1. Introduction
As Earth’s highest terrestrial elevation area, the Tibetan Plateau (TP) naturally features complex dynamic and thermal climate effects (Valdes and Hoskins 1991; Abe et al. 2003; Shaw and Voigt 2015; Liu et al. 2020). For example, in winter, the blocking effect of the TP leads to a division of the westerly flow to the northern and southern branches (Hoskins and Karoly 1981; Son et al. 2020). In summer, the TP is a large and elevated heat source for the atmosphere and can impact the local climate and the climate in East Asian monsoon regions (e.g., Duan and Wu 2005; Wu et al. 2007, 2012; Chen et al. 2020).
During the cold seasons, large areas of snow over the TP can modulate the local energy–hydrology balance and impact climate variation globally (e.g., Si and Ding 2013; Xiao and Duan 2016; Li et al. 2018; Qian et al. 2019; Zhang et al. 2019; Jia et al. 2021). Recently, some studies revealed that even in summertime, snow can be found in some high-altitude regions of the TP, which impacts the climate of the downstream mei-yu regions and midlatitude Asia (e.g., Xiao and Duan 2016; Wang et al. 2018; Zhang et al. 2021; Zhang and Jia 2022). In general, TP snow accumulates starting in autumn and melts after late spring (Barnett et al. 1988; Wang et al. 2019; Jiang et al. 2019). Spring is a transitional season for TP snow that is accompanied by complex climate effects. Both observed and simulated evidence show that spring snow-cover extent (SCE) (SSCE) over the TP (SSC_TP) can impact the climate over East Asia and downstream regions on multiple time scales (e.g., Zhang et al. 2004; Fan et al. 2014; Liu et al. 2014; Zhang et al. 2021; Zhang and Jia 2022). For instance, Xiao and Duan (2016) advocated that winter snow anomalies over the western TP can persist until the following spring and summer, which further impacts the East Asian summer monsoon (EASM) by eastward-propagating synoptic disturbances and moisture flux originating from the TP. Wang et al. (2020) revealed that when the Pacific North American pattern is weak in winter, the spring snow cover over the eastern TP can simultaneously impact the North American temperature by stimulating a half-hemispheric large-scale wave train spanning from the eastern TP to North America. Zhang et al. (2004) and Fan et al. (2014) argued that on the interdecadal time scale, the increased spring TP snow causes increased summer rainfall over southern China owing to TP snow-induced cooling over the TP and its surrounding area, favoring a northwestward shift in the subtropical high.
The formation of the SSC_TP has also been investigated (e.g., Bamzai 2003; Yuan et al. 2009; Mao 2010; Bao et al. 2018; Wang et al. 2018, 2019; Bao and You 2019; Jiang et al. 2019; Zhang et al. 2019). For example, some work revealed that the positive Arctic Oscillation (AO), together with the negative Western Pacific pattern, can contribute to the SSC_TP variation by triggering an Arctic–TP teleconnection wave train (Bamzai 2003; Bao et al. 2018; Zhang et al. 2019). Some studies have proposed that in association with an enhanced and southwestward-shifted subtropical westerly jet, positive vorticity anomalies and vertical upward motion over the TP favor the formation of the SSC_TP (Mao 2010; Bao and You 2019). North Atlantic sea surface temperature (SST) anomalies (SSTAs) can also modulate the SSC_TP variation by forcing a North Atlantic–TP atmospheric wave train (Wang et al. 2018, 2019). The dipole-like winter SSTAs over the tropical Indian Ocean are associated with anomalous TP snow, which can be prolonged from winter to the subsequent spring and can contribute to the accumulation of spring TP snow (Yuan et al. 2009, 2012; Jiang et al. 2019). Yuan et al. (2009) illustrated that the correlation of winter ENSO and SSC_TP was nonsignificant during the period 1973–99, while Wang and Xu (2018) documented that winter–spring snow over the TP was significantly correlated with ENSO during the period 1987–2005. In addition, the impact of southern hemispheric Antarctic Oscillation (AAO) in May could be prolonged and extended to the TP summer snow via with the help of Indian Ocean SST anomalies through air–sea interactions (Dou and Wu 2018). Recently, Tang et al. (2022) revealed that the AAO can influence the TP through a northeast-propagating wave train and tropical air–sea interaction.
Most of these works examined the impact and formation of spring snow over the entire TP domain, while some studies have found that spring snow cover changes differently over the western and eastern TP (Fan et al. 2014; Wang et al. 2018, 2020). Our recent study shows the interdecadal changes in the impacts of SSC_TP on precipitation over eastern China in the early 1990s (Zhang et al. 2022). However, the physical mechanism of the interdecadal changes in the interannual variability in SSC_TP is not yet clear. This study aims to explore the possible dynamic mechanism for the change in the leading empirical orthogonal function (EOF) pattern of SSC_TP.
This work is arranged as follows. Section 2 describes the data, methodology, and model used in this study. Section 3 examines the interdecadal changes in the leading EOF pattern of the SSC_TP. The possible reasons for the interdecadal changes in the leading SSC_TP EOF patterns are explored in section 4. Finally, the conclusions and discussion are provided in section 5.
2. Data, methodology, and model
a. Datasets
The datasets employed in this work cover the period 1970–2020 and are described as follows:
-
The monthly mean Northern Hemisphere SCE is obtained from the National Oceanic and Atmospheric Administration (NOAA) climate data record (Robinson et al. 2012). The original weekly SCE data products are produced by high-resolution radiometers and multiple satellites, which cover the period of October 1966 to the present. The original data were converted into monthly mean data with a spatial resolution of 1° longitude × 1° latitude (https://www.ncei.noaa.gov/access/metadata/landing-page/bin/iso?id=gov.noaa.ncdc:C00756).
-
Monthly mean variables, for example, wind, vertical velocity, geopotential height, specific humidity, vorticity, and air temperature, are provided by the fifth major global reanalysis produced by the European Centre for Medium-Range Weather Forecasts (ECMWF) (ERA5; Hersbach et al. 2020). These variables have a horizontal resolution of 1° × 1° (https://www.ecmwf.int/en/forecasts/datasets/reanalysis-datasets/era5).
-
A monthly mean global precipitation dataset with a resolution of 2.5° longitude × 2.5° latitude is obtained from the Global Precipitation Climatology Project Version 2.2 (Adler et al. 2003). This dataset is produced by both satellite and observed precipitation (https://www.esrl.noaa.gov/psd/data/gridded/).
-
Monthly mean global SST data with a resolution of 1° × 1° are acquired from the Hadley Centre Sea Ice and Sea Surface Temperature dataset (HadISST) (Rayner et al. 2003) (https://www.metoffice.gov.uk/hadobs/hadisst/data/).
b. Methods
The spring AO index is defined as the time series associated with the first EOF pattern of the spring 1000-hPa height poleward of 20°N (Qian et al. 2020). The spring AAO index is defined as the time series associated with the leading EOF pattern of the spring 700-hPa height poleward of 20°S (Alves et al. 2017). Here, the AO and AAO indices are calculated using data for the period 1970–2020. The indices are then divided into two epochs, that is, P1 (1970–89) and P2 (1991–2020). The conclusions are not sensitive to the indices if they are calculated using the data of the two epochs separately. In the current work, the winter–spring (December–May) Niño-3.4 index is calculated by normalizing the area-averaged SST anomalies over the Niño-3.4 region (5°S–5°N, 170°–120°W; Hafez 2017).
The current work focuses on examining the variations on the interannual time scale, and therefore, all physical variables are linearly detrended and an 8-yr high-pass Gaussian filter are applied to remove the interdecadal variation and long-term trends, respectively. To obtain the main spatial structures of the SSC_TP variation, an EOF analysis is performed. To verify the impact of TP snow independent of ENSO, a partial regression analysis is utilized by removing winter–spring ENSO-related signals. As revealed by previous work, the 1-yr lag autocorrelation of the interannual component of the time series associated with TP snow is nonsignificant; therefore, the degree of freedom about the interannual variation in TP snow and its origins are chosen to be the number of sample years minus 2 (Qian et al. 2019; Wang et al. 2022). A two-tailed Student’s t test was used to estimate the statistical significance.
c. Model
The linear baroclinic model (LBM) was jointly developed by the Center for Climate System Research, University of Tokyo, and the National Institute for Environmental Studies, Japan. Watanabe and Kimoto (2000) provided the LBM based on a dynamical core of the atmospheric general circulation model (AGCM). This model has been utilized to simulate forcings, for example, vorticity, divergence, pressure, heating, and specific humidity, according to the corresponding observational anomalies. Here, the climatological boreal spring atmospheric circulation is used as the basic state. Idealized forcings are added to the basic state, and the model is set to run for a certain time. In the current study, the model is integrated for 25 days, and the average from day 20 to day 25 is analyzed; this represents the model’s steady response to the idealized forcings.
3. Interdecadal changes in the principal patterns of SSC_TP
The ratio of the climatology (shading) and standard deviation (contours) fields for the SSCE to those fields of winter over the TP are presented in Fig. 1. Two regions with large values of both climatology and standard deviation ratio occur over the western and eastern TP and account for more than 65% compared with that of winter over much of the western and eastern TP areas. This result indicates that the excessive SSCE and strong interannual variability in SSC_TP over the western and eastern TP are consistent with previous studies (e.g., Wang et al. 2019, 2020). Two TP snow indices are constructed by normalizing the area-weighted average of the SSCE over the two rectangles in Fig. 1 to represent the time evolutions of the SCE_TP over the western and eastern TP. To concentrate on the interannual variation, the two snow indices are high-pass filtered with a Gaussian filter where variations longer than 8 years are removed. The eastern (transparent bars in Fig. 2a) and western (solid bars in Fig. 2a) TP snow index switch from an out-phase relationship before the 1990s to an in-phase relationship after the 1990s. The sliding temporal correlation coefficients (TCCs) with a 15-yr moving window between the two TP snow indices are shown in Fig. 2b. The TCCs clearly change from negative to positive values after 1990, confirming that the relationship of the SSCE over the western and eastern TP experienced significant changes. Thus, in the following, the datasets are divided into two subperiods, that is, P1 and P2, and the possible reasons accounting for the changes in the relationship of the SSCE between the western and eastern TP are examined.
The ratio of climatology (shading) and interannual component of standard deviation (contours) maps of spring (March–May) snow-cover extent (SCE) over the TP (SSC_TP) to the counterpart of winter (December–February) during 1970–2020. The blue rectangle (29°–41°N, 68°–84°E) and purple rectangle (26°–36°N, 88°–103°E) refer to the key regions of the western and eastern TP that are utilized to construct the snow indices.
Citation: Journal of Climate 36, 11; 10.1175/JCLI-D-22-0487.1
(a) The high-filtered SSCE indices over the western (solid bars) and eastern (transparent bars) TP. (b) The sliding TCCs with a 15-yr sliding window between the high-filtered snow indices over the western and eastern TP. The horizontal purple dashed lines indicate that the TCCs are significant at a confidence level of 90%.
Citation: Journal of Climate 36, 11; 10.1175/JCLI-D-22-0487.1
The EOF analysis is applied to the SSC_TP for P1 and P2 to obtain their main variation modes. Based on the criterion of North et al. (1982), the first EOF pattern (EOF1) and the second EOF pattern (EOF2) can be mutually separated from each other and from the rest of the EOF patterns for both periods. The spatial distributions of the first two EOFs of the SSC_TP for P1 and P2 are shown in Fig. 3. The positive phase of EOF1 of the SSC_TP in P1 displays an east–west dipole structure, with the largest positive and negative anomalies prevailing over the eastern and western TP, respectively (Fig. 3a). The positive of EOF2 of the SSC_TP in P1 features a monosign pattern, with significant positive SSCE anomalies dominating the whole TP (Fig. 3b). Compared with P1, the leading two EOFs of SSC_TP in P2 (Figs. 3c,d) switch in order. Specifically, in P2, EOF1 of the SSC_TP becomes a monosign pattern, while EOF2 of the SSC_TP is an east–west dipole pattern. In other words, the EOF1 of SSC_TP (snow_EOF1) transformed from an eastern–western dipole pattern to a monopole pattern after 1990, consistent with Fig. 2b, which shows that the snow indices of the western and eastern TP changed from an out-phase to an in-phase relationship after 1990. To examine the significance of snow_EOF1, two other snow indices are calculated. In P1, a snow index is constructed by normalizing the differences in the area-weighted average of the SSCE over the purple and blue rectangles shown in Fig. 1. In P2, a snow index is constructed by normalizing the sum of the area-weighted average of the SSCE over the two rectangles. The TCCs between the corresponding time series of snow_EOF1 (PC1) and the two new snow indices are 0.95 and 0.87 for P1 and P2, respectively, passing the 99% confidence level, suggesting that snow_EOF1 can well represent the main features of the SSC_TP in both epochs.
The spatial patterns of the first two EOF patterns of the interannual variability of SSC_TP for (a),(c) EOF1 and (b),(d) EOF2 during (a),(b) P1 and (c),(d) P2. The transparent and solid dots indicate that the anomalous SSCE is significant at the 90% and 95% confidence levels, respectively.
Citation: Journal of Climate 36, 11; 10.1175/JCLI-D-22-0487.1
The changes in the local winds and geopotential height at 200 hPa and the moisture flux at 500 hPa associated with the variation in snow_EOF1 in P1 and P2 are obtained by regressing these fields onto PC1 and are presented in Fig. 4. The snow_EOF1-related anomalous local atmospheric circulations in P1 obviously differ from those in P2. During P1, corresponding to positive snow_EOF1, anomalies of high (anticyclone) and low (cyclone) heights prevail over the southwestern and southeastern TP in the upper troposphere (Fig. 4a), respectively. Anomalous southerly winds are observed on the eastern TP, which bring moist air northward from the lower-latitude oceans to the eastern TP, and water vapor convergence is also noticed (Fig. 4c). In contrast, over the western TP, anomalous northerly winds prevail, which bring dry and cold air from high latitudes southward to the western TP, accompanied by water vapor divergence anomalies (Fig. 4c). The distribution of the wind and moisture transport associated with snow_EOF1 in P1 favors more snowfall over the eastern TP and less snowfall over the western TP, an east-wet–west-dry dipole of snow_EOF1. In P2, when snow_EOF1 is in a positive phase, anomalous low (cyclone) dominates the entire TP (Fig. 4b). Along the southern and eastern flanks of this anomalous cyclone, southerlies transport moisture from the Bay of Bengal (BOB) to the TP and lead to a monopole pattern of snow_EOF1 (Fig. 4d). Additional diagnosis of the moist potential vorticity also shows that the anomalous atmospheric circulation may contribute to snow_EOF1 through vorticity advection in both epochs.
Anomalous spring (a),(b) 200-hPa geopotential height (shading; unit: gpm) and wind [vector; scale at the top-right corner of (b); unit: m s−1], (c),(d) 500-hPa moisture [vectors; scale at the top-right corner of (d); unit: kg (m s)−1] and divergence [shading; unit: 10−6 (kg2 s)−1] obtained by regression onto snow_PC1 during (a),(c) P1 and (b),(d) P2. Areas where the anomalous geopotential height in (a) and (b) and divergence in (c) and (d) are significant at the 95% confidence level are stippled. The black and purple vectors refer to the anomalous winds in (a) and (b) and moisture in (c) and (d) that are significant at the 90% and 95% confidence levels, respectively.
Citation: Journal of Climate 36, 11; 10.1175/JCLI-D-22-0487.1
Figure 4 indicates that the anomalous circulation associated with snow_EOF1 is beyond the TP region. To see the snow_EOF1-related anomalous atmospheric circulations from the global perspective, the height anomalies in the upper and lower troposphere are displayed in Fig. 5. During P1, pronounced geopotential height anomalies can be observed with alternative signs spanning Antarctica, the southern Indian Ocean, and Australia (Figs. 5a,c), a structure similar to the negative phase of AAO (−AAO). In a previous work, Dou and Wu (2018) also revealed that the AAO in May contributes to summer snow-cover variability over the western TP. In comparison, in P2, the height anomalies over the Antarctic are weak (Figs. 5b,d). Significant negative and positive height anomalies dominate the Arctic and the extratropical region between 45° and 60°N. Although some differences exist, especially over the North Pacific, where the positive anomalies are not significant, this structure bears certain similarities to the positive phase of AO. A wave train–like atmospheric pattern with alternative positive–negative–positive–negative height anomalies is also noticed, spanning from western Europe to the TP. These results suggest that the snow_EOF1 in P2 is likely linked to the climate systems over mid–high latitudes over the Northern Hemisphere. Some previous work has reported a close relationship link between the AO and SSC_TP (Bamzai 2003; Bao et al. 2018; Zhang et al. 2019; You et al. 2020), suggesting that the AO could impact TP snow variability by triggering an Arctic–TP teleconnection wave train. In the current work, we found that the AO–SSC_TP relationship experiences obvious interdecadal changes. The sliding TCC, with a 21-yr window, between the AAO/AO index and TP SCE index shows that the TCC between AAO and SSC_TP was pronounced before the 1990s, while the AO and SSC_TP increased after the 1990s (Fig. S1 in the online supplemental material).
Anomalous spring geopotential height at (a),(b) 300 and (c),(d) 850 hPa (shading; unit: gpm) obtained by regression onto PC1 during (a),(c) P1 and (b),(d) P2. Areas where the geopotential height anomalies are significant at the 95% confidence level are stippled.
Citation: Journal of Climate 36, 11; 10.1175/JCLI-D-22-0487.1
The above analysis suggests that the SSC_TP variation is closely related to −AAO over the Southern Hemisphere (SH) before 1990 and is more related to AO after 1990. It should be noted that the interdecadal changes in the AAO/AO, which may be impacted by the South (North) Atlantic decadal-to-multidecadal SST oscillation (Xue et al. 2018; Ting et al. 2014), are not exactly in accordance with the climate shift time of SSC_TP, implying that other factors may also contribute to the interdecadal changes in SSC_TP around the 1990s. In the following section, the possible reasons accounting for this change will be examined.
4. Possible mechanisms of the interdecadal changes in the SSC_TP
The AAO and AO indices are presented in Figs. 6a and 6b, respectively. The year-to-year variation in the AAO index changed from an active period to a relatively quiet period after 1990. In contrast, the magnitudes of the year-to-year variation in the AO index clearly increased after 1990. The standard deviations of the AAO index are 1.17 and 0.83 before and after 1990, respectively, whereas those of the AO index are 0.71 and 1.03 during P1 and P2, respectively. During P1, the TCC between the AAO index and PC1 is −0.48, while during P2, the TCC between the AO index and PC1 is 0.38; both can pass the confidence level of 95%. However, the TCC between the AO index and PC1 in P1 and that between the AAO index and PC1 in P2 is not significant (−0.11 and 0.02, respectively). The above results show that the AAO weakens while the AO signal intensifies after 1990, which is consistent with the interdecadal changes in snow_EOF1. This result is also consistent with the changes in the relationship between snow_EOF1 and AAO, as well as the changes between snow_EOF1 and AO.
(a) The AAO index and (b) the AO index during 1970–2020. The values at the bottom of each panel denote the standard deviation of the (a) AAO index and (b) AO index during P1 and P2.
Citation: Journal of Climate 36, 11; 10.1175/JCLI-D-22-0487.1
a. The relationship between the AAO and snow_EOF1 in P1
In this subsection, focus is placed on investigating how the AAO is related to the east-wet–west-dry dipole structure of snow_EOF1 in P1. Given that AAO and snow_EOF1 are significantly negatively correlated, in the following section, the climate anomalies associated with the AAO index are reversed in sign for a better comparison, and in the following, we use “−AAO” to represent this reverse. The anomalous SSC_TP and moisture at 500 hPa associated with −AAO are presented in Fig. 7. During P1, the −AAO is related to significant positive and negative SSC_TP anomalies over the eastern and western TP (Fig. 7a), with a pattern similar to that shown in Fig. 3a. The moisture associated with −AAO transport northward from the lower latitudes of the oceans to the TP (Fig. 7c) bears many similarities to Fig. 4c. In contrast, during P2, the SSC_TP and moisture anomalies associated with −AAO are weak and nonsignificant (Figs. 7b,d). In consideration of the possible influence of ENSO on the −AAO–SSC_TP relationship, partial regression is also calculated after removing the winter–spring ENSO signals (Fig. S2). Previous work revealed that ENSO plays an important role in the variation of spring snow cover over the TP (Wang et al. 2022); therefore, we removed the Niño-3.4 signal from the snow index and recalculated the regression map. The results are similar to those in Fig. 7, implying that the −AAO–SSC_TP relationship is independent of ENSO.
Anomalous spring (a),(b) snow-cover extent (shading; unit: %) and (c),(d) water vapor transport at 500 hPa [vectos; scale at the top-right corner of (d); unit: kg (m s)−1] and divergence [shading; unit: 10−6 (kg2 s)−1] obtained by regression onto the negative AAO index for (a),(c) P1 and (b),(d) P2. The circled and stippled regions indicate the anomalous SCE and divergence significant at the 90% and 95% confidence levels, respectively. The black and purple vectors in (c) and (d) refer to the anomalous moisture fluxes that are significant at the 90% and 95% confidence levels, respectively.
Citation: Journal of Climate 36, 11; 10.1175/JCLI-D-22-0487.1
To shed light on how the −AAO over the Southern Hemisphere can influence the SSC_TP variation in P1, the −AAO-associated atmospheric circulation anomalies are given in Fig. 8. In the lower troposphere, corresponding to −AAO, a significant anticyclone–cyclone–anticyclone in the form of a wave train can be observed spanning the Antarctic–southern Indian Ocean–Australian regions (Fig. 8a). The two centers of the wave train over the Antarctic and the southern Indian Ocean can also be clearly seen in the upper troposphere (Fig. 8b), suggesting a barotropic structure in the mid- to high-latitude Southern Hemisphere. However, the wave train transforms into a baroclinic structure when it reaches the tropical region, where an anticyclone and a cyclone prevail in Australia in the lower and upper troposphere, respectively. The baroclinic structure of the wave train component over lower latitudes favors a sinking motion around Australia (18°S, 130°E; Fig. 8d), where significant divergent wind can be found in the low troposphere (Fig. 8c). The southeasterly along the east flank of this low-level anticyclonic system can cross the Bay of Bengal and reach the eastern TP (Figs. 8a,c), which extends from the lower troposphere to the midlevel (Fig. 8d). The anomalous southeasterly is in accordance with the climatological cross-equatorial airflow, whereby the strengthened cross-equatorial airflow is more conducive to transporting water vapor from the tropical Pacific Ocean to the eastern TP and causing water vapor convergence, consistent with Fig. 7c. Moreover, significant upward motion can be observed above the eastern TP (Fig. 8d), which is also a condition favoring more precipitation there. In the upper level, the western TP is dominated by anomalous northerly winds that bring cold and dry flow from high latitudes to the western TP (Fig. 8b), causing dry conditions over the western TP. Figure 8 suggests that the anomalous circulation associated with −AAO favors more and less than normal snowfall over the eastern and western TP, respectively, thereby contributing to the east–west dipole structure of snow_EOF1 in P1.
Anomalous spring wind at (a) 850 and (b) 150 hPa (vectors; unit: m s−1) and (c) velocity potential at 850 hPa (shading; unit: 105 × m2 s−1) and divergent winds [vectors; scale at the top-right corner of (c); unit: m s−1] obtained by regressing against the negative AAO index during P1. (d) The vertical winds (vectors; vertical velocity × (scale factor −100); unit: m s−1) and velocity (shading; unit: 10−3 × Pa s−1) along the blue line in (c) obtained by regressing against the negative AAO index during P1. The black and purple vectors refer to the anomalous wind in (a) and (b), divergent winds in (c) and vertical winds in (d) that are significant at the 90% and 95% confidence levels, respectively.
Citation: Journal of Climate 36, 11; 10.1175/JCLI-D-22-0487.1
Given that excessive SSC_TP can also induce a TP cyclone system by the diabatic cooling effect (Wang et al. 2020), the −AAO-associated atmospheric circulation after removing the SSC_TP signals is also investigated by partial regression after removing PC1 of SSC_TP. It is found that signals over the northwestern Pacific and coast of East Asia are decreased, implying that the SSC_TP exerts impacts on surrounding circulations. However, the low-level cross-equatorial southeasterly and mid–high-level TP cyclone can be clearly observed, which indicates that the relationship between −AAO and the dipole structure for snow_EOF1 is robust. We also examined the −AAO-associated atmospheric circulation anomalies for P2 (Fig. S3). In P2, the anomalous atmospheric circulations associated with −AAO are weak; therefore, the airflow cannot cross the equator. Hence, the relationship between −AAO and SSC_TP is not established in P2 (Fig. 5b).
As the memory of the atmosphere is not more than half a month, the formation and maintenance of the hemispheric scale wave train could be maintained by external forcing or the basis flow. To determine how the −AAO is related to the atmospheric wave train in the Southern Hemisphere in P1, the RWS and divergent winds associated with −AAO are calculated (Fig. 9a). Associated with −AAO, pronounced divergent winds are observed spanning the Antarctic (Fig. 9a; vectors). As proposed by previous works (e.g., Watanabe 2004; Wu et al. 2020), divergent winds are vital contributors to the potential vorticity density, which can be considered an effective RWS, consistent with Fig. 9a, where pronounced positive RWS anomalies are noticed over the Antarctic region. This RWS indicates the source of the AAO-related Southern Hemisphere atmospheric wave train. Note that strong climatological winds prevail over the Southern Hemisphere approximately 60°S (Fig. 9b). To examine the possible contributions of the basic flow to the propagation of the AAO-related RWS and the wave train, the conversion of CK and CP associated with −AAO are examined and shown in Figs. 9b and 9c: these represent the barotropic and baroclinic energies obtained from the basic flow. Large values of CP and CK lie over the Southern Ocean and the edge of the Antarctic continent. CP is obviously larger than CK, indicating that the atmospheric wave train is mainly maintained by CP. In contrast, CK and CP are feeble in tropical regions, implying that other processes may play a role in the further northward propagation of the wave train, for example, upper-tropospheric divergence motion associated with tropical precipitation (Dou and Wu 2018).
Anomalous spring (a) RWS at 850 hPa (shading; unit: 10−11 s−2) and divergent wind [vectors; scale at the top-right corner of (a); unit: m s−1], (b) barotropic energy conversion (CK; unit: 10−5 × W m−2), and (c) baroclinic energy conversion (CP; unit: 10−5 × W m−2) obtained by regressing against the negative AAO index during P1. The vectors [scale at the top-right corner of (b); unit: m s−1] in (b) refer to the spring climatological-mean 850-hPa wind during P1, and blue vectors refer to velocities greater than 9 m s−1.
Citation: Journal of Climate 36, 11; 10.1175/JCLI-D-22-0487.1
Then, we further performed an Antarctic forcing experiment with the LBM to confirm the relationship between the −AAO and the atmospheric wave train. In this experiment, the basic state used is the observational climatology of the spring circulation field during P1. An idealized divergence forcing (Fig. 10a) is imposed on the first day of the integration and is persistent during the model run. The idealized forcing is used to represent the anomalous divergence induced by −AAO over the Antarctic (Fig. 10a) and is designed based on the observational divergent wind and divergence anomalies related to −AAO. Details of the model set, for example, forcing domains and peak value at the sigma level, can be found in Table 1. The model response of the 850-hPa wind from day 25 to day 30 is displayed in Fig. 10b. A clear anticyclone–cyclone–anticyclone with alternative signs can be found spanning Antarctic–Southern Ocean–Australian regions, a distribution similar to that in the observations (Fig. 8a). Differences can also be found between the model response and the observational results: for example, the cyclone over the south Indian Ocean in Fig. 10b is located more eastward than the observations; the weaker cross-equatorial flow and the upper-tropospheric eastward shifting of the Australian cyclone (Fig. S4); the southeasterly winds near the TP bear differences compared with the observations, probably due to the absence of a nonlinear process in the LBM and the simplicity of the idealized forcing designed in the experiment. In general, the Antarctic model experiments provide additional evidence that the −AAO can generate an atmospheric wave train response in the SH. Specifically, the −AAO-associated upper divergent wind anomalies in the Antarctic play the role of the RWS there and propagate northward in the form of a Rossby wave train (Fig. 9).
(a) Spatial distributions of the idealized convergence forcing (shading, unit: 10−6 s−1) in the numerical experiments for the basic state of P1. (b) Model response of 850-hPa wind averaged for day 20 to day 25 (vectors; unit: m s−1).
Citation: Journal of Climate 36, 11; 10.1175/JCLI-D-22-0487.1
Details of Antarctic and Greenland forcing experiments.
b. The relationship between the AO and snow_EOF1 in P2
In this subsection, we attempt to determine the possible mechanisms through which the AO contributes to the monopole pattern of the SSC_TP after 1990. Figure 11 illustrates the SSC_TP and the moisture anomalies at 500 hPa associated with the AO index for P1 and P2. Corresponding to the positive AO in P1, the anomalous SSC_TP and moisture associated with AO are generally weak over the TP (Figs. 11a,c), implying that the AO has a weak impact on the SSC_TP variation during this period. In contrast, corresponding to the positive AO in P2, the AO is related to significant positive SSC_TP anomalies dominating the whole TP (Fig. 11b), similar to Fig. 3c. The moisture flux transported from the Bay of Bengal to the TP (Fig. 11d) bears many similarities to that shown in Fig. 4d. Considering the possible impact of ENSO on the AO–SSC_TP relationship, a partial regression similar to Fig. 11 is also calculated after removing the winter–spring ENSO signal (Fig. S5). We found that the results resemble Fig. 11, indicating that the relationship between the AO and the monopole structure of SSC_TP in P2 is independent of ENSO.
Anomalous spring (a),(b) snow-cover extent (shading; unit: %) and (c),(d) 500-hPa moisture [vectors; scale at the top-right corner of (d); unit: kg (m s)−1] and divergence [shading; unit: 10−6 (kg2 s)−1] obtained by regressing onto the AO index for (a),(c) P1 and (b),(d) P2. The circled and stippled regions indicate the anomalous SCEs that are significant at the 90% and 95% confidence levels, respectively. The black and purple vectors in (c) and (d) refer to the anomalous moisture at the 90% and 95% confidence levels, respectively.
Citation: Journal of Climate 36, 11; 10.1175/JCLI-D-22-0487.1
The AO-associated atmospheric circulation anomalies in P1 are presented in Fig. 12a. Corresponding to a positive AO, two branches of wave trains stemming from southern Greenland can be detected dominating the Eurasian continent along 60° and 30°N (marked by the black vectors), consistent with Zhang et al. (2019). The higher-latitude branch of the wave train spans the North Atlantic, western Europe, and northern Russia and terminates south of Lake Baikal. The lower-latitude branch of the wave train propagates southeastward, crosses northern Africa, and reaches the Mideast area. Neither of the two branches of wave trains reach the TP area. The magnitudes of the WAF are weaker than its counterpart in P2, which might be due to the weakened AO in P1, as shown in Fig. 6b.
Anomalous spring WAF at 200 hPa (vectors; unit: m2 s−2) and vorticity (shading; unit: 10−6 s−1) obtained by regressing onto the AO index for (a) P1 and (b) P2. The stippled regions indicate that the anomalous vorticity is significant at the 95% confidence level.
Citation: Journal of Climate 36, 11; 10.1175/JCLI-D-22-0487.1
Positive vorticity anomalies can be noticed around the TP and nearby Bay of Bengal regions, which are conducive to significant TP ascending airflow. Meanwhile, a local cyclone system sweeps over the positive vorticity regions extending from the middle to upper troposphere (Fig. S6). Along the eastern flank of the cyclone, anomalous southerlies bring moisture from the BOB to the TP, favoring the above-normal SSC_TP during P2.
To determine the relationship between the AO and the atmospheric wave train–like pattern spanning the North Atlantic–TP during P2, we checked the AO-associated vertical velocity and precipitation over the North Atlantic, as shown in Figs. 13a and 13b. The AO-related anomalous significant ascending airflow (Fig. 13a), which is associated with positive precipitation over southeastern Greenland (Fig. 13b), plays an essential role in generating the pronounced negative RWS there during P2. The AO-induced CK (Fig. 13c) and CP (Fig. 13d) in P2 were examined. The magnitudes of the CP are much larger than CK, but its distribution is inconsistent with the route of the wave train, whereas the alternative signs of CK along the subtropical westerly jet are just in accordance with the route of the wave train, suggesting that the CK plays an important role in maintaining the atmospheric wave train by extracting barotropic energy from the basic flow during P2. Meanwhile, both CK and CP are feeble over the TP and its surroundings, implying that other processes also play a role in the wave train, for example, transient eddies (Liu et al. 2020).
Anomalous spring (a) vertical velocity at 500 hPa (shading; unit: 10−3 × Pa s−1), (b) precipitation (shading; unit: mm day−1), (c) CK at 200 hPa (shading; unit: 10−5 × W m−2), and (d) CP at 200 hPa (shading; unit: 10−5 × W m−2) obtained by regressing onto the AO index during P2. The contours in (c) and (d) refer to the climatological means of spring zonal wind that are larger than 15 m s−1 during P2. The stippled regions indicate that the anomalous vertical velocity in (a) and precipitation in (b) are significant at the 95% confidence level.
Citation: Journal of Climate 36, 11; 10.1175/JCLI-D-22-0487.1
Then, the LBM is used to perform experiments to further verify the role of the AO in the atmospheric wave train. In the experiment, a prescribed positive vorticity forcing over southern Greenland is imposed (Fig. 14a) according to Fig. 12b. The basic state used is the climatological mean of the spring circulation field during P2. The forcing has a gamma profile in the vertical direction, and its details can be found in Table 1. The model response of the 200-hPa wind and vorticity averaged from day 20 to day 25 is depicted in Fig. 14b. An obvious wave train can be noticed, with alternative signs of vorticity originating from southern Greenland, propagating to the North Atlantic, crossing northern Africa, and terminating over the TP, consistent with the observed result in Fig. 12b. At the middle level, the cyclone lying over the northern Bay of Bengal could transport moisture to the TP, thereby favoring the SSC_TP in P2. Differences can also be found between the model response and the observational results. For example, the negative vorticity over the Barents Sea in Fig. 14b is not clear in the observations. However, generally, the model results reproduce the basic structure of the atmospheric wave train and can provide additional support for the impact of the AO on the SSC_TP.
(a) Spatial distributions of the prescribed vorticity forcing (shading; unit: 10−6 s−1) in the numerical experiments for the basic state of P2 and (b) model response of 200-hPa wind (vectors; unit: m s−1) and vorticity (shading; unit: 10−6 s−1) averaged from day 20 to day 25.
Citation: Journal of Climate 36, 11; 10.1175/JCLI-D-22-0487.1
5. Conclusions and discussion
This study investigated the causes of interdecadal changes in the leading EOF pattern of the SSC_TP (snow_EOF1) by analyzing observational data and performing numerical experiments using an atmospheric model. The results show that snow_EOF1 experienced a prominent shift in 1990. The datasets are divided into two subperiods: before 1990 (P1) and after 1990 (P2). In P1, snow_EOF1 features a dipole-like pattern, whereas it converts into a monosign structure in P2. The possible mechanisms accounting for the interdecadal changes in the SSC_TP pattern are examined from the atmospheric circulation viewpoint.
Analysis shows that the TCCs between the −AAO index and the time series associated with snow_EOF1 (snow_PC1) are only significant in P1 and that the amplitudes of the AAO index are obviously stronger during this period compared with P2. During P1, the −AAO induced an atmospheric wave train that propagates northward from the Antarctic, spanning the southern Indian Ocean to Australia (Figs. 8a,b). The components of the wave train over Australia appear to be anticyclones and cyclones in the lower and upper troposphere (Figs. 8a,b), respectively, favoring a sinking motion there. In the lower troposphere (Fig. 8d), along the east flank of the Australian anticyclone, southeasterly winds cross the equator and transport water vapor to the eastern TP (Figs. 7c and 8a). Moreover, northerly winds transport dry flow from high latitudes to the western TP (Fig. 8b). The above circulation anomalies favor an east–west dipole-like SSC_TP pattern in P1. In contrast, during P2, the AAO is weak, which results in a weak AAO–SSC_TP relationship. Further examination shows that the −AAO induces anomalous divergence over the Antarctic, which plays an important role in the formation of the positive RWS there (Fig. 9a). The RWS can induce atmospheric wave trains and propagate northward to Australia. The wave train is maintained mainly by the CP by gaining baroclinic energy from the basic flow (Figs. 9b,c).
In P2, TCCs between the AO index and snow_PC1 are significant, and the amplitudes of the AO index are stronger in this period than in P1. The AO is associated with a wave train spanning southern Greenland, propagates eastward, crosses the North Atlantic and northern Africa, and terminates over the TP (Fig. 12b). The atmospheric wave train leads to pronounced positive vorticity anomalies over the TP and nearby regions. Southerly winds associated with the system transport moisture from the BOB to the TP, favoring more SSC_TP (Fig. 11d). In contrast, during P1, the AO-induced wave trains cannot reach the TP and therefore have no clear impact on SSC_TP. In addition, the formation and maintenance of the wave pattern associated with AO in P2 are examined. The AO is associated with anomalous divergence over southeastern Greenland, which plays a crucial role in forming the RWS there (Figs. 13a,b). The RWS is responsible for the cross-Eurasian continent wave train (Fig. 12b). Further analysis shows that the CK plays an essential role in maintaining the wave train by obtaining barotropic energy from the basic subtropical westerly jet during P2 (Figs. 13c,d).
In addition, we also identified SSC_TP-related SST signals for both periods (not shown). The result shows that ENSO and IOD (Indian Ocean dipole) could not be linked with the spring snow cover over the entire TP domain for both periods, since ENSO and IOD could only influence eastern and western TP snow, respectively (Jiang et al. 2019). We also analyzed the possible linkage of the pronounced SST anomalies associated with SSC_TP. The tripole-like North Atlantic SST anomalies were statistically correlated with the dipole-like SSC_TP, but they failed to connect with the western SSC_TP for P1. Tropical Atlantic SST anomalies could only link with the SSC_TP for P2.
It is true that the correlation between the AO/AAO and the SSC_TP is not very high due to other factors, for example, local precipitation and temperature (You et al. 2020). However, they are still significant at the 95% confidence level. Although the current work focuses on the interannual variation of the TP snow and its relationship with AAO and AO, we also examined the changes in the unfiltered TP snow indices and their relationship with the AAO and AO. It appears that the interdecadal changes of the TP snow between the eastern and western TP, as well as the changed relationship among the TP snow, the AAO, and AO, are generally consistent with the results presented in this work, suggesting that the changes in the TP snow revealed here are stable. In contrast, although the AO/AAO may not be the greatest potential contributor to SSC_TP, it favors the primary information in the patterns of spring snow-cover anomalies over the entire TP domain for both periods. We also explored whether other patterns with longer time scales could modulate the TP snow. It was found that the preceding January–February–March North Pacific Oscillation (NPO) may also contribute to the changes in the spring TP snow. Further examinations of the relative contributions to the interdecadal changes of the TP snow from multiple factors need to be carried out in the future.
This work attempts to interpret the mechanisms underlying the changes in the SSC_TP pattern from the atmospheric perspective. However, other factors, for example., North Atlantic SSTA, tropical ocean SSTA, and Arctic sea ice, may also exert influences on the changes; this is not clear and needs further investigation in future work.
Acknowledgments.
This research is funded by the National Natural Science Foundation of China (Grants 42030602 and 42075050) and by Fundamental Research Funds for the Central Universities (K20220232). Zhang Chao is also supported by the Outstanding Postdoctoral Scholarship, State Key Laboratory of Marine Environmental Science at Xiamen University.
Data availability statement.
The monthly mean Northern Hemisphere SCE from the NOAA climate data record is available at https://www.ncei.noaa.gov/access/metadata/landing-page/bin/iso?id=gov.noaa.ncdc:C00756. The monthly mean reanalysis datasets from ERA5 can be derived from https://www.ecmwf.int/en/forecasts/datasets/reanalysis-datasets/era5. The monthly mean global precipitation dataset from the Global Precipitation Climatology Project version 2.2 is provided by https://www.esrl.noaa.gov/psd/data/gridded/. The monthly mean global SST data are from HadISST, which is available at https://www.metoffice.gov.uk/hadobs/hadisst/data/.
REFERENCES
Abe, M., A. Kitoh, and T. Yasunari, 2003: An evolution of the Asian summer monsoon associated with mountain uplift—Simulation with the MRI atmosphere-ocean coupled GCM. J. Meteor. Soc. Japan, 81, 909–933, https://doi.org/10.2151/jmsj.81.909.
Adler, R. F., and Coauthors, 2003: The Version-2 Global Precipitation Climatology Project (GPCP) monthly precipitation analysis (1979–present). J. Hydrometeor., 4, 1147–1167, https://doi.org/10.1175/1525-7541(2003)004<1147:TVGPCP>2.0.CO;2.
Alves, M. P. A., R. B. Silveira, R. B. Minuzzi, and A. E. Franke, 2017: The influence of the Antarctic Oscillation (AAO) on cold waves and occurrence of frosts in the state of Santa Catarina, Brazil. Climate, 5, 17, https://doi.org/10.3390/cli5010017.
Bamzai, A. S., 2003: Relationship between snow cover variability and Arctic Oscillation index on a hierarchy of time scales. Int. J. Climatol., 23, 131–142, https://doi.org/10.1002/joc.854.
Bao, Y., and Q. You, 2019: How do westerly jet streams regulate the winter snow depth over the Tibetan Plateau? Climate Dyn., 53, 353–370, https://doi.org/10.1007/s00382-018-4589-1.
Bao, Y., Q. You, and X. Xie, 2018: Spatial-temporal variability of the snow depth over the Qinghai-Tibetan Plateau and the cause of its interannual variation. Plateau Meteor., 37, 899–910.
Barnett, T. P., L. Dümenil, U. Schlese, and E. Roeckner, 1988: The effect of Eurasian snow cover on global climate. Science, 239, 504–507, https://doi.org/10.1126/science.239.4839.504.
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.
Dou, J., and Z. Wu, 2018: Southern Hemisphere origins for interannual variations of Tibetan Plateau snow cover in boreal summer. J. Climate, 31, 7701–7718, https://doi.org/10.1175/JCLI-D-17-0327.1.
Duan, A. M., and G. X. Wu, 2005: Role of the Tibetan Plateau thermal forcing in the summer climate patterns over subtropical Asia. Climate Dyn., 24, 793–807, https://doi.org/10.1007/s00382-004-0488-8.
Fan, K., Z. Xu, and B. Tian, 2014: Has the intensity of the interannual variability in summer rainfall over South China remarkably increased? Meteor. Atmos. Phys., 124, 23–32, https://doi.org/10.1007/s00703-013-0301-5.
Hafez, Y., 2017: On the relationship between heat waves over the western and central Europe and NAO, SOI, El-Nino 3.4 in summer 2015. J. Geosci. Environ. Prot., 5, 31–45, https://doi.org/10.4236/gep.2017.54004.
Hersbach, H., and Coauthors, 2020: The ERA5 global reanalysis. Quart. J. Roy. Meteor. Soc., 146, 1999–2049, https://doi.org/10.1002/qj.3803.
Hoskins, B. J., and D. J. Karoly, 1981: The steady linear response of a spherical atmosphere to thermal and orographic forcing. J. Atmos. Sci., 38, 1179–1196, https://doi.org/10.1175/1520‐0469(1981)038<1179:TSLROA>2.0.CO;2.
Hoskins, B. J., I. N. James, and G. H. White, 1983: The shape, propagation and mean-flow interaction of large-scale weather systems. J. Atmos. Sci., 40, 1595–1612, https://doi.org/10.1175/1520-0469(1983)040<1595:TSPAMF>2.0.CO;2.
Jia, X., C. Zhang, R. Wu, and Q. Qian, 2021: Influence of Tibetan Plateau autumn snow cover on interannual variations in spring precipitation over southern China. Climate Dyn., 56, 767–782, https://doi.org/10.1007/s00382-020-05497-8.
Jiang, X., T. Zhang, C.-Y. Tam, J. Chen, N.-C. Lau, S. Yang, and Z. Wang, 2019: Impacts of ENSO and IOD on snow depth over the Tibetan Plateau: Roles of convections over the western North Pacific and Indian Ocean. J. Geophys. Res. Atmos., 124, 11 961–11 975, https://doi.org/10.1029/2019JD031384.
Kosaka, Y., and H. Nakamura, 2006: Structure and dynamics of the summertime Pacific–Japan teleconnection pattern. Quart. J. Roy. Meteor. Soc., 132, 2009–2030, https://doi.org/10.1256/qj.05.204.
Li, W., W. Guo, B. Qiu, Y. Xue, P.-C. Hsu, and J. Wei, 2018: Influence of Tibetan Plateau snow cover on East Asian atmospheric circulation at medium-range time scales. Nat. Commun., 9, 4243, https://doi.org/10.1038/s41467-018-06762-5.
Liu, G., R. Wu, Y. Zhang, and S. Nan, 2014: The summer snow cover anomaly over the Tibetan Plateau and its association with simultaneous precipitation over the mei-yu-baiu region. Adv. Atmos. Sci., 31, 755–764, https://doi.org/10.1007/s00376-013-3183-z.
Liu, Y., M. Lu, H. Yang, A. Duan, B. He, S. Yang, and G. Wu, 2020: Land–atmosphere–ocean coupling associated with the Tibetan Plateau and its climate impacts. Natl. Sci. Rev., 7, 534–552, https://doi.org/10.1093/nsr/nwaa011.
Mao, J.-Y., 2010: Interannual variability of snow depth over the Tibetan Plateau and its associated atmospheric circulation anomalies. Atmos. Ocean. Sci. Lett., 3, 213–218, https://doi.org/10.1080/16742834.2010.11446875.
North, G. R., T. L. Bell, R. F. Cahalan, and F. J. Moeng, 1982: Sampling errors in the estimation of empirical orthogonal functions. Mon. Wea. Rev., 110, 699–706, https://doi.org/10.1175/1520-0493(1982)110<0699:SEITEO>2.0.CO;2.
Qian, Q. F., X. J. Jia, and R. G. Wu, 2019: Changes in the impact of the autumn Tibetan Plateau snow cover on the winter temperature over North America in the mid-1990s. J. Geophys. Res. Atmos., 124, 10 321–10 343, https://doi.org/10.1029/2019JD030245.
Qian, Q. F., X. J. Jia, and R. Wu, 2020: On the interdecadal change in the interannual variation in autumn snow cover over the central eastern Tibetan Plateau in the mid‐1990s. J. Geophys. Res. Atmos., 125, e2020JD032685, https://doi.org/10.1029/2020JD032685.
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.
Robinson, D. A., T. W. Estilow, and NOAA CDR Program, 2012: NOAA Climate Data Record (CDR) of Northern Hemisphere (NH) snow cover extent (SSCE), version 1 [1972–2009]. NOAA National Centers for Environmental Information, accessed 4 August 2020, https://doi.org/10.7289/V5N014G9.
Sardeshmukh, P. D., and B. J. Hoskins, 1988: The generation of global rotational flow by steady idealized tropical divergence. J. Atmos. Sci., 45, 1228–1251, https://doi.org/10.1175/1520-0469(1988)045<1228:TGOGRF>2.0.CO;2.
Shaw, T. A., and A. Voigt, 2015: Tug of war on summertime circulation between radiative forcing and sea surface warming. Nat. Geosci., 8, 560–566, https://doi.org/10.1038/ngeo2449.
Si, D., and Y. H. Ding, 2013: Decadal change in the correlation pattern between the Tibetan Plateau winter snow and the East Asian summer precipitation during 1979–2011. J. Climate, 26, 7622–7634, https://doi.org/10.1175/JCLI-D-12-00587.1.
Son, J.-H., K.-H. Seo, and B. Wang, 2020: How does the Tibetan Plateau dynamically affect downstream monsoon precipitation? Geophys. Res. Lett., 47, e2020GL090543, https://doi.org/10.1029/2020GL090543.
Takaya, K., and H. Nakamura, 2001: A formulation of a phase‐independent wave‐activity flux for stationary and migratory quasigeostrophic eddies on a zonally varying basic flow. J. Atmos. Sci., 58, 608–627, https://doi.org/10.1175/1520-0469(2001)058<0608:AFOAPI>2.0.CO;2.
Tang, Y., A. Duan, and J. Hu, 2022: Surface heating over the Tibetan Plateau associated with the Antarctic Oscillation. J. Geophys. Res. Atmos., 127, e2022JD036851, https://doi.org/10.1029/2022JD036851.
Ting, M., Y. Kushnir, and C. Li, 2014: North Atlantic multidecadal SST oscillation: External forcing versus internal variability. J. Mar. Syst., 133, 27–38, https://doi.org/10.1016/j.jmarsys.2013.07.006.
Valdes, P. J., and B. J. Hoskins, 1991: Nonlinear orographically forced planetary waves. J. Atmos. Sci., 48, 2089–2106, https://doi.org/10.1175/1520-0469(1991)048<2089:NOFPW>2.0.CO;2.
Wang, Y., and X. Xu, 2018: Impact of ENSO on the thermal condition over the Tibetan Plateau. J. Meteor. Soc. Japan, 96, 269–281, https://doi.org/10.2151/jmsj.2018-032.
Wang, Z., R. Wu, S. Chen, G. Huang, G. Liu, and L. Zhu, 2018: Influence of western Tibetan Plateau summer snow cover on East Asian summer rainfall. J. Geophys. Res. Atmos., 123, 2371–2386, https://doi.org/10.1002/2017JD028016.
Wang, Z., R. Wu, P. Zhao, S.-L. Yao, and X. Jia, 2019: Formation of snow cover anomalies over the Tibetan Plateau in cold seasons. J. Geophys. Res. Atmos., 124, 4873–4890, https://doi.org/10.1029/2018JD029525.
Wang, Z., R. Wu, A. Duan, and X. Qu, 2020: Influence of eastern Tibetan Plateau spring snow cover on North American air temperature and its interdecadal change. J. Climate, 33, 5123–5139, https://doi.org/10.1175/JCLI-D-19-0455.1.
Wang, Z., R. Wu, S. Yang, and M. Lu, 2022: An interdecadal change in the influence of ENSO on the spring Tibetan Plateau snow-cover variability in the early 2000s. J. Climate, 35, 725–743, https://doi.org/10.1175/JCLI-D-21-0348.1.
Watanabe, M., 2004: Asian jet waveguide and a downstream extension of the North Atlantic Oscillation. J. Climate, 17, 4674–4691, https://doi.org/10.1175/JCLI-3228.1.
Watanabe, M., and M. Kimoto, 2000: Atmosphere‐ocean thermal coupling in the North Atlantic: A positive feedback. Quart. J. Roy. Meteor. Soc., 126, 3343–3369, https://doi.org/10.1002/qj.49712657017.
Wu, G., and Coauthors, 2007: The influence of mechanical and thermal forcing by the Tibetan Plateau on Asian climate. J. Hydrometeor., 8, 770–789, https://doi.org/10.1175/JHM609.1.
Wu, G., Y. Liu, B. Dong, X. Liang, A. Duan, Q. Bao, and J. Yu, 2012: Revisiting Asian monsoon formation and change associated with Tibetan Plateau forcing. I: Formation. Climate Dyn., 39, 1169–1181, https://doi.org/10.1007/s00382-012-1334-z.
Wu, G., T. Ma, Y. Liu, and Z. Jiang, 2020: PV-Q perspective of cyclogenesis and vertical velocity development downstream of the Tibetan Plateau. J. Geophys. Res. Atmos., 125, e2019JD03912, https://doi.org/10.1029/2019JD030912.
Xiao, Z., and A. Duan, 2016: Impacts of Tibetan Plateau snow cover on the interannual variability of the East Asian summer monsoon. J. Climate, 29, 8495–8514, https://doi.org/10.1175/JCLI-D-16-0029.1.
Xue, J., J. Li, C. Sun, S. Zhao, J. Mao, D. Dong, Y. Li, and J. Feng, 2018: Decadal-scale teleconnection between South Atlantic SST and southeast Australia surface air temperature in austral summer. Climate Dyn., 50, 2687–2703, https://doi.org/10.1007/s00382-017-3764-0.
You, Q., and Coauthors, 2020: Review of snow cover variation over the Tibetan Plateau and its influence on the broad climate system. Earth-Sci. Rev., 201, 103043, https://doi.org/10.1016/j.earscirev.2019.103043.
Yuan, C., T. Tozuka, T. Miyasaka, and T. Yamagata, 2009: Respective influences of IOD and ENSO on the Tibetan snow cover in early winter. Climate Dyn., 33, 509–520, https://doi.org/10.1007/s00382-008-0495-2.
Yuan, C., T. Tozuka, and T. Yamagata, 2012: IOD influence on the early winter Tibetan Plateau snow cover: Diagnostic analyses and an AGCM simulation. Climate Dyn., 39, 1643–1660, https://doi.org/10.1007/s00382-011-1204-0.
Zhang, C., and X. Jia, 2022: The seasonal evolution of the Tibetan Plateau snow cover related moisture during spring-to-summer. J. Geophys. Res. Atmos., 127, e2022JD036560, https://doi.org/10.1029/2022JD036560.
Zhang, C., X. Jia, and Z. Wen, 2021: Increased impact of the Tibetan Plateau spring snow cover to the mei-yu rainfall over the Yangtze River valley after 1990s. J. Climate, 34, 5985–5997, https://doi.org/10.1175/JCLI-D-21-0009.1.
Zhang, C., Y. Guo, and Z. Wen, 2022: Interdecadal change in the effect of Tibetan Plateau snow cover on spring precipitation over eastern China around the early 1990s. Climate Dyn., 58, 2807–2824, https://doi.org/10.1007/s00382-021-06035-w.
Zhang, Y., T. Li, and B. Wang, 2004: Decadal change of the spring snow depth over the Tibetan Plateau: The associated circulation and influence on the East Asian summer monsoon. J. Climate, 17, 2780–2793, https://doi.org/10.1175/1520-0442(2004)017<2780:DCOTSS>2.0.CO;2.
Zhang, Y., T. Zou, and Y. Xue, 2019: An Arctic‐Tibetan connection on subseasonal to seasonal time scale. Geophys. Res. Lett., 46, 2790–2799, https://doi.org/10.1029/2018GL081476.