AK4 CHS jets pT cut for Run3

[clacaputo]

This issues is intended as a reminder for @cms-sw/reconstruction-l2 , @cms-sw/xpog-l2 and @cms-sw/jetmet-pog-l2, about the discussion we had in #36890.

Context
Switching from AK4 CHS to AK4 PUPPI jet for Run3 as recommended jet collection.
The PR #36890 is decreasing the AK4 PUPPI jet pT cut from 15 to 10. This effect the AOD event size. Three possible scenarios:

1. keeping AK4 CHS jets pT cut at 10, leads to an MiniAOD event size increase of 0.37%
2. increasing AK4 CHS jets pT cut at 15, leads to an MiniAOD event size decrease of 1.0%
3. keeping AK4 CHS jets pT cut at 10, but slimming down the collection, not yet investigated.

Scenario 1 is favoured by xpog and jmet so we could recalibration of the type1MET with both collections, without invalidating pfMET (#36890 (comment)).

This is a solution just until the end of 2022. To be revised with either scenario 2 or 3

NOTA: potential decrease in MiniAOD event size ~1%

[clacaputo]

assign reconstruction

[cmsbuild]

New categories assigned: reconstruction

@slava77,@jpata,@clacaputo you have been requested to review this Pull request/Issue and eventually sign? Thanks

[cmsbuild]

A new Issue was created by @clacaputo Claudio Caputo.

@Dr15Jones, @perrotta, @dpiparo, @makortel, @smuzaffar, @qliphy can you please review it and eventually sign/assign? Thanks.

cms-bot commands are listed here

[clacaputo]

assign xpog

[cmsbuild]

New categories assigned: xpog

@mariadalfonso,@gouskos,@fgolf you have been requested to review this Pull request/Issue and eventually sign? Thanks

[jpata]

type jetmet

[vlimant]

@kirschen : is the need for keeping CHS AK4 Jets at >10 still justified ? am I getting this correctly that such CHS Jets are needed to allow one to use pfMET (by allowing for type1 jet correction to be propagated to it) ?

[kirschen]

Hi @vlimant, good to bring this up. Yes, the "solution 1" was chosen for maximum flexibility during startup, and being able to do type-1 jet corrections "with CHS jets" is one aspect. I will bring it up for discussion within JME again.

[kirschen]

Hi @vlimant, good to bring this up. Yes, the "solution 1" was chosen for maximum flexibility during startup, and being able to do type-1 jet corrections "with CHS jets" is one aspect. I will bring it up for discussion within JME again.

Pinging @alkaloge @miquork @cms-sw/jetmet-pog-l2 , cf. coordination chat post in December to wrap this up.

[vlimant]

can we close this, eventually ?

[alkaloge]

@nurfikri89 @abenecke @gagarwal @anmalara @dsavoiu

[nurfikri89]

@vlimant This is on my to do list. I am looking into doing option (2) and (3) at the same time.

[nurfikri89]

I have made more studies with a TTto4Q Run3Summer23DRPremix inputs. Here are the event size reduction for the various options.

1. AK4 CHS jet pt > 15 GeV: -1.5%
2. AK4 CHS jet pt > 15 GeV, remove PileupID & QGL: -1.6%
3. AK4 CHS jetpt > 15 GeV, remove PileupID, QGL & DeepCSV: -1.8%
4. AK4 CHS jet pt > 15 GeV, remove PileupID, QGL, DeepCSV & DeepJet: -2.0%.

If the JetMET convenors can agree on an option, I can make a PR for this ASAP.

cc JetMET convenors (@abenecke, @alkaloge, @gagarwal, @dsavoiu, @fiemmi, @rkogler, @wajidalikhan)

[vlimant]

can you please prepare a summary of this topic for next XPOG meeting https://indico.cern.ch/event/1357023/ so that we move on

[nurfikri89]

@vlimant yeah I can give the talk.

[nurfikri89]

This is now resolved and merged in 14_1_X and 14_0_X.

[vlimant]

please close