1. Introduction
The tropical cyclone (TC) activity in the western North Pacific (WNP), the region of the most frequent TC genesis in the world, exhibits a strong interannual fluctuation that is affected significantly by El Niño–Southern Oscillation (ENSO; e.g., Chan 2000; Chia and Ropelewski 2002; Wang and Chan 2002). The mean TC genesis location (MGL) shifts substantially southeastward (northwestward) from the climatological location during the El Niño (La Niña)–developing boreal summer. The southeastward shift of MGL substantially prolongs the TC’s lifetime, enhances the peak intensity, and enlarges the coverage of influenced regions during the El Niño–developing summer (Camargo and Sobel 2005; Camargo et al. 2007a,b). The reversed effect tends to occur in La Niña–developing summers. The changes associated with TCs mentioned above have significant impacts on the weather and climate systems in the WNP. It is therefore important to understand the factors influencing the MGL.
A strong El Niño event occurred in the boreal summer of 2015 and had a strong effect on the shifting of the MGL. Whereas the TC genesis number in the WNP was approximately normal during June–August (JJA) 2015, the MGL shifted unprecedentedly eastward to 150°E, approximately 10° longitude farther to the east compared with the El Niño event in 1997 (Fig. 1a). This eastward shift prolonged the average lifetime of TCs in JJA 2015 by approximately 65 h compared to that in JJA 1997. One extreme example is Supertyphoon Soudelor in August 2015. Its long lifetime (348 h, approximately 15 days) might have contributed to the extreme intensity, with minimum pressure and maximum wind speed reaching 900 hPa and 115 kt (1 kt ≈ 0.51 m s−1), respectively. The Niño-3.4 sea surface temperature (SST) index in summer 2015 was comparable to, but not stronger than, that in 1997. Why, therefore, did the 2015 El Niño exert such a distinctive influence on the MGL in the WNP during JJA?
(a) JJA MGL in the WNP for each year since 1981. El Niño, La Niña, and other years are marked in red, blue, and green circles, respectively, and the plus symbols indicate the MGL of the corresponding groups. (c) Correlation coefficients between SST and TC distance. Here the TC distance is defined as the distance between the reference point [marked by ⊕ in (a)] and MGL in each summer. The PMM-like and ENSO-like SSTAs represent the subtropical and tropical SSTAs, respectively. (e),(g) As in (c), but for correlation coefficients between SST and the longitude and latitude of MGL. Only the signals exceeding the 95% confidence level are plotted. (b),(d),(f),(h) As in (a),(c),(e),(g), but for the MGL in the ENP.
Citation: Journal of Climate 31, 8; 10.1175/JCLI-D-17-0504.1
In addition to the WNP, the MGL in the eastern North Pacific (ENP) during JJA 2015 also shifted unprecedentedly westward toward the central Pacific, more than 20° longitude farther than that in 1997 (Fig. 1b), and resulted in an unusual TC landfall in Hawaii. Murakami et al. (2017) proposed that the subtropical North Pacific SST warming rather than the warming in the equatorial eastern Pacific contributed to the westward shift of MGL.
The observation reveals that the subtropical eastern North Pacific SST anomaly (SSTA) was the major distinction between the two strong El Niño events of 1997 and 2015 (Murakami et al. 2017): a positive SSTA extending from the California coastline southwestward to the equatorial central Pacific in summer 2015. This northeast–southwest-orientated SSTA resembles the Pacific meridional mode (PMM; Chiang and Vimont 2004; Chang et al. 2007). The 2015 El Niño is a mixed-type El Niño (Paek et al. 2017), that is, a combination of the central Pacific El Niño (Kao and Yu 2009) and the PMM-like subtropical SSTA. Whereas the effect of the PMM SSTA on the TC activity in the WNP had been documented in Zhang et al. (2016), its impact was estimated without considering other contributing factors, such as the ENSO SSTA. The ENSO and PMM SSTAs tend to coexist in boreal summer with significant correlation of 0.61. This was also the case in the 2015 event. The relative effects of the ENSO and PMM SSTAs on the TC activity in the North Pacific have not been thoroughly investigated. In this study, we demonstrate that the MGL in the WNP is predominantly determined by two major SSTAs in the Pacific: the ENSO-like tropical SSTA (located in the equatorial central-eastern Pacific) and the PMM-like subtropical SSTA (located in the subtropical eastern North Pacific). The ENSO-like SSTA controls the southward shift of MGL but the eastward shift is primarily induced by the PMM-like SSTA. Both the ENSO-like and PMM-like SSTAs reached unusually high values in the summer of 2015 and resulted in the extremely southeastward shift of the MGL in the WNP.
2. Data, model, and analysis procedure
The observational data used in this study are monthly atmospheric fields from the NCEP–NCAR reanalysis (Kalnay et al. 1996) and SSTs from the National Oceanic and Atmospheric Administration (NOAA) Extended Reconstructed Sea Surface Temperature, version 4 (Huang et al. 2015). The WNP and ENP tropical cyclone data are from the Joint Typhoon Warning Center (JTWC) and Unisys [http://weather.unisys.com/hurricane/; originally from the National Hurricane Center best-track northeast and north-central Pacific hurricane database (HURDAT2)], respectively. The TC genesis location is defined as the point when the maximum sustained wind speed of a tropical cyclone reaches 34 kt. The analyses applied in this study are repeated based on the genesis location of tropical depressions (maximum sustained wind speed greater than 17 kt). The results based on two definitions are essentially the same, indicating the insensitivity of our results to the definition of TC genesis.
The atmospheric general circulation model (AGCM) used was ECHAM5 (Roeckner et al. 2003), with spectral T42 horizontal resolution and 31 vertical sigma levels. ECHAM5 was used in the numerical experiments to infer the relative influence of the ENSO-like and PMM-like SSTAs on the MGL. Although the low-resolution ECHAM5 is not designed to simulate TC activity, it captures the changes well in the large-scale circulation that modulates the MGL and TC track in response to the SST anomaly associated with El Niño (Hong et al. 2011). A similar approach is taken in this study to evaluate the large-scale circulation features (e.g., low-level circulation, vertical motion, and midlevel moisture) and the TC genesis potential index (GPI; Emanuel and Nolan 2004; Camargo et al. 2007b) that are essential in creating a favorable environment for TC genesis and inferring the relative influences of the ENSO-like and PMM-like SSTAs.
3. SSTs correlated with the mean TC genesis location in the WNP
The JJA MGL in the WNP for each year since 1981 is marked in Fig. 1a. The mean location during the El Niño–developing summer (marked with red circles) shifts remarkably southward and eastward from the climatological mean location. Conversely, the MGL shifts northward and westward (marked with blue circles) during the La Niña–developing summer. The impact of ENSO on MGL is consistent with Chia and Ropelewski (2002). ENSO also exerts a significant influence on the MGL in the ENP: a westward and eastward shift during the El Niño– and La Niña–developing summers (Fig. 1b), respectively, as reported in previous studies (Whitney and Hobgood 1997; Irwin and Davis 1999; Collins and Mason 2000; Chu and Zhao 2007). In contrast to the WNP, the south–north shift of MGL in the ENP is not as significant.
A comparison of two strong El Niño events in 2015 and 1997 indicates that the southward shift of MGL in the WNP was approximately equal, but the MGL in 2015 is shifted farther eastward by approximately 10° longitude more than that in 1997. Because the Niño-3.4 (5°S–5°N, 120°–170°W) SSTAs in 1997 and 2015 were nearly the same (Fig. 2), SSTAs other than the Niño-3.4 likely contributed to the eastward shift of the MGL.
Spatial distribution of JJA mean SSTA (K) in (a) 1997 and (b) 2015. (c) Time series of normalized JJA mean PMM-like (red dashed line) and ENSO-like (blue dashed line) indices. The black contour in (a),(b) denotes the domain of the PMM-like SSTA. The PMM-like and ENSO-like indices are defined as the normalized area-averaged SSTAs in the corresponding regions marked in Fig. 1c.
Citation: Journal of Climate 31, 8; 10.1175/JCLI-D-17-0504.1
The correlation between the SSTA and the MGL was calculated to reveal the relationship. Here, a TC distance is defined as the distance from the JJA MGL to a reference point located on the East Asian coast (marked with a plus symbol inside a circle in Fig. 1a). The correlation map between the TC distance and SSTA shown in Fig. 1c depicts that the TC distance from East Asian coast is positively correlated with an SSTA in the tropical central Pacific and the subtropical ENP. Both positive central Pacific and ENP SSTAs seem to be responsible for the southeastward shift of MGL. Other reference points were chosen to examine the sensitivity of the correlation map (figure not shown). The relationship between the TC distance and SST was found relatively insensitive to the exact location of reference points. TC distance was further separated into longitude and latitude to estimate the relative contribution of zonal and meridional shifts of the MGL. The zonal shift is highly correlated (0.73 correlation coefficient) with the southwest–northeast-oriented subtropical SSTA in the ENP, which resembles the PMM (Fig. 1e). By contrast, the meridional shift is significantly correlated (0.96 correlation coefficient) with the Niño-3.4 SSTA (Fig. 1g). Based on this result, the SSTA in the ENP shown in Fig. 1c was divided into two parts: the tropical part (termed ENSO-like) and the subtropical part (termed PMM-like). Figure 1 also reveals that other regions, such as Niño-1+2, do not has significant effect on the meridional and zonal shifts of MGL in the WNP and ENP. Time series of ENSO-like and PMM-like SSTAs show that both indices reached unusually high values in JJA 2015 (Fig. 2c), which might have jointly resulted in the unusually southward and eastward shift of WNP MGL in 2015. The same analysis was applied to the MGL in the ENP. The results shown in Figs. 1b,d,f,h reveal that the PMM-like SSTA, as it does in the WNP, determines the east–west shift of TC distance in the ENP. This result is consistent with the result of Murakami et al. (2017), who argued that the subtropical warm Pacific rather than the tropical Pacific resulted in the unusually westward shift of the TC genesis in ENP in 2015.
The above results are derived from JJA. The same analysis applied to September–November (SON) reveals that the eastward (westward) shift of MGL in the WNP (ENP) during the El Niño yeas is much weaker than that in JJA (Fig. 3). Furthermore, the eastward shift of MGL in 2015 compared with 1997 was not observed in SON. The correlation analysis indicates that the effect of the PMM-like SSTA on the MGL is insignificant in SON (Fig. 3). By contrast, the ENSO-like SST dominates the MGL in SON. The exact reason for the seasonal dependence is not clear. We speculate that it may be rooted in the characteristics of the PMM SSTA, which peaks in the boreal spring and decays rapidly after the boreal summer (Chiang and Vimont 2004). By contrast, the ENSO-like SSTA usually becomes stronger in SON during the developing phase of ENSO and therefore maintains stronger influences on MGL.
4. Effects of the ENSO-like and PMM-like SSTAs on the mean TC genesis location
Regressions of the large-scale circulations on the ENSO-like and PMM-like SSTAs were computed to investigate the effect of each SSTA pattern on the MGL. Because the ENSO-like and PMM-like SSTAs are significantly correlated (0.67), partial correlation was computed. In the correlation map for the PMM-like SSTA (Figs. 4a,b), an east–west dipole structure, that is, a low-level cyclonic (anticyclonic) anomaly associated with a negative (positive) vertical zonal wind shear anomaly, is clearly seen in the subtropical central (western North) Pacific. It follows that the positive PMM-like SSTA has an effect on shifting the favorable condition (i.e., weaker vertical shear and cyclonic circulation) for TC genesis toward the tropical central Pacific. Further calculations reveal that the east–west dipole circulation has been enhanced and become more significant since the 1990s (Fig. 5a). The east–west dipole circulation anomaly creates an east–west overturning circulation in the subtropical North Pacific (Fig. 5b). The associated ascending (descending) motion likely further enhances (suppresses) the TC activity in the central (western) Pacific and therefore shifts the MGL in the WNP eastward. By contrast, the ENSO-like SSTA exerts an impact on the MGL through a pair of low-level cyclonic circulation anomalies covering almost the entire western Pacific, resembling a typical Gill-type response to the equatorial central–eastern positive SSTA (Fig. 4d). The pair of cyclone anomalies straddling the equator is associated with a negative vertical zonal wind shear anomaly covering the equatorial western–central Pacific between 15°N and 15°S (Fig. 4c). Regions of anomalies of vertical zonal wind shear with reversed signs are located in the poleward side of the equatorial anomaly. The pattern suggests that the favorable environment for TC genesis shifts southward (northward) under the influence of positive (negative) ENSO-like SSTA.
Partial correlation coefficients between the PMM-like SSTA and (a) vertical zonal wind shear (between 200- and 850-hPa zonal winds) and (b) 850-hPa streamfunction during JJA 1991–2016. (c),(d) As in (a),(b), but for the ENSO-like SSTA. (e) Vertical zonal wind shear (shaded; m s−1) and 200-hPa wind anomalies in JJA 2015 and (f) 850-hPa streamfunction (shaded; 105 m2 s−1) and wind anomalies in JJA 2015. The black contours in (a)–(d) denote the domains for the PMM-like and ENSO-like SSTAs. The period 1991–2016 is chosen because the effect of the PMM-like SSTA on the large-scale circulation in the WNP has become more significant since the 1990s.
Citation: Journal of Climate 31, 8; 10.1175/JCLI-D-17-0504.1
(a) The 19-yr sliding partial correlation coefficient between the PMM-like SSTA and the 850-hPa streamfunction averaged over 15°–25°N, 110°–140°E (red line) and vertical zonal wind shear averaged over 5°–15°N, 110°–140°E (blue line). The horizontal dashed line indicates the threshold of correlation coefficient exceeding the 95% confidence level. (b) Cross section of partial correlation between the PMM-like SSTA and vertical velocity (ω; shaded) and zonal wind (contours) anomaly averaged over 10°–20°N.
Citation: Journal of Climate 31, 8; 10.1175/JCLI-D-17-0504.1
A comparison with the correlation map indicates that the vertical zonal wind shear anomaly in 2015 (Fig. 4e) is a combination of Figs. 4a and 4c and can be attributed to the compounding effect of the ENSO-like and PMM-like SSTAs. More specifically, the negative vertical zonal wind shear in the tropical central Pacific, which is associated with the ENSO-like SSTA, is enhanced and broadened by the negative vertical zonal wind shear related to the PMM-like SSTA in the subtropical North Pacific. In addition, the observed anticyclone and positive vertical zonal wind shear anomalies in the western Philippine Sea and South China Sea (Figs. 4e,f), which are not correlated with the ENSO-like SSTA (Figs. 4c,d), carry the footprint of the positive PMM-like SSTA (Figs. 4a,b). The observed positive PMM-like SSTA induces an east–west dipole of vertical zonal wind shear anomaly in the subtropical North Pacific, which in turn provides a favorable (unfavorable) environment for TC genesis in the central (western) Pacific and shifts the TC genesis location eastward. The results discussed above are consistent with Zhang et al. (2016), in which the effect of PMM-like SST on the vertical zonal wind shear was suggested as a responsible mechanism for the shift of TC genesis position.
The midlevel relative humidity correlated with the PMM-like and ENSO-like SSTs is further analyzed to investigate the role of thermodynamic effects on the TC genesis location. As shown in Fig. 6a, the PMM-like SST is negatively (positively) correlated with relative humidity in the tropical western (eastern) Pacific, and the correlation in the western Pacific was especially pronounced (Fig. 6). By contrast, the positive (negative) ENSO-like SSTA is associated with a positive (negative) relative humidity anomaly in the tropical central–western (eastern) Pacific. In other words, the major effect of the positive PMM-like SSTA is to reduce the relative humidity in the Philippine Sea, whereas the positive ENSO-like SSTA acts to enhance the relative humidity in the equatorial western–central Pacific between 140°E and 170°W. A combined effect of both SSTAs is therefore to shift the maximum relative humidity region southeastward. A comparison with the observed relative humidity anomalies in JJA 2015 yields the information about the relative influences of the ENSO-like and PMM-like SSTAs. The negative midlevel relative humidity anomaly in the Philippine Sea in 2015 was apparently associated with the positive PMM-like SST. The observed positive humidity anomaly in central–western Pacific, on the other hand, was primarily correlated with ENSO-like SSTAs. Figure 6 also indicates that the observed eastward extension of the positive humidity anomaly from the central to eastern Pacific (approximately in 90°W) resulted from the PMM-like SST. It is evident (Figs. 4 and 6) that both dynamic and thermodynamic effects of the PMM-like and ENSO-like SSTAs contributed to the unusual southeastward shift of MGL in JJA 2015.
As in Fig. 4, but for the partial correlation coefficients between the 700-hPa relative humidity anomaly and (a) PMM-like SSTA and (b) ENSO-like SSTA . (c) The observed 700-hPa relative humidity anomaly (%) in JJA 2015.
Citation: Journal of Climate 31, 8; 10.1175/JCLI-D-17-0504.1
Whereas the ENSO-like SSTAs in the summers of 1997 and 2015 were approximately equal, the PMM-like SSTA in 2015 was approximately twice as large as in 1997 (Fig. 2c). The similarity in the ENSO-like SSTAs explains why the TC genesis locations in 1997 and 2015 both shifted southward. However, the distinctive eastward shift in 2015 was likely induced by the PMM-like SSTA. In view of the distinctly different influence of the ENSO-like and PMM-like SSTAs, we hypothesize that the extreme southeastward shift of MGL in summer 2015 can be attributed to the compounding effect of the two SST anomaly patterns.
5. Numerical experiments
To test our hypothesis, we conduct a series of numerical experiments using ECHAM5 to simulate the atmospheric circulation in JJA 2015 under the forcing of observed SSTAs. Each experiment consists of 10 member simulations from January to December, with the observed initial conditions from 1 to 10 January of arbitrary years. Three sets of experiment are conducted, namely, control (CTL), ENSO-like, and PMM-like experiments. The CTL experiment is forced with climatological monthly SST. The ENSO-like and PMM-like experiments are forced by observed monthly SST in JJA 2015 in the designated ENSO-like and PMM-like regions (Fig. 1c), respectively, and climatological monthly SST elsewhere. The description of experiment designs is presented in Table 1.
Design of numerical experiments.
The black contours in Figs. 7a–d indicate the domain of the prescribed SSTA in the PMM-like and ENSO-like experiments. The SSTA is added to the observed climatological monthly SST to drive the simulations. The differences between the aforementioned experiments and the control experiment are compared with observed thermodynamic and dynamic anomalies to attribute the relative effect of the prescribed SSTA on the large-scale factors that serve as the background for TC genesis. The simulated 850-hPa streamfunction and vertical zonal wind shear are shown in Fig. 7. Low-level cyclonic (anticyclonic) anomalies associated with negative (positive) vertical zonal wind shear in the central (western) Pacific, respectively, are simulated in the PMM-like SSTA experiment (Figs. 7a,b). In addition, the observed east–west overturning circulation anomaly (averaged over 10°–20°N; Fig. 8c), ascending in the central Pacific and descending in the western Pacific, is also realistically captured (Fig. 8a).
As in Figs. 4e,f, but for the (a),(b) PMM-like and (c),(d) ENSO-like simulations. The black contours denote the prescribed SSTA domains in the respective experiments. Note that the black contours are slightly different from that in Fig. 4 because of the difference in horizontal resolution between model and observation.
Citation: Journal of Climate 31, 8; 10.1175/JCLI-D-17-0504.1
Longitude–pressure (hPa) cross sections of the 10°–20°N-averaged overturning circulation anomaly for the (a) PMM-like and (b) ENSO-like experiments and (c) the observation in JJA 2015. Vectors indicate overturning circulation and shading denotes vertical motion (Pa s−1). Only the signals exceeding the statistical Student’s t test at the 5% significance level are plotted in (a),(b).
Citation: Journal of Climate 31, 8; 10.1175/JCLI-D-17-0504.1
The response to the ENSO-like SSTA features a pair of huge cyclonic circulation anomalies straddling the equator in the western and central Pacific, and a pair of weaker anticyclonic circulation anomalies in the eastern Pacific (Fig. 7d). South–north dipole-like vertical zonal wind shear anomalies (Fig. 7c), resembling the observation, are also simulated. The negative zonal wind vertical shear in the central Pacific induced by the PMM-like SSTA acts to expand northward the negative shear induced by the ENSO-like SSTA. Besides the dynamic processes (e.g., wind shear and low-level circulation), the thermodynamic effect (viz., 700-hPa relative humidity anomalies) of the PMM-like and ENSO-like SSTAs is investigated. A comparison between the simulations and observation indicates that the PMM-like experiment realistically simulates the observed negative 700-hPa relative humidity anomaly in the WNP (figure not shown). And the PMM-like and ENSO-like SSTAs jointly contribute to the observed positive 700-hPa humidity anomalies in the subtropical and tropical central Pacific, respectively. Numerical experiments confirm that the dynamic and thermodynamic processes induced by the PMM-like and ENSO-like SSTAs jointly contribute to the observed southeastward shift of MGL in JJA 2015.
The combined effect of dynamic and thermodynamic processes on MGL is demonstrated by GPI in Fig. 9. The observed zonally elongated positive GPI anomaly in the subtropical North Pacific (approximately from 160°E to 120°W) and the negative GPI anomaly in the South China Sea and Philippine Sea in JJA 2015 are realistically simulated with the PMM-like SSTA as forcing (Figs. 9a,b). By contrast, the observed positive GPI anomaly in the WNP (25°N, 150°E) and the negative GPI anomaly in the extratropical North Pacific (zonally elongated from south of Japan to the date line) was successfully captured in the ENSO-like experiment (Figs. 9a,c). The GPI analysis clearly indicates that the PMM-like SSTA contributed to the weakened (enhanced) TC genesis potential in the far western Pacific (the eastern part of the subtropical WNP) in JJA 2015, whereas the ENSO-like SSTA primarily contributed to the southeastward shift of the genesis potential region to the tropical central Pacific. This result again confirms our hypothesis that the PMM-like and ENSO-like SSTAs led to the eastward and southward shifts of MGL, respectively.
Spatial distribution of GPI anomaly for the (a) PMM-like and (b) ENSO-like experiments and (c) the observation in JJA 2015. Only the signals exceeding the statistical Student’s t test at the 5% significance level are plotted in (a),(b).
Citation: Journal of Climate 31, 8; 10.1175/JCLI-D-17-0504.1
6. Summary and discussion
It is well known that the MGL in the WNP shifts significantly southeastward from the climatological location during the El Niño–developing summer, and substantially modulates the lifetime, intensity, and track of TCs. The 2015/16 event was identified as a strong El Niño with the Niño-3.4 SST strength equivalent to that in 1997. However, the MGL over the WNP in 2015 exhibited an unprecedented eastward shift by approximately 10° longitude relative to that in 1997. This eastward shift resulted in the average lifetime of TCs in JJA 2015 being approximately 3 days longer than that in JJA 1997. The key SSTA feature determining this unusual eastward shift of MGL in JJA 2015 and the responsible mechanisms are identified in this study. The main findings are described as follows.
Statistical analysis reveals that the MGL in the WNP is significantly affected by both the ENSO-like and PMM-like SSTAs. The ENSO-like SSTA determines the south–north shift of MGL, whereas the east–west shift of MGL is primarily controlled by the PMM-like SSTA. A similar result for the MGL in the ENP is identified, except the south–north shift of MGL is much weaker than the east–west shift.
Observations and numerical experiments yield consistent characteristics of the basinwide circulation anomaly that explains the distinctly different influences of the ENSO-like and PMM-like SSTAs on the MGL. The positive PMM-like SSTA generates an east–west overturning circulation anomaly and an ascending (descending) anomaly accompanied with negative (positive) vertical zonal wind shear anomaly and positive (negative) midlevel relative humidity anomaly in the central (western) Pacific, which likely shift the MGL eastward. By contrast, the positive ENSO-like SSTA induces negative and positive vertical zonal wind shear anomalies in the tropical and subtropical WNP, respectively, and cyclonic circulation and positive relative humidity anomalies in the tropical central Pacific, and therefore tends to shift the MGL southward.
Both positive ENSO-like and PMM-like SSTAs in the summer of 2015 broke the high temperature records for the 1980–2015 period. The extreme ENSO-like and PMM-like SSTAs in 2015 jointly led to the record-breaking southeastward shift of the MGL in the WNP. But it was the PMM-like SSTA that contributed to the much farther eastward shift of MGL in 2015 than in 1997.
The influence of the PMM SSTA on the TC activity in the WNP was documented in Zhang et al. (2016). The ENSO-like and PMM-like SSTAs are significantly correlated and tend to coexist in boreal summer but their relative contribution to the MGL has not been adequately investigated. This study, based on statistical analysis and numerical simulation, explores the relative influence of ENSO and PMM SSTAs on the TC activity and the seasonal dependence. The effect of the PMM-like and ENSO-like SSTAs on the MGL, which is clearly revealed in statistical analysis, is not explicitly simulated. Instead, we focus our discussion on the SSTA-forced variation of large-scale circulations, such as the subtropical anticyclone, vertical wind shear, and relative humidity that strongly modulate the MGL. We demonstrate that the positive ENSO-like SSTA did not drive an anomalous large-scale subsidence in the WNP as observed in JJA 2015. It is the positive PMM-like SSTA that forced a west–east overturning circulation anomaly, with descending (ascending) in the subtropical western (central-eastern) North Pacific, and induced an environment favorable for the eastward shift of the MGL in the WNP. A calculation of GPI based on numerical simulations also supports the differentiated effects of the ENSO-like and PMM-like SSTAs. This finding supports the hypothesis that the extreme positive PMM-like SSTA in the summer of 2015 caused the unprecedented eastward shift of the MGL in the WNP.
The effect of the positive PMM-like SSTA is similar to the impact of warm SST in the eastern North Pacific on the unusual absence of TC in the WNP in August 2014 (Hong et al. 2016). Additionally, this large-scale subsidence substantially enhances the WNP subtropical high, which plays a crucial role in the interannual variation of East Asian summer monsoons (Chang et al. 2000). A recent study suggested that the PMM-like SSTA exhibits a significant interdecadal fluctuation (Sullivan et al. 2016). This study also identifies the enhanced statistical relationship between the PMM-like SSTA and the circulation and vertical zonal wind shear in the WNP since the late 1990s (Fig. 5). The possible influence of the PMM-like SSTA on the interdecadal influence of TC activity in the WNP and East Asian summer monsoons is an intriguing topic that deserves further attention.
In view of the warming trend in both the tropical and subtropical Pacific, there is a need to more accurately identify the relative influences of the ENSO-like and PMM-like SSTAs. It follows that the climate models, which have been used to project future changes in the large-scale circulation and TC activity in the Pacific, need to be able to robustly simulate the relative effects of the ENSO-like (Timmermann et al. 1999; Yeh et al. 2009) and PMM-like SSTAs (Manganello et al. 2014; Nakamura et al. 2017; Murakami et al. 2017) to narrow down the uncertainty.
Acknowledgments
We are grateful to three anonymous reviewers for their valuable comments. The observational data used in this study are from the NOAA’s Earth System Research Laboratory, NOAA/NCEP, and NCAR. This study is supported by the Ministry of Science and Technology, Taiwan, under Grants MOST-106-2111-M-001-005, and MOST-105-2111-M845-002. The National Center for High-Performance Computing provided computing resources for this study.
REFERENCES
Camargo, S. J., and A. H. Sobel, 2005: Western North Pacific tropical cyclone intensity and ENSO. J. Climate, 18, 2996–3006, https://doi.org/10.1175/JCLI3457.1.
Camargo, S. J., A. W. Robertson, S. J. Gaffney, P. Smyth, and M. Ghil, 2007a: Cluster analysis of typhoon tracks. Part II: Large-scale circulation and ENSO. J. Climate, 20, 3654–3676, https://doi.org/10.1175/JCLI4203.1.
Camargo, S. J., K. A. Emanuel, and A. H. Sobel, 2007b: Use of a genesis potential index to diagnose ENSO effects on tropical cyclone genesis. J. Climate, 20, 4819–4834, https://doi.org/10.1175/JCLI4282.1.
Chan, J. C. L., 2000: Tropical cyclone activity over the western North Pacific associated with El Niño and La Niña events. J. Climate, 13, 2960–2972, https://doi.org/10.1175/1520-0442(2000)013<2960:TCAOTW>2.0.CO;2.
Chang, C.-P., Y. Zhang, and T. Li, 2000: Interannual and interdecadal variations of the East Asian summer monsoon rainfall and tropical Pacific SSTs. Part I: Roles of the subtropical ridge. J. Climate, 13, 4310–4325, https://doi.org/10.1175/1520-0442(2000)013<4310:IAIVOT>2.0.CO;2.
Chang, P., L. Zhang, R. Saravanan, D. J. Vimont, J. C. H. Chiang, L. Ji, H. Seidel, and M. K. Tippett, 2007: Pacific meridional mode and El Niño–Southern Oscillation. Geophys. Res. Lett., 34, L16608, https://doi.org/10.1029/2007GL030302.
Chia, H. H., and C. F. Ropelewski, 2002: The interannual variability in the genesis location of tropical cyclones in the northwest Pacific. J. Climate, 15, 2934–2944, https://doi.org/10.1175/1520-0442(2002)015<2934:TIVITG>2.0.CO;2.
Chiang, J. C. H., and D. J. Vimont, 2004: Analogous Pacific and Atlantic meridional modes of tropical atmosphere–ocean variability. J. Climate, 17, 4143–4158, https://doi.org/10.1175/JCLI4953.1.
Chu, P.-S., and X. Zhao, 2007: A Bayesian regression approach for predicting seasonal tropical cyclone activity over the central North Pacific. J. Climate, 20, 4002–4013, https://doi.org/10.1175/JCLI4214.1.
Collins, J. M., and I. M. Mason, 2000: Local environmental conditions related to seasonal tropical cyclone activity in the northeast Pacific basin. Geophys. Res. Lett., 27, 3881–3884, https://doi.org/10.1029/2000GL011614.
Emanuel, K. A., and D. S. Nolan, 2004: Tropical cyclone activity and the global climate system. 26th Conf. on Hurricanes and Tropical Meteorology, Miami, FL, Amer. Meteor. Soc., 10A.2, http://ams.confex.com/ams/pdfpapers/75463.pdf.
Hong, C.-C., Y.-H. Li, T. Li, and M.-Y. Lee, 2011: Impacts of central Pacific and eastern Pacific El Niños on tropical cyclone tracks over the western North Pacific. Geophys. Res. Lett., 38, L16712, https://doi.org/10.1029/2011GL048821.
Hong, C.-C., M.-Y. Lee, H.-H. Hsu, and T.-C. Chang, 2016: Compounding factors causing the unusual absence of tropical cyclones in the western North Pacific during August 2014. J. Geophys. Res. Atmos., 121, 9964–9976, https://doi.org/10.1002/2016JD025507.
Huang, B., and Coauthors, 2015: Extended Reconstructed Sea Surface Temperature version 4 (ERSST.v4). Part I: Upgrades and intercomparisons. J. Climate, 28, 911–930, https://doi.org/10.1175/JCLI-D-14-00006.1.
Irwin, R. P., III, and R. E. Davis, 1999: The relationship between the Southern Oscillation index and tropical cyclone tracks in the eastern North Pacific. Geophys. Res. Lett., 26, 2251–2254, https://doi.org/10.1029/1999GL900533.
Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77, 437–471, https://doi.org/10.1175/1520-0477(1996)077<0437:TNYRP>2.0.CO;2.
Kao, H.-Y., and J.-Y. Yu, 2009: Contrasting eastern-Pacific and central-Pacific types of ENSO. J. Climate, 22, 615–631, https://doi.org/10.1175/2008JCLI2309.1.
Manganello, J. V., and Coauthors, 2014: Future changes in the western North Pacific tropical cyclone activity projected by a multidecadal simulation with a 16-km global atmospheric GCM. J. Climate, 27, 7622–7646, https://doi.org/10.1175/JCLI-D-13-00678.1.
Murakami, H., and Coauthors, 2017: Dominant role of subtropical Pacific warming in extreme eastern Pacific hurricane seasons: 2015 and the future. J. Climate, 30, 243–264, https://doi.org/10.1175/JCLI-D-16-0424.1.
Nakamura, J., and Coauthors, 2017: Western North Pacific tropical cyclone model tracks in present and future climates. J. Geophys. Res. Atmos., 122, 9721–9744, https://doi.org/10.1002/2017JD027007.
Paek, H., J.-Y. Yu, and C. Qian, 2017: Why were the 2015/2016 and 1997/1998 extreme El Niños different? Geophys. Res. Lett., 44, 1848–1856, https://doi.org/10.1002/2016GL071515.
Roeckner, E., and Coauthors, 2003: The atmospheric general circulation model ECHAM5: Model description. Max Planck Institute for Meteorology Rep. 349, 127 pp.
Sullivan, A., J.-J. Luo, A. C. Hirst, D. Bi, W. Cai, and J. He, 2016: Robust contribution of decadal anomalies to the frequency of central-Pacific El Niño. Sci. Rep., 6, 38540, https://doi.org/10.1038/srep38540.
Timmermann, A., J. Oberhuber, A. Bacher, M. Esch, M. Latif, and E. Roeckner, 1999: Increased El Niño frequency in a climate model forced by future greenhouse warming. Nature, 398, 694–697, https://doi.org/10.1038/19505.
Wang, B., and J. C. L. Chan, 2002: How strong ENSO events affect tropical storm activity over the western North Pacific. J. Climate, 15, 1643–1658, https://doi.org/10.1175/1520-0442(2002)015<1643:HSEEAT>2.0.CO;2.
Whitney, L. D., and J. S. Hobgood, 1997: The relationship between sea surface temperatures and maximum intensities of tropical cyclones in the eastern North Pacific Ocean. J. Climate, 10, 2921–2930, https://doi.org/10.1175/1520-0442(1997)010<2921:TRBSST>2.0.CO;2.
Yeh, S.-W., J.-S. Kug, B. Dewitte, M.-H. Kwon, B. P. Kirtman, and F.-F. Jin, 2009: El Niño in a changing climate. Nature, 461, 511–514, https://doi.org/10.1038/nature08316.
Zhang, W., G. A. Vecchi, H. Murakami, G. Villarini, and L. Jia, 2016: The Pacific meridional mode and the occurrence of tropical cyclones in the western North Pacific. J. Climate, 29, 381–398, https://doi.org/10.1175/JCLI-D-15-0282.1.
