1. Introduction
The Köppen climate classification system (e.g., Rohli and Vega 2011) defines polar climates as those regions where the mean near-surface air temperature of the warmest month is below 10°C. Köppen further divided the polar climates into Polar Tundra (ET), where the warmest month was above 0°C, and Polar Ice Cap (EF), in which the warmest month was below 0°C. The upper limit of 10°C for the warmest month corresponds roughly with the poleward limit of tree growth. The warmest month limit of 0°C corresponds roughly with the equatorward limit of permanent snow and ice on land.
Shear (1964) suggested that the Polar Tundra climate be further divided into the Polar Marine (EM) climate, in which the mean temperature of the coldest month is above −6.7°C (20°F), and the remaining Polar Tundra (ET), in which the coldest month is below −6.7°C. Shear chose 20°F as the lower limit of coldest month in the EM climate to limit its occurrence to marine environments. Shear expected that the poleward margins of the EM climate would coincide with the maximum winter extent of pack ice, as the pack ice boundary is the seasonal projection of the pseudocontinent whose role in terms of energy exchange processes is more like snow-covered land than open water. He used data from about 20 weather stations on islands or continental coasts and various marine climate atlases from the early twentieth century to define the EM regions and noted that an absence of data in many polar regions made the location of boundaries difficult to determine.
Based on the data available in the early and mid-twentieth century, Shear (1964) identified three primary regions with EM climate: (a) a southwest-to-northeast trending region in the North Atlantic from south of Greenland through the Denmark Strait across much of Iceland and to the Barents Sea north of Norway and the Kola Peninsula, (b) the southern Bering Sea and Aleutian Islands in the North Pacific, and (c) a circumpolar zone over the Southern Ocean and sub-Antarctic islands between roughly 49° and 60°S (Fig. 1). Isolated areas of EM climate also exist in high-altitude mountain areas outside of the polar regions, including portions of New Zealand (Mark et al. 2000) and Hawaii (Noguchi et al. 1987).
The EM climate as depicted by Shear (1964) for the (a) North Atlantic, (b) North Pacific, and (c) Southern Ocean (reprinted by permission of Taylor & Francis Ltd).
Citation: Journal of Climate 26, 11; 10.1175/JCLI-D-12-00660.1
Other characteristics of the EM climate, as described by Shear, distinguish it from the continental ET climate. The EM climate is wetter than ET, has a winter precipitation maximum rather than a summer maximum, a larger proportion of the annual precipitation falls as rain, and there are more days of precipitation. The EM climate also has greater storm frequency, more cloud cover, and stronger winds than the ET climate.
The EM domains have experienced varying degrees of temperature increases over recent decades. Warming trends have been especially robust in the Arctic where observations have shown terrestrial Arctic (64°–70°N) annual temperature trends +0.38°C decade−1 from 1970 to 2008 (Chylek et al. 2009), while temperature trends over land north of 60°N have been estimated at nearly +0.64°C decade−1 from 1979 to 2008 (Bekryaev et al. 2010). Arctic Ocean SSTs have likewise risen over roughly similar periods, especially in the western Arctic during the last two decades (Steele et al. 2008). These ocean/atmosphere warming trends have coincided with declining terrestrial Arctic snow cover (Liston and Hiemstra 2011) and rapidly deteriorating sea ice extent (Maslanik et al. 2007) estimated at −4.1 ± 0.3% decade−1 over the period 1979–2010 (Cavalieri and Parkinson 2012). This has led to anomalous upward latent and sensible heat fluxes during late fall and winter and lower troposphere warming (Serreze et al. 2009; Kumar et al. 2010; Screen and Simmonds 2010; Porter et al. 2012), along with delayed refreeze and earlier melt onset (Markus et al. 2009) and decreased March maximum ice cover (Nghiem et al. 2007).
Temperature trends in the high southern latitudes are much more spatially and temporally variable. Chapman and Walsh (2007) found warming trends over 60°–90°S during all seasons from 1958–2002, especially winter (+0.17°C decade−1), most pronounced over the Antarctic Peninsula, while Steig et al. (2009) constructed a 50-yr climatology (1957–2006) that revealed significant annual warming (+0.18°C decade−1) over West Antarctica, largely focused on winter and spring seasons. Large-scale Southern Ocean SST increases (Levitus et al. 2000; Gille 2002) and rising air temperatures over some Southern Hemisphere EM areas (Quayle et al. 2002; Smith 2002) have also been observed. Despite oceanic heating, circum-Antarctic sea ice response has been nonlinear and regionally variable. Studies of Antarctic sea ice extent have shown a slight positive annual trend during the past few decades (Cavalieri et al. 2003) and increases of nearly 1% decade−1 from 1979 to 2006 (Cavalieri and Parkinson 2008). Parkinson and Cavalieri (2012) recently found that Southern Hemisphere sea ice extent has been increasing at a rate of +1.5 ± 0.4% decade−1 from 1979 to 2010, with notable gains in the Ross Sea outweighing well-documented losses in the Bellingshausen and Amundsen Seas (Vaughn et al. 2003; Schneider et al. 2012). Antarctic temperature variability has been partly tied to phases of the southern annular mode (SAM; Thompson and Solomon 2002) and ENSO (Bromwich et al. 2004), as have sea ice trends in the Southern Ocean (Stammerjohn et al. 2008).
These climatically sensitive EM areas are biologically rich and diverse, making their study important in the midst of an ever-changing climate system. The purpose of our study is to map and explore the modern EM climate using contemporary reanalyses to expand Shear’s original study. Recent decades have witnessed increases and advancements in observational networks, especially in terms of satellite-derived products, and climate models that have greatly benefitted polar climate research since Shear’s analysis. Global coupled atmosphere–ocean reanalyses optimally utilize a multitude of datasets from seemingly disparate sources, including weather stations, buoys, aircrafts, rawinsondes, and satellites in order to depict atmospheric and oceanic conditions over large, remote areas where direct observations are lacking. Our study employs current reanalysis products because they offer the best opportunity to map the modern EM climatology and its characteristics while assessing its dynamic nature with respect to time. Further sections of the paper are organized as follows. In section 2, data and methodology will be addressed. In section 3, we compare reanalyses, justify selection of one reanalysis for comparison with Shear’s original outputs and forthcoming plots, determine the poleward and equatorward EM boundaries, look at some variables that impact this climate regime, and examine EM interannual variability over the period of study. Section 4 will discuss differences from Shear’s study, examine potential EM ecological responses, and briefly address the possibility of future EM changes. Section 5 will briefly summarize our results and their implications.
2. Data and methodology
Global reanalyses use fixed numerical weather prediction models and data assimilation schemes to produce gridded fields over time periods suitable for climate research. Reanalyses are particularly useful over polar regions, providing a coherent representation of weather and climate where relatively short temporal spans of data records and areas of sparse observations exist. However, caution must be exercised when using reanalyses to study climate trends, as output is sensitive to changes of the observing system and how observations are processed (Bengtsson et al. 2004a,b; Sterl 2004; Thorne and Vose 2010; Screen and Simmonds 2011). Such changes result in erroneous trends, particularly over Southern Hemisphere polar regions (e.g., Hines et al. 2000; Marshall and Harangozo 2000; Marshall 2002), limiting the viability of reanalysis products in polar regions to the post-1978 modern satellite era (e.g., Bromwich and Fogt 2004; Renwick 2004; Trenberth et al. 2005; Bromwich et al. 2007; Serreze et al. 2009; Serreze and Barrett 2011). There are also substantial differences between reanalyses, based on different models and parameterizations, observations, and data assimilation systems (e.g., Bromwich and Fogt 2004; Bromwich et al. 2007; Walsh et al. 2009; Bromwich et al. 2011; Screen and Simmonds 2011). In this study we use three reanalyses, described below, for the 32-yr period from 1979 to 2010.
The European Centre for Medium-Range Weather Forecasts (ECMWF) Interim Re-Analysis (ERA-Interim; Dee et al. 2011) supersedes the 40-yr ECMWF Re-Analysis (ERA-40; Uppala et al. 2005) and improves upon ERA-40 in several regards (see Dee et al. 2011). ERA-Interim uses a 12-hourly four-dimensional variational data assimilation (4D-Var) system, as well as an automated satellite radiance variational bias correction scheme (Dee and Uppala 2009). The observational sources for polar regions are listed in Andersson (2007) and Dee et al. (2011). ERA-Interim features spectral T255 (~0.7°) horizontal resolution and 60 vertical levels. Output on a regular 512 × 256 0.7° Gaussian grid from the National Center for Atmospheric Research Data Support Section (NCAR DSS) is used in this study.
The National Centers for Environmental Prediction (NCEP) Climate Forecast System Reanalysis (CFSR; Saha et al. 2010) is a coupled atmosphere–ocean–land surface–sea ice model that supersedes the NCEP–Department of Energy Atmospheric Model Intercomparison Project 2 (AMIP2) reanalysis (Kanamitsu et al. 2002). CFSR uses a 3D-Var gridpoint statistical interpolation (GSI) data assimilation system (Kleist et al. 2009) and ingests a wide array of satellite observations in radiance form. The CFSR atmospheric component features spectral T382 (~0.31°) horizontal resolution with 64 vertical levels. Output on a 720 × 361 0.5° latitude–longitude grid from the NCAR DSS is used here.
The Japan Meteorological Agency (JMA) 25-yr reanalysis (JRA-25; Onogi et al. 2007) uses the JMA numerical weather prediction and data assimilation systems. JRA-25 uses a 3D-Var data assimilation scheme that ingests satellite radiances and features spectral T106 (~1.125°) horizontal resolution with 40 vertical levels. Output on a regular 320 × 160 1.125° Gaussian grid from the NCAR DSS is used in this study.
The previously described Polar Marine (EM) climate classification from Shear (1964) was applied to each grid point of monthly 2-m air temperature from the three reanalyses. The union of the regions of warmest month mean temperature greater than 0°C (32°F) but less than 10°C (50°F) and coldest month mean temperature greater than −6.7°C (20°F) represents EM climate. In addition to 2-m air temperature, precipitation, mean sea level pressure, sea surface temperature, sea ice fraction, 10-m wind, and total cloud fraction are also analyzed to provide a more complete description of EM climate. While all three reanalyses are used to characterize the spatial distribution of EM area, to simplify the analysis we solely use ERA-Interim for the detailed description of EM climate. ERA-Interim is the only reanalysis to use 4D-Var data assimilation, and it along with its predecessor ERA-40 compare favorably against other global reanalyses in both the Northern Hemisphere (e.g., Bromwich and Wang 2005; Bromwich et al. 2007; Walsh et al. 2009; Screen and Simmonds 2011; Serreze et al. 2012) and Southern Hemisphere (Bromwich and Fogt 2004; Monaghan et al. 2006; Bromwich et al. 2011; Hodges et al. 2011) high latitudes. The primary findings of this study are not critically dependent upon which reanalysis is used for detailed analysis.
3. Results
a. Intercomparison of the reanalyses
Figure 2a shows EM area for all three reanalyses over the Northern Hemisphere. The area east of Newfoundland extending south of Greenland, across Iceland, and over the Norwegian and Barents Seas matches well between all three reanalyses. The second area over the Bering Sea also shows general agreement, although the JRA-25 and CFSR areas are slightly larger than ERA-Interim. There are also scattered small EM regions along the southern Alaska and western Canadian coastlines. Additional small high-altitude midlatitude regions primarily show up in CFSR, likely due to the enhanced horizontal resolution compared to the other reanalyses. The two large-scale EM areas in Fig. 2a are along the primary high-latitude storm tracks (e.g., Hoskins and Hodges 2002), where warm and moist air is advected into these areas from the south. Notice the eastern offset of EM areas from the Canadian, Greenland, and Siberian coasts, where continental effects prevent establishment of EM climate until a marine influence dominates offshore. The latitudinal position of the southernmost North Atlantic EM (~55°N) almost matches the northernmost EM in the North Pacific, emphasizing the generally warmer environment in the North Atlantic compared to the North Pacific for equivalent latitudes.
(a) Overplots of EM area for the Northern Hemisphere (NH; 1979–2010) from ERA-Interim, CFSR, and JRA-25 reanalyses. (b) As in (a), but for the Southern Hemisphere (SH).
Citation: Journal of Climate 26, 11; 10.1175/JCLI-D-12-00660.1
Figure 2b shows EM area for the Southern Hemisphere from all three reanalyses. The Southern Hemisphere contains 90% of global EM climate area. EM area exists over much of the Southern Ocean, and extends farther equatorward (to ~50°S) in the Southern Hemisphere compared to the Northern Hemisphere. Differences between reanalyses are again small, with CFSR extending slightly farther south, ERA-Interim extending slightly farther north, and some discrepancies over Patagonia in southern Chile. CFSR also identifies EM over the mountains of southwestern New Zealand. The Southern Hemisphere EM area generally follows the Southern Hemisphere storm track, which dips poleward from the South Atlantic eastward to regions south of New Zealand and into the South Pacific. However, the EM area occurs on the northern edge of the primary circumpolar storm track (e.g., Simmonds et al. 2003; Hoskins and Hodges 2005), where equatorward incursions of Antarctic air masses allow for establishment of EM climate in otherwise marine environments (e.g., Parish and Bromwich 1998).
b. Assessment of the EM boundaries
The ERA-Interim EM areas are mapped in Figs. 3a and 3b along with the winter sea ice maximum, represented by the 15% ice concentration extent as the poleward boundary, and the warmest-month SST 10°C isotherm as the equatorward boundary. Figure 3a shows that the SST threshold values nearly match the southern EM boundary in the North Atlantic from approximately 30°W northeastward until the northwestern Norwegian coastline at about 15°E before ending just north of Scandinavia in the Norwegian Sea. The maximum sea ice extent fits the poleward EM boundary well just east of Greenland across the Denmark Strait and Greenland Sea, and represents an acceptable border west and southeast of Svalbard and southwest of Novaya Zemlya in the Barents Sea. The SST boundary in the North Pacific (Fig. 3a) is just south of the EM near the Aleutian Islands. These air–sea temperature contrasts could be the result of warm air advection from eastern Siberia onto the EM areas during the EM warmest month (August), which would likely make the EM and surrounding areas warmer than the 10°C SST isotherm to the south. In contrast, the poleward boundary roughly matches the northern extent of this region’s EM at 60°N with the fit improving westward across the Bering Sea.
(a) ERA-Interim EM area (green) from 1979–2010 with warmest month SST (red) and maximum winter 15% sea ice concentration (blue) for the NH. (b) As in (a), but for the SH.
Citation: Journal of Climate 26, 11; 10.1175/JCLI-D-12-00660.1
The Antarctic EM boundaries are also well represented by SST and maximum sea ice as shown in Fig. 3b. The SST boundary is slightly farther south in the South Atlantic between 30° and 60°W, but otherwise stays consistent between 40° and 60°S around the Southern Ocean as it mirrors the equatorward (northern) EM area. The maximum sea ice poleward (southern) boundary is generally within 1°–2° of the EM climatic extent roughly between 58° and 65°S around the Antarctic continent, confirming the assessment of Shear (1964). EM boundaries in both hemispheres are fairly well defined by the aforementioned SST and sea ice edge parameters, making the accurate monitoring of these variables critical to mapping the variability of the EM climate, which will be addressed in section 3d.
c. EM climatic characteristics
Shear (1964) argued that the Polar Marine (EM) climate had distinct characteristics compared to the continental Polar Tundra (ET) climate, which occurs roughly at the same latitudes as EM but over land. The mild winter and smaller annual temperature range were the chief distinguishing features upon which Shear defined the EM climate. In addition, he noted the EM climate has more precipitation than the ET, the EM climate has a tendency toward a winter maximum of precipitation instead of a summer maximum, and EM precipitation is dominantly rain. Shear noted that the EM climate has greater storm frequency, more cloud cover, and stronger winds than the continental ET. Our analysis supports Shear’s claim that the EM has more precipitation than ET (shown later in this section), and a late fall/early winter (November/February) precipitation maximum (90–100 mm month−1) is apparent over the Northern Hemisphere EM (Fig. 4a). This is not the case in the Southern Hemisphere where maximum precipitation (90–93 mm month−1) clearly occurs in austral autumn (April/May; Fig. 4b), although it broadly peaks during the coldest half of the year that includes winter. In agreement with Shear, Figs. 5a and 5b indicate that rain is the dominant form of precipitation over EM areas since snowfall typically comprises less than 40% of total annual precipitation with small exceptions along the poleward EM boundaries in the North Atlantic and circum-Antarctic where snowfall accounts for 50%–60% of total annual precipitation.
Time series of ERA-Interim EM (solid line) and ET (dashed line) monthly mean precipitation (1979–2010) for the (a) NH and (b) SH.
Citation: Journal of Climate 26, 11; 10.1175/JCLI-D-12-00660.1
ERA-Interim plots of the annual snow fraction (%) of total precipitation (1979–2010) for the (a) NH and (b) SH. The ERA-Interim EM area is outlined by solid black contours.
Citation: Journal of Climate 26, 11; 10.1175/JCLI-D-12-00660.1
Additional characteristics of the EM climate are also explored using ERA-Interim data including patterns of annual mean sea level pressure, wind speed, precipitation, and cloud over the Northern Hemisphere (Figs. 6a–d) and Southern Hemisphere (Figs. 7a–d). The Polar Marine climates of the North Atlantic and North Pacific oceans are clearly regions of low mean sea level pressure, indicating prevailing tracks of cyclonic storms (Fig. 6a). Strong air–sea temperature differences over the oceanic EM regions produce higher storm frequencies relative to the northern continental ET climates, and coincide with greater precipitation, wind speed, and cloud cover. Wind speeds are higher in the EM climates than in the continental ET climates; however, even greater wind speeds occur south of the EM climates in the Atlantic and Pacific (Fig. 6b). The EM climate region northeast of Iceland displays a regional wind speed minimum. Annual precipitation is greater in the EM climates than in the continental ET climates (Fig. 6c). Precipitation is greater in the Atlantic EM climate (600–900 mm east of Iceland, 900–1200 mm west of Iceland) than in the Pacific EM climate (600–900 mm) because of orographically induced precipitation near the southern tip of Greenland and the unique path of weather systems in the region. In both ocean basins the precipitation increases southward from the EM climates. Cloud cover is 80%–90% in the EM climates, higher than over the adjacent continents (Fig. 6d).
ERA-Interim annual mean (1979–2010) (a) mean sea level pressure (MSLP; hPa), (b) 10-m wind speed (m s−1), (c) precipitation (mm yr−1), and (d) cloud cover (%) for the NH. The ERA-Interim EM area is outlined by solid black contours.
Citation: Journal of Climate 26, 11; 10.1175/JCLI-D-12-00660.1
As in Fig. 6, but for the SH. MSLP is not plotted at elevations over 1000 m over Antarctica.
Citation: Journal of Climate 26, 11; 10.1175/JCLI-D-12-00660.1
In the Southern Hemisphere, the circumpolar trough of low pressure is on the poleward margins of the EM climate region (Fig. 7a). There is scant area of ET climate in the Southern Hemisphere for comparison. The southern storm track appears to be displaced southward rather than situated in the core of the EM climate as occurs in the NH EM climates. Mean wind speeds reach a maximum through the core of the southern EM climate associated with the circumpolar westerlies (Fig. 7b) and are stronger than in the northern EM climates. Wind speeds are strongest in the Eastern Hemisphere and reach a peak between 50° and 100°E. Precipitation is 600–1200 mm in the southern EM climate and decreases toward the pole (Fig. 7c). The regions of greatest cloud cover (>90%), similar to mean sea level pressure, are on the poleward margins of the southern EM climate or even south of the EM climate (Fig. 7d). The southern EM climate has 80%–90% cloud cover, similar to the northern EM climates.
As expected and as predicted by Shear (1964), the EM climates are stormy, windy, wet, and cloudy. Where comparisons can be made in the Northern Hemisphere, these EM climates are distinct from their continental ET counterparts in all of these parameters.
d. Interannual variability
Figure 8a shows the general downward trend of the total (NH + SH) EM area (−42.4 × 109 m2 yr−1 (>95% significance according to a two-tailed Student’s t test) from 1979 to 2010. However, the Northern and Southern Hemisphere EM areas have remarkably contrasting trends. The Northern Hemisphere EM area (Fig. 8b) annual trend of −45.7 × 109 m2 yr−1 (99% significance) solely represents the total EM decline, as the Southern Hemisphere EM area (Fig. 8c) displays an insignificant positive trend of +3.3 × 109 m2 yr−1.
Time series of ERA-Interim EM (a) total area (NH and SH; 1013 m2 yr−1), (b) NH area (1013 m2 yr−1), and (c) SH area (1013 m2 yr−1) from 1979 to 2010. Bold lines represent least squares linear regression. Linear trends and statistical significance are shown in bottom left of each plot.
Citation: Journal of Climate 26, 11; 10.1175/JCLI-D-12-00660.1
EM interannual variability differs both spatially and temporally in the respective hemispheres. Higher standard deviations of EM area, indicative of more substantial interannual area variation, are found in the Northern Hemisphere (σ = 9.6 × 1011 m2) versus the Southern Hemisphere (σ = 6.0 × 1011 m2), despite the fact that the total area of the NH EM regions is much smaller than that of the SH EM regions. Figures 9a and 9b show the high (1985) and low (2003) EM years for the Northern Hemisphere. During 1985, compared to the 32-yr climatology, the EM extended a few degrees farther south into the Labrador Sea and North Atlantic and also slightly farther north into the Davis Strait. Slight increases (red) are also found just east of Iceland and southeast of Svalbard. The largest Northern Hemisphere EM increases are found in the North Pacific where the EM expands to cover the Aleutians as well as the southern tip of the Kamchatka Peninsula, encroaching into the Sea of Okhotsk. In contrast, 2003 shows EM declines (blue) around both the equatorward and poleward boundaries in the North Atlantic. The largest declines of this low area year undoubtedly occur in the North Pacific where almost the entire EM area is lacking with the exception of a sliver at 60°N near the Bering Strait. Southern Hemisphere EM variability during high/low years is much less distinct (Figs. 9c,d). During the high EM (1986) there are slight poleward increases in area near 60°S just west of the Antarctic Peninsula and Ross Sea and equatorward increases in the South Pacific between 90° and 150°W. The low EM year (1983) shows a decline in a similar region of the South Pacific that expressed growth in 1986.
ERA-Interim representation of the (a) highest EM area year (1985) in the NH and (b) lowest EM area year (2003) in the NH. (c),(d) As in (a),(b), but for the SH (1986 and 1983, respectively). Green represents the EM area (1979–2010), red represents additional EM area for that specific year, and blue represents missing EM area for that year, relative to 1979–2010 monthly average temperatures.
Citation: Journal of Climate 26, 11; 10.1175/JCLI-D-12-00660.1
Variability of winter sea ice extent and summer SSTs controls EM areas (Figs. 10 and 11). In particular, the NH SST boundary, in terms of its latitudinal position, is more interannually variable over the study period (σ = 1.50°) than winter ice extent (σ = 1.08°), indicating that the EM variability may be more affected by the SSTs since sea ice is constrained by the continents. These findings are consistent across both NH EM areas. In the SH, the overall variability in both parameters is smaller relative to the NH. The ice maximum is slightly more variable (σ = 0.96°) than SST (σ = 0.78°), indicating that sea ice variability is both larger and more regional than SST variability.
(a) ERA-Interim mean winter (JFM) sea ice concentration anomalies for 1985 minus the 1979–2010 climatology in the NH. The ERA-Interim EM area for the 1979–2010 average is outlined by solid black contours, and the EM area for 1985 is outlined by dashed black contours. (b) As in (a), but for 2003. (c) Mean summer (JAS) SST (°C) anomalies for 1985 minus the 1979–2010 climatology in the NH. The ERA-Interim EM area for 1979–2010 average is outlined by solid black contours, and the EM area for 1985 is outlined by dashed black contours. (d) As in (c), but for 2003.
Citation: Journal of Climate 26, 11; 10.1175/JCLI-D-12-00660.1
(a) ERA-Interim mean winter (JAS) sea ice concentration anomalies for 1986 minus the 1979–2010 climatology in the SH. The ERA-Interim EM area for the 1979–2010 average is outlined by solid black contours, and the EM area for 1986 is outlined by dashed black contours. (b) As in (a), but for 1983. (c) Mean summer (JFM) SST anomalies for 1986 minus the 1979–2010 climatology in the SH. The ERA-Interim EM area for the 1979–2010 average is outlined by solid black contours, and the EM area for 1986 is outlined by dashed black contours. (d) As in (c), but for 1983.
Citation: Journal of Climate 26, 11; 10.1175/JCLI-D-12-00660.1
Figures 10a and 10b show the winter ice cover of the high (1985) and low (2003) years relative to 1979–2010. Temporal offsets created by the use of traditional seasonal classifications are avoided by using consecutive monthly averages for winter [January–March (JFM)] and summer [July–September (JAS)] variables. The negative anomalies during 1985 in EM areas are much more pronounced versus the smaller negative anomalies of 2003, especially in the North Pacific, which would indicate that the ice concentration was anomalously low during winter 1985, allowing the possibility for northward expansion of EM area. However, the overall lack of EM poleward growth during 1985 would indicate that winter maximum ice extent may not the best controller of this climatic regime during this particular year. SSTs during the summer of 1985 (Fig. 10c) were also slightly lower than average around the equatorward EM limit (60°N) across most of the North Atlantic and North Pacific, which may have allowed the EM to expand southward.
In contrast, during summer 2003 similar areas exhibited SSTs up to 3°C warmer than average (Fig. 10d), while positive winter sea ice anomalies that year over the Barents Sea (30°–60°W) likely cocontributed to the decreased EM area. Despite warming trends, the 2003 event should not be surprising given that positive Arctic sea ice extents are possible on various time scales depending on the strength of natural variability (Kay et al. 2011). Arctic winter sea ice variability has previously been tied to other atmospheric and oceanic factors, such as the Arctic Oscillation (Rigor et al. 2002) and Pacific decadal oscillation (Francis and Hunter 2007). Annual time series of these teleconnections (not shown) indicate that positive phases during 2003 may be partly responsible for the areal NH EM loss. Poleward shifting ecosystems due to climatic warming may also cause the EM area to decrease and/or shift over time (Wang and Overland 2004). We found that global EM climates are shrinking over time at the expense of poleward shifting A–D climates in the low and midlatitudes (not shown). However, oceanic EM climates are changing more slowly (−0.13 × 1013 m2 yr−1) than other polar Köppen climates such as the tundra ET (−0.55 × 1013 m2 yr−1) and ice cap EF (−0.35 × 1013 m2 yr−1) over the specified study period, in part due to the concentrated warming over northern high latitude terrestrial areas especially during transition seasons (Solomon et al. 2007, Fig. 3.10) and changes over the high southern latitudes (indicated below).
Southern Hemisphere ice cover during high (1986)/low (1983) EM years (Figs. 11a,b) shows strong negative anomalies in sea ice concentration in the Ross Sea during 1986 that prompted a slight increase in EM area, whereas positive anomalies just west of the Antarctic Peninsula coincided with the EM declines during 1986 (Fig. 9c). The circum-Antarctic sea ice concentration anomalies are less pronounced in 1983 and the equatorward boundary declines cannot be conclusively tied to sea ice behavior. On the other hand, the EM declines in both high and low years are associated with regional SST anomalies ≥1°C (Figs. 11c,d). During 1986, a warm SST anomaly pocket between 40° and 60°S and 150° and 170°E is likely responsible for slight EM declines, yet cool SSTs (negative anomalies) at similar latitudes explain EM growth between 120° and 150°W and offset some of these losses. During the low year of 1983, moderate area declines within 40°–60°S, 90°W–180° coincide with strong SST warming along the equatorward EM boundary. The SST variability between these years in the South Pacific suggests an ENSO influence (Yuan 2004). Despite stronger ENSO events in the 1990s, ENSO-related variability was concentrated over the South Pacific in the 1980s, rather than farther south near the Antarctic coast in the 1990s, because of the connection between ENSO and SAM (Fogt and Bromwich 2006; Fogt et al. 2011). Moreover, a negative SAM coupled with a stronger El Niño during late summer/early fall of 1982–83 relative to the 1986–88 event could partly explain the SH EM areal minimum (Stammerjohn et al. 2008).
4. Discussion
a. Differences from Shear (1964)
The modern EM climate regions shown in Figs. 3a and 3b are similar in general to the Shear’s (1964) regions presented in Fig. 1, but differ in some aspects as would be expected given the different time periods, data availability, and methods used to develop the maps. In the North Atlantic, the southern margin of the EM climate depicted by Shear (Fig. 1) extends from off the coast of Newfoundland at about 53°N, 51°W, northeastward to southern Iceland, and then northeastward to the North Cape of Norway and ending in the Barents Sea at 72°N, 41°E. The modern EM region (Fig. 3a) has a similar position off the coast of Newfoundland but extends farther westward into the Labrador Sea than shown by Shear. The southern boundary is similar to Shear’s across Iceland to North Cape, but the modern EM climate extends farther eastward along the Kola Peninsula and northward into the Barents Sea. The northern boundary in the modern EM climate (Fig. 3a) is similar to Shear’s around southern Greenland, but it extends hundreds of kilometers north of Shear’s northern boundary over the Greenland Sea to 79°N near Svalbard.
In the North Pacific (Fig. 1b), Shear showed the EM climate in the Bering Sea extending southward to 50°N across the Aleutian Islands from Umnak Island (167°W) westward beyond Attu Island (173°E) to the Komandorskyie Islands and ending along the 50th parallel at about 153°E. The EM climate depicted in Fig. 3a does not extend as far south or west. It does not include the Aleutians, except near 180°, and does not extend west of about 172°E. The northern extent of the EM climate in the Bering Sea depicted by Shear was just south of 60°N from Kuskokwim Bay westward. Figure 3a shows a similar position in Kuskokwim Bay but the EM climate extends westward while trending north of 60°N to near 175°E, well north of Shear’s region along the Russian coast. The modern EM region in the Bering Sea is somewhat smaller and displaced northward from Shear’s region. These differences could be attributable to strong climate change in the region since the 1960s when Shear first analyzed the EM, the recent 1979–2010 NH EM trend (Fig. 8b), differences in data availability, or a combination of these factors.
The poleward (southern) boundary of the EM climate in the Southern Hemisphere (Fig. 3b) is near 60°S from 30° to 100°E, where Shear depicts the boundary 2° to 3° north of 60°S (Fig. 1c). From 30°E to 40°W, the modern EM boundary is north of 60°S, similar to Shear’s. From the Antarctic Peninsula (~50°W) westward to 120°E, the modern EM boundary is well south of 60°S and south of Shear’s southern boundary. The equatorward (northern) boundary of the EM region is near 50°S in the Western Hemisphere, but near 45°S elsewhere, in general agreement with Shear’s northern boundary.
b. The EM ecosystem
These marine ecosystems are rich in biomass, are important for human subsistence and as commercial resources, and have global ocean connections. O’Harra (2005) referred to the Bering Sea as a “northern Galápagos” for the extraordinary collection of wildlife, including 10 species of seals, sea lions, and walruses, 17 types of whales and dolphins, and millions of nesting seabirds. Populations of northern fur seals, Steller sea lions, and sea otters have decreased in recent decades and shifting climates may have a role in the changes, in part by affecting fish populations (Miller et al. 2005). The Bering Sea is home to about 300 species of fish and these populations are affected by changing climates (Livingston and Tjelmeland 2000). The management of the commercial fisheries depends on understanding these relationships and trends. The Barents and Norwegian Seas in the North Atlantic are also rich ecosystems (Livingston and Tjelmeland 2000). The Icelandic marine ecosystem is highly sensitive to climate variations as seen in fluctuating biomass during alternating warm and cold periods of the twentieth century (Astthorsson et al. 2007). Climatic variability in the EM North Atlantic influences distribution of fish species with northward migration during warm periods and southward shifts in cooling periods (Loeng and Drinkwater 2007). The EM climate of the Southern Hemisphere is a similarly rich marine ecosystem with Antarctic krill playing a central role (Clarke and Harris 2003). There are large populations of squid and fish along with top predators, such as penguins, petrels, and marine mammals. Most biogeographical patterns are circumpolar with sub-Antarctic islands forming distinct ecoregions (Clarke and Harris 2003). The terrestrial EM ecosystems in the Northern and Southern Hemispheres occur on isolated islands and are adapted to a windy, wet, and low-light environment with frequent freeze–thaw activity. These islands include some endemic species and are important for bird migration and breeding (Winker et al. 2002). Soil frost conditions are critical for some vascular plant species and for lichens, moss, and grasses (Cannone et al. 2006; Haussmann et al. 2009). A reduction and shift in NH EM area, in particular, could especially impact biodiversity in the region.
c. Future projections
The modern depiction of EM climate in Figs. 3a and 3b can serve as a baseline against which future changes in the area and distribution of the climate may be assessed for potential impacts on marine and terrestrial ecosystems. Feng et al. (2012) used a suite of climate models and three Special Report on Emissions Scenarios (SRES) scenarios (B1, A1B, and A2) to estimate the 1900–2099 mean total annual warming of 3°–5°C at 50°–60°N, which is expected to be more pronounced during the winter than summer season. Similarly, Bracegirdle and Stephenson (2012) also employ ensemble approaches and find strong, yet inhomogenous wintertime warming over Arctic seas and the Southern Ocean through the twenty-first century. Continued air and ocean temperature increases over EM domains are likely to impact sea ice cover in both hemispheres (Arzel et al. 2006; Overland and Wang 2007; Bracegirdle et al. 2008; Serreze et al. 2009; Neelin 2011) and potentially create wetter, cloudier, and stormier conditions, especially in the Arctic where strong warming and sea ice decline are already long underway (Vavrus et al. 2012). Should oceanic warming and sea ice decline persist under warming conditions, we would expect the EM area to potentially shrink and shift poleward [as the comparison between the Shear (1964) EM maps and our updated EM maps already suggest may be occurring]. However, as we have seen in the strong downward trend in NH EM area since 1979, and from comparing high/low EM years, these parameters do not necessarily cooperate interannually. Therefore, while we would expect the strong interannual variability of EM area to persist, a substantial EM climatic (poleward) shift that would allow the areal extent of EM areas to be preserved may be less likely.
5. Conclusions
The modern EM climate is roughly consistent with that originally mapped by Shear (1964). Similarities between three reanalyses (ERA-Interim, CFSR, and JRA-25; Fig. 2) show that EM climatic results can be suitably portrayed regardless of model selection. Using ERA-Interim, the distinct differences of the EM from its continental ET counterpart are shown in both hemispheres and its climatic boundaries can be traced with warmest month 10°C SSTs and coldest month 15% sea ice concentration extent in most instances. Interannual variability of the EM climate, as measured by fluctuations in areal extent, is especially impacted by SST anomalies, particularly in the NH where EM variability dominates the downward temporal trend of global EM area.
Because of its mild oceanic climate, the EM domain supports a unique, sensitive, and globally important ecosystem. However, recent decades have witnessed a decline in Northern Hemisphere EM area, especially, as climate variability and change have prompted regional increases in air and sea surface temperatures and sea ice decline. The EM has undoubtedly been affected as a result, as recent literature suggests. Therefore, it is critical to continue monitoring the impacts of climate change on EM regions and the associated biogeographical response in the forthcoming century.
Acknowledgments
The authors thank two anonymous reviewers and Andrew Monaghan for their constructive comments. Reanalysis output were provided by the Data Support Section of the Computational and Information Systems Laboratory at the National Center for Atmospheric Research.
REFERENCES
Andersson, E., 2007: Data assimilation in the polar regions. ECMWF Newsletter, No. 112, ECMWF, Reading, United Kingdom, 10–15.
Arzel, O., T. Fichefet, and H. Goosse, 2006: Sea ice evolution over the 20th and 21st centuries as simulated by AOGCMs. Ocean Modell., 12, 401–415.
Astthorsson, O. S., A. Gislason, and S. Jonsson, 2007: Climate variability and the Icelandic marine ecosystem. Deep-Sea Res. II, 54, 2456–2477.
Bekryaev, R. V., I. V. Polyakov, and V. A. Alexeev, 2010: Role of polar amplification in long-term surface air temperature variations and modern Arctic warming. J. Climate, 23, 3888–3906.
Bengtsson, L., S. Hagemann, and K. I. Hodges, 2004a: Can climate trends be calculated from reanalysis data? J. Geophys. Res., 109, D11111, doi:10.1029/2004JD004536.
Bengtsson, L., K. I. Hodges, and S. Hagemann, 2004b: Sensitivity of the ERA-40 reanalysis to the observing system: Determination of the global atmospheric circulation from reduced observations. Tellus, 56A, 456–471.
Bracegirdle, T. J., and D. B. Stephenson, 2012: Higher precision estimates of regional polar warming by ensemble regression of climate model projections. Climate Dyn., 39, 2805–2821, doi:10.1007/s00382-012-1330-3.
Bracegirdle, T. J., W. M. Connolley, and J. Turner, 2008: Antarctic climate change over the twenty first century. J. Geophys. Res., 113, D03103, doi:10.1029/2007JD008933.
Bromwich, D. H., and R. L. Fogt, 2004: Strong trends in the skill of the ERA-40 and NCEP–NCAR Reanalyses in the high and midlatitudes of the Southern Hemisphere, 1958–2001. J. Climate, 17, 4603–4619.
Bromwich, D. H., and S.-H. Wang, 2005: Evaluation of the NCEP–NCAR and ECMWF 15- and 40-yr reanalyses using rawinsonde data from two independent Arctic field experiments. Mon. Wea. Rev., 133, 3562–3578.
Bromwich, D. H., A. J. Monaghan, and Z. Guo, 2004: Modeling the ENSO modulation of Antarctic climate in the late 1990s with the Polar MM5. J. Climate, 17, 109–132.
Bromwich, D. H., R. L. Fogt, K. I. Hodges, and J. E. Walsh, 2007: A tropospheric assessment of the ERA-40, NCEP, and JRA-25 global reanalyses in the polar regions. J. Geophys. Res., 112, D10111, doi:10.1029/2006JD007859.
Bromwich, D. H., J. P. Nicolas, and A. J. Monaghan, 2011: An assessment of precipitation changes over Antarctica and the Southern Ocean since 1989 in contemporary global reanalyses. J. Climate, 24, 4189–4209.
Cannone, N., J. C. Ellis Evans, R. Strachan, and M. Guglielmin, 2006: Interactions between climate, vegetation and the active layer in soils at two maritime Antarctic sites. Antarct. Sci., 18, 323–333.
Cavalieri, D. J., and C. L. Parkinson, 2008: Antarctic sea ice variability and trends, 1979– 2006. J. Geophys. Res., 113, C07004, doi:10.1029/2007JC004564.
Cavalieri, D. J., and C. L. Parkinson, 2012: Arctic sea ice variability and trends, 1979–2010. Cryosphere Discuss., 6, 957–979, doi:10.5194/tcd-6-957-2012.
Cavalieri, D. J., C. L. Parkinson, and K. Y. Vinnikov, 2003: 30-year satellite record reveals contrasting Arctic and Antarctic decadal sea ice variability. Geophys. Res. Lett., 30, 1970, doi:10.1029/2003GL018031.
Chapman, W. L., and J. E. Walsh, 2007: A synthesis of Antarctic temperatures. J. Climate, 20, 4096–4117.
Chylek, P., C. K. Folland, G. Lesins, M. K. Dubey, and M. Wang, 2009: Arctic air temperature change amplification and the Atlantic multidecadal oscillation. Geophys. Res. Lett., 36, L14801, doi:10.1029/2009GL03877.
Clarke, A., and C. M. Harris, 2003: Polar marine ecosystems: Major threats and future changes. Environ. Conserv., 30, 1–25.
Dee, D. P., and S. Uppala, 2009: Variational bias correction of satellite radiance data in the ERA-Interim reanalysis. Quart. J. Roy. Meteor. Soc., 135, 1830–1841.
Dee, D. P., and Coauthors, 2011: The ERA-Interim reanalysis: Configuration and performance of the data assimilation system. Quart. J. Roy. Meteor. Soc., 137, 553–597.
Feng, S., C.-H. Ho, Q. Hu, R. J. Oglesby, S.-J. Jeong, and B.-M. Kim, 2012: Evaluating observed and projected future climate changes for the Arctic using the Köppen–Trewartha climate classification. Climate Dyn., 38, 1359–1373.
Fogt, R. L., and D. H. Bromwich, 2006: Decadal variability of the ENSO teleconnection to the high-latitude South Pacific governed by coupling with the southern annular mode. J. Climate, 19, 979–997.
Fogt, R. L., D. H. Bromwich, and K. M. Hines, 2011: Understanding the SAM influence on the South Pacific ENSO teleconnection. Climate Dyn., 36, 1555–1576.
Francis, J. A., and E. Hunter, 2007: Drivers of declining sea ice in Arctic winter: A tale of two seas. Geophys. Res. Lett., 34, L17503, doi:10.1029/2007GL030995.
Gille, S. T., 2002: Warming of the Southern Ocean since the 1950s. Science, 295, 1275–1277.
Haussmann, N. S., J. C. Boelhouwers, and M. A. McGeoch, 2009: Fine scale variability in soil frost dynamics surrounding cushions of the dominant vascular plant species (Azorella selago) on sub-Antarctic Marion Island. Geogr. Ann., 91, 257–268.
Hines, K. M., D. H. Bromwich, and G. J. Marshall, 2000: Artificial surface pressure trends in the NCEP–NCAR reanalysis over the Southern Ocean and Antarctica. J. Climate, 13, 3940–3952.
Hodges, K. I., R. W. Lee, and L. Bengtsson, 2011: A comparison of extratropical cyclones in recent reanalyses ERA-Interim, NASA MERRA, NCEP CFSR, and JRA-25. J. Climate, 24, 4888–4906.
Hoskins, B. J., and K. I. Hodges, 2002: New perspectives on the Northern Hemisphere winter storm tracks. J. Atmos. Sci., 59, 1041–1061.
Hoskins, B. J., and K. I. Hodges, 2005: A new perspective on Southern Hemisphere storm tracks. J. Climate, 18, 4108–4129.
Kanamitsu, M., W. Ebisuzaki, J. Woollen, S.-K. Yang, J. J. Hnilo, M. Fiorino, and G. L. Potter, 2002: NCEP–DOE AMIP-II Reanalysis (R-2). Bull. Amer. Meteor. Soc., 83, 1631–1643.
Kay, J. E., M. M. Holland, and A. Jahn, 2011: Inter-annual to multi-decadal Arctic sea ice extent trends in a warming world. Geophys. Res. Lett., 38, L15708, doi:10.1029/2011GL048008.
Kleist, D. T., D. F. Parrish, J. C. Derber, R. Treadon, W.-S. Wu, and S. Lord, 2009: Introduction of the GSI into the NCEP Global Data Assimilation System. Wea. Forecasting, 24, 1691–1705.
Kumar, A., and Coauthors, 2010: Contribution of sea ice loss to Arctic amplification. Geophys. Res. Lett., 37, L21701, doi:10.1029/2010GL045022.
Levitus, S., J. I. Antonov, T. B. Boyer, and C. Stephens, 2000: Warming of the world ocean. Science, 287, 2225–2229.
Liston, G. E., and C. A. Hiemstra, 2011: The changing cryosphere: Pan-Arctic snow trends (1979–2009). J. Climate, 24, 5691–5712.
Livingston, P. A., and S. Tjelmeland, 2000: Fisheries in boreal ecosystems. ICES J. Mar. Sci., 57, 619–627.
Loeng, H., and K. Drinkwater, 2007: An overview of the ecosystems of the Barents and Norwegian Seas and their response to climate variability. Deep-Sea Res. II, 54, 2478–2500.
Mark, A. F., K. J. M. Dickinson, and R. G. M. Hofstede, 2000: Alpine vegetation, plant distribution, life forms, and environments in a perhumid New Zealand region: Oceanic and tropical high mountain affinities. Arct. Antarct. Alp. Res., 32, 240–254.
Markus, T., J. C. Stroeve, and J. Miller, 2009: Recent changes in Arctic sea ice melt onset, freezeup, and melt season length. J. Geophys. Res., 114, C12024, doi:10.1029/2009JC005436.
Marshall, G. J., 2002: Trends in Antarctic geopotential height and temperature: A comparison between radiosonde and NCEP–NCAR reanalysis data. J. Climate, 15, 659–674.
Marshall, G. J., and S. A. Harangozo, 2000: An appraisal of NCEP/NCAR reanalysis MSLP data viability for climate studies in the South Pacific. Geophys. Res. Lett., 27, 3057–3060.
Maslanik, J. A., C. Fowler, J. Stroeve, S. Drobot, J. Zwally, D. Yi, and W. Emery, 2007: A younger, thinner Arctic ice cover: Increased potential for rapid extensive sea-ice loss. Geophys. Res. Lett., 34, L24501, doi:10.1029/2007GL032043.
Miller, A. J., A. W. Trites, and H. D. G. Maschner, 2005: Ocean climate changes and the Steller sea lion decline. Antarct. Res. USA, 19, 54–63.
Monaghan, A. J., D. H. Bromwich, and S.-H. Wang, 2006: Recent trends in Antarctic snow accumulation from polar MM5 simulations. Philos. Trans. Roy. Soc. London, 364A, 1683–1708.
Neelin, J. D., 2011: Climate Change and Climate Modeling. Cambridge University Press, 282 pp.
Nghiem, S. V., I. G. Rigor, D. K. Perovich, P. Clemente-Colón, J. W. Weatherly, and G. Neumann, 2007: Rapid reduction of Arctic perennial sea ice. Geophys. Res. Lett., 34, L19504, doi:10.1029/2007GL031138.
Noguchi, Y., H. Tabuchi, and H. Hasegawa, 1987: Physical factors controlling the formation of patterned ground on Haleakala, Maui. Geogr. Ann., 69A, 329–342.
O’Harra, D., 2005: A puzzle in the Pribilofs. Smithsonian, 2005 (March), 65–70.
Onogi, K., and Coauthors, 2007: The JRA-25 Reanalysis. J. Meteor. Soc. Japan, 85, 369–432.
Overland, J. E., and M. Wang, 2007: Future regional Arctic sea ice declines. Geophys. Res. Lett., 34, L17705, doi:10.1029/2007GL030808.
Parish, T. R., and D. H. Bromwich, 1998: A case study of Antarctic katabatic wind interaction with large-scale forcing. Mon. Wea. Rev., 126, 199–209.
Parkinson, C. L., and D. J. Cavalieri, 2012: Antarctic sea ice variability and trends, 1979–2010. Cryosphere Discuss., 6, 931–956, doi:10.5194/tcd-6-931-2012.
Porter, D. F., J. C. Cassano, and M. C. Serreze, 2012: Local and large-scale atmospheric responses to reduced Arctic sea ice and ocean warming in the WRF model. J. Geophys. Res., 117, D11115, doi:10.1029/2011JD016969.
Quayle, W. C., L. S. Peck, H. Peat, J. C. Ellis-Evans, and P. R. Harrigan, 2002: Extreme responses to climate change in Antarctic lakes. Science, 295, 645.
Renwick, J. A., 2004: Trends in the Southern Hemisphere polar vortex in NCEP and ECMWF reanalyses. Geophys. Res. Lett., 31, L07209, doi:10.1029/2003GL019302.
Rigor, I. G., J. M. Wallace, and R. L. Colony, 2002: Response of sea ice to the Arctic Oscillation. J. Climate, 15, 2648–2663.
Rohli, R. V., and A. J. Vega, 2011: Climatology. 2nd ed. Jones and Bartlett, 432 pp.
Saha, S., and Coauthors, 2010: The NCEP Climate Forecast System Reanalysis. Bull. Amer. Meteor. Soc., 91, 1015–1057.
Schneider, D. P., C. Deser, and Y. Okumura, 2012: An assessment and interpretation of the observed warming of West Antarctica in austral spring. Climate Dyn., 38, 323–347.
Screen, J. A., and I. Simmonds, 2010: The central role of diminishing sea ice in recent Arctic temperature amplification. Nature, 464, 1334–1337.
Screen, J. A., and I. Simmonds, 2011: Erroneous Arctic temperature trends in the ERA-40 reanalysis: A closer look. J. Climate, 24, 2620–2627.
Serreze, M. C., and A. P. Barrett, 2011: Characteristics of the Beaufort Sea high. J. Climate, 24, 159–182.
Serreze, M. C., A. P. Barrett, J. C. Stroeve, D. N. Kindig, and M. M. Holland, 2009: The emergence of surface-based Arctic amplification. Cryosphere, 3, 11–19.
Serreze, M. C., A. P. Barrett, and J. Stroeve, 2012: Recent changes in tropospheric water vapor over the Arctic as assessed from radiosondes and atmospheric reanalyses. J. Geophys. Res., 117, D10104, doi:10.1029/2011JD017421.
Shear, J. A., 1964: The polar marine climate. Ann. Assoc. Amer. Geogr., 54, 310–317.
Simmonds, I., K. Keay, and E.-P. Lim, 2003: Synoptic activity in the seas around Antarctica. Mon. Wea. Rev., 131, 272–288.
Smith, V. R., 2002: Climate change in the sub-Antarctic: An illustration from Marion Island. Climatic Change, 52, 345–357.
Solomon, S., D. Qin, M. Manning, M. Marquis, K. Averyt, M. M. B. Tignor, H. L. Miller Jr., and Z. Chen, Eds., 2007: Climate Change 2007: The Physical Science Basis. Cambridge University Press, 996 pp.
Stammerjohn, S. E., D. G. Martinson, R. C. Smith, X. Yuan, and D. Rind, 2008: Trends in Antarctic annual sea ice retreat and advance and their relation to El Niño–Southern Oscillation and southern annular mode variability. J. Geophys. Res., 113, C03S90, doi:10.1029/2007JC004269.
Steele, M., W. Ermold, and J. Zhang, 2008: Arctic Ocean surface warming trends over the past 100 years. Geophys. Res. Lett., 35, L02614, doi:10.1029/2007GL031651.
Steig, E. J., D. P. Schneider, S. D. Rutherford, M. E. Mann, J. C. Comiso, and D. T. Shindell, 2009: Warming of the Antarctic ice-sheet surface since the 1957 International Geophysical Year. Nature, 457, 459–462.
Sterl, A., 2004: On the (in)homogeneity of reanalysis products. J. Climate, 17, 3866–3873.
Thompson, D. W. J., and S. Solomon, 2002: Interpretation of recent Southern Hemisphere climate change. Science, 296, 895–899.
Thorne, P. W., and R. S. Vose, 2010: Reanalyses suitable for characterizing long-term trends: Are they really achievable? Bull. Amer. Meteor. Soc., 91, 353–361.
Trenberth, K. E., D. P. Stepaniak, and L. Smith, 2005: The mass of the atmosphere: A constraint on global analyses. J. Climate, 18, 864–875.
Uppala, S. M., and Coauthors, 2005: The ERA-40 Re-Analysis. Quart. J. Roy. Meteor. Soc., 131, 2961–3012.
Vaughn, D. G., and Coauthors, 2003: Recent rapid regional climate warming on the Antarctic Peninsula. Climatic Change, 60, 243–274.
Vavrus, S. J., M. M. Holland, A. Jahn, D. A. Bailey, and B. A. Blazey, 2012: Twenty-first-century Arctic climate change in CCSM4. J. Climate, 25, 2696–2710.
Walsh, J. E., W. L. Chapman, and D. H. Portis, 2009: Arctic cloud fraction and radiative fluxes in atmospheric reanalyses. J. Climate, 22, 2316–2334.
Wang, M., and J. E. Overland, 2004: Detecting Arctic climate change using Köppen climate classification. Climatic Change, 67, 43–62.
Winker, K., D. D. Gibson, A. L. Sowls, B. E. Lawhead, P. D. Martin, E. P. Hoberg, and D. Causey, 2002: The birds of St. Matthew Island, Bering Sea. Wilson Bull., 114, 491–509.
Yuan, X., 2004: ENSO-related impacts on Antarctic sea ice: A synthesis of phenomenon and mechanisms. Antarct. Sci., 16, 415–425.
