The Grand Challengers Podcast Episode #1

Engineers never say never and never say always – a new take on flood models

Guest: João Paulo Leitão

January 3rd, 2023


“Engineers never say never and never say always”

In the face of an unpredictable future, we are all called upon each and every day to think of out-of-the box ways to tackle challenges to our environment, wellbeing or how to find your submerged bicycle by scrolling through social media.

My guest today is Dr João Paulo Leitão, a senior researcher at the Swiss Federal Institute of Aquatic Science & Technology and lecturer at ETH Zurich. Joao has been active in the areas of urban flood risk assessment and urban water asset management research for over 13 years and actively looks for new ways to combine numerical models, spatial planning and a good dose of storytelling to create future cities.

Today on the show, we explore João’s early research journey, how his Portuguese heritage has actively inspired his work and how he is using flood models, unconventional data sources and machine learning beyond engineering design.


João Paulo Leitão is a Senior Scientist at the Department of Urban Water Management in the Swiss Federal Institute of Aquatic Science and Technology (Eawag). He studied Environmental Engineering at the Technical University of Lisbon (2001) and obtained a Master of Science degree in Geographic Information Systems (2004) from the same University. After a couple of years as Research Assistant at the Technical University of Lisbon, investigating spatial analysis methods, he moved to Imperial College London for a PhD in Civil and Environmental Engineering focusing on Enhancing elevation models for advanced urban flood modelling (2009). After his PhD, João moved back to Lisbon to work as Postdoctoral fellow (2010-2013) at the Portuguese Civil Engineering Laboratory (LNEC); there he developed new methods and coordinated projects aiming to assess the performance and the risks associated with urban water systems, with a particular focus on pluvial floods. In 2013 João joined Eawag where he established himself as Group Leader on Urban Flood Risk. Since 2018, he is also lecturing Infrastructure Systems in Urban Water Management at ETH Zurich.

João has more than 10 years of experience in urban flood risk assessment and urban water asset management, and has published over 50 ISI-cited articles since 2005. He specialises in exploring novel data sources, such as images and videos, to estimate flood data to support flood risk management in urban areas. In particular, he has been developing new methods to extract flood depth and velocity from videos and images obtained from social media and surveillance cameras. He has also been pioneering the development of data-driven two-dimensional flood prediction models that can be used to inform rescue services during urban pluvial events and help protecting citizens and assets threatened by these events.

During João’s scientific career, he has been the recipient of a few fellowships and grants. He serves frequently as reviewer for various journals and of scientific proposals. Currently, João is Associate Editor of the Urban Water Journal and Chair of the International Working Group on Data and Models (IWGDM) of the IWA/IAHR Joint Committee on Urban Drainage (JCUD).

Resources related to the Episode

(Disclosure: Links on this page to “View on Amazon” are Affiliate links. This means that, at zero cost to you, I will earn an affiliate commission if you click through the link and finalize a purchase.)

  • Salted Cod Recipes: Bacalhau à Brás – João has suggested four possible recipes, feel free to experiment and optimise these for your taste! [Recipe 1] [Recipe 2] [Recipe 3] [Recipe 4]
  • The Azores: Learn more… – key islands João mentioned include Terceira, São Jorge, Faial
  • Historical study on Terceira: Flood modelling to inform and validate historical documents on the development of the town of Angra on the island of Terceira, The Azores
    • Leite, A. and Leitão, J., 2021. The new town of Angra (Terceira, the Azores): confirming a contested urban planning history using reverse historical analysis and flood modelling tools. Urban History48 [Link] (1), pp.20-36.
  • Data Driven Flood Models: Training a ‘structure’ or convolutional neural network on data to reproduce flooding
    • Guo, Z., Leitão, J.P., Simões, N.E. and Moosavi, V., 2021. Data‐driven flood emulation: Speeding up urban flood predictions by deep convolutional neural networks. Journal of Flood Risk Management14 (1), p.e12684. [Link]
    • Guo, Z., Moosavi, V. and Leitão, J.P., 2022. Data-driven rapid flood prediction mapping with catchment generalizability. Journal of Hydrology609, p.127726. [Link]
  • Water Level Predictions using social media images: Using alternative data sources to help validate or inform flood modelling – data for floods!
    • Chaudhary, P., D’Aronco, S., Leitão, J.P., Schindler, K. and Wegner, J.D., 2020. Water level prediction from social media images with a multi-task ranking approach. ISPRS Journal of Photogrammetry and Remote Sensing167, pp.252-262. [Open Access]
    • de Vitry, M.M. and Leitão, J.P., 2020. The potential of proxy water level measurements for calibrating urban pluvial flood models. Water Research175, p.115669. [Link]
  • The Adventure of Life by Joël de Rosnay, a French biologist born on June 12th 1937, original title „L’aventure du vivant“, Paris, 1988, Seuil [View on Amazon]

Connect with João Paulo Leitão