Arak University
Abstract: (53 Views)
Introduction and Objective: In recent decades, growing water demand and frequent droughts—particularly in arid and semi-arid regions—have made the development of sustainable water resource solutions a critical necessity. One effective approach is rainwater harvesting through the construction of small-scale farm dams. Optimal siting of these structures requires the use of robust analytical methods within the framework of Geographic Information Systems (GIS) and Multi-Criteria Decision Analysis (MCDA). As the weighting of criteria plays a pivotal role in spatial model outputs, sensitivity analysis of input weights is a crucial tool for assessing model stability and reliability. The aim of this study is to evaluate the spatial sensitivity of criteria weights in the process of rainwater harvesting site selection, using the AHP-SA2 tool and One-At-a-Time (OAT) method in the Hawkesbury-Nepean catchment, located in western Sydney, Australia.
Materials and Methods: The study area encompasses approximately 174 km² in the western part of the Hawkesbury-Nepean catchment, a hydrologically diverse region with various land use types including urban, agricultural, pastoral, and industrial zones. Six key criteria were selected to evaluate site suitability: runoff, slope, stream order, hydrologic soil group (HSG), land capability, and land use. Runoff was estimated using the standard SCS-CN method. Criteria weights were calculated through the Analytic Hierarchy Process (AHP), and the Consistency Ratio (CR) of the pairwise comparison matrix was computed to validate the weights. Sensitivity analysis was performed using the AHP-SA2 tool, which enables spatial evaluation of how weight variations affect model outcomes. A relative percentage change of ±20% with 5% incremental steps was applied to each criterion independently, resulting in 48 simulation runs. Variations in the spatial classification of suitability were analyzed across all scenarios.
Materials and Methods: The study area encompasses approximately 174 km² in the western part of the Hawkesbury-Nepean catchment, a hydrologically diverse region with various land use types including urban, agricultural, pastoral, and industrial zones. Six key criteria were selected to evaluate site suitability: runoff, slope, stream order, hydrologic soil group (HSG), land capability, and land use. Runoff was estimated using the standard SCS-CN method. Criteria weights were calculated through the Analytic Hierarchy Process (AHP), and the Consistency Ratio (CR) of the pairwise comparison matrix was computed to validate the weights. Sensitivity analysis was performed using the AHP-SA2 tool, which enables spatial evaluation of how weight variations affect model outcomes. A relative percentage change of ±20% with 5% incremental steps was applied to each criterion independently, resulting in 48 simulation runs. Variations in the spatial classification of suitability were analyzed across all scenarios.
Results: The results indicated that the runoff criterion exhibited the highest sensitivity to weight changes, with even minor modifications significantly impacting the spatial suitability classification. Following runoff, slope, stream order, and HSG demonstrated the next highest sensitivities. In contrast, land use had the lowest sensitivity among the criteria. These findings were consistent with the initial weight rankings, reinforcing the reliability of the AHP weighting process. Spatially, suitability classes 1 (most suitable) and 4 (less suitable) were found to be the most stable, whereas intermediate classes 2 and 3 were more sensitive to weight variations. Areas with steep slopes or distinct hydrological characteristics exhibited the greatest fluctuation in classification. Runoff estimation using the SCS-CN method highlighted the significant influence of both soil hydrological group and antecedent moisture condition on final suitability scores. The spatial sensitivity maps generated during this analysis enabled the identification of areas with high uncertainty—those most affected by variations in criteria weights—thus indicating regions where more careful planning and data validation are needed.
Conclusion: This study demonstrated that spatial sensitivity analysis of criteria weights is essential for enhancing the robustness and credibility of MCDA-based spatial models. In the applied model for selecting farm dam locations (SSMFD), the majority of the catchment retained its original suitability classification despite changes in input weights, indicating strong model stability. Nonetheless, high-sensitivity criteria—particularly runoff—should receive special attention during the early stages of decision-making. Overall, the use of the AHP-SA2 tool, combined with spatial analysis, offers valuable insights for weight sensitivity modeling and can be extended to a wide range of natural resource management applications.
Keywords: Spatial sensitivity analysis, Multi-Criteria Decision Analysis (MCDA), farm dam siting, weight evaluation, rainwater harvesting, GIS.
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
