Escaping Smoke is Hard

During winter, the Salt Lake Valley often sees pollution episodes with elevated particulate matter due to the presence of low-elevation cold pools that persist due to the limited solar heating.  This enables particulate matter that forms from the emissions from automobiles, industry, home heating, and other sources build up within the cold pools, resulting in poor air quality.  This poor air quality is reflected in elevated levels of particulate matter with a diameter of 2.5 micrometers or less, or PM2.5.  Such events are known colloquially as inversions.

Because the PM2.5 is typically trapped in the cold pool, one can usually find clean air by moving to higher elevations.  The depth of the pollution varies from event to event and sometimes from hour to hour as the cold pool sloshes around in the valley.  At times, the benches may be in the clear air.  At other times, elevated PM2.5 can extend higher.  

An example, from December 15, 2025, is shown below.  Taken from about 6000 feet in the Avenues foothills, one can very clearly see the layer of pollution within the valley, but clear air at upper elevations.  

PM2.5 associated with wildfire smoke in the summer is more complicated for a number of reasons.  First, the heat and intense vertical motions associated with wildfires can inject wildfire smoke anywhere from the surface to more than 15 km.  Because the wind often varies in strength and direction with height, the transport of that wildfire smoke can also vary in direction.  As an example, note how smoke from the Grasshopper Fire in northern Oregon spread in two different directions yesterday morning. 

Source: College of DuPage

As a result, smoke can be found even at the highest elevations and PM2.5 concentrations do not necessarily decrease with height.  Sometimes they can even increase with height.  

I had planned to hike up Snowbird this morning, but thought this might not be a good idea after looking out the window and taking a look at the PM2.5 estimates from Purple Air.  Unlike in winter, elevated PM2.5 wasn't confined to the Salt Lake Valley.  One could find values > 35 ug/m3 (i.e., unhealthy for sensitive groups) even at some locations in the Snyderville Basin.  Although there are fewer upper-elevation sits, at 8:57 AM MDT the highest PM2.5 was actually near the mouth of Little Cottonwood (53-68 ug/m3), Red Pine Lodge at Canyons (83 ug/m3), and Silverfork in Big Cottonwood Canyon (61 ug/m3).   

Source: Purple Air

It's hard to say if the top of the tram would have been above the smoke. The 9 AM webcam suggested it was close, but smoke clearly enveloped the terrain to high elevations.   

Source Snowbird

I ended up going for a less strenuous walk in the Avenues foothills.  Although the PM was still elevated, my back likes having the kinks worked out. Visually inferring PM2.5 concentrations can be tricky for a number of reasons, but the view toward the Oquirrhs seemed to confirm that this morning concentrations may have been higher above the valley floor as one could see the base of the Oquirrhs (red arrow) but not the mid or upper elevations (blue arrow).  This doesn't mean the valley floor had low PM2.5 values, but just that they appeared to be lower than at mid elevations.  

The National Weather Service has a modeling system to predict wildfire emissions and smoke transport and concentrations known as HRRR smoke. Its analysis for 1300 UTC (0700 MDT) this morning actually showed higher concentrations in upper elevation areas (e.g., Stansbury, Oquirrh, and Wasatch Ranges) compared to the intervening valleys. Although concentrations were lower than observed, it at least had the right idea. 

The HRRR smoke forecasts suggests perhaps some improvement later today and tonight.  We will see if that pans out.  Smoke forecasts are tricky.  

Note that the situation this morning with lower PM2.5 in the valley cannot be generalized to all wildfire situations.  There are times when PM2.5 concentrations from smoke can be higher at lower elevations.  Escaping wildfire smoke requires more investigation (and maybe a bit of luck).  The HRRR smoke forecast can be helpful.  I like to access them at https://sites.gsl.noaa.gov/desi. Select "HRRR-CONUS" as the dataset, an appropriate runtime (those starting at 0, 6, 12, or 18Z will extend out 48 hours whereas the other go to 18 hours but are more recent), and then "surface smoke" from the surface drop down menu at top.