Demand for groundwater resources is drastically increasing with increasing population. Hence, management, assessment and planning of groundwater resources is very crucial. Remote Sensing (RS) and Geographic Information Systems (GIS) plays a major role in delineation, management, planning, assessment and monitoring of groundwater resources and related studies. The study area belongs to the Wilgerivier Formation, Middelburg Basin. This research entails mapping of groundwater recharge potential zones in the Wilgerivier Formation using the Weighted Index Overlay Analysis (WIOA) with an aid of RS and GIS. Different factor maps namely; geology, geomorphology, lineament density, drainage density and slope were generated using ArcGIS 10.4 software. To assign weights and scores to factor maps and classes of each factor map, respectively, the WIOA technique was used. The weights and scores assigned ranged from 1 to 10 depending on the importance to groundwater recharge occurrence. The factor maps were integrated to produce a groundwater recharge potential map which was categorized into five distinct zones; excellent, good, moderate, poor and very poor. The map generated was validated with existing yield data of existing boreholes in the study area. The output map will help researchers, local authorities and planners in formulating proper planning, management and sustainability of groundwater resources for future generations. Freshwater is one of the most important resources on earth, which make it hospitable to life since living things depend on it. It covers about 2.5% of the earth surface with about 0.75% of it stored as groundwater. Groundwater is a dynamic and replenishing natural resource but its availability is limited in hard rock terrains. However, with drastic increases in population, the demand for freshwater, exceeds the supply and, this leads to an increase in over-exploitation of groundwater (Dhakate et al., 2008; Arkropovo et al., 2012). This puts a strain on natural freshwater resources since more of it is needed to meet the demands of not just for domestic activities but also for industrial and agricultural activities (Arkropovo et al., 2012).
South Africa is one of the semi-arid countries which receive minimal annual rainfall, therefore a substantial amount of groundwater extraction is needed to supplement available freshwater resources to meet the demand (Van Camp et al., 2013). Freshwater in the Middelburg Basin, which is the site for this study is located in Mpumalanga Province, remains a critical yet scarce resource and needs to be properly managed. Demand for water exceeds water supply which places a burden on water management agencies. With the implementation of the groundwater reserve, remaining available water resources need to be properly managed to prevent over-exploitation of water resources, thereby compromising the groundwater reserve (Mpumalanga State of the Environment Report, 2003).
To ensure sustainability of water resources for generations to come, exploration of groundwater recharge potential zones is vital. The conventional methods used to delineate groundwater recharge potential zones in hard rock terrains have proven to be time-consuming. Therefore, with the integration of geologic and hydrogeological surveys through weighted index overlay analysis (WIOA) with the aid of Remote Sensing (RS) and Geographic Information System (GIS) techniques, groundwater mapping in hard rock terrains may be tackled in a rapid and time-effective manner.
The combination of RS and GIS techniques is very efficient for delineating groundwater prospecting zones anywhere. Remote sensing is one of the main techniques for obtaining surface feature data of groundwater; hence, data can be processed in GIS to delineate groundwater prospecting zones (Saraf and Choudhury, 1998).
There are several factors that need to be taken into consideration for delineating groundwater resources, as they affect the occurrence, origin and movement of groundwater. These factors may include slope, geology, lineament density, drainage density, geomorphology, and the interrelationships among them. There are several parameters that need to be taken into consideration for delineating potential groundwater recharge zones as they affect the occurrence, origin and movement of groundwater from the surface and these include topography, lithology, soil texture, lineament density, drainage density, geomorphology, land cover, rainfall and the interrelationships among these factors (Jasrotia, 2007; Chowdhury et al., 2009; Chowdhury et al., 2010; Gupta and Srivastava, 2010; Elewa and Qaddah, 2011; Arkropovo et al., 2012; Hammouri, 2012; Awawdeh et al. 2014). According to Zaidi et al. (2015), geomorphology describes land forms and their features, and topography is one of the geomorphological features that influences runoff and infiltration. Topography is one of the parameters that needs to be taken into consideration when delineating groundwater recharge potential zones since it tells us more about the surface features and shape of an area to be mapped. In areas with steep slopes there is usually high runoff leading to less or no recharge for groundwater; however, in areas with gentle slopes there is more infiltration, less runoff leading to high recharge for groundwater. Water resources are mostly found in flat slopes. Slope/ elevation maps may be acquired from satellite imagery such as Digital Elevation Model (DEM) data from Shuttle Radar Topography Mission (SRTM) (Kaliraj et al., 2014).
It is also important to know the types of rocks (geology) found in an area so that we could deduce possible aquifers. For instance, permeable, highly fractured and more porous rocks make good aquifers like sandstone, conglomerates, fractured limestone, unconsolidated sand and gravel as well as columnar basalts. This information may be obtained from Geological Survey of India (GSI) maps.
Study of soil texture maps helps to identify which soils are more capable of holding most water, thereby might make good aquifers. Sand particles are more porous and permeable compared to clay particles since some have large sand particle sizes and the space between the particles is also larger compared to that of clay, therefore can store more water and could make a good aquifer. However, clay particles have low permeability and are less porous compared to sand particles due to small particles sizes that stack together, therefore, limiting the size and connectivity of the pore spaces leading to no or less water being stored. Studies have shown that loamy and sandy soils are more suitable for groundwater recharge, thus make good groundwater recharge potential zones (Zaidi et al. 2015). The aforementioned information about soil texture may be obtained from soil maps. Study of soil texture maps helps to identify which soils are more capable of holding most water, thereby might make good aquifers. Sand particles are more porous and permeable compared to clay particles since some have large sand particle sizes and the space between the particles is also larger compared to that of clay, therefore can store more water and could make a good aquifer. However, clay particles have low permeability and are less porous compared to sand particles due to small particles sizes that stack together, therefore, limiting the size and connectivity of the pore spaces leading to no or less water being stored. Studies have shown that loamy and sandy soils are more suitable for groundwater recharge, thus make good groundwater recharge potential zones (Zaidi et al. 2015). The aforementioned information about soil texture may be obtained from soil maps. Locations of linear features (lineaments) such as faults, folds and dykes is important to know whether the underlying rock has any secondary permeability or not (Pothiraj and Rajagopalan, 2013). For these features to store and transmit groundwater, the presence of fractures and joints is necessary (Pinto et al., 2017). Therefore, studies have shown that areas with higher lineament density would make possible aquifers (Pothiraj and Rajagopalan, 2013). This information data can be extracted from both aerial photographs and satellite imagery.On steep impermeable soil or rocks, surface runoff occurs since infiltration does not take place; therefore, higher drainage density. However, gentle permeable soils or rocks have less runoff and more infiltration leading to low drainage density. According to Chowdhury et al. (2009) and Chenini et al. (2010), areas with high drainage density tend to make good groundwater recharge potential zones since runoff occurs instead of infiltration and this give rise to a well-developed drainage system. However, Rahmati et al. (2015) argue that actually areas with low drainage density make good groundwater recharge potential zones since infiltration exceeds runoff. With the use of Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) DEM through spatial analyst tool in ArcGIS we can acquire information about drainage density of an area. Study of various features at the surface of the earth helps to identify zones or areas with potential for groundwater recharge. These features include agricultural land, water bodies, forest plantation, waste land, and built-up land. However, the most common features are vegetation and concrete surfaces (Dar et al., 2011). Vegetation enhances infiltration and reduces runoff; hence, more groundwater recharge. However, concrete surfaces enhance runoff and reduce infiltration, hence, no groundwater recharge. Therefore, the presence of vegetation makes good groundwater recharge potential zone of an area (Kaliraj et al., 2014). Land cover data may be obtained from both field survey and analysis of remotely sensed imagery. Rainfall is also considered to be one of the most important factor that influences groundwater recharge since groundwater recharge mainly comes from rainfall. Less or no rainfall leads to no recharge, whereas more rainfall leads to more recharge. Therefore, groundwater resources are more likely to occur in areas that receive more rainfall (Pinto et al., 2017). The rainfall data can be obtained from the Meteorological Department. Studies over the past years have shown that methods used to delineate groundwater recharge potential zones using GIS and RS have been almost the same. Each parameter is evaluated using different RS and GIS techniques depending on the geological features and scale of the area to be mapped (Arkoprovo et al., 2012). Arkoprovo et al., (2012) stated that with the use of ArcGIS software, the most common datasets include the use of Landsat Thematic Mapper (TM) data, Digital Image Processing (DIP) and Survey of India (SOI) Toposheets on a 1:50 000 scale. These data have been used for analysis and mapping of the individual layers; hence, integrated to produce a final map showing the possible groundwater recharge zones (Abdalla, 2012). Landsat TM consists of seven bands and produce high image resolution for the derived layers whereas the SRTM DEM provides the most complete high-resolution digital topographic data of a particular area. Over the years, Landsat TM has been enhanced to Landsat Enhanced Thematic Mapper Plus (ETM+) which has eight spectral bands consisting of an additional panchromatic band which is sensitive to a wide range of wavelengths of light, typically most of the visible spectrum (Awawdeh et al., 2014; Adham et al., 2010).
The thematic layers are generated from Indian Remote Sensing Satellite using Linear Imaging and Self-Scanning Sensor (IRS LISS III) on a 1:50 000 scale. Each layer is processed and analysed using the Spatial Tool Analysis (STA) on ArcGIS software.
Some of the most common decision-making techniques that researchers have used to assign weights to each factor maps and integrate them include; watershed modeling system, weighted spatial probability modelling, weighted index overlay analysis (Jasrotia et al., 2007; Nag and Ghosh, 2013; Danee and Helen, 2014), sensitivity analysis, analytical hierarchy process (Srivastava and Bhattacharya, 2006; Pradhan, 2009; Chowdhury et al., 2010; Gupta and Srivastava, 2010; Elewa and Qaddah, 2011; Hammouri et al., 2012; Awawdeh et al., 2013; Al-Abadi and Al-Shamma’a, 2014; Kaliraj et al., 2014; Mallick et al., 2015; Rahmati et al., 2015; Razandi et al., 2015; Jain and Singh, 2017), Boolean logical analysis (Dar et al., 2011; Pothiraj and Rajagopalan, 2013; Zaidi et al., 2015) and spatial modelling, multi-influencing factor, evidential belief function model (Nampak et al., 2014; Pourghasemi and Beheshtirad, 2015) were added to enhance the validity of the results (Gumma and Pavelic, 2013; Awawdeh et al., 2014; Jain and Singh, 2017).
Factor maps are converted to raster format and then assigned map weights and class weights to find the product of these weights and then classify them according to the influence each layer has on groundwater recharge (Dar et al., 2011). This is done for each decision-making technique used, for instance;