Hacker News Reader

Comment by mindslight | Hacker News Reader

FWIW "100 year flood" refers to the flooding that is expected to occur in that specific area on average once every hundred years, not an event that will happen only once over the whole country once per hundred years. So regardless of climate change it makes a lot of sense to be seeing many hundred year floods since mass media has us caring about many areas in parallel, not just the specific area we live.

View on HN

Also, as people have a life expectancy in the order of 100 years (75 perhaps?), there is a pretty good chance you will experience a 1:100 year event at some point in your life. Also a lot of infrastructure is built for 50-100 year life (or is still used after 100 years, even though it was designed for less).

The 1:100 year safety level should thus be considered a bare minimum for high impact events. In the Netherlands water infrastructure is designed to withstand the 1:10.000 year storm for catastrophic events, and e.g. 1:1.000 year storm for high impact events.

But even with higher safety standards, it is key to design not only for the current climate, but for the future climate over the design life of whatever you are building. Extrapolating 1:10.000 year events from 30 years of observations might seem sketchy already, but with all the uncertainties in climate change predictions is a bit of a stretch.

There's a pretty good chance you'll experience any given 1:100 year event. You're basically guaranteed to experience quite a lot of 1:100 year events, since there are so many options.

This is a misleading definition as it makes people believe that if there is a 100-year flood this year, that equal or worse flooding will not happen next year.

Long term observed water levels are fed into a model with normal distribution, and a 100-year flood is then defined as 99-th percentile of the model. This means that there's a 1% chance of a 100-year flooding each year.

Global warming and climate change have an effect that there's a lot more water vapour in air. Same air mass can deliver more rainfall on a given territory, pushing observed water levels into higher extremes, meaning what used to be 100-year flood is now 25 (4% yearly chance), 10 (10% yearly chance) or even 5-year (20% yearly chance) flood.

Yeah, that is another issue with the way the media frames things and commonly invokes "100 year flood". In my comment, "on average" is doing a lot of work for those not decently versed in probability.

https://www.usgs.gov/special-topics/water-science-school/sci...

Ah, I wasn't aware of that definition. That makes a lot more sense, actually.

Reply on news.ycombinator.com