Flagship 3: Astronomy Knowledge and Skills for Development






This flagship aims to make use of knowledge and skills used in astronomy, such as programming, data handling, data analysis and machine learning, as well as infrastructure such as cloud computing platforms, to advance development objectives through either the teaching or application of skills. This may be executed in the form of educational programmes such as advanced schools/workshops, or in the form of hackathons/competitions which bring together skilled professionals or students with the goal of solving development issues using such skills.

The implementation of such schools/workshops could lead to  an increase in the number of students/individuals who are confident in applying the skills learned to their own studies, research or careers, as well as  motivated to further the skills learned, leading to an enhanced employability and greater contribution to the economy. It could also lead to a growing number of cross disciplinary and cross sector data intensive projects, further partnerships with development and data science organisations and wider use of infrastructure such as high performance and cloud computing, especially in contexts where such tools are used infrequently and could be beneficial.

N.B. Educational or outreach programmes at the schooling level, as well as general astro teacher training programmes will fall under the scope of the IAU Office for Astronomy Education (OAE http://www.haus-der-astronomie.de/OAE) and the IAU Office for Astronomy Outreach (OAO https://www.iau.org/public/oao/).

Expected Outputs:

  • Example projects for in-person or virtual schools, workshops or hackathon events at which participants gain transferable skills such as programming, data analysis, machine learning and deep learning techniques.
  • Best practice guidelines for the implementation of such events.
  • Reusable and openly accessible teaching or learning resources and assessments.
  • A searchable database of data science tools & teaching methods (refer to the OAD Data Science Portal)
  • Use of the OAD volunteer portal to either solve challenges using data science or to develop hackathon projects.

Key steps

  1. Review and consolidate existing initiatives, resources and projects.
  2. Produce a searchable database of programming/data science resources and teaching materials linked to astronomy.
  3. Identify platforms for hosting online events.
  4. Develop example projects/packages for both in-person and online implementations. These would include best practice guidelines for various models for different contexts, e.g. for hackathons held in countries with limited resources.
  5. Secure funding for the global rollout of such events.
  6. Partner with development organisations to identify development needs which could be addressed using data science skills.
  7. Volunteer participation: Utilise the volunteer portal to enlist skilled individuals who will source data, solve/tackle development challenges using data science or design hackathons which target a development need.
  8. Partner with data science organisations such as DataKind (https://www.datakind.org/) and research groups such as Data Science for Social Impact (https://dsfsi.github.io/) to identify potential projects or individuals who can assist in developing solutions.

Resources and Example Applications of Data Science:

Pilot Project Guidelines:

The astronomy discipline and those in the profession can contribute to development through skills transfer, or the application of these skills to solve development problems. That is, these applications can be direct, e.g. applying data analysis techniques used by astronomers to solve development-related issues, or indirect, e.g. the transfer of skills in order to grow capacity in data science. We provide guidelines below for these two approaches, with the caveat that these are still being refined, and that other approaches to this flagship may be developed in due course.

Indirect Approach:

e.g. Educational programmes such as data analysis, machine learning and deep learning schools, workshops, hackathons etc.

  1. Determine which individuals would benefit from the event in mind, i.e. would the school/workshop/hackathon be targeted at students (if so, at what level), educators, working individuals who would benefit from upskilling, or a combination thereof.
  2. The event may be physically attended by participants, or could be a virtual event, e.g. virtual hackathons. If the event is to be held at a venue, identify accessible institutions/venues at which sufficient space and resources (such as computers/laptops) are available.
  3. For machine learning / deep learning workshops and hackathons, an available computing platform such as a server or cloud should be identified for participants to access during the event. In these scenarios, the capabilities of computing infrastructure originally developed for astronomy purposes are brought to light, broadening horizons in terms of what can be achieved with programming.
  4. In cases where necessary, there should be access to a reliable internet connection with sufficient speed.
  5. In the case of hackathons/competitions, it may be beneficial for participants to work in teams.
  6. Prepare/source the necessary resources such as handouts, notebooks, slides, videos, tutorials etc.
  7. Identify volunteer tutors who have the expertise to assist participants throughout the event.
  8. In the case of data science events, projects may be chosen such that the participants are exposed to the various different applications of machine learning / deep learning, for example, commercial, science/research and development.
  9. Where applicable, participants may present their solutions nearing the end of the event, showing their understanding of the techniques used, or demonstrating the accuracy of their solutions. Prizes may also be awarded where possible.
  10. The success/impact of the event should be assessed with the aid of feedback from participants and tutors, and/or the quality/effectiveness of the solutions produced.
  11. Organisers may create a network for participants to keep in touch with each other, as well as the organisers, in order to assess the long term impact of the event or to carry out participant assessments post event.

Direct Approach: 

e.g. Applying the data analysis or machine learning techniques used by astronomers to analyse development-related data or hosting a hackathon event to tackle a development-related issue.

  1. Identify a development problem/need which could be addressed or contributed to by the application of skills such as programming, data handling, machine learning, deep learning etc. The issue could be a global problem, or one specific to a particular country or community.
  2. All or some of the data necessary to tackle the problem may be acquired and pre-processed prior to the event. Else, sourcing and cleaning/preparing data could also be part of the challenge for participants.
  3. Call upon individuals with a mixture of skills across data science and domain specific knowledge, as well as students or others who would like to develop particular skills through such an event, to come together for the event to work on solutions either individually or as part of a team.
  4. The event could take place physically or could be held online. If held at a venue, there should be access to a reliable internet connection with sufficient speed.
  5. An available computing platform such as a server or cloud should be identified for participants to access during the event.
  6. At the end of the event, participants may present their solutions and begin planning the implementation phase, unless the event itself involves the implementation of a solution/project.
  7. Teams could possibly work with various organisations, allowing for their solutions to be applied on a larger scale.


Bornmann, L. (2013). What is societal impact of research and how can it be assessed? a literature survey. Journal of the American Society for Information Science andTechnology, 64(2):217–233.

Fabian, A. (2010). The impact of astronomy. Astronomy & Geophysics, 51(3):3.25–3.30.

Davoust, E. (1995). The Purpose of Astronomy. Technical report.

Rosenberg, M., Russo, P., Bladon, G., and Christensen, L. L. (2013). Why is Astronomy Important?

Djorgovski, S. G., Mahabal, A. A., Crichton, D. J., and Chaudhry, B.(2015). From stars to patients: Lessons from space science and astrophysics for health care informatics. In 2015 IEEE International Conference on Big Data (Big Data), pages 2957–2959.

Meijer, I. (2012). Societal returns of scientific research: How can we measure it?


Big Data Hackathons

The Office of Astronomy for Development (OAD) and DARA Big Data (Development in Africa through Radio Astronomy) in partnership with the Inter-University Institute For Data Intensive Astronomy (IDiA) organised a number of Big Data Hackathons in Africa. The hackathons were part of a multi-year program that aimed to develop data science, programming and related skills in Africa in the context of huge radio astronomy projects.


Hackathons for Development aims to develop data science skills from science and use those skills to address development challenges. The team runs hackathons for learning, with science, development and industry contexts.



Hackathon projects

Guidelines to organise a hackathon


Please contact the team at hack4dev@idia.ac.za.