https://www.isi.edu/~hardaker/news/20221106-ietf-c02-analysis/

UP | HOME

Analyzing Carbon Dioxide levels while attending IETF-115 in London,
UK

Table of Contents

  * 1. Introduction
      + 1.1. Virus Risk Measurement Approach
  * 2. 2022-11-05 Fri: Getting there - SMF -> LHR
      + 2.1. SMF to DEN
      + 2.2. DEN to LHR
  * 3. 2022-11-06 Sun: Day 0 - the early work day before the real
    start
  * 4. 2022-11-07 Mon: Day 1 - the meeting really begins
      + 4.1. special news bulletin: turn on your room fan. always.
  * 5. 2022-11-08 Tue: Day 2 - buses are a big problem
  * 6. 2022-11-09 Wed: Day 3 - a generally cleaner air day
  * 7. 2022-11-10 Thu: Day 4 - chairing two meetings
  * 8. Conclusions

1. Introduction

I went 2.5 years with no plane fights and minimal in-person contact
to suddenly "everyone must get back to normal! get on those planes!
let's have lots of face to face meetings!". I, like many others,
wanted to continue minimizing my risk of getting sick from airborne
viruses while acknowledging being successful in my career required a
lot of social networking. So, I've been vaccinated, I wear N-95
masks, and I make judgments along the way about what activities are
acceptable to participate in. But... until this trip to IETF-115 in
London, UK, I haven't had numbers to back up my decisions.

1.1. Virus Risk Measurement Approach

Just before this 6th-ish trip for the year I decided to purchase an
aranet4 \(CO_2\) sensor. There is some belief that by measuring \
(CO_2\) you can indirectly measure your risk to airborne viruses,
since more \(CO_2\) means you may be around more exhaled air. The
aranet4 is a factory calibrated, Non-Dispersive InfraRed sensor
(NDIR), and is considered to be one of the more accurate sensors for
consumer use. I used the aranet4 Android App to configure and gather
the results from the sensor (using its csv export feature).

The meter is kept in a cheap padded case (it doesn't come with one
and is known to be fragile). I either put it next to me where I'm
sitting, or have it otherwise near the open air sticking out of a
backpack pocket or clothing pocket. The picture below shows the meter
sitting in the Richmond 2 meeting room.

meter.jpg

Generally, I had the sensor recording \(CO_2\) levels every minute or
two. To analyze, produce and draw conclusions from the graphs below,
I ended up writing a python program that let me graph and annotate
the results.

For some level setting about what are reasonable numbers, I note that
my house itself (where just my wife and I live) has \(CO_2\) levels
of around 750 Parts Per Million (PPM) most of the time, while walking
around outside showed levels of 450-ish. When sitting in my
home-office, the \(CO_2\) levels often slowly climb over a few hours
to reach 950 PPM (it turns out my office really needs a fan as the
single plant on the opposite side of the room is failing to keep up
even with my sedentary exhaling).

Note that \(CO_2\) sensors do suffer from changes in altitude, but
the math shows that you can adjust readings based on the altitude
such that sensor readings are lower than they should be at higher
altitudes. Note that the graphs below aren't calculated for altitude
adjustment. Since planes are generally pressurized to an equivalent
of 8000 feet, and senor drift is typically \(3%\) per 1000 feet of
elevation, it means that while in the air the \(CO_2\) levels are
likely $24% higher than shown on the graph.

2. 2022-11-05 Fri: Getting there - SMF -> LHR

One of the largest fears of trying to avoid virus exposure while
traveling involves airports, airplanes, subways, and taxis (oh my).
My trip to London, Uk is divided below into two parts: the short-haul
flight from Sacramento, CA, US to Denver, CO, US followed by the
longer-haul flight from Denver, CO, US to London, UK.

2.1. SMF to DEN

The graph with marked interesting points along the way is below:

2022-11-05-smf-to-den.png

A number of things surprised me in these results:

 1. Airport gates/hallways are significantly better ventilated than I
    had thought they would be. Even in a crowded airport, I could
    find clean air space somewhere reasonably away from others.
 2. Tightly packed airline club lounges, on the other hand, were a
    nightmare of too many people and, thus, high levels of \(CO_2\).
    Though I found spots along the edge that were away from most
    people, it was impossible to find anywhere with a reasonably low
    level of \(CO_2\). The boarding gate I went to afterward was
    significantly better (though still not as good as the more empty
    Sacramento gate I started my trip at).
 3. I always get sleepy right when taking off on a plane, and now I
    know why: The recorded \(CO_2\) level of 2500 as we were taking
    off was significantly higher than desired. Note that at 1400 PPM
    it is commonly believed that mental cognition begins to be
    impaired.
 4. Though the measured \(CO_2\) levels were significantly higher
    than I would have liked, I actually rate my risk of viral
    exposure here much lower than the united lounge. My reasoning: I
    had the air vent streaming directly across my face, and the air
    from it supposedly comes through a HEPA filter which means it was
    likely fairly virus free even if the \(CO_2\) was that high.

    In the end, you must use logic combined with measurements to
    evaluate potential virus exposure risk.

Note: the airplane was a Boeing 737-900.

2.2. DEN to LHR

2022-11-05-den-to-lhr.png

Like the national flight, my international leg was less than ideal:

 1. The takeoff again had a high \(CO_2\) peak, as did the arrival.
 2. I had heard that international flights might have better
    in-flight \(CO_2\) levels. Nope.
 3. After arriving and getting onto the exit tunnel, the \(CO_2\)
    dropped to nearly outside levels, and stayed very low even to the
    point of waiting for the express train to London.
 4. Once on the train (a 15 minute journey), the \(CO_2\) began
    slowly climbing until I finally arrived at Paddington Station.
    I'm glad the ride wasn't longer.

Note: the airplane was a Boeing 787-9.

3. 2022-11-06 Sun: Day 0 - the early work day before the real start

Sunday at the IETF, at least for me, is full of greeting long-time
friends, attending side meetings, and the welcome reception, etc. The
graph below shows the extent of my day:

2022-11-06-ietf-sunday.png

Some thoughts:

 1. The levels of \(CO_2\) were fairly high while I was sleeping at
    night. Fortunately, my virus exposure was low (being alone), but
    I was unhappy with not having fresher air - my windows were
    locked shut unfortunately. (More on this problem later.)
 2. Early in the day (when few people were there yet, at least), the
    breakfast area had reasonably low \(CO_2\) levels.
 3. I have a feeling that the huge spike seen just after breakfast
    and before entering the side meeting was from an elevator ride
    that ended up getting fairly full and stopping at nearly every
    floor. By my memory of that time is hazy, unfortunately.
 4. The meeting room I was in all day had near outdoor levels of \
    (CO_2\), which was pleasantly surprising. (This turned out to be
    generally consistent throughout the week. Big rooms with lots of
    air circulation is certainly helpful.)
 5. The newcommers quick connections gathering, however, was held in
    a smaller room with shorter ceilings. With a generally high
    density of people, the \(CO_2\) levels climbed quickly in the
    measurement graph. I'm glad I wore my mask the entire time,
    though my voice got sore from shouting over the din in the room.
 6. The welcome reception, although in a very large room with high
    ceilings, also exhibited ramping and high levels of \(CO_2\).
    This was likely my highest risk point in the day, even though I
    spent much of the time in the nearby less crowded hallways.

4. 2022-11-07 Mon: Day 1 - the meeting really begins

2022-11-07-ietf-monday.png

Take aways from the first real day:

 1. I was generally impressed by the meeting rooms for air
    circulation.
 2. The RADEXTRA BOF was an exception as it was held in a small,
    fairly full room, and thus the \(CO_2\) levels kept rising. This
    was in the "meazine 1-4" room, and the ramping started primarily
    after someone shut the doors to dampen the noise from the
    hallway. I wonder what it would have been like if the doors
    remained open.
 3. I got tired of seeing my hotel room continuing to be high (even
    though the virus risk is low), so I decided to do something about
    it...

4.1. special news bulletin: turn on your room fan. always.

We interrupt this report for a special news bulletin...

One thing I've realized is that air circulation in my room is pretty
poor because my fan was set to "auto" (naively thinking that "auto"
means "do whatever you think is best for my health and comfort").
However, the \(CO_2\) levels creep up and up at this setting, though,
(likely the room temperature didn't require much heating/cooling so
the fan rarely turned on). Here's a graph of what happened in the
morning after I turned the fan from "auto" to the medium (2) setting.

room-fan.png

5. 2022-11-08 Tue: Day 2 - buses are a big problem

2022-11-08-ietf-tuesday.png

Notes from today:

 1. The meeting rooms I was in today relatively low \(CO_2\) levels
    all day.
 2. The social event required riding a bus there. Or walking. I
    should have walked. I was on the top of a double-decker bus next
    to a 1/4 opened roll-down window. I had my face generally pointed
    to the window the entire time trying to get the freshest air. My
    sensor, however, was sticking out of the top of my pants pocket.
    Hopefully the fresh air helped me avoid any contamination.
 3. The museum had a wonderfully fresh air blowing gently through its
    open front doors. That combined with the high, lofty ceilings
    meant that the \(CO_2\) rates were low until...
 4. My word that spike on the bus on the way home was depressing.
    (Note this was a more modern single-decker bus.) I can only hope
    that there was a good filter on the bus. But I bet not. I should
    have walked and risked getting wet in the London rain.

6. 2022-11-09 Wed: Day 3 - a generally cleaner air day

2022-11-09-ietf-wednesday.png

Notes from today:

 1. Today in general was fairly clean $CO[2]$-wise, except for the
    wgchairs lunch (and of course my hotel room, even with a fan on).
 2. I went to dinner and intended to take my meter to a restaurant,
    but failed to remember it in the rush to get there.

7. 2022-11-10 Thu: Day 4 - chairing two meetings

2022-11-09-ietf-wednesday.png

Notes from today:

 1. I chaired two meetings today, and not surprisingly the front of
    the room well away from most people had nice, clear air. The
    DANCE room was a smaller room too, so it's a touch higher but
    still generally low.
 2. The hotel room fan keeps thinking it's helping me by switching
    back to "auto". I wish it would stop trying to help me.

8. Conclusions

TBD after the week is over... though don't ride the buses is probably
an early and obvious one.

Author: Wes Hardaker

Created: 2022-11-11 Fri 02:19

Validate