1. Introduction
Heat waves (HWs) are meteorological disasters that pose significant threats to human health and socioeconomic systems due to their extremely high temperatures. In recent decades, numerous record-breaking HWs have been reported globally, including in western Europe (2003) (Robine et al. 2008), Russia (2010) (Barriopedro et al. 2011), China (2013) (Zhang et al. 2014), India (2015) (Ghatak et al. 2017), and North America (2021) (Thompson et al. 2022). These events, particularly in densely populated regions, have resulted in billions of dollars in economic losses and thousands of severe illnesses and deaths.
China, as a rapidly developing country with a vast population, is particularly vulnerable to the impacts of HWs. Studies have shown a notable increase in the frequency of HWs in China in recent decades (Ding et al. 2010; Sun et al. 2014), with projections indicating a further increase in the future due to global warming (Sun et al. 2017; Guo et al. 2017; Dong et al. 2023). This has sparked considerable interest in the scientific community to investigate the spatiotemporal variations and underlying causes of HWs in China (e.g., Miao et al. 2016; Chen et al. 2016; Luo and Lau 2017; Chen and Li 2017; Deng et al. 2019; Wu et al. 2021; Zheng et al. 2022; Liang et al. 2022).
The occurrence of HWs typically results from the combined action of multiple physical processes. Climate change is considered the primary driver of the increased frequency of HWs in recent decades (Horton et al. 2016). Human activities, for instance, urbanization leads to increased absorption of solar energy in urban areas, contributing to the rise in HWs (Liao et al. 2018). Additionally, land–atmosphere process through the interaction between soil moisture and turbulent heat fluxes is strongly associated with occurrence of high temperatures (Fischer et al. 2007; Zheng et al. 2022). Various atmospheric and atmosphere–ocean coupled modes, such as the Atlantic multidecadal oscillation (Zhang et al. 2020; Wei et al. 2023), El Niño–Southern Oscillation (ENSO; Luo and Lau 2019), monsoon anomalies (Deng et al. 2020), and intraseasonal oscillations (Hsu et al. 2017), also influence HWs. Atmospheric circulation changes can significantly impact the diabatic and adiabatic processes, affecting local and remote air temperatures and facilitating the occurrence of HWs (e.g., Lee and Lee 2016; Wang et al. 2017; Li et al. 2022; Faranda et al. 2023).
Accurate short–medium-range forecasting of HWs is crucial for implementing effective mitigation strategies. In China, HWs are significantly modulated by various atmospheric subseasonal modes. Among these, the 10–30-day quasi-biweekly oscillation (QBWO) originating from the tropical Pacific plays a prominent role (Chen et al. 2016; Hsu et al. 2017; Gao et al. 2018; Chen et al. 2018, 2020; Zheng et al. 2022; Lu et al. 2022; Li and Chen 2023; Qin et al. 2023; Zheng et al. 2024). During summer, the QBWO is initiated in the equatorial Pacific and moves northwestward toward East Asia as a large-scale convection–circulation complex (Li et al. 2020). Numerous studies have highlighted the impact of QBWO on HWs in China, particularly in the densely populated regions of the middle reaches of the Yangtze River (MYR) and South China (SC). QBWO is found to be critical for the occurrence of HWs in the SC, accounting for approximately half of the original temperature anomaly during its onset (Chen et al. 2016). Summer HWs are typically triggered by anomalous lower-level anticyclones and positive pressure anomalies associated with the westward extension of the western North Pacific subtropical high (WNPSH; Lu et al. 2022; Wu et al. 2023), whose quasi-biweekly variability is closely linked to the northwestward-propagating QBWO from the western North Pacific (Yang and Li 2020). The anticyclonic anomaly and subsidence induce excess solar radiation and significant diabatic heating, favoring the formation of HWs (Zheng et al. 2022). The anomalous warm advection driven by the westward extension of the WNPSH also contributes to the development of HWs, especially in the MYR (Hsu et al. 2017; Lu et al. 2022).
Previous studies have primarily focused on the general relationship between QBWO and HWs in China. However, the ENSO event, a prominent interannual mode in the tropical Pacific Ocean, significantly influences summer air temperature and HWs in China (Luo and Lau 2019). The anomalous anticyclone persisting in the western North Pacific following the peak phase of El Niño strongly modulates summer air temperatures in China, and some discrepancies have been observed regarding different types of El Niño event (Yuan and Yang 2012). Additionally, an interdecadal shift in the relationship between the summer temperature anomaly and ENSO has been documented (Zhu et al. 2007; Wu et al. 2010).
Meanwhile, ENSO exerts a significant impact on QBWO in the tropical northwestern Pacific. The intensity and northward propagation of QBWO are enhanced in the northwestern Pacific during the developing summer of El Niño compared to the decaying summer of El Niño (Lin and Li 2008; Wu and Cao 2017; Xu and Fan 2023). The year-to-year changes in atmospheric background fields, such as vertical shear of zonal wind and low-level moisture, are primarily responsible for the asymmetry of summertime QBWO regarding El Niño evolution. Model experiments suggest that this asymmetry is mainly due to the asymmetric response of these atmospheric background fields to opposite tropical central–eastern Pacific sea surface temperature (SST) anomalies (Wang et al. 2022).
Given that ENSO significantly affects both summer air temperature in China and the QBWO from the tropical Pacific, their complex interactions may result in nonlinear effects on the relationship between QBWO and HWs in China. Now, will the statistical relationship between QBWO and HWs in China exhibit year-to-year variations influenced by ENSO events, and to what extent will the relationship be affected? This study aims to provide a quantitative statistical analysis of the relationship between QBWO and HWs in southeastern China and its year-to-year changes in relation to ENSO event, which potentially support operational HWs forecasting based on monitoring and prediction of QBWO.
The rest of the paper is organized as follows. The data and methods used in the study are introduced in section 2. Section 3 documents the composite QBWO–HWs relationship across different ENSO episodes. In section 4, possible mechanism is proposed to understand the change of QBWO–HWs relationship in an ENSO cycle. A summary is presented in the final section.
2. Data and method
The main dataset is observed daily maximum temperature (Tmax) at 2419 meteorological stations in China acquired from the National Meteorological Information Center of the China Meteorological Administration (CMA) (Cao et al. 2016). In addition, daily and monthly reanalysis data on a 0.25° × 0.25° spatial grid from the fifth major global reanalysis produced by the European Centre for Medium-Range Weather Forecasts (ERA5; Hersbach et al. 2020) are employed. The analyzed variables include the three-dimensional wind velocity (u, υ, and w), specific humidity q, air temperature T, and SST. To identify convective signals associated with the QBWO, daily interpolated outgoing longwave radiation (OLR) with a spatial resolution of 2.5° × 2.5° from National Oceanic and Atmospheric Administration (NOAA) (Liebmann and Smith 1996) is selected. The temporal coverage for the aforementioned variables spans from 1980 to 2019. Then, a 10–30-day bandpass filter is applied to the raw daily data to isolate QBWO-related anomalies, and these anomalies are composited in reference to a near-real-time monitoring index for the QBWO in northwestern Pacific proposed by Qian et al. (2019). The index is constructed based on the extended empirical orthogonal function (EEOF) of OLR anomalies. The first two EEOF modes display a quadrature phase relationship, representing a northwestward-propagating wave. Based on the principal components (PCs) of these two modes, the amplitude and phase of a QWBO event are expressed as [PC1(t)2 + PC2(t)2]1/2 and tan−1[PC2(t)/PC1(t)], respectively. The composite in this study is based on active QBWO days, in which the amplitude exceeds 1. It has been demonstrated that the index more faithfully captures the state of the QBWO and its connections with extreme events.
HWs represent extremely prolonged high temperatures in specific regions. However, there is no universally accepted definition of HWs. For better prospect in operational forecast practice, the current analysis adheres to the CMA’s operational criteria, which defines HWs as a period when Tmax is equal to or exceeds 35°C for 3 consecutive days or more. Here, the HW day frequency (HWDF) is used as a property of HWs. HWDF is expressed as the percentage of days within HW events over the entire statistical time series. For instance, the annual summer [June–August (JJA)] average number of HW days divided by the total number of JJA days (92 days) yields the JJA mean HWDF (Fig. 1). Moreover, considering the uneven duration of each QBWO phase, directly comparing the number of HW days within each phase would be inappropriate. By introducing HWDF, which normalizes the number of HW days by total number of active days within each QBWO phase, we eliminate the impact of uneven phase duration, allowing for a more accurate representation of the differential influence of different QBWO phases on HWs.
Climatological JJA mean HWDF in China for 1980–2019 based on meteorological station (dots) data. The value is calculated as the percentage of HW days divided by the total days in JJA.
Citation: Journal of Climate 38, 7; 10.1175/JCLI-D-24-0191.1
The definition of ENSO event is referred to the oceanic Niño index (ONI) derived from NOAA. An event is classified as El Niño when the ONI remains at or above +0.5°C for a minimum of five consecutive overlapping seasons. It is classified as La Niña when the ONI remains at or below −0.5°C for the same duration. Therefore, 13 El Niño and 14 La Niña events are identified during 1980–2019. Given that some events persist more than 1 year (e.g., 1986–88 El Niño; 1998–2001 La Niña), or some events are neighbors of the same type (e.g., 2010–11 and 2011–12 La Niña; 2016–17 and 2017–18 La Niña), those overlapping years are discarded to avoid the confusion when we classify the developing and decaying year of El Niño/La Niña event. Accordingly, the selected ENSO episodes are presented in Table 1.
Classification of different ENSO episodes during 1980–2019.
3. Composite QBWO–HWDF relationship in different ENSO episodes
Figure 1 shows the annual mean HWDF in China during the summer months (JJA) from 1980 to 2019. Regions with higher HWDF values, exceeding 20%, are primarily concentrated in two regions: southeastern China and arid desert in the northwest. Southeastern China, in particular, is densely populated, housing over 40% of the country’s population and contributing more than 50% of China’s gross domestic product. Therefore, further research on HWs in this vulnerable region is warranted.
Actually, the temperature variability in southeastern China is significantly modulated by the QBWO from the tropical northwestern Pacific. The QBWO, originating in the equatorial Pacific, is likely maintained by the interactions among anomalously equatorial moisture, Walker-like circulation, and mean trade winds (Li et al. 2021). After formation, the northeast–southwest titling band successively moves toward East Asia, periodically regulating the weather systems through alternation of large-scale dry and wet spells. As seen from Fig. 2, sandwich-type convective anomalies persist in the northwestern Pacific, with the northernmost part reaching approximately 30°N, i.e., the MYR. It is also evident that the low-level air temperature is strongly influenced by the evolution of QBWO. Accompanying the enhanced (suppressed) convective anomalies are low-level negative (positive) temperature anomalies, which are most pronounced in landmass and adjoining areas. Specifically, in phases 7–8 of QBWO, positive temperature anomalies are located in the SC, and they extend northward to the MYR in phases 1–2, coinciding with the northwestward propagation of suppressed convective anomalies. Negative temperature anomalies primarily occur in the southeastern China during phases 3–4 and 5–6, accompanying the enhanced convective anomalies. It is suggested that diabatic heating and temperature advection are of importance in contributing to the temperature anomalies associated with QBWO in Asian monsoon regions (Hsu et al. 2017).
Composite spatiotemporal evolution of the QBWO in the northwestern Pacific. Shadings are 10–30-day bandpass-filtered OLR anomalies (W m−2), contours are 925-hPa temperature anomalies (°C), and vectors are 925-hPa wind anomalies (m s−1). The green (blue) box denotes the MYR (SC) in the following analysis. The red line in bottom-right panel marks the moving route of QBWO analyzed in Fig. 8. The QBWO phases are indicated in the upper left of each panel.
Citation: Journal of Climate 38, 7; 10.1175/JCLI-D-24-0191.1
In association with the QBWO-induced temperature anomalies, the probability of HWs occurrence in southeastern China may vary dramatically across different QBWO phases. To explore the influence of QBWO on HWDF, we calculated HWDF for each QBWO phase through dividing the number of HW days by the total number of days in that phase. Note that only active QBWO days are taken into consideration. The resulting HWDF values, which may incorporate the influence of lower-frequency variability, are then adjusted by subtracting the mean HWDF across all QBWO phases. This adjustment underscores the fluctuations in HWDF with positive and negative anomalies, which we interpret as the degree of QBWO’s influence on HW variations, or the QBWO–HWDF relationship. As shown in Fig. 3, positive HWDF anomalies are evident in the SC (MYR) during phases 7–8 (phases 1–2), which are concurrent with the arrival of the QBWO dry spell depicted in Fig. 2. Since negative HWDF anomalies prevail during phases 3–4 and 5–6, the subsequent analysis will mainly focus on the positive HWDF anomalies in phases 7–8 and 1–2. To facilitate a quantitative evaluation, strong anomalous centers in (107°–118°E, 22°–27°N) and (104°–115°E, 28°–33°N) are elaborated and will henceforth represent the regions of the SC and the MYR, respectively. Table 2 illustrates the HWDF of each QBWO phase, all phases mean, and JJA mean in the MYR and the SC. Recognizing that direct comparisons of absolute HWDF fluctuation across QBWO phases may not be straightforward due to the different climatological baselines between the MYR and the SC, we introduce HWDF percentage change rate. These rates are calculated by dividing the HWDF fluctuation value in different QBWO phases by the climatological JJA mean HWDF. Since the same climatological JJA mean HWDF is used as the denominator for 40-yr mean and different ENSO episodes, HWDF percentage change rates can be meaningfully compared among them. From Table 2, it can be seen that HWDF significantly increases 38.7% (36.2%) in the SC (MYR) during phases 7–8 (phases 1–2), relative to the JJA mean value of 14.2% (13.0%) for 1980–2019.
HWDF anomalies in each QBWO phase, i.e., HWDF in each QBWO phase minus mean HWDF across all QBWO phases.
Citation: Journal of Climate 38, 7; 10.1175/JCLI-D-24-0191.1
HWDF in each QBWO phase, i.e., the percentage of HW days divided by total active days within each QBWO phase. The value with parentheses is the percentage change rate relative to the climatological JJA mean, indicating the percentage of the HWDF anomaly in each QBWO phase divided by the climatological JJA mean HWDF.
The aforementioned quantitative relationship between the QBWO and HWDF anomalies potentially supports the operational HWs forecasting based on monitoring and prediction of QBWO. Given the significant impacts of ENSO event on summer air temperature and the QBWO, it is worth investigating whether this statistical relationship varies with ENSO evolution. It has been demonstrated that there is an asymmetry in dynamic and thermodynamic processes between the developing and decaying summers regarding El Niño/La Niña events (Lin and Li 2008; Wu and Cao 2017; Tao et al. 2017). This kind of asymmetry significantly influences the propagation and strength of QBWO in the northwestern Pacific, which may further alter the quantitative QBWO–HWDF relationship in southeastern China. Therefore, the composite QBWO–HWDF relationship in an ENSO cycle is examined in Fig. 4. Interestingly, the composite results reveal that the quantitative relationship obtained from the climatological mean (Fig. 3 and Table 2) changes dramatically and exhibits a distinction between the MYR and the SC during an ENSO cycle. In the MYR, the significantly positive HWDF anomalies occurring in phases 1–2 are much stronger during El Niño developing (El Dev) and La Niña decaying (La Dec) summers than during the El Niño decaying (El Dec) and La Niña developing (La Dev) summers. In the SC, however, the significantly positive HWDF anomalies appearing in phases 7–8 are much stronger during El Niño decaying and La Niña developing summers compared to El Niño developing and La Niña decaying summers.
HWDF anomalies in each QBWO phase across different ENSO episodes. (a) El Dev summers, (b) El Dec summers, (c) La Dev summers, and (d) La Dec summers.
Citation: Journal of Climate 38, 7; 10.1175/JCLI-D-24-0191.1
The quantitative evaluation from Tables 3 and 4 provides further details. For the positive HWDF anomalies in the MYR during phases 1–2, the QBWO induces a larger value during El Niño developing summers (57.2%), which is more than twice that during El Niño decaying summers (25.5%). For the positive HWDF anomalies in the SC during phases 7–8, a larger value is produced by the QBWO during El Niño decaying summers (80.9%), which is more than 3 times that during El Niño developing summers (23.7%). The results regarding the HWDF difference between La Niña decaying and developing summers are similar to those between El Niño developing and decaying summers. This is largely because that all La Niña developing years match El Niño decaying years, and three out of eight La Niña decaying years correspond to El Niño developing years (Table 1).
The HWDF percentage change rate relative to climatological JJA mean in the MYR in different ENSO episodes.
4. Understanding the QBWO–HWDF relationship in an ENSO cycle
In the previous section, we described in detail the significant differences in the quantitative QBWO–HWDF relationship in the MYR and the SC during different episodes of an ENSO cycle. We also highlighted the distinct patterns between these two regions. In this section, we intend to explore the underlying mechanism responsible for these diverse QBWO–HWDF relationships.
From a physical perspective, the occurrence of QBWO-related HWs could be viewed as the sum of the background temperature and QBWO temperature perturbation exceeding a threshold (35°C in this study). A higher background temperature and a larger QBWO temperature perturbation will both increase the occurrence probability of HWs, or in other words, increase the HWDF in a specific QBWO phase. Therefore, the distinct QBWO–HWDF relationship can be understood in terms of the year-to-year difference in the background temperature and QBWO temperature perturbation. Since the differences in the QBWO–HWDF relationship between El Niño developing and decaying summers are similar to those between La Niña decaying and developing summers, the following discussion will primarily focus on the differences between El Niño developing and decaying summers for simplicity.
Figure 5 presents boxplots of Tmax in the MYR and the SC during different ENSO episodes with the outliers removed. In the MYR, Tmax during El Niño developing summers is higher than that during El Niño decaying summers in all quartile levels except for 100th percentile, with the former being approximately 0.3°C higher than the latter. Therefore, given the same QBWO perturbation, the higher background temperature is more conducive to the formation of HWs during El Niño developing summers compared to El Niño decaying summers. This is in accordance with the observation of a larger HWDF in phases 1–2 during El Niño developing summers. In the SC, however, we observe an opposite scenario. The Tmax during El Niño decaying summers is higher than that during El Niño developing summers in all quartile levels, with a difference of approximately 0.5°C. This is also consistent with the higher HWDF in phases 7–8 during El Niño decaying summers. The situations regarding the La Niña event exhibit a striking similarity to those of the El Niño event. Overall, for both the MYR and the SC, the difference in the background mean air temperature aligns well with the distinct patterns of the QBWO–HWDF relationship during an ENSO cycle.
Boxplots of Tmax in (a) the MYR and (b) the SC in different ENSO episodes. The quartiles values from 0% to 100% are shown aside. The unit is degrees Celsius (unit: °C).
Citation: Journal of Climate 38, 7; 10.1175/JCLI-D-24-0191.1
In addition to the variation in the background mean air temperature, the 10–30-day air temperature perturbation also plays a significant role in shaping the distinct QBWO–HWDF relationship. To address this issue, the 10–30-day Tmax anomalies in HWs peak phases (phases 1–2 for the MYR and phases 7–8 for the SC) based on 75th percentile of Tmax in different ENSO episodes are illustrated in Fig. 6. The selection of the 75th percentile as the base value is not grounded in any particular rationale but merely serves as a means to illustrate the variability in background mean temperature across distinct ENSO episodes. The adoption of an alternative choice, such as the 50th percentile, would not alter the discussions or conclusions presented in this section.
The 10–30-day Tmax perturbations (red bars; °C) and the 75th percentiles of Tmax (blue bars; °C) in (a) the MYR and (b) the SC in different ENSO episodes. In (a), Tmax perturbations are averaged in phases 1–2, and in (b), Tmax perturbations are averaged in phases 7–8. The values are indicated within each bar.
Citation: Journal of Climate 38, 7; 10.1175/JCLI-D-24-0191.1
In the MYR, the QBWO temperature perturbation is much stronger during El Niño developing summers, which induces a Tmax anomaly of approximately 0.75°C in phases 1–2. In contrast, the QBWO only induces a Tmax anomaly of approximately 0.14°C during El Niño decaying summers, which is 0.6°C lower than that during El Niño developing summers. This is consistent with the observed higher HWDF in the QBWO phases 1–2 during El Niño developing summers. Although both the background temperature and the QBWO temperature perturbation contribute positively to a higher HWDF in phases 1–2 during El Niño developing summers, the latter appears to play the major role because it induces a year-to-year temperature difference of around 0.6°C, while the former only induces a difference of around 0.2°–0.3°C (Fig. 5a).
Compared to the MYR, the situation in the SC is somewhat more complex. The QBWO temperature perturbation is much stronger during El Niño decaying summers, inducing a Tmax anomaly of approximately 0.5°C in phases 7–8. In contrast, the QBWO-induced Tmax perturbation is only 0.06°C during El Niño developing summers, which is approximately 0.4°C lower than that during El Niño decaying summers. This is also consistent with the observed higher HWDF in QBWO phases 7–8 during El Niño decaying summers. Thus, a similar conclusion can be drawn for the SC: both the background temperature and the QBWO temperature perturbation contribute positively to a higher HWDF in phases 7–8 during El Niño decaying summers compared to El Niño developing summers. A minor distinction lies in the fact that the contributions from both factors are roughly equivalent, with a year-to-year temperature difference of approximately 0.4°–0.5°C.
The aforementioned analysis indicates that the distinction in the quantitative QBWO-HWDF relationship with respect to ENSO evolution can be reasonably attributed to the year-to-year changes in both the QBWO temperature perturbations and summer mean temperature. To more clearly describe the interannual variation of QBWO, we simply characterize its regional intensity according to the amplitude of QBWO-related temperature perturbations in both regions. For the MYR, the year-to-year change in QBWO intensity plays the dominant role, while for the SC, both factors are approximately equally important. The question now arises: what causes the year-to-year change in QBWO temperature perturbations and summer mean temperature in the MYR and the SC?
To investigate the underlying physical processes responsible for the year-to-year variability in QBWO temperature perturbations, we conducted a temperature budget analysis at the near-surface level (925 hPa) utilizing the daily ERA5 reanalysis data. Our calculation reveals a strong temporal correlation coefficient of approximately 0.82 (0.74) between daily temperatures over the MYR (SC) derived from ERA5 and surface meteorological stations during the summers spanning 1980–2019, highlighting the remarkable consistency between the two datasets.
Prior to the application of Eq. (1), we preprocessed each parameter by a 10-day low-passed filter to eliminate the synoptic-scale signal (with periods less than 10 days) and calculated its mean value across the active days within each QBWO phase. Additionally, we adopt the assumption that QBWO perturbations arise consistently against a lower-frequency background field (with periods exceeding 30 days), suggesting that the phase-mean value closely approximates the superposition of QBWO perturbations and an invariant low-frequency background. Thus, the phase transitions in the composite temperature are primarily attributed to the temporal evolution of QBWO temperature perturbations. Since the power spectrum analysis of the 10–30-day filtered OLR over southeastern China (104°–118°E, 22°–33°N) indicates that the dominant period of QBWO is 23 days, we hypothesize that the QBWO phase transition time is 5.75 days (one quarter cycle).
In light of the above analysis, the time tendency in preprocessed temperature across different QBWO phases in the MYR and the SC is given in Figs. 7a and 7b. In the MYR, prior to the peak HWDF in phases 1–2 (i.e., the phase transition of 7–8 → 1–2), a distinct positive temperature tendency appears. This positive value is much larger during El Niño developing summers compared to decaying summers, aligning with the year-to-year changes in the QBWO–HWDF relationship. A similar pattern is observed in the SC, where a pronounced positive temperature tendency precedes the maximum HWDF in phases 7–8 (i.e., the phase transition of 5–6 → 7–8). Obviously, the positive tendency is more pronounced during El Niño decaying summers. In one word, in the MYR (SC), the much higher temperature tendency prior to the peak HWDF phase during El Niño developing years (El Niño decaying years) significantly contributes to the year-to-year changes in the QBWO–HWDF relationship. Now, which processes are responsible for the above difference in temperature time tendency across different ENSO episodes?
Temperature time tendency and each budget term at 925-hPa during QBWO phase transition in (left) the MYR and (right) the SC across different ENSO episodes. See the legend in each panel for details. The unit is degrees Celsius per day.
Citation: Journal of Climate 38, 7; 10.1175/JCLI-D-24-0191.1
For the prepeak temperature increase in the MYR, its notable year-to-year contrast is accompanied by a coherent interannual variability in three budget terms (Figs. 7c,e,g). Specifically, during El Niño developing (decaying) years, significantly higher (lower) values are observed in the MYR for the phase transition of 7–8 → 1–2. Among three budget terms, vertical motion–driven adiabatic process and diabatic heating are major terms, and horizontal advection is secondary. In the MYR, the positive OLR anomalies associated with QBWO dry phases are much stronger in El Niño developing years compared to decaying years (Figs. 8a,b). Therefore, adiabatic heating related to a stronger subsidence anomaly could contribute positively to a higher temperature increase. Meanwhile, a stronger subsidence anomaly would enhance the local diabatic heating due to upward surface longwave radiation and sensible and latent heat fluxes (Hsu et al. 2017). However, for the prepeak temperature increase in the SC, i.e., phase transition of 5–6 → 7–8, the difference of three budget terms between El Niño’s developing and decaying years appears distinctly with the MYR. There is no obvious year-to-year discrepancy in the adiabatic heating related to local subsidence and diabatic heating due to local surface heat fluxes. This is consistent with the fact that the positive OLR anomalies associated with QBWO dry phases are similar between El Niño’s developing and decaying years in the SC. It shows that the horizontal advection term is primarily responsible for the year-to-year contrast in the prepeak temperature increase in the SC (Fig. 7d). Generally, the adiabatic heating driven by vertical motion and diabatic heating due to surface heat fluxes are mainly sensitive to local processes associated with convective anomalies. In contrast, the horizontal advection term, which is driven by the interaction between temperature gradient and horizontal winds, seems less dependent on local processes. For instance, wind variations in the SC may be forced by external factors beyond the SC region. This is a possible reason why, in the SC, the stronger positive temperature anomalies during El Niño decaying years do not correspond to locally stronger convective anomalies. Further analysis regarding the horizontal advection term indicates that the interaction between lower-frequency background winds and QBWO temperature anomalies primarily accounts for the year-to-year difference in the horizontal advection term between El Niño’s developing and decaying years in the SC (figure not shown).
Composite phase-meridional evolutions of 10–30-day bandpass-filtered OLR anomalies (shadings; W m−2) and 925-hPa temperature anomalies (contours; °C) along the QBWO moving route (averaged between 10°W and 10°E of the selected line in Fig. 2). The labels of y axis indicate the red dots marked along the selected line. The green (blue) box denotes the latitude of the MYR (SC).
Citation: Journal of Climate 38, 7; 10.1175/JCLI-D-24-0191.1
In view of the QBWO temperature perturbations, its fluctuation changes dramatically across different ENSO episodes. Now, what causes the change in QBWO intensity in the MYR and the SC during an ENSO cycle? As reported in numerous previous studies, boundary layer moisture and the vertical shear of zonal wind are considered to be the critical dynamic and thermodynamic environmental factors responsible for the change in QBWO intensity (e.g., Wang and Xie 1997; Lin and Li 2008; Wu and Cao 2017; Xu and Fan 2023). A moist boundary layer naturally favors the development of deep convection, and an easterly shear environment could enhance the lower-tropospheric Rossby wave response, leading to greater perturbation growth through convection–circulation–moisture feedback. It has been shown that both factors are sensitive to an ENSO cycle (Wang et al. 2022). To identify their relative roles in causing the QBWO intensity change in southeastern China, the anomalous 925-hPa specific humidity and the vertical shear of zonal wind relative to JJA climatological mean in the MYR and the SC are compared in Figs. 9 and 10.
Anomalies of 925-hPa specific humidity (contours; g kg−1) and air temperature (shadings; °C) relative to climatological JJA mean.
Citation: Journal of Climate 38, 7; 10.1175/JCLI-D-24-0191.1
Anomalies of vertical shear of zonal wind (U200–U850; contours; m s−1), 850-hPa winds (vectors; m s−1), and SST (shadings; °C) relative to climatological JJA mean.
Citation: Journal of Climate 38, 7; 10.1175/JCLI-D-24-0191.1
For the MYR, there are positive boundary layer moisture anomalies during both El Niño summers (Figs. 9a,b). However, their values seem much larger (smaller) during El Niño decaying (developing) summers, which is inconsistent with the observed weaker (stronger) QBWO at that time (Figs. 6–8). For the SC, there are weak humidity anomalies during both El Niño summers. In this regard, it suggests that the background boundary layer moisture is not the key factor for the observed change in QBWO intensity in both regions. With respect to the vertical shear of zonal wind, the MYR and the SC show opposite patterns during both El Niño summers. In the MYR, easterly shear is stronger during El Niño developing summers, which is consistent with the observational stronger QBWO there at that time. In the SC, easterly shear is stronger during El Niño decaying summers, which potentially favors the stronger QBWO there. Though the year-to-year contrast is obscure in terms of the convective anomalies, the temperature anomalies in relation to QBWO are indeed much stronger during El Niño decaying summers (Fig. 8). Therefore, it is likely that the interannual variation in the vertical shear of zonal wind is the dominant factor causing the change in QBWO intensity and the distinction between the MYR and the SC during an ENSO cycle.
The anomalies of the vertical shear of zonal wind during four ENSO episodes are further examined. In Fig. 10a, one can observe that strong positive values occupy the zonal belt around 20°–30°N in the Northern Hemisphere, which is the northern portion of an anomalous low-level anticyclonic circulation during El Niño developing summers. This pattern can be easily interpreted because in summer, negative SST anomalies have already developed in the northwestern Pacific associated with an El Niño event. It is evident that the SC is entirely embedded in the positive region, while the MYR generally lies to its northwest, with only a small portion exhibiting positive values. The aforementioned meridional contrast in the vertical shear of zonal wind may favor a much stronger (weaker) QBWO in the MYR (SC) during El Niño developing summers. Similar conclusion is easily derived for other ENSO episodes.
As previously documented, the interannual variation in the background mean temperature can significantly contribute to the year-to-year changes in the QBWO–HWDF relationship. Its positive role is particularly evident when contrasting the developing and decaying years of El Niño (La Niña) in the SC (MYR), as depicted in Fig. 6. As shown in Fig. 9, the 925-hPa air temperature anomalies exhibit distinct behaviors in the MYR and the SC during the four ENSO summers. Specifically, in the MYR (SC), pronounced contrasts in temperature anomalies are observed between La Niña (El Niño) developing and decaying years. To elucidate the governing mechanisms behind the summer temperature anomalies, a temperature budget analysis based on ERA5 monthly data is conducted, encompassing temperature variations from the preceding winter (December–February) through to summer (JJA). As illustrated in Fig. 11, in the MYR, the temperature differences between La Niña’s developing and decaying years are predominantly attributed to the diabatic heating process (Fig. 11g), whereas in the SC, the temperature differences between El Niño’s developing and decaying years are primarily influenced by horizontal temperature advection (Fig. 11d). The regional characteristics of summer mean temperature in relation to the ENSO event require further in-depth study, which is temporarily beyond the scope of the current work.
Time tendency of 925-hPa air temperature and each budget term from the preceding winter through to summer in (left) the MYR and (right) the SC during four ENSO episodes. See the legend in each panel for details. The unit is degrees Celsius per day.
Citation: Journal of Climate 38, 7; 10.1175/JCLI-D-24-0191.1
5. Summary
The summertime HWs have become increasingly frequent in China in recent decades and are projected to intensify further in the warming future. Southeastern China, with its dense population, is particularly vulnerable to HWs due to their severe societal and economic impacts. Therefore, accurate prediction of HWs based on a comprehensive understanding of the physical mechanism is an essential operational task. As one of the key factors for subseasonal prediction, the 10–30-day QBWO significantly modulates the occurrence of HWs in southeastern China. However, the extent to which the QBWO influences HWs remains unclear until now. Furthermore, as ENSO significantly affects both the QBWO activity and China’s climate, it is still unknown how the statistical QBWO–HWs relationship changes in an ENSO cycle. Clarifying these issues has the potential to improve the subseasonal prediction of HWs in southeastern China.
Based on observations from meteorological stations and reanalysis data from ERA5, this study investigates the influences of QBWO on the southeastern China HWs during different ENSO episodes. The results indicate that HWs in the MYR and the SC are remarkably influenced by the QBWO from the tropical northwestern Pacific. Accompanying the stepwise northwestward evolution of the QBWO dry phase, HWDF significantly increases by 38.5% (36.3%) in the SC (MYR) during phases 7–8 (phases 1–2), relative to a summer mean value of 14.2% (13.0%). Further analysis reveals that the above quantitative QBWO–HWDF relationship changes dramatically in an ENSO cycle, and a notable distinction appears between the MYR and the SC. In the MYR (SC), the QBWO induces a larger HWDF perturbation during El Niño developing (El Niño decaying) summers which is more than twice (3 times) that during El Niño decaying (El Niño developing) summers in phases 1–2 (7–8). This distinct QBWO–HWDF relationship regarding El Niño event can be understood in terms of the year-to-year difference in background mean temperature and QBWO temperature perturbation. It is shown that the above change is primarily due to the variation in QBWO intensity in the MYR, while it is attributed to the changes in both QBWO intensity and summer mean temperature in the SC. The QBWO–HWDF relationship between La Niña’s decaying and developing summers is similar to that between El Niño’s developing and decaying summers.
The temperature budget analysis on QBWO perturbation is conducted to understand the difference of prepeak temperature increase between El Niño’s developing and decaying summers. In the MYR, its notable year-to-year contrast is mainly contributed by vertical motion–driven adiabatic process and diabatic heating. In comparison, the horizontal advection term is primarily responsible for the year-to-year contrast in the SC. For the significant variation in background mean temperature, the notable difference between La Niña’s developing and decaying summers in the MYR is predominantly attributed to the diabatic heating process, whereas the striking differences between El Niño’s developing and decaying summers in the SC are primarily influenced by horizontal temperature advection. Further examination indicates that the QBWO change in an ENSO cycle is likely related to the variation in background vertical wind shear, which behaves differently between the MYR and the SC during four typical summers. It is supposed that the anomalous atmospheric circulation in response to the different summer SST anomalies may play the underlying role (as indicated by Fig. 12).
Schematic maps showing QBWO–HWDF relationship during El Dev and El Dec summers.
Citation: Journal of Climate 38, 7; 10.1175/JCLI-D-24-0191.1
The observed QBWO–HWDF relationship, particularly the varying quantitative statistics in the MYR and the SC across different episodes of an ENSO cycle has established a benchmark for future operational HWs prediction in southeast China. However, scientific explanation of the underlying mechanisms requires further model validation. Additionally, as a parallel subseasonal mode to the QBWO, the 30–60-day intraseasonal oscillation (ISO) will be investigated in our next study to determine its quantitative statistical relationship with HWDF and the extent to which this quantitative relationship varies with the ENSO cycle. Furthermore, different flavors of ENSO events have been clarified in recent decade (e.g., Kao and Yu 2009; Kug et al. 2009), and they influence the climate in China in distinct ways (Yuan and Yang 2012). Therefore, it is necessary to investigate whether these different flavors exert distinct impacts on the QBWO–HWDF relationship. There are some other properties describing HWs, such as duration and magnitude, and the analysis of their statistical relationship with QBWO is also needed in the future.
Acknowledgments.
This work was jointly supported by the State Key Program of National Natural Science Foundation of China (42230408), the Basic Scientific Fund for National Public Research Institutes of China (2024Q04), and the National Natural Science Foundation of China (42475066). The authors declare that they do not have any competing interests.
Data availability statement.
The data that support the findings of this study are derived from the following sources: CMA data (http://data.cma.cn/), ONI (https://ggweather.com/enso/oni.htm), ERA5 reanalysis (https://www.ecmwf.int/en/forecasts/dataset/ecmwf-reanalysis-v5), and OLR (ftp://ftp2.psl.noaa.gov/Datasets/interp_OLR).
REFERENCES
Barriopedro, D., E. M. Fischer, J. Luterbacher, R. M. Trigo, and R. García-Herrera, 2011: The hot summer of 2010: Redrawing the temperature record map of Europe. Science, 332, 220–224, https://doi.org/10.1126/science.1201224.
Cao, L., Y. Zhu, G. Tang, F. Yuan, and Z. Yan, 2016: Climatic warming in China according to a homogenized data set from 2419 stations. Int. J. Climatol., 36, 4384–4392, https://doi.org/10.1002/joc.4639.
Chen, R., Z. Wen, and R. Lu, 2016: Evolution of the circulation anomalies and the quasi-biweekly oscillations associated with extreme heat events in southern China. J. Climate, 29, 6909–6921, https://doi.org/10.1175/JCLI-D-16-0160.1.
Chen, R., Z. Wen, and R. Lu, 2018: Large-scale circulation anomalies and intraseasonal oscillations associated with long-lived extreme heat events in South China. J. Climate, 31, 213–232, https://doi.org/10.1175/JCLI-D-17-0232.1.
Chen, W., J.-X. Fu, and G. Li, 2020: Objective identification and multi-scale controlling factors of extreme heat-wave events in Southern China. Atmosphere, 11, 668, https://doi.org/10.3390/atmos11060668.
Chen, Y., and Y. Li, 2017: An inter-comparison of three heat wave types in China during 1961–2010: Observed basic features and linear trends. Sci. Rep., 7, 45619, https://doi.org/10.1038/srep45619.
Deng, K., S. Yang, M. Ting, P. Zhao, and Z. Wang, 2019: Dominant modes of China summer heat waves driven by global sea surface temperature and atmospheric internal variability. J. Climate, 32, 3761–3775, https://doi.org/10.1175/JCLI-D-18-0256.1.
Deng, K., S. Yang, D. Gu, A. Lin, and C. Li, 2020: Record-breaking heat wave in southern China and delayed onset of South China Sea summer monsoon driven by the Pacific subtropical high. Climate Dyn., 54, 3751–3764, https://doi.org/10.1007/s00382-020-05203-8.
Ding, T., W. Qian, and Z. Yan, 2010: Changes in hot days and heat waves in China during 1961–2007. Int. J. Climatol., 30, 1452–1462, https://doi.org/10.1002/joc.1989.
Dong, W., X. Jia, Q. Qian, and X. Li, 2023: Rapid acceleration of dangerous compound heatwaves and their impacts in a warmer China. Geophys. Res. Lett., 50, e2023GL104850, https://doi.org/10.1029/2023GL104850.
Faranda, D., G. Messori, A. Jezequel, M. Vrac, and P. Yiou, 2023: Atmospheric circulation compounds anthropogenic warming and impacts of climate extremes in Europe. Proc. Natl. Acad. Sci. USA, 120, e2214525120, https://doi.org/10.1073/pnas.2214525120.
Fischer, E. M., S. I. Seneviratne, P. L. Vidale, D. Lüthi, and C. Schär, 2007: Soil moisture–atmosphere interactions during the 2003 European summer heat wave. J. Climate, 20, 5081–5099, https://doi.org/10.1175/JCLI4288.1.
Gao, M., J. Yang, B. Wang, S. Zhou, D. Gong, and S.-J. Kim, 2018: How are heat waves over Yangtze River valley associated with atmospheric quasi-biweekly oscillation? Climate Dyn., 51, 4421–4437, https://doi.org/10.1007/s00382-017-3526-z.
Ghatak, D., B. Zaitchik, C. Hain, and M. Anderson, 2017: The role of local heating in the 2015 Indian heat wave. Sci. Rep., 7, 7707, https://doi.org/10.1038/s41598-017-07956-5.
Guo, X., J. Huang, Y. Luo, Z. Zhao, and Y. Xu, 2017: Projection of heat waves over China for eight different global warming targets using 12 CMIP5 models. Theor. Appl. Climatol., 128, 507–522, https://doi.org/10.1007/s00704-015-1718-1.
Hersbach, H., and Coauthors, 2020: The ERA5 global reanalysis. Quart. J. Roy. Meteor. Soc., 146, 1999–2049, https://doi.org/10.1002/qj.3803.
Horton, R. M., J. S. Mankin, C. Lesk, E. Coffel, and C. Raymond, 2016: A review of recent advances in research on extreme heat events. Curr. Climate Change Rep., 2, 242–259, https://doi.org/10.1007/s40641-016-0042-x.
Hsu, P.-C., J.-Y. Lee, K.-J. Ha, and C.-H. Tsou, 2017: Influences of boreal summer intraseasonal oscillation on heat waves in monsoon Asia. J. Climate, 30, 7191–7211, https://doi.org/10.1175/JCLI-D-16-0505.1.
Kao, H.-Y., and J.-Y. Yu, 2009: Contrasting eastern-Pacific and central-Pacific types of ENSO. J. Climate, 22, 615–632, https://doi.org/10.1175/2008JCLI2309.1.
Kug, J.-S., F.-F. Jin, and S.-I. An, 2009: Two types of El Niño events: Cold tongue El Niño and warm pool El Niño. J. Climate, 22, 1499–1515, https://doi.org/10.1175/2008JCLI2624.1.
Lee, W.-S., and M.-I. Lee, 2016: Interannual variability of heat waves in South Korea and their connection with large‐scale atmospheric circulation patterns. Int. J. Climatol., 36, 4815–4830, https://doi.org/10.1002/joc.4671.
Li, K., Y. Yang, L. Feng, W. Yu, and S. Liu, 2020: Structures and northward propagation of the quasi-biweekly oscillation in the western North Pacific. J. Climate, 33, 6873–6888, https://doi.org/10.1175/JCLI-D-19-0752.1.
Li, K., Q. He, Y. Yang, and W. Yu, 2021: Equatorial moisture dynamics of the quasi‐biweekly oscillation in the tropical northwestern Pacific during boreal summer. Geophys. Res. Lett., 48, e2020GL090929, https://doi.org/10.1029/2020GL090929.
Li, X., and R. Chen, 2023: Diversity in the quasi-biweekly oscillation sources of the daily maximum temperature over South China during summer. Int. J. Climatol., 43, 7099–7118, https://doi.org/10.1002/joc.8254.
Li, Y., G. Ren, Q. You, Q. Wang, Q. Niu, and L. Mu, 2022: The 2016 record-breaking marine heatwave in the Yellow Sea and associated atmospheric circulation anomalies. Atmos. Res., 268, 106011, https://doi.org/10.1016/j.atmosres.2021.106011.
Liang, L., L. Yu, and Z. Wang, 2022: Identifying the dominant impact factors and their contributions to heatwave events over mainland China. Sci. Total Environ., 848, 157527, https://doi.org/10.1016/j.scitotenv.2022.157527.
Liao, W., and Coauthors, 2018: Stronger contributions of urbanization to heat wave trends in wet climates. Geophys. Res. Lett., 45, 11 310–11 317, https://doi.org/10.1029/2018GL079679.
Liebmann, B., and C. A. Smith, 1996: Description of a complete (interpolated) outgoing longwave radiation dataset. Bull. Amer. Meteor. Soc., 77, 1275–1277, https://doi.org/10.1175/1520-0477-77.6.1274.
Lin, A., and T. Li, 2008: Energy spectrum characteristics of boreal summer intraseasonal oscillations: Climatology and variations during the ENSO developing and decaying phases. J. Climate, 21, 6304–6320, https://doi.org/10.1175/2008JCLI2331.1.
Lu, C., Y. Shen, Y. Li, B. Xiang, and Y. Qin, 2022: Role of intraseasonal oscillation in a compound drought and heat event over the middle of the Yangtze River basin during midsummer 2018. J. Meteor. Res., 36, 643–657, https://doi.org/10.1007/s13351-022-2008-3.
Luo, M., and N.-C. Lau, 2017: Heat waves in southern China: Synoptic behavior, long-term change, and urbanization effects. J. Climate, 30, 703–720, https://doi.org/10.1175/JCLI-D-16-0269.1.
Luo, M., and N.-C. Lau, 2019: Amplifying effect of ENSO on heat waves in China. Climate Dyn., 52, 3277–3289, https://doi.org/10.1007/s00382-018-4322-0.
Miao, C., Q. Sun, D. Kong, and Q. Duan, 2016: Record-breaking heat in northwest China in July 2015: Analysis of the severity and underlying causes. Bull. Amer. Meteor. Soc., 97, S97–S101, https://doi.org/10.1175/BAMS-D-16-0142.1.
Qian, Y., P.-C. Hsu, and K. Kazuyoshi, 2019: New real-time indices for the quasi-biweekly oscillation over the Asian summer monsoon region. Climate Dyn., 53, 2603–2624, https://doi.org/10.1007/s00382-019-04644-0.
Qin, Y., Y. Qin, Y. Shen, Y. Li, and B. Xiang, 2023: Numerical study on the effects of intraseasonal oscillations for a persistent drought and hot event in South China summer 2022. Remote Sens., 15, 892, https://doi.org/10.3390/rs15040892.
Robine, J.-M., S. L. K. Cheung, S. Le Roy, H. Van Oyen, C. Griffiths, J.-P. Michel, and F. R. Herrmann, 2008: Death toll exceeded 70,000 in Europe during the summer of 2003. C. R. Biol., 331, 171–178, https://doi.org/10.1016/j.crvi.2007.12.001.
Sun, Q., C. Miao, A. AghaKouchak, and Q. Duan, 2017: Unraveling anthropogenic influence on the changing risk of heat waves in China. Geophys. Res. Lett., 44, 5078–5085, https://doi.org/10.1002/2017GL073531.
Sun, Y., X. Zhang, F. W. Zwiers, L. Song, H. Wan, T. Hu, H. Yin, and G. Ren, 2014: Rapid increase in the risk of extreme summer heat in Eastern China. Nat. Climate Change, 4, 1082–1085, https://doi.org/10.1038/nclimate2410.
Tao, W., G. Huang, R. Wu, K. Hu, P. Wang, and D. Chen, 2017: Asymmetry in summertime atmospheric circulation anomalies over the northwest Pacific during decaying phase of El Niño and La Niña. Climate Dyn., 49, 2007–2023, https://doi.org/10.1007/s00382-016-3432-9.
Thompson, V., and Coauthors, 2022: The 2021 western North America heat wave among the most extreme events ever recorded globally. Sci. Adv., 8, eabm6860, https://doi.org/10.1126/sciadv.abm6860.
Wang, B., and X. Xie, 1997: A model for the boreal summer intraseasonal oscillation. J. Atmos. Sci., 54, 72–86, https://doi.org/10.1175/1520-0469(1997)054<0072:AMFTBS>2.0.CO;2.
Wang, P., J. Tang, X. Sun, S. Wang, J. Wu, X. Dong, and J. Fang, 2017: Heat waves in China: Definitions, leading patterns, and connections to large‐scale atmospheric circulation and SSTs. J. Geophys. Res. Atmos., 122, 10 679–10 699, https://doi.org/10.1002/2017JD027180.
Wang, Y., R. Wu, and Q. Gu, 2022: Asymmetric changes in intraseasonal oscillation intensity over the tropical western North Pacific in El Niño and La Niña developing summers. J. Climate, 35, 2617–2631, https://doi.org/10.1175/JCLI-D-21-0615.1.
Wei, J., W. Han, W. Wang, L. Zhang, and B. Rajagopalan, 2023: Intensification of heatwaves in China in recent decades: Roles of climate modes. npj Climate Atmos. Sci., 6, 98, https://doi.org/10.1038/s41612-023-00428-w.
Wu, R., and X. Cao, 2017: Relationship of boreal summer 10–20-day and 30–60-day intraseasonal oscillation intensity over the tropical western North Pacific to tropical Indo-Pacific SST. Climate Dyn., 48, 3529–3546, https://doi.org/10.1007/s00382-016-3282-5.
Wu, R., S. Yang, S. Liu, L. Sun, Y. Lian, and Z. Gao, 2010: Changes in the relationship between Northeast China summer temperature and ENSO. J. Geophys. Res., 115, D21107, https://doi.org/10.1029/2010JD014422.
Wu, S., M. Luo, X. Wang, E. Ge, W. Zhang, X. Gu, and J. Liu, 2023: Season-dependent heatwave mechanisms: A study of Southern China. Wea. Climate Extremes, 42, 100603, https://doi.org/10.1016/j.wace.2023.100603.
Wu, X., L. Wang, R. Yao, M. Luo, and X. Li, 2021: Identifying the dominant driving factors of heat waves in the North China Plain. Atmos. Res., 252, 105458, https://doi.org/10.1016/j.atmosres.2021.105458.
Xu, Z., and K. Fan, 2023: Comparison of changes in quasi-biweekly oscillation intensity over the western North Pacific during the developing late-summer of super El Niño events. Atmos. Res., 295, 107039, https://doi.org/10.1016/j.atmosres.2023.107039.
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, S., and T. Li, 2020: Cause for quasi-biweekly oscillation of zonal location of western Pacific subtropical high during boreal summer. Atmos. Res., 245, 105079, https://doi.org/10.1016/j.atmosres.2020.105079.
Yuan, Y., and S. Yang, 2012: Impacts of different types of El Niño on the East Asian climate: Focus on ENSO cycles. J. Climate, 25, 7702–7722, https://doi.org/10.1175/JCLI-D-11-00576.1.
Zhang, G., G. Zeng, C. Li, and X. Yang, 2020: Impact of PDO and AMO on interdecadal variability in extreme high temperatures in North China over the most recent 40-year period. Climate Dyn., 54, 3003–3020, https://doi.org/10.1007/s00382-020-05155-z.
Zhang, L.-S., J.-J. Xue, W.-G. Wang, and J. Sun, 2014: Comparative analysis of extreme high temperature weather in the summers of 2013 and 2003. Atmos. Ocean. Sci. Lett., 7, 132–136, https://doi.org/10.3878/j.issn.1674-2834.13.0073.
Zheng, B., D. Gu, A. Lin, D. Peng, C. Li, and Y. Huang, 2022: Structures and mechanisms of heatwaves related to quasi-biweekly variability over Southern China. J. Climate, 35, 7981–7994, https://doi.org/10.1175/JCLI-D-22-0282.1.
Zheng, B., Y. Huang, L. Li, and H. Wang, 2024: Heat waves related to quasi-biweekly variability over Southern China in the FGOALS-g3 model. Dyn. Atmos. Oceans, 106, 101443, https://doi.org/10.1016/j.dynatmoce.2024.101443.
Zhu, Y.-m., X.-q. Yang, X.-y. Chen, S.-s. Zhao, and X.-g. Sun, 2007: Interdecadal variation of the relationship between ENSO and summer interannual climate variability in China. J. Trop. Meteor., 13, 132–136.
