برآورد آبدهی قنات‌ها دشت مشهد-چناران با استفاده از الگوریتم جنگل تصادفی

نوع مقاله : مقاله پژوهشی

نویسندگان
حمزه نور
محمد رستمی خلج
علی دسترنج
بخش حفاظت خاک و آبخیزداری، مرکز تحقیقات و آموزش کشاورزی و منابع طبیعی خراسان رضوی، سازمان تحقیقات، آموزش و تروج کشاورزی، مشهد، ایران
10.22034/wmji.2026.2063456.1118
چکیده
آبدهی قنات به عوامل متعددی بستگی دارد که به دلیل پیچیدگی و روابط غیرخطی بین عوامل، الگوریتم‌های یادگیری ماشین مانند جنگل تصادفی، می‌توانند در مدل‌سازی آبدهی قنات‌ها کاربرد مناسبی داشته باشند. در این راستا تحقیق حاضر باهدف تعیین دقت الگوریتم جنگل تصادفی در برآورد آبدهی قنات‌ها دشت مشهد – چناران طرح‌ریزی‌شده است. برای این منظور از اطلاعات 78 قنات موجود در آبخوان مشهد-چناران استفاده شد. از این تعداد 70 درصد برای آموزش و 30 درصد داده‌ها برای ارزیابی مدل جنگل تصادفی در محیط نرم‌افزار R استفاده شد. نتایج تعیین اهمیت عوامل مؤثر بر آبدهی قنات‌ها آبخوان مشهد-چناران نشان داد که از بین هفت عامل موردبررسی، مهم‌ترین عوامل در آبدهی قنات‌ها دشت مشهد-چناران، خصوصیات نوع و هندسه قنات (طول قنات و عمق مادر چاه) و ضریب آبگذری ویژه آبخوان است. درنهایت نتایج ارزیابی دقت الگوریتم جنگل تصادفی در برآورد آبدهی قنات‌ها آبخوان مشهد-چناران نشان داد که مدل جنگل تصادفی با ضریب تبیین (R2) 0/91 و 0/85 به ترتیب در مرحله آموزش و آزمون و هم‌چنین ضریب نش ساتکلیف 0/83 و 0/76 برای این دو مرحله، دارای دقت قابل قبولی می‌باشد. با استفاده از مدل RF ایجادشده در این تحقیق، می‌توان تأثیر سناریوهای مختلف احیا و مدیریت قنات مانند افزایش طول تره‌کار و هم‌چنین عمق مادر چاه را بر دبی قنات قبل از اجرای اقدامات شبیه‌سازی کرد.
کلیدواژه‌ها
سطح آب زیرزمینی
ویژگی‌های هندسی قنات
یادگیری ماشینی
مدل‌سازی

عنوان مقاله English

Predicting Qanats water flow of Mashhad-Chenaran plain using the random forest algorithm

نویسندگان English

Hamzeh Noor
Mohamaad Rostami Khalaj
Ali Dastranj
Soil Conservation and Watershed Management Department, Khorasan Agricultural and Natural Resources Research Centre, Mashhad, Iran
چکیده English

Qanat discharge depends on several factors, and due to the complexity and nonlinear relationships between factors, machine learning algorithms such as random forest can be used appropriately in modeling qanat discharge. In this regard, the present study was designed to determine the accuracy of the random forest algorithm in estimating the discharge of qanats in the Mashhad-Chenaran plain. For this purpose, data from 78 qanats in the Mashhad-Chenaran aquifer were used. Of these, 70% were used for training and 30% of the data were used for testing the random forest model in the R software environment. The results of determining the importance of factors affecting the discharge of qanats in the Mashhad-Chenaran aquifer showed that the most important factors in the discharge of qanats in the Mashhad-Chenaran plain are the characteristics of the type and geometry of the qanat (qanat length and depth of the mother well) and the specific water permeability coefficient. Finally, the results of the assessment of the accuracy of the random forest algorithm in estimating the water yield of the Mashhad-Chenaran aquifer qanats showed that the random forest model has acceptable accuracy with R2 of 0.91 and 0.85 in the training and evaluation stages, respectively, and also the Nash-Sutcliffe coefficient of 0.83 and 0.76 for these two stages. Using the RF model created in this study, the impact of different qanat discharge scenarios, such as increasing the length and depth of the well head, on the qanat discharge can be simulated before any costs.

کلیدواژه‌ها English

Groundwater Level
Geometric Characteristics of Qanat
Machine Learning
Modelling
1.       Avand, M. T., Moradi, H. R., and Ramazanzadeh, M. 2020. Flood Susceptibility Mapping Using Random Forest Machine Learning and Generalized Bayesian Linear Model. Environment and Water Engineering, 6(1): 83-95. (In Persian)
2.       Babran, S. 2008. Water Crisis in in Iran and the World. Rahbord Journal, 15 (48): 193. (In Persian)
3.       Beaumont, P. 1971. Qanat systems in Iran. Hydrological Sciences Journal, 16(1): 39-50
4.       Bhusal, A., Parajuli, U., Regmi, S., and Kalra, A. 2022. Application of machine learning and process-based models for rainfall-runoff simulation in Dupage River basin, Illinois. Hydrology, 9(7): 117.
5.       Breiman, L. 2001. Random forests. Machine learning, 45, 5-32.
6.       Dastranj, A., and Shadfar, S. 2022. Predicting the susceptibility of landslide occurrence in order to manage landslide risk in Bar Neyshabur watershed, Journal of Disaster Prevention and Management Knowledge, 12(3): 253-268. (In Persian)
7.       Dastranj, A., and Rajaei, H. 2023. Investigating the factors affecting the Qanat discharge as a method for sustainable harvesting water from underground water resources. Journal of Rainwater Catchment Systems; 10 (4): 5. (In Persian)
8.       Dou, J., Yunus, A. P., Bui, D. T., Merghadi, A., Sahana, M., Zhu, Z., and Pham, B. T. 2019. Assessment of advanced random forest and decision tree algorithms for modeling rainfall-induced landslide susceptibility in the Izu-Oshima Volcanic Island, Japan. Science of the total environment, 662, 332-346.
9.       Ghasemi, A. 2012. A brief look at the historical course of destruction of the country's water resources, causes, consequences and solutions. Tehran: Bureau of Research and Policies of Production Sectors. (In Persian)
10.    Hemasinghe H, Rangali R. S, Deshapriya N. L, Samarakoon L. 2018. Landslide susceptibility mapping using logistic regression model (a case study in Badulla District, Sri Lanka). Procedia engineering, 212, 1046-1053.
11.    Hong, H., Pradhan, B., Xu, C., Bui, D. 2015. Spatial prediction of landslide hazard at the Yihuang area (China) using two-class kernel logistic regression, alternating decision tree and support vector machines. Catena, 133: 266-281.
12.    Khosravi, K., E. Nohani, E. Maroufinia and Pourghasemi, H.R. 2016. A GIS -based flood susceptibility assessment and its mapping in Iran: a comparison between frequency ratio and weights - of-evidence bivariate statistical models with multi -criteria decision -making technique. Natural Hazards, 83(2): 947 -987.
13.    Kim, H. I., and Kim, B. H. 2020. Flood hazard rating prediction for urban areas using random forest and LSTM. KSCE Journal of Civil Engineering, 24(12): 3884-3896.
14.    Merghadi, A., Yunus, A. P., Dou, J., Whiteley, J., ThaiPham, B., Bui, D. T. and Abderrahmane, B. 2020. Machine learning methods for landslide susceptibility studies: A comparative overview of algorithm performance. Earth-Science Reviews, 207, 103225.
15.    Mirani Moghadam, Karami, Gh. H., Bagheri, R. 2020. Hydrogeological Assessment of the Gonabad Plain Qanats, Journal of Hydrogeology, 4(2): 34-47. (In Persian)
16.    Moriasi, D.N., Arnold, J.G., Van Liew, M.W., Bingner, R.L., Harmel, R.D. and Veith, T.L., 2007. Model evaluation guidelines for systematic quantification of accuracy in watershed simulations. Transactions of the ASABE, 50(3): pp.885-900.
17.     Naghibi, S. A Ahmadi, K., and Daneshi, A. 2017. Application of support vector machine, random forest, and genetic algorithm optimized random forest models in groundwater potential mapping. Water Resources Management, 31: 2761-2775.
18.    Naghibi, S.A., Pourghasemi, H.R., and Abbaspour, K. 2018. A comparison between ten advanced and soft computing models for groundwater qanat potential assessment in Iran using R and GIS. Theoretical and applied Climatology, 131: 967-984.
19.    Nejad Ebrahimi, A., Keynejad, M., and Heydari, M. 2021. An Analysis on the Role of. Geographical Planning of Space, 11(40): 103-118.
20.    Noor, H. 2017. Analysis of groundwater resources utilization and their current condition in Iran. Iranian Journal of Rainwater Catchment Systems, 5(2): 29-38. (In Persian)
21.    Noor, H., Fazli, S., Rostami, M., and Kalat, A. 2017. Cost-effectiveness analysis of different watershed management scenarios developed by simulation–optimization model. Water Science and Technology: Water Supply, 17(5): 1316-1324.
22.    Noor, H., Vafakhah, M., Taheriyoun, M., and Moghadasi, M. 2014. Hydrology modelling in Taleghan mountainous watershed using SWAT. Journal of Water and Land Development.
23.    Papoli Yazdi, M., and Vosoghi, F. 2021. Cultural Analysis of Indigenous Knowledge in the Aqueduct Tool Chain, Journal of Indigenous Knowledge, 6(12): 9-40. (In Persian)
24.    Rahmati, O., Choubin, B., Fathabadi, A., Coulon, F., Soltani, E., Shahabi, H., Mollaefar, E., Tiefenbacher, J., Cipullo, S., Ahmad, B.B.; Bui, D.T. 2019. Predicting uncertainty of machine learning models for modelling nitrate pollution of groundwater using quantile regression and UNEEC methods. Science of the Total Environment, 688: 855-866.
25.    Rostami Khalaj, M. 2023. Modeling groundwater surface changes and its impact on Qanat discharge using the GMS model in Selected Mashhad-Chenaran Qanats, Final Report of Research project, Soil Conservation and Watershed Management Research Institute. (In Persian)
26.    Safaie, F., Yazdi, F., Shojaii, M. 2021. Archaeology of water structures with a focus on the aqueducts of Khorasan. Biannual Journal of Iranian Archaeology, 20: 69-81.
27.    Samani, S., Vadiati, M., Delkash, M., Bonakdari, H. 2023. A hybrid wavelet–machine learning model for qanat water flow prediction. ActaGeophysica, 71(4): 1895-1913
28.    Sedghi, M.M., and Zhan, H. 2020. Semi-analytical solutions of discharge variation of a qanat in an unconfined aquifer subjected to general areal recharge and nearby pumping well discharge. Journal of Hydrology, 584: 124691.
29.    SemsarYazdi, A.A., and LabbafKhaneiki, M. 2016. Qanat knowledge: Construction and maintenance. Springer.
30.    Sheikhi, D., and Pazoki, M. 2021. Study of social functions of Qanat (subterranean) in rural areas with emphasis on social capital indicators (Case study: Hamadan township), Journal of Economic geography research, 2(3): 59-73. (In Persian)
31.    Soltnai Mohammadi, M. 2018. The Impact of Qanat in Physical Form Case Study: Mohammadiyeh Village, Nain, Iran, Journal of Architecture Hot and dry climate, 5(6): 97-115. (In Persian)
32.    Tabatabaei, S. M., and Khozeymehnezhad, H. 2019. Investigating the Effect of Indigenous Knowledge of Qanat on the Dimensions of Sustainable Development, Iranian Journal of Rainwater Catchment Systems, 7(1): 31-44. (In Persian)
33.    Veicy, H. 2021. The Water Crisis in the Central Plateau of Iran and the Necessity of Paying Attention to the Indigenous Knowledge of Qanat in Iran (Case study: Joupar’s GuharRiz Qanat), Geography, 18(67): 104-116. (In Persian)
34.    Vrettas, M.D., and Fung, I.Y. 2015. Toward a new parameterization of hydraulic conductivity in climate models: simulation of rapid groundwater fluctuations in Northern California. Journal of Advances in Modeling Earth Systems. 7 (4): 2105–2135.
35.    Wang, X., Liu, T., Zheng, X., Peng, H., Xin, J., and Zhang, B. 2018. Short-term prediction of groundwater level using improved random forest regression with a combination of random features. Applied Water Science, 8, 1-12.
36.    Yao, J., Qin, S., Qiao, S., Liu, X., Zhang, and Chen, J. 2022. Application of a two-step sampling strategy based on deep neural network for landslide susceptibility mapping. Bulletin of Engineering Geology and the Environment, 81(4): 1-20.
37.    Yousefirad, M., Mokhtar, S., Mahbod, A. 2012. The influential factors on the qanat hydrogeology. J Food Agric Environ, 10(2): 843-848
38.    Zhou, H., Gómez-Hernández, J.J., and Li, L. 2014. Inverse methods in hydrogeology: evolution and recent trends. Advance Water Resource, 63: 22.