یک مدل شبکة عصبی پیچشی برای پیش‌بینی مسیر حرکت طوفان‌های گردوغبار

نوع مقاله : مقاله پژوهشی

نویسندگان

1 دانشجوی کارشناسی ارشد، دانشکدة مهندسی نقشه‌برداری، دانشگاه صنعتی خواجه نصیرالدین طوسی، تهران

2 استاد دانشکدة مهندسی نقشه‌برداری، دانشگاه صنعتی خواجه نصیرالدین طوسی، تهران

3 استادیار دانشکدة علوم انسانی، دانشگاه هرمزگان، بندرعباس

چکیده

طوفان‌های گردوغبار بلایایی طبیعی‌اند که در زندگی انسان و محیط‌زیست تأثیر چشمگیری گذاشته‌اند. توسعة مدل‌هایی، به‌منظور پیش‌بینی مسیر حرکت این طوفان‌ها، در پیشگیری و مدیریت طوفان‌های گردوغبار نقش بسزایی ایفا می‌کند زیرا مسیر انتقال آنها را آشکار و مناطق آسیب‌پذیر بعدی در برابر طوفان را مشخص می‌کنند. به‌لطف امکانات روش‌های یادگیری عمیق در حل مسائل مبتنی‌بر سری زمانی و یافتن الگوهای پنهان از حجم دادة کلان، در این پژوهش، یک مدل ترکیبی شبکة عصبی پیچشی (CNN) به‌منظور پیش‌بینی مسیر حرکت طوفان گردوغبار، براساس دادة عمق نوری هواویز (AOD) محصول MERRA-2 برای دوازده ساعت آینده، توسعه داده ‌شده است. همچنین چهل رویداد طوفان، شامل 2489 ساعت طوفان در منطقه‌ای خشک در مرکز و جنوب آسیا، به‌منظور آموزش مدل به‌کار رفته است. نتایج نشان می‌دهد که مدل پیشنهادی پیش‌بینی دقیقی از مسیر حرکت طوفان به‌دست می‌دهد؛ به‌گونه‌ای‌که درمورد گام‌های زمانی 3، 6، 9 و 12 ساعت آینده، مقادیر دقت کلی به‌ترتیب برابر با 9806/0، 9810/0، 9813/0 و 9790/0، مقادیر امتیاز F1 به‌ترتیب برابر با 8490/0، 8524/0، 8530/0 و 8384/0 و مقادیر ضریب کاپا به‌ترتیب برابر با 8387/0، 8424/0، 8431/0 و 8273/0 است.

کلیدواژه‌ها


عنوان مقاله [English]

A Convolutional Neural Network Model for Predicting the Transport Pathway of Dust Storms

نویسندگان [English]

  • Mahdis Yarmohamadi 1
  • Ali Asghar Alesheikh 2
  • Mohammad Sharif 3
1 M.Sc. Student, Dep. of Geospatial Information Systems, K. N. Toosi University of Technology, Tehran
2 Professor, Dep. of Geospatial Information Systems, K. N. Toosi University of Technology, Tehran
3 Assistant Prof., Dep. of Geography, University of Hormozgan, Bandar Abbas
چکیده [English]

Dust storms are natural disasters that have severely affected human life and the environment. The majority of research in dust storm has been dedicated to the forecasting of storm-prone areas. However, developing models to predict the movement of these storms plays a significant role in the prevention and management of dust storms, because they reveal the transport pathway and identify the next vulnerable areas against the storm. In this research, a hybrid convolutional neural network (CNN) model has been developed to predict the path of dust storms based on airborne optical depth (AOD) data of MERRA-2 product for the next 12 hours. 40 storm events including 2489 storm hours in a dry region in Central and South Asia have been used for training the model. The results show that the proposed model provides an accurate prediction of the storm's path, so that for the time steps of 3, 6, 9, and 12 hours, the overall accuracy values are 0.9806, 0.9810, 0.9813, and 0.9790, respectively, the F1 score values are 0.8490, 0.8524, 0.8530, and 0.8384, respectively, and the Kappa coefficient values are 0.8387, 0.8424, 0.8431, and 0.8273, respectively.

کلیدواژه‌ها [English]

  • Deep learning
  • Movement prediction
  • Moving process
  • Dust storms
  • MERRA-2
Aili, A., Xu, H., Kasim, T. & Abulikemu, A., 2021, Origin and Transport Pathway of Dust Storm and Its Contribution to Particulate Air Pollution in Northeast Edge of Taklimakan Desert, China, Atmosphere, 12, P. 113.
Al-Yahyai, S. & Charabi, Y., 2014, Trajectory Calculation as Forecasting Support Tool for Dust Storms, Advances in Meteorology, 2014.
Alzubaidi, L., Zhang, J., Humaidi, A.J., Al-dujaili, A., Duan, Y., Al-Shamma, O., Santamaría, J., Fadhel, M.A., Al-Amidie, M. & Farhan, L., 2021, Review of Deep Learning: Concepts, CNN Architectures, Challenges, Applications, Future Directions, Journal of Big Data, 8.
Ansari Ghojghar, M., Pourgholam-Amiji, M., Bazrafshan, J., Liaghat, A. & Araghinejad, S., 2020, Performance Comparison of Statistical, Fuzzy and Perceptron Neural Network Models in Forecasting Dust Storms in Critical Regions in Iran, Iranian Journal of Soil and Water Research, 51(8), PP. 2051-2063.
Boloorani, A.D., Shorabeh, S.N., Neysani Samany, N., Mousivand, A., Kazemi, Y., Jaafarzadeh, N., Zahedi, A. & Rabiei, J., 2021, Vulnerability Mapping and Risk Analysis of Sand and Dust Storms in Ahvaz, IRAN, Environmental Pollution, 279, P. 116859.
Boroumand, F., Alesheikh, A.A., Sharif, M. & Farnaghi, M., 2022, FLCSS: A FuzzyBased Longest Common Subsequence Method for Uncertainty Management in Trajectory Similarity Measures, Transactions in GIS, 26, PP. 2244-2262.
Burge, J., Bonanni, M., Ihme, M. & Hu, L., 2020, Convolutional LSTM Neural Networks for Modeling Wildland Fire Dynamics, arXiv preprint arXiv:2012.06679.
Chollet, F., 2021, Deep Learning with Python, Simon and Schuster.
Giovanni, 2022, The Bridge Between Data and Science. https://giovanni.gsfc.nasa.gov/ giovanni/.
Goudarzi, S., Sharif, M. & Karimipour, F., 2022, A Context-Aware Dimension Reduction Framework for Trajectory and Health Signal Analyses, J. Ambient Intell. Humaniz. Comput., 13, PP. 2621-2635.
Goudie, A.S. & Middleton, N.J., 2006, Desert Dust in the Global System, Springer Science & Business Media.
Guan, Q., Yang, Y., Luo, H., Zhao, R., Pan, N., Lin, J. & Yang, L., 2019, Transport Pathways of PM10 During the Spring in Northwest China and Its Characteristics of Potential Dust Sources, Journal of Cleaner Production, 237(10), P. 117746.
Jiao, Z., Jia, G. & Cai, Y., 2019, A New Approach to Oil Spill Detection that Combines Deep Learning with Unmanned Aerial Vehicles, Comput. Ind. Eng., 135, PP. 1300-1311.
Kattenborn, T., Leitloff, J., Schiefer, F. & Hinz, S., 2021, Review on Convolutional Neural Networks (CNN) in Vegetation Remote Sensing, ISPRS Journal of Photogrammetry and Remote Sensing.
Kolmonen, P., Sundström, A.M., Sogacheva, L., Rodriguez, E., Virtanen, T. & de Leeuw, G., 2013, Uncertainty Characterization of AOD for the AATSR Dual and Single View Retrieval Algorithms, Atmos. Meas. Tech. Discuss., 6, PP. 4039-4075.
Krizhevsky, A., Sutskever, I. & Hinton, G.E., 2012, ImageNet Classification with Deep Convolutional Neural Networks, Communications of the ACM, 60, PP. 84-90.
LeCun, Y., Boser, B.E., Denker, J.S., Henderson, D., Howard, R.E., Hubbard, W.E. & Jackel, L.D., 1989, Backpropagation Applied to Handwritten Zip Code Recognition, Neural Computation, 1, PP. 541-551.
Liu, S., Wang, T. & Mouat, D., 2013, Temporal and Spatial Characteristics of Dust Storms in the Xilingol Grassland, Northern China, During 1954–2007, Regional Environmental Change, 13(1), PP. 43-52.
Mahdianpari, M., Salehi, B., Rezaee, M., Mohammadimanesh, F. & Zhang, Y., 2018, Very Deep Convolutional Neural Networks for Complex Land Cover Mapping Using Multispectral Remote Sensing Imagery, Remote. Sens., 10, P. 1119.
Mobarak Hassan, E., Saadatabadi, A. & Fattahi, E., 2020, Dust Investigation by MERRA-2 Model in Iran: (during 2007- 2017), Iranian Journal of Soil and Water Research, 51(9), PP. 2203-2219.
Mohammadpour, K., Rashki, A., Sciortino, M., Kaskaoutis, D.G. & Boloorani, A.D., 2022, A Statistical Approach for Identification of Dust-AOD Hotspots Climatology and Clustering of Dust Regimes over Southwest Asia and the Arabian Sea, Atmospheric Pollution Research, 13(4), P. 101395.
Perumal, R. & van Zyl, T.L., 2020, Comparison of Recurrent Neural Network Architectures for Wildfire Spread Modelling, 2020 International SAUPEC/ RobMech/PRASA Conference.
Rashki, A., Middleton, N. & Goudie, A.S., 2021, Dust Storms in Iran – Distribution, Causes, Frequencies and Impacts, Aeolian Research, 48, P. 100655.
Safriel, U., Adeel, Z., Niemeijer, D., Puigdefabregas, J., White, R., Lal, R., Winslow, M., Ziedler, J., Prince, S. & Archer, E., 2005, Dryland Systems, In Ecosystems and Human Well-being: Current State and Trends.: Findings of the Condition and Trends Working Group (PP. 623-662), Island Press.
Sharif, M., Alesheikh, A.A. & Tashayo, B., 2019, CaFIRST: A Context-Aware Hybrid Fuzzy Inference System for the Similarity Measure of Multivariate Trajectories, Journal of Intelligent & Fuzzy Systems, 36(6), PP. 5383-5395.
Stein, A.F., Draxler, R.R., Rolph, G., Stunder, B., Cohen, M.D. & Ngan, F., 2015, NOAA’s HYSPLIT Atmospheric Transport and Dispersion Modeling System, Bulletin of the Amerigcan Meteorological Society, 96, PP. 2059-2077.
Tiancheng, L., Qing-dao-er-ji, R. & Ying, Q., 2019, Application of Improved Naive Bayesian-CNN Classification Algorithm in Sandstorm Prediction in Inner Mongolia, Advances in Meteorology, 2019.
Wang, R., Zhu, Z., Zhu, W.-h., Fu, X. & Xing, S., 2021, A Dynamic Marine Oil Spill Prediction Model Based on Deep Learning, Journal of Coastal Research, 37, PP. 716-725.
Yousefi, R., Wang, F., Ge, Q.-s. & Shaheen, A., 2020, Long-Term Aerosol Optical Depth Trend over Iran and Identification of Dominant Aerosol Types, The Science of the Total Environment, 722, P. 137906.