Science news

FMI's researchers publish about 350 peer-reviewed articles annually.

In Science News we publish current information about FMI's studies on the weather, the sea and the climate.

Machine learning methods improve accuracy of seasonal weather forecasts

Machine learning methods improve accuracy of seasonal weather forecasts

A study made use of novel statistical machine learning applications in order to make three-month temperature forecasts.

The results showed that predictability in Scandinavia was surprisingly good and even better than previously thought, especially as regards the summer and autumn seasons.

A new finding that emerged from the study pointed to an improvement in the accuracy of three-month mean temperature forecasts when the machine learning model utilised forecast signals from further away in the past as input data

Further infomation:

Researcher Matti Kämäräinen, Finnish Meterorological Institute, tel. +358 50 380 2868, matti.kamarainen@fmi.fi
Head of group Antti Mäkelä, Finnish Meterorological Institute, tel. +358 50 301 1988, antti.makela@fmi.fi

Kämäräinen, M., P. Uotila, A. Y. Karpechko, O. Hyvärinen, I. Lehtonen, and J. Räisänen, 2019: Statistical Learning Methods as a Basis for Skillful Seasonal Temperature Forecasts in Europe. J. Clim., 32, 5363–5379, https://doi.org/10.1175/JCLI-D-18-0765.1


Science news archive