Tellus Series A, published online 28 October 2009; DOI: 10.1111/j.1600-0870.2009.00421.x
Large-scale atmospheric circulation changes are associated with the recent loss of Arctic sea ice
Abstract
Recent loss of summer sea ice in the Arctic is directly connected to shifts in northern wind patterns in the following autumn, which has the potential of altering the heat budget at the cold end of the global heat engine. With continuing loss of summer sea ice to less than 20% of its climatological mean over the next decades, we anticipate increased modification of atmospheric circulation patterns. While a shift to a more meridional atmospheric climate pattern, the Arctic Dipole (AD), over the last decade contributed to recent reductions in summer Arctic sea ice extent, the increase in late summer open water area is, in turn, directly contributing to a modification of large scale atmospheric circulation patterns through the additional heat stored in the Arctic Ocean and released to the atmosphere during the autumn season. Extensive regions in the Arctic during late autumn beginning in 2002 have surface air temperature anomalies of greater than 3 °C and temperature anomalies above 850 hPa of 1 °C. These temperatures contribute to an increase in the 1000–500 hPa thickness field in every recent year with reduced sea ice cover. While gradients in this thickness field can be considered a baroclinic contribution to the flow field from loss of sea ice, atmospheric circulation also has a more variable barotropic contribution. Thus, reduction in sea ice has a direct connection to increased thickness fields in every year, but not necessarily to the sea level pressure (SLP) fields. Compositing wind fields for late autumn 2002–2008 helps to highlight the baroclinic contribution; for the years with diminished sea ice cover there were composite anomalous tropospheric easterly winds of ∼1.4 m s–1, relative to climatological easterly winds near the surface and upper tropospheric westerlies of ∼3 m s–1. Loss of summer sea ice is supported by decadal shifts in atmospheric climate patterns. A persistent positive Arctic Oscillation pattern in late autumn (OND) during 1988–1994 and in winter (JFM) during 1989–1997 shifted to more interannual variability in the following years. An anomalous meridional wind pattern with high SLP on the North American side of the Arctic—the AD pattern, shifted from primarily small interannual variability to a persistent phase during spring (AMJ) beginning in 1997 (except for 2006) and extending to summer (JAS) beginning in 2005.
1. Introduction
In previous papers we discuss that recent Arctic surface air temperature (SAT), sea level pressure (SLP) and summer sea ice loss often have decidedly different spatial patterns at the beginning of the 21st century compared to most of the 20th century, an Arctic Paradox and suggest calling this recent interval the Arctic Warm period (Overland and Wang, 2005; Overland et al., 2008). In the past we have followed the conceptual model of Quadrelli and Wallace (2004) where, based on Empirical Orthogonal Function (EOF)/Principal Component Analysis of SLP north of 20°N, the main patterns of winter climatic variability are the Arctic Oscillation (AO) and a near approximation to the Pacific North American pattern, labeled PNA*. A recent Arctic climate pattern is the 3rd EOF mode of SLP north of 20°N, and has been referred to as the third Arctic pattern, a Dipole pattern, the Arctic Warm pattern, the Barents Oscillation or the meridional pattern, as it contrasts to the more zonal pattern of the AO (Skeie, 2000; Overland and Wang, 2005; Zhang et al., 2008). Several authors relate the recent major reduction in summer Arctic sea ice extent in part to the presence and phasing of this meridional atmospheric climate pattern, which resulted in a much faster than the expected loss of sea ice in the real world compared to the timing of ensemble mean projections found in climate models based on anthropogenic forcing alone (e.g. Wu et al., 2006; Maslanik et al., 2007; Stroeve et al., 2007; Holland et al., 2008; L’Heureux et al., 2008; Wang et al., 2009). Current usage supports calling this pattern the Arctic Dipole (AD).
Most authors including the present ones considered that the persistent AO and AD patterns were mainly representative of natural variability of the chaotic climate system in the northern latitudes. However, due to recent sea ice loss at the end of summer there is a direct feedback to shifts in the broader atmospheric circulation in late autumn and winter. Heat being stored in the upper Arctic Ocean due to reduction in the area of Arctic summer sea ice is given back to the atmosphere in the following autumn, which has a direct impact on the temperature of the troposphere (Schweiger et al., 2008; Serreze et al., 2009) and thus on geopotential height and thickness fields.Chapman and Walsh (2007) based on Intergovernmental Panel on Climate Change (IPCC) models, Singarayer et al. (2006) based on the Hadley Center climate model, and Francis et al. (2009) based on the NCEP/NCAR Reanalysis (NNR) suggest that regional loss of sea ice can have hemispheric consequences in atmospheric circulation. Sokolova et al. (2007), Seierstad and Bader (2008) and Honda et al. (2009) show a relation in models between years with minimum sea ice cover and the negative phase of the AO (weaker zonal wind), although regional details are complicated by storm track and atmospheric long-wave/low-frequency dynamic processes. Thus model results and recent atmospheric conditions suggest that the continued reduction of sea ice in summer and other Arctic climate changes are not simply be due to natural chaotic and anthropogenic processes alone, but include an important feedback component to the atmospheric circulation mediated by sea ice loss and forcing from the surface.
We investigate recent observations for evidence of variations in the frequency of climate patterns in the Arctic and for potential atmospheric impacts from reductions in summer Arctic sea ice cover. In the next section we evaluate the persistence and seasonality of the AO and AD patterns more completely than has been done previously. In Section 3, the main part of the paper, we document the late autumn (OND) tropospheric temperature and dynamic impact through the geopotential height and 500–1000 hPa thickness fields associated with reduced summer sea ice cover during recent years in the western (Pacific side) Arctic.
2. AO and AD climatology
We make use of a limited area EOF analysis to compare seasonal time series of the more zonal AO pattern with the more meridional AD pattern. To put this regional analysis into physical context, we begin by showing the standard first three EOF patterns of individual months from the extended winter season (NDJFM) for SLP anomaly fields over nearly the entire northern hemisphere (north of 20° N) from 1959 to 1998, based on the NNR (Fig. 1 left-hand panel). The first three patterns are the AO and the PNA* as discussed by Quadrelli and Wallace (2004) and the more meridional ‘third’ pattern across the central Arctic discussed by Overland and Wang (2005). Sign conventions are arbitrary, but the positive phase of the AO is considered to be associated with a negative SLP anomaly over the central Arctic and the third or AD pattern has a dipole field with the positive phase associated with a negative SLP anomaly on the North American side of the Arctic.
More at link below.
Link to free, open-access article: http://onlinelibrary.wiley.com/doi/10.1111/j.1600-0870.2009.00421.x/full
No comments:
Post a Comment