Keywords

Site Suitability , Remote Sensing , Irrigation , GIS , AHP

1 . INTRODUCTION

Due to heavily reliance on rain-fed agriculture and the unpredictable nature of rainfall, alternative strategies for boosting agricultural productivity must be developed in Ethiopia and the irrigation development should be maximized (Wubalem, 2021). Ethiopia is currently working to establish effective water resource management strategies and to take advantage of the potential irrigation in its river basin (Kassie, 2020). One of Ethiopia’s longest basins with the potential for surface water irrigation is the Upper Blue Nile River Basin (Teshome and Halefom, 2020; Dawit et al., 2020). However, there is a need for clear and flexible irrigation management. The river basin features in the Wama watershed are vital to know the entire potential of the land resources and available surface water for irrigation (Burayu, 2022). Growing irrigation demand combined with shifting land use patterns, population growth, climate change and variability is a serious problem for surface water irrigation in the Wama watershed. Irrigation is one of the solutions to the food shortage in the region where rain-fed agriculture practiced (Wubalem, 2021; Gonfa et al., 2021). It moves water from a source with plenty of water to fields with little water because of constructed structures. It can contribute to the reduction of destitution and the achievement of food self-sufficiency by giving farmers more income during the dry seasons (Kebede and Ademe, 2016).

In the Wama Watershed, rain-fed agriculture is the only practiced method for growing crops (Wubalem, 2021). Because of the region’s high level of regional and temporal rainfall variability, food insecurity is a prevalent problem (Nigussie et al., 2019). Uncertainty in rainfall distribution resulted in crop failure and droughts. As a result, in the face of even small setbacks, food insecurity can soon turn into famine, wreaking havoc on the existence of the poor rural community (Shitu and Berhanu, 2020). The most important strategy for transforming low-output rain-fed agriculture into the most effective and efficient irrigated agriculture is to use water to quadruple crop output during the arid season (Aybehon and Tiku, 2022; Negasa and Wakjira, 2021; Shitu and Berhanu, 2020). However, there hasn’t been any study done on-site appropriateness evaluations for surface irrigation (Burayu, 2022). To sustain rain-fed agriculture through irrigation, which will help to reduce poverty in the watershed, site suitability assessment for surface irrigation is essential in this area (Nigussie et al., 2019). A necessity for better land resource usage, which helps in the optimization and sustainability of these land resources’ production, is proper site suitability appraisal of the available land resources in the irrigation command area (Negasa and Wakjira, 2021). Access to irrigation influences both changes in land use and the adoption of intensive farming techniques. On the other hand, irrigation is not always appropriate for all regions (Teshome and Halefom, 2020). The AHP method is commonly employed  (Saaty, 1980) to account a number of factors including land use, topography, and various soil properties (physical and chemical) for classifying surface irrigation sites according to their suitability as one of the crucial steps in planning for many objectives. The main goal of this study is to detect the suitable lands for surface irrigation in the Wama watershed using the AHP, RS and GIS techniques. This analysis will help in selecting appropriate sites for surface irrigation and planning and managing the land use.

2 . MATERIALS

1.1 Study Area

Didessa sub-basin of Wama (Figure 1) is one of the most significant watersheds according to the yearly contribution to runoff. Geographically, the sub-basin is located between 08º10ʹ and 9º10ʹ N and 36º 05ʹ and 36º 05ʹE in Western part of Ethiopia (Burayu, 2022). The elevation in the basin ranges between 1174m and 3083m. The catchment area is of about 3344.4 km² with mean annual rainfall of 1934.64 mm and the maxima in Didessa sub-watershed (Burayu, 2022). The mean annual runoff is about 448mm. The humid tropical climate observed in the watershed with high yearly rainfall during rainy periods. The temperature ranges from 21-31ºC and 10-15 ºC. Wama is dominated by the Jimma Volcanic, the Wellega Basalt in the middle, and the undifferentiated lower complex, Wellega Basalt, and Adigrat Sandstone in the lower parts.

Figure 1. Study Area: Wama watershed in Ethiopian river basin

1.2 Data

Site suitability study for surface irrigation requires a sufficient amount of high-quality data.  As a result, relevant data about the topography, accessibility, soils, and water source was acquired from a variety of sources. These parameters were compiled by the FAO and USGA. Depending on the availability of data, the regional circumstances in the watershed, the agricultural background, and the appraisal of the literature, eight input factors namely slope, elevation, soil types, soil textures, soil drainage, soil depth, and LULC were selected for investigation. The information about physical characteristics of soil including soil depth, soil drainage, soil texture, and soil type were procured from the Food and Agriculture Organization (FAO, 1985). The suitability of land for agricultural use will depend on several factors, including the topographic features of the area. While elevation can determine the local climate, a slope can control how quickly erosion and deposition occur. The slope and elevation are topographic characteristics, which were retrieved from the DEM. The distance from streams was calculated using Euclidian distance. The map showing LULC of the region was created using and Landsat ETM+ data and supervised classification technique in GIS software.

3 . METHODOLOGY

In this study, suitable sites for surface irrigation were detected based on datasets of eight land surface characteristics viz. slope, soil types, soil textures, soil drainage, soil depth, LULC, elevation, and distance from streams. The data of input parameters for a GIS-based site suitability were transformed into the same pixel size of 30m x 30m with an equivalent projection. The significance of each sub-criteria was used to reclassify and standardize the rasterized characteristics. The weight assigned to each criterion was then determined using a pairwise comparison matrix. The weighted overlay algorithms in ArcGIS were used to add the weighted criteria and provide a site appropriateness index. After all, natural break method in the ArcGIS was used to reclassify the site suitability indexed map into four suitability classes (Figure 2).

Figure 2. Methodology

3.1 Weight Assessment and Normalization

A pairwise comparison matrix was created for each map based on the weighting of each criterion (Saaty 1980)  using AHP based Multi-Criteria Decision Making (MCDM) technique along with the rates for classes in  input layers namely slope, elevation, soil types, soil textures, soil drainage, soil depth and LULC. The matrix was used to compare the criteria based on Saaty’s scale of 1 to 9 (Table 1). The relative scores of their classes were taken into account while calculating the cumulative weightages of the criteria. Weightages calculated using AHP technique (Saaty, 1980) and normalized considering the relative importance of criteria and classes to determine the potential suitable sites for surface irrigation. The Consistency Ratio (CR) was calculated in accordance to evaluate the normalized weights of distinct thematic layers and their specific classes.

3.2 Overlay Analysis

After allocating rates for classes within a layer and weights for thematic layers, the final map (Figure 11) was produced by superimposing all thematic maps  (Burayu, 2022; Neissi et al., 2019; Girma et al., 2020) with the help spatial analysis tool in GIS software.

4 . RESULTS AND DISCUSSION

4.1 Thematic layers

One of the crucial elements in a site suitability analysis for surface irrigation is the preparation of thematic layers. For this study, a single database for criteria maps was created. The characteristics were transformed into a projection system and equal resolution.

4.1.1 Slope

The slope, which is typically given in percent, is the inclination or gradient of a surface. Due to its impact on runoff, drainage, erosion, and crop selection, the slope is crucial for the creation and maintenance of soil. The gradient has a significant impact on duration of irrigation cycle, crop adaptation, erosion prevention and irrigation technique. As the gradient rises, the risk of erosion increases in surface irrigation. So, the water management is useful to reduce the length of irrigation runs and crop options. The irrigation management become more significant in steep sloped area. In general, steep gradients lead to reduce productivity and greater production costs. Therefore, slopes were derived using digital elevation model (DEM) with 30m resolution from the Shuttle Radar Topography Mission (SRTM) and used for this site suitability analysis for surface irrigation. The raster layer was classified into four groups with corresponding suitability ratings of 1 to 4: (4 = Highly suitable, 3 = Moderately suitable, 2 = Marginally suitable, and 1 = Unsuitable).The result obtained after reclassifying the slope map was that 15% (448.1km²) of the watershed was less than 2%, 39 % (1289km²) of the watershed, 2-5%, 23% (756.3km²), 5-8% and 24% (811.5km²) of the watershed is higher than 8% slope. This shows that 76% (2532.9km²) of the study area was revealed to be appropriate for surface irrigation.

Figure 3. Slope

4.1.2 Elevation

The elevation was rescaled and classified into four classes: extremely suitable class (4) for elevation ranges from 1170m to 1500m, moderately suitable class (3) for elevation ranges from 1500m to 1750m, marginally suitable class (2) for elevation ranges from 1750m to 2060m, and unsuitable class (1) for elevation ranges from 2060m to 3080m. The highly suitable land class is covered by approximately 35% (1174km²), the moderately suitable land class (3) is covered by approximately 31% (1032km²), and the marginally suitable land class (2) is covered by approximately 21% (694km²), and the unsuitable land class (1) is covered by approximately 13% (174.2km²).

Figure 4. Elevation

4.1.3 Soil Depth

Based on the required soil depth for the majority of common crops to have the option of surface irrigation, the soil depth was classified into acceptable classes. Rating parameters for soil depth were supplied and weighted to determine if the study area was suitable for surface (gravity) irrigation. Based on the suitability score assigned for each soil depth, a map showing the soil depth suitability after being converted to raster and divided into four categories with accompanying suitability scores of 1 to 4. According to the suitability map, 32% of the area (1078.5km²) is classified as highly appropriate, compared to 21% (691.1km²) for moderate suitability and 47% (1574.7km²) for marginal suitability.

Figure 5. Soil depth

4.1.4 Soil Drainage

Normal plant development is possible due to soil drainage. Elevation of the soil drainage is important when choosing sites for surface irrigation, especially with diverse highland crops. Ample soil drainage is crucial to ensuring continued production and enabling productivity in agricultural operations. The efficient removal of the additional water and salts that irrigation contributes requires consideration of new irrigation facilities. A area with efficency of drainage may be classified as well-drained, moderately drained, imperfectly drained, severely drained, or extremely poorly drained depending on the techniques used to measure a  permeability of soils. Therefore, the soil drainage properties were classified into well drained, moderately drained, imperfectly drained, and poorly drained. In our case, we give suitability class and score for each of the soil drainage classes as  4 (highly suitable), 3 (moderately suitable), 2 (marginally suitable), and 1 (unsuitable). We convert soil drainage map to raster and reclassify it to know area coverage of each soil drainage classes. From drainage suitability map, 75.4% (2522km²) is classified as  4 (highly suitable), 7.5 % (250km²)  is 3 (moderately suitable), 16% (535.6km²) is 2 (marginally) and 1.1%  (36.4km²) is unsuitable (Figure 6).

Figure 6. Soil drainage

4.1.5 Soil Type

There are four types of soils observed in the Wama watershed: Humic Nitisols, Humic Alisols, Eutric Vertisols, and Lithic Leptosols. Most of the Wama watershed is covered by Humic Nitisols. The soil type mentioned above are standardized for land suitability assessment as  Humic Nitisols are 4 (highly suitable), Eutric Vertisols are 3 (moderately suitable), Humic Alisols are 2 (marginally suitable) and lithic leptosols are 1 (unsuitable). We convert the soil type map to raster and reclassify it to know the area coverage of each soil type class. From soil type suitability map 77.24% (2583km²) (highly suitable), 17.93% (599.72km²) (moderately suitable), 4.79% (160.33km²) (marginally suitable) and the left 0.0004% ( 1.3km²) (unsuitable) (Figure 7).

Figure 7. Soil types

5.1.6 Soil Texture

According to the FAO (1985) guidelines for site suitability classification, soil textural classes like clay, clay loam, and loam soil were classified as highly suitable (4), silt and silty loam were classified as moderately suitable (3), and sandy loam soil texture was classified as marginally suitable (2) for surface irrigation with the restriction factor of high infiltration rate. The classes of soil texture in the study region were determined by the percentage of the soil hydrological group, which includes loam, clay loam, and clay. The soil texture classes and their respective suitability levels are highly suitable (4), clay soil is classified as moderately suitable (3) and sandy loam is as marginally suitable (2). From soil texture suitability map 82% (2743km²) (highly suitable), 17.93% (599.72km²) (moderately suitable), and 0.0004% (1.3km²)  (marginally suitable) (Figure 8). 

Figure 8. Soil texture

4.1.7 Land Use Land Cover

Another element used to assess the suitability of the area for irrigation is land use land cover (LULC). The Landsat 8 image was used to create the LULC map. It was categorized as very suitable (4), moderately suitable (3), marginally suitable (2), and not suitable (1). While grassland is rated as somewhat suitable (3), cropland was categorized as highly suitable (4). Due to the initial land investment required, woodland, shrub, and bush were rated as only slightly acceptable (2). The classification of the forest, barren land, water body, and settlements is unsuitable (1) (Figure 9). LULC categorized under highly suitable class covers 47% (1582.4km²), moderately suitable 15% (500km²), marginally suitable class 15% (500km²) and 23% (775km²) is under unsuitable class. 

Figure 9. LULC

4.1.8  Distance to Streams

One of the most crucial aspects of site suitability evaluations for surface irrigation is the assessment and determination of the distance to the water source. Euclidean distance in ArcGIS was used to calculate the distance from the water source, which was then categorized into four categories: 0-1.5km (very suitable, 4), 1.5-3km (moderately suitable, 3), 3-5km (marginally suitable, 2), and >5km (unsuitable, 1) (Figure 10).  Distance to streams categorized under highly suitable class covers 30.7% (1026.6km²), moderately suitable 24.6% (823.46km²), marginally suitable class 19.5% (652km²) and 25.2% (842.25km²) is under unsuitable class.

Figure 10. Distance to streams

4.2 Weighting and Combining

The parameters used to generate the rasterized and categorized surface irrigation site suitability maps have been weighted. Saaty’s approaches were used in this study for each map and a pairwise comparison was prepared using 9-point significance scale in AHP. AHP is a multi-criteria decision-making strategy that offers a logical way of assessing and taking into consideration the consequences of numerous issues by connecting various level-dependent, qualitative and quantitative information. It was a method for accurately estimating the pairwise comparison-based qualified significance of a set of performances. Prioritizing the relative importance of each element that was qualified for a particular factor was done using weighting procedures. The weighted overlay element is more significant concerning other factors the higher the burden. A consistency ratio of 0.1 was considered acceptable. The calculated consistency ratio of 0.099 was accepted to demonstrate a certain matched weight (Saaty, 1978). The slope, elevation, soil texture, soil drainage, soil depth, LULC, and soil types all receive substantial weights in the pairwise evaluation, but the distance to the stream is assigned the largest weight (Saaty, 2004). The effectiveness of careful effect on the origin of contribution requirements is explained by numerical algorithms, and this influence is mutually influenced by various mathematical or logical methods of assessing disagreement. In this method, the influence value range of 1 to 9 was assigned to each aspect, and relationship outcomes were replicated by professionals. Using the weighted linear combination technique, each map layer was a critical step in the AHP. On a scale from 1 to 1/9, map was stacked in GIS spatial assessment for an appropriate site zone imitation (Burayu, 2022). The pairwise comparison matrix was prepared to estimate the relative importance of the elements (Saaty, 1980). The relationship between a value and intensity of less importance was as follows: 1= equal importance, 1/3 = moderate, 1/5 = strong, 1/7 = very strong, 1/9= extremely at the contrary, high significant variables were rated between 1 and 9. Each map layer was piled in a final GIS spatial assessment for suitable site zone imitation using the weighted linear combination technique.

4.3 Suitable Site Mapping for Surface Irrigation

Analyzing the suitability of a site is crucial to identifying the best locations for surface irrigation in light of numerous considerations. The goal of the current study was to identify the likely viable farmlands for surface irrigation within the Wama watershed. In this work, the association between soil physical characteristics, topography (slope and elevation), and LULC determinants on the appropriateness of locations for surface irrigation was assessed using an AHP method with GIS. An 8x8 pairwise comparison matrix was used to establish the relationship between governing criteria and site suitability (Table 1). Eight governing characteristics, including topographic factors (slope and elevation), land use, physical factors (soil texture, drainage, type, and depth), and distance to water source were taken into account for this study. Eight important factors were compared one to at least one in the pairwise comparison matrix, and the Saaty’s scale was used to score the results (Saaty, 1980). The normalized weights for each parameter were calculated using the matrix’s average row sum (Table 3). An overview of the relative significance of each parameter is given in Table 4. The distance from streams (44%), slope (18%), elevation (13%), LULC (13%), and soil drainage (5%) are the most important parameters in evaluating ideal sites for surface irrigation (Table 4). Soil depth (4%), soil type (3%), and soil texture (3%) are the next three factors (2%). Weighted overlay analysis (WOA) is carried out in an ArcGIS environment following the generation of weights for each raster layer using AHP. During suitability analysis, the weighted overlay is created by intersecting standardized and variably weighted layers. The weights quantify the relative weights of the taken into account appropriateness criteria. The suitability scores assigned for the sub-criteria inside each criteria layer were multiplied by the weights assigned for each criterion to create the final suitability map using the WOA approach.

\(S= { \sum_{i=1}^n Wi \ Xi}\)

where, S is the overall suitability score, Wi is the weight allocated to the layer of suitability criterion that was chosen, Xi is the sub-criteria score that was assigned to layer i of suitability criteria, and n is the total number of suitability criteria.

\(S=0.437×[Distance \ to \ stream]+0.177×[Slope]+0.128×[Elevation]\)

\(+0.128×[LULC]+0.0461×[Soil \ drainage]+0.0354×[Soil \ depth]\)

\(+0.0275×[Soil \ types]+0.021×[Soil \ texture]\)

A combined suitability map was produced by the weighted overlay analysis that was done utilizing the factors that were taken into consideration and their relative weights. The final map of suitability displayed the geographic distribution of the land types that were appropriate for surface irrigation (Figure 11). According to the surface irrigation suitability map, 16% (530km²) of the area is inappropriate, 49% (1648km²) is moderately acceptable, 34% (1128km²) is marginally suitable, and 16% (530 km²) is severely unsuitable. From the total study, area 1% (39km²) of the study area is covered by a not suitable area. According to the study, this is frequently caused by being far from the source of water, a steep gradient, excessive drainage, minimal soil depth, high elevation, settlements, forests, water bodies, leptosols, and poor drainage which is consistent with the work of (Aybehon and Tiku, 2022; Hussien et al., 2019). The 34% (1128km²) area was covered by a marginally suitable class (2).  Because of its uneven soil drainage, moderate elevation, location just on the edge of a water body, sandy loam soil type, relatively shallow soil depth, and bush land, this region is consistent with the work of (Aybehon and Tiku, 2022; Wubalem, 2021; Hussien et al., 2019). Approximately, 49% (1648 km²) of the overall research region is covered by a class that is just somewhat appropriate (3). Due to the existence of grassland, nitisols, somewhat deep soil, and soil mass that is moderately drained, which is consistent with the study of (Hussien et al., 2019; Negasa and Wakjira, 2021; Wubalem, 2021). Due to the existence of well-draining soil, nitisols type, low raised land, closest to water sources, clay soil texture, and already cultivated lands, the highly suitable class (4) is covered by roughly 16% (530 km²), which is consistent with the work of (Aybehon and Tiku, 2022; Negasa and Wakjira, 2021; Hussien et al., 2019).  The overall findings demonstrated that the feasibility of land for surface irrigation within the study region was significantly influenced by physical characteristics including slope, soil depth, soil drainage, soil texture, soil type and altitude.

Figure 11. Site suitability for surface irrigation

5 . CONCLUSION

Using the integration of GIS, remote sensing, and the AHP approach, a final map of the site suitability for surface irrigation was prepared in the this study by assessing eight features. In the normalized pairwise comparison matrix, distance from a water source (44 %), slope (18 %), elevation (13 %), LULC (13 %), and soil drainage (5 %) received high ratings. They are also the most important factors in determining whether a piece of land is eligible for surface irrigation, ahead of factors like soil depth (4 %), soil type (3 %), and soil texture (2 %). The final suitable map was constructed with four classifications: extremely suitable (4), fairly suitable (3), marginally suitable (2), and not suitable (1) after a load of each parameter was chosen. This map will aid in surface irrigation ability to support rain-fed agriculture. According to the results, extremely suitable, moderately suitable, marginally suitable and not suitable land uses account for 16%, 49%, 34%, and 1% area of region, respectively. This study demonstrates how GIS may be used at different scales and is an effective tool for emphasizing the suitability of land for agriculture and assessing cross-tabulations between classes of thematic maps. The map created using this platform might therefore be used as a rough guide when choosing appropriate sites for surface irrigation in the area.

This investigation showed the spatial distribution of suitable and unsuitable lands for surface irrigation, the relevant body may also be at the Zonal/Woreda level and should measure to apply surface irrigation is highly appropriate regions (4) and moderately suitable regions (3). The land suitability for surface irrigation is the sole focus of this study. It is recommended that future studies take into account the  water supply, the chemical composition of the soil, sociological conditions and economic variables for the region.

Tables

Figures

Conflict of Interest

Acknowledgements

Abbreviations

References