Calculation of Wilson Coefficients in SPheno

Questions about the interface between SARAH and SPheno as well as the FlavorKit functionality
SoundSonic1
Posts: 11
Joined: 10. Jan 2018, 14:53

Calculation of Wilson Coefficients in SPheno

Postby SoundSonic1 » 10. Jan 2018, 15:16

Hello,

i tried to verify the coeffC8sm = 0.16443888E-10 (real part) at Q = 160 GeV taken from the SPheno output with the formula given in hep-ph/9805303v2 Eq. 10 on page 5,
but what i got was C8 = -0.0964357.
Then i normalized my C8 with n_cq^-1 = 8.87289*10^-9 as stated in arxiv:1405.1434 and got C8norm = -8.55663*10^-10.

Could you tell me what kind of normalization i should use to get the appropriate value?

FStaub
Site Admin
Posts: 719
Joined: 13. Apr 2016, 14:05

Re: Calculation of Wilson Coefficients in SPheno

Postby FStaub » 12. Jan 2018, 14:11

Hi,

the normalisation was chosen to be in agreement with the conventions of flav-io. I would need to check what the exact factor is, but I'm still on parental leave. I hope that checking the flav-io documentation is of some help for you.

Best,
Florian

Avelino
Posts: 21
Joined: 13. Apr 2016, 14:57
Contact:

Re: Calculation of Wilson Coefficients in SPheno

Postby Avelino » 12. Jan 2018, 14:41

If this is not solved earlier, I will have a look next week.

In the meantime, perhaps it would be useful to point out that the most recent version of SARAH/SPheno produces output for the Wilson coefficients in WCxf format (see 1712.05298 and https://wcxf.github.io/). This format was recently defined to avoid confusions regarding normalization (and other conventions) and to allow for an easier exchange of I/O files between different flavor codes). In this case the normalization of the operator you are interested about was unambiguously fixed when defining the FlavorKit basis (https://wcxf.github.io/assets/pdf/WET.FlavorKit.pdf).

SoundSonic1
Posts: 11
Joined: 10. Jan 2018, 14:53

Re: Calculation of Wilson Coefficients in SPheno

Postby SoundSonic1 » 12. Jan 2018, 15:32

Hi,

@Staub
it seems like flav-io and the paper i referred to use the same basis. Is it possible to output the Wilson Coefficients of flav-io?
On a different note: i was able to confirm coeffC7sm using the steps i mentioned in the first post.

@Avelino
The json files from SPheno/FlavorKit contain only the chirality flipped coeffC7p and coeffC8p as Q1R_32 and Q2R_32?
This was testet with SARAH-4.12.3 and SPheno-4.0.3 and the THDM-II.

Best regards,
SoundSonic1

Avelino
Posts: 21
Joined: 13. Apr 2016, 14:57
Contact:

Re: Calculation of Wilson Coefficients in SPheno

Postby Avelino » 12. Jan 2018, 15:43

SoundSonic1 wrote:Hi,

The json files from SPheno/FlavorKit contain only the chirality flipped coeffC7p and coeffC8p as Q1R_32 and Q2R_32?


You also have the operators obtained by exchanging the flavor indices: Q1R_23 and Q2R_23. With those you should be able to get the coefficients you are interested in.

SoundSonic1
Posts: 11
Joined: 10. Jan 2018, 14:53

Re: Calculation of Wilson Coefficients in SPheno

Postby SoundSonic1 » 12. Jan 2018, 15:54

Hi Avelino,

they are zero :/
You can have a look at my files:

https://pastebin.com/PTZk2ATv SPheno.spc.THDMII
https://pastebin.com/DFurKQeH WC.THDMII_1.json

SoundSonic1
Posts: 11
Joined: 10. Jan 2018, 14:53

Re: Calculation of Wilson Coefficients in SPheno

Postby SoundSonic1 » 25. Jan 2018, 12:55

Hi Florian,

i have looked into an older spectrum file taken from the flavio website:
https://raw.githubusercontent.com/flav-io/flavio/master/flavio/data/test/SPheno.spc.MSSM
In the old version, the coeff8sm is correctly calculated. You can see
coeff8sm = -0.86377623E-09 vs 0.16439750E-10
from the current version.

It would be nice if you could figure out where the different factor comes from.

Best regards,
SoundSonic1

Avelino
Posts: 21
Joined: 13. Apr 2016, 14:57
Contact:

Re: Calculation of Wilson Coefficients in SPheno

Postby Avelino » 25. Jan 2018, 14:27

SoundSonic1 wrote:Hi Avelino,

they are zero :/
You can have a look at my files


Hi again and sorry for the long delay in my answer. Perhaps I do not understand the issue, so please let me try to clarify it.

As I wrote in a previous message, it is perhaps convenient for you to focus on the json files in WCxf format, since their normalization has been checked recently to avoid any confusion. For instance, the standard WCs C8 and C8p can be derived from the WCs Q2R_23 and Q2R_32 in the FlavorKit basis, see their definition and exact normalization here: https://wcxf.github.io/assets/pdf/WET.FlavorKit.pdf. By looking at your example json file, I see that while Q2R_23 is zero, Q2R_32 is not. However, please note that Q2R_23 is zero only because by default FlavorKit does not compute this WC. If you want this WC to be computed you need to follow the instructions given in the FlavorKit manual and use the PreSARAH package. Basically, you must (1) modify the file PreSARAH-1.0.X/Input/GluonQQp.m by replacing the line

CombinationGenerations = {{3,2}};

by

CombinationGenerations = {{3,2},{2,3}};

so that both index combinations are computed, (2) run PreSARAH to generate the input for SARAH, and (3) copy the results of PreSARAH to the SARAH/FlavorKit folder. Then you are ready to produce your new SPheno code, which will now compute both WCs. As I said, this is much better explained in the FlavorKit manual.

With this, you should get non-zero results for both Q2R_23 and Q2R_32. Do they agree with your expectations?

SoundSonic1
Posts: 11
Joined: 10. Jan 2018, 14:53

Re: Calculation of Wilson Coefficients in SPheno

Postby SoundSonic1 » 25. Jan 2018, 14:52

Hi Avelino,

ok i didn't know they are not turned on in WCxf by default. The coeffC8 value is calculated in the spectrum file which should correspond to Q2R_23, so i don't see a reason why it shouldn't be exported by default.

Anyway, i want to know why the calculation of coeff8 has changed in SPheno.

Best regards,
SoundSonic1

SoundSonic1
Posts: 11
Joined: 10. Jan 2018, 14:53

Re: Calculation of Wilson Coefficients in SPheno

Postby SoundSonic1 » 20. Feb 2018, 11:41

Hey Florian,

can you please check why the current SPheno version calculates a different coeffC8sm opposed to this one
https://raw.githubusercontent.com/flav-io/flavio/master/flavio/data/test/SPheno.spc.MSSM ?

Btw this switch to SuperCKM basis does not work for the MSSM. There are some divisions by zero after transforming between bases.


Return to “SPheno, FlavorKit”

Who is online

Users browsing this forum: No registered users and 1 guest