Source: The New York Times.
In my just-prior post, I asked why the sidewalks reek of marijuana smoke in some parts of York City. Vastly more than, say, the smell of cigarette smoke. This, despite almost never actually seeing anyone smoking dope, when I was in Manhattan last week.
It was a conundrum. How could the sidewalk air be saturated with the odor of marijuana, but actual, observable marijuana smokers were few and far between?
I think the answer to that riddle is simple: Burning marijuana really stinks. That is, the odor of burning marijuana is potent. It takes a large volume of fresh air to dilute it down to the point where you can no longer smell it. Turning that on its head, a lungful of marijuana smoke can stink up a far larger volume of air than the equivalent amount of cigarette smoke.
Based on a single measurement, casually reported in a single article, using “odor units”, the odor of marijuana smoke was maybe 300 times as potent as the odor of cigarette smoke.
But is there more to the story
After thinking that over for a bit, there are two more things I’d like to quantify, if possible.
First, just how much fresh air is needed, to dilute a lungful of marijuana smoke to down to one “odor unit”, that is, to the point where you can just barely smell that something is in the air.
Second, in a cityscape dominated by skyscrapers and narrow streets, how much fresh air typically mixes with the street-level air? I’m guessing that part of the puzzle is that whatever odors are emitted at street level stay trapped there, in a mid-town Manhattan cityscape. But I’d like some confirmation of that.
If I can find the answers, and they turn out to be “a lot” and “not much”, then I think that’s a fairly complete explanation of why mid-town Manhattan reeks of dope. That’s not (necessarily) the product of huge numbers of daily dope smokers. That could be the result of a few individuals, whose potent emissions remain trapped at street level.
How much air can a single dope smoker taint?
One “odor unit” is defined as the concentration of an odor that is barely noticeable. It’s the detection threshold for a substance. Typically, this might be measured by using a panel of randomly chosen individuals, exposing them to a substance in some known concentration, and asking if they smell something.
The concentration or number of odor units, for a given substance, present in a given parcel of air, is measured by how much you’d have to dilute that with fresh air, before you could no longer smell anything.
For example, if you take some particular parcel of air that stinks, and you find that if you dilute it 20 to 1 with fresh air, the odor finally goes away, then the original parcel of air was said to have a concentration of 20 odor units (typically expressed per cubic meter) of that smelly substance.
The higher the reading in odor units, the more fresh air you need to dilute the smell to the point where you can’t notice it. Turning that on its head, the higher the reading in odor units, the greater the volume of air that a given bit of tainted air can stink up.
My single Canadian measurement found that dope smoke, 30 meters downwind of the source, measured 5000 odor units/cubic meter, (reference).
... Thirty metres upwind, a man and woman lean on tree trunks, passing a cigarette that is not tobacco. Medina recalibrates his nose by breathing clean oxygen through his mask. The apparatus mixes the oxygen with a minuscule amount of outside air, in increasing intervals. Very little is needed before Medina detects terpenes, the odourous molecules in cannabis. Startled by what appears on his tablet, he redoes the test, only to confirm that, at 5:38 pm, the ambient smell measures 5,000 odour units per cubic meter, meaning it’s 5,000 times stronger than required for an average human nose to detect.
In other words, for every cubic meter of air in that smoke plume, at 30 yards (meters) downwind from the source, you’d have to mix that with 5000 cubic meters of fresh air, before you could no longer smell marijuana smoke.
Typical side-streets in mid-town Manhattan are about 50′ wide, from building face to building face. (Verified with Google Earth, for 47th street). Doing the math, 5000 cubic yards is about 250 feet of the 50′ wide street, filled to a depth of 10 feet.
In other words, a single cubic meter of air tainted with dope smoke, 30 yards downwind of the smoker, is enough to stink up 250 feet of a typical Manhattan side street. That’s assuming the stink remains trapped within 10 feet of the ground. And that’s assuming the air is well-mixed within that 10-foot ground layer. But that only accounts for a small portion of the smoke plume emitted by the marijuana smoker.
For reference, 47th street in Manhattan, between 8th and 9th avenues, is only about 850 feet long.
When I put all that together, I vaguely come to the conclusion that a single lungful of marijuana smoke, emitted at ground level, is probably enough to generate a detectable street-level odor of marijuana in at least an entire city block in Manhattan. That’s assuming that ground-level air mostly does not mix with the air above it, and that the ground-level air itself is reasonably well-mixed.
I would love to pin this down a bit more tightly, but I’m drawing a blank on other measurements of odor units for marijuana smoke. That said, I think that one Canadian measurement is enough to show that the intensity of the smell of marijuana smoke is key. Note, in the quote above, the professional doing the odor measurement more-or-less did a double-take at how strong the odor was, at 30 yards distance, in odor units.
How much does street-level air mix in urban-canyon landscapes.
The final piece of the puzzle is to pin down the extent to which street-level air mixes with the air above it, in a cityscape with narrow streets and tall buildings. A lot of mixing will dissipate the odor of marijuana. Little mixing will keep that odor trapped at street level.
First, it’s well-established that automobile emissions levels are much higher at street level than they are just a few floors above. You can find any number of reference that will tell you that. But that could be due, in large part, to the continuous emission of those pollutants by the cars at ground level. So that’s not the same as saying that there’s little-to-no mixing of street-level air with the air above the street. It might be a consequence of that, and it might not be.
Weirdly enough, there actually was a study of Manhattan street-level air flows, conducted using rare trace gasses, circa 2005. This was the Urban Dispersion Program (UDP) NYC MSG05 Experiment. That appears, however, conducted in the context of modern U.S. civil defense, and if any clear results from that study have ever been published, I was unable to find them via Google search.
And in hindsight, that’s probably a good thing.
After spending half an hour reading what I could find of the scholarly literature on air flow in cities, I think the answer is “it depends”. There seems to be some general agreement that in most conditions, air tends to get trapped against the leeward sides of tall buildings. There’s also a general consensus that the air at the bottom of an urban canyon remains dirtier than it would be in the absence of all those tall buildings (duh).
In any case, I never came across a study that literal answered my key question. Good enough to say that there is no one, good, quantitative, general-purpose answer on the extent to which street-level air mixes with air at higher altitudes, in an urban-canyon landscape. Which I read as “it depends” on a lot of stuff.
So that part — just how thick is the layer of air into which street-level marijuana smoke may disperse — remains an unknown. In some circumstances, street-level air is trapped at street level (leeward sides of building with moderate winds). In others, the wind blows the air out of the street canyon entirely. Likely, it depends on the wind speed and direction, air temperature, and the actual configuration of the surrounding buildings.
Conclusion
It seems pretty clear at this point that if the weather conditions are right, a low density of on-street marijuana smokers may fill Manhattan streets with the odor of dope smoke. This is due to the pungency of marijuana smoke, compared to other common smells such as tobacco smoke. Dope smoke has to be diluted many-thousand-fold to reach the point where the odor is no longer noticeable by the average person.
As to whether or not the particular configuration of streets and buildings in mid-town Manhattan contributes to this effect, I could not say. The issue of air flow in urban-canyon environments was too complex for any simple generalization.
The only practical point is that even though the area reeked of dope smoke, it’s not correct to assume that there are large numbers of dope smokers on the streets of Manhattan at any given time of the day. That’s not necessary to generate the pervasive odor of marijuana found there. It may only take a few, due to the potency of the smell of burning marijuana.
Addendum: Trees maybe?
The only bit of the academic literature that stuck to me is that trees do something. They are a “porous barrier”, and in some circumstances increases mixing of air along the street.
I don’t remember there being a lot of trees in mid-town Manhattan. Maybe street-level air might mix less with above-street air in treeless concrete canyons compared to other cityscapes.
