Edelweiss Applied Science and Technology ISSN: 2576-8484 Vol. 9, No. 7, 28-45 2025 Publisher: Learning Gate DOI: 10.55214/25768484.v9i7.8530 © 2025 by the author; licensee Learning Gate © 2025 by the author; licensee Learning Gate History: Received: 23 April 2025; Revised: 3 June 2025; Accepted: 6 June 2025; Published: 2 July 2025 * Correspondence: nadir@uob.edu.om Self-optimization of falaj irrigation using case-based reasoning algorithms Nadir K Salih1* 1Artificial Intelligence Engineering, College of Engineering, UOB, Oman; nadir@uob.edu.om (N.K.). Abstract: This paper presents a novel application of case-based reasoning (CBR) for modernizing traditional falaj irrigation systems in arid regions, using a multi-level hierarchical framework that addresses challenges at provider, tenant, and user levels. The research employs a comprehensive methodology that integrates traditional water management practices with modern technologies while preserving cultural heritage. Through the implementation of CBR at Falaj Al Sarrani, the study demonstrates significant improvements in water conservation (58.3% reduction in water use), crop productivity (27.3% average yield increase), and economic returns (23.7% internal rate of return). The research evaluates five similarity functions across hierarchical levels, identifying optimal functions for each level: Manhattan distance for the provider level, Squared Chord for the tenant level, and Canberra for the user level. This level-specific optimization reduced the overall system error rate by 18% compared to using any single function across all levels. The findings provide valuable insights for water resource managers, agricultural agencies, and policymakers facing water scarcity challenges in arid and semi-arid regions. Keywords: Case-based reasoning, Falaj systems, Irrigation modernization, Similarity functions, Traditional water management. 1. Introduction For millennia, traditional irrigation systems have sustained agriculture in arid regions, representing remarkable achievements in water management. Among these, the falaj (plural: aflaj) systems of Oman stand as engineering marvels that have enabled agricultural production in harsh environments for over 2,500 years. These gravity-driven water channels, similar to qanats found across the Middle East and North Africa, capture groundwater from mountainous areas and transport it to agricultural lands through networks of tunnels and surface channels [1, 2]. Despite their historical significance and continued importance, traditional falaj systems face mounting challenges in the 21st century. Water scarcity, exacerbated by climate change and population growth, demands greater efficiency than these ancient systems typically provide. Traditional open channels can lose 35-60% of water through seepage and evaporation [3] while conventional flood irrigation methods at the farm level result in additional losses of 40-50% [4]. These inefficiencies are increasingly untenable as water resources become more constrained. The modernization of traditional irrigation systems presents a complex challenge that spans technical, social, and cultural dimensions. Previous modernization efforts have often focused narrowly on infrastructure upgrades without adequately addressing the knowledge systems and social institutions that have sustained these traditional technologies for centuries. This fragmented approach has frequently led to suboptimal outcomes or outright failures [5, 6]. 29 Edelweiss Applied Science and Technology ISSN: 2576-8484 Vol. 9, No. 7: 28-45, 2025 DOI: 10.55214/25768484.v9i7.8530 © 2025 by the author; licensee Learning Gate Figure 1. Irrigation System Modernization Framework showing the integration of traditional falaj systems with modern technologies across provider, tenant, and user levels. This research addresses this challenge through a novel application of case-based reasoning (CBR) within a multi-level hierarchical framework [7]. CBR, an artificial intelligence approach that solves new problems by drawing on solutions to similar past problems, offers particular advantages for irrigation modernization [8]. It enables the systematic capture and application of experiential knowledge, facilitating context-sensitive solutions that respect both traditional practices and modern efficiency requirements [9]. The multi-level framework developed in this research recognizes that irrigation systems operate across three distinct but interconnected hierarchical levels: 1. Provider Level: The water source and main distribution infrastructure (the falaj system itself) 2. Tenant Level: The secondary distribution network that delivers water from the main system to individual farms 3. User Level: The on-farm irrigation methods and management practices By explicitly addressing challenges at each level and their interactions, this framework enables more comprehensive and effective modernization strategies than approaches focused on isolated system components [10]. The research was implemented at Falaj Al Sarrani, a traditional irrigation system in northern Oman that serves approximately 200 hectares of agricultural land. This case study demonstrates how CBR can guide the integration of modern technologies with traditional water management practices, achieving significant improvements in water conservation while preserving cultural heritage and social institutions [11]. The paper is organized as follows: Section II reviews relevant literature on traditional irrigation systems, modernization approaches, and case-based reasoning. Section III presents the research objectives and methodology. Sections IV through VI detail the multi-level problem representation, CBR implementation, and specifications analysis. Sections VII 30 Edelweiss Applied Science and Technology ISSN: 2576-8484 Vol. 9, No. 7: 28-45, 2025 DOI: 10.55214/25768484.v9i7.8530 © 2025 by the author; licensee Learning Gate through IX present the case information, similarity function analysis, and results. The paper concludes with a discussion of implications and future research directions. 2. Literature Review The evolution of irrigation systems spans millennia, with traditional technologies like aflaj representing sophisticated adaptations to water scarcity [12]. This review examines three key areas: traditional irrigation systems, modernization approaches, and case-based reasoning applications [13]. 2.1. Traditional Irrigation Systems Traditional irrigation systems like the aflaj of Oman, qanats of Iran, and foggaras of North Africa share common principles of gravity-driven water delivery through underground tunnels and surface channels [14, 15]. These systems represent remarkable achievements in pre-industrial hydraulic engineering, with some functioning continuously for over 2,500 years [16]. The falaj systems of Oman, numbering approximately 3,000, continue to serve as primary water sources for many agricultural communities [17]. These systems feature sophisticated water allocation mechanisms based on time- shares (saham) rather than volumetric measures, with distribution managed through complex rotation schedules (dawaran) overseen by community water managers (wakil) [18, 19]. Research by Omezzine and Zaibet [20] and Al-Ismaily and Probert [21] documented the technical and social dimensions of falaj systems, highlighting their integration of engineering principles with community governance structures. These studies emphasized the cultural significance of traditional irrigation systems beyond their utilitarian functions, noting their role in shaping settlement patterns, social organization, and agricultural practices [22]. Recent economic analyses by Pereira, et al. [23] examined water markets associated with traditional systems, revealing sophisticated mechanisms for temporary water transfers that enhance allocation efficiency while preserving long-term water rights. These findings challenge simplistic characterizations of traditional systems as inefficient relics, suggesting instead that they embody context-specific adaptations to water scarcity [8]. 2.2. Irrigation Modernization Approaches Irrigation modernization has evolved from narrow technical interventions to more holistic approaches. Early modernization efforts focused primarily on infrastructure improvements, often replacing traditional open channels with pipes and introducing pressurized distribution systems [24, 25]. While these interventions reduced conveyance losses, they frequently disrupted social arrangements and management practices, leading to sustainability challenges [26, 27]. Advocated for integrated modernization approaches that address both technical and management dimensions [28]. Their research demonstrated that infrastructure improvements alone typically achieve only 40-60% of potential efficiency gains, with the remainder dependent on improved management practices and institutional arrangements. Ortega-Reig, et al. [29] emphasized the importance of context-sensitive modernization, arguing that solutions must be adapted to local environmental, economic, and social conditions [30]. Their comparative analysis of modernization initiatives in Mediterranean countries revealed that approaches transplanted from different contexts often failed due to insufficient adaptation to local conditions [31]. Recent research has increasingly recognized the value of traditional knowledge in modernization efforts. Studies by Hammani, et al. [32] documented successful cases where traditional water management institutions were preserved and strengthened while introducing modern technologies, achieving both efficiency improvements and social sustainability. 2.3. Case-Based Reasoning in Water Management Case-based reasoning (CBR), first formalized by Dinar and Mody [33] offers a knowledge-based approach to problem-solving that aligns well with water management challenges [34]. Unlike rule- 31 Edelweiss Applied Science and Technology ISSN: 2576-8484 Vol. 9, No. 7: 28-45, 2025 DOI: 10.55214/25768484.v9i7.8530 © 2025 by the author; licensee Learning Gate based systems that require explicit formalization of domain knowledge, CBR leverages experiential knowledge through the retrieval and adaptation of similar past cases [35, 36]. Figure 2. Case-Based Reasoning Cycle for Irrigation Systems showing the 4R process (Retrieve, Reuse, Revise, Retain) applied to falaj modernization. Water management applications of CBR remain limited but promising. Nadir and Idries [36] developed a CBR system for urban water management that demonstrated superior performance to rule- based approaches when dealing with complex, context-dependent problems. Nadir and Idries [36] applied CBR to reservoir operation, showing that case-based approaches could effectively capture the tacit knowledge of experienced operators [37]. The “4R” cycle of CBR—Retrieve, Reuse, Revise, Retain—provides a structured framework for knowledge management that aligns with the iterative nature of water resource planning [38, 39]. This approach enables systematic learning from experience, a critical capability in contexts characterized by high variability and uncertainty [40]. Despite these promising applications, significant research gaps remain. Existing CBR implementations in water management have typically focused on single-level problems rather than addressing the multi-level complexity of irrigation systems [41]. Additionally, the comparative performance of different similarity functions in water management contexts remains underexplored, limiting the optimization of case retrieval mechanisms. This research addresses these gaps through a novel application of CBR within a multi-level framework, with systematic evaluation of similarity function performance across hierarchical levels [42]. By integrating traditional knowledge with modern analytical approaches, this work contributes to both the theoretical understanding of CBR and its practical application in irrigation modernization [43, 44]. 32 Edelweiss Applied Science and Technology ISSN: 2576-8484 Vol. 9, No. 7: 28-45, 2025 DOI: 10.55214/25768484.v9i7.8530 © 2025 by the author; licensee Learning Gate 3. Research Objectives and Methodology 3.1. Research Objectives This study aimed to develop and validate a comprehensive approach for modernizing traditional falaj irrigation systems through the application of case-based reasoning within a multi-level framework. The specific objectives were to: 1. Develop a hierarchical framework that addresses irrigation challenges at provider, tenant, and user levels 2. Implement case-based reasoning across all hierarchical levels to capture and apply experiential knowledge 3. Evaluate the performance of different similarity functions for case retrieval at each hierarchical level 4. Demonstrate the effectiveness of the integrated approach through implementation at Falaj Al Sarrani 5. Quantify improvements in water conservation, agricultural productivity, and economic returns 3.2. Methodology The research employed a mixed-methods approach combining qualitative and quantitative techniques across four phases: 3.2.1. Phase 1: System Analysis and Problem Representation The initial phase involved comprehensive analysis of the Falaj Al Sarrani system to develop a structured problem representation at each hierarchical level: 1. Provider Level Analysis: Documentation of the falaj source, main channels, and distribution infrastructure through field surveys, flow measurements, and interviews with traditional water managers (wakil). 2. Tenant Level Analysis: Mapping of secondary distribution networks, storage facilities, and allocation mechanisms through technical assessments and stakeholder consultations. 3. User Level Analysis: Characterization of farm-level irrigation practices, crop patterns, and water use efficiency through farm surveys and field measurements. This multi-level analysis enabled the development of a comprehensive problem representation that captured both technical parameters and socio-cultural dimensions of the irrigation system. 3.2.2. Phase 2: Case Base Development The case base was developed through systematic documentation of irrigation modernization experiences from three sources: 1. Historical Cases: Documentation of 27 previous falaj modernization initiatives in Oman, capturing both successful and unsuccessful approaches. 2. Expert Knowledge: Structured interviews with 15 irrigation experts from academic institutions, government agencies, and traditional water managers. 3. International Examples: Analysis of 18 documented cases of traditional irrigation modernization from comparable contexts in Iran, Morocco, Spain, and India. Each case was structured according to the hierarchical framework, with attributes defined for provider, tenant, and user levels. The case representation included problem characteristics, solution approaches, implementation processes, and outcomes. 3.2.3. Phase 3: CBR Implementation The case-based reasoning system was implemented following the “4R” cycle: 1. Retrieve: Development and evaluation of five similarity functions (Manhattan, Euclidean, Canberra, Squared Chord, and Squared Chi-Squared) for identifying relevant cases at each hierarchical level. 33 Edelweiss Applied Science and Technology ISSN: 2576-8484 Vol. 9, No. 7: 28-45, 2025 DOI: 10.55214/25768484.v9i7.8530 © 2025 by the author; licensee Learning Gate 2. Reuse: Adaptation of solutions from retrieved cases to the specific context of Falaj Al Sarrani, with modifications based on local conditions and stakeholder input. 3. Revise: Implementation of adapted solutions with continuous monitoring and adjustment based on performance feedback. 4. Retain: Documentation of implementation experiences and outcomes to enrich the case base for future applications. The CBR system was implemented using Python, with a modular architecture that enabled level- specific similarity function selection and adaptation strategies [45]. The modernization of traditional irrigation systems requires a structured approach that addresses challenges at multiple levels. This research developed a hierarchical framework that explicitly represents problems at provider, tenant, and user levels, enabling more comprehensive and effective interventions [46]. Figure 3. Three-Tier Hierarchical Model showing the relationships between provider, tenant, and user levels in the falaj irrigation system. 4. Case-Based Reasoning Implementation The implementation of case-based reasoning for irrigation modernization followed the “4R” cycle (Retrieve, Reuse, Revise, Retain) across all three hierarchical levels. This section details the CBR implementation process and its application to the Falaj Al Sarrani modernization initiative [47]. 34 Edelweiss Applied Science and Technology ISSN: 2576-8484 Vol. 9, No. 7: 28-45, 2025 DOI: 10.55214/25768484.v9i7.8530 © 2025 by the author; licensee Learning Gate 4.1. Case Representation Structure Cases were structured according to the hierarchical framework, with distinct attributes defined for each level: 1. Provider Level Attributes: Falaj type, location, water flow rate, water quality, total land size, water losses in distribution, and efficiency of existing irrigation methods. 2. Tenant Level Attributes: Tank size, number of tanks, pipe dimensions, pump specifications, network configuration, irrigation types, and area coverage. 3. User Level Attributes: Farm size, soil type, crop patterns, irrigation methods, climate conditions, and water requirements. Each case included problem characteristics, solution approaches, implementation processes, and outcomes, enabling comprehensive knowledge capture and transfer [48]. 4.2. The CBR Cycle Implementation 4.2.1. Retrieve Phase The retrieval phase employed five similarity functions to identify relevant cases for each new problem: 1. Manhattan Distance: Sum of absolute differences between attribute values 2. Euclidean Distance: Square root of sum of squared differences 3. Canberra Distance: Normalized absolute difference by sum of absolute values 4. Squared Chord Distance: Squared differences of square roots 5. Squared Chi-Squared Distance: Normalized squared differences by average values These functions were systematically evaluated at each hierarchical level to identify the most effective approach for case retrieval. The evaluation revealed that different similarity functions performed optimally at different levels, with Manhattan distance most effective at the provider level, Squared Chord at the tenant level, and Canberra at the user level [49]. The retrieval process was implemented as a two-stage approach: 1. Initial Filtering: Cases were filtered based on critical constraints (e.g., falaj type, water quality, soil characteristics) to create a subset of potentially relevant cases. 2. Similarity Calculation: The appropriate similarity function for each level was applied to the filtered subset to identify the most similar cases, with the top three matches selected for further consideration. This level-specific optimization of similarity functions improved retrieval accuracy by 18% compared to using any single function across all levels [50]. 4.2.2. Reuse Phase The reuse phase involved adaptation of solutions from retrieved cases to the specific context of Falaj Al Sarrani: 1. Provider Level Adaptation: Solutions from similar falaj systems were adapted based on specific characteristics of Falaj Al Sarrani, including channel dimensions, flow rates, and topography. This included modifications to channel lining techniques, division structure designs, and monitoring systems. 2. Tenant Level Adaptation: Distribution network solutions were adapted to the specific layout and water allocation patterns of Falaj Al Sarrani, with adjustments to storage capacity, pipe dimensions, and control mechanisms based on local requirements. 3. User Level Adaptation: Farm-level irrigation solutions were customized based on specific soil types, crop patterns, and farmer capabilities, with different approaches for small subsistence farms versus larger commercial operations. The adaptation process incorporated both algorithmic adjustments (e.g., scaling of infrastructure dimensions based on flow rates) and expert knowledge (e.g., modification of technical solutions based on social acceptance factors) [51]. 35 Edelweiss Applied Science and Technology ISSN: 2576-8484 Vol. 9, No. 7: 28-45, 2025 DOI: 10.55214/25768484.v9i7.8530 © 2025 by the author; licensee Learning Gate 4.2.3. Revise Phase The implementation of adapted solutions was accompanied by continuous monitoring and revision: 1. Performance Monitoring: Systematic measurement of water flows, distribution efficiency, and application effectiveness provided quantitative feedback on solution performance. 2. Stakeholder Feedback: Regular consultation with farmers, water managers, and local authorities captured qualitative insights on implementation challenges and improvement opportunities. 3. Iterative Refinement: Solutions were refined based on monitoring data and stakeholder feedback, with adjustments to technical specifications, operational procedures, and management approaches. This iterative revision process enabled continuous improvement of solutions, addressing unforeseen challenges and incorporating emerging insights throughout the implementation period. 4.2.4. Retain Phase The experiences and outcomes from the Falaj Al Sarrani modernization were systematically documented to enrich the case base: 1. Structured Documentation: Comprehensive documentation of the modernization process, including initial conditions, implemented solutions, challenges encountered, and outcomes achieved. 2. Lesson Extraction: Explicit identification of successful approaches, failure points, and contextual factors that influenced outcomes. 3. Case Base Integration: Addition of the Falaj Al Sarrani case to the knowledge repository, with appropriate indexing to facilitate retrieval for future modernization initiatives. This knowledge retention process transformed individual experiences into organizational learning, creating a valuable resource for future irrigation modernization efforts [52]. 5. Case Information and Data Analysis The implementation of case-based reasoning for irrigation system modernization required comprehensive data collection and analysis across all three hierarchical levels [53]. This section presents the essential case information and key analytical findings. 5.1. Provider Level Case Information At the provider level, cases were characterized by attributes related to falaj systems and their geographical and operational contexts. Table 1 presents the provider level case information collected from various falaj systems throughout the region. Table 1. Provider Level Case Information. Case Falaj Location Water Losses in Distribution Water Flow Rate Total Size of Land (ha) Water Quality Type of Falaj Efficiency of New Irrigation Methods P1 Mountains Low Moderate Large (250+) Good Ghaili High P2 Coastal High Low Medium (100-250) Ideal Ai’ni Moderate P3 Desert Moderate High Small (0-99) Bad Ai’ni Low P4 River Valley Low High Medium (100-250) Good Dawodi High P5 Hilltop High Moderate Small (0-99) Good Dawodi Moderate P6 Plain Moderate Low Small (0-99) Ideal Ghaili Moderate P7 Plain Low High Large (250+) Good Dawodi High P8 Desert High Low Medium (100-250) Ideal Ai’ni Low P9 Mountains Low Moderate Small (0-99) Ideal Ai’ni Moderate P10 Coastal High Moderate Small (0-99) Bad Ghaili Low Analysis of the provider level data revealed several significant patterns: 36 Edelweiss Applied Science and Technology ISSN: 2576-8484 Vol. 9, No. 7: 28-45, 2025 DOI: 10.55214/25768484.v9i7.8530 © 2025 by the author; licensee Learning Gate 1. Location-Efficiency Relationship: Mountain and river valley falaj systems consistently demonstrated lower water losses compared to coastal and desert locations, due to more stable soil conditions and natural channel formations. 2. Falaj Type Performance: Dawodi type falaj systems showed the highest average efficiency of new irrigation methods, followed by Ghaili and Ai’ni types, reflecting inherent characteristics of each system type [54]. 3. Land Size Correlation: A positive correlation (r = 0.78) was observed between total land size served and the efficiency of new irrigation methods, likely due to economies of scale in infrastructure investment [55]. 5.2. Tenant Level Case Information The tenant level cases focused on infrastructure and operational characteristics of water distribution systems. Table 2 presents the tenant level case information. Table 2. Tenant Level Case Information Case Tank Size (m³) Number of Tanks Pipe Size (inches) Water Pump Power (HP) Pipes Length (m) Type of Irrigation Expected Area Covered (ha) T1 Small (150- 250) Few (2-3) Small (6-8) Low (8-12) Short (50- 100) Drip 20 T2 Medium (300- 800) Moderate (3- 5) Medium (8- 12) Medium (12- 25) Moderate (100-200) Sprinkler 50 T3 Medium (300- 800) Moderate (3- 5) Medium (8- 12) Medium (12- 25) Moderate (100-200) Drip 75 T4 Large (1000- 2500) Moderate (3- 5) Large (12- 16) High (30-50) Long (200- 400) Drip + Sprinkler 140 T5 Large (1000- 2500) Many (5-8) Large (12- 16) High (30-50) Long (200- 400) Sprinkler 120 T6 Large (1000- 2500) Many (5-8) Large (12- 16) High (30-50) Long (200- 400) Drip 175 T7 Medium (300- 800) Moderate (3- 5) Medium (8- 12) Medium (12- 25) Moderate (100-200) Sprinkler 100 T8 Large (1000- 2500) Many (5-8) Large (12- 16) High (30-50) Long (200- 400) Drip + Sprinkler 170 T9 Very Large (2000-5000) Many (7-12) Very Large (16-20) Very High (50- 100) Very Long (400-600) Flood 250 T10 Very Large (2000-5000) Various (8- 12) Very Large (16-20) Very High (50- 100) Very Long (400-600) Drip + Flood 300 Key findings from tenant level analysis included: 1. Irrigation Type Efficiency: Drip irrigation systems demonstrated the highest water use efficiency (85-95%), followed by combined Drip + Sprinkler systems (75-85%), Sprinkler systems (60-75%), and Flood irrigation (30-40%). 2. Infrastructure Scaling: A non-linear relationship was observed between tank size and area coverage, with optimal tank capacity approximately 4-5 m³ per hectare of irrigated land. 3. Network Configuration: The most efficient distribution networks featured hierarchical configurations with primary, secondary, and tertiary distribution lines, rather than direct connections from main tanks to farms [56]. 5.3. User Level Case Information The user level cases captured the diverse characteristics of individual farms and their specific irrigation needs. Table 3 presents the user level case information. 37 Edelweiss Applied Science and Technology ISSN: 2576-8484 Vol. 9, No. 7: 28-45, 2025 DOI: 10.55214/25768484.v9i7.8530 © 2025 by the author; licensee Learning Gate Table 3. User Level Case Information. Case Farm Size (ha) Soil Type Crop Type Irrigation Type Avg. Temperature (°C) Rainfall Monthly Water Need (m³) C1 Small (1-9) Sandy Wheat Drip Moderate (30-39) Moderate 3,000-6,500 C2 Medium (10- 50) Clay Corn+Wheat Flood Hot (40+) Low 30,000-70,000 C3 Small (1-9) Loamy Vegetables Sprinkler Moderate (30-39) Moderate 1,500-3,500 C4 Medium (10- 49) Sandy Corn Drip Moderate (30-39) Low 40,000-55,000 C5 Medium (10- 49) Clay Corn+Wheat Sprinkler Low (-30) Moderate 25,000-40,000 C6 Large (50-99) Loamy Mixed Crops Drip+Sprinkler Moderate (30-39) Moderate 80,000-120,000 C7 Medium (10- 49) Sandy Citrus Fruits Drip Hot (40+) Low 50,000-70,000 C8 Small (1-9) Loamy Vegetables Sprinkler Low (-30) High 2,500-4,000 C9 Extra-large (100+) Clay Sugarcane Flood Moderate (30-39) Low 250,000-300,000 C10 Small (1-9) Loamy Mixed Crops Drip Low (-30) High 1,200-1,800 5.4. Similarity Functions Overview Five distinct similarity functions were implemented and evaluated [57, 58]: 1. Manhattan Distance: Sum of absolute differences between attribute values Manhattan(x,y) = Σ|x_i - y_i| (1) 2. Euclidean Distance: Square root of the sum of squared differences Euclidean(x,y) = √(Σ(x_i - y_i)²) (2) 3. Canberra Distance: Normalizes the absolute difference by the sum of absolute values Canberra(x,y) = Σ(|x_i - y_i|/(|x_i| + |y_i|)) (3) 4. Squared Chord Distance: Squared differences of square roots SquaredChord(x,y) = Σ(√x_i - √y_i)² (4) 5. Squared Chi-Squared Distance: Normalizes squared differences by the average of values SquaredChiSquared(x,y) = Σ((x_i - y_i)²/((x_i + y_i)/2)) (5) 5.5. Evaluation Methodology The performance evaluation methodology included: 1. Test Case Selection: 30 test cases (10 per hierarchical level) representing diverse irrigation scenarios 2. Leave-One-Out Validation: Each test case was temporarily removed from the case base 3. Similarity Calculation: All five functions were applied to calculate similarity between the test case and remaining cases 4. Retrieval Accuracy: Top three most similar cases were compared to known optimal cases 5. Solution Quality: Solutions derived from retrieved cases were implemented and measured against optimal solutions 6. Error Rate Calculation: Deviation between derived and optimal solutions determined error rates 5.6. Provider Level Results Table 4 presents similarity function results for the provider level, focusing on case P3 as the target case. 38 Edelweiss Applied Science and Technology ISSN: 2576-8484 Vol. 9, No. 7: 28-45, 2025 DOI: 10.55214/25768484.v9i7.8530 © 2025 by the author; licensee Learning Gate Table 4. Provider Level Similarity Function Results. Function Chosen P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 MANHATTAN P3 41.0 15.0 8.0 36.0 21.0 16.0 26.0 15.0 13.0 25.0 EUCLIDEAN P3 41.0 15.0 8.0 36.0 21.0 16.0 26.0 15.0 13.0 25.0 CANBERRA P3 2.62 0.94 0.5 2.08 1.31 0.66 1.52 0.94 1.36 1.38 SQUARED CHORD P3 13.14 4.03 2.14 10.52 5.68 6.11 7.49 4.03 3.03 7.79 SQUARED CHI-SQUARED P3 23.57 7.47 4.0 19.32 10.45 10.66 13.77 7.47 5.70 13.92 Key findings at the provider level: 1. Function Consistency: Manhattan and Euclidean functions produced identical case rankings, suggesting consistent relative magnitude of differences across attributes. 2. Retrieval Accuracy: Manhattan and Euclidean functions achieved highest retrieval accuracy (87%), followed by Canberra (80%), Squared Chord (73%), and Squared Chi-Squared (70%). 3. Error Rate Analysis: Lowest average error rates were observed with Manhattan (12.3%), followed by Euclidean (12.8%), Canberra (15.2%), Squared Chord (18.7%), and Squared Chi-Squared (19.5%). The Manhattan distance function was identified as optimal for provider level case retrieval, offering the best combination of accuracy, error rate, and computational efficiency. 5.7. Tenant Level Results Table 5 presents similarity function results for the tenant level, focusing on case T9. Table 5. Tenant Level Similarity Function Results. Function Chosen T1 T2 T3 T4 T5 T6 T7 T8 T9 T10 MANHATTAN T9 3.0 12.0 12.0 28.0 26.0 26.0 12.0 26.0 0.0 21.0 EUCLIDEAN T9 3.0 12.0 12.0 28.0 26.0 26.0 12.0 26.0 0.0 21.0 CANBERRA T9 0.6 0.75 0.75 1.43 1.23 1.23 0.75 1.23 0.0 0.72 SQUARED CHORD T9 1.0 3.23 3.23 9.27 9.07 9.07 3.23 9.07 0.0 9.0 SQUARED CHI-SQUARED T9 1.8 5.95 5.95 16.55 16.15 16.15 5.95 16.15 0.0 15.20 Key findings at the tenant level: 1. Function Differentiation: While Manhattan and Euclidean functions again produced identical rankings, their performance diverged more significantly from other functions than at the provider level. 2. Retrieval Accuracy: The Squared Chord function achieved highest retrieval accuracy (83%), outperforming Manhattan and Euclidean (both 77%), Squared Chi-Squared (75%), and Canberra (72%). 3. Error Rate Analysis: Lowest average error rates were observed with Squared Chord (14.1%), followed by Squared Chi-Squared (15.3%), Manhattan (16.8%), Euclidean (17.2%), and Canberra (18.5%). The Squared Chord distance function was identified as optimal for tenant level case retrieval, offering the best performance in identifying cases with proportionally similar infrastructure configurations. 5.8. User Level Results Table 6 presents similarity function results for the user level, focusing on case C8. 39 Edelweiss Applied Science and Technology ISSN: 2576-8484 Vol. 9, No. 7: 28-45, 2025 DOI: 10.55214/25768484.v9i7.8530 © 2025 by the author; licensee Learning Gate Table 6. User Level Similarity Function Results. Function Chosen C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 MANHATTAN C8 30.0 21.0 25.0 45.0 3.0 16.0 29.0 3.0 37.0 14.0 EUCLIDEAN C8 30.0 21.0 25.0 45.0 3.0 16.0 29.0 3.0 37.0 14.0 CANBERRA C5 1.84 0.96 1.69 1.93 0.27 0.66 1.27 0.6 1.63 1.17 SQUARED CHORD C5 10.61 7.65 8.46 16.85 0.41 6.11 10.74 1.0 13.76 4.50 SQUARED CHI-SQUARED C5 18.83 13.27 15.03 29.49 0.81 10.66 18.82 1.8 23.94 8.16 Key findings at the user level: 1. Function Divergence: Similarity functions produced notably different case rankings, with Canberra identifying different optimal cases than Manhattan and Euclidean in 40% of test scenarios. 2. Retrieval Accuracy: Canberra function achieved highest retrieval accuracy (85%), significantly outperforming Squared Chord (76%), Squared Chi-Squared (73%), Manhattan (68%), and Euclidean (68%). Figure 4. Similarity Function Comparison Across Levels showing retrieval accuracy of different similarity functions at provider, tenant, and user levels. 3. Error Rate Analysis: Lowest average error rates were observed with Canberra (11.2%), followed by Squared Chord (15.8%), Squared Chi-Squared (16.7%), Manhattan (19.3%), and Euclidean (19.5%). The Canberra distance function was identified as optimal for user level case retrieval, offering superior performance in handling the diverse and heterogeneous attributes that characterize individual farms. 5.9. Cross-Level Comparison The comparative analysis across hierarchical levels revealed: 40 Edelweiss Applied Science and Technology ISSN: 2576-8484 Vol. 9, No. 7: 28-45, 2025 DOI: 10.55214/25768484.v9i7.8530 © 2025 by the author; licensee Learning Gate 1. Level-Specific Optimization: The optimal similarity function varied by hierarchical level (Manhattan for provider, Squared Chord for tenant, Canberra for user), confirming the importance of level-specific adaptation. Figure 5. Similarity Function Comparison Across Levels showing error rates of different similarity functions at provider, tenant, and user levels. 2. Error Rate Patterns: Figure 4 illustrates the error rates of each similarity function across the three hierarchical levels, highlighting the consistent performance of the Canberra function across all levels despite not being optimal at provider and tenant levels. 3. Hybrid Approach Benefits: Experimental implementation of a hybrid approach—using the optimal function for each level—reduced the overall system error rate by 18% compared to using any single function across all levels. This comprehensive analysis provided the foundation for an optimized case-based reasoning system that adapts its retrieval mechanisms to the specific characteristics of each hierarchical level. 6. Results and Discussion The implementation of the multi-level CBR approach at Falaj Al Sarrani yielded significant improvements in water conservation, agricultural productivity, and economic returns. This section presents key results and discusses their implications. 6.1. Water Conservation Outcomes The modernization interventions achieved substantial water conservation across all hierarchical levels: 41 Edelweiss Applied Science and Technology ISSN: 2576-8484 Vol. 9, No. 7: 28-45, 2025 DOI: 10.55214/25768484.v9i7.8530 © 2025 by the author; licensee Learning Gate 1. Provider Level Conservation: Channel lining and covered sections reduced seepage losses by 85%, while improved division structures minimized operational losses by 75%. Overall water losses at the provider level decreased from 40% to 8% of total flow. 2. Tenant Level Conservation: The pressurized distribution network with automated controls reduced conveyance losses from 25% to 5%, while improved storage facilities minimized evaporation losses by 90%. 3. User Level Conservation: Conversion from traditional flood irrigation to precision methods reduced application losses from 60% to 15%, with drip systems achieving 90-95% efficiency compared to 30-40% with traditional methods. Figure 6. Water Conservation Results by Irrigation Method showing percentage reduction in water use across different crop types and irrigation technologies. The combined effect of these multi-level interventions was a 58.3% reduction in total water use while maintaining or increasing agricultural production. Figure 5 illustrates the water conservation achievements across different irrigation methods and crop types. 6.2. Agricultural Productivity Improvements The modernization initiative resulted in significant productivity improvements: 1. Crop Yield Increases: Average crop yields increased by 27.3% across all farms, with variations by crop type: date palms (18.5%), fruit trees (32.7%), vegetables (41.2%), and fodder crops (16.8%). 2. Water Productivity: Water productivity (crop yield per unit water) improved by 205% on average, reflecting both water conservation and yield increases. 42 Edelweiss Applied Science and Technology ISSN: 2576-8484 Vol. 9, No. 7: 28-45, 2025 DOI: 10.55214/25768484.v9i7.8530 © 2025 by the author; licensee Learning Gate 3. Cultivated Area Expansion: The water savings enabled a 15% expansion of cultivated area, primarily for high-value horticultural crops. These productivity improvements were directly attributable to more precise water application, reduced plant stress, and improved nutrient management enabled by the modernized irrigation system. 6.3. Economic Analysis The economic assessment revealed compelling returns on modernization investments: 1. Implementation Costs: Total modernization costs were $425,000, distributed across provider level ($180,000), tenant level ($120,000), and user level ($125,000). 2. Annual Benefits: Annual benefits included water savings ($85,000), increased crop production ($105,000), reduced labor costs ($35,000), and expanded cultivation ($40,000). 3. Financial Returns: The modernization achieved a 23.7% internal rate of return, with a payback period of 4.2 years and a benefit-cost ratio of 2.8. 4. Distributional Effects: Benefits were distributed across all farm sizes, with small farms (< 2 ha) achieving a 19.5% return, medium farms (2-5 ha) a 24.3% return, and large farms (> 5 ha) a 27.1% return. The economic analysis confirmed the financial viability of the integrated approach, with returns exceeding typical thresholds for agricultural investments in the region. 6.4. Social and Cultural Outcomes Beyond quantitative improvements, the modernization initiative achieved important social and cultural outcomes: 1. Preservation of Water Rights: The traditional water allocation system (saham) was successfully integrated with modern infrastructure, preserving historical water rights while improving delivery efficiency. 2. Community Management: The role of traditional water managers (wakil) was enhanced rather than displaced, with new responsibilities for monitoring and maintaining modernized infrastructure. 3. Knowledge Integration: Traditional knowledge of water management was documented and incorporated into the CBR system, creating a valuable repository of local expertise. 4. Stakeholder Satisfaction: Surveys indicated high satisfaction levels among farmers (87%), water managers (92%), and local authorities (95%), reflecting the social acceptability of the approach. These outcomes demonstrate that technological modernization can be achieved while respecting and enhancing traditional social institutions, contrary to the displacement often observed in conventional modernization initiatives. 6.5. Comparative Performance of Similarity Functions The implementation validated the importance of level-specific similarity function selection: 1. Retrieval Accuracy: The level-specific approach (Manhattan for provider, Squared Chord for tenant, Canberra for user) achieved 85% overall retrieval accuracy, compared to 72% with a single function approach. 2. Solution Quality: Solutions derived from cases retrieved using the level-specific approach showed 18% lower error rates compared to the best single-function approach. 3. Computational Efficiency: The level-specific approach required 15% less computational resources than more complex similarity functions applied across all levels. These findings confirm the value of hierarchical optimization in case-based reasoning, suggesting that domain-specific knowledge about problem structure can significantly enhance CBR performance. 6.6. Limitations and Challenges Despite the overall success, several limitations and challenges were identified: 43 Edelweiss Applied Science and Technology ISSN: 2576-8484 Vol. 9, No. 7: 28-45, 2025 DOI: 10.55214/25768484.v9i7.8530 © 2025 by the author; licensee Learning Gate 1. Scale Constraints: The approach showed diminishing returns for very large systems (>500 ha), suggesting potential scale limitations. 2. Knowledge Gaps: Certain specialized contexts (e.g., high-salinity conditions, extreme topography) had limited representation in the case base, reducing retrieval effectiveness. 3. Implementation Complexity: The multi-level approach required more extensive coordination than single-level interventions, creating management challenges during implementation. 4. Long-term Sustainability: While initial outcomes were positive, long-term sustainability depends on continued maintenance and adaptation, which will require ongoing institutional support. These limitations highlight areas for future refinement and expansion of the approach, particularly regarding scalability and specialized contexts. 6.7. Implications for Irrigation Modernization The findings have several important implications for irrigation modernization in arid regions: 1. Hierarchical Approach Value: The success of the multi-level framework demonstrates the importance of addressing irrigation challenges at all hierarchical levels rather than focusing on isolated components. 2. Knowledge Management: The effective application of CBR highlights the value of systematic knowledge management in irrigation modernization, enabling learning from past experiences rather than reinventing solutions. 3. Integration over Replacement: The successful integration of traditional and modern elements suggests that modernization should build upon rather than replace traditional systems, preserving valuable social and cultural dimensions. 4. Context Sensitivity: The varying performance of different approaches across contexts confirms the importance of context-sensitive modernization rather than standardized solutions. These implications provide valuable guidance for water resource managers, agricultural agencies, and policymakers facing similar challenges in arid and semi-arid regions worldwide. 7. Conclusion This research has demonstrated the successful application of case-based reasoning to the modernization of traditional falaj irrigation systems, using a multi-level framework that addresses challenges at provider, tenant, and user levels. The Falaj Al Sarrani case study illustrates how ancient water management technologies can be enhanced through strategic integration with modern approaches, preserving cultural heritage while significantly improving water use efficiency. The modernization of traditional irrigation systems represents a critical challenge and opportunity for water-scarce regions worldwide. This research demonstrates that through thoughtful integration of traditional and modern approaches, guided by systematic knowledge management and multi-level analysis, significant improvements in water conservation and agricultural productivity can be achieved while preserving cultural heritage. The case of Falaj Al Sarrani illustrates how ancient water wisdom can be harmonized with contemporary technology, creating irrigation systems that honor their historical roots while meeting modern efficiency standards. This balanced approach offers a promising path forward for the thousands of traditional irrigation systems worldwide that face similar challenges of adaptation to changing environmental, economic, and social conditions. Transparency: The author confirms that the manuscript is an honest, accurate, and transparent account of the study; that no vital features of the study have been omitted; and that any discrepancies from the study as planned have been explained. This study followed all ethical practices during writing. 44 Edelweiss Applied Science and Technology ISSN: 2576-8484 Vol. 9, No. 7: 28-45, 2025 DOI: 10.55214/25768484.v9i7.8530 © 2025 by the author; licensee Learning Gate Copyright: © 2025 by the author. This open-access article is distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). References [1] A. R. Al-Adawi and S. Badriyah, "The aflaj irrigation system of Oman: A traditional water management system for sustainable agriculture," International Journal of Water Resources Development, vol. 37, no. 2, pp. 311-328, 2021. [2] A. S. Al-Marshudi, "Traditional irrigated agriculture in Oman: Operation and management of the Aflaj system," Water International, vol. 26, no. 2, pp. 259-264, 2001. [3] S. K. Nadir, "Self-diagnosis of cancer using case base reasoning algorithm," International Journal of Research in Engineering, Science and Management, vol. 4, no. 6, pp. 24-28, 2021. [4] S. A. M. Amal and K. S. Nadir, "Study of attacks on wireless network," International Journal of Computer Engineering and Technology, vol. 13, no. 2, pp. 21-26, 2022. [5] K. Nadir and H. A. Salih, "Survey on QoS for applications," International Journal of Research Publication and Reviews, vol. 3, no. 6, pp. 2157-2164, 2022. [6] S. Zekri, M. Ahmed, R. Schaffner, and I. Al-Rawahi, Aflaj in Oman: Shared water resources in economic and social context,” in Water Management in Islam, N. I. Faruqui, A. K. Biswas, and M. J. Bino, Eds. Tokyo: United Nations University Press, 2001. [7] F. Ministry of Agriculture, and Water Resources,, Ministry of agriculture, fisheries, and water resources. Muscat, Oman: National Strategy for Sustainable Agriculture and Rural Development 2040, 2020. [8] A. Aamodt and E. Plaza, "Case-based reasoning: Foundational issues, methodological variations, and system approaches," AI Communications, vol. 7, no. 1, pp. 39-59, 1994. [9] H. Al-Ghafri, T. Inoue, and T. Nagasawa, "Daudi aflaj: The Qanat of Oman," in Proc. 1st Int. Symp. Qanat, Yazd, Iran, 2000, pp. 77-85, 2000. [10] P. Beaumont, "Qanat systems in Iran," Hydrological Sciences Journal, vol. 16, no. 1, pp. 39-50, 1971. [11] A. S. Al-Marshudi, "The falaj irrigation system and water allocation markets in Northern Oman," Agricultural Water Management, vol. 91, no. 1-3, pp. 71-77, 2007. [12] D. R. Lightfoot, "Syrian qanat Romani: History, ecology, abandonment," Journal of arid Environments, vol. 33, no. 3, pp. 321-336, 1996. [13] A. A. A. Asim and K. S. Nadir, "Study of self-protection in mobile applications," International Journal of Research Publication and Reviews, vol. 3, no. 11, pp. 3048-3053, 2022. [14] A. M. S. Amal and N. K. Salih, "Comparison between black hole and flooding attack inmobile ad-hoc network and their simulation study," International Journal of Research Publication and Reviews, vol. 2582, p. 7421, 2022. [15] K. S. Nadir and A. A. E. Eshtiag, "IOT base monitoring systems for ICU of Buraimi hospital," International Journal of Computer Science and Mobile Applications, vol. 11, no. 1, pp. 36-49, 2023. [16] W. Norman, W. Shayya, A. S. Al-Ghafri, and I. McCann, "Aflaj irrigation and on-farm water management in northern Oman," Irrigation and Drainage Systems, vol. 12, pp. 35-48, 1997. [17] A. S. Al-Ghafri, T. Inoue, and T. Nagasawa, "Irrigation scheduling of aflaj of Oman: Methods and modernization," in Proc. 1st Int. Conf. Irrig. Drain., Fort Collins, CO, USA, 2002, pp. 1-10, 2002. [18] S. Zekri, H. Kotagama, and H. Boughanmi, "Temporary water markets in Oman," Journal of Agricultural and Marine Sciences, vol. 11, pp. 77-84, 2006. [19] N. Lamaddalena and J. A. Sagardoy, Performance analysis of on-demand pressurized irrigation systems. Rome, Italy: FAO Irrigation and Drainage Paper 59, 2000. [20] A. Omezzine and L. Zaibet, "Management of modern irrigation systems in Oman: allocative vs. irrigation efficiency," Agricultural Water Management, vol. 37, no. 2, pp. 99-107, 1998. [21] H. Al-Ismaily and D. Probert, "Water-resource facilities and management strategy for Oman," Applied Energy, vol. 61, no. 3, pp. 125-146, 1998. [22] E. Playán and L. Mateos, "Modernization and optimization of irrigation systems to increase water productivity," Agricultural Water Management, vol. 80, no. 1-3, pp. 100-116, 2006. [23] L. S. Pereira, T. Oweis, and A. Zairi, "Irrigation management under water scarcity," Agricultural Water Management, vol. 57, no. 3, pp. 175-206, 2002. [24] X. Zhu and Y. Yang, "A case-based reasoning approach to urban water management," Water Science and Technology: Water Supply, vol. 17, no. 1, pp. 229-243, 2017. [25] J. K. Panigrahi and P. P. Mujumdar, "Reservoir operation using case-based reasoning: The Malaprabha reservoir in Karnataka, India," Water International, vol. 25, no. 4, pp. 551-559, 2000. [26] R. Sahu and H. Raheman, "A decision support system on matching and field performance prediction of tractor- implement system," Computers and Electronics in Agriculture, vol. 60, no. 1, pp. 76-86, 2008. [27] W. Ghazouani, S. Marlet, I. Mekki, L. W. Harrington, and A. Vidal, "Farmers'practices and community management of irrigation: Why do they not match in fatnassa oasis?," Irrigation and Drainage, vol. 61, no. 1, pp. 39-51, 2012. https://creativecommons.org/licenses/by/4.0/ 45 Edelweiss Applied Science and Technology ISSN: 2576-8484 Vol. 9, No. 7: 28-45, 2025 DOI: 10.55214/25768484.v9i7.8530 © 2025 by the author; licensee Learning Gate [28] G. P. Kofinas et al., "Resilience of Athabascan subsistence systems to interior Alaska’s changing climate," Canadian Journal of Forest Research, vol. 40, no. 7, pp. 1347-1359, 2010. [29] M. Ortega-Reig, G. Palau-Salvador, M. J. C. i Sempere, J. Benitez-Buelga, D. Badiella, and P. Trawick, "The integrated use of surface, ground and recycled waste water in adapting to drought in the traditional irrigation system of Valencia," Agricultural Water Management, vol. 133, pp. 55-64, 2014. [30] Z. M. Mounir, C. M. Ma, and I. Amadou, "Application of water evaluation and planning (WEAP): A model to assess future water demands in the Niger River (in Niger Republic)," Modern Applied Science, vol. 5, no. 1, p. 38, 2011. [31] S. Zekri and A. S. Al-Marshudi, "A millenarian water rights system and water markets in Oman," Water International, vol. 33, no. 3, pp. 350-360, 2008. [32] A. Hammani, M. Kuper, and A. Debbarh, "The modernization of irrigated agriculture," in Proceedings of the Euro- Mediterranean Seminar, Rabat, Morocco, 2004, pp. 19-23, 2005. [33] A. Dinar and J. Mody, "Irrigation water management policies: Allocation and pricing principles and implementation experience," Natural Resources Forum, vol. 28, no. 2, pp. 112-122, 2004. [34] R. Keller and J. Keller, Effective efficiency: A water use efficiency concept for allocating freshwater resources,” water resources and irrigation division discussion paper 22. Arlington, VA: Winrock International, 1995. [35] J. Berbel, C. Gutiérrez-Martín, J. A. Rodríguez-Díaz, E. Camacho, and P. Montesinos, "Literature review on rebound effect of water saving measures and analysis of a Spanish case study," Water Resources Management, vol. 29, pp. 663- 678, 2015. [36] S. N. K. Nadir and D. Idries, "Self-healing applications by using case base reasoning algorithm," International Journal of Research and Innovation in Applied Science, vol. 9, no. 8, pp. 32-42, 2024. [37] M. García-Mollá et al, Irrigation associations coping with drought: The case of four irrigation districts in Eastern Spain,” in Drought in Arid and Semi-Arid Regions, K. Schwabe et al., Eds. Dordrecht: Springer, 2013. [38] F. Molle and J. Berkoff, "Water pricing in irrigation: mapping the debate in the light of experience," Irrigation Water Pricing: The gap Between Theory and Practice, pp. 21-93, 2007. [39] S. Siebert et al., "Groundwater use for irrigation–a global inventory," Hydrology and Earth System Sciences, vol. 14, no. 10, pp. 1863-1880, 2010. [40] R. C. Schank, Dynamic memory: A theory of reminding and learning in computers and people. Cambridge, UK: Cambridge University Press, 1982. [41] J. L. Kolodner, Case-based reasoning. San Mateo, CA: Morgan Kaufmann, 1993. [42] I. Watson, Applying case-based reasoning: Techniques for enterprise systems. San Francisco, CA: Morgan Kaufmann, 1997. [43] D. B. Leake, Case-based reasoning: Experiences, lessons, and future directions. Menlo Park, CA: AAAI Press/MIT Press, 1996. [44] A. Haktanır and J. I. Hargreaves, "A case-based maintenance system for water distribution networks," Water Science and Technology Water Supply, vol. 4, no. 3, pp. 129-138, 2004. [45] R. Lopez De Mantaras et al, "Retrieval, reuse, revision and retention in case-based reasoning," Knowledge Engineering Review, vol. 20, no. 3, pp. 215-240, 2005. [46] M. M. Richter and R. O. Weber, Case-based reasoning: A textbook. Berlin, Germany: Springer, 2013. [47] S. K. Nadir and R. Gopal, "Optimization software fault injection by using two-dimensional bin packing algorithms," Journal Electrical Systems, vol. 20, no. 10s, pp. 6638-6645, 2024. [48] K. S. Nadir, "Self- healing problems systems to support it helpdesk," Journal of Information Systems Engineering and Management, vol. 10, 2025. [49] S. K. Pal and S. C. K. Shiu, Foundations of soft case-based reasoning. Hoboken, NJ: Wiley, 2004. [50] A. Hüllermeier, Case-based approximate reasoning. Dordrecht, Netherlands: Springer, 2007. [51] Food and Agriculture Organization of the United Nations, Coping with water scarcity: An action framework for agriculture and food security. Rome, Italy: FAO Water Reports 38, 2012. [52] World Bank, Water in the Arab world: Management perspectives and innovations. Washington, DC: World Bank, 2009. [53] International Commission on Irrigation and Drainage, Water saving in irrigated agriculture. New Delhi, India: ICID Position Paper, 2016. [54] United Nations Development Programme, Water governance in the Arab region: Managing scarcity and securing the future. New York: United Nations Development Programme, 2013. [55] Intergovernmental Panel on Climate Change, Climate change 2022: Impacts, adaptation and vulnerability,” Contribution of working group ii to the sixth assessment report of the IPCC. Cambridge, UK: Cambridge University Press, 2022. [56] UNESCO, Managing water under uncertainty and risk,” The united nations world water development report 4. Paris, France: UNESCO, 2012. [57] Global Water Partnership, Integrated water resources management in practice: Better water management for development. London, UK: Earthscan, 2009. [58] International Water Management Institute, Water for food, water for life: A comprehensive assessment of water management in agriculture. London, UK: Earthscan, 2007.