Ryan Hisner Profile picture
Apr 22, 2021 17 tweets 7 min read Read on X
1/ The @nytimes has a good interactive story on the safety of flying during the pandemic, but they didn’t discuss boarding & deboarding. Judging by CO2 readings I saw on recent flights, this is the most dangerous part of flying. #covidco2 #COVIDisairborne
nytimes.com/interactive/20…
2/ By now there is overwhelming evidence #COVIDisAirborne. It’s transmitted mainly by shared air, i.e. inhaling the air others have exhaled, which contain aerosols—tiny, liquid particles that float suspended in air for seconds to hours, depending on size.
3/ There’s no easy way to measure virus levels in the air, but CO2 is a good proxy measure of risk. Outdoor CO2 levels are ~410 ppm, but since we exhale CO2, indoor concentrations are higher. The higher the CO2 level, the more air you’re sharing.
4/ I used an Aranet CO2 meter, which was tested & recommended by @jljcolorado, a top aerosol expert. I flew from Ft. Wayne, IN (FWA), to Rochester, MN (RST) with a layover at Chicago. FWA was not busy, & the CO2 ranged from 516 to 546 ppm.
5/ When we first sat down on the plane, the CO2 level was at 1424. Four minutes later, it was 1470. Over the course of the next 6 minutes, the CO2 concentration rose steadily.
6/ In the next four minutes, the CO2 level rose from 1654 ppm to 1899 to 2027 to 2255. Most IAQ experts, including @linseymarr, recommend keeping CO2 below 700 for a multitude of reasons.
7/ The next two pictures are quite blurry because the plane started to move, but you can tell the CO2 rose to 2343, then 2351 ppm.
8/ I think the ventilation system started bringing in outdoor air when the plane began taxiing but that it took a couple minutes for the CO2 to begin to decline. Two minutes after peaking at 2351 ppm, it had decreased to 2026, and two minutes after that it was 1738.
9/ By the time we took off the CO2 was around 1400 ppm, and it stayed right around that level for the entire flight.
10/ Chicago’s O’Hare Airport appeared to have quite good ventilation. The CO2 level stayed in the low-600 range where I was. I did not venture into the crowded cafeteria area nearby, where I imagine the levels were significantly higher.
11/ The next flight was to Rochester, MN, also on American Airlines. Immediately after boarding, the CO2 was already at 2096 ppm.
12/ Five minutes later, the CO2 concentration had risen to 2393 ppm, and in the next two minutes it rose further, to 2548 before peaking at 2650 ppm.
13/ At this point there was a revving noise, and CO2 levels started to decline, falling to 1820 ppm seven minutes later. 20 minutes after the peak 2650 reading, CO2 had declined to 1292 when we started taxiing.
14/ After taxiing for several minutes, we stopped. CO2 once again rose, but only to 1546, after which the plane started rolling again and levels declined.
15/ During this flight, which seemed equally crowded as the first flight and on a similarly small plane, CO2 concentrations were lower, ranging between 900-1030 ppm. Notably, relative humidity got as low as 12% later on in the flight.
16/ The Rochester Airport wasn’t very crowded, but with CO2 at 492 ppm, it still must have had good ventilation. Doors about 10 meters away likely helped.
17/ I’ll stop there for now. Later I’ll add info about the CO2 levels at Mayo Clinic and our hotel. Our return flights, again with a layover at O’Hare, are later today, and it will be interesting to see how they compare.

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with Ryan Hisner

Ryan Hisner Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @LongDesertTrain

Jul 2
Quick BA.3.2 update. Another BA.3.2.2 (S:K356T+S:A575S branch) from South Africa via pneumonia surveillance.

This means that 40% of SARS-CoV-2 sequences from SA collected since April 1 (2/5) and 50% collected after May 1 (1/2) are BA.3.2. Its foothold seems strong there. 1/3
2 interesting aspects of the new BA.3.2:
1. ORF1b:R1315C (NSP13_R392C)—This mut is in all Omicron *except* BA.3. So this may well be adaptive.

2. S:Q183H—First known antigenic spike mut seen in BA.3.2, not a major one, but one we've seen before—eg, LB.1/JN.1.9.2.1 2/3 Image
I think the unusually long branches in the BA.3.2 tree indicate 2 things:
1. Slow growth globally—fast growth results in many identical sequences, if surveillance is sufficient

2. Undersampling—BA.3.2 most common in poorer world regions with little sequencing of late. 3/3
Read 5 tweets
Jun 29
BA.3.2 update, Chapter: "I'm Not Quite Dead, Sir"

A new sequence from a traveler to the USA from the Netherlands was uploaded yesterday, with a collection date of June 17. 1/10 Image
This was a BA.3.2.1, the branch with S:H681R + S:P1162R (not S:K356T + S:A575S).

An updated, annotated version of the BA.3.2 Usher tree pictured below.

This sequence has the first new spike mutation since BA.3.2 emerged in November 2024—S:V227L. 2/10 Image
It has an extremely rare NSP5 mutation, ORF1a:T3487S (NSP5:T224S), only in 4 of ~17 million SARS-2 seqs

Intriguingly, 3 of these 4 share something in common w/this BA.3.2.

The first—and most remarkable—is a BA.2 from England that, like BA.3.2, has the ORF7ab-ORF8 deletion. 3/10 Image
Read 11 tweets
Jun 27
@yaem98684142 @TBM4_JP This analysis is extremely flawed.

There is nothing abnormal about BA.2.86 appearing in multiple countries shortly after discovery. This has been the norm lately w/reduced surveillance. 1/
@yaem98684142 @TBM4_JP The mutational spectrum analysis is poorly done. It cites a single study looking at the mutational spectrum in *three* immunocompromised individuals. Needless to say, this sample size is WAY too small. 3/
@yaem98684142 @TBM4_JP Furthermore, the IC people examined did not give rise to highly divergent variants with a large number of spike mutations. They appear to have accumulated a very modest number of mutations, with few substitutions in spike. The sequences themselves are apparently not published. 4/
Read 7 tweets
Jun 19
Interesting recombinant showed up today from Texas. It's a mixture of B.1.595, BA.1, and some flavor of JN.1. Most of the genome is from B.1.595. The ancestry of this one is clear: it directly descends from a B.1.595 sequence collected in January 2023, also in Texas. 1/11 Image
When the B.1.595 was collected this infection was >1 yr old, w/no sign of Omicron. BA.1 ceased circulating ~1 year prior.
Now a BA.1 spike appears w/just 5 changes from baseline BA.1, none in the RBD—S12F, T76I, Q271K, R765H, S939F.

This is a zombie BA.1 spike. 2/ Image
There are only a few signs of JN.1, & they're scattered. In ORF1a, we see JN.1's V3593F, P3395H, & R3821K, but the NSP6 deletion btwn these—universal in Omicron—is absent. In
M has JN.1's D3H + T30A & E19Q (in JN.1 & BA.1), yet A63T—also in both BA.1 & JN.1 is absent. 3/11 Image
Read 11 tweets
May 31
An awesome preprint on the novel, unsung SARS-CoV-2 N* protein came out recently, authored by @corcoran_lab & Rory Mulloy. I’ve previously written on N*’s demise in XEC, the top variant in late 2024/early 2025. But…
1/34
…this preprint, along with another great study by the @DavidLVBauer, @theosanderson, @PeacockFlu & others prompted me to take a closer look...
2/34biorxiv.org/content/10.110…
...and for reasons I’ll describe below, I now believe rumors of N*’s death are exaggerated.

First, XEC is in terminal decline, replaced by variants with full N* expression, so N* is back in fashion.
3/34
journals.plos.org/plosbiology/ar…
Read 35 tweets
May 15
@DameSunshine @SharonBurnabyBC B.1.1.529 wasn't/isn't a real variant; it's a placeholder that represents a putative ancestor of BA.1/BA.2/BA.3.

Bad sequences and/or coinfections tend to get categorized as B.1.1.529:—they have enough Omicron muts to be ID'd as Omicron but so much dropout/mixed signals...
1/
@DameSunshine @SharonBurnabyBC ...that a specific designation isn't possible. Travel sequencing in the US is done by Ginkgo Bioworks. Their sequences are generally poor quality & they upload *pooled* sequences—against database guidelines. The B.1.1.529 here are likely low-quality/pooled sequences from GBW.
2/
@DameSunshine @SharonBurnabyBC I think it's entirely possible that a new, divergent variant will emerge this summer. There are hints with BA.3.2 & a 50-spike-mutation BQ.1.1 that has transmitted at least once. Other similar chronic infection-derived variants are undoubtedly lurking all over, unsequenced.... 3/
Read 4 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us!

:(