Events Calendar

MISU Seminar | Sebastian Scher

Predicting weather forecast uncertainty with machine learning

by Sebastian Scher
PhD student at the Department of Meteorology (MISU)

Time: 23 October 2018, 11h15–12h15
Venue: Rossbysalen C609, Arrhenius Laboratory, 6th floor

Abstract
Weather forecasts are inherently uncertain. Therefore, for many applications, weather forecasts are only considered valuable if an uncertainty estimate can be assigned to them. Currently, the only widely used method to provide a confidence estimate for individual forecasts are ensemble weather forecast, which are computationally very expensive. I present an alternative machine-learning approach that can predict the uncertainty of a weather forecast given the  large-scale weather situation at initialisation. The method is based on deep learning with artificial convolutional neural networks and is trained on past weather forecasts. Given a new weather situation, it assigns a scalar value of confidence to medium range forecasts initialized from said weather situation, indicating whether the predictability is higher or lower than usual for the time of the year.

Welcome!

Back

JEvents v2.2.3   Copyright © 2006-2012

Bolin Centre for Climate Research
A collaboration between Stockholm University, KTH and the Swedish Meteorological and Hydrological Institute | Web administrator This email address is being protected from spambots. You need JavaScript enabled to view it.