My first two Katrina posts announced the following "lessons": 1) science doesn't confer certainty about hurricanes and global warming; but scientific uncertainty doesn't justify inaction, either; 2) the issue of hurricane risks is much bigger than New Orleans. In other words, there are many, many other disaster-prone places. I only named a few possible worst case scenarios, involving Tampa Bay, Houston, Miami, and New York. Lesson three, I think, arises inevitably from these first two. We don't know precisely what global warming is doing to hurricanes; it would be foolhardy to claim otherwise. But we do know that we have scores of population centers that are highly exposed to to these storms. And if only due to sea level rise and nothing else, the risks to these population centers are changing--probably worsening. In this context, here's what's truly amazing to me: There is no national project to study changing hurricane risks to U.S. cities in light of the future scenarios that global warming may bring. Sure, this research has been done in some places in an episodic way; and the U.S. Army Corps of Engineers is apparently doing it for New Orleans in particular. But it ought to be a coordinated national project, so that communities can get at least some sense of what they may be facing, and so protective measures can be taken before the disaster happens. Why don't we have such a project? That will be the subject of my final "Katrina Lessons" post, so stand by...