• 1 Introduction
• 2 New LSST:UK Science Centre team members
• 3 Plans for future data rights calls
• 4 New paper: Systematics mitigation for catalogue-based angular power spectra
• 5 JAN Science Communications Training report
• 6 New paper: Isophote shape analysis and the unfortunate subtlety of dwarf galaxy structure
• 7 Junior Associates Network update
• 8 Communications news
• 9 Forthcoming meetings of interest

Open

Introduction

The flurry of activity reported last month following the first Rubin alerts has eased over the past few weeks as the continuing Early Operations System Optimization work has focussed on engineering observations. Keith Bechtol continues to provide very informative weekly updates on the Community forum, but the bottom line is that the image quality is still not reaching, on a routine, repeatable basis, the level that it is seen to be capable of achieving. It appears that there optical aberrations caused by higher-order mirroring bending modes, and it is taking some time to obtain the observational data needed to characterise these and determine corrections for them.

The programme for our All Hands Meeting in April is now being finalised, and we’re starting to plan our sessions at NAM2026. As usual, these will include a range of content, with plenty of room for contributed talks. All are welcome to submit abstracts for those, and we would particularly welcome submissions from Early Career Researchers. The deadline for submitting abstracts is 23.45 on 2 April and the form to do so for our sessions can be found here. We will soon be setting up an organising committee for those sessions and would welcome volunteers to help with that; again, all are welcome, but we would particularly encourage ECRs to apply - please email @Graham Smith and myself, if you are interested. In addition to the three sessions in the parallels programme, the Junior Associates Network will be holding a lunch-time discussion and information session, following the successful pattern from last year’s NAM.

Looking ahead, registration for the 2026 Rubin Community Workshop (27-31 July) and for LSST@Europe8 (27 September - 2 October) is now open and we have opened calls for applications to the Pool Travel Fund to support attendance of these meetings. Since both of these meetings have abstract submission deadlines in April - on 13 April for LSST@Europe8 and on 24 April for RCW2026 - we have set the application deadline for these calls to be 21 April, although later applications will be considered (subject to available funding).

Finally, please note that as detailed below, the next LSST:UK data rights call will open on 21 April (TBC, via an announcement on the lusc-announce list) and close on 19 May.

 

Open

The Magellanic Clouds and the Rubin dome.
Credit: NSF–DOE Rubin Observatory/NOIRLab/SLAC/AURA/W. O'Mullane

 

 

 

@Bob Mann

New LSST:UK Science Centre team members

We’re pleased to introduce a number of new staff who have joined the LSST:UK Science Centre (LUSC) team in recent months.

Manu Antony is a Senior Research Software Engineer working with the Data Access Centre Team in Edinburgh to facilitate data management and build platforms for users to be able to interact with the data.

His background includes Computer and Electronics engineering and a PhD in developing Grammar-based systems for 3-D Visualizations.

Open

Cordelia Dashwood Brown recently joined LSST as a post-doctoral researcher at the University of Southampton, where she is working as part of the TiDES team to develop and maintain observation and data pipelines for following up transients identified by LSST. Part of Cordelia’s role focuses on improving the speed and accuracy of transient classification to support cosmological studies using Type 1a supernovae.

Beyond TiDES, she also works on black holes and core-collapse supernovae, and is part of the COBALT collaboration, which focuses on using optical surveys such as LSST to identify non-interacting black holes through lensing. 

Laura Moran is a Project Manager on LSST:UK, and is based in EPCC, University of Edinburgh. She joined EPCC in January 2022 and has worked on a myriad of projects. These include ARCHER2 support, medical imaging data research, metascience work on research council funding, Rust, and outreach. Her first degree was an MPhys Astrophysics from the University of St Andrews, followed by a Physics PhD from the University of Exeter where she adapted an MCRT code developed for modelling circumstellar environments into a model for Raman spectroscopy to aid in breast cancer detection. Her hobbies include reading and gardening.

Eduardo Toledo Romero is a software engineer with a background in Earth Science and a Masters of Astrophysics, now working as part of the Data Access Centre team in Edinburgh. He has focused his career on building computational solutions to address scientific questions in different fields. Eduardo led a Scientific Computing team for 7 years at Kew Botanic Gardens focusing in phylogenomic and geographic information before joining us, to pursue his passion for astrophysics.

Chris Winterburn recently joined the LUSC as a Scientific Computing Graduate with STFC at the Rutherford Appleton Laboratory, following an MSc in Data Science from the University of Liverpool and a BSc in Natural Sciences from the University of Birmingham. Over his first six months with the team, Chris has been contributing to the LSST Data Release Processing (DRP) pipeline, focusing on migrating the messaging stack to Kubernetes to ensure it can handle the high throughput expected from the US Data Facility. In the coming months, he'll be researching ways to enhance the performance of the DRP Butler database.

Outside the day job Chris enjoys getting out and about running and hiking. He took this photo last week up Latrigg in Keswick for any frequenters of the Lake District.

@Bob Mann

Plans for future data rights calls

One indirect casualty of the STFC PPAN Prioritisation Exercise has been that the progress made over recent months towards the long-awaited signing of the UK’s LSST Data Rights Agreement (DRA) has stalled, as STFC are unable to sign an agreement that includes a Statement of Work detailing our in-kind contributions to Rubin operations until they are confident that they know what level of funding they can now commit to that programme. This delay with the signing of the DRA means that we will have to run at least one more interim data rights round, before we can return to awarding three-year data rights terms.

In preparation for the return to awarding longer data rights terms, the Consortium Board has updated our policy for awarding data rights and has appointed a new Data Rights Allocation Committee (DRAC) to assess applications once we have returned to awarding three-year terms. I am delighted to report that Michelle Collins, Suhail Dharwan, Poshak Gandhi, Cyrielle Opitom and Brooke Simmons have agreed to serve on the DRAC, alongside Nial Tanvir, who will chair it, as he did its previous incarnation, the PI Selection Committee – and thanks to Sarah Bridle, Paul Crowther, Pedro Ferreira, Wes Fraser, Nina Hatch, Kathy Romer, Amelie Saintonge, Stephen Smartt, Colin Snodgrass, Vivienne Wild (and others I may have forgotten) for serving on that.

The next interim data rights round is due to open on 21 April and an announcement confirming its details will be circulated via the lusc-announce email list before then. It will award data rights to 30 September 2027; existing data rights holders will be contacted and asked to confirm whether they want to have their current term extended to that date, so the only people who need make a full application in this coming round are those who currently do not hold data rights. All applications received by 19 May will be considered as part of this round, and successful applicants informed by early July, to enable timely access to Data Preview 2.

One difference from previous data rights rounds is that the application forms will continue to remain open after that closing date of 19 May; applications received after that date will then be considered at the next review date, which will be in the autumn. The motivation for this change is that people should be able to apply for data rights whenever they want to - so, for example, if someone hears a talk about LSST and thinks “I should look into applying for data rights”, they should be able to do so straight away, rather than, potentially, being put off from doing so by discovering that the next call is some months away. We will continue to review data rights applications twice a year - as we have agreed with Rubin that we will make only two bulk updates to our list of data rights holders per year - and we intend to publish those review dates on the wiki and in the Newsletter well in advance. The intention is that these procedural changes should remove barriers that might prevent interested people from applying for data rights.

The forms through which people will apply for data rights, as either PIs or Junior Associates, will also be changing, in response to a review of the criteria that should be used for awarding data rights, which will now focus on the plans that applicants have for making use of data rights, rather than their past track record. Applicants will also be asked for more personal information than before, to support the enhanced EDI monitoring now being required of us by UKRI, but this will be done in such a way that no personal information about named individuals will be retained after summary statistics are generated from the applications processed at a review point and every question will continue to include a “prefer not to say” option.

@Bob Mann

New paper: Systematics mitigation for catalogue-based angular power spectra

@Thomas Cornish and collaborators recently had a paper published in the Monthly Notices of the Royal Astronomical Society, titled Systematics mitigation for catalogue-based angular power spectra. This work extends a recently introduced formalism for computing angular power spectra directly from catalogue-based data, by incorporating a means of mitigating systematic effects which can significantly bias the measured signal.

Background: catalogue-based angular power spectra

The angular power spectrum has become an essential summary statistic in cosmology: in essence, it quantifies how common fluctuations of a given angular scale are in a field (e.g. the CMB temperature field, or the galaxy overdensity field), and as such has seen frequent use in studies of the cosmic microwave background (CMB) and of the large-scale structure (LSS).

Computation of these power spectra typically requires data in the form of pixelised maps showing the fields of interest (e.g. the CMB temperature field), but there are many scenarios in which the field is sampled at discrete positions on the sky, corresponding to the positions of sources in a catalogue – for example, the galaxy overdensity field used in clustering analyses is sampled at the positions of individual galaxies. The traditional course of action in these cases is to bin these discrete points into pixels and produce a map showing the mean field value in each pixel. This pixelisation naturally introduces a bias at scales comparable to or smaller than the pixel scale, inhibiting the use of angular power spectra to probe small-scale anisotropies: in order to do so, one has to produce maps which have sufficiently high resolution while also being low-resolution enough to avoid being dominated by Poisson noise, which are not easy criteria to meet simultaneously. This has significant implications in LSS cosmology, where being able to probe small scales enables valuable constraints on the gravity-dominated, nonlinear regime of cosmic structure formation.

Recent work by Wolz et al. (2025) addresses this problem by introducing a new formalism which allows one to compute angular power spectra directly from catalogue-based data, circumventing the need to make pixelised maps. They find that their formalism performs as well as the traditional map-based formalism at large scales and, crucially, outperforms it at smaller angular scales due to the absence of pixelisation. They make this formalism available in the public angular power spectrum estimator codes such as NaMaster (Alonso et al. 2019).

This work: incorporating systematics mitigation

However, pixelisation is not the only source of systematic error, and the catalogue-baed formalism is still susceptible to others. These systematic effects arise from observational or instrumental phenomena which cause the observed field values to fluctuate in ways that are not cosmological in origin, thereby biasing the measured power spectrum. These effects can include (among many others) variations in airmass or seeing across a survey footprint, or the attenuation of light from distant objects by dust from the Milky Way. They often dominate at larger scales, necessitating some means of removing their contribution to the observed power spectrum to achieve unbiased results.

Several techniques have been explored in the literature to mitigate these effects. One technique, known as mode deprojection, models the contribution of each systematic to the observed signal as linear, determines the best-fit linear coefficients for each one, and subsequently removes that contribution from the signal. While the assumptions involved are quite simplistic, this technique has proven effective at mitigating systematic effects in many cases, and also has the advantage of being readily available in widely-used codes such as NaMaster (Alonso et al. 2019).

Open

Figure 1: Top panel: Comparison of the mean clustering angular power spectra measured from our simulated catalogues, before (green dots, solid line) and after (teal circles, dashed line) correcting for contamination by systematics. The black line shows the input angular power spectrum used to generate the catalogues, prior to contamination. For reference, the results of using the traditional map-based approach to compute the power spectra and apply mode deprojection are also shown (blue diamonds, dotted line). Bottom panel: Normalised residuals of the cleaned (i.e. post-mode deprojection) power spectra with respect to the true power spectrum. Our catalogue-based approach is able to robustly recover the input angular power spectrum, and is seen to outperform the map-based approach at both the smallest and largest angular scales.

In this work, we adapt the mode deprojection technique to the catalogue-based formalism for computing angular power spectra, and validate its performance using simulations, both for discretely sampled fields (such as the weak lensing shear field sampled at the positions of galaxies) and for the unique case of galaxy clustering, where the density of sources is itself the field of interest. Briefly, these simulations involve first generating many realisations of a catalogue corresponding to some input angular power spectrum, after adding some artificial systematic contamination. We then compute the angular power spectra from these catalogues, first with no correction for these systematics, and then with a correction applied using our adapted mode deprojection technique. Repeating this for 1000 simulations, we find that our catalogue-based approach is able to robustly recover the input angular power spectrum (see Figure 1). Combined with the other advantages of the catalogue-based formalism, this approach thus enables unbiased measurements of angular power spectra even at small scales where pixelisation effects would bias map-based measurements, and at large scales where observational/instrumental systematic effects tend to dominate. Having verified its efficacy, we implemented this technique in the NaMaster, where it is now publicly available.

Implications for LSST

While beyond the scope of this paper, we have also incorporated the catalogue-based formalism for computing angular power spectra (complete with mode deprojection) into the LSST:DESC pipeline for joint galaxy clustering and weak lensing analysis, TXPipe (Prat et al. 2023). This update extends the regime of angular scales that LSST will be able to reliably probe for galaxy clustering and weak lensing, and is an important step towards obtaining precise, unbiased constraints on the dark matter distribution in the Universe.

@Thomas Cornish

JAN Science Communications Training report

I recently participated in LSST:UK Science Communications Training, March 2026. This training was conducted over two online sessions delivered by @Eleanor O'Kane for members of the Junior Associates Network.

As an early career researcher I went into the training with limited science communication experience, mainly giving occasional outreach talks. I was very interested in learning about best practices and what other opportunities there are in the world of science communication such as using social media and talking to the press. I also learned how science communication helps fulfil a variety of objectives, from promoting your research (and yourself), to engaging and inspiring the public.

This training was very useful and I will be referring back to the slides when I next need to think about science communication. My main takeaways were tips on how to cater to a modern screen-based audience. Namely how to get your core message across in the first minute(s) of communication so that if someone decides to ‘scroll past’ or ‘switch off’, you've at least imparted the important bits that they might think about or follow up on later.

@James Robinson

New paper: Isophote shape analysis and the unfortunate subtlety of dwarf galaxy structure

MNRAS recently accepted a paper by @Aaron Watkins and collaborators titled Isophote shape analysis and the unfortunate subtlety of dwarf galaxy structure. All galaxies occupy a fairly constrained set of morphologies – spirals like the Milky Way, ellipticals like M87 in Virgo, and so on – which result from those galaxies' formation histories.

Among massive galaxies, these morphologies are distinct enough that classification can be largely automated using simple parametrisations of their light profiles, like how centrally concentrated or evenly distributed the starlight is. Among dwarfs, however, the differences are much subtler: too subtle, in fact, for these automated methods to work, which spells trouble for dwarf classification in large surveys like LSST (where the dwarf sample will number in the tens of millions of objects). Watkins et al. thus investigated whether alternatives to the traditional parameters would suffice. Dwarfs are much fainter than massive galaxies, so in our modern era of deep imaging, they occupy a similar place as massive galaxies did in the days of photographic plates. Watkins et al. thus took inspiration from those early photographic studies by quantifying the shapes, orientations, and variability of dwarf galaxy isophotes (annuli of constant brightness) as a function of radius from the center, which directly trace the distribution of starlight within galaxies.

Unfortunately, as the paper title suggests, they found that these parameters also can't easily distinguish dwarfs from each other, even if said dwarfs are visually quite different. In fact, dwarfs seem to be self-similar in more ways than just quantified light profiles. In addition to looking for trends between these isophotal parameters and morphology or stellar mass, Watkins et al. employed some high-dimensional clustering algorithms using a wider parameter space to determine which characteristics could actually separate dwarfs into distinct categories, but found few noteworthy natural clusters. It may be that dwarfs are nearly structurally indistinguishable, except perhaps for the presence or absence of young, massive, blue stars. In some ways this is expected, as dwarfs are considered the building blocks of massive galaxies in our current cosmological model, but it does mean that studying dwarf galaxy evolution on its own terms may be a tricky endeavour. Expanding the parameter space even more, and using a larger dwarf galaxy sample, will likely be needed to tease out the fine machinations behind dwarf galaxy evolution.

Open

Left panel: showing the mean measured isophotal twistiness (quantifying variability in the isophotes' on-sky orientation) as a function of morphological class (from left to right: elliptical, lenticular or lenslike, spirals, and featureless). Black squares show values for dwarf galaxies, while gray circles show a comparison with a more massive galaxy sample. Within the errorbars, the dwarfs are nearly indistinguishable using this parameter, whereas massive spirals are clearly much twistier than ellipticals or lenticulars. Right panel: histograms showing the full distributions of twistiness for each morphological class. Blue histograms show distributions for dwarfs, while gold histograms show them for massive galaxies. Again, the blue histograms are all quite similar.

@Aaron Watkins

Junior Associates Network update

Here are a few short updates from the LSST:UK Junior Associates Network (JAN) – a network organised by junior associates for junior associates.

JAN sessions at the All-Hands Meeting

The LSST:UK All-Hands Meeting is rapidly approaching, and will provide a great opportunity for Junior Associates to interact with other members of the LSST:UK Science Consortium and to learn more about the collaboration. To help facilitate this, the JAN has been granted several time slots across the first two days (13-14 April), including icebreaker activities, a JAN social, and a presentation/discussion session hosted by the JAN’s very own chairs, Clara Pennock and Tom Wilson. These are scheduled at the following times:

• Mon 13th 12:00-12:30 – Icebreaker activity

• Mon 13th 15:00-15:30 – Icebreaker activity

• Mon 13th 16:30-17:00 – Presentation/discussion 

• Mon 13th 18:00-??? – JAN social

• Tue 14th 10:30-11:00 – Icebreaker activity

The icebreaker activities will have a “speed meet” format, in which participants will engage in short (~2 min) conversations with each other before moving on to the next person. The social will involve a trip to one of the local pubs (TBD). We look forward to seeing many of you there!

Communications training

This update is just to say a huge thank you to Eleanor O’Kane for delivering two excellent training sessions on science communications for members of the JAN earlier this month. If you were unable to attend and would like to know more about it, see Jamie Robinson’s report above, where you can also find a link to recordings of the sessions.

JANEC elections

As previously announced, we intend to hold elections for three positions on the JAN Executive Committee (see the Wiki page for details). Since the All-Hands Meeting is just around the corner, we have decided to extend the deadline for nominations until sometime after the meeting (date TBD), so that people have more time to familiarise themselves with the JAN and meet some of its members before submitting any nominations.

JANOC Membership

If you are interested in helping to run the JAN in any capacity (e.g. keeping track of membership, updating the Wiki, managing the Slack, organising online workshops and talks,  or anything else you would like to do) then please consider joining the JAN Organising Committee (JANOC) by contacting either of our current chairs, Clara Pennock (cpennock@roe.ac.uk) and Tom Wilson (T.J.Wilson@exeter.ac.uk). We are always happy to welcome new volunteers!

Additional information

More information on the JAN can be found on the Wiki, including sign-up forms for the mailing list, details of the various Slack channels available to JAs in the LSST:UK workspace, and minutes from the meetings of the JAN leadership team.

@Thomas Cornish, Steve Ardern, @Garreth Martin, @Clara Pennock, Andres Ponte Perez, @James Robinson, @Matthew Temple, @Andrew Wilson, @Tom J Wilson

Communications news

In the Spotlight this month

Research software engineer research @Eduardo Toledo is the latest team member to tell us what he enjoys about his work and how he’s contributing to LSST:UK.

@Eleanor O'Kane

Read Eduardo’s interview on the website:
https://www.lsst.ac.uk/news/2026-03-26-lsstuk-in-the-spotlight-eduardo-toledo-romero

Get in touch if you’d like to take part in our In the Spotlight series | email: eokane@roe.ac.uk

Open

Communications discussion at the All Hands Meeting

There will be a presentation and discussion on LSST:UK Communications at the All Hands Meeting in the afternoon on 13 April. If you have any views on how we communicate and what we cover, please do use the opportunity to tell us what you think.

Forthcoming meetings of interest

Dates, locations and links… The current list of forthcoming meetings is always available on the Relevant Meetings page. You may also wish to check information held on the LSST organisation website LSST-organised events and the LSST Corporation website.