Wednesday, February 10, 2016

Can Mother Nature Crack This Terrible Inversion?

The situation in the Salt Lake Valley continues to deteriorate and is becoming quite serious.

Here on campus, we are currently fogged in.  Although depressing, it doesn't look as ugly as the pollution does under clear skies, but don't be fooled.

PM2.5 concentrations this morning are sky high.  The University of Utah's Trax-mounted PM2.5 sampler is measuring concentrations this morning in the 85-115 ug/m3 range (maroon filled circles below).

I did a quick inspection of the data and found a maximum concentration of 111 ug/m3, which is well into the unhealthy category.  

The terrible situation we are in is one of our own doing.  Persistent cold pools (a.k.a., inversions) are naturally occurring phenomenon in the western United States during the winter.  One shouldn't equate inversions with air pollution.  Inversions happen all over the west.  Pollution happens in those areas that experience inversions and where emissions are concentrated.  

At this point, we're up to our eyeballs in alligators.  We have absolutely terrible air quality and things are not going to improve until Mother Nature cracks this inversion.  What will that take?
The situation we have right now is one where dense, cold air is pooled in the Salt Lake Valley (and other basins of northern Utah).  From March to October, there is typically enough energy provided by the sun each day to warm that airmass and allow it to mix with the air aloft, limiting pollution concentrations.  This time of year, however, we don't have enough solar energy to mix out the cold, dense airmass.  We're basically mired in an oil and water situation.  I did a quick calculation and found that the air near the valley floor is about 25% more dense than that at crest level.  That's a big number.  

Thus, there are only two ways to get rid of this inversion.  One is to bring in even denser, colder air at upper levels, allowing what meteorologists call buoyancy driven turbulence to scour out the valley.  The other is to increase the winds near the top of the cold pool, allowing what meteorologist call mechanically driven turbulence to scour out the valley.  

One can do another quick calculation and find that for a valley temperature of 0ºC one needs to bring in an airmass with a crest-level temperature of about -17ºC to remove the cold pool with no help from the wind.  You can get a way with a higher crest-level temperature if the valley is a bit warmer (such as might be found in the afternoon), or if clouds are present (the release of heat in clouds helps invigorate turbulence).  So, perhaps we could do some damage if we got down to say -12ºC or so. 

Of course if you add wind, you can sometimes get a bit more bang for the buck.  It's very hard, however, to determine how wind will influence a cold pool unless it is very strong.  

Which brings us to our sole glimmer of hope, the trough that is forecast to brush by Utah on Saturday evening.  This will drop our crest-level (700-mb, 10,000 ft) temperatures to about -6ºC to -8ºC.

That by itself is not enough to crack the inversion, but there is also an increase in flow aloft and at low levels.  

We really don't have the tools today to determine if this will yield a full mix out (I doubt it, but can't rule it out), partial mix out at all elevations, a scouring out of the pollution from the top down, leaving a lens of pollution near the valley floor, or no change.  This is in my view a critical area for research as it is forecasting of these weaker trough passages and their influence on inversion and air pollution strength that is the hardest part of forecasting these events. 

As things stand now, all we can do is hope for the best, but I'm concerned that even if we mix out some, we're still going to see this event persist into early next week.  


  1. Does hygroscopic growth of particles in high humidity conditions (like we have this morning) inflate the PM2.5 mass concentrations in that water is condensing onto the aerosol particles?

    1. When the humidity is high enough, soluble particulate matter (such as ammonium nitrate, NH4NO3) deliquesces into the aqueous phase, meaning that it "absorbs moisture from the atmosphere until it dissolves in the absorbed water and forms a solution" (encyclopaedia britannica). The deliquesced particulate matter is no longer in its typical, solid phase, and can be transported in the valley atmosphere by fog water. Air quality measurements in fog can be skewed a couple of different ways depending on the methodology of the instrument; an optical spectrometer that measures concentrations using a laser (as DAQ uses) won't register anything not tuned programmatically in its absorption band, and an instrument that measures weights directly (such as an E-Sampler) won't pick them up fully after it spins the sample through the vortex, and the solids fall out. Most of the weight measuring devices are tuned for certain RH values, and lose considerable accuracy at RH higher than is set in the instrument. So, long story short, kind of - but not really. A skew in the data can become present depending on the device, however concentrations reported by an agency such as DAQ shouldn't really register deliquesced particulate matter, as the absorption bands within the instruments are tuned for solid particulate matter and not particles that have absorbed water into the aqueous phase.

      Or at least, this is the extent to which I understand the workings of these instruments, someone with more experience on mass spectrometers may comment further...

    2. Thanks for commenting Will, especially since PM measurements are above my pay grade.

    3. Thanks. That is great information. I wonder if the measurement technique is producing the discrepancy between the Hawthorne site and other sites mentioned below. It sounds like the Hawthorne site uses a laser spectrometer and the MTMET and NAA sites use E-samplers. They agreed fairly well until this morning when we had much more extensive fog than previous days.

  2. The Trax readings along 200 West are now around 120. Any idea why the DEQ station is so much lower (67.5)?

    For SL school district recess is indoors when PM 2.5 > 90. I wonder which data various schools look at.

    1. I suspect (and hope) that 90 is the air quality index, rather than the mass concentration. If you go to, the weird bar graph on the right shows the mass concentration on the left and the air quality index on the right.

    2. No, 90 does not refer to the AQI. 90 ug/m3 is the level at which all kids are kept inside for recess. Below 90 only "sensitive students and students with respiratory symptoms should remain indoors for recess and PE classes."
      The policy refers to the Hawthorne site and thus today's air was nowhere near dirty enough to keep kids inside at recess.

      Thanks again,

  3. TRX02
    2016-02-10 10:35 Local
    129.98 ug/m3
    (highest current)
    2016-02-10 11:15 Local
    122.00 ug/m3
    2016-02-10 09:00 Local
    67.50 ug/m3

  4. With regards to these sensor questions, I have to defer to others with regards to some of the issues at play. I will simply note that PM is a mess of solids, liquids, and semi-volatile compounds and it is very difficult to measure. Different instruments and techniques have their strengths and weaknesses and some variability amongst instruments is to be expected. We have enough data to confirm that the air quality is in the unhealthy category, even if there is some uncertainty in the absolute magnitude. In addition, the spatial structure inferred from Trax is certainly good qualitatively.

    As noted above, this mornings observations show large differences between QHW and the Trax measurements that I've seen. Some analysis of why is warranted.


  5. Further divergence:

    200 West ~2000 South
    2016-02-10 15:10 Local
    145.20 ug/m3

    2016-02-10 14:00 Local
    49.00 ug/m3