Modelling Precipitation as a Function of Several Variables in Northern Madagascar Using the ANFIS Network

RABENIAINA Anjara Davio Ulrick

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

^a.[email protected]; ^b.[email protected]; ^c.[email protected]; ^d.[email protected] 

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

This article presents a study on the modeling of precipitation in northwest Madagascar as a function of several variables. The method is based on the use of the Granger test to assess the causal relationship between precipitation, temperature, atmospheric pressure, wind at 925 hPa and Outgoing Longwave Radiation (OLR). The model is developed using the multiple linear regression method, followed by the use of the Sobol index to identify the most influential of these variables. Forecasts are based on the ANFIS model. The results show that precipitation in this area (Sofia, Diana and Sava regions) is influenced by pressure, wind and OLR. The explanatory model obtained highlights the relationship between precipitation and these climatic variables, showing a 7 mm decrease in precipitation according to the ANFIS model developed.

Precipitation, Granger causality test, Modeling, Multiple linear regression, ANFIS

RABENIAINA Anjara Davio Ulrick, RAKOTOVAO Niry Arinavalona, Ambeondahy, Hary Jean (2024). Modelling Precipitation as a Function of Several Variables in Northern Madagascar Using the ANFIS Network. Journal of Artificial Intelligence and Systems, 6, 135–145. https://doi.org/10.33969/AIS.2024060110.

[1] Salim Ali : Études de la variabilité climatique et de l’impact des modes ENSO et IOD au Nord de Madagascar et aux Comores. 2018 

[2] Rajaonarisoa : Caractérisation de l’évolution et analyse de l’influence des paramètres climatiques température, humidité et pression atmosphérique sur la précipitation annuelle au sud de Madagascar. 2022.

[3] Rousson, Valentin : Régression linéaire multiple. In : Statistique appliquée aux

Sciences de la vie. Springer, 2013, S. 219–258

[4] KHABBAZ, Mohamed ; BENHAMOU, Brahim ; LIMAM, Karim ; HAMDI, Hassan ; BENNOUNA, Amin : ÉTUDE DE SENSIBILITÉ ET PARAMÉTRIQUE COMPLETE D’UN ÉCHANGEUR AIR-SOL POUR LE RAFRAICISSEMENT DE L’AIR SOUS LE CLIMAT CHAUD ET SEMI-ARIDE DE MARRAKECH. (2016)

[5] D.J. Cho, S. Kim, and K. Lee, "An adaptive-network-based fuzzy inference system," IEEE Transactions on Fuzzy Systems, vol. 6, no. 2, pp. 210-221, May 1998.

[6] Jang, J.-S. R. (1993). ANFIS: Adaptive-Network-Based Fuzzy Inference System. IEEE Transactions on Systems, Man, and Cybernetics, 23(3), 665-685.