Deconstructing the climate change response of the Northern 1 Hemisphere wintertime storm tracks

There are large uncertainties in the circulation response of the atmosphere to climate change. 6 One manifestation of this is the substantial spread in projections for the extratropical storm tracks made 7 by different state-of-the-art climate models. In this study we perform a series of sensitivity experiments, 8 with the atmosphere component of a single climate model, in order to identify the causes of the differences 9 between storm track responses in different models. In particular, the Northern Hemisphere wintertime 10 storm tracks in the CMIP3 multi-model ensemble are considered. A number of potential physical drivers 11 of storm track change are identified and their influence on the storm tracks is assessed. The experimental 12 design aims to perturb the different physical drivers independently, by magnitudes representative of the 13 range of values present in the CMIP3 model runs, and this is achieved via perturbations to the sea 14 surface temperature and the sea-ice concentration forcing fields. We ask the question: can the spread of 15 B. J. Harvey and L. C. Shaffrey National Centre for Atmospheric Science, Department of Meteorology, University of Reading, Reading, UK. E-mail: [email protected] T. J. Woollings Atmospheric, Oceanic and Planetary Physics, Department of Physics, University of Oxford, Oxford, UK. 2 B. J. Harvey et al. projections for the extratropical storm tracks present in the CMIP3 models be accounted for in a simple 16 way by any of the identified drivers? The results suggest that, whilst the changes in the upper-tropospheric 17 equator-to-pole temperature difference have an influence on the storm track response to climate change, 18 the large spread of projections for the extratropical storm track present in the northern North Atlantic in 19 particular is more strongly associated with changes in the lower-tropospheric equator-to-pole temperature 20