WEBVTT

1
00:01:21.930 --> 00:01:23.539
David Schlegel: Hey, good morning.

2
00:01:25.850 --> 00:01:27.440
Lynne Jones: That's on mute. Marty.

3
00:02:01.150 --> 00:02:02.109
Eric Gawiser (he/him): Hello, Lynn.

4
00:02:02.400 --> 00:02:04.140
Lynne Jones: Hey, good morning!

5
00:02:06.000 --> 00:02:07.880
Lynne Jones: Okay, there is mine.

6
00:02:09.530 --> 00:02:16.750
Eric Gawiser (he/him): We didn't get any kind of official notification, but I do have a desk Summary ready.

7
00:02:16.750 --> 00:02:24.350
Lynne Jones: Oh, perfect. When you say you didn't get an official notification, you mean, like… I mean, like.

8
00:02:24.350 --> 00:02:28.659
Eric Gawiser (he/him): It wasn't clear who was supposed to do it, and this kind of thing, it's all informal.

9
00:02:29.590 --> 00:02:30.759
Lynne Jones: Yeah, I am…

10
00:02:30.760 --> 00:02:31.630
Eric Gawiser (he/him): last minute.

11
00:02:31.820 --> 00:02:36.610
Lynne Jones: Yeah, I'm sorry about that. I think that, between…

12
00:02:37.700 --> 00:02:42.090
Lynne Jones: Oh, a variety of things. We'll have to see how it goes today, yeah.

13
00:02:48.370 --> 00:02:50.910
Lynne Jones: I'm glad to hear you.

14
00:02:51.440 --> 00:02:56.280
Lynne Jones: You have something to… To talk about.

15
00:02:56.870 --> 00:02:58.560
Lynne Jones: Hey, good morning, Fred.

16
00:03:07.320 --> 00:03:09.329
Lynne Jones: So, I know,

17
00:03:13.260 --> 00:03:19.390
Lynne Jones: We'll wait a few more minutes here for… For people to come in.

18
00:03:50.170 --> 00:03:53.160
Lynne Jones: We…

19
00:03:53.580 --> 00:04:07.630
Lynne Jones: We don't have, a lot today. Like, the agenda is primarily, well, entirely, to, give a chance for people to, to talk

20
00:04:08.000 --> 00:04:12.030
Lynne Jones: About their thoughts, and have another set of discussions.

21
00:04:12.140 --> 00:04:15.450
Lynne Jones: Because the whole workshop was… was…

22
00:04:16.790 --> 00:04:24.549
Lynne Jones: put together relatively quickly, and we only just came out with these simulations for 5.3. I'm not entirely certain,

23
00:04:25.080 --> 00:04:37.979
Lynne Jones: how much feedback we'll get, and I didn't really get a lot of feedback from the Science Collaboration chairs about what… whether the times worked for them, and so on. So, we'll see how it goes today.

24
00:04:38.130 --> 00:04:43.060
Lynne Jones: But thank you all for joining us for Day 3 of the SEOC Workshop No. 6.

25
00:04:43.730 --> 00:04:51.480
Lynne Jones: Do I get a list of how many people are here yet?

26
00:04:55.210 --> 00:04:57.380
Lynne Jones: Yeah, okay, so we have about…

27
00:04:57.990 --> 00:05:00.569
Lynne Jones: We still have some people coming in here.

28
00:05:09.230 --> 00:05:13.829
Lynne Jones: But I imagine, we probably had a few more people last week, so…

29
00:05:34.790 --> 00:05:35.640
Lynne Jones: Very cool.

30
00:05:42.900 --> 00:05:48.829
Lynne Jones: Go ahead, hear it, and… Start up in 2 more minutes at 5 after.

31
00:06:59.840 --> 00:07:00.980
Lynne Jones: Alrighty then.

32
00:07:02.690 --> 00:07:17.430
Lynne Jones: I think we picked up two more people while we made it. Let's see, anyway, so, good morning, everybody, bright and early for me. Maybe not so much for the rest of you, but today is Day 3 of the SEOC workshop.

33
00:07:17.780 --> 00:07:24.080
Lynne Jones: Again, this is a workshop,

34
00:07:25.310 --> 00:07:32.440
Lynne Jones: from the SCOC, the Survey Cadence Optimization Committee, and today, the big focus is to get feedback.

35
00:07:32.560 --> 00:07:36.230
Lynne Jones: From, the community and the science collaborations.

36
00:07:37.960 --> 00:07:56.909
Lynne Jones: As a reminder, as always for the Code of Conduct, let's just make sure we be kind. If you do have any issues, please report to either Federico Bianco or myself, or you can reach out to your Science Collaboration's Code of Conduct contact person.

37
00:07:57.180 --> 00:07:59.890
Lynne Jones: Yep.

38
00:08:00.780 --> 00:08:08.610
Lynne Jones: as a quick reminder, on Day 1, We, this is…

39
00:08:09.310 --> 00:08:17.340
Lynne Jones: I will post the link to these slides in the Slack channel, but we do have links to the slides from day one.

40
00:08:18.410 --> 00:08:25.610
Lynne Jones: We went through the original… that was just the overview of the 5.3 simulations,

41
00:08:26.340 --> 00:08:35.660
Lynne Jones: And we talked about the SEOC assessments by science areas, so the SEOC review of the… of some high-level metrics.

42
00:08:36.669 --> 00:08:42.510
Lynne Jones: And day two, we did a lot more, sort of, detailed evaluation of different aspects.

43
00:08:43.049 --> 00:08:44.750
Lynne Jones: These slides are here.

44
00:08:45.920 --> 00:08:47.130
Lynne Jones: And…

45
00:08:48.320 --> 00:09:04.520
Lynne Jones: You know, we can… we can talk about any of those things. Today, the plan is to give the science collaborations time to… to talk about what they found, or looked at, or saw in the 5.3 simulations.

46
00:09:04.630 --> 00:09:23.130
Lynne Jones: I did not get a lot of feedback from people about how much time they needed, or if these time slots worked, so I suspect there's a good chance that we'll be a little under on the time we need, and we'll have lots of time for discussion. So, we shall see. Now, I see Eric is here.

47
00:09:23.370 --> 00:09:28.340
Lynne Jones: And had some, some words to pass from the desk.

48
00:09:28.640 --> 00:09:30.800
Lynne Jones: group, right?

49
00:09:33.880 --> 00:09:38.260
Lynne Jones: Is it… do we also have AGN and Galaxies here for this morning?

50
00:09:42.210 --> 00:09:45.270
Roberto Assef: I'm here, but I don't really have much to ask.

51
00:09:45.270 --> 00:09:51.170
Lynne Jones: Yeah, and, yes, I imagine, and galaxies…

52
00:09:52.070 --> 00:09:58.900
Lynne Jones: maybe… we did have the SEOC overview from the Galaxy's point of view already, so…

53
00:09:59.000 --> 00:10:00.409
Lynne Jones: We may or may not have there.

54
00:10:00.820 --> 00:10:04.360
Lynne Jones: Okay, in which case, Eric, do you want to start?

55
00:10:06.140 --> 00:10:10.190
Lynne Jones: And I can just hand over sharing to you if you have slides.

56
00:10:10.190 --> 00:10:12.930
Eric Gawiser (he/him): Yeah, sure, I'll… I'll share here.

57
00:10:25.530 --> 00:10:27.629
Lynne Jones: And thanks for the link in the…

58
00:10:28.710 --> 00:10:39.349
Eric Gawiser (he/him): Yeah, if there's a central repository, we can add them in later, but hopefully this way, people can follow along in real time if they want in the slides. I hope you're seeing the slides in a slideshow mode now.

59
00:10:39.700 --> 00:10:49.330
Eric Gawiser (he/him): Great. So, this is just a few slides summarizing what we've been able to glean so far in the Dark Energy Science Collaboration from V5.3.

60
00:10:49.830 --> 00:10:56.560
Eric Gawiser (he/him): As a reminder, the main metrics we pay attention to, although we look at a lot more than this, but these are the most important ones for us.

61
00:10:56.670 --> 00:11:02.399
Eric Gawiser (he/him): Our, Dark Energy Task Force figure of merit that combines weak lensing and large-scale structure.

62
00:11:02.970 --> 00:11:12.730
Eric Gawiser (he/him): A weak lensing average GRI visits, that really is just how many total visits you get, and that's because the more visits there are, the more you beat down some of the systematics for weak lensing.

63
00:11:12.920 --> 00:11:20.170
Eric Gawiser (he/him): The third one is the number of Type Ia supernovae below a limiting redshift, and this relates to what…

64
00:11:20.180 --> 00:11:35.570
Eric Gawiser (he/him): Philippe Grease works on. The fourth one looks similar, but it's really about strongly lensed supernovae, so that's into the strong lensing science. And the fifth one is about kilonova population counts over 10 years. And there's links here to the GitHub and a paper describing these.

65
00:11:38.040 --> 00:11:43.530
Eric Gawiser (he/him): A reminder, as I set out to give this quick summary, that not all metrics are included right now.

66
00:11:43.560 --> 00:12:01.169
Eric Gawiser (he/him): For some simulations, and unfortunately it's all the ones that Lynn recommended we take a look at, the strong lensing metric is a NAN. We don't understand that yet, but I'm sure we'll be able to solve that and get real results on that in the future. And then, as some of you are aware, Philippe Gries has a custom supernova metric that he has to run.

67
00:12:01.170 --> 00:12:10.999
Eric Gawiser (he/him): The math team has very helpfully coded an approximation for that, and so we're using that at the moment, pending the chance for Philippe to more fully run the full version of the metric.

68
00:12:12.310 --> 00:12:14.450
Eric Gawiser (he/him): So I guess this was the caveat slide.

69
00:12:15.360 --> 00:12:21.169
Eric Gawiser (he/him): Alright, so we've got some really nice analysis that Michelle Lochner put together that I'm going to present on her behalf.

70
00:12:21.430 --> 00:12:28.399
Eric Gawiser (he/him): The first plot here is evolution of baseline simulations, going all the way back to V2.

71
00:12:28.930 --> 00:12:45.660
Eric Gawiser (he/him): through V3, V4, V5, and now, most relevantly, V5.3, with a couple of versions of that, one that goes to the 11-year survey, and then a couple of comparisons. And so, of course, because everything here is referenced to

72
00:12:45.660 --> 00:12:58.570
Eric Gawiser (he/him): Baseline V5.3, all four metrics for that one sit at zero, they have no difference from it. The quantity being plotted is metric minus baseline divided by baseline, so we're seeing percentage changes for other simulations.

73
00:12:58.640 --> 00:13:07.060
Eric Gawiser (he/him): Most things were better in the recent past. This is no surprise. We're being realistic here with the as-built observatory and what we can do.

74
00:13:07.150 --> 00:13:25.580
Eric Gawiser (he/him): Interestingly, the static science weakensing plus LSS figure of merit, has much smaller fluctuations across these changes than the time domain metrics. And for us, the bottom line here is that the 10% reduction in on-sky visits hits the time domain metrics by basically 10-20%,

75
00:13:25.950 --> 00:13:28.310
Eric Gawiser (he/him): In the direction we would not like to have.

76
00:13:29.220 --> 00:13:39.480
Eric Gawiser (he/him): It's a pretty robust effect, and I don't think a big surprise, when you take away 10% of the visits, you could expect a 10% loss in things that depend on time domain. I think the ones that are 20% worse

77
00:13:39.480 --> 00:13:53.740
Eric Gawiser (he/him): we should look at and think if we can be clever about, because sometimes it's about having expected something, and that's what our metric is looking for, and now we're just below that threshold, and so we think all the data's bad. So, everything here needs an iteration before we fully understand it.

78
00:13:55.800 --> 00:13:57.950
Eric Gawiser (he/him): Set of weather simulations.

79
00:13:57.960 --> 00:14:12.150
Eric Gawiser (he/him): These I don't understand in detail, but the main point is, you know, that they can be used to estimate the uncertainty, because any one of these weather realizations could end up happening. And it's still unclear why, on average, these simulations are better than baseline.

80
00:14:12.150 --> 00:14:19.540
Eric Gawiser (he/him): Maybe Lynn knows, maybe they tend to be more optimistic about the weather, but I wouldn't have thought so, so it might require a little bit of discussion here.

81
00:14:19.630 --> 00:14:21.459
Eric Gawiser (he/him): Probably later and offline.

82
00:14:22.230 --> 00:14:37.779
Eric Gawiser (he/him): And then the third plot from Michelle is about some of the other simulations released in this set. So there's faster templates. You'll notice that seems to help a bit with the time domain metrics, that's good. There's adding in the DESE UNG band coverage.

83
00:14:37.780 --> 00:14:44.079
Eric Gawiser (he/him): Now, I want to be clear, we, you know, as a cosmology collaboration, would love to help enable DESI cosmology as well.

84
00:14:44.110 --> 00:15:02.139
Eric Gawiser (he/him): Where we're at right now is we don't yet have in place the metrics needed to figure out the impact of these changes on the desk science, because this basically shifts things in early years, and most of the metrics we're looking at are 10 years. We need to understand what this does to our cosmology results after year one, after 4 years.

85
00:15:02.350 --> 00:15:20.469
Eric Gawiser (he/him): And these particular metrics, things look worse, but again, we need to develop additional metrics that are looking at the impacts after 1 in 4 years. These are the delayed rolling metrics, then extra blue band in DDF for AGN science, and then removing the TOOs, which seems pretty unrealistic.

86
00:15:20.910 --> 00:15:24.100
Eric Gawiser (he/him): none of those are providing a big help. Everything's going in the…

87
00:15:24.240 --> 00:15:27.089
Eric Gawiser (he/him): Wrong direction, but notice that the amplitudes are not huge.

88
00:15:27.240 --> 00:15:31.840
Eric Gawiser (he/him): So nothing here is too scary, these are down by sort of 5%, even when it looks bad.

89
00:15:34.770 --> 00:15:40.020
Eric Gawiser (he/him): I've got a few more slides, but if there are burning questions, it's fine to interrupt me, as far as I'm concerned.

90
00:15:40.760 --> 00:15:45.669
Lynne Jones: You know, I wanted to, to bring up a comment that, that,

91
00:15:46.810 --> 00:15:51.469
Lynne Jones: just about the no TOO, was that, and this is what,

92
00:15:51.610 --> 00:16:06.259
Lynne Jones: just… just… it's that we took them out of the database, but the time that they were involved with… so this way, it looks like it falls and you don't get any gain back. It's just… And the reason we took it out is just because of

93
00:16:08.950 --> 00:16:16.340
Lynne Jones: it's kind of like, okay, and what happens if the TO observations weren't exactly like we thought they were?

94
00:16:16.480 --> 00:16:22.680
Lynne Jones: I honestly, I don't know if this is… that's the best way to analyze that, but it's… it's one thing to look at.

95
00:16:23.230 --> 00:16:24.770
Lynne Jones: Yeah, true. Yeah.

96
00:16:25.490 --> 00:16:28.929
Lynne Jones: You know, Ashley, if you can go back a second just to the weather.

97
00:16:29.670 --> 00:16:37.899
Lynne Jones: I just want to make a comment about this, because the… I think the weather runs are very useful, but it is also… they're a little bit harder to understand.

98
00:16:38.090 --> 00:16:44.429
Lynne Jones: So… In gen… so we… in general…

99
00:16:44.740 --> 00:16:50.360
Lynne Jones: The weather runs… what happens is, we have this big…

100
00:16:51.920 --> 00:17:09.069
Lynne Jones: I think it's 50 years block of weather at this point, maybe 60. And we take the 10 years of the survey, and we are simulating those 10 years of the survey at different points in that weather block, basically. So this is the… the,

101
00:17:09.839 --> 00:17:17.609
Lynne Jones: seeing database, because remember, we used this… sorry, not the seeing database. The seeing database also gets… should get shifted,

102
00:17:17.900 --> 00:17:28.450
Lynne Jones: But the thing is that the… is the cloud coverage database that gets shifted, and so it's, like, the… when the observatory is closed, and how long it's closed.

103
00:17:28.490 --> 00:17:41.780
Lynne Jones: changes depending on which of those years out of the database history you sample. And so that's, the little O and the number is, like, the number of years that were shifted.

104
00:17:41.900 --> 00:17:44.270
Lynne Jones: So 0… O0?

105
00:17:44.730 --> 00:17:49.710
Lynne Jones: In theory, that should be, I believe, the same as our…

106
00:17:50.110 --> 00:17:54.589
Lynne Jones: standard baseline, because it's a shift of zero. And Peter should confirm on that.

107
00:17:54.900 --> 00:18:00.060
Lynne Jones: The fact that it's not exactly the same,

108
00:18:00.180 --> 00:18:03.450
Lynne Jones: Makes me actually wonder about that, if that's really true.

109
00:18:05.090 --> 00:18:06.600
Lynne Jones: I don't know if Peter had a comment.

110
00:18:09.070 --> 00:18:13.009
Peter Yoachim: Yeah, I'm not sure why that one's totally different.

111
00:18:13.180 --> 00:18:26.629
Lynne Jones: So, so another thing that happens is, as you know, you may have noticed that we have different, we have, like, baseline 5.3.0, baseline 5.3.1, and they are slightly, slightly different.

112
00:18:26.630 --> 00:18:33.809
Lynne Jones: In that, what this means is the strategy overall is the same, but we may have changed something in Ribbon Scheduler.

113
00:18:34.120 --> 00:18:50.899
Lynne Jones: So one of the things we did that is different between 5.3.1 and 5.3.0 is we actually made some of our code run faster because Peter did some profiling and said, oh, like, this is, this is not…

114
00:18:51.030 --> 00:18:54.840
Lynne Jones: Running as fast as it could be, and so… so changed a few things.

115
00:18:55.130 --> 00:19:00.299
Lynne Jones: In changing, it really should have been the same, and statistically, overall, it's the same, but…

116
00:19:00.460 --> 00:19:06.230
Lynne Jones: The thing about the visit history is, If some visits are different.

117
00:19:07.130 --> 00:19:24.939
Lynne Jones: then you can have a little bit more of a knock-on effect. So it's not necessarily that, you end up with completely different behavior, it's just you don't end up with exactly the same visit history. And some of these metrics are quite sensitive to the details of

118
00:19:25.460 --> 00:19:29.719
Lynne Jones: what comes up in the visit history, and the Killenovo metric is one of those, because

119
00:19:30.780 --> 00:19:36.620
Lynne Jones: In the kilonova metric, we're injecting a lot of kilonova, like, light curves.

120
00:19:36.740 --> 00:19:42.749
Lynne Jones: But only a very, very small fraction of those are successfully detected.

121
00:19:43.090 --> 00:19:44.430
Lynne Jones: And so…

122
00:19:44.920 --> 00:19:53.330
Lynne Jones: When you have small perturbations in the visit history, the metric itself can change a lot, and so this is why

123
00:19:53.720 --> 00:20:02.329
Lynne Jones: In these runs, this is one estimate of the uncertainty, but there are more things that will contribute to uncertainty in the metric.

124
00:20:02.530 --> 00:20:08.570
Lynne Jones: Number… like, the numbers you get out that… Are harder to capture.

125
00:20:08.920 --> 00:20:17.309
Lynne Jones: So, like, the weather history is one thing. There's the questions… some of the questions I mentioned at the end of the first session,

126
00:20:17.820 --> 00:20:30.400
Lynne Jones: on Thursday, like, we don't have… there is some… some overhead that varies, variability in the visit gap rate, and there is… is going to be variability in the weather, but there's also going to be…

127
00:20:30.520 --> 00:20:34.209
Lynne Jones: you know, variability in our faults and things like that. And so.

128
00:20:35.710 --> 00:20:53.709
Lynne Jones: These weather runs are good because they help give us at least some estimate of uncertainty, but I don't think it captures all of it, and so I think we can look at this and say, okay, this is at least what the uncertainty is probably likely to be, but it also is

129
00:20:54.310 --> 00:21:01.960
Lynne Jones: there's… there's more to it. And I think one of the things you're seeing here, where you look at, weather clouds zero.

130
00:21:02.560 --> 00:21:03.620
Lynne Jones: is…

131
00:21:04.590 --> 00:21:21.139
Lynne Jones: the fact that that's not exactly the same as the baseline is telling you that it was probably run at a slightly different time, and there was some slightly different… slight difference in riven scheduler, and essentially, I think this comes out to be… that's kind of part of the noise, too.

132
00:21:21.390 --> 00:21:21.990
Eric Gawiser (he/him): Okay, that means…

133
00:21:21.990 --> 00:21:22.410
Lynne Jones: sense.

134
00:21:22.410 --> 00:21:22.730
Eric Gawiser (he/him): to me.

135
00:21:22.730 --> 00:21:33.749
Lynne Jones: Yeah, I mean, it's just, it's really hard to get a handle on what the actual uncertainty is, and some metrics are really steady, like you said, the static figure of merit.

136
00:21:34.190 --> 00:21:47.830
Lynne Jones: pretty steady, right? And even here, it varies a bit, but not as much. And then other metrics are very sensitive to some of the details of the simulation, and you find more uncertainty, and the Kilanova one is definitely one like that.

137
00:21:48.910 --> 00:21:54.849
Lynne Jones: I'm actually not sure quite why the supernova one is as sensitive.

138
00:21:55.580 --> 00:22:00.189
Lynne Jones: But… It is, it is a thing.

139
00:22:01.150 --> 00:22:08.130
Lynne Jones: Bill has a question about climate change and so on. I'd be happy to talk about that, but you said you had, like, maybe a few more slides, so…

140
00:22:08.160 --> 00:22:09.050
Eric Gawiser (he/him): Yeah.

141
00:22:09.050 --> 00:22:09.450
Lynne Jones: I will.

142
00:22:09.450 --> 00:22:10.769
Eric Gawiser (he/him): Vaughn, just so I don't…

143
00:22:11.060 --> 00:22:29.949
Eric Gawiser (he/him): end up taking all the time. We can certainly come back to that interesting topic. So this is another analysis that Mi Dai, who's on the line, ran, using light curve links to get a quick simulation. And it's as complementary to Philippe's more involved supernova metric, but also informative, and so across a different

144
00:22:30.000 --> 00:22:32.770
Eric Gawiser (he/him): Set of several different key.

145
00:22:32.880 --> 00:22:45.880
Eric Gawiser (he/him): versions here of the simulations. We're finding a 4-17% reduction in the number of Type 1A supernovae in V5.3. We, of course, expect some reduction because, again, this is

146
00:22:45.880 --> 00:22:57.019
Eric Gawiser (he/him): an estimate for the as-built observatory, we know we're losing 10% of the visits, and there's a larger impact for the more stringent detection criteria. So this comes back to the issue

147
00:22:57.440 --> 00:23:10.449
Eric Gawiser (he/him): you know, as you start to get worse than what we've been planning on, are we being too stringent in our criteria, or is that just what it takes, and this more… the 15% hit is more realistic? I think that's something we need to hash out and be sure about.

148
00:23:10.830 --> 00:23:13.580
Eric Gawiser (he/him): Mee, did you want to say anything more about this? What's on screen?

149
00:23:15.330 --> 00:23:28.359
Dai, Mi: Maybe just a quick clarification. It's, like, one set of simulation, but applying different, like, detection criteria. So, like, with all these different labels, it's, like.

150
00:23:28.360 --> 00:23:42.509
Dai, Mi: For example, for more phases, basically, we are requiring, like, more, like, her points to be selected in our sample, so that makes, like, a more, like, effectively, change in the number, in the end.

151
00:23:43.260 --> 00:23:44.390
Eric Gawiser (he/him): Okay, thank you.

152
00:23:44.630 --> 00:23:56.480
Eric Gawiser (he/him): Alright, so then let me move on to comments that you've seen in a similar format on Day 1 from our official SCOC liaisons, that's Sarb and Rachel, who are both on the line.

153
00:23:56.500 --> 00:24:11.100
Eric Gawiser (he/him): And, I've bold-faced, that's my own commentary, the first bullet here being really important, the desk uniformity metrics were not run. These are essential to understand what's going on in V5.3 and beyond. The metrics aren't…

154
00:24:11.100 --> 00:24:23.909
Eric Gawiser (he/him): as I mentioned earlier, sufficient to assess the DESI sim. We're going to have to add metrics to understand what happens to our photometric redshifts because of this unexpected possible change in the sequencing of the UNG observations.

155
00:24:23.950 --> 00:24:42.200
Eric Gawiser (he/him): And then for early cosmology, defined here is up to year four, so I don't mean exactly official early science. These could be important. We have to look at it very carefully. We're expecting to analyze year one, and probably year four as being highly uniform data releases upon which we can do cosmology analysis.

156
00:24:42.430 --> 00:24:48.699
Eric Gawiser (he/him): Also, TOO issue, and then reminder to ensure that the rolling starts only after templates for that region are available.

157
00:24:50.900 --> 00:24:55.710
Eric Gawiser (he/him): There are some suggestions made by the SEOC liaisons that I want to reiterate here.

158
00:24:55.710 --> 00:25:17.479
Eric Gawiser (he/him): to look at running the simulations with a decreased WFD footprint. In our view, this would be to take areas at high Milky Way extinction and shrink them down a bit to see if we can recover a higher number of visits per area by reducing the area in areas where it's harder to do science. This would, of course, need feedback from galactic science folks as to which regions would be tolerable to lose.

159
00:25:17.480 --> 00:25:23.440
Eric Gawiser (he/him): And, of course, changing the footprints and decision needs to be made early, and impacts the cadence across subsequent years.

160
00:25:23.580 --> 00:25:30.260
Eric Gawiser (he/him): We need to routinely run those uniformity metrics and the year one to year four science metrics, some of which we need to update and add.

161
00:25:30.260 --> 00:25:45.599
Eric Gawiser (he/him): And then there's further work going on, led by Matt Becker, to explore uniform rolling options. I know that Peter has also looked at this in terms of things we might change. Instead of coming to uniform at 8 years 4, 7, 10, maybe it should be 5, 8, 10, or…

162
00:25:45.600 --> 00:25:52.450
Eric Gawiser (he/him): 5, 6, or 5, 7, 10, well, you know, some small changes are possible in that, and so that's being explored.

163
00:25:53.070 --> 00:25:55.860
Eric Gawiser (he/him): I have a slide that's just for reference.

164
00:25:55.960 --> 00:26:03.789
Eric Gawiser (he/him): So, I'm not going to go through this, but these are the desk publications and call responses related to survey strategy, in case anyone wants to track them down.

165
00:26:03.970 --> 00:26:09.139
Eric Gawiser (he/him): And then just a list of summary of what we learned, particularly from Michelle and me's work.

166
00:26:09.620 --> 00:26:24.040
Eric Gawiser (he/him): The biggest hit for us comes to the loss of 10% on-sky visits. No surprise. It's partially mitigated by moving to an 11-year survey. Definitely helps, but doesn't solve everything for us. Other simulations have smaller effects, 5-10% reductions.

167
00:26:24.040 --> 00:26:32.799
Eric Gawiser (he/him): The faster templates results in more visits, so then we do a little bit better on the time domain. Increasing DDF visits in the bluer band has some impact on the transit metrics.

168
00:26:32.800 --> 00:26:56.519
Eric Gawiser (he/him): But it wasn't catastrophic, and we should compare with the full supernova WFD and DDF metrics that Philippe can run to make sure. Obtaining extra U and G band visits and DESI footprint could harm our transient metrics, but we have to figure out how to check for that in intermediate years in a more robust way. And then delaying rolling cadence has an impact of the transient metrics as well, as well as uniformity that we need to cross-check and come up with a…

169
00:26:56.530 --> 00:27:01.240
Eric Gawiser (he/him): Perhaps improved uniform rolling scheme, given what we've now learned about the observatory.

170
00:27:01.630 --> 00:27:07.280
Eric Gawiser (he/him): So that's all I have from desk. Thanks for listening. Rachel and Sarba, if you want to add anything

171
00:27:07.390 --> 00:27:09.390
Eric Gawiser (he/him): Please just speak up.

172
00:27:17.620 --> 00:27:19.989
Eric Gawiser (he/him): Or we can take more general questions or comments.

173
00:27:24.370 --> 00:27:25.840
Lynne Jones: Alright.

174
00:27:27.780 --> 00:27:36.159
Bob Blum (NOIRLab/Rubin): Hey, I have a… I have a question, for Eric. Thanks, Eric, this is really helpful. When you say that the,

175
00:27:37.770 --> 00:27:57.380
Bob Blum (NOIRLab/Rubin): you know, the possible, changing cadence to support DESI would… could harm transients. Do you specifically mean transients in aid of desk science, or in generally? And if it's desk science, does… do you think that desk will

176
00:27:58.290 --> 00:28:05.839
Bob Blum (NOIRLab/Rubin): Attempt to weigh the pluses and minuses of that versus what they might gain from having all those spectra?

177
00:28:08.100 --> 00:28:18.150
Eric Gawiser (he/him): Alright. Thank you, Bob. There are a few things in there. So, the first is that, yes, I mean transit metrics that are relevant to DESC, in particular supernovae.

178
00:28:18.530 --> 00:28:28.549
Eric Gawiser (he/him): Of course, if we're concerned about harm to those transient metrics, there are other transients of interest to a wide variety of science collaborations that might be affected, so we should look for everybody.

179
00:28:28.890 --> 00:28:44.610
Eric Gawiser (he/him): We're also needing to look at the uniformity impacts on photometric redshifts, as was mentioned on a previous slide. So that's a non-transient science issue that comes in from sort of getting deeper in some bands, much shallower in others, than we had anticipated in previous analyses.

180
00:28:44.780 --> 00:28:48.220
Eric Gawiser (he/him): Obviously, we see benefit from lots of DESI spectra.

181
00:28:48.420 --> 00:28:53.610
Eric Gawiser (he/him): But the objects being targeted here from that imaging for DESI2

182
00:28:53.880 --> 00:29:04.899
Eric Gawiser (he/him): have not been a central part of the science plan. There is a Lyman Break Galaxy Science component desk that would benefit from more Lyman Break Galaxy Spectra down the line, so that is something we would weigh in a trade-off.

183
00:29:05.210 --> 00:29:06.990
Eric Gawiser (he/him): So, I would say yes to that.

184
00:29:07.410 --> 00:29:14.490
Eric Gawiser (he/him): But I think we have to be careful not to do something too quickly here before we have a chance to develop the metrics and iterate.

185
00:29:15.120 --> 00:29:18.279
Bob Blum (NOIRLab/Rubin): Thanks, that's helpful, and I agree, it's complicated.

186
00:29:24.980 --> 00:29:29.620
Lynne Jones: Yeah, thanks again, Eric. I…

187
00:29:29.860 --> 00:29:34.959
Lynne Jones: I did have a question, actually, about simulating with a smaller footprint.

188
00:29:35.710 --> 00:29:38.769
Lynne Jones: So I know, I know this request is,

189
00:29:39.480 --> 00:29:43.099
Lynne Jones: It's sort of, coming from multiple places.

190
00:29:44.110 --> 00:29:49.669
Lynne Jones: One of the things that… that this does get a little tricky is that

191
00:29:50.020 --> 00:29:57.410
Lynne Jones: We can reduce the area, in the White Fast Deep and still meet the SRD area requirement.

192
00:29:57.630 --> 00:30:05.360
Lynne Jones: Because… because there's the design requirement, which is at, 18,000 square degrees, and then there's, like, a minimum requirement

193
00:30:05.870 --> 00:30:07.649
Lynne Jones: Which is closer to…

194
00:30:08.480 --> 00:30:15.959
Lynne Jones: I should be careful about this, because I always confuse it whether it's actually 15,000 or 17,000, but it's a smaller number.

195
00:30:16.470 --> 00:30:18.199
Lynne Jones: Slightly smaller.

196
00:30:18.480 --> 00:30:28.759
Lynne Jones: And so, in theory, we can reduce the footprint in WideFast Deep. It then leaves the question, like, what do we do about the part of the footprint that we just

197
00:30:29.370 --> 00:30:31.240
Lynne Jones: moved out of White Festiv.

198
00:30:32.780 --> 00:30:39.659
Lynne Jones: Do we still cover that part of the sky with a reduced number of visits, or do we start leaving holes, right?

199
00:30:39.810 --> 00:30:42.250
Lynne Jones: The other question is,

200
00:30:43.360 --> 00:30:56.730
Lynne Jones: Desk, for a long time, said that the area that is currently at… included in the low dust, wide, fast, deep, is the minimum area that's possible, because it is…

201
00:30:56.960 --> 00:31:00.040
Lynne Jones: The low dust part is about 15,000.

202
00:31:00.290 --> 00:31:05.150
Lynne Jones: square degrees, maybe a little bit more, right? So…

203
00:31:06.690 --> 00:31:14.770
Lynne Jones: If we reduce by, like, 10% and do 5% in high dust and 5% in the low dust area.

204
00:31:15.930 --> 00:31:20.350
Lynne Jones: We're starting to go below that threshold, but is that…

205
00:31:20.960 --> 00:31:35.339
Lynne Jones: is that really as, like, hard cut and dry as it has sometimes sounded in the past? Or is that… well, okay, yeah, you know, if we have 5% less area, we'll still be doing okay, because I'm not sure we actually have a metric about that.

206
00:31:35.520 --> 00:31:36.640
Lynne Jones: To be honest.

207
00:31:38.090 --> 00:31:40.249
Eric Gawiser (he/him): I have a feeling Sarab has a response to this.

208
00:31:40.980 --> 00:31:48.780
Saurabh Jha: Yeah, so I… Lynn, could you remind me, what is the total footprint area? It's, like, 21… something around there?

209
00:31:48.780 --> 00:31:50.920
Lynne Jones: Yeah, it's something like that.

210
00:31:50.920 --> 00:31:55.439
Saurabh Jha: Is that the number that needs to be above 18 in the science requirements, or is it…

211
00:31:55.440 --> 00:31:59.649
Lynne Jones: No, the science requirements number is about the…

212
00:32:00.200 --> 00:32:06.580
Lynne Jones: So, it's the Wide Fast Deep area, which is, area that's covered to

213
00:32:06.720 --> 00:32:15.769
Lynne Jones: the other… so there's two numbers that go together. There's an area and number of visits, and the number of visits is a minimum of 750.

214
00:32:16.480 --> 00:32:23.110
Lynne Jones: In this… in this big area, and the area itself is,

215
00:32:23.560 --> 00:32:27.140
Lynne Jones: See, I should look it up again, but the design number is…

216
00:32:27.920 --> 00:32:36.530
Lynne Jones: 18,000. Just, like, the design number of visits is, like, 825, but we're at the minimum, which is 750, and so…

217
00:32:37.660 --> 00:32:42.699
Lynne Jones: You know, the minimum area is also a smaller number.

218
00:32:43.020 --> 00:32:43.389
Saurabh Jha: Yeah, so…

219
00:32:43.390 --> 00:32:44.839
Lynne Jones: I'm gonna go look it up while you talk.

220
00:32:44.840 --> 00:32:47.320
Saurabh Jha: Okay, yeah, I think the idea…

221
00:32:47.690 --> 00:33:06.849
Saurabh Jha: behind this would be… so, yeah, so that 21 is already… is big, and so, you know, presumably that there's other stuff there, but yeah, so if… I think you're right that the 15 in the low dust region, you know, ideally we wouldn't want to go below that, but that does leave this, like, sort of 3,000 square degrees that is

222
00:33:06.850 --> 00:33:09.279
Saurabh Jha: Too dusty for…

223
00:33:09.320 --> 00:33:22.580
Saurabh Jha: desk extragalactic science, but is getting wide, fast, deep level visit coverage right now. And maybe the galactic folks are like, yes, that's what, absolutely, that region needs that. But…

224
00:33:22.580 --> 00:33:37.640
Saurabh Jha: I suspect, you know, because in the actual plane, they get the galactic plane coverage, which is less, right? And it's, you know, redder and, you know, all the other things. So, if some of that 3,000 square degrees could shift from wide, fast, deep-like coverage to galactic

225
00:33:38.080 --> 00:33:54.229
Saurabh Jha: plane-like coverage, maybe that would be a win for everybody. Or not, you know, not a win, I mean, it'd be less visits, but it wouldn't be a detriment to, you know, people's signs significantly. So those are the kinds of directions we were sort of suggesting to just look at and see if that was, you know.

226
00:33:54.450 --> 00:33:56.040
Saurabh Jha: You know, palatable or not.

227
00:34:05.160 --> 00:34:10.909
Lynne Jones: Yeah, that… it does get tricky, because, as you know, the footprint

228
00:34:11.409 --> 00:34:19.739
Lynne Jones: Took a long time to evolve to where it is. But, yeah…

229
00:34:21.239 --> 00:34:30.519
Eric Gawiser (he/him): But I think what Saurabh just described is a well-defined simulation that can be done, and then everyone's metrics can be run on it, and they can look at it qualitatively, and see if it flies.

230
00:34:38.190 --> 00:35:02.139
Saurabh Jha: you know, like, there may be some very important science that needs wide, vast, deep level visits, and still applies to this dusty region, and then, you know, we'd love to hear what that is, and then, okay, this may not work. But I'm not sure what that science case is, because presumably, for the galactic science, the galactic plane level coverage is already adequate to do galactic science. So, that's the issue, just, you know, people would have to look at it. I totally agree with what Eric just said.

231
00:35:02.700 --> 00:35:06.480
Lynne Jones: So I, I think we have… hopefully we have,

232
00:35:06.970 --> 00:35:09.630
Lynne Jones: Some people to talk about this later.

233
00:35:09.820 --> 00:35:12.130
Lynne Jones: Oh, oh, I see Fed has a handout, too.

234
00:35:12.550 --> 00:35:25.810
Federica Bianco (she/her): Yeah, I think this is a really important point, because of the urgency of the decision, which, as we said, if we want to modify the footprint, we should do that sooner so that we don't cover the footprint with something that then becomes obsolete.

235
00:35:25.850 --> 00:35:39.760
Federica Bianco (she/her): And the multi-science aspect of it. I think this would be a good task force project. We could set up a task force that primarily… certainly you should involve task folks.

236
00:35:40.060 --> 00:35:57.160
Federica Bianco (she/her): And Galaxy Science Collaboration folks and anybody else, you know, representation from any other science collaboration would also obviously be welcome. We could instantiate it and include, people from the community and the SCOC and obviously the survey strategy team.

237
00:35:58.710 --> 00:36:07.970
Saurabh Jha: Yeah, so one thing is, I agree, if we actually reduce the actual area that LCSD points at, then, yeah, that's a decision, you know, you don't want to waste

238
00:36:07.970 --> 00:36:27.140
Saurabh Jha: time pointing at regions that you're not going to actually do later. But if we do it the way I was suggesting, which is change some fraction of the area from wide, fast, deep lake cadence to galactic plane-like cadence, then the footprint basically still stays the same. So, the time to make that decision becomes less urgent, because it's, like, it's just moving visits

239
00:36:27.140 --> 00:36:29.909
Saurabh Jha: Around more than changing area.

240
00:36:31.510 --> 00:36:33.469
Lynne Jones: Yeah,

241
00:36:36.350 --> 00:36:45.410
Lynne Jones: Okay, I… so, first of all, I wanted to make a comment. The minimum area required in Wide Fast Deep over

242
00:36:45.610 --> 00:36:49.120
Lynne Jones: any kind of void festive is 15,000 square degrees, so…

243
00:36:49.780 --> 00:37:05.860
Lynne Jones: certainly within the SRD guidelines, we can reduce the amount of area in Wide Fast Deep. If we move it from Wide Fast Deep coverage into Galactic plane coverage, you don't get all of those visits back, so instead of, like, a 5% gain, if you reduce that, you would get…

244
00:37:08.330 --> 00:37:14.789
Lynne Jones: It's not… it's not as low as half, but you're, you know, the number of visits in that area, it's…

245
00:37:16.550 --> 00:37:21.949
Lynne Jones: just under a third, I think. So you would get some, but not all.

246
00:37:22.250 --> 00:37:26.469
Lynne Jones: And also, John had a comment

247
00:37:26.830 --> 00:37:38.990
Lynne Jones: And do you want to make your comment here, but I also don't want to get too… I think this is something we should study, we could have a look at. And, also, when we do this, we have to be

248
00:37:39.280 --> 00:37:45.840
Lynne Jones: very careful, we're… because, like I said, deciding the footprint was very hard, and there was a lot of…

249
00:37:45.960 --> 00:37:50.439
Lynne Jones: Negotiation and work to get to that point, so…

250
00:37:52.400 --> 00:37:55.280
Lynne Jones: It is true, reopening the footprint is a tough one.

251
00:37:55.770 --> 00:38:04.429
Lynne Jones: Yeah, and the, the Stellar Ops Milky Way Local Volume Group will have.

252
00:38:04.660 --> 00:38:07.499
Lynne Jones: To have a look at what happens as well.

253
00:38:09.390 --> 00:38:10.130
Lynne Jones: So…

254
00:38:13.220 --> 00:38:14.030
Saurabh Jha: Rangers move.

255
00:38:14.030 --> 00:38:17.920
Lynne Jones: Yeah, given… given the reduced number of visits available.

256
00:38:17.920 --> 00:38:31.480
Saurabh Jha: Yeah, that's the thing, that's the thing I wanted to just point out, you know, we're getting close to that 750, which was, like, that minimum requirement for a wide, fast, deep, right? And, you know, I guess what the 11-year survey can help with that, but if it, you know, if we can't count on that, then, you know.

257
00:38:31.810 --> 00:38:42.510
Saurabh Jha: it is getting, you know, it's getting close to, you know, a minimum design requirement, right? And so, you know, I think… I'm not trying to… please, I'm not trying to say, like.

258
00:38:42.510 --> 00:38:43.660
Lynne Jones: Like, you know…

259
00:38:43.660 --> 00:38:58.030
Saurabh Jha: Galactic Science, you know, in any way is… you should make more compromises or anything like that. I just, you know, feel like, oh, you know, this is… it's less visits than people are expecting, and, you know, that's going to be true everywhere, and so, like, you know, just

260
00:38:58.150 --> 00:39:02.949
Saurabh Jha: having some simulations to look at the effects, you know, I think that's all we are proposing right now.

261
00:39:02.950 --> 00:39:03.600
Lynne Jones: Yeah.

262
00:39:04.030 --> 00:39:20.989
Lynne Jones: Yeah, and if you remember, as part of the work to come to that footprint, we did expand the low dust coverage for White Fast Deep as well, so it's… it's a tricky question. How to… how do we do this again?

263
00:39:24.810 --> 00:39:27.370
Lynne Jones: So… All right.

264
00:39:28.430 --> 00:39:33.190
Lynne Jones: great areas for discussion. I do think we should go ahead and… and,

265
00:39:34.110 --> 00:39:38.440
Lynne Jones: Look at a task force with… you know, fairly…

266
00:39:38.910 --> 00:39:44.299
Lynne Jones: Quick turnaround, presumably. Okay, sorry, my agenda…

267
00:39:44.650 --> 00:39:55.460
Lynne Jones: disappeared. So, okay, do we have, Roberto, did you want to say something from the AGN? And did we have anybody from Galaxies who wanted to present anything?

268
00:39:58.050 --> 00:40:07.649
Roberto Assef: For… so from the AGM side, there's not much else to add to what was mentioned, and I think, a lot of the… what…

269
00:40:07.670 --> 00:40:26.880
Roberto Assef: has been, discussed is very much along the lines of the conclusions that Eric just presented. So there will be… so the 10% does affect a lot of the, transient or the variability science, and it's unclear that everything would recover with an 11th, year.

270
00:40:27.860 --> 00:40:32.759
Roberto Assef: And there hasn't been really…

271
00:40:32.760 --> 00:40:48.959
Roberto Assef: much support for the DESE strategy, for adopting the DESE strategy. But one thing that could be interesting to see or to discuss would be if there's any way to find a sort of midpoint

272
00:40:48.960 --> 00:40:52.140
Roberto Assef: Between, so if it goes, ahead.

273
00:40:52.140 --> 00:41:06.729
Roberto Assef: would there be a possibility of not going as deep in U and G, such that you can, even with a reduced cadence, you can still keep more U and G observations throughout the rest of the survey in those areas?

274
00:41:07.260 --> 00:41:23.199
Roberto Assef: And other than that, the only other thing was just to reiterate what Neil mentioned, that there's a strong interest that particularly the ocean strategy is not… does not change to reaccommodate the survey.

275
00:41:23.220 --> 00:41:31.250
Roberto Assef: And I don't think that's part of the discussion. I don't think it has been part of the discussion, but just to… to make sure that it doesn't become part of the discussion, too.

276
00:41:32.370 --> 00:41:40.119
Lynne Jones: That's actually a really good point. We haven't discussed it, but it is true that the deep drilling fields

277
00:41:40.250 --> 00:41:44.430
Lynne Jones: In the current, simulations are coming in a bit over 7%.

278
00:41:44.610 --> 00:41:54.189
Lynne Jones: This is one of the places where I think it's just… there's still a lot of uncertainty. The deep drilling field visits don't tend to…

279
00:41:54.410 --> 00:42:01.690
Lynne Jones: get reduced the way that visits over the rest of the survey do. Like, if we have…

280
00:42:02.630 --> 00:42:09.219
Lynne Jones: If we have bad weather that lasts for a single night, the deep drilling fields will reschedule themselves to the next night.

281
00:42:09.720 --> 00:42:15.869
Lynne Jones: If we have bad weather for several nights in a row, you do lose that time period.

282
00:42:16.060 --> 00:42:23.710
Lynne Jones: So they, they experience, bad weather, or faults, or, or other things.

283
00:42:24.100 --> 00:42:28.769
Lynne Jones: That reduced the number of visits in the rest of the footprint in a slightly different way.

284
00:42:29.020 --> 00:42:37.270
Lynne Jones: And that's why you see them go up in this… in this… these new simulations as a fraction of the total.

285
00:42:40.090 --> 00:42:46.700
Lynne Jones: My inclination would be to say we're still uncertain enough that we are not… we shouldn't modify it now.

286
00:42:47.410 --> 00:43:02.069
Lynne Jones: But I do think this is something we'll have to keep an eye on and think about how we can modify it in the future, especially if the deep drilling fields start getting further ahead of the rest of the survey. So it would definitely be worth

287
00:43:03.660 --> 00:43:12.589
Lynne Jones: thinking about how do we respond if we see that happening? Like, where do we go to figure out, like, what's the next thing?

288
00:43:14.580 --> 00:43:27.660
Lynne Jones: presumably drop, because, you know, if we have more visits, we don't have a problem, but if we have fewer visits, that's when we have a problem. So, yeah, so understanding what to do there, like, if it's,

289
00:43:28.280 --> 00:43:32.669
Lynne Jones: Drop the cadence on short timescales, drop the cadence, like.

290
00:43:32.840 --> 00:43:38.750
Lynne Jones: For some weeks, or something like that, it would be… Get to start,

291
00:43:39.170 --> 00:43:41.180
Lynne Jones: Getting ahead of that problem, I imagine.

292
00:43:42.310 --> 00:43:54.010
Roberto Assef: Yeah, I would imagine there would be some support for the SD version of the DDF simulations, but I think that starts getting even farther away from the nominal 7%, right?

293
00:43:55.550 --> 00:43:58.809
Roberto Assef: And I guess a lot of these questions

294
00:43:58.980 --> 00:44:01.470
Roberto Assef: will be things that we'll have to…

295
00:44:01.770 --> 00:44:11.000
Roberto Assef: to look into, sort of, as well as the survey progresses, but I was wondering, is there a particular, expectation of when

296
00:44:11.160 --> 00:44:28.510
Roberto Assef: we should expect, sort of, new simulations once the first year is underway. So… so with the simulation is what we expect, but once we know how the first year is going, I'm guessing we'll be updating things to… because that will propagate for the rest of the 10 years.

297
00:44:28.670 --> 00:44:33.090
Lynne Jones: Yeah, actually, I love this question because it would give me a chance to…

298
00:44:33.290 --> 00:44:39.709
Lynne Jones: bring up the idea of survey progress reporting, right? Because they are tied together.

299
00:44:40.010 --> 00:44:52.029
Lynne Jones: So, during commissioning, with the SV survey, we had the websitesurvey-strategy.lsst.io, that had periodic updates.

300
00:44:52.350 --> 00:44:55.350
Lynne Jones: During early operations, we…

301
00:44:55.690 --> 00:45:05.360
Lynne Jones: Did not post updates there, for a variety of reasons, and those reasons are not really good reasons anymore, so it…

302
00:45:05.810 --> 00:45:11.650
Lynne Jones: I do have it on my list to update that website and put in some updates on what we've been doing to date.

303
00:45:11.860 --> 00:45:18.889
Lynne Jones: So… and then as we start the survey, those regular updates would have to… will have to start up again.

304
00:45:19.400 --> 00:45:23.160
Lynne Jones: As part of those regular updates,

305
00:45:23.320 --> 00:45:30.710
Lynne Jones: My initial thought was to include not just the visits we've had to date, But also…

306
00:45:30.900 --> 00:45:37.879
Lynne Jones: from those visits, like, what would we simulate for the next ongoing period? And…

307
00:45:39.230 --> 00:45:48.540
Lynne Jones: It may be that only simulating for the next year is useful. It may be that running for the rest of a 10-year simulation is more useful.

308
00:45:49.030 --> 00:46:08.350
Lynne Jones: it's kind of a trade-off, because we can post updates there every month or so, but I don't think people would actually really want to have… I mean, sorry, an updated simulation, an updated 10-year simulation, but I'm not sure people would really necessarily have the bandwidth to take those,

309
00:46:08.690 --> 00:46:21.599
Lynne Jones: monthly updates and do much with them. On the other hand, some people might, so it's likely worthwhile for us to do that anyway, because as part of some of our metrics that we want to report, like.

310
00:46:21.600 --> 00:46:28.800
Lynne Jones: oh, given the current pace of the survey, what do we think we'll do at the end of 10 years? We'll have additional,

311
00:46:29.390 --> 00:46:39.319
Lynne Jones: We'll have to run those kind of simulations, and then we'll have the metrics output, and then we might as well make those simulations available to people as well.

312
00:46:41.780 --> 00:46:56.460
Lynne Jones: The other reporting we are currently doing is nightly reporting. So we did this during the SV survey and commissioning, and we quietly continued doing it during, early operations, and now we're

313
00:46:56.590 --> 00:47:03.109
Lynne Jones: talking about it more again. This site… let me grab the link for you.

314
00:47:06.900 --> 00:47:13.269
Lynne Jones: So this site… is, here, I'll put it in the Zoom chat.

315
00:47:17.130 --> 00:47:24.849
Lynne Jones: So, we've been working on these daily reports. If you looked at them during the SV survey, and you look at them now, you can see that

316
00:47:25.060 --> 00:47:32.929
Lynne Jones: there are some improvements, and I… we're working on it further, so if you have feedback to give us about these.

317
00:47:33.040 --> 00:47:37.890
Lynne Jones: Please go ahead, you can send it to… the…

318
00:47:39.070 --> 00:47:47.549
Lynne Jones: the, the Sims channel on… that's probably the best place to do it. Sims channel on the LSSTDA Slack.

319
00:47:47.930 --> 00:47:58.100
Lynne Jones: This… at the moment, these just have nightly reports, but Eric Nielsen is working on standing up a

320
00:47:58.310 --> 00:48:18.170
Lynne Jones: daily progress report, so this will be somewhere in between. We made a new simulation of the whole rest of the survey, but instead it will be like, okay, we take the visits we've had till now, and we take our most recent simulation of the rest of the survey, and we just paste them together at the interface. So then you can have daily updates on

321
00:48:18.170 --> 00:48:22.399
Lynne Jones: What's… where we look like we are compared to the rest of the survey?

322
00:48:23.110 --> 00:48:27.180
Lynne Jones: So that's, still a bit of a work in progress.

323
00:48:27.460 --> 00:48:31.740
Lynne Jones: And we're still trying to figure out what's the most useful things to have for people on the…

324
00:48:33.210 --> 00:48:36.659
Lynne Jones: Sort of handwritten survey strategy website.

325
00:48:38.830 --> 00:48:47.030
Lynne Jones: But yes, we will be able to make those simulations available to people. We wouldn't, in general for those, have a lot of

326
00:48:47.460 --> 00:48:48.880
Lynne Jones: Variations?

327
00:48:50.730 --> 00:48:56.049
Lynne Jones: we might run the weather variations, because I think that might be useful, but we would…

328
00:48:56.310 --> 00:48:59.640
Lynne Jones: We would probably just stick to, like, here's the baseline.

329
00:49:00.010 --> 00:49:11.170
Lynne Jones: So there will still be the kinds of, okay, and now what about different survey strategy options that come in a different set, like a coherent set, like we're doing now for the V5.3?

330
00:49:11.480 --> 00:49:17.089
Lynne Jones: And at some point, I think we'll have to look and see what is…

331
00:49:17.640 --> 00:49:26.400
Lynne Jones: Helpful for the community, and how we can help people… guide people to, like, what's the most useful version of these simulations for… for their work.

332
00:49:29.870 --> 00:49:37.309
Lynne Jones: So that's a very good question, and thank you for giving me a chance to talk about that, which we didn't have on the schedule, but

333
00:49:38.070 --> 00:49:50.320
Lynne Jones: It's, yeah, that's the plan. We'll have things like this, and I don't know what the most useful timescale is for that yet, and so if you have feedback about that, that's useful to know, too.

334
00:49:56.540 --> 00:50:10.569
Roberto Assef: Well, thank you very much for the… for the answer, and yeah, I don't have any… anything else to… to add. I'm guessing with respect to the question you were just saying about when would it be most useful, I'm guessing

335
00:50:10.640 --> 00:50:27.019
Roberto Assef: it shouldn't be too fast, because things vary in timescales and average out, and I think, you know, we don't want to, sort of, panic modes, you know, every now and then, but I'm… but yeah, finding what the right cadence is might be… it's not as tricky.

336
00:50:29.930 --> 00:50:31.350
Lynne Jones: Yeah, the…

337
00:50:33.220 --> 00:50:40.479
Lynne Jones: Yes, I agree. It's a little tricky, and you don't want to perturb somebody when we've had a bunch of downtime, but…

338
00:50:41.270 --> 00:50:42.030
Lynne Jones: Yeah.

339
00:50:44.480 --> 00:50:51.240
Lynne Jones: We should… so thank you, Roberto. We should open it up to more general discussion, because I don't think…

340
00:50:51.810 --> 00:51:02.400
Lynne Jones: that we have a report from the Galaxy Science Collaboration. I'll just remind you from the SEOC analysis the other day, the concern is uniformity,

341
00:51:02.440 --> 00:51:15.329
Lynne Jones: Certainly also the loss of visits and loss of depth, but, bringing up the question of uniformity and what we can do about that is important. And I was going to say that, Leanne…

342
00:51:15.720 --> 00:51:23.249
Lynne Jones: said she was going to file an RFC, that's a request for comments, in… with, ribbon to see what…

343
00:51:23.460 --> 00:51:29.330
Lynne Jones: How we could do… Some kind of uniform co-add, and what would be the options around that?

344
00:51:30.680 --> 00:51:32.150
Lynne Jones: Bindu.

345
00:51:33.410 --> 00:51:45.069
Bindu Rani: Hi, yeah, Roberto, I think there is a question about, the mattresses, and I recall that, I don't recall that we discussed this in the EGN meeting, did we?

346
00:51:46.520 --> 00:51:48.090
Roberto Assef: Which, sir.

347
00:51:48.090 --> 00:51:52.430
Bindu Rani: So, that is the question, from Federica. The question is,

348
00:51:53.070 --> 00:52:03.840
Bindu Rani: Are we happy with the mattresses? Do we understand? I mean, it's related to, like, how do we feel about the current mattresses?

349
00:52:04.350 --> 00:52:07.040
Roberto Assef: Okay, yeah, so that's… yeah, we haven't discussed…

350
00:52:07.040 --> 00:52:08.070
Bindu Rani: We haven't, discussed.

351
00:52:08.070 --> 00:52:21.030
Roberto Assef: I think there's some questions about the time lag metrics, that there's some things that… that I don't fully, or that it's hard to understand exactly why they are reacting the way they are to the changes in the metrics.

352
00:52:21.030 --> 00:52:28.590
Roberto Assef: But I think that has been happening for some time now. So yeah, we should discuss this more in detail, and I think…

353
00:52:29.370 --> 00:52:45.610
Roberto Assef: I think it would be interesting to know also about this sort of future updates, because that means that writing metrics is not necessarily something that is finished, but we could be updating metrics and adding more metrics as, sort of, science interests keep changing.

354
00:52:47.240 --> 00:52:48.120
Bindu Rani: Thank you.

355
00:52:49.450 --> 00:52:56.560
Lynne Jones: Yes, also, thank you, Roberto. Writing metrics is not finished, and we do hope people will continue writing metrics.

356
00:52:56.710 --> 00:53:03.139
Lynne Jones: So… Yeah, that's, that's, definitely a good comment.

357
00:53:04.650 --> 00:53:10.660
Lynne Jones: But you had your hand up earlier, I think you put it down now, though, so…

358
00:53:11.200 --> 00:53:16.080
Federica Bianco (she/her): Yes, I was gonna bring up the question of the metrics, so, done.

359
00:53:19.170 --> 00:53:21.690
Lynne Jones: Good, good. Okay,

360
00:53:25.020 --> 00:53:33.009
Lynne Jones: Are there other… there was some discussion about the weather, and I'll just recap that,

361
00:53:33.510 --> 00:53:35.440
Lynne Jones: So Phil's question was…

362
00:53:35.870 --> 00:53:47.319
Lynne Jones: with the metrics being better, looking better with the… when we were offset by 30 years. So, first of all, you have to be careful about which way that offset works, because,

363
00:53:47.800 --> 00:53:54.800
Lynne Jones: I think the O30 is actually 30 years backwards, but… It's not.

364
00:53:54.960 --> 00:54:01.279
Lynne Jones: Is that right, Peter? The O number is an offset backwards in time, or is it offset forwards?

365
00:54:01.810 --> 00:54:07.639
Peter Yoachim: Oh, I don't have no idea. Probably forwards. I think we started at the beginning and then shifted forwards.

366
00:54:07.840 --> 00:54:19.769
Lynne Jones: Yeah, so I think that it's very hard to interpret which way that shift went, and remember that we wrap around in the cloud coverage database if we hit the end.

367
00:54:21.150 --> 00:54:30.160
Lynne Jones: I do think it went forward, because I always thought it went backwards, and when I looked at the actual code, it went the other way than I thought I did, so I think, yeah, it goes forwards.

368
00:54:30.190 --> 00:54:42.859
Lynne Jones: And Peter's comment was, it's hard to tell from what we have, and the year-to-year variation is just very big. And Eric Nielsen, who's done a lot more work on this and has included a link to, some of his analysis.

369
00:54:42.920 --> 00:54:48.770
Lynne Jones: said, It's… it's not… it's not really obvious.

370
00:54:49.140 --> 00:54:49.959
Lynne Jones: On what we have.

371
00:54:51.290 --> 00:54:56.059
Phil Marshall (SLAC/Rubin): Yeah, I followed those, those comments through, so thanks… thanks all for that.

372
00:54:56.670 --> 00:55:04.769
Phil Marshall (SLAC/Rubin): I guess we, I guess when you choose a 10-year chunk to choose to be the fiducial.

373
00:55:05.300 --> 00:55:09.839
Phil Marshall (SLAC/Rubin): We didn't necessarily choose… we didn't know which was gonna be the…

374
00:55:10.160 --> 00:55:12.669
Phil Marshall (SLAC/Rubin): The… the median value in the end.

375
00:55:12.900 --> 00:55:13.570
Phil Marshall (SLAC/Rubin): True.

376
00:55:13.660 --> 00:55:23.980
Lynne Jones: Yeah, that is true, because, and we've actually expanded the database since the time when we picked that initial tenure chunk.

377
00:55:24.110 --> 00:55:41.619
Lynne Jones: So, we definitely didn't pick a chunk that was the median overall, but we did pick a chunk where, when we set a threshold, it roughly matched the numbers of… the numbers of downtime nights due to weather.

378
00:55:42.110 --> 00:55:46.150
Lynne Jones: that we… that were coming from, I think it was from Gemini.

379
00:55:46.480 --> 00:55:54.620
Lynne Jones: Now… The time when we did that calibration was already, over 10 years ago.

380
00:55:55.650 --> 00:55:57.769
Lynne Jones: Because we've been working on this for a while.

381
00:55:57.770 --> 00:56:20.259
Lynne Jones: And so, it is also true that when, last summer, I was trying to compare, do our numbers match Gemini's numbers, it wasn't exactly one-to-one, like, a clear and obvious match. So I do think this is something where, as we go forward in the survey, we will just have to look at what we're actually getting on the sky compared to what our records say.

382
00:56:20.260 --> 00:56:26.770
Lynne Jones: And the other thing is that it is also different from observatory to observatory, what kind of weather you actually shut down for.

383
00:56:26.980 --> 00:56:27.770
Phil Marshall (SLAC/Rubin): Hmm.

384
00:56:27.770 --> 00:56:30.210
Lynne Jones: Which we saw… last…

385
00:56:30.390 --> 00:56:38.160
Lynne Jones: you know, southern winter, we were shutting down under conditions that we won't be shutting down for this next summer, for example.

386
00:56:38.880 --> 00:56:40.820
Lynne Jones: So…

387
00:56:42.570 --> 00:56:52.850
Lynne Jones: It's hard to say, and I, you know, we'll just have to keep looking and checking. This is not an obvious sign that we're wrong at the moment.

388
00:56:53.410 --> 00:56:54.359
Phil Marshall (SLAC/Rubin): Yeah, thank you.

389
00:57:02.420 --> 00:57:07.670
Lynne Jones: So, I didn't get feedback from other groups, about…

390
00:57:08.120 --> 00:57:10.680
Lynne Jones: Wanting to make sure they had time to present.

391
00:57:10.890 --> 00:57:14.920
Lynne Jones: I don't know if anybody contacted you directly, Fed.

392
00:57:16.530 --> 00:57:17.820
Federica Bianco (she/her): Not that I know.

393
00:57:18.100 --> 00:57:18.820
Lynne Jones: Okay.

394
00:57:19.200 --> 00:57:23.519
Lynne Jones: And, so… But maybe we can…

395
00:57:23.520 --> 00:57:30.090
Federica Bianco (she/her): Just, you know, renew the invitation if anybody wants to speak on behalf of their science or a science collaboration.

396
00:57:30.230 --> 00:57:35.510
Federica Bianco (she/her): without slides, just any comments would be very appreciated by the SCOC.

397
00:57:36.100 --> 00:57:44.159
Lynne Jones: Well, and I know Kat had slides for the, Solar System Science Collaboration, which were scheduled for 10 o'clock.

398
00:57:44.410 --> 00:57:47.980
Lynne Jones: But, if you want to go ahead now, Kat, I think this is.

399
00:57:47.980 --> 00:57:48.930
Kat Volk: Yeah.

400
00:57:49.180 --> 00:57:53.720
Kat Volk: I don't actually have slides, I just have comments to pass along.

401
00:57:54.510 --> 00:58:01.879
Kat Volk: Yeah, we kind of had a Google Doc, and I think, Lynn, you've seen the Google Doc of the feedback, so… But mostly…

402
00:58:02.380 --> 00:58:13.820
Kat Volk: The biggest concern I've heard from the solar system is that they are unsure about these, shorter template observations, the 22nd, template tier.

403
00:58:14.130 --> 00:58:17.919
Kat Volk: Just because they're not…

404
00:58:18.380 --> 00:58:29.219
Kat Volk: Convinced it's been demonstrated that the depth will be comparable to the 30-second exposures, and they are concerned about losing depth during year one, if we go to the shorter template tier.

405
00:58:29.590 --> 00:58:32.510
Kat Volk: And then…

406
00:58:33.130 --> 00:58:44.880
Kat Volk: you know, just the same assessment that I kind of gave the other day, that if we add that 11th year, that really does help recover a lot of the solar system metrics. So those are kind of the two comments I got overall, you know.

407
00:58:45.560 --> 00:58:51.649
Kat Volk: The baseline looks like the best of the options, given the kind of uncontrolled

408
00:58:51.800 --> 00:58:57.110
Kat Volk: The things we can't change, which is the downtime and the actual performance.

409
00:58:59.870 --> 00:59:12.910
Kat Volk: So, Fed, they're concerned about, losing depth in the exposures themselves, in terms of using them for science. Yeah, so Solar System has figured out how to detect the objects without the templates.

410
00:59:13.230 --> 00:59:19.870
Kat Volk: So yeah, I think it's a concern about losing depth during year one, since so many of the observations will be in that template tier.

411
00:59:20.590 --> 00:59:23.530
Lynne Jones: Yeah, I mean, I think this is a really

412
00:59:23.960 --> 00:59:27.049
Lynne Jones: It… it is a valid concern.

413
00:59:31.220 --> 00:59:46.839
Lynne Jones: it's not super obvious… yeah, I don't know if that… that's a good thing to keep in mind. We have a long list of concerns in the RFC that's in progress in the project that are mostly around, like, the technical questions,

414
00:59:47.190 --> 00:59:49.879
Lynne Jones: Of, of what… whether or not we can…

415
00:59:50.950 --> 00:59:59.950
Lynne Jones: Whether we can sustain the 20-second exposures, given various other things that are happening with the observatory.

416
01:00:00.120 --> 01:00:04.550
Lynne Jones: So I think it is a really… Good question.

417
01:00:04.680 --> 01:00:07.500
Lynne Jones: I suppose.

418
01:00:07.500 --> 01:00:08.189
Kat Volk: Yeah, and I know that.

419
01:00:08.190 --> 01:00:08.580
Lynne Jones: in November.

420
01:00:08.580 --> 01:00:27.400
Kat Volk: You know, the template tier is supposed to be only when the sky conditions are really good, but then it's also that, like, 70% of the observations need to be in the template tier during year one, so that would, I think, in my mind, kind of question, are they all… is 70% going to be under the best conditions? So, yeah, I think that's just something to keep in mind.

421
01:00:27.920 --> 01:00:29.090
Lynne Jones: Yeah.

422
01:00:29.820 --> 01:00:32.070
Federica Bianco (she/her): Is there a trade-off?

423
01:00:32.490 --> 01:00:48.449
Federica Bianco (she/her): in getting the template… because, you know, solar system doesn't necessarily need the templates, but the templates still benefit solar system science. So, is there a trade-off between getting shorter, and therefore less deep exposures in year one, but getting

424
01:00:48.800 --> 01:00:57.380
Federica Bianco (she/her): better, you know, more comprehensive coverage, and earlier on with the templates, and can sort of system sort of weigh the two

425
01:00:58.060 --> 01:01:00.380
Federica Bianco (she/her): The two… the two issues together.

426
01:01:01.400 --> 01:01:09.469
Kat Volk: I mean, that's something I can ask the collaboration to think very carefully about, so I will… I will pass that question along.

427
01:01:11.510 --> 01:01:20.059
Kat Volk: I guess my other comment, and just looking at the feedback I have gotten, I don't think anybody's really looked at the DESI simulations in detail.

428
01:01:20.340 --> 01:01:22.870
Kat Volk: So, I think there are still…

429
01:01:23.110 --> 01:01:25.770
Kat Volk: Thus, something that will need to be investigated.

430
01:01:25.950 --> 01:01:31.570
Kat Volk: Especially because, as I said the other day, I don't know that our metrics fully capture that.

431
01:01:31.680 --> 01:01:37.919
Kat Volk: And then something else I was thinking about since, we talked about it, like, I'm not sure what the…

432
01:01:38.450 --> 01:01:53.439
Kat Volk: benefits would be for solar system from some of the data and other things that have been offered in exchange for the cadence changes for DESE, so that's just something to keep in mind, that I'm not sure what could be traded to the SSSC if there are

433
01:01:53.560 --> 01:02:02.500
Kat Volk: things that hurt the science for solar system, for that particular cadence. That's coming from my opinion. We haven't really talked about the DESI stuff.

434
01:02:02.660 --> 01:02:06.470
Kat Volk: in the Solar System Science Collaboration, just because there hasn't been time

435
01:02:06.600 --> 01:02:11.769
Kat Volk: to look at that in detail. So, that last bit is my personal question.

436
01:02:12.160 --> 01:02:20.210
Kat Volk: But, yeah, I will… I will ask them about the trade-off, faster templates versus depth to think about that in a little more detail.

437
01:02:21.930 --> 01:02:25.030
Kat Volk: So that's pretty much what I had from the solar system.

438
01:02:30.580 --> 01:02:32.719
Lynne Jones: Yeah, great. Thanks, Kat.

439
01:02:41.400 --> 01:02:45.499
Lynne Jones: If there's, no more comments about these in particular.

440
01:02:46.000 --> 01:02:59.809
Lynne Jones: I see Kathy said that John has some results to talk about for the, SMWLV science collaboration. Science… STARS Milky Way Local Volume, yeah.

441
01:03:00.030 --> 01:03:02.459
Lynne Jones: I always get the acronym the wrong way around, so…

442
01:03:03.700 --> 01:03:10.569
John Gizis: Well, this was only for the parallaxes, because I'd been looking at them and checking them, so it's not representing

443
01:03:12.160 --> 01:03:18.800
John Gizis: All of the countless other, things. But, I'm sorry, Anna.

444
01:03:18.910 --> 01:03:20.040
John Gizis: Hitting the wrong button.

445
01:03:21.250 --> 01:03:23.420
Lynne Jones: You know, and I was just gonna say… I'm just gonna show…

446
01:03:23.420 --> 01:03:26.700
John Gizis: No, but without doing the full presenter view, I think.

447
01:03:26.870 --> 01:03:28.989
John Gizis: Hopefully this is fine, the way it is.

448
01:03:31.220 --> 01:03:32.160
Lynne Jones: That looks good.

449
01:03:32.970 --> 01:03:35.040
John Gizis: I interrupted you, I'm sorry, go ahead.

450
01:03:35.370 --> 01:03:40.039
Lynne Jones: I was just gonna say, and I know there's a lot of…

451
01:03:40.260 --> 01:03:48.410
Lynne Jones: diverse science and diverse metrics in, both this group and the TVS group, so…

452
01:03:49.320 --> 01:03:53.400
John Gizis: Yeah, so, anyway, for the parallax issue, why we always,

453
01:03:53.800 --> 01:04:05.240
John Gizis: brought it up, and why it's in the… some of the requirements and so forth, is you do have to be careful how you sample the parallax curve, and maybe that's illustrated on the bottom right of this plot that I have from an earlier

454
01:04:05.370 --> 01:04:20.309
John Gizis: version, right? Which is because as the sky becomes available, you want to observe the source, and as it disappears, you want to observe the source, you know, near, you know, near its transit. And the sort of the data in between is sort of sampling

455
01:04:21.660 --> 01:04:40.510
John Gizis: you know, regions where there's no parallax signal, and then, of course, there's the time when it's up in the daytime, which also has no useful information. So you just have to be a little careful that you're not chasing things that are rising and setting at the beginnings and ends of nights, because that's a way of removing your parallax signal.

456
01:04:40.510 --> 01:04:55.859
John Gizis: As it turns out, you know, the version 5.3, you know, it's perfectly good and will produce lots of great science. I was interested in the rolling cadence question, and so on the left sort of shows you

457
01:04:55.930 --> 01:05:09.839
John Gizis: sort of how non-uniform rolling cadence is. I chose a different color stretch than the traditional, observatory one, which sort of shows you that there'll be, like, regions of the sky. Of course, they'll vary, you know, depending

458
01:05:10.010 --> 01:05:29.059
John Gizis: on, you know, reality, where you really have sampled things pretty well, and so you'll be competitive, you know, it'll be quite a good test, I think, of the parallaxes, of small parallaxes that you've added together. And of course, other parts of the wide, fast, deep footprint, you can see, are kind of bluish. They're not going to do as well due to the rolling cadence.

459
01:05:29.230 --> 01:05:37.560
John Gizis: None of that is a reason to design the survey around. Over 10 years, it all gets averaged out, as you can see here on the right. And so,

460
01:05:37.560 --> 01:05:50.000
John Gizis: Everything looks quite nice. And so the overall thoughts is that version 5.3, like 5.0 and many of the other major releases, has preserved sampling of the parallax curve and the reddest filters.

461
01:05:50.180 --> 01:05:58.500
John Gizis: Thank you. DESI, has no impact whatsoever on this, right? Because, it's not removing IZY observations.

462
01:05:58.650 --> 01:06:02.039
John Gizis: Of course, I just wrote… put this in, but…

463
01:06:02.190 --> 01:06:08.319
John Gizis: Losing wide, fast, deep area, of course, proportionally loses the number of objects you can measure.

464
01:06:08.480 --> 01:06:21.490
John Gizis: On the other hand, as you make it, the sampling worse, you don't do as well in the parallax, so you don't go as far out. So that's a bit complicated, you know, we can always look at the simulations, but, you know, very likely it's going to be fine.

465
01:06:22.880 --> 01:06:41.369
John Gizis: 5.3 compared to 5.0 is sort of 6% worse for the brighter objects, where you can use IZY, and for the intrinsically fainter objects, it's sort of 10% worse. On the other hand, 11 years of 5.3 is better than 10 years of 5.0 for this science case.

466
01:06:41.370 --> 01:06:59.210
John Gizis: You know, we, you know, astrometry would really, you know, really does love doing just an extra year of data and, and averaging it all together. The deep drilling fields are an example of being on the wrong side of the trade-off of area versus sampling. Of course, they will get wonderful, wonderful astrometry.

467
01:06:59.210 --> 01:07:07.800
John Gizis: but it's not a big enough area of the sky to add up to anything. There won't be… there's not that many things within 100 or 200 parsecs in those particular fields of view.

468
01:07:07.930 --> 01:07:15.960
John Gizis: And then just as we put in the Slack earlier, someone should just double-check that the 20-second exposures

469
01:07:16.120 --> 01:07:29.640
John Gizis: are not somehow a disaster for astrometry due to the astronomical turbulence and so forth. Under the existing metrics, faster templates, especially over 10 years, has no impact whatsoever. There's no problem at all.

470
01:07:29.730 --> 01:07:45.149
John Gizis: With doing it from that point of view. It's just a question that, you know, traditionally, you try to take longer exposures to average out over this stuff. Remember that it's not necessarily the seeing, it's how well you can measure the angle between two objects an arcminute apart.

471
01:07:45.150 --> 01:07:55.710
John Gizis: And so, evidently, experience is that that's not necessarily the same as having good seeing, right? And so different nights are different ways. But the modern software corrects that very well.

472
01:07:55.770 --> 01:08:03.380
John Gizis: And so maybe it's… maybe it's not an issue at all. It's something that perhaps someone should just double check, because if it turns out that you…

473
01:08:03.670 --> 01:08:07.139
John Gizis: Have twice as much uncertainty unexpectedly than it would be.

474
01:08:07.470 --> 01:08:15.289
John Gizis: Then you really would be wiping out the first year of the survey for proper motions and astrometry, but most likely you aren't, so it'll be fun.

475
01:08:15.480 --> 01:08:18.560
John Gizis: That's all I had to say, so I will try to stop sharing.

476
01:08:20.240 --> 01:08:29.850
John Gizis: It's certainly a narrow science case, and but as far as I can tell, everything is great, and we thank the SCOC for keeping in mind these metrics.

477
01:08:32.109 --> 01:08:33.569
Lynne Jones: Perfect. Thank you, John.

478
01:08:34.729 --> 01:08:38.830
Lynne Jones: Alright.

479
01:08:41.510 --> 01:08:49.030
Lynne Jones: It is entirely possible that nobody else has anything else to share, but I'm not sure,

480
01:08:50.080 --> 01:08:55.180
Lynne Jones: Fed, do you know if anybody was going to present from the TVS collaboration?

481
01:08:55.180 --> 01:08:57.739
Federica Bianco (she/her): I did not hear from anybody.

482
01:08:58.000 --> 01:08:58.720
Lynne Jones: Okay.

483
01:08:59.830 --> 01:09:02.460
Federica Bianco (she/her): But we should ask Sarah.

484
01:09:02.580 --> 01:09:04.740
Federica Bianco (she/her): Sarah, one of the co-chairs, is here.

485
01:09:04.740 --> 01:09:06.380
Lynne Jones: Oh, okay, great.

486
01:09:12.220 --> 01:09:20.340
Sara Bonito: I, I don't have, update from, TVS members.

487
01:09:20.590 --> 01:09:22.360
Sara Bonito: to share today.

488
01:09:22.939 --> 01:09:27.329
Sara Bonito: With you. But of course, we…

489
01:09:27.490 --> 01:09:32.209
Sara Bonito: We will discuss after the conclusion of this meeting more.

490
01:09:32.310 --> 01:09:45.349
Sara Bonito: During the TVS regular meetings, and, we have, we will have, a collection of, inputs more after that.

491
01:09:45.790 --> 01:10:02.500
Sara Bonito: And, last time, Rachel has, highlighted, the, the important, influence that the new, simulations can have, and the impact on, TVS science.

492
01:10:05.110 --> 01:10:08.560
Lynne Jones: Yeah, it's very valid. We didn't… The…

493
01:10:08.870 --> 01:10:17.689
Lynne Jones: Our timescale for releasing these simulations and then asking for feedback is not what we would have hoped in general.

494
01:10:17.800 --> 01:10:19.910
Lynne Jones: There are various reasons.

495
01:10:20.010 --> 01:10:28.010
Lynne Jones: Around that, having to do with how quickly things are evolving.

496
01:10:28.140 --> 01:10:34.410
Lynne Jones: With, sort of, performance of the observatory in general, and then how, how…

497
01:10:34.610 --> 01:10:38.639
Lynne Jones: Quickly, we would also like to, update

498
01:10:38.980 --> 01:10:41.709
Lynne Jones: Our survey strategy for the start of the survey.

499
01:10:41.830 --> 01:10:51.090
Lynne Jones: And again, like I said in the very first day, this is primarily an update to V5.0, and not necessarily…

500
01:10:52.260 --> 01:10:57.109
Lynne Jones: a very big change in the overall survey strategy.

501
01:10:57.380 --> 01:11:03.940
Lynne Jones: The major changes being the fact that we see how the templates

502
01:11:04.710 --> 01:11:20.090
Lynne Jones: what the requirements are for the templates, and it's a little bit more stringent than we were expecting, and so we have also increased the likelihood of being able to acquire those template images by changing the survey strategy somewhat.

503
01:11:22.190 --> 01:11:33.429
Lynne Jones: So, yes, it is… it is a very tough timescale for… to… to turn around and say, okay, actually, everybody give us this feedback, and this is actually why we have the SEOC, so that they can…

504
01:11:33.570 --> 01:11:37.580
Lynne Jones: Also buffer those decisions, and then make the…

505
01:11:37.840 --> 01:11:48.639
Lynne Jones: make these decisions on shorter timescales, working with the Rubin Observatory, and then still give options for community feedback.

506
01:11:50.700 --> 01:11:57.430
Sara Bonito: Sorry, I will just, like to add that also for Trans and Variable Star Science Collaboration is,

507
01:11:57.470 --> 01:12:10.900
Sara Bonito: is not so straightforward as far as so many scientific cases, and galactic and astrogalactic science is involved. So, the ideal plan

508
01:12:10.900 --> 01:12:22.570
Sara Bonito: is to meet at least with all the leaders of the two groups of transient variable stars, and once all the material from the ICOC meeting,

509
01:12:22.590 --> 01:12:43.990
Sara Bonito: Which, by the way, this is a question for you. Are the recordings already available? Sorry for the question, maybe you already have the material. This material will be extremely useful for people that cannot join and to discuss further during the general meetings of TBS.

510
01:12:47.950 --> 01:12:51.980
Lynne Jones: I'm sorry, I'm not sure I quite understood the…

511
01:12:51.980 --> 01:12:54.930
Sara Bonito: The recording material from…

512
01:12:54.930 --> 01:13:04.730
Lynne Jones: recordings from this, yes, yes, I will have to talk to the Nora Lab, because it's the Noora Lab soon, so I will have to talk to them to get the recordings. I…

513
01:13:05.280 --> 01:13:09.799
Lynne Jones: I actually don't know the answer to that, I'm sorry. I should have checked before today, and I…

514
01:13:10.210 --> 01:13:23.489
Federica Bianco (she/her): I don't know that answer either, but let me add, we post them, as a reminder, we traditionally have posted them on the webpage of the meeting as links, and we will do the same, and we will let you know when they're posted, but also keep an eye on the webpage.

515
01:13:24.400 --> 01:13:25.589
Sara Bonito: Perfect, thank you.

516
01:13:29.170 --> 01:13:30.120
Lynne Jones: That's right.

517
01:13:32.670 --> 01:13:36.410
Lynne Jones: So, I think we should take…

518
01:13:38.440 --> 01:13:43.550
Lynne Jones: We sort of had a break scheduled for 9.20.

519
01:13:45.650 --> 01:13:54.709
Lynne Jones: And then we had other presentations for… or discussion sections for… for the rest of the afternoon… oh, for the rest of this time. Fed has a hand up.

520
01:13:54.710 --> 01:13:55.200
Federica Bianco (she/her): Yeah.

521
01:13:56.110 --> 01:14:00.520
Federica Bianco (she/her): No, I just wanted to say, as liaison into the,

522
01:14:00.670 --> 01:14:17.410
Federica Bianco (she/her): Informatic and Statistics Science Collaboration, that they will be excited to contribute to the metrics work and the evaluation work, and so to reach out to the ISSC, there's a number of ways in which we have thought we might be doing it, for example, hackathons.

523
01:14:17.410 --> 01:14:25.009
Federica Bianco (she/her): Because the timeline is so rapid, it's a bit hard to decide what is the best path to help.

524
01:14:25.010 --> 01:14:46.229
Federica Bianco (she/her): people get involved with the ISSC folks who are on the technical side, but some of the software that has been developed that falls under the umbrella of Lynx and ISSC, that recently has been published, like the Lynx, like, framework for Liker simulations within OpSense, can be really helpful. So if you have concerns about your metrics.

525
01:14:46.230 --> 01:15:00.010
Federica Bianco (she/her): Reach out to your liaison in the SEOC, to the chairs of the ISSC, to me, and we can see if the ISSC can help you realize metrics that are more effective to measure the impact of different options.

526
01:15:06.280 --> 01:15:08.580
Lynne Jones: Yes.

527
01:15:08.750 --> 01:15:26.739
Lynne Jones: My proposal was going to be… we take, some time now, we have our discussion, and then when we're done, we can be done for today, and not worry about the later sessions, since I think we've already covered all the groups who I know will be presenting.

528
01:15:26.960 --> 01:15:31.900
Lynne Jones: Is that… would that be,

529
01:15:32.130 --> 01:15:34.259
Lynne Jones: In line with what you would…

530
01:15:34.490 --> 01:15:37.060
Lynne Jones: expect to do here, Fed, because,

531
01:15:37.480 --> 01:15:39.180
Lynne Jones: I know you had lined up that.

532
01:15:39.450 --> 01:15:46.729
Lynne Jones: the schedule for today, so I want to make sure I'm not missing something, but I also think that we've covered all the things we needed to, and we can take our time back.

533
01:15:48.110 --> 01:15:55.699
Lynne Jones: Excuse my dog. Alright, so… Yeah, are there any…

534
01:15:56.040 --> 01:15:59.560
Lynne Jones: Other questions or things you wanted to discuss?

535
01:15:59.970 --> 01:16:05.999
Lynne Jones: about the V5.3 simulations, or the start of the survey, or what information you might be able to get back.

536
01:16:06.140 --> 01:16:09.449
Lynne Jones: As we go through the survey,

537
01:16:11.600 --> 01:16:14.920
Lynne Jones: Or maybe you have questions about the SEOC process?

538
01:16:19.760 --> 01:16:21.609
Lynne Jones: What do you have a hand?

539
01:16:21.950 --> 01:16:23.040
Lynne Jones: Vincenzo.

540
01:16:24.040 --> 01:16:33.050
Vincenzo Petrecca: I have just a short question, maybe it was already asked. Are you going to prepare a summary document as after every workshop?

541
01:16:33.400 --> 01:16:37.530
Vincenzo Petrecca: Describing the changes in the observation schedule.

542
01:16:39.250 --> 01:16:41.669
Lynne Jones: the…

543
01:16:41.670 --> 01:16:42.420
Federica Bianco (she/her): Thanks for that, Lynn.

544
01:16:42.420 --> 01:16:45.190
Lynne Jones: Yeah, I was gonna say, you're working on this already.

545
01:16:45.360 --> 01:16:59.379
Federica Bianco (she/her): Yeah, so the SCSC is preparing one of our recommendations for the start of the LSST, so what will be implemented at the start of LSST's survey, and how it is consistent and inconsistent with previous

546
01:16:59.380 --> 01:17:20.879
Federica Bianco (she/her): observation, simulations, and with 5.3. The spoiler alert is that, basically, we're going to recommend to start with 5.3, but keep enough flexibility to be able to change details of the survey, even during year one, as we hear feedback from the community, and assessment of the decisions that are being made.

547
01:17:21.030 --> 01:17:38.570
Federica Bianco (she/her): so that usually we had produced a draft of the document before the workshop, then had the workshop, and then released the final document. We obviously could not do this because of the fact that as we enter operation, decisions are gonna need to be made in a much more rapid

548
01:17:38.570 --> 01:17:43.949
Federica Bianco (she/her): much more rapid flow, but we will have a draft, and we will incorporate

549
01:17:43.950 --> 01:17:46.650
Federica Bianco (she/her): Comments and feedback from this workshop in the draft.

550
01:17:49.080 --> 01:18:01.769
Vincenzo Petrecca: Thank you very much. So, actually, this is very useful, especially when you need to plan for grants, apply for telescope time and whatsoever, so you have a reference of what is the expected observing plan for LSSD.

551
01:18:01.770 --> 01:18:18.630
Federica Bianco (she/her): Yeah, and sorry, I used the word draft in my last sentence, I didn't mean that. We will release the recommendation to the director, as we usually do. That definitely will be public, and you should expect it to be public in, you know, very short time, like, days, not weeks.

552
01:18:19.420 --> 01:18:21.110
Federica Bianco (she/her): Or maybe just a couple of weeks.

553
01:18:31.360 --> 01:18:32.140
Lynne Jones: Alright.

554
01:18:40.310 --> 01:18:48.400
Lynne Jones: Maybe we have… Talked ourselves out about all the… Upcoming start of the survey.

555
01:18:49.090 --> 01:18:55.669
Lynne Jones: Phil, I don't know if you had any… any…

556
01:18:56.170 --> 01:19:05.809
Lynne Jones: closing comments you wanted to make, or… or just comments about where we're standing right now as we're about to start? Oh, actually, sorry, I see Keith actually has a hand up, so…

557
01:19:06.120 --> 01:19:09.230
Lynne Jones: If you want to defer to him and let him go first, that's good.

558
01:19:09.230 --> 01:19:10.450
Phil Marshall (SLAC/Rubin): Yeah, go ahead, Keith.

559
01:19:11.250 --> 01:19:24.579
Keith Bechtol: Okay, yeah, this is Keith just asking as scientists, not in a special position. What is the earliest time that the SEOC would consider making

560
01:19:24.880 --> 01:19:30.850
Keith Bechtol: a recommendation of an adjustment relative to version 5.3? Does that… does that set a…

561
01:19:31.330 --> 01:19:37.180
Keith Bechtol: an approximate timeline for science collaborations providing input…

562
01:19:40.260 --> 01:19:45.300
Federica Bianco (she/her): So I want to say the SE… as soon as the SEOC has…

563
01:19:46.480 --> 01:19:53.949
Federica Bianco (she/her): Time to ponder an idea, and ensure that it's viable, and…

564
01:19:54.510 --> 01:20:06.280
Federica Bianco (she/her): leads to scientific improvements, we would release a recommendation, or anyhow, start communicating with the director's office that we think that's a good idea and it should be implemented.

565
01:20:06.280 --> 01:20:20.510
Federica Bianco (she/her): we don't have a schedule for when the scheduler can be, like, how frequently can the scheduler be changed in year one? I don't think we should actually make that schedule, because in year one, things will change rapidly, and…

566
01:20:20.600 --> 01:20:23.770
Federica Bianco (she/her): It… I don't think we want to constrain that.

567
01:20:24.140 --> 01:20:25.350
Federica Bianco (she/her): Of course.

568
01:20:25.350 --> 01:20:46.390
Federica Bianco (she/her): at the same time, everybody should understand that once we get an idea, on our… on our, you know, on… under our eyes, then maybe we need to make a simulation, and we need to deliberate on a simulation, and we need to ensure that everybody has seen the simulation and assessed all the impacts of that idea and that change. So that process itself takes time.

569
01:20:46.390 --> 01:20:57.670
Federica Bianco (she/her): So I don't expect that we will be able to make recommendations on time scales of days or weeks. It will more be, like, a month or two before we can recommend make a change.

570
01:20:58.370 --> 01:21:01.660
Federica Bianco (she/her): Any… anybody on the SEOC that may want to add, or…

571
01:21:01.660 --> 01:21:05.380
Lynne Jones: Actually, Fred… Fred, can I give, like, a very specific…

572
01:21:05.480 --> 01:21:19.880
Lynne Jones: example here, because, I mean, I think you're probably right, that that's a more realistic timescale for the SEOC, but I also want to push back that some of these we may actually be looking at trying to do that sooner, like…

573
01:21:19.960 --> 01:21:34.380
Lynne Jones: say the RFC for 20-second exposures comes back. Okay, yes, we've tested it out, we've actually taken some on-sky data, and it all looks okay, so we think we'd like to try this, if the SEOC says that's the change they want to make.

574
01:21:35.550 --> 01:21:42.699
Lynne Jones: what's the soonest you think you might say, yes, go ahead, please flip over to 20-second exposures? And then a second example of…

575
01:21:43.060 --> 01:21:47.060
Lynne Jones: We haven't looked at reducing the footprint, but when do you think you might

576
01:21:47.600 --> 01:21:52.830
Lynne Jones: come back with something that says, yes, actually, go ahead and reduce the footprint, if we were going to.

577
01:21:55.730 --> 01:22:02.709
Lynne Jones: And I know those are tough questions, so I'm just like, what's the ballpark, like, from the people in the SEOC?

578
01:22:02.980 --> 01:22:12.849
Lynne Jones: Personally, I think, like, the faster templates, if we were able to validate that that was possible, that you could probably switch over to those within a month.

579
01:22:13.010 --> 01:22:22.650
Lynne Jones: And… if… We were reducing the footprint that that would… might take, more like, 2 months.

580
01:22:22.920 --> 01:22:23.609
Lynne Jones: But I don't know.

581
01:22:23.610 --> 01:22:39.050
Federica Bianco (she/her): Right, and I think the key difference is some of those ideas are already something that we're evaluating, and obviously the process will be faster for those, but if it's something new that we have not considered yet, then it needs some physiological time to make sure that it's a safe choice.

582
01:22:46.810 --> 01:22:48.049
Lynne Jones: Keith, didn't you have a…

583
01:22:48.050 --> 01:22:56.910
Keith Bechtol: Yeah, so maybe the follow-up to my question then is, should we consider any of the options that have been discussed at this workshop as…

584
01:22:57.060 --> 01:22:59.540
Keith Bechtol: Formal proposals, so to speak.

585
01:23:04.280 --> 01:23:10.590
Lynne Jones: Are you… are you asking that in terms of, like, from the project side, they are…

586
01:23:10.730 --> 01:23:16.049
Lynne Jones: there are things that we should… we should go forward in investigating? Is that… is that what you mean?

587
01:23:16.050 --> 01:23:21.070
Keith Bechtol: No, I'm asking from the SCOC perspective if any of the…

588
01:23:21.920 --> 01:23:25.999
Keith Bechtol: Any of the options that we've discussed this workshop, are they…

589
01:23:28.380 --> 01:23:33.129
Keith Bechtol: Is it, I mean, is it, is it really, like, a concrete proposal that…

590
01:23:33.790 --> 01:23:47.279
Keith Bechtol: that the SCOC is considering and is prepared to move forward on in terms of a recommendation, right? Like, will there be another juncture for the science collaborations to provide input before that decision is made?

591
01:23:50.140 --> 01:23:56.609
Federica Bianco (she/her): So let me, so, the current set of simulations that we have…

592
01:23:57.070 --> 01:24:05.089
Federica Bianco (she/her): The recommendation that we make is gonna be based, obviously, on simulations that we have, and that we have seen, and we had time to evaluate.

593
01:24:05.110 --> 01:24:21.019
Federica Bianco (she/her): So baseline 5.3, or something very similar to that, is what we are recommending to start the survey with if the survey starts on a short time scale, which we all think it will.

594
01:24:22.400 --> 01:24:39.249
Federica Bianco (she/her): As far as, time for feedback from the science collaborations, the next official point… meeting point, where we are all in the same room, or virtual room, and we can present and entertain presentations from the audience and the SCOC, is the RCW session, which I think this year is on Tuesday.

595
01:24:39.250 --> 01:24:55.849
Federica Bianco (she/her): And in between, every science collaboration and every, every science collaboration should, contact their, liaisons and go through their liaisons, and also everybody, in the community can communicate with the SCSC directly, me on Slack, etc.

596
01:24:55.850 --> 01:24:57.750
Federica Bianco (she/her): So,

597
01:24:57.750 --> 01:25:12.680
Federica Bianco (she/her): If the survey starts in order of days, then we would only entertain, simulations, things that we have seen in simulations and that are very close to what we're looking at right now with version 5.3.

598
01:25:17.720 --> 01:25:21.400
Lynne Jones: I will also say there is… there is…

599
01:25:21.550 --> 01:25:25.429
Lynne Jones: Obviously, some room for short-term variation, so…

600
01:25:25.830 --> 01:25:31.309
Lynne Jones: Let's say, for whatever reason, we can only observe in I-band when we go on Skype.

601
01:25:32.230 --> 01:25:39.959
Lynne Jones: we're only going to observe an I-band. That's not going to be a question, right? We're going to observe, and it won't be the strategy that we've seen.

602
01:25:40.460 --> 01:25:42.659
Federica Bianco (she/her): We don't need an SEOC recommendation for that.

603
01:25:42.660 --> 01:25:47.130
Lynne Jones: We don't. We absolutely don't. That would be crazy.

604
01:25:47.430 --> 01:25:59.520
Lynne Jones: Where we might need an, depending on how long we were stuck in that mode, we might need an SEOC recommendation for what to do afterwards, and because that timescale might be quite short.

605
01:25:59.890 --> 01:26:07.190
Lynne Jones: we may… that may be a thing where the SCOC would end up recommending something that maybe didn't manage to get.

606
01:26:07.610 --> 01:26:10.959
Lynne Jones: a lot of input from the community.

607
01:26:12.030 --> 01:26:21.959
Lynne Jones: I think there's just, like, some corner cases here, right, that we… that we haven't run into, we may never run into, but that are… that are there, and so I think the SEC…

608
01:26:22.750 --> 01:26:33.839
Lynne Jones: Absolutely should always vet everything through the community as much as possible, but you still have to reserve the right to make a decision when you need to make a decision on a shorter timescale.

609
01:26:35.350 --> 01:26:36.150
Lynne Jones: No.

610
01:26:39.360 --> 01:26:43.399
Phil Marshall (SLAC/Rubin): I think one thing to keep in mind when, you know, thinking through these

611
01:26:43.530 --> 01:26:47.580
Phil Marshall (SLAC/Rubin): Decision-making things, that there is a difference between

612
01:26:47.680 --> 01:27:03.749
Phil Marshall (SLAC/Rubin): survey strategy that we want to get right in the… in the long term, which means making good decisions now, but also survey tactics, which is more kind of responding to, you know, realities at the observatory, and I think… I think we're going to have to

613
01:27:04.030 --> 01:27:10.929
Phil Marshall (SLAC/Rubin): Respond tactically as planned on… on those short timescales with… with,

614
01:27:11.450 --> 01:27:21.199
Phil Marshall (SLAC/Rubin): with limited input from the SCOC. I think where… where the SCOC process really helps… helps Ruben out, most is in,

615
01:27:21.220 --> 01:27:31.760
Phil Marshall (SLAC/Rubin): you know, getting good information out to you in the form of simulations, allowing time for… for considered feedback, and then working it in thoughtfully. So, I think… I think we…

616
01:27:31.990 --> 01:27:50.660
Phil Marshall (SLAC/Rubin): we can imagine, benefiting from SCOC, you know, rapid feedback during year one, but it feels to me as though most of the benefit is going to come from, you know, how to adjust the survey strategy, in… in years two and beyond. And so if there… if there are

617
01:27:50.760 --> 01:27:51.760
Phil Marshall (SLAC/Rubin): M…

618
01:27:52.160 --> 01:28:03.300
Phil Marshall (SLAC/Rubin): kind of difficult things there about, you know, the details of Year 1 in order to try and make Year 2 and beyond go well, then it's worth bringing them up. Otherwise, we're going to need to be

619
01:28:03.630 --> 01:28:11.730
Phil Marshall (SLAC/Rubin): making tactical choices on the fly while… while the SCOC and the science collaborations think hard about how to make the best of the rest of the survey.

620
01:28:18.920 --> 01:28:19.620
Lynne Jones: Yeah.

621
01:28:20.690 --> 01:28:25.640
Lynne Jones: And if things go smoothly, we don't have to worry about much, but… We'll see.

622
01:28:26.100 --> 01:28:31.889
Lynne Jones: Thanks, Phil. That's a good perspective. May?

623
01:28:34.490 --> 01:28:43.049
Meg Schwamb: Sorry, I've only been able to join in the last few minutes. I just wanted to raise, you know, I was talking about feedback and community feedback at that

624
01:28:43.220 --> 01:28:46.999
Meg Schwamb: and SDRC feedback, is that we also need transparency.

625
01:28:47.340 --> 01:29:03.169
Meg Schwamb: And we need time to be able to just digest things. One of the reasons I think there's less or lower attendance, or lower, at least in terms of solo science collaboration attendance, is that very little warning, and I understand why, about this meeting and getting feedback and getting people to…

626
01:29:03.450 --> 01:29:08.300
Meg Schwamb: Be able to engage, because it's end of semester for many people.

627
01:29:08.450 --> 01:29:13.090
Meg Schwamb: Two-week notice of simulations going out, and timing of this workshop, and…

628
01:29:13.530 --> 01:29:29.439
Meg Schwamb: So, I think, as I understand it, there will have to be tactical changes made, right, based on what the observatory needs are, but how is that going to be communicated to the community? Because even this SCOC process, depending on who your liaison or what they're able to say.

629
01:29:29.700 --> 01:29:36.500
Meg Schwamb: is how you're getting information, and I think it shouldn't have to rely on one person in the community to be able to

630
01:29:37.150 --> 01:29:41.719
Meg Schwamb: tell us what's going on. And so I'm just wondering if that is, again.

631
01:29:42.130 --> 01:29:58.860
Meg Schwamb: Survey Strategy website, again, using community better so that we are… because the last update of notes, you know, I think… I don't know if the last few meetings were updated on the SCOC website, or the SEOC community posts, because those things help us understand what's going on.

632
01:29:58.860 --> 01:30:03.789
Meg Schwamb: And so, I know there will be rapid things that will have to be decided where there won't be time for community input.

633
01:30:03.840 --> 01:30:12.229
Meg Schwamb: But finding a way for the observatory to tell us about those things, or those decisions as they go would be helpful. And I don't know where that would be.

634
01:30:12.300 --> 01:30:15.850
Meg Schwamb: In terms of things right now, and so I think that would be really important.

635
01:30:16.020 --> 01:30:20.809
Meg Schwamb: Like, just, like, knowing that nobody knows what the timescale roughly is for templates.

636
01:30:22.460 --> 01:30:24.489
Lynne Jones: Yes, and I…

637
01:30:24.710 --> 01:30:31.120
Lynne Jones: agree on all of those points, and I also think it's really been very difficult over the last

638
01:30:31.320 --> 01:30:38.370
Lynne Jones: Early operations phase, and I hope it gets clearer and smoother.

639
01:30:38.630 --> 01:30:52.650
Lynne Jones: My initial thought is that, yes, we will be providing updates on the current survey strategy, and progress on surveystrategy.alices.t.io, and we can also,

640
01:30:53.260 --> 01:30:58.989
Lynne Jones: When we have, sort of, approved changes in the survey strategy, we can update it there.

641
01:30:59.180 --> 01:31:15.660
Lynne Jones: think that there's another place that we probably need for… when we're talking about, like, what are the options for survey strategy changes, and that's currently as the SUC and the liaisons, and community, and if that's not…

642
01:31:16.250 --> 01:31:21.239
Lynne Jones: Not working well, then we probably need to have some updates in that process.

643
01:31:29.890 --> 01:31:31.240
Lynne Jones: Go ahead, Mike.

644
01:31:32.150 --> 01:31:41.809
Meg Schwamb: I guess just to respond to that, I think also it's that we have a lot of new people. Like, the people who've done the cadence are tired and busy, and I can't get them to look at things.

645
01:31:42.060 --> 01:31:49.650
Meg Schwamb: And there's a lot of new people who are overwhelmed and don't understand anything. And so I think I want to push in the observatory a little bit that

646
01:31:49.930 --> 01:31:52.130
Meg Schwamb: Documentation is your job.

647
01:31:52.300 --> 01:32:10.879
Meg Schwamb: not saying just Yulin, but in general, and I think there's a lot of information that's not available in places to get people up to speed, to then be able to help understand, to then understand those more detailed documents about things. And I know that this workshop, as part of that reason for having it, is to help with that.

648
01:32:11.140 --> 01:32:16.860
Meg Schwamb: But what… but videos and workshops can't be the only way that this information is available.

649
01:32:17.040 --> 01:32:25.879
Meg Schwamb: Because that's not long-term sustainability, and so I want to just push it that as we go into that this is the cadence, or whatever is the cadence for the start.

650
01:32:26.030 --> 01:32:29.340
Meg Schwamb: As quickly as possible, getting that up in different ways.

651
01:32:29.560 --> 01:32:44.209
Meg Schwamb: on websites and things, so people can digest that, because I think, you know, that helps with these more detailed things or information that's coming out. That's what I'm finding, is people are finding it difficult to get new people in to know enough to understand NASH.

652
01:32:44.310 --> 01:32:49.399
Meg Schwamb: Or understand what a math metric is, and then the people that did are now like, I'm busy, and I'm teaching, and…

653
01:32:49.530 --> 01:32:53.089
Meg Schwamb: you know, my kids are on vacation, so I'm off.

654
01:32:53.630 --> 01:32:59.040
Meg Schwamb: So I'm wondering if there's a balance, and that's maybe where the CST can come in and help, of, like.

655
01:32:59.240 --> 01:33:18.070
Meg Schwamb: you know, walking us through those more basic information, lots more basics about how the scheduler works and things like that, to take that off of the liaisons, because I don't think that's their job. I think their job's to help interpret some of those things when you have enough information to have those discussions, but how do we get people in the door to understand

656
01:33:18.140 --> 01:33:32.670
Meg Schwamb: you know, what the mini-surveys are, or when can we apply, talk about things for year two, because we're talking about DESE, but can we start talking about solar system detrilling fields? We didn't think that was available to year two, but DESE seems to be talking to people now.

657
01:33:32.770 --> 01:33:37.220
Meg Schwamb: So how do we get those things on that? And maybe that's somewhere the CST could help.

658
01:33:40.220 --> 01:33:45.889
Phil Marshall (SLAC/Rubin): Yeah, thanks, Meg. That's a good suggestion. And I, you know, I see the problem that you're…

659
01:33:46.230 --> 01:33:55.429
Phil Marshall (SLAC/Rubin): that you're flagging. We do need to help new people come in and understand the process and the scheduler in order to be able to make good contributions to this discussion. So, yeah, thank you.

660
01:33:55.570 --> 01:33:57.040
Phil Marshall (SLAC/Rubin): Fed, you had a hand up.

661
01:33:57.520 --> 01:34:03.770
Federica Bianco (she/her): Yeah, to clarify, it is not the job of the liaisons to train people on how to run the metrics.

662
01:34:03.790 --> 01:34:22.970
Federica Bianco (she/her): The job of the liaison is to maintain communication between each science collaboration and the SEOC, let the science collaboration know what the SCOC is doing, let the SCOC know what the science collaborations are doing, and help us interpret the metrics of the science collaborations, in addition to bringing their own expertise.

663
01:34:22.970 --> 01:34:26.879
Federica Bianco (she/her): Their own specific scientific expertise on the table.

664
01:34:26.920 --> 01:34:39.479
Federica Bianco (she/her): A question, about ways of, trying to help people learning, relearning, or remind themselves how to use the maps. Would a hackathon.

665
01:34:39.480 --> 01:34:50.680
Federica Bianco (she/her): Be a good way to do that, where people can be live in person, or would videos, be a better option?

666
01:34:50.680 --> 01:34:59.520
Federica Bianco (she/her): would maybe take in some times at the RCW, although I don't know that there is enough, time to do that on the schedule, but maybe during the unconference.

667
01:34:59.520 --> 01:35:01.960
Federica Bianco (she/her): be an option to do that. And also.

668
01:35:02.070 --> 01:35:17.919
Federica Bianco (she/her): I want to point out that the experts on math are the survey strategy team, so they are, like, Lynn and Peter and very few other people, and the community science team is not… they're not experts on math, so while they may help in a number of ways, they can't coach people

669
01:35:17.920 --> 01:35:22.880
Federica Bianco (she/her): On how to write a map that's outside of the scope or their expertise.

670
01:35:24.220 --> 01:35:25.380
Lynne Jones: I think…

671
01:35:25.690 --> 01:35:32.310
Lynne Jones: So I think we do have documentation on some of this, but I suspect that it's just not very accessible.

672
01:35:32.660 --> 01:35:40.580
Lynne Jones: And that… When we did… we did actually make a couple of,

673
01:35:41.010 --> 01:35:47.599
Lynne Jones: slightly more accessible slideshows and stuff, and those are probably lost.

674
01:35:47.710 --> 01:35:51.340
Lynne Jones: in the seas of Google Drive at this point, and…

675
01:35:51.470 --> 01:35:58.110
Lynne Jones: And that we… so one place the CST could perhaps help us is in pull…

676
01:36:01.250 --> 01:36:05.330
Lynne Jones: They have a better process for documentation.

677
01:36:05.440 --> 01:36:08.410
Lynne Jones: And so perhaps we… what…

678
01:36:08.670 --> 01:36:18.609
Lynne Jones: needs to happen here is that we need to lean on them a little bit more for, like, a better process, so we can keep it and keep it up to date a little bit better.

679
01:36:19.110 --> 01:36:26.929
Lynne Jones: It is also true that… The overhead of that documentation is a…

680
01:36:27.800 --> 01:36:36.410
Lynne Jones: Little bit large for our team right now, given that we don't actually have new people coming in saying we want to write new metrics.

681
01:36:36.600 --> 01:36:41.999
Lynne Jones: And I think we are a little bit of a… A point where…

682
01:36:42.540 --> 01:36:55.510
Lynne Jones: we need to look at, like, what we have in our codebase, what other people are starting to develop through Lync, primarily, and figure out how we can… we can join these together, or, like.

683
01:36:55.740 --> 01:36:57.689
Lynne Jones: How… how we bridge that gap.

684
01:36:59.400 --> 01:37:08.300
Lynne Jones: Because I do see… also see a risk that, as you've heard today, there are people who run metrics in their own way.

685
01:37:08.450 --> 01:37:26.020
Lynne Jones: using the outputs that we have, which is totally fine. We certainly want to encourage, like, people to… to evaluate, survey simulations however they need to. However, if we don't know how to do that, and we don't get that contributed, and we did have this framework for trying to… to…

686
01:37:26.100 --> 01:37:30.570
Lynne Jones: Accept those contributions, then we can't run them?

687
01:37:31.030 --> 01:37:48.760
Lynne Jones: And so… so there's… there's, I think this is not the thing we've been focusing on in the last year or so, because we've been focusing very much on, like, getting reporting and monitoring stuff set up, and we've been focusing on getting the scheduler working and working at the summit.

688
01:37:49.330 --> 01:37:58.769
Lynne Jones: So it's probably just part of our cycle that we need to come and look at community metrics again and see what we need to do to get, sort of, to the next stage.

689
01:38:01.250 --> 01:38:04.859
Lynne Jones: But yeah, it's a… it's a really good question.

690
01:38:04.970 --> 01:38:06.780
Lynne Jones: Fed, you have your hand up?

691
01:38:06.780 --> 01:38:25.519
Federica Bianco (she/her): Yeah, quick… just quickly, perhaps we can run… if people are still paying attention to the Slack channel of this workshop, perhaps we can run a quick poll on what people think would be the most effective way for them to learn, and also help us base the decision on what to do, whether they are video tutorials or just

692
01:38:25.520 --> 01:38:31.359
Federica Bianco (she/her): Asynchronous pages that, show tutorials.

693
01:38:31.360 --> 01:38:44.020
Federica Bianco (she/her): As written documentation, or if an interaction, like a hackathon or at the RCW, perhaps, could be, could be valuable, or later, as soon as we could possibly arrange.

694
01:38:46.830 --> 01:38:47.370
Lynne Jones: Yeah.

695
01:38:48.620 --> 01:38:50.230
Lynne Jones: Indeed.

696
01:38:50.450 --> 01:39:06.430
Bindu Rani: Yeah, I see many people have their hands up, so I want to be very brief. So what we do for SWIFT is, we organize, like, 2-3 hour online training session, because we know every year there may be new researchers coming in. For them, it's useful.

697
01:39:06.660 --> 01:39:10.239
Bindu Rani: And then we record it, and we post it as well, so…

698
01:39:10.580 --> 01:39:13.930
Bindu Rani: That way, their old videos are there all the time.

699
01:39:14.150 --> 01:39:19.129
Bindu Rani: And we share slides, so I think we find that it's really effective in that way.

700
01:39:22.340 --> 01:39:31.859
Lynne Jones: Yeah, and to be honest, we do still have our old workshops and RCW sessions are accessible, and so we need to

701
01:39:33.640 --> 01:39:40.850
Lynne Jones: I think pull those out into a more accessible place, because the problem with posting them on the workshop websites is that

702
01:39:40.960 --> 01:39:45.470
Lynne Jones: they're harder to find, I think.

703
01:39:47.130 --> 01:39:48.380
Lynne Jones: David?

704
01:39:49.590 --> 01:39:58.389
David Schlegel: Yeah, I do apologize first, I can't show up to a Rubin workshop and not bring up that you are leaving something on the table

705
01:39:58.580 --> 01:40:04.799
David Schlegel: if you did dynamically controlled exposure time, that would…

706
01:40:05.210 --> 01:40:18.710
David Schlegel: both effectively gain more survey time, improve uniformity. I mean, there are no downsides at all, and, you know, people are always asking me, how did Desi get ahead of schedule?

707
01:40:18.950 --> 01:40:21.360
David Schlegel: You know, such a big survey, like…

708
01:40:21.490 --> 01:40:25.970
David Schlegel: That… that's absolutely how. I mean, that's, you know, you…

709
01:40:25.970 --> 01:40:26.480
Lynne Jones: I love the.

710
01:40:27.380 --> 01:40:32.469
Lynne Jones: Yeah, sorry, sorry, I shouldn't cut you off. I'm so enthusiastic that you brought it up again, though.

711
01:40:32.770 --> 01:40:36.929
Lynne Jones: We actually did look at this in an early… in an early simulation.

712
01:40:37.560 --> 01:40:46.809
Lynne Jones: And… The problem we have is that our,

713
01:40:49.640 --> 01:40:54.849
Lynne Jones: So, so there's a couple of problems, but the… the…

714
01:40:55.890 --> 01:41:03.059
Lynne Jones: It basically came down to, we can't… if we make our exposure time longer on average.

715
01:41:03.720 --> 01:41:05.810
Lynne Jones: Then we lose the number of epics.

716
01:41:05.810 --> 01:41:06.450
David Schlegel: Right.

717
01:41:06.450 --> 01:41:07.100
Lynne Jones: No pictures.

718
01:41:07.100 --> 01:41:09.799
David Schlegel: But you don't do that. You can keep…

719
01:41:09.800 --> 01:41:11.829
Lynne Jones: It's very hard to predict ahead of time.

720
01:41:12.870 --> 01:41:15.420
David Schlegel: Well, you can make exposures shorter as well.

721
01:41:15.620 --> 01:41:28.480
Lynne Jones: Yes, and we did. But we can't make them too much shorter, because we do have this problem, like, we… we already have… when we ran these simulations, we set a floor at, I think, 20 seconds for the visit.

722
01:41:28.610 --> 01:41:35.949
Lynne Jones: And as you see today, we may… we may have a problem even at running at 20 seconds much of the time.

723
01:41:38.120 --> 01:41:47.740
Lynne Jones: In our simulations, it was hard to set the set points, because you have to have some way to control what is your current exposure time, and it was hard to set them

724
01:41:47.930 --> 01:42:05.240
Lynne Jones: ahead of time, when you didn't know what you were going to be getting for the rest of the survey, so that you ended up with the same number of epics, the same average exposure time. And so it did turn out to be quite difficult, and when we simulated it, we kept turning out with,

725
01:42:05.350 --> 01:42:07.960
Lynne Jones: Slightly longer exposure times.

726
01:42:08.450 --> 01:42:16.449
Lynne Jones: And so all of the transient metrics were having a little bit of a problem with that.

727
01:42:17.420 --> 01:42:30.149
Lynne Jones: So it's… so first of all, it's just kind of hard to predict ahead of time. You know, maybe we can work around that and be, like, be smarter and, like, continually adjust what those thresholds are so that we're… we're sort of continually…

728
01:42:30.410 --> 01:42:37.589
Lynne Jones: Resetting so that we have an average exposure time that's about 30 seconds.

729
01:42:37.740 --> 01:42:42.940
Lynne Jones: The… the problem is we don't… also don't necessarily have a lot of window we can work with.

730
01:42:44.470 --> 01:42:52.530
Lynne Jones: It's much easier to make the exposure slightly longer, especially in clouds, than it is to make them shorter, since we can't go much shorter than,

731
01:42:52.900 --> 01:42:57.160
Lynne Jones: Well, maybe we can go to 20 seconds, but I'm…

732
01:42:58.500 --> 01:42:59.160
David Schlegel: Well, already go.

733
01:42:59.160 --> 01:43:02.130
Lynne Jones: And it certainly can't go shorter than 15 seconds, so…

734
01:43:02.130 --> 01:43:04.860
David Schlegel: Yeah, I mean, the overheads, yep.

735
01:43:04.860 --> 01:43:05.440
Lynne Jones: Yeah.

736
01:43:05.440 --> 01:43:13.930
David Schlegel: Desi has approximately the same overheads as a fractional time, so it's 2 minutes out of 15. So it's about the same, and…

737
01:43:14.320 --> 01:43:16.650
David Schlegel: Just… it's just a huge win.

738
01:43:18.840 --> 01:43:20.100
Lynne Jones: Yeah, no.

739
01:43:20.100 --> 01:43:23.460
David Schlegel: You're never… you're never throwing data out, effectively.

740
01:43:26.920 --> 01:43:34.459
Lynne Jones: Yeah, and all I can say in… Here, is that… We did look at it.

741
01:43:34.700 --> 01:43:36.639
Lynne Jones: And it was…

742
01:43:36.980 --> 01:43:45.509
Lynne Jones: very hard to turn into something that we could use on Sky effectively and not make our main exposure time longer.

743
01:43:46.990 --> 01:43:57.429
Lynne Jones: And yeah, the duty cycle does get to be a bit of a problem. We can't make it much… we can't make it… we certainly can't go shorter than 15 seconds. It's not obvious that we can go to 20 seconds.

744
01:43:57.740 --> 01:44:08.989
Lynne Jones: And so… you know, if the SEOC direct us to look at this again, we absolutely would.

745
01:44:10.120 --> 01:44:14.119
Lynne Jones: But we did look at it, and it was hard.

746
01:44:14.510 --> 01:44:16.140
Lynne Jones: Hard to make Ashley work.

747
01:44:19.470 --> 01:44:26.889
David Schlegel: But, I mean, I'll tell you, we're running this now with Dark Energy Camera on a big survey there. That's great. That's,

748
01:44:27.320 --> 01:44:36.170
David Schlegel: You know, so there we're getting a uniform survey. I, you know, I don't look forward to Rubin data, that's…

749
01:44:36.600 --> 01:44:37.900
David Schlegel: non-uniform.

750
01:44:43.220 --> 01:44:48.449
David Schlegel: Anyway, I think that's the obvious thing on the table here, and it would…

751
01:44:49.000 --> 01:44:51.129
David Schlegel: prove things for everyone, I think.

752
01:44:58.120 --> 01:44:58.750
Lynne Jones: May?

753
01:45:02.200 --> 01:45:04.750
Meg Schwamb: Sort of a related question,

754
01:45:04.850 --> 01:45:12.090
Meg Schwamb: There was a question about, why the Solar System Science Collaboration is less happy with going to 22nd

755
01:45:12.230 --> 01:45:16.830
Meg Schwamb: template generation observation, and one of the… sort of related to this is…

756
01:45:17.100 --> 01:45:23.709
Meg Schwamb: What is the quality that we're going to get? There's some concern that, especially as we go into the winter months in Chile.

757
01:45:23.840 --> 01:45:26.100
Meg Schwamb: And the fact that dome venting is not

758
01:45:26.240 --> 01:45:33.709
Meg Schwamb: where it should be? Will we have issues… will there be thermal issues with temperature? Will the seeing that was getting in April

759
01:45:34.160 --> 01:45:41.650
Meg Schwamb: be there over the summer, or in the northern summer, rather, in the winter in Chile. And we're concerned that those things are not the same.

760
01:45:41.820 --> 01:45:43.590
Meg Schwamb: And that, you know.

761
01:45:43.720 --> 01:45:52.349
Meg Schwamb: Yes, the scheduling, good scheduling, good seeing, but does anyone know what the performance of the facility is, such that

762
01:45:52.490 --> 01:45:58.700
Meg Schwamb: The… is the mirror thermalizing well to the outside temperature with only 12…

763
01:45:58.960 --> 01:46:12.519
Meg Schwamb: Louvers that are active. Are there other mitigations, or are all the heating, cooling, whatever, all those systems working at the level that that's not an issue for the next couple months, because…

764
01:46:12.920 --> 01:46:23.840
Meg Schwamb: there was bad seeing last year, and some of that must be some of those effects, because there was no dome venting at all. So that's one of our concerns, is that we… I don't know what the image performance is.

765
01:46:24.140 --> 01:46:28.740
Meg Schwamb: to know, or anyone, I think, in the community really knows, to know

766
01:46:28.930 --> 01:46:31.890
Meg Schwamb: If we cut to 20 seconds, and…

767
01:46:32.620 --> 01:46:41.180
Meg Schwamb: the, you know, things are worse. Will that ex… get going… will going to 30 help us enough to make, you know, a difference in year two?

768
01:46:41.640 --> 01:46:54.769
Meg Schwamb: rather than cutting to 20 seconds just to try to get template coverage, and then finding out that, well, we also got the C, you know, performance improved, and now we're at LSST in year 2, and so now we have templates that are even worse than we expected.

769
01:46:56.100 --> 01:47:04.760
Meg Schwamb: I'm wondering is how is all this going in together? Because I just kind of noticed over the past, you know, in the last few reports from

770
01:47:05.320 --> 01:47:16.939
Meg Schwamb: early operations that the seeing seems to be worse, and I would imagine that is something involving the worst weather, winter months, normalization, so how is all this going to be disentangled

771
01:47:17.070 --> 01:47:19.879
Meg Schwamb: Make that decision about going from 30-second

772
01:47:20.310 --> 01:47:29.330
Meg Schwamb: to 20 seconds, if DM allows it. I think… I don't know if that has already been studied enough that DM is happy with 20-second template observations.

773
01:47:29.700 --> 01:47:38.579
Meg Schwamb: I just want to raise if that's where our concern's coming from, because we don't hear that information to know how does that all interplay, or just 10 seconds not matter. I don't know.

774
01:47:39.460 --> 01:47:47.280
Lynne Jones: So, I think the crux of this, you're saying, is, like, what if… Regardless of the exposure time.

775
01:47:47.400 --> 01:47:51.420
Lynne Jones: because I think that… so it sounds like the question is.

776
01:47:51.850 --> 01:47:55.790
Lynne Jones: When we get images for templates out of year one.

777
01:47:56.030 --> 01:48:00.520
Lynne Jones: If the template saying is worse than

778
01:48:00.780 --> 01:48:03.869
Lynne Jones: What we actually achieve in year two.

779
01:48:04.130 --> 01:48:10.729
Lynne Jones: Is… how much of a problem? I think that's kind of where this ends up, because it…

780
01:48:11.080 --> 01:48:14.949
Lynne Jones: That part doesn't matter if it's 20-second or 30-second visits, right?

781
01:48:16.620 --> 01:48:17.290
Lynne Jones: Well, thank you.

782
01:48:17.290 --> 01:48:26.850
Meg Schwamb: more flux and more flux, right? We're seeing a little more flux. Does that help you? How much does it help you? I don't know.

783
01:48:26.850 --> 01:48:27.410
Lynne Jones: I don't know the performance.

784
01:48:27.410 --> 01:48:37.060
Meg Schwamb: The focal plane. I have yet to be… there's only a database that's available as of yesterday that shows me what the limiting magnitude was in the past month.

785
01:48:37.220 --> 01:48:46.170
Meg Schwamb: So I don't know. And so there's just concern of, we're making decisions on a system that is not 100% fair.

786
01:48:47.370 --> 01:48:53.629
Meg Schwamb: And if that 20, you know, does those 10 seconds actually matter? I don't know. But there's concern as it might.

787
01:48:54.550 --> 01:48:55.240
Lynne Jones: No.

788
01:48:55.350 --> 01:49:01.290
Meg Schwamb: Instead of going, whoa, but we take templates and good images and good seeing, so we'll get that back, and it'll be just as deep.

789
01:49:01.720 --> 01:49:03.980
Meg Schwamb: But that's an assuming good condition, right?

790
01:49:04.430 --> 01:49:11.670
Lynne Jones: Yeah, I think the answer is that the depth of the templates is somewhat secondary to the pixel coverage.

791
01:49:12.900 --> 01:49:21.040
Lynne Jones: And… not only the pixel coverage, but that you have a more continuous PSF, so…

792
01:49:22.830 --> 01:49:36.660
Lynne Jones: more epics helps them with making the template co-add have a nicer PSF that's more continuous over the sky, I believe. There's parts of this here that Eric Balum really is the one I should talk about, but

793
01:49:37.620 --> 01:49:42.479
Lynne Jones: But yes, the crux of it is, is that it's not the depth.

794
01:49:42.580 --> 01:49:50.790
Lynne Jones: so much in the template image that's the issue, it's the number of epics and the coverage on the sky, so the pixel coverage.

795
01:49:51.160 --> 01:50:00.140
Lynne Jones: Which is why we would say, okay, we could go to slightly smaller depth in each individual image. Now.

796
01:50:00.440 --> 01:50:04.040
Lynne Jones: I will say… I don't think that…

797
01:50:04.200 --> 01:50:07.160
Lynne Jones: The creation of the template from

798
01:50:07.660 --> 01:50:12.960
Lynne Jones: the 20-second images is the big stumbling block, and I don't see that, data management

799
01:50:13.350 --> 01:50:15.269
Lynne Jones: Putting up a complaint about that.

800
01:50:15.870 --> 01:50:28.480
Lynne Jones: I think the big stumbling blocks are going to be whether or not the atmosphere is really, stable enough in the 20-second exposure versus the 30-second exposure, whether the AOS systems are happy.

801
01:50:28.860 --> 01:50:33.659
Lynne Jones: Running with 20-second exposures and at that rate, rather than 30-second exposures.

802
01:50:33.890 --> 01:50:35.790
Lynne Jones: And,

803
01:50:36.270 --> 01:50:42.409
Lynne Jones: a little bit in U-band, we may not be able to reduce the exposure time anyway because of the read noise.

804
01:50:43.630 --> 01:50:51.620
Lynne Jones: Because the U-band is… gets read noise limited very, very fast, and so if we reduce the,

805
01:50:52.610 --> 01:51:01.369
Lynne Jones: If we… if we are read noise limited, the noise characteristics in the images are different, and apparently that prints through to the templates, so that might be bad.

806
01:51:02.040 --> 01:51:04.559
Lynne Jones: And maybe Keith has something to say here, too.

807
01:51:05.880 --> 01:51:11.920
Keith Bechtol: Yeah, there's another issue related to the active optics system, in that…

808
01:51:12.130 --> 01:51:15.470
Keith Bechtol: Right now, the time to compute the…

809
01:51:16.410 --> 01:51:30.210
Keith Bechtol: basically the characterization of the wavefront, and to provide the corrections. We're right on the cusp of being able to, apply those corrections with N plus 2 latency, because that time is around 30 seconds.

810
01:51:30.210 --> 01:51:40.250
Keith Bechtol: And so it's… we have just enough time, basically, to… to determine what those corrections are from the wavefront sensors, and then apply them to the… to the,

811
01:51:40.430 --> 01:51:56.099
Keith Bechtol: two visits later. If we drop down to 20 second exposures, then we will need, you know, substantial improvements in the speed of the AOS wavefront estimation algorithms to be able to achieve that N plus 2 latency.

812
01:51:56.320 --> 01:52:06.889
Keith Bechtol: Otherwise, we're running at N plus 3, and it just makes the, the control loop more challenging. There, there are…

813
01:52:07.070 --> 01:52:15.380
Keith Bechtol: Multiple approaches that we are pursuing for… to improve that control, but it… it… it does, it does add additional complications.

814
01:52:20.460 --> 01:52:25.810
Federica Bianco (she/her): That is also something that the data management system is evaluating in the request for comments.

815
01:52:33.040 --> 01:52:35.080
Meg Schwamb: Can I ask a follow-up question on this?

816
01:52:35.830 --> 01:52:40.400
Meg Schwamb: If the survey started in the next 2 months, our template's starting with 30 seconds.

817
01:52:41.260 --> 01:52:48.139
Meg Schwamb: Because people, like, if 5.3 is out, these are decisions that are happening now, right?

818
01:52:48.550 --> 01:52:49.050
Federica Bianco (she/her): the…

819
01:52:49.050 --> 01:52:58.249
Meg Schwamb: some of this is great discussion, but can… is there a plan to take 5.3 and use that while these decisions are happening? Because…

820
01:52:58.590 --> 01:53:03.449
Meg Schwamb: It sounds like this is not something that's gonna get decided today, or next week, or in two weeks.

821
01:53:03.620 --> 01:53:05.190
Meg Schwamb: This may be longer.

822
01:53:05.720 --> 01:53:10.539
Meg Schwamb: So is that the decision to go in with what is 5.3 as…

823
01:53:11.200 --> 01:53:17.809
Meg Schwamb: As the survey strategy, if the survey started in, say, the next 2 months, or 3 months?

824
01:53:18.660 --> 01:53:25.969
Lynne Jones: But I just wanted to say that the 20-second AOS stuff would be something we'd have to test on Sky.

825
01:53:26.090 --> 01:53:31.999
Lynne Jones: And that could be done fairly rapidly, I wouldn't think. Anyway, Fred has more comments on the philosophy.

826
01:53:32.000 --> 01:53:47.520
Federica Bianco (she/her): Yeah, as a reminder, I cannot speak for the decision, because I chair the SEOC, which makes a recommendation. The SEOC plans to recommend that we start with version 5.3, so with the templates, at the regular exposure of 30 seconds.

827
01:53:47.520 --> 01:53:52.780
Federica Bianco (she/her): And wait for feasibility assessments from the data management team.

828
01:53:52.780 --> 01:53:59.539
Federica Bianco (she/her): To see if all of these issues that we have discussed about taking 20-second exposures

829
01:53:59.540 --> 01:54:18.889
Federica Bianco (she/her): can be resolved. From a scientific perspective, what we have seen is that the majority of the feedback, now I hear the feedback from the SCOC, and this would also be considered, I hadn't heard before about the exposure of the depth of the individual exposures. All the rest of the feedback that we got was

830
01:54:18.950 --> 01:54:35.209
Federica Bianco (she/her): mostly positive. So, scientifically, we think it might be… there might be a gain, but not to start with it, because we don't have a feasibility study, and to keep an eye on this, and this may be one of the changes that we would implement throughout year one.

831
01:54:36.470 --> 01:54:37.320
Federica Bianco (she/her): Potentially.

832
01:54:50.210 --> 01:54:52.969
Lynne Jones: I think we should add to the…

833
01:54:53.380 --> 01:54:59.850
Lynne Jones: RSC the point that some of it might be in 30 seconds, and some of it might be in 20 seconds, and if

834
01:55:00.140 --> 01:55:05.570
Lynne Jones: I don't think that makes any difference, as far as creating the co-add, but we should make sure we explain that.

835
01:55:05.570 --> 01:55:11.919
Federica Bianco (she/her): Yeah, I think that's on the request for comments. That question was asked in the request for comments already.

836
01:55:14.440 --> 01:55:23.650
Phil Marshall (SLAC/Rubin): Yeah, and that was always going to be the way, right? Because there are many visits already taken with 30-second exposure times, so… so part of the deal, I think, was always going to be a mixed

837
01:55:23.830 --> 01:55:28.040
Phil Marshall (SLAC/Rubin): Mixed set of exposures. Or at least part of the opportunity.

838
01:55:28.260 --> 01:55:38.739
Federica Bianco (she/her): Right, and in the co-ads, you know, aside from the templates, in the data release co-ads, we would have those images, those exposures from the templates, and we would have regular exposures.

839
01:55:39.130 --> 01:55:44.289
Federica Bianco (she/her): Obviously, this would not be a viable solution if they couldn't be combined together.

840
01:55:44.630 --> 01:55:45.170
Phil Marshall (SLAC/Rubin): Yep.

841
01:55:54.150 --> 01:56:00.779
Lynne Jones: Well, we kind of went straight through the break, which is okay, because now we're not gonna come back from the break, I think.

842
01:56:01.070 --> 01:56:08.470
Lynne Jones: I think we've… Covered all of the things that people have questions about.

843
01:56:08.530 --> 01:56:27.310
Lynne Jones: We've accepted all of the feedback that we have so far, and I'm going to leave it up to Fed to say, you know, how… actually, sorry, I will say for her, because I know that the, you know, the liaisons are always open to feedback from the science collaborations, and then they bring this to the SEOC.

844
01:56:27.810 --> 01:56:31.120
Lynne Jones: And…

845
01:56:32.840 --> 01:56:38.299
Lynne Jones: My… yeah, and like Fez said, the recommendation is going to be to start the survey with

846
01:56:38.450 --> 01:56:40.530
Lynne Jones: the V5.3 baseline?

847
01:56:41.170 --> 01:56:44.240
Lynne Jones: And… Hopefully we start very soon.

848
01:56:44.860 --> 01:56:49.179
Lynne Jones: Phil, did you want to cap it off with anything, or Jocko? I see Jocko managed to just…

849
01:56:49.180 --> 01:57:05.660
Phil Marshall (SLAC/Rubin): I could start, and then if Joko wants to say something as well. I just wanted to say thank you to you all for, your patience waiting for the simulations, and then… and then, jumping on them when they came out to analyze them. I mean, the… the SCOC process at Rubin

850
01:57:05.750 --> 01:57:24.960
Phil Marshall (SLAC/Rubin): is a real strength, I think, that we have you… you here committed to helping making the survey strategy work well, providing that critical interface to the science collaborations in the community beyond that, and… and we've seen over the years that it does work well. I think we need to,

851
01:57:25.240 --> 01:57:38.740
Phil Marshall (SLAC/Rubin): keep hold of it, and keep working to try and continue having it work well. And so we can work on that from our side, the observatory, getting information out there, and, trying to allow

852
01:57:38.800 --> 01:57:47.960
Phil Marshall (SLAC/Rubin): time for analysis and deliberation, and I think we'll emerge with a better and continually improving survey as a result, so…

853
01:57:47.960 --> 01:58:04.479
Phil Marshall (SLAC/Rubin): Thank you very much. It's been really good to see you all here and contributing to such a good workshop. I will say thanks very much to Fed and Lynn for all their work organising it. Not easy to pull things off at short notice, but thank you very much.

854
01:58:04.850 --> 01:58:07.000
Phil Marshall (SLAC/Rubin): Joko, did you want to add something?

855
01:58:07.310 --> 01:58:11.929
Zeljko: I wanted to say something, but you said it all, so I have nothing to add.

856
01:58:13.410 --> 01:58:14.540
Zeljko: Thank you all.

857
01:58:15.480 --> 01:58:21.919
Lynne Jones: Yeah, thank you all for coming out, for your patience with the rapid turnaround at this point, and

858
01:58:23.210 --> 01:58:24.000
Lynne Jones: Yeah.

859
01:58:24.630 --> 01:58:26.800
Lynne Jones: Let's go start a survey soon,

860
01:58:27.770 --> 01:58:35.310
Lynne Jones: I thank you all for your time. We're canceled for the rest of today, no more, no more, SEOC workshop today.

861
01:58:35.640 --> 01:58:40.969
Lynne Jones: If you have slides to send to me, I'll add them to the,

862
01:58:41.110 --> 01:58:45.700
Lynne Jones: Slack pages and put a copy in the slides for today, if you like.

863
01:58:46.580 --> 01:58:47.490
Lynne Jones: Thanks.

864
01:58:48.190 --> 01:58:49.060
Phil Marshall (SLAC/Rubin): Thanks, Will.

865
01:58:49.280 --> 01:58:50.170
Andjelka Kovacevic: Thank you.