Closeout and breakout session reports���LSST2018 Project and Community Workshop�August 17, 2018

​

http://ls.st/l3p

​

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

1

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Stack Club (Mon 1:30pm)

Message box example for short highlighted messages.

​

• The Stack Club is an emerging online community of LSST science collaboration members committed to producing high quality Jupyter notebook tutorials as a way of learning about the LSST software pipelines.

You don’t really understand something until you have to teach it

• We meet weekly to show what we’ve made, get feedback and help, work together, and learn from each other. We have 40 accounts on the splendid new LSST Science Platform for club members to use, thanks to LSSTC and the DM team

​

• Slides from Monday’s session: http://ls.st/xn1 (video here, starts 43m in)
• Hang out with us on Slack: #stack-club
• Browse our notebooks and projects at https://github.com/LSSTScienceCollaborations/StackClub

Ready to commit? Fill out the Stack Club Membership Application Form!

​

2

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Summit Base ITC Design (Mon 1:30pm)

• Covered
• Contents of LSE-309 Kian-Tat Lim (Summit, Base networks Luis Corral)
• Contents of LSE-239 Kian-Tat Lim
• Contents of LSE-299, Test and V&V plans Jeff Kantor
• Contents of LSE-78 Ron Lambert (Jeff Kantor will present)
• IT Team and IT Related Meetings Jeff Kantor
• DevOps, Infrastructure configuration, service delivery requirements Andres Villalobos
• Attendance very light, appears that all subsystems being represented on Tiger Team means they feel covered
• Relying on on Tiger Team members to communicate requirements from above documents as necessary
• Anyone sending servers, networked equipment to Chile needs to be familiar with LSE-309 in particular

3

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Software Pathfinders and Integration (Mon 1:30pm)

• Discussed schedule of upcoming EIA (pathfinder) exercises
• Discussed opportunities for new exercises (AuxTel)
• Robustness testing
• Fault state studies
• Discussed leveraging NCSA L1 Test Stand for automated testing
• Discussed ideas on how to tighten up software deployment

4

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

M1M3 Thrm Ctrl Sys - FCU Test Status (Mon 1:30pm)

• Presented overview of M1M3 mirror thermal control system
• 96 Fan Coil Units (FCU) deliver conditioned air to 1650 mirror cores to cool the M1M3 mirror
• Presented status of functional testing of the individual FCUs
• This is not a system level test of the thermal control system
• Test setup exists and testing has begun but…
• SW development effort to automate some of the tests is expected to be ready in ~1 week.
• Thermal Inner Loop Controllers (ILC) will be used to test FCUs and are expected to be ready in ~1 week
• Discussion of concerns about the temperature measurement and control strategy and have agreed on a clear path forward

​

5

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

TUESDAY

6

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Scheduler/Survey Strategy (Tue 11am)

• Discussed plan for scheduler development
• Discussed scheduler operation modes
• Adoption of new Feature Based Scheduler
• Discussed use-cases for Scheduler during commissioning
• AuxTel
• ComCam

​

​

​

​

7

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Sensor Characterization, Analysis & Sim (Tue 11am)

A session focused on sensor work and communications between the Camera, DM and SE/Simulations teams.�

• Overview of the focal plane construction status with a focus on sensors - Steve R.
• Summary of the various Camera information expected from I&T - Aaron
• Sensor & Readout effects - Pierre A
• Overview of the signature removal plan/status in DM - Merlin
• Documenting/validation of imSim sensor models - Serge
• B-F effect correction as in the Coulton, et.al. paper - Craig
• LSST sensor characterization study updates - Bela / Emily
• Discussion - How can we effectively work together and transfer information between the camera, DM and simulation teams efficiently

8

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Telescope & Site S/W Testing (Tue 11am)

• Introduced the TSSW Testing team
• Discussed testing methods and strategies
• Examined two tools used to test software
• Robot-Framework - automated testing tool
• Jenkins - continuous integration
• Deep dive on reading and analyzing test results
• Presented current status and future goals
• Requested more feedback and interaction

Software testing: providing data used to assess the quality of a software product

9

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Verification Software Tooling (Tue 11 am)

• Provided an overview of the Verification Architecture.
• Provided a live demo of the tooling and worked with the tooling live in the commissioning sessions.
• Took action item from feedback to make it easier to see the details of the system requirements and their traceability in Jira.
• Action item has been closed. Four new fields have been added to the Verification issues in Jira from the Requirement in MagicDraw.
• Requirement Discussion- This is the additional information to further clarify the Requirement Text.
• Requirement Parameters- This field contains the definition(s) of the parameter(s) that are referenced in the Requirement Text field.
• Higher Level Requirement- This defines any requirement that is higher in the specification tree that has been flowed down or allocated to this requirement.
• Lower Level Requirement- This defines any requirements lower in the specification tree that this requirement is flowed down or allocated to.

10

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Administrative Optimization (N & S)

• Found areas of improvement and developed plans of action for differences in administrative processes
• Reimbursements
• Licenses: procurement and renewals
• Travel to Chile and the Summit
• Communication, compliance, calendaring
• Admin vision and mission statements�

11

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Observing Strategy White Paper Workshop (Tue 1:30pm)

• Call for White papers:
• http://ls.st/c66
• Deadline November 30, 2018
• Please use LSST Community (community.lsst.org) to talk to us and each other!
• http://ls.st/m6u

​

​

​

Released 11 simulated survey strategies with white paper; since added 4 new runs.

Will add a few more before November, and move to Feature Based Scheduler.

12

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

ComCam Planning and Scheduling (Tue 1:30pm)

• Brief overview, status, and schedule
• Discussion on what is needed to get Tucson integration area up and running
• Discussion on boundary condition on when to ship to Chile
• Discussion on test plan in Tucson
• ComCam team answered questions from the community

​

13

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Science Communication & Media Training (Tue 1:30pm)

• Give great presentations and interviews
• Really important to have a big story
• Story/analogy to explain complicated concepts
• Press: engaging story, emotion, pictures
• Record and watch yourself to eliminate unwanted habits
• It’s ok if you don’t have all the answers
• PRACTICE, PRACTICE and PRACTICE - out loud!

​

14

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Cadences, DDF Locations, Mini-Surveys.. (Tue 3:30pm)

• Heard from a number of studies on new observing strategies, cadences
• Wide range of impact from different strategies already released
• Looking at new strategies that vary visit pairs, filter changes
• May be solutions that are compromises between different science areas
• Certain visit-pairs near galactic plane, others for not near
• Rolling cadence for certain years..

​

Message box example for short highlighted messages.

15

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Assembly, Integration and Verification (Tue 3:30pm)

• Overview of Telescope and Site AIV:
• Organization of the Team in Chile
• Schedule Driven by Deliveries
• AIV Test Plan Build Structure
• Tools to Measure Key Parameters
• Open Discussion:
• Safety is paramount

​

Coating Chamber Arrival in Belgium on its way to Chile

16

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

CMMS tool discussion (Tue 3:30pm)

Message box example for short highlighted messages.

17

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

WEDNESDAY

18

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Cadence Options 1 - Visits and Area

(Wed 11am)

19

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

PSF Estimation (Wed 11am)

• Four talks:
• Sensor effects on PSF ellipticity and mitigation through observing strategy (Tyson)
• Double Zernike and Nodal Aberration Theory (Bauman)
• PSF modeling for DES (Roodman)
• PSF modeling for HSC (Meyers)
• Valuable to have PSF experts from DES, HSC, LSST all in the same room.
• Discussion of relative merits of modeling choices.
• Inspired many followup conversations later in the week.

​

20

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Science Verification Planning (Tues 1:30am & 2:30)

Commissioning Verification Planning (Wed 11am & 1:30pm)

Develop verification procedures for a few specific requirements from the LSR (http://ls.st/lse-29) and OSS (http://ls.st/lse-30) as a means to:

(1) educate members of LSST Project System Engineering and Commissioning teams on the use of of tools comprising the LSST verification architecture (including our adopted best practices and conventions);

(2) better understand and clarify the scope of detailed verification planning -- including required input data, procedures, and post processing -- necessary to satisfy verification requirements; and

(3) identify personnel roles and locations where the specific verification activities will be conducted.

​

21

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Science Verification Planning (Tues 1:30am & 2:30)

Commissioning Verification Planning (Wed 11am & 1:30pm)

• 20-30 active participants in each session
• Reviewed resources for how to define verification and tests cases
• Action items for improving the workflow and process (e.g. traceability in verification JIRA tickets)
• Reviewed methodology, architecture, terminology and nomenclature for verification
• Afternoon Hack Sessions had ~5 groups working in real time developing verification definitions and test cases on several selected requirements:

OSS-REQ-0219: Wavefront Sensing on Sky Efficiency OSS-REQ-0222: Ghost Image Control

OSS-REQ-0237: Filter Out of Band Constraints OSS-REQ-0247: Total Optical Throughput

OSS-REQ-0334: Flat Fielding Allocations OSS-REQ-0298: Absolute Pointing

OSS-REQ-0276: Calibration of the Atmospheric Transmission LSR-REQ-008: Image Degradation over FOV

LSR-REQ-0090: Filter Depths LSR-REQ-0014: Collection of Science Data as Visits

• Development was captured in verification JIRA tickets and test cases.

22

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Communications Office update (Wed 11am)

• Good attendance (25+ in the room including ZI, DMcF)
• Partner presentations online: LSSTPO, SLAC, IN2P3, NCSA
• Described individual structure and partner role with project
• Discussion to identify issues and ways to leverage/coordinate:
• Camera itself is distributed; shared repository for story of LSST; concerns about branding; best ways for COMM to communicate
• Identified five session outcomes:
• Annual face-to-face meeting
• Begin quarterly virtual meetings
• Set up Slack channel
• Share milestone plans
• Coordinate conference presence (SC18)
• ​

23

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Pipeline Development and Cadence Optimization for Solar System Science (Wed 11am)

• Overview of LSST Solar System Processing
• Review of data products and timescales
• Current plan is to work with Minor Planet Center to compute new/updated orbits from LSST-reported tracklets (and adopt full MPC catalog as the basis for the Daily Orbit Catalog)
• Details of LSST-MPC plan still under discussion; more work needed�
• Discussion about Opsim, MAF, and cadence evaluation
• Showed results of impacts of various cadences on discovery rates for different solar system populations
• Opsim runs with rolling cadences need more�work and further evaluation

​

Time is now to talk to MOPS and MPC teams about special needs!

24

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Cadence Options 2 –Strategy

(Wed 1:30pm)

25

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

EPO Open House (Wed 1:30pm)

• What would you like to know about LSST EPO? How would you like to get involved? pollev.com/lsst results:
• Cloud technology opportunities. Discussion: investigate free options, such as education grants, free research credits, and public dataset hosting.
• More info on formal education investigations:
• We will leverage JupyterCon to influence change and wait for 1.0 GA release to address styling challenges to creating EPO interface.
• Formal Ed investigations have two versions: ready-to-go and widget only + editable doc. Discussion: code-free UI is less error prone and potentially less costly.
• Coloring the Universe demo: https://lsst-epo.github.io/properties-of-light/
• Design: very important! on roadmap and part of ongoing branding discussion
• Group interest in promoting the engineering aspects of the telescope and staff (STEM careers)

26

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Unconference (Wed 3:30pm)

• Low Surface Brightness Observations

Highlighted LSB. Heard about scattered light tests. Thoughts on image detection. How to pass ideas to DM. Session notes uploaded to agenda so public.

• Instrument Signature Removal Validation

We met and discussed the status of ISR in the stack: the corrections and the creation of input data products from raw data. We made progress in planning how to effectively improve knowledge transfer between the various teams and people involved in camera construction, integration, testing, and commissioning. (Merlin)

• Cadence White Paper planning (Community post)
• Cadence for Supernova Physics Studies

We discussed light curve sampling for explosion physics such as shock break-out, and how the DESC and TVS might collaborate on their cadence white papers.

• South Pole Observations
• Commissioning Observations

Thank you to all session organizers and participants!

https://community.lsst.org/t/lets-coordinate-observing-cadence-white-papers/3144

27

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Short time assignment in Chile (Wed 3:30pm)

• Logistics of working in Chile discussion
• Business trips
• Short-term assignment
• Long-term assignment
• Relocation
• Info on lodging, transport, visas, per diem, medical
• Debut of new resources developed since 2017
• Document-28242
• Webpages: https://project.lsst.org/services/working-chile�

​

28

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

THURSDAY

29

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Interfacing with the Prompt Products Database

(Thu 11am)

• The PPDB is a key interface in the DM system: database reads & writes during real-time pipeline execution.
• Goal: identify and plan to resolve perf. and interface issues.
• Presentations:
• Eric Bellm: Intro to PPDB
• Andy Salnikov: Prototype database�design & performance
• Chris Morrison: Pipeline design
• Discussion:
• Science requirements for out-of-order processing
• Pre-loading database results to improve performance
• API refinement

30

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Visualization Tools for Analysis and Commissioning

(Thu 11am)

We learned about the pyvis ecosystem and saw LSST specific tools like Firefly.

We also saw demos from the audience of relevant tools from DESC/DES/ZTF etc.

31

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Preparing to Do Solar System Science with LSST

(Thu 11am)

• Overview of Solar System Science Collaboration (SSSC)
• Science roadmap published on arXiv
• Next steps include working on cadence white papers and software development roadmap/details
• Particularly want to characterize significance of North Ecliptic Spur

​

• Overview of solar system science tool development
• Want to identify common needs to save redundant effort and free time to work on more interesting science
• Scientist funding is an issue, both to support�time spent on LSST prep and travel to PCWs

​

Key priorities: cadence white papers, finalizing new MOPS schema, software white papers, funding proposals

32

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Perfect Pixels. Accurate Astrophysics. Correct Cosmology.

(Thu 1:30pm)

A summary and lively discussion of topics from an Aspen workshop earlier this summer:

• Coadds (Jim Bosch): properties of different types of coadds, and the impact on various LSST science cases
• Theoretical uncertainties (Tim Eifler): making progress on controlling their impact on cosmology, separately and jointly
• Tension metrics (Eric Huff): how do we know whether our experiments agree with each other, and what can we learn from these tension metrics?

Message box example for short highlighted messages.

33

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Plans and Policies for LSST Alert Distribution

(Thu 1:30pm)

Presented an overview of the LSST alert stream:

a powerful real-time view of the dynamic sky, which

requires downstream tools to identify objects of interest.

​

Discussed capabilities and limitations of the LSST mini-broker.

​

Described draft policies to guide selection of community brokers, to be released later this year.

Message box example for short highlighted messages.

34

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Astrophysical Probes of Dark Matter with LSST

(Thu 3:30pm)

• Guidelines
• Delete notes and save this template as your presentation
• No drop shadows
• Only use the font Calibri or Carlito
• Please do not alter template

Message box example for short highlighted messages.

35

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Upcoming DM Middleware (Thu 3:30pm)

• DM has been developing big upgrades to the way we define pipelines (“PipelineTask”) and to the data access abstraction (the “Butler”) over the last ~year.
• This session gave an overview to pipeline developers and interested others about the changes that are soon going to become available.
• Lots of excitement in the room�about the new features and�better interfaces!
• Thanks to Jim Bosch & Nate�Lust for presenting & demoing.

36

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Active Optics Validation using Simulation (Thu 3:30pm)

Goal of the breakout: Coordinate work between construction of the active optics system, its validation of the active optics system, and the optimization of the algorithms needed to achieve the image quality required by the SRD

If interested to join the effort email sthomas@lsst.org

• Template definition for donut fitting
• Noise influence
• Optimization

​

Definition of best strategy to include history

Validation and optimization tool = Phosim / Imsim Others?

​

37

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Following-up LSST Alerts (Thu 3:30pm)

Presentations available https://goo.gl/YVadHR

Talks by Soraisam on NOAO’s ANTARES, Graham on the TVS broker survey.

Talk on TOMs/ toolkit by Bianco

ZTF example: talk by Ridgway

Talk on AURA/LCOgt collaboration by Miller

38

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

DAQ system delivery to NCSA (Thu 3:30pm)

• How to manage the DAQ system & its corresponding infrastructure (the NCSA Teststand) as a project wide resource?
• Short Presentations by:
• M. Huffer (Overview of the DAQ System)
• J. Parsons (Overview of the corresponding NCSA infrastructure)
• M. Reuter (Use cases)
• Resulting discussion:
• System still on track for November delivery
• Highlighted importance of Image emulation
• Agreed usage is a natural extension of current path-finding exercises
• Agreed systems engineering will oversee its development + usage

39

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

FRIDAY

40

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

T&S Subsystem Verification Status (Fri 9am)

• Guidelines
• Delete notes and save this template as your presentation
• No drop shadows
• Only use the font Calibri or Carlito
• Please do not alter template

Message box example for short highlighted messages.

41

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Time-Domain Spectroscopic Observatory (Fri 9am)

• Guidelines
• Delete notes and save this template as your presentation
• No drop shadows
• Only use the font Calibri or Carlito
• Please do not alter template

Message box example for short highlighted messages.

42

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Commissioning Review Action Items Discussion (Fri 9am)

Commissioning team met to review comments and recommendations from the recent “CD-2 like” review.

​

Internal action items from the comments were discussed.

​

Formal responses to recommendations were also discussed.

43

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

LSST Code of Conduct (Fri 9am)

How do we ensure LSST is creating an inclusive, collaborative environment?

What documents exists - high level summary

Activity to solicit feedback regarding CoC’s

Recap of what was captured in the discussion

A next-steps process that will include review by an appropriate subgroup and communication of final decisions

Join #lsst2018coc to follow progress

​

44

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

DM Hack & LSST Stack (Mon -Fri)

• A chance for DM developers (& interested others) to work together to address cross-team issues.
• Lots of fun & useful projects. Some highlights:
• Removing support for Python 2 & upgrades to support Python 3 in the LSST “software stack” (Russell Owen & Tim Jenness)
• Deblending on DcrCoadd subfilters (Ian Sullivan)
• New features for the PipelineTask preflight solver (Andy Salnikov)
• Developed a demonstration notebook looking at calibration images in the full focal plane; aim to use this with Camera test-stand data at NCSA (Michael Wood-Vasey & Robert Lupton)
• Converting old documentation to new (Numpydoc) standards (David Staker)
• ...many other things that I forgot.

45

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

46

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Special Programs

47

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

LSSTC Summer Intern Program (Mon - Tue)

Let’s turn this pilot program into a tradition!

( Fundraising challenge

opportunity)

Mirror Lab Tour with Buddy Martin

Poster Pitches

• Supported by LSST Corp (~$1200/student for 2 days / 3 nights).
• Informal feedback extremely positive.

• Career-advising breakfasts & dinner.
• Two student-focused sessions.

48

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

StoryTime Domain (Tue-Thu)

​

​

​

We interviewed 21 people!

​

Thank you to everyone who participated.

​

49

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

HISTORY of Workshop on Impact of Blending

on LSST Science (Tue-Thu)

• A bit of history: how did blending workshop end up at LSST2018?
• As follow-up to blending discussions at LSST2017, Sarah Brough (Galaxies) was put in touch with Pat Burchat & David Kirkby (DESC Blending Task Force); Alexie Leauthaud and others @ UCSC also expressed interest.
• We formed a multi-science-collaboration SOC and applied for LSSTC support:
• granted $25k; used for providing partial travel support for 14 grad students & postdocs and 10 senior scientists who needed support to attend.
• Highest availability during LSST2018 so approached Beth Willman and Ranpal Gill.
• Did it work as a joint workshop? Yes!
• Very relevant DM & Scarlet expertise was available when needed (e.g., Jim Bosch, Peter Melchior, Fred Moolekamp).
• Broad range of SCs were represented.
• Brought people to LSST2018 who would otherwise likely not

have attended: LSSTC support was extremely important for this!

P&CW is a great space for cross-SC/DM engagement

50

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

GOALS of Workshop on Impact of Blending

on LSST Science

From the Workshop page:

The workshop will focus expertise across the LSST community on assessing our current understanding of blending issues and addressing the following questions:

• What are the figures of merit in quantifying the impacts of blending?
• What tools and data products exist to evaluate these metrics -- e.g., simulations, data sets, software frameworks?
• Which tools and data products are missing?
• What are the requirements on these metrics for different science cases?

Message box example for short highlighted messages.

51

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Why a Workshop on Impact of Blending

on LSST Science?

Blending affects all science collaborations!

52

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

OUTCOMES of Workshop on Impact of Blending

on LSST Science

Message box example for short highlighted messages.

• Scribe notes for every one of 8 sessions => 17 pages to be synthesized!!

​

• Tutorial session:
• Participants were given two Jupyter notebook tutorials, running in the LSST Science Platform Notebook Aspect (https://nb.lsst.io/).
• How to use multi-band deblender Scarlet as a single package.
• Walk-through of source detection, deblending with Scarlet, and running measurements in the LSST stack.
• Tutorial will live on the Stack Club github page:

https://github.com/LSSTScienceCollaborations/StackClub

• Many thanks to Adam Thornton and SQuaRE for setting up the notebook environment and helping during the session!

​

53

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

OUTCOMES of Workshop on Impact of Blending

on LSST Science

Session #8:

Small group breakouts.

​

​

Message box example for short highlighted messages.

54

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

OUTCOMES of Workshop on Impact of Blending

on LSST Science

Strong desire for cross science collaboration communication & sharing of already available resources:

• Web area to document existence of tools, data sets,...
• use LSSTC-sponsored SC pages (under redesign)?
• Communication options: new #lsst-blending slack channel? LSST Community?
• Presentations at DESC Blending Task Force meetings (with permission of presenters).

Lots of desire for working across science collaborations!

• Tools for generating pixel-level simulations of isolated or blended objects. Examples:
• GalSim & WeakLensingDeblending open, well documented, widely used in DESC.
• Blending Tool Kit (under development).
• Public data sets with catalogs (e.g., HST images).

55

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Why a Workshop on Impact of Blending

on LSST Science ?

• Space-based “truth” is limited.

=> Curate and share catalogs across science collaborations.

Data useful to many Science Collaborations

HST Frontier Fields

56

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

OUTCOMES of Workshop on Impact of Blending

on LSST Science

Shared training & test data sets: “Wish List” examples

• Simulated postage stamps with truth (including seg maps for each object) with or without noise, for training (e.g., ML) and testing algorithms
• Simulated objects “injected” into HSC images and run through DM processing?
• DESC Data Challenge 3 (being specified during next ~year) - bigger volume, improved realism, specific science questions?

Shared near term goals across different science collaborations?

• Can make sure they are consistent with DM schedule for

Scarlet development, object injection, etc.

Lots of desire for working across science collaborations!

57

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

OUTCOMES of Workshop on Impact of Blending

on LSST Science

Injecting targets:

Balrog: The injection simulation package in the Dark Energy Survey (Eric Huff et al.)

Similar tool developed for HSC (SynPipe). DM plans to implement in next ~6 months.

58

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

OUTCOMES of Workshop on Impact of Blending

on LSST Science

Injecting targets:

Balrog: The injection simulation package in the Dark Energy Survey (Eric Huff et al.)

Similar tool developed for HSC (SynPipe). DM plans to implement in next ~6 months.

Anticipate this object injection tool will be very useful!

59

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

OUTCOMES of Workshop on Impact of Blending

on LSST Science

Ideas for future workshops:

• Machine learning focus for detection, classification, deblending, generative models**, … (all SC’s?)
• Morphometrics workshop (Galaxies SC).
• Hack/Sprint with specific data sets / goals.

When/Where? Append to another meeting? Examples:

• LSST DESC meeting, Feb 25 - Mar 1, 2019 at SLAC (or Bay Area)
• LSST@Asia, May 20-23, 2019, in Sydney

​

** See Blending Session #5.

60

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

OUTCOMES of Workshop on Impact of Blending

on LSST Science

Is it a star? Or is it a galaxy?

Generate “look alikes”?

Generative adversarial networks (GAN), variational autoencoders (VAE),

VAEGANs, and all that...

​

Message box example for short highlighted messages.

61

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

OUTCOMES of Workshop on Impact of Blending

on LSST Science

Is it a star? Or is it a galaxy?

Generate “look alikes”?

Generative adversarial networks (GAN), variational autoencoders (VAE),

VAEGANs, and all that...

​

Message box example for short highlighted messages.

Example:

arXiv/1807.00734

62

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

OUTCOMES of Workshop on Impact of Blending

on LSST Science

1 ADULT corgi ~ 46 litres

Message box example for short highlighted messages.

63

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Soccer

• 20+ people every day
• 3 women played
• Balanced teams (only a few games decided by more than 1 goal)

​

Thanks to Robin and El Conquistador staff for all the support!

64

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Game Nights etc.

Captain Sonar, dinosaurs,

fun, team-building,

and space pants/suit!

65

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Questions?

66

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018

Blank Template

• Guidelines
• Delete notes and save this template as your presentation
• No drop shadows
• Only use the font Calibri or Carlito
• Please do not alter template

Message box example for short highlighted messages.

67

LSST Project and Community Workshop 2018 • Tucson • August 13 - 17

#lsst2018