WP 1.4 Coordination of LSST:UK Contributions to Commissioning

A session on early science and commissioning opportunities with Rubin was held at NAM2021, coinciding with the publication of a call for proposals for UK commissioning contributions. The proposals submitted in response to that call were reviewed by the Exec on behalf of the Board in October, and forwarded to the Rubin Commissioning team in November.

WP1.5: External Liaison

The 2021 LSST:UK All Hands Meeting attracted more than 150 registrants for its three half-days of events on 11-13 May 2021. Attendees heard a range of talks, including updates on the Rubin Observatory status and schedule, LSST:UK plans and a series of contributed talks from members of the LSST:UK community. Monthly LSST:UK Newsletters were produced each month during the reporting period and circulated via the lusc-announce email list, in addition to being posted on the LSST:UK wiki. Newsletter items keep LSST:UK Consortium members apprised of progress with Rubin construction and negotiations regarding the UK in-kind package, as well as disseminating news from the LUSC programme to the wider UK LSST community.

WP1.7: International Program Coordinator.

The revised programme of in-kind contributions was delivered. The Manual for in-kind contributors and recipients was developed, as well as the contribution profile explorer system. Contributions were made to the in-kind session at the Project and Community Workshop and the Rubin session at NAM2021. The work plan system for the in-kind program was delivered, and advice provided to contributors and recipients to support execution of the Fall 2021 contribution review period.

WP 2.1 DAC Management

The UK DAC Roadmap was revised in the light of the updated schedule for the LSST Data Previews; this will drive the planning for WP2.2, 2.4 and 2.5 for the remainder of Phase B. A selection round was run for Junior Associates, and discussions were held with Rubin Observatory staff over the mechanisms for updating the UK data rights list in future.

WP2.2 Data Ingestion and Publication

An updated instance of the qserv database system was installed on DAC resources in Edinburgh and loaded with the Data Preview 0.1 dataset. A collaboration was launched with French colleagues working with qserv at CC-IN2P3 in Lyon, with the intention of sharing experience of deploying qserv on different types of infrastructure (in France, the US and the UK) and of benchmarking its performance on these different systems

WP2.3/3.2 Lasair

Lasair selected as a Community Broker, see https://www.lsst.ac.uk/news/2021/lasair-selected-community-broker-21-07-28.

WP2.4 Provision of the DAC Platform

The latest version of the Rubin Science Platform has been installed on UK DAC resources and the Rubin SQuaRE team have recognised this IRIS-based deployment, undertaking to ensure that future updates to the RSP codebase will be consistent with it.

WP2.5 Science Support

DAC staff collaborated with WP3.11 team to complete a first data challenge, exercising DAC resources to create cross-match catalogues for a subset of Gaia and WISE, with outcomes documented in an updated DAC-DEV interface requirements document. DAC staff working with WP3.5 team to plan and resource the data-fusion pipeline, in operations, augmenting Rubin data with additional insight from VISTA (or other near-IR surveys).

WP3.3: Spectroscopic classification of transients

Delivery of software package to process any new LSST cadence strategy (opsim) and automatically produce metrics for TiDES SN and TiDES host galaxies survey performance. Presentations to the DESC time-domain working group and to the 4MOST All-Hands Meeting. Began development work on the interface between the Lasair broker and 4MOST.

WP3.5 LSST and near-infrared data fusion

Completion of a full run of the pipeline in the overlap region between HSC PDR2 and VISTA. First test multi-wavelength catalogues currently being used to produce photometric redshifts and for spectral energy distribution modelling, which will aid in their validation. Work has begun on the generation 3 version of the pipeline, which involves a completely restructured ‘Butler’ database.

WP3.7 Low-surface-brightness science using LSST

Our work on quantifying the accuracy of the sky-subtraction in the current LSST pipeline has been gaining attention throughout the LSST landscape. Since the outputs of this work package will determine LSST's ability to access the low-surface-brightness regime, it will affect the discovery space of all science collaborations. Aaron Watkins has given several invited talks on the work e.g. at Symposium 12 at the EAS meeting, the low-surface-brightness session at the Rubin PCW, two sessions at the UK NAM, the Synthetic Source Injection Workshop organised by the DM team and a science talk at a Galaxies Science Collaboration meeting. His work was featured and discussed in the DM team's presentation (by Lee Kelvin) at the Rubin PCW on the properties of the LSST pipeline. Our focus now is to develop new sky-subtraction algorithms with a view to inserting them into the pipeline further down the line. Aaron's current work is developing an idea that was first proposed by Robert Lupton (DM team lead) to perform sky subtraction that preserves LSB flux at scale for the LSST. Initial tests on small deep fields have shown promise. A paper describing these new algorithms, tested on deep fields from a wide variety of telescopes (including the Hyper Suprime-Cam) is expected be published in the summer of 2022 (the co-authors will comprise the members of WP 3.7 and the DM team).

WP3.9 LSST Point Spread Function, sensor characterisation and modelling

Initial resistivity measurement technique on CCDs performed (part of WP 3.9.2). Sentaurus model of CCD250 pixel started (part of WP3.9.2). New design for the interferometrically aligned PSF projector completed (preliminary work for WP3.9.3).

WP3.10 UK Contributions to DESC Operations

The paper on tomography was published: The LSST-DESC 3x2pt Tomography Optimization Challenge | Published in The Open Journal of Astrophysics . The team is now assisting with the STAR project within DESC, the Static Analysis Round Table, which will collect together everything needed for a lensing + clustering analysis. Joe Zuntz started new role as Deputy Computing Manager for DESC. James Perry completed his contribution to GalSim to improve how pixel boundaries are handled, making it ready for LSST's camera behaviour. James has improved the serial run-time of GalSim by a factor of 5.

WP3.11 Cross matching and astrometry at LSST depths

The WP3.11 team have continued to build links with the wider community. One result was their being asked by the Rubin SMWLV (Stars, Milky Way, and Local Volume) Science Collaboration to present the work of this collaboration at the LSST:UK all-hands meeting. Within this reporting period the deliverable D3.11.3 was also completed. This extends the options available in the cross-matches to include two important aspects for the matching of LSST to ancillary datasets, mostly infrared catalogues such as WISE. WP3.11 now have a robust and accurate cross-match tool, and will be able to provide LSST:UK users with robust matches between LSST and other photometric catalogues. The WP3.11 team have also undertaken scaling and efficiency tests in preparation for executing their cross-matching software on DAC hardware. Finally, an initial version of a DAC-DEV Interface Definition has been successfully validated by a data challenge to cross-match two precursor surveys – namely, Wise and Gaia. Lessons learned from the data challenge will feed into a revised Interface Definition in the coming period of work.

D3.7.3 Software to Output Metrics That Keep Track of Improvements to the Pipeline Sky Subtraction

Aaron Watkins, Chris Collins, Sugata Kaviraj

Low surface brightness science contains much of the potential discovery space for LSST, in galaxies, solar system, and Milky Way research. Unfortunately, the authors (i.e. LSST:UK work package WP3.7) have previously demonstrated that the current LSST data reduction pipeline’s sky subtraction routine over-subtracts flux in the outskirts of extended objects, making low surface brightness science with LSST a potentially very difficult enterprise. Changes to the pipeline are thus required to ensure that LSST’s scope is not limited in this way. As WP3.7 develop an alternative sky-subtraction routine to resolve this issue, it is necessary to track improvements using quantitative metrics. This in turn requires software that can measure and output these metrics alongside regular runs of the pipeline. This report details the current state of such software, which WP3.7 are now making available to the LSST:UK community. The authors also describe future plans for the software, to address its current slow speed and to more easily integrate some version of it into the LSST pipeline itself for use by the data management team.