A didactic approach to the Machine Learning application to weather forecast

TL;DR

This paper presents a didactic approach to applying machine learning for weather forecasting in complex volcanic regions, emphasizing the importance of local topography and proposing a simple thermodynamic model alongside ML techniques.

Contribution

It introduces a combined thermodynamic and machine learning methodology tailored for weather prediction in challenging orographic areas, with a focus on educational clarity.

Findings

Machine learning accurately predicts common dry conditions.

Model performance can improve with more data sources.

Thermodynamic approach aids understanding of local weather phenomena.

Abstract

We propose a didactic approach to use the Machine Learning protocol in order to perform weather forecast. This study is motivated by the possibility to apply this method to predict weather conditions in proximity of the Etna and Stromboli volcanic areas, located in Sicily (south Italy). Here the complex orography may significantly influence the weather conditions due to Stau and Foehn effects, with possible impact on the air traffic of the nearby Catania and Reggio Calabria airports. We first introduce a simple thermodynamic approach, suited to provide information on temperature and pressure when the Stau and Foehn effect takes place. In order to gain information to the rainfall accumulation, the Machine Learning approach is presented: according to this protocol, the model is able to ``learn'' from a set of input data which are the meteorological conditions (in our case dry, light rain, moderate rain and heavy rain) associated to the rainfall, measured in mm. We observe that, since in the input dataset provided by the Salina weather station the dry condition was the most common, the algorithm is very accurate in predicting it. Further improvements can be obtained by increasing the number of considered weather stations and time interval.