Dominant Role of Meridional Circulation in Regulating the Anomalous Subsidence of the Western Pacific Subtropical High in Early Summer 2020

Anomalous subsidence over the western part of the western Pacific subtropical high (WPSH) caused record-breaking precipitation anomalies over the Yangtze-Huaihe River catchment in early summer 2020 (June–July 2020). The meridional circulation (MC) made a positive contribution to this anomalous subsidence, while the zonal circulation (ZC) made a negative contribution. The quantitative contributions of the MC and ZC to this anomalous subsidence were approximately 110% and –10% in June, 130% and –30% in July, and 120% and –20% for the mean of June and July, respectively, suggesting that the MC played a dominant role in the anomalous subsidence of the western part of the WPSH. The anomalous MC, with a rising branch located at the Maritime Continent and a descending branch located over South China, was forced by the warming of the northern tropical Indian Ocean and the rapidly developed La Niña event, which further resulted in the intensification and southwestward expansion of the WPSH and thus in heavy rainfall over the Yangtze River region.


INTRODUCTION
The Meiyu in China or the Baiu in Japan and Changma in Korea is the major rainy season over the East Asian with duration from early-June to mid-July, which can have vital importance on the economic development and human society of East Asian regions [1][2][3][4][5][6][7]. The Meiyu rain season in 2020 started early (began on June 1, 7 days earlier than usual) and ended late (ended on August 2, 15 days later than usual), lasting for 62 days, which was the longest since 1961 ( Figure 1A) [1][2][3][4][5][6][7]. The amount of precipitation averaged over the Yangtze and Huaihe Rivers during this season reached 759.2 mm, recording a precipitation anomaly that was higher than the precipitation anomaly in the second year of a strong El Niño, ranking the highest amount since 1961. In early summer 2020 (June-July), the mid-lower reaches of the Yangtze River catchment experienced more than 10 heavy rainfall events, leading to direct economic losses of approximately 116 billion Chinese yuan [3].
To investigate the underlying causes of the Meiyu in summer 2020, many efforts have been made. Takaya et al. [2] and Zhou et al. [7] proposed that the extraordinary rainfall was originated from the tropical Indian Ocean warming, and this warming was possibly resulted from the enhanced western Pacific subtropical high (WPSH) [5]. Zheng and Wang [6] suggested that all three oceans of the Pacific, Indian, and Atlantic Oceans contributed to the Meiyu, and the Atlantic Ocean made the dominate contribution. Liu et al. [1] and Qiao et al. [5] investigated the causes of the Meiyu from the subseasonal time scale. Liu et al. [1] divided the duration of Meiyu into two periods and claimed that the sequential warm and cold Meiyu front regulated by the North Atlantic Oscillation was responsible for this unexpected extreme Meiyu event. Qiao et al. [5] divided the duration of Meiyu into three stages: advanced-onset, strong-persisting, and delayed-withdrawal, and suggested different causes in different periods. Although these studies suggested different causes of the extraordinary Meiyu in summer 2020, they consistently emphasized the role of the anomalous WPSH.
Climatologically, the WPSH can transport the water vapor from the western Pacific toward mainland China by the low-level southwesterly jet along the edge of the WPSH, which can influence the rainfall over the Yangtze River catchment [8][9][10][11][12][13][14][15]. The above-normal (subnormal) precipitation over the Yangtze River catchment in early summer is commonly accompanied by an intensification and westward shift (a weakening and eastward shift) of the WPSH [11][12][13][14][15]. Thus, the location, intensity, and variability of the WPSH are vitally important to the precipitation anomaly over the Yangtze River catchment in early summer [8][9][10]. In June and July of 2020, the area of exceptionally strong WPSH showed a strong anomalous subsidence, especially in the western part of the WPSH (Figures 2A,D,G), which caused record-breaking precipitation anomalies over the Yangtze River catchment in early summer 2020. Therefore, it is necessary to investigate the source of the anomalous subsidence in the western part of the WPSH.
Recently, to achieve a uniform description of the general circulation of the atmosphere from a global perspective and to reveal the mechanism of the complicated interactions and connections of the circulations between the low latitudes and mid-high latitudes, a novel three-pattern decomposition of global atmospheric circulation (3P-DGAC) method was proposed [16][17][18][19][20][21]. Hu et al. [17] suggested that tropical overturning circulations consist of a couple of orthogonal overturning circulations, i.e., meridional circulation (MC) and zonal circulation (ZC). Climatologically, the MC averaged over 135°E-160°E is characterized as the anticlockwise circulation in both hemispheres, while the ZC averaged over 15°N-25°N is characterized as the anticlockwise circulation in Indian Ocean and the clockwise circulation in the Pacific and Atlantic Ocean (Supplementary Figure S1). The sinking motion of the MC between 15°N and 25°N makes positive contribution to the WPSH, while the rising motion of the ZC between 135°E-160°E makes the negative contribution (Supplementary Figure S1). Since the tropical overturning circulation can be decomposed into the MC and ZC, the vertical wind contains two components, i.e., the vertical winds of the MC and ZC (Supplementary Figure S2). Supplementary Figure S2 shows that when analyzing the MC (ZC), the vertical velocity of the MC (ZC) should be used, and the vertical velocity of the ZC (MC) is regarded as the deviation if the total vertical velocity is used. Thus, there may be a bias in studying the MC and ZC in some previous studies because the total vertical velocity has commonly been used. Additionally, the anomalous subsidence in the western part of the WPSH in early summer 2020 could be decomposed into two parts that corresponded to anomalous MC and anomalous ZC (Figure 2), and thus the quantitative contribution of the MC and ZC to the anomalous WPSH can be clarified, which is not investigated in previous studies. Therefore, we would like to address the question: what were the effects of the MC and the ZC on the anomalous subsidence of the western part of the

Data
In this study, we employed monthly horizontal winds and vertical velocity data from five reanalysis datasets as follows: the Climate Forecast System Reanalysis Version 1 and Version 2 (hereafter CSFR) [22,23], the fifth generation of European Centre for Medium-Range Weather Forecast (ECMWF) atmospheric reanalysis of the global climate (hereafter ERA5) [24], the Japanese Meteorological Agency 55-years reanalysis (hereafter JRA-55) [25], the National Centers for Environmental Prediction (NCEP)/National Center for Atmospheric Research (NCAR) reanalysis (hereafter NCEP1) [26], and the NCEP/Department of Energy (DOE) reanalysis (hereafter NCEP2) [27]. The daily precipitation data were obtained from the Climate Prediction Center (CPC) Global Daily Unified Gauge-Based Analysis of Precipitation [28], and the monthly precipitation data were obtained from the Global Precipitation Climatology Project monthly precipitation dataset [29]. The monthly geopotential height data were obtained from ERA5 reanalysis datasets. The monthly sea surface temperature (SST) data were taken from the Extended Reconstructed SST version 5 (ERSST5) dataset [30]. For consistency, all datasets used in this study were interpolated to a 2.5°× 2.5°horizontal resolution. The time period analyzed in this study was early summer 2020 (i.e., June-July 2020), and the climatology of variables was defined as the climatological mean from 1981 to 2010.

Three-Pattern Decomposition of Global Atmospheric Circulation
A simple introduction of the 3P-DGAC method is offered in this section. To solve the unit discrepancy in calculating the threedimensional (3D) vorticity vector in the pressure coordinates, the spherical σ-coordinate system was adopted [17], namely: where a is the Earth's radius, p is the pressure, and P s 1000 hPa is the pressure at the Earth's surface. (u′, v′, _ σ) and (u, v, ω) represent the three velocity components in the spherical σ-coordinate system and spherical p-coordinate system, respectively. Thus, the 3D velocity field in the spherical σ-coordinate system can be represented as follows: which satisfies the following continuity equation: Based on the features of the Rossby wave in the middle-high latitudes and the Hadley and Walker circulations in the low latitudes, Hu et al. [17] defined the 3D horizontal circulation V′ R , MC V′ H , and ZC V′ W as follows: and the following continuity equations were satisfied: Eq. 5 is the sufficient condition that the components of V′ R , V′ H , and V′ W can be represented by the stream functions R(λ, θ, σ), H(λ, θ, σ), and W(λ, θ, σ), respectively, as follows: Because three-pattern circulations (horizontal circulation, MC, and ZC) exist in both the low and the middle-high latitudes, the global atmospheric circulation can be expressed as the superposition of the horizontal circulation, MC, and ZC, as follows: with the following components: The following restriction condition is needed to pick up the correct decomposition: Eq. 9 guarantees the uniqueness of the stream functions R, H, and W.
By combining Eqs. 8, 9, the following equations were obtained: where Δ 3 1 sin 2 θ Since the MC and ZC can be effectively separated from the tropical atmospheric circulation by using the 3P-DGAC method (Eq. 8, Supplementary Figure S2), the 3P-DGAC method is potentially useful for analyzing the relative contributions of the MC and the ZC to the anomalous sinking motion of the western part of the WPSH. Therefore, in this study, the 3P-DGAC method was used to investigate the effects of the MC and the ZC on the anomalous subsidence of the western part of the WPSH.
In addition to the 3P-DGAC method, by using the traditional two-dimensional (2D) decomposition method of the vortex and divergent circulations [31,32], the vertical velocity could also be decomposed into two parts, i.e., the vertical velocity of the regional Hadley circulation (RHC) and regional Walker circulation (RWC) (Supplementary Figure S3, Supplementary  Figure S4). Thus, the traditional 2D method was also adopted in this study. However, the results derived from the traditional 2D method were displayed in the Supplementary Materials. More information of the traditional 2D method and 3P-DGAC method can be found in [17][18][19]. Figure 2 displays the horizontal distribution of the vertical velocity anomaly at 500 hPa in early summer 2020. The anomalous subsidence of the western part of the WPSH can be decomposed into two parts that correspond to the anomalous MC and ZC by using the 3P-DGAC method, and the vertical velocity anomaly of MC was positive, while the vertical velocity anomaly of ZC was negative in the region of the western part of the WPSH from June to July, implying that the MC (ZC) made a positive (negative) contribution to the anomalous subsidence of the western part of the WPSH (Figure 2). Additionally, by comparing the first and second rows of Figure 2 (i.e., comparing Figures 2A,D,G and Figures 2B,E,H), it was found that the original vertical velocity anomaly and the vertical velocity anomaly of MC were quite similar, suggesting that the MC played a dominating role in the anomalous subsidence of the western part of the WPSH. Although Figure 2 displays the qualitative contributions of the MC and the ZC to the anomalous subsidence, further quantitative analysis is needed. Figure 3 displays the quantitative contributions of the MC and the ZC to the anomalous subsidence of the western part of the WPSH at 500 hPa in early summer 2020 based on the 3P-DGAC method derived from five reanalysis datasets. The quantitative contributions of the MC and the ZC to the anomalous subsidence of the western part of the WPSH were approximately 110% and -10% in June, 130% and -30% in July, and 120% and -20% for the mean of June and July, respectively ( Figure 3). Specifically, the quantitative contributions of the MC (ZC) were 110, 109, 108, 114, and 114% (-10%, -9%, -8%, -14%, and -14%) in June for the CFSR, ERA5, JRA-55, NCEP1, and NCEP2 reanalysis datasets, respectively. In July, the quantitative contributions of the MC (ZC) were 131, 136, 121, 133, and 146% (-31%, -36%, -21%, -33%, and -46%) for the five reanalysis datasets. For the mean of June and July, the quantitative contributions of the MC (ZC) were 119, 121, 114, 123, and 129% (-19%, -21%, -14%, -23%, and -29%) for the five reanalysis datasets. Although discrepancies existed in the quantitative contributions of the MC and the ZC to the anomalous subsidence based on the five different reanalysis datasets, the main results obtained from all five reanalysis datasets indicated that the MC played a dominant role in the anomalous subsidence of the western part of WPSH in early summer 2020. Additionally, although the quantitative contributions of the RHC and the RWC to the anomalous subsidence of the western part of the WPSH based on the traditional 2D decomposition method were different from those based on the 3P-DGAC method, the results derived from the two methods both supported the conclusion that meridional circulation played a dominant role in the anomalous subsidence of the western part of the WPSH in early summer 2020 (Figure 3, Supplementary Figure S5).

RESULTS
Since the MC played a dominating role in the anomalous subsidence of the western part of the WPSH in early summer 2020, we then wondered how the MC caused this anomalous subsidence. Figures 4A-D display the correlation between the regionally averaged vertical velocity anomaly of the MC over the region of the western part of the WPSH in June-July and the SST anomaly (SSTA) from December-January to June-July based on the seasonal means for the period 1979-2020. The correlation maps show that the correlated SSTA in the tropics is characterized by the persistent warming of the northern tropical Indian Ocean and the La Niña developing phase ( Figures 4A-D), which was also observed from the actual SSTA (Supplementary Figures S6A-D). The warming of the northern tropical Indian Ocean is strengthened from December-January to June-July, and the northern tropical Indian Ocean warming can heat the troposphere and force the equatorial Kelvin wave to propagate eastward [33,34], which triggers a positive Pacific-Japan (PJ) pattern ( Figure 4E) and strengthens the northwest Pacific anomalous anticyclone (vectors in Figure 4D). Additionally, an enhanced zonal SST gradient is caused by the rapidly developed La Niña event, forcing anomalous easterly winds in the equatorial western Pacific (vectors in Figure 4D), which generates anomalous convection and precipitation anomalies over the Maritime Continent ( Figures 4F,  Supplementary Figure S6F) [5]. These two processes can lead to anomalous MC with a rising branch located on the Maritime Continent and a descending branch located over South China ( Figures 2B,E,H), which can enhance the subsidence of the western part of the WPSH and lead to the southwestward expansion of the WPSH ( Figure 4E, Supplementary Figure S6E) [33,34].

DISCUSSION
The results in this study show that the anomalous subsidence over the western part of the WPSH is vitally important because the WPSH was stronger and inclined more toward the west, which caused record-breaking precipitation anomalies over the Yangtze River catchment in early summer 2020. The MC made a positive contribution to the anomalous subsidence of the western part of the WPSH, while the ZC made a negative contribution. This finding is also obtained by previous studies [1][2][3][4][5][6][7]. However, in this study, the quantitative contribution of the MC and ZC to the anomalous WPSH is obtained by using the 3P-DGAC method, which is not investigated in previous studies. The quantitative contributions of the MC and the ZC to the anomalous subsidence of the western part of the WPSH were approximately 110% and -10% in June, 130% and -30% in July, and 120% and -20% for the mean of June and July, respectively, suggesting that the MC played a dominant role in the anomalous subsidence of the western part of the WPSH. It should be noted that these results are obtained from the seasonal time scale and results may not the same in different periods from the subseasonal time scale, as Liu et al. [1] and Qiao et al. [5] suggested that the causes of the extraordinary Meiyu rain vary in different periods. Thus, the relative contribution of the MC and ZC to the extraordinary Meiyu rain from the subseasonal time scale should be investigated in the future. Chen et al. [35] proposed that the regional MC over the 110°E-160°E has intensified from 1979 to 2010, and the intensification of the regional MC is maintained under global warming [36]. Whether the rainfall over the Yangtze River catchment has intensified accompanied by the enhancement of the MC under global warming should also be explored. The warming of the northern tropical Indian Ocean and the rapidly developed La Niña event forced anomalous MC with a rising branch located on the Maritime Continent and a descending branch located over South China, which further resulted in the intensification and southwestward expansion of the WPSH and anomalous precipitation over the Yangtze and Huaihe River catchment. The relative role of the warming of the northern tropical Indian Ocean and the rapidly developed La Niña events in forcing anomalous MC is not clear and should be investigated in the future.
According to Hu et al. [19], the dynamical equations of the three-pattern circulations have been established by combining the primitive equations and the 3P-DGAC method. The novel dynamical equations can be used to diagnose and predict the changes of the horizontal, meridional, and zonal circulations. Thus, the novel dynamical equations are potentially useful for prediction of the anomalous WPSH since the anomalous WPSH is closely related to the anomalous MC.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding authors.