A Machine Learning Approach to Climate Forecasting - Archive ouverte HAL Accéder directement au contenu
Poster De Conférence Année : 2023

A Machine Learning Approach to Climate Forecasting

Résumé

Climate change poses a major threat, with impacts including water scarcity, extreme weather, and rising sea levels. Machine learning offers a powerful approach for climate modeling and prediction to support policymaking. This research developed a machine learning model to investigate climate variables in Saudi Arabia. Historical data on temperature, precipitation, pressure, and wind from 1980-2015 was used to train a Random Forest model, which then predicted temperature from 2016-2020 based on the other variables. After preprocessing and compatibility checks, the model achieved a 2.69% mean-squared error, demonstrating its accuracy. The unifunctional model successfully uncovered interdependencies between variables. Next steps involve integrating it into a multifunctional tool with broader predictive capabilities. Overall, this work produced an accurate machine learning model for a key climate variable. With further development, such models can generate actionable insights and early warnings related to weather disasters, agriculture, air quality, sea level rise, and other impacts in Saudi Arabia and beyond.
Fichier principal
Vignette du fichier
final draft of poster.pdf (407.25 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04409406 , version 1 (22-01-2024)

Identifiants

  • HAL Id : hal-04409406 , version 1

Citer

Ahmed Almeslemani, Saad Aldoihi. A Machine Learning Approach to Climate Forecasting. GRI Enrichment Program, Aug 2023, Riyadh, Saudi Arabia. ⟨hal-04409406⟩
4 Consultations
4 Téléchargements

Partager

Gmail Facebook X LinkedIn More