1. Introduction
Precipitation strongly affects the human society by supplying freshwater to land surface (Oki and Kanae 2006). Much literature regarding the precipitation climatology in East Asia has focused on summer when the majority of annual precipitation occurs (e.g., Lau et al. 1988; Kripalani and Kulkarni 2001; Gong and Ho 2002; Ho et al. 2005; Ueda et al. 2006; Wei et al. 2015). Although the spring precipitation accounts for a smaller portion of the annual total, it is no less important than the summer precipitation (Han and Byun 2006; Huang et al. 2015). The amount of spring precipitation is critical in replenishing the water resources in East Asian countries after dry winters (Lansigan et al. 2000; Sun and Yang 2012; Koide et al. 2013). Thus, understanding the spatiotemporal variation in the spring precipitation and related environmental mechanism is essential.
Large amounts of spring precipitation mainly occur in southern China and Southeast Asia (the Indochina Peninsula and the Philippines) from a climatological perspective (Chen et al. 2014; Qu et al. 2017). The precipitation amount in these two regions shows a significant interannual variation (resembling a seesaw pattern) related to the location of the mei-yu front in spring (Juneng and Tangang 2005; Xu et al. 2010; Pan et al. 2013; Seo et al. 2013; Qu et al. 2017). Recent asymmetric long-term negative and positive trends in the spring precipitation between Northeast Asia (China, the Korean Peninsula, and Japan) (Zhai et al. 2005; Choi et al. 2010; Duan et al. 2015; Li et al. 2016) and Southeast Asia (Vu and Mishra 2016), respectively, imply significant decadal and interannual variations in the spring precipitation over those regions. The dipole-like pattern of various time scales suggests a possible existence of leading spatiotemporal variability of the spring precipitation in East Asia. However, many previous studies focused on limited regions, such as southern and eastern China (e.g., Wu et al. 2003; Yang and Lau 2004; Feng and Li 2011; Ying et al. 2017; Wu and Mao 2018). Other studies on the Southeast Asian region also concerned limited region such as the Indochina Peninsula (e.g., Duc et al. 2018; Ruiz-Barradas and Nigam 2018). Accordingly, a comprehensive investigation covering the entire East Asian region is needed.
It has been well known that the spring precipitation variation in East Asia is influenced by various large-scale circulations. The westerly quasi-biennial oscillation displaces the major rainband to lower latitudes, which in turn leads to a decrease in the spring precipitation in the Korean Peninsula and Japan (Seo et al. 2013). The positive North Atlantic Oscillation in winter affects the pressure system over East Asia in the following spring by altering the upper-tropospheric circulations to result in driving below-normal precipitation in southern China (Xin et al. 2006; Sun and Yang 2012). Although the influence of Arctic Oscillation on the spring precipitation in the Southeast Asia remains unclear, it is positively (negatively) related to the spring precipitation in China (North Korea) (Gong and Ho 2003; Liu and Ding 2007; Choi et al. 2013; He et al. 2017).
Among various large-scale circulations, the sea surface temperature (SST) variation in the tropical Pacific related with El Niño–Southern Oscillation (ENSO) and associated tropospheric circulations over the western Pacific are likely the most important factors controlling the spring precipitation anomalies in East Asia (e.g., Wang et al. 2000; Yang and Lau 2004; Yoo et al. 2004: Feng et al. 2011; Chen et al. 2014; Kim et al. 2017; Duc et al. 2018; Stephan et al. 2018). The spring precipitation is above normal in southern China and the Korean Peninsula (Wu and Mao 2016; Lee and Julien 2017) and below normal in Southeast Asia (Juneng and Tangang 2005; Ruiz-Barradas and Nigam 2018) during the positive ENSO phase (i.e., El Niño). In recent decades, the central Pacific (CP) ENSO has become as noticeable as the canonical ENSO (Lee and McPhaden 2010; Liu et al. 2017). One remarkable characteristic of the CP ENSO is that its decadal variation is much stronger than the canonical ENSO. During strong CP El Niño years, a significant reduction (enhancement) in precipitation over southern (eastern) China can be observed, different from the pattern related to the canonical El Niño (Feng and Li 2011; Jia and Ge 2017). However, how the leading spatiotemporal variability of the spring precipitation in East Asia is connected to canonical and to CP ENSO in the tropical Pacific has not been investigated in detail yet.
The effects of the SST variations in the extratropical North Pacific like the Pacific decadal oscillation (PDO) on the variation in the East Asian precipitation were recently highlighted (e.g., Chen et al. 2013; Yang et al. 2017). The PDO leads to an asymmetric north–south precipitation anomaly in China (Yang et al. 2017). The link between the spring precipitation in China and ENSO can be modulated by the PDO phase (Wu and Mao 2016, 2018): for example, when the PDO is in phase with El Niño, the increasing trend of spring precipitation in southern China is enhanced. These findings imply that the leading spatiotemporal variability of the spring precipitation in East Asia might be affected by the SST variation in the extratropical North Pacific. Thus, it is necessary to disentangle the effect of SST in the extratropical North Pacific on the spring precipitation in order to explicitly reveal the intrinsic effect of the tropical SST variation in the canonical and CP ENSO.
This study investigated the leading spatiotemporal variability of spring precipitation over the entire East Asian region on interannual and decadal time scales. The associated large-scale circulations were also analyzed considering ENSO variability focusing on the tropical Pacific. This paper is organized as follows. The data, methodology, and configurations of the model simulations are described in sections 2 and 3. The dominant mode of spring precipitation and related large-scale circulations obtained from the statistical analyses and model simulations are presented in section 4. We also discuss the future changes in the dominant mode and spring precipitation based on model data from phase 5 of the Coupled Model Intercomparison Project. Finally, the main results of this study are summarized and discussed in section 5.
2. Data and method
The Global Precipitation Climatology Project (GPCP) monthly precipitation version 2.3 data (Adler et al. 2003) with a horizontal resolution of 2.5° × 2.5° for boreal spring (March–May) from 1979–2017 was analyzed. This gridded data were constructed by various observations obtained from low- and geosynchronous-orbit satellites and about 7000 rain gauge stations. To investigate large-scale environments, the zonal and meridional winds, vertical velocity, and specific humidity data from 1000 to 200 hPa were obtained from the 2.5° × 2.5° European Centre for Medium-Range Weather Forecasts interim reanalysis (ERA-Interim; Dee et al. 2011). The SST data were obtained from the 1° × 1° Met Office Hadley Centre Sea Ice and Sea Surface Temperature version 1.1 (HadISST 1.1; Rayner et al. 2003). The ENSO indices (e.g., Niño 1+2, Niño-3, Niño-4, and Niño-3.4) were calculated using the HadISST 1.1. The definitions of the ENSO indices used in this study are the same as the original Niño indices, which are domain-averaged SST anomalies within certain Pacific ocean areas; Niño-1+2 (0°–10°S, 90°–80°W), Niño-3 (5°N–5°S, 150°–90°W), Niño-4 (5°N–5°S, 160°E–150°W), and Niño-3.4 (5°N–5°S, 170°–120°W) (Trenberth and Stepaniak 2001). Based on these indices, we selected ENSO years from HadISST 1.1 for the composite analysis (Table 1). When the mean values of Niño-1+2, Niño-3, and Niño-3.4 during March–May (MAM) of a certain year showed the same sign and exceeded the 0.5 standard deviation, the year was defined as eastern Pacific (EP) El Niño year for positive and EP La Niña year for negative. Similarly, the CP El Niño and CP La Niña years were selected by using the Niño-4 and Niño-3.4 indices, but only if the value of Niño-1+2 was within 0.5 standard deviations. To investigate future changes in the spring precipitation and associated large-scale circulation, the historical (1979–2005) and representative concentration pathway (RCP) 8.5 scenario (2006–99) runs of the CMIP5 models were used. All CMIP5 model data were interpolated to a common 2.5° × 2.5° grid.
Years of each ENSO type’s SST condition.
Empirical orthogonal function (EOF) analysis was applied to extract the leading spatiotemporal variability of the spring precipitation over East Asia. The EOF analysis allows to investigate each individual spatiotemporal variation from mixed original variation of spring precipitation. Although the dominant variability can be extracted by EOF analysis, the result may still contain complex temporal variations at multiple time scales. To extract each temporal variation on interannual and decadal time scales, high- and low-pass filter analyses were applied to the principal component (PC) time series of the first leading EOF mode. The cutoff frequency for the high- and low-pass filter analyses was obtained through the power spectrum analysis (Torrence and Compo 1998) to the PC time series. The Student’s t test was used to examine the statistical significance of all analyses. To consider the degree of freedom, a method using the e-folding decay time of autocorrelation was applied (Panofsky and Brier 1958; Leith 1973; Hartmann 2016).
3. Model simulation
To confirm the influence of large-scale circulation on the leading spatiotemporal variability of spring precipitation in East Asia, experiments forced by nine different SST boundaries were performed using the Community Atmosphere Model version 5.1 (CAM5), one of the components of the Community Earth System Model (CESM). The experiment sets consist of one control and eight experimental runs (Table 2). The control run was forced by the climatological annual cycle of the SST from 1900 to 2014 using the HadOIBl, the boundary SST data of CAM5. The four experimental runs were forced by the mean annual cycles of the SST of four different ENSO types (i.e., EP El Niño, EP La Niña, CP El Niño, and CP La Niña), which are monthly SST composites averaged for the years of each of ENSO types. Herein, we used the same definition as presented in section 2 to select the years of the four ENSO types from HadOIBl for the period of 1900–2014. The remaining four experimental runs are identical to the aforementioned ENSO-type experimental runs, but they are only for the tropical Pacific domain (15°S–15°N, 110°E–80°W). The climatological annual cycle of the SST was forced for the other remaining regions. Hereafter, we refer to these runs as tropical domain runs for EP and CP ENSO (i.e., Tro_EP El Niño, Tro_EP La Niña, Tro_CP El Niño, and Tro_CP La Niña). Each of the control and experimental runs include 30 subsets for the ensemble average. To set 30 atmospheric initial conditions for the subsets, we performed a CAM5 atmospheric experiment for 45 years using the F_AMIP_CAM5 component set of CESM. We excluded the first 15 years as the spinup period. The model output on 1 January for each of the remaining 30 years was used as the initial condition of each subset. The sea ice condition was configured by the prescribed mode of the Community Ice Code (CICE), which is the ice component of CESM (for details, please see http://www.cesm.ucar.edu/models/cesm1.0/cice/doc/index.html). The model configuration is that the horizontal and vertical resolutions are 1.9° × 2.5° latitude–longitude and 26 layers of sigma–pressure coordinate, respectively.
SST boundary and atmospheric initial conditions for each experiment.
4. Results
a. Dipole mode of the spring precipitation and associated large-scale circulations
To identify the leading spatiotemporal variability of spring precipitation in East Asia, we applied EOF analysis to the MAM-mean precipitation over the region 5°–45°N, 95°–145°E. A dipole mode between southern China and Southeast Asia is identified as the first leading EOF mode (Fig. 1a). This mode accounts for 37.6% of the total variance, and differs from the remaining modes (Fig. S1 in the online supplemental material). Its PC time series shows complex temporal variations with both strong interannual and decadal variations (bars in Fig. 1b). The power spectrum analysis of the PC time series reveals that prominent periodicities can be grouped into high and low frequencies based on a 7.1-yr periodicity (red circle in Fig. 1c). Based on the 7.1-yr periodicity, we applied the high- and low-pass filter analyses to the PC time series (symbols with lines in Fig. 1b). The variation in the high-pass-filtered (low-pass-filtered) time series represents the variation in the PC time series for the periodicity window smaller (larger) than 7.1 years. Accordingly, the high- and low-pass-filtered PC time series can be considered as the interannual and decadal variations of the dipole mode, respectively. Hereafter, analyses were performed based on these high- and low-pass-filtered PC time series.
Normalized (a) eigenvector and (b) associated principal component (PC) time series of the first leading EOF mode of March–May (MAM)-mean precipitation for 1979–2017. (c) The power of periodicity for the PC time series in (b). The number 37.6% in (a) is the explained variance of the EOF mode. The high- and low-pass filters in (b) are based on 7.1-yr periodicity. The red dashed line in (c) is the red-noise spectrum.
Citation: Journal of Climate 33, 23; 10.1175/JCLI-D-19-0625.1
To identify atmospheric circulations related to the dipole mode of the spring precipitation between southern China and Southeast Asia, the regression coefficients onto the original and filtered PC time series in Fig. 1b were calculated for the MAM-mean vertically integrated moisture divergence and flux from 1000 to 500 hPa, 850-hPa wind, and SST (Fig. 2). The original, high-pass-filtered, and low-pass-filtered variables were utilized when looking into the regression onto the relevant environmental variables. In all cases, significant meridional moisture convergence and divergence anomalies are observed over southern China and Southeast Asia, respectively (Figs. 2a–c). In addition, anomalous anticyclonic moisture fluxes over the South China Sea and the Philippines lead the moisture influx from Southeast Asia into southern China. These moisture patterns account for the positive phase of the dipole mode where spring precipitation increases in southern China and decreases in Southeast Asia. Similar moisture conditions related to the interannual and decadal variations may be due to the analogous large-scale environmental forcing at different time scales.
Regression coefficients of four climate variables with respect to (a),(d) unfiltered, (b),(e) high-pass-filtered, and (c),(f) low-pass-filtered PC time series in Fig. 1b: the MAM-mean vertically integrated (a)–(c) moisture flux (qVh,1000–500; vectors; unit: kg m−1 s−1) and moisture divergence (∇ ⋅ qVh,1000–500; shaded; unit: 10−8 kg m−2 s−1) along 1000 to 500 hPa and (d)–(f) sea surface temperature (SST; shaded; unit: °C) and wind field at 850 hPa (Vh,850; vectors; unit: m s−1). The area with the diagonal line indicates that the regression coefficient is statistically significant at the 95% confidence level. The vectors were only plotted if the grid values satisfied the statistical significance at the 95% confidence level.
Citation: Journal of Climate 33, 23; 10.1175/JCLI-D-19-0625.1
The regression coefficients of the SST onto the PC time series suggest that the above mentioned large-scale circulation in the western North Pacific is closely related to the El Niño–like SST (Figs. 2d–f). The regressed SST pattern against the high-pass-filtered PC time series resembles the EP El Niño, whereas that against the low-pass-filtered PC time series resembles the CP El Niño. Note that the decadal variation of ENSO predominantly appears in the CP (Timmermann et al. 2018); that is, the decadal variation of the CP ENSO is related to the decadal dipole mode of the spring precipitation in East Asia. The significant decadal phase shift of the dipole mode between 1998 and 1999 shown in Fig. 1b may be related to the recent activation of CP ENSO (Lee and McPhaden 2010; Liu et al. 2017). Despite the different SST patterns, an anomalous lower-tropospheric anticyclonic circulation over the South China Sea and Philippines can be observed in both the high- and low-pass-filtered cases (Figs. 2e,f). This seems to be important in inducing the anticyclonic moisture circulation over those regions (see Figs. 2b,c). Meanwhile, there are relatively warmer SST and stronger winds over the central and eastern Pacific along 10°–30°N in the low-pass-filtered case than those in the high-pass-filtered case (Fig. 2e vs Fig. 2f), indicating the wind–evaporation–SST feedback process in the CP El Niño (Yeh et al. 2015).
Figure 3 shows the regression coefficients of the zonal cross sections for the MAM-mean horizontal divergence and vertical circulation between 1000 and 200 hPa over the tropical North Pacific (0°–10°N) onto the PC time series. In both high- and low-pass-filtered cases, anomalous lower-tropospheric divergence and convergence (i.e., upper-tropospheric convergence and divergence) occurs in the western and the eastern/central regions, respectively. In addition, the cyclonic vertical circulation indicates a weakened Walker circulation, which is due to the zonal SST gradient in the tropical Pacific characterized by a warm-east and cold-west pattern (i.e., a well-known feature of El Niño) (e.g., Wang et al. 2000; Chen et al. 2014; Kim et al. 2017). Because the main SST warming regions of the EP and CP El Niño differ, the core of the lower-tropospheric convergence zone and rising motion in the low-pass-filtered case appear west of those in the high-pass-filtered case (Figs. 3b,c). This is because the relatively higher SST over the eastern (central) tropical Pacific may lead stronger convection therein in the EP El Niño (CP El Niño) compared to that over the central (eastern) tropical Pacific (Sun and Yu 2009; Timmermann et al. 2018). Nevertheless, the downward motion between the upper-tropospheric convergence and lower-tropospheric divergence zones is similarly observed in the western Pacific region (about 110°–140°E) for both the low- and high-pass-filtered cases. This high pressure pattern (i.e., western North Pacific anticyclone; Wang et al. 2000) induces the anomalous lower-tropospheric anticyclonic circulation over the western Pacific region shown in Figs. 2e and 2f.
Regression coefficients of the zonal cross sections for the MAM-mean horizontal divergence (shaded; unit: 10−7 s−1) and vertical circulation (vectors; unit: m s−1) between 1000 and 200 hPa over the tropical North Pacific (0°–10°N). The black vectors indicate the statistical significance at the 95% confidence level. The horizontal divergence was only plotted if the grid value was statistically significant at the 95% confidence level.
Citation: Journal of Climate 33, 23; 10.1175/JCLI-D-19-0625.1
b. CAM5 model experiments
Before performing the ensemble experiments using CAM5 to identify the effect of EP and CP ENSO-type SST in the tropical Pacific on the dipole mode of the spring precipitation in East Asia, we checked whether the dipole mode and its associated large-scale circulations can be observed in actual EP and CP ENSO years. We investigated anomalous atmospheric patterns during the four phases of the EP and CP ENSO years (Table 1). Figure 4 presents the composite map of the anomalies of the MAM-mean precipitation, vertically integrated moisture flux from 1000 to 500 hPa, 850-hPa wind, and SST for each EP and CP ENSO phase. The dipole mode of the MAM-mean precipitation between southern China and Southeast Asia can be observed in all phases (Figs. 4a,c,e,g). As seen in Fig. 2, the dipole mode is induced by anomalous anticyclonic or cyclonic moisture circulations over the South China Sea and the Philippines. It can be also found that these moisture circulations are driven by large-scale circulations related to the tropical SST variations in the EP and CP ENSO (Figs. 4b,d,f,h).
Composite map of the anomalies for (a),(c),(e),(g) the MAM-mean qVh,1000–500 (vectors; unit: kg m−1 s−1) and precipitation (shaded; unit: mm day−1) and (b),(d),(f),(h) SST (shaded; unit: °C) and Vh,850 (vectors; unit: m s−1) for each ENSO phase. The hatched areas and the black vectors are statistically significant at the 95% confidence level. Corr. represents the spatial correlation coefficients of precipitation and SST against the eigenvector in Fig. 1a and the regressed SST patterns in Figs. 2e and 2f, respectively. An asterisk (*) indicates that the Corr. is statistically significant at the 95% confidence level.
Citation: Journal of Climate 33, 23; 10.1175/JCLI-D-19-0625.1
Figure 5 shows the same climatic variables as Fig. 4, but for the composites of the differences between each ENSO-type SST experimental run and the control run of the CAM5 simulation. Compared with the patterns in Fig. 4, spatial correlation coefficients obtained for the precipitation and SST patterns in all four ENSO experimental runs show the similarity between the models and observations at the 95% statistical significance level. This implies that the CAM5 model replicates the observed dipole mode reasonably well. Although the general spatial pattern of the precipitation at the 30°N latitude line somewhat differs from the observations, the dipole pattern is still identified between southern China and Southeast Asia. As seen in Figs. 2 and 4, it is also found that the dipole pattern is related to anomalous moisture circulations over the South China Sea and Philippines, which are formed by the large-scale circulations related to the EP and CP ENSO. These results suggest that the EP and CP ENSO-type SST variations in the tropical Pacific are major factors inducing the dipole mode of the spring precipitation in East Asia.
As in Fig. 4, but for the differences between each ENSO type’s SST experimental run and the control run of the CAM5 simulations. Corr. indicates the spatial correlation coefficients of precipitation and SST against the observed patterns shown in Fig. 4.
Citation: Journal of Climate 33, 23; 10.1175/JCLI-D-19-0625.1
Some previous studies suggested that the SST variation in the extratropical North Pacific–like PDO can also be another important factor for the spatiotemporal variation in East Asian spring precipitation (Chen et al. 2013; Yang et al. 2017; Wu and Mao 2016, 2018). It should be checked whether the EP and CP ENSO-type SST in the tropical Pacific can directly induce the dipole mode of the spring precipitation in East Asia without the influence of the SST variation in the extratropical North Pacific. Accordingly, we checked the results of experiments that the SST forcing domain is limited in the tropical Pacific region (Fig. 6). As shown in Fig. 5, the dipole patterns of the spring precipitation between southern China and Southeast Asia and their related anomalous environmental circulations are also found. In summary, all experimental results suggest that the SSTs of the EP and CP ENSO types in the tropical Pacific can directly induce the dipole mode of the spring precipitation in East Asia without the contribution of the SST variation in the extratropical North Pacific. Meanwhile, the SST variation in the extratropical North Pacific may amplify the dipole mode if they are in phase with ENSO, as discussed in the introduction (Wu and Mao 2016, 2018).
As in Fig. 5, but for the results of the CAM5 simulations for experiments with limited SST forcing domain.
Citation: Journal of Climate 33, 23; 10.1175/JCLI-D-19-0625.1
c. Future projections using CMIP5 models
To project future changes in the dipole mode and spring precipitation, we applied EOF analysis to the MAM-mean precipitation for the period of 1979–2099 over the East Asian domain shown in Fig. 1a for each CMIP5 model. Subsequently, the regressed patterns of SST and 850-hPa wind against the PC time series of the first leading EOF mode for each CMIP5 model were gathered. In this study, we used 15 CMIP5 models which replicate the observed patterns well (Table 3). These models satisfy the criterion that the spatial correlation coefficients of the first leading EOF eigenvector and the regressed SST pattern against the observations shown in Figs. 1a and 2d exceed 0.5 at the 95% statistical significance level. Figure 7 shows that the eigenvectors of the first leading EOF mode of the 15 selected CMIP5 models exhibit the dipole mode between southern China and Southeast Asia. The regressed patterns of SST and 850-hPa wind of all models show the El Niño pattern. In addition, when we separately checked the spatial patterns of the first leading EOF mode’s eigenvector and the related SST and 850-hPa winds for the historical period (1979–2005) and future period (2070–99), most of the CMIP5 models show results consistent with Fig. 7 (Figs. S2 and S3). This indicates that the CMIP5 models’ EOFs are not sensitive to the time period used for analysis. Hence, all these results suggest that the selected CMIP5 models can replicate the observational dipole mode and potentially capture the associated mechanisms. Figure 8 presents the PC time series associated with the first leading EOF modes. Although only five models (CESM1-CAM5, FGOALS-g2, MRI-CGCM3, NorESM1-M, and NorESM1-ME) satisfy the 95% statistical significance level regarding their linear regression slopes, eight models (red slopes in Fig. 8) show a general increasing trend and the others (blue slopes in Fig. 8) present a decreasing trend.
Information about 15 CMIP5 models.
Spatial distributions of (first column),(third column) the first leading EOF mode’s normalized eigenvector for the MAM-mean precipitation (shaded) and (second column),(fourth column) the regression coefficients onto the associated PC time series for the MAM-mean SST (shaded; unit: °C) and wind field at 850 hPa (vector; unit: m s−1) during 1979–2099 for each CMIP5 model. Var. represents the explained variance of the first leading EOF mode. Corr. indicates the spatial correlation coefficients of eigenvector and SST to the observed patterns shown in Figs. 1a and 2d. The hatched area indicates that the regression coefficients are statistically significant at the 95% confidence level. The wind field at 850 hPa was only plotted if the grid value was statistically significant at the 95% confidence level.
Citation: Journal of Climate 33, 23; 10.1175/JCLI-D-19-0625.1
Principal component (PC) time series of the first leading EOF mode for each CMIP5 model. The red line represents the linear regression line. Positive and negative linear regression slopes are marked red and blue, respectively. An asterisk (*) indicates that the linear regression slope is statistically significant at the 95% confidence level.
Citation: Journal of Climate 33, 23; 10.1175/JCLI-D-19-0625.1
Figure 9 shows the map of the differences in the MAM-mean precipitation and SST between the future (2070–99) and present (1979–2005) periods for each CMIP5 model. Table 4 presents their spatial correlation coefficients against the first leading EOF eigenvector and its regressed SST pattern in the tropical Pacific (15°S–15°N, 110°E–80°W) (i.e., Fig. 9 vs Fig. 7) and the linear regression slopes of the first leading EOF mode’s PC time series for each of the corresponding CMIP5 models. Based on the signals of the spatial correlation coefficients and linear regression slopes, the 15 selected CMIP5 models can be divided into three groups (Table 4). In the first group, seven models (MRI-CGCM3, IPSL-CM5B-LR, CESM1-CAM5, CMCC-CMS, GISS-E2-H, MPI-ESM-MR, and IPSL-CM5A-LR) show consistent positive signals for the spatial correlation coefficients and linear regression slope, indicating a future change in the spring precipitation over East Asia related to the positive phase of the dipole mode and El Niño–like SST warming in the tropical Pacific (Fig. 9). An opposite pattern is observed for the three models (FGOALS-g2, NorESM1-ME, and NorESM1-M) of the second group, showing consistent negative signals (i.e., a future change in the spring precipitation related to the negative phase of the dipole mode and La Niña–like SST warming in the tropical Pacific).
Map of the differences in the (first column),(third column) MAM-mean precipitation (unit: mm month−1) and (second column),(fourth column) SST (unit: °C) between the future (2070–99) and present (1979–2005) periods for each CMIP5 model. The hatched area indicates statistical significance at the 95% confidence level.
Citation: Journal of Climate 33, 23; 10.1175/JCLI-D-19-0625.1
Spatial correlation coefficients (Corr.) of the future anomalies of the spring precipitation in East Asia and SST in the tropical Pacific (15°S–15°N, 110°E–80°W) (Fig. 9) against the first leading EOF eigenvectors and their regressed SST patterns (Fig. 7) for each of CMIP5 models. The linear regression slopes of the first leading EOF mode’s PC time series of the CMIP5 models shown in Fig. 8 are also presented. An asterisk (*) indicates statistical significance at the 95% confidence level.
To clarify the relationship between the variation in the dipole mode and total future change in the spring precipitation, we calculated the future change in the spring precipitation solely based on the variation in the dipole mode for each CMIP5 model (Fig. S4). Notably, their magnitudes are much smaller than those of the total future changes in the spring precipitation. This may be the case because the total future change is due to the combined influence of various factors, such as changes in other EOF variabilities and local Hadley circulation due to global warming (Lee and Wang 2014; Tao et al. 2016). On the other hand, the spatial correlation coefficients of them against to the spatial patterns of total future changes in the spring precipitation shown in Fig. 9 (Corr. in Fig. S4) are generally as high as those of the first leading EOF mode’s eigenvectors shown in Table 4. This implies that the dipole mode generally strengthens the spatial pattern of the total future change in the spring precipitation. Thus, according to the projections from about half of the analyzed models belonging to the first group, the spatial pattern of future change in spring precipitation in East Asia would be partially related with the positive phase of dipole mode, possibly due to the El Niño–like SST warming in the tropical Pacific (Cai et al. 2018).
The remaining five models of the third group show inconsistent signals for the dipole mode and SST warming in the tropical Pacific (Table 4). In addition, although three models, MPI-ESM-MR, IPSL-CM5A-LR, and NorESM1-M, show consistent signals for the spatial correlation coefficients and linear regression slope, those values are relatively small. These results may imply that, in some cases, other EOF modes might be more important than the dipole mode in the future. Another reason might be that future changes in the spring precipitation simulated by these models may be unrelated to the present-day climate variability. Meanwhile, several models (e.g., CESM1-CAM5, CMCC-CMS, HadGEM2-AO, and IPSL-CM5A-LR) project a significant SST warming in the extratropical North Pacific, which is an out-of-phase spatial pattern of the El Niño (Fig. 9). It is difficult to determine the effects of this SST warming on the future spring precipitation and dipole mode based on the results of the present study. Further SST-forced model experiments may be needed to identify the individual impact of the SST variation in each Pacific region; however, this is beyond the scope of the present study. As shown in previous works of Wu and Mao (2016, 2018), the out-of-phase SST warming in the extratropical North Pacific in the future may weaken the signal of the dipole mode induced by the El Niño–like SST warming in the tropical Pacific.
5. Summary and discussion
This study investigated the leading spatiotemporal variability of the spring precipitation in East Asia and associated environmental mechanisms. According to an EOF analysis, the first leading mode is a dipole pattern between southern China and Southeast Asia. The temporal variation of this mode includes both of strong interannual and decadal components. Although the dipole mode is a part of the total variance, its spatiotemporal variation explains the actual change in the spring precipitation in East Asia. The decrease (increase) in the spring precipitation in southern China (Southeast Asia) during the recent period until the early 2010s reported in previous studies (e.g., Choi et al. 2010; Huang et al. 2015; Li et al. 2016; Vu and Mishra 2016) corresponds to the negative phase of the dipole mode. Our results based on an extended observational period up to 2017 indicate that the dipole mode becomes positive phase after the early 2010s due to decadal variations. This implies that the changing trend of spring precipitation in East Asia might have reversed recently.
The results show that the interannual and decadal variations of the dipole mode are affected by large-scale environmental circulations related to the EP and CP ENSO variabilities, respectively. Anomalous anticyclonic and cyclonic moisture circulations over the South China Sea and Philippines induce a dipole mode between southern China and Southeast Asia. Anomalous air–sea interactions due to the EP and CP ENSO-type SST variations lead to these moisture conditions over the western North Pacific region. The SST forced ensemble experiments using CAM5 confirm that the EP and CP ENSO-type SST variations in the tropical Pacific directly induce the dipole mode. Thus, the SST evolution in the tropical Pacific related to the ENSO during winter and spring can be used as a predictor for the spring precipitation in East Asia.
In CMIP5 models, about half of 15 models show that the positive phase of the dipole mode become dominant, possibly due to the El Niño–like tropical SST warming (Cai et al. 2018). Previous studies presented that the precipitation in the East Asian region will significantly increase under future warmer climate condition (e.g., Scoccimarro et al. 2013; Chen and Frauenfeld 2014; Cha et al. 2016). Hence, the dipole mode may enhance such an increasing precipitation trend over southern China. However, the future projections between individual models considerably vary as shown in section 4c, suggesting that it is still hard to be confident at future changes in dipole mode and precipitation. Meanwhile, the effects of SST warming in the extratropical North Pacific on the future spring precipitation and dipole mode remain uncertain as well. Therefore, further investigation on the other EOF modes (particularly, the second mode) and SST warming in the extratropical North Pacific may be necessary to make sure these uncertainties.
Acknowledgments
This work was supported by the National Research Foundation of Korea (NRF) grant funded by the South Korean government (MSIT) (2019R1A2C208429412). Tae-Won Park and Doo-Sun R. Park were supported by the National Research Foundation of the South Korean government (NRF-2020R1A4A3079510). Doo-Sun R. Park was supported by the research fund of the Korea Environment Industry & Technology Institute (KEITI) through the Climate Change Correspondence Program, funded by the Korea Ministry of Environment (MOE) (2018001310004).
REFERENCES
Adler, R. F., and Coauthors, 2003: The version-2 Global Precipitation Climatology Project (GPCP) monthly precipitation analysis (1979–present). J. Hydrometeor., 4, 1147–1167, https://doi.org/10.1175/1525-7541(2003)004%3c1147:TVGPCP%3e2.0.CO;2.
Cai, W., and Coauthors, 2018: Increased variability of eastern Pacific El Niño under greenhouse warming. Nature, 564, 201–206, https://doi.org/10.1038/s41586-018-0776-9.
Cha, D.-H., and Coauthors, 2016: Future changes in summer precipitation in regional climate simulations over the Korean Peninsula forced by multi-RCP scenarios of HadGEM2-AO. Asia-Pac. J. Atmos. Sci., 52, 139–149, https://doi.org/10.1007/s13143-016-0015-y.
Chen, J., Z. Wen, R. Wu, Z. Chen, and P. Zhao, 2014: Interdecadal changes in the relationship between southern China winter–spring precipitation and ENSO. Climate Dyn., 43, 1327–1338, https://doi.org/10.1007/s00382-013-1947-x.
Chen, L., and O. W. Frauenfeld, 2014: A comprehensive evaluation of precipitation simulations over China based on CMIP5 multimodel ensemble projections. J. Geophys. Res. Atmos., 119, 5767–5786, https://doi.org/10.1002/2013JD021190.
Chen, W., J. Feng, and R. Wu, 2013: Role of ENSO and PDO in the link of the East Asian winter monsoon to the following summer monsoon. J. Climate, 26, 622–635, https://doi.org/10.1175/JCLI-D-12-00021.1.
Choi, K.-S., D.-W. Kim, and H.-R. Byun, 2010: The regime shift in the early 1980s of spring precipitation in Korea. Int. J. Climatol., 30, 721–732, https://doi.org/10.1002/joc.1927.
Choi, K.-S., S.-D. Kang, and H.-D. Kim, 2013: Possible relationship between North Korean total rainfall and Arctic Oscillation in May. Theor. Appl. Climatol., 112, 483–494, https://doi.org/10.1007/s00704-012-0738-3.
Dee, D. P., and Coauthors, 2011: The ERA-Interim reanalysis: Configuration and performance of the data assimilation system. Quart. J. Roy. Meteor. Soc., 137, 553–597, https://doi.org/10.1002/qj.828.
Duan, W., B. He, K. Takara, P. Luo, M. Hu, N. E. Alias, and D. Nover, 2015: Changes of precipitation amounts and extremes over Japan between 1901 and 2012 and their connection to climate indices. Climate Dyn., 45, 2273–2292, https://doi.org/10.1007/s00382-015-2778-8.
Duc, H. N., H. Q. Bang, and N. X. Quang, 2018: Influence of the Pacific and Indian Ocean climate drivers on the rainfall in Vietnam. Int. J. Climatol., 38, 5717–5732, https://doi.org/10.1002/joc.5774.
Feng, J., and J. Li, 2011: Influence of El Niño Modoki on spring rainfall over south China. J. Geophys. Res., 116, D13102, https://doi.org/10.1029/2010JD015160.
Feng, J., W. Chen, C.-Y. Tam, and W. Zhou, 2011: Different impacts of El Niño and El Niño Modoki on China rainfall in the decaying phases. Int. J. Climatol., 31, 2091–2101, https://doi.org/10.1002/joc.2217.
Gong, D.-Y., and C.-H. Ho, 2002: Shift in the summer rainfall over the Yangtze River valley in the late 1970s. Geophys. Res. Lett., 29, 1436, https://doi.org/10.1029/2001GL014523.
Gong, D.-Y., and C.-H. Ho, 2003: Arctic Oscillation signals in the East Asian summer monsoon. J. Geophys. Res., 108, 4066, https://doi.org/10.1029/2002JD002193.
Han, S.-U., and H.-R. Byun, 2006: The existence and the climatological characteristics of the spring rainy period in Korea. Int. J. Climatol., 26, 637–654, https://doi.org/10.1002/joc.1274.
Hartmann, D. L., 2016: ATM 552 Notes: Time Series Analysis. University of Washington, 275 pp.
He, S., Y. Gao, F. Li, H. Wang, and Y. He, 2017: Impact of Arctic Oscillation on the East Asian climate: A review. Earth-Sci. Rev., 164, 48–62, https://doi.org/10.1016/j.earscirev.2016.10.014.
Ho, C.-H., J.-H. Kim, K.-M. Lau, K.-M. Kim, D. Gong, and Y.-B. Lee, 2005: Interdecadal changes in heavy rainfall in China during the northern summer. Terr. Atmos. Oceanic Sci., 16, 1163–1176, https://doi.org/10.3319/TAO.2005.16.5.1163(A).
Huang, D.-Q., J. Zhu, Y.-C. Zhang, J. Wang, and X.-Y. Kuang, 2015: The impact of the East Asian subtropical jet and polar front jet on the frequency of spring persistent rainfall over southern China in 1997–2011. J. Climate, 28, 6054–6066, https://doi.org/10.1175/JCLI-D-14-00641.1.
Jia, X., and J. Ge, 2017: Interdecadal changes in the relationship between ENSO, EAWM, and the wintertime precipitation over China at the end of the twentieth century. J. Climate, 30, 1923–1937, https://doi.org/10.1175/JCLI-D-16-0422.1.
Juneng, L., and F. T. Tangang, 2005: Evolution of ENSO-related rainfall anomalies in Southeast Asia region and its relationship with atmosphere–ocean variations in Indo-Pacific sector. Climate Dyn., 25, 337–350, https://doi.org/10.1007/s00382-005-0031-6.
Kim, J.-W., S.-I. An, S.-Y. Jun, H.-J. Park, and S.-W. Yeh, 2017: ENSO and East Asian winter monsoon relationship modulation associated with the anomalous northwest Pacific anticyclone. Climate Dyn., 49, 1157–1179, https://doi.org/10.1007/s00382-016-3371-5.
Koide, N., A. W. Robertson, A. V. M. Ines, J.-H. Qian, D. G. DeWitt, and A. Lucero, 2013: Prediction of rice production in the Philippines using seasonal climate forecasts. J. Appl. Meteor. Climatol., 52, 552–569, https://doi.org/10.1175/JAMC-D-11-0254.1.
Kripalani, R. H., and A. Kulkarni, 2001: Monsoon rainfall variations and teleconnections over South and East Asia. Int. J. Climatol., 21, 603–616, https://doi.org/10.1002/joc.625.
Lansigan, F. P., W. L. de los Santos, and J. O. Coladilla, 2000: Agronomic impacts of climate variability on rice production in the Philippines. Agric. Ecosyst. Environ., 82, 129–137, https://doi.org/10.1016/S0167-8809(00)00222-X.
Lau, K.-M., G. J. Yang, and S. H. Shen, 1988: Seasonal and intraseasonal climatology of summer monsoon rainfall over East Asia. Mon. Wea. Rev., 116, 18–37, https://doi.org/10.1175/1520-0493(1988)116%3c0018:SAICOS%3e2.0.CO;2.
Lee, J. H., and P. Y. Julien, 2017: Influence of the El Niño/Southern Oscillation on South Korean streamflow variability. Hydrol. Processes, 31, 2162–2178, https://doi.org/10.1002/hyp.11168.
Lee, J.-Y., and B. Wang, 2014: Future change of global monsoon in the CMIP5. Climate Dyn., 42, 101–119, https://doi.org/10.1007/s00382-012-1564-0.
Lee, T., and M. J. McPhaden, 2010: Increasing intensity of El Niño in the central-equatorial Pacific. Geophys. Res. Lett., 37, L14603, https://doi.org/10.1029/2010GL044007.
Leith, C. E., 1973: The standard error of time-averaged estimates of climatic means. J. Appl. Meteor., 12, 1066–1069, https://doi.org/10.1175/1520-0450(1973)012%3c1066:TSEOTA%3e2.0.CO;2.
Li, Z., S. Yang, B. He, and C. Hu, 2016: Intensified springtime deep convection over the South China Sea and the Philippine Sea dries Southern China. Sci. Rep., 6, 30470, https://doi.org/10.1038/srep30470.
Liu, X. H., and R. Q. Ding, 2007: The relationship between the spring Asian atmospheric circulation and the previous winter Northern Hemisphere annular mode. Theor. Appl. Climatol., 88, 71–81, https://doi.org/10.1007/s00704-006-0231-y.
Liu, Y., and Coauthors, 2017: Recent enhancement of central Pacific El Niño variability relative to last eight centuries. Nat. Commun., 8, 15386, https://doi.org/10.1038/ncomms15386.
Oki, T., and S. Kanae, 2006: Global hydrological cycles and world water resources. Science, 313, 1068–1072, https://doi.org/10.1126/science.1128845.
Pan, W., J. Mao, and G. Wu, 2013: Characteristics and mechanism of the 10–20-day oscillation of spring rainfall over southern China. J. Climate, 26, 5072–5087, https://doi.org/10.1175/JCLI-D-12-00618.1.
Panofsky, H. A., and G. W. Brier, 1958: Some Application of Statistics to Meteorology. Pennsylvania State University, 224 pp.
Qu, J., D. Gong, R. Mao, J. Yang, and S. Li, 2017: Possible influence of Arctic oscillation on precipitation along the East Asian rain belt during boreal spring. Theor. Appl. Climatol., 130, 487–495, https://doi.org/10.1007/s00704-016-1900-0.
Rayner, N. A., D. E. Parker, E. B. Horton, C. K. Folland, L. V. Alexander, D. P. Rowell, E. C. Kent, and A. Kaplan, 2003: Global analyses of sea surface temperature, sea ice, and night marine air temperature since the late nineteenth century. J. Geophys. Res., 108, 4407, https://doi.org/10.1029/2002JD002670.
Ruiz-Barradas, A., and S. Nigam, 2018: Hydroclimate variability and change over the Mekong River basin: Modeling and predictability and policy implications. J. Hydrometeor., 19, 849–869, https://doi.org/10.1175/JHM-D-17-0195.1.
Scoccimarro, E., S. Gualdi, A. Bellucci, M. Zampieri, and A. Navarra, 2013: Heavy precipitation events in a warmer climate: Results from CMIP5 models. J. Climate, 26, 7902–7911, https://doi.org/10.1175/JCLI-D-12-00850.1.
Seo, J., W. Choi, D. Youn, D.-S. R. Park, and J. Y. Kim, 2013: Relationship between the stratospheric quasi-biennial oscillation and the spring rainfall in the western North Pacific. Geophys. Res. Lett., 40, 5949–5953, https://doi.org/10.1002/2013GL058266.
Stephan, C. C., N. P. Klingaman, P. L. Vidale, A. G. Turner, M.-E. Demory, and L. Guo, 2018: A comprehensive analysis of coherent rainfall patterns in China and potential drivers. Part I: Interannual variability. Climate Dyn., 50, 4405–4424, https://doi.org/10.1007/s00382-017-3882-8.
Sun, C., and S. Yang, 2012: Persistent severe drought in southern China during winter–spring 2011: Large-scale circulation patterns and possible impacting factors. J. Geophys. Res., 117, D10112, https://doi.org/10.1029/2012JD017500.
Sun, F., and J.-Y. Yu, 2009: A 10–15-yr modulation cycle of ENSO intensity. J. Climate, 22, 1718–1735, https://doi.org/10.1175/2008JCLI2285.1.
Tao, L., Y. Hu, and J. Liu, 2016: Anthropogenic forcing on the Hadley circulation in CMIP5 simulations. Climate Dyn., 46, 3337–3350, https://doi.org/10.1007/s00382-015-2772-1.
Timmermann, A., and Coauthors, 2018: El Niño–Southern Oscillation complexity. Nature, 559, 535–545, https://doi.org/10.1038/s41586-018-0252-6.
Torrence, C., and G. P. Compo, 1998: A practical guide to wavelet analysis. Bull. Amer. Meteor. Soc., 79, 61–78, https://doi.org/10.1175/1520-0477(1998)079%3c0061:APGTWA%3e2.0.CO;2.
Trenberth, K. E., and D. P. Stepaniak, 2001: Indices of El Niño evolution. J. Climate, 14, 1697–1701, https://doi.org/10.1175/1520-0442(2001)014%3c1697:LIOENO%3e2.0.CO;2.
Ueda, H., A. Iwai, K. Kuwako, and M. E. Hori, 2006: Impact of anthropogenic forcing on the Asian summer monsoon as simulated by eight GCMs. Geophys. Res. Lett., 33, L06703, https://doi.org/10.1029/2005GL025336.
Vu, T. M., and A. K. Mishra, 2016: Spatial and temporal variability of standardized precipitation index over Indochina Peninsula. Cuad. Invest. Geogr., 42, 221–232, https://doi.org/10.18172/cig.2928.
Wang, B., R. Wu, and X. Fu, 2000: Pacific–East Asian teleconnection: How does ENSO affect East Asian climate? J. Climate, 13, 1517–1536, https://doi.org/10.1175/1520-0442(2000)013%3c1517:PEATHD%3e2.0.CO;2.
Wei, W., R. Zhang, M. Wen, B.-J. Kim, and J.-C. Nam, 2015: Interannual variation of the South Asian high and its relation with Indian and East Asian summer monsoon rainfall. J. Climate, 28, 2623–2634, https://doi.org/10.1175/JCLI-D-14-00454.1.
Wu, R., Z.-Z. Hu, and B. P. Kirtman, 2003: Evolution of ENSO-related rainfall anomalies in East Asia. J. Climate, 16, 3742–3758, https://doi.org/10.1175/1520-0442(2003)016%3c3742:EOERAI%3e2.0.CO;2.
Wu, X., and J. Mao, 2016: Interdecadal modulation of ENSO-related spring rainfall over South China by the Pacific decadal oscillation. Climate Dyn., 47, 3203–3220, https://doi.org/10.1007/s00382-016-3021-y.
Wu, X., and J. Mao, 2018: Spatial and interannual variations of spring rainfall over eastern China in association with PDO–ENSO events. Theor. Appl. Climatol., 134, 935–953, https://doi.org/10.1007/s00704-017-2323-2.
Xin, X., R. Yu, T. Zhou, and B. Wang, 2006: Drought in late spring of South China in recent decades. J. Climate, 19, 3197–3206, https://doi.org/10.1175/JCLI3794.1.
Xu, X., C. Lu, X. Shi, and Y. Ding, 2010: Large-scale topography of China: A factor for the seasonal progression of the Meiyu rainband? J. Geophys. Res., 115, D02110, https://doi.org/10.1029/2009JD012444.
Yang, F., and K.-M. Lau, 2004: Trend and variability of China precipitation in spring and summer: Linkage to sea-surface temperatures. Int. J. Climatol., 24, 1625–1644, https://doi.org/10.1002/joc.1094.
Yang, Q., Z. Ma, and B. Xu, 2017: Modulation of monthly precipitation patterns over East China by the Pacific decadal oscillation. Climatic Change, 144, 405–417, https://doi.org/10.1007/s10584-016-1662-9.
Yeh, S.-W., X. Wang, C. Wang, and B. Dewitte, 2015: On the relationship between the North Pacific climate variability and the central Pacific El Niño. J. Climate, 28, 663–677, https://doi.org/10.1175/JCLI-D-14-00137.1.
Ying, K., X. Zheng, T. Zhao, C. S. Frederiksen, and X.-W. Quan, 2017: Identifying the predictable and unpredictable patterns of spring-to-autumn precipitation over eastern China. Climate Dyn., 48, 3183–3206, https://doi.org/10.1007/s00382-016-3258-5.
Yoo, S.-H., C.-H. Ho, S. Yang, H.-J. Choi, and J.-G. Jhun, 2004: Influences of tropical western and extratropical Pacific SST on East and Southeast Asian climate in the summers of 1993–94. J. Climate, 17, 2673–2687, https://doi.org/10.1175/1520-0442(2004)017%3c2673:IOTWAE%3e2.0.CO;2.
Zhai, P., X. Zhang, H. Wan, and X. Pan, 2005: Trends in total precipitation and frequency of daily precipitation extremes over China. J. Climate, 18, 1096–1108, https://doi.org/10.1175/JCLI-3318.1.