Geospatial and Multi-Criteria Analysis for Identifying Groundwater Potential Zones in the Oltu Basin, Turkey

Abstract

This study was conducted to determine potential groundwater storage areas in the semi-arid Oltu Basin in northeastern Turkey. The groundwater potential of the basin was analyzed by evaluating eight geographical factors: lithology, linear density, soil depth, land use, precipitation, geomorphology, slope, and drainage density. These factors were classified and weighted using remote sensing, geographical information systems (GIS), and the analytic hierarchy process (AHP). The obtained data were modeled using ArcGIS software, and a potential groundwater storage map of the Oltu Basin was created. The results show that there is a high groundwater potential in areas of the basin close to the stream bed, while the groundwater potential is low in mountainous and steeply sloped regions. The study provides significant findings for sustainable water resource management in the region and future water resources planning.

1. Introduction

The availability of water resources on Earth has a critical importance for the continuity of life. However, since the majority of water on Earth is salt water, with 97% of all water being found in oceans and seas, the amount of freshwater that humans can directly use is very limited. The remaining 3% of water is freshwater, but most of this freshwater is not accessible to humans. Approximately 68.7% of freshwater is hidden in glaciers and mountain glaciers in polar regions, and 30.1% is found in the form of groundwater; these waters are the most essential source for agriculture, drinking water, and industrial uses. The remaining 1.2% of freshwater is found in lakes, rivers, and the atmosphere. This distribution represents the amount of water directly available to the world’s population for daily use. The amount of usable freshwater on Earth is about 200,000 km³ and is not evenly distributed due to geographical and climatic conditions. Moreover, the possible effects of climate change and environmental factors, such as unconscious water use and the increase in human population, are causing the amount of usable water to decrease significantly daily. Therefore, properly managing available water resources is essential for a sustainable environment and human wellbeing. The limited directly accessible water resources in the world have led to a shift towards exploiting groundwater resources, which are of great importance in terms of human use. According to the United Nations World Water Development Report, global groundwater use increased from 312 km³ in 1960 to 959 km³ in 2017 [1]. Moreover, studies show that groundwater has decreased significantly in the last 40 years [2]. Therefore, to ensure the sustainable use of groundwater, groundwater volume must be estimated accurately and with the most straightforward approach to reduce costs.

In order to determine an area’s groundwater potential, some basic concepts need to be better understood. In a hydrological context, infiltration, which describes the movement of water in the subsurface, is defined as the process by which surface water passes through soil and rock layers to reach groundwater reservoirs. Infiltration is a fundamental component of the hydrological cycle and plays a critical role in groundwater recharge. This process is influenced by factors such as surface soil type, topography, vegetation cover, and the amount and intensity of rainfall. For example, sandy soils allow water to infiltrate quickly due to their high permeability, while clay soils can cause water to accumulate at the surface due to their low permeability. In addition, sloping terrain has an increased passage of water to surface runoff, while infiltration rates are generally higher in flat areas. Human activities can also affect infiltration: agricultural irrigation can increase groundwater infiltration, while urbanization and paved surfaces greatly reduce infiltration. Water that passes underground through infiltration is stored in aquifers, and the quantity and quality of this water is of great importance for the management and sustainability of local water resources. Fetter [3] discusses hydrogeological processes in groundwater storage, especially in terms of aquifer properties and water movement. The storage potential of groundwater is influenced by aquifer properties, water movement and storage capacity, and biological and chemical interactions. Seiler and Gat [4] state that groundwater, which represents the most important resource for continental ecosystems, applies not only to semi-arid and arid areas but also to all rock formations such as karst and gravelly formations in other climates with unlimited infiltration capacity. According to Schiavo [5], the most favorable areas for groundwater storage are found where permeable rocks and aquifers are present. In particular, areas with high groundwater storage capacity are concentrated in areas with features such as sedimentary rocks, lava flows, or fractures and cracks. The infiltration rates of natural water are also important to the replenishment of aquifers over time. Aquifers can be naturally fed from surface water sources. Therefore, it is important to understand the specific environmental interactions and hydrological cycle of groundwater storage areas [6]. The concept of groundwater recharge is also important in understanding where groundwater can be found. Groundwater recharge refers to the process by which surface water seeps underground to reach aquifers and recharge these reservoirs. Recharge is a critical component of the hydrological cycle and is vital to the sustainability of aquifers. This process usually occurs through natural phenomena such as rainfall, snowmelt, and infiltration from rivers, lakes, and dams. Human interventions such as agricultural irrigation and artificial injection can also contribute to groundwater recharge. Rates of recharge vary depending on soil permeability, characteristics of geological formations, topography, vegetation density, climatic conditions, and the amount of surface water. For example, sandy and gravelly soils offer favorable conditions for recharge due to their high permeability, while low-permeability soils such as clay limit this process. Geological structures with fractures or faults can facilitate water reaching deeper depths. Finally, potential groundwater zones are defined as geographical areas where groundwater resources may be abundant and easily accessible. Identification of these areas is an important step in groundwater management and planning.

Various direct and indirect methods can be used to determine areas with groundwater potential, including statistical methods, expert evaluation, geophysical, deterministic, and hydrogeological methods, as well as drilling, GIS, and remote sensing techniques [7,8,9,10]. Many researchers around the world use logistic regression [11], artificial neural network models [12], frequency ratios [13], machine learning [14], and hydrogeological techniques [15], and geophysical methods [16] for the detection of potential groundwater factors. However, these methods are generally time-consuming and costly.

Remote sensing and GIS are powerful tools that can estimate groundwater resources quickly and at low cost before starting more comprehensive and costly studies [17]. In recent years, remote sensing, geographical information systems, and analytical hierarchy processes have increasingly been used to determine potential groundwater collection areas by considering cost–benefit analyses. Various factors influence the development of groundwater potential. These factors include elements such as geomorphological structure, lineament, slope, altitude, soil structure, drainage pattern, land use, land cover, and rainfall amount [18]. Each component plays a vital role in determining the water potential of the region by directly affecting groundwater storage and flow capacity. The combination of these factors shapes the dynamics of groundwater systems and constitutes the key elements that must be considered to evaluate groundwater resources effectively.

Remote sensing, geographic information systems, and the analytical hierarchy process have been used previously in various studies to determine the relative impact of these elements and create thematic maps. In this context, a study was conducted by Aslan and Çelik [19] to determine potential groundwater storage areas for the Harran Plain in southeast Turkey. It evaluated the following geographical factors: land use, soil, geomorphology, geology, aquifer, drainage density, precipitation, slope, and linear density. By creating 10 thematic layers, such as land class, the groundwater potential of the region was determined in five different categories. Also, Beden et al. [20] carried out a study to assess the groundwater potential for the Kızılırmak Delta located in the north of Turkey by using the thematic layers of potential groundwater storage areas, drainage density, linear density, precipitation, slope, soil types, geology, land cover, and groundwater level. In his study of the Upper Tigris sub-basin within the borders of Batman province in the southeast of Turkey, Çelik [21] evaluated potential groundwater storage areas based on eight different criteria, namely geomorphology, geology, linear density, slope, drainage density, land use, and soil properties. In the study conducted by Dilekoğlu and Aslan [22] for the Ceylanpınar region in the southeast of Turkey, the fuzzy analytical hierarchy process and multi-criteria decision-making methods were evaluated together, and potential groundwater storage areas were classified into five different categories. In his study for the upper Çoruh River basin in Turkey in 2021, Yıldırım [23] identified a groundwater potential zone using 10 different geoenvironmental factors with the help of remote sensing, AHP, and geographic information systems. In their study, Kavurmacı and Üstün [24] used an analytical hierarchy process and data envelopment analysis to evaluate groundwater quality. Muratlıharan and Palanivel [25] estimated groundwater potential zones in five different categories for hard rock regions in Tamil Nadu, India, using seven thematic criteria using analytical hierarchy process and geographic information systems. Mohammadi-Behzad et al. [26] estimated groundwater potential using five thematic layers, such as linear density, precipitation, lithology, slope, and drainage density, in their study in the southwest of Iran. The authors showed that tectonic structures play an essential role in the fracture and crushing of limestone units, which affects groundwater storage areas. Achu et al. [27] revealed the effects of geoenvironmental factors on groundwater storage. Also, Shao et al. [28] conducted a study in China wherein they proved that groundwater storage areas identified in a semi-arid region by evaluating geoenvironmental factors using remote sensing, geographic information systems, and fuzzy analytical hierarchy process are compatible with the data collected from the field. In a study on the Brandenburg region of Germany, storage areas were determined using remote sensing, geographical information systems, and analytical hierarchy process and using 21-year data to estimate seasonal groundwater [29]. In the study by Opoku et al. [30] in Jinan Karst Spring Basin, China, groundwater potential zones were identified by integrating geographic information systems, remote sensing, and AHP for a complex topographic structure. Finally, in the study conducted by Hossain et al. [31] in the Barind region of Bangladesh in 2024, regional groundwater volume calculations were made using a combination of geographic information systems, AHP, and remote sensing for sustainable groundwater resource use.

Many studies in this context have been carried out in Central Asia, India, and the Middle East, and the common point of these regions is that they include arid and semi-arid areas. Groundwater has great importance for these regions. This study was conducted in the Oltu basin, which is a sub-basin of the Çoruh basin in the northeast of Turkey and which is a semi-arid region. The combined use of remote sensing, geographic information systems, and analytical hierarchy process, which have been successfully applied in many previous studies, has been evaluated for the Oltu basin. For this purpose, eight different geoenvironmental factors were evaluated together, and potential groundwater storage areas were determined for the Oltu basin, a large part of which consists of steeply sloped and rocky lands.

2. Study Area

The Oltu basin is located within the borders of the Çoruh basin, which is one of the twenty-six water collection basins in the northeastern region of Turkey, between 40°10′ and 40°50′ North latitude, and 41°30′ and 42°40′ East longitude. The drainage area of the basin, which includes the Oltu, Narman, Şenkaya districts and part of the Olur district, is about 3518.50 km² (Figure 1a,b).

When the geology of the basin was examined, which is located on the North Anatolian orogenic belt, it was determined that it consisted of Oligocene-aged conglomerate, sandstone, silt-clay, gypsum limestone, tuff, and alluvial sediments [32]. Examining the region’s topography, it is seen that most of the basin consists of mountains and hills. It is observed that there are valley formations in the areas between the mountains. Most of the region has a slope greater than 15%, which can cause floods, landslides, erosion, and so on. In addition, the high slope negatively affects the storage of rainwater falling in the region (Figure 1c).

In various climate classification systems, the Oltu basin is included in the semi-arid–semi-humid climate class. These climate characteristics deeply affect the region’s agricultural activities and the local ecosystem. While only a very small part of the basin area consists of forest, residential, and agricultural areas, the remaining part is covered by pastures and bare areas. The scarcity of forest areas limits agricultural productivity and biodiversity. Most people in the region earn their living from agriculture and animal husbandry. Agriculture focuses mainly on the cultivation of grain and livestock-related products. In addition, it is seen that the local people primarily engage in animal husbandry under the given climatic conditions. In addition, handicrafts, especially Oltu stone processing, are another activity that significantly shapes the economy of the region. Oltu stone is processed by the local people and turned into various jewelry and ornaments, which are valuable both as a cultural heritage and an economic resource (Figure 1d).

When the aquifer structure of the study area is examined, it is seen that a small part of the basin consists of karst aquifers (sedimentary and metamorphic carbonate rocks) with poor and shallow aquifers (other metamorphic and igneous rocks), while a large part of the basin contains various hydrogeological aquifer formations (other sedimentary and volcanic formations) (Figure 2). This shows that the region should be studied more locally rather than on a regional level. For this purpose, we aimed to identify areas with high groundwater potential via remote sensing.

In the study conducted by Yarbaşı [32] on the Oltu district center, where the study area is located, it was found that the groundwater level in areas containing alluvial units close to the stream bed varied between 2 and 10 m. No groundwater was encountered at these depths during drilling in areas in the north and south of the Oltu stream where sedimentary and volcanic rocks were located [32].

For this reason, the Oltu basin offers essential opportunities in agriculture and handicrafts, which in turn affects the living standards of the region’s people. This social and economic structure creates a vital need for sustainable management and protection of the basin’s natural resources. In addition to agriculture and animal husbandry, the protection and management of pastures is critical for the continuity of the regional economy.

Using the region’s natural resources efficiently to continue livestock and agricultural activities, which are crucial to the livelihoods of the local population, requires careful planning and the simultaneous evaluation of climatic features, geological features, and topography. The sustainable management of groundwater resources is of great importance for both the continuation of agricultural and livestock activities in the region and for the protection of the local ecosystem. In this context, developing water management strategies appropriate to the climate and geographical characteristics of the basin is critical to ensuring the welfare of the people of the region and ensuring environmental balance.

3. Materials and Methods

The steps for determining groundwater potential zones are shown in Figure 3. The study includes the analysis of eight different layers of geographical data in ArcGIS environment. These layers are an average annual precipitation map created with annual precipitation data obtained from the Turkish State Meteorological Service; land use data obtained from Sentinel-2 satellite images; lineation density, expressing the density of fault lines in the unit area; slope, expressing the slope of the land; drainage density, expressing the length of the river in the unit area; geomorphology, analyzing landforms; lithology, expressing the properties of rock types; and soil depth, expressing the thickness of the soil layer. Following the steps below, weight calculations were made with analytical hierarchy process (AHP) using these layers, and a final map showing the groundwater potential zones was created as a result of the weighted overlay analysis (Figure 3).

Eight geographical data layers were generated using data collected from various sources to determine the potential groundwater storage zones in the study area, and necessary projection transformations and adjustments to the layers were made with ArcGIS 10.6 software. Classification was carried out using the maps in the study and evaluating them as individual layers to estimate groundwater storage areas. Lithology, lineament density, soil depth map, land use map, precipitation, geomorphology, slope, and drainage density layers were used in the study. In this context, the Digital Elevation Model (DEM) was obtained from search.earthdata.nasa.gov and adapted for the study area. A geomorphological map was prepared with DEM data using the Jenness Landform Classification toolbox, which is offered as an open source for ArcGIS software, and the classification process was performed through the ArcGIS 10.6/Reclassify toolbar (Figure 4a). A lineament density map (https://yerbilimleri.mta.gov.tr/anasayfa.aspx (accessed on 5 October 2024)) was created with the help of the site drawing editor and via the ArcGIS Line Density menu. The classification process on the resulting map was performed with the Reclassify toolbar (Figure 4b). A land use map was produced for the study area from 10 m resolution images from the Sentinel-2 satellite for the year 2023, and a column was added to the attribute table for the classification process, classified for groundwater storage capacities based on land class (Figure 4c). For the lithology map, geological maps of the study area at a 1:100,000 scale, which were obtained from the General Directorate of Mineral Research and Exploration (MTA), were arranged in ArcGIS, and the formations were obtained by performing symbology analysis according to lithology codes. The resulting lithology map was classified according to formation permeability values in the attribute table (Figure 4d).

A slope map was prepared in the ArcGIS program based on DEM data, and classification was performed with ArcGIS/Reclassify according to the slope ratio (Figure 5a). The map of large soil groups was obtained from the General Directorate of Agricultural Reform (TRGM) of the Ministry of Agriculture and Forestry (2001), and the classification process was segmented into 3 groups: deep soil, medium-deep soil, and shallow soil (Figure 5b). A drainage density map was prepared in ArcGIS based on DEM data, and classification between 1 and 5 was made to indicate the drainage network per km². The classification was performed with the ArcGIS/Reclassify toolbar (Figure 5c). According to the annual average precipitation values received from the General Directorate of Meteorological Affairs (MGM), a precipitation map of the study area was generated within the ArcGIS environment through the application of the Kriging method, and the classification process was carried out via the ArcGIS/Reclassify toolbar (Figure 5d). WGS-1984-UTM Zone 37 projection system was used in the study, and projection transformations were made for all maps.

The classification of the relevant layers according to their potential groundwater storage status in the study was determined by adding classification categories to the ArcGIS/Reclassify module and the attribute tables of each of the layers. The classification used for each layer was evaluated with a value of 1–5 according to the potential groundwater capacity. (5, high water retention capacity; 1, low water retention capacity) (

Table 1. Classification of thematic layers according to potential groundwater retention values.

Thematic Map	Classification	Potential Groundwater Capacity	Software Used
Lithology	Alluvium, old alluvium	5	Created on the attribute table of the thematic layer
	Andesite-basalt, undifferentiated volcanics, conglomerate, sandstone, siltstone, limestone, slope accumulation	4
	Andesite, undifferentiated terrestrial sediments, flysch, conglomerate, sandstone, ophiolitic units, volcanosedimentary	3
	Undifferentiated metamorphics, gypsum undifferentiated continental sediments	2
	Gabro, granite, marble	1
Lineament Density	0.84 km−1–1.4 km−1	5	ArcGIS/Line Density-Reclassify
	0.56 km−1–0.83 km−1	4
	0.33 km−1–0.55 km−1	3
	0.12 km−1–0.32 km−1	2
	0 km−1–0.11 km−1	1
Soil Depth	Very deep	5	Created on the attribute table of the thematic layer
	Moderate	3
	Shallow	2
Land Use	Water	5	Created on the attribute table of the thematic layer
	Crops	5
	Trees	3
	Built area	2
	Rangeland	2
	Bare ground	1
Precipitation	220–488 mm	5	ArcGIS/Reclassify
	489–659 mm	4
	660–810 mm	3
	811–952 mm	2
	953–1260 mm	1
Geomorphology	Valleys	5	ArcGIS/Reclassify
	Flat Area	4
	Ridges	2
	Hills	1
Slope	20.1°–69.6°	5	ArcGIS/Reclassify
	15.1°–20°	4
	10.1°–15°	3
	5.01°–10°	2
	0°–5°	1
Drainage Density	0.73 km−1–1.3 km−1	5	ArcGIS/Reclassify
	0.54 km−1–0.72 km−1	4
	0.37 km−1–0.53 km−1	3
	0.19 km−1–0.36 km−1	2
	0 km−1–0.18 km−1	1

).

The analytical hierarchy process (AHP) determines the weights which the classified maps will be assigned. The AHP is a mathematical method used in measurement and decision-making, developed by Thomas L. SAATY in the 1970s [33]. Decision makers frequently prefer the AHP because it considers subjective criteria when making multi-criteria decisions and is a method that can combine qualitative and quantitative factors when evaluating alternatives [34]. The AHP method is a versatile decision-making approach incorporating quantitative and qualitative criteria, allowing decision-makers to factor in individual or group preferences, experiences, intuitions, knowledge, and judgments. The AHP method enables effective solutions, accommodating objective and subjective perspectives within decision-making by structuring complex problems hierarchically, thus allowing the decision-maker to familiarize themself with their own decision-making mechanisms. In the AHP method, the problem is first defined, and a hierarchical structure is then created to solve the problem. In this hierarchical structure, the main target is located on the top tab, and the basic criteria and sub-criteria are listed below. Then, a pairwise comparison matrix is created. Matrices are created to compare decision options and generate importance levels ranging from 1 to 9.

$$A=\left[\begin{array}{ccc}1& a_{12}& \dots \dots .. a_{1n}\\ 1/a_{12}& 1& \dots \dots .. a_{2n}\\ .& .& \dots \dots .. .\\ .& .& \dots \dots .. .\\ 1/a_{1n}& 1/a_{2n}& \dots \dots .. 1\end{array}\right]$$

(1)

Decision matrices are created using the 1–9 comparison scale below (

Table 2. Comparison scale.

Importance	Definition	Explanation
1	Equally significant	Both options are equally significant
2	Weak or mild
3	Somewhat substantial	One criterion was deemed slightly more substantial than the other
4	Reasonable plus
5	Too notable	One criterion was deemed much more notable than the other
6	Strong plus
7	Too important	The criterion was definitely considered much more important than the other criterion.
8	Very very strong
9	Extremely crucial	Based on various information, one criterion is determined to be significantly more crucial than the other.

), prepared by Saaty [35].

Pairwise comparison matrices are then normalized by dividing by their column total.

$$a_{i{j}^{\prime }}={\displaystyle \frac{a_{ij}}{ {{\displaystyle \sum }}_{i=1}^n{a}_{ij}}}$$

(2)

Then, the priority vector for the weights is calculated by using Equation (3) for the priority vector.

$$w_i=\left(1/n\right){{\displaystyle \sum }}_{i=1}^n{a_{ij}}^{\prime } i, j=1, 2, \dots \dots \dots .., n$$

(3)

The consistency index (CI) is determined by calculating consistency ratios, and a random index comparison is made.

$$CI={\displaystyle \frac{{\lambda }_{max}-n}{n-1}}$$

(4)

$${\lambda }_{max}={\displaystyle \frac{1}{n}} {{\displaystyle \sum }}_{i=1}^n({\displaystyle \frac{{{\displaystyle \sum }}_{i=1}^n{a}_{ij }{w}_{ij}}{w_i}})$$

(5)

$$A=\left[\begin{array}{ccc}1& a_{12}& \dots \dots .. a_{1n}\\ 1/a_{12}& 1& \dots \dots .. a_{2n}\\ .& .& \dots \dots .. . \\ .& .& \dots \dots .. .\\ 1/a_{1n}& 1/a_{2n}& \dots \dots .. 1\end{array}\right]x\left[\begin{array}{c}{w}_1\\ w_2\\ .\\ .\\ w_n\end{array}\right]=\left[\begin{array}{c}{x}_1\\ x_2\\ .\\ .\\ x_n\end{array}\right]$$

(6)

$$d_i={\displaystyle \frac{x_i}{w_i}} , i=1, 2, \dots \dots \dots n$$

(7)

$${\lambda }_{max}= {\displaystyle \frac{{{\displaystyle \sum }}_{i=1}^n{d}_i}{n}}$$

(8)

To assess consistency, it is essential to know the value of the random index (RI) and the consistency ratio (CR); the CR is determined to assess whether the comparison matrix is consistent.

Table 3. RI values according to the dimensions of the comparison matrices.

Matrix Size (N)	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15
RI	0	0	0.58	0.90	1.12	1.24	1.32	1.41	1.45	1.49	1.51	1.53	1.56	1.57	1.59

is used to determine the random index value, and the consistency ratio was calculated via the following formula:

$$CR={\displaystyle \frac{CI}{RI}}$$

(9)

The AHP matrices were created using Table 2 and the criteria table generated for the study area (

Table 4. Comparative AHP table.

Criterion	a	b	c	d	e	f	g	h
Geomorphology (a)	1	2	2	4	3	3	2	3
Lineament Density (b)	1/2	1	2	2	3	2	3	4
Land Use (c)	1/2	1/2	1	3	4	4	3	3
Geology (d)	1/4	1/2	1/3	1	2	3	2	3
Slope (e)	1/3	1/3	1/4	1/2	1	3	4	2
Soil Depth (f)	1/3	1/2	1/4	1/3	1/3	1	2	2
Drainage Density (g)	1/2	1/3	1/3	1/2	1/4	1/2	1	3
Precipitation (h)	1/3	1/4	1/3	1/3	1/2	1/2	1/3	1
Total	3.75	5.42	6.50	11.67	14.08	17.00	17.33	21.00

). The weight percentages of the maps to be used according to the comparative AHP table are shown in

Table 5. Weight percentages (normalized).

Total	3.75	5.42	6.50	11.67	14.08	17.00	17.33	21.00	Total	Weight (wi)	xi	di
Geomorphology (a)	0.27	0.37	0.31	0.34	0.21	0.18	0.12	0.14	1.93	0.24	2.1881	9.0500
Lineament Den. (b)	0.13	0.18	0.31	0.17	0.21	0.12	0.17	0.19	1.49	0.19	1.7053	9.1482
Land Use (c)	0.13	0.09	0.15	0.26	0.28	0.24	0.17	0.14	1.47	0.18	1.7299	9.4023
Geology (d)	0.07	0.09	0.05	0.09	0.14	0.18	0.12	0.14	0.87	0.11	0.9926	9.0987
Slope (e)	0.09	0.06	0.04	0.04	0.07	0.18	0.23	0.10	0.81	0.10	0.9020	8.9609
Soil Depth (f)	0.09	0.09	0.04	0.03	0.02	0.06	0.12	0.10	0.54	0.07	0.5782	8.5449
Drainage Den. (g)	0.13	0.06	0.05	0.04	0.02	0.03	0.06	0.14	0.54	0.07	0.5550	8.2721
Precipitation (h)	0.09	0.05	0.05	0.03	0.04	0.03	0.02	0.05	0.35	0.04	0.3747	8.6480

.

Equation (9) was used to evaluate the accuracy of normalized weight percentages, and when evaluated together with the random index specified in Table 2, the ${\lambda }_{max}$ was 8.8906, the CI was 0.1272, and the CR was calculated as 0.0902. As the resulting consistency ratio is less than 0.10, the comparison matrix is consistent.

4. Results and Discussion

In groundwater research, classified thematic maps have an extremely critical role in understanding the hydrogeological characteristics of the area and evaluating the groundwater potential. These maps allow different environmental and geological factors to be combined and analyzed. Thematic maps are essential in estimating, managing, and protecting groundwater potential and can be applied in the following ways:

• Revealing regional differences;
• Integration for sustainable management;
• Evaluation of hydrological parameters;
• Creating infrastructure for weighted layer analysis;
• Use in decision support systems;
• Increasing precision and model accuracy;
• Determining groundwater potential zones.

By using remote sensing and geographical information systems together, the spatial distributions of different parameters of the basin were visualized. As a result of our study, the areal distributions and concentration areas of the thematic maps that affected the decision mechanism were determined. In this context, when the geomorphology map shown in Figure 4a is examined, it is determined that approximately 41% of the basin consists of valleys, while 3% is flat areas, 16% is ridges, and 40% is hills. When Figure 4b is examined, it is seen that the faults and fractures in the basin are concentrated in the northwest of the Narman district (the area between the northeast of the Oltu district and the northwest of the Şenkaya district) and the south of the Şenkaya district. When we look at the land use of the basin in Figure 4c, we see that approximately 12% of the basin consists of forest cover, 3% arable land, 1% construction area, and 83% pasture, while the remaining very small part consists of water surface and bare regions. Looking at Figure 4d, 72% of the basin consists of high-permeability rock, such as andesite and alluvium; 20% consists of medium-permeability rock, such as flysch, andesite, and sandstone; and 8% consists of gabbro, granite, marble, and metamorphic rocks with low permeability. Looking at Figure 5a, 60% of the basin has a slope of 0–5°, 17% has a slope of 5–10°, 11% has a slope of 10–15°, 8% has a slope of 15–20°, and the remaining 4% has a slope of 15–20°. In addition, it was determined that the third group had a slope of more than 20°. Examining Figure 5b, it is seen that 5% of the basin has a very deep soil structure, 15% has a medium-depth soil structure, and 80% has a shallow soil structure. Figure 5c shows the drainage of the basin, and the stream density per unit area was calculated according to the distribution of stream beds and tributaries over the basin. Looking at Figure 5d, based on the annual average rainfall in the basin, 16% of the basin receives precipitation between 276 and 423 mm, 29% receives precipitation between 424 and 571 mm, and 55% receives precipitation between 572 and 679 mm.

The areas classified according to their potential groundwater values are shown in Table 1. AHP matrices in Table 4 were determined according to the importance levels specified in Table 2. The comparison matrices are defined in Table 3, and the weighted and normalized weight percentages are calculated in Table 5. When all these thematic maps and the percentage values obtained from the AHP and the parameters affecting the groundwater potential zone of the basin are evaluated, as seen in Table 5, geomorphology ranks first at a rate of 24%, followed by lineament density with a rate of 19%. Land use follows with a rate of 18%, lithology with 11%, slope with 10%, soil depth and drainage density with 7%, and, finally, precipitation with 4%. When we compare the calculated weight ratio percentages calculated and those used in determining the groundwater potential zones of the basin, the study reveals the existence of a groundwater potential zone parallel to the percentage densities. Five appropriateness classifications can be identified from the groundwater potential zone map: poor potential, low potential, moderate potential, good potential, and excellent potential (Figure 6). The findings showed that the topography and hydrogeological features of the basin affect the amount of water that may be stored in various locations. An area-by-area analysis of the distribution of possible groundwater storage areas across the basin revealed that 0.86% of the study area had poor groundwater potential, 0.12% had excellent groundwater potential, 2.60% had good groundwater potential, 34.74% had medium groundwater potential, and 61.68% had low groundwater potential.

Groundwater recharge areas are generally determined by geological features and soil or rock types with high vertical conductivity. Recharge areas are associated with permeable formations that allow water to percolate from the surface to the subsurface and reach aquifers. For example, geological formations such as sand, gravel, and fractured rock have high permeability and allow water to percolate easily and recharge aquifers. Karstic structures, such as limestone, offer high vertical conductivity through fractures and dissolved voids, and these areas are typically identified as the main recharge areas. When the potential groundwater storage area map presented in Figure 6 is examined, it is seen that the areas with high potential values generally coincide with geological units with high permeability and advantageous regions in terms of precipitation. Examining the correlation of this map with the lithological layer, it can be seen that high-potential areas are mostly compatible with sand, gravel, and karstic structures. Groundwater potential zones should also be evaluated in terms of their hydraulic conductivity. Hydraulic conductivity determines the rate at which water travels underground per unit time and is expressed in cm/s or m/s. For example, the conductivity value of a sandy unit generally ranges from 10⁻³ to 10⁻⁵ m/s, while clayey units have values lower than 10⁻⁸ m/s. The map of potential groundwater storage areas in Figure 6 overlaps with units with high vertical conductivity. The areas showing low potential on the map generally correspond to units with low permeability (e.g., clay or compact rock). Furthermore, the integration of potential groundwater map with other layers, such as slope, drainage density, and rainfall, allows us to better understand the locations and extents of these feeding areas. In conclusion, the potential groundwater map in Figure 6 is an important tool for the identification of the main groundwater recharge areas and shows a particularly strong correlation with the high vertical conductivity values of the geological formations. Detailed analysis of the map confirms that these areas are critical for groundwater management.

The study assessing the groundwater potential of the Oltu basin reveals a strong linkage between geological patterns and potential groundwater areas in Figure 6. Lithological data are among the main determinants of groundwater storage and mobility; for example, highly permeable units such as alluvium are directly related to the areas classified as having ‘excellent’ and ‘good’ groundwater potential in Figure 6. Similarly, geomorphological structures (such as valleys) and drainage density are also key factors contributing to this distribution, favoring groundwater flow and recharge. From a spatial perspective, these patterns show strong potential in valleys and areas close to stream beds, and low potential in areas with high slopes and impermeable lithologies. By integrating different types of information (e.g., lithology, geomorphology, rainfall, and soil depth) within the analytical hierarchy process, the study obtained a clearer understanding of groundwater aquifer boundaries and piezometric patterns. This combined analysis not only provides information on the hydrogeological characteristics of the basin, but also a basic framework for sustainable water management and efficient use of groundwater reserves. The detailed analysis of Figure 6 clearly demonstrates the interaction of geographical and geological factors affects water storage capacity, as well as this interaction’s contribution to water management in the region.

5. Conclusions

This study analyzed eight geographical data layers to determine groundwater potential zones in the Oltu basin. Model Builder in the ArcGIS program was used to perform a weighted analysis of the maps to determine classifications, weight percentages, integrate layers, and determine potential groundwater storage areas based on the classification values. At this stage, all reclassified maps were moved into Model Builder, and the model was run by defining the weight ratios for each layer. A potential groundwater storage map was created using raster data.

This study provides essential data for the sustainable management of water resources, especially in semi-arid regions. The results revealed that water storage capacity varies in different areas depending on the topography and hydrogeological characteristics of the basin. When the distribution of potential groundwater storage areas over the basin was examined, 0.12% of the study area was considered to have very high potential, 2.60% was considered to have high potential, 34.74% showed medium potential, 61.68% was considered to have low potential, and 0.86% was determined to have very low groundwater potential.

The results show that groundwater either does not exist or has minimal potential in regions with topographies characterized by high slopes and rocky lands forming the basin. It was observed that the groundwater potential is higher in areas close to the stream bed. This reveals that the river systems in the region play a critical role in feeding groundwater. In addition, a study conducted by Şenocak and Taşci [36] throughout the Çoruh basin determined that the region’s base flow index was around 78%. A high base flow in a basin indicates that the groundwater potential has a positive impact. Baseflow refers to the contribution to a river or stream coming directly from groundwater. High base flow values suggest that the basin has a significant amount of groundwater reserves, and this water constantly contributes to rivers and streams. Still, when the region’s dynamics are examined, it is observed that the groundwater recharge zones in question are limited only to the areas that form the stream bed. Freeze and Cherry [37] in their book groundwater stated that the only constants for groundwater resources are the recharge area of high zones and the discharge area of low zones. This shows that surface flow direction and groundwater flow direction are interdependent. Edery et al. [38] and Schiavo et al. [39] indirectly support this statement by stochastic analysis of surface flow paths. In this respect, when the potential groundwater areas given in Figure 6 and the cross-section shown in Figure 6 with overlapping river networks are evaluated together, it is seen that the regions with potential groundwater and the river networks overlap to a great extent. This information is very helpful in understanding the groundwater flow in the basin, especially in cases where piezometric data are not available.

Groundwater can serve as a continuous source of water, especially during the dry summer months, usable as both agricultural and drinking water due to the high base flow in the parts of the basin close to the stream bed. This high base flow also supports the continuous supply of water to river ecosystems, maintaining ecological balance. However, it has been observed that the hydroelectric power plants built on the basin are unsuitable for fish passage, negatively affecting the environmental balance.

The research method was largely successful in combining GIS and AHP techniques. Modeling the data and weighting the layers with ArcGIS software created an accurate and reliable groundwater potential map. Many studies have demonstrated that GIS-based methods can be used as effective tools to identify groundwater potential zones without direct access to piezometric patterns and that these methods provide reliable results by integrating and analyzing various surface data such as lithology, drainage density, slope, soil type, linearity density, and precipitation in a GIS environment. The results obtained in these studies were often validated by field studies or the status of existing groundwater resources, and it was emphasized that GIS-based methods offer viable alternatives, especially where access to piezometric data is limited [40,41,42].