1. Introduction
During the past decade or so, many researchers have suggested that the buildup of greenhouse gases (Watson et al. 2001) will likely result in a rise in sea surface temperature (SST), subsequently increasing both the number and maximum intensity of tropical cyclones (TCs). In most of the studies that made use of numerical climate models, the conclusion is that TC activity will increase when the local or underlying SST increases (Broccoli and Manabe 1990; Haarsma et al. 1993; Krishnamurti et al. 1998). However, Bengtsson et al. (1996) found that the number of cyclones decreases in a doubled-CO2 world. More recent studies using models with higher resolution further suggested that in a doubled-CO2 world, the maximum intensity of TCs will likely increase (Krishnamurti et al. 1998; Knutson et al. 1998; Knutson and Tuleya 1999). In simulations of TC intensity, an important consideration is its interaction with the underlying ocean. The strong winds associated with a TC tend to upwell cool waters from below to the ocean surface, which reduces the amount of energy available for the TC to intensify (Shay et al. 1992; Chan et al. 2001). Therefore, some studies have also incorporated ocean coupling into the model formulation (Shen et al. 2000; Knutson et al. 2001). However, the conclusions are almost completely contradictory in these two studies. The discrepancies among the various model simulations highlight the fact that depending on the characteristics of the model (horizontal and vertical resolution, physical processes, etc.), the simulated results could be dramatically different. Therefore, drawing conclusions of what may happen in the future based on these numerical simulations would be difficult.
An alternative to running numerical simulations of future climate scenarios is to examine past data. If the frequency of TC occurrence were to increase with increasing global air temperature, one would expect to see an increase in the number of TCs during the past few decades. While Landsea et al. (1996) found a decrease in the number of intense Atlantic hurricanes between 1991 and 1994, Goldenberg et al. (2001) showed a doubling of overall hurricane activity in the North Atlantic during the period 1995–2000. In the western North Pacific (WNP), Chan and Shi (1996) did identify an upward trend in the number of TCs in the early 1990s, but that trend has recently reversed (see example in Fig. 1a). In other words, interdecadal variations in TC activity are apparently quite significant. Indeed, the large interdecadal variations shown in Fig. 1 would likely overwhelm any modification of the frequency due to global warming, which is the conclusion of Lighthill et al. (1994).
In terms of TC intensity, Evans (1993) found no statistically significant relationship between SST and the average intensity of TCs, although she did find that individual TCs were generally more intense over waters of higher temperature. DeMaria and Kaplan (1994) also obtained a statistical relationship between maximum intensity and SST for Atlantic TCs. Similar conclusions have been drawn for WNP typhoons (Kuroda et al. 1998) and eastern North Pacific hurricanes (Whitney and Hobgood 1997). However, all such observational studies investigated the relationship between individual TCs and the underlying ocean rather than the variations in the average intensity of TCs under oceanwide changes in SST, as are usually the focus of climate models.
In order to make a meaningful comparison with the model results, one would need to study the relationship between SST over the entire ocean basin and the maximum intensity of TCs on an individual year basis, which is the objective of this paper. Tropical cyclones in the western North Pacific are chosen here because more intense TCs tend to occur in that ocean basin (e.g., Hope 1979). As the focus is on intense TCs, only the frequency of typhoon (maximum sustained 1-min mean surface winds >33 m s−1) occurrence is examined. Parameters representing the typhoon activity and SST are first defined in section 2. Temporal variations of these parameters are examined in section 3 to identify trends on interdecadal time scales. In section 4, statistical correlations between these time series are then presented. In addition, variations in typhoon activity over the western North Pacific are correlated with those of SST over the equatorial eastern Pacific Ocean, the latter often being used as an indicator of whether an El Niño event is taking place. The impact of large-scale parameters on annual typhoon activity and their relationship with SST are then described in section 5. The paper concludes in section 6 with a discussion of what one should consider in studying how future climate change might affect TC frequency and intensity.
2. Data and definitions
a. TC data
TCs that occurred during the period 1960–20031 over the WNP form the basic dataset for this study. These were obtained from the Web site of the Joint Typhoon Warning Center (http://www.npmoc.navy.mil/jtwc/ best_tracks/;). As the focus of this study is on typhoon activity, the following parameters are used as measures of such activity:
Number of typhoons (NTY).
The ratio of NTY (RTY) to the total number of TCs (NTC) during the same year, that is, RTY = NTY/ NTC.
Typhoon destruction potential (TDP), defined as the sum of the squares of a typhoon's maximum wind speed for each 6-h period of its existence.
b. SST data
The SST data are from the National Centers for Environmental Prediction–National Center for Atmospheric Research (NCEP–NCAR) reanalysis dataset. The horizontal resolution of the dataset is 2° latitude × 2° longitude. Since most of the TCs over the WNP generally reached their maximum intensity south of 30°N and east of 120°E (Xue and Neumann 1984), the SST within the domain 5°–30°N, 120°E–180° are considered. In addition, because over 85% of the WNP TCs occur between May and November (Xue and Neumann 1984), only data within these seven months are used.
c. Atmospheric data
Monthly atmospheric data from the NCEP–NCAR reanalysis dataset are used to compute the large-scale parameters including relative vorticity, vertical shear of zonal wind, and moist static energy. The horizontal resolution of the dataset is 2.5° latitude × 2.5° longitude.
3. Variations in typhoon activity parameters and SST
a. Variations in typhoon activity parameters
The time series of the three parameters describing annual typhoon activity show significant interannual as well as interdecadal variations. Typhoon activity was very high during 1960–68, with an average of 21 typhoons (Fig. 1a). Although a large variability was found during 1969–72, a rather long inactive period (1973– 88), with an average of 15 typhoons, was observed. Typhoon activity increased again from 1989. However, a sharp decrease was observed in 1998 and 1999 although the number of typhoons was close to the mean during the period 2000–03. These observations suggest that the typhoon activity exhibits a significant interdecadal variation. The year-to-year variation can also be very large. For example, 1970 had 12 typhoons, but 24 typhoons occurred the next year. SST variations in this region certainly do not have similar amplitudes (see Fig. 2).
The values of TDP exhibit similar interdecadal and interannual variations (Fig. 1b). Because the value of TDP increases as the square of the maximum wind speed, an increase in the maximum intensity of TCs in response to a rise in SSTs should be reflected in an increase in the TDP value. However, the time series of TDP actually shows a slightly decreasing trend although the values in the last 3 yr are close to the long-term (1960–2003) mean value.
The interdecadal variation of the RTY values is less significant (Fig. 1c) probably due to the fact that the value of RTY depends not only on the number of typhoons but also on the total number of TCs. Nevertheless, no significant upward trend is observed during the period of study. If TCs tend to become more intense/ numerous with a rise in SSTs, this value should increase with time.
b. Variations in SST
SSTs over the WNP generally have rather small variations. A time series of these temperatures averaged over the prescribed domain for the months of May to November shows less than a 1°C variation over the period 1960–99 (Fig. 2). Nevertheless, an upward trend is apparent.
4. Relationship between typhoon activity parameters and SST
a. Correlation between SST and typhoon activity during the main typhoon season
Given the variations of the typhoon activity parameters, it is prudent to investigate the correlation between the SST and typhoon activity. To minimize the effect of the variation of SST during the winter season, which should have no significant influence on annual typhoon activity, only the SST in the main typhoon season (May– November) is examined. The typhoon activity between May and November represents ∼90% of the annual activity, and the values of the three typhoon activity parameters in this period are highly correlated with their annual values (correlation coefficients >0.95). Therefore, correlations between the May–November SST and the typhoon activity parameters during the same period can be used to infer the relationship between SST and annual typhoon activity.
The value of RTY was found to have a negative correlation with the SST over the western North Pacific (5°–30°N, 120°E–180°; hereafter referred as local SST) between May and November (Table 1; Fig. 3a). Such correlations are more prominent in the domain 10°– 30°N, 125°–150°E with the strongest correlation (∼−0.6) around 20°N, 140°E. As suggested earlier, both the RTY and SST exhibit interdecadal variations (see Figs. 1b and 2). To minimize this effect, the interdecadal variations of both SST and RTY are removed (Fig. 4). Each time series is first fitted to a fourth-order polynomial. The fitted values obtained from the polynomial are then subtracted from the original value, which are then used for the computation of correlation. The resulting correlations are found to be similar (Table 1). Table 1 also shows that while the TDP and NTY are not significantly correlated with the local SST, the correlations become significant after the removal of their interdecadal variations.
The correlation patterns in Fig. 3a further suggest that RTY is not only negatively related to the variation of local SST, it is also positively correlated with the mean May–November SST over the equatorial central and eastern Pacific Ocean, with a mean correlation of 0.58 in the Niño-3.4 region (5°S–5°N, 170°–120°W). Such a result suggests that variations in typhoon activity may be related to the El Niño–Southern Oscillation (ENSO) phenomenon such that a higher (lower) typhoon activity corresponds to above-normal (below-normal) SST in this region, that is, an El Niño (La Niña) condition. Indeed, of the 11 El Niño years since 1960, 9 were associated with above-normal TDP (Table 2). In contrast, 8 of the 11 La Niña years were associated with below-normal TDP (Table 3). The difference in the TDP during El Niño and La Niña years (301.5 and 182.7, respectively) is statistically significant at the 95% confidence level. Similar results can be seen for the other two parameters NTY and RTY except for the difference in NTY during El Niño and La Niña years.
To summarize, the WNP typhoon activity, as measured by the various parameters, does not show any significant increase in response to a rise in local SST. Indeed, a decrease in the typhoon activity is apparently observed in the years with above-normal local SST. On the other hand, the ENSO phenomenon is shown to have a significant relationship with typhoon activity.
b. Effect of ENSO on western North Pacific SST
While an El Niño (La Niña) event involves the anomalous warming (cooling) of the equatorial central and eastern Pacific, it also has a significant influence on SST in the global oceans through the “atmospheric bridge” (Alexander et al. 2002). Since the focus of this study is the WNP, the impact of ENSO on the SST in this region is examined here.
The mean Niño-3.4 SST anomaly between July and November is used as a measure of the strength of an ENSO event during the TC season. The correlation pattern suggests that SSTs in most parts of the WNP are negatively correlated with the ENSO index (Fig. 5). Such correlations are more prominent (with a maximum correlation coefficient of >−0.6) in the domain 5°– 20°N, 130°–155°E, which is also the region with most occurrence of TC activity over the WNP (Xue and Neumann 1984). However, the interannual variation of SST in this region is much smaller than that in the eastern equatorial Pacific. During the period 1960–99, the year with the warmest SST within this domain was 1998, with an SST anomaly of ∼0.6°C, while the coldest year was 1972 with an SST anomaly of ∼−0.5°C, which represents about a 1°C variation. Of the 10 yr with the highest SST in this region, 5 were associated with La Niña events while 6 of the 10 yr with the lowest SST were El Niño years. These results suggest that the interannual variation of SST in the WNP, particularly for the region 5°–20°N, 130°–155°E, is strongly related to the ENSO phenomenon. Therefore, in the examination of the impact of local SST on WNP TC activity, the influence of ENSO on local SST should be removed.
c. Correlation between local SST and typhoon activity after removal of the ENSO effect
The partial correlation (after removal of the correlation with ENSO) between local SST and typhoon activity suggests that the negative correlation previously observed over the WNP is reduced in both magnitude and areal extent (cf. Figs. 3a and 3b), which demonstrates the significant effect of ENSO. With such a removal of the contribution by ENSO, the correlations between local SST and various typhoon activity parameters are reexamined. The negative correlations previously obtained become insignificant after the ENSO effect is removed (Table 1). Therefore it may be concluded that the increase in local SST has no significant effect on typhoon activity, the latter being mainly constrained by the ENSO event through the alteration of the atmospheric circulation, details of which are examined in the next section.
5. Impact of large-scale parameters on typhoon activity
Climatologically, TC genesis and development are closely related to six large-scale parameters including three dynamical factors: Coriolis parameter, low-level (850 hPa) relative vorticity, and vertical shear of the horizontal wind; and three thermodynamic factors: midlevel moisture, moist instability of the low- to midlevel (500 hPa) atmosphere, and SST (Gray 1979). To investigate the impact of these parameters on typhoon activity, the empirical orthogonal function (EOF) analysis is applied to the 850-hPa relative vorticity, vertical zonal wind shear, and moist static energy fields. Correlations are then computed among these large-scale parameters, SSTs (both local and remote), and various measures of typhoon activity. The procedure is as follows.
First, the large-scale parameters are averaged over the peak TC season (July–November). An EOF analysis is then performed on the anomalous fields of these averaged parameters. The time coefficients of each mode are then correlated with the typhoon activity parameters. Only those modes that are significantly correlated with the typhoon activity will be discussed, and the sign of the EOF is chosen such that it is positively correlated to the Niño-3.4 SST anomaly. The correlations between these time coefficients and SST are then computed, and the effect of ENSO events on these correlations is also investigated.
a. 850-hPa relative vorticity
The first EOF of 850-hPa relative vorticity reveals the interannual variability of the strength of the monsoon trough extending from the eastern part of the tropical WNP to the Philippine Sea, with the maximum amplitude in the region east of 155°E (Fig. 6). This mode explains ∼35% of the total variance. The region with the largest amplitude reflects that the low-level relative vorticity has the largest interannual variation there, which is consistent with the fact that this is the area in which more TCs form during El Niño years, but very few or none form during La Niña years (Chan 1985, 2000; Wang and Chan 2002).
Time series of the first principal component (PC1) of 850-hPa relative vorticity is positively correlated with that of the typhoon activity parameters (correlation coefficients ranging from 0.32 to 0.59, which are all significant at the 95% confidence level), and these correlations are slightly higher after the removal of the interdecadal variation of these parameters (not shown). These results suggest that a stronger monsoon trough results in increased typhoon activity, which is reasonable since such a situation favors TC genesis and development (Briegel and Frank 1997). Furthermore, PC1 is highly correlated with the Niño-3.4 SST anomaly (r = 0.86), which suggests that the monsoon trough is generally stronger (weaker) in an El Niño (La Niña) year, which is consistent with the previous results of Chan (2000) and Wang and Chan (2002).
The PC1 of the 850-hPa relative vorticity is generally negatively correlated with the SST in the western part of the WNP (Fig. 7a). Such correlations are more prominent in the domain 5°–20°N, 130°–155°E. This result suggests that a weaker monsoon trough is generally observed in years with higher local SST. However, since the local SST is also influenced by the ENSO event (section 4b) and the monsoon trough strength is highly correlated with the ENSO index, the partial correlation between local SST and PC1 after the removal of the ENSO effect needs to be examined. It is found that the resulting partial correlations are significantly reduced and become insignificant in most parts of the WNP (Fig. 7b), which confirms that the interannual variation of the monsoon trough strength is mostly contributed by the ENSO event rather than the local SST. As typhoon activity is related to the monsoon trough strength, this result further suggests that the increase in local SST has no significant impact on typhoon activity.
b. Vertical shear of the zonal wind
The first EOF of vertical shear of the zonal wind between 200 and 850 hPa shows a north–south dipole over the tropical WNP with a positive (negative) sign north (south) of 18°N (Fig. 8) and explains ∼61% of the total variance. This mode generally represents the alternation of westerly shear and easterly shear over the southern and northern parts of the WNP, and such a variation is larger in the tropical region.
The PC1 is positively correlated with the typhoon activity parameters (correlation coefficients ranging from 0.44 to 0.57, which are significant at the 95% confidence level), and these correlations are slightly lower after the removal of the interdecadal variation of these parameters. The positive correlation suggests that a westerly shear north of 18°N and an easterly shear south of 18°N in the WNP generally result in higher typhoon activity, which is consistent with the horizontal distribution of zonal vertical shear that is favorable for TC development (McBride and Zehr 1981). On the other hand, the PC1 coefficients are also highly correlated with the Niño-3.4 SST anomaly (r = 0.82). In other words, during an El Niño (La Niña) year, westerly and easterly (easterly and westerly) shears are generally found over the northern and southern parts of the WNP, respectively, which corresponds to a distribution of zonal vertical shear that is favorable (not favorable) for TC development. Furthermore, such a variation of north– south distribution of zonal vertical shear is more prominent over the eastern part of the WNP, which is consistent with the eastward (westward) shift of the mean TC genesis location during an El Niño (La Niña) year (Chan 1985, 2000; Wang and Chan 2002).
In addition, the PC1 is negatively correlated with the local SST in the western part of the WNP (Fig. 9a). A higher local SST in this area may be associated with a distribution of zonal vertical shear that is not favorable for TC development. After the removal of the ENSO effect, the correlation between local SST and vertical wind shear is reduced but still significant near the Philippine Sea and the South China Sea (Fig. 9b). However, the correlation is not significant if the mean local SST over the entire WNP is considered.
c. Moist static energy
The first EOF of moist static energy2 (pressure-weighted mean over the layer 1000–500 hPa) is not significantly correlated to the typhoon activity and therefore will not be discussed. The second EOF shows a southeast–northwest dipole over the WNP (Fig. 10) and explains ∼24% of the total variance. The area of maximum amplitude is again in the southeastern part of the WNP. This suggests that in this area, the moist static energy can have a large variability, consistent with that of TC activity here.
The second principal component (PC2) of moist static energy is positively correlated with the typhoon activity parameters (correlation coefficients ranging from 0.37 to 0.52, all significant at the 95% confidence level). These correlations are slightly lower after the removal of the interdecadal variation of these parameters (not shown). A higher value of moist static energy in the southeast quadrant and a lower value of moist static energy in the northwest quadrant of the WNP generally give a higher amount of typhoon activity. This pattern resembles the El Niño situation for TC activity (Wang and Chan 2002). Indeed, PC2 is highly correlated with the Niño-3.4 SST anomaly (r = 0.73), which means that more (less) moist static energy is available in the southeast quadrant during an El Niño (La Niña) year, but the opposite is true in the northwest quadrant.
The PC2 is positively correlated with the local SST in the southeastern part of the WNP but has weak negative correlation with the SST in other parts of the WNP (Fig. 11a). This means that in most parts of the WNP, an increase in SST actually correlates with a decrease in moist static energy. A lower moist static energy is likely to be associated with stronger subsidence, a dry midtroposphere, and stronger low-level inversion, all of which are probably due to the forcing from the ENSO condition. In other words, even though local air–sea interaction may provide favorable thermodynamic conditions near the surface, the large-scale conditions do not support deep convection and hence less TC activity. After the removal of the ENSO effect, the correlation between local SST and moist static energy is reduced significantly (Fig. 11b), which further supports the above hypothesis.
d. Summary
In this section, the impact of various large-scale dynamic and thermodynamic factors on WNP typhoon activity is examined and includes low-level relative vorticity, vertical zonal wind shear, and moist static energy. The change of SST in the tropical central and eastern Pacific associated with an ENSO event has a significant impact on these large-scale parameters, which is consistent with previous results. During an El Niño year, the enhanced low-level relative vorticity, weaker vertical wind shear, and higher moist static energy provide a favorable atmospheric environment for TC development and intensification in the WNP, especially over the southeastern quadrant of the NWP. In other words, in such a year, the monsoon trough tends to be stronger and extends farther east and south relative to its climatological position, which results in an eastward shift of the mean genesis location and more time for a TC to intensify before making landfall. Hence, the annual typhoon activity is generally higher in an El Niño year.
To conclude, the interannual variations of typhoon activity over the WNP appear to be mainly constrained by the ENSO phenomenon through the alteration of the large-scale circulation induced by ENSO events, which have been discussed by Chan (2000) and Wang and Chan (2002). Therefore, in projecting the future intensity of TCs in response to a rise in SST as a result of the buildup of greenhouse gases, the effect of ENSO must be considered. Unfortunately, it is still uncertain how the behavior of such events would change in a doubled-CO2 world (Cubasch and Meehl 2001).
6. Discussion and conclusions
This paper investigates the validity of the assertion based on numerical simulation results that an increase in sea surface temperature (SST) due to global warming will likely lead to more intense tropical cyclones (TCs) by examining the data in the past. Analyses of the annual number of typhoons in the western North Pacific, its ratio to the total number of TCs in the basin, and its destruction potential show that during the period 1960– 2003, these parameters have gone through large interannual as well as interdecadal variations. They also show a slight decreasing trend. On the other hand, only very small changes and a slight increasing trend in SST have been observed. No significant correlation was found between the typhoon activity parameters and local SST. In other words, an increase in local SST does not lead to a significant change of the number of intense TCs in the NWP, which is contrary to the results produced by many of the numerical climate models. Indeed, the interannual variation of annual typhoon activity is mainly constrained by the ENSO phenomenon through the alteration of the large-scale circulation induced by the ENSO event. During an El Niño year, the enhanced low-level relative vorticity, weaker vertical wind shear, and higher moist static energy provide a favorable atmospheric environment for TC development and intensification in the WNP. Such an environment is more prominent over the southeastern quadrant of the NWP, which may result in the eastward shift of mean genesis location and therefore more time for a TC to intensify before making landfall. Hence, the annual typhoon activity is generally higher in an El Niño year.
The reason for the discrepancy between the present results and those from many of the numerical climate models likely lies in the assumptions made in most of these models. Generation of TCs is assumed to result primarily from energy from the oceans and mainly depends on SST. Thus, a higher SST would lead to more energy being transferred from the ocean to the atmosphere, and hence the model cyclones become more intense. In other words, the typhoon activity predicted in these models is almost solely determined by thermodynamic processes, as advocated by Emanuel (1999). However, in the real atmosphere, dynamic factors, such as the vertical variation of the atmospheric flow (vertical wind shear) and the juxtaposition of various flow patterns that lead to different angular momentum transports, often outweigh the thermodynamic control in limiting the intensification process, perhaps except for landfall or when a TC moves to colder waters in the midlatitudes. Both modeling and data analysis studies have emphasized the importance of these dynamic factors (Frank and Ritchie 2001; Titley and Elsberry 2000). In fact, DeMaria and Kaplan (1994) showed that only 20% of Atlantic TCs reached 80% or more of their “maximum potential intensity,” calculated purely on upper-tropospheric temperatures and SSTs. This means that in most cases, it is the dynamic factors, and perhaps the atmospheric moisture, that control TC intensity.
Indeed, the significant correlations between the overall TC intensity and the SST over the equatorial eastern and central Pacific demonstrates the crucial importance in considering the changes of the large-scale atmospheric flow patterns over the western North Pacific due to variations in these ocean temperatures, which in turn modify the locations of TC formation as well as the positions of troughs and ridges relative to the TCs and thus the maximum intensities to which the latter could attain.
Therefore, in such modeling efforts, the dynamic factors must be examined to see if the predictions in terms of the dynamic forcings of the TC intensification process indeed resemble those in the real atmosphere. A step in the right direction is the work by Bengtsson et al. (1996), who analyzed the large-scale circulation associated with an increase in SST due to a doubling of CO2. They found that stronger vertical wind shear, among other unfavorable factors in the large-scale atmospheric circulation, leads to a reduction in the number of TCs (see also Gray 1968). The results from the present study might be explained from a similar perspective, although more detailed analyses need to be made.
To conclude, at least for the western North Pacific, observational evidence does not support the notion that increased typhoon activity will occur with higher local SSTs. Rather, any such increase is likely due to the enhancement of the monsoon trough, reduction of the vertical wind shear, and increase in availability of moist static energy, all of which are largely due to variations in the SSTs in the equatorial central and eastern Pacific. Diagnostics of the output from climate models should be made to determine whether such results could be reproduced. Based on such diagnostics, modifications of the model design can be made with the eventual aim of simulating the observed variations in the large-scale fields. Once this is achieved, the model will then have the credibility to give projections on future climate trends.
Acknowledgments
This research is supported by City University of Hong Kong Grant 7010010. Constructive and meticulous comments from the two reviewers contributed toward improvements in the manuscript and are much appreciated.
REFERENCES
Alexander, M. A., I. Bladé, M. Newman, J. R. Lanzante, N. C. Lau, and J. D. Scott, 2002: The atmospheric bridge: The influence of ENSO teleconnections on air–sea interaction over the global oceans. J. Climate, 15 , 2205–2231.
Bengtsson, L., M. Botzet, and M. Esch, 1996: Will greenhouse gas– induced warming over the next 50 years lead to higher frequency and greater intensity of hurricanes? Tellus, 48A , 57–73.
Briegel, L. M., and W. M. Frank, 1997: Large-scale influences on tropical cyclogenesis in the western North Pacific. Mon. Wea. Rev, 125 , 1397–1413.
Broccoli, A. J., and S. Manabe, 1990: Can existing climate models be used to study anthropogenic changes in tropical cyclone climate? Geophys. Res. Lett, 17 , 1917–1920.
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 , 2960–2972.
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.
Chan, J. C. L., Y. Duan, and L. K. Shay, 2001: Tropical cyclone intensity change from a simple ocean–atmosphere coupled model. J. Atmos. Sci, 58 , 154–172.
Cubasch, U., and G. A. Meehl, 2001: Projections for future climate change. Climate Change 2001: The Scientific Basis, J. T. Houghton et al., Eds., Cambridge University Press, 525–582.
DeMaria, M., and J. Kaplan, 1994: Sea surface temperature and the maximum intensity of Atlantic tropical cyclones. J. Climate, 7 , 1324–1334.
Emanuel, K. A., 1999: Thermodynamic control of hurricane intensity. Nature, 401 , 665–669.
Evans, J. L., 1993: Sensitivity of tropical cyclone intensity to sea surface temperature. J. Climate, 6 , 1133–1140.
Frank, W. M., and E. A. Ritchie, 2001: Effects of vertical wind shear on the intensity and structure of numerically simulated hurricanes. Mon. Wea. Rev, 129 , 2249–2269.
Goldenberg, S. B., C. W. Landsea, A. M. Mestas-Nuñez, and W. M. Gray, 2001: The recent increase in Atlantic hurricane activity: Causes and implications. Science, 293 , 474–479.
Gray, W. M., 1968: Global view of the origin of tropical disturbances and storms. Mon. Wea. Rev, 96 , 669–700.
Gray, W. M., 1979: Hurricanes: Their formation, structure, and likely role in the tropical circulation. Meteorology over the Tropical Oceans, D. B. Shaw, Ed., Royal Meteorological Society, 155– 218.
Gray, W. M., C. W. Landsea, P. W. Mielke, and K. J. Berry, 1992: Predicting Atlantic seasonal hurricane activity 6–11 months in advance. Wea. Forecasting, 7 , 440–455.
Haarsma, R. J., J. F. B. Mitchell, and C. A. Senior, 1993: Tropical disturbances in a GCM. Climate Dyn, 8 , 247–257.
Hope, J. R., 1979: Tropical cyclone climatology. Operational Techniques for Forecasting Tropical Cyclone Intensity and Movement, World Meteorological Organization, I.2.1–I.2.10.
Knutson, T. R., and R. E. Tuleya, 1999: Increased hurricane intensities with CO2-induced warming as simulated using the GFDL hurricane prediction system. Climate Dyn, 15 , 503–519.
Knutson, T. R., R. E. Tuleya, and Y. Kurihara, 1998: Simulated increase of hurricane intensities in a CO2-warmed climate. Science, 279 , 1018–1020.
Knutson, T. R., R. E. Tuleya, W. Shen, and I. Ginis, 2001: Impact of CO2-induced warming on hurricane intensities as simulated in a hurricane model with ocean coupling. J. Climate, 14 , 2458–2468.
Krishnamurti, T. N., R. Correa-Torres, M. Latif, and G. Daughenbaugh, 1998: The impact of current and possibly future sea surface temperature anomalies on the frequency of Atlantic hurricanes. Tellus, 50A , 186–210.
Kuroda, M., A. Harada, and K. Tomine, 1998: Some aspects on sensitivity of typhoon intensity to sea-surface temperature. J. Meteor. Soc. Japan, 76 , 1145–1151.
Landsea, C. W., N. Nicholls, W. M. Gray, and L. A. Avila, 1996: Downward trends in the frequency of intense Atlantic hurricanes during the past five decades. Geophys. Res. Lett, 23 , 1697–1700.
Lighthill, J., and Coauthors, 1994: Global climate change and tropical cyclones. Bull. Amer. Meteor. Soc, 75 , 2147–2157.
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.
Shay, L. K., P. G. Black, A. J. Mariano, J. D. Hawkins, and R. L. Elsberry, 1992: Upper ocean response to Hurricane Gilbert. J. Geophys. Res, 97 , 20227–20248.
Shen, W., R. E. Tuleya, and I. Ginis, 2000: A sensitivity study of the thermodynamic environment on GFDL model hurricane intensity: Implications for global warming. J. Climate, 13 , 109–121.
Titley, D. W., and R. L. Elsberry, 2000: Large intensity changes in tropical cyclones: A case study of Supertyphoon Flo during TCM-90. Mon. Wea. Rev, 128 , 3556–3573.
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.
Watson, R. T., the Core Writing Team, Eds.,. 2001: Climate Change 2001: Synthesis Report. Cambridge University Press, 398 pp.
Whitney, L. D., and J. S. Hobgood, 1997: The relationship between sea surface temperatures and maximum intensities of tropical cyclones in the eastern North Pacific Ocean. J. Climate, 10 , 2921–2930.
Xue, Z., and C. J. Neumann, 1984: Frequency and motion of western North Pacific tropical cyclones. NOAA Tech. Memo. NWS NHC 23, 89 pp.
Time series of (a) the NTY, (b) the TDP, and (c) the ratio of NTY to the total number of tropical cyclones in the period 1960– 2003. Open circles and closed squares indicate the El Niño and La Niña years, respectively. The thick lines and horizontal lines indicate the fourth-order polynomial fits and the 1960–2003 means, respectively
Citation: Journal of Climate 17, 23; 10.1175/3240.1
Time series of the mean May–Nov SST averaged over the area 5°–30°N, 120°E–180°. The thick line indicates the fourth-order polynomial fit
Citation: Journal of Climate 17, 23; 10.1175/3240.1
(a) Spatial distribution of correlation coefficients between RTY and SST in the period May–Nov. Shaded areas indicate that the correlations are statistically significant at the 95% confidence level. (b) Same as in (a), except for the partial correlation coefficients
Citation: Journal of Climate 17, 23; 10.1175/3240.1
Time series of the RTY (closed circles with solid lines) and mean SST averaged over the area 5°–30°N, 120°E–180° (open squares with dashed lines) in the period May–Nov after removal of the interdecadal variations
Citation: Journal of Climate 17, 23; 10.1175/3240.1
Correlation between SST anomalies in the North Pacific and Niño-3.4 SST anomalies in the period Jul–Nov. Shaded areas indicate that the correlations are statistically significant at the 95% confidence level. The rectangular box indicates the Niño-3.4 region
Citation: Journal of Climate 17, 23; 10.1175/3240.1
EOF1 of the 850-hPa relative vorticity averaged over the period May–Nov during 1960–99. The sign of the EOF is chosen such that it is positively correlated to the Niño-3.4 SST anomaly
Citation: Journal of Climate 17, 23; 10.1175/3240.1
(a) Correlation between the time series of EOF1 of the 850-hPa relative vorticity and SST averaged over the period May–Nov during 1960–99. Shaded areas indicate that the correlations are statistically significant at the 95% confidence level. (b) Same as in (a), except for the partial correlation
Citation: Journal of Climate 17, 23; 10.1175/3240.1
Same as in Fig. 6, except for the vertical wind shear
Citation: Journal of Climate 17, 23; 10.1175/3240.1
Same as in Fig. 7, except for the vertical wind shear
Citation: Journal of Climate 17, 23; 10.1175/3240.1
Same as in Fig. 6, except for the EOF2 of the moist static energy
Citation: Journal of Climate 17, 23; 10.1175/3240.1
Same as in Fig. 7, except for EOF2 of the moist static energy
Citation: Journal of Climate 17, 23; 10.1175/3240.1
Correlation coefficients between local SST and various typhoon activity parameters (see text for definition) in the period May–Nov. Parameters with an asterisk indicate that the interdecadal variations have been removed. Numbers in parentheses show the partial correlations with the removal of the contribution from ENSO. Numbers in italic indicate correlations that are statistically significant at the 95% confidence level
Variation of typhoon activity parameters (see text for definition) in El Niño years. Numbers in italic indicate values above the climatological mean (1960–99)
Same as in Table 2, except for La Niña years. Numbers in italic indicate values below the climatological mean (1960–99)
