Monday, December 11, 2017

Diurnal Intricacies of the Inversion

PM2.5 concentrations during our current inversion event have shown remarkable variations from day to night.

Below is a time series of PM2.5 measured at our mountain meteorology lab at the University of Utah showing a clear long-term upward trend, but also a tendency for PM2.5 concentrations to spike just before noon, remain elevated until mid to late afternoon, and then decline.

Source: MesoWest
What are the causes of this diurnal behavior.  There are several possible contributors.

First, there is the possibility that photochemistry - chemical reactions occurring in the presence of sunlight, are contributing.  Comparison of the above plot with the incoming solar radiation below shows some relationship, with the PM2.5 exhibiting a bit of a lag relative to the solar radiation.

Source: MesoWest

Another possibility is that temperature is playing a role since it also affects the PM2.5 chemistry.  Again, there is some correlation.  

Source: MesoWest
Finally, there is the transport possibility as the winds are also changing diurnally, with a good correlation between wind direction and PM2.5 concentrations.  

Source: MesoWest
Another perspective is provided by the someone hacked-up graphs below, based on data collected at the University of Utah by our MesoWest team over the 24-hour period ending this morning at 10 AM (the hacking reflects my splicing of their multiple graphs together).  The top figure is derived using a laser that is shot vertically through the pollution.  The color fill is backscatter, a measure of how much of the laser light is reflected back to the ground, with higher values roughly correlated with greater PM2.5 concentrations (brown-white being the dirtiest air).

The plot begins on the left at 10 AM on Sunday when the local flow just shifted to predominantly westerly (some variability from SW-NW).  Surface PM2.5 concentrations during this period are quite high and, in addition, the pollution is quite deep.  At just after 1700 MST (5 PM), the flow shifts abruptly to ENE, which reflects the onset of down valley flow from Red Butte Canyon.  This marks the beginning of a gradual decline of surface PM2.5 concentrations, as well as a decrease in PM2.5 concentrations aloft.

Source: MesoWest
There is a brief lull in the wind that occurs just before 2300 MST (11 PM MDT), with the flow becoming somewhat erratic.  Without the inflow of cleaner air from Red Butte Canyon during this period, the surface PM2.5 values climb, although things don't change too much aloft.  Finally, after midnight, the ENE flow returns and PM2.5 values drop again, although there are a few spikes during the night that may correlate with declines in wind speed (I haven't bothered to check take this comment for what it is worth).

At the end of the time period, the PM2.5 values climb again, abruptly, when the flow shifts to westerly.

All of this illustrates some of the intricacies of these inversion events.  Pollution concentrations vary in the vertical (yes, there is clean air up there), although if you look carefully at the plot above, you can see that it's not as simple as polluted air near the ground and non-polluted air aloft.  There are layers.  In addition, pollutant concentrations vary horizontally and at the University of Utah one can clearly see the migration of pollutant-laden air onto campus when the wind shifts to westerly in the morning.

What role photochemistry and temperature play in all of this is unclear to me.  I suspect it plays a secondary role compared to meteorological factors, but I am not an atmospheric chemist and over the years I've learned that when all you have is a hammer, everything looks like a nail.  In other words, as a meteorologist, I might be guilty of placing too much weight on meteorological factors.

One thing to keep in mind is that not all inversions look or behave like this and even this one might behave differently in the days to come.  As a scientist, I think what we see over the next few days will be "interesting."  As a citizen, I wish the damn thing would just blow away.


  1. Can you clarify something for me? Some people assert that while today's inversions suck and are terrible, they are not nearly as bad as what SLC used to experience 30-40 years ago, in the 70s and 80s.

    I know real data is hard to come by relative to the inversion that far back, but do you have an opinion or position on that?


    1. In responding to this question, it is important to distinguish between an inversion, which is a meteorological phenomenon, and air pollution.

      Recent studies have shown that there have been no long-term trends in inversions over the past 40 years. There are year-to-year variations in the severity and intensity, but no evidence that inversions are becoming worse with time.

      PM2.5 was not measured in the 70s, but trends in a mixture of particulate matter measures suggest that there has been some improvement in air quality over the past 40 years or so.

      So, the view that "inversions", or more correctly air pollution, was worse in the 70s is probably accurate. Certainly the PM10 record, which covers the period since 1985 or so, shows this pretty clearly.

      Reference: Whiteman et al. 2014 (


    2. dug,

      For what it's worth, the trend seems to be *slight* improvement or steady: The black dashed line is the old federal standard. The red line is the new federal standard. Though in the past few years Salt Lake has definitely exceeded the red dashed line.

      That the trend is relatively steady despite huge population growth is somewhat impressive. Moving forward the switch to Tier 3 gasoline is projected to help slightly:

      For fun, here is a clip from the 1940s when Salt Lake had a "smoke control division" to help with inversions: The problem now isn't direct emissions like it was back in the 1940s. Now most of our pollution particles get created by the sun in mid air through chemical reactions. So we have to stop the precursors to those particles.

  2. This is an extremely interesting discussion. From personal experience (I attended the U of U Meteorology program 77-84), there appeared to be more visual particulates in the 70’s. The winter of 80-81 was extremely bad...very long periods of inversion weather. The Kennecott smelter and stack spewed (although they had built the larger stack by the early 80's), as did the Geneva Steel Plant in Utah County...and many other uncontrolled industrial companies throughout the Wasatch Front, including the Murray Smelter (Superfund site). What seemed especially noticeable were the refineries along the north side of Salt Lake City and in Davis County. I knew at the time there was a lot of burn-off going on, but they didn't have the controls and filtering systems we have today. There was lead in gasoline fuel and almost everyone had a fireplace or a wood burning stove (and they used them). I can still remember the smell of burning wood everywhere during the cold months in the 60's, 70's and 80's. From the turn of the Century through the 1950’s, until Natural Gas became widely available, everyone burned coal and wood to heat their homes, or used all three; coal, wood and Natural Gas. Though the population was much smaller back then, I would venture to say the amount of soot, particulates and other pollution was probably at its worst historically in the Salt Lake Valley between 1920 and 1950. Visually, the coal burning caused the snow to turn a blackish color (accounts from my father and grandparents). 1932-33 was a brutally cold winter, especially in February of 1933. Everyone was doing everything they could to keep their homes warm.

    Dan Pope