The way to look at it is - imagine every year, you have jar with 99 red M&Ms and 1 green M&M. You pick 1 M&M. How long before you pick that green one?
In a previous job, I worked with a team building chemical-process plants, and were looking at a site, 1/100 year flood potential was the bare minimum, and that was for industry - not urban/housing/residential construction.
I’m not a hydrologist, but I have heard that when the Dutch build or repair their levees (to keep the Baltic Sea out) they look at a minimum of of 1/10,000 year event possibility.
“In a previous job, I worked with a team building chemical-process plants, and were looking at a site, 1/100 year flood potential was the bare minimum, and that was for industry - not urban/housing/residential construction.”
Agree, and the thing, what is 1/100 versus 1/1000 - actually very little difference? It might mean designing heating systems for 100 hours below freezing temperature with a minimum temperature of 10F, versus designing for 130 hours below freezing with a minimum temperature of 5F.
In other words, by the time you’ve designed for 1 in a 100 years, you’re generally 90% of the way to 1 in a 1000 years. Also keep in mind that good data likely doesn’t go back more than 100 years, as engines didn’t even have a written language (in most cases).