Evaluation of the groundwater recharge potential zone by using GIS and remote sensing in Ziway Abijata sub-basin, Central Rift Valley of Ethiopia Open Access

About 71% of the Earth's surface is covered with water, including groundwater (aquifers) and surface water sources such as lakes, rivers, and reservoirs (Huang et al. 2021; Chandnani et al. 2022; Cheng et al. 2022; L. Zhang et al. 2022; Y. Zhang et al. 2022). Ethiopia is the water tower of Africa by having 12 river basins with an annual runoff volume of 122 billion m³ of water and an estimated volume of 2.6–6.5 billion m³ of groundwater potential (Varady et al. 2023). Groundwater is fresh water that flows within aquifers below the water table, is less vulnerable to pollution than surface water, and is communally used for public water supply (Suciu et al. 2020; Tamiru & Wagari 2022; W. Zhu et al. 2022). It is the most important natural water resource stored in a saturated underground zone and moves slowly in the form of aquifers (Yeh et al. 2016; Chen et al. 2022; Tamiru et al. 2022). Groundwater recharge occurs when additional water seeps occur into underground aquifers either incidentally or intentionally from the surface area (Mishra & Dubey 2015; H. Liu et al. 2022, 2023b). In addition, Alrawi et al. (2022) reported about 46% of high potential groundwater recharge zone and 54% and 49.4% areas of poor potential groundwater recharge zone in the Al-Qalamoun region of Syria by using the analytical hierarchy process (AHP) and the multi-influence factor (MIF) method, respectively.

The occurrence and intensity of groundwater recharge zone vary from place to place due to determinant factors like soil texture, infiltration capacity, precipitation rate, climate condition, and vegetation cover on the surface area (Mengistu et al. 2022; H. Liu et al. 2023a; Pei et al. 2023). About 185 billion m³ of groundwater held in sedimentary, volcanic, and Quaternary rocks covers 924,140 km² of Ethiopian highland and Rift Valley (Alemayehu et al. 2006). Consequently, Seifu et al. (2022) reported that about 84% of moderate groundwater potential zones, 14% of high groundwater potential zones, and 2% of low and high potential zones were identified in Fafen-Jerer of the Ethiopian sub-basin. Moreover, about 33.6% and 16.8% were classified as very good and good groundwater recharge potential zones, while about 23.3%, 20.2%, and 22.9% were classified as very poor, poor, and moderate groundwater recharge potential zones, respectively, in the Guder River Basin (Duguma & Duguma 2022). The previous studies indicate that the status and distribution of groundwater resources with their ecological, social, and economic aspects in different parts of Ethiopia (Wada et al. 2016; Xie et al. 2019). According to Berhanu et al. (2014), groundwater provides 90% of the industrial supply and around 70% of the rural water supply. In addition, pastoralists use groundwater for livestock watering and small-scale agricultural practices (Tamiru & Wagari 2021).

Groundwater plays a critical role in reducing food insecurity in drought-susceptible areas around the Ethiopian Rift Valley. The groundwater potential zone particularly around the Ethiopian Rift Valley faces significant challenges due to human-induced factors. Several studies highlighted the effects of climate change, land use land cover change (LULC), and agricultural drought on groundwater resources (Berhanu et al. 2014; Bambrick et al. 2015; Desta & Lemma 2017; Godebo et al. 2021; Li et al. 2021; Ayalew et al. 2022; Daniel & Abate 2022; M.Yang et al. 2023). However, the identification of the groundwater recharge potential zone has received little attention in the study area to date.

The lack of sufficient information on the availabilities and accessibilities of groundwater recharge potential zones causes a serious challenge for prioritizing and applying conservation action. Scientific investigations were recommended as the core solution in order to restore and improve the sustainable management of groundwater resources. Geographical information systems (GIS) and remote sensing were advanced technology used to analyze and visualize the groundwater recharge potential zone. Therefore, the present study aimed to fill the existing research gap by evaluating the groundwater recharge potential zone by using GIS and remote sensing with the AHP in the Ziway Abijata sub-basin, Central Rift Valley of Ethiopia.

Several software packages were used to analyze the groundwater recharge potential zone in the study area. For instance, ArcGIS 10.3 software was used to analyze and visualize all the factors represented by GIS thematic layers and to produce the groundwater recharge potential zone map. ERDAS 2015 and Google Earth Pro were applied for the LULC classification and accuracy assessment. IDRISI Selva 17 was also used to calculate pairwise comparisons and weights of the parameters in this study, whereas Arc SWAT was used for watershed delineation of the study area.

In the present study, the six factors of soil texture, soil drainage, slope, lineament density, drainage density, and LULC were selected. The parameter ratings and ranges for this study are indicated in Table 2.

Soil texture is the main parameter used for assessing the groundwater recharge potential zone. There are five types of soil texture in the Ziway Abijata sub-basin. These are sandy loam, loam, sandy clay loam, sandy clay, and clay loam. According to the study by Githinji et al. (2022), soil textures sandy loam and loam were excellent and good, respectively, to assess groundwater recharge. In addition, sand clay loam and sandy clay were categorized under moderate and low, respectively, while clay loam was classified under the poor groundwater recharge potential zone.

Soil drainage is another essential factor, and five soil drainage types existed in the study area. These are: well, moderate, somewhat excessive, imperfect, and poor. According to a previous study, from the types of soil drainage, well and moderate soil drainage were excellent and good, respectively, to assess groundwater recharge potential zones. Consequently, somewhat excessive and imperfect soil drainage were categorized into moderate and low groundwater recharge potential zones, respectively. However, poor soil drainage was classified under the poor groundwater recharge potential zone (Bera et al. 2020; Xu et al. 2023).

The slope was an essential parameter to assess the groundwater recharge potential zone and it was generated from a DEM. According to the study by Pande et al. (2018), the slope variation for determining the groundwater potential zone ranges from 0° to 2°, 2° to 5°, 5° to 12°, 12° to 22°,^, and greater than 22°. In this classification, 0°–2° and 2°–5° were excellent and good groundwater recharge, respectively. In addition, 5°–12° and 12°–22° were classified as moderate and low groundwater recharge potential zone, respectively, whereas a slope greater than 22° ranked as poor for groundwater recharge potential zone mapping.

The land use and land cover (LULC) was classified from Landsat OLI/TIRS 2022 by using supervised classification with a maximum likelihood algorithm. The classified LULC types include built-up area, agricultural land, water body, forest land, and bare land. According to the study by Senapati & Das (2022), water bodies and forest land were categorized under excellent and good groundwater recharge potential zone mapping, respectively. Consequently, grassland and agricultural land were classified as moderate and low groundwater recharge potential zones, whereas built-up areas and bare land were classified as poor groundwater recharge potential zone mapping. The accuracy assessment of LULC classes of the study area was validated by using Google Earth Pro.

The AHP method-based multi-criteria evaluation (MCE) analysis was used to compute the criteria weights of spatial data to determine the groundwater recharge potential zone by a scientific ratio scale of 1–9 (Moisa et al. 2023). To model the potential zone of groundwater recharge, targeted parameters of soil texture, soil drainage, slope, lineament density, drainage density, and LULC, and the relative values of each factor were calculated in IDRISI Selva 17 environments (Table 3). A pairwise comparison matrix was applied to reclassify weight parameters based on their relative importance and the degree of influence for mapping the groundwater recharge potential zone in the study area (Moisa et al. 2022a; Gao et al. 2023). According to Moisa et al. (2022b), the consistency ratio was computed from the consistency index (Equation (3)). The consistency ratio of this study was 0.03, which was less than 0.1 (Moisa et al. 2022c, 2022d; Z. Liu et al. 2023).

Table 3

Pairwise comparison of the parameters

Factors .	Stxt .	Sdrg .	Slope .	Lmd .	Dad .	LULC .	Weight .
Stxt	1	2	2	2	2	3	0.29
Sdrg	1/2	1	2	2	2	3	0.25
Slope	1/2	1/2	1	2	2	2	0.15
Lmd	1/2	1/2	1/2	1	2	2	0.13
Dad	1/2	1/2	1/2	1/2	1	2	0.1
LULC	1/3	1/3	1/2	1/2	1/2	1	0.08
=	3.3	4.83	6.5	8	9.5	13	1

Factors .	Stxt .	Sdrg .	Slope .	Lmd .	Dad .	LULC .	Weight .
Stxt	1	2	2	2	2	3	0.29
Sdrg	1/2	1	2	2	2	3	0.25
Slope	1/2	1/2	1	2	2	2	0.15
Lmd	1/2	1/2	1/2	1	2	2	0.13
Dad	1/2	1/2	1/2	1/2	1	2	0.1
LULC	1/3	1/3	1/2	1/2	1/2	1	0.08
=	3.3	4.83	6.5	8	9.5	13	1

Stxt, soil texture; Sdrg, soil drainage; Lmd, lineament density; Dad, drainage density; LULC, land use land cover.

View Large

Slopes with gentle (0°–2°) and moderate (2°–5°) landform classes were considered excellent and good for the groundwater recharge potential zones. Flat land with a minimum slope was characterized by a lower speed of surface runoff and higher water retention capacity, which assists groundwater recharge. The result shows that about 31.9% and 36.9% of the study area were classified under excellent and moderate groundwater potential zones, respectively. Besides this, the remaining 2.7% of the study area was occupied by a steep slope, which is categorized under a poor groundwater potential zone. Based on slope classification, larger parts of the study area were occupied by excellent and moderate groundwater recharge potential zones (Table 5). Spatially, northern and southern parts particularly around Shala Abijata and Hawassa Lake were considered as the excellent groundwater potential zone. However, southern and western parts of the study area were occupied by the poor groundwater recharge zone (Figure 3(b)). The results of this finding are more in line with those of Hassini et al.(2022), where a gentle slope was more excellent for groundwater recharge than a steep slope in the Regueb basin of central Tunisia.

Lineament density was calculated from lineament data, which is more important for determining the groundwater potential zone. An area with high lineament density was ranked as excellent for the groundwater recharge zone. However, low lineament density was classified as a poor groundwater recharge potential zone. Geographically, the southern and southwestern parts were characterized by an excellent groundwater recharge potential zone, whereas the central and northern parts of the study area were indicated as poor for groundwater recharge (Figure 3(c)). The previous studies that have been conducted in the Ajani-Jhiri watershed of north Maharashtra of India showed that high lineament density was excellent for groundwater potential zone mapping (Ardakani et al. 2022; Sahu et al. 2022; X. Zhu et al. 2022; Yin et al. 2023).

Spatially, the central part of the study area was ranked as excellent for the groundwater recharge potential zone. However, the northern and southern parts were classified as poor groundwater potential zone (Figure 4(a)).

Soil texture is another determinant factor, which may limit the groundwater recharge capacity of an area. Soil texture classifies soil into clay, sandy loam, and loam based on the proportion of sand, clay, and silt-sized particles in water-holding capacity (J. Yang et al. 2022; C. Yang et al. 2023; Zhou et al. 2022). The rate of the groundwater recharge was influenced by soil texture. An area with sandy loam, loamy sand, and sandy soil texture was characterized as having high capacity for groundwater recharge due to the presence of a larger porosity size for water infiltration (Yue et al. 2021; Parker et al. 2022; Zhuo et al. 2022). The results show that about 424.9 km² (15.6%) of the study area was classified as an excellent groundwater recharge zone due to the domination of sandy loam soil texture. Similarly, about 2,254 km² (82.7%) of the area was covered with sandy clay loam soil texture and a moderate potential zone for recharging groundwater. However, the remaining 12 km² (0.4%) of the study area was covered with clay loam soil texture and ranked as poor for recharging groundwater. Hence, a moderate potential groundwater zone occupies larger parts of the study area than the other classes (Table 6).

Geographically, a central part of the study area was an excellent groundwater recharge potential zone, while a southern part was characterized by a poor groundwater recharging zone (Figure 4(b)). The result of this study was more in line with Wadi et al. (2022), which confirmed that sandy soils have higher water infiltration than clay soil and that there is a high potential zone for groundwater recharge in the semi-arid crystalline rock context of Biteira district, Sudan.

Land use land cover types have an influence on the groundwater recharge capability. LULC classes in this study area include agricultural land, bare land, built-up area, forest land, grassland, and water body. From classified LULC classes, water body, forest land, grassland, and agricultural land were ranked as excellent, good, moderate, and lower potential zones for recharging groundwater, respectively. However, bare land and built-up area were ranked as poor for groundwater recharge potential zone mapping. Bare land was characterized by a high rate of surface runoff and soil evaporation, which limited groundwater recharge. Additionally, a built-up area degrades the forest, which absorbs and slows the rate of surface runoff that seeps into the soil to enhance groundwater recharge (Table 7). Water bodies, grassland, and forest land were more suitable than agricultural land and bare land for recharging groundwater due to their capacity for increasing infiltrations by holding rainwater for a certain period of time.

Spatially, the southern and south eastern parts of the study area were excellent and good for groundwater recharge, whereas the central and western parts were low and poor in groundwater recharge, respectively (Figure 4(c)). The result of this study is more consistent with Siddik et al. (2022), which indicated that LULC change has an impact on groundwater recharge in northwestern Bangladesh. In addition, a recent study (Warku et al. 2022) conducted in the upper Gibe watershed stated that LULC has an impact on groundwater recharge potential zone mapping.

The identification of a groundwater recharge potential zone plays a crucial role within a critical drought-vulnerable area due to surface water resources and rainfall scarcity. It provides clues on how and what extent the groundwater resource is to be utilized to minimize wastage of water resources. In the present study, geospatial technologies with the AHP were applied to evaluate the groundwater recharge potential zone in the Ziway Abijata sub-basin of the Central Ethiopian Rift Valley. High groundwater potential areas required continual follow-up and appropriate modes of utility to improve the sustainability of existing resources. In addition, the area covered with low and poor groundwater recharge potential zones requires a sustainable management. Based on the result, we suggested that decision-makers, environmentalists, water resource management offices, geologists, and other concerned stakeholders will have a great responsibility for sustainable utilization and proper management of the identified groundwater recharge potential zone in the study area. In addition, further studies can evaluate the groundwater recharge potential zone by using ecological parameters and socio-economic data, which were not included in the present study.

The authors acknowledge Wollega University Shambu Campus, Raya University, Mattu University, Bedele Campus, and Jimma University College of Agriculture and Veterinary Medicine for the existing facilities to conduct this study.

M.B.M. and M.M.G. participated in research design, document analysis, and manuscript writing. G.F.N., D.G.O. and M.L.D. participated in data collection, methodology, data analysis, and interpretation. K.T.D., Z.R.R., and D.O.G. participated in the research design, literature review, data analysis, and final draft edition. All authors read and approved the final manuscript for publication.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.