Satellite picture for western Washington around 1 PM Saturday
The National Weather Service's forecast for Saturday had a problem. For central Puget Sound it was predicting clouds in the morning and then becoming mostly sunny (see below)
But instead fog held in ALL day, something shown by the Space Needle PanoCam at 3:10 PM Saturday. You don't see conditions like that holding in all day very much.
And the UW high resolution forecasts of fog/log clouds for 5 PM Saturday started with fog in the morning and rapidly burned it back (see forecast at that time).
In fact, this error--the unrealistic burn off of low-level cloud--has happened several times during the past few week, with fog and low stratus showing unusual persistence.
Local meteorologists and some of the NOAA modelers have been musing that perhaps, just perhaps, there is a reason for this uber-persistent low clouds and our model' inability to get it right: wildfire smoke.
And there are good physical reasons why smoke might be the explanation.
For example, smoke scatters some of the solar radiation back to space, lessening the solar heating available to burn off the fog.
In earlier blogs I showed the effect clearly by presenting the solar radiation on the roof of my building before and after the smoke moved in on Sept 11th (see below).
And there is one.
Cloud droplets generally form on small particles in the atmosphere called Cloud Condensation Nuclei (CCN). Smoke particles can act cloud condensation nuclei. So adding smoke can lead to MORE cloud condensation nuclei and MORE cloud droplets (see left side below). If the amount of water available doesn't change (which is normally the case) then having more nuclei leads to more, but smaller cloud droplets . And it turns out the having more particles leads to the cloud being more reflective, which REDUCES the solar radiation available to burn off the cloud.
And there is more! Having a lot of small particles works against producing drizzle, which takes moisture out of the fog/stratus. Bigger droplets become heavier and fall out more quickly, collecting other droplet as they fall. Thus. having the moisture in smaller droplets helps maintain the fog!
Now here is the key issue. Most models, including the UW WRF and the NOAA HRRR model do not properly simulate these cloud effects and thus fail to forecast the ability of smoke to keep the clouds around.
The above is a physically plausible explanation, and certainly there were enhanced values of small particles in the lower atmosphere over Puget Sound on Saturday (see the small particles concentration at Seattle's Beacon Hill below).
Some of my colleagues at the NOAA Lab in Boulder, CO believe the above is a real possibility, and during the next month UW graduate student Robert Conrick will be exploring this mechanism with sophisticated modeling experiments. Will let you know what we find out.
Such situations are why science is so much fun, with interesting theories to test and real practical value when we solve the mystery.
_______________________________________________
Stream my podcast from your favorite services:
Want to support the creation of the blog and podcast?
____________________________________________________
My blog on KNKX and the Undermining of American Freedom is found here.
An extraordinary story about Matt Martinez, Program Director of KNKX, is found here. All KNKX listeners should read it.
Cliff, this is great and something I was wondering about! Can you explain the process of fog getting burned off by the sun sometime? I’m a little lost on how that process works.
ReplyDeleteI believe there are two mechanisms by which this happens. First, as the sun warms the foggy air from the top down, the increase in temperature increases the vapor pressure of the fog droplets (or, is popular terms, it increases the amount of vapor that the air can hold) and the droplets evaporate. But also, once some of the sunlight reaches the ground, it starts to warm, and this starts the convection process which allows the air at ground level to mix with air higher up, which is usually drier.
DeleteThis is very interesting. Thanks!
ReplyDeleteOne would think this was self-evident given the layman's common understanding that as the day progresses sunshine burns off the fog, so if there is less sunshine due to a layer of smoke, there is less burning off of the fog.
ReplyDeleteAnother observation from Saturday night: the air smelled somewhat salty in Kirkland as the fog returned. I was wondering whether this could be due to the fog carrying Elliott Bay salty air better than less moist air would.
Is there a way to measure fog droplet size directly?
ReplyDeleteHow interesting!
ReplyDeleteI am back to walking and part of it was due to the fog (I love fog) I went for a walk late in the morning and it was not as heavy as it was earlier. Friday I went out early and both mornings the advisory was fog to lift by noon at the latest.
Friday, we did fully lift and had a rather sunny afternoon, but Saturday, the fog never did completely lift by late morning and remained cloudy all day before the fog began to roll back in. By 6:22pm, it was definitely rolling back in, not thick as yet, but it was obvious we had fog. It was misty Saturday morning for sure, not enough to warrant a parka, but cool enough that my ski jacket was necessary as I went for a walk.
Awoke yesterday morning to light rain and the roads wet, ultimately, it did dry out by the afternoon so no walk ensued.
This morning at 0 Dark thirty, still dark out, no fog as I can tell, but dew for sure as my car has lots of it on its surface out front but the streets are dry so no rain. The Accuweather says, 54 and mostly cloudy right now. The forecast had been saying in the low to mid 60's but both Saturday and yesterday, we didn't get beyond 57-58 all day and I think Friday too. We are expected to hit 67 by end of day, I'll believe that when I see the official high later today. The lows feel about correct, low to mid 50's.
Not to mention the burning cities thanks to Antifa...
ReplyDeleteHi Cliff. We have had unseasonably warm nights lately. Has there been a string of record high lows as I’m guessing? Where can one find info on that?
ReplyDeleteBlog readers may find these of interest:
ReplyDeletehttps://www.weather.gov/media/rah/science/2016.NWA.RAH.Hartfield.Superfog.pdf
https://www.google.com/url?q=https://youtu.be/X0iwFi8muOw?t%3D7916&sa=D&ust=1601915693995000&usg=AFQjCNFO-q1UBj7LT9umBTh_y7kxfIChOg
Thank you for these fascinating insights, Cliff. I would love to hear your thoughts on a theory of mine. My empirical observations during our week of smoke left me with the impression that per unit mass, smoke seems to reflect / absorb more solar radiation than water vapor does. My poor man's thermometer (my bare skin) just "feels" much cooler than I would expect when it's smoky compared to a similar density of clouds. Do you know if there is any scientific basis to this? If so I'm sure that could be applied as a "correction factor" to the models.
ReplyDeleteI went hiking near Mt. Baker on Saturday (the beautiful "Chain Lakes Trail"); it was sunny and very warm there, even at elevation (my car gave a reading of 72 degrees as the outdoor temperature at the Bagley Lakes Trailhead (elevation ~4300 feet); then afterwards I paid a visit to Blaine, where it was also sunny and warm. Then, driving home southbound on I-5, leaving Blaine at about 4 PM, once I got south of Bellingham, the sky went from bright and blue to dark and overcast, and the temperature dropped to about 59 degrees. Very interesting.
ReplyDeleteWasn't the explanation of such "pea-souper" London fogs that there was so much coal dust/smoke in the air (adding fog nucleation)? And that also explains why there aren't such London fogs anymore - as they've shifted from coal use? (we have pollution to thank for such a setting for mysteries)
ReplyDelete