The Washington Post has an interesting story today about the National Infrastructure Simulation and Analysis Center’s (NISAC) reporting to the Administration on the potential impacts of Katrina in the very early hours of Monday, August 29, 2005:
In the 48 hours before Hurricane Katrina hit, the White House received detailed warnings about the storm’s likely impact, including eerily prescient predictions of breached levees, massive flooding, and major losses of life and property, documents show.
A 41-page assessment by the Department of Homeland Security’s National Infrastructure Simulation and Analysis Center (NISAC), was delivered by e-mail to the White House’s “situation room,” the nerve center where crises are handled, at 1:47 a.m. on Aug. 29, the day the storm hit, according to an e-mail cover sheet accompanying the document….
The documents shed new light on the extent on the administration’s foreknowledge about Katrina’s potential for unleashing epic destruction on New Orleans and other Gulf Coast cities and towns. President Bush, in a televised interview three days after Katrina hit, suggested that the scale of the flooding in New Orleans was unexpected.
While this was probabilistic analysis, it still should have set off a lot more alarm bells. I’m curious to know what the distribution chain was for the report on 8/29 after NISAC sent it. This page on a University of California website provides a partial answer to this question:
The NISAC report was provided to the Homeland Security Operations Center for posting on the Homeland Security Information Network portal for local, state and regional assistance and consideration.
It’s also important to note that the key revelations in the WaPo piece are in the content of report: not its existence. Sandia, one of the two national labs where NISAC is run (Los Alamos is the other), wrote this article in its internal newsletter on September 16, 2005 that discussed the report and NISAC’s work in modeling Katrina:
NISAC began its work on the Saturday prior to Katrina hitting the Gulf Coast region. Sandiaâ€™s main goal is to look at flooding and infrastructure failures; LANL focuses on electricity systems.
â€œWe help determine the long-range industrial consequences regardless of the cause,â€ says Kevin Stamber, a NISAC member.
The idea is to use the models to warn officials of potential problems before they happen. The team often has less than one day from the time they are notified by DOE or DHS to generate reports concerning an eventâ€™s potential effects on infrastructure.
Nancy Brodsky, the Fast Analysis and Simulation Team lead, says the reports are not made public but are given to DHS for analysis and planning. Reports are not issued to the public because they could be misrepresented.
Perhaps these reports could be misrepresented. But in my opinion, a worse outcome is that they’re not made available to the people on the frontline who could benefit from the warning. In the future, I think NISAC should err on the side of disclosure in the cases of natural disasters (but not their models of terrorist attacks). For example, why not release the NISAC avian flu study mentioned in this Wired story? It sounds like something that could be useful to officials in states, cities, and the private sector that are developing their avian flu contingency plans today.
Update (1/24): This Times-Picayune story adds more details about the report.