| ... | ... | @@ -115,6 +115,31 @@ Char / Minute Taker / Focus Session [Rota](https://git.ligo.org/groups/gstlal/-/ |
|
|
|
* Focus session
|
|
|
|
- MDC retrospective (Becca) - [slides](https://docs.google.com/presentation/d/1mWF8YuL1uxePYr1WFtAyPnSzh2VlX5mTQBKpWzUTp-w/edit?usp=sharing)
|
|
|
|
- **Discussion:**
|
|
|
|
- Chad: Slide 9, it looks like the number of injections are changing, do you have any idea of the efficiencies ie ratio of found to total inj.
|
|
|
|
- Leo: It would be interesting to see which loud injections are missed, can we add that to these plots?
|
|
|
|
- Becca: can do both.
|
|
|
|
- Sarah: It looks like the missed injections turn on linearly at low mass ratios, is this an edge effect?
|
|
|
|
- Jolien: Looks like there's smaller mass ratios that you're doing injections in but then missing those. Does it have to do sth with the edge effect of the bank? this is a weird set of coordinates to be thinking in. That wedge must be NSBHs? At some fixed mass, you're missing things with small mass ratios compared to large mass ratios.
|
|
|
|
- Jolien: Slide 3, there's a line at 2 M_sun which must correspond to this wedge.
|
|
|
|
- Chad: maybe this is sth to do with max mass of NS.
|
|
|
|
- Jolien: are these injection masses or trigger masses on the plots?
|
|
|
|
- Becca: Slide 3 and 9 are both injected masses for both found and missed.
|
|
|
|
- Jolien: were sub-banks used to analyze these injections?
|
|
|
|
- Becca: No, this is the full MDC injection set and the results are coming from the Edward portion of the full bank.
|
|
|
|
- Surabhi: IMRPhenomPv2_NRtidal is used for all injections which could be causing discrepancies in the NSBH region, but not sure if that's contributing to this wedge.
|
|
|
|
- Jolien: Looks like it could be some book-keeping - somehow some triggers are not being reported. OR there's a hole int he bank?
|
|
|
|
- Chad: it's definitely all about m2. What's the minimum BH mass? What masses are allowed to be fully spinning. cAnnot make this a fully spin assumption, coz thus clearly looks like an m2 issue. Also, @becca do you somehow sum or classify these sources?
|
|
|
|
- Becca: 2 measurements - missed triggers and triggers, and they all include injected m1, and injected m2. Don't know immediately what's causing this but can look into it more.
|
|
|
|
- Jolien: Looks like it's in the BNS as well.
|
|
|
|
- Chad: this would be explained if there was somehow a cut-off on 2 M_sun; looks like a book-keeping issue.
|
|
|
|
- Jolien: there is an effect on m1 as well, which I assumed was from the template bank earlier, but could this somehow be related to detector mass, source frame mass?
|
|
|
|
- Chad: from Ron's point, maybe we are not sensitive to inj between 2-3 M_sun because of the high spin injections in this region.
|
|
|
|
- Jolien: what about the back hole spins in the NSBHs, that should get detected coz it will have a significant chi.
|
|
|
|
- Discuss this at the scheduled Friday call.
|
|
|
|
- Jolien: There is a weird feature about this: If you just look at the right edge, it is sloped, do we understand that? Is that a q-cut or sth like that?
|
|
|
|
- Becca: https://docs.google.com/presentation/d/1IEpIV7pVKbFgYkYzjiRAqbZXrvI7KBV5kFqBV44JTI0/edit#slide=id.g19fc6f08965_0_12 MDC7 update and slide 8 shows this is a feature for all pipelines.
|
|
|
|
- Cody: Maybe sth is going to wrong with the injection generation code unless all matches filtering pipelines are insensitive to this particular region in the parameter space.
|
|
|
|
- Jolien: This could be a 1+z effect. It is negligible for low mass but it might be pronounced for the high mass.
|
|
|
|
|
|
|
|
|
|
|
|
* AOB
|
| ... | ... | |
