Hurricane seasons can be deceiving because they encourage us to think on a yearly basis when longer-term trends can be more revealing.

It is true that last year's season broke all kinds of records and that Katrina has had a lasting impact. The repercussions are everywhere - from a new approach to catastrophe modelling, to more stringent capital requirements, increased diversification, reduced reinsurance capacity and the growth of alternative solutions. But these changes shouldn't be viewed in isolation. Behind them lurks a more poignant issue: why are the hurricane seasons getting worse?

The science behind the latest research and theories can be truly baffling, and certainly isn't helped by a lack of unity within the scientific community. Put simply, one camp believes global warming is behind the recent increase in hurricane activity, while the other believes it is down to a natural cycle - the Atlantic Multi-decadal Oscillation (AMO). Rumour has it the debate is as stormy as the Gulf of Mexico in September, with experts from rival camps refusing to attend the same conferences and publishing scathing attacks on opposing theories.

In his article "The great debate" (page 16) Munich Re's Dr Eberhard Faust takes an admirably balanced look at some of the current research supporting both arguments. But it's hard not to have an opinion. "A year ago the AMO was seen as the answer to everything, but looking at historical data we've only seen two cycles" says ReAdvisory's atmospheric physicist Dr Steve Smith. "The problem is it doesn't really have an explanation. El Nino, for example, has a physical explanation but there isn't a good one for the AMO."

If the AMO cycle is behind current activity levels then we would rightly expect to return to a period of lower storm activity at some point over the next few decades. But if global warming is the overriding cause it's likely that the hurricane seasons will become incrementally worse. And can we really afford to sit back and wait to find out which scenario we're dealing with? "As far as the reinsurance industry is concerned, in the short term it doesn't matter which theory is right - the next five to ten years will have heightened hurricane activity no matter which theory is ultimately proved correct," warns Dr Smith.

Scientists do agree that even small increases in sea surface temperatures can have a huge impact on the ferocity of tropical storms. And stronger winds mean more destructive hurricanes. According to a recent analysis conducted by AIR Worldwide for the Association of British Insurers, just small increases in hurricane wind speeds can have a devastating impact. AIR determined that if US hurricane wind speeds increased by as little as 6% on average, insured losses could increase by as much as 75% per year.

Ridiculously, as hurricanes get stronger and more frequent, millions of people continue to flock to the coast every year. Even if there is no increase in hurricane activity over the next few years, future hurricane losses will be amplified by rising property values (which doubled in the last decade according to AIR) and more development. The combined effect of this increasing population density and stormier seasons does not bode well for the industry.

Insurers are already feeling the heat, no pun intended. Allstate is no longer writing homeowners' policies in Louisiana, Florida and coastal parts of Texas and New York and, more recently, St Paul Travelers decided not to renew 3,000 commercial property policies in Florida. With a squeeze on reinsurance capacity there are growing calls for federal help in setting up state-run catastrophe funds.

It is likely these trends will continue if we experience another active season in 2006. Setting aside unhelpful wranglings in the scientific community, surely it's time to heed Katrina's wake-up call and think about a long-term strategy.