We have presented a physically based model of the ocean environment which has been used to predict the occurrence of tropical cyclones. The analysis leads to realistic predictions not only of the initiation of TCs, but also of their growth rate. The model is simple to apply both to reanalysis data and also to the results of climate models. The essential physical process is the interplay between evaporation and SST fields which are controlled by the large scale dynamics. In the tropics rising air occurs in the relatively cloudy regions of lower E and of higher SST


Fig. 9 Annual cycle of monthly average count for the observational period (1979-2005) and for the climate model predictions for 2051-2080


Fig. 9 Annual cycle of monthly average count for the observational period (1979-2005) and for the climate model predictions for 2051-2080

nearer the equator, and sinks over the relatively cloud free regions of higher E and lower SST further from the equator. This convective regime gives rise to a meridional region which extends from about 5°N (or 5°S) to the latitudes at which T ^26° C, which correspond approximately with the peak in evaporation shown in Figs. 1 and 2. Note that the data points with lnE less than its maximum value with respect to T, occur in the range, 5°N-5°S (Bye and Keay, 2006).

The results of our analysis show however that the standard deviation of H is much greater than the magnitude of the mean of H. The enhanced negative regions of H are due to localized intense convection and the enhanced positive regions are due to localized subsidence; the two processes being on average in balance.

We emphasize that the H-index predicts the regions of potential TC development in each generation region, which during any month are many. Other conditions such as vertical wind shear and the instability of the lower atmosphere would then select the locations at which the TC actually develops. The comparison between observations and predictions from the H-index model (Fig. 7), however, suggests that on a monthly time scale at least, the proportion of favorable sites (as predicted by the H-index), which realize TCs does not vary greatly over the TC season or between TC generation regions.

This is the link between the H-index and other indices such as the seasonal genesis parameter (SGP) originally proposed by Gray (1975), and often applied to make seasonal predictions of TC occurrence, see for example Watterson et al (1995). The SGP consists of a product of a thermal potential and a dynamic potential, in which the variability of the former is largely controlled by the ocean, and of the latter by the atmosphere. The modeling study of Royer et al. (1998) in which the results of a control run (1 x CO2) were compared with those of a doubled CO2 environment (2 x CO2) clearly showed that the increases in TC numbers were mainly brought about by changes in the thermal potential, i.e. the ocean. Royer et al. (1998) also replaced the thermal potential (which relies on the ocean temperature) by a convective potential (which relies ultimately on the ocean evaporation) to give a modified seasonal genesis parameter and found that the increases in TC numbers, although much more modest, were also mainly due to changes in the convective potential brought about by the ocean.

The success of the H-index supports the view that the major player in the TC variability is the ocean; in our case through the interaction of the ocean temperature and evaporation fields. This dual importance has also been recently recognized by Camargo et al. (2007) in a new generation potential, in which the ocean component (the potential intensity) depends on the sea surface temperature and pressure and vertical profiles of temperature and specific humidity, i.e. essentially on SST and E. The two prediction models, however, differ very significantly as the H-index is evaluated from the horizontal structure of the environment whereas the potential intensity is evaluated from its vertical structure.

In concluding this discussion, we suggest that the H-index may have the advantage over the oceanic component of GP and SGP and its convective modification, as it is based on robust elementary physical reasoning and is simple to calculate. We also note that the differences between the observations and the predictions from the H-index highlighted in the discussion in Results: (II) Comparison Between Observed Tropical Cyclone Numbers and the Standard Deviation of the H-index may be attributed to the variability of the dynamic potential, i.e. the atmosphere, rather than data inadequacies in the TC archives and the SST and reanalysis fields. This possibility, however, remains to be investigated.

We have used monthly mean data averaged over the generation region throughout. This gave rise, when further averaged over the record period (1979-2005) to almost symmetrical histograms of H (Fig. 3). The symmetry of the histograms, which was possibly the most unexpected finding of the study, signifies that the H-index captures the two-way energy exchange, characterized by the release of KE to the atmosphere from the mixed layer and its subsequent re-absorption, which lies at the centre of the tropical dynamics. This is an analogous process to the release of KE and its subsequent dissipation, which underlies the baroclinic instability mechanism in the subtropics.

The histograms for individual months, from which the inter-monthly time series (Fig. 7) were constructed, however, each showed an individual structure due to the synoptic evolution or suppression of the tropical cyclones It is anticipated that this structure would be even more marked over shorter averaging periods. This however remains to be checked.

With regard to climate models that do not explicitly resolve TCs, it was found that the standard deviation of H was smaller than for the reanalysis data, due to the poorer resolution. It is planned in a future study to obtain similar statistics using a climate model that does resolve TCs., which will enable the spatial structure of H to be modeled on a finer scale, and compared with that derived from synoptic analyses supported by high resolution SST fields.

Acknowledgments J.A.T.B. would like to express his thanks to colleagues at the 1st International Summit on Hurricanes and Climate Change for many useful discussions. The authors also thank a reviewer for many useful and instructive comments, and Kevin Keay of the University of Melbourne for assistance with the processing of the text.

Was this article helpful?

0 0
Survival Treasure

Survival Treasure

This is a collection of 3 guides all about survival. Within this collection you find the following titles: Outdoor Survival Skills, Survival Basics and The Wilderness Survival Guide.

Get My Free Ebook

Post a comment