This article specifically focuses on introducing GIS spatial analysis and the subsequent thematic mapping to informal settlement upgrading based on information gathered from the residents of the informal settlement, Monwabisi Park.
The use of geographic information systems (GIS) in informal settlement upgrading has rapidly increased throughout the world [1,2 and 3]. This is because structured planning makes use of both spatial and attribute data which are supported by GIS. Nayak et al. [4 p. 290] describes GIS as: “A computer assisted system for the acquisition, storage, analysis and display of geographic data”. Shoba and Ra sappan [5 p. 1] also describe GIS as follows: “A computer tool for capturing, storing, querying, analysing and displaying spatial data from the real world for a particular set of purposes”.
Fig. 1: Geographical location of Monwabisi Park.
Data storage, management and processing are powerful capabilities of GIS that can be utilised in informal settlement upgrading. GIS software is capable of collection, storage, management, retrieval and analysis of geographic information [6]. It is also able to output data in various formats e.g. maps which can be useful in decision making and urban upgrading processes.
This study utilises indigenous or local community knowledge which have been integrated with other data (such as stakeholders‘ perspectives) which would eventually be analysed by GIS. This more inclusive access to data, as well as provision of data, empowers communities who would otherwise not have had any say in the process [7].
Spatial analysis is used to process this data into useful and meaningful information which can be used in making effective decisions during the planning process. Evaluation of environmental impacts caused by informal occupation and flood mapping are some of the applications of GIS in informal settlements that have been carried out in previous studies [1, 2 and 8]. The focus tended to be on GIS mapping, with less emphasis being placed on the application of spatial analysis. This study was requested by Violence Prevention through Urban Upgrading (VPUU), an organisation which works closely with the community being analysed. It showcases the use of various types of spatial analysis in informal settlement upgrading.
Background and motivation
Khayelitsha is Cape Town’s biggest township, and was created for the black population as a result of South Africa‘s apartheid legislation. It is located in the province of the Western Cape, on the south-east of Cape Town’s metropolitan area. The informal settlement of Monwabisi Park is the site of this study, and lies on the south-east side of Khayelitsha (Fig.1).
Fig. 2: Map showing vulnerability based on type of exposure to a hazard in section M, C, B and A of Monwabisi Park.
Initially, Monwabisi Park was reserved for landfill purposes. It was occupied in 1997 by some residents from Harare, a suburb of Khayelitsha. The exact population of Monwabisi Park is unknown because of its dynamic nature as an informal settlement. In 2012, a study showed that the population was approximately 6000 households.
Like most informal settlements in South Africa, Monwabisi Park faces service provision challenges. These include the following:
In early 2013, Monwabisi Park was identified as the study area by VPUU, the research partners and initiators of this project. VPUU is a group based in the City of Cape Town, and uses social engagement and town planning as tools in fighting crime in Khayelitsha, mainly to improve living conditions of residents. Having done GIS mapping in previous projects, they approached the authors to explore further GIS mapping and spatial analysis that could improve on what they had done already. VPUU, together with the community of Monwabisi Park, collected the data through questionnaires and mapping each household structure. This is a more democratic application of GIS, which allowed the community to participate and contribute to the body of knowledge known as Participatory GIS (PGIS).
| Data | Data type | Content |
| Spatial location of households | Polygons | Geographical location of households |
| Socio-economic demographic data | Text/attribute data | Socio-economic and demographic information |
| Monwabisi park sections | Polygons | Polygons dividing Monwabisi Park into four sections |
| Contour line | Line | Elevation of Monwabisi Park terrain |
Laituri [9 p. 25] describes PGIS as: “A confluence of social activity such as grassroots organisations and government decision making with technology in specific places or grounded geographies”. Quan et al. [10] describes PGIS as “the integration of local knowledge as well as stakeholders’ perspectives in a GIS”. The incorporation of indigenous knowledge of the area in the decision making process is emphasised so as to empower the community [7].
GIS in informal settlement upgrading
In 1983, the Voluntary Association for International Service was invited by the Catholic Church to provide technical support in the Belo Horizonte town upgrading, in which they decided to use GIS. This resulted in Brazil being the first country to utilise GIS in informal settlement upgrading [3]. It was effective for alternative decision making during the upgrading process. The successful application of GIS in Brazil created more interest in its application as a planning tool in informal settlement upgrading. In addition, the use of GIS in urban upgrading expanded throughout the world, in both developed and developing countries [3, 11 and 12].
In 1996, an adaption of the modified method used in Brazil was applied in Cape Town informal settlement upgrading, and it was the first time GIS was used in South Africa to this effect [3]. The project was successful and led to more recognition of GIS in upgrading informal settlements. This was followed by other scholars who have used GIS in informal settlement upgrading [1, 2, 8 and 13]
GIS analysis
The following section introduces different spatial analysis techniques that can be used during the informal settlement upgrading process. This is by no means an exhaustive list, but was chosen so as to showcase techniques that were not previously considered by VPUU.
Fig. 2: Map showing vulnerability based on type of exposure to a hazard in section M, C, B and A of Monwabisi Park.
Approach
The main aim of this study was to apply additional analysis on the spatial data captured by VPUU, so as to optimise the use of GIS on the rich data that came out of the participatory mapping of Monwabisi Park. The existing analysis that was done was largely restricted to thematic mapping of individual attributes.
After consultation with VPUU to understand their broad operational aims, data capture methods, and the GIS methods employed, different types of spatial analyses that they had not used before were identified that might be useful. These were: buffer analysis, Thiessen polygons, multi-criteria evaluation analysis, and thematic mapping.
Enumeration and mapping
For this project, the data was captured by VPUU, using field- and office work. The data was collected from all the households in Monwabisi Park; furthermore some of the residents were involved in the data collection process. Before commencement of fieldwork, outlines of households were digitised, to create a polygon shapefile. During the fieldwork, each team of four data collectors had the following responsibilities:
The data had to be cleaned and the household shapefile then had to be joined to the attribute data. Due to irregularities between the handwritten responses, the captured attribute data, and the numbering of the polygons in the shapefile, 1127 records (either lines in the attribute table, or polygons in the shapefile) could not be used. This resulted in a shapefile containing 5125 polygons representing households with attribute data.A preliminary exploration of the questionnaire data showed the characteristics in Table 1.
The selection of the GIS analysis method was dependent on a particular problem that was to be analysed. For instance, using a 100 m buffer to analyse how many households of children between ages of 0 to 5 are within a 100 m from crèches.
Results and discussion
MCE investigating disaster vulnerability
The following alternatives were derived for types of exposure to hazards, based on responses to the questionnaire:
Fig. 3: Two-metre contour lines overlaid on a vulnerability surface.
After discussions with VPUU, who represented the community for input on this analysis, the levels of exposure to these hazards in Monwabisi Park were ranked in order of preference. After completion of ranking, a pairwise comparison was carried out in order to derive weights for each alternative. In this analysis, the highest weight was allocated to the best case scenario and the lowest weight to the worst case scenario. Disaster vulnerability weights are illustrated in Table 2. The weights were then allocated to the individual households based on their responses. For instance, if a particular household experienced no fire, a weight of 0,246 was allocated to that household. Fig. 2 shows the resulting map, covering section M, C, B and A of Monwabisi Park (these sections were demarcated by the community), which used a nearest neighbour interpolation method to create a raster vulnerability surface.
Like most informal settlements in Cape Town, the residents of Monwabisi Park are exposed to fire and flooding. This finding was in agreement with that of the study carried out by Musungu et al. [13] and Douglas et al. (2008). Fig. 2 shows the vulnerability of households to flooding and fire. It was noted that the majority of households were prone to flooding, and that they were also randomly dispersed.
In discussion with VPUU, it emerged that heavy rainfall contributes to flooding as the material used to construct households is of poor quality. These households are constructed from timber and corrugated metal. It appears that households nearby the surrounding main roads are regularly flooded during heavy rainfall, due to poor construction material. This is the case because the roads are situated on higher altitude terrain (Fig. 3), and a proper drainage system along the roads assists in channeling storm water away from them, towards the shacks in Monwabisi Park.
| Exposure to hazards | |
| Alternatives | Weights |
| No disaster | 0,492 |
| Only fire | 0,246 |
| Only flooding | 0,164 |
| Fire and flooding | 0,098 |
| Sum | 1,000 |
Buffer analysis and thematic mapping investigating fire disaster
After being advised by VPUU, fire disasters were categorised into two groups. These are: fires that occurred where formal electricity was used (this constituted 26% of households that experienced a fire); and fires that resulted where informal electricity was used (this constituted 74% of households that experienced a fire). An informal electricity connection is the result of electricity being illegally acquired.
Thematic mapping was used to represent different combinations of factors contributing to fire, and this is shown in Fig. 4. Houses that did not experience fire were colour coded according to the energy source they utilised. Light blue and dark blue polygons represented houses that did not have a formal electricity connection, whereas yellow polygons represented households that had formal electricity. Households that experienced fires were either coded as pink (with formal electricity) or red (with informal electricity or used flammable material such as paraffin, gas or candles for lighting or cooking). Fifty percent of Monwabisi Park households used formal electricity, most of them being residents who arrived first at the settlement.
A 100 m buffer zone was executed around a household that experienced fire, to check the number of households that would be affected if the spread of fire is within this range. There was a correlation between fire and flammable energy sources used at households (Fig. 4).
Fig. 4: 100 m buffer, fire disasters mapped to the type of energy source used. Section M, C, B and A of Monwabisi Park.
Thiessen polygons, distance matrix and statistical analysis investigating crèche accessibility
Households were identified that had children between the ages of 0 to 5 years (with an assumption that all these children went to crèche). A Thiessen polygon analysis was employed to identify the number of these households that could be accommodated by the nearest crèche. This analysis was combined with a distance matrix, showing the distances that these children would have to travel (by foot in the majority of cases) to get to their nearest crèche. The distance matrix results are presented in Table 3. It is evident from the mean distances that crèche 1316 has more households being located away from it, with an average distance of 327 m travelled, while crèche 6752 has more households closer to it, with an average travel distance of 102 m. Moreover an additional crèche is needed in the vicinity of crèche 1703 to reduce the large number of children being accommodated by it.
Conclusions
GIS technology has the ability to assist in improving the quality, and efficiency of informal settlements. It offers a broad range of benefits, which are realised from data collection processes and the display of results as maps, graphs and tables.
Fig. 5: Thiessen Polygons, 0-5 years old children households and crèches.
Five different methods of GIS spatial analysis were applied on Monwabisi Park spatial data, in order to assist with decision making for urban upgrading. The different types of analyses were buffer analysis, Thiessen polygons, MCE, distance matrix and thematic mapping. The resultant maps produced different results, depending on the type of spatial analysis employed and the purpose of the resultant map.
During the analysis, the attribute data and the polygons that did not join were excluded, representing a waste of effort. A more sustainable and reliable method of capturing data from the communities should be developed, so as to prevent wastage in the data collection process.
Discussions with VPUU assisted the principal author to get a clear understanding of the data captured, moreover the indigenous knowledge that the residents had about the study area assisted in the production and interpretation of maps. This study assisted VPUU in understanding the complexity of the participatory data collection process and its flaws. This could help them to create a more robust data collection process. The data analysis, on the other hand, showed VPUU and their community partners the power of GIS analysis. Being able to identify trends at different scales allows for a “big picture” understanding of various factors (such as vulnerability across the whole informal settlement), together with a finer understanding at an individual level (such as the feasibility of specific crèche locations).
This study has confirmed that useful analysis can be achieved through interaction between GIS professionals and communities. The knowledge and participation of communities could be used to produce sophisticated mapping and analysis, and ultimately could assist in the upgrading of the community.
| Thiessen Polygons | Distance matrix | |
| Crèche | % of 0-5 years old households within polygon | Mean (metres) |
| 1316 | 6% | 327 |
| 1703 | 22% | 216 |
| 2568 | 8% | 149 |
| 2774 | 21% | 141 |
| 5528 | 9% | 121 |
| 6752 | 10% | 107 |
| 5791 | 20% | 267 |
| 6476 | 4% | 102 |
Acknowledgment
This paper was presented at AfricaGEO 2014 and is republished here with permission.
References
[1] P Zeilhofera and V Topanottib: GIS and ordination techniques for evaluation of environmental impacts in informal settlements: A case study from Cuiabá, central Brazil‘, Applied Geography, Vol. 28, no. 1, pp. 1–15, 2008.
[2] M Huchzermeyer: A legacy of control? The capital subsidy for housing, and informal settlement intervention in South Africa‘, International Journal of Urban and Regional Research, vol. 27, no. 3, pp. 591–612, 2003.
[3] J Abbott: The Use of GIS in Informal Settlement Upgrading: Its Role and Impact on the Community and on Local Government‘, Habitat International vol. 27, no. 4, pp. 575-593, 2003.
[4] SK Nayak, SB, Thorat, and NV Kalyankar: GIS: (Geographic Information System) An application for socio-economical data collection for rural area‘, (IJCSIS) International Journal of Computer Science and Information Security, vol. 7, no. 3, March 2009.
[5] B Shoba and K Ra sappan: Application of GIS in Solid Waste Management for Coimbatore City‘, International Journal of Scientific and Research Publications, vol. 3, no. 10, pp. 1-4, 2013.
[6] A Toosi, M Delavar, H Rezayan: Spatial Development Infrastructure Linkages with urban planning and Infrastructure management‘, FIG working week April 16-21, Cairo, Egypt, 2005
[7] K Musungu, and S Motala: Maps and blocks: Using GIS to enhance community participation in planning informal settlements‘, GISSA Ukubuzana 2012 Conference, Kempton Park, 2 – 4 October, 2012.
[8] J Abbott, and D Douglas: The use of longitudinal spatial analyses of informal settlements in urban development planning‘, Development Southern Africa vol. 20, no. 1,pp. 3-19, 2003.
[9] M Laituri: The Issue of Access: An Assessment Guide for Evaluating Public Participation Geographic Information Science Case Studies. ‘Journal of the Urban and Regional Information Systems Association, Vol. 15, no. 2, pp. 25–31, 2003.
[10] J Quan, N Oudwater, J Pender and A Martin: GIS and Participatory approaches in natural Resources Research‘, Socio-Economic Methodologies for Natural Resources Research. Best Practice Guidelines. Natural Resources Institute: Chatham, UK. Available at: www.nri.org/publications/bpg/bpg09.pdf. Accessed on 12 December 2013, 2001.
[11] M Brown-Luthango, P Makanga, and J Smit: Towards Effective City Planning – The Case of Cape Town in Identifying Potential Housing Land‘, Urban Forum, vol. 24, no. 2, pp. 189-203, 2013.
[12] O Dubovyk, R Sliuzas, and J Flacke: Spatio-temporal modelling of informal settlement development in Sancaktepe district, Istanbul, Turkey‘, ISPRS Journal of Photogrammetry and Remote Sensing, vol. 66, no. 2, pp. 235–246, 2011.
[13] K Musungu, S Motala, and J Smit: Using Multi-criteria Evaluation and GIS for Flood Risk Analysis in Informal Settlements of Cape Town: The Case of Graveyard Pond‘, South African Journal of Geomatics, Vol. 1, no.1, pp. 77-91, 2012.
[14] V Jovanovic, and A Njegus: The Application of GIS and its components in Tourism‘, Yugoslav Journal of Operations Research, vol. 18, no. 2, pp. 261-272, 2008.
[15] J Malczewski: GIS and Multi-criteria Decision Analysis, John Wiley and Sons, New York, 1999.
[16] S Yahaya, N Ahmad and RF Abdalla: Multi-criteria analysis for flood vulnerable areas in Hadejia-Jama‘are River basin, Nigeria‘, European Journal of Scientific Research, Vol. 42, no. 1, pp. 71–83, 2010.
[17] L Brabyn and C Skelly: Geographical Access to Services, Health (GASH): Modelling Population Access to New Zealand Public Hospitals‘, The 13th Annual Colloquium of the Spatial Information Research Centre University of Otago, Dunedin, New Zealand December 2nd-5th, 2001.
[18] MP Kwan, and J Weber: Individual accessibility revisited: Implications for geographical analysis in the twenty-first century‘, Geographical Analysis, vol. 35, no.4, pp. 341–353, 2003.
Contact Bosiu Lefulebe, Cape Peninsula University of Technology Cape Town, bosiulefulebe@gmail.com
