Jenni-Louise Evans

Pennsylvania State University, Department of Meteorology

Tropical cyclones, storms characterized by winds of 17 meters per second or greater, occur at a frequency of 80 per year globally, plus or minus 7% (about 5 storms). There is very little variation in their annual occurrence. Globally, Evans believes this number is unlikely to change, though regionally, we do not understand enough to say.

If the wind speed of a cyclone is doubled, the destructive potential increases approximately four times. We currently have little skill at forecasting cyclone size, though we can forecast track. General circulation models (GCMs) don't include tropical cyclones because they are smaller in area than the grid scale.

Worldwide, cyclones cause about $10 billion in damage each year and are responsible for 10,000 human deaths on annual average. These numbers can be much higher in individual bad years; 300,000 Bangladeshis died in one year (1970) from storm surges.

Emphasizing the need to conduct regional climate studies, Evans summarized an array of techniques. Among these, listed in order from less useful to more useful, she discussed:

1. extrapolation of recent observations
2. direct inference from GCMs
3. analogs of past climate
4. nested climate modeling
5. theoretical models
6. inferring changes from indirect GCM diagnostics
7. sensitivity studies using process models
8. detailed data analysis, including interannual variability studies
9. combination of (5), (7), and (8) with GCM information on foreshadowed large-scale changes

Current theoretical models generally only look at one variable, sea surface temperature (SST), even though we know that tropical cyclones are affected by other variables as well. Warm water is rare, and Evans' data show little correlation between storm intensity and warm water. Not enough people are working on the problem, and most people working on it are in the United States, focusing mainly on the Atlantic. Gray looks at other factors and has studied, for example, long-term cycling in western Sahel rain storms in Africa and their possible association with hurricanes in Florida. SST may be too limiting as the focus of study, as opposed to atmospheric variables.

A policy statement from a World Meteorological Organization/International Council of Scientific Unions (WMO/ICSU) panel on tropical cyclones and climate change says that first-order effects of SST on tropical cyclone frequency and intensity should not be expected. Second order effects are expected to be less significant than El Niñ o effects. We actually know very little about tropical cyclones and, in particular, how they interact with the climate system. This statement prompted group discussion with Evans saying that it is the most responsible statement that can be made at this time, whereas some maintained that in light of the uncertainty, it is irresponsible to say that increases in frequency and intensity should not be expected.

In conclusion, Evans believes that the earlier speculation of some experts that tropical storms could increase in frequency and intensity has become dogma in the media and policy arena, as well as with the general public, despite the fact that most experts on this issue feel that their data do not support this speculation, and that they are more uncertain now than before.