Phase B Preparation Workshop

Introduction

We are planning to hold a workshop in late April to kick-start the preparation of the science cases for Level 3 development effort in the Phase B proposal. As you know, Phase B is due to run from April 2019 to March 2023, thereby spanning LSST Commissioning and the start of survey operations. The DEV side of the Phase B proposal should, therefore, focus on what we could do (R&D tasks, not exploitation) with the commissioning data and what is needed for early science with LSST. The intention is that this workshop should inform the development of those cases by providing a more detailed description of what we can expect from the Project - in terms of data products, software and computing infrastructure - and comparing that with the requirements for some of our high priority science cases, thereby yielding a better idea of what we must provide ourselves, funded through the Phase B award. 

From initial discussion with Mario Juric and Andy Connolly in the LSST Project team, we have come up with the following initial plan.

Logistics

Registration: Please add your name and details on the Phase B Prep W/shop Participants page to help us coordinate catering, venue, etc.

Dates: 26th—28th April 2017

Location: Royal Observatory Edinburgh

Accommodation: Accommodation options in the vicinity of the Royal Observatory are listed on the ROE website. Of particular note, the Travel Lodge at Cameron Toll is conveniently located on major public transport links to/ from city, around a 15-minutes walk from the observatory.

Social events: We have not arranged any group dinners, to enable you to explore Edinburgh a little, but here are some suggested restaurants.

Pre-meeting preparation

• The recommended Wi-Fi network for use during the meeting is Eduroam. You may wish to check you have your current Eduroam credentials, before travelling.

• Read the Data Products Definition Document.

• Read the Juric et al paper on The LSST Data Management System from ADASS XXV  [https://arxiv.org/abs/1512.07914]

• Install some Python software on laptop — LSST Software Tutorial Notes

Final Programme:

Slides from talks are (will be) available from here, as well as being linked from the programme below. 

Day One: What to expect from the LSST Project

• 09.00: (Bob Mann)

• 09.30: Session 1 (1.5hr): (Eric Bellm)

  • Difference Image Analysis pipeline data products

  • Alert stream and Level 1 database: what to do with which and how

  • Brokers - what will the Project deliver? What are the constraints on developments by others (e.g. LSST:UK)?

• 11.00: coffee

• 11.30: Session 2 (1.5 hr): (Robert Lupton)

  • Outline of direct imaging pipeline, highlighting key algorithms

  • Specification of direct imaging data products

  • Availability of data products (policies and practicalities) - e.g. which images and how soon?

  • Data Release production and Data Access Centre capabilities

• 13.00: lunch

• 14.00: Session 3 (1.5 hr): (Mario Juric and Simon Krughoff)

  • DAC concepts and plans

    • Availability of LSST software stack to use in Level 3 re-reductions and analysis

    • Data access software

    • Environment to support development of Level 3 software (and constraints on that software)

    • Standards-compliance - e.g. how easy to use astropy, VO tools, etc.

  • Video of Simon K's Juptyer Lab demos are available from https://github.com/lsst-dm/sciplat-examples; the notebooks and instructions will be added there soon. The videos may also be viewed on YouTube: https://www.youtube.com/playlist?list=PLOVYFNUhDEm_gTJXLFhxgh1bgu-pCI0s1

• 15.30: tea

• 16.00: Session 4 (1 hr): (Lynne Jones, remotely) 

  • Introduction to scope and aims of different packages - PhoSim, CatSim, AlertSim, OpSim, MAF, GalSim. 

  • More detailed descriptions of use of some: i.e. MAF, GalSim/PhoSim, CatSim (possibly also AlertSim?)

• 17.00: Session 5 (0.5 hr):  and (Pat Eliason, remotely, with slides from Lucianne Walkowicz)

  • Role of LSSTC in operations and before - e.g. supporting scientific activities

  • Science Collaborations: roles

• 17.30: close

Day Two: Requirements from high-priority UK science cases

This will be a walk-through of a number of high-priority UK science cases, describing the requirements on data (quality and quantity), software and hardware, so that we could contrast those with what had been described on Day One. The cases should be drawn from a range of science areas, so as to span the range of dependencies on the Project deliverables. Since the goal is identifying gaps for development in Phase B, the focus should be on “early science”, but it will be left to each speaker to decide whether that means use of commissioning data, first year survey data, etc

The format for these will be as follows*:

1. 20 min talk: 

   • What is the science? 

   • What data is required? 

   • What is the computational scale of the analysis? - e.g. workstation v small cluster v supercomputer/grid

2. 10 min discussion:

     *N.B. the stellar talks have a different timing, but the same split between talk and discussion.

• 09.00: (Bob Mann)

• 09.15: (Carole Mundell)

• 09.45: (Aprajita Verma)

• 10.15: Weak lensing and clustering (Benjamin Joachimi)

• 10.45: Coffee

• 11.30:  (Stephen Smartt)

• 12.00:  (Mark Sullivan)

• 12.30: (Hiranya Peiris)

• 13.00: Lunch

• 13.45: (Vasily Belokurov)

• 14.09: Resolved stellar populations in nearby galaxies (Vicky Scowcroft)

• 14.33:(Ben Burningham) 

• 14.57: (Tim Naylor)

• 15.08: (Aleks Scholz)

• 15.19: Discussion of Naylor and Scholz talks

• 15.30: Tea

• 16.00: Solar System science: Activity, collisions and extreme objects (Alan Fitzsimmons)

• 16.30: (Wes Fraser)

• 17.00: (Eamonn Kerins)

• 17.30: Close