1. Introduction
Climatologically, the frequency of Northern Hemisphere blocking is higher in the extended winter season. As a function of longitude, blocking frequency features a bimodal distribution peaking over the Euro-Atlantic region and the Pacific Ocean, both of which are located downstream of the climatological storm tracks (e.g., Rex 1950; Tibaldi and Molteni 1990; Lupo and Smith 1995; Pelly and Hoskins 2003; Barriopedro et al. 2006; Tyrlis and Hoskins 2008a). In addition, the Ural Mountains (western Russia) are a secondary blocking sector, and blocking occurs over this region is called Ural blocking (UB) in the literature. The occurrence of UB is related to the eastward shift of storm activity over the Euro-Atlantic region in boreal winter (Diao et al. 2006; Tyrlis and Hoskins 2008b; Luo et al. 2010; Cheung et al. 2013). Under a warming climate, the blocking frequency likely decreases over the Euro-Atlantic and Pacific sectors and undergoes no apparent change or a slight increase over the Ural Mountains (Dunn-Sigouin and Son 2013; Masato et al. 2013). Yet, there have been limited studies analyzing the implications of blocking for the present and future regional climate (Masato et al. 2014; Woollings et al. 2014).
The objective of our work is to systematically evaluate the impact of wintertime UB on the East Asian winter climate. In Part I of this work (Cheung and Zhou 2015, hereinafter Part I), we demonstrated the ability of 25 GCMs participating in the fifth phase of the Coupled Model Intercomparison Project (CMIP5) to simulate the linkage between UB and the East Asian winter climate in the historical run (HIST), which agrees with earlier observational work (e.g., Takaya and Nakamura 2005; Zhou et al. 2009; L. Wang et al. 2010; Cheung et al. 2012; Cheung et al. 2013). Basically, the occurrence of UB strengthens the advection of cold polar air toward western Siberia and reinforces the Siberian high (SH) near the surface. Because of the persistence of blocking in the extratropical atmosphere, UB potentially triggers severe cold air outbreaks or even persistent snowstorms in East Asia, where the impact can be large in the subtropical regions. For example, the extraordinarily high UB frequency in January 2008 was one of the dynamic causes of an unusually long icy rain event and severe snowstorm in southern China (Zhou et al. 2009). Given the crucial role of UB in the East Asian winter climate, the projection of UB in the state-of-the-art CMIP5 GCMs may be vital for the government and policy makers in improving preparedness for cold extremes in East Asia even under a warming climate.
Compared to earlier generations of GCMs, the CMIP5 GCMs do not show great improvement in the simulation of blocking. As mentioned in chapter 14 of the IPCC Fifth Assessment Report (AR5; Christensen et al. 2014), there is medium confidence that blocking frequency will not increase, and the impact of blocking on the regional climate remains uncertain. Undoubtedly, the uncertainties in the occurrence of blocking could present a great challenge for assessing regional climate change, especially the occurrence of extreme weather. These uncertainties should be identified before further work is carried out in order to increase our confidence in projecting the future changes of blocking frequency.
It is reasonable to assume that climate change in the twenty-first century is a result of anthropogenic forcing superimposed on natural variability. Whereas natural forcing consists of strong interannual and multidecadal signals, anthropogenic forcing is a slowly varying signal. In Part II we attempt to investigate how likely it is that the long-term mean and variance of UB will undergo systematic changes and to determine whether these changes are related to changes in the East Asian winter circulation in the future climate scenarios. The results in Part II are derived from two of the representative concentration pathways (RCPs), the RCP4.5 and RCP8.5 scenarios (Moss et al. 2010), where the value denotes the strongest radiative forcing (W m−2) in the twenty-first century. If the long-term mean of the large-scale circulation varies in a similar manner and differs only in magnitude between the two scenarios, then we can deduce the role of anthropogenic forcing in the basic state of the large-scale circulation. On the other hand, if the projection exhibits a large spread across individual GCMs, we will attempt to identify the cause of the uncertainty.
We seek to answer the questions below based on multiple model ensembles (MMEs) and the spread of 20 GCMs in the RCP4.5 and RCP8.5 scenarios:
Will there be any systematic and significant changes in the frequency of Ural blocking?
What are the possible causes for the future changes of UB?
What will be the likely implication of UB for the East Asian winter climate?
2. Data and methods
a. Data
We analyze the outputs of 20 CMIP5 GCMs where daily data are available in the historical (HIST), RCP4.5, and RCP8.5 runs, as summarized in Table 1. The data were extracted from the Earth System Grid Federation (ESGF) data portal (http://pcmdi9.llnl.gov/esgf-web-fe/). The raw data include daily and monthly fields of mean sea level pressure (MSLP), 250-hPa and 500-hPa zonal wind (U250 and U250), and 500-hPa geopotential height (Z500). To be consistent with Part I, we bilinearly interpolated the CMIP5 data into a horizontal resolution of 2.5° latitude × 2.5° longitude.
Description of the 25 CMIP5 GCMs used in Part I, where the letters V and X represent whether the data was (V) or was not (X) available for a specific scenario. The five GCMs with an X in either RCP run, which are in italics, are not adopted in this study because of data availability. (Expansions of institutions and model names are available online at http://www.ametsoc.org/PubsAcronymList.)
The entire study period for future climate conditions is December–February (DJF) from 2006 to 2099, where 2006 represents the DJF period from 2006 to 2007. The present climate condition is based on the 30-yr period of 1971–2000.
b. Detection of blocking
A blocking-type circulation at λ (“blocking longitude”) is characterized by the reversal of the daily meridional Z500 gradient over the extratropical region. This has to be satisfied by ZIN < −10 gpm (°lat)−1 and ZIS > 0.
Second, a blocking region consists of at least five consecutive blocking longitude grid points (12.5° longitude) because blocking is a large-scale system. The longitude and latitude center of a blocking region (λctr, ϕctr) is defined as the grid with the largest positive Z500 anomaly within the region.
Third, only the blocking regions persisting for at least four consecutive days are retained, and these are called blocking events. The persistence criterion is determined by the characteristic time of the blocking region (Pelly and Hoskins 2003).
Fourth, the blocking frequency at each longitude in a period (say, DJF) is counted as the number of days in which a blocking event can be identified over that longitude throughout the period.
c. Ural blocking and Ural blocking index
In each winter, the UB frequency is taken as the DJF-averaged blocking frequency over 45°–90°E, which is called the Ural blocking index (UBI). The climatological statistics of UBI, including the long-term mean and variance, are derived from the UBI of individual winters.
The spatiotemporal characteristics are analyzed by the composites of UB events. Wintertime UB events are defined as those events whose centered longitude (λctr) is located at 45°–90°E during their establishment (day 0). Furthermore, all of these events should start in the DJF period.
d. Siberian high intensity
As mentioned in section 2d of Part I, the East Asian winter climate is depicted by the Siberian high intensity. It is defined as the area-averaged MSLP over 40°–65°N, 80°–120°E, which encloses the climatological region of the SH.
e. Projection in the twenty-first century
Considering a period of 30 years to be the most common length to define the climatology (e.g., 1961–90, 1971–2000, and 1981–2010), a 30-yr sliding window is applied for the period 2006–2100 in the RCP runs to project the long-term mean and variance. To test whether the changes in blocking frequency and large-scale circulation features are systematic and statistically significant in different periods of the twenty-first century, the projection and uncertainty analyses are assessed by the running difference, the time-slice difference, and the trend of the long-term mean and variance:
The running difference is carried out for the 65 running means of a time series throughout the entire study period, where the first period is 2006–35 and the last is 2070–99.
The time-slice difference is used for composite and regression analyses, which are based on three 30-yr base periods: (i) 2010–39, (ii) 2040–69, and (iii) 2070–99.
3. Frequency of UB occurrence
To identify the most notable change in response to anthropogenic forcing, the MME of the blocking frequency climatology in the RCP4.5 and RCP8.5 runs is shown together with the HIST run and NCEP data in Fig. 1. In the HIST run, although the majority of the CMIP5 models can reproduce the bimodal blocking frequency distribution over the Northern Hemisphere, the primary peak over the Euro-Atlantic sector becomes much weaker and the secondary peak over the Ural Mountains appears to be more distinctive when compared to NCEP. As discussed in section 7b of Part I and in Scaife et al. (2010), this discrepancy is due to biases in the long-term mean circulation over the Euro-Atlantic region.
The climatology of winter-mean Northern Hemisphere blocking frequency. (a),(b) NCEP–NCAR reanalysis datasets (NCEP), the multiple model ensembles (MME) of the historical (HIST) run, and 30-yr running mean (from 2010–39 to 2070–99) of the RCP4.5 and 8.5 runs. (c),(d) Difference between the RCP runs and the HIST run. Unit: % of winter days. Note that the square box encloses the Ural sector (45°–90°E).
Citation: Journal of Climate 28, 6; 10.1175/JCLI-D-14-00309.1
In the two RCP runs, the 65 MMEs of the 30-yr climatology from 2006–35 to 2070–99 are illustrated by a map with 65 different colors. Whereas the bimodal distribution remains unchanged (Figs. 1a,b), the mean blocking frequency over the two climatological peaks decreases systematically in the twenty-first century, and such a decrease is more pronounced in the Pacific sector and in the RCP8.5 run (Figs. 1c,d). This is consistent with previous studies analyzing the change of blocking frequency (Matsueda et al. 2009; Dunn-Sigouin and Son 2013; Masato et al. 2013). Over the Ural sector, however, there is no significant change in any period in the two RCP runs. Whereas the MME of UB frequency decreases slightly during the entire period of the RCP4.5 run, it increases slightly in the RCP8.5 run over the western and eastern boundary of the UB sector in the middle and late twenty-first century.
Because the blocking occurrence in a GCM is closely related to the climatological mean state of the GCM (Scaife et al. 2010), the change in the mean blocking occurrence over various sectors can be deduced by U500, as shown in Fig. 2. Compared to the HIST run, no significant change in U500 can be seen in the period 2010–39 (Figs. 2a,b). In the later periods, U500 exhibits a stronger tendency in the midlatitudes (south of ϕ0) and a weakening tendency in the subpolar region (north of ϕ0) over the North Pacific and Euro-Atlantic sectors (Figs. 2c–f), where these changes are more apparent in the period 2070–99 (Figs. 2e,f). The above changes suggest that there are 1) a weaker climatological meridional gradient of Z500 between ϕN and ϕ0 and a less negative tendency of ZIN in Eq. (1), and 2) a stronger climatological meridional gradient of Z500 between ϕ0 and ϕS and a stronger negative tendency of ZIS in Eq. (2). Because blocking has to be achieved by ZIN < −10 gpm (°lat)−1 and ZIS > 0, these tendencies are related to a decrease of blocking frequency over the Euro-Atlantic and Pacific sectors. On the other hand, U500 over the latitudes defining UB does not show a significant change in any period (Fig. 2). Based on the MME of the climatological mean state, the UB frequency is not likely to increase or decrease significantly in the twenty-first century.
Composited 500-hPa zonal wind (U500) climatology in three subperiods of the twenty-first century. (a),(b) 2010–39, (c),(d) 2040–69, and (e),(f) 2070–99 in (left) RCP4.5 and (right) RCP8.5 runs. The MME (shading; m s−1) and its difference with the HIST run (black; contour interval: 1 m s−1). Stippling indicates where at least 75% of CMIP5 GCMs agree with the sign change of the MME. The blue box encloses the UB sector (55°–65°N, 45°–90°E).
Citation: Journal of Climate 28, 6; 10.1175/JCLI-D-14-00309.1
The MME and the spread of the long-term mean of the UBI are shown in Figs. 3a and 3b. In the entire study period, although the MME is fairly constant (~6% of winter days; Figs. 3a,b) and is comparable to the HIST run (Figs. 3c,d), a large spread exists across the 20 CMIP5 GCMs. Such a spread might be partially related to the spread of large-scale circulation features over the Euro-Atlantic sector, as mentioned in Part I. Moreover, the difference in the long-term mean of UBI between the RCP runs and the HIST run also has a notable spread across the 20 CMIP5 GCMs (Figs. 3c,d), where the magnitude appears to be larger in the RCP8.5 run (Fig. 3d). The larger spread of the RCP8.5 run is probably due to the atmospheric response of different GCMs to the effect of climate change, including the long-term mean and variance of UBI, which will be investigated in section 4.
Time series of 30-yr running mean of Ural blocking index for the MME (black) and the spread (shading) of 20 CMIP5 GCMs. (a),(b) Year-to-year variation in the RCP runs, where the gray lines represent the long-term mean and ±1 standard deviation of the UBI obtained from the 20 CMIP5 GCMs for the period 1971–2000. (c),(d) Difference between the RCP runs and the long-term mean in the HIST run of each GCM. Unit: % of winter days.
Citation: Journal of Climate 28, 6; 10.1175/JCLI-D-14-00309.1
4. Trend of UBI and SHI
In boreal winter, the occurrence of UB is accompanied by the intensification of the SH. During the HIST period, a lower occurrence of UB in a given winter likely corresponds to a weaker SH and a warmer East Asian winter climate, and vice versa (L. Wang et al. 2010). In the two RCP runs, the long-term relationship between UB and the SH is studied by using the 30-yr running correlation between UBI and SHI throughout the twenty-first century, where each model has 65 linear correlation coefficients from 2006–35 to 2070–99 (Fig. 4). Although the running correlation of the two RCP runs has a large spread across the 20 CMIP5 GCMs, the majority of the GCMs have a significant correlation coefficient at the 95% confidence level throughout the twenty-first century (Fig. 4). This suggests a stable long-term relationship between UBI and SHI in the two RCP runs. Accordingly, we would like to address whether there is any relationship between the trend of UBI (representing the UB frequency) and SHI (as a proxy of the East Asian winter climate) in the future climate conditions.
Time series of 30-yr moving correlation between UBI and SHI in the MME (black solid line) and the spread across the 20 CMIP5 GCMs (shading) of the (a) RCP4.5 and (b) RCP8.5 runs. The black dashed line indicates the 95% confidence level with 29 degrees of freedom.
Citation: Journal of Climate 28, 6; 10.1175/JCLI-D-14-00309.1
a. Long-term mean
In section 3, based on the results of MME, we do not identify any robust changes of the UB frequency, unlike the remarkable decrease of the blocking frequency over the Euro-Atlantic and Pacific sectors. On the other hand, the mean state of the East Asian winter climate is expected to undergo a weakening trend under the global warming trend, which results from the weakened land-sea thermal contrast between the East Asian continent and the Pacific Ocean (Hu et al. 2000). The inconsistency of the MME trends of the long-term mean of UBI and SHI can be seen in Figs. 5a and 5b: whereas the weakening trend of SHI is pronounced in both runs, the trend of UBI is insignificant in the RCP4.5 run and positive in the RCP 8.5 run.
Time series of 30-yr (a),(b) running mean and (c),(d) running variance of (black solid line) UBI and (gray solid line) SHI in the (left) RCP4.5 and (right) RCP8.5 runs. The linear trend is represented by the dotted line, and its level of significance using the least squares fit is shown at the top left for UBI and at the top right for SHI.
Citation: Journal of Climate 28, 6; 10.1175/JCLI-D-14-00309.1
Although the MME trend of UBI does not follow that of SHI, the two trends across the 20 GCMs in the RCP8.5 run appear to be closely related to each other. When the trend of the 30-yr running mean of SHI is plotted against that of UBI across the GCMs (Fig. 6a), we obtain a statistically significant relationship between the two long-term means. The long-term mean trend of SHI in a given GCM tends to be positive (negative) when the long-term mean trend of UBI exhibits a strong positive (negative) trend. It is noteworthy that five of the GCMs show a significant positive trend of SHI (Fig. 6a), in contrast to a remarkable weakening trend of the MME (Fig. 5b). Four of the five GCMs also show a positive trend of UBI (Fig. 6a). The above result suggests that the projection uncertainty of UBI might affect that of SHI.
Linear trend of 30-yr (a) running mean and (b) running variance of SHI against that of UBI for the 20 CMIP5 GCMs in the RCP8.5 run, where the dot indicates that the two trends of a GCM are significant at the 95% confidence level, and the cross indicates either one of the two trends of a GCM is not significant at the 95% confidence level. The star represents the trend of the MME. Both the linear correlation coefficient between the linear trends of the 20 GCMs and the corresponding significance level are shown at the top.
Citation: Journal of Climate 28, 6; 10.1175/JCLI-D-14-00309.1
It is necessary to depict the trend of large-scale circulation features related to the trend of UBI in order to deduce whether the trend of UBI and SHI could be linked. As shown in Figs. 7a,c,e, the trend of the 30-yr running mean of UBI is regressed against that of large-scale parameters (MSLP, Z500, and U250) across the 20 CMIP5 GCMs in the RCP 8.5 run. Associated with a larger long-term mean trend of UBI in a given GCM, positive and negative regression coefficients can be found over the SH and the Aleutian low (~160°–170°E) near the surface in the GCM (Fig. 7a), corresponding to a stronger trend of the mean state of the East Asian winter climate, and vice versa. Furthermore, the regression coefficients of the trend of Z500 are positive north of 45°N and negative south of 45°N in the extratropical region (Fig. 7c). Because the Z500 mainly decreases with latitude, the regression map of Z500 seems to imply that the GCMs showing a positive trend of UBI have a weaker trend in the meridional gradient of Z500 over the midlatitudes, and vice versa.
Regression of the linear trend of the 30-yr (left) running mean and (right) running variance in the large-scale circulation parameters against that of UBI: (a),(b) MSLP, (c),(d) Z500, and (e),(f) U250. The contour interval is indicated at the bottom right of each plot, and shading indicates the 95% confidence level. The solid black box encloses the UB sector and the dashed black box encloses the SH region.
Citation: Journal of Climate 28, 6; 10.1175/JCLI-D-14-00309.1
Specifically, the positive trend of UBI is likely associated with the positive trend of Z500 near the Urals and the negative trend of Z500 near the Mediterranean Sea (~30°E; Fig. 7c). These trends are accompanied by the weakening trend of U250 south of the UB sector and the strengthening trend of U250 southeast of the Mediterranean Sea (Fig. 7e). The difference in the trends of U250 between the GCMs with an opposite trend of UBI is further depicted in Fig. 8. Comparatively, the GCMs showing a negative trend of UBI tend to have a stronger strengthening trend of U250 over the midlatitudes (Fig. 8b). These GCMs also tend to have a stronger weakening trend of U250 southeast of the Mediterranean Sea. As a result, the trends of these GCMs likely contribute to the significant regression coefficients of the trend of U250 upstream of the Urals in Fig. 7e. In addition, the positive trend of U250 attains a maximum over the North Atlantic and extends northeastward toward Europe (Figs. 8a,b), which are more pronounced in the GCMs showing a negative trend of UBI (Fig. 8b). Hence, the trend of U250 in Fig. 7e is related to the future change of the Atlantic jet stream (Barnes and Polvani 2013). In other words, the change in the long-term mean of UBI is closely related to the change in the climatological mean state, as discussed in Scaife et al. (2010).
Trend of U250 for the GCMs showing a significant (a) positive or (b) negative trend of UBI in Fig. 6a. Unit: m s−1 (100 yr)−1.
Citation: Journal of Climate 28, 6; 10.1175/JCLI-D-14-00309.1
b. Long-term variance
In addition to the trend of the long-term mean, it is also valuable to investigate the trend of the long-term variance of UBI and SHI, where the positive sign of the trend suggests a larger year-to-year variability in the late twenty-first century. As shown in Figs. 5c and 5d, the MME trend of the long-term variance of UBI is opposite in sign in the two RCP runs, being negative in the RCP4.5 run and positive in the RCP8.5 run. The variance trend of SHI, on the other hand, is insignificant in the RCP4.5 run and positive in the RCP8.5 run. In contrast to the consistent weakening of the long-term mean trend of SHI in the two RCP runs, the sign of its long-term variance trend is not consistent between the two RCP runs, suggesting a lower confidence in projecting the interannual variability of SHI.
In the RCP8.5 run, the trends of the long-term variance of UBI and SHI tend to be positive, and these trends seem to be moderately correlated across the 20 GCMs (Fig. 6b). When the long-term variance trend of UBI is regressed against that of MSLP and Z500 across the GCMs, alternating positive and negative regression centers appear across Eurasia (Figs. 7b,d). On one hand, both Z500 and MSLP have a positive center over the Atlantic, a negative center over Scandinavia, and another positive center over the Ural–Siberia region (Figs. 7b,d). Downstream of the UB sector, on the other hand, the regression coefficient of MSLP is positive and extends southeastward over the climatological SH region (Fig. 7b). In addition, the regression coefficient of Z500 is negative east of Lake Baikal (~120°E) and positive in the vicinity of Japan (~130°–140°E), where the latter is centered over the climatological East Asian trough region (Fig. 7d). These regression maps suggest that the long-term variance trend of UBI is related to the quasi-stationary Rossby wave train signal propagating from the Euro-Atlantic region to the Urals, and such a signal further propagates from the Urals to East Asia and the western North Pacific.
The positive trend of the long-term variance of UBI is further studied by the long-term mean and variance trends of MSLP (Fig. 9). In Fig. 9a, we can identify a robust falling trend of the long-term mean of MSLP over the polar region, which suggests a deepening trend of the Arctic low. Such a falling trend in DJF can also be reproduced by the numerical simulations in response to the reduction of Arctic sea ice (Higgins and Cassano 2009; Screen et al. 2014). As mentioned in Alexander et al. (2004), the reduction of Arctic sea ice increases the local surface heat fluxes, which in turn increases near-surface air temperature, enhances precipitation, and lowers the MSLP locally. Moreover, the stronger heat fluxes during a period of lower sea ice concentrations reduce the vertical static stability of the atmosphere (Jaiser et al. 2012). Because the atmospheric response to sea ice might depend on the mean state of the atmosphere, such as the Euro-Atlantic circulation (Balmaseda et al. 2010), we speculate that the stronger year-to-year variability of the UBI is related to the stronger atmospheric response of the extratropical circulation to the climatological mean state under the reduction of sea ice. As can be seen in Fig. 9b, the long-term variance trend of MSLP is positive over the Barents, Kara, and Laptev Seas, and such a positive trend is robust over the UB sector. The underlying mechanisms linking the reduction of sea ice to the occurrence of UB is still unknown, and this is considered in our future work.
Trend of MSLP in the MME of the RCP 8.5 runs: (a) 30-yr running mean [unit: hPa (100 yr)−1] and (b) 30-yr running variance [unit: hPa2 (100 yr)−2], where shading indicates at least 75% of GCMs have the same sign of change as the MME. The solid black box encloses the UB sector.
Citation: Journal of Climate 28, 6; 10.1175/JCLI-D-14-00309.1
5. Implications of UB for the East Asian winter climate
A significant correlation between UBI and SHI, as well as their trends, was identified in section 4, suggesting that UB plays a crucial role in the mean state and interannual variation of the East Asian winter climate in the twenty-first century. Will the anomalous large-scale circulation features associated with UB strengthen or weaken on synoptic and seasonal time scales in response to increasing anthropogenic forcing? Because the anthropogenic forcing is strongest at the end of the twenty-first century, the analysis is performed for the period 2070–99 only.
a. Synoptic time scales
As in Part I, the development and mature stages are illustrated by the composites of Z500 and the 10-day high-pass filtered E vector at 250 hPa (Hoskins et al. 1983), whereas the mature and decay stages are demonstrated by the composites of MSLP tendencies and anomalies. The composite difference between the RCP runs and the HIST run is used to determine whether the large-scale circulation features are changed and whether the sign of these changes is the same in the majority of GCMs. Because the composite differences are comparable between the RCP4.5 run and the RCP8.5 run, differing mainly in magnitude only, we present only the results of the RCP8.5 run.
The large-scale circulation features in the RCP8.5 run are analogous to those in the HIST run (Figs. 6d–f of Part I). In the development and mature stages of UB, the momentum fluxes propagate eastward along the quasi-stationary Rossby wave train over the Atlantic and Eurasia (Figs. 10a–c). This is related to the eastward shift of an Atlantic storm track. The westerly momentum fluxes diverge from the Euro-Atlantic region and converge over the Ural Mountains. The convergence of the momentum flux over the Ural Mountains corresponds to deceleration of the westerly flow. This is essential for the build-up and maintenance of UB (Figs. 10b,c). In the meantime, the stronger meridional flow of UB enhances the northerly cold air advection from the polar region to western Siberia (downstream of UB). The SH undergoes intensification and the positive MSLP anomaly extends southeastward from the Urals toward the East Asian coast (Figs. 11a–c). In the decay phase of UB, the breakdown of the SH and the migration of a cold surge can be deduced by the negative MSLP tendencies over the Ural–Siberia region and positive MSLP tendencies over the East Asian continent (Figs. 11b,c). Apparently, the dynamics linking UB and the SH do not show significant changes under a warmer climate in the late twenty-first century.
(a)–(c) Composite maps of the Z500 anomalies (shading; unit: gpm) and the 250-hPa E vector (vectors; unit: m2 s−2) on days −2, 0, and 2 with respect to the UB onset obtained from the MME of 20 CMIP5 GCMs for the period of 2070–99 of the RCP8.5 run. (d)–(f) As in (a)–(c), but for the composite difference between the RCP8.5 run and the HIST run. Only the difference significant at the 95% confidence level is plotted.
Citation: Journal of Climate 28, 6; 10.1175/JCLI-D-14-00309.1
As in Fig. 10, but for MSLP anomalies (contours; hPa) and MSLP tendencies (shading; hPa day−1) on days 0, 2, and 4 with respect to the UB onset. The zero contours are omitted and the negative contours are dashed.
Citation: Journal of Climate 28, 6; 10.1175/JCLI-D-14-00309.1
Compared to the HIST run, the eastward propagation of the momentum fluxes from the Euro-Atlantic region becomes stronger in the majority of GCMs (Figs. 10d–f). In the development stage, there are stronger momentum fluxes traveling from Europe to the Urals (Fig. 10d). This is likely due to changes in the background flow over the Euro-Atlantic region, such as the intensification of the eddy-driven jet over the North Atlantic (Fig. 8) and the lower MSLP over the high-latitude region (Fig. 9a). These changes are apparent in the MME differences between the RCP8.5 run and the HIST run, as can be inferred from the change of U500 (Fig. 2f).
Associated with stronger momentum fluxes propagating from Europe and converging near the Urals, the positive MSLP anomalies over the Ural–Siberia region appear to be stronger in the majority of GCMs, where the larger MSLP anomalies have two local maxima over the UB center and the SH (Fig. 11d). Besides larger MSLP anomalies, the positive MSLP tendencies are larger over the SH region, suggesting stronger intensification of the SH (Fig. 11d). Following that, the positive MSLP tendencies are larger over China, suggesting the stronger passage of a cold surge associated with the breakdown of the SH (Figs. 11e,f).
b. Seasonal time scale
On the seasonal time scale, the large-scale circulation features associated with UB are depicted by regression analysis, where the 30-yr DJF-mean time series of UBI is regressed against that of Z500 and MSLP for each of the 20 GCMs. The MMEs of the regression pattern of the HIST run and the RCP runs are shown in Fig. 12.
Regression of the UBI against the MSLP and Z500 (a),(b) in the HIST run for the period 1971–2000, and in the (c),(d) RCP4.5 and (e),(f) RCP8.5 runs for the period 2070–99. The zero regression coefficients are omitted, the negative contours are dashed, and the coefficients significant at the 95% confidence level are shaded. The contour interval is 0.2 hPa %−1 for the MSLP and 2 gpm %−1 for the Z500. The solid black box and dashed black box enclose the UB sector and the climatological SH region, respectively. In the left panel, the Tibetan Plateau is dark shaded.
Citation: Journal of Climate 28, 6; 10.1175/JCLI-D-14-00309.1
The large-scale teleconnections of UB is comparable in the HIST run and the RCP runs (Figs. 12a–f). Corresponding to the more frequent occurrence of UB, the stronger high pressure system over the Urals can be inferred from the positive regression coefficients of both MSLP and Z500 over the region (Figs. 12a–f). Upstream of UB, the regression coefficients of Z500 are negative near the Mediterranean Sea (~10°–30°E; Figs. 12b,d,f). Downstream of UB, the regression coefficients of MSLP are positive in sign (Figs. 12a,c,e), and those of Z500 are negative in sign over the southern flank of the climatological SH region, suggesting a stronger baroclinicity over the East Asian winter monsoon region (Fig. 12b,d,f). Such a teleconnection pattern is comparable to that associated with the dominant modes of the East Asian winter monsoon and UB (B. Wang et al. 2010; Sohn et al. 2011; Cheung et al. 2012).
Compared to the HIST run (Figs. 12a,b), the regression coefficients of the two RCP runs are larger in magnitude near the regression centers, and most of the significant regions expand (Figs. 12c–f). In particular, the positive regression coefficients of MSLP and Z500 over the Urals expand poleward to the Kara Sea and Laptev Sea. Wang and Chen (2014) also found a stronger relationship between the UBI and the MSLP and Z500 of the subpolar region. Their study showed that the recent strong East Asian winter monsoon epoch (2004–12) was characterized by pronounced Arctic warming and stronger high anomaly regions over the Kara Sea and Laptev Sea, where these characteristics could not be identified in the strong East Asian winter monsoon epoch (1976–87). Therefore, the difference in the regression patterns between the RCP runs and the HIST run might arise from the change in the atmospheric response over the extratropical region under Arctic warming, which needs to be further explored in future work.
6. Summary and discussion
In Part II of this study, we have projected the UB frequency and 30-yr running (long-term) mean and variance trends of UBI (Ural blocking index) and SHI (Siberian high intensity) in 20 CMIP5 GCMs in RCP4.5 and 8.5 runs. Our results show that UB still plays an important role in the East Asian winter climate on both synoptic and seasonal time scales in the late twenty-first century when anthropogenic forcing is strongest. An accurate projection of UB might be considered one of the crucial factors for the evaluation of future changes of the East Asian winter climate. The main findings of this study are summarized as follows:
The MME of the UB frequency does not show a significant change in the twenty-first century in the two RCP runs, but a large spread is identified across the GCMs and is even larger in the RCP8.5 run. This suggests a larger uncertainty in the UB frequency under a mild climate, where the long-term variance of UBI tends to increase in the RCP8.5 run. The uncertainty might be related to a change in the atmospheric response of extratropical regions under climate change. This might also be related to the uncertainty of the central blocking longitude over Eurasia, which tends to shift eastward toward Asia associated with the eastward shift of the Atlantic storm tracks in a warmer climate (de Vries et al. 2013). The characteristics of blocking should also be assessed in future work.
The long-term mean and variance trends of the UBI and the SHI (a proxy of the East Asian winter climate) across the GCMs are related. On one hand, the long-term mean trend of UBI might be related to the trend of the Atlantic jet stream. A significant negative trend of UBI is likely associated with a strengthening trend and a northeastward displacement of the Atlantic jet stream. On the other hand, the long-term variance trend of UBI might be related to the quasi-stationary Rossby wave train across Eurasia. The uncertainty of UBI seems to be related to how the climatological mean state over the Euro-Atlantic region will change. Based on these results, we also hypothesize that the uncertainty of UBI will cause uncertainty in the East Asian winter climate.
The linkage between the long-term mean of UBI and the East Asian winter climate appears to be stable in the twenty-first century, but the large-scale circulation features associated with UB in the late twenty-first century might be different from those in the HIST period. On interannual time scales, regression analyses reveal that UB shows a significant correlation with the Z500 and MSLP over the Kara and Laptev Seas, where their correlations are weak in the HIST period. Indeed, the reduction of autumn Arctic sea ice in recent years has been linked to the warm Arctic–cold Siberia temperature pattern (Cohen et al. 2012). This also likely contributes to more frequent UB and more frequent cold winters in East Asia (Wu et al. 2011, 2013; Wang and Chen 2014). Therefore, the effect of Arctic sea ice on the UB frequency under a warmer climate is noteworthy.
The underlying dynamics of UB do not have notable changes in the late century. The formation and maintenance of UB is related to the eastward shift of an Atlantic storm track. The associated momentum fluxes converge near the Ural Mountains result in deceleration of the westerly flow over there. However, because of apparent changes of the large-scale circulation over the Euro-Atlantic region, such eastward propagating momentum fluxes appear to be stronger and these result in a stronger intensification of UB. Then, UB might exert a stronger impact on the synoptic evolution of SH and the cold air outbreak in East Asia.
Acknowledgments
The first author is a recipient of a research studentship provided by the City University of Hong Kong (CityU). This study is partly supported by National Nature Science Foundation of China Grants 41375096 and 41175079, and CityU Strategic Research Grant 7004004. Last but not least, we greatly appreciate the critical and detailed comments provided by the three anonymous reviewers.
REFERENCES
Alexander, M. A., U. S. Bhatt, J. E. Walsh, M. S. Timlin, J. S. Miller, and J. D. Scott, 2004: The atmospheric response to realistic Arctic sea ice anomalies in an AGCM during winter. J. Climate, 17, 890–905, doi:10.1175/1520-0442(2004)017<0890:TARTRA>2.0.CO;2.
Balmaseda, M. A., L. Ferranti, F. Molteni, and T. N. Palmer, 2010: Impact of 2007 and 2008 Arctic ice anomalies on the atmospheric circulation: Implications for long-range predictions. Quart. J. Roy. Meteor. Soc., 136, 1655–1664, doi:10.1002/qj.661.
Barnes, E. A., and L. Polvani, 2013: Response of the midlatitude jets, and of their variability, to increased greenhouse gases in the CMIP5 models. J. Climate, 26, 7117–7135, doi:10.1175/JCLI-D-12-00536.1.
Barnes, E. A., J. Slingo, and T. Woollings, 2011: A methodology for the comparison of blocking climatologies across indices, models and climate scenarios. Climate Dyn., 38, 2467–2481, doi:10.1007/s00382-011-1243-6.
Barriopedro, D., R. García-Herrera, A. R. Lupo, and E. Hernández, 2006: A climatology of Northern Hemisphere blocking. J. Climate, 19, 1042–1063, doi:10.1175/JCLI3678.1.
Cheung, H. H. N., and W. Zhou, 2015: Implications of Ural blocking for the East Asian winter climate in CMIP5 GCMs. Part I: Biases in the historical scenario. J. Climate, 28, 2203–2216, doi:10.1175/JCLI-D-14-00308.1.
Cheung, H. N., W. Zhou, H. Y. Mok, and M. C. Wu, 2012: Relationship between Ural–Siberian blocking and the East Asian winter monsoon in relation to the Arctic Oscillation and the El Niño–Southern Oscillation. J. Climate, 25, 4242–4257, doi:10.1175/JCLI-D-11-00225.1.
Cheung, H. N., W. Zhou, Y. Shao, W. Chen, H. Y. Mok, and M. C. Wu, 2013: Climatology and characteristics of wintertime atmospheric blocking over Ural–Siberia. Climate Dyn., 41, 63–79, doi:10.1007/s00382-012-1587-6.
Christensen, J. H., and Coauthors, 2014: Climate phenomena and their relevance for future regional climate change. Climate Change 2013: The Physical Science Basis, T. F. Stocker et al., Eds., Cambridge University Press, 1217–1308.
Cohen, J. L., J. C. Furtado, M. A. Barlow, V. A. Alexeev, and J. E. Cherry, 2012: Arctic warming, increasing snow cover and widespread boreal winter cooling. Environ. Res. Lett., 7, 014007, doi:10.1088/1748-9326/7/1/014007.
de Vries, H., T. Woollings, J. Anstey, R. J. Haarsma, and W. Hazeleger, 2013: Atmospheric blocking and its relation to jet changes in a future climate. Climate Dyn., 41, 2643–2654, doi:10.1007/s00382-013-1699-7.
Diao, Y., J. Li, and D. Luo, 2006: A new blocking index and its application: Blocking action in the Northern Hemisphere. J. Climate, 19, 4819–4839, doi:10.1175/JCLI3886.1.
Dunn-Sigouin, E., and S.-W. Son, 2013: Northern Hemisphere blocking frequency and duration in the CMIP5 models. J. Geophys. Res. Atmos., 118, 1179–1188, doi:10.1002/jgrd.50143.
Francis, J. A., and S. J. Vavrus, 2012: Evidence linking Arctic amplification to extreme weather in mid-latitudes. Geophys. Res. Lett., 39, L06801, doi:10.1029/2012GL051000.
Higgins, M. E., and J. J. Cassano, 2009: Impacts of reduced sea ice on winter Arctic atmospheric circulation, precipitation, and temperature. J. Geophys. Res. Atmos., 114, D16107, doi:10.1029/2009JD011884.
Hoskins, B. J., I. N. James, and G. H. White, 1983: The shape, propagation, and mean-flow interaction of large-scale weather system. J. Atmos. Sci., 40, 1595–1612, doi:10.1175/1520-0469(1983)040<1595:TSPAMF>2.0.CO;2.
Hu, Z., L. Bengtsson, and K. Arpe, 2000: Impact of global warming on the Asian winter monsoon. J. Geophys. Res., 105, 4607–4624, doi:10.1029/1999JD901031.
Jaiser, R., K. Dethloff, D. Handorf, A. Rinke, and J. Cohen, 2012: Impact of sea ice cover changes on the Northern Hemisphere atmospheric winter circulation. Tellus, 64A, 11595, doi:10.3402/tellusa.v64i0.11595.
Lejenäs, H., and H. Økland, 1983: Characteristics of Northern Hemisphere blocking as determined from long time series of observational data. Tellus, 35A, 350–362, doi:10.1111/j.1600-0870.1983.tb00210.x.
Luo, D., W. Zhou, and K. Wei, 2010: Dynamics of eddy-driven North Atlantic Oscillations in a localized shifting jet: Zonal structure and downstream blocking. Climate Dyn., 34, 73–100, doi:10.1007/s00382-009-0559-y.
Lupo, A. R., and P. J. Smith, 1995: Climatological features of blocking anticyclones in the Northern Hemisphere. Tellus, 47A, 439–456, doi:10.1034/j.1600-0870.1995.t01-3-00004.x.
Masato, G., B. J. Hoskins, and T. Woollings, 2013: Winter and summer Northern Hemisphere blocking in CMIP5 models. J. Climate, 26, 7044–7059, doi:10.1175/JCLI-D-12-00466.1.
Masato, G., T. Woollings, and B. J. Hoskins, 2014: Structure and impact of atmospheric blocking over the Euro-Atlantic region in present-day and future simulations. Geophys. Res. Lett., 41, 1051–1058, doi:10.1002/2013GL058570.
Matsueda, M., R. Mizuta, and S. Kusunoki, 2009: Future change in wintertime atmospheric blocking simulated using a 20-km-mesh atmospheric global circulation model. J. Geophys. Res., 114, D12114, doi:10.1029/2009JD011919.
Moss, R. H., and Coauthors, 2010: The next generation of scenarios for climate change research and assessment. Nature, 463, 747–756, doi:10.1038/nature08823.
Omrani, N.-E., N. S. Keenlyside, J. Bader, and E. Manzini, 2014: Stratospheric key for wintertime atmospheric response to warm Atlantic decadal conditions. Climate Dyn., 42, 649–663, doi:10.1007/s00382-013-1860-3.
Pelly, J. L., and B. J. Hoskins, 2003: A new perspective on blocking. J. Atmos. Sci., 60, 743–755, doi:10.1175/1520-0469(2003)060<0743:ANPOB>2.0.CO;2.
Rex, D. F., 1950: Blocking action in the middle troposphere and its effect upon regional climate. Part II: The climatology of blocking action. Tellus, 2A, 275–301, doi:10.1111/j.2153-3490.1950.tb00339.x.
Scaife, A. A., T. Woollings, J. Knight, G. Martin, and T. Hinton, 2010: Atmospheric blocking and mean biases in climate models. J. Climate, 23, 6143–6152, doi:10.1175/2010JCLI3728.1.
Scaife, A. A., and Coauthors, 2011: Improved Atlantic winter blocking in a climate model. Geophys. Res. Lett., 38, L23703, doi:10.1029/2011GL049573.
Screen, J. A., C. Deser, I. Simmonds, and R. Tomas, 2014: Atmospheric impacts of Arctic sea-ice loss, 1979–2009: Separating forced change from atmospheric internal variability. Climate Dyn., 43, 333–344, doi:10.1007/s00382-013-1830-9.
Sohn, S.-J., C.-Y. Tam, and C.-K. Park, 2011: Leading modes of East Asian winter climate variability and their predictability: An assessment of the APCC multi-model ensemble. J. Meteor. Soc. Japan, 89, 455–474, doi:10.2151/jmsj.2011-504.
Takaya, K., and H. Nakamura, 2005: Mechanisms of intraseasonal amplification of the cold Siberian high. J. Atmos. Sci., 62, 4423–4440, doi:10.1175/JAS3629.1.
Tibaldi, S., and F. Molteni, 1990: On the operational predictability of blocking. Tellus, 42A, 343–365, doi:10.1034/j.1600-0870.1990.t01-2-00003.x.
Tyrlis, E., and B. J. Hoskins, 2008a: Aspects of a Northern Hemisphere atmospheric blocking climatology. J. Atmos. Sci., 65, 1638–1652, doi:10.1175/2007JAS2337.1.
Tyrlis, E., and B. J. Hoskins, 2008b: The morphology of Northern Hemisphere blocking. J. Atmos. Sci., 65, 1653–1665, doi:10.1175/2007JAS2338.1.
Wang, B., Z. Wu, C. Chang, J. Liu, J. Li, and T. Zhou, 2010: Another look at interannual-to-interdecadal variations of the East Asian winter monsoon: The northern and southern temperature modes. J. Climate, 23, 1495–1512, doi:10.1175/2009JCLI3243.1.
Wang, L., and W. Chen, 2014: The East Asian winter monsoon: Re-amplification in the mid-2000s. Chin. Sci. Bull., 59, 430–436, doi:10.1007/s11434-013-0029-0.
Wang, L., W. Chen, W. Zhou, J. C. L. Chan, D. Barriopedro, and R. Huang, 2010: Effect of the climate shift around mid 1970s on the relationship between wintertime Ural blocking circulation and East Asian climate. Int. J. Climatol., 30, 153–158, doi:10.1002/joc.1876.
Woollings, T., J. M. Gregory, J. G. Pinto, M. Reyers, and D. J. Brayshaw, 2012: Response of the North Atlantic storm track to climate change shaped by ocean–atmosphere coupling. Nat. Geosci., 5, 313–317, doi:10.1038/ngeo1438.
Woollings, T., B. Harvey, and G. Masato, 2014: Arctic warming, atmospheric blocking and cold European winters in CMIP5 models. Environ. Res. Lett., 9, 014002, doi:10.1088/1748-9326/9/1/014002.
Wu, B., J. Su, and R. Zhang, 2011: Effects of autumn–winter Arctic sea ice on winter Siberian high. Chin. Sci. Bull., 56, 3220–3228, doi:10.1007/s11434-011-4696-4.
Wu, B., D. Handorf, K. Dethloff, A. Rinke, and A. Hu, 2013: Winter weather patterns over northern Eurasia and Arctic sea ice loss. Mon. Wea. Rev., 141, 3786–3800, doi:10.1175/MWR-D-13-00046.1.
Zhou, W., J. C. L. Chan, W. Chen, J. Ling, J. G. Pinto, and Y. Shao, 2009: Synoptic-scale controls of persistent low temperature and icy weather over southern China in January 2008. Mon. Wea. Rev., 137, 3978–3991, doi:10.1175/2009MWR2952.1.
