Different mechanisms for the extremely hot central-eastern China in July–August 2022 from a Eurasian large-scale circulation perspective

In the summer of 2022, unprecedented and consecutive heat events occurred in many regions in China. Especially in July and August, heatwaves spreading across central and eastern China (CEC) were ascertained to be the longest-running and most intense at least since 1961 (Lu et al 2022, Mallapaty 2022). The most affected area was in the Yangtze River (YR) basin, a highly populated region in China. A large portion of the YR basin reached record high temperatures (figures 1(a) and (b)), with surface air temperature (SAT) anomalies greater than 5 °C in both upper and lower reaches of the YR in August (figure 1(b)). According to Nature News, about 360 million people experienced temperatures exceeding 40 °C at some point during this summer in China (Mallapaty 2022). Accompanying the extreme heat, severe droughts and wildfires also attacked CEC. For instance, the water level of Poyang Lake, China's largest freshwater lake in the YR basin, dropped about 10 m by the end of August; and in the same month, worst wildfires occurred in Chongqing, a municipality upstream of the YR basin. This extreme heat presented a profound social impact, including harms on human health, food security, and terrestrial ecosystems, which urgently call for a careful investigation on the formation mechanisms.

Zoom In Zoom Out Reset image size

Extreme events, in essence, are strongly influenced by internal climate variability, such as changes in atmospheric circulation (Horton et al 2015). Many studies have indicated that heatwaves can be easily driven by persistent large-scale anomalous anticyclone through adiabatic heating from descent and incoming solar radiation (e.g. Yiou and Nogaj 2004, Barriopedro et al 2011, Horton et al 2015, Zheng and Wang 2019, Wang et al 2022a). In East Asia, the local anomalous anticyclones associated with heatwaves are usually associated with the western Pacific subtropical high (WPSH) (Chen et al 2019, Liu et al 2019, Ren et al 2020) and mid-latitude atmospheric teleconnections (Lee et al 2017, Deng et al 2020, Zhang et al 2021). It is worth noting that teleconnections can sometimes lead to simultaneous extreme events in multiple regions (Kornhuber et al 2020, Rogers et al 2022). For example, concurrent heatwaves or amplified warmings over Europe and North Asia were observed under the influence of mid-latitude wave trains, such as the silk road pattern (SRP, Lu et al 2002, Enomoto et al 2003) over the Eurasian continent (Hong et al 2017, Deng et al 2018). Moreover, teleconnections may induce simultaneous heatwaves and floods across the Northern Hemisphere (Lau and Kim 2012, Schubert et al 2014, Kornhuber et al 2019, Capua et al 2021). In fact, simultaneous extremes also appeared in summer 2022. During the extremely hot CEC, severe floods in Pakistan and record-breaking heatwaves in Europe occurred in July and August (see the left panel of figure 1), which caused massive losses of life and economy. However, little is known about whether these extreme events over the Eurasian continent are linked.

Besides the changes in atmospheric circulation, human-induced global warming can also influence extreme events, particularly heatwaves (e.g. Stott et al 2004, Rahmstorf and Coumou 2011, Sun et al 2015, Ma et al 2017, Wang et al 2020, Zittis et al 2022). The increase in extreme heat is regarded as one of the greatest threats posed by global warming (McMichael et al 2006). According to the time series of monthly mean daily maximum SAT, significant linear warming trends existed in both YR basin (figures 1(b) and (d)) and Europe (figure S1) during the past decades, suggesting the highest temperatures in 2022 were partly due to the global warming. Note that these values in 2022 (except those over the selected European region in July) remain the maximum even after we eliminate the linear trend. Moreover, same features are found in the time series of precipitation anomalies in northwestern South Asia (NWSA), where record-breaking downpour occurred in July and August (figures 1(f) and (h)). Thus, interannual climate variability might have been mainly responsible for the occurrence of the extreme events in 2022 over the Eurasian continent. In the present study, we will focus on physical processes of the CEC extreme heat from a Eurasian large-scale circulation perspective based on the observational and reanalysis datasets, which can link with the European heatwaves and NWSA downpour. Particularly, we highlight the different mechanisms responsible for the CEC heatwaves in July and August.

The following observational and reanalysis datasets are used in this study: (a) daily maximum SAT from the Climate Prediction Center Global Unified Temperature data (at 0.5° × 0.5° resolution) provided by the National Oceanic and Atmospheric Administration (https://psl.noaa.gov/data/gridded/data.cpc.globaltemp.html); (b) monthly precipitation from the Global Precipitation Climatology Project version 2.3 at 2.5° × 2.5° resolution (Adler et al 2003); and (c) monthly atmospheric variables from the Japanese 55 year Reanalysis (JRA-55; Kobayashi et al 2015) at 1.25° × 1.25° resolution. All the datasets used here cover the period of 1979–2022. Our analysis focuses on midsummer, because the CEC's extreme heat in 2022 was the worst during July and August. Anomalies are calculated relative to the climatology of 1979–2022.

The Eurasian teleconnection SRP index is defined as the principle component of the leading mode (PC1) of 200 hPa meridional wind over the domain of 0°–150°E, 20°–60°N (Yasui and Watanabe 2010). The wave activity flux derived from Takaya and Nakamura (2001) is used to analyze atmospheric teleconnection, and detailed definition can be referred to the supplementary text S1. Diabatic heating (Q1) is employed to estimate the intensity of atmospheric thermal forcing over NWSA, which is calculated from the sum of heating rates in the JRA-55, including convective heating rate, large-scale condensation heating rate, vertical diffusion heating rate, solar radiative heating rate, and long wave radiative heating rate. The NWSA-Q1 index is defined as the column integral Q1 averaged in NWSA. Furthermore, we diagnose the moist static energy (MSE) budget equation (Neelin and Held 1987), in which the wind-induced moist enthalpy advection ($- {{\mathbf{V}}^{\prime} } \cdot {\nabla _{\text{h}}}\overline {({C_{\text{p}}}T + {L_{\text{v}}}q)}$) has been extensively shown to be the critical mechanism for anomalous vertical motion in monsoon regions (e.g. Wu et al 2017, Hu et al 2021, Wang et al 2022b). Here, ${\mathbf{V^{\prime}}}$ is anomalous horizontal wind; T and q are the climatological air temperature and specific humidity, respectively; C_p and L_v are the specific heat at constant pressure and latent heat of vaporization, respectively; and $\overline {{C_{\text{p}}}T + {L_{\text{v}}}q}$ is the climatological moist enthalpy. More details about the MSE equation can be found in Wu et al (2017).

3.1. Observed characteristics of 2022 extreme heat in CEC

3.2. Different mechanisms for the extreme heat in July and August 2022

Figures 2(a) and (c) show the anomalies of upper-level geopotential height and winds in July and August 2022, respectively. A common feature in the two month is: anomalous deep anticyclone appeared in the mid-upper troposphere over northern CEC, although the intensity differed (much stronger in August). In fact, the YR basin is located just below the southern flank of the anomalous anticyclone, where easterly anomalies prevailed, and significant atmospheric descent was generated (figures 2(b) and (d)), corresponding with the occurrence of the extreme heat. Thus, we hypothesize that the upper-level anomalous anticyclone over northern CEC was critical to the extreme heat in 2022. However, why did strong anomalous sinking motions occur at the southern flank of the anomalous anticyclone?

Zoom In Zoom Out Reset image size

Further investigation indicates that the easterly anomalies at the southern flank of the anomalous anticyclone were responsible for the significant descent over the YR basin. Climatologically, both air temperature (also potential temperature) and specific humidity peaked over the TP in the mid-upper levels (figure S2), supporting the plateau as a warm and moist center in summer (e.g. Ye and Wu 1998, Wu et al 2007, Xu et al 2008). Thus, according to the vertical distribution of potential temperature (green contours in figures 2(b) and (d)), the isentropic lines/surfaces sloped downward from east to west over the YR basin, especially in the mid-upper troposphere. Naturally, the air transported by the anomalous easterlies at the anticyclone's southern flank adiabatically moved along the isentropic surfaces, inducing descent to some extent (This process was recently emphasized by He et al (2022)). On the other hand, due to the warm and moist center over the TP, the anomalous easterlies brought cold and dry (low moist enthalpy) air westward into the YR basin (figure S2), characterized by the significantly negative moist enthalpy advection ($- {\mathbf{V}}^{\prime} \cdot {\nabla _{\text{h}}}\overline {({C_{\text{p}}}T + {L_{\text{v}}}q)}$ < 0, see the blue dashed contours in figures 2(b) and (d)). The negative moist enthalpy advection further intensified anomalous sinking motions under the constraint of the MSE budget (Neelin and Held 1987), and thus suppressed convection over the YR basin. As is well known, deep convection associated with the East Asian summer monsoon often occurs in the YR basin in July–August. With this background, the suppressed convection could induce obvious diabatic cooling (Q1 < 0), which further built up anomalous sinking motions and gradually formed a positive feedback between them.

Thus, we conclude that the easterly anomalies at the southern flank of the upper-level anomalous anticyclone over northern CEC were critical for the occurrence of heatwaves in both July and August. The easterly anomalies led to sinking motions through air isentropic sliding and a more important mechanism called 'wind-induced moist enthalpy advection'. The latter actually has been shown in one of our recent works (Wang et al 2022b), which was used to explain the weakened ascent and drying tendency over the TP's southern slope. In fact, significant drought was also observed over the southern TP in summer 2022 (figures 1(e) and (g)), a similar critical factor is the westward-extended easterlies associated with the above-mentioned anomalous anticyclone over northern CEC and the obvious invasion of negative moist enthalpy advection (figures 2(b) and (d)). It should also be pointed out that the anomalous anticyclone over northern CEC in August was much stronger than that in July (figures 2(a) and (c)), corresponding with more significant abnormal easterlies and sinking motions, and more serious heatwaves in the YR basin. The question is what caused the difference in anomalous anticyclone in the mid-upper troposphere of northern CEC between July and August in 2022?

As presented in figures 2(a) and (c), the patterns of wave activity flux indicate that the anomalous anticyclone over northern CEC was linked to different upstream wave trains. According to the path of wave activity flux transport in July, the anomalous anticyclone related to CEC's heat event seemed to originate from another anticyclone in the western TP, with an anomalous cyclone north of the TP linking the two (figure 2(a)). Such wave propagation was analogous to the classical pathway for connecting Indian and East Asian summer rainfalls (e.g. Lau et al 2000, Wu 2002, 2017), in which the anomalous diabatic heating associated with the Indian monsoon rainfall was regarded as a critical driving force (Rodwell and Hoskins 1996, Wei et al 2014). As described in section 3.1, a record-breaking downpour occurred in NWSA in July 2022 (figure 1(e)), which generated extremely anomalous diabatic heating (figure 1(f)) and potentially stimulated the change in atmospheric circulation over mid-latitude Asia. Furthermore, we calculate the regressed pattern of upper-level atmospheric circulation against the NWSA-Q1 index in July of 1979–2022 (figure 3(a)). Clearly, the mid-latitude Asian teleconnection originating from the western TP in July 2022 (figure 2(a)) was similar to the regressed pattern in figure 3(a). Both the wave activity fluxes in figures 2(a) and 3(a) diverged from the centers of high-pressure anomalies west of the TP, suggesting the effect of anomalous NWSA heating. Indeed, the role of anomalous diabatic heating over South Asia in the development of wave pattern has been demonstrated by many numerical simulations (e.g. Wu et al 2003, Wei et al 2015). We also successfully reproduced similar teleconnection in July 2022 forced by the observed NWSA anomalous heating in a linear baroclinic model (LBM; Watanabe and Kimoto 2000) (figure not shown). Moreover, there existed an observed significant interannual correlation between the time series of daily maximum SAT in the YR basin and NWSA diabatic heating in July during 1979–2022 (R= 0.50, P< 0.01). Therefore, the upper-level anomalous anticyclone over northern CEC related to the hot YR basin in July was largely caused by the strong diabatic heating (extreme rainfall) over NWSA. Under this mechanism, plus the extreme heat in the YR basin, significantly more and even extreme precipitation also occurred in northern China (figure 1(e)), consistent with the in-phase variation between South Asian and northern China summer rainfall noted in previous studies (e.g. Wu 2002, 2017, Ha et al 2018).

Zoom In Zoom Out Reset image size

Differently, the wave activity flux associated with strong anticyclonic anomaly over northern CEC in August emanated from Europe (figure 2(c)), as exhibited by a SRP-like pattern. Such pattern is a well-known teleconnection over the Eurasian continent in summer, which is also called the circumglobal teleconnection in some studies (e.g. Ding and Wang 2005). In August 2022, there actually existed an extremely negative SRP event (figure 3(d)). The spatial characteristics shown in the regressed pattern of upper-level atmospheric circulation against the SRP index in August of 1979–2022 (figure 3(c)) were consistent with the anomalies in August 2022 (figure 2(c)). Nevertheless, it should be noted that the weaker cyclonic anomaly over the western TP and the stronger anticyclonic anomaly over northern CEC in 2022 were somewhat different from those in the SRP-regressed pattern. We speculate that, besides the extreme SRP, other factors such as NWSA diabatic heating might also contribute, since the heavy rainfall in NWSA continued into August (figure 1(h)). To support our hypothesis, we decompose the components related to the SRP and NWSA heating in the 2022 anomalies of 200 hPa geopotential height and winds (figure 4). Here, we define the SRP-related component as the product of the SRP index in 2022 and the linear regression coefficients at each grid point onto the SRP index during 1979–2022. We also define the NWSA-Q1-related component similarly. Figure 4 clearly shows that the sum of the two components is much close to the raw anomalies presented in figure 2(c), indicating that both the extreme SRP event and the NWSA heating contributed to the formation of the Eurasian teleconnection in August 2022. The NWSA heating triggered anticyclonic anomalies over both western TP and northern CEC (figure 4(a)), but the negative SRP brought an anomalous cyclone and a more significant anticyclone respectively over the two regions (figure 4(b)). Compared with the SRP-regressed pattern (figure 3(c)), a relatively weaker cyclone and a much stronger anticyclone appeared zonally over the two sides of the TP as observed in August 2022 (figures 2(c) and 4(c)).

Zoom In Zoom Out Reset image size

Based on the above analysis, the extreme SRP dominantly contributed to the formation of anomalous anticyclone over northern CEC in August, and the diabatic heating due to persistent downpour in NWSA further amplified the anomalous anticyclone. These two factors conspired to generate the extremely strong anomalous anticyclone in August, which induced more significant anomalous easterlies, more significant sinking motions, and more serious heatwaves in the YR basin. Such superimposed effect of the SRP and NWSA heating in observations is supported by the numerical experiments with the LBM (figure S3). From the simulated results, the upper-level anomalous anticyclone over northern CEC, triggered by the atmospheric perturbation over Europe (SRP), was obviously intensified after the superposition of the NWSA thermal forcing. It should be pointed out that concurrent extreme heat also occurred in Europe in August (figure 1(c)), associated with the upstream anticyclonic node in the SRP (figure 2(c)). Statistically, a significant interannual correlation between the time series of daily maximum SATs in the YR basin and northern Europe existed in August during 1979–2022 (R = 0.48, P < 0.01).

We propose a schematic diagram to conclude the salient results obtained from this study (figure 5). The unprecedented heatwaves in the YR basin during July and August in 2022 were regulated by the anomalous anticyclonic circulation in the mid-upper troposphere over northern CEC, although the intensity differed in each month. The anomalous easterlies at the southern flank of the anomalous anticyclone caused obvious sinking motions over the YR basin through air isentropic sliding and a mechanism of 'wind-induced moist enthalpy advection'. In comparison, the anomalous descent and resulted heatwaves in August were more significant due to the stronger upper-level anomalous anticyclone and associated easterlies. Importantly, different physical processes were responsible for the formation of above-mentioned anomalous anticyclones in the two months. In July, the relatively weak anticyclone was dominated by the diabatic heating over NWSA, corresponding to the record-breaking rainfall in and around Pakistan. Whereas in August, a strong anticyclonic condition for the CEC heatwaves mainly originated from a Eurasian teleconnection, exhibited as an extreme SRP, superposing the effect of NWSA heating due to the persistent downpour. Note that another upstream anticyclonic node in the SRP teleconnection also caused heatwaves over Europe in August. In summary, we highlight different mechanisms for the extremely hot CEC in July and August 2022 from a Eurasian large-scale circulation perspective in this study. We also point out that the CEC extreme heat was actually associated with other concurrent extremes over the Eurasian continent through mid-latitude teleconnections.

Zoom In Zoom Out Reset image size

The SRP index in July was relatively low, which exerted a limited effect on the formation of anomalous anticyclone over northern CEC. However, an extreme SRP developed in August, emanating from the rapidly intensified anticyclonic node over Europe. Such a great anticyclonic node over Europe induced the local extreme heat (figures 1(c) and S1(b)); and its underlying cause is also of substantial interest. In fact, a relatively weaker and more poleward located anticyclonic anomaly and surface warming already appeared in July (figures 1(a) and 2(a)). Thus, we speculate that the local positive feedback between the warming and barotropic anticyclonic anomaly might contribute to the intensified anticyclone over Europe to some extent. Moreover, the North Atlantic Oscillation in August 2022 was observed as the third strongest case since 1979 (www.cpc.ncep.noaa.gov/products/precip/CWlink/pna/nao.shtml), also favoring the anticyclonic condition over Europe (Folland et al 2009, Hong et al 2022). However, further investigation for comprehensively explaining the European rapidly amplified anticyclonic anomaly is needed.

As indicated in the introduction, there also exists close association between the WPSH and heatwaves over East Asia. Figures 2(a) and (c) show that the WPSH in 2022 was stronger and located more westward than in the climatology. With the enhancement of mid-upper-level anomalous anticyclone over northern CEC from July to August (figure 2), the WPSH also rapidly intensified and extended farther to the west, and then occupied the entire YR basin by August. In fact, previous studies clarified that the eastward propagating wave train over the Eurasian continent (when the anticyclonic node reaches East Asia) can lead to strengthening and westward extending WPSH (e.g. Enomoto et al 2009, Liu et al 2019). Moreover, the position and intensity of the WPSH may also be directly influenced by NWSA heating (e.g. Wei et al 2014, Wu 2017), but specific process involved needs to be confirmed by further numerical study. On the other hand, both tropical convection and sea-surface temperature anomalies may also play roles in WPSH variation (e.g. Huang and Li 1992, Wang et al 2000, Chen et al 2019), and the possible combined role of the tropical forcing and mid-latitude teleconnection on heatwaves is another interesting topic for investigation.

Last but not least, previous studies recognized that local land-atmosphere coupling plays an important role in influencing surface climate (e.g. Koster et al 2004). Preceding low soil moisture always amplifies heatwaves through surface energy balance (Zhang and Wu 2011, Mueller and Seneviratne 2012). Although there was no obvious soil moisture deficit in previous months (figure not shown), dry soil induced by the concurrent drought in the YR basin during July–August 2022, especially in August, may have amplified the heatwaves. The specific feedback from local concurrent dry soil should be revealed through the simulations with and without local interactive soil moisture.

Authors appreciate the constructive comments from the anonymous reviewers. Ziqian Wang wishes to thank Dr Boqi Liu in China Meteorological Administration and Dr Chao He in Jinan University for useful discussions. This work was supported jointly by the Guangdong Major Project of Basic and Applied Basic Research (Grant 2020B0301030004), the National Natural Science Foundation of China (Grant 41975080), the Innovation Group of Southern Marine Science and Engineering Guangdong Laboratory (Zhuhai) (Grant 311021001), and the Guangdong Province Key Laboratory for Climate Change and Natural Disaster Studies (Grant 2020B1212060025). The authors declare no competing financial interests.

The data that support the findings of this study are available upon reasonable request from the authors.