Supervised Learning Approaches to Classify Stratospheric Warming Events

Blume, Christian, Matthes, Katja and Horenko, I. (2012) Supervised Learning Approaches to Classify Stratospheric Warming Events Journal of the Atmospheric Sciences, 69 . pp. 1824-1840. DOI 10.1175/JAS-D-11-0194.1.

[img]
Preview
Text
jas-d-11-0194.1.pdf - Published Version

Download (1204Kb) | Preview

Supplementary data:

Abstract

Sudden stratospheric warmings are prominent examples of dynamical wave–mean flow interactions in the Arctic stratosphere during Northern Hemisphere winter. They are characterized by a strong temperature increase on time scales of a few days and a strongly disturbed stratospheric vortex. This work investigates a wide class of supervised learning methods with respect to their ability to classify stratospheric warmings, using temperature anomalies from the Arctic stratosphere and atmospheric forcings such as ENSO, the quasi-biennial oscillation (QBO), and the solar cycle. It is demonstrated that one representative of the supervised learning methods family, namely nonlinear neural networks, is able to reliably classify stratospheric warmings. Within this framework, one can estimate temporal onset, duration, and intensity of stratospheric warming events independently of a particular pressure level. In contrast to classification methods based on the zonal-mean zonal wind, the approach herein distinguishes major, minor, and final warmings. Instead of a binary measure, it provides continuous conditional probabilities for each warming event representing the amount of deviation from an undisturbed polar vortex. Additionally, the statistical importance of the atmospheric factors is estimated. It is shown how marginalized probability distributions can give insights into the interrelationships between external factors. This approach is applied to 40-yr and interim ECMWF (ERA-40/ERA-Interim) and NCEP–NCAR reanalysis data for the period from 1958 through 2010.

Document Type: Article
Keywords: Stratosphere, Classification, Neural networks, Pattern detection, Statistical techniques, Time series; Meteorology
Research affiliation: OceanRep > GEOMAR > FB1 Ocean Circulation and Climate Dynamics > FB1-ME Maritime Meteorology
OceanRep > The Future Ocean - Cluster of Excellence
Refereed: Yes
DOI etc.: 10.1175/JAS-D-11-0194.1
ISSN: 0022-4928
Projects: Future Ocean
Date Deposited: 20 Feb 2012 13:53
Last Modified: 30 Aug 2013 09:24
URI: http://eprints.uni-kiel.de/id/eprint/13743

Actions (login required)

View Item View Item

Document Downloads

More statistics for this item...