KWR takes Vienna, overview from a first time at EGU

The first day started with a dark grey sky and with our amazement at the size of the venue and the number of interesting sessions (with some of us feeling very stressed in having to make choices on which sessions to attend…). Again we opted for a divide-and-conquer approach. Joost and I ended up together at a session on “Bringing together climate scientists, impact modellers, and economists to build knowledge to effectively deal with climate change” for an out-of-the-box experience. Since this is not my area of expertise, it would be easy to be flabbergasted. And this was exactly what happened right during the day’s first (and by far the best) presentation. This was a talk from Ashish Sharma (UNSW Sydney) entitled “Can annual streamflow volumes be characterized by flood events alone” (Abstract EGU23-10210 (copernicus.org)). The simplicity and elegance of the proposed approach inspired me. In short, I’ve learned that while different climate scenarios lead to very different predictions on the average precipitation (and thus a lot of uncertainty), they agree very well on the extreme precipitation values. This means that predictions on the annual maximum flows are quite robust. The authors, therefore, chose to predict extremes of precipitation and link that to extreme event flows, and this ultimately to the total annual river flow for water supply. I’m curious to dive into this and see if and how we can use the same philosophy to predict extremes in future water demand, an input we need for the design of future-proof water distribution networks.

Later in the morning, Amin gave a presentation on “Error-Uncertainty Trade-off Judgment for Ensemble Member Selection of Regional Climate Models and Climate Change Impacts Modelling”. And in the afternoon, Xin got several questions on his poster “Classifying polymers with mid-IR spectra and machine learning: From monitoring to detection”.

On Wednesday, I convened a session on “Water Resources Policy and Management: digital water and interconnected urban infrastructure” (Session HS5.12 (copernicus.org)), together with David Steffelbauer (KWB), Riccardo Taormina (TU Delft), Andrea Cominola (TU Berlin) and Nadia Lund (Envidan). Bias aside, this was a very interesting session, addressing a variety (but at the same time a niche) of topics. It started with a keynote on scientific machine learning for speeding up distributed simulations by Roland Löwe (DTU). This was followed by presentations addressing stakeholder engagement, understanding water demand drivers, leakage detection, and heavy on the topics related to hydraulic modelling. Temperature modelling, numerical optimization, testbeds and gamification were addressed on the latter. The poster session in the afternoon was also very inspiring and I was impressed with the quality of the presented work and the dynamic and interactive vibe of the session. Topics ranged from data management ecosystems for decision-making, probabilistic hydraulic simulations, and short and long-term water demand forecasts and optimization (design and location problems for water distribution networks, optimizing chlorination in emergency water supplies and water reuse and service allocation).

Poster presentaions.

We ended one of the days with a KWR dinner at a nice restaurant in the city centre, where we had to fight the huge Austrian portion sizes. Conversations flowed naturally from topic to topic, but at some point, touched on “and why do you work at KWR?”. And this, again (recalling IWA Copenhagen here for a moment), proved that attending a conference with your colleagues is much more than the presentations and the networking, but also the realization of how inspiring it is to work with this bunch of intelligent, fun and passionate people.

I’ll keep this post relatively short; for more technical and reflective overviews of EGU, I refer to the posts of our colleagues Karel van Laarhoven, Peter van Thienen and Janine de Wit.

Overview of our contribution at EGU23 (presenting colleagues in bold)