The clean-up operations in the wake of Katrina are a nightmare for all concerned, including scientists at Tulane University and other research institutions on the Gulf coast (see Nature 437, 177; 2005). They need all the help they can get, and other institutions must endeavour to provide it and get them back on their feet.

The ramifications of this tragedy will run deep. There are already signs that national priorities are changing: President George W. Bush, speaking in New Orleans on 16 September, broached some themes that he had previously avoided. “We have a duty to confront poverty with bold action,” he said. If that pledge is to be followed through, it will involve changes in research priorities.

Policy-makers need good information if they are to tackle poverty and racial division, and in many instances research can provide it.

For better or for worse, the US federal government — particularly Congress — has a propensity to adjust the government's spending portfolio quickly in response to particular events. After the attacks of 11 September 2001, for example, the government created the Department of Homeland Security, with a large and ill-defined research programme, and diverted resources at the National Institutes of Health towards activities related to bioterrorism. Four years later, there is scant evidence that either shift has achieved much for science or for national security. After that precedent, some would argue that the government should avoid overreacting to Katrina.

Yet Katrina has brought to the surface some critical issues that have been wantonly ignored in Washington in recent years and now deserve some attention. The most significant of these relate to poverty, as Bush has now acknowledged, and racial division. Policy-makers need good information if they are to tackle these issues, and in many instances research can provide it. The aftermath of Katrina will push poverty, at least, up the agendas of agencies such as the National Science Foundation and the National Institutes of Health that support research in the social sciences and environmental health.

At the same time, the disaster raises the profile of two very different spheres of environmental research — water management and climate change. The former is quite well understood, although a great deal of existing knowledge about rivers and wetlands, for example, is frequently ignored by policy-makers. Despite the Bush administration's scepticism about the latter, it has maintained a powerful climate-change science programme, which in time may shed valuable light on the complex relationship between global warming and extreme weather events, including hurricanes.