A Spatiotemporal CNN–LSTM Framework for Daily Precipitation Bias Correction Using Satellite and Reanalysis Data

Miri, Seyed Mohammad; Kavianpour, Mohammad Reza

doi:10.61882/NMCE.2601.1124

A Spatiotemporal CNN–LSTM Framework for Daily Precipitation Bias Correction Using Satellite and Reanalysis Data

Document Type : Research

Authors

Seyed Mohammad Miri ¹

Mohammad Reza Kavianpour ²

¹ Ph.D. Candidate, Faculty of Civil Engineering, K.N. Toosi University of Technology, Tehran, Iran

² Professor, Faculty of Civil Engineering, K.N. Toosi University of Technology, Tehran, Iran

10.61882/NMCE.2601.1124

Abstract

Accurate daily precipitation estimates are critical for hydrological and civil engineering applications, especially in regions affected by short-duration rainfall events. Satellite-based products provide wide coverage but often show systematic biases compared with ground observations. This study presents a spatiotemporal convolutional neural network–long short-term memory (CNN–LSTM) framework for daily precipitation bias correction in Golestan Province, northern Iran, using satellite and reanalysis data. The proposed framework integrates PERSIANN (Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks) satellite precipitation with ERA5 reanalysis precipitation and 2-m air temperature to improve station-scale rainfall estimation at several synoptic stations. In order to represent the spatial organization of precipitation systems vividly, gridded precipitation from neighboring upstream cells was incorporated into the model structure. This configuration accounts for the dominant west-to-east movement of rainfall systems across the region, where precipitation commonly arrives with temporal delay and reduced intensity. In addition, temporal variations in near-surface air temperature were included to provide supplementary physical information associated with rainfall development. Model performance is evaluated using root mean square error (RMSE), correlation coefficient (CC), and mean error (ME). Categorical skill is assessed using probability of detection (POD) and false alarm ratio (FAR). The results indicate that the CNN–LSTM model consistently outperformed both conventional bias-correction approaches and baseline deep learning models. Incorporating spatial information from adjacent grid cells reduces errors and improves temporal consistency. Across the evaluated stations, the proposed model achieved the lowest RMSE values, with error reductions reaching 23% relative to the DNN benchmark. The framework also demonstrated improved capability in capturing the spatiotemporal evolution of precipitation events. Overall, the proposed approach provides a transferable and operationally practical framework for enhancing daily precipitation estimates in data-scarce regions. The improved precipitation estimates can support more reliable hydrological simulations, flood assessment, and water resources planning.

Keywords

Satellite precipitation

Rainfall estimation

CNN-LSTM

ERA5 reanalysis

Spatiotemporal Deep Learning

Subjects

Water Resources Engineering

[1] Center for Hydrometeorology and Remote Sensing (CHRS). CHRS Data Portal: PERSIANN-UNet precipitation product. University of California, Irvine. Available at: https://chrsdata.eng.uci.edu/.

[2] Copernicus Climate Change Service (C3S). ERA5 hourly data on single levels from 1940 to present. Copernicus Climate Data Store; 2018.

[3] Wang, F., Tian, D., & Carroll, M. (2023). Customized deep learning for precipitation bias correction and downscaling. Geoscientific Model Development, 16(2), 535–556.

[4] Apaza-Vilca, Cristhian, Maria Liz Mamani-Yupanqui, and Efrain Lujano. "Performance Evaluation of the ERA5, MERRA-2, and PERSIANN-CDR Gridded Products in the Tambo Basin." Environmental and Earth Sciences Proceedings 32.1 (2025): 17.

[5] Putro, Danurendro Ramadhani, and Obaja Triputera Wijaya. "Evaluating PERSIANN-CDR and ERA5 Precipitation Products: Insights from Bandung Raya." Jurnal Teknik Sipil 25.1 (2025): 1623-1633.

[6] Saravanan, Aatralarasi, et al. "Evaluation of remote-sensing-and reanalysis-based precipitation products for agro-hydrological studies in the semi-arid tropics of Tamil Nadu." Hydrology and Earth System Sciences 29.19 (2025): 4847-4870.

[7] Bagiliko, John, et al. "Validation of satellite and reanalysis rainfall products against rain gauge observations in Ghana and Zambia." Theoretical and Applied Climatology 156.5 (2025): 1-20.

[8] Assamnew, Abera Debebe, Birhan Gessese Gobie, and Birhanu Asmerom Habtemicheal. "Evaluating the accuracy of gridded precipitation datasets and CMIP6 GCM ensemble means against gauge observations in South Wollo, Ethiopia." Scientific Reports 15.1 (2025): 34197.

[9] Zhang, Jianbin, Zhiqiu Gao, and Yubin Li. "Deep-learning correction methods for weather research and forecasting (WRF) model precipitation forecasting: a case study over zhengzhou, China." Atmosphere 15.6 (2024): 631.

[10] Hu, Yi‐Fan, Fu‐Kang Yin, and Wei‐Min Zhang. "Deep learning‐based precipitation bias correction approach for Yin–He global spectral model." Meteorological Applications 28.5 (2021): e2032.

[11] Miri, S. M., M. R. Kavianpour, and M. J. Alizadeh. "Short-term rainfall forecasting using multi-task learning and Weibull based postprocessing technique." International Journal of Environmental Science and Technology 22.14 (2025): 13571-13584.

[12] Yang, Xiaoying, et al. "Correcting the bias of daily satellite precipitation estimates in tropical regions using deep neural network." Journal of Hydrology 608 (2022): 127656.

[13] Moraux, Arthur, et al. "A deep learning multimodal method for precipitation estimation." Remote Sensing 13.16 (2021): 3278.

[14] Tao, Yumeng, et al. "A deep neural network modeling framework to reduce bias in satellite precipitation products." Journal of Hydrometeorology 17.3 (2016): 931-945.

[15] Akbari Asanjan, Ata, et al. "Short‐term precipitation forecast based on the PERSIANN system and LSTM recurrent neural networks." Journal of Geophysical Research: Atmospheres 123.22 (2018): 12-543.

[16] Sadeghi, Mojtaba, et al. "Improving near real-time precipitation estimation using a U-Net convolutional neural network and geographical information." Environmental Modelling & Software 134 (2020): 104856.

[17] Amini, Amirmasoud, Mehri Dolatshahi, and Reza Kerachian. "Effects of automatic hyperparameter tuning on the performance of multi‐variate deep learning‐based rainfall nowcasting." Water Resources Research 59.1 (2023): e2022WR032789.

[18] Amini, Amirmasoud, Mehri Dolatshahi, and Reza Kerachian. "Real-time rainfall and runoff prediction by integrating BC-MODWT and automatically-tuned DNNs: Comparing different deep learning models." Journal of Hydrology 631 (2024): 130804.

[19] Aderyani, Fatemeh Rezaei, S. Jamshid Mousavi, and Fatemeh Jafari. "Short-term rainfall forecasting using machine learning-based approaches of PSO-SVR, LSTM and CNN." Journal of Hydrology 614 (2022): 128463.

[20] Li, Peifeng, Jin Zhang, and Peter Krebs. "Prediction of flow based on a CNN-LSTM combined deep learning approach." Water 14, no. 6 (2022): 993.

[21] Ahmed, AA Masrur, Shahida Akther, Thong Nguyen-Huy, Nawin Raj, S. Janifer Jabin Jui, and S. Z. Farzana. "Real-time prediction of the week-ahead flood index using hybrid deep learning algorithms with synoptic climate mode indices." Journal of Hydro-Environment Research 57 (2024): 12-26.

[22] Wu, H., Yang, Q., Liu, J., & Wang, G. (2020). A spatiotemporal deep fusion model for merging satellite and gauge precipitation in China. Journal of Hydrology, 584, 124664.

[23] Ishida, K., Ercan, A., Nagasato, T., Kiyama, M., & Amagasaki, M. (2024). Use of one-dimensional CNN for input data size reduction in LSTM for improved computational efficiency and accuracy in hourly rainfall-runoff modeling. Journal of Environmental Management, 359, 120931.

[24] Gavahi, K., Foroumandi, E., & Moradkhani, H. (2023). A deep learning-based framework for multi-source precipitation fusion. Remote Sensing of Environment, 295, 113723.

[25] Mousavimehr, SM., & Kavianpour, MR. (2025). A non-stationary downscaling and gap-filling approach for GRACE/GRACE-FO data under climatic and anthropogenic influences. Applied Water Science, 15, 91.

[26] Alizadeh, MJ., & Nourani, V. (2024). Multivariate GRU and LSTM models for wave forecasting and hindcasting in the southern Caspian Sea. Ocean Engineering, 298, 117193.

[27] Ko, J., Lee, K., Hwang, H., Oh, S. G., Son, S. W., & Shin, K. (2022). Effective training strategies for deep-learning-based precipitation nowcasting and estimation. Computers & Geosciences, 161, 105072.