Saturday afternoon was a perfect example of urban flooding. Streets and a few homes/businesses saw flooding in parts of the metro area…apparently mainly in the downtown and eastside areas since almost all the pics/video seems to be from there. Yet there were no flood warnings for area rivers (none were needed) nor was there any flooding on any sizeable streams in our area. The NWS did issue an Urban & Small Stream Flood Advisory. The event was almost perfectly forecast by our models…take a look at our RPM forecast from Saturday morning with the hourly rain accumulation highlighted:
Sheets of rain from 2-3pm (it was an hour or so late), and for several days models had been advertising this very intense band of rain passing through the metro area in the afternoon. I didn’t work Friday-Sunday, but highlighted in the middle of the Thursday evening blog post was this:
8:51am Saturday I posted this on our FOX12WEATHER Facebook page:
Seems pretty clear-cut…Portland airport was doused with .67″ rain from 3-4pm, and .50″ in the two hours leading up to it. Great forecast by models and forecasters.
But you can’t have that much rain within such a short time in urban areas and not expect flooding. Urban locations are covered in concrete and water can’t do down into the soil. Plus we had leaves blown off trees that were blocking drains. I’m pretty sure it wouldn’t have been so bad to see the same thing anytime after mid-November when the leaves are mainly gone.
Notice I keep saying URBAN flooding? I drove about 12 miles during the peak of the rain (kid shuttle) in far east Gresham and Corbett areas and only saw some small pockets of water here/there. In areas where there isn’t much concrete it generally isn’t a huge problem unless something is blocking drains.
So a great forecast, but I think communication could have been a bit better. As forecasters we tend to focus on widespread flooding, when rivers and streams rise and flood large areas. In the few times around here when it’s obvious that we’re going to see extremely heavy rain in a short period (this time it was obvious) we need to hit that harder and make a clear distinction between the localized urban flooding and the big-time river stuff.
Now of course the elephant in the living room is the missed cutoff time for the rain. The rain was about 2 hours later to leave on Halloween evening than forecast. We probably should have told kids to delay 2 hours and most of them in the metro area would have been dry. Note the 8pm radar image, just about the entire metro area was fine by then:
This is what that same Saturday morning RPM model was showing for 5pm Halloween evening:
and the 5pm actual radar image:
Models were too fast with the movement of the main rain band. Our models are far better than 10-15 years ago. Back then we would have just said DECREASING RAIN IN THE EVENING, which would have been quite a bit more vague. Nowadays we’re often able to pin things down to within an hour or two, but by giving these more detailed forecasts we’re opening ourselves up to a miss once in a while. Too bad it was during the prime Trick-or-Treat hours!
By the way, total rainfall accumulation was a spot-on forecast as well. Looks like metro area was in the 1.50-3.00″ range and Cascades were 6-12″…not so bad for an extreme rain event! Also note how well all the mountain rivers handled the rain. If rivers/reservoirs would have been full already, we would have seen big flood issues! As expected, the big totals were in the Cascades with lighter amounts in the lower Coast Range.
Rain totals are from 5pm Friday to 5pm Sunday (72 hr totals)
June Lk (nr Mt St Helens) 12.00″
North Fork (nr Mult. Falls) 11.40″
Mt. Hood SNOTEL (Tmbrlne Ski Area) 6.60″
Larch Mtn (WA) 6.70″
Saddle Mtn (nr F. Grove) 6.80″
Chief Meteorologist Mark Nelsen