1/Pb?
September 14, 2011 12:44 PM Subscribe
Why would data coming out of the LHC be reported as Pb^-1?
A friend of mine in grad school is at the LHC for the year. She sent me a message this morning telling me that yesterday the collider produced "over 100 pb^-1" of data. So okay, the Pb is petabits, which I get. But why the exponent of -1? She's confused on this too.
As of right now, when I hear about the data that is being produced over there, I just remove the -1 exponent. But it feels like there should be a reason behind it which I am completely missing. So, can someone provide an explanation? My google-fu is failing me completely on this.
A friend of mine in grad school is at the LHC for the year. She sent me a message this morning telling me that yesterday the collider produced "over 100 pb^-1" of data. So okay, the Pb is petabits, which I get. But why the exponent of -1? She's confused on this too.
As of right now, when I hear about the data that is being produced over there, I just remove the -1 exponent. But it feels like there should be a reason behind it which I am completely missing. So, can someone provide an explanation? My google-fu is failing me completely on this.
Response by poster: Sorry, to thread sit, but I'm trying to figure this out in one question- the information she's talking about is the number of collisions?
posted by Hactar at 12:53 PM on September 14, 2011
posted by Hactar at 12:53 PM on September 14, 2011
Best answer: Luminosity is the word you're looking for. It's a subtle concept, and it's tough to explain.
Physicists measure interactions by the size of their "cross-section". A large cross-section means that the interaction is likely to occur, and a small cross-section means it's unlikely. It's basically an analogy to the old billiard ball theory of particles: big particles are more likely to collide. The way the term is used today, though, has nothing to do with the physical size of the particle. It does, however, have units of area. A "barn" (b) is a tiny unit of area. A picobarn (pb) is even tinier.
So, to measure how much data you have, you need a number to multiply with the cross section to get the number of each type of interaction. Hence inverse picobarns. For example:
proton + proton -> Higgs + other stuff has a cross-section of 0.0053 pb (*WARNING TOTALLY MADE UP*)
The LHC has so far collected 3000 pb^-1 of data (*ALSO MADE UP*)
Therefore we would expect to have seen (0.0053 pb)*(3000 pb^-1)=~ 16 of these events so far.
An important thing to remember is that because of the negative exponent, smaller numbers means more data. Femto is smaller than pico, so inverse femtobarns is more data than inverse picobarns.
Knowing exactly how much luminosity your experiment has collected is tricky business, and it adds uncertainty to your experimental results.
Again, the concept of luminosity is subtle, and it's always a good question to make grad students answer during their defence. Tell your friend to study up!
posted by auto-correct at 1:13 PM on September 14, 2011 [4 favorites]
Physicists measure interactions by the size of their "cross-section". A large cross-section means that the interaction is likely to occur, and a small cross-section means it's unlikely. It's basically an analogy to the old billiard ball theory of particles: big particles are more likely to collide. The way the term is used today, though, has nothing to do with the physical size of the particle. It does, however, have units of area. A "barn" (b) is a tiny unit of area. A picobarn (pb) is even tinier.
So, to measure how much data you have, you need a number to multiply with the cross section to get the number of each type of interaction. Hence inverse picobarns. For example:
proton + proton -> Higgs + other stuff has a cross-section of 0.0053 pb (*WARNING TOTALLY MADE UP*)
The LHC has so far collected 3000 pb^-1 of data (*ALSO MADE UP*)
Therefore we would expect to have seen (0.0053 pb)*(3000 pb^-1)=~ 16 of these events so far.
An important thing to remember is that because of the negative exponent, smaller numbers means more data. Femto is smaller than pico, so inverse femtobarns is more data than inverse picobarns.
Knowing exactly how much luminosity your experiment has collected is tricky business, and it adds uncertainty to your experimental results.
Again, the concept of luminosity is subtle, and it's always a good question to make grad students answer during their defence. Tell your friend to study up!
posted by auto-correct at 1:13 PM on September 14, 2011 [4 favorites]
Also, you can tell how tricky a concept it is by how crappy my explanation was.
posted by auto-correct at 1:24 PM on September 14, 2011
posted by auto-correct at 1:24 PM on September 14, 2011
yup, inverse picobarns.
1 PB: 1 petabyte
1 Pb: 1 petabit
1 pb: 1 picobarn
1 pb-1: 1 inverse picobarn
posted by zamboni at 1:30 PM on September 14, 2011
1 PB: 1 petabyte
1 Pb: 1 petabit
1 pb: 1 picobarn
1 pb-1: 1 inverse picobarn
posted by zamboni at 1:30 PM on September 14, 2011
Just for scale, the LHC generates 15 Petabytes of data per year.
http://en.wikipedia.org/wiki/Petabyte
posted by jeffch at 6:17 PM on September 14, 2011
http://en.wikipedia.org/wiki/Petabyte
posted by jeffch at 6:17 PM on September 14, 2011
This thread is closed to new comments.
posted by Jacen Solo at 12:46 PM on September 14, 2011 [3 favorites]