New study of sense of smell: 1 triilion?!

There is an oft quoted study that found humans can distinguish 10,000 scents. This article is about a new study that claims the number is actually one trillion! It appeared on National Public Radio's medical page.
Now, the researchers are not saying we can distinguish one trillion individually named odors. Rather, that we can detect small differences in perfume-like concoctions (my term). The researchers then extrapolated that information mathematically and reached the figure of one trillion.
What's interesting is that the results led the researchers to a separate conclusion and it's something perfumers/noses have known all along: that it's possible to create a particular scent from an accord of notes which are seemingly unrelated to that scent.

One other curious thing, from the original article in Science, is how different individual participants were in their performances.
From the initial group of 28 people, two were excluded because they failed to correctly identify the wrong vial in more than three out of four trials. From the remaining 26 (17 females, 9 males), the worst sniffer was calculated to be able to distinguish between some 78 million odorous mixtures. The best, a whopping 1.03 × 10^28 (ten octillion). Or, written out as decimal numbers:

Worst: 78,400,000
Best: 10,300,000,000,000,000,000,000,000,000

Makes me wonder if the ability to tell scents apart is why some people are more likely to become perfumistas...

Thanks for posting the info from Science, Epimedes. It seems much more complete than the NPR article. Honestly, I'm surprised that subjects were able to detect differences between such complex blends. Did the Science article mention if subjects could identify the specific odors that made the test fragrances different or was it that they could simply tell there was a difference?

Their test was set up in a way that the subjects would only have to tell which vial out of three was different. Their vials contained blends of 10, 20 or 30 odorous molecules. Practically no one could tell apart 20- or 30-component mixtures that differed only by a single component. Some blends were trickier than others; the participants could only tell apart one half of their 20-component blends that were 50% similar. Not always easy it seems! There are many other results that are quite difficult to digest and the math they used to come up with these huge numbers is simply beyond my comprehension. But I kind of agree with their conclusion that humans can distinguish more olfactory combinations than colors or tones.

One other interesting thing they mention is that the scent of a rose is produced by 275 components, much more than what their subjects were tested with. Must be very difficult to reproduce this in a perfume, if possible at all.

Another article that appeared in Popular Science concerns "the smell of space" (link below).
www.popsci.com.au/technology/space/what-does-s pace-smell-like
Apparently, the carbon-rich, oxygen-poor atmosphere of space smells a lot like a Nascar race or barbequed meat. The moon's atmosphere smells like gunpowder. Astronauts noticed the odours which coming from molecules adhering to their suits after returning from space walks.
There's a perfume idea here for sure!

Triffid, I was trying to think of a clever perfume name based on this article when I realized there was at least one "carbon based" fragrance. However, I didn't notice notes like burned tires, barbecued meat or gunpowder...

"6C Carbon" by nu_be
"Sweet & Sour Carbon" by Linn Young

nu-be seem to have the elements covered!
When I was reading the description of the way space smells, I immediately thought of Tauer's Lonestar Memories with it's petrochemical, burnt exhaust, mechanic's workshop smell.