DIS: charm pdf in F3charm at CC

felixhekhorn commented 4 years ago

We believe there is a bug in combination of CC coefficient functions with the respective PDFs
When computing F3charm with the conly pdf (which is defined by c(x)=x(1-x) and 0 otherwise ) APFEL is 0, while we're not (see table below)
looking at the PDG expression for the LO expressions (=parton-model) the charm pdf can contribute to F3 when scattering electrons or anti-neutrinos (what we are doing here) (see also Eq. 51 of CC_observables_def.pdf)
keep in mind that the APFEL definition of F3charm relies on the seperation of the CKM matrix

output:

------------------------------F3charm------------------------------

           x           Q2  APFEL    yadism  yadism_error  rel_err[%]
0   0.000317    90.000000    0.0  0.001465  6.438219e-12         inf
1   0.001007    90.000000    0.0  0.003751  9.234736e-12         inf
2   0.003195    90.000000    0.0  0.009510  4.041613e-11         inf
3   0.010138    90.000000    0.0  0.024013  1.321150e-10         inf
4   0.032169    90.000000    0.0  0.060758  4.250026e-10         inf
5   0.102077    90.000000    0.0  0.153750  1.063813e-09         inf
6   0.304545    90.000000    0.0  0.346403  1.647991e-09         inf
7   0.536364    90.000000    0.0  0.450932  2.779282e-09         inf
8   0.768182    90.000000    0.0  0.395607  6.832703e-10         inf
9   1.000000    90.000000    0.0  0.000000  0.000000e+00         NaN
10  0.001000     4.000000    0.0  0.008233  9.639907e-12         inf
11  0.001000     7.387505    0.0  0.005930  2.837073e-11         inf
12  0.001000    13.643806    0.0  0.004952  1.346871e-11         inf
13  0.001000    25.198421    0.0  0.004389  1.252950e-11         inf
14  0.001000    46.538363    0.0  0.004015  1.220218e-11         inf
15  0.001000    85.950595    0.0  0.003747  1.481200e-11         inf
16  0.001000   158.740105    0.0  0.003545  1.100350e-11         inf
17  0.001000   293.173318    0.0  0.003386  1.215494e-11         inf
18  0.001000   541.454816    0.0  0.003258  1.348017e-11         inf
19  0.001000  1000.000000    0.0  0.003152  1.028605e-11         inf
--------------------------------------------------------------------

APFEL setup:


 WARNING: FFNS is a FFN scheme
          ... setting NF = 3 FFNS PDF evolution
 WARNING: if there are external grids they cannot be locked
          ... unlocking subgrids

 Welcome to 
      _/_/_/    _/_/_/_/   _/_/_/_/   _/_/_/_/   _/
    _/    _/   _/    _/   _/         _/         _/
   _/_/_/_/   _/_/_/_/   _/_/_/     _/_/_/     _/
  _/    _/   _/         _/         _/         _/
 _/    _/   _/         _/         _/_/_/_/   _/_/_/_/
 _____v3.0.4 A PDF Evolution Library, arXiv:1310.1394
      Authors: V. Bertone, S. Carrazza, J. Rojo

 Report of the evolution parameters:

 QCD evolution
 Space-like evolution (PDFs)
 Unpolarized evolution
 Evolution scheme: FFNS with 3 active flavours at N1LO
 Solution of the DGLAP equation: 'exactalpha'
 Solution of the coupling equations: 'exact'
 Coupling reference value:
 - AlphaQCD( 91.2000 GeV) =  0.118000
 Pole heavy quark masses:
 - Mc =   2.0000 GeV
 - Mb =   4.0000 GeV
 - Mt = 173.0700 GeV
 The matching thresholds coincide with the physical masses
 muR / muF =  1.0000

 Allowed evolution range [   1.0000 :  10000.0000 ] GeV
 Fast evolution enabled

 Initialization of the evolution completed in   1.309 s

 Report of the electroweak parameters:

 Mass of the Z = 91.188 GeV
 Mass of the W = 80.398 GeV
 Mass of the proton = 0.9380 GeV
 sin^2(thetaW) = 0.2313
 GFermi = 1.16638E-05
       | 0.9743 0.2253 0.0035 |
 CKM = | 0.2252 0.9735 0.0410 |
       | 0.0086 0.0403 0.9992 |
 Z propagator correction = 0.00000

 Report of the DIS parameters:

 Computation in the FFNS3 mass scheme
 Charged Current (CC) process
 Scattering electron - proton   
 muR / Q =  0.5000
 muF / Q =  1.0000
 (No scale variation in the evolution)
 Target Mass corrections disabled
 Intrinsic charm disabled

juanrojochacon commented 4 years ago

Hi @felixhekhorn but according to the PDF, Eq. (18.18), F3 depends only on q-qbar right? And we never have a non-zero charm asymmetry, at least at the input evolution scale

juanrojochacon commented 4 years ago

In your model conly is charm the same as anticharm or different? I am confused here

felixhekhorn commented 4 years ago

according to the PDF, Eq. (18.18),

are we talking about the same file? we meant this one: https://github.com/scarrazza/apfel/blob/master/doc/pdfs/CC_observables_def.pdf and there seems to be no equation (18.18)

And we never have a non-zero charm asymmetry, at least at the input evolution scale

maybe in a true physical setup, yes, but here we think of the PDF as an external property (of which we know nothing) (and also c and cbar can diverge at higher orders) ... so here for benchmarking purpose to isolate one feature from another, we've chosen the (unphysical) conly with

In your model conly is charm the same as anticharm or different?

c=x(1-x), cbar=0=u=ubar=g=...

juanrojochacon commented 4 years ago

I meant the PDG ;)

It is true that for benchmarking purposes we can assume that c and cbar are different (although this will not happen in a realistic setup). In this case yes, APFEL should return a non-zero value, I agree. So this needs to be fixed, I guess that somewhere in APFEL one assumes that c_V=0?

vbertone commented 4 years ago

@felixhekhorn from the banner I see that you are using the NF = 3 FFNS that assumes that there is no charm in the initial state. Therefore it is perfectly consistent that you get zero.

felixhekhorn commented 4 years ago

@felixhekhorn from the banner I see that you are using the NF = 3 FFNS that assumes that there is no charm in the initial state. Therefore it is perfectly consistent that you get zero.

I guess you're right -> it is a bug in our implementation

scarrazza / apfel

DIS: charm pdf in F3charm at CC #26