Recap of last year's wildfire smoke season

It has been awhile since I posted, and I really don't want this blog to wither. But the absence has been for good reason. Last summer, I went on maternity leave and had high hopes of working on this blog during my "time off." That didn't happen. I made a few drafts, but never sat down and finished any draft. So, I wanted to do a recap of last season as I start thinking about 2019.

Below is a plot showing the number of days when there was a fire and the number of days when there was smoke in the atmospheric column for the wildfire season (May - October) of 2018. A couple of notes, I'm using satellite products here that I have "gridded." This doesn't represent the number of fires, but the number of days when there was fire in that grid (a defined area). A grid could have multiple fires at the same time and fires could burn for multiple days. But, it gives you a sense of where fires were occurring. For the smoke plot, this is again from a satellite product. It notes when there was smoke in the atmospheric column, so this does not equate necessarily with surface air quality. Also, this satellite smoke product is only for daytime, so it doesn't show plumes at night, and it might miss very small plumes. But, again, it gives you a sense of where smoke was. I also like to compare these to different years to kind of get a sense of severity (you can look back at my analysis of 2017).

Maps showing the number of days with a fire (left) in a gridbox or smoke plume (right) in a gridbox for May - October 2018. These are created using the Hazard Mapping System Fire and Smoke Product. 

What these plots show is that there were lots of fires in the Southeast, which is normal. These are generally smaller fires, and many of the fires are prescribed fires. If you compare to the smoke plot, you can see that these fires aren't creating a lot of smoke. Most of the smoke comes for the large fires in the Pacific Northwest and Canada. These fires create large plumes that can travel long distances and impact a lot of the US. But again, this is smoke in the atmospheric column, not necessarily at the surface.

To see the impact on surface air quality, we can look at the PM!d2.5!n measurements in Fort Collins. The following is a time series of the 24-hour (daily) average.

Concentrations were pretty high last summer, and we had a lot of days with smoke. Early in the summer, we had smoke from some local fires (Wyoming and Colorado). Later in the summer, we had smoke transported from the Northwest.

Smoke plumes on 14 June 2018, 29 June 2018, and 1 July 2018. Note the different scales (first picture is zoomed in on Northern Colorado).

On the plot, the two days with the highest concentrations were 20 August 2018 and 24 August 2018. Even though I was on maternity leave, I was still taking smoke pictures as you can see below. You can tell just from comparing the pictures (and seeing how much the foothills were obscured) that concentrations were higher on the 24th of August.

Looking west from Galway Dr. in Laporte, CO on 20 August and 24 August 2018. 

We can again use the HMS product to determine where this smoke originated. On the 20th, there were fires all over the Northwest and plumes coming from all over. In Fort Collins, surface concentrations were lower overnight and then ramped up and stayed high throughout the day (shown in the hourly time series below).

HMS Smoke Plumes and Fire Locations for morning, early afternoon, and late afternoon on 20 August 2018.

Hourly PM!d2.5!n in Fort Collins, CO for 19 August - 22 August 2018. 

On the 24th, smoke was also coming from fires in the Northwest and Canada. The plumes were pretty broad, covering a large portion of the western US, but there weren't as many as on the 20th.

HMS Smoke Plumes and Fire Locations for morning, early afternoon, and late afternoon on 24 August 2018.

The WE-CAN field campaign (https://www.eol.ucar.edu/field_projects/we-can) was happening during this time period, and they actually sampled a lot of these plumes on their flights. It will be pretty interesting to see all that they learned about smoke and air quality!

May-October (wildfire season) summary for Fort Collins

The wildfire season has come to a close, so I thought I would just give a quick overview of the summer air quality as it pertains to PM_2.5.

We had several days of smoke this season, with very little of it being from local fires.  The following is a map of the CONUS showing the number of days with HMS smoke plumes from May-October 2017.

Number of days with an HMS smoke plume for the period of May 1^st - October 31^st, 2017. Data from satepsanone.nesdis.noaa.gov.

We obviously did not experience as many smoke days as the Northwest, but Fort Collins still had a significant number of days where the HMS product noted smoke compared to previous years. According to the HMS product, there were 41 days when smoke was noted over Fort Collins (2016 had 30; 2015 had 23; 2014 had 21; 2013 had 34; and 2012 had 69 days).

The maximum hourly concentration at the Fort Collins site was 106 ug/m³ and the maximum daily concentration was 60 ug/m³. (as a note, all PM data I show on this sites is from the real-time reports and have not been corrected or validated, meaning that these values may change). Both of these occurred on Labor Day (4^th of September). You can read the blog post on that event here. This was the only day where the 24-hr concentration was greater than the EPA's standard of 35 ug/m³. However, as this was due to transported wildfire smoke, it will likely be classified as an exceptional event (which means it will be excluded when determining attainment status). Below is the time series of the 24-hour average PM_2.5 concentrations in Fort Collins.

Time series of PM_2.5 concentrations in Fort Collins, CO from May through October 2017. Data from https://www.colorado.gov/airquality/.

You can see from the time series that our air quality (as it pertains to particulate matter) is generally pretty good here in Fort Collins. Most of the days with elevated PM_2.5 concentrations had smoke present. Earlier in the season (May and June), we had a few high PM days which we can't specifically attribute to smoke because the satellites do not clearly show plumes over Fort Collins. However, there were lots of fires in Mexico, Arizona, Utah, and New Mexico that were producing large smoke plumes. It is feasible that we got a little of that smoke here but that the plumes were much more diluted by the time they reached us so as not to be as distinct in the satellite images. There's also the possibility that these were due to more local sources as other monitoring sites in Colorado did not see the same PM increases. One exception would be June 27^th. The HMS plumes did not extend as far north as Fort Collins, but all of the sites around Colorado did show increases in PM_2.5 concentrations, and the HMS product did show smoke from a fire in Utah transported over all of Colorado on the 28th (we weren't on top of our blogging game during that event).

Some side thoughts: Environmental regulations have really helped improve air quality overall in the US. We have seen decreases in anthropogenic (meaning from human activity) emissions that have led to better "average" air quality. However, at the same time, we've also had lots more large wildfires that have produced extremely poor air quality for much of the western US. Reducing wildfire emissions is obviously difficult as it means stopping large wildfires before they start (land management, prescribed burns, people being more responsible, etc.). Thus, wildfire smoke is becoming an increasing health concern as it could be offsetting some of the air quality benefits gained from reductions in other emission sources. Know how to protect your health during smoke events and stay aware of current  air quality conditions!