Shaping Research Software: An Interview with Sharon Tickell

The ARDC is working with the research community to drive recognition of research software. Our purpose is to provide Australian researchers with competitive advantage through data. Each month, we talk to a leading research software engineer (RSE), sharing their experience and tips on creating, sustaining and improving software for research. 

This month, we spoke with Sharon Tickell, Senior Software Engineer at CSIRO. Sharon specialises in design, implementation and operation of several distributed web applications, improving their interoperability. 

Sharon will walk us through her involvement in eReefs, a platform developed to support reporting and decision making around Great Barrier Reef preservation in a collaboration between the Great Barrier Reef Foundation, CSIRO, Australian Institute of Marine Science, Bureau of Meteorology, and Queensland Government. Currently funded by the Reef Trust Partnership, eReefs has users at universities all over Australia in addition to government decision makers and the Great Barrier Reef Marine Park Authority (GBRMPA). It is also now one of the foundation modelling platforms for the Reef Restoration and Adaptation Program (RRAP) and delivers data into many other research efforts. 

The ARDC has supported eReefs for many years by hosting part of the platform on the ARDC Nectar Research Cloud. The platform was shortlisted for the 2018 Australian Museum Eureka Prize.

What problems does eReefs solve?

The Great Barrier Reef is a hugely significant region for many people, but like almost all seas and oceans, the observational data that is available for use in research and for decision support is very sparse and/or localised. It is also very expensive and time-consuming to collect. 

eReefs seeks to model the physical, chemical and optical properties of the Great Barrier Reef waters. Subsequently, it makes the model results, along with data and visualisation products derived from these results, accessible to anyone who needs them. eReefs has been running since 2012, and has been delivering open-access data products and visualisations since about 2014. Data products include 4-km- and 1-km-spatial-resolution hydrodynamic, biogeochemical, sediment and optical model results for the entire Great Barrier Reef from back as far as 2010. We’ve also published a number of remote sensing datasets derived from Sentinel-3 OLCI and SAR observations, calibrated for Great Barrier Reef waters.

To help people discover and use the eReefs data, we host several public data discovery, analysis and visualisation websites that present the eReefs data in different ways. We also publish code libraries to help people query and visualise the data in their own applications without needing to download it first.

In addition to all that, we also have an online modelling platform called RECOM, which allows researchers in the reef science field to build even higher-resolution models for small parts of the reef. It uses public eReefs datasets as forcing data.

What is the impact of eReefs?

eReefs is used to:

• support the scientific consensus statement and track progress towards goals for the Reef 2050 Water Quality Improvement Plan
• contribute to UNESCO reporting for the Great Barrier Reef World Heritage Area
• contribute to an annual Regional Report Card for 5 Natural Resource Management areas
• assess the efficacy of nutrient, sediment and pesticide reduction policies for catchments adjacent to the reef
• assess which factors control the initiation and spread of crown-of-thorns starfish outbreaks on the reef.

eReefs data and visualisation products have been used to:

• detect oil spills and track flood plumes in the reef lagoon, and provide imagery to relevant authorities and researchers (using remote sensing data)
• generate freshwater exposure calculations for the GBRMPA Reef Snapshot report
• generate high-resolution RECOM and connectivity forcing for the RRAP Modelling and Decision Support modelling suite
• identify water depth and environmental conditions at locations of dugong and turtle sightings.

eReefs has also been referenced in 89 peer-reviewed publications so far.

Over the years we’ve also deployed sibling versions of the eReefs model and data explorer for other regions, including Storm Bay in Tasmania and Los Lagos in Chile.

What’s your role in eReefs?

I’m currently leading development on and support of the eReefs web platforms, including data services for hosting and delivering the eReefs datasets, the Data Explorer website and RECOM. I work with a small team of developers on those applications,  and coordinate with the catchment modelling, marine modelling and remote sensing teams to get their results published and integrated into our various visualisation tools.

Where does eReefs run? What infrastructure is being utilised?

eReefs runs in many places!   

• The big models are running on NCI Gadi, and we publish the results of those models via the NCI THREDDS server (with much appreciated assistance from the NCI National Collections team).
• The remote sensing algorithms and datasets are published from CSIRO’s internal network.
• Our data discovery tools and the eReefs Data Explorer platform are hosted on the ARDC Nectar Research Cloud’s QCIF node, also known as the QRISCloud.
• RECOM is currently hosted on AWS, as are the visualisation products maintained by our colleagues at AIMS.
• We rely on the Australian Access Federation (AAF) for authentication.
• The collaboration’s website and our static content are GitHub pages.

How does the ARDC Nectar Research Cloud support the eReefs platform?

We’ve been hosting eReefs data access and visualisation applications on Nectar right back since our first prototypes went live back around 2012. I believe we were quite early adopters of the QRISCloud node, and that work was actually my own first attempt at hosting anything on a public cloud platform and my first exposure to OpenStack.

I believe that Nectar is an incredible resource for enabling public delivery of science applications.  It’s allowed us to have a hosting platform that is independent of any of the partner organisations in the collaboration. Most critically, it is one that lets us keep our applications online when we have those pesky gaps between funded stages of the project! Also, the Nectar support teams have always been really helpful, and the training resources they provide are just getting better and better.

What are your key lessons of administering such infrastructure?

• For software development, standardising integrated development environments (IDEs) doesn’t matter at all. If you’re working with containers (which we currently are), then neither do operating systems. At one point, every developer on my team was using a different combination of operating system and IDE for their development work, and our deployment-hosts were different again, and it all still worked. 

• For application deployment, configuration-as-code is the key rule of everything. The ideal goal is to be able to deploy (or re-deploy!) your entire application from a blank tenancy without touching a graphical interface, because then it’s easy to make copies if you find you want a test or staging system. 

• Backups, backups, backups – automated backups which are stored offsite or on separate instances and include scripts for restoring backups into the development environment (for debugging) or a new tenancy.

• For supporting downstream users of your applications, Swagger documentation for APIs is a must-have, and Jupyter notebooks for live examples of using your libraries. 

• For project planning, if you’re operating a live platform, you need to plan in all sorts of extra time for upgrading things: operating systems, OpenStack API client versions, server software versions, server library versions, shinier JavaScript libraries. 

Keep In Touch

You can connect with:

• Sharon via GitHub
• the eReefs Team via email, the CSIRO website and GitHub.

If you’d like to be part of a growing community of RSEs in Australia, become a member of RSE-AUNZ – it’s free!

Updates on ARDC-Sponsored Research Software Awards

The ARDC is proud to sponsor awards for research software and research software engineers in all stages of their careers. The goal of the awards is to strengthen the recognition of research software and those who develop and maintain it as being vital to modern research.

The ARDC continues to sponsor a wide range of research software awards for 2024, including:

Venables Award for New Developers of Open Source Software for Data Analytics

The 2024 Statistical Society of Australia (SSA) Bill Venables Award is open for application until Friday 1 November to early to mid-career researchers/developers of new open source software primarily developed in Australia. A later-career researcher application would be considered if it was a first software project.

Learn more and apply now.

Read about the:

• 2023 winners:
  • Associate Professor Andrew Zammit-Mangion and Matthew Sainsbury-Dale (winners)
  • predictNMB team (runner-up)
• 2022 winners, Dr Lydia Lucchesi and Sam Nelson and the Vizumap team.

Eureka Prize for Excellence in Research Software

The finalists for the 2024 Australian Museum Eureka Prize for Excellence in Research Software has been announced. The prize is awarded for the development, maintenance or extension of software that has enabled significant new scientific research.

The prize will be presented on Wednesday 4 September. Register to watch the ceremony live. 

The ARDC is funded through the National Collaborative Research Infrastructure Strategy (NCRIS) to support national digital research infrastructure for Australian researchers.