[Rgc_analysis] [Rgc] RGC-analysis meeting tomorrow

[Sebastian Kuhn]

Hi Patrick,

thanks for your comments. Here are some replies - I am still in the process of updating the PAC presentation.

On Jul 2, 2024, at 9:14 PM, Patrick Achenbach <patricka at jlab.org> wrote:

Good evening Sebastian,

I don't think that it is particularly useful to emphasize the fact that the individual experiments add up to "a total of 916 PAC days worth of new data". This notion of one experiment = one data set is common at very small facilities and is historically embedded in the JLab user community. However, I don't think that there was ever a chance to get ~ 1000 PAC days awarded for CLAS12 running with the polarized target. What do you mean with "Including 120 PAC days on NH3 just for pDVCS"? To me, these bullet points are misleading. In this jeopardy process you are defending the remaining 40 PAC days to finish a program of approved 120 PAC days.

I agree that my wording may have been a bit off. Here is what the slide says now:
•8 experiments with a total of “916 PAC days worth” of proposed new data
•Due to simultaneous data collection for all channels with CLAS12, original request was 185 PAC days
•Including 120 PAC days on NH3 originally requested for pDVCS (E12-06-119(b))
•Plus 60 more PAC days on ND3 originally requested for nDVCS (E12-06-109A)

It is true that “high energy experiments” are used to having many channels measured at once, but at JLab, this is still not necessarily the norm (in Halls A and C). I put this information here for 2 reasons:
1) To drive home the point that all of the great Physics from RG-C will come in, regardless of what (some) members of the PAC may think is “unimportant"
2) In particular, the original program for DVCS required 120 days on p and 60 days on d, and hence, when PAC48 cut the total from 185 to 120 “with emphasis on DVCS”, they just showed their misunderstanding of the situation. So I want to be very clear this time around.
Of course, I am not trying to re-litigate PAC48, but I think it should be absolutely clear that we already lost 1/3 of THEIR preferred Physics at PAC48, and they should not cut another 1/3 from that.

I am happy to show the conditional run schedule in my presentation on Monday. I am not sure if the PAC really appreciates a discussion on the beam-time scheduling by the proponents.

I am not trying to discuss beam-time scheduling - I am just rying to show that by combining RG-C and RG-G, the overhead is minimized and that the price to pay for 50% more PAC days is relatively small. I gave this slide a new header: “Possibility to Minimize Overhead - Combine w/ RG-G”. Of course I’m open to other suggestions.

On page 20, you specify "• 40 PAC days on NH3 and ND3". How is this distributed? The reader is also asking for clarifying the ND3-NH3 split. I don't fully understand the following bullet point "•If run in conjunction with RG-G: Common auxiliary targets and systematic measurements shared (C, CH2, CD2, 4He, Empty)". Do you mean that the auxiliary target measurements are only important when running in conjunction with RG-G? How many PAC days are needed on this for RG-C? Personally, I think that the need for systematic measurements to complete the program is more convincing that the need for < 20 PAC days for NH3 and < 20 PAC days for ND3.

I tried to make it a bit clearer - now it says the following:
•40 PAC days split as previously between NH3 and ND3 (+ 15% auxiliary targets)
–In addition, higher average target polarization (especially for ND3) and a modest increase in beam current will yield about 70%-100% (FoM) more data
–Improved systematics and extended kinematic reach by running BOTH solenoid and torus polarities & optimal complement of background measurements (interrupted previously by extended downtime), plus potentially higher e- energy
If run in conjunction with RG-G: Recommissioning, common auxiliary targets and systematic measurements shared => optimal use of time

(The split was roughly 50-50 ND3/NH3, with a bit more of the former). We need the auxiliary targets either way - just for RG-C or just for RG-G - what I was trying to say is that if we can run both in succession, we can share most of them between both RGs and hence they would use up less time from either.
I agree that systematics are one strong reason for more data, but I disagree that more statistics is not significant. We added more slides that show this - see also below.

On page 21, the highlighted necessity of multi-dimensional binning has - in my view - not been substantiated in this presentation. I get the point on "improved systematics".

The last bullet is now 2 bullets that read as follows:
•More statistics crucial for multi-dimensional binning, statistics-starved channels (DVCS, TCS, K-SIDIS), and all Deuteron asymmetries.
•Completing the set of originally planned run configurations and auxiliary measurements will lead to improved systematics.

We added some slides (from Samy) specifically addressing the “multidimensional binning” argument (see elsewhere in this email chain).

Best,
Patrick

Greetings, and have a nice rest of the weekend. - Sebastian

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/rgc_analysis/attachments/20240706/96c6b4ee/attachment-0001.html>