This interview is part of an ongoing series exploring the evolving role of interdisciplinarity across the geosciences. As environmental challenges grow more complex, addressing them requires not only disciplinary expertise but also meaningful collaboration and innovation across fields, methodologies, and communities. In each conversation, I ask Division Presidents to reflect on how cross-divisional work is currently practiced, where it falls short, and what more transformative forms of collaboration could look like. Through this series, I aim to surface both practical pathways and significant tensions in interdisciplinary work to highlight why it matters for research, policy, and the broader societal relevance of the geosciences.
Hello Philip, and thank you so much for joining us. This conversation is part of our new interview series, Cross-Divisional Perspectives: Insights from the Presidents. In this series, we are exploring the unique scope of each EGU division and identifying new opportunities for interdisciplinary collaboration across the union. So, you’ve been serving as the AS Division President for a while now, right in the thick of your term. Looking back to when you first stepped into this leadership role, what has surprised you the most about steering the AS division?
So many things! For most people EGU is simply a conference, but once you look deeper you realise the full range of activities EGU as a geoscience union is leading, from conferences (more than just the General Assembly) to publications, webinars, early career events, teacher training – and the list is growing. What people also often don’t realise is the real bottom-up and volunteer-led spirit of EGU, which is fantastic to experience. As for Atmospheric Sciences in particular, it was always clear that we cover a wide range of science, but I had no idea how big the division really is – almost 2700 members in 2026, 12% of EGU’s total membership. This is amazing, but also a challenge to represent effectively.
The AS division covers an extraordinary spectrum that spans spatial scales from small-scale turbulent mixing to global planetary circulation, and timescales from seconds in chemical kinetics to centuries in climate modeling. How do you maintain a cohesive identity for a division where two members’ daily workflows can look very different from each other?
Diversity is good, in so many ways, so I don’t think we need to strive for too much unification; each of our many scientific communities should be able to express their own identity and their own ways of working. What should unite us are the broader scientific questions, and this is something we can always improve on. For example, as Earth warms and air pollution decreases, we struggle to close Earth’s radiation budget under emerging cloud feedbacks and aerosol trends, which also projects on large-scale dynamics. Traditional separation by disciplinary communities slows down progress in our understanding, and we still need to improve the links. Bringing everyone together at the General Assembly is a great start.
While the General Assemblies are EGU’s most popular and well-known events, I know from experience that the real work happens in the other 51 weeks of the year. During your presidency, what has been your primary strategy for keeping the AS community engaged, inclusive, and collaborative outside of the annual General Assembly?
Here I should really point out that EGU divisions are a team effort. Over the last years we have refreshed our entire team including our scientific officers, OSPP organisers and, importantly, a team of new early career representatives who also lead on our blog and social media (the full team can be found here). We also introduced the role of a policy officer. And when I myself may get stuck and distracted by EGU council business and the organisation of the AS GA programme, it has really been the early career team who pushed our AS communications and have revamped our blog with many interesting articles, including interviews with our AS award winners.
Your own research leans heavily into emerging tools like AI, machine learning, and advanced data compression. Within the broader AS division, do you see these rapidly evolving computational tools acting as a bridge that brings disparate sub-disciplines together, or are they risking a new digital divide?
It’s an exciting and slightly scary time to do atmospheric sciences! While AI is transforming all areas of science, for much of what we do the underlying physics is key and should not be forgotten. But aside from the actual science, we also see a rapid transformation of methodologies, with agentic coding taking over, and this is where I see a real risk of a digital divide between groups that embrace this change and others that don’t. This is particularly important for Early Career Scientists, but we will all need to learn how to navigate this space.
Atmospheric science naturally bleeds into Ocean Sciences, Biogeosciences, Hydrological Sciences, and Climate. Where do you currently see the most organic collaborative bridges between AS and these neighboring divisions?
Many atmospheric scientists, including myself, would naturally consider themselves climate scientists as well, but we of course have close links and overlaps with many other divisions – the question is: How best EGU can support this? One attempt to do this is cross-listing General Assembly sessions between divisions, but we are also considering more ambitious ways to group sessions and allow attendees to put together tailored programmes that really fit their interests. But personally, I believe the historic division structure of EGU would benefit from a major overhaul by grouping into its major components and introducing cross-cutting themes.
If you could point to one planetary blind spot that exists simply because atmospheric scientists and researchers from other earth science divisions aren’t talking to each other enough, what would it be? What are we missing out on by staying in our comfort zones?
Good question! To be honest, I am not entirely sure, as I believe we tend to be very open-minded and collaborative. But at the same time the current division structure does not serve cross-cutting topics well; for example, extreme events: a flooding event would be in Hydrology, is caused by the Atmosphere, possibly with Oceanic precursors, and we are interested in the change in risk under climate change…
Interdisciplinarity is a popular buzzword in grant proposals, but at conferences, it tends to manifest as just adding another division’s acronym as a co-sponsor to a session. When dealing with system-wide challenges like climate tipping points or the risks of solar radiation modification, how can the AS division move past superficial collaboration and work towards structural and efficient co-production of knowledge with other disciplines?
The current EGU division structures do not help, but divisions will exist in any structure. But I believe that, in terms of the General Assembly programme, there are many innovative ways to address this, which we are actively discussing in EGU council. This could go as far as the submission of abstracts without a target session and clustering sessions with AI – but this also risks losing community, so all of this needs quite a bit of thought.
Let’s say you are handed a blank slate to completely restructure how EGU divisions operate, throw out the traditional categories, and design a system optimised solely for cross-disciplinary breakthroughs. What is one change you would introduce, and what role should the Atmospheric Sciences play in leading that charge?
This is controversial but, personally, I would significantly reduce the number of divisions so that they represent major Earth or space “spheres”. This does not need to mean that any community loses its identity, as structures can be retained as sub-divisions or programme areas, just as is done in Atmospheric Sciences division. Combined with cross-cutting themes across all divisions, e.g. extreme events and machine learning, this would naturally support cross-disciplinary exchange – and make the organisation, and also the GA programme, much more manageable. It would also give EGU members more equal representation on council. Although division structure does not need to match the GA programme, it feels much more natural for it to do so.
