1-in-100 year rains cause extreme flooding in NY, PA

[phantom power]

An extreme rainfall event unprecedented in recorded history has hit the Binghamton, New York area, where 7.49" fell yesterday. This is the second year in a row Binghamton has recorded a 1-in-100 year rain event; their previous all-time record was set last September, when 4.68" fell on Sep 30 - Oct. 1, 2010. Records go back to 1890 in the city. The skies have now cleared in Binghamton, with this morning's rain bringing the city's total rainfall for the 40-hour event to 9.02".

However, another large region of rain lies just to the south in Pennsylvania, and all of the rivers in the surrounding region are in major or record flood. The Susquehanna River at Binghamton is at 25.18', its highest level since records began in 1847, and is expected to overtop the flood walls protecting the city this afternoon. In Hershey, Pennsylvania, Swatara Creek is 18' over flood stage, and more than 8' above its record flood crest. Widespread flash flooding is occurring across the entire area, and over 125,000 people have been evacuated from their homes.

...

The extreme rains are due the the remains of Tropical Storm Lee interacting with a stationary front draped along the Eastern U.S. Adding to the potent moisture mix last night was a stream of tropical moisture associated with Hurricane Katia that collided with the stationary front. You don't often see a major city break its all-time 24-hour precipitation record by a 60% margin, according to wunderground's weather historian, Christopher C. Burt, and he can't recall ever seeing it happen before. It's worth noting that the Susquehanna River Binghamton stream gage, which has been in operation since 1847, is due to be shut off in 3 weeks due to budget cuts. Here's the note at the USGS web site:

http://www.wunderground.com/blog/JeffMasters/comment.html?entrynum=1921

[Duer 157099]

1. "This is the second year in a row Binghamton has recorded a 1-in-100 year rain event"

Um, what?

[Eddie Haskell]

4. It's like flipping a coin.

The odds don't change. Still, it does seem like the odds of having a major storm are changing ... doesn't it.

[Duer 157099]

5. I think the odds changed to 2-in-100

How can they call it a once in a 100 year event when it just happened last year?

[phantom power]

7. A "one-in-100-year" event means...

that if you look at any time interval of duration T, you will on average see T/100 of those events.

So, when a meteorologist says "this was a 1-in-100 year flood" they mean that our historical data indicates that in any given 100-year interval, you would expect to see a flood like that once. Or, if you looked at a 1000-year interval, you'd expect to see 10.

you would expect to see one of these storms only rarely in any given 1-year interval (about 1% of them). Seeing two such floods in a year interval is that much more rare: 0.00005, or "5 in 100,000" odds.

[Duer 157099]

12. I get it. The more frequent it is, the more rare it is

Wait, whut? :shrug:

[nadinbrzezinski]

2. Holly mother of God

That is ugly...

This is what climactic change looks like btw.

[givemebackmycountry]

3. Susquehanna River Binghamton stream gage in operation since 1847, is due to be shut down...

Budget cuts.
Nice.

[phantom power]

6. I make out the probability of these storms in consecutive years at about 0.00005

Modeling it as a Poisson process with time unit of one year and lambda = 1/100.

[dtexdem]

8. Wow, and they could go for a third next year.

Or even another one this year.

[phantom power]

9. That would be bad. I'd have to recompute the probability.

And, also flooding.

[disillusioned73]

10. Yeah, we got our buts kicked pretty good..

mother nature surely reminded us of whos in charge - I had to reroute my trek to work this morning 3 times before getting through. I am located SW of Allentown on that map.

[jberryhill]

11. Three Mile Island's Flood Defenses Are Based On An Obsolete Model

I posted some screenshots I grabbed from an Exelon engineer's powerpoints a while back (shoulder surfed on a flight), which suggested the 100 year flood stage calculation used in TMI's flood defense system was no longer valid, and that the existing dam system (remember that plant in the Midwest recently) may be several feet low. TMI is built on a silt island smack dab in the middle of the Susquehanna.

There are some vague references to that in some NRC docs from last year, but the gravity of the situation seems understated.