Rubin2020_PCW slack archives day2-tue-slot2g-rrb-survey-strategy-and-solar-system 2020-07-31---2020-08-11

Fri 2020-07-31 08:14PM
@Melissa Graham has joined the channel
Melissa Graham Fri 2020-07-31 08:14PM
@Melissa Graham set the channel purpose: Rubin Research Bytes Session G. Survey Strategy and Solar System Science
Melissa Graham Fri 2020-07-31 08:37PM
@Melissa Graham has renamed the channel from "day2-slot-2g-rrb-survey-strategy-and-solar-system" to "day2-tue-slot2g-rrb-survey-strategy-and-solar-system"
Melissa Graham Sat 2020-08-01 01:16PM
@Melissa Graham set the channel topic: Rubin Research Bytes Session G. Survey Strategy and Solar System Science
RRB Webpage: <https://project.lsst.org/meetings/rubin2020/agenda/session/rubin-research-bytes>
Mon 2020-08-03 05:12AM
@Laud Bentil has joined the channel
Mon 2020-08-03 11:39AM
@K-T Lim has joined the channel
Thu 2020-08-06 12:09AM
@Michele Bannister has joined the channel
Thu 2020-08-06 07:59PM
@Tim Lister has joined the channel
Sat 2020-08-08 02:33PM
@Michael Solontoi has joined the channel
Mon 2020-08-10 06:44AM
@quasar_panda has joined the channel
Mon 2020-08-10 01:11PM
@Jeyhan Kartaltepe has joined the channel
Mon 2020-08-10 04:47PM
@drphilmarshall has joined the channel
drphilmarshall Mon 2020-08-10 04:50PM
Session G. Survey Strategy and Solar System Science session page: https://project.lsst.org/meetings/rubin2020/agenda/session/rubin-research-bytes/sessg
drphilmarshall Mon 2020-08-10 04:51PM
For a sneak preview, here's the https://docs.google.com/presentation/d/1KKej_y5FOkSILMWWtcPsQ295NX4uI0z6aqkEqCqMed0/edit#slide=id.g8d57fbcff6_0_156 ! Looks like we're in for a real treat :-)
drphilmarshall Mon 2020-08-10 04:53PM
@Lynne Jones will be running the show in Session G, sharing screen, introducing speakers, advancing the slides, and keeping time.
Mon 2020-08-10 04:53PM
@Lynne Jones has joined the channel
drphilmarshall Mon 2020-08-10 04:56PM
I'll be monitoring this channel and the zoom chat for questions. Our speakers are:

David Trilling ( David Trilling ) Andjelka Kovacevic ( Andjelka ) Humna Awan ( Humna Awan ) Fabio Ragosta ( Fabio Ragosta )
We'll save all the questions until the end of the last talk, so you can help me out, and also increase your chances of an interaction with the speakers themselves, by mentioning the speaker's name in your question, please! And also, please only ask your question in one place - I'll hop back and forth between the zoom chat and here to try and catch all your Qs.
Mon 2020-08-10 04:56PM
@David Trilling has joined the channel
Mon 2020-08-10 04:56PM
@Andjelka has joined the channel
Mon 2020-08-10 04:56PM
@Humna Awan has joined the channel
Mon 2020-08-10 04:56PM
@Fabio Ragosta has joined the channel
drphilmarshall Mon 2020-08-10 05:10PM
Session G Program:

SNAPS: The Solar System Notification Alert Processing System, David Trilling Chasing the loudest gravitational waves sources: the LSST perspective, Andjelka Kovacevic Optimizing the LSST WFD for Dark Energy Science, Humna Awan Unknowns in the Rubin Observatory LSST era, Fabio Ragosta
Program starts Tuesday at 10:30am and then again on Tuesday at 11am, in the https://stanford.zoom.us/j/94581680305?pwd=R0ZYZm8zNmRLNlFySGZBd2xTZ1JLdz09 . Don't miss it!
Mon 2020-08-10 05:13PM
@Ed has joined the channel
Mon 2020-08-10 08:46PM
@Andrés Plazas (he/him/his) has joined the channel
Mon 2020-08-10 09:45PM
@Lucianne Walkowicz has joined the channel
Mon 2020-08-10 10:18PM
@Sukanya Chakrabarti has joined the channel
Tue 2020-08-11 01:57AM
@Rseaman has joined the channel
Tue 2020-08-11 09:31AM
@Will Clarkson has joined the channel
Andjelka Tue 2020-08-11 01:15PM
Hello I can not join the room
Andjelka Tue 2020-08-11 01:15PM
Is it open now
drphilmarshall Tue 2020-08-11 01:19PM
Yes! We are in the room and getting set up. Come on in, everyone :slightly_smiling_face:
Andjelka Tue 2020-08-11 01:20PM
I am out of the room now
drphilmarshall Tue 2020-08-11 01:20PM
Hmm. Me too.
Humna Awan Tue 2020-08-11 01:20PM
so im kicked out with a message that said that someone else started a meeting with the same ID, and this is what i currently see:
Lynne Jones Tue 2020-08-11 01:20PM
Ok - so we all got kicked out.
drphilmarshall Tue 2020-08-11 01:20PM
Let me try and restart it.
drphilmarshall Tue 2020-08-11 01:21PM
We're back!
Eric Gawiser Tue 2020-08-11 01:25PM
If troubles persist, see discussion in_ #help
drphilmarshall Tue 2020-08-11 01:30PM
Audience! Feel free to ask questions here (or on the zoom chat, eiether works) during the talks, and we'll try and get as many answered as possible in the last 10 mins.
drphilmarshall Tue 2020-08-11 01:35PM
Don't forget to @mention the speaker you are asking!
Tim Lister Tue 2020-08-11 01:38PM
@David Trilling what size DB/number of objects have you tested SNAPS and the periodogram determination with so far ?
Carlos Gonzalez-Fernandez (He/Him) Tue 2020-08-11 01:44PM
@Humna Awan would you gain anything by adding wide band NIR data on your survey area (either helping with MW extinction or adding data for photo-z)
Sukanya Chakrabarti Tue 2020-08-11 01:46PM
@Fabio Ragosta to what accuracy will the proper motions be measured?
Michelle Lochner (she/her) Tue 2020-08-11 01:47PM
:clap: to all speakers!
Gareth Francis Tue 2020-08-11 01:47PM
@David Trilling What technology/technologies are you anticipating using to publilsh outgoing alerts?
Melissa Brucker (she/her) Tue 2020-08-11 01:50PM
is there a current ZTF alert system, what tech does it use?
Tim Lister Tue 2020-08-11 01:51PM
It uses Kafka, it's approx. 10% scale model of LSST. Patterson et al 2019 has the details ( https://arxiv.org/pdf/1902.02227.pdf )
Melissa Brucker (she/her) Tue 2020-08-11 01:53PM
thanks
Meg Schwamb Tue 2020-08-11 01:56PM
This doesn't seem public yet though. The lasair broker is currently ingesting LSST Solar System alerts if you're looking for something now https://lasair.roe.ac.uk/objlist/#
Stephen Smartt Tue 2020-08-11 01:56PM
@David Trilling - followup to @Gareth Francis question. Is there a strong reason you are chosing Kafka as the outgoing stream ? You noted it's not yet built and you are open, but other than the obvious reason that you want to keep it similar format to incoming (so people are familiar) are there other reasons ? We have the same issue for stationary transients, In what format do users want to interact with value added data (such as yours). Is kafka really the most useful for users ?
Tim Lister Tue 2020-08-11 01:59PM
@Meg Schwamb I think that's true. The alerts from the public part of the ZTF survey are going to ANTARES but unknown asteroids are not going there IIRC when I discussed this with @Michael Kelley
Meg Schwamb Tue 2020-08-11 02:00PM
I meant that the output from SNAPS is not public right now.
Meg Schwamb Tue 2020-08-11 02:00PM
Or at least that's what I gathered from the talk.
Tim Lister Tue 2020-08-11 02:01PM
Sorry. misunderstood which alert producer you meant. @David Trilling did say the alert part of SNAPS had not been built yet
Melissa Graham Tue 2020-08-11 01:49PM
@here new transition time 11:15am due to technical difficulties
Carlos Gonzalez-Fernandez (He/Him) Tue 2020-08-11 01:54PM
Thanks for the answer!
Carlos Gonzalez-Fernandez (He/Him) Tue 2020-08-11 01:54PM
We are working on pixel-matching VISTA data (in particular VHS survey) that should overlap nicely with your footprint
Carlos Gonzalez-Fernandez (He/Him) Tue 2020-08-11 01:55PM
that is mostly J and Ks filters
Rseaman Tue 2020-08-11 01:58PM
Regarding SNAPS, an atypical point may result from a false linkage as well as an active asteroid, etc. How will these be distinguished?
drphilmarshall Tue 2020-08-11 01:59PM
https://rubin2020pcw.slack.com/archives/C017VF7MU3A for info about RRB session zoom rooms - many have changed!
drphilmarshall Tue 2020-08-11 01:59PM
(Thats #day2-tue-slot2-rrb-general )
Michelle Lochner (she/her) Tue 2020-08-11 02:04PM
@Andjelka recent rolling cadence simulations include some amount of time that continues to cover the rest of the sky so no part of the sky will be completely unobserved for such a long time. Will this help improve detection?
Andjelka Tue 2020-08-11 03:04PM
Hi Michelle thank you for input. It would be great to have cadence as fine as it is possible :+1:
Lynne Jones Tue 2020-08-11 03:36PM
Is it important to consider the time between visits in the same filter or would visits in different filters be ok?
Andjelka Tue 2020-08-11 04:17PM
Hi @Lynne Jones thank you. Different filters will produce different light curves, simultaneous filter alternation would produce sparse curve per band.
Lynne Jones Tue 2020-08-11 04:18PM
Ok - so the interval between visits should be counted in the same bandpass, it sounds like it (unless you can calibrate between bands using colors .. but my understanding is maybe the colors change too?)
Gareth Francis Tue 2020-08-11 02:14PM
Thanks again to all the speakers :clap:
Hiranya Peiris Tue 2020-08-11 02:18PM
A general question following up on current discussion to all speakers - how easy has it been for you to utilise the existing survey simulations from the Project to do detailed assessments of the impact of observing strategy on your science cases? Is there something that can be done to help improve the usefulness or accessibility of these products, especially if you are not part of a big collaboration like DESC?
drphilmarshall Tue 2020-08-11 02:20PM
Great question for all of @David Trilling @Andjelka @Humna Awan @Fabio Ragosta (and I am sure @Lynne Jones will be interested in the answer!)
Meg Schwamb Tue 2020-08-11 02:21PM
I think that's a great question for tomorrow's https://project.lsst.org/meetings/rubin2020/agenda/session/community-evaluation-rubin-survey-strategies session. Each of the SCs have been asked to put together slides on this point/current status
Hiranya Peiris Tue 2020-08-11 02:22PM
@Meg Schwamb sadly can't make it (conflicting session) but I am interested in the answer particularly for people/groups who may not be working as part of an SC yet. I will watch the recording though!
Meg Schwamb Tue 2020-08-11 02:23PM
Definitely something for us to think about re the SCOC
Fabio Ragosta Tue 2020-08-11 02:57PM
From my experience I can say that It is not too hard to access the existing survey simulation, It is quite straightforward. The example notebooks on the lsst community worked well for me. I found a little bit harder access the product of the metric after you ran the simulation.
Andjelka Tue 2020-08-11 03:20PM
Thank you Hiranya. We used simulated HORIZON-AGN.fits photometric catalogues below redshift 4 with LSST-like filters.
Hiranya Peiris Tue 2020-08-11 03:23PM
@Andjelka thanks - I was wondering if you built on top of OpSim simulations? Is there anything that could be done to make them more accessible or easier to use?
Andjelka Tue 2020-08-11 03:26PM
If OpSim can produce photometric catalogues like HorizonAGN or to use them it would be great :slightly_smiling_face:
Hiranya Peiris Tue 2020-08-11 03:28PM
ok - I think that's unlikely to happen? - but I could not work out if you were taking an OpSim simulation and building a mock catalogue on top of that or not (i.e. I could not understand if your "rolling cadence" was an actual OpSim simulation or not).
Andjelka Tue 2020-08-11 03:38PM
In our analysis we used normal cadence, we do not propose rolling cadence . Instead of OpSim we used HORIZON simulation mock photometric catalogues of AGN with LSST filters https://www.horizon-simulation.org/data.html
Hiranya Peiris Tue 2020-08-11 03:39PM
OK, clearly some kind of communications barrier here - thanks for your answers anyway!
Andjelka Tue 2020-08-11 04:02PM
Hi @Hiranya Peiris Apologies for unclear answer, It is edited now. Hope it is better. We are happy to have zoom chat anytime :slightly_smiling_face: :+1:
Lynne Jones Tue 2020-08-11 04:22PM
@Andjelka Looking at the HORIZON simulation catalogs online (without downloading them though) I can't tell if they're producing time series photometry or just a single magnitude per bandpass for each object. When you add what you call "normal cadence", are you adding the lightcurve information or does that come from HORIZON? If you're adding it, how are you doing that?
Andjelka Tue 2020-08-11 04:55PM
@Lynne Jones HORIZON does not produce time series but produces spectra, magnitudes and masses of objects for different redshift, these data are also helpful for estimates of AGN binaries which can be observed. Work with these data is in progress and not presented in our talk. In our talk we presented normal cadences based on our long term observed light curves of AGN. :slightly_smiling_face:
Lynne Jones Tue 2020-08-11 05:08PM
I am a bit worried that the 'normal cadences' you're referring to may not match the distribution of visits in particular filters .. there may be some complications beyond simply rolling / not-rolling.
It seems like, however, that there may be useful approximations in looking at the interval between visits in the same filter, perhaps within each observing season. Differences between simulations where we vary the filter distribution or how filters are observed throughout the month would show up, as would the varying cadences in rolling vs. non-rolling. Given that it seems like you have approximations about intervals of X number of days leads to Y percent error in identifying binary fraction.
What happens if the interval between visits changes over time?
Andjelka Tue 2020-08-11 05:31PM
@Lynne Jones yes that is great point.Thanks! Perhaps changing visits would produce non homogenous density of data such as in the light curve PG1302-102 (Liu et al. 2018 https://arxiv.org/abs/1803.05448 ) . Because of this machine learning methods could help in overcoming these and similar issues (see e.g. our Kovacevic at al. 2019, https://ui.adsabs.harvard.edu/link_gateway/2019ApJ...871...32K/doi:10.3847/1538-4357/aaf731 )
Will Clarkson Tue 2020-08-11 02:42PM
Comment on @Humna Awan 's question about feedback between the Science collaborations on the dust-avoidance candidate strategy: while I am not aware of all that much direct contact regarding the various tradeoffs (e.g. with greater number of observations towards the inner Plane), I think the more interaction between the SC's on this issue, the better. That, and once we get all the metrics coded and evaluated, we'll be in a better position to quantitatively address the tradeoffs for our (with my SMWLV hat on) science.
Lynne Jones Tue 2020-08-11 02:47PM
Indeed!
Lynne Jones Tue 2020-08-11 02:49PM
Here's my view of the metrics related to the footprint .. (each panel is shaded by general 'kind' of footprint .. misty rose = traditional baseline style, with variation, green = newer 'big sky' style variations without additional regions for the plane, etc., salmon = big sky + additional regions for the plane/bulge/north).
Lynne Jones Tue 2020-08-11 02:50PM
The metrics are grouped approximately by panel ... TVS is currently standing in for a lot.
Lynne Jones Tue 2020-08-11 02:51PM
What I see there is that the SRD metrics (top panel) like the old baseline footprint (because we spend more visits on the WFD, and less on the 'rest of the sky'). The big sky+ variations have lower SRD metric values, but they still pass requirements.
Lynne Jones Tue 2020-08-11 02:51PM
The solar system metrics (bottom panel) like big sky+ metrics, although it's primarily an issue for TNOs .. this comes down to sky coverage.
Lynne Jones Tue 2020-08-11 02:52PM
The DESC metrics (near middle) are actually a bit mixed .. the area with low dust extinction and depth over that area is better with big sky. But the number of visits in the WFD area is then lower, so the weak lensing is a bit less happy. Overall, I suspect this is a win.
Lynne Jones Tue 2020-08-11 02:53PM
The TVS metrics, in particular metrics related to microlensing which depend on bulge coverage, are happy with footprints with lots of bulge coverage. (the bulges_bs footprint has big sky+ but lower levels of bulge coverage, more like the current standard baseline).
Lynne Jones Tue 2020-08-11 02:54PM
The number of galaxies across the entire sky (our current standin for all metrics for 'galaxies') likes more area coverage, so big sky+ does better.
David Trilling Tue 2020-08-11 02:44PM
thank you @Lynne Jones and @drphilmarshall for running these sessions.
drphilmarshall Tue 2020-08-11 02:46PM
You are very welcome! Excellent talks and questions, thanks very much everyone who participated live. Do please feel free to use this channel to follow up with the speakers or each other - and enjoy the rest of the PCW!