Modeling the evolution of temperature in the Boeny region using the convolutional neural network method

RABENIAINA Anjara Davio Ulrick

¹ Laboratory of Atmosphere, Climate and Ocean Dynamics, University of Antananarivo, Science and Technology, Antananarivo, Madagascar.

² Terre, Ruralite et Eau, Mahajanga, Madagascar

Email: [email protected]; [email protected]

*Corresponding Author: RABENIAINA Anjara Davio Ulrick, Email: [email protected]

This article focuses on the modeling and analysis of temperature evolution in the Boeny region between 1979 and 2024. To achieve this objective, we first performed an annual temperature analysis, followed by a break test to identify significant changes in the trend. The anomaly was then used to examine deviations from the annual averages. Finally, a prediction model based on a convolutional neural network was developed. The study showed that the temperature in the Boeny region has followed a general upward trend since 1979. Over the past 25 years, positive anomalies have occurred consecutively, with a major shift observed in 2022, corresponding to an increase of 1.3°C. The model thus created allowed for the prediction of a continuous temperature rise over the next five years.

Artificial Intelligence, Temperature, Anomaly, Modeling, MAPE

RABENIAINA Anjara Davio Ulrick, RAMAHEFARISON Heriniaina (2025). Modeling the evolution of temperature in the Boeny region using the convolutional neural network method. Journal of Artificial Intelligence and Systems, 7, 1–10. https://doi.org/10.33969/AIS.2025070101.

[1] OMM (2025). 2024 est l’année la plus chaude jamais enregistrée. Organisation météorologique mondiale. https://wmo.int/fr

[2] Greenpeace (2024). 2024 : l’année la plus chaude jamais connue. https://www.greenpeace.fr

[3] TF1 Info (2025). Changement climatique : janvier 2025 bat un nouveau record. https://www.tf1info.fr

[4] Le Monde (2024a). Les nuits tropicales s’intensifient en France. https://www.lemonde.fr

[5] Killick, R., Fearnhead, P., & Eckley, I. A. (2012). Optimal detection of changepoints with a linear model. Journal of the American Statistical Association, 107(500), 1590-1598. https://doi.org/10.1080/01621459.2012.737745

[6] LeCun, Y., Bengio, Y., & Hinton, G. (1998). Gradient-based learning applied to document recognition. Proceedings of the IEEE, 86(11), 2278-2324. https://doi.org/10.1109/5.726791

[7] Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep learning. MIT Press

[8] He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. Proceedings of the IEEE conference on computer vision and pattern recognition (CVPR), 770-778. https://doi.org/10.1109/CVPR.2016.90.

[9] Kingma, D. P., & Ba, J. (2015). Adam: A method for stochastic optimization. Proceedings of the 3rd International Conference on Learning Representations (ICLR 2015).