Slashdot Mirror
Running out of Hurricane Names
fm6 writes "LiveScience is reporting that the 21 names reserved for tropical storms and hurricanes in Atlantic Basic are almost used up. If there are more than 21 storms, they'll start using the Greek alphabet. The most storms ever recorded was 21 in 1933, before they started giving them official names. The connection between this record-breaking storm year and global warming remains controversial."
According to this article, they currently think the main effect of global warming will be stronger hurricanes, not more hurricanes.
Of course, that's the current theory. If it turns out that we consistently get more, we'll end up with some new theories. Global warming is a big uncontrolled experiment, so it's hard to say. That's pretty sloppy science; I say we should have waited until we had two planets so we could try this side by side. And really, 20 or 30 would be better, so we could get a good statistical sample.
To disclose, I'm a chemist/statistician, and I drive a prius. I'm in favor of hedging our policy on the side of safety - but purely as a scientist, claiming any sort of accuracy in terms of climate prediction seems ridiculous given the current models.
You say "all us scientists" as if you have 100% consensus, and as if you're a climatologist. Are you?
They like to set up models, but their climate models can't prove a connection either, because they're all based on a lot of assumptions, abstractions, and potentially erroneous inputs. We're a long way off from weather models with any level of certainty. When they can give spot-on weather reports for a month out, then it'll be time to start paying attention to the models.
All they have now are measurements, where they hope to see a correlation. But no matter what correlation they saw, even if it was the most beautiful curve you've ever seen, with a curve fit with an R value of 1.0, correlation does not imply cause and effect. But at least it wouldn't contradict their theory.
What they actually have is a tiny sample, where nearly any conceivable data set would mean nothing. The problem is that there are so many factors. While their actual data set is really jumpy and shows no really strong trend, suppose it were different- suppose they got their "ideal" data set over the past 30 years. Suppose it showed that the number of hurricanes is trending up sharply and steadily. If they had seen this trend, which would most strongly support the hypothesis of global warming, it would equally strongly support all of the following hypothesis:
1. We're in a natural cycle of hurricanes increasing, which global warming is making worse.
2. We're in a natural cycle of hurricanes increasing, which global warming is having no effect on.
3. We're in a natural cycle of hurricanes increasing, which global warming is partially alleviating.
4. There is no natural trend, and global warming is causing a rise in hurricane activity.
5. We are in a natural cycle of reduced hurricanes, and global warming is counteracting that entirely and actually increasing the number of hurricanes.
6. There is no actual trend at all. The number of hurricanes every year is entirely random, with no natural tendency or influence from global warming, and our 30-year sample happens to look like it has some trends, because any series of random numbers will appear to have some trends over certain samples.
Furthermore, with so many factors that affect weather, less than two dozen hurricanes per year, an apparently large natural variability, the probability there are many natural trends that could be working in conflict or in concert, using a mere 30-year sample is like trying to estimate global warming with a 30-day temperature sample. It would make all the difference in the world if you take your sample during spring or fall, and time you take it at all, it's extremely unlikely it would give you an accurate picture of what's going on at all. If they had 1,000 years of data, I might expect them to find something more convincing there.
Can anyone tell me how to set my sig on Slashdot?
Because of the question as to whether this is from a natural cycle, or whether from global warming effects causing increased baseline ocean temperatures, or simply a statistical fluke year. As a first pass, either of the first two sounds credible as a cause. (If you RTFA, the last sounds less so.)
We know climate moves in cycles; we also know that hurricanes are formed by (and get their energy from) warm water. We don't have detailed records for a long enough time frame to readily determine if it's just a natural swing in the cycle. Ergo, we should be doing climate research, perhaps specifically focused on what affects hurricane formation.
Perhaps it's Global Warming; perhaps it's a natural oscilation in the deep ocean currents; perhaps it's just a statistical outlier event. Depending on which, the responses might be different. If it's an outlier, we can just plan for a short term headache with the rebuilding. If it's caused by human-induced global warming, we should start taking measures to ameliorate it. If it's just an unstoppable natural cycle unrelated to human influence, we should start considering what extent build-up of coastal developments ought to be insurable/taxed/regulated/&c, and considering how to minimize the impact on our national transportation infrastructure.
The fact that we are headed for a record year and don't know the cause suggests we should be doing more research into climate and oceanography, in order to determine the best reactions... preferably backed more by clearly stated measurements and mathematically calculated confidence intervals, rather than more by political pre-evaluation of what the implications might be for Senator Bedfellow's congressional district. Mother nature doesn't give a damn what we think the world ought to be like; she's going to hit us with the way it is.
//Information does not want to be free; it wants to breed.
We like to think we know everything. How can we say there is global warming when we have maybe 100 years on the subject. Same thing for Hurricanes.
Global warming is 'controversial' only as long as one forgets three undeniable facts: melted water lakes in the middle of Greenland, glacier melting and permafrost melting. We have more than 100 years on documented data on the length of glaciers and they have been getting smaller at an accelarated pace.
These phenomena can not be explained by anything else than a long term change in climatic conditions.
Fight Frist Psoting!
Browse Slashdot with 'Newest First'!
Assuming that Global Warming is undisputed fact, what do you propose doing about it? The entire world is built on a foundation of burning things to power other things, you can't change that overnight. 40-mpg cars and wind turbines? It will never power the world of 20-30 years from now. The global GDP has been advancing at a 3-5 percent rate for quite some time. That means that in 20 years we will have probably more than doubled our energy consumption. It's nice to think that ocean waves and hydrogen fuel cells will change the world; but remove burning things, and releasing CO2, and you throw the world into economic chaos. Economic depressions and hyperinflation are the types of things that start global wars; not the sorts of things that help the environment. Although I suppose a world conflict brought on by an energy shortage will solve the energy usage problem anyway...