Articles | Volume 20, issue 2
https://doi.org/10.5194/os-20-417-2024
https://doi.org/10.5194/os-20-417-2024
Research article
 | 
22 Mar 2024
Research article |  | 22 Mar 2024

Machine learning methods to predict sea surface temperature and marine heatwave occurrence: a case study of the Mediterranean Sea

Giulia Bonino, Giuliano Galimberti, Simona Masina, Ronan McAdam, and Emanuela Clementi

Data sets

ERA5 hourly data on single levels from 1940 to present H. Hersbach et al. https://doi.org/10.24381/cds.adbb2d47

ESA Sea Surface Temperature Climate Change Initiative (SST_cci): Level 4 Analysis Climate Data Record, version 2.1 S. A. Good et al. https://doi.org/10.5285/62c0f97b1eac4e0197a674870afe1ee6

Model code and software

Machine learning methods to predict Sea Surface Temperature and Marine Heatwave occurrence: a case study of the Mediterranean Sea Giulia Bonino https://doi.org/10.5281/zenodo.8335345

Download
Short summary
This study employs machine learning to predict marine heatwaves (MHWs) in the Mediterranean Sea. MHWs have far-reaching impacts on society and ecosystems. Using data from ESA and ECMWF, the research develops accurate prediction models for sea surface temperature (SST) and MHWs across the region. Notably, machine learning methods outperform existing forecasting systems, showing promise in early MHW predictions. The study also highlights the importance of solar radiation as a predictor of SST.