Water resources, especially groundwater, are a valuable component of the natural hydrological process, as they are stored below the water table in the voids of rocks and soil in the main areas of the earth's crust [1,2]. In many regions of the world, groundwater is the main source and is extensively used for drinking, household, industrial, and agricultural needs [3,4]. It is the safest and most dependable supply of water, utilized for household, agricultural, industrial, and municipal needs. According to Nebiyou and Mebruk [5], Ethiopia, one of the most hydrologically promising nations in East Africa, is thought to contain a sizable groundwater potential reserve. Lithology, geomorphology, drainage density, geology, slope, drainage network, land use/land cover pattern, and meteorological conditions are some of the variables that affect groundwater potential [6-8].

Thus, it is essential to evaluate and map the zones that have the potential for groundwater using GIS and remote sensing. Regarding the management and development of groundwater, in-depth knowledge of aquifers and their potential mapping is essential. The groundwater potential zones can be mapped and assessed using a variety of approaches. Numerous nations have evaluated groundwater potential zone mapping using different techniques [9-12]. However, the traditional methods, including hydrogeological fieldwork and geophysical surveys, take time and are too expensive [13].

Recent years have seen an increase in the use of geographic information systems (GIS) and remote sensing (RS) for mapping, generating valuable data quickly and inexpensively, and identifying groundwater potential zones that provide massive scale in space-time and save time and money [14-16]. It can produce data in the spatial and temporal domains, which has a significant role in successful analysis, prediction, and validation. It can also give all the parameters that affect a region's groundwater potential zones. The huge amount and quality of data processed in geospatial software produce groundwater potential zones [17,18].

Additionally, by combining multi-criteria decision analysis (MCDA) with RS and GIS approaches, AHP has been successfully implemented in various studies for groundwater recharge potential zone mapping and water resource management [19,20]. Numerous studies with encouraging findings have effectively combined RS, GIS, and AHP methods to map groundwater potential zones [21-24].

In Ethiopia, groundwater is not adequately used due to higher development and operational costs and a lack of understanding of resource dynamics [25]. The LULC, or climate change, brought challenges and the loss of available surface water, which alarmingly increased the demand for groundwater [26-28].

The dramatic increment in human population, the LULC change, the dryness of deep and shallow springs and wells, and limited research studies brought unwise utilization and declines in groundwater potential in Bale Zone, Genale-Dawa Sub-basin, creating competition over surface available water sources for multi-purpose. Additionally, alleviating problems in water demand and failure related to groundwater exploitation is vital within the study area. Furthermore, the lack or limited research-based study using the integrated geospatial techniques (GIS and RS) with multi-criteria decision analysis (MCDA-AHP) is a main limitation for the planner, decision-makers, investment, management options, selecting suitable sites for drilling new boreholes, and current status of groundwater potential in Bale Zone, Genale-Dawa Sub-basin. 

Description of the Study Area

The research region is located 430 kilometers from Addis Ababa, the capital city of Ethiopia, in the Oromia Regional State's Bale Zone in the southeast of the country. The Genale-Dawa basin contains the Bale Zone Genale-Dawa Sub-Basin. According to Figure 1, the research area covers 1,510,426.32ha and is located between the latitudes of 5°57'40"N and 7°33'30"N and the longitudes of 39°53'50"E and 41°19'50"E. The majority of the districts, including Agarfa, Dinsho, Sinana, Gobba, Goro, Ginnir, Dawe Kechane, Gasara, Gololcha, Berbere, Delomana, Sawena, and Raitu, are covered by this Bale Zone Genale-Dawa Sub-basin, which has elevations ranging from 670 m to 4463 m above mean sea level (amsl).

Figure 1: Map of the Study Area Bale Zone Genale Dawa Sub-Basin.

Topography

Between the sub-basin upstream and downstream ends, there is a significant height difference. Because of its unusual topographical steepness and importance as a supply of water for the Bale zone, other regions, and countries further downstream, the uppermost part of the sub-catchment needs to be safeguarded with great care. Bale has a wide range of physiographic features. It is made up of flat-topped plateaus, lowlands, mountainous terrain, deeply cut river valleys, and deep gorges. Southeast Rayitu, Guradamole, and Dawe Qachen are the surfaces that rise from below 300 meters above sea level to high ranges that culminate in Tulu Dimtu, the highest peak in the area at 4377 meters. The Sannate plateaus (Bale Mountain National Parks) and Mount Tulu Dimtu are incorporated into the high land plateaus. Flat plains, river basins, and gorges are all features of the lowlands, which are divided by hills and ridges.

Climate and Agro-Ecologies

Bale Zone is separated into Dega (highlands), Waine Dega (midlands), and Kola (lowlands) according to topography. It also has a bimodal rainfall pattern. In accordance with this, the region has two cropping seasons: Ganna (March to June) and Bona (July to December). It displays extensive temporal and geographical climate variability, which is mostly influenced by variations in height. The large highland plateau and surrounding mountains are known for their cool climate and heavy rains, and high peaks like the Sanetti plateau and Tullu Dimtu may have winter snowfalls.

A tropical, hot, and dry climate predominates in the lowlands and farther south of the mountains. The region has a bimodal local climate, with two wet seasons that feature both heavy and light precipitation. The bimodal rainfall pattern has light rains from March to June with a peak in April and strong rains from July to October with the highest peak in August. In the region with the monomodal pattern, there are typically four dry months (November–February) and eight rainy months (March–October). Annual rainfall in this lower-altitude area ranges from 600 to 1000 mm, while it ranges from 1000 to 140 mm in higher-altitude places. In the course of the dry season, there is a lot of variation in the daily temperatures. 18.4°C is the average annual high temperature, and 1.4°C is the average annual minimum.

Soil Types

Soils, the result of climate, topography, and geology, greatly control the rate of infiltration and infiltration into aquifers. Sub-basin soil type maps were clipped from digital soil maps [40] using ArcGIS 10.3 software (Figure 2 and Table 1). 

Figure 2: Soil Type Map of Bale Zone Genale Dawa Sub-Basin.

Table 1: Soil Types and Its Area Coverage of Bale Zone Genale Dawa Sub-Basin.

Methods

Geospatial techniques and MCDA (AHP) were applied to create a map of groundwater potential for the study area. This research work includes criteria identification, evaluation, pre-processing, reclassification, pairwise comparison of criteria, weight assignment, and ranking using the AHP process and ArcGIS 10.3 software. Finally, groundwater potential zone maps for the Bale Zone Genale-Dawa sub-basin were developed using weighted overlay analysis in ArcGIS 10.3 software.

Data Collection and Description

This section describes the data sources, purpose, description, and data processing techniques used to establish the study area's groundwater potential zones. The required data were collected from various government agencies, field surveys, and satellite imagery published on the United States Geological Survey (USGS) Earth Explorer website. All data were resampled after acquisition and processing to a spatial resolution, row, and column sample suitable for overlay analysis of groundwater potential maps and descriptions (Table 2).

Table 2: Summary of Data Collected Descriptions.

Types of Software

Different software was used for data pre-processing, preparation, data analysis, editing, and the final output of the zone where groundwater is possible. Generally, detailed descriptions of the software used and their purpose in the groundwater potential zones map were described (Table 3).

Table 3: Types and Purposes of Software.

Factor Identification and Preparations for the Groundwater Potential Map

Rainfall

The rainfall map was created using an annual average of 40 years (1981-2021) of historical rainfall data collected from 11 nearby weather stations, and the Ethiopia National Meteorological Agency (Figure, 3). Precipitation data were spatially interpolated using the IDW interpolation method using ArcGIS 10.3 software to obtain rainfall distribution maps. Similarly, the IDW interpolation method has been adopted by several authors due to the uneven distribution of stations [25,29-33]. Finally, the interpolated rainfall data were classified using this IDW interpolation technique and then divided into five classes, and weightings were assigned based on intensity and groundwater potential as the standard suggested by [33] (Table 4).

Figure 3: Meteorological Stations Points.

Geomorphology

The Bale Zone's Genale-Dawa sub-basin geomorphological features were clipped from the geomorphology map of a geological survey of Ethiopia. Based on the views of groundwater potential, geomorphological classification, weight, and ratings were made according to the standard rate suggested by Bane [19] (Table 4).

Land Use Land Cover

Landsat 8 was downloaded from the United States Geological Survey (USGS) for the study area to create a LULC map, added to ERDAS 2015 software, processed for image pre-processing, and integrated into ArcGIS 10.3 software. According to Mahalingam and Vinay [34], standard LULC was classified into five classes based on groundwater potential (Table 4). 

Lithology

The lithology map was developed from a 1:2,000,000 geologic map published by an Ethiopian geologic survey. These maps were geo-referenced and clipped to the study area’s shapefile. The shapefile for the lithological units inside the study area was sketched to create a vector layer, and the vector layer was converted to a raster layer of the same in ArcMap 10.3. According to the possibility of groundwater potential points of view, lithological classification, weight assignment, and ranking were conducted as standard rates suggested by Takorabt et al. [35], Fauzia et al. [36], and Abdessamed et al. [37].

Soil Texture

Soil samples from 0 to 20 cm in depth were collected using a stratified random sampling technique using Auger sampling points (Figure 4). Soil samples were air-dried, ground using a mortar and pestle, and passed through a 2 mm mesh sieve. Soil texture analysis was performed at the Sinana Agricultural Research Center Soil Laboratory using the Bouyoucos hydrometer method [38]. Finally, soil texture classes were assigned using his USDA classification system of texture triangles [39]. The laboratory analysis results were further encoded, and IDW was spatially interpolated using ArcGIS 10.3 software to obtain the soil texture map. Finally, the soil texture was reclassified into five classes [19,31,40,41] (Table 4). 

Slope

Slope maps were developed from Shuttle Radar Topography Missions (SRTM) DEM 30m resolution downloaded from USGS using ArcGIS 10.3 software. As a result, the slope maps were rearranged into five classes according to his ranking of groundwater potential suitability. Then the slopes were classified by degrees according to the standard set by Julla et al. [33] (Table 4). Therefore, the lower the slope, the higher the potential of groundwater and the lower the runoff, hence the higher rank. 

Topographic Wetness Index

TWI (Topographic Wetness Index) was calculated from SRTM DEM 30 m spatial resolution using ArcGIS 10.3 software. Finally, classification, weight, and ratings were made according to the standard rate set by Abdessamed et al. [37] as per the suitability groundwater potential (Table 4).

The lowest rank was assigned to low TWI values, and the highest rank was assigned to high TWI values, indicating a trend of soil moisture accumulation.

Elevation

Elevation was classified by the Shuttle Radar Terrain Mission (SRTM) with a spatial resolution of 30 m using ArcGIS 10.3 software. Next, classification was conducted into five categories based on groundwater potential standards as stated by Mojtaba et al. [42] (Table 14). 

Drainage Density

The DEM was used to extract the study area's drainage density map at 30 m spatial resolution using a boundary shapefile after filling with ArcGIS 10.3 software. The resulting maps of drainage density were classified into five categories as suitable for groundwater potential according to standard rates given by Abdessamed et al. [37] (Table 4). Drainage density (Dd) was calculated according to the following equation (1): 

Where,represents the length of drainage and A represents the area of catchment.

Lineament Density

Lineament densities were calculated using Landsat-8 using the Geomatica (Principal Component Imaging) (PCI) 17 software supporting ArcGIS 10.3 software. Similarly, the method and procedure for extracting lineaments from Landsat 8 OLI using ArcGIS software and PCI Geomatica 17 version integration have been adopted by several authors in previous studies [25,27,30,33,36,43-48]. Finally, the lineament density maps were categorized into five categories as a basis for the groundwater potential given by Muhammad et al. [47] (Table 4). Therefore, low weights have low linear densities, and high weights have high linear densities. Lineament density (LD) was calculated as follows (equation 2):

Where, represents the length of lineament lines, and A represents the area of catchment.

Table 4: A Standard Classification Rate and Ranks of Factors Determines Groundwater Potential.

Analytical Hierarchy Process

They were based on multi-criteria decision analysis (MCDA) using the Analysis Hierarchy Process (AHP), and the thematic layer maps were weighted. The GIS software for the groundwater potential zones map was integrated with an analytical hierarchical process (AHP). The various thematic layers selected include rainfall, geomorphology, LULC, lithology, soil texture, slope, elevation, topographic wetness index, drainage, and lineament density. The study used large-scale thematic layers that have a significant influence on the groundwater potential zones. The weighting of these factors was based on the literature review, expert opinion, and multi-discipline field survey local condition experience on groundwater resources. Comparisons were made utilizing the 1–9 scale, indicating how often one shift is more important than another. Saaty [49] shows the scaling used in AHP (Table 5). 

If the matrix formed is equal to bij, then aij = wi/wj, where w is the weight of each parameter, the element of all elements of each positive number i, j=1,...n, and the reciprocal property bnij = i/bij, what is called the matrix inverse. 

Table 5: Saatty's Scale of Intensity Relative Importance.

The consistency Index (CI), which defines the consistency coefficient of the pairwise comparison matrix, was estimated using (Equation 3).

The calculation of the consistency index relies on the λmax value using Equation 1 [49]. The weights of each factor were calculated by the pairwise comparison matrix, and the maximum eigenvalue (λmax) of the normalized matrix was calculated (Equation 4). 

A random consistency index (RI) served as a means of determining the degree of consistency, or a consistency ratio (CR) was calculated using (equation 5 and table 5).

Table 6: Random Consistency Index.

Weight Assignment and Normalization

Apply the AHP technique to normalize the weights assigned to different thematic layers. As shown in Table 6, a value of 1 indicates equal importance for the two factors, and a value of 9 indicates that one factor is very important compared to the other. According to Saaty [49], the tolerance or value of CR should be less than 0.1. 

Overlay-Weighted Analysis

A map of the study area's groundwater potential zones was mapped using the weighted index overlay method in ArcGIS 10.3. Weight assignment was done by assigning new weight values to map sub-units (sub-criteria) calculated from the AHP. The reclassified tools in ArcGIS 10.3 Spatial Analyst were used for this task. Finally, a map of groundwater potential zones was created by overlaying all thematic layers using the weighted overlay analysis tool.

Validation of the Groundwater Potential Map 

To confirm probable groundwater zones, groundwater inventory data from the regional, district, and Bale zone water and energy sectors was also gathered in addition to field survey data. As a result, the groundwater potential zones of the Bale Zone Genale-Dawa sub-basin were validated in the current study using a total of 100 well yield data points (Table 7). The observed groundwater data were mapped using ArcGIS 10.3 software, and the analysis was overlaid on the map of the groundwater potential zone. In this case, a higher overlay analysis indicates that the produced map is considered more dependable. Model reliability and well-yield data are also true indicators of potential zone availability. 

Similarly, several authors, including Andualem and Demeke [50], Berhanu and Hatiye [51]; Lentswe and Molwalefhe [52]; Murmu et al. [53]; Saravanan et al. [54]; Doke et al. [55]; Sapkota et al. [56]; and Sarwar et al. [57], used groundwater inventory data such as borehole data, wells, and hand digging yield to validate the developed groundwater potential zone.

Table 7: Groundwater (Springs and Wells) Yield Classification by Different Authors.

Groundwater Potential Mapping Criteria and Determining Factors

Rainfall

The mean rainfall map of the Bale zone of the Genale-Dawa sub-basin varies from 374.6 mm to 2236 mm and was classified into five classes based on the groundwater perspective: very low, low, moderate, high, and very high (Table 8 and Figure 4). Similarly, those with the highest rainfall were assigned the highest weights and had the highest groundwater potential, and vice versa. In this study, the highest area of about 501738.16 ha (33.22%) received rainfall varied from 1561.3 to 2236 mm, followed by an area of about 289588.83 ha (19.17%) rainfall ranged from 1090.8 - 1281 mm, considered very high, and moderate from groundwater potential perspective views (Table 8).  On the other hand, 233064.85 ha (15.43%) with rainfall ranging from 374.6 to 940.7 mm and 217454.12 ha (14.50%) with rainfall varying from 940.7 to 1090.8 mm were considered very low and low, respectively, from the groundwater potential point of view (Table 8). Several studies confirmed that higher rainfall leads to higher groundwater potential and vice versa [19,29,31,37,47,50,58,59]. 

Table 8: Rainfall Class and Its Rank as per Suitable for Groundwater Potential.

Figure 4: Rainfall Map of Bale Zone Genale-Dawa Sub-Basin.

Geomorphology

The geomorphology map of the studied Bale Zone Genale-Dawa sub-basin consists of five classes (Table 9 and Figure 5). Accordingly, volcanic landform, structural landform, residual landform, alluvial landform, and flat/flood plain have 509125.22 ha (33.71%), 20862.12 ha (1.38%), 523800.97 ha (34.68%), 3769.21 ha (0.25%), and 452666.57 ha (29.97%) area coverage, respectively (Table 9). Therefore, in terms of groundwater potential, alluvial landforms and flat/flood plain lands have high and very high, respectively, while volcanic landforms and structural landforms have very low and low, respectively (Table 9). Bane [19] reported similar potential groundwater conditions in the geomorphological categories of volcanic landform, structural landform, residual landform, alluvial landform, and flat/flood plain. This means that geomorphology is an important part of the groundwater potential as it describes zones of porosity and permeability. Several studies have also included geomorphological features that reflect different landforms and structural features as important factors in determining groundwater potential [1,19,44,45,60,61-63]. 

Table 9: Geomorphology Type and Its Rank as per Suitable for Groundwater Potential.

Figure 5: Geomorphology Typemap of Bale Zone Genale-Dawa Sub-Basin.

For full article please go through this link: https://www.pubtexto.com/pdf/?groundwater-potential-zones-assessment-and-mapping-using-integrated-geospatial-techniques-and-multicriteria-decision-analysis-mcda