is a free flow of information to the people of South Africa (Harris et al, 1995). The model should ensure that spatial information gathered is sufficiently comprehensive to ensure that most information needed for effective decision making is available. To meet the needs of processes such as the RDP, priority data sets should be gathered first and then other information that has not already been collected. This will necessitate the development of data sets that allow for integration so that the complexity of questions being posed can be answered (de Man, 1988). Care must be taken to assure decision-makers that the process to develop such a model can take place within the immediate short term rather than the long term. This is because the delivery of spatial information is required now to facilitate answers needed to development questions. For the successful implementation of the model it is imperative that its components are clearly defined and are recognised within the decision making circles of government. This will require an effective process of informing and educating officials at different levels of government about the spatial information model. 2. A Spatial Information Model The aim of the model is to provide macro spatial information needed at a national and provincial level for effective decision making in the reconstruction and development process in South Africa. In addition, a multi-layered and incremental approach is followed in the development of the spatial information model which ensures the full integration and integrity of the different data sets (Al-Ankary, 1991; Jones et al, 1996). An important component of the model is the development of mechanisms for the dissemination of the spatial on. The belief is that a comprehensive baseline set of information is needed first before any spatial analyses can be undertaken or effective decision making can take place. In the development of the model the emphasis is on a national focus which entails developing the spatial data sets at relevant spatial levels for the entire country. However, to fine tune the methodological aspects it is necessary in certain circumstances to undertake pilot studies which can later lead to national initiatives (de Man, 1988). The essence of the spatial information model (Figure 1) is the identification of core data sets needed for government decision making (Campbell, 1994). Not only must these data sets be required for decision making but they must be effectively used in showing decision makers the application of such information in the implementation and monitoring of the reconstruction and development process. For the RDP in South Africa three core data sets were identified, namely, socio-economic, service need and provision, and development funding. From this core set, financial models can be generated to predict the extent of further funding required to meet the basic needs of people in South Africa and development funding. I
Figure 1: The Spatial Information Model | Although the core data sets provide most of the information needed to address issues associated with the reconstruction and development process there are other secondary data sets which are required. The secondary data sets are needed to show the influence of other factors on the reconstruction and development process and to provide a truly holistic picture of the situation in South Africa. In addition, the secondary data sets form linkages with the core data sets which necessitate that they are fully integrated and results in enhancing the overall use of the spatial information model (Chorley, 1988). These data sets include spatial information on crime, education, health, welfare, environmental sustainability, labour markets and migration, economics and poverty, mortality and fertility, cadastra and even indicators on whether the people see the RDP as being successful in achieving their objectives. It may be argued that some of these data sets, for example crime, could be considered to be part of the core data sets. However, with the emphasis of the RDP being on an integrated development process to meet the basic needs of the people, it was necessary to focus on those data sets specifically needed in the development process. It must be emphasised that the core and secondary data sets are equally important but because of the focus of the spatial model, they integrate with each other in a particular fashion. In other countries where the emphasis is on other issues, it would mean that the core and secondary data sets might differ in order to meet the primary needs of government. The development of the model was based on communications with decision makers from regions throughout the country and learning from projects undertaken in the past especially in relation to their acceptance as an important component in decision making and their use in development planning. The model has also been developed from having a feeling for what questions will need to be answered now and in the future. This is guided by the vision and policy provided by government in their plans to alleviate poverty in South Africa and bring out some form of equality in access to services and infrastructure. The development of the spatial information models needs to be a conscious and deliberate act (de Man, 1988). The model should also be sufficiently flexible so that it can change with the demands of the reconstruction and development process. The development of the model has both a top-down and bottom-up approach. The top-down approach entails decision-makers identifying their information needs and developing the necessary baseline spatial information. Decisions on development activities are taken daily which necessitates that the information is readily available and as a consequence it is not always possible to determine what information communities feel should be used by government officials in making decisions about development activities. The bottom-up approach requires identifying what information communities feel should be used in decision making at a national level and what information is required for their own needs. 2.1 The Core data sets 2.1.1 Socio-economic information The socio-economic database, which is broadly perceived both locally and internationally to be the most important data set for planning purposes, was the first to be developed (Chorley, 1988). A thorough understanding of the population distribution of a country and other socio-economic characteristics (e.g. per capita income, literacy, population density) are imperative to address the needs of communities. Socio-economic information is primarily accessible from census data such as the 1985, 1991 and 1996 censuses in South Africa. The approach to the development of the socio-economic databases is to first capture the Enumerator Area (EA) boundaries used by South Africa's Central Statistical Services (CSS). An EA is the area within which people will be enumerated by one census official and in the urban areas, an EA approximates 120-180 households. In developing any spatial database it is important to always try and access the information at the smallest building block possible. In the case of socio-economic information in South Africa it is the EA. Having captured the EA boundaries in GIS, socio-economic information can be linked to each EA for purposes of further analysis and the creation of thematic maps. The EA is one of the most important building blocks in any spatial database because it provides on a regular basis up-to-date socio-economic information from the censuses that are usually held every ten years. From the EA layer, several other spatial layers may be incorporated into the data set and the socio-economic information from the EA's aggregated to provide census statistics for spatial units in the other layers. In South Africa the layers that are included in a comprehensive socio-economic database would be metropolitan suburbs, metropolitan sub-structures, municipalities, magisterial districts, district councils, tribal authorities and provinces. Attributes that would be typically linked to each layer would include population by age (Figure 2), race and gender, income (household and per capita), education statistics, household information, language and religion. Census information at an EA level is without a doubt one of the most important spatial data sets in any information model. It is used with most other data sets whether it being education, health, crime, development, services provision or business applications for gaining a greater understanding of the socio-economic environment within which decisions are being made. Another important aspect is that many internationally recognised indicators are often a ratio of population (e.g. murders per 100 000 population) and, therefore, it is imperative that this important component of the core data sets be obtained. 2.1.2 Service need and provision An understanding of what services and infrastructure people have access to is imperative for prioritising areas for development funding and addressing the backlog which has arisen as a result of policies such as Apartheid. In addition, this data set creates a benchmark against which government can monitor their progress in developing greater equity in the access to services and infrastructure (Harris et al, 1995). The data set created for the South African situation was termed the Service Need and Provision database. The first component comprised of information on the social needs of communities at a magisterial district level. This included information on the total population, population density, employment, functional literacy, poverty (Poverty Gap Index - Figure 3), dependency ratio and health. Combining the ranked scores of social variables generated an index of social need. The second component of the database comprised of information on the provision of services and infrastructure. These included factors such as access to water, sanitation, electricity, housing, policing, health, education, postal facilities, telecommunications, roads and retirement facilities. Information was obtained firstly and primarily, from existing databases, secondly, from service providers and only as a last resort, was new baseline information collected. This approach facilitated easy access to spatial information and reduced the duplication of effort and expenditure. Combining the ranked scores of service and infrastructure variables also generated an index of service need. Finally, by combining the social and service need indices a composite social and service need index was created which identifies areas within provinces and South Africa that are underdeveloped from both a social and service perspective. Regions which require intervention from government, in order to address the inequalities, are easily located (Figure 4). Products which have emanated from this database are atlases for all nine provinces in South Africa which include maps, tabular data and commentaries by local experts on the different social and service variables. By disseminating these atlases to senior government officials, the mechanism of providing spatial information for decision making is being implemented. This requires a commitment from the developers of the spatial data sets in terms of both man-hours and financial resources. 2.1.3 Development Funding To effectively implement an integrated reconstruction and development programme, it is not enough to just have an understanding of characteristics of the population or for that matter whether people have equality in access to services. An understanding of how much money is already being spent by funding agencies in regions is important because it illustrates whether government is targeting the high priority areas and whether the backlog is being addressed. Having an understanding of the population profile, service access and the extent of expenditure on development projects provides the basis upon which further financial projections can be made to alleviate the backlog of services and infrastructure. In tam, monitoring of the provision of services and infrastructure can be undertaken in addition to indices being developed to monitor the social upliftment of the population. Development funding revolves around the gathering of information on the extent of expenditure by funding agencies in South Africa at different spatial levels. Development funding information from national and provincial government line departments, district councils, the RDP and other major funding agencies is collected at a project specific level. The location of the project is determined within GIS before the information is aggregated to show the extent of expenditure at a magisterial district level. This level is used because it provides decision-makers with the most appropriate spatial level of detailed information for macro planning purposes. In addition, it is the level at which other data sets are represented. The development funding information has several important applications. Firstly, in relation to service provision information it can be determined whether under developed areas are being effectively targeted (Figure 5). Secondly, whether the allocation of funds for provinces are equitable considering indices of service need and the population profile of an area. Thirdly, whether there are areas that are either getting no financial allocation for development projects or there is the duplication of development funding for particular projects from different funding agencies. Lastly, comparisons of which types of projects are being funded and where there is the greatest service need will indicate whether the allocation of finances is the most appropriate. 2.2 The Secondary data sets 2.2.1 Crime and the justice system In South Africa with its high crime rates (Figure 6), the secondary data set that probably is the most important is the crime information. Crime is perceived by many people to be the most important issue that the GNU must tackle. Pressure is being placed on the Department of Safety and Security and the South African Police Services (SAPS) to reduce the crime rate throughout South Africa. One mechanism that has been put into place is the development of a national crime GIS database at a police station level. Information on crime, population characteristics and police resource allocation at the police station level is felt to be an important first step in providing information for decision making. The people of South Africa also need to be informed of the crime situation so that through the Community Police Forums (CPF) they can contribute to the National Crime Prevention Strategy (NCPS). The cause of the high crime rate is alternatively placed before the door of the police or the Department of Justice. The feeling is that either the police are unable to prevent crime or place criminals before the law courts or the judicial system is unable to effectively prosecute the criminals brought before them. In order to truly identify where the problem lies requires information on crime as well as statistics from the courts on the cases prosecuted. As the capacity of the prisons to house convicted criminals is also an important dimension of the criminal-justice system in South Africa, information pertaining to prison facilities in the country need also to be gathered. Only once spatial information on all three components of the criminal-justice system is available will government truly be able to develop strategies to overcome crime and improve the quality of life of all South Africans. 2.2.2 Education The growth of the South African economy will require the future labour force of the country to be better educated and skilled. An understanding of the education system in South Africa is required to improve the access of the population to education. This in turn will empower the people and with an upturn in the economy will create employment opportunities. Knowing the location of schools, their resources and facilities is imperative if people are to be provided with equal education opportunities. A GIS databases exists which shows the location of every school in South Africa. Associated with every spatial feature representing a school are more than 200 education variables (Figure 7). Using GIS, catchment areas have been defined using Thiesson polygons to examine the socio-economic profile of the communities that are potentially being serviced by the school. 2.2.3 Health The Department of Health has for several years had a fairly detailed spatial database of health facilities in South Africa. The database known as REHMIS (Regional Health Management Information System) has the geographic location of many health facilities and other important statistics on the facilities (e.g. number of beds, personnel). The provincial health departments largely co-ordinate the database. Unfortunately, at a national level it is not being effectively co-ordinated and it requires a "champion" to drive the process of developing a comprehensive and accurate health information system. Included in such a spatial data set must be epidemiological information that is critically needed to have an understanding how health impacts on the reconstruction and development process. 2.2.4 Welfare Although there is very little information on welfare aspects of the South African society in GIS it is beginning to be recognised as an important area especially in relation to family and child welfare. Data sets are already available on aspects such as child, drug and alcohol abuse. Significantly more information on aspects such as welfare facilities (e.g. homes, drug abuse centres, hospices) are needed. Welfare encompasses all aspects of a society and is directly linked to one of the RDP's main aims of improving the quality of life of people. Consequently, more emphasis should be placed on gathering information on welfare issues in South Africa. 2.2.5 Environmental South Africa is quite fortunate to have a fairly detailed environmental data set of the country and the major metropolitan areas. The Department of Environment Affairs and Tourism develop the database known as ENPAT (Environmental Potential Atlas). Included in the data set is a wide variety of environmental variables including geology, soils, hydrology, land suitability, vegetation, conservation areas and many more. The only criticism that can be laid against ENPAT is the lack of consultation in the development of the spatial database and the format of the GIS data which makes it extremely difficult to use and cumbersome. 2.2.6 Labour market The labour market in South Africa is an important component of the South African economy. It is also very diverse in nature which necessitates that spatial information be made available for decision making purposes. There is some information which has been developed by the Development Bank of Southern Africa (DBSA) which gives statistics on labour demand and supply. However, a great deal more is needed to truly be able to provide for the basic needs of the people who are geographically located in different regions. Information such as migration rates and paths, labour force by economic sector, education levels and employment is all needed. By knowing this information, the government, labour unions and business can begin to develop strategies of developing the labour market in South Africa. 2.2.7 Economics Information on how well the economy of the country is doing in the different regions is very limited and requires a concerted effort on the part of government and business to develop such information. Variables that are available are the Gross Geographic Product (GGP) per sector, per capita and household income and poverty statistics. More economic variables need to be collected including information on the creation of new businesses, productivity, location of businesses, collapse of businesses, consumption and the annual turnover per sector by region. 2.2.8 Cadastra South Africa is fortunate to have a wealth of cadastral information covering the entire country mainly as paper maps with some GIS information. The main cadastral data set that is available is the 1:500 000 topocadastral maps. World-wide, cadastral information is recognised as being a very important data set and needs to be rapidly converted into a GIS format (Chorley, 1988). This applies equally to the situation in South Africa where the Surveyor General has recognised the need for this information and is presently completing the 1:50 000 cadastral coverage of the country. The important information that can be accessed via these data sets are roads, railways, rivers, farm boundaries, administrative boundaries, settlements, contours, simple land cover information and other point, line and area features. Once all the map series have been placed in GIS, a recent and complete coverage of the country on ortho-imagery is required. 2.2.9 Mortality and fertility Having an understanding of how a country's population is changing is critical for the planning of services and infrastructure. Information on the mortality and fertility of a population is key components together with the existing population size and migration, in determining the growth or decline of a population over the years. South Africa has relatively good information from the 1991 census, and more recently from the 1996 census, on its population size. However, to estimate changes over time, information on migration, fertility and mortality is needed. Information on mortality and fertility from the DHS surveys conducted in 1991 is presently being placed into GIS. Originally the study consisted of a sample survey and, thus, the information does not provide a full coverage of the country at a sub-provincial level. However, using GIS and Artificial Neural Networks (ANN) methods of extrapolating the information to cover the whole country at a magisterial district level are being investigated. Migration information is still lacking and requires methodologies to be developed to model the movement of people from one place to another. 2.2.10 Spatial databases used for geocoding GIS databases of telephone exchanges and postal areas are recognised to be powerful data sets for the geocoding of database records which do not have geographic coordinates. An example of how these databases may be used in development planning is to obtain telephone numbers or a postal code of a project that has no geographic location and then to use the postal area or telephone exchange database to give that project geographic coordinates. In South Africa both these data sets exist to a lesser or greater degree but require a commitment from the service providers and government to improve the quality and accuracy of the information. In addition, there needs to be the integration of the telephone exchange and postal areas databases with other important data sets such as the EA database (Chorley, 1988). Another important data set is the street address database. This database consists essentially of all the streets and addresses of properties in the urban centres in the country. Again the street address database can be used to geocode the location of planning projects but its application is more towards the business sector in identifying the location of clients and new potential markets. A factor which inhibits the broader use of the street address database in the reconstruction and development process is the exorbitant costs of these databases and, to some extent, their quality. A detailed list of major towns, villages and places is also a very useful tool for geocoding development projects. In South Africa a database of 88 000 records exist on placenames which is a very powerful data source for locating RDP projects being funded throughout the country. Through organisations such as the Surveyor General, HSRC and other institutions this information can be accessed in South Africa. 2.2.11 Lifestyle Segmentation The clustering of EAs into classes based on their homogeneity from a socio-economic perspective is a very useful data set for planning, business analysis and surveying purposes. To accomplish the development of the lifestyle segmentation data sets for provinces in South Africa entailed linking 150 socio-economic variables from the 1991 census to each EA. With the use of ANN the EA are clustered into twenty or more homogeneous categories. Each category is classified in relation to the dominant socio-economic characteristics of that category. By knowing that a particular area has the lowest per capita income, has on average more than fifteen people per dwelling, the dwelling type being informal, the people are predominantly Xhosa and Zulu speaking and have an average education level of Standard 8 gives decision makers a important perspective of the type of community that they are dealing with and assists in the prioritising of areas for funding for reconstruction and development. For business analysis purposes this information is used to define the types of communities within which clients are residing and to identify new potential markets. Lifestyle segmentation information is also useful for fine tuning the marketing of specific products to specific communities based on their socio-economic profile. Categories defined in a lifestyle segmentation database form the strata that can be used for sampling purposes. The lifestyle segmentation categories used as strata provide a detailed understanding of the socio-economic profile of communities and provide strata based on more than a couple of socio-economic variables. The value of using this methodology for identifying sampling strata is that it may potentially reduce the size of the sample needed to get a more comprehensive coverage of an area based on a fairly small sample. However, more scientific research is required to determine the validity of this suggested methodology for use in survey sampling. 2.2.12 Other useful spatial databases Having all the administrative boundaries of a country is a must. These boundaries often link into each other and are useful for analysing different sets of information (e.g. crime and socio-economic data). A good understanding of the development environment within a country cannot be obtained if there are no administrative boundaries available. In South Africa, there are many different administrative boundaries with the provincial and magisterial districts being the main ones. A problem that has recently arisen in the country is the proliferation of boundaries for education, health, police, telecommunications and postal services with each set being very different to the next. This is a major problem as it hinders the management of areas in a cost-effective manner and complicates the analysis of spatial information. There is a need to standardise on a small common spatial unit (Chorley, 1988) such an the EA boundary in South Africa which hopefully will take place once a spatial information model is accepted in the country. Transportation networks of a country is useful for several different purposes. The main one is to use the road and railways to provide some form of context to the spatial information being mapped. From an analysis perspective, road networks are used for undertaking shortest path analysis and undertaking route optimisation. 2.3 Dissemination of spatial information The model suggested that for the effective dissemination of spatial information in South Africa there must be a central data processing unit those responsibilities are to gather and develop spatial databases. This information is then presented to specialised consultancy sections, based in the different provinces, whose function it is to facilitate the dissemination of the spatial information. In instances that the consultancy sections are not represented within particular regions, partnerships need to be developed with government and university departments to ensure the dissemination of information. However, care must be taken to ensure that people providing information are given training in GIS, facilitating clients and in understanding of the different spatial data sets (Chorley, 1988). The information that is disseminated includes maps, tables, digital data and customised map viewing software. The purpose of disseminating spatial information is to make it available to as many people as possible so that they may be empowered to make their own decisions. Analyses of the spatial information will be required to meet the exact needs of clients and, therefore, GIS software used in the dissemination of spatial information must be chosen after careful consideration. To ensure the effective and sustainable dissemination of information it is imperative that the consultancy section be managed along correct business principles. One of the most important aspects in ensuring the successful dissemination of information is the ability of GIS analysts, responsible for disseminating information, to facilitate the provision of spatial information to a client. Feedback from clients is also an important part of improving the dissemination of spatial information. To be able to achieve this requires a professional approach as well as a thorough understanding of GIS, analyses and the spatial data sets available. A critical issue in the provision of information is the response time to a request for maps, tables or digital data (Campbell, 1994). In general, a turnaround time of to three days for the provision of information is considered necessary to meet the time deadlines of clients and to maintain their interest in the information. However, the time to provide information is often dependent on the complexity of the information, the products being produced and the capacity of GIS personnel. 2.3.1 Spatial Decision Support Software To ensure growth in the dissemination of spatial information it is imperative that greater access be provided to decision-makers and the people of South Africa. The problem is to develop mechanisms that will encourage decision-makers and community members to use spatial information. This requires that the information is accessible in an easy-to-use format and at different spatial levels (Campbell, 1994). One mechanism that must be further used is the development of easy-to-use Spatial Decision Support Software (SDSS). According to Campbell (1994) this type of software must ensure that the "complexity and sophistication of the technology must be virtually invisible". An important aspect of improving the use of spatial information by decision makers and communities is the undertaking of research into how these people use the technology and which must be done to improve their use of the spatial information and the technology (Chorley, 1988). The design of SDSS requires several aspects to be considered. Probably the most important aspect is to make the user interface and the functions of the software as easy to use as possible. The intention should be to continually research the interaction of users with the SDSS and to obtain information from them as to how it should be modified to improve the access of spatial information to decision-makers and community members. Existing GIS software on the market, whether it is a full GIS or desktop packages are still too difficult to use for a majority of people. In South Africa, where there are eleven official languages and 64% of the South African population is considered functionally illiterate, there is a need to improve the access of these people to spatial information. Thus, the SDSS must be available in the different languages and its use by illiterate people encouraged. The SDSS must make maps and its associated tabular information available literally by a click of a button. 'Me development and structuring of the database must be in such a manner that the difficulty in accessing spatial information is totally removed from the user but the full extent of the information remains and is accessible for further analysis. When accessing the spatial information the user must be focused towards the variety of maps and information that is available within the SDSS (Armstrong and Densham, 1990). The production of the information should allow users to quickly produce high quality output or to integrate the information into other reports and documents. The integration of a user-friendly interface with a well-structured database management system and spatial analysis capabilities will determine the success of a SDSS (Armstrong and Densham, 1990). In addition, it must be possible to modify the SDSS to meet the specific needs of decision-makers and communities. Attribute information, as well as descriptions on the meaning of maps and metadata information must be easily accessible. Functions that should be included in the SDSS are zoomed views, undertaking spatial and database queries, creating new colour maps, switching spatial layers of information on and off, printing maps and viewing images of selected features. Ideally, in the future the SDSS should be able to provide verbal instructions, in the user's language, to users to access the spatial information. 2.4 The development of the spatial information model According to de Man (1988), there are three main components in the development of a successful spatial information system. These are (1) a conceptual base or a spatial information model as it is referred to in this paper (2) the collection, storage and management of data or the development of spatial databases and (3) analysis and modeling. It is felt that to ensure the effective implementation of a spatial information system in South Africa requires additional components including those of (1) information dissemination (2) making decision makers aware of what is available and the applications of GIS (3) educating and developing capacity in the use of spatial information and GIS and finally (4) the development of value-added products. The first step in developing the spatial information model is to ensure the development of national baseline data sets. This is assuming that the organisational issues, setting up the GIS software and hardware, employing of skilled GIS personnel and the financial aspects have already been attended to and implemented. For the reconstruction and development process the focus needs to be on the provision of data sets needed by decision-makers at a national and provincial level. However, a good principle in developing such data sets is to always try to capture the information at the smallest spatial unit possible and to then aggregate the information to spatial levels needed at different decision making levels (Chorley, 1988). This will in turn facilitate the provision of spatial information to people at a more local level. The development of the baseline data sets should follow the process of undertaking extensive surveys of what spatial information is in existence and then if possible, to use this information (Mikhova and Pickles, 1994). However, confidentially of information, cost and bureaucratic controls could prevent this information from being accessed and this may lead to a duplication of effort and database development. The most cost-effective approach to the development of spatial databases should always be followed and this may have to be facilitated by a GIS specialist. Once the spatial information model, with its core and secondary data sets have been completed can further more detailed analyses of the data be undertaken for purposes of the reconstruction and development process. It is argued that there is a need for government departments and national research institutions to develop spatial databases at risk based on the possible broad scale needs of decision-makers and planners in a country. By doing this, the information becomes available in the short term rather than in the long term and it means that potential users of the spatial information do not have to go through a long learning curve in the development of the data sets. It also means that a more holistic analysis of the information can take place and an understanding of the information can be developed once all the baseline information is available (Burrough and Frank, 1995). The dissemination model must at this stage be well developed and the distribution channels of the spatial information clearly defined. Products in the form of maps, tabular data, GIS data and customised map information viewing software can then be developed for dissemination. The provision of spatial information to local communities will be a much harder task than providing it to decision-makers but it is imperative in the light of the GNU's commitment to transparency and the empowering of people that this takes. Mechanisms by which this can happen is to develop information centres at strategically position facilities such as community centres, clinics, post offices, schools and police stations. Participatory research will be required to determine what spatial information communities require, to evaluate the effectiveness of providing spatial information to communities and to generate recommendations for improving the dissemination of spatial information to them (Harris et al, 1995). The Department of Communications is taking visionary steps in the evaluation of mechanisms by which information can be made available to the people of South Africa. One option is the development of Public Internet Terminals (PIT) at post offices while another is the development of Telecentres throughout the country. It is at these centres that spatial information can be disseminated to the public which in turn will empower them with their own decision making and provide them with an understanding of what is happening in the reconstruction and development process in South Africa. A process of informing decision-makers and planners throughout the country then needs to be implemented. There is a lack of awareness within both national and provincial government departments, not to mention local communities, as to what spatial information is available for South Africa and the capabilities of GIS for use in the reconstruction and development process (Chorley, 1988). It requires providers of spatial information to inform decision makers at all government levels as to what spatial information is available and how it can be used in a GIS environment to answer many of the planning decisions that are having to be made. Not only is the generation of awareness required but there is a need for capacity building in the application of GIS especially amongst officials in the new provincial governments (Chorley, 1988). To ensure the use of GIS and spatial information in the reconstruction and development process requires the education and development of capacity amongst (1) academics in the tertiary institutions, especially previously disadvantages universities (2) planners in the government departments and (3) the broader public with greater emphasis being placed on users with access to such facilities in rural communities. Amongst academics and planners the capacity building needs to revolve around principles of GIS and the practical use of the technology and spatial data (Harris et al, 1995). Users in local communities must be given the most basic training in using the technology in order to access the spatial information. Having gone through the process of conceptualising the information needs for the reconstruction and development process, developing the spatial databases, implementing the dissemination of spatial information, communicating with government about the model and educating and training people, eventually provides time for the development of value added products. This is mainly in the form of bundling spatial information with map and information viewing software to create SDSS, developing other customised software, atlases and reports. Examples of the SDSS would be demographic, crime, service provision, development funding, education and business data products. Information in the form of atlases and reports provides an important medium for disseminating information especially as it allows more in depth investigations of the spatial data and for contributions to be made by selected experts in their fields. 2.5 Some conclusions To effectively implement a spatial information model for the reconstruction and development process in South Africa requires the development of a partnership between government, service providers and the major research councils in the country. The purpose of the partnership would be to co-ordinate the gathering of the core and secondary data sets identified within the spatial information model. The value of such a partnership is that it combines the research expertise of the research councils with the resources of government and the service providers. The capacity of the members of the partnership will provide the necessary expertise and experience needed to undertake the spatial analysis of the information. In addition, a culture of sharing information, especially that funded by government, will be created which will have financial and resource benefits for members of the partnership. There is also the need for a. central body that coordinates and regulates the activities of the GIS industry in South Africa and assists with the developing of policy. The central body would play a leading role in defining the spatial information needs of government and the reconstruction and development process. It would also need to act as a forum where government, service providers and institutions involved in the development of spatial information could openly discuss their information needs, work in progress and communicate with each other about the development of new spatial databases. The central body would also play an important part in promoting the use of GIS and spatial information within the public as well as the private sector. In order for the central body to maintain its legitimacy amongst all parties involved in the use of spatial information it is imperative that it remains independent of government. This was the conclusion of the Select Committee on Science and Technology of the House of Lords in the United Kingdom in 1983 and organisations that have been set up in the U.S.A. and France have followed similar principles (Chorley, 1988). It is suggested that members be elected from institutions that represent the GIS and spatial information industries in South Africa and should include people from government departments and the research councils. It is suggested that the CSS play a leading role in the establishment of such a body. Government should do the implementation of the spatial information model. However, research councils who have been given the mandate by government to provide information for decision making can play an important role in kick-starting the process especially as they have the research and GIS expertise. Nevertheless, it should be done in partnership with government as this will result in them having ownership in the development of the spatial information model. It will also encourage government to make a long-term commitment to the process and making finances available for the development of the spatial information needed for decision making. Some of the problems that will be faced in the development of the spatial information model is the restricted access placed by organisations on their data sets, the lack of suitably qualified and skilled GIS personnel, the information not being current and it not being sufficiently accurate. Information is power and as a consequence, institutions continue to keep their information very close to them. This is a major problem as it flies in the face of government's attempts to make information available to all citizens of South Africa. Harris et al (1995) feels that "the free flow of information is essential to truly democratise the implementation of GIS" in South Africa. There are several factors that need to be met if the spatial information model is to be sustainable. Probably the most important is that there is a long-term commitment to the development of spatial information in the country (de Man, 1988). This commitment must include making sufficient funds available for the development of the model. There must also be a strong commitment to the development of experienced and highly trained GIS personnel who have the mix of skills to analyse the spatial information on behalf of decision-makers. For any institution to be sustainable requires a commitment to the training of staff and maintaining a stable working environment (Campbell, 1994). In addition, a culture of sharing information needs to be created amongst the providers of spatial information. Some of the lessons that can be learnt from the U.S.A is that to encourage the use of GIS and spatial information in decision-making requires information to be priced at a level which makes it more accessible to people. Consequently, a differential pricing policy needs to be developed which makes information available to communities at an affordable cost or even free of charge. However, corporations that can afford to purchase the information at its market-related price should be encouraged to do so. Nevertheless, an attempt should be made to provide spatial information at a price that encourages more people into using spatial information and GIS technology in their decision making. The implementing of appropriate marketing strategies is required to define the benefits of using spatial information and creating a market for spatial information. By the development of the spatial information model it is hoped that South Africa will become an example of what can be done to address issues of poverty, inequality and to ensure economic growth as South Africa is transformed into a democratic, non-racial and non-sexist state. Acknowledgements Although the opinions expressed in this paper are solely those of the authors, they would like to acknowledge the significant contribution made by the Human Sciences Research Council (HSRC) and the South African government in providing the necessary funds to focus on the development of a spatial information model for the reconstruction and development process. Members of the HSRC's GIS Centre are thanked for their continued contribution in developing the much-needed spatial information needed for decision making in South Africa.
References AL-ANKARY, K.M., 1991, An incremental approach for establishing a geographical information system in a developing country: Saudi Arabia. International Journal of Geographical Information Systems, 5,85-98. ARMSTRONG, M.P. AND DENSHAM, P.J., 1990, Database organisation strategies for spatial decision support systems. International Journal of Geographical Information Systems, 4,3-20. BURROUGH, P.A. AND FRANK, A.U., 1995, Concepts and paradigms in spatial information: are current geographical information systems truly generic? International Journal of Geographical Information Systems, 9, 101-116. CAMPBELL, H., 1994, How effective are GIS in practice? A case study of British local governments. International Journal of Geographical Information Systems, 8,309-325. CHORLEY, R., 1988, Some reflections on the handling of geographical information. International Journal of Geographical Information Systems, 2, 3-9. DE MAN, W.H.E., 1988, Establishing a geographical information system in relation to its use: A process of strategic choices. International Journal of Geographical Information Systems, 2, 245-261. HARRIS, T.M., WEINER, D., WARNER, T.A., AND LEVIN, R., 1995, Pursuing social goals through participatory geographic information systems: Redressing South Africa's historical political ecology. In: Pickles (Ed.) Ground Truth: the social implications of Geographic Information Systems, (ppl96-222). The Guilford Press, New York JONES, C.B., KIDNER, D.B., LUO, L.Q., BUNDY, G.L. AND WARE, J.M., 1996, database design for a multi-scale spatial information system. International Journal of Geographical Information Systems, 10,901-920. MIKHOVA, D. AND PICKLES, J., 1994, GIS in Bulgaria: development and perspectives. International Journal of Geographical Information Systems, 8, 471-477. RDP White paper, 1994, RDP White Paper: Discussion Document. GNU Web site, South African Government.
|
