Urban Development Policy Making

Urban Development Policy Making

The Application of Artificial Intelligence in Developing a Local Climate Change Adaptation Strategy for Qanats: Evaluating the Performance of XGBoost in Estimating the Discharge of Qaen Qanats

Document Type : Original Article

Authors
Mohamad Fouladi Nasrabad 1
Mohammad Reza Farzaneh 2
Sepideh Zeraati Neyshabouri 3
1 Ph.D. Student in Water Resources, Department of Water Engineering, University of Birjand, Birjand, Iran
2 Assistant Prof. at Research Group of Environmental Engineering and Pollution Monitoring, Research Center for Environment and Sustainable Development, RCESD, Tehran, Iran
3 Ph.D. in Water Resources, Department of Water Engineering, Faculty of Agriculture, University of Birjand, Birjand, Iran
10.22034/judpm.2026.579182.1096
Abstract
Climate change, as a fundamental challenge of the present century, has made it inevitable for urban and regional policymakers to revise adaptation strategies and develop accurate prediction tools by affecting water resources in arid regions. Qanats, as strategic and vital infrastructures for water supply in settlements of Iran's arid regions, are currently facing threats due to declining groundwater levels, which doubles the need for applying artificial intelligence-based decision support systems. The present study aims to strengthen the foundations of water resources management planning by evaluating the performance of the XGBoost machine learning model in predicting the discharge of qanats in the Qaen Plain. To this end, 12-year data on qanat discharge (2007–2018), climatic variables (precipitation and temperature), and elevation were used. After preprocessing, the data were split into training (70%) and testing (30%) sets, and the model hyperparameters were tuned using RandomizedSearchCV with RMSE as the optimization objective. The evaluation results showed that XGBoost, with R² ≈ 0.88 and NS = 0.76 in the testing phase, has high reliability for use in hydrological analyses. Based on sensitivity analysis, the optimal combination of hyperparameters was n_estimators=500, max_depth=6, learning_rate=0.08, and min_child_weight=4. However, it was observed that the model suffers from poor performance in predicting discharges exceeding 20 L/s due to the RMSE loss function and data imbalance. The findings of this study, while confirming the effectiveness of machine learning models in smartening water management, emphasize the need for caution in policymaking based on extreme flow predictions and provide an operational framework for enhancing the resilience of traditional water infrastructures against climate change.
Keywords
Machine Learning in HydrologyوClimate Change Adaptation
Sensitivity Analysis
Qanat Management
XGBoost Modeling
Subjects
1.    Samaniego L. Permanent shifts in the global water cycle. Science. 2025;387(6741):1348-1350.
2.    Ogunrinde AT, Adeyeri OE, Xian X, Yu H, Jing Q, Faloye OT. Long-term spatiotemporal trends in precipitation, temperature, and evapotranspiration across arid Asia and Africa. Water. 2024;16(22):3161.
3.    Abdelkhalek AM, Awadallah AG, Awadallah NA. Rainfall Extremes Analysis in Arid Regions Under Climate Change: A Structured Review of Methods and Approaches. Climate. 2026 May 3;14(5):100.
4.    Scanlon BR, Fakhreddine S, Rateb A, de Graaf I, Famiglietti J, Gleeson T, et al. Global water resources and the role of groundwater in a resilient water future. Nat Rev Earth Environ. 2023;4(2):87-101.
5.    Wang D, Li K, Li H, Zhang Y, Fu T, Sun L, Wang Y, Zhang J. Water resource utilization and future supply–demand scenarios in energy cities of semi-arid regions. Scientific reports. 2025 Feb 11;15(1):5005.
6.    Granata F, Di Nunno F. Pathways for hydrological resilience: strategies for adaptation in a changing climate. Earth Systems and Environment. 2026 Feb;10(1):203-31.
7.    Bartlett JA, Dedekorkut-Howes A. Adaptation strategies for climate change impacts on water quality: a systematic review of the literature. J Water Clim Change. 2023;14(3):651-675.
8.    Hosen N, Nakamura H, Hamzah A. Adaptation to Climate Change: Does Traditional Ecological Knowledge Hold the Key? Sustainability. 2020;12(2):676.
9.    Akhondi Z, Khayat A, Yaghoobzadeh M. Investigating the Impact of Climate Change on Qanat Discharge: A Case Study of Qanats in Birjand County. J Aquifer Qanat. 2026;5(2):1-18. (In Persian with English abstract)
10.              Ferdowsi A, Piadeh F, Behzadian K, Mousavi SF, Ehteram M. Urban water infrastructure: A critical review on climate change impacts and adaptation strategies. Urban Clim. 2024;58:102132.
11.              Agricultural Research, Education, and Extension Organization. Research on groundwater resources management in Iran (Report No. 1401). Tehran, Iran: Agricultural Research, Education and Extension Organization; 2022.
12.              Ministry of Energy, Office of Basic Water Studies. Annual report on the status of qanats in Iran (Report No. 1402). Tehran, Iran: Ministry of Energy; 2023.
13.              Foruzanmehr M, Khozeymehnezhad H. Investigation of spatial and temporal changes of qualitative parameters of Qaen plain aquifer using interpolation methods. J Aquifer Qanat. 2023;4(1):1-15. (In Persian with English abstract)
14.              Hosseini Fahraji RS, Sharifzadeh M. Qanat maintenance from key informants' viewpoint: A qualitative research in Taft County. J Community Dev (Rural-Urban). 2016;8(2):295-312. (In Persian with English abstract)
15.              Niazkar M, Menapace A, Brentan B, Piraei R, Jimenez D, Dhawan P, Righetti M. Applications of XGBoost in water resources engineering: A systematic literature review (Dec 2018–May 2023). Environ Model Softw. 2024;174:105971.
16.              Samani S, Vadiati M, Kisi O, Ghasemi L, Farajzadeh R. Qanat discharge prediction using a comparative analysis of machine learning methods. Earth Sci Inform. 2024;17(5):4597-4618.
17.              Piraei R, Afzali SH, Niazkar M. Assessment of XGBoost to estimate total sediment loads in rivers. Water Resour Manag. 2023;37(13):5289-5306.
18.              Safira RAD, Anjasmara IM, Awange JL. Predictive Modeling of Terrestrial Water Storage Anomalies in Kalimantan Basins: Bridging the GRACE and GRACE-FO Data Gap with Extreme Gradient Boosting. GEOID. 2024;19(3):508-518.
19.              Kosravi S, Khozeimeh Nejad H, Khayat A, Akhoondi Z. Forecasting the water level of Baldeh Qanat in Ferdows Using Time Series Methods. J Aquifer Qanat. 2026;5(2):189-202. (In Persian with English abstract)
20.              Jiang X, Fan C, Liu K, Chen T, Cao Z, Song C. Centenary covariations of water salinity and storage of the largest lake of Northwest China reconstructed by machine learning. J Hydrol. 2022;612:128095.
21.              Wu J, Ma D, Wang W. Leakage identification in water distribution networks based on XGBoost algorithm. J Water Resour Plan Manag. 2022;148(3):04021107.
22.              Zhou S, Liu Z, Wang M, Gan W, Zhao Z, Wu Z. Impacts of building configurations on urban stormwater management at a block scale using XGBoost. Sustain Cities Soc. 2022;87:104235.
23.              Ali S, Khorrami B, Jehanzaib M, Tariq A, Ajmal M, Arshad A, et al. Spatial downscaling of GRACE data based on XGBoost model for improved understanding of hydrological droughts in the Indus Basin Irrigation System (IBIS). Remote Sens. 2023;15(4):873.
24.              Zhang T, Wang Z, Huang L, He L, Yao C. A novel XGBoost-based approach for reconstruction terrestrial water storage variations with GNSS in the Northeastern Tibetan Plateau. J Hydrol. 2025;659:133255.
25.              Dastourani M, Mamirabadizadeh M, Mohammadie K. The Impact Assessment of Changes in Agricultural Cropping Patterns on the Quantitative Status of Groundwater Resources in the Qaen Plain Aquifer Using the MODFLOW Model. Iran J Irrig Drain. 2025;e223369. (In Persian with English abstract)
26.              Chen T, Guestrin C. Xgboost: A scalable tree boosting system. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining; 2016 Aug 13-17; San Francisco, CA. New York: ACM; 2016. p. 785-794.
27.              Knoben WJ, Freer JE, Woods RA. Inherent benchmark or not? Comparing Nash–Sutcliffe and Kling–Gupta efficiency scores. Hydrol Earth Syst Sci. 2019;23(10):4323-4331.
28.              Falah F, Tahmasbipour N. Identification of outliers in flood data analysis using range and statistical methods. In: Proceedings of the 2nd National Conference on Hydrology of Iran; 2017; (location unknown). (publisher unknown); 2017.
29.              Mohammadi M. Statistics and its applications in experimental sciences. Tehran, Iran: Daneshgah Publishing; 2016. (In Persian with English abstract)
30.              Karimi M, Hosseini A. Time series analysis and its applications in engineering. Tehran, Iran: Daneshgah Publishing; 2015. (In Persian with English abstract)
31.              Willmott CJ, Matsuura K. Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance. Clim Res. 2005;30(1):79-82.
32.              Samantaray S, Sahoo A, Yaseen ZM, Al-Suwaiyan MS. River discharge prediction based multivariate climatological variables using hybridized long short-term memory with nature inspired algorithm. Journal of Hydrology. 2025 Mar 1;649:132453
33.              Ebrahimzadeh Z, Abdollahi K, Bidaki RZ, Eslamian S. A comparative analysis of machine learning models for predicting groundwater and surface water in a stressed semi-arid watershed: The Khanmirza case study. Environmental Earth Sciences. 2026 Mar 17;85(7):159.
34.              Gupta HV, Kling H, Yilmaz KK, Martinez GF. Decomposition of the mean squared error and NSE performance metrics: Implications for improving hydrological modelling. J Hydrol. 2009;377(1-2):80-91.
 
Urban Development Policy Making
Volume 3, Issue 3 - Serial Number 8
Autumn 2026
Pages 455-469

  • Receive Date 27 February 2026
  • Revise Date 07 April 2026
  • Accept Date 26 May 2026
  • Publish Date 23 September 2026