Deep Learning-Based Forecasting of Significant Wave Height in Padang Coastal Waters: A Comparative Study of RNN, LSTM, and GRU Architectures
DOI:
https://doi.org/10.24191/jcrinn.v11i1.587Keywords:
Significant Wave Height, Deep Learning, Recurrent Neural Network, Long Short-Term Memory, Gated Recurrent Unit, Time Series ForecastingAbstract
This study investigates deep learning–based forecasting of significant wave height (SWH) in Padang coastal waters, western Sumatra, a region exposed to wave-induced hazards that affect navigation, fisheries, and coastal infrastructure. SWH data are obtained from the ERA5 reanalysis single-levels product using the variable significant_height_of_combined_wind_waves_and_swell at hourly resolution from January to November 2025. After conversion from NetCDF to CSV, the time series is cleaned, normalized using Min-Max scaling, and transformed into supervised samples through a sliding-window approach. The dataset is split chronologically into training (January–September), validation (October), and testing (November) subsets. Three recurrent neural architectures are compared under a consistent experimental setup: a vanilla Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM), and Gated Recurrent Unit (GRU). All models use the same input window length, optimization settings, and early-stopping criterion, and are evaluated with Mean Squared Error (MSE), Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), and the coefficient of determination . Results show that all architectures capture SWH dynamics with very high skill, yielding test values above 0.997 and MAPE below 0.7%. The RNN achieves the lowest test RMSE (0.007551) and MAE (0.003894), while the GRU delivers slightly higher errors but a modeling compromise between accuracy and training time. The LSTM attains the largest error among the three models (RMSE 0.008308, MAE 0.004423) yet is the most computationally efficient. Residual analyses indicate that the largest errors occur during sharp transitions and high-energy events. Overall, the study demonstrates that recurrent deep learning models driven solely by ERA5 reanalysis can provide accurate short-term SWH forecasts to support coastal and maritime decision-making in Padang and other data-sparse regions.
Downloads
References
Alqushaibi, A., Abdulkadir, S. J., Rais, H., & Al-tashi, Q. (2021). Enhanced weight-optimized recurrent neural networks based on sine cosine algorithm for wave height prediction. Marine Science and Engineering, 9(524). https://doi.org/10.3390/jmse9050524
Arleth, S., Ludeña, R., Collazos, S. Z., Antonio, J., & Chavez, C. (2024). Model for reducing mean absolute percentage error through smoothing and time series forecasting in a tourism SME : A case study. Journal of Machine Intelligence and Data Science (JMIDS), 5, 109–116. https://doi.org/10.11159/jmids.2024.012
Bruno, M. F., Molfetta, M. G., Totaro, V., & Mossa, M. (2020). Performance assessment of ERA5 wave data in a swell dominated region. Journal of Marine Science and Engineering, 2–19. https://doi.org/10.3390/jmse8030214
Campos-Caba, R., Alessandri, J., Camus, P., Mazzino, A., Ferrari, F., Federico, I., Vousdoukas, M., Tondello, M., & Mentaschi, L. (2024). Assessing storm surge model performance: What error indicators can measure the model’s skill? Ocean Science, 20(6), 1513–1526. https://doi.org/10.5194/os-20-1513-2024
Chan, F. R., Annas, F., Yuspita, Y. E., & Darmawati, G. (2024). Implementation of Convolutional Neural Networks ( CNN ) in an emotion detection system for measuring learning concentration levels. Knowbase: International Journal of Knowledge in Database, 04(01), 49–61. https://doi.org/10.0.121.7/knowbase.v4i1.8429
Eluri, P. R. R., Annapurnaiah, K., Uday Bhaskar, T. V. S., Kiran Kumar, N., Padmanabham, J., Karthik, T. N. C., Vighneshwar, S. P., Reddem, V. S., Venugopala Rao, V., Murali Krishna, A., Ganesh, M. V, Lotliker, A., Nagaraja Kumar, M., Francis, P. A., Joseph, S., Balakrishnan Nair, T. M., & Srinivasa Kumar, T. (2025). Empowering the Blue Economy: The role of ICT in ocean observations, data, information and advisory services. CSI Transactions on ICT, 13(1), 3–25. https://doi.org/10.1007/s40012-025-00407-x
Fang, H., & Li, T. (2025). Comparative analysis of machine / deep learning models for single-step and multi-step forecasting in river water quality time series. Water, 17(13), 1–14. https://doi.org/10.3390/w17131866
Feng, Z., Hu, P., Li, S., & Mo, D. (2022). Prediction of significant wave height in Offshore China based on the machine learning method. Marine Science and Engineering, 10(836). https://doi.org/10.3390/jmse10060836
Haditiar, Y., Ikhwan, M., Nanda, M., & Haridhi, H. A. (2024). Understanding sea wave height conditions in sumatra waters. BIO Web of Conferences, 02014. https://doi.org/10.1051/bioconf/20248702014
Han, Y., Zhao, R., Wu, F., Yan, J., & Dong, C. (2025). A two channel optimized SWH deep learning forecast model coupled with dimensionality reduction scheme and attention mechanism. Ocean Engineering, 330, 121217. https://doi.org/10.1016/j.oceaneng.2025.121217
Hao, P., Li, S., & Gao, Y. (2023). Significant wave height prediction based on deep learning in the South China Sea. Frontiers in Marine Science, 9(February), 1–12. https://doi.org/10.3389/fmars.2022.1113788
Ji, Q., Han, L., Jiang, L., Zhang, Y., Xie, M., & Liu, Y. (2023). Short-term prediction of the significant wave height and average wave period based on the variational mode decomposition – temporal convolutional network – long short-term memory ( VMD – TCN – LSTM ) algorithm. Ocean Science, 19, 1561–1578. https://doi.org/10.5194/os-19-1561-2023
Lee, H., & Ahn, Y. (2025). Comparative study of RNN-based deep learning models for practical 6-DOF ship motion prediction. Journal of Marine Science and Engineering, 13(9), 1–35. https://doi.org/10.3390/jmse13091792
Li, G., Gao, Y., & Yang, H. (2025). Multi-step significant wave height prediction model based on feature enhancement compression , mode decomposition , multi-path convolutional recurrent network and regression correction. Measurement, 256(PD), 118401. https://doi.org/10.1016/j.measurement.2025.118401
Ma, J., Song, N., Nie, J., Ye, M., Liu, X., Yuan, Y., & Wei, Z. (2025). MPST : Effective significant wave height prediction method based on multiscale physical space ‑ time ‑ frequency fusion. Intelligent Marine Technology and Systems. https://doi.org/10.1007/s44295-025-00080-5
Maria, E., Wahyono, T., Hartomo, K. D., & Arthur, C. (2025). BiLSTM OptiFlow : An enhanced LSTM model for cooperative financial health forecasting. Bulletin of Electrical Engineering and Informatics, 14(3), 2004–2016. https://doi.org/10.11591/eei.v14i3.8653
Minuzzi, F. C., & Farina, L. (2023). A deep learning approach to predict significant wave height using long short-term memory. Ocean Modelling, 181, 102151. https://doi.org/10.1016/j.ocemod.2022.102151
Munyao, J. N., Oluoch, L. A., Iftikhar, H., & Rodrigues, P. C. (2025). Recurrent Neural Networks for hierarchical Time Series Forecasting: An application to the S&P 500 market value. Physica A: Statistical Mechanics and Its Applications, 678, 130869. https://doi.org/10.1016/j.physa.2025.130869
Omar, M., Yakub, F., Abdullah, S. S., Rahim, M. S. A., Zuhairi, A. H., & Govindan, N. (2024). One-step vs horizon-step training strategies for multi-step traffic flow forecasting with direct particle swarm optimization grid search support vector regression and long short-term memory. Expert Systems with Applications, 252, 124154. https://doi.org/10.1016/j.eswa.2024.124154
Paul, A., Maheswaran, P. A., Satheesan, K., & Kottayil, A. (2025). Assessing ocean reanalysis accuracy for marine extremes in the Indian ocean using in-situ observations. Climate Dynamics, 63(9), 343. https://doi.org/10.1007/s00382-025-07780-y
Sathe, A. M., Supraja, R., & Thomas, A. A. (2025). Results in engineering enhancing agricultural sustainability: Time Series Forecasting with ICEEMDAN-VMD-GRU for economic-resilience. Results in Engineering, 27(June), 106423. https://doi.org/10.1016/j.rineng.2025.106423
Singh, L. K., Garg, H., & Khanna, M. (2023). An artificial intelligence-based smart system for early glaucoma recognition using OCT images. International Journal of E-Health and Medical Communications (IJEHMC), 12(4), 32-59. https://doi.org/10.4018/IJEHMC.20210701.oa3
Unlu, A. (2025). Comparative analysis of hybrid deep learning models for electricity load forecasting during extreme weather. Energies, 18(12), 1–27. https://doi.org/10.3390/en18123068
Yadav, H., Padavale, V., & Parate, H. (2025). Early natural disaster prediction using machine learning : A comprehensive review. International Journal of Scientific Research in Engineering and Management (IJSREM), 09(04), 1–14. https://doi.org/10.55041/IJSREM44416
Yao, R., Shao, W., Hu, Y., Xu, H., & Zou, Q. (2025). Numeric modeling of sea surface wave using WAVEWATCH-III and SWAN during tropical cyclones : An overview. Journal of Marine Science and Engineering, 13(8), 1–34. https://doi.org/10.3390/ jmse13081450
Yunita, A., Iqbal, M. H. D., Zaki, M., Ramadhan, H., Akashah, E., Akhir, P., Besse, A., Mansur, F., & Hoirul, A. (2025). MethodsX performance analysis of Neural Network architectures for Time Series Forecasting : A comparative study of RNN, LSTM, GRU, and hybrid models. MethodsX, 15, 103462. https://doi.org/10.1016/j.mex.2025.103462
Zelios, V., Mastorocostas, P., Kandilogiannakis, G., Kesidis, A., Tselenti, P., & Voulodimos, A. (2025). Short-Term electric load forecasting using deep learning : A case study in Greece with RNN, LSTM, and GRU Networks. Electronics, 14(14), 1–24. https://doi.org/10.3390/ electronics14142820
Zhou, J., Liu, Z., Li, C., Du, K., & Yang, H. (2025). Cutting-edge approaches to specific energy prediction in TBM disc cutters: Integrating COSSA-RF model with three interpretative techniques. Underground Space, 22, 241–262. https://doi.org/10.1016/j.undsp.2024.11.004
Zhou, S., Bethel, B. J., Sun, W., Zhao, Y., Xie, W., & Dong, C. (2021). Improving significant wave height forecasts using a joint empirical mode decomposition – Long Short-Term Memory network. Journal of Marine Science and Engineering, 9(744). https://doi.org/10.3390/jmse9070744
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Fajri Rinaldi Chan, Rahma Yanti, Aldi Hidayat, Agung Ramadhanu, Firdaus Annas (Author)
This work is licensed under a Creative Commons Attribution 4.0 International License.
