Developing suitability maps for rainwater harvesting in South Africa

https://doi.org/10.1016/j.pce.2008.06.047Get rights and content

Abstract

Dry spells are a direct consequence of spatial and temporal variability of rainfall, and these jeopardise the success of rainfed agriculture by causing crop yield reduction and crop failure in rural South Africa. The potential of rainwater harvesting (RWH) to mitigate the spatial and temporal variability of rainfall has brought about its revival during the last two decades. For planning and implementation purposes, it is critical to be able to identify areas suitable for RWH. The paper presents a methodology that enable water managers to assess the suitability of RWH for any given area of South Africa. Previous methodologies developed to assess RWH suitability recognised the importance of the socio-economic factors but did not incorporate them in their assessment. This non-integration of socio-economic factors is pointed as the main cause of failure of rainwater harvesting projects. In this study, in-field RWH and ex-field RWH suitability maps are developed based on a combination of physical, ecological and socio-economic factors. Model Builder, an extension of ArcView 3.3 that enables a weighted overlay of datasets, is used to create the suitability model, comprising the physical, ecological and vulnerability sub-models from which the physical, the ecological and the vulnerability maps are derived respectively. Results indicate that about 30% is highly suitable for in-field RWH and 25% is highly suitable for ex-field RWH. Details of the proposed method as well as the suitability maps produced are presented in this paper. The implementation of this method is envisaged to support any policy shifts towards wide spread adoption of RWH.

Introduction

To improve the reliability of rural water supply and the productivity of small-scale rainfed agriculture, South Africa needs to further investigate unconventional water sources. Aridity and climatic uncertainty are the major challenges faced by small-scale farmers who rely on rainfed agriculture as the low crop yield they experience is mainly attributed to poor temporal and spatial rainfall distribution rather than acute water shortage.

Rainwater harvesting (RWH) is described as the collection, storage and use of rainwater for small-scale productive purposes. It has been identified at a number of international fora as one of the important interventions necessary towards meeting the Millennium Development Goals in Africa. RWH enhances water productivity by mitigating temporal and spatial variability of rainfall (Mwenge Kahinda et al., 2007a, Rockström and Barron, 2007) and provide water for basic human needs and other small-scale productive activities (Mwenge Kahinda et al., 2007b). In areas with dispersed populations or where the costs of developing surface or groundwater resources are high, RWH and storage have proved to be an affordable and sustainable intervention (Mati et al., 2006).

Three broad categories of RWH can be distinguished when it is classified according to the type of catchment surface used: in-field RWH (IRWH), ex-field RWH (XRWH), and Domestic RWH (DRWH). DRWH systems collect water from rooftops, courtyards, compacted or treated surfaces, store it in RWH tanks for domestic uses. IRWH systems use part of the target area as the catchment area, while XRWH systems use an uncultivated area as its catchment area. The focus of this paper is on both IRWH and XRWH that will refer to as field RWH (FRWH).

Although non-governmental organisations, faith-based groups and networks are advocating the use of RWH, its adoption rate is slow. The reasons might be that, firstly, there is inadequate attention paid to social factors (Patrick, 1997, Rockström, 2000) and secondly, there is lack of scientifically verified information which can be used to indicate areas where rainwater harvesting can be applied (Mati et al., 2006). DRWH is the least used water source in South Africa with only 1% of rural households currently using it as their main water source (Mwenge Kahinda et al., 2007c). IRWH is mostly practiced at household level in the backyards while XRWH is rarely implemented. The South African Agricultural Research Council (ARC) has had a programme of IRWH in the Taba Nchu area for over a decade but the technique has not extended beyond small plots around homestead with total adoption of about 0.5% of the Upper Middle Modder River Basin.

FAO (2003) lists six key factors when identifying RWH sites: climate (rainfall), hydrology (rainfall–runoff relationship and intermittent watercourses), topography (slope), agronomy (crop characteristics), soils (texture, structure and depth) and socio-economic (population density, work force, people’s priority, experience with RWH, land tenure, water laws, accessibility and related costs). A number of studies present methods for assessing RWH suitability of a given area. Those studies commonly use physical factors such as rainfall, land cover/use, soil characteristics and topography for the assessment of suitability. For instance, Mbilinyi et al. (2006) used rainfall, soil depth, soil texture, differential global positioning system points, aerial photos, ground truthing and topographic maps while Mati et al. (2006) used baseline thematic maps (rainfall, topography, soils, population density and land use) to produce composite maps that show attributes or “development domains” that serve as indicators of suitability for targeted RWH interventions. To determine index maps of RWH potential (ponds) in Africa, Senay and Verdin (2004) used runoff data derived from rainfall data using the SCS curve number methods and delineated watersheds from the Africa-wide Hydro-1K digital elevation model. Mou et al. (1999) used rainfall (average rainfall, annual rain days, annual rainfall fluctuation rate) and topography. Prinz et al. (1998) used remote sensed data and thematic maps in conjunction with field investigations. In his iterative decision tree, Patrick (1997) incorporated socio-economic factors after the mapping. The need to integrate socio-economic factors into deciding the suitability of an area to RWH largely underscores the failure or success of RWH systems (Critchley and Siegert, 1991, Oweis et al., 2001).

The quality, reliability and availability of data often limit the setup of Geographic Information systems (GIS). Errors in the final results may originate from any stage of the process; from the collection of the source data to the interpretation of the final results (Store and Kangas, 2001). Also error propagation or the accumulation of errors from various sources affects the results of analysis. Therefore, the accuracy of the RWH suitability maps depends on the quality of the different layers used as well as the quality of the spatial data analysis.

This paper presents a GIS-based model, which combines physical, ecological and socio-economic attributes, to assess the suitability of a given area for FRWH in South Africa.

Section snippets

Methods

The RWH suitability model (RSM) was developed using model builder, an extension of ArcView 3.3. The physical, ecological, socio-economic as well as constraint layers used in the RSM are listed in Table 1. Only the best available datasets at national level were used in the RSM. As model builder works in the raster environment with grid format layers, vector themes were converted into grid themes of cell size 100 m × 100 m.

Different suitability values were assigned to the physical, ecological and

The vulnerability map

The vulnerability map was generated by the WOP of the four socio-economic layers. The vulnerability map indicates that the most vulnerable communities are mainly located in the eastern part of the country, in Kwazulu–Natal and the Eastern Cape provinces (Fig. 7). Some vulnerable communities are also located in the Limpopo and the North-West provinces. It is in those communities that interventions such as RWH are most needed.

The ecological map

The ecological model (Fig. 8) indicates areas to the South-West and

Conclusion and recommendations

The objective of this study was to develop a GIS-based rainwater harvesting model (RSM) that combines through a MCE process the physical, ecological, socio-economic and constraint layers. The model produces three types of RWH maps for IRWH and XRWH: physical maps, potential maps and suitability maps.

The RSM has a high degree of customizability in that it enables the user to add, remove layers and change the relative importance weights of the layers. It should be noted that determining the

Acknowledgments

This study is part of the Water Research Commission funded project K5/1563: “Water Resources Management in Rainwater Harvesting an integrated systems approach”. The authors would like to thank the Department of Water Affairs and Forestry, Statistics South Africa, the Agricultural Research Council (ARC), the Council for Scientific and Industrial Research (CSIR), the Smallholder system innovations in integrated watershed management: (SSI-program) as well as the School of Bioresources Engineering

References (33)

  • DWAF, 2004. National Water Resources Strategy. Our Blue Print for Survival, first ed. Department of Water Affaires and...
  • FAO, 2003. Land and Water Digital Media Series, 26. Training Course on RWH (CD-ROM). Planning of Water Harvesting...
  • P.H. Gleick

    Basic water requirements for human activities: meeting basic needs

    Water International

    (1996)
  • T. Hoffman et al.

    Nature Divided: Land Degradation in South Africa

    (2001)
  • J. Malczewski

    GIS and Multicriteria Decision Analysis

    (1999)
  • Mati, B., De Bock, T., Malesu, M., Khaka, E., Oduor, A., Meshack, M., Oduor, V., 2006. Mapping the Potential of...
  • Cited by (137)

    View all citing articles on Scopus
    View full text