Modélisation et prévision mensuelle de la salinité par réseaux de neurones de type perceptron dans la partie centrale du canal du Mozambique

Audiat Miller Miller, Jean Eugène RANDRIANANTENAINA, Jacques Chrysologue RATSIMAVO, Pierre Ruphin FATIANY, Sahoby LALAOHARISOA, Adolphe A. RATIARISON

Abstract


This study presents a hybrid methodology combining multiple linear regression and Multilayer Perceptron (MLP) for monthly forecasting of surface salinity in the Mozambique Channel. Applied to data from the region (1980-2018), the approach identifies precipitation as the dominant forcing factor in the regional water balance. The final neural model, optimized according to a 4-layer hidden architecture, shows a correlation of R=0.67 and an RMSE of 0.060 PSU with observations, demonstrating its quantitative accuracy. Projections for 2019-2028 suggest a trend toward increasing salinity in the channel. This method provides a robust framework for monitoring and forecasting hydro-climatic variability in this strategic sea lane.

Keywords : Surface salinity, Mozambique Channel, Hybrid model, Neural network, Linear regression, Ocean forecasting.


Keywords


Salinité de surface, Canal du Mozambique, Modèle hybride, Réseau de neurones, Régression linéaire, Prévision océanique

Full Text:

PDF

References


Bishop, 2006. « PATTERN RECOGNITION and MACHINE LEARNING »

Chatfield, C. (2004). The Analysis of Time Series: An Introduction. Chapman & Hall.

Coulibaly, P., Anctil, F., & Bobée, B. (2000). Daily reservoir inflow forecasting using artificial neural networks. Journal of Hydrology, 230, 244–257.

Cybenko, 1989 « Approximation par superpositions d'une fonction sigmoïde »

Emery, W. J., & Thomson, R. E. (2001). Data Analysis Methods in Physical Oceanography. Elsevier.

Glorot et al, 2011 « Réseaux neuronaux à redresseur clairsemé profond »

Hornik et al, 1989 “Multilayer feedforward networks are universal approximators”

R. R. (2014). Root mean square error (RMSE) or mean absolute error (MAE) ? Geoscientific Model Development, 7, 1247–1250.

Talley, L. D., Pickard, G. L., Emery, W. J., & Swift, J. H. (2011). Descriptive Physical Oceanography (6ᵉ éd.). Academic Press.

Tomczak, M., & Godfrey, J. S. (2003). Regional Oceanography: An Introduction. Pergamon.

UNESCO (1981). Background papers and supporting data on the Practical Salinity Scale 1978.




DOI: http://dx.doi.org/10.52155/ijpsat.v56.2.7922

Refbacks

  • There are currently no refbacks.


Copyright (c) 2026 Audiat Miller Miller

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.