1. Introduction
The northwest Pacific (NWP) is the basin where tropical cyclones (TCs) are the most active (e.g., Chan and Shi 1996; Chia and Ropelewski 2002; Camargo and Sobel 2005; Lin et al. 2020). Each year, millions of people in Southeast and East Asia are vulnerable to the hazards and damages caused by these violent storms (e.g., Zhang et al. 2009; Peduzzi et al. 2012; Woodruff et al. 2013; Geiger et al. 2021). Thus, it is crucial to improve our understanding and prediction of NWP TC activity in order to minimize the impacts of TCs on human society (e.g., Knutson et al. 2010; Zhao and Held 2012; Wang et al. 2013; Kossin et al. 2016; Mei and Xie 2016; Yan et al. 2016; Korty et al. 2017; Camargo et al. 2019; Lee et al. 2020; Studholme et al. 2022; Zhao et al. 2022). In this study, we attempt to advance our knowledge of the spatiotemporal variations in TC genesis density over the NWP using both observations and a large ensemble (i.e., 100 members) of high-resolution atmospheric simulations.
Despite the NWP being the most active basin for TC activity, the sample size in the observations is still too small to characterize the spatiotemporal variability in TC genesis density of this basin. Accordingly, most of the previous studies focus on basinwide NWP TC number, which exhibits strong variations on both interannual and decadal time scales (e.g., Chan 1985, 2005; Chu 2004; Yan et al. 2015; Wang and Wang 2019; Zhan et al. 2019; K. Song et al. 2022). The basinwide NWP TC number stayed at high levels from the early 1960s to the early 1970s and from the late 1980s to the mid-1990s, and moved to a relatively inactive stage around the late 1990s (e.g., Matsuura et al. 2003; Hu et al. 2018; Li et al. 2022). These decadal changes in TC genesis frequency (TCGF) have been attributed to the phase shifts of internal climate modes [e.g., the Pacific decadal oscillation (PDO), interdecadal Pacific oscillation (IPO), and Atlantic multidecadal oscillation (AMO); Liu and Chan 2013; Zhang et al. 2018; Zhao et al. 2018] and anthropogenic forcing (e.g., aerosol changes; Takahashi et al. 2017). During the inactive period of NWP TC activity (e.g., after the late 1990s), unfavorable large-scale environmental conditions (e.g., strong vertical wind shear and subtropical high, and weak monsoon trough) can suppress basinwide TCGF by approximately 20% in comparison with that during the active period (e.g., between the late 1980s and the mid-1990s; Aiyyer and Thorncroft 2011; Liu and Chan 2013; Hsu et al. 2014; Zhao and Wang 2019; Kim et al. 2020).
The interannual variability of basinwide NWP TCGF has been linked to sea surface temperature (SST) variations in the tropical Indian Ocean, tropical North Atlantic, and tropical Pacific (e.g., Chan 1985; Chen and Tam 2010; Du et al. 2011; Zhan et al. 2011; Yu et al. 2016; Zhang et al. 2016, 2017). Specifically, anomalous SST warming in the tropical Indian Ocean tends to induce low-level anticyclonic circulation anomalies and above-normal vertical wind shear in the NWP, inhibiting TC genesis in most parts of the basin (e.g., Xie et al. 2009; Du et al. 2011; Zhan et al. 2011; Tao et al. 2012; Ha et al. 2015). Meanwhile, Yu et al. (2016) suggested that SST anomalies in the tropical North Atlantic can modulate NWP TCGF via the route of the Indian Ocean. Another modulator that has been identified to affect NWP TC number and/or its spatial distribution is the SST anomalies in the tropical Pacific. The warm anomalies in the eastern equatorial Pacific associated with the eastern Pacific (EP) El Niño–Southern Oscillation (ENSO) primarily act to displace TC genesis in a southeast–northwest direction without a significant effect on basinwide TC number (e.g., Wang and Chan 2002). On the contrary, the anomalous warming in the central equatorial Pacific in association with the central Pacific (CP) ENSO (e.g., Larkin and Harrison 2005; Ashok et al. 2007; Kao and Yu 2009) favors TC genesis in the NWP primarily by enhancing low-level vorticity (e.g., Chen and Tam 2010; Kim et al. 2011; Bell et al. 2014; Chung and Li 2015; Mei et al. 2015; Patricola et al. 2018; Wu et al. 2018; J. Song et al. 2022).
Recently, a few studies have been devoted to understanding the variability of NWP TCGF on a subbasin scale, achieved by dividing the open ocean of the NWP into four regions using a fixed latitude and longitude (e.g., Wu et al. 2019; Mei and Li 2022). Wu et al. (2019) found that in the observations TCGF variations in both the southeastern and northwestern NWP are controlled by SST anomalies in the central-to-eastern equatorial Pacific and tropical Indian Ocean, while those in the northeastern NWP are subject to the influence of SST variability in the tropical North Atlantic. More recently, using both observations and a large ensemble of atmospheric simulations, Mei and Li (2022) identified the key regions of relative SSTs (i.e., local SST minus tropical-mean SST) for TCGF in each subbasin region of the NWP. Results showed that above-normal TCGF in each subbasin region is tied to enhanced relative SSTs either locally or to the southeast of the corresponding regions, which themselves are associated with changes in both local and remote SSTs. Further, they indicated that the predictability of TCGF varies substantially among the subbasin regions, with the South China Sea (SCS) and the southwest quadrant of the NWP showing relatively low predictability. This can be attributed to the significant differences in the noise level in the atmospheric processes (including TC genesis) over different subbasin regions.
While the two studies mentioned above explored the variability of NWP TCGF at a subbasin scale using fixed latitude and longitude, more knowledge is needed on the spatiotemporal variations of NWP TC genesis density on various time scales. In addition, several studies examined the responses of TC activity (e.g., TC counts, genesis, and track) to different El Niño flavors using observations and/or simulations with a relatively small ensemble size (e.g., Kim et al. 2011; Chung and Li 2015; Patricola et al. 2018). In those simulations and the observations, TC genesis is sparse and large noise exists, and thus a detailed and robust characterization of the spatiotemporal variations in TC genesis density is difficult. In this study, we take advantage of a large ensemble of high-resolution atmospheric simulations to fully address this issue. Specifically, we extract the leading modes of the variations in annual TC genesis density over the NWP using an ensemble of 100-member simulations, examine the physical mechanisms associated with these modes, and compare them with the observations (section 3). We then explore the impacts of the seasonal evolution of ENSO on NWP TC genesis in section 4. Conclusions are provided in section 5.
2. Data and methods
a. Observed and reanalysis data
Observed TC data are from the TC best track datasets, compiled and provided by the International Best Track Archive for Climate Stewardship (IBTrACS) project (Knapp et al. 2018). Because of the discrepancies among the best track datasets produced by different agencies for the NWP, here we consider three best track datasets (i.e., the Joint Typhoon Warning Center, Shanghai Typhoon Institute, and Japan Meteorological Agency) and use their average to represent the observations. All the datasets contain the information of TC intensity and location at 6-h intervals. Only TCs reaching at least tropical storm intensity and lasting for more than two days during 1951–2010 are considered in this study (e.g., Landsea et al. 2010).
Monthly SSTs during 1951–2010 with a horizontal resolution of 1° × 1° from the Centennial in situ Observation-Based Estimates SST (COBE-SST; Ishii et al. 2005) and monthly atmospheric fields during 1958–2010 with a horizontal resolution of 1.25° × 1.25° from the Japanese 55-year Reanalysis (JRA-55; Kobayashi et al. 2015) are adopted to understand the physical mechanisms underlying the variations in the observed TC genesis density. In section 3, TCs are analyzed on a yearly basis with SSTs and atmospheric fields being averaged during June–November (i.e., the TC peak season in the NWP). In section 4, TCs are analyzed for the four seasons [i.e., March–May (MAM), June–August (JJA), September–November (SON), and December–February (DJF)] with SSTs and atmospheric fields being averaged during the corresponding seasons.
b. Simulated data
The simulated TCs and atmospheric variables are from the historical simulations (i.e., 1951–2010) of the Database for Policy Decision Making for Future Climate Change (d4PDF; Mizuta et al. 2017). These simulations were run with the Meteorological Research Institute atmospheric general circulation model, version 3.2, and forced with observed monthly SSTs and sea ice concentration (COBE-SST2; Hirahara et al. 2014) as well as climatological monthly sea ice thickness. The simulations have a horizontal resolution of approximately 60 km, and consist of 100 ensemble members that differ in initial conditions and slightly in the imposed SSTs.
The simulated TCs are detected and tracked following an algorithm that combines Murakami et al. (2012) and Mei et al. (2014), and have been used in Mei and Li (2022) to study the variability and predictability of both basinwide and subbasin TCGF in the NWP. As shown in Yoshida et al. (2017) and Mei and Li (2022), the simulations faithfully reproduce the climatology and statistics of the observed TC genesis, including the climatological spatial distribution and seasonal evolution as well as the interannual-to-decadal variability at a basinwide scale.
c. Methods
TC genesis density is calculated as the total number of TCs forming in each 2° × 2° (longitude × latitude) grid for simulations and 8° × 5° (longitude × latitude) grid for observations on a yearly basis. To characterize the spatiotemporal variability of TC genesis density, we apply an empirical orthogonal function (EOF) analysis to the simulated TC genesis density to extract its leading modes. Considering that the variations in TC genesis density on different time scales may be controlled by different factors, we divide them into the interannual and decadal components using an 8-yr high-pass filter, and then study the two components separately. Simple linear regression and correlation analyses are used to identify the atmospheric conditions that connect SST variations with TC genesis variations. The standard two-tailed Student’s t test is used to estimate the significance level of the results. The IPO index is calculated following Henley et al. (2015), which is defined as the difference between the SST anomalies averaged over the central equatorial Pacific (10°S–10°N, 170°E–90°W) and the average of the SST anomalies in the northwest (25°–45°N, 140°E–145°W) and southwest (50°–15°S, 150°E–160°W) Pacific.
3. Interannual and decadal variations in simulated TC genesis density and comparisons with the observations
In this section, we shall examine separately the interannual and decadal variations of NWP TC genesis density using the 100-member ensemble of the simulated TC genesis density. The 100-member simulations dramatically increase the sample size of TC genesis in the NWP, which is too small in the observations (Fig. S1 in the online supplemental material). This allows us to robustly examine the SST-forced spatiotemporal variations of TC genesis density in the NWP. The underlying physical mechanisms will be diagnosed by analyzing the corresponding anomalies in both SSTs and the large-scale atmospheric environment. The variations in the simulated TC genesis density will also be compared with those in the observations.
a. Interannual variability in simulated TC genesis density
Applying an EOF analysis to the interannual component of the TC genesis density in the 100-member ensemble mean, we obtain two physically meaningful leading modes, which explain 26.5% and 10.4% of the total variance, respectively (Figs. 1 and 2). They are statistically separated based on North’s rule of thumb (North et al. 1982). The first leading mode is characterized by a dipole structure, with a positive loading in the southeastern NWP (i.e., south of 16°N and east of 144°E; Fig. 1a) and a negative loading west of 144°E (including the northwestern NWP, the SCS, and the southwestern NWP; Fig. 1a).1 The division between the areas of positive and negative loadings appears to well correspond to that used in Mei and Li (2022) to define the subbasin regions.
(a) The spatial pattern (TC number per 2° × 2° grid) of the first leading mode extracted from the interannual component of the simulated TC genesis density in the NWP during 1951–2010. Stippled areas denote regression coefficients statistically significant at the 0.05 level. Blue dotted lines delineate the boundaries of the five subbasin regions of the NWP as defined in Mei and Li (2022): the SCS and the four quadrants of the open ocean. (b) Regression map of observed TC genesis density anomalies (TC number per 2° × 2° grid) onto the normalized PC1. Stippled areas denote regression coefficients statistically significant at the 0.05 level. Blue dotted lines are the same as those in (a), which delineate the boundaries of the five subbasin regions. (c) The normalized PC1 (black curve) and normalized Niño-3.4 index during June–November (red curve; r = 0.85). (d) The boreal warm-season (i.e., June–November) SST anomalies (shading; °C) and 850-hPa wind anomalies (vectors; m s−1) in the 100-member ensemble regressed against the normalized PC1. Stippled areas denote regression coefficients statistically significant at the 0.05 level. Only the wind vectors statistically significant at the 0.05 level are shown. (e) The boreal warm-season (i.e., June–November) 500-hPa vertical pressure velocity anomalies (shading; Pa s−1) and SST anomalies (contours; °C) in the 100-member ensemble regressed against the normalized PC1. Stippled areas denote regression coefficients statistically significant at the 0.05 level.
Citation: Journal of Climate 37, 4; 10.1175/JCLI-D-22-0868.1
As in Fig. 1, but for the second leading mode extracted from the interannual component of the simulated TC genesis density in the NWP during 1951–2010. The red curve in (c) is the normalized Niño-3.4 index during January–March and its correlation coefficient with the normalized PC2 is 0.52.
Citation: Journal of Climate 37, 4; 10.1175/JCLI-D-22-0868.1
The southeast–northwest dipole structure resembles the typical anomalous spatial pattern of TC genesis in response to a developing El Niño (e.g., Wang and Chan 2002). This is supported by the strong correlation between the principal component (PC) of the first leading mode (i.e., PC1) and the Niño-3.4 index of the NWP TC peak season (i.e., June–November; the linear correlation coefficient r = 0.85; Fig. 1c) and the map of global SST anomalies regressed onto the PC1: strong warm anomalies extend from the west coasts of South America toward the central equatorial Pacific and cold anomalies are present in the western Pacific (Fig. 1d). Concurrently, a dipole structure also exists in the Indian Ocean SST, resembling the positive phase of the Indian Ocean dipole that often occurs during the developing El Niño years (e.g., Saji et al. 1999; Webster et al. 1999; Kao and Yu 2009).
We then proceed to use the GPI and its four components (section 2c) computed using the ensemble mean of the simulated fields to represent the large-scale atmospheric environment and identify the environmental conditions associated with the first leading mode via linear regression and correlation analyses. Figures 3a–e display the obtained regression patterns for the GPI and its components, respectively. The anomalous pattern of the GPI resembles that of TC genesis density shown in Fig. 1a, with the atmospheric state favoring TC genesis in the southeast quadrant of the basin and not favoring TC genesis in other subbasin regions (Fig. 3a). The favorable condition in the southeast quadrant is contributed primarily by dynamic factors, that is, increased low-level vorticity and reduced vertical wind shear (Figs. 3d,e).2 On the contrary, the hostile conditions outside this quadrant are dominated by an above-normal midlevel saturation deficit (Fig. 3c). This is in line with the typical responses of the atmospheric condition to the SST anomalies in the tropical Pacific associated with a developing El Niño. Specifically, positive SST anomalies over the central-to-eastern equatorial Pacific tend to enhance convective activity in the deep tropics of the central Pacific (Fig. 1e). The promoted convective heating generates an anomalously cyclonic circulation in the lower troposphere over the low latitudes of the NWP, as a part of the Rossby wave response (Fig. 1d). This circulation anomaly increases low-level vorticity over the tropical NWP and reduces vertical wind shear over the southeastern NWP (Figs. 3d,e). These dynamic changes favor TC genesis and development in the southeastern NWP. Outside this subbasin region, anomalous descending motion and increased midtroposphere saturation deficit (Figs. 1e and 3c), associated with anomalous upper-troposphere convergence and cooling in the lower boundary, suppress TC genesis (e.g., Wang and Chan 2002; Camargo et al. 2007).
Coefficients of GPI and its four components (i.e., potential intensity, midlevel saturation deficit, vertical wind shear, and low-level vorticity) regressed on the PC of (a)–(e) EOF1 and (f)–(j) EOF2 extracted from the interannual component of the simulated TC genesis density in the NWP during 1951–2010, and on the PC of (k)–(o) EOF1 extracted from the decadal component of the simulated TC genesis density in the NWP during 1951–2010. Stippling denotes regression coefficients statistically significant at the 0.05 level in (a)–(j) and correlation coefficients greater than 0.4 in (k)–(o). Gray dotted lines delineate the boundaries of the five subbasin regions of the NWP as defined in Mei and Li (2022): the SCS and the four quadrants of the open ocean.
Citation: Journal of Climate 37, 4; 10.1175/JCLI-D-22-0868.1
The second leading mode of the NWP TC genesis density variability on the interannual time scales features high loadings in the central part of the basin, with changes of opposite sign primarily near the equator and date line (Fig. 2a). The anomalous SST pattern associated with this mode resembles the decay phase of an El Niño, showing dissipated SST warming over the eastern tropical Pacific, developed SST warming over the tropical Indian Ocean and tropical North Atlantic, and SST cooling in the central tropical North Pacific (Fig. 2d). The connection of this mode with preceding-winter El Niño is confirmed by the statistically significant correlation coefficient between the PC of this mode (i.e., PC2) and the Niño-3.4 index of January–March (r = 0.52; Fig. 2c).
Analyses of dynamic and thermodynamic fields and GPI show that the above configuration of SST anomalies in the Indian and Pacific Oceans generates an anomalous anticyclonic circulation at the low level and anomalous subsidence at the midlevel over the central part of the NWP (Figs. 2d,e), reducing low-level vorticity and increasing midlevel saturation deficit (Figs. 3h,j). These changes tend to suppress TC genesis over most parts of the basin (Fig. 2a). These results are in line with previous studies discussing the negative effect of Indian Ocean warming on NWP TC counts (e.g., Du et al. 2011; Zhan et al. 2011). The remnant of the SST warming in the eastern equatorial Pacific also maintains a cyclonic circulation anomaly to the southeast of the anomalously anticyclonic circulation, enhancing low-level vorticity and thereby encouraging TC genesis near the equator and date line (Figs. 2a,d and 3j). It is worth noting that over the eastern Philippine Sea the GPI shows the opposite sign of anomaly in comparison with TC genesis, because TC genesis in this region may be more subject to the influence of saturation deficit and synoptic-scale disturbance activity.
b. Decadal variability in simulated TC genesis density
Only one physically meaningful mode is identified for the decadal variability of the simulated NWP TC genesis density, and it accounts for 29.7% of the variance on decadal time scales (Fig. 4). High loadings are primarily located south of 20°N and show an asymmetric east–west dipole structure with the eastern lobe having larger amplitude and covering a broader area (Fig. 4a). During a positive phase of this mode, more TCs tend to occur east of 140°E and fewer TCs are generated near the Philippines and the SCS, leading to an eastward shift in the mean genesis location. The dominance of the eastern lobe leads to above-normal basinwide TCGF during the positive phase of this mode, as evidenced by the high correlation coefficient between the PC of this mode and the low-pass-filtered basinwide TCGF (i.e., r = 0.82).
As in Fig. 1, but for the first leading mode extracted from the decadal component of the simulated TC genesis density in the NWP during 1951–2010. The red curve in (c) is the filtered (i.e., 8-yr low-pass) and normalized IPO index during June–November and its correlation coefficient with the normalized PC1 is 0.60. Stippled areas in (a), (b), (d), and (e) denote correlation coefficients greater than 0.4. Only the wind vectors with correlation coefficients greater than 0.4 are shown in (d).
Citation: Journal of Climate 37, 4; 10.1175/JCLI-D-22-0868.1
The associated anomalous SST pattern is dominated by warm anomalies in the central tropical Pacific and central-to-eastern subtropical North Pacific and cold anomalies in the western Pacific and central North and South Pacific (Fig. 4d). This SST pattern resembles the spatial pattern of the IPO, and the time series of the PC is correlated with the low-pass-filtered normalized IPO index of June–November at 0.60 (Fig. 4c). Previous studies have also suggested a linkage between the basinwide NWP TCGF and the IPO/PDO on decadal time scales: the IPO/PDO can impact the low-frequency variations in the NWP TC counts by modulating the subtropical high and tropospheric vertical wind shear in this basin (e.g., Liu and Chan 2013; Hsu et al. 2014; Lin and Chan 2015; Zhao et al. 2018; Wang and Wang 2022).
Analysis of large-scale environmental conditions suggests that the positive SST anomalies in the central tropical Pacific and central-to-eastern subtropical North Pacific generate a low-level cyclonic circulation anomaly to their northwest (Fig. 4d) and strengthen the convective activity in the southeast quadrant of the NWP (Fig. 4e). These changes enhance low-level relative vorticity and moisten the midtroposphere in the region east and south of 145°E and 20°N (Figs. 3m,o). Meanwhile, the westerly anomalies in the southern flank of the anomalously cyclonic circulation weaken the easterly trade winds in the southeastern NWP and thereby reduce vertical wind shear there (Fig. 3n). All these conditions favor TC genesis in the southeast quadrant of the NWP (Fig. 4a). In the areas surrounding the Philippines and the SCS, local cold SST anomalies appear to discourage deep convection (Fig. 4e), which corresponds to increased midlevel saturation deficit and decreased potential intensity (Figs. 3l,m). In addition, the low-level westerly anomalies in these areas (Fig. 4d) strengthen the climatological westerly winds, resulting in enhanced vertical wind shear, especially east of the Philippines (Fig. 3n). These factors dominate over the increased low-level relative vorticity that is tied to the anomalously cyclonic circulation in the lower troposphere (Figs. 3o and 4d), leading to suppressed TC genesis west of 140°E (Fig. 4a).
c. Comparisons with the observations
For each of the three modes discussed above, we further regress the interannual or decadal components of observed TC genesis density anomalies during 1951–2010 onto the corresponding PC (Figs. 1b, 2b, and 4b).3 We perform the regression analysis for observed TCs using a much larger grid size (i.e., 8° × 5° instead of 2° × 2°) because of the much smaller TC sample size in the observations. The regression maps (Figs. 1b, 2b, and 4b) show anomalous patterns very similar to those in the simulations extracted via EOF analysis (Figs. 1a, 2a, and 4a). For example, both the observed TC genesis density anomalies shown in Figs. 1b and 4b exhibit dipole structures that are the same as the patterns from the simulations (Figs. 1a and 4a). The regression map of observed TC genesis density anomalies corresponding to PC2 shown in Fig. 2c also resembles the EOF pattern in the simulations, except for the difference in the area west of the Philippines (cf. Figs. 2a,b). These results from the observations demonstrate the fidelity of the simulations in reproducing the spatiotemporal variations of observed TC genesis density.
We also compare the EOF results from the observations (Fig. S3) and those from two member simulations (Figs. S4 and S5). In both the observations and individual member simulations, because of the presence of large noise, the variances explained by the first two leading modes are comparable, and thus the two leading modes are not statistically separated based on North’s rule of thumb (North et al. 1982). But the second leading mode in the observations does depict the response of TC genesis density to El Niño: increased TC genesis in the southeastern NWP and decreased TC genesis in other subbasin regions of the NWP (e.g., the northwest quadrant; Fig. S3b). In some member simulations (e.g., member 89; Fig. S4), one of the first two leading modes does capture the response to ENSO, resembling the second leading mode in the observations. However, in some member simulations (e.g., member 54; Fig. S5), the response to ENSO does not emerge in the leading modes, because of the strong randomness in individual member simulations. This is why we apply EOF analysis to the 100-member ensemble mean to characterize the spatiotemporal variations of TC genesis density, through which we are able to uncover the two physically meaningful modes shown in Figs. 1 and 2.
d. Number of ensemble members needed to obtain robust results
To determine the minimum number of ensemble members needed to obtain a robust relationship, we calculate the EOF results for the ensembles consisting of various numbers of member simulations. First, using a random sampling method, we independently draw N member (N = 1, 2, 3, …, 100) simulations from the entire 100 member simulations, forming an N-member ensemble. We then apply EOF analysis to the mean of the obtained ensemble, and extract the leading EOF modes of the NWP TC genesis density from the ensemble mean. Third, after pairing the newly obtained modes with those from the 100-member ensemble mean, for each of the leading modes described in sections 3a and 3b, we compute the correlation coefficients of the PC and spatial pattern of the mode between the N-member ensemble mean and 100-member ensemble mean. For each number of N, we repeat the above calculations 2000 times, yielding a collection of 2000 correlation coefficients. We then use a box-and-whisker plot to visualize the distribution of the collection of the correlation coefficients (Fig. 5).
Box-and-whisker plots of the correlation coefficients between the EOF mode from the N-member ensemble mean and that from the 100-member ensemble mean as a function of ensemble size for the (top) PC and (bottom) spatial pattern. (a),(b) The first EOF mode on interannual time scales, (c),(d) the second EOF mode on interannual time scales, and (e),(f) the first EOF mode on decadal time scales.
Citation: Journal of Climate 37, 4; 10.1175/JCLI-D-22-0868.1
For both the PC and spatial pattern of a specific mode, as the ensemble size increases, the mean value of the correlation coefficients increases, with their range narrowing down. The mean value (ren) increases dramatically at the beginning and tends to level off as the ensemble size (N) further increases. To quantify the number of ensemble members needed to capture the spatiotemporal variability associated with the mode (Nmin), we calculate the rate of change in ren and determine Nmin such that when N > Nmin the rate of change in ren is smaller than p × ren_max [ren_max is the maximum value that ren can reach (i.e., 1 in Fig. 5); the value of p is more or less subjective and here we set it to 2.5 × 10−3; see details in section 5c of Mei and Li (2022)]. The obtained Nmin for each subplot in Fig. 5 is as follows: 8 and 16 for the PC and spatial pattern of the first mode on interannual time scales, respectively; 20 and 31 for the second mode on interannual time scales; and 11 and 24 for the first mode on decadal time scales. Overall, we conclude that ensembles of 15, 30, and 20 member simulations are needed to capture the spatiotemporal variability associated with the first mode on interannual time scales, the second mode on interannual time scales, and the first mode on decadal time scales, respectively.
To explain the big difference in the needed ensemble size between the first and second modes on interannual time scales, we calculate the signal-to-noise ratio (SNR) of TC genesis density following Mei and Li (2022) and show the results in Fig. 6. The values of SNR are the largest in the southeastern NWP, where high loadings are located for the first EOF mode on interannual time scales (Fig. 1a). On the contrary, high loadings in the second mode on interannual time scales (Fig. 2a) are in the areas with small SNRs (Fig. 6). Thus, a greater number of ensemble members are needed for the second mode to reduce the noise.
Spatial distribution of the signal-to-noise ratio (SNR; Mei and Li 2022) of NWP TC genesis density in the 100-member simulations.
Citation: Journal of Climate 37, 4; 10.1175/JCLI-D-22-0868.1
4. Seasonal evolution of the ENSO effect
The analyses in section 3 indicate that the variations of NWP TC genesis density on interannual time scales are strongly tied to ENSO-related SST anomalies, with the two leading modes emerging respectively during the developing and decay stages of El Niño. Considering the prominent seasonal dependency of both TC genesis and ENSO (e.g., Chia and Ropelewski 2002; Mei et al. 2015; Xie et al. 2016), it is of great interest to explore how the responses in TC genesis density to ENSO evolve with season, which has important implications for seasonal TC prediction in the NWP.
a. Seasonality of climatological TC genesis density
Figures 7a–d display the climatology of TC genesis density in the four seasons [i.e., March–May (MAM), June–August (JJA), September–November (SON), and December–February (DJF)] during 1951–2010 in the simulations. TC genesis exhibits distinct seasonal cycles in terms of both the amplitude and location. It is much more active during JJA and SON than during MAM and DJF. Specifically, TC genesis is confined to south of 10°N in MAM, migrates northward from MAM to JJA when most TCs form within the latitudinal band of 8°–24°N, and then retreats southward to be primarily located within 6°–20°N during SON and further southward to be concentrated south of 10°N during DJF. It extends from the SCS to the date line in a southeast–northwest orientation during MAM and DJF, but in a much more zonal direction during JJA and SON. The observations show very similar seasonality in TC genesis as the simulations (cf. Figs. 7a–d,e–h). The simulated and observed seasonal evolutions of the climatological TCGF in the entire NWP and its five subbasin regions are compared in Fig. 2 of Mei and Li (2022). The model performs better at replicating the observed climatologically seasonal cycle of TCGF in the northwest, northeast, and southeast quadrants than in the SCS and the southwest quadrant regarding the magnitude and phase of the seasonal cycle. In the SCS, the simulated seasonal cycle has a magnitude comparable with that in the observations, but lags in phase by around one month, with more TCs forming during October–March and fewer TCs during May–August in the simulations. In the southwest quadrant, the seasonal cycle in the simulations has a magnitude only around half of that in the observations, and shows more TCs during December–March and fewer TCs during June–October. Next we shall examine the responses in TC genesis density to ENSO during different seasons using TCs from the ensemble simulations.
Climatology of TC genesis density (TC number per year per 2° × 2° grid) during the four seasons [i.e., March–May (MAM), June–August (JJA), September–November (SON), and December–February (DJF)] of 1951–2010 for (a)–(d) simulated and (e)–(h) observed TCs. Blue dotted lines delineate the boundaries of the five subbasin regions of the NWP as defined in Mei and Li (2022): the SCS and the four quadrants of the open ocean.
Citation: Journal of Climate 37, 4; 10.1175/JCLI-D-22-0868.1
b. Seasonal evolution of the ENSO effect on TC genesis density
1) CP ENSO
Figure 8 shows the seasonal evolution of SST anomalies associated with the CP El Niño (i.e., C-index), and the accompanying changes in the large-scale atmospheric circulation at 850 hPa. The corresponding evolution of the responses in TC genesis density is displayed in Fig. 9. The warm anomalies associated with the CP El Niño are initiated over the central-to-eastern tropical North Pacific and central equatorial Pacific in spring [i.e., MAM(0); Fig. 8a]. These SST anomalies produce an anomalously cyclonic circulation in the lower troposphere over the tropical NWP with strong westerly anomalies along the equator (Fig. 8a). The enhanced low-level vorticity promotes TC genesis in the NWP, primarily south of 10°N (Fig. 9a and Fig. S6a).
Coefficients of SST anomalies (shading; °C) and 850-hPa wind anomalies (vectors; m s−1) during six consecutive seasons [i.e., MAM(0), JJA(0), SON(0), D(0)JF(1), MAM(1), and JJA(1)] regressed on the C-index during 1951–2010. Stippled areas denote regression coefficients statistically significant at the 0.05 level. Only the wind vectors statistically significant at the 0.05 level are shown.
Citation: Journal of Climate 37, 4; 10.1175/JCLI-D-22-0868.1
Coefficients of TC genesis density (TC number per 2° × 2° grid) during six consecutive seasons [i.e., MAM(0), JJA(0), SON(0), D(0)JF(1), MAM(1), and JJA(1)] regressed on the C-index during 1951–2010. Stippled areas denote regression coefficients statistically significant at the 0.05 level. Blue dotted lines delineate the boundaries of the five subbasin regions of the NWP as defined in Mei and Li (2022): the SCS and the four quadrants of the open ocean.
Citation: Journal of Climate 37, 4; 10.1175/JCLI-D-22-0868.1
In JJA(0), the warm anomalies in the central equatorial Pacific develop and extend eastward, reaching the Pacific coasts of South America (Fig. 8b). The anomalously cyclonic circulation in the NWP expands westward and northward to cover nearly the entire basin, resulting in increased low-level vorticity south of 25°N including the SCS (Fig. 8b and Fig. S6b). Because of the northward migration of the climatological TC genesis (Fig. 7b), the area with significant response in TC genesis also shifts northward, with a prominent increase in TC genesis south of 20°N, primarily in the southeastern NWP and SCS (Fig. 9b). The reduced vertical wind shear tied to the low-level westerly anomalies also contributes to promoting TC genesis in the southeastern NWP (Fig. 9b and Fig. S6b).
During SON(0), the SST warming along the central-to-eastern equatorial Pacific continues growing in strength; meanwhile, cold anomalies develop over the western tropical Pacific (Fig. 8c). Such an anomalous SST pattern induces an anomalously anticyclonic circulation in the lower troposphere extending from the tropical north Indian Ocean to the Philippines, accompanied by an eastward retreat of the anomalously cyclonic circulation in the NWP (Fig. 8c). The changes in low-level vorticity and midlevel saturation deficit (Fig. S6c) associated with the dipole pattern in low-level circulation anomaly lead to a dipole structure in the TC genesis anomaly, with above-normal cyclogenesis confined to the southeast quadrant of the basin and below-normal genesis west of 145°E (Fig. 9c).
In D(0)JF(1), the SST warming in the Niño regions peaks, with the anomaly strongest in the central equatorial Pacific (Fig. 8d). The concurrent intensification of cold SST anomaly in the tropical NWP, together with the development of warm SST anomaly along the equatorial Indian Ocean, leads to an eastward and northeastward expansion of the anomalously low-level anticyclonic circulation that exists over the tropical north Indian Ocean and the SCS in SON(0) (cf. Figs. 8c,d). The accompanying changes in low-level vorticity and midlevel saturation deficit (Fig. S6d) cause the dipole structure of the anomalies in TC genesis to shift eastward slightly (Fig. 9d). The genesis anomalies also exhibit a southward displacement in comparison with those in SON(0) because of the southward retreat of TC genesis climatology (cf. Figs. 7c,d).
In MAM(1), the SST anomalies in the tropical Pacific and the circulation anomalies over the NWP begin to decay, while their patterns persist (Fig. 8e). Consistently, the changes in the large-scale environment and TC genesis show a dipole pattern that is similar to those during D(0)JF(1) but weaker in magnitude (cf. Fig. 9d and Fig. S6d, and Fig. 9e and Fig. S6e). During JJA(1), the SST anomalies in the tropical Pacific continue to dissipate, with the cold anomaly in the tropical NWP almost disappearing (Fig. 8f). The warm anomaly in the tropical Indian Ocean has shifted to the SCS. Such a configuration of SST anomalies produces a cyclonic circulation anomaly at the low level over the SCS, while the dipole pattern of the circulation anomalies over the open ocean of the NWP has substantially weakened (Fig. 8f). The resultant changes in low-level vorticity and midlevel saturation deficit (Fig. S6f) lead to a tripole pattern in TC genesis, with the southeast quadrant of the basin and the SCS experiencing more genesis and the northwest quadrant witnessing below-normal genesis (Fig. 9f). The northward shift of the TC genesis anomalies over the open ocean is due to the northward migration of TC genesis climatology (Fig. 7b). We note that all these changes are rather modest, compared to those in previous seasons.
2) EP ENSO
This subsection discusses the seasonal evolution of another type of ENSO, which is defined by the E-index in Eq. (3) and features the EP El Niño (Fig. 10). During the spring of the developing year [i.e., MAM(0)], warm SST anomalies of considerable magnitude have been developing over the eastern equatorial Pacific and along the coast of South America (Fig. 10a). These warm anomalies are located to the east of the warm anomalies emerging along the equator during MAM(0) of the CP El Niño (cf. Figs. 8a and 10a). Correspondingly, the anomalously cyclonic circulation in the lower troposphere also shifts eastward4 to the southeastern NWP, as do the positive low-level relative vorticity anomaly and TC genesis density anomaly (cf. Figs. 8a and 9a and Fig. S6a, and Figs. 10a and 11a and Fig. S7a).
Coefficients of SST anomalies (shading; °C) and 850-hPa wind anomalies (vectors; m s−1) during six consecutive seasons [i.e., MAM(0), JJA(0), SON(0), D(0)JF(1), MAM(1), and JJA(1)] regressed on the E-index during 1951–2010. Stippled areas denote regression coefficients statistically significant at the 0.05 level. Only the wind vectors statistically significant at the 0.05 level are shown.
Citation: Journal of Climate 37, 4; 10.1175/JCLI-D-22-0868.1
Coefficients of TC genesis density (TC number per 2° × 2° grid) during six consecutive seasons [i.e., MAM(0), JJA(0), SON(0), D(0)JF(1), MAM(1), and JJA(1)] regressed on the E-index during 1951–2010. Stippled areas denote regression coefficients statistically significant at the 0.05 level. Blue dotted lines delineate the boundaries of the five subbasin regions of the NWP as defined in Mei and Li (2022): the SCS and the four quadrants of the open ocean.
Citation: Journal of Climate 37, 4; 10.1175/JCLI-D-22-0868.1
In JJA(0), the positive SST anomalies in the eastern equatorial Pacific have substantially strengthened and extended to the central equatorial Pacific (Fig. 10b). The strengthened and westward-extended SST anomalies generate a prominent wave train in the anomalous low-level circulation in the meridional direction over the western North Pacific, with a cyclonic anomaly south of 15°N and an anticyclonic anomaly between 15° and 32°N (Fig. 10b). The corresponding changes in low-level vorticity and accompanying changes in midlevel saturation deficit (Fig. S7b) promote and suppress TC genesis respectively south and north of 15°N in the NWP (Fig. 11b). We also note that the warm SST anomalies in the Indian Ocean are developing and extending northward (Fig. 10b).
During SON(0), with the continuous development of the warm SST anomalies in the central-to-eastern equatorial Pacific and tropical Indian Ocean, the anticyclonic circulation anomaly in the lower troposphere expands and covers nearly the entire NWP south of 30°N, except the southeast quadrant of the basin where the SST warming in the central-to-eastern equatorial Pacific generates an anomalously cyclonic circulation (Fig. 10c). Consistently, reduced TC genesis is observed over most parts of the basin while above-normal TC genesis is present in the southeastern NWP (Fig. 11c). This is also reflected in the anomalous pattern of the GPI, with below-normal low-level vorticity and midlevel relative humidity leading to decreased GPI in the majority of the NWP except the southeast corner of the basin where enhanced low-level vorticity dominates (Fig. S7c).
The SST warming over the eastern equatorial Pacific peaks in the winter season [i.e., D(0)JF(1); Fig. 10d] and decays afterward [i.e., MAM(1); Fig. 10e]. Meanwhile, the warming over the Indian Ocean continues to develop (Figs. 10d,e). In response to the strengthened warming in the Indian Ocean, the anticyclonic circulation anomaly in the lower troposphere intensifies and expands eastward (Figs. 10d,e). In comparison with the situation during CP El Niños, this anomalously anticyclonic circulation occupies a larger area, and as a result of this and the accompanying changes in midlevel saturation deficit (Figs. S7d,e), the tropical NWP experiences below-normal TC genesis nearly everywhere (Figs. 11d,e) instead of the east–west dipole pattern seen during CP El Niños (Figs. 9d,e).
During JJA(1), the warm anomaly in the eastern equatorial Pacific undergoes a transition to a La Niña phase, while the warming in the Indian Ocean and the SCS is sustained (Fig. 10f). The warm SST anomalies in the latter regions produce a compact anticyclonic circulation anomaly in the NWP, with its center displaced northward relative to the previous two seasons (cf. Figs. 10d,e,f). This results in suppressed low-level vorticity and TC genesis between 10° and 25°N, primarily in the southeast and northeast quadrants of the basin (Fig. 11f and Fig. S7f).
3) Comparisons with the observations
We have also examined the seasonal evolution of the two types of ENSO and their impacts on TC genesis density in the observations following the same procedures. Because of the much smaller sample size of the observed TCs, the regression analysis is performed using a much larger grid size (i.e., 8° × 5° instead of 2° × 2°). The obtained results (Figs. S8–S11) are very similar to those from the simulations (Figs. 8–11), except for some slight differences in the responses of circulations and TCs. For example, the anticyclonic circulation anomalies over the open ocean of the NWP associated with the E-index (i.e., the EP El Niño) during JJA(1) are located more eastward in the simulations than in the observations (cf. Fig. 10f and Fig. S10f), and correspondingly the anomalies in the simulated TC genesis density are also shifted eastward compared with those in the observations (cf. Fig. 11f and Fig. S11f).
5. Summary and conclusions
While the temporal variations in basinwide tropical cyclone (TC) genesis frequency (TCGF) have been extensively explored, our knowledge of the spatiotemporal variations in TC genesis density is still lacking, primarily owing to the small sample size in the observations. In this study, we fill the gap and investigate the variability of annual TC genesis density over the northwest Pacific (NWP) during 1951–2010, taking advantage of a large ensemble (i.e., 100 members) of 60-km-resolution atmospheric simulations that well reproduce the spatial distribution and seasonal evolution of climatological TC genesis and capture the year-to-year variations of basinwide TC counts in the observations. We also explore the seasonality of the El Niño–Southern Oscillation (ENSO) effects on TC genesis density.
The spatiotemporal variability in TC genesis density is studied via the empirical orthogonal function (EOF) analysis and separately on interannual and decadal time scales. The interannual variations of the simulated NWP TC genesis density are dominated by two leading modes (Figs. 1 and 2). The first leading mode is characterized by a dipole pattern and can be linked to the developing ENSO (Fig. 1). Specifically, developing El Niños tend to produce favorable environmental conditions (i.e., increased low-level vorticity and decreased vertical wind shear; Figs. 3d,e) and thus above-normal TC genesis in the southeast quadrant of the NWP (Fig. 1a), and hostile environment (i.e., dry midtroposphere; Fig. 3c) and suppressed TC genesis outside this quadrant (Fig. 1a). The second mode features high loadings over most parts of the open ocean, except near the equator and date line where relatively low loadings of opposite sign exist (Fig. 2a). This mode primarily takes place during ENSO decay years: SST warming in the tropical Indian Ocean and the SCS and the remnant warming in the eastern tropical Pacific associated with the decaying El Niño (Fig. 2d) discourage TC genesis over most of the NWP by affecting low-level vorticity and midlevel saturation deficit (Figs. 3h,j); the latter SST warming favors TC genesis near the equator and date line through its effect on low-level vorticity (Fig. 3j). We note that the Indo-Pacific SST gradient is also an important modulator for TC genesis in the NWP (e.g., Zhan et al. 2013, 2019). The correlation between the basinwide TCGF and the Indo-Pacific SST gradient is −0.39 (p < 0.01) during 1951–2010 and −0.56 (p < 0.01) during 1979–2010 in the simulations. We will thoroughly explore the connections between the Indo-Pacific SST gradient and the developing and decaying ENSO and their respective contributions to TCGF variations in the NWP in a future study.
On decadal time scales, the simulated TC genesis density is mainly controlled by one mode, which exhibits an east–west dipole structure with strong signals confined to south of 20°N (Fig. 4a). This mode is driven by low-frequency SST variations over the central tropical Pacific and central-to-eastern subtropical North Pacific (Fig. 4d): SST warming in these regions tends to generate a low-level cyclonic circulation anomaly to the northwest, enhancing low-level vorticity nearly over the entire basin and decreasing and increasing vertical wind shear respectively east and west of around 150°E south of 20°N (Figs. 3n,o and 4d). These warm SST anomalies also promote local convective activity and midlevel relative humidity (Figs. 3m and 4e), while the thermodynamic factors are considerably suppressed to the west owing to anomalous downward motion and presence of cold SST anomalies (Figs. 3l,m and 4d,e). All these environmental changes lead to above-normal and below-normal TC genesis approximately east and west of 140°E.5
Further, we explore the seasonal evolution of ENSO effect on TC genesis density over consecutive seasons extending from March–May (MAM) of the El Niño developing year to June–August (JJA) of the decay year. We categorize ENSO into two types: the eastern Pacific (EP) and central Pacific (CP) ENSO, using the first two leading EOF modes of the tropical Pacific SST variability following Takahashi et al. (2011), and then examine the seasonality of the respective effects of these two types of ENSO. The two types of ENSO differ in various aspects, including the amplitude and location of the maximum SST anomaly in the tropical Pacific and the pattern and pace of the decay. The differences in the SST evolution in the tropical Pacific and other basins (primarily the tropical Indian Ocean) produce remarkable differences in the response of the large-scale atmospheric environment to the two types of ENSO (Figs. 8 and 10). Correspondingly, the NWP TC genesis density anomaly responds very differently to the seasonal evolution of the two types of ENSO (Figs. 9 and 11), which is further complicated by the seasonal shifts of TC genesis climatology (Fig. 7). Overall, the results highlight distinct impacts of different ENSO flavors on TC genesis density in the NWP and show strong seasonal dependency of the TC genesis responses to ENSO.
We also examine the interannual and decadal variations in TC genesis density as well as the seasonality of the ENSO effect using observations. For the interannual and decadal variability, regression maps of observed TC genesis density (Figs. 1b, 2b, and 4b) are calculated based on the principal components (PCs) in the simulations. The obtained anomalous patterns in observed TC genesis density are very similar to those in the simulations extracted via the EOF analysis (Figs. 1a, 2a, and 4a). High consistency is also found between the observations (Figs. S8–S11) and simulations (Figs. 8–11) regarding the seasonal evolution of TC genesis density response to the two types of ENSO. Although the observational results are not as robust as those from the simulations because of the much smaller sample size, their consistency with the results from the simulations lends confidence to our results from the simulations and demonstrates the capability of the model at reproducing the variability and statistics of TC genesis density in the observations.
In short, we have investigated the spatial and temporal variations in TC genesis density over the NWP and their connections to the seasonal evolution of different flavors of ENSO, using an unprecedentedly large ensemble of atmospheric simulations that remedies the undersampling issue of the observations. The novelty of this work is that we thoroughly examined the forced variability of TC genesis density in the NWP using the 100-member ensemble, extracted its leading modes respectively on the interannual and decadal time scales, and explored how the spatial structure of TC genesis responds to different types of ENSO during different seasons. These have not been addressed before, primarily because of the limited TC sample size in the observations. The results contribute to our understanding of the SST and climate control of TC genesis and development in the NWP with important implications for seasonal TC predictions.
Blue dotted lines in Fig. 1a delineate the boundaries of the five subbasin regions of the NWP as defined in Mei and Li (2022).
Positive anomalies in the subplots for the four components of the GPI shown in Fig. 3 all indicate positive effects of the components on GPI, which are produced by increased potential intensity, decreased midlevel saturation deficit, decreased vertical wind shear, and increased low-level vorticity, respectively.
The anomalously cyclonic circulation in the lower troposphere is not statistically significant in the simulations and thus not visible in Fig. 10a. It is, however, statistically significant in the observations (Fig. S10a). We note that the positive anomalies in low-level relative vorticity and thus the GPI are statistically significant in the simulations (Fig. S7a).
Acknowledgments.
This work was supported by National Science Foundation Grant AGS-2047721. This study used d4PDF produced with the Earth Simulator jointly by science programs (SOUSEI, TOUGOU, SI-CAT, DIAS) of the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan. We thank the editor and three anonymous reviewers for their comments, which greatly helped improve the manuscript.
Data availability statement.
The TC best-track datasets provided by the IBTrACS are available at https://www.ncdc.noaa.gov/ibtracs/. The monthly atmospheric fields from the JRA-55 are available at https://rda.ucar.edu/datasets/ds628.1/. The monthly SSTs from the COBE-SST can be accessed at https://psl.noaa.gov/data/gridded/data.cobe.html. Historical simulations of the Database for Policy Decision Making for Future Climate Change (d4PDF) are available at https://www.miroc-gcm.jp/d4PDF/index_en.html.
REFERENCES
Aiyyer, A., and C. Thorncroft, 2011: Interannual-to-multidecadal variability of vertical shear and tropical cyclone activity. J. Climate, 24, 2949–2962, https://doi.org/10.1175/2010JCLI3698.1.
Ashok, K., S. K. Behera, S. A. Rao, H. Weng, and T. Yamagata, 2007: El Niño Modoki and its possible teleconnection. J. Geophys. Res., 112, C11007, https://doi.org/10.1029/2006JC003798.
Bell, R., K. Hodges, P. L. Vidale, J. Strachan, and M. Roberts, 2014: Simulation of the global ENSO–tropical cyclone teleconnection by a high-resolution coupled general circulation model. J. Climate, 27, 6404–6422, https://doi.org/10.1175/JCLI-D-13-00559.1.
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., K. A. Emanuel, and A. H. Sobel, 2007: 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.
Camargo, S. J., and Coauthors, 2019: Tropical cyclone prediction on subseasonal time-scales. Trop. Cyclone Res. Rev., 8, 150–165, https://doi.org/10.1016/j.tcrr.2019.10.004.
Chan, J. C. L., 1985: Tropical cyclone activity in the northwest Pacific in relation to the El Niño/Southern Oscillation phenomenon. Mon. Wea. Rev., 113, 599–606, https://doi.org/10.1175/1520-0493(1985)113<0599:TCAITN>2.0.CO;2.
Chan, J. C. L., 2005: Interannual and interdecadal variations of tropical cyclone activity over the western North Pacific. Meteor. Atmos. Phys., 89, 143–152, https://doi.org/10.1007/s00703-005-0126-y.
Chan, J. C. L., and J.-e. Shi, 1996: Long-term trends and interannual variability in tropical cyclone activity over the western North Pacific. Geophys. Res. Lett., 23, 2765–2767, https://doi.org/10.1029/96GL02637.
Chen, G., and C.-Y. Tam, 2010: Different impacts of two kinds of Pacific Ocean warming on tropical cyclone frequency over the western North Pacific. Geophys. Res. Lett., 37, L01803, https://doi.org/10.1029/2009GL041708.
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.
Chu, P.-S., 2004: ENSO and tropical cyclone activity. Hurricanes and Typhoons: Past, Present, and Future, R. J. Murnane and K.-B. Liu, Eds., Columbia University Press, 297–332.
Chung, P.-H., and T. Li, 2015: Characteristics of tropical cyclone genesis in the western North Pacific during the developing and decaying phases of two types of El Niño. J. Trop. Meteor., 21, 14–22, https://www.proquest.com/docview/1648091544?accountid=14229.
Du, Y., L. Yang, and S.-P. Xie, 2011: Tropical Indian Ocean influence on northwest Pacific tropical cyclones in summer following strong El Niño. J. Climate, 24, 315–322, https://doi.org/10.1175/2010JCLI3890.1.
Emanuel, K. A., 2010: Tropical cyclone activity downscaled from NOAA-CIRES reanalysis, 1908–1958. J. Adv. Model. Earth Syst., 2 (1), https://doi.org/10.3894/JAMES.2010.2.1.
Geiger, T., J. Gütschow, D. N. Bresch, K. Emanuel, and K. Frieler, 2021: Double benefit of limiting global warming for tropical cyclone exposure. Nat. Climate Change, 11, 861–866, https://doi.org/10.1038/s41558-021-01157-9.
Ha, Y., Z. Zhong, X. Yang, and Y. Sun, 2015: Contribution of East Indian Ocean SSTA to western North Pacific tropical cyclone activity under El Niño/La Niña conditions. Int. J. Climatol., 35, 506–519, https://doi.org/10.1002/joc.3997.
Henley, B. J., J. Gergis, D. J. Karoly, S. Power, J. Kennedy, and C. K. Folland, 2015: A tripole index for the interdecadal Pacific oscillation. Climate Dyn., 45, 3077–3090, https://doi.org/10.1007/s00382-015-2525-1.
Hirahara, S., M. Ishii, and Y. Fukuda, 2014: Centennial-scale sea surface temperature analysis and its uncertainty. J. Climate, 27, 57–75, https://doi.org/10.1175/JCLI-D-12-00837.1.
Hsu, P.-C., P.-S. Chu, H. Murakami, and X. Zhao, 2014: An abrupt decrease in the late-season typhoon activity over the western North Pacific. J. Climate, 27, 4296–4312, https://doi.org/10.1175/JCLI-D-13-00417.1.
Hu, F., T. Li, J. Liu, and M. Peng, 2018: Cause of interdecadal change of tropical cyclone controlling parameter in the western North Pacific. Climate Dyn., 51, 719–732, https://doi.org/10.1007/s00382-017-3951-z.
Ishii, M., A. Shouji, S. Sugimoto, and T. Matsumoto, 2005: Objective analyses of sea-surface temperature and marine meteorological variables for the 20th century using ICOADS and the Kobe collection. Int. J. Climatol., 25, 865–879, https://doi.org/10.1002/joc.1169.
Kao, H.-Y., and J.-Y. Yu, 2009: Contrasting eastern-Pacific and central-Pacific types of ENSO. J. Climate, 22, 615–632, https://doi.org/10.1175/2008JCLI2309.1.
Kim, H.-K., K.-H. Seo, S.-W. Yeh, N.-Y. Kang, and B.-K. Moon, 2020: Asymmetric impact of central Pacific ENSO on the reduction of tropical cyclone genesis frequency over the western North Pacific since the late 1990s. Climate Dyn., 54, 661–673, https://doi.org/10.1007/s00382-019-05020-8.
Kim, H.-M., P. J. Webster, and J. A. Curry, 2011: Modulation of North Pacific tropical cyclone activity by three phases of ENSO. J. Climate, 24, 1839–1849, https://doi.org/10.1175/2010JCLI3939.1.
Knapp, K. R., H. J. Diamond, J. P. Kossin, M. C. Kruk, and C. J. Schreck, 2018: International Best Track Archive for Climate Stewardship (IBTrACS) Project, version 4. NOAA/National Centers for Environmental Information, accessed October 2021, https://www.ncei.noaa.gov/products/international-best-track-archive.
Knutson, T. R., and Coauthors, 2010: Tropical cyclones and climate change. Nat. Geosci., 3, 157–163, https://doi.org/10.1038/ngeo779.
Kobayashi, S., and Coauthors, 2015: The JRA-55 reanalysis: General specifications and basic characteristics. J. Meteor. Soc. Japan, 93, 5–48, https://doi.org/10.2151/jmsj.2015-001.
Korty, R. L., S. J. Camargo, and J. Galewsky, 2012: Tropical cyclone genesis factors in simulations of the last glacial maximum. J. Climate, 25, 4348–4365, https://doi.org/10.1175/JCLI-D-11-00517.1.
Korty, R. L., K. A. Emanuel, M. Huber, and R. A. Zamora, 2017: Tropical cyclones downscaled from simulations with very high carbon dioxide levels. J. Climate, 30, 649–667, https://doi.org/10.1175/JCLI-D-16-0256.1.
Kossin, J. P., K. A. Emanuel, and S. J. Camargo, 2016: Past and projected changes in western North Pacific tropical cyclone exposure. J. Climate, 29, 5725–5739, https://doi.org/10.1175/JCLI-D-16-0076.1.
Landsea, C. W., G. A. Vecchi, L. Bengtsson, and T. R. Knutson, 2010: Impact of duration thresholds on Atlantic tropical cyclone counts. J. Climate, 23, 2508–2519, https://doi.org/10.1175/2009JCLI3034.1.
Larkin, N. K., and D. E. Harrison, 2005: Global seasonal temperature and precipitation anomalies during El Niño autumn and winter. Geophys. Res. Lett., 32, L16705, https://doi.org/10.1029/2005GL022860.
Lee, T.-C., T. R. Knutson, T. Nakaegawa, M. Ying, and E. J. Cha, 2020: Third assessment on impacts of climate change on tropical cyclones in the typhoon committee region—Part I: Observed changes, detection and attribution. Trop. Cyclone Res. Rev., 9 (1), 1–22, https://doi.org/10.1016/j.tcrr.2020.03.001.
Li, S., W. Mei, and S.-P. Xie, 2022: Effects of tropical sea surface temperature variability on Northern Hemisphere tropical cyclone genesis. J. Climate, 35, 4719–4739, https://doi.org/10.1175/JCLI-D-21-0084.1.
Lin, I.-I., and J. C. L. Chan, 2015: Recent decrease in typhoon destructive potential and global warming implications. Nat. Commun., 6, 7182, https://doi.org/10.1038/ncomms8182.
Lin, I.-I., and Coauthors, 2020: ENSO and tropical cyclones. El Niño Southern Oscillation in a Changing Climate, Geophys. Monogr., Vol. 253, Amer. Geophys. Union, 377–408, https://doi.org/10.1002/9781119548164.ch17.
Liu, K. S., and J. C. L. Chan, 2013: Inactive period of western North Pacific tropical cyclone activity in 1998–2011. J. Climate, 26, 2614–2630, https://doi.org/10.1175/JCLI-D-12-00053.1.
Matsuura, T., M. Yumoto, and S. Iizuka, 2003: A mechanism of interdecadal variability of tropical cyclone activity over the western North Pacific. Climate Dyn., 21, 105–117, https://doi.org/10.1007/s00382-003-0327-3.
Mei, W., and S.-P. Xie, 2016: Intensification of landfalling typhoons over the northwest Pacific since the late 1970s. Nat. Geosci., 9, 753–757, https://doi.org/10.1038/ngeo2792.
Mei, W., and S. Li, 2022: Variability and predictability of basinwide and sub-basin tropical cyclone genesis frequency in the northwest Pacific. J. Climate, 35, 6865–6884, https://doi.org/10.1175/JCLI-D-21-0232.1.
Mei, W., S.-P. Xie, and M. Zhao, 2014: Variability of tropical cyclone track density in the North Atlantic: Observations and high-resolution simulations. J. Climate, 27, 4797–4814, https://doi.org/10.1175/JCLI-D-13-00587.1.
Mei, W., S.-P. Xie, M. Zhao, and Y. Wang, 2015: Forced and internal variability of tropical cyclone track density in the western North Pacific. J. Climate, 28, 143–167, https://doi.org/10.1175/JCLI-D-14-00164.1.
Mei, W., Y. Kamae, S.-P. Xie, and K. Yoshida, 2019: Variability and predictability of North Atlantic hurricane frequency in a large ensemble of high-resolution atmospheric simulations. J. Climate, 32, 3153–3167, https://doi.org/10.1175/JCLI-D-18-0554.1.
Menkes, C. E., M. Lengaigne, P. Marchesiello, N. C. Jourdain, E. M. Vincent, J. Lefevre, F. Chauvin, and J.-F. Royer, 2012: Comparison of tropical cyclogenesis indices on seasonal to interannual timescales. Climate Dyn., 38, 301–321, https://doi.org/10.1007/s00382-011-1126-x.
Mizuta, R., and Coauthors, 2017: Over 5,000 years of ensemble future climate simulations by 60-km global and 20-km regional atmospheric models. Bull. Amer. Meteor. Soc., 98, 1383–1398, https://doi.org/10.1175/BAMS-D-16-0099.1.
Murakami, H., R. Mizuta, and E. Shindo, 2012: Future changes in tropical cyclone activity projected by multi-physics and multi-SST ensemble experiments using the 60-km-mesh MRIAGCM. Climate Dyn., 39, 2569–2584, https://doi.org/10.1007/s00382-011-1223-x.
North, G. R., T. L. Bell, R. F. Cahalan, and F. J. Moeng, 1982: Sampling errors in the estimation of empirical orthogonal functions. Mon. Wea. Rev., 110, 699–706, https://doi.org/10.1175/1520-0493(1982)110<0699:SEITEO>2.0.CO;2.
Patricola, C. M., S. J. Camargo, P. J. Klotzbach, R. Saravanan, and P. Chang, 2018: The influence of ENSO flavors on western North Pacific tropical cyclone activity. J. Climate, 31, 5395–5416, https://doi.org/10.1175/JCLI-D-17-0678.1.
Peduzzi, P., B. Chatenou, H. Dao, A. De Bono, C. Herold, J. Kossin, F. Mouton, and O. Nordbeck, 2012: Global trends in tropical cyclone risk. Nat. Climate Change, 2, 289–294, https://doi.org/10.1038/nclimate1410.
Saji, N. H., B. N. Goswami, P. N. Vinayachandran, and T. Yamagata, 1999: A dipole mode in the tropical Indian Ocean. Nature, 401, 360–363, https://doi.org/10.1038/43854.
Song, J., P. J. Klotzbach, N. Wei, and Y. Duan, 2022: Modulation of western North Pacific tropical cyclone formation by central Pacific El Niño–Southern Oscillation on decadal and interannual timescales. Int. J. Climatol., 43, 426–437, https://doi.org/10.1002/joc.7777.
Song, K., J. Zhao, R. Zhan, L. Tao, and L. Chen, 2022: Confidence and uncertainty in simulating tropical cyclone long-term variability using the CMIP6-HighResMIP. J. Climate, 35, 6431–6451, https://doi.org/10.1175/JCLI-D-21-0875.1.
Studholme, J., A. V. Fedorov, S. K. Gulev, K. Emanuel, and K. Hodges, 2022: Poleward expansion of tropical cyclone latitudes in warming climates. Nat. Geosci., 15, 14–28, https://doi.org/10.1038/s41561-021-00859-1.
Takahashi, C., M. Watanabe, and M. Mori, 2017: Significant aerosol influence on the recent decadal decrease in tropical cyclone activity over the western North Pacific. Geophys. Res. Lett., 44, 9496–9504, https://doi.org/10.1002/2017GL075369.
Takahashi, K., A. Montecinos, K. Goubanova, and B. Dewitte, 2011: ENSO regimes: Reinterpreting the canonical and Modoki El Niño. Geophys. Res. Lett., 38, L10704, https://doi.org/10.1029/2011GL047364.
Tang, B., and K. Emanuel, 2012: A ventilation index for tropical cyclones. Bull. Amer. Meteor. Soc., 93, 1901–1912, https://doi.org/10.1175/BAMS-D-11-00165.1.
Tao, L., L. Wu, Y. Wang, and J. Yang, 2012: Influence of tropical Indian Ocean warming and ENSO on tropical cyclone activity over the western North Pacific. J. Meteor. Soc. Japan, 90, 127–144, https://doi.org/10.2151/jmsj.2012-107.
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.
Wang, B., B. Xiang, and J.-Y. Lee, 2013: Subtropical high predictability establishes a promising way for monsoon and tropical storm predictions. Proc. Natl. Acad. Sci. USA, 110, 2718–2722, https://doi.org/10.1073/pnas.1214626110.
Wang, C., and B. Wang, 2019: Tropical cyclone predictability shaped by western Pacific subtropical high: Integration of trans-basin sea surface temperature effects. Climate Dyn., 53, 2697–2714, https://doi.org/10.1007/s00382-019-04651-1.
Wang, H., and C. Wang, 2022: What caused the increase of tropical cyclones in the western North Pacific during the period of 2011–2020? Climate Dyn., 60, 165–177, https://doi.org/10.1007/s00382-022-06299-w.
Webster, P. J., A. M. Moore, J. P. Loschnigg, and R. R. Leben, 1999: Coupled ocean–atmosphere dynamics in the Indian Ocean during 1997–98. Nature, 401, 356–360, https://doi.org/10.1038/43848.
Woodruff, J. D., J. L. Irish, and S. J. Camargo, 2013: Coastal flooding by tropical cyclones and sea level rise. Nature, 504, 44–52, https://doi.org/10.1038/nature12855.
Wu, L., H. Zhang, J.-M. Chen, and T. Feng, 2018: Impact of two types of El Niño on tropical cyclones over the western North Pacific: Sensitivity to location and intensity of Pacific warming. J. Climate, 31, 1725–1742, https://doi.org/10.1175/JCLI-D-17-0298.1.
Wu, R., Y. Yang, and X. Cao, 2019: Respective and combined impacts of regional SST anomalies on tropical cyclogenesis in different sectors of the western North Pacific. J. Geophys. Res. Atmos., 124, 8917–8934, https://doi.org/10.1029/2019JD030736.
Xie, S.-P., K. Hu, J. Hafner, H. Tokinaga, Y. Du, G. Huang, and T. Sampe, 2009: Indian Ocean capacitor effect on Indo-western Pacific climate during the summer following El Niño. J. Climate, 22, 730–747, https://doi.org/10.1175/2008JCLI2544.1.
Xie, S.-P., Y. Kosaka, Y. Du, K. Hu, J. S. Chowdary, and G. Huang, 2016: Indo-western Pacific Ocean capacitor and coherent climate anomalies in post-ENSO summer: A review. Adv. Atmos. Sci., 33, 411–432, https://doi.org/10.1007/s00376-015-5192-6.
Yan, Q., R. Korty, and Z. Zhang, 2015: Tropical cyclone genesis factors in a simulation of the last two millennia: Results from the Community Earth System Model. J. Climate, 28, 7182–7202, https://doi.org/10.1175/JCLI-D-15-0054.1.
Yan, Q., T. Wei, R. L. Korty, J. P. Kossin, Z. Zhang, and H. Wang, 2016: Enhanced intensity of global tropical cyclones during the mid-Pliocene warm period. Proc. Natl. Acad. Sci. USA, 113, 12 963–12 967, https://doi.org/10.1073/pnas.1608950113.
Yoshida, K., M. Sugi, R. Mizuta, H. Murakami, and M. Ishii, 2017: Future changes in tropical cyclone activity in high-resolution large-ensemble simulations. Geophys. Res. Lett., 44, 9910–9917, https://doi.org/10.1002/2017GL075058.
Yu, J., T. Li, Z. Tan, and Z. Zhu, 2016: Effects of tropical North Atlantic SST on tropical cyclone genesis in the western North Pacific. Climate Dyn., 46, 865–877, https://doi.org/10.1007/s00382-015-2618-x.
Zhan, R., Y. Wang, and X. Lei, 2011: Contributions of ENSO and east Indian Ocean SSTA to the interannual variability of northwest Pacific tropical cyclone frequency. J. Climate, 24, 509–521, https://doi.org/10.1175/2010JCLI3808.1.
Zhan, R., Y. Wang, and M. Wen, 2013: The SST gradient between the southwestern Pacific and the western Pacific warm pool: A new factor controlling the northwestern Pacific tropical cyclone genesis frequency. J. Climate, 26, 2408–2415, https://doi.org/10.1175/JCLI-D-12-00798.1.
Zhan, R., Y. Wang, and J. Zhao, 2019: Contributions of SST anomalies in the Indo-Pacific Ocean to the interannual variability of tropical cyclone genesis frequency over the western North Pacific. J. Climate, 32, 3357–3372, https://doi.org/10.1175/JCLI-D-18-0439.1.
Zhang, Q., L. Wu, and Q. Liu, 2009: Tropical cyclone damages in China 1983–2006. Bull. Amer. Meteor. Soc., 90, 489–496, https://doi.org/10.1175/2008BAMS2631.1.
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.
Zhang, W., G. A. Vecchi, G. Villarini, H. Murakami, A. Rosati, X. Yang, L. Jia, and F. Zeng, 2017: Modulation of western North Pacific tropical cyclone activity by the Atlantic meridional mode. Climate Dyn., 48, 631–647, https://doi.org/10.1007/s00382-016-3099-2.
Zhang, W., G. A. Vecchi, H. Murakami, G. Villarini, T. L. Delworth, X. Yang, and L. Jia, 2018: Dominant role of Atlantic multidecadal oscillation in the recent decadal changes in western North Pacific tropical cyclone activity. Geophys. Res. Lett., 45, 354–362, https://doi.org/10.1002/2017GL076397.
Zhao, H., and C. Wang, 2019: On the relationship between ENSO and tropical cyclones in the western North Pacific during the boreal summer. Climate Dyn., 52, 275–288, https://doi.org/10.1007/s00382-018-4136-0.
Zhao, J., R. Zhan, Y. Wang, and H. Xu, 2018: Contribution of interdecadal Pacific oscillation to the recent abrupt decrease in tropical cyclone genesis frequency over the western North Pacific since 1998. J. Climate, 31, 8211–8224, https://doi.org/10.1175/JCLI-D-18-0202.1.
Zhao, J., R. Zhan, Y. Wang, L. Jiang, and X. Huang, 2022: A multiscale-model based near-term prediction of tropical cyclone genesis frequency in the Northern Hemisphere. J. Geophys. Res. Atmos., 127, e2022JD037267, https://doi.org/10.1029/2022JD037267.
Zhao, M., and I. M. Held, 2012: TC-permitting GCM simulations of hurricane frequency response to sea surface temperature anomalies projected for the late twenty-first century. J. Climate, 25, 2995–3009, https://doi.org/10.1175/JCLI-D-11-00313.1.
