1. Introduction
East Asian winter monsoon (EAWM) is one of the most active systems in boreal winter, and the EAWM variations can significantly influence the climate over East Asia and its surrounding areas (Ding 1994; Chang et al. 2006; Chen et al. 2019; Ma and Chen 2021). The anomalously strong (weak) EAWM causes colder (warmer) winters over East Asia, and very strong EAWM can lead to extreme cold events, such as cold surges, freezing, and snowstorm hazards in East Asia (Gu et al. 2008; Sun et al. 2010; Huang et al. 2016; Chen and Dong 2019), consequently exerting serious impacts on people’s livelihoods, society, and natural environment over the region. Therefore, exploring EAWM variations and the mechanisms is of high significance.
EAWM variations on different time scales are hot topics of previous studies, among which the decadal time scale is an important one. Observational analysis indicates that the EAWM presents a decadal weakness around the mid-1980s (Ding et al. 2014; Sun et al. 2016; Miao et al. 2018; Ma and Chen 2021) and an enhancement around the early twenty-first century (Wang and Chen 2014; Xiao et al. 2016). Many studies have focused on the impact factors and mechanism of the decadal EAWM variation, such as the Arctic Oscillation (AO; Jhun and Lee 2004; Miao et al. 2020), quasi-stationary planetary wave activity (Wang et al. 2009), the Pacific decadal oscillation (PDO; Zhou et al. 2007; Ding et al. 2014; Wang et al. 2018), Arctic sea ice (Wang and Chen 2014), and thermal conditions over the Tibetan Plateau (L. Chen et al. 2020). However, from the perspective of long-term variation, the PDO and AO cannot well explain the decadal EAWM variation (Sun et al. 2016; Miao et al. 2020). According to Sun et al. (2016), the northwestern North Pacific sea surface temperature (NPSST) anomaly is significantly responsible for the decadal EAWM variation from the 1950s to 2010s. However, limited by the data length and the compound of climate internal variability and external forcings in the reanalysis, two questions still remain which were not addressed by Sun et al. (2016). The first is whether the relationship between the EAWM and the northwestern NPSST anomaly is stable. The other is whether the decadal variation of the EAWM originates from climate internal variability. The two questions are valuable for deeply understanding and predicting the decadal EAWM variation and merit being researched.
The limited instrumental observations hinder the comprehensive investigation of the climate internal variability on the decadal time scale. In addition, the complex influence of external forcings adds difficulty to the study of climate internal variability using the observed data. Model simulation is a useful tool to explore these questions. Previous studies have shown that current coupled models can well reproduce the characteristics of EAWM-related circulations (He and Wang 2012; Miao et al. 2018; Jiang et al. 2020), as well as the decadal variations of the EAWM-related circulation around the 1980s (Wang and He 2012; Miao et al. 2018). In addition, Jiang et al. (2020) concluded that phase 6 of the Coupled Model Intercomparison Project (CMIP6) models simulate EAWM variability better than CMIP5 models. Therefore, some studies have used coupled climate models to distinguish the influence of climate internal variability and external forcings on the East Asian winter climate. Miao and Jiang (2022) used 12 models’ all-forcing and single-forcing simulations from CMIP6, respectively, to quantify the contribution of climate internal variability and external forcings to the East Asian surface air temperature variations. They found that climate internal variability can account for 59% of multidecadal variation in the air temperature. Though several studies have investigated the performance of climate models in simulating the decadal EAWM variation, there is still a lack of comprehensive knowledge on the decadal EAWM variation and the related mechanism.
To address the abovementioned issues, in this study, we investigate the decadal variation of the EAWM and the possible mechanism using multiple long-term preindustrial simulations within CMIP6. The paper is constructed as follows. The data and analysis methods are introduced in section 2. Section 3 evaluates the performance of CMIP6 models in simulating the decadal EAWM variation. The possible mechanism responsible for the decadal EAWM variation is further explored in section 4. The discussion and conclusions are drawn in section 5.
2. Data and methods
a. Data
The atmospheric circulation dataset used in this study is the monthly and daily mean Twentieth Century Reanalysis, version 3 (20CRv3) (Slivinski et al. 2019), from the National Oceanic and Atmospheric Administration (NOAA) Earth System Research Laboratory (ESRL). The NOAA-20CRv3 data are available from 1836 to 2015 and gridded at a resolution of 2° × 2°. The monthly mean SST data are from the Hadley Centre Sea Ice and Sea Surface Temperature dataset (HadISST; Rayner et al. 2003), gridded at a resolution of 1° × 1° over 1870–2021. To verify the robustness of our results, the European Centre for Medium-Range Weather Forecasts (ECMWF) twentieth-century reanalysis (ERA-20C) (Poli et al. 2015, 2016) and the Extended Reconstructed SST, version 5 (ERSSTv5), from the NOAA (Huang et al. 2017) are also used. The ERA-20C data are gridded at a resolution of 2.5° × 2.5° over 1900–2010, and the ERSSTv5 are gridded at a resolution of 2° × 2° over 1854–2021. In this study, we analyze the data from 1930 because the available observed data used in Twentieth-Century Reanalysis are relatively sparse before 1930 (Befort et al. 2016; Slivinski et al. 2021).
The preindustrial control (piControl) simulations from 30 coupled models of CMIP6 are used (Eyring et al. 2016). Detailed information on the selected models, including the institutes, horizontal and vertical resolutions, and time spans, are given in Table 1. The conditions prior to the onset of large-scale industrialization are imposed on the piControl simulations, with 1850 being the reference year. The piControl simulation excludes natural and anthropogenic forcings, so it only exhibits unforced climate internal variability of the climate system. All selected coupled models have simulations over 450 years. We use the monthly mean data and only analyze one ensemble from each coupled model. Because these models and reanalysis have different horizontal resolutions, for convenience, the CMIP6’s datasets are linearly interpolated onto the same grid resolution of 2° × 2°.
List of the CMIP6 models used in this study.
b. Definition
In this paper, the definition of the EAWM index (EAWMI) is the negative mean geopotential height at 500 hPa over the region of 25°–45°N, 110°–150°E, where the East Asian trough is located (Wang and He 2012). The positive EAWMI indicates the enhancement of the East Asian trough, which corresponds to a stronger-than-normal EAWM. We focus on the decadal variation in the boreal winter (December–February). The winter of 1931 refers to the average of December 1930–February 1931.
c. Methods
3. Evaluation of the CMIP6 models
a. Time period
Figure 1a shows the 11-yr low-pass-filtered EAWMI calculated from NOAA-20CRv3 and ERA-20C during 1931–2010. The EAWM shows a decadal variation and a phase shift around 1986/87, which is consistent with previous studies (Ding et al. 2014; Sun et al. 2016; Ma and Chen 2021). The correlation coefficient of the decadal EAWMI between the two datasets is 0.83, significant at the 99% confidence level. The power spectrum analysis shows that the EAWMI has a significant 10–30-yr period in both NOAA-20CRv3 and ERA-20C (Figs. 1b,c). Moreover, we repeat the spectrum analysis by the unfiltered data. Both the NOAA-20CRv3 and ERA-20C indicate a significant periodicity of about 20-yr in the EAWM indices (figure not shown). So, the above-revealed 10–30-yr period is not an artifact of low-pass filtering.
(a) The series of normalized 11-yr low-pass-filtered EAWMI from NOAA-20CRv3 (red line) and ERA-20C (black line). The power spectrum (black line) of normalized 11-yr low-pass-filtered EAWMI from (b) NOAA-20CRv3 and (c) ERA-20C. The red and blue lines in (b) and (c) indicate the red noise spectrum and 95% confidence level bound, respectively.
Citation: Journal of Climate 38, 3; 10.1175/JCLI-D-23-0704.1
To explore the robustness of the EAWMI in describing the monsoon variation, we calculate another EAWM index defined by Li and Yang (2010) (marked as EAWMI_LY). Figure S1a in the online supplemental material shows the EAWMI_LY derived from the NOAA-20CRv3 and ERA-20C. The correlation coefficient is 0.71 (0.70) between the decadal EAWMI and EAWMI_LY derived from the NOAA-20CRv3 (ERA-20C), indicating the high consistency of the two indices in representing the decadal EAWM variation. The power spectra of the NOAA-20CRv3 and ERA-20C EAWMI_LY are shown in Figs. S1b and S1c, also exhibiting a dominant period of 10–30 years, albeit with a smaller significant peak period than the EAWMI. Considering the high consistency of the two EAWM indices, we mainly focus on the results derived from the EAWMI in the following analysis.
Such a 10–30-yr period can also be found in the model simulations (Fig. 2), so the decadal variation of observed EAWMI can be well captured by the CMIP6 piControl simulations. The piControl simulation only exhibits unforced climate internal variability of the climate system because it excludes natural and human-induced forcings. Therefore, we deduce that the decadal variation of the EAWM may originate from climate internal variability and external forcings may only modulate the decadal EAWM variation. Additionally, 22 CMIP6 models (the last 22 in Fig. 2) exhibit a second decadal peak of EAWMI with a longer period, which is not found in either NOAA-20CRv3 or ERA-20C. The possible reasons could be that the length of observed data is too short to detect the longer period variation or the models incorrectly simulate the longer period variation of the EAWM.
Power spectrum (black lines) of normalized 11-yr low-pass-filtered EAWMI in the CMIP6 models. The red and blue lines indicate the red noise spectrum and 95% confidence level bound, respectively.
Citation: Journal of Climate 38, 3; 10.1175/JCLI-D-23-0704.1
b. Spatial structure
We evaluate the CMIP6 models’ performances in simulating the winter climatology of the large-scale monsoonal circulation over 25°–50°N, 100°E–180°, which is an extensive region covering the key region of the East Asian trough. The Taylor diagram includes the pattern correlation coefficient (PCC) and the ratio of standard deviation (RSD) between the models and the reanalysis (Fig. 3). The PCCs of all used models exceed 0.95 and RSDs of all models range from 0.75 to 1.2, indicating that all used models exhibit a good performance in simulating the spatial pattern of the East Asian trough. We also calculate the ratio of the standardized deviation of the decadal EAWMI between the model simulations and reanalysis. All of the ratios are below 1.0, and the 25th and 75th percentiles are 0.74 and 0.86 (0.70 and 0.81), referring to the NOAA-20CRv3 (ERA-20C) reanalysis, respectively, indicating that the decadal EAWM variation forced by the solo climate internal variability is somewhat weak. The above analysis qualitatively measures the contribution of the climate internal variability to the decadal EAWM variation. In future, we shall compare the EAWM variability in the CMIP and Atmospheric Model Intercomparison Project (AMIP) experiments and in the stand-alone atmospheric model simulations forced with climatological SSTs, to deepen our understanding of the oceanic impacts on the EAWM variation.
Taylor diagram of climatological winter geopotential height at 500 hPa over the key region (25°–50°N, 100°E–180°) of the East Asian trough calculated from CMIP6 models and (a) NOAA-20CRv3, (b) ERA20C, and (c) average of NOAA-20CRv3 and ERA20C.
Citation: Journal of Climate 38, 3; 10.1175/JCLI-D-23-0704.1
The CMIP6 piControl simulations can well capture the decadal EAWM variation and spatial structure, implying that the decadal EAWM variation mainly originates from internal variability of the climate system. And all piControl simulations used in this study span over 450 years, compensating for the limited length of observed data. Therefore, they are suitable for further investigating the mechanism of the decadal EAWM variation.
We further evaluate the models’ skill in simulating the spatial structure of the EAWM-related atmospheric circulation anomalies. Figures 4a and 4b show the regressions of the 500-hPa geopotential height and horizontal wind anomalies against the decadal NOAA-20CRv3 and ERA-20C EAWMI, respectively. The two reanalysis results display the significant negative geopotential height anomalies over the key region of the East Asian trough and the intensified northerly wind along the East Asian coast, with respect to the positive EAWMI (Figs. 4a,b). In addition, the aforementioned negative geopotential height anomalies extend eastward to the midlatitude North Pacific, constituting a dipole pattern with the positive ones over the high-latitude North Pacific.
Regression of winter geopotential height (color; m) and horizontal wind anomalies (vector; m s−1) at 500 hPa against the EAWMI derived from the (a) NOAA-20CRv3 and (b) ERA-20C data. The dotted areas and green vectors are significant at the 95% confidence level. All data are 11-yr low-pass filtered before analysis.
Citation: Journal of Climate 38, 3; 10.1175/JCLI-D-23-0704.1
Similar regression results in the piControl simulations are analyzed and shown in Fig. S2. All the models could simulate that the strengthened EAWM is associated with intensified northerly wind along the East Asian coast and significant negative geopotential height anomalies extending from East Asia to the midlatitude North Pacific at 500 hPa, consistent with the reanalysis results (Figs. 4a,b). To objectively evaluate the models’ skill in reproducing the EAWM-related atmospheric structure, we calculate the PCC of the 500-hPa geopotential height anomalies between the reanalysis results and model simulations over 25°–50°N, 100°E–180°. Such an extent is larger than that defining the EAWM key region to reflect the integrated EAWM-related atmospheric circulation. The results show that PCCs of each model contain at least one value greater than 0.70. Accordingly, the models are capable of reproducing the EAWM-related atmospheric structure.
4. Influence of the northwestern NPSST anomaly on the EAWM
a. Relationship and the physical processes
On a long time scale, the sources of climate decadal variability are complex. Some observational evidence and model simulations have validated the ocean’s role in contributing to the atmospheric decadal variability (Latif and Barnett 1994; An 2008; Gulev et al. 2013). For example, the Atlantic multidecadal oscillation (AMO) and tropical Pacific and Indian Ocean SST decadal variation contribute to the decadal weakening of the East Asian summer monsoon around the late 1970s (Lu et al. 2006; Li et al. 2008; Li et al. 2010). The PDO and AMO are regarded as the important SST factors responsible for the decadal EAWM variation through the triggered atmospheric teleconnection patterns (Li and Xian 2003; Yang et al. 2004; Ding et al. 2014).
To investigate the SST anomalies associated with the decadal variation of the EAWM, we perform 11-yr low-pass filtering on EAWMI and SST anomaly field and then calculate the correlations between them. The correlation results show significant negative correlations over the northwestern North Pacific (Fig. 5). Accordingly, an SST index is defined as the regional mean SST over the northwestern North Pacific (26°–40°N, 120°E–180°), marked as northwestern NPSST index (NPSSTI). After 11-yr low-pass filtering, the correlation coefficient between the NOAA-20CRv3 EAWMI and the HadISST SST index is −0.75, significant at the 99% confidence level. And the correlation coefficients from other data are significant at the 95% confidence level. This result is consistent with Sun et al. (2016), which indicates that the northwestern NPSST anomaly plays an important role in the decadal variation of the EAWM.
The correlation distributions between the EAWMI and SST anomalies. SST datasets are (a),(b) HadISST and (c),(d) ERSSTv5. EAWMI are calculated separately from (a), (c) NOAA-20CRv3 and (b), (d) ERA-20C. The dotted areas are significant at 95% confidence level. The rectangle is the area used to calculate the SST index. The value on the top-right corner of each panel indicates the correlation coefficient between EAWMI and northwestern NPSSTI, and the value in the parenthesis indicates the confidence level. All data are 11-yr low-pass filtered before analysis.
Citation: Journal of Climate 38, 3; 10.1175/JCLI-D-23-0704.1
In addition, we perform power spectrum analysis on the northwestern NPSSTI from HadISST and ERSSTv5 (Fig. S3). Both the northwestern NPSSTIs have a dominant periodicity of 10–30 years. The consistency of the dominant period between the EAWMI and SST indices further confirms the northwestern NPSST anomaly’s contribution to the decadal EAWM variation. Because the results from the two Twentieth-Century Reanalysis and two SST datasets are consistent; therefore, we mainly use the NOAA-20CRv3 and HadISST data in the following analysis. We have also checked the correlation coefficient between the EAWMI_LY and the northwestern NPSSTI. The correlation coefficient is −0.58 between the NOAA-20CRv3 EAWMI_LY and the HadISST northwestern NPSSTI, significant at the 95% confidence level, which further confirms the connection between the northwestern NPSST anomaly and the EAWM.
Along with the close relationship between the northwestern NPSST anomaly and the EAWM, the next question is how can the SST anomaly influence the EAWM. To answer this question, we first diagnose the thermodynamical processes related to the northwestern NPSST anomaly. To facilitate analysis, we provide a regression map of the SST anomalies onto the decadal northwestern NPSSTI, which shows significant positive signals over the North Pacific region (Fig. 6a). Subsequently, the regressions of the precipitation and Q1 anomalies against the decadal northwestern NPSSTI are shown in Figs. 6b and 6c. More-than-normal precipitation is over the warm northwestern North Pacific, especially over its northern part (Fig. 6b). Associated with the increasing precipitation are the latent heat release and positive Q1 at the 700–300 hPa in the latitude belt of 30°–40°N (Fig. 6c). By this way, the northwestern NPSST anomaly could impact the overlying air temperature. The warm northwestern NPSST anomaly–related surface heat flux and radiation anomalies are shown in Figs. 6d–g. At the air–sea interface, approximately homogeneous positive latent heat flux anomalies are over the northwestern North Pacific, indicating the warm SST could heat the above air by releasing latent heat (Fig. 6d). An east–west dipole pattern of the sensible heat flux anomalies is over the northwestern North Pacific, while the intensity is weaker than the latent heat flux (Fig. 6e). The overall positive turbulent heat flux (the sum of the latent and sensible heat flux) anomalies over the warm SST region suggest the potential SST’s forcing to the overlying atmosphere. In addition, we can find significantly negative shortwave radiation anomalies over the western Pacific from the equator to around 30°N (Fig. 6f), indicating the ocean absorbs more energy. In turn, the ocean will heat the above air column through emitting more longwave radiation (Fig. 6g). Collectively, the northwestern NPSST anomaly could change the air temperature via the latent heat process and longwave radiation in its northwestern and southern parts, respectively.
(a) Regression of winter SST anomalies (°C) against the northwestern NPSSTI derived from the HadISST data. (b) Regression of winter NOAA-20CRv3 precipitation anomaly (mm day−1) against the northwestern NPSSTI derived from the HadISST data. (c) As in (b), but for the latitude–pressure cross section of the regression of Q1 (K day−1) averaged along 120°E–180°. (d)–(g) As in (b), but for the regression of (d) latent heat flux, (e) sensible heat flux, (f) shortwave radiation, and (g) longwave radiation anomalies. The positive values in (d)–(g) are directed upward (W m−2). The dotted areas are significant at the 95% confidence level. All data are 11-yr low-pass filtered before analysis.
Citation: Journal of Climate 38, 3; 10.1175/JCLI-D-23-0704.1
Through the above processes, the warm northwestern NPSST anomaly could heat the overlying atmosphere, resulting in enhanced (weakened) meridional temperature gradients over mid- to low (mid- to high) latitudes (Fig. 7a). The meridional temperature gradient can change the zonal wind according to the thermal wind theory. Therefore, the zonal westerlies are significantly enhanced to the north of 40°N and weakened to the south of 40°N (Fig. 7b), which leads to a weakened East Asian trough and EAWM (Fig. 7c). This observational physical process is consistent with previous research (Sun et al. 2016), while they used the short-term atmospheric reanalysis of the National Centers for Environmental Prediction Reanalysis-1 (NCEP-1) and 40-yr ECMWF Re-Analysis (ERA-40). The above analysis results are still not enough to discern the cause and effect. Sun et al. (2016) have further conducted numerical simulations by the Community Atmosphere Model, version 5 (CAM5), to explore the role of the northwestern North Pacific SST anomaly. They designed two experiments: One was forced by the climatological SST and sea ice concentration, which was referred to as the control run. The other was an SST-sensitivity experiment, which had the same model configuration as the control run but was forced by the combined SST field that was obtained by superimposing the warm northwestern NPSST anomaly on the climatological SST in winter. The atmospheric circulation difference between the SST-sensitivity experiment and control run showed the significant positive 500-hPa geopotential height anomalies over East Asia, validating the SST’s forcing over the northwestern North Pacific on the East Asia winter atmospheric circulation. The above analysis demonstrates the impact of the northwestern NPSST anomaly on the EAWM via the thermodynamic process. Theoretically, due to the maximum Q1 anomalies at the middle troposphere, the underlying warm SST anomaly possibly leads to an atmospheric anomaly pattern with a baroclinic structure. However, when inspecting the vertical structure of the atmospheric anomalies, we can find an equivalent barotropic ridge over the northwestern North Pacific (Fig. 7d). Such a difference hints at the important role of the atmospheric dynamical processes in shaping the connection between the EAWM and the northwestern NPSST anomaly.
(a) The latitude–pressure cross section of the regression of NOAA-20CRv3 meridional temperature gradient (10−2 °C/lat-degree) averaged along 120°E–180° against the northwestern NPSSTI. Regression of the winter NOAA-20CRv3 (b) 300-hPa zonal wind (m s−1), (c) 500-hPa geopotential height (m) against the northwestern NPSSTI. (d) As in (a), but for the regression of geopotential height (m). (e),(f) As in (b), but for the regression of (e) 300-hPa transient vorticity forcings (color; 10−11·s−2) and 300-hPa (red contours) and 850-hPa (black contours) geopotential height (m) and (f) 850-hPa wind (vector; m s−1), and surface air temperature anomalies (color; °C). The dotted areas and green vectors are significant at the 95% confidence level. Only the regressed TE vorticity forcings significant at the 90% confidence level are drawn in (e). All data are 11-yr low-pass filtered before analysis.
Citation: Journal of Climate 38, 3; 10.1175/JCLI-D-23-0704.1
The northwestern North Pacific is the oceanic frontal zone, and the regional SST anomaly could impact the EAWM through regulating the TE activity. Next, we diagnose the TE vorticity forcing to investigate the atmospheric dynamical processes related to the northwestern NPSST anomaly, which are not considered by Sun et al. (2016). The regressions of the winter 300-hPa TE forcing and 300/850-hPa geopotential height anomalies against the decadal northwestern NPSSTI are shown in Fig. 7e. We can see the consistency of the centers of anomalous anticyclone and those of the large negative TE vorticity forcing over the northwestern North Pacific. Because compared to the initial baroclinic responses, the subsequent barotropic adjustments in the atmosphere are stronger and last longer (Lau and Holopainen 1984; Deser et al. 2007; Fang and Yang 2016), the TE vorticity forcing is critical in forming an atmospheric equivalent barotropic structure (Kushnir et al. 2002; Zhou et al. 2017; Xue et al. 2018). On the one hand, the induced southerly wind anomalies over the west flank of the low-level anomalous anticyclone could cause positive air temperature anomalies over East Asia (Fig. 7f). On the other hand, the anomalous wind could strengthen the warm current from the tropical areas to the northwestern North Pacific and favor maintaining the regional warm SST anomalies. By this air–sea interaction, the EAWM and the northwestern NPSST anomaly are physically linked.
To further investigate the regional air–sea interaction, we examine the results in Figs. 6 and 7 using the unfiltered seasonal anomalies, as shown in Figs. S4 and S5. We can see that the regressed SST, precipitation, Q1, and net radiation anomalies (Figs. S4a–c,f,g) by the unfiltered data show similar spatial patterns to the counterparts regressed by the low-pass-filtered data, albeit with relatively larger amplitudes and higher confidence levels over the northwestern North Pacific. Concerning the latent and sensible heat fluxes, we can see the positive latent and sensible heat flux anomalies over the northern part of the northwestern North Pacific (Figs. S4d,e), consistent with those by the low-pass-filtered data (Figs. 6d,e). The positive latent and sensible heat flux anomalies concurrent with the warm SST anomalies over the northern part of the northwestern North Pacific indicate the SST forcing on the overlying atmosphere, consistent with previous studies (Qiu et al. 2014; Joh et al. 2023). The consistency of the results between the unfiltered and low-pass-filtered data further confirms the important effect of SST’s diabatic forcing on the overlying atmosphere. On the other hand, we can still note the visible differences over the southern part of the northwestern North Pacific. For instance, the negative latent heat flux anomalies regressed on unfiltered data cover the southern part of the northwestern North Pacific (Fig. S4d), while the anomalies regressed on low-pass-filtered data shrink southward with a relatively smaller domain (Fig. 6d). In general, the latent and sensible heat flux anomalies regressed on unfiltered and low-pass-filtered data are consistent.
Besides the diabatic forcing, the regressed unfiltered atmospheric anomalies are shown in Fig. S5. The regression spatial patterns of the unfiltered meridional temperature gradient, atmospheric circulation anomalies, TE vorticity forcings, and surface air temperature anomalies are generally consistent with the counterparts by the low-pass-filtered data (Fig. 7). According to the above analysis, the consistency between the results derived from the unfiltered and low-pass-filtered data further confirms the influence of the winter northwestern NPSST anomaly on the overlying atmosphere on the decadal time scale.
In addition to the tropospheric processes, we have also analyzed the stratospheric circulation anomalies. With respect to the positive decadal northwestern NPSSTI, there are no significant geopotential height anomalies at 50 hPa over the EAWM key region (figure not shown), indicating the weak contribution of the stratosphere–troposphere interaction to the linkage between the EAWM and the northwestern NPSST anomaly.
Winter is the season when a large amount of turbulent heat flux transfers from the ocean to atmosphere over the northwestern North Pacific (e.g., Joh et al. 2023), but Tao et al. (2022) have demonstrated the different status of the local air–sea interaction in different winter months on the interannual time scale. Diagnosing the changes of the midlatitude air–sea coupled processes in winter is valuable for deeply comprehending the linkage between the northwestern NPSST anomaly and the EAWM. The anomalous precipitation, diabatic heat flux, and TE vorticity forcing related to the decadal northwestern NPSST anomaly in the three winter months are shown in Figs. S6–S8. We can note very interesting phenomena when comparing the results in different months. In December, the regional precipitation anomalies related to the northwestern NPSST anomaly are weak, as well as the Q1 (Figs. S6a,b). In contrast, there are significantly positive latent and sensible heat flux anomalies at the air–sea interface (Figs. S6c,d), indicating that the above air gains energy from the warm SST, which is consistent with previous studies (Czaja et al. 2019; Xu and Guan 2019; Seo et al. 2023). In addition, the ocean absorbs more shortwave radiation and warms the above air column through emitting longwave radiation over the western Pacific from the equator to around 30°N (Figs. S6e,f). However, at the same time, the anomalous TE vorticity forcing is weak over the northwestern North Pacific, indicating weak atmospheric feedback (Fig. S6g). The consistently positive geopotential height anomalies at 850 and 300 hPa are possibly due to the heating over the northwestern North Pacific (Fig. S6g). Therefore, regarding the linkage between the northwestern NPSST anomaly and the EAWM in December, the SST’s forcing could play an initiative role in the regional air–sea interaction through the surface diabatic forcing.
Situations change in January. Associated with the warm northwestern NPSST anomaly is the regional more-than-normal precipitation over the northwestern and central North Pacific (Fig. S7a), which can significantly warm the middle troposphere (Fig. S7b) and cause the positive geopotential height anomalies at the mid-to-upper troposphere (Fig. S7g). At the air–sea interface, there are positive latent heat flux anomalies over the more precipitation region and negative anomalies over the southwestern North Pacific (Fig. S7c). For the sensible heat flux, there are negative anomalies over the western North Pacific and positive over middle to eastern North Pacific (Fig. S7d). For the radiation flux, the ocean absorbs more shortwave radiation over the southwestern North Pacific and emits longwave radiation to heat the overlying atmosphere (Figs. S7e,f). Therefore, the influence of the SST on the atmosphere in January shows a strong regional feature. Meanwhile, we can find that the significantly positive TE vorticity forcing is overlapped with an anomalous anticyclone at 300 hPa over the northwestern North Pacific (Fig. S7g). At the low-level troposphere, an anomalous anticyclone extends from the North Pacific to East Asia, and atmospheric circulation anomalies with an equivalent barotropic structure are formed. The resultant anomalous southerly wind over the west flank of the abovementioned anticyclone favors strengthening the warm current and maintaining the warm SST anomaly. Therefore, the linkage between the northwestern NPSST anomaly and the EAWM in January is associated with the active air–sea coupled processes. The processes related to the warm northwestern North Pacific SST anomaly in February are similar to those in January (Fig. S8).
Overall, the analyses above for each winter month further highlight the essential regional air–sea coupled processes contributing to the physical linkage between the EAWM and the northwestern NPSST anomaly, confirming the results derived from the winter mean.
b. Simulated relationship and its stability in the CMIP6 models
The above section has elucidated the close relationship between the EAWM and the northwestern NPSST anomaly and the possible mechanism. However, it still remains uncertain whether the relationship between the EAWM and SST factor is significant and stable over a longer period. To further verify the role of the northwestern NPSST anomaly in the decadal EAWM variation, we use the long-term piControl simulations to explore their relationship and mechanism.
The correlation results derived from the CMIP6 simulations are shown in Fig. 8. The common significant area in all 30 selected models is the northwestern North Pacific, consistent with the observation (Fig. 5). All of the simulated correlation coefficients between the decadal EAWMI and the northwestern NPSSTI are significant at the 99% confidence level. Therefore, the decadal variation of the northwestern NPSST significantly correlates with the EAWM not only in short-term observation but also in long-term piControl simulations. In addition, we analyze the power spectrum of the simulated SST indices (Fig. S9). All the simulated northwestern NPSST indices have a significant 10–30-yr period, highlighting the models’ skill in reproducing the northwestern NPSST decadal variation and consolidating the linkage between the EAWM and regional SST anomaly.
The correlation distributions between the EAWMI and SST anomalies in the CMIP6 models. The dotted areas are significant at 95% confidence level. The correlation coefficients between the EAWMI and northwestern NPSSTI are given over the Eurasian land. All data are 11-yr low-pass filtered before analysis.
Citation: Journal of Climate 38, 3; 10.1175/JCLI-D-23-0704.1
To check the stability of the decadal relationship between the EAWM and the northwestern NPSST anomaly in long-term piControl simulations, the running correlations with a 201-yr running window between the decadal EAWMI and northwestern NPSSTI are calculated. As shown in Fig. 9, except for five models (CESM2, E3SM-1-0, CESM2-WACCM-FV2, CESM2-WACCM-FV2, and CMCC-CM2-SR5), the remaining 25 models show significantly negative correlations in all the running windows. On the other hand, for the abovementioned five models, the simulated insignificant correlations only occupy less than 20% of the running windows. Therefore, we further verify that the decadal relationship between the EAWM and the northwestern NPSST anomaly is significant and stable in long-term simulations. This result is an extension and supplement of the conclusion surmised from limited observed data, revealing the stability of the abovementioned decadal relationship which is not analyzed by Sun et al. (2016) and then providing valuable insights for the EAWM future projection.
Running correlation coefficients between 11-yr low-pass-filtered EAWMI and northwestern NPSSTI with a 201-yr running window in each CMIP6 model. The blue lines indicate the correlation coefficients significant at the 95% confidence level. The value on right-top corner is the percentage of the running windows with a significant negative correlation.
Citation: Journal of Climate 38, 3; 10.1175/JCLI-D-23-0704.1
The regression patterns of the meridional temperature gradient against the northwestern NPSSTI in the CMIP6 models (Fig. 10) agree with observational analysis, although there are some visible differences in the amplitudes among the models. Therefore, the long-term piControl simulations can effectively capture the influence of the northwestern NPSST anomaly on its overlying atmosphere meridional temperature gradient on the decadal time scale.
The latitude–pressure cross section of the regression of meridional temperature gradient (10−2 °C/lat-degree) averaged along 120°E–180° against the northwestern NPSSTI in the CMIP6 model. The dotted areas are significant at 95% confidence level. All data are 11-yr low-pass filtered before analysis, and the negative value indicates the meridional temperature gradient is enhanced.
Citation: Journal of Climate 38, 3; 10.1175/JCLI-D-23-0704.1
The CMIP6 models can well reproduce the zonal wind response to the warm SST anomaly over the northwestern North Pacific, as shown in Fig. 11, with significant positive (negative) values to the south (north) of 40°N. Compared with the reanalysis results, the significant zonal wind anomalies are extended more eastward, which could be due to the difference in the SST patterns between the observation and models. In the observation, the EAWM-related significant SST anomalies are mainly located over the northwestern North Pacific (Fig. 5), while in the simulations, the SST anomalies can be found over the whole North Pacific (Fig. 8). Such a result indicates that the model still has some simulation biases in the NPSST anomaly, which should be improved in the future.
Regression of 300-hPa zonal wind anomalies (m s−1) against the northwestern NPSSTI in the CMIP6 models. The dotted areas are significant at 95% confidence level. All data are 11-yr low-pass filtered before analysis.
Citation: Journal of Climate 38, 3; 10.1175/JCLI-D-23-0704.1
Therefore, consistent with the observations, the East Asian trough in the long-term simulations is also significantly weakened, with anomalous positive geopotential heights over the North Pacific covering the East Asian trough region (Fig. 12). Similar to the situation of the zonal wind anomalies, the significant geopotential height anomalies are also extended more eastward. Therefore, CMIP6 can generally well simulate the East Asian trough’s response to the SST anomalies over the northwestern North Pacific on the decadal time scale.
Regression of 500-hPa geopotential height anomalies (m) against the northwestern NPSSTI in the CMIP6 models. The dotted areas are significant at the 95% confidence level. All data are 11-yr low-pass filtered before analysis.
Citation: Journal of Climate 38, 3; 10.1175/JCLI-D-23-0704.1
The wind anomalies at 850 hPa associated with decadal variation of northwestern NPSST in the CMIP6 models are shown in Fig. 13. Although there are some visible differences in the intensity and location of the wind anomalies between the reanalysis results and simulations, the spatial patterns of winds in the CMIP6 models are generally consistent with that in NOAA-20CRv3, with anomalous southeasterly winds over East China and southwesterly winds over northern of East Asia. Therefore, in 29 of 30 models (except for INM-CM4-8), the surface air temperatures exhibit significant warm anomalies over East Asia (especially over North China) (Fig. 14). The simulated air temperature responses are consistent with the reanalysis results (Fig. 7f). Nevertheless, the amplitude of response over East Asia in some models (e.g., CESM-WACCM-FV2, FGOALS-f3-l, MCM-UA-1-0) is relatively weak. Such results confirm that the northwestern NPSST anomaly is also a stable indicator for the East Asian winter climate anomaly on the decadal time scale.
Regression of 850-hPa wind anomalies (m s−1) against the northwestern NPSSTI in the CMIP6 models. The gray shading areas are significant at the 95% confidence level. All data are 11-yr low-pass filtered before analysis.
Citation: Journal of Climate 38, 3; 10.1175/JCLI-D-23-0704.1
Regression of surface air temperature anomalies (°C) against the northwestern NPSSTI in the CMIP6 models. The gray shading areas are significant at the 95% confidence level. All data are 11-yr low-pass filtered before analysis.
Citation: Journal of Climate 38, 3; 10.1175/JCLI-D-23-0704.1
The CMIP6’s piControl simulations spanning over 450 years can capture the response of meridional temperature gradients and East Asian atmosphere circulation and temperature to the northwestern NPSST anomaly. We can further verify that the decadal variation of the northwestern NPSST anomaly can influence the East Asian atmosphere circulation, weakening the EAWM and warming East Asia. Therefore, northwestern NPSST anomaly plays an important role in the decadal EAWM variation.
5. Discussion and conclusions
a. Discussion
Previous studies have pointed out that the EAWM is significantly influenced by ENSO, and El Niño (La Niña) is generally associated with a weakened (strengthened) EAWM (e.g., Zhang et al. 1996; Wang et al. 2000). Therefore, ENSO’s impact on the EAWM on the decadal time scale is investigated. In this study, ENSO is measured by the Niño-3.4 index, which is defined as the area-averaged SST anomalies over 5°S–5°N, 120°–170°W. Considering the dominance of interannual variability of ENSO, we first calculate the correlation coefficient between the unfiltered Niño-3.4 index and the EAWMI. The correlation coefficient between the unfiltered HadISST Niño-3.4 index and the NOAA-20CRv3 (ERA-20C) EAWMI is −0.26 (−0.18); the correlation coefficient between the ERASSTv5 Niño-3.4 index and the NOAA-20CRv3 (ERA-20C) EAWMI is −0.30 (−0.21). All of the correlation coefficients are significant at the 90% confidence level. We further examine the relationship between ENSO and the EAWM and its stability in the piControl simulations. As shown in Fig. S10a, 23 of 30 (76.7%) models simulate significant negative correlations between the unfiltered Niño-3.4 index and the EAWMI during the whole period. The significance is estimated according to the effective degree of freedom. Furthermore, we calculate the percentage of the running windows with significant negative correlations between the unfiltered Niño-3.4 index and the EAWMI relative to all running windows in each model. The result shows that there are 18 of 30 (60%) models with a percentage equal to or greater than 75%. Such a result suggests a significant and relatively stable negative relationship between the unfiltered Niño-3.4 index and the EAWMI.
After 11-yr low-pass filtering, the correlation coefficient between the HadISST Niño-3.4 index and the NOAA-20CRv3 (ERA-20C) EAWMI is 0.17 (0.30); the correlation coefficient between the ERASSTv5 Niño-3.4 index and the NOAA-20CRv3 (ERA-20C) EAWMI is −0.11 (0.05). All of these correlation coefficients are weak and insignificant. For the piControl simulations, we find that 18 of 30 (60%) models simulate insignificant correlation coefficients between the Niño-3.4 index and the EAWMI during the whole period (Fig. S10b). We also calculate the percentage of the running windows with a significant negative correlation coefficient between the low-pass-filtered Niño-3.4 index and the EAWMI in each model. The percentages in 19 of 30 (63.3%) models are less than 25%, especially in AWI-CM-1-MR, BCC-CSM2-MR, BCC-ESM1, CESM2-FV2, CESM2-WACCM-FV2, EC-Earth3-AerChem, EC-Earth-CC, FGOALS-f3-L, FGOALS-g3, INM-CM4-8, and INM-CM5-0 with the values of 0. We can still note that the percentages are more than 50% in the seven models: NorESM2-LM, NorESM2-MM, CESM2-WACCM, EC-Earth3-Veg-LR, CAS-ESM2-0, CMCC-CM2-SR5, and MCM-UA-1-0. Such results hint at the models’ different performances in climate simulations, but the majority of the simulation suggests the weak correlation between the low-pass-filtered Niño-3.4 index and the EAWMI.
The above analyses reveal the significant (weak) correlation between the unfiltered (low-pass-filtered) Niño-3.4 index and the EAWMI. Next, after removing the ENSO signal, we investigate the relationship between the EAWM and the northwestern NPSST anomaly. According to previous studies (e.g., Frankignoul et al. 2011; Chen et al. 2013), to objectively remove the ENSO’s impact, we subtract the signals associated with the winter Niño-3.4 index from the unfiltered EAWMI (marked as EAWMI_res) by the linear regression method and repeat the analyses in Figs. 5, 8, and 9. The results are consistent (Figs. S11–S13) compared to Figs. 5, 8, and 9. Therefore, the reanalysis results and numerical simulations confirm that the northwestern NPSST anomaly exhibits a stably significant relationship with the EAWM independent of ENSO.
Since previous research suggested that PDO contributes to the decadal variation of the EAWM (Ding et al. 2014), we analyze the PDO’s role in the decadal EAWM variation. According to Mantua et al. (1997), the PDO index is defined as the principal component time series of the first empirical orthogonal function of the North Pacific (20°–70°N) SST anomalies. After 11-yr low-pass filtering, the correlation coefficients of the NOAA-20CRv3 EAWMI with the HadISST and ERSSTv5 PDO index are 0.37 and 0.36, respectively, which are insignificant. The correlation coefficients of the ERA-20C EAWMI with the HadISST and ERSSTv5 PDO index are 0.56 and 0.56, respectively, which are significant at the 95% confidence level. The differences in the correlation analysis imply the uncertainty of the EAWM–PDO connection. Therefore, we explore the relationship between the EAWM and PDO in the CMIP6 piControl simulations (Fig. S14). Although 22 of 30 models could simulate a significant relationship between the EAWM and PDO, there still remain eight models that cannot. When investigating the stability of the simulated relationship between the EAWM and PDO, we can see that the percentages of the running windows with significant negative correlation coefficients between the EAWMI and PDO index in 8 of 30 (26.7%) models are less than 25%, especially in CESM2-WACCM and MPI-ESM1-2-HR with the values of 0. Referring to the results in Figs. 8 and 9, we can conclude that compared to the PDO, the connection of the EAWM with the northwestern NPSST anomaly is more significant and robust.
When revisiting the results in Fig. 5, we can see that the northwestern North Pacific is also the oceanic frontal zone with energetic eddy activity. The oceanic eddy needs to be resolved from a fine-resolution (at least 0.25°) field (Sugimoto and Hanawa 2011; Siqueira et al. 2021). Previous numerical simulations demonstrate that, compared to the low-resolution model, the eddy-resolving climate model can reproduce the mesoscale SST anomalies over the western boundary region, simulating more realistic regional convection and precipitation (Smirnov et al. 2015; Ma et al. 2017; Siqueira et al. 2021). Consequently, the mesoscale SST anomalies associated with the oceanic eddy could change the storm track and cause basin-scale atmospheric circulation anomalies (Smirnov et al. 2015; Ma et al. 2017). Besides the contribution of the small-scale oceanic processes, the northwestern NPSST anomaly could influence the overlying EAWM via other pathways. In our study, the large-scale SST-related diabatic forcing, temperature gradient, and atmospheric feedback could also significantly impact the EAWM. On this basis, despite the relatively coarse resolution (compared to the eddy-resolving resolution) of the current CMIP6 models, the relationship between the winter northwestern NPSST anomaly and the EAWM can still be well simulated. Also, it is necessary to explore the role of oceanic eddy using high-resolution simulation in future studies.
b. Conclusions
In this study, we use the CMIP6 piControl simulations spanning over 450 years to examine the climate internal variability of EAWM and its possible mechanism on the decadal time scale. The result indicates that the EAWM has a significant 10–30-yr period in observation, and CMIP6 models can well capture the time period and spatial pattern of EAWM. Provided that the piControl simulations only have internal climate variability, we deduce that the decadal EAWM variation could originate from the decadal SST variation.
According to Sun et al. (2016), northwestern NPSST anomaly plays an important role in the decadal EAWM variation. However, such a conclusion is obtained from the limited observed and reanalysis data. In this study, we further investigate the relationship between the two variables using long-term simulations from CMIP6 models. The results show that the EAWM and the northwestern NPSST anomaly are significantly and stably correlated with each other in all CMIP6 models. The warm northwestern NPSST can release energy from the ocean and heat the overlying atmosphere. The meridional temperature gradients are consequently weakened (enhanced) to the south (north) of 40°N, leading to the suppressed (strengthened) westerlies to the south (north) of 40°N according to the thermal wind theory. Thus, the anomalous anticyclone occurs over East Asia, weakening the East Asian trough. Therefore, the EAWM is weakened and warmer-than-normal winter occurs over East Asia. In turn, the atmospheric feedback in the local air–sea interaction could favor maintaining the SST anomaly. Such revealed relationship and physical processes are generally captured by the CMIP6 piControl simulations, which further verify the influence of the northwestern NPSST anomaly on the EAWM revealed by Sun et al. (2016) using short-term observations. Therefore, the decadal EAWM variation could originate from climate internal variability and the northwestern NPSST anomaly could play an important role in the decadal EAWM variation.
Our analysis implies the potential contribution of the northwestern NPSST anomaly to the decadal EAWM variation. Therefore, comprehending the decadal SST variability in the northwestern North Pacific is meaningful. Currently, there are two main viewpoints regarding the source of decadal variability. One emphasizes the significant role of air–sea interactions or oceanic forcing in decadal variability. Upon the observational and coupled model results, some studies have revealed the roles of the regional air–sea interactions and oceanic dynamical adjustments to the decadal SST variability (e.g., Latif and Barnett 1994; An 2008; Gu et al. 2024). On the other hand, alternative viewpoints suggest that atmospheric stochastic forcing may be crucial for climate decadal variation. For example, Frankignoul et al. (1997) proposed that stochastic wind stress forcing could explain a significant portion of the decadal variation in oceanic circulation. A review by Liu (2012) contended that stochastic forcing plays a vital role in the decadal SST variability. In detail, the North Pacific Oscillation (NPO) could modulate decadal SST variability in the northwestern North Pacific through impacting the thermocline depth and ocean heat content (Newman et al. 2016; Di Lorenzo et al. 2023). Collectively, when exploring the source of the decadal SST variability in the northwestern Pacific, we shall not only focus on the ocean itself but also pay attention to the atmospheric stochastic forcing, which merits further careful studies.
Acknowledgments.
This study was supported by the National Natural Science Foundation of China (Grant 42221004). We thank the World Climate Research Programme for providing the CMIP6 datasets.
Data availability statement.
The information of the datasets used in this study are present here. The piControl simulations from CMIP6 listed in Table 1 are available at https://esgf-node.llnl.gov/search/cmip6/. The NOAA-20CRv3 is available at https://psl.noaa.gov/data/gridded/data.20thC_ReanV3.html. The HadISST is available at https://www.metoffice.gov.uk/hadobs/hadisst/data/download.html. The Public Datasets Service for ERA-20C closed on 1 June 2023, and we accessed it in October 2022. For more information and possible alternatives, we can consult the dedicated page at https://confluence.ecmwf.int/display/DAC/Decommissioning+of+ECMWF+Public+Datasets+Service. The ERSSTv5 is available at https://climatedataguide.ucar.edu/climate-data/sst-data-noaa-extended-reconstruction-ssts-version-5-ersstv5.
REFERENCES
An, S.-I., 2008: A mechanism for the multi-decadal climate oscillation in the North Pacific. Theor. Appl. Climatol., 91, 77–84, https://doi.org/10.1007/s00704-006-0288-7.
Befort, D. J., S. Wild, T. Kruschke, U. Ulbrich, and G. C. Leckebusch, 2016: Different long-term trends of extra-tropical cyclones and windstorms in ERA-20C and NOAA-20CR reanalyses. Atmos. Sci. Lett., 17, 586–595, https://doi.org/10.1002/asl.694.
Chang, C.-P., Z. Wang, and H. Hendon, 2006: The Asian winter monsoon. The Asian Monsoon, B. Wang, Ed., Springer, 89–127, https://doi.org/10.1007/3-540-37722-0_3.
Chen, D., Y. Gao, J. Sun, H. Wang, and J. Ma, 2020: Interdecadal variation and causes of drought in Northeast China in recent decades. J. Geophys. Res. Atmos., 125, e2019JD032069, https://doi.org/10.1029/2019JD032069.
Chen, L., R. Zhang, S. C. Pryor, X. Li, and H. Wang, 2020: Influence of wintertime surface sensible heat flux variability over the central and eastern Tibetan Plateau on the East Asian winter monsoon. Climate Dyn., 54, 4589–4603, https://doi.org/10.1007/s00382-020-05246-x.
Chen, W., and B. Dong, 2019: Anthropogenic impacts on recent decadal change in temperature extremes over China: Relative roles of greenhouse gases and anthropogenic aerosols. Climate Dyn., 52, 3643–3660, https://doi.org/10.1007/s00382-018-4342-9.
Chen, W., J. Feng, and R. Wu, 2013: Roles of ENSO and PDO in the link of the East Asian winter monsoon to the following summer monsoon. J. Climate, 26, 622–635, https://doi.org/10.1175/JCLI-D-12-00021.1.
Chen, W., L. Wang, J. Feng, Z. Wen, T. Ma, X. Yang, and C. Wang, 2019: Recent progress in studies of the variabilities and mechanisms of the East Asian monsoon in a changing climate. Adv. Atmos. Sci., 36, 887–901, https://doi.org/10.1007/s00376-019-8230-y.
Czaja, A., C. Frankignoul, S. Minobe, and B. Vannière, 2019: Simulating the midlatitude atmospheric circulation: What might we gain from high-resolution modeling of air-sea interactions? Curr. Climate Change Rep., 5, 390–406, https://doi.org/10.1007/s40641-019-00148-5.
Deser, C., R. A. Tomas, and S. Peng, 2007: The transient atmospheric circulation response to North Atlantic SST and sea ice anomalies. J. Climate, 20, 4751–4767, https://doi.org/10.1175/JCLI4278.1.
Di Lorenzo, E., and Coauthors, 2023: Modes and mechanisms of Pacific decadal-scale variability. Annu. Rev. Mar. Sci., 15, 249–275, https://doi.org/10.1146/annurev-marine-040422-084555.
Ding, Y., 1994: Monsoon over China. Kluwer Academic Publishers, 420 pp.
Ding, Y., and Coauthors, 2014: Interdecadal variability of the East Asian winter monsoon and its possible links to global climate change. J. Meteor. Res., 28, 693–713, https://doi.org/10.1007/s13351-014-4046-y.
Eyring, V., S. Bony, G. A. Meehl, C. A. Senior, B. Stevens, R. J. Stouffer, and K. E. Taylor, 2016: Overview of the Coupled Model Intercomparison Project Phase 6 (CMIP6) experimental design and organization. Geosci. Model Dev., 9, 1937–1958, https://doi.org/10.5194/gmd-9-1937-2016.
Fang, J., and X.-Q. Yang, 2016: Structure and dynamics of decadal anomalies in the wintertime midlatitude North Pacific ocean–atmosphere system. Climate Dyn., 47, 1989–2007, https://doi.org/10.1007/s00382-015-2946-x.
Frankignoul, C., P. Müller, and E. Zorita, 1997: A simple model of the decadal response of the ocean to stochastic wind forcing. J. Phys. Oceanogr., 27, 1533–1546, https://doi.org/10.1175/1520-0485(1997)027<1533:ASMOTD>2.0.CO;2.
Frankignoul, C., N. Sennéchael, Y.-O. Kwon, and M. A. Alexander, 2011: Influence of the meridional shifts of the Kuroshio and the Oyashio Extensions on the atmospheric circulation. J. Climate, 24, 762–777, https://doi.org/10.1175/2010JCLI3731.1.
Gu, L., K. Wei, and R. H. Huang, 2008: Severe disaster of blizzard, freezing rain and low temperature in January 2008 in China and its association with the anomalies of East Asian monsoon system (in Chinese). Climatic Environ. Res., 13, 405–418.
Gu, P., Z. Liu, and T. L. Delworth, 2024: Strong oceanic forcing on decadal surface temperature variability over global ocean. Geophys. Res. Lett., 51, e2023GL107401, https://doi.org/10.1029/2023GL107401.
Gulev, S. K., M. Latif, N. Keenlyside, W. Park, and K. P. Koltermann, 2013: North Atlantic Ocean control on surface heat flux on multidecadal timescales. Nature, 499, 464–467, https://doi.org/10.1038/nature12268.
He, S., and H. Wang, 2012: Analysis of the decadal and interdecadal variations of the East Asian winter monsoon as simulated by 20 coupled models in IPCC AR4. Acta Meteor. Sin., 26, 476–488, https://doi.org/10.1007/s13351-012-0407-6.
Huang, B., and Coauthors, 2017: Extended Reconstructed Sea Surface Temperature, version 5 (ERSSTv5): Upgrades, validations, and intercomparisons. J. Climate, 30, 8179–8205, https://doi.org/10.1175/JCLI-D-16-0836.1.
Huang, J, J. He, H. Xu, and Q. Jin, 2016: Relationship between the seasonal transition of East Asian monsoon circulation and Asian-Pacific thermal field and possible mechanisms. J. Trop. Meteor., 22, 466–478, https://doi.org/10.16555/j.1006-8775.2016.04.003.
Jhun, J.-G., and E.-J. Lee, 2004: A new East Asian winter monsoon index and associated characteristics of the winter monsoon. J. Climate, 17, 711–726, https://doi.org/10.1175/1520-0442(2004)017<0711:ANEAWM>2.0.CO;2.
Jiang, D., D. Hu, Z. Tian, and X. Lang, 2020: Differences between CMIP6 and CMIP5 models in simulating climate over China and the East Asian monsoon. Adv. Atmos. Sci., 37, 1102–1118, https://doi.org/10.1007/s00376-020-2034-y.
Joh, Y., T. L. Delworth, A. T. Wittenberg, X. Yang, A. Rosati, N. C. Johnson, and L. Jia, 2023: The role of upper-ocean variations of the Kuroshio-Oyashio Extension in seasonal-to-decadal air-sea heat flux variability. npj Climate Atmos. Sci., 6, 123, https://doi.org/10.1038/s41612-023-00453-9.
Kushnir, Y., W. A. Robinson, I. Bladé, N. M. J. Hall, S. Peng, and R. Sutton, 2002: Atmospheric GCM response to extratropical SST anomalies: Synthesis and evaluation. J. Climate, 15, 2233–2256, https://doi.org/10.1175/1520-0442(2002)015<2233:AGRTES>2.0.CO;2.
Latif, M., and T. P. Barnett, 1994: Causes of decadal climate variability over the North Pacific and North America. Science, 266, 634–637, https://doi.org/10.1126/science.266.5185.634.
Lau, N.-C., and E. O. Holopainen, 1984: Transient eddy forcing of the time-mean flow as identified by geopotential tendencies. J. Atmos. Sci., 41, 313–328, https://doi.org/10.1175/1520-0469(1984)041<0313:TEFOTT>2.0.CO;2.
Li, C., and P. Xian, 2003: Atmospheric anomalies related to interdecadal variability of SST in the North Pacific. Adv. Atmos. Sci., 20, 859–874, https://doi.org/10.1007/BF02915510.
Li, H., A. Dai, T. Zhou, and J. Lu, 2010: Responses of East Asian summer monsoon to historical SST and atmospheric forcing during 1950–2000. Climate Dyn., 34, 501–514, https://doi.org/10.1007/s00382-008-0482-7.
Li, S., J. Lu, G. Huang, and K. Hu, 2008: Tropical Indian Ocean basin warming and East Asian summer monsoon: A multiple AGCM study. J. Climate, 21, 6080–6088, https://doi.org/10.1175/2008JCLI2433.1.
Li, Y., and S. Yang, 2010: A dynamical index for the East Asian winter monsoon. J. Climate, 23, 4255–4262, https://doi.org/10.1175/2010JCLI3375.1.
Liu, Z., 2012: Dynamics of interdecadal climate variability: A historical perspective. J. Climate, 25, 1963–1995, https://doi.org/10.1175/2011JCLI3980.1.
Lu, R., B. Dong, and H. Ding, 2006: Impact of the Atlantic Multidecadal Oscillation on the Asian summer monsoon. Geophys. Res. Lett., 33, L24701, https://doi.org/10.1029/2006GL027655.
Ma, T., and W. Chen, 2021: Climate variability of the East Asian winter monsoon and associated extratropical–tropical interaction: A review. Ann. N. Y. Acad. Sci., 1504, 44–62, https://doi.org/10.1111/nyas.14620.
Ma, X., P. Chang, R. Saravanan, R. Montuoro, H. Nakamura, D. Wu, X. Lin, and L. Wu, 2017: Importance of resolving Kuroshio Front and eddy influence in simulating the North Pacific storm track. J. Climate, 30, 1861–1880, https://doi.org/10.1175/JCLI-D-16-0154.1.
Mantua, N. J., S. R. Hare, Y. Zhang, J. M. Wallace, and R. C. Francis, 1997: A Pacific interdecadal climate oscillation with impacts on salmon production. Bull. Amer. Meteor. Soc., 78, 1069–1080, https://doi.org/10.1175/1520-0477(1997)078<1069:APICOW>2.0.CO;2.
Miao, J., and D. Jiang, 2022: Multidecadal variations in East Asian winter temperature since 1880: Internal variability versus external forcing. Geophys. Res. Lett., 49, e2022GL099597, https://doi.org/10.1029/2022GL099597.
Miao, J., T. Wang, H. Wang, Y. Zhu, and J. Sun, 2018: Interdecadal weakening of the East Asian winter monsoon in the mid-1980s: The roles of external forcings. J. Climate, 31, 8985–9000, https://doi.org/10.1175/JCLI-D-17-0868.1.
Miao, J., T. Wang, and H. Wang, 2020: Interdecadal variations of the East Asian winter monsoon in CMIP5 preindustrial simulations. J. Climate, 33, 559–575, https://doi.org/10.1175/JCLI-D-19-0148.1.
Newman, M., and Coauthors, 2016: The Pacific decadal oscillation, revisited. J. Climate, 29, 4399–4427, https://doi.org/10.1175/JCLI-D-15-0508.1.
Poli, P., H. Hersbach, P. Berrisford, D. Dee, A. Simmons, and P. Laloyaux, 2015: ERA-20C deterministic. ERA Rep. Series 20, 53 pp., https://www.ecmwf.int/sites/default/files/elibrary/2015/11700-era-20c-deterministic.pdf.
Poli, P., and Coauthors, 2016: ERA-20C: An atmospheric reanalysis of the twentieth century. J. Climate, 29, 4083–4097, https://doi.org/10.1175/JCLI-D-15-0556.1.
Qiu, B., S. Chen, N. Schneider, and B. Taguchi, 2014: A coupled decadal prediction of the dynamic state of the Kuroshio Extension System. J. Climate, 27, 1751–1764, https://doi.org/10.1175/JCLI-D-13-00318.1.
Quenouille, M. H., 1952: Associated Measurements. Academic Press, 242 pp.
Rayner, N. A., D. E. Parker, E. B. Horton, C. K. Folland, L. V. Alexander, D. P. Rowell, E. C. Kent, and A. Kaplan, 2003: Global analyses of sea surface temperature, sea ice, and night marine air temperature since the late nineteenth century. J. Geophys. Res., 108, 4407, https://doi.org/10.1029/2002JD002670.
Sardeshmukh, P. D., and B. I. Hoskins, 1984: Spatial smoothing on the sphere. Mon. Wea. Rev., 112, 2524–2529, https://doi.org/10.1175/1520-0493(1984)112<2524:SSOTS>2.0.CO;2.
Seo, H., and Coauthors, 2023: Ocean mesoscale and Frontal-Scale Ocean–Atmosphere interactions and influence on large-scale climate: A review. J. Climate, 36, 1981–2013, https://doi.org/10.1175/JCLI-D-21-0982.1.
Siqueira, L., B. P. Kirtman, and L. C. Laurindo, 2021: Forecasting remote atmospheric responses to decadal Kuroshio stability transitions. J. Climate, 34, 379–395, https://doi.org/10.1175/JCLI-D-20-0139.1.
Slivinski, L. C., and Coauthors, 2019: Towards a more reliable historical reanalysis: Improvements for version 3 of the twentieth century reanalysis system. Quart. J. Roy. Meteor. Soc., 145, 2876–2908, https://doi.org/10.1002/qj.3598.
Slivinski, L. C., and Coauthors, 2021: An evaluation of the performance of the twentieth century reanalysis version 3. J. Climate, 34, 1417–1438, https://doi.org/10.1175/JCLI-D-20-0505.1.
Smirnov, D., M. Newman, M. A. Alexander, Y.-O. Kwon, and C. Frankignoul, 2015: Investigating the local atmospheric response to a realistic shift in the Oyashio sea surface temperature front. J. Climate, 28, 1126–1147, https://doi.org/10.1175/JCLI-D-14-00285.1.
Sugimoto, S., and K. Hanawa, 2011: Roles of SST anomalies on the wintertime turbulent heat fluxes in the Kuroshio–Oyashio Confluence Region: Influences of warm eddies detached from the Kuroshio Extension. J. Climate, 24, 6551–6561, https://doi.org/10.1175/2011JCLI4023.1.
Sun, J., H. Wang, W. Yuan, and H. Chen, 2010: Spatial-temporal features of intense snowfall events in China and their possible change. J. Geophys. Res., 115, D16110, https://doi.org/10.1029/2009JD013541.
Sun, J., S. Wu, and J. Ao, 2016: Role of the North Pacific sea surface temperature in the East Asian winter monsoon decadal variability. Climate Dyn., 46, 3793–3805, https://doi.org/10.1007/s00382-015-2805-9.
Tao, L., J. Fang, X.-Q. Yang, X. Sun, and D. Cai, 2022: Midwinter reversal of the atmospheric anomalies caused by the North Pacific Mode-related air-sea coupling. Geophys. Res. Lett., 49, e2022GL100307, https://doi.org/10.1029/2022GL100307.
Wang, B., R. Wu, and X. Fu, 2000: Pacific–East Asian teleconnection: How does ENSO affect East Asian climate? J. Climate, 13, 1517–1536, https://doi.org/10.1175/1520-0442(2000)013<1517:PEATHD>2.0.CO;2.
Wang, H., and S. He, 2012: Weakening relationship between East Asian winter monsoon and ENSO after mid-1970s. Chin. Sci. Bull., 57, 3535–3540, https://doi.org/10.1007/s11434-012-5285-x.
Wang, L., and W. Chen, 2014: The East Asian winter monsoon: Re-amplification in the mid-2000s. Chin. Sci. Bull., 59, 430–436, https://doi.org/10.1007/s11434-013-0029-0.
Wang, L., R. Huang, L. Gu, W. Chen, and L. Kang, 2009: Interdecadal variations of the East Asian winter monsoon and their association with quasi-stationary planetary wave activity. J. Climate, 22, 4860–4872, https://doi.org/10.1175/2009JCLI2973.1.
Wang, X., S. Qian, and G. Feng, 2018: A study of the winter-winter recurrence mechanism of the temperatures in Northern East Asia. J. Trop. Meteor., 24, 49–59, https://doi.org/10.16555/j.1006-8775.2018.01.005.
Xiao, X., W. Chen, G. Fan, and D. Zhou, 2016: Possible external forcing factors for the interdecadal change in the East Asian winter monsoon around the late 1990s (in Chinese). Climatic Environ. Res., 21, 97–209.
Xu, Q., and Z.-Y. Guan, 2019: Interdecadal change of diabatic forcing over key region of the Maritime Continent and its possible relations with the East Asian summer monsoon anomalies. J. Trop. Meteor., 25, 54–62, https://doi.org/10.16555/j.1006-8775.2019.01.006.
Xue, J., C. Sun, J. Li, J. Mao, H. Nakamura, T. Miyasaka, and Y. Xu, 2018: Divergent responses of extratropical atmospheric circulation to interhemispheric dipolar SST forcing over the two hemispheres in boreal winter. J. Climate, 31, 7599–7619, https://doi.org/10.1175/JCLI-D-17-0817.1.
Yanai, M., S. Esbensen, and J.-H. Chu, 1973: Determination of bulk properties of tropical cloud clusters from large-scale heat and moisture budgets. J. Atmos. Sci., 30, 611–627, https://doi.org/10.1175/1520-0469(1973)030<0611:DOBPOT>2.0.CO;2.
Yang, X.-Q., Y.-M. Zhu, Q. Xie, X.-J. Ren, and G.-Y. Xu, 2004: Advances in studies of Pacific decadal oscillation (in Chinese). Chin. J. Atmos. Sci., 28, 979–992, https://doi.org/10.3878/j.issn.1006-9895.2004.06.15.
Zhang, R., A. Sumi, and M. Kimoto, 1996: Impact of El Nino on the East Asian monsoon: A diagnostic study of the ‘86/87 and ‘91/92 events. J. Meteor. Soc. Japan, 74, 49–62, https://doi.org/10.2151/jmsj1965.74.1_49.
Zhou, G., M. Latif, R. J. Greatbatch, and W. Park, 2017: State dependence of atmospheric response to extratropical North Pacific SST anomalies. J. Climate, 30, 509–525, https://doi.org/10.1175/JCLI-D-15-0672.1.
Zhou, W., C. Li, and X. Wang, 2007: Possible connection between Pacific Oceanic interdecadal pathway and East Asian winter monsoon. Geophys. Res. Lett., 34, L01701, https://doi.org/10.1029/2006GL027809.
