Evaluation of land suitability areas for irrigation using GIS and AHP-based tools in the case of Zenti River Catchment, Gofa district, Ethiopia

Abstract

Land potential assessment has enormous use for smallholder farmers for the sustainable utilization of scarce physical land resources. The study area is faced with a significant lack of available studies on irrigation land suitability due to limited information regarding soil and land resources. This scarcity of data poses a significant challenge in effectively assessing the suitability of land for irrigation purposes. The study aimed to assess the land suitability for irrigation in the case of the Zenti River Catchment, Ethiopia. To achieve the objectives, soil properties, land use/cover, slope, and proximity to a perennial river were used. This was accomplished by using the analytical hierarchical process (AHP) and GIS-based tools. Soil samples were collected and used as criteria for irrigation suitability analysis: soil salinity, soil reaction (pH), soil organic matter, soil cation exchange capacity, and soil moisture content. To evaluate the land suitability for irrigation potential, the weighted linear combination analysis of AHP and reclassified thematic layers were overlaid in the Arc GIS tool. The result showed that soil suitability and land slope factors shared higher weights of 40% and 30%, respectively. A large portion of the study area falls in a high (19%) to moderate (37%) irrigation potential zone. Decision-makers can utilize this information to plan and allocate resources for irrigation projects in the most suitable locations. Moreover, smallholder farmers can benefit from understanding, which areas are suitable for irrigation, allowing them to make informed decisions when selecting land for cultivation and adopting proper management.

Keywords:

• Irrigation potential
• land suitability
• ArcGIS
• AHP
• Zenti river catchment

Reviewing Editor:

• Marco Montemurro, Arts et Metiers ParisTech – Centre de Bordeaux-Talence, France

Subjects:

• Hydraulic Engineering
• Water Engineering
• Water Science

1. Introduction

Agricultural production is one of the most significant pillars on which the Ethiopian economy is built, as it contributes significantly to the advancement and development of the national economy as a whole. In order to achieve self-sufficiency in some strategic crops such as wheat, and maize, the state pays great attention to getting its own agriculture to be self-sufficient. According to Hussien et al. (2019, Irrigation is a crucial investment for raising rural income by boosting agricultural output. Through increased production, irrigation projects could raise agricultural productivity and contribute to the nation’s socioeconomic development. Concerns about food security are becoming more widespread throughout Africa, particularly in sub-Saharan Africa, where agriculture provides the majority of the continent’s income (more than 80% of foreign exchange earnings and more than 43% of GDP) (Belete et al., 2011; IWMI, 2010) yet agriculture is practiced in a traditional and very inefficient way.

Not only traditional practice but also the rain-fed dependent agriculture, scarcity, and uneven distribution of rainfall was leading to famine and decreasing the economic development of Ethiopians, and exposed them to innumerable consequences (Awulachew & Ayana, 2011). The majorities of Ethiopians live in rural areas and depend on agriculture for their livelihoods. Yet agricultural production hasn’t kept up with population growth, causing severe chronic malnutrition and hunger, as well as periodic drought-induced crises. Drought susceptibility is also becoming a bigger issue as a result of unfavorable climate change, fast population increase, shrinking land holding sizes, environmental degradation, subsistence farming, and rain-fed agricultural output.

It is believed that Ethiopia has 3.5 million hectares of potential irrigation land (Awulachew et al., 2007). Numerous studies have shown that the nation possesses a significant amount of arable land. Unfortunately, due to a lack of infrastructure (pumps, water conveyance structures, etc.), irrigation development, and water storage facilities, less than 5% of this potential is irrigated. The shift of the agricultural sector to agriculture development-led industrialization a more complete agricultural production system has attracted special attention recently. The agriculture industry, which is presently dominated by single-crop, rain-fed crops, could undergo significant change if irrigation is used. Using a variety of irrigation technologies, the Ethiopian government aims to increase food production per unit area of land by increasing irrigated areas Dawit et al. (2020). It is essential to have a well-developed database to help plan and design irrigation systems and manage land more efficiently. For this purpose, geographic information system (GIS) techniques can easily be employed to generate the required database via maps and attribute tables (Birhanu et al., 2019). The application of GIS can help in agricultural water management in an irrigated area to enhance water productivity. While some approaches like the fuzzy logic techniques, machine learning models, analytical hierarchy process (AHP) were preferred over others to determine the suitability of land for irrigation potential (Ozsahin et al., 2022; Ozsahin & Ozdes, 2022). AHP is a technique widely used in decision-making processes that involve multiple criteria and alternatives. Decision-makers can systematically evaluate and rank various parameters according to their relative significance from the perspective of irrigation potential by integrating AHP with GIS-based multi-criteria decision analysis (MCDA). In the context of land suitability assessment, AHP can be applied to establish a hierarchy of criteria and assign weights to each criterion based on expert knowledge. Land suitability is the evaluation and categorization of individual land areas for a particular purpose. The land potential for irrigation can be evaluated using spatial factors including soil, topography, land use and cover (LULC), and proximity to water bodies. Using GIS tools, the relative importance of each of these factors can be assessed in relation to the predetermined criteria for the appropriateness of the area for irrigation. Meron (2007) conducted research using a GIS approach to analyze the southern Abay Basin’s suitability for surface irrigation. In order to locate suitable land for irrigation in relation to the location of accessible water resources, this study took into account soil, slope, and land cover/use characteristics. Weighted overlay analysis was utilized in Arc GIS to evaluate the combined influence of these parameters for irrigation suitability analysis. Mandal et al. (2018), used four factors like soil physical properties (depth, texture, and drainage), topographic factor (slope), land cover/use and distance from water sources for land suitability assessment for potential surface irrigation of river catchment for irrigation development.

One of the problems in the study area is the scarcity of precise data and information about the suitability of various land areas for irrigation purpose. Zenti River catchment area has significant agricultural development possibilities. However, the region suffers from a dearth of comprehensive studies that effectively integrate Geographic Information Systems (GIS) and Analytic Hierarchy Process (AHP) tools to assess the suitability of land for irrigation purposes. This research gap hinders the efficient and sustainable utilization of land resources for agricultural purposes in the region. Effective planning and decision-making for agricultural activities are hampered by this information gap. The current land use practices within the study area may not be optimally aligned with the potential for irrigation. Soil degradation, water scarcity, and diminished agricultural productivity are potential outcomes of employing inappropriate land areas for irrigation, indicating unsustainable agricultural practices. In addition, the absence of a systematic method for assessing the suitability of land for irrigation may result in inadequate land use planning and decision-making procedures in the region. As a result, a better understanding of the land potential of the river is essential for better resource management and utilization. Yet, there is a lack of studies conducted on the assessment of irrigation land potential in the study area that analyse the land suitability for irrigation so that the local communities and decision makers can utilise them for resource planning and management activities to develop irrigation projects in the study area. Due to this, the major users of the river are facing a challenge to allocate the potential land suitability for irrigation development. To ensure sustainable irrigation development, one has to know the limitations and suitability of the potential land for irrigation in the area. To this end, it becomes necessary to identify land suitability so as to evaluate the status of the irrigation potential of the Zenti River catchment in the Gofa district, and propose reliable recommendations. Moreover, the objectives of this study was formulated as; to assess and identify potential land suitability areas for irrigation with regard to agriculturally influential factors such as soil chemical and physical properties, topography(slope), LULC, and river proximity, and to delineate and evaluate a possible physical irrigation suitability area of the river catchment.

2. Materials and methods

2.1. Description of study area

The study was conducted in the Zenti River catchment, which is located in the Gofa zone of the southern nations. Gofa district is one of the zones in the southern nations, consisting of seven woredas and two towns administratively, and having a population of 1.5 million. The Zenti River catchment is located in the lower Omo-Gibe river basin, situated in southern Ethiopia along the central rift valley. The Zenti River Catchment is one of the small tributaries of the Omo Gibe River Basin. It is located about 468 kilometers south of Addis Ababa on an asphalt road from Addis Ababa to Butajira-Sodo-Sawla. Geographically, it is located approximately between 6˚8'N to 6˚40'N longitude and 36˚40'E to 37˚30'E latitude of the bounding coordinate system (Figure 1). The study area lies within the altitude range of 609 m to 3303 m a.s.l. The spatial extent of the study area covers about 178,509ha.

Figure 1. Location map of the study area.

2.2. Methods of data collection and sources

To achieve the goals of the study, a variety of input data were acquired from various sources in addition to field observations. Both primary and secondary data were used in this investigation. A shape file containing the Ethiopian soil map was obtained from the Ministry of Water, Irrigation, and Electricity (FAO, 2003). The Shuttle Radar Topographic Mission (SRTM) of 30*30 m resolution digital elevation model (DEM) of the area from the Geospatial Information Institute of Ethiopia was used to generate a slope suitability area for irrigation (Gonfa et al., 2021). Soil-related factors such as soil pH, soil textures, soil moisture content, soil organic matter, soil cation exchange capacity and soil salinity were obtained from laboratory analysis. A satellite image (SPOT5) was purchased from the Geospatial Information Institute of Ethiopia. The primary data was collected via reconnaissance survey in order to identify key features of the catchment, including the existing irrigated area. Field surveys were conducted in order to collect geographic information: ground truth points for image classification, river outlets, and agricultural fields using GPS tool. The coordinates that were recorded using GPS were also used to delineate the watershed from the DEM. The collected data were then analyzed to estimate irrigation’s contribution of Zenti river catchments for small holder farmers. Table 1 shows the various data used, their sources, and their purpose in tabular form.

Table 1. Data used for this study.

No	Data Used	Purpose	Source	Date
1	Soil data	Soil suitability analysis	Ethio soil map	2003
2	Land use land cover	Land suitability analysis	SPOT5 satellite image	2016
3	DEM	Delineate watershed and generate slope	Ethiopia DEM map	2023

2.3. Land suitability and soil properties analysis for irrigation

The most suitable land for surface irrigation was selected based on each property’s specific suitability in order to assess the possibilities for sustainable land use and productivity without adversely affecting natural resources. In addition to other considerations, LULC was utilized in the land suitability class to look into possibly irrigable lands. Weighted overlay tools in the ArcGIS environment were used to classify and overlay the irrigation suitability parameters taken into consideration in this study, including the slope factor, soil factor, water proximity, and land use/cover factor. This process was in accordance with (Fikadie et al., 2022). A slope analysis was carried out in a GIS environment (Arc Tool Box). The LULC map was generated from satellite imagery produced by the SPOT satellite sensor at high resolution (1.5 m by1.5m) to identify different classes of land use/cover. As reported in Abdelkareem et al. (2018), the Kappa coefficient was utilized to account for chance agreement in the classification and give an indication of how well the classification performed relative to the likelihood of randomly allocating pixels to the appropriate categories. One of the factors used to determine whether soil is appropriate for irrigation or not is particle size. To determine soil-water mobility, one can use data from a particle size study. To know the influence of soil moisture content on irrigation suitability, the oven dry method was conducted in the laboratory. The most influential soil chemical properties like salinity, soil pH, organic matter, and cation exchange capacity were also determined in the laboratory. Moisture content, salinity, soil pH, organic matter (Mani et al., 2018; Van Beek et al., 2019), and cation exchange capacity (Olakayode et al., 2020; Yunan et al., 2018) were identified as influential parameters with most frequent crop type in the area Maize, Onion, Sugarcane and Cabbage. Soil sample spots for irrigation purposes were chosen using grid and free survey methods, and their locations were also taken using the Global Positioning System (GPS) tool based on the results of earlier soil investigations conducted by Gyekye et al. (2021). Total of twenty-five (25) soil samples were collected from irrigation sites at a depth of 20–60 cm. The composite soil samples were taken to the Arba Minch University main campus and the Abaya campus for laboratory investigation of chemical and physical soil analysis. Figure 2 shows soil sample points of the study area.

Figure 2. Soil sample points of the study area.

2.4. Analytical hierarchy process of land suitability factors for irrigation

The AHP technique helped in assigning relative weights of irrigation suitability influencing factors based on their importance and making decisions accordingly. The most relevant factors for choosing suitable land for irrigation were found by applying AHP to the perspective of land suitability factors. Satellite imagery, the Digital Elevation Model (DEM), and map information were used to identify potential irrigable sites in the study area. Each thematic map was clipped with a delineated catchment area boundary. The weighted sum of each factor was analyzed independently and used to calculate overall suitability using multi-criteria of the analytical hierarchy process (AHP) in ArcGIS. Several authors (AbdelRahman et al., 2022; Gonfa et al., 2021) used this method to assess the suitability of the land. Evaluation of land potential suitability for irrigation was conducted based on soil factors, slope, irrigable land availability, and the proximity to water resources in the area. The descriptions of the recommended scales are given in Tables 2.

Table 2. Saaty’s 9-point continuous rating scale (Saaty, 1990).

	Rating
	1	3	5	7	9
Qualitative description	Equal	Moderate	Strong	very strong	Extreme
Level of importance	Equal	More
	1/9	1/7		1/5	1/3
Qualitative description	Extreme	very strong		strong	Moderate
Level of importance	Less
	2	4		6	8
Qualitative description	Intermediate values between two adjacent judgments

Steps followed in analytical hierarchy process (AHP) in making decision using pairwise comparison matrix.

Random index for different number of criteria is shown in Table 3.

Table 3. Saaty’s Ratio Index (RI) value of consistency (Saaty, 2016).

n	1	2	3	4	5	6	7	8	9	10
RI	0	0	0.58	0.89	1.12	1.24	1.32	1.41	1.45	1.49

1. Developing pairwise comparison matrix (M) based on the numbers of factors (n), (1) $$M=\left(\begin{array}{cccc}{M}_{11}& M_{12}& \dots & M_{1n}\\ M_{21}& M_{22}& \dots & M_{2n}\\ \dots & \dots & \dots & \dots \\ M_{n1}& M_{n2}& \dots & M_{\mathit{\text{nn}}}\end{array}\right)$$(1)

Where, M is a pair wise comparison matrix. Mij, i = 1, 2, 3…n, j = 1, 2, 3…n: i is the row component and j is the column component of the matrix.

• 2. Determining the relative importance of factors with respect to irrigation potential influencing,

• 3. Calculating the consistency ratio (CR) to check whether the weights assigned are correct or not. (2) $$\mathrm{\text{CR}}=\frac{\mathrm{\text{CI}}}{\mathrm{\text{RI}}}$$(2)

Where CI is the consistency index and RI is a random index. Since the comparison was carried out through subjective judgement, some degree of inconsistency may have occurred. In order to ensure consistency with the initial preference ratings, the normalized weights of the thematic layers and those of their features were examined for consistency as recommended by Saaty (Saaty, 1990). If a consistency ratio is less than 0.1 for larger matrices, then the pairwise comparison matrix is said to be consistent, and if a CR > 0.1, the judgement weights have to be revised due to their inconsistency. When CR = 0, the constructed pairwise comparison matrix is perfectly consistent (Saaty, 1990). The consistency index may be given by: (3) $$\mathit{\text{CI}}=\frac{{\lambda }_{\mathrm{\text{max}}}-n}{n-1}$$(3)

Where, λ-is a maximum eigenvalue of the matrix and n-is the number of compared elements.

2.5. Delineation of a possible physical irrigation suitability map

Spatial analysis toolsets in the GIS Arc tools box were used to develop a suitability model and overlay the factors to map the suitable area in order to determine an overall appropriate site for irrigation. Multi-criteria analysis (MCA) was conducted using GIS with weights for each factor overlay, as mentioned in Ahmed et al. (2016) and Doyo et al. (2022). This was established using weighted linear combination (WLC) techniques that took into account all variables in an integrated layer (weighted overlay analysis in Arc GIS). WLC is the summarization of the results of the assigned final weights of each thematic map (w) and the normalized rate of each class in a thematic map (Nr). Weighted linear combination techniques can be determined using equation below: (4) $${\mathrm{S}}_{\mathrm{i}}=\left(\text{LULC}.\mathrm{w}*\text{Nr}\right)+\left(\text{PWS}.\mathrm{w}*\text{Nr}\right)+\left(\text{STw}*\text{Nr}\right)+\left(\text{SLw}*\text{Nr}\right)$$(4)

Where, Si, SL, LULC, ST, and PWS, indicate suitability index, slope factor, land use land cover, soil type, and proximity to a water source, respectively, and w is the final weight of a thematic layer and Nr is normalized rate of each class in each thematic layer. Based on the degree of suitability, the overlaid map revealed irrigation suitability, which is categorised as highly suitable (S1), moderately suitable (S2), slightly suitable (S3), and not suitable (N) (Table 4).

Table 4. Irrigation Suitability Rating.

Suitability rating range	Class	Definition
4	S1	Highly suitable
3	S2	Moderately suitable
2	S3	Slightly suitable
1	N	Not suitable

3. Results and discussion

Multiple factors that might influence the land suitability potential of the research area were gathered from various sources based on the relevant data’s accessibility and availability. In addition to the chemical and physical properties of the soils (pH, organic carbon, cation exchange capacity (CEC), pH, electrical conductivity (EC), and soil type, drainage, depth, and texture), the land suitability potentials for irrigation of the Zenti River catchment were assessed and evaluated using a number of factors, including slope, river proximity, land use, and land cover (LULC). These parameters were analyzed using ArcGIS and AHP.

3.1. Assessment and identification of land suitability area for irrigation

3.1.1. Slope suitability assessment results

Regarding the preparation of the land for irrigation as well as the operation and efficiency of irrigation, the slope of the field is a significant factor on irrigation suitability. According to the FAO land suitability classification, land slopes of the study area (Figure 3) were classified into S1, S2, S3 and N from the surface irrigation suitability perspective, i.e. from 0 to 2% as highly suitable (S1), 2–5% as moderately suitable (S2), 5–8% as slightly suitable (S3) and >8% as not suitable (N). According to the suitability results, 34% of the land was extremely suitable, 35% was moderately acceptable, 24% was only slightly suitable, and 7% was not suitable for irrigation development. Table 5 presents the suitability of slope land and its classes for irrigation purpose in hectares in the entire catchment area.

Figure 3. Reclassified slope suitability map of the study area.

Table 5. Slope suitability for irrigation and their areal coverage.

Slope (%)	Area(ha)	Area (%)	Definition	Rating	Class
0-2	57,833	34	Highly suitable	4	S1
2-5	58,645	35	Moderately suitable	3	S2
5-8	39,574	24	Slightly suitable	2	S3
>8	11,612	7	Not suitable	1	N

3.1.2. Soil suitability factors for irrigation

Soil physiochemical properties like soil salinity, soil reaction (pH), soil organic matter, soil cation exchange capacity, soil textures, soil drainage, soil depth, soil moisture content, and soil type were taken as indicators to assess general land suitability for irrigation purposes. This is so that the suitability of a piece of land for irrigation cannot be determined only based on a single factor in line with (Gonfa et al., 2021). The major soil groups identified in the study area were eutric cambisols and dystric cambisols, which have good natural fertility and are considerable for agriculture were categorised as moderately suitable (S2). Dystric nitosols (S1), eutric fluvisols and dystric fluvisols (S3) were also observed in the study area. Leptosols and orthic acrisols(N) have low moisture holding capacity, low production potential, rocky soil, poor fertility, and are also classified as not suitable (N) for surface irrigation, as shown in Figure 4. The following new values, numbered from 1 to 4, were assigned: S1 stands for ‘highly suitable (4)’ S2 for ‘moderately suitable (3)’, S3 for ‘marginally suitable (2)’ and N is ‘currently not suitable (1)’. Summary of soil suitability classification results and areal coverage are given in Table 6.

Figure 4. Reclassified soil type suitability map of the study area.

Table 6. Soil type suitability and their area coverage.

Soil type	Area(ha)	Area (%)	Definition	Rating	Class
dystric nitisols	51,016.6	31	Highly suitable	4	S1
Eutric cambisols/dystric cambisols	33,468.6	20	Moderately suitable	3	S2
dystric fluvisols/eutric fluvisols	7547.74	9	Slightly suitable	2	S3
leptosols/orthic acrisols	75,475.6	40	Not suitable	1	N

3.1.2.1. Soil texture

According to FAO, water-holding capacity was taken into consideration while classifying soil textures into four groups depending on their suitability for irrigation practices. How water travels through the soil and how much water it can hold depends on the type of soil texture. In accordance with Balew et al. (2021) and Getahun et al. (2023), which validates our findings, these are highly suitable (4) (silt loam, silty clay loam, and clay), moderately suitable (3) (silty clay and clay loam), marginally suitable (2) (sandy clay loam), and not suitable (1) (sand clay, silt). For instance, the improved nutrient and moisture content of clay soil texture makes it ideal for surface irrigation. Figure 5 shows soil textural suitability map of the study area.

Figure 5. Reclassified Soil Texture Suitability Map of the Study Area.

3.1.2.2. Soil depth

Using FAO standard guidelines, the appropriateness of the soil depth for irrigation purposes was determined. For irrigation purposes, a deeper soil depth was preferred since it will enable the provision of more water and nutrients to plants than a shallow soil depth. The soil depths of the study area were categorized as shallow (less than 10 cm) not suitable, medium (below 50 cm) moderately suitable, and deep up to 150 cm (highly suitable), as shown in Figure 6. Shallow soil depth indicates the minimum availability of nutrients and lower moisture-holding capacity of the soil. The soil depth, which is less than 10 cm, limits the rooting depth, and the available soil water for plants is decreased.

Figure 6. Reclassified soil depth suitability map of the study.

3.1.2.3. Soil drainage

One of the key considerations when assessing irrigation potential is soil drainage. The growth and productivity of irrigated crops depend on the soil’s ability to drain properly, which is essential for efficient root aeration in plants. Three classes of soil drainage (well-drained, imperfectly drained, and poorly drained) were identified in the study area based on FAO guidelines criteria. According to Birhanu et al. (2019), well-drained soils facilitate crop nutrient and water infiltration, cultivation, and aeration. As depicted in Figure 7, the majority of the research region was determined to possess well-drained soil, which is crucial for the development of irrigation.

Figure 7. Reclassified Soil Drainage of the Study Area.

3.1.2.4. Overall soil- related factors suitability for irrigation

The soil in the Zenti River catchment was generally characterised in the weighted overlay analysis as moderately to slightly suitable in about 85% of the study area and not suitable to highly suitable in about 15% of the area. The area distribution of the watershed’s overall soil suitability map is depicted in Figure 8.

Figure 8. Overall soil- related factors suitability map for irrigation.

3.1.2.5. Acidity or pH of soil

Due to its influence on practically all soil system processes, pH is referred to as a ‘master variable’ in soil science. Nutrient availability, soil life, and activity all affect the health of crops and other soil life (McFarland et al., 2015). Low soil pH has an impact on every aspect of crop productivity. Low-pH has an impact on the availability of some nutrients, which impacts how quickly weeds and crops may grow. In low-pH soil, pesticides, including herbicides, may be less effective or persist longer. Previous research identified the well-known yet significant pH levels for a variety of plant nutrients. The UK and most other nations are advised to keep soil pH values at optimal levels of 6.5 (5.8 in peaty soils) for cultivated land and 6.0 (5.3 in peaty soils) for grassland. Managing soils to attain the whole range of crop-specific pH values in a crop rotation is not practicable. In general, crops vary in their compatibility with soil pH range. Some crops may be intolerant to a specific soil pH due to a specific mechanism. Soil pH 5.5, for example, is unsuitable for soybean plants when molybdenum levels are low, but it is ideal for soybeans when molybdenum levels are high. Most agricultural crops grow best when soil pH is about 7.0 (neutral) (Goulding, 2016). This demonstrates the critical importance of bringing both acidic and alkaline soils to neutral soil pH values for crop performance. For most agricultural crops, the ideal range is between 5.5 and 7.5. Moreover, some investigations had made different classes with regard to pH; the value in range 5–8.5, 4.5–5 and 8.5–9, 4–4.5 and 9–9.2, <4 and >9.2 were classified as highly suitable (S1), moderately suitable (S2), marginal suitable (S3) and not suitable (N) respectively (Dinku & Habtamu, 2022). According to investigations results in terms of soil pH, soil samples collected from various agricultural areas for this study were found to be highly suitable values.

3.1.2.6. Moisture content

A crucial factor in crop productivity is the relative humidity of the surrounding air since it affects the water balance and photosynthetic processes in plants. The soil provides nutrients, water, anchorage, and air for plant growth. Hence, knowing the ideal moisture for various crop types is one important issue which needs attention (Chia & Lim, 2022). Soil moisture content before planting is a crucial aspect of agronomic management since it affects the emergence and roots of seedlings, which in turn can affect crop yield. But the precise relationship between starting soil moisture and crop yield remains unclear, especially in developing nations where food security is a major concern and there is a dearth of field data on soil moisture and climate in line with (Robinson et al., 2008). Crop plants are susceptible to climate change; a lack of or excess water, as well as sub-optimal temperatures, can produce significant abiotic stressors. A 40% water deficiency in relation to wheat and maize needs may reduce yield by 20% and 39%, respectively (Rosenzweig et al., 2014). For sustainable agriculture, efficient use of finite water resources is both a marketing goal and an environmental obligation. Soil moisture monitoring is essential in current agricultural production, which is intensive while still being environmentally benign (Osakabe et al., 2014). The ideal range of soil moisture content for crops varies according to plant species, although for most crops, the range is between 20% and 60%. The laboratory results of moisture content revealed that the soil types identified in these particular areas were suitable or ideal for crop production as they were in the range of 20–60%. Table 7 shows the laboratory analysis of soil chemical properties of the study area.

Table 7. Soil chemical properties results from lab analysis.

pH	Moisture Content (%)	CEC(Cmol (p+)/kg)	Organic Matter (%)	Salinity (ds/m)
7.12	43.60	27.1	2.48	1.40
7.11	43.61	26.5	2.30	1.23
7.06	43.55	26.4	2.52	1.52
7.05	35.14	28.3	2.98	1.35
7.06	35.38	29	3.09	1.30
7.1	34.93	28.5	3.28	1.44
7.24	37.39	18.56	3.10	3.44
7.26	36.97	17.67	2.98	3.50
7.28	37.36	18.39	3.23	3.65
7.46	45.54	49.56	3.56	4.90
7.43	44.76	50.23	3.23	4.67
7.23	45.23	49.24	3.41	4.75
7.44	35.42	39.06	2.77	2.99
7.42	35.09	38.12	2.57	3.20
7.4	34.48	38.43	2.58	3.12
5.25	35.58	17.89	3.18	3.23
5.24	36.52	18.24	3.38	3.44
5.26	36.98	18.45	3.27	3.12

Source: Laboratory data (2015).

3.1.2.7. Salinity of soil

Around the world, soil salinity has a significant impact on crop production. 33% of irrigated land and 20% of all cultivated land are damaged and degraded by salt. Over 800 million hectares of arable land are worldwide; however, managing them sustainably could be difficult due to soil salinity, especially in flooded, dry and semiarid areas. The salinity of soil were classified in different classes according the range < 4, 4–8, 8–12, > 12 which were classified as highly suitable (S1), moderately suitable (S2), marginal suitable (S3) and not suitable (N) respectively as per (Getahun et al., 2023). Therefore, all soil samples were highly suitable except one area, which was probable due to various factors like topography, crop type, frequency of agriculture and so on.

3.1.2.8. Cation exchange capacity and organic matter

The loss of organic matter from agricultural soils limits our ability to feed a growing population and ameliorate the effects of climate change. Addressing these issues necessitates land use practices that increase soil carbon (C) and contribute to food production. With regard to organic matter, similar classifications were made. The values in the >2.18, 1.15–2.18, 0.74–1.15, and < 0.74 were classified as highly suitable (S1), moderately suitable (S2), marginally suitable (S3), and not suitable (N), respectively, as per (Getahun et al., 2023). Numerous investigations had made different classes with regard to CEC; the value in range 18–60, 15–18, 9–15, <9 were classified as highly suitable (S1), moderately suitable (S2), marginal suitable (S3), and not suitable (N), respectively, as per (Dinku & Habtamu, 2022; Hagos et al., 2022; Mandal et al., 2018). With regard to CEC, almost all soil from the study area was categorized as highly suitable whereas regarding organic matter all samples were greater than 2.18 being highly suitable for agriculture.

3.1.3. Land use/land cover suitability for irrigation

According to the land use/land cover reclassification result in Figure 9, the major land use/land cover of the area was categorised as farmland, water bodies, wetland, forests, settlements, bare land, closed and open shrub land, and grassland. From the perspective of irrigation suitability, farmland is categorised as highly suitable for irrigation, whereas grassland is moderately suitable for irrigation potential. The areas covered by bare land, closed, and open shrub land are slightly suitable for irrigation, whereas water bodies, wetlands, forests and settlements are currently not suitable for irrigation potential. The overall accuracy assessment of the classified images in Arc Map was 91%. According to Abdulkareem et al. (2017), there is a high level of acceptability indicated by the high kappa coefficient and the overall accuracy of the image classification. When the categorised image’s overall accuracy is compared to the land cover conditions derived from the associated ground truth, great accuracy is attained. This shows that the classified LULC image is accurate. Table 8 depicted the error matrix used to evaluate the LULC's correctness.

Figure 9. Reclassified land use/cover suitability map of the study area.

Table 8. Error matrix to check overall accuracy of LULC classification.

Classes	Water body	Forest	Shrub land	Grass land	Settle Ment	Farmland	Wetland	Total	User accuracy
Water body	6	0	1	0	0	0	0	7	86
Forest	0	20	0	1	0	0	0	21	95
shrub land	0	0	14	0	1	0	0	15	93
Grassland	0	0	0	12	1	0	0	13	92
Settlement	0	1	0	0	20	0	0	21	95
Farmland	1	1	1	0	0	35	2	40	88
Wetland	0	0	0	1	0	0	11	12	92
Total	7	22	16	14	22	35	13	129
Producer accuracy	86	91	88	86	91	100	85

Overall classification accuracy = 91% and Kappa coefficient (K)= 89%.

3.1.4. Proximity to water source (river)

The physical closeness to water sources has been calculated using the spatial analysis tools of Arc GIS layers, as shown in Figure 10. The result of the analysis to estimate the suitability of the land from the stream revealed that 69,043.84 ha of it are highly suitable (<2 km), while 48,078.08 ha are moderately suitable (2–4 km), 18,743.36 ha of land are slightly suitable for irrigation development (4–6 km), whereas 31,765.28 ha are currently not suitable for irrigation (>6 km).

Figure 10. Proximity to water source suitability map reclassified.

3.2. Delineation and evaluation of suitable sites for irrigation potential

The multiple influencing factors of land suitability for irrigation, including slope, land use/land cover, soil factors, and proximity to water sources, were examined and given relative importance values based on Saaty’s rating scale (1–9) through judgement based on field conditions and data gathered from scientific research reviewed. Equation 1(1) $$M=\left(\begin{array}{cccc}{M}_{11}& M_{12}& \dots & M_{1n}\\ M_{21}& M_{22}& \dots & M_{2n}\\ \dots & \dots & \dots & \dots \\ M_{n1}& M_{n2}& \dots & M_{\mathit{\text{nn}}}\end{array}\right)$$(1) was used to pair up all the factors. Table 9 shows land suitability factors and their comparison matrix.

Table 9. Land suitability factors and pairwise comparison matrix.

Factors	Soil	Slope	LULC	PWR
Soil	1	2	2	3
Slope	½	1	3	2
LULC	½	1/3	1	2
PWR	1/3	1/2	1/2	1
Sum	2.33	3.83	6.50	8.00

LULC: land use/land cover: PWS: proximity to water source.

The lower triangular matrix was obtained from the values for the inverse comparison of the upper triangular matrix. The normalized weight of the irrigation suitability factors was calculated by dividing all the elements in each column by the sum of the columns as shown in Table 10.

Table 10. Normalized land suitability factor comparison matrixes.

Factors	Soil	Slope	LULC	PWS	Eigen vector	Weight (%)
Soil	0.43	0.52	0.31	0.38	0.401	40
Slope	0.21	0.26	0.46	0.25	0.30	30
LULC	0.21	0.09	0.15	0.25	0.18	18
PWS	0.14	0.13	0.08	0.13	0.12	12
Sum	1.00	1.00	1.00	1.00	1.00	100

CR = CI/RI where, CI= (λmax − n)/(n − 1) and λmax= 4.083 and for n = 4, RI = 0.89 then, CI = 0.028 and CR= 0.031.

Upon checked a paired matrix, the consistency ratio (CR) was found to be 0.031 using equation 2(2) $$\mathrm{\text{CR}}=\frac{\mathrm{\text{CI}}}{\mathrm{\text{RI}}}$$(2) , which is considered acceptable since it is less than 0.1 (10%) (Saaty, 1990). Our results are within acceptable limits as in accordance with Saaty, a pairwise comparison’s judgement loses credibility if the consistence ratio (CR) is more than 0.1. The resulting pairwise comparison matrix’s judgement was therefore consistent, and the weights that were assigned were acceptable and having an influence on irrigation potential. This indicates that the weights of various factors are in the acceptance limit and exhibit good consistency. Below are the soil-related factors displayed in Tables 11 and 12, along with their respective pairwise comparison matrix and normalized comparison matrix of sub-criteria.

Table 11. Soil-related factors and their pairwise comparison matrix.

Factors	Soil types	Depth	Texture	Drainage
Soil types	1	2	3	4
Depth	½	1	2	3
Texture	1/3	½	1	4
Drainage	¼	1/3	¼	1
Sum	2.08	3.83	6.25	12.00

Table 12. Normalized comparison matrix of sub-criteria (soil-related factors).

Factors	Soil types	Depth	Texture	Drainage	Eigen vectors	Weight (%)
Soil types	0.48	0.52	0.48	0.33	0.45	45
Depth	0.24	0.26	0.32	0.25	0.27	27
Texture	0.16	0.13	0.16	0.33	0.20	20
Drainage	0.12	0.09	0.04	0.08	0.08	8
Sum	1.00	1.00	1.00	1.00	1.00	100

CR = CI/RI where, CI= (λmax − n)/(n − 1) and λmax = 4.09, for n = 4, RI = 0.89 then, CI = 0.0158 and CR= 0.012.

The summary of weights assigned to each pair of factors in the overlay thematic map used to assess the study area’s appropriateness for irrigation are shown in Table 13.

Table 13. The weight of influencing factors in determining the irrigation potentials.

Factors	Weight (%)
Soil suitability	40
Slope suitability	30
Land use land cover suitability	18
Distance from water source suitability	12
Sum	100

From the result indicated in Table 13, the analytical hierarchy process (AHP) prioritizing result shows that soil factors and slope share the highest weightage for the identification of irrigation potential areas. Among these, the soil factor was assigned the highest weight because it plays a dominant role in irrigation potential sites and its values in the row are greater than 1, and ‘slope suitability’ is the second-most crucial factor because it only has one less-important value. The result also shows that all criteria are significant for determining the irrigation potential area, even though their relative contributions vary. In order to create the final suitable irrigation zones, the weight overlay analysis model in ArcGIS used the pair-wise cross-comparison matrix from the AHP technique together with their weights. The study region has been classified into four categories based on the analytical results: highly suitable irrigation land, moderately suitable irrigation land, slightly suitable irrigation land, and currently not suitable for irrigation.

3.2.1. Weighted overlay analysis in ArcGIS

Arc GIS 10.4's weighted overlay analysis using equation 4(4) $${\mathrm{S}}_{\mathrm{i}}=\left(\text{LULC}.\mathrm{w}*\text{Nr}\right)+\left(\text{PWS}.\mathrm{w}*\text{Nr}\right)+\left(\text{STw}*\text{Nr}\right)+\left(\text{SLw}*\text{Nr}\right)$$(4) was used to evaluate a land potential site for irrigation purpose. Each parameter was rasterized first, and a reclassified suitability map was used as input for the analysis. From the potentially suitable irrigable area suitability map depicted in Figure 11, 19% (34,551 ha) of the study area would be highly suitable for agricultural production, 37% (67,733 ha) would be moderately suitable, 41% (74,557 ha) would be marginally suitable land, and 3% (1666 ha) is currently not suitable land for irrigated agriculture or agricultural production. Highly Suitable agriculture land was characterized by a flat zone comparatively, in which the majority of the area was below 1000 m, the slope varied from 0 to 2%, the most dominant land cover was grasses, and most of the agriculture part was cultivating. Moderately suitable zones have elevations of 700–1000 m, and the slope varies from 2–5%. Marginally suitable zones have an elevation of the major part above 1000 m, a dominated rocky surface, and slopes ranging from 5–8%. Currently not suitable for agriculture due to dense forest cover, water bodies, and the dominant spread of not suitable soil factors. Mostly, the elevation varied from 2600 to 3303 m, with exposed rocky lands and precipitous slopes. The results of the irrigation suitability model’s weighted overlay analysis are shown in Table 14 and Figure 11 below.

Figure 11. Overall irrigation suitability map of the study area.

Table 14. Overall suitable sites and their area coverage in the catchment.

No	Potentially suitable class	Area coverage
		Ha	%
1	Highly suitable	34,511.27	19
2	Moderately suitable	67,733.13	37
3	Slightly suitable	74,556.73	41
4	Not suitable	1666.13	3
Total		178,507.26	100

3.3. Validation

To validate the evaluation of land suitability areas for irrigation, we performed field surveys and experimental methods. We have conducted on-site visits to gather ground truth data related to land suitability for irrigation purposes. Field surveys were conducted to collect coordinate points from the irrigation site and compared with land suitability result obtained from ArcGIS tools.

In addition to ground truth data, soil samples were collected within the study area that were identified as highly suitable, moderately suitable, slightly suitable, and not suitable for irrigation based on the ArcGIS and AHP-based analysis. Soil samples collected from these locations were taken for laboratory analysis to determine their suitability for irrigation. Finally, validating the experimental findings by comparing them with real-world irrigation practices in the study areas was conducted. The areas that were not part of the experimental parts were extrapolated with the findings to other parts of the study area with similar characteristics. The identified suitable areas for irrigation purposes using ArcGIS have been cross-checked with the locations of existing irrigation schemes in the study area. The model results were compared with ground-truth data and existing irrigation practices to assess the accuracy and reliability of the results. The overall accuracy of the land suitability evaluation result was conducted to validate the accuracy and reliability of the results obtained through the ArcGIS and AHP-based tool assessments. The validation result showed that around 85% of the experimental and field observation results are in agreement with the land suitability potential map produced in this study using ArcGIS.

4. Conclusions

The study was conducted to assess the land potential suitability of the Zenti River catchment, for irrigation development in the area. Land suitability was analyzed using ArcGIS- based multi-criteria AHP. The results of AHP revealed that soil suitability, land slope, and LULC factors shared higher weights of 40%, 30%, and 18%, respectively. The land’s slope has a higher effect on surface irrigation for gravity flow, and its suitability influences land preparation, affecting a sizeable initial expenditure and decreasing economic viability. Such information is needed to sustain the new projects and the livelihoods of communities that rely on the subsistence farming system. Based on the LULC, soil, slope, and proximity to water resources, the study results showed that approximately 56% of the study areas are in the range of high irrigation potential to moderate irrigation potential for irrigation development. Based on the availability of possible irrigation land, the research area’s irrigation potential sites have been estimated and mapped. The results revealed that sufficient land is available to expand irrigable areas, improve the livelihoods of local communities by implementing an enhanced irrigation system, and improve the productivity of crops in the area. There were good correlations found between the model output and field observations and experiments. The validation result indicates that approximately 85% of the field inventory data aligns with the land suitability map generated by ArcGIS in this investigation. Future irrigation projects and activities will therefore employ these updated resources in conjunction with ArcGIS to increase land utilization in the study area.

Acknowledgements

Authors would like to Arba Minch University Sawla Campus, Arba Minch University for their valuable assistances. All reference materials used in the researches are duly acknowledged.

Data availability statement

All data that support these findings are provided in the manuscript.

Disclosure statement

All reference and material used are acknowledged and cited well. Arba Minch University funded this study with funding code GOV/AMU/WRAM/CEAT/CE/62/14.

Additional information

Funding

Arba Minch University funded this study with funding code GOV/AMU/WRAM/CEAT/CE/62/14.

Notes on contributors

Diriba Worku

Diriba Worku is currently works at the department of Civil Engineering, Arba Minch University as a lecturer for more than three years. Diriba does research in ground water prospective mapping using and irrigation potential assessment using ArcGIS and AHP based tools. His research interests encompass, improving water resource management, assess the effects of climate change on water resource, sediment transport modeling in rivers and estuaries system, design and optimization of dam and reservoir systems, groundwater modeling, and others. In parallel with teaching and learning processes, he has been participating in different research works and community engagement activities.

Abuye Boja

Abuye Boja is currently pursuing a PhD at Yokohama National University, Japan. Prior to this, he served as a university lecturer for over four years at Arba Minch University. His research interests encompass soil stabilization, embankment construction, the application of geosynthetic materials in civil engineering structures, soil testing, pavement design, permeability, and seepage analysis, among others.

Adugna Fantu

Adugna Fantu is a lecturer and researcher, currently working at the department of civil Engineering, Arba Minch University, Ethiopia. I received my MSc degree in 2020 at Arba Minch University. I conducted several researches at various areas like highway drainage, soil properties, construction materials, groundwater hydrology, mapping and irrigation potential assessment using ArcGIS and AHP-based tools. my research interests includes evaluation of highway drainage problems, capacity and performance evaluation, investigation of soil properties of rural high way, irrigation potential, ground water and other construction materials characterization and modeling. In addition to this I have been participating in different research and community service projects under university.