بررسی قابلیت‌های رویکرد یادگیری ماشین در پیش‌بینی جریان سطحی روزانه با استفاده از برخی داده‌های هواشناسی و شاخص تفاضلی نرمال شده برف (مطالعه موردی: حوضه آبخیز لتیان و ناورود)

نوع مقاله : مقاله پژوهشی

نویسندگان

1 گروه علوم و مهندسی خاک، دانشکده کشاورزی، دانشگاه تربیت مدرس، تهران، ایران

2 گروه علوم و مهندسی خاک، دانشکده مهندسی و فناوری کشاورزی، دانشگاه تهران، تهران، ایران

10.22059/ijswr.2022.338986.669207

چکیده

پیش­بینی دقیق جریان سطحی برای مدیریت منابع آب به ویژه پیش­بینی سیل و فرسایش خاک بسیار مهم است. در مطالعه حاضر، قابلیت سه روش یادگیری ماشین (ML) شامل رگرسیون بردار پشتیبان (SVR)، شبکه عصبی مصنوعی با پس انتشار خطا (ANN-BP) و رگرسیون تقویت گرادیان (GBR) با استفاده از داده­های هواشناسی و پوشش برف سنجنده MODIS برای پیش­بینی جریان سطحی روزانه در دو حوضه مختلف لتیان و ناورود بررسی شد. برای توسعه مدل، چهار متغیر اصلی شامل باران روزانه (P)، دمای حداکثر(Tmax) ، دمای حداقل (Tmin) و شاخص تفاضلی نرمال شده برف (NDSI) از سنجنده MODIS در طول سال‌های 1379-1397 استفاده شد. کارایی این مدل‌ها با استفاده از شاخص‌های آماری مورد ارزیابی قرار گرفت. نتایج شبیه‌سازی نشان داد که همه مدل‌ها نتایج رضایت‌بخشی را در شبیه‌سازی جریان سطحی روزانه با استفاده از متغیرهای هواشناسی به عنوان پارامترهای ورودی مدل­ها ارائه کردند. همچنین، کارایی همه مدل‌های ML مورد مطالعه، زمانی که شاخص NDSI به ‌عنوان متغیر تخمین­گر در شبیه­سازی اعمال شد، بهبود یافت. بهترین کارایی را در بین تمام مدل‌های مورد مطالعه در هر دو حوضه، مدل GBR نشان داد. مدل SVR  پایین­ترین کارایی را در پیش‌بینی جریان سطحی روزانه برای هر دو مرحله آموزش و اعتبارسنجی در اکثر موارد نشان داد. به­طور کلی، نتایج شبیه­سازی در حوضه لتیان نسبت به حوضه ناورود در هر دو مرحله آموزش و اعتبارسنجی بهتر بود و نسبت به دو مدل دیگر، بهترین کارایی در مدل GBR با ضریب همبستگی (85/0R=)، ضریب کارایی نش-ساتکلیف )72/0  (NS=و جذر میانگین مربعات خطا ( m3/s43/3(RMSE= با استفاده از شاخص NDSI در حوضه لتیان مشاهده شده است که نشان­دهنده تاثیر زیاد ذوب برف در ایجاد جریان سطحی در مناطق برف­خیز است.

کلیدواژه‌ها


عنوان مقاله [English]

Investigating Capabilities of Machine Learning Techniques in Forecasting Daily Streamflow Using Some Meteorological Data and Normalized Difference Snow Index (Case Study: Latian and Navroud Basins)

نویسندگان [English]

  • Mahboobeh Fallah 1
  • Hosseinali Bahrami 1
  • Hossein Asadi 2
1 Department of Soil Science, Faculty of Agriculture, University of Tarbiat Modares, Tehran, Iran
2 Department of Soil Science, Faculty of Agricultural Engineering and Technology, University of Tehran, Tehran, Iran
چکیده [English]

Accurate prediction of streamflow is crucial for water resources management, especially for the prediction of floods and soil erosion. In the current study, the capability of three machine learning (ML) methods, including Support Vector Regression (SVR), Artificial Neural Network with Backpropagation (ANN-BP), and Gradient Boosting Regression (GBR) was investigated using meteorological observations and MODIS snow cover data to forecast daily streamflow in two different basins, namely Latian and Navroud. For model development, four major predictors, including daily rainfall (P), maximum temperature (Tmax), minimum temperature (Tmin), and the Normalized Difference Snow Index (NDSI) from the MODIS satellite were used from 2000 to 2018. The performance of these models was evaluated using statistical indices. Simulation results revealed that all models presented satisfactory results in simulating daily streamflow using meteorological predictors, and the efficiency of all applied models was improved when the NDSI index was applied as an additional predictor. The best performance was observed in GBR among all studied models in both basins, whereas SVR revealed the lowest performance in forecasting streamflow for both validation and calibration steps in most cases. In general, the simulation results demonstrated higher accuracy in Latian basin than Navroud basin in both calibration and validation steps. The best performance among all models was observed using GBR with R = 0.85, NS=0.72, and RMSE = 3.43 m3/s using the NDSI index in Latian basin indicating the significant effect of snowmelt on streamflow generation in snowmelt-dominated regions.

کلیدواژه‌ها [English]

  • Artificial Neural Networks
  • Machine Learning Model
  • NDSI
  • Support Vector Regression
Ali, Z., Hussain, I., Faisal, M., Nazir, M.H., Hussain, T., Shad, M.Y., Shoukry, M.A. and Gani, S. (2017). Forecasting drought using multilayer perceptron artificial neural network model. Advances in Meteorology, 1–9.
 Ali, M., Prasad, R., Xiang, Y. and Yaseen, Z. (2020). Complete ensemble empirical mode decomposition hybridized with random forest and kernel ridge regression model for monthly rainfall forecasts. Journal of Hydrology, 584, 124647.
Ali, M., Deo, R.C., Maraseni, T. and Downs, N.J. (2019). Improving SPI-derived drought forecasts incorporating synoptic-scale climate indices in multi-phase multivariate empirical mode decomposition model hybridized with simulated annealing and kernel ridge regression algorithms. Journal of Hydrology, 576, 164–184.
Avand, M., Janizadeh, S. and Jafari, F. (2020). Evaluating the Efficiency of Machine Learning Models in Preparing Flood Probability Mapping. Degradation and Rehabilitation of Natural Land. 1(1):19-32. (In Farsi)
Adnan, R.M., Liang, Z., Trajkovic, S., Zounemat-Kermani, M., Li, B. and Kisi, O. (2019). Daily streamflow prediction using optimally pruned extreme learning machine. Journal of Hydrology, 577, 123981.
Ahmadi, H., Malekian, A. and Abedi, R. (2012). The most appropriate statistical method for estimating suspended sediment of Jajrud River (Case study: Rudak station of Jajrud watershed). Journal of Environmental Erosion Research, 2, 88-78.
Bennett, K., Cherry, J., Balk, B. and Lindsey, S. (2019). Using MODIS estimates of fractional snow cover area to improve streamflow forecasts in interior Alaska. Hydrology and Earth System Sciences, 23, 2439-2459.
Boser B.E., Guyon I.M. and Vapnik V.N. (1992). A training algorithm for optimal margin classiers. In D.Haussler, editor, 5th Annual ACM Workshop on COLT, pp. 144 -152, Pittsburgh, PA.
Bilandi M., Khashei siouki E. and Sadeghi Tabas S. (2014). Daily rainfall-runoff modeling with Least Square Support Vector Machine (LS-SVM). Journal of Soil and Water Conservation Research, 6, 293 -304. (In Farsi)
Benimam H., Si -Moussa C., Laidi M. and Hanini S. (2020). Modeling the activity coefficient at infinite dilution of water in ionic liquids using artificial neural networks and support vector machines. Neural Comput. Appl. 32 (12), 8635 –8653.
Cheng, C., Zhao, M.Y., Chau, K. and Wu, X.Y. (2006). Using genetic algorithm and TOPSIS for Xinanjiang model calibration with a single procedure. Journal of Hydrology, 316, 129-140.
Cheng, M., Fang, F., Kinouchi, T., Navon, I.M. and Pain, C.C. (2020). Long lead-time daily and monthly streamflow forecasting using machine learning methods. Journal of Hydrology, 590, 125376.
Dikshit, A., Pradhan, B. and Alamri, A.M. (2020). Temporal Hydrological Drought Index Forecasting for New South Wales, Australia Using Machine Learning Approaches. Atmosphere, 11(6), 585.
Dibike, Y.B., Velickov, S., Solomatine, D. and Abbott, M.B. (2001). Model induction with support vector machines: introduction and applications. Journal of Computing in Civil Engineering, 15(3), 208-216.
Ebrahimi, H. and Rajaee, T. (2017). Simulation of groundwater level variations using wavelet combined with neural network, linear regression and support vector machine. Global and Planetary Change, 148, 181–191. 
Guo, W.D., Chen, W.B., Yeh, S.H., Chang, C.H. and Chen, H, (2021). Prediction of River Stage Using Multistep-Ahead Machine Learning Techniques for a Tidal River of Taiwan. Water, 13(7), 920
Garmdareh, E., Vafakhah, M. and Eslamian, S. (2019). Assessment the Performance of Support Vector Machine and Artificial Neural Network Systems for Regional Flood Frequency Analysis (A Case Study: Namak Lake Watershed). Journal of Water and Soil Science. 23. 351-366. (In Farsi).
Hall, D.K., Riggs, G.A. and Salomonson, V.V. (1995). Development of methods for mapping global snow cover using moderate resolution imaging spectro-radiometer data. Remote Sensing of Environment, 54(2), 127–140.
Ha, S., Liu, D. and Mu, L. (2021). Prediction of Yangtze River stream flow based on deep learning neural network with El Niño–Southern Oscillation. Scientific Reports, 11, 11738.
Hosseini, S.M. and Mahjouri, N. (2016). Integrating Support Vector Regression and a geo-morphologic Artificial Neural Network for daily rainfall-runoff modeling. Applied Soft Computing, 38, 329–345.
He, S., Gu, L., Tian, J., Deng, L., Yin, J., Liao, Z., Zeng, Z., Shen, Y. and Hui, Y. (2021). Machine Learning Improvement of Stream Flow Simulation by Utilizing Remote Sensing Data and Potential Application in Guiding Reservoir Operation. Sustainability, 13, 3645.
Hadi, S.J. and Tombul, M. (2018). Forecasting daily stream flow for basins with different physical characteristics through data-driven methods. Water Resources Management, 32, 3405–3422.
Hay, L.E., Wilby, R.L. and Leavesley, G.H. (2000). A comparison of delta changes and downscaled GCM scenarios for three mountainous basins in the United States 1. JAWRA Journal of the American Water Resources Association, 36, 387-397
Kashid, S.S., Ghosh, S. and Maity, R. (2020). Stream flow prediction using multi-site rainfall obtained from hydroclimatic teleconnection. Journal of Hydrology, 395, 23–38.
Keteklahijani, V.K., Alimohammadi, S. and Fattahi, E. (2019). Predicting changes in monthly streamflow to Karaj dam reservoir, Iran, in climate change condition and assessing its uncertainty. Ain Shams Engineering Journal, 10, 669–679.
Khan, N., Sachindra, D.A., Shahid, S., Ahmed, K., Shiru, M.S. and Nawaz, N. (2020). Prediction of droughts over Pakistan using machine learning algorithms. Advances in Water Resources, 139, 103562.
Konapala, G., Kao, S.C., Painter, S. and Lu, D. (2020). Machine learning assisted hybrid models can improve stream flow simulation in diverse catchments across the conterminous US. Environmental Research Letters, 15(10), 104022.
Lin, J.Y., Cheng, C.T. and Chau, K.W. (2006). Using support vector machines for long-term discharge prediction. Hydrological Sciences Journal, 51(4), 599-612.
Liu, Y., Sang, Y.F., Li, X., Hu, J. and Liang, K. (2017). Long-term stream flow forecasting based on relevance vector machine model. Water, 9(1), 9.
Liao, S., Liu, Z., Liu, B., Cheng, C., Jin, X. and Zhao, Z. (2020). Multistep-ahead daily inflow forecasting using the ERA-Interim reanalysis data set based on gradient-boosting regression trees. Hydrology and Earth System Sciences, 24, 2343–2363.
Meng, E., Huang, S., Huang, Q., Fang, W., Wu, L. and Wang, L. (2019). A robust method for non-stationary stream flow prediction based on improved EMD-SVM model. Journal of Hydrology, 568, 462-478.
Malik, A., Tikhamarine, Y., Souag-Gamane, D., Kisi, O. and Pham, Q.B. (2020a), Support vector regression optimized by meta-heuristic algorithms for daily stream flow prediction. Stochastic Environmental Research and Risk Assessment, 34, 1755–1773.
Malik, A., Kumar, A., Salih, S.Q., Kim, S., Kim, N.W., Yaseen, Z.M. and Singh, V.P. (2020b).  Drought index prediction using advanced fuzzy logic model: Regional case study over Kumaon in India. PLoS ONE, 15 (5), e0233280.
Mahjouri, N. and Kerachian, R. (2011). Revising river water quality monitoring networks using discrete entropy theory: The Jajrood River experience. Environmental Monitoring and Assessment, 175, 291-302.
Mokari, E., DuBois, D., Samani, Z., Mohebzadeh, H. and Djaman, K. (2021). Estimation of daily reference evapotranspiration with limited climatic data using machine learning approaches across different climate zones in New Mexico. Theoretical and Applied Climatology, 1-13.
Nolin, A.W. and Liang, S. (2002). Progress in bidirectional reflectance modeling and applications for surface particulate media: snow and soils. Remote Sensing Reviews, 18, 307–342.
Niu, W.J., Feng, Z.K., Zeng, M., Feng, B.F., Min, Y.W. and Cheng, C. (2019). Forecasting reservoir monthly runoff via ensemble empirical mode decomposition and extreme learning machine optimized by an improved gravitational search algorithm. Applied Soft Computing, 105589.
Ni, Q., Wang, L., Ye, R., Yang, F. and Sivakumar, M. (2010). Evolutionary modeling for stream flow forecasting with minimal datasets: a case study in the West Malian River, China. Environmental Engineering Science, 27,377–385.
Prasad, R., Ali, M., Xiang, Y. and Khan, H. (2020). A double decomposition-based modelling approach to forecast weekly solar radiation. Renewable Energy, 152, 9–22.
Parisouj, P., Mohebzadeh, H. and Lee, T. (2020). Employing Machine Learning Algorithms for Streamflow Prediction: A Case Study of Four River Basins with Different Climatic Zones in the United States. Water Resources Management, 34, 4113–4131.
Roy, D.K., Barzegar, R., Quilty, J. and Adamowski, J. (2020). Using ensembles of adaptive neuro-fuzzy inference system and optimization algorithms to predict reference evapotranspiration in subtropical climatic zones. Journal of Hydrology, 591, 125509.
Rakwatin, P., Takeuchi, W. and Yasuoka, Y. (2008). Restoration of Aqua MODIS band 6 using histogram matching and local least squares fitting. IEEE Transactions on Geoscience and Remote Sensing, 47, 613–627.
Steele, C., Dialesandro, J., James, D., Elias, E., Rango, A. and Bleiweiss, M. (2017). Evaluating MODIS snow products for modelling snowmelt runoff: Case study of the Rio Grande headwaters. International Journal of Applied Earth Observation and Geoinformation, 63, 234-243.
Shahabi, H., Khezri, S., Ahmad, B.B. and Musa, T. (2014). Application of moderate resolution imaging spectroradiometer snow cover maps in modeling snowmelt runoff process in the central Zab basin, Iran. Journal of Applied Remote Sensing, 8(1), 084699.
Shortridge, J.E., Guikema, S.D. and Zaitchik, B.F. (2016). Machine learning methods for empirical streamflow simulation: a comparison of model accuracy, interpretability, and uncertainty in seasonal watersheds. Hydrology and Earth System Sciences, 20, 2611-2628.
Tyralis, H., Papacharalampous, G. and Langousis, A. (2021). Super ensemble learning for daily stream flow forecasting: large-scale demonstration and comparison with multiple machine learning algorithms. Neural Computing and Applications, 33, 3053–3068.
Tongal, H. and Booij, M.J. (2018). Simulation and forecasting of stream flows using machine learning models coupled with base flow separation. Journal of hydrology, 564, 266-282.
Venkatesan, E. and Mahindrakar, A.B. (2019). Forecasting floods using extreme gradient boosting – A new approach. International Journal of Civil Engineering and Technology, 10, 1336-1346.
Wang, L., Li, X., Ma, C. and Bai, Y. (2019). Improving the prediction accuracy of monthly stream flow using a data-driven model based on a double-processing strategy. Journal of Hydrology, 573, 733-745.
Weier, J. and Herring, D. (2011). Measuring Vegetation (NDVI & EVI)” (http://earthobservatory.nasa.gov/Features/MeasuringVegetation/).
Xiang, Z. and Demir, I. (2020). Distributed long-term hourly stream flow predictions using deep learning—a case study for State of Iowa. Environmental Modelling & Software, 131, 10476.
Xiang, Z., Yan, J. and Demir, I. (2020). A rainfall-runoff model with LSTM-based sequenceto-sequence learning. Water Resources Research, 56(1), e2019WR025326.
Yu, P., Yang, T., Chen, S., Kuo, C. and Tseg, H. (2017). Comparison of random forests and support vector machine for real-time radar-derived rainfall forecasting. Journal of Hydrology, 552, 92–104.
Zhang, H., Yang, Q., Shao, J., and Wang, G. (2019). Dynamic Stream Flow Simulation via Online Gradient-Boosted Regression Tree. Journal of Hydrologic Engineering, 24(10), 0401.