نیوار

نیوار

Forecasting short-term precipitation using artificial intelligence on the southern coast of the Caspian Sea

نوع مقاله : مقاله پژوهشی

نویسندگان
Donya Sadeghnezhad 1
Gholamali Kamali 2
Ebrahim Fattahi 3
Zahra Ghassabi 4
majid vazifedoust 5
1 Doctoral student of Meteorology, Islamic Azad University, Science and Research Unit, Tehran, Iran
2 Associate Professor of Climatology, Islamic Azad University, Science and Research Unit, Tehran, Iran
3 Associate Professor of Climatology, Research Institute of Meteorology and Atmospheric Sciences, Tehran, Iran
4 Assistant Professor of Meteorology, Research Institute of Meteorology and Atmospheric Sciences, Tehran, Iran
5 Assistant Professor, Department of Water Engineering, University of Guilan
10.30467/nivar.2025.521906.1335
چکیده
Rainfall is considered one of the most important input data for hydrological systems, and studying and measuring it is often necessary for runoff, groundwater, flooding, and sediment studies. Traditionally, rainfall measurement is conducted using rain gauges. Rain gauges are the most reliable source of rainfall observations and are used in most studies as a reference for comparing and validating satellite data. However, they have limited and irregular spatial coverage. In recent decades, advancements in remote sensing technologies for rainfall have led to a significant reduction in the chronic lack of spatial information, making vast datasets on precipitation available with unprecedented resolution in both space and time. Today, various meteorological satellites such as TRMM, ERA5, CHIRPS, and GPM are actively monitoring the Earth's atmospheric conditions, cloud cover, and humidity, with a high spatial resolution of around kilometers. Precipitation maps from these datasets have been continuously and consistently available on a global scale, but the resulting output images have a low spatial resolution of 25×25 kilometers. This spatial resolution is not highly applicable in small basins and irrigation and drainage networks. In this study, in order to develop an algorithm to increase the spatial resolution of satellite-based precipitation products, an algorithm was extracted using downscaling methods SVM ,RF, KNN,GBR with the help of the relationship between the physical characteristics of clouds and precipitation. Between September 2018 and December 2018 and September 2019 to December 2019 on the GEE platform, precipitation data were downscaled to 1 kilometer using cloud physical properties and data downscaling methods. The downscaled data RF has a higher coefficient of determination and better agreement with ground station data.
کلیدواژه‌ها
Physical characteristics of the cloud
Short-term forecast
Machine learning model
precipitation
موضوعات

عنوان مقاله English

Forecasting short-term precipitation using artificial intelligence on the southern coast of the Caspian Sea

نویسندگان English

Donya Sadeghnezhad 1
Gholamali Kamali 2
Ebrahim Fattahi 3
Zahra Ghassabi 4
Majid Vazifedoust 5
1 Doctoral student of Meteorology, Islamic Azad University, Science and Research Unit, Tehran, Iran
2 Associate Professor of Climatology, Islamic Azad University, Science and Research Unit, Tehran, Iran
3 Associate Professor of Climatology, Research Institute of Meteorology and Atmospheric Sciences, Tehran, Iran
4 Assistant Professor of Meteorology, Research Institute of Meteorology and Atmospheric Sciences, Tehran, Iran
5 Assistant Professor, Department of Water Engineering, University of Guilan
چکیده English

Rainfall is considered one of the most important input data for hydrological systems, and studying and measuring it is often necessary for runoff, groundwater, flooding, and sediment studies. Traditionally, rainfall measurement is conducted using rain gauges. Rain gauges are the most reliable source of rainfall observations and are used in most studies as a reference for comparing and validating satellite data. However, they have limited and irregular spatial coverage. In recent decades, advancements in remote sensing technologies for rainfall have led to a significant reduction in the chronic lack of spatial information, making vast datasets on precipitation available with unprecedented resolution in both space and time. Today, various meteorological satellites such as TRMM, ERA5, CHIRPS, and GPM are actively monitoring the Earth's atmospheric conditions, cloud cover, and humidity, with a high spatial resolution of around kilometers. Precipitation maps from these datasets have been continuously and consistently available on a global scale, but the resulting output images have a low spatial resolution of 25×25 kilometers. This spatial resolution is not highly applicable in small basins and irrigation and drainage networks. In this study, in order to develop an algorithm to increase the spatial resolution of satellite-based precipitation products, an algorithm was extracted using downscaling methods SVM ,RF, KNN,GBR with the help of the relationship between the physical characteristics of clouds and precipitation. Between September 2018 and December 2018 and September 2019 to December 2019 on the GEE platform, precipitation data were downscaled to 1 kilometer using cloud physical properties and data downscaling methods. The downscaled data RF has a higher coefficient of determination and better agreement with ground station data.

کلیدواژه‌ها English

Physical characteristics of the cloud
Short-term forecast
Machine learning model
precipitation
1.    Ahmed, K., Mosa, Z., Al-Ansari, N., Al-Hussaini, M., & Al-Dabbagh, M. (2020). Multi-model ensemble predictions of precipitation and temperature using machine learning algorithms. Atmospheric Research, 236, 104806.
2.    Assamnew, A. D., & Mengistu Tsidu, G. (2023). Assessing improvement in the fifth-generation ecmwf atmospheric reanalysis precipitation over east africa. International Journal of Climatology, 43(1), 17–37.
3.    Bi, K., Foster, J. G., Shan, Q., Li, S., & Yang, Y. (2023). Accurate medium-range global weather forecasting with 3d neural networks. Nature, 619(7971), 533–538.
4.    Bližnák, V., Kašpar, M., Müller, M., & Zacharov, P. (2019). Sub-daily temporal reconstruction of extreme precipitation events using nwp model simulations. Atmospheric Research, 224, 65–80.
5.    Bremnes, J. B. (2004). Probabilistic forecasts of precipitation in terms of quantiles using nwp model output. Monthly Weather Review, 132(2), 338–347.
6.    Buizza, R. (2014). The TIGGE global, medium-range ensembles (ECMWF Tech. Memo. No. 710). DOI: 10.21957/khygx7grk.
7.    Charlton-Perez, A. J., Pappenberger, F., Lipton, R., Bauer, P., & Cripe, D. (2024). Do ai models produce better weather forecasts than physics-based models? a quantitative evaluation case study of storm ciarán. npj Climate and Atmospheric Science, 7(1), 93.
8.    Chen, S., Liu, B., Tan, X., & Wu, Y. (2020). Inter-comparison of spatiotemporal features of precipitation extremes within six daily precipitation products. Climate Dynamics, 54(5-6), 1057–1076.
9.    Davenport, F. V., & Diffenbaugh, N. S. (2021). Using machine learning to analyze physical causes of climate change: A case study of us midwest extreme precipitation. Geophysical Research Letters, 48(14), e2021GL093787.
 10. Du, M., Wu, G., & Zhou, T. (2021). Water vapor anomaly over the tropical western pacific in el niño winters from radiosonde and satellite observations and era5 reanalysis data. Atmospheric Chemistry and Physics, 21(19), 13553–13569. DOI: 10.5194/acp-21-13553-2021
11. Feng, X., & Chen, X. (2021). Feasibility of era5 reanalysis wind dataset on wave simulation for the western inner-shelf of yellow sea. Ocean Engineering, 236, 109413. DOI: https://doi.org/10.1016/j.oceaneng.2021.109413
12. Huang, H., & Huang, Y. (2023). Radiative sensitivity quantified by a new set of radiation flux kernels based on the ecmwf reanalysis v5 (era5). Earth System Science Data, 15(7), 3001–3021. DOI: 10.5194/essd-15-3001-2023
13. Huang, Z., Liu, C., Wang, H., Zhang, R., & Li, P. (2023). Paired satellite and nwp precipitation for global flood forecasting. Journal of Hydrometeorology, 24(10), 2191–2205.
14. Hu, Y., Chen, L., Wang, Z., & Li, H. (2023). Swinvrnn: A data-driven ensemble forecasting model via learned distribution perturbation. Journal of Advances in Modeling Earth Systems, 15, e2022MS003211.
15. Li, L., Carver, R., Lopez-Gomez, I., Sha, F., & Anderson, J. (2024). Generative emulation of weather forecast ensembles with diffusion models. Science Advances, 10(46), eadk4489.
16. Liu, C., Allan, R. P., Brooks, M., & Milton, S. (2014). Comparing tropical precipitation simulated by the met office nwp and climate models with satellite observations. Journal of Applied Meteorology and Climatology, 53(1), 200–214.
17. Liu, Q., Li, X., Cui, Y., Wang, Y., & Chen, Z. (2023). Deep-learning post-processing of short-term station precipitation based on nwp forecasts. Atmospheric Research, 295, 107032.
18. Liu, C., Sun, J., Yang, X., Jin, S., & Fu, S. (2021). Evaluation of ecmwf precipitation predictions in china during 2015–18. Weather and Forecasting, 36(4), 1043–1060.
19. Lam, R., Xiao, H., Guo, D., Dong, W., Chen, Y., Li, J., ... & Chen, S. (2023). Learning skillful medium-range global weather forecasting. Science, 382(6673), 1416–1421.
20. Lewis, W. R., Steenburgh, W. J., Alcott, T. I., & Rutz, J. J. (2017). Gcfs precipitation forecasts and the implications of statistical downscaling over the western united states. Weather and Forecasting, 32(3), 1007–1028.
21. Li, W., Gao, X., Hao, Z., & Sun, R. (2022). Using deep learning for precipitation forecasting based on spatio-temporal information: a case study. Climate Dynamics, 58(3-4), 443–457.
22. Medina, H., Tian, D., Marin, F. R., & Chirico, G. B. (2019). Comparing gefs, ecmwf, and postprocessing methods for ensemble precipitation forecasts over brazil. Journal of Hydrometeorology, 20(4), 773–790.
23. Pathak, J., Selskar, G., Huyan, B., Sircar, S., Nanda, S., Raskar, R., ... & Menon, S. (2022). Fourcastnet: A global data-driven high-resolution weather model using adaptive fourier neural operators. arXiv preprint arXiv:2202.11214.
24. Patra, A., Min, S.-K., & Seong, M.-G. (2020). Climate variability impacts on global extreme wave heights: Seasonal assessment using satellite data and era5 reanalysis. Journal of Geophysical Research: Oceans, 125(11), e2020JC016754. DOI: https://doi.org/10.1029/2020JC016754
25. Qi, Q., Liu, R., Zhang, Q., Wang, W., & Liu, S. (2024). Mapping of 10-km daily diffuse solar radiation across china from reanalysis data and a machine-learning method. Scientific Data, 11(1), 756.
26. Ran, Q., Wang, L., & Zhang, Z. (2018). Evaluation of quantitative precipitation predictions by ecmwf, cma, and ukmo for flood forecasting: Application to two basins in china. Natural Hazards Review, 19(4), 05018003.
27. Rojas-Campos, A., Pérez-Martín, M., Soria-Comas, J., & Sánchez-Arrieta, A. (2023). Postprocessing of nwp precipitation forecasts using deep learning. Weather and Forecasting, 38(3), 487–497.
28. Sachindra, D., Ahmed, K., Rashid, M. M., Shahid, S., & Perera, B. (2018). Statistical downscaling of precipitation using machine learning techniques. Atmospheric Research, 212, 240–258.
29. Senocak, A. U. G., Yilmaz, M. T., Kalkan, S., Yucel, I., & Amjad, M. (2023). An explainable two-stage machine learning approach for precipitation forecast. Journal of Hydrology, 627, 130375.
30. Shi, X., Chen, Z., Wang, W., Yeung, D. I., & Wong, W. K. (2015). Convolutional lstm network: A machine learning approach for precipitation nowcasting. In Advances in Neural Information Processing Systems 28.
31. Shi, X., Chen, Z., Wang, W., Yeung, D. I., & Wong, W. K. (2017). Deep learning for precipitation nowcasting: A benchmark and a new model. In Advances in Neural Information Processing Systems 30.
32. Song, C., & Chen, X. (2021). Performance comparison of machine learning models for annual precipitation prediction using different decomposition methods. Remote Sensing, 13(5), 1018.
 
 
 
 
 
 
 
 
 
 
 
 
 
33. Sun, J., Xiao, H., Chen, Z., Chen, S., & Zhang, S. (2014). Use of nwp for nowcasting convective precipitation: Recent progress and challenges. Bulletin of the American Meteorological Society, 95(3), 409–426.
34. Weyn, J. A., Durran, D. R., & Caruana, R. (2019). Can machines learn to predict weather? using deep learning to predict gridded 500-hpa geopotential height from historical weather data. Journal of Advances in Modeling Earth Systems, 11(9), 2680–2693.
35. Xu, H., Zhao, Y., Zhao, D., Duan, Y., & Xu, X. (2024). Improvement of disastrous extreme precipitation forecasting in north china by pangu-weather ai-driven regional wrf model. Environmental Research Letters, 19(5), 054051.
36. Zhao, S. (1983). A new scheme for comprehensive physical regionalization in china. Acta Geographica Sinica, 38(1), 1–10.

  • تاریخ دریافت 17 اردیبهشت 1404
  • تاریخ بازنگری 16 آذر 1404
  • تاریخ پذیرش 29 آذر 1404
  • تاریخ انتشار 01 مهر 1404