By Richard Black
Environment Correspondent, BBC News website
Here is a recipe for an explosive news cocktail.
Scientists need more data - and that only comes with time
Take the president of the world's most powerful nation. Add two intense and damaging natural storms which bring destruction to that country; then mix in the widely held view that the same nation's environmental policies are partially responsible for those storms.
In the polarised world of climate change, this cocktail has proved an irresistible temptation to organisations which campaign against President Bush's administration in support of enhanced action to curb greenhouse gas emissions.
The latest to succumb was the British newspaper The Independent, which screamed on its front page: "This is global warming", above an alarmingly portentous graphic of Hurricane Rita's projected path.
But is it global warming? What is the evidence that the growing concentrations of carbon dioxide and other greenhouse gases in the atmosphere are changing weather systems in such a way that hurricanes become more powerful, or more frequent?
Well above average
Certainly, 2005 appears to have been an unusually active year.
The US National Hurricane Center/Tropical Prediction Center comments in its August summary that "thus far in 2005, there have been 12 named storms and four hurricanes.
"These numbers are well above the long-term averages of 4.4 storms and 2.1 hurricanes that would normally have formed by this date."
But a single year's observation does not permit the divination of a long-term trend, or the attribution of that trend to a cause such as climatic warming.
"Based on recent research, the consensus view is that we don't expect global warming to make a difference to the frequency of hurricanes," explains Julian Heming, from the UK Meteorological Office.
"Activity is naturally very variable in terms of frequency, intensity and regional occurrence; in the Atlantic, there are active phases and not so active phases, and currently we're in the middle of an active phase.
"It's very dangerous to explain Rita or Katrina through global warming, because we have always had strong hurricanes in the USA - the strongest one on record dates back to 1935."
Records from the 20th Century suggest that hurricane formation over the Atlantic has changed phase every few decades: the 1940s and 50s were active, the 70s and 80s less so, while the currently active phase appears to have commenced in 1995.
A key factor in the formation of a tropical cyclone - a low-pressure region that can turn into a hurricane - is sea-surface temperature, which has to be above about 27 degrees Celsius.
So anything which changes the sea-surface temperature in the right parts of the world could theoretically affect hurricane formation.
Population growth means there is now more property to damage
The most recent study on the issue, published this month in the journal Science, found that while the incidence of hurricanes and tropical storms has remained roughly constant over the last 30 years, there has been a rise in the number of intense hurricanes with wind speeds above 211km/h (131mph).
The leader of that research project, Dr Peter Webster, believes there may be a link to climate change.
"What I think we can say is that the increase in intensity is probably accounted for by the increase in sea-surface temperature," he told the BBC News website, "and I think probably the sea-surface temperature increase is a manifestation of global warming."
"The problem is," observes Julian Heming, "that we can only look back about 35 years with satellite data; before that the record is somewhat unreliable, and 35 years isn't long enough to draw a definite conclusion.
"Before global satellite coverage, we're pretty sure there are gaps in the record; storms would start at sea and die out at sea, so we never knew about them."
The changing phases of Atlantic hurricane activity are not completely understood; but there appears to be a link to fluctuations in the thermohaline circulation, the global pattern of ocean currents which in western Europe appears as the Gulf Stream.
By causing the sea-surface temperature in the tropical Atlantic to change by even a degree Celsius, these fluctuations can bring major differences to the number of hurricanes generated in a particular year.
Other natural climate cycles such as the El Nino Southern Oscillation and the North Atlantic Oscillation may also play a role.
The other crucial factor with Katrina and Rita is where they landed.
Some hurricanes never reach land; others will hit a sparsely-populated area, causing minimal damage.
This also appears to be determined by weather systems, in particular the location of a region of high atmospheric pressure, the sub-tropical ridge.
"In the Atlantic, storms form in the east and move towards the west," says Julian Heming, "and at some point they turn northwards.
"Where they turn northwards has much to do with the weather conditions further north; 1995, for example, was a very active season with 19 named storms, but the US got away very lightly because the sub-tropical ridge did not extend right across the Atlantic, and many storms turned north before reaching the US.
"Last year it did extend across the Atlantic, and so hurricanes were forced much further west - hence Ivan, Jean, Charlie and Francis all hit the US."
Bigger and bolder
Every time a hurricane comes along - or a flood, or a drought, or a freeze, or a heatwave - the question is now asked "is it linked to global warming?"
A decade ago, that was not the case - a clear signal that climate change is now firmly established in the public mind and in the political arena.
Now that climate scientists are being taken seriously, they are also under pressure to produce instant answers.
One problem is that not all of those answers exist. Another problem is that some scientists - not to mention lobby groups, environmental organisations, politicians, newspapers and commentators - will go much further in their public statements than the data allow.
With such incendiary material, that is unlikely to change; but it is difficult to avoid the conclusion that we would all benefit from people on both wings of the issue looking rather more to research, however laboured its progress, and rather less to screaming headlines and easy quotes.