1. Introduction
It has long been recognized that tropical cyclone (TC) activity in all ocean basins has a strong interannual variability (e.g., see the discussion in Landsea 2000). The El Niño–Southern Oscillation (ENSO) and to some extent, quasi-biennial oscillation (QBO) are two large-scale phenomena documented to have some statistical relationship with the annual number of TCs in some of the ocean basins, with the most pronounced influences on Atlantic basin storm frequency (Gray and Sheaffer 1991; Lander and Guard 1998).
Gray (1979) found that the climatological frequency of TC genesis is related to six environmental factors. They include three thermodynamic parameters and three dynamic parameters. As noted by Frank (1987), the parameters cited above are not completely independent. In the Tropics, above-average midlevel moisture and conditional instability through a deep layer are strongly correlated with high sea surface temperature (SST). While the necessary thermodynamic and Coriolis conditions exist over the western North Pacific (WNP) for long periods of time, the low-level vorticity and vertical wind shear parameters can change significantly on much shorter timescales and smaller space scales (McBride and Zehr 1981). Thus, it has been hypothesized that tropical cyclogenesis occurs when above-normal low-level vorticity and locally weak vertical wind shear occurred within a thermodynamically favorable environment (Gray 1979).
The relative vorticity associated with the monsoon trough in the WNP is a vital factor for TC formation in that region. The favorable locations for TC genesis are in, or just poleward of, the intertropical convergence zone (ITCZ) or a monsoon trough (Gray 1968; Ramage 1974). In long-term averages of low-level wind flow and sea level pressure (Sadler et al. 1987), the monsoon trough of the WNP (during Northern Hemisphere summer) extends eastward from the south Asian low pressure trough and is accompanied by low-level southwesterly winds to the south of the trough axis. Ritchie (1995) found that slightly more than 75% of the classifiable genesis cases from 1990 to 1992 occurred in the monsoon trough.
The mean evolution of the monsoon trough and the western Pacific subtropical high (WPSH) are closely related (Lau and Li 1984; Wang and Wu 1997). From these studies and others it is well known that the WPSH moves to the southeast in winter. Conversely, the WPSH moves northward in June and is at its northernmost position near 40°N in August and September. The WPSH tends to withdraw in October and shrinks to its winter location in November (Ding 1994). When the WPSH retreats from the South China Sea (SCS) to the Philippine Sea and the western Pacific, the monsoon westerly winds penetrate from the Indian Ocean to SCS, the Philippine Sea, and the western Pacific. These conditions are more favorable for TC genesis over the western Pacific (Frank 1987).
It is well accepted that strong local vertical wind shear inhibits the formation and intensification of TCs. Modeling studies of tropical cyclogenesis confirm this effect (Kurihara and Tuleya 1981). A number of empirical studies (e.g., Gray 1979; Shapiro 1987) have also found that TC activity is strongly influenced by the vertical wind shear between the upper and lower levels of the atmosphere, for example, 200 and 850 hPa, respectively. In a review of TC genesis, McBride (1995) suggested that the seasonal distribution of TC genesis location was determined primarily by two factors: SST greater that 26.5°C and the location of the monsoon trough. This is consistent with other studies (e.g., Sadler 1967), which show that most TCs form in the shear zone between monsoon westerly winds and the trade easterly winds.
There is a rich and somewhat controversial background literature concerning interannual variability of TC activity over the WNP. Atkinson (1977) noted that an above-average number of TCs developed in the extreme eastern part of the WNP during 1972, an El Niño year. Using spectral analysis, Chan (1985) found that ENSO events are significantly correlated with TC activity in the northwest Pacific. Wu and Lau (1992) found a statistically significant ENSO-related signal in mean genesis locations of numerical model–simulated TCs. On the other hand, Ramage and Hori (1981) did not find any significant correlation between observed TC frequency and ENSO.
The relationship between ENSO events and TC genesis position in the northwest Pacific has also been studied. Chan (1985), Dong (1988), and Lander (1994) have documented reduced numbers of TCs west of 160°E, but increased cyclogenesis events east of 160°E and south of 20°N during El Niño events. These studies suggest that the opposite occurs during La Niña events. Lander (1994) further suggests that during a given year, the TC distribution and the preferred areas for genesis are governed primarily by the location and the behavior of the monsoon trough. His study, however, found no relationship between annual TC totals and ENSO.
Chen et al. (1998) found substantial variations in the TC genesis locations especially during the Northern Hemisphere summers (June–August) and attributed this to zonal and meridional shifts of the monsoon trough. Chan (2000) further found that TC activity over the northwest Pacific has large variations not only in El Niño and La Niña years, but also during the years before and the years after the events. He related variations in TC activity to variations of large-scale flow patterns associated with these events.
Many of these previous studies hypothesized relationships between ENSO and TC activity but the overall large-scale climate background associated with interannual variability of TC genesis remains unclear. In this study, we investigate the relationship of TC genesis position in the WNP from the viewpoint of large-scale circulation without making a priori assumptions of significant ENSO impact. We generalized the study to examine whether there are systematic variations in the large-scale atmospheric circulation associated with interannual variability in the region of TC genesis. We also performed a preliminary investigation of precursors in the large-scale circulation pattern for the purpose of monitoring and as an aid to seasonal prediction of TC activity.
The data used in this study and the analysis procedures are described in section 2. The mean annual cycle of TC genesis positions in the WNP is presented in section 3. Section 4 presents the composite of the various large-scale circulation patterns associated with interannual variability in the TC genesis region. The results are discussed in section 5.
2. Data and methodology
The historical record of TC activity in the WNP was obtained from the best-track archives of the Joint Typhoon Warning Center (JTWC), Guam. Each Annual Tropical Cyclone Report (ATCR) issued by the JTWC contains a summary of the best-track positions for all storms for which the JTWC issued a warning during that calendar year (McPherson and Stapler 1999). The JTWC summary contains measurements of latitude, longitude, and maximum wind speed at 6-h intervals from 1945 to 1999. Because satellite observations of TC activity and tracks are unavailable in significant quantities prior to 1966 (Frank 1987) and routine coverage of the western North Pacific by geostationary satellite did not start until 1979, this study is limited to the period from 1979 to 1999.
The following data are also used: the monthly National Centers for Environmental Prediction–National Center for Atmospheric Research (NCEP–NCAR) reanalysis for the period 1979–99 (Kalnay et al. 1996), the monthly interpolated outgoing longwave radiation (OLR) obtained from the Climate Diagnostics Center (Liebmann and Smith 1996), and the reconstructed Reynolds SST produced by NCEP (Smith et al. 1996). ENSO is characterized by the Niño-3.4 (5°S–5°N, 120°–70°W) index, as suggested by Barnston et al. (1997), and obtained from the NCEP Climate Prediction Center (CPC).
The historical records show that western North Pacific TCs have occurred in every month of the calendar year, but the mean minimum and absolute minimum in the number of TCs occur in February. In the mean, the peak TC season over the WNP runs from July to October (Fig. 1). Almost 70% of the TCs form during this peak season. In this study, we focus the analysis on the July–October “season,” rather than the annual TC occurrences as done in several other studies. The seasonal mean TC genesis position (MGP) in the western North Pacific for the period 1979–99 was determined from the JTWC best-track dataset. Genesis is defined as the first TC position, for a particular TC that attains tropical depression (TD) strength. In the JTWC dataset a TD is defined as a TC having a maximum sustained (1-min mean) surface wind <18 m s−1. The category of typhoon intensity is based on the Saffir–Simpson scale (Simpson 1974). Intense typhoons correspond to categories 3, 4, and 5 on the Saffir–Simpson scale (Table 1).
The seasonal mean circulation, OLR, and SST were derived from the 1979–99 data. Anomalies from the 1979–99 base period means were used in computations of the composites discussed here.
3. Mean TC genesis positions
The monthly MGP for the period 1979–99, Fig. 2, and the numbers of observed TCs, during this period, Table 2, suggests a well-defined annual cycle. The February mean position appears to be an outlier but it is based on only four TC genesis positions (Table 2) over the entire 21-yr period. January, March, and April also have a relatively small number of TCs, averaging less than one TC per year during the 1979–99 period. Other months varied from a minimum of 26 TCs in May to a maximum of 137 TCs in August. In this map projection, the monthly MGP travels linearly from its southernmost and easternmost position, near 5°N and 158°E in March, to the westernmost mean TC genesis position, near 135°E in June. This evolution parallels the seasonal movement of the principal tropical convection system over the western North Pacific (Atkinson 1977; Murakami and Matsumoto 1994; Chen and Chen 1995). The MGP travels northeast from June to August, reaching to near 17°N then moves roughly south-southeast, reaching to about 8°N in December.
Atkinson (1977) noted a strong seasonality in the mean axis of the monsoon trough in the western North Pacific (Fig. 3). The relationship between the mean position of the monsoon trough and the MGP is apparent during the core TC season (June–October) and even into November (i.e., Figs. 2 and 3). The monthly MGP appears to be closely associated with the movement of the mean monsoon trough from June to August. The mean trough axis reaches its highest latitude during August coincident with the northernmost excursion of the MGP. However, the easternmost penetration of monsoon westerly wind occurs in November (Fig. 3). It seems that the monthly MGP tends to lag the north-to-south shift of monsoon trough from August to November by a few degrees of latitude. The seasonal MGP for the 1979–99 period, near 15°N and 143°E (see Fig. 4), is only a few hundred kilometers to the northwest of the long-term mean eastward penetration of the monsoon trough of the WNP during summer.
4. Interannual variability of TC genesis
The seasonal (July–October) MGP for each year (Fig. 4) provides one way to characterize interannual variability. The overall 1979–99 seasonal MGP, based on 446 TCs, is near 15°N, 143°E (Fig. 4). The seasonal MGPs tend to align along a northwest–southeast axis with respect to the 1979–99 mean seasonal position. Years in which the seasonal MGP lies within one standard deviation of the seasonal MGP, a circle of radius 333 km or approximately 3° of latitude, are defined here as normal years. Two distinct TC genesis groups can be defined by dividing the domain to quadrants with the mean seasonal MGP as the origin (see Fig. 4). Group B contains the years when the MGP falls in the upper-left-hand quadrant (1981, 1985, 1996, 1998, and 1999). Group D contains the years where the MGP falls in the lower-right-hand quadrant, which includes 1987, and the relatively prolonged warm period years of 1991–95 (Trenberth and Hoar 1996). We note, however, than in 1995, the monthly Niño-3.4 index switched to negative in August, so the seasonal mean Niño-3.4 was negative, that is, slightly on the La Niña side for more than half the season of interest to this study. The mean positions of the group B years and group D years are both roughly a distance of 6° of latitude (i.e., two standard deviations) from the overall mean position.
The tendency toward a northwest–southeast alignment of the seasonal mean TC genesis positions suggests that some physical mechanisms are likely associated with the interannual variability in TC genesis position. Some of these are examined below, starting with SST. The Spearman's rank cross correlation between seasonal mean longitude (latitude) and the Niño-3.4 index is 0.469 (−0.512), both significant at the 95% level (Fig. 5). Thus, when the July–October Niño-3.4 index is positive (negative), the seasonal MGP tends to be southeast (northwest) of the long-term mean genesis position.
The southeastward shift of the genesis position associated with the positive Niño-3.4 index is clear for some ENSO episodes; for example, during the 1987 El Niño a Niño-3.4 index of 1.7 was associated with an extreme southeastward displacement of the MGP. However, during the strongest El Niño events of the twentieth century (1982, 1997), when the seasonal Niño-3.4 indices reached values of 1.5 and 2.3, respectively, the seasonal MGP remained near the long-term mean. Both in late 1982 and 1997, episodes were associated with an eastward displacement of the monsoon trough far to the east of the date line (not shown) and the genesis of TCs far to the east of our analysis area, which was arbitrarily restricted to west of the international date line. If our analysis area had included TC genesis much farther to the east and included January–December TCs, then the annual MGP for both 1982 and 1997 would have shifted eastward outside of the “normal” area, consistent with the observed displacements of MGP during other El Niño years.
The northwestward shift of the seasonal MGP with the negative values of the Niño-3.4 index is also clear for 1998 and 1999, with Niño-3.4 index values of −1.2 and −1.0, respectively. We note that with the exception of 1983, a transition year, and 1988, with a seasonal Niño-3.4 index value of −1.5, all of seasonal MGPs that fall outside of the 3° standard deviation radius are in the third and first quadrant.
In the mean there are more occurrences of intense typhoons (Saffir–Simpson scale categories 3–5, Table 1) in TCs forming to the southeast (group D) than to the northwest (group B). This is significant at the 95% level according to the t test (Fig. 6). However, there is no significant intergroup difference for the occurrences of the less intense typhoons, TD through category-2 typhoons. In addition, the storm tracks for the intense typhoons (categories 3–5, not shown) show large differences between the tracks for B and the tracks for D. Compared to B, the intense typhoons from D last longer and have longer tracks. Those differences in the seasonal MGP, TC intensity, and tracking suggest that there may be differences in large-scale circulation patterns associated these two genesis regions. The relationships between seasonal MGP and large-scale circulation features are discussed further in section 4.
5. Large-scale features
a. Vertical wind shear
The vertical wind shear is defined here as the difference of zonal winds between 200 and 850 hPa. Climatologically, the WNP basin contains a region of low vertical wind shear during the July–October season (Fig. 7), with the mean zero shear line extending to 178°E. The magnitude of the area mean vertical wind shear over the WNP basin is less than 4 m s−1 over the mean TC genesis regions for the July–October period. The climatologically low values of vertical wind shear in the WNP combined with the very warm SSTs contribute to the reason that this area is a more active region for TC formation compared to other basins.
Further analysis (Fig. 8) reveals large differences of vertical wind shear anomaly structure in the WNP between composites for group B and D. For group B (Fig. 8a), strong vertical wind shear anomalies dominate the southern part of the WNP, inhibiting TC genesis to the southeast of the 1979–99 MGP. In contrast, the group D composite (Fig. 8b) shows a weak vertical wind shear anomaly to the southeast of the 1979–99 MGP. Figure 8 also shows that the mean seasonal MGP of each group is located in the weak negative vertical wind shear anomaly area; that is, strong vertical shear inhibits tropical cyclogenesis while weak vertical shear favors tropical cyclogenesis and possible typhoon development (Goldenberg and Shapiro 1996).
For group B TCs, the vertical wind shear anomaly pattern over the western Pacific region is associated with strong upper-level westerly wind anomalies from about 140°E to the date line (Fig. 9a) and low-level easterly wind anomalies (Fig. 10a) that appear to be associated with tropical heating anomalies (Fig. 9a). Large east-to-west shifts in the equatorial Pacific convection patterns are suggested in the positive–negative anomaly “dipole” pattern in the OLR (Fig. 9a). A region of enhanced convection extends from about 85° to 135°E in the Southern Hemisphere, with a drier area and reduced convective activity from 150°E to 160°W (Fig. 9a).
The development of the anomalous low-level anticyclone in the Philippine Sea (Fig. 10a) associated with the easterly wind anomaly in the lower-level wind field indicates that the monsoon trough is weaker and retreats westward thus favoring the northwest quadrant for TC genesis as would be expected for a group B composite. In contrast, an anomalous cyclone in the Philippine Sea is associated with low-level westerly wind anomaly and can be seen in the group D composites (Fig. 10b), indicating an intensification and eastward extension of the monsoon trough. As a result, favorable conditions for TC genesis tend to shift southeast. Another notable feature in the group B (D) composite (Fig. 10) is the tropical twin anomalous anticyclone (cyclone) features with reduced (enhanced) cross-equator flow from the Southern Hemisphere indicating a weak (strong) Southern Hemisphere winter southeast monsoon. This is consistent with Love (1985) who noted that the cross-equatorial flow from the winter hemisphere could influence the genesis of TCs in the summer hemisphere.
Another feature of the low-level flow (Fig. 10) is the anomalous negative relative vorticity (anticyclone cell) near 35°N in the group B composite, indicating that the WPSH ridge extends northward. This is in contrast to the anomalous negative relative vorticity near 30°N in the group D composite, indicating that the WPSH ridge is stronger and extends southward. The strengthened (weakened) WPSH ridge along 30°N for group D (B) and the enhanced (reduced) monsoon trough in the Philippine Sea lead to an eastward (westward) and southward (northward) displacement of the major TC genesis pattern. These patterns are reminiscent of ENSO composites of Wang (1995).
An inspection of the monthly mean wind shear anomaly patterns, within groups B and D (not shown), reveals considerable similarities between the seasonal mean shear anomalies and the anomalies for each of the months that went into the respective composites, suggesting that these are very stable and robust large-scale circulation features. Further, for group D (B) the wind shear anomaly patterns become established in the preceding January (April), well in advance of the main TC season. This further suggests that the vertical wind shear anomalies over the area are strongly influenced by low-frequency variability lasting for more than one or two seasons, again consistent with an ENSO influence, and providing a potential precursor index for monitoring and the prediction of major TC genesis regions.
In the next section, the variation of the SST and monsoon trough with respect to TC genesis is examined through composites of the SST and westerly wind at 850 hPa.
b. Sea surface temperature and 850-hPa zonal winds
Composite July–October seasonal SST anomalies for group B and D years are presented in Figs. 11a,b. The composite for the group B TCs features negative SST anomalies across the eastern tropical Pacific to central tropical Pacific Ocean (Fig. 11a) straddled by near zero or weak positive SST anomalies to the north and south. As expected, this pattern is reminiscent of composite La Niña conditions with the additional feature of relatively strong positive SST anomalies along the coast of east Asia down to the South China Sea (Fig. 11a).
In contrast to the negative central equatorial SST anomalies for group B, the composites for group D show the large positive SST anomaly along the date line (Fig. 11b). As expected, this pattern is reminiscent of El Niño composites but with the largest positive SST anomalies situated near the date line and almost no anomalies in the eastern Pacific. This SST anomaly pattern does not occur in the evolution of every El Niño, suggesting that El Niño is not always a significant factor for TC genesis during the July–October season. The equatorial central Pacific SST warming enhances convection near the equatorial date line (e.g., Figs. 10b and 11b), which is consistent with a Rossby wave–like response (Gill 1980) associated with the anomalous cyclone over the Philippine Sea (e.g., Fig. 10b). The differences between SST anomalies for the group B TCs and group D TCs are large for the equatorial central Pacific around the date line but negligible for the east Asian coast and the western North Pacific.
The climatology of the seasonal SST shows that very warm SSTs (i.e., >29°C) generally characterize the western North Pacific during the July–October season (not shown). The composite 29°C contour shows a significant westward retreat of the warmest waters in the group B years and an eastward extension of the warmest water associated with the group D years, but no changes in the genesis regions (Fig. 12a). Thus, in the mean, the local SSTs are very favorable for TC formation according to McBride's (1995) criteria and SST variations in the genesis areas are likely not a factor in the interannual variability of MGP.
The 850-hPa wind composites show a large eastward extension of the westerly winds in the group D composite (Fig. 12b). The composite westerlies for the group D years extend 20° of longitude farther east than the westerlies for the group B composite. This suggests an enhancement of the monsoon trough for group D versus a reduction of the monsoon trough for group B. These differences in the monsoon trough make conditions more favorable for TC genesis in the group D cases and, in contrast, tend to move the major TC genesis region more westward in the group B. The intensification and eastward extension of monsoon trough in the group D is also apparent in the sea level pressure (not shown).
6. Discussion
The western North Pacific is a very active region for TC formation particularly during the July–October season. The seasonal MGP shows considerable interannual variability. Analysis of the TCs during the 1979–99 period suggest that the seasonal MGP are related by the 200–850-hPa wind shear, the position and strength of the monsoon trough, the position and strength of the WPSH, and the SST. Each of these circulation features as well as the SST are, in turn, related to ENSO suggesting that, while ENSO may not be the sole determinant of interannual variability in the MGP, it is a major factor during the July–October season.
The analysis of the July–October mean MGP shows two contrasting anomalous regions for TC genesis and one cluster around the climatological seasonal MGP. One set of TCs (group B) tends to form to the northwest of the climatological seasonal TC genesis position. The other set of TCs (group D) tends to form to the southeast of the climatological seasonal TC genesis position. The group B (D) TCs tend to be associated with La Niña (El Niño) but the correspondence is not perfect. In particular, the MGP for 1988, a strong La Niña year, does not conform to the overall pattern, being the only major outlier that falls in quadrant A, that is, to the northeast of the overall mean MGP. Our analysis also suggests that more group D TCs tend to become intense typhoons (Saffir–Simpson categories 3–5) than group B TCs. The group D storms also tend to be longer lived and with longer storm tracks.
The interannual modulation of the TC genesis regions, represented by groups B and D in this study may be related to the interannual variation of the equatorial central Pacific heating and its Rossby wave response. This response may manifest itself as changes in the circulation features over the Philippine Sea as suggested in the composite wind fields (Figs. 9 and 10); for example, the Rossby wave response may be related to a tendency for low-level anomalous cyclone or anticyclone development.
While large SST anomalies in the central equatorial Pacific appear to be a factor in TC genesis the composite analysis suggests that any relationship to ENSO and TC genesis is not simple. This is manifest most clearly in the group D composite SST (Fig. 11b), which resembles the Rasmusson and Carpenter (1982) mature El Niño phase. Thus the relationship between ENSO and MGP is likely complicated by the differences in the timing and evolution of individual ENSOs with respect to the peak of the TC season; for example, the mature phase of ENSO tends to be in winter (Rasmusson and Carpenter 1982; Wang 1995), but the peak of the typhoon season runs from July to October.
In recent years the interdecadal climate shift in the late 1970s exhibits global-scale variations in SST and has a significant impact on ENSO onset structure (Wang 1995). Based on composites, Wang (1995) noted that the composite El Niño episodes after the late 1970s were characterized by an anomalous cyclone over the Philippine Sea and associated anomalous westerlies in the western equatorial Pacific in the antecedent (June–October of year −1) and onset phase (November of year −1 to January of year 0). It is possible that the same interdecadal climate shift has influenced the TC activity in the western North Pacific during the period of our study, a topic that needs further investigation.
The interannual variability of TC genesis positions appears to be complex and related to wind shear, the monsoon trough–WPSH circulation system, as well as the SST. All of these elements of the climate system are interrelated. It seems likely that the same factors that influence the evolution of the individual ENSOs and/or the interdecadal changes in ENSO are also responsible for observed shifts in seasonal typhoon genesis position, a topic for further investigation.
Acknowledgments
This work was performed while the lead author was a visiting research scientist at the IRI and was supported by the Taiwan Grant NSC 89-2119-M-002-003. We wish to thank Benno Blumenthal, IRI, for his help with the data. We also wish to thank our other colleagues at the IRI and Mong-Ming Lu at the CWB for their helpful discussions and suggestions. Thanks to both reviewers and the editor for their helpful comments. We especially wish to thank one of the anonymous reviewers for the insightful comments and suggestions with respect to the 1982 and 1997 El Niño episodes.
The IRI was established as a cooperative agreement between the NOAA Office of Global Programs and Columbia University. Taiwan joined the IRI in 2000.
REFERENCES
Atkinson, G. D., 1977: Forecasters guide to tropical meteorology. U.S. Air Force Tech. Rep. 240, Air Weather Service [Military Airlift Command (MAC)], 360 pp.
Barnston, A. G., M. Chelliah, and S. B. Goldenberg, 1997: Documentation of a highly ENSO-related SST region in the equatorial Pacific. Atmos.–Ocean, 35 , 367–383.
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.
Chan, J. C. L., . 2000: Tropical cyclone activity over the western North Pacific associated with El Niño and La Niña events. J. Climate, 13 , 2960–2927.
Chen, T-C., and J-M. Chen, 1995: An observational study of the South China Sea monsoon during the 1979 summer: Onset and life cycle. Mon. Wea. Rev., 123 , 2295–2318.
Chen, T-C., S-P. Weng, N. Yamazaki, and S. Kiehne, 1998: Interannual variation in the tropical cyclone formation over the western North Pacific. Mon. Wea. Rev., 126 , 1080–1090.
Ding, Y., 1994: Monsoons over China. Atmospheric Sciences Library, Vol. 16, Kluwer, 419 pp.
Dong, K., 1988: El Niño and tropical cyclone frequency in the Australian region and the northwest Pacific. Aust. Meteor. Mag., 36 , 219–255.
Frank, W. M., 1987: Tropical cyclone formation. A Global View of Tropical Cyclone, R. L. Elsberry et al., Eds., Naval Postgraduate School, 53–90.
Gill, A. E., 1980: Some simple solutions for heat-induced tropical circulation. Quart. J. Roy. Meteor. Soc., 106 , 447–462.
Goldenberg, S. B., and L. J. Shapiro, 1996: Physical mechanisms for the association of El Niño and West African rainfall with Atlantic major hurricanes. J. Climate, 9 , 1169–1187.
Gray, W. M., 1968: Global view of the origin of tropical disturbance and storms. Mon. Wea. Rev., 96 , 669–700.
Gray, W. M., . 1979: Hurricanes: Their formation, structure and likely role in the general circulation. Meteorology over the Tropical Oceans, D. B. Shaw, Ed., Royal Meteorological Society, 155–218.
Gray, W. M., and J. D. Sheaffer, 1991: El Niño and QBO influences on tropical cyclone activity. Teleconnections Linking Worldwide Climate Anomalies, M. H. Glantz, R. W. Katz, and N. Nicholls, Eds., Cambridge University Press, 257–284.
Kalnay, E., and Coauthors. 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77 , 437–471.
Kurihara, Y., and R. E. Tuleya, 1981: A numerical simulation study on the genesis of a tropical storm. Mon. Wea. Rev., 109 , 1629–1653.
Lander, M. A., 1994: An exploratory analysis of the relationship between tropical storm formation in the western North Pacific and ENSO. Mon. Wea. Rev., 122 , 636–651.
Lander, M. A., and C. P. Guard, 1998: A look at global tropical cyclone activity during 1995: Contrasting high Atlantic activity with low activity in other basins. Mon. Wea. Rev., 126 , 1163–1173.
Landsea, C. W., 2000: El Niño–Southern Oscillation and the seasonal predictability of tropical cyclones. El Niño: Impacts of Multiscale Variability on Natural Ecosystems and Society, H. F. Diaz and V. Markgraf, Eds., Cambridge University Press, 149–181.
Lau, K-M., and M. T. Li, 1984: The monsoon of East Asia and its global associations—A survey. Bull. Amer. Meteor. Soc., 65 , 114–125.
Liebmann, B., and C. A. Smith, 1996: Description of a complete (interpolated) outgoing longwave radiation dataset. Bull. Amer. Meteor. Soc., 77 , 1275–1277.
Love, G., 1985: Cross-equatorial interactions during tropical cyclogenesis. Mon. Wea. Rev., 113 , 1499–1509.
McBride, J. L., 1995: Tropical cyclone formation. Global Perspective on Tropical Cyclones, WMO Tech Doc. 693, World Meteorological Organization, 63–105.
McBride, J. L., and R. M. Zehr, 1981: Observational analysis of tropical cyclone formation. Part II: Comparison of non-developing versus developing systems. J. Atmos. Sci., 38 , 1132–1151.
McPherson, T., and W. Stapler, Eds.,. 1999: Summary of Northwest Pacific tropical cyclones, 1999. Annual Tropical Cyclone Report, Joint Typhoon Warning Center, 5–6.
Murakami, T., and J. Matsumoto, 1994: Summer monsoon over the Asia continent and western North Pacific. J. Meteor. Soc. Japan, 72 , 719–745.
Ramage, C. S., 1974: Monsoonal influences on the annual variation of tropical cyclone development over the Indian and Pacific Oceans. Mon. Wea. Rev., 102 , 745–753.
Ramage, C. S., and A. M. Hori, 1981: Meteorological aspects of El Niño. Mon. Wea. Rev., 109 , 1827–1835.
Rasmusson, E. M., and T. H. Carpenter, 1982: Variations in tropical sea surface temperature and surface wind fields associated with the Southern Oscillation/El Niño. Mon. Wea. Rev., 110 , 354–384.
Ritchie, E. A., 1995: Mesoscale aspects of tropical cyclone formation. Ph.D. dissertation, Center for Dynamical Meteorology and Oceanography, Monash University, Melbourne, Australia, 167 pp.
Sadler, J. C., 1967: On the origin of tropical vortices. Proc. Working Panel on Tropical Dynamic Meteorology, Norfolk, VA, Naval Weather Research Facility, 39–75.
Sadler, J. C., M. A. Lander, A. M. Hori, and L. K. Oda, 1987: Pacific Ocean. Vol. 2, Tropical Marine Climate Atlas. UHMET Publ. 87-02, Department of Meteorology, University of Hawaii, Honolulu, HI, 27 pp.
Shapiro, L. J., 1987: Month-to-month variability of the Atlantic tropical circulation and its relationship to tropical storm formation. Mon. Wea. Rev., 115 , 2598–2614.
Simpson, R. H., 1974: The hurricane disaster potential scale. Weatherwise, 27 , 169–186.
Smith, T. M., R. M. Reynolds, R. E. Livezey, and D. C. Stokes, 1996: Reconstruction of historical sea surface temperature using empirical orthogonal function. J. Climate, 9 , 1403–1420.
Trenberth, K. E., and T. J. Hoar, 1996: The 1990–1995 El Niño–Southern Oscillation event: Longest on record. Geophys. Res. Lett., 23 , 57–60.
Wang, B., 1995: Interdecadal changes in El Niño onset in the last four decades. J. Climate, 8 , 267–285.
Wang, B., and R. Wu, 1997: Peculiar temporal structure of the South China Sea summer monsoon. Adv. Atmos. Sci., 14 , 177–194.
Wu, G., and N-C. Lau, 1992: A GCM simulation of the relationship between tropical storm formation and ENSO. Mon. Wea. Rev., 120 , 958–977.
The average monthly number of TCs over the WNP from 1959 to 1999 (McPherson and Stapler 1999). The Jul–Oct season accounts for an average of almost 70% of the annual total
Citation: Journal of Climate 15, 20; 10.1175/1520-0442(2002)015<2934:TIVITG>2.0.CO;2
Monthly mean TC genesis position from 1979 to 1999. The number indicates the month, e.g., 1 = Jan
Citation: Journal of Climate 15, 20; 10.1175/1520-0442(2002)015<2934:TIVITG>2.0.CO;2
Typical migration of the axis of the monsoon trough indicated by its mean monthly positions during Jun–Nov (after Atkinson 1971)
Citation: Journal of Climate 15, 20; 10.1175/1520-0442(2002)015<2934:TIVITG>2.0.CO;2
The seasonal mean (Jul–Oct) TC genesis position from 1979 to 1999. (“87” is the seasonal mean genesis position for 1987, etc.) The triangle shows the long-term (1979–99) seasonal (Jul–Oct) mean genesis position at the center of circle. The radius (333 km, i.e., near 3° of lat) is the std dev of the seasonal TC genesis positions. Letters B and D indicate the mean TC genesis positions for each of these composites
Citation: Journal of Climate 15, 20; 10.1175/1520-0442(2002)015<2934:TIVITG>2.0.CO;2
The (a) lon and (b) lat of the seasonal mean TC position as a function of the seasonal mean Niño-3.4 index. The linear correlation coefficient is given in both (a) and (b). The letter M indicates the mean lon, lat, and Niño-3.4 index for the 1979–99 period
Citation: Journal of Climate 15, 20; 10.1175/1520-0442(2002)015<2934:TIVITG>2.0.CO;2
Average number of TCs (1979–99) based on the Saffir–Simpson scale for each of the genesis groups, B, D, and N for neither. TD: tropical depression; TS: tropical storms; C1–C5: category 1–5 typhoons
Citation: Journal of Climate 15, 20; 10.1175/1520-0442(2002)015<2934:TIVITG>2.0.CO;2
Climatological (Jul–Oct) vertical wind shear pattern (m s−1) based on NCEP–NCAR reanalysis (Kalnay et al. 1996). The vertical wind shear is defined here as the difference of zonal winds between 200 and 850 hPa. The contour interval is 4 m s−1. The 1979–99 mean TC genesis region is indicated by the triangle–dot (shower) symbol
Citation: Journal of Climate 15, 20; 10.1175/1520-0442(2002)015<2934:TIVITG>2.0.CO;2
Composite seasonal vertical wind shear anomaly based on NCEP–NCAR reanalysis (Kalnay et al. 1996) for (a) group B (1981, 1985, 1996, 1998, and 1999) and (b) group D (1987, 1991–95), respectively. Areas where a t test indicates significant differences from the mean shear at the 95% level or above are shaded. The 1979–99 mean TC genesis region is indicated by the triangle–dot (shower) symbol in both (a) and (b). The letters B and D indicate the composite mean TC genesis position for each group
Citation: Journal of Climate 15, 20; 10.1175/1520-0442(2002)015<2934:TIVITG>2.0.CO;2
Composite of the seasonal (Jul–Oct) 200-hPa wind vector anomaly (m s−1) based on NCEP–NCAR reanalysis (Kalnay et al. 1996) and OLR anomaly (Liebmann and Smith 1996) for (a) group B and (b) for group D. Solid contours indicate reduced convection regions and dashed contours indicate enhanced convection regions. The contour interval of the OLR anomaly is 5 W m−2. Areas where a t test indicates significant differences from the period mean winds at the 95% level or above are shaded. The 1979–99 mean TC genesis region is indicated by the triangle–dot (shower) symbol. The letters B and D indicate the composite mean TC position in each group
Citation: Journal of Climate 15, 20; 10.1175/1520-0442(2002)015<2934:TIVITG>2.0.CO;2
Composite of the seasonal 850-hPa wind vector anomaly (m s−1) and relative vorticity anomaly (10−6 s−1; contours) based on NCEP–NCAR reanalysis (Kalnay et al. 1996). Areas where a t test indicates significant differences from the mean winds at the 95% level or above are shaded. The 1979–99 mean TC genesis region is indicated by the triangle–dot (shower) symbol. The letters B and D indicate the composite mean TC position in each group
Citation: Journal of Climate 15, 20; 10.1175/1520-0442(2002)015<2934:TIVITG>2.0.CO;2
Composite of the seasonal sea surface temperature (SST) anomaly based on SST from Smith et al. (1996): (a) group B and (b) group D. The contour interval is 0.2°C. Areas where a t test indicates significant differences from the mean shear at the 95% level or above are shaded. The 1979–99 mean TC genesis region is indicated by the triangle–dot (shower) symbol. The letters B and D indicate the composite mean TC position in each group.
Citation: Journal of Climate 15, 20; 10.1175/1520-0442(2002)015<2934:TIVITG>2.0.CO;2
Composite seasonal (Jul–Oct): (a) 29°C SST contours, and (b) zero line of the zonal wind at 850 hPa. The solid line represents the group D composite; the short dashed line indicates the group B composite, and long dashed line indicates the climatology. The 1979–99 mean TC genesis region is indicated by the triangle–dot (shower) symbol. The letters B and D indicate the composite mean TC position for the groups B and D
Citation: Journal of Climate 15, 20; 10.1175/1520-0442(2002)015<2934:TIVITG>2.0.CO;2
Minimum surface pressure and maximum sustained wind speed for the Saffir–Simpson (Simpson 1974) typhoon-scale values
The number of TCs by month from 1979 to 1999
