Scatterometers and millibars: Inside the forecast

The National Weather Service posted an explanation on their Facebook page

A tornado caused damage throughout Manzanita, Oregon on October 14, 2016. (KOIN)
A tornado caused damage throughout Manzanita, Oregon on October 14, 2016. (KOIN)

PORTLAND, Ore. (KOIN) — There is no doubt the storms that hit Oregon — bringing tornadoes to the coast, downing trees and leaving thousands without power — were big storms, significant events that merited widespread warning and attention.

But the storm on Saturday was more of a glancing blow than a direct hit. That left some people — as people are prone to do — complaining that the storm was oversold and overhyped and over-everything.

Let’s be honest. If you hadn’t been warned at all, it would have been a lot worse.

Still, the National Weather Service wanted to explain what happened with their forecast. Below is their entire Facebook post from late Saturday. It includes great words like “scatterometer” and “millibars.”

The National Weather Service Facebook page:

Why were winds not as strong as forecast along the coast?

In an effort to improve our forecasting, meteorologists commonly reassess what happened after each weather event. This process becomes even more important when a forecast doesn’t work out the way we expect it to. The impacts were still quite significant, especially inland where deciduous trees still had full foliage. We received several dozen reports of trees down throughout the Portland metro area. Therefore, we are glad we issued a high wind warning for the north Willamette Valley. However, winds with this storm were significantly less along the coast than we were expecting. We will spend the next several weeks digging into the details of this storm, but for now, here is a quick look at our first assessment of why winds were not as strong as we expected, especially along the coast.

First, the strength of a low pressure center is directly proportional to the strength of the winds. Most forecast models, even this morning, had the low pressure center between 968 and 970 millibars off the Astoria coast this afternoon. While the models did a fairly good job with the track of the low, the strength of the low was not well forecast by models. The surface low west of Astoria this afternoon was around 978 millibars. This is 8 to 10 millibars less intense than forecast. This is VERY significant.

Second, an advanced scatterometer from satellite passed over our low pressure system at 10:45 AM PDT. This scatterometer shows the winds near the surface. What we saw left all of us here scratching our heads. There were 2 centers of circulation! In a typical low pressure center there is only one center of circulation and there were no forecast models forecasting this second low. When the energy is spread between 2 low pressure centers instead of one, the central pressure and winds associated with the system are typically significantly lower than they would be if it were one center of circulation.

Overall, our forecast wind speeds were pretty close (maybe 5 mph too low) in valley locations. Also, we are glad conditions were not as extreme as we were forecasting for the coast. This is better for everyone who may have been negatively impacted who now came through unscathed. However, when a forecast does not work out as expected, it is frustrating as a forecaster. Weather science and model forecasts are getting better every day, but this is just another reminder that Mother Nature will always keep a certain level of unpredictability. We will continue to research and improve our forecasts to provide the best forecast possible for the great people of northwest Oregon and southwest Washington.

Until the next big storm,

NWS Portland