makent solution: inclusive DIS yields

Goal

Plot the relative yields of inclusive and semi-inclusive DIS events. The semi-inclusive (SIDIS) events are those where a hadron was detected in coincidence with the scattered beam lepton. In practice, a normal spectrometer can only detect 6 hadron types directly: pi+, pi-, K+, K-, proton, and anti-proton. These are the only hadrons that make tracks (they're charged, and they live a very long time). We'll consider only these 6 guys in our SIDIS plots.

We dealt with absolute normalization in Example 1, so we'll skip it here and just bother with relative yields. The challenge is to produce these 3 plots:

Plot 1 Overlay plot vs the variable-of-your-choice (x, Q2, or W2) showing the relative yields of (a) inclusive and (b,c,d,e,f,g) the six semi-inclusive hadron types.
Plot 2 Two simple plots: pi+ yield vs z and vs transverse momentum p_T (where 'transverse' means 'transverse to the q vector', as usual)
Plot 3 Just semi-inclusive pi+ yields again, this time vs x and showing the quark purities: how often the pi+ was produced from the 6 different flavours of struck quark
Plot 4 Make a nice plot of all 6 parton distribution functions q(x) (at the average Q² of your simulation)

We'll apply these standard DIS cuts to all events:

Q² > 1
W² > 4
y < 0.85

We'll impose the standard HERMES 'box acceptance' cuts on all tracks:

p > 1.5
|theta_x| < 170 mrad
40 mrad < |theta_y| < 140 mrad

And we won't consider any event if the scattered lepton was not accepted. That includes a higher minimum-momentum cut because of our DIS trigger:

p of scattered lepton > 3.5

As in example 1, you can leave radiative corrections turned off for simplicity.

Generate events with gmc_disNG

Guess what? The GMC kumac you need is precisely the same as that from Example 1 -- just DIS events with an accepted scattered lepton. That means the events file dis_accep_1.devents.gz that you generated before can be used for this example too.

Run makent to produce an ntuple

The ntuple-descriptor file for this set of plots is trickier than in the inclusive case. You now need information about the hadrons within your DIS events, so you'll need makent to scroll through the mcTrack table. That means a track-level ntuple ... give it a try ...

Download: semi.in

(The DAD parser rocks, eh!)

Once you have your ntuple-descriptor, Uun makent:

       makent -in semi -out semi_1 dis_accep_1.devents.gz

That'll produce the ntuple file semi_1.rz. You'll notice makent taking a lot longer to run than in Example 1, as hexe and dad are doing a lot more work. Finally, we'll ignore the normalization file dis_accep_1.norm.kumac this time since we're only interested in relative yields.

Plots the results in PAW

Start PAW and read in your ntuple:

       h/file 1 semi_1.rz

With the ntuple in hand, making the 4 requested plots is pretty straightforward:

Download: semi-plot.kumac

Uh, for this script to work, you also need my little labelling utility:

Download: label.kumac

Just drop it into your working directory. (I can't stand doing PAW labels 'by hand').