The 23rd Conference on Hurricanes and Tropical Meteorology

4D.2
INTERANNUAL VARIABILITY OF TROPICAL CYCLONES IN THE CENTRAL NORTH PACIFIC

James D. Clark, University of Hawaii, Honolulu, HI; and P. S. Chu

Interannual variability of tropical cyclones in the Central North Pacific basin is investigated by analyzing the annual cyclone tracks, the annual cyclone numbers, and correlations between tropical cyclone numbers and both SST anomalies and the Southern Oscillation Index (SOI). Interannual fluctuations of various circulation parameters are also examined to identify relationships with tropical cyclone frequency. Tropical cyclone data from the National Hurricane Center's Eastern and Central North Pacific best track data set during the years 1966-1997 are used. A 32-year data set is chosen because of the implementation of satellite reconnaissance in 1966. The SOI, SST, and wind data are obtained from NCEP.

Although the annual number of tropical cyclones in the Central North Pacific basin is low in comparison to other basins, the interannual variability of tropical cyclones in this basin is high. The number of tropical cyclones, which includes tropical depressions, tropical storms and hurricanes, in the Central North Pacific in this period ranges from 0 to 11, with an annual mean of 4.5 tropical cyclones.

The effect of El Nino on the Central North Pacific tropical cyclone season is investigated by calculating correlations between various counts of tropical cyclones in the basin and both the SOI values and SST anomalies with different lags. Numerous correlations were significant at the 95% level by taking climatological persistence into account. More tropical cyclones occur in the Central North Pacific in warm years than in non-warm years. Because of the small data set, 144 tropical cyclones in 32 years, it is difficult to determine whether this difference is significant using conventional statistical methods. For this reason the data set was bootstrap resampled, that is, resampled with replacement, to examine the difference in tropical cyclone numbers between warm and non-warm years. In addition, a two-sample permutation test suggests that the mean number and variance of tropical cyclones in warm versus non-warm years are significantly different from each other. Finally, tropical cyclone tracks are compared between warm and non-warm years to investigate their spatial distribution in the Central North Pacific.

Currently, surface winds in the Central North Pacific during the same time period, 1966-1997, are being analyzed to investigate the interannual variability of the displacement of the monsoon trough. It is hypothesized that this eastward extension of the monsoon trough plays a major role in the Central North Pacific's tropical cyclone season by enhancing equatorial westerlies, which, in turn, increase the low-level cyclonic relative vorticity, one factor creating an environment favorable for tropical cyclone formation. The fluctuations in the low-level vorticity fields as well as the tropospheric vertical wind shear will be analyzed to find possible relationships with interannual tropical cyclone variability in the Central North Pacific. A discussion of how these possible predictive signals affect the circulation patterns and the differences in the numbers and locations of tropical cyclones in the Central North Pacific will then follow

The 23rd Conference on Hurricanes and Tropical Meteorology