Science, Politics, Variability, Change, Learning, Uncertainty
February 27th, 2007Posted by: Roger Pielke, Jr.
The issue of floodplain management in the city of Boulder reflects in microcosm many of the themes that we discuss on this site. Here is an excerpt from an article in the Daily Camera today:
Boulder’s water board approved a flood plan Monday that predicts hundreds more homes and businesses will be inundated in a 100-year flood than previously believed.
But the new flood study predicts the University of Colorado’s South Campus property will stay dry in a 100-year flood, worrying residents who don’t want to see the former gravel mine developed.
The city’s current map places 363 structures in the flood plain. The new study predicts more than three times as many buildings — 1,137 — would take on some level of water in a 100-year flood, which has a 1 percent chance of occurring in any year.
Some issues raised by this circumstance:
1. The climate varies and changes faster than the built environment. Yesterday’s “100-year flood” is today’s “50-year flood.” Any flood policy based on the assumption of long-term stasis in climate is bound to fail.
2. Scientific understandings change faster than the built environment. Policies should be flexible to the possibility that we may learn more in the future, and such learning may result in revisions to our expectations for risks and vulnerabilities. Any policy that is based on an assumption that we know all we are going to is likely to fail.
3. People have different vested interests in particular scientific outcomes. In Boulder, people with different views about development have strong feelings about how the floodplain should be designated, based on how they think that will affect the chances of development. It would be foolish to think that such considerations can be ignored or kept separate from the political process of designating floodplain restrictions.
4. All important decisions are characterized by some degree of uncertainty. An important analytical question is not whether we can remove uncertainty (we can of course by chose to ignore it), but to design decision processes that are robust in the face of uncertainties.
The case study of flood policy in Boulder, Colorado reflects all of these issues.
February 27th, 2007 at 11:03 am
Roger -
Last summer, we had heavy summer rainfall here in New Mexico that was, by any measure, a statistical outlier. This triggered a fascinating series of discussions among the meteorology, civil engineering, political, policy communities and the public, often mediated by the news media. It became clear over the course of these discussions that there was not a clear and common understanding of the terms of discussion: that is, the notion of a “100 year flood” did not have a universally understood definition.
February 28th, 2007 at 2:11 pm
Consider the confluence of micro geographies and micro climates common throughout the US West. Indeed, attempting to create a “100 year flood map” is a fool’s errand.
March 1st, 2007 at 8:13 am
In the SW US, the summer thunderstorms are small and intense. The “100 year storm” can be over an area as small as two square miles. If a city is 20 miles square, it will average two of those per year. If it hits the wrong place or maybe there are four in a year (hardly a statistical outlier), there are alarmist reports. This is not to say the current intensity distribution maps are not sadly innaccurate…
There is also shown to be an increase in urban rainfall due to the UHI. The UHI, of course, is under the same attack as the MWP – the carbon traders (and their dupes) have to get rid of it to enhance their bottom line…