Multi-criteria decision support system for selecting the rainwater harvesting technique in wadi systems, Sinai, Egypt

Water scarcity in several countries around the world is a significant issue, especially in developing countries (Singh et al. 2009). Sustainable water solutions are the best way to address the issue of water scarcity. The potable water could be used for many purposes, such as agricultural and industrial activities, and could be substituted with harvested rainwater (Kaposztasova et al. 2014). Rainwater harvesting (RWH) is the action of collecting rainfall for direct usage or conservation through recharging into groundwater for future use (Naseef & Thomas 2016). It is an ancient technique still being used for rainwater gathering. Many countries in the Middle East, Southeast Asia, and Africa have developed some RWH techniques to store and use rainwater for agricultural purposes (Li et al. 2000).

RWH is a practical solution that is widely used in arid and semi-arid areas. Also, floodwater harvesting in the wadis reduces the potential risks to the existing infrastructure, especially damage due to floods (Al Zayed et al. 2013; Jamali et al. 2020). In Tanzania, the RWH techniques were used to provide the residents in some villages with a suitable amount of water for agriculture and livestock. The results of this study revealed that using the canal system as an RWH technique is profitable for agricultural purposes (Senkondo et al. 2004). In Iraq, the RWH techniques were evaluated using various evaluation methods, which are incorporated with ArcGIS tools to assign the appropriate RWH locations and techniques and the results revealed that the Jessor system and wadi bed cultivation are the appropriate RWH techniques in the study area (Ammar et al. 2016).

Selection of the appropriate RWH site and technique is a challenge because of the different variables, such as unreliable rainfall, which has a negative impact on the fit design (Ndomba & Wambura 2010; Zhang et al. 2018), rainwater storage tanks that can take up valuable space (Souza & Ghisi 2012; Matos et al. 2013), inability to install an efficient and effective system (Xu et al. 2018; Corvaro 2019), and the capital cost of the RWH systems might be high and requires some technical skills to install (Farreny et al. 2011; Chiu et al. 2015; Shiguang et al. 2022). The challenges of implementing appropriate RWH techniques require a multi-objectives performance assessment of RWH systems. However, the sustainable RWH could be achieved in four steps, identifying the suitable comparison criteria, evaluating the appropriateness of the identified criteria, selecting the RWH sites, and creating the suitability degree map for the selected RWH sites (Mkiramwinyi et al. 2007). Surface slope, runoff volume, drainage streams, residential settlements, infrastructures, and soil texture are the basic criteria to identify the potential sites for RWH (Isioye et al. 2012; Khan et al. 2022). Socioeconomic evaluation is an important criterion in the evaluation of the RWH sites and techniques (Al Zayed et al. 2013). The selection criteria should be combined in a multi-criteria framework to be analyzed, evaluated and weighted from all desired aspects to determine the suitability degree of the RWH site and technique (Sarkar & Biswas 2022).

The analytical hierarchy process (AHP) was used in this study to evaluate the RWH techniques. It is a method for making multi-criteria evaluations that are based on mathematics and professional experience. It provides a systematic framework for planning and assessing challenging decisions (Adamcsek 2008). This method examines the alternatives based on some comparison criteria that should be defined and has an important input to select the appropriate technique for RWH (Ammar et al. 2016). For evaluating multiple indexes, the AHP has a distinct advantage, and spatial analysis is a strength of the Geographic Information System (GIS). An efficient method for studies of regional eco-environmental evaluation is the combination of the AHP and GIS (Ying et al. 2007). It was used to identify the potential zone for RWH in integration with the geospatial techniques in drought-prone catchments in India (Sarkar & Biswas 2022).

The above-mentioned challenges are found in Egypt. Egypt faces many challenges because of population growth, economic development, global warming, and potential climate change (Abdel-Gawad 2008). Concerning water resources, the Nile River represents the main renewable freshwater resource with 95% of the total water resources (Negm et al. 2019). In regard to the arid and semi-arid areas, desert and mountainous regions that neither the Nile River nor its branches pass through, groundwater and RWH are the main water resources (Othman et al. 2012).

One of Egypt's arid areas is the Sinai Peninsula, which composes nearly 6% of the country's total land and has strategic importance (Badreldin & Goossens 2013). Due to the desert nature of this region, there are no water resources rather than groundwater and some RWH facilities. However, rainfall is rare in these areas (Bjerg et al. 2000). For the study purpose, the Wadi Watier basin was selected, which is located in the southeast of South Sinai governorate and suffers from water scarcity and flash floods. It is one of the most active wadis in Sinai and has various economical activities represented in agriculture, livestock, tourism, and some hand-making industries (Al Zayed et al. 2013). The sustainable water resources enhance Bedouins' settlements, which have positive social, economic, and environmental impacts on the region. Some residents inside the wadi have their RWH structures or use the retained water in front of the existing dams after the flood events. If the stored water is not used, it infiltrates and recharges the groundwater storage in the study area (EEAA 2003).

This study aimed to identify the appropriate RWH sites and techniques in wadi systems and define the suitability degree of each alternative to be considered in the decision-making actions in cooperation with the stakeholders and experts to be acceptable technically, socially, and economically.

• The rainfall data were obtained from five rainfall stations for the events that occurred between the years 2000 and 2016. The maximum annual precipitation is recorded by 51 mm in Saint Catherine station because of the high elevation while the average ranges between 5 and 7 mm.

• The topographic data (DEM) were obtained as a raster image, which was developed by NASA. The maximum elevation is 1,646 m in the Saint Catherine area while the lowest is 44 m at the outlet of the wadi in Nuweiba city.

• The geological information was obtained from WRRI. It was created in 1999 during the Japanese cooperation project in the study area to obtain the geological analysis map.

HEC-1 is the hydrological model used to simulate the runoff generated from rainfall events. It works under a WMS 7.1 program and simulates the whole basin as an interconnected and comprehensive system (US Army Corps of Engineers 1998). The model was applied to mitigate flash floods in Petra, Jordan (Al-Weshah & El-Khoury 1999). Also, it was applied for selecting the appropriate RWH sites and techniques in Lebanon as well as generating the rainfall–runoff volume at some RWH locations in a case study in Egypt (Makke 2002; Al Zayed et al. 2013).

SCS-Curve Number (CN) method was used in the model to estimate the initial and secondary losses due to infiltration. The CN value depends on the geological formation in each sub-catchment in the wadi which was previously classified by the joint venture of Japan International Cooperation Authority and WRRI in 1999 and a CN was given to each geological formation.

The equality of rainfall distribution should be considered in determining the average precipitation in the sub-catchment. Therefore, the Thiessen polygon method was used to define the affected area of each rainfall station in the wadi. This method is widely used to determine the weight of the rainfall gauges for a corresponding area (Chow et al. 1988). The most weighted station is El-Sheikh Attia, which is located in the middle with an effective weight of 64% of the wadi area while the El-themed station has the lowest influence with 3% of the total area of the wadi.

Flood harvesting is one of the RWH techniques defined as the collection and storage of the wadi bed flow for irrigation use (Prinz & Singh 2000). This technique is applied to large catchments. A large catchment comprises many square kilometres in size, from which runoff water flows through a major wadi. There are two forms of floodwater harvesting: floodwater harvesting within streambed and floodwater diversion (Prinz & Singh 2000).

• Small farm reservoirs: by constructing a dam with a reservoir, pond, or underground tank. The stored water could be used for irrigation. These reservoirs typically have volumes between 1,000 and 500,000 m³. However, the ponds and underground tanks are normally selected in case of a small amount of runoff water.

• Wadi bed cultivation: by constructing a small dam or dike crossing the wadi bed to reduce the flow velocity and allow settlement of the soil sediments. The structure is usually not higher than 1 m preferably made of permeable stone and may be reinforced with gabions.

• Jessor system: is a terrace wadi system with earth dikes supported by dry stone walls. The sediments accumulating behind the dikes are used for cropping. The three alternatives are tabulated and coded in Table 1.

Sub-criteria were selected according to the relationship to the primary criteria, field research, expert discussions, and previous studies. In the majority of the research reviewed, the most technical criteria widely used to identify appropriate RWH techniques were, as a percent of the reviewed studies; slope (83%), soil texture (75%), and rainfall volume (56%) (Adham Ali 2016). Also, the most socioeconomic factors used were cost (8%) and distance to communities (25%) as well as road accessibility (15%) (Adham Ali 2016).

Other sub-criteria were defined during the consultation and discussion with the experienced is the wadi width in addition to some socioeconomic and financial factors; social acceptability, annual O/M cost, and the cost revenue. The technical data was collected from the proposed RWH locations. The revenue is evaluated according to the expected economic activities based on water availability as the expected revenue ratio is 1.25 of the total cost.

A comparable scale across criteria must be identified due to the variety of measurements and scales for the different criteria. To illuminate their applicability, the chosen criteria were reclassified into the following five categories: 5 (very high suitability), 4 (high suitability), 3 (moderate suitability), 2 (low suitability), and 1 (very low suitability) (Ammar et al. 2016).

The main objective is at the top level, and the most crucial criteria that are related to it are at the lower levels. For each level, pairwise comparison matrices are built and scaled from 1 to 9 in terms of preference. Then, a consistency ratio (CR) is calculated to determine whether each matrix is consistent CR. The CR needs to be lower or equal to 10% (Ying et al. 2007).

The pairwise comparison focuses on the weightings of two factors that go into evaluating suitability for a particular objective. For both the primary and secondary evaluation criteria, a pairwise matrix is initially created. The odd values of 1, 3, 5, 7, and 9 correspond to equally, moderately, highly, very strongly, and extremely important criteria, respectively, when two criteria are assessed and rated on a continuous scale of 9. Even numbers 2, 4, 6, and 8 come in the intermediate range (Saaty 2008). Based on the data from the field survey and consultations with stakeholders and experts, the criteria were ranked.

The sum of all sub-criteria weights should be equal to 100%. Thereafter, it will be evaluated in regards to the RWH alternative to evaluate the alternative weight as a percentage of 100%.

The overall weight is a percentage of 100% (0% means no suitability, and 100% means very high suitability).

It is important to check the results with the stakeholders and experts, including the preliminary conclusions and recommendations.

After running the storm events that occurred during the period (2000–2016) on the hydrologic model, the probability analysis method was applied to the resulting runoff volume at the selected RWH sites. As a result, the rainfall–runoff volumes were calculated for different return periods (T). For achieving the sustainability objectives, the runoff water volumes for the return period of 2 years (T2) were considered in the evaluation of the RWH alternatives (Table 3).

The suitability degree of the RWH alternatives was determined and presented in Table 4. The alternatives A1, A2, and A3 are the most suitable alternatives for the flat wadis. Likewise, the soil texture suitability indicates that the clayey soil is the most suitable type for alternatives A2, A4, and A5. Also, the rainfall volume and wadi width have an important input in the evaluation. The second and third alternatives, A2 and A3, are the best and cheapest options for the narrow wadis and small amount of water while the other alternatives will not be feasible in such cases. The suitability analysis indicates that alternatives (A1, A4, A5) are the most appropriate alternatives in case of RWH sites remote from the Bedouins, settlements as the water is stored and spread over the wadi bed while the other alternatives are preferable in case of communities close to the proposed RWH sites. It can be noted that all alternatives have the same suitability degree regarding financial and other social evaluation aspects.

Table 4

Suitability degree of the technical criteria with the RWH alternatives

View Large

The pairwise comparison matrices were created for the main and sub-evaluation criteria. It can be noted that the technical criteria have the highest importance in the evaluation procedure by twice both social and financial criteria as shown in Table 5.

Regarding the technical evaluation criteria, the wadi bed slope is the most important criterion in the technical evaluation by five times the soil texture weight and three times the rainfall volume weight while it has the same importance as the wadi width because both of them are the most effected parameters in the identification of the structure height and the construction cost. Also, the rainfall volume importance is represented by four times the soil texture importance in the evaluation process as presented in Table 6.

For the social criteria, society acceptability is the most important criterion in the evaluation by twice the weight of the settlement remoteness and three times the accessibility weight. Also, society acceptability is in the second rank while settlement remoteness is the least important as the Bedouins use the RWH facilities mainly for agriculture purposes, which should be close to the agriculture, not the settlements (Table 7).

In regards to the financial and economic aspects, the initial cost has the highest importance in the selection of the RWH technique at seven times the O&M cost and three times the cost revenue. Also, the cost revenue has economic importance of four times the O&M, which reflects that O&M has little importance because of no mechanical or electrical equipment in the proposed RWH techniques except for some maintenance costs from time to time (Table 8).

• This research was carried out to develop a multi-criteria decision system to measure the suitability of the RWH alternatives.

• A hydrological model was built and verified to identify the volume of the runoff generated at each proposed RWH site.

• Five RWH alternatives were identified for evaluation and selection. The main and sub-criteria of comparison were identified and classified according to suitability with each RWH alternative. The AHP was used to identify the evaluation weights for the main criteria as well as the sub-criteria.

• The technical aspect has the maximum weight of 49% followed by the financial and economic aspects with 31 and 20% for the social aspect. The slope, wadi width, and initial cost have the highest weights with 20, 15, and 20%, respectively. They are followed by the water volume, social acceptability, and cost revenue with percentages of 11, 11, and 8%, respectively, while the rest have low evaluation weights.

• The suitability degree for each RWH alternative was determined at each proposed site and resulted in different suitability degrees of the alternatives and the conclusion is the storage ponds are the most suitable alternative with a suitability degree of 65.40% followed by the underground tank at 62.70%, dam with a reservoir 49.30% while the Jessor system and wadi bed cultivation have a low suitability degree with 42.70 and 41.70%, respectively.

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

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

The authors declare there is no conflict.