Seasonal differences in the response of Arctic cyclones to climate change in CESM1

Clim Dyn DOI 10.1007/s00382-017-3767-x Seasonal differences in the response of Arctic cyclones to climate change in CESM1 Jonathan J. Day1 · Marika M. Holland2 · Kevin I. Hodges3 Received: 23 January 2017 / Accepted: 11 June 2017 © The Author(s) 2017. This article is an open access publication Abstract The dramatic warming of the Arctic over the last three decades has reduced both the thickness and extent of sea ice, opening opportunities for business in diverse sectors and increasing human exposure to meteorological hazards in the Arctic. It has been suggested that these changes in environmental conditions have led to an increase in extreme cyclones in the region, therefore increasing this hazard. In this study, we investigate the response of Arctic synoptic scale cyclones to climate change in a large initial value ensemble of future climate projections with the CESM1-CAM5 climate model (CESM-LE). We find that the response of Arctic cyclones in these simulations varies with season, with significant reductions in cyclone dynamic intensity across the Arctic basin in winter, but with contrasting increases in summer intensity within the region known as the Arctic Ocean cyclone maximum. There is also a significant reduction in winter cyclogenesis events within the Greenland–Iceland–Norwegian sea region. We conclude that these differences in the response of cyclone intensity and cyclogenesis, with season, appear to be closely linked to changes in surface temperature gradients in the high latitudes, with Arctic poleward temperature gradients increasing in summer, but decreasing in winter. Electronic supplementary material The online version of this article (doi:10.1007/s00382-017-3767-x) contains supplementary material, which is available to authorized users. * Jonathan J. Day 1 National Centre for Atmospheric Science, Department of Meteorology, University of Reading, Reading, UK 2 National Center for Atmospheric Research, Boulder, USA 3 Department of Meteorology, University of Reading, Reading, UK Keywords Arctic cyclones · Extreme weather · Polar climate · Climate change · Storm risk 1 Introduction Unprecedented warming in the Arctic has led to a dramatic reduction in both the extent and thickness of Arctic sea ice (Stroeve et al. 2011), opening up opportunities for business in diverse sectors such as fossil fuel and mineral extraction, shipping and tourism (Jung et al. 2016). Industrial activities in the Arctic are expected to be subject to high levels of investment over the coming decades (Emmerson and Lahn 2012). As a result, there has been an increase in the exposure of humans and infrastructure to environmental risks in the Arctic. Unlike the mid-latitude storm tracks of the North Atlantic and Pacific, which are most active during winter, in an area of the central Arctic, known as the Arctic Ocean cyclone maximum (AOCM) synoptic scale cyclones are most numerous during summer (Serreze 1995; Serreze and Barrett 2008) (see Figs. 1, 2). However, Arctic cyclones are most dynamically intense during winter (Zhang et al. 2004). The source region of Arctic cyclones also differs depending on the season, with summer cyclones largely originating over the Eurasian continent (Reed and Kunkel 1960; Crawford and Serreze 2016) and winter cyclones largely originate from the North Atlantic and North Pacific (Sorteberg and Walsh 2008; Simmonds et al. 2008) (Figure S1, S2 & S3). The dramatic warming of the Arctic over the last three decades has reduced both the thickness and area covered by summer sea ice, leaving Arctic waters navigable by shipping exactly during this period of seasonally enhanced cyclone activity in the AOCM region. Therefore, understanding changes in storminess in the region is 13 Vol.:(0123456789) J. J. Day et al. Fig. 1  Track density in ERA-Interim (1990–2005) (top row) and ensemble mean track density bias of historical simulations relative to ERA-Interim, for (left) DJF and (right) JJA. Units are in number of cyclones per month per unit area, where unit area is equivalent to a 5° spherical cap. In c, d, stippling shows where more than 80% of ensemble members have a bias of the same sign an important factor required to manage the risks associated with these storms. Investigations into Arctic cyclone trends using the National Center for Environmental Prediction (NCEP) atmospheric reanalyses (Kistler et al. 2001) have found a significant positive trend in the number of cyclones entering the Arctic from lower latitudes in spring and summer months, but little change in winter (Zhang et al. 2004; Sorteberg and Walsh 2008), however these studies also indicate significant inter-annual and longer timescale variability in these trends. Sepp and Jaagus (2011) confirmed that the number of cyclones entering the Arctic had increased, but found that the total number of cyclones inside the Arctic had not changed significantly, suggesting a reduction in local cyclogenesis. However, analysis of trends appears to be sensitive to the choice of reanalysis used. For example, Simmonds et al. (2008) found a statistically significant trend in the total number of summer cyclones in the Arctic in the NCEP reanalysis for the period (1948–2002) but not in ERA-40 (Uppala et al. 2005). To measure trends in maximum cyclone intensity, these studies have largely focussed on minimum mean sea level pressure (MSLP), and studies generally agree that there is a general reduction in this quantity. A recent intercomparison of cyclone tracking methods by Simmonds and Rudeva (2014) concluded that different methods generally agree on which Arctic cyclones are most intense over the ERA-Interim reanalysis period (1979–2009). However, Chang (2014) cautions that differences in the definitions used in cyclone identification algorithms can lead to differing conclusions when investigating the response of strong cyclones to climate change. The apparent temporal changes in cyclone statistics have occurred against a backdrop of changes in the mean climate. The warming in the Arctic is amplified with respect to the global mean, particularly outside of the summer months when solar energy absorbed during 13 Seasonal differences in the response of Arctic cyclones to climate change in CESM1 Fig. 2  Arctic seasonal mean cyclone frequency for ERAInterim (dots) and box plots showing quartiles of mean cyclone frequency for historical simulations (1990–2005) (left) and RCP8.5 simulations (2071– 2080) (right) (#/month). The bottom panel is as top except for the mean maximum wind speed achieved by cyclones within the Arctic summer is released back to the atmosphere (Rigor et al. 2000). This pattern is expected to continue into the future (Manabe and Stouffer 1980; Holland and Bitz 2003). The impact of these changes on both the time-mean and transient large scale atmospheric circulation is an active area of research (e.g. Shepherd 2016) and significant disagreements remain over the role of Arctic warming (...truncated)