MODELING ANNUAL PRECIPITATION AS A FUNCTION OF CLIMATIC VARIABLES BY FUZZY LOGIC

climatology

Currently, the use of artificial intelligence method for forecasting purposes is still lacking in precipitation forecasting research. In this paper, a fuzzy logic-based model is proposed for annual precipitation forecasting in the northwestern part of Madagascar. Several climatic variables were used to build the models. The method is based on the analysis of the correlation coefficient to evaluate the existing relationship between precipitation, temperature, sea surface temperature (SST), atmospheric pressure, specific humidities at 700 hPa and 850 hPa. The models are developed using the multiple linear regression method. Then, the fuzzy inference system is used for series prediction. The study area is subdivided into three rainfall homogeneous sub-zones. Thus, we obtained three different models. The performance of each of these models is evaluated by the value of the mean absolute deviation. All models showed excellent predictive performance. Finally, fuzzy models show an increase in precipitation in all three study sub-areas.

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