TLThomas Lutonsky
Thomas Lutonsky
*This is an algorithmically generated author record based on publicly available information.
Research Subjects
Thomas Lutonsky has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Andreas Buttinger-Kreuzhuber
VRVis Zentrum für Virtual Reality und Visualisierung Forschungs‐GmbH Vienna Austria
2 shared publications - Daniel Cornel
VRVis GmbH
2 shared publications - Jürgen Waser
VRVis GmbH
2 shared publications - Michael Wimmer
Technische Universität Wien (TU Wien), Vienna, Austria
2 shared publications - Nils Wandel
University of Bonn, Bonn, Germany
2 shared publications - Reinhard Klein
University of Bonn Germany
2 shared publications
Latest Publications
- Open Access
- 25 September 2026
- Published by Elsevier in Journal of Hydrology
- Vol. 680
Deep learning with upstream-aware terrain representations and rainfall adapter for generalizable pluvial flood prediction
- 25 September 2026
- Published by Elsevier in Journal of Hydrology
- Vol. 680
Open AccessDeep learning with upstream-aware terrain representations and rainfall adapter for generalizable pluvial flood prediction
- Open Access
- 4 April 2026
- Published by Elsevier in SSRN Electronic Journal
Deep Learning with Upstream-Aware Terrain Representations and Rainfall Adapter for Generalizable Pluvial Flood Prediction
Deep learning-based surrogate models can substantially accelerate pluvial flood simulations. However, most existing approaches struggle to generalize across heterogeneous terrains or modified topographies, limiting their applicability for real-world flood risk assessment unless retrained for each specific terrain.In this work, we investigate how deep learning flood surrogates can be designed to achieve robust spatial generalization across previously unseen terrain configurations without retraining. We train models on diverse terrain and introduce upstream-aware input maps that propagate terrain properties into downstream regions, allowing the model to capture long-range spatial dependencies and supporting the learning of more generalizable spatial features. In addition, we add architectural design choices to improve the model’s capacity to generalize beyond the training domain. We further propose a two-stage training strategy with an adapter-like rainfall encoder that enables efficient adaptation to previously unseen rainfall intensities while preserving spatial generalizability.The resulting model predicts maximum water depths and is evaluated on unseen catchments in Austria and Denmark, as well as on artificially modified terrains such as dikes. Our results demonstrate that combining a large training dataset with upstream-aware terrain representations and a generalization-oriented architectural design enables accurate flood predictions across unseen terrains and rainfall scenarios, reaching a RMSE below $0.05$ and CSI$_{0.3}$ of up to $0.91$ in test cases, outperforming previous models in both prediction accuracy and Critical Success Index.- 4 April 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessDeep Learning with Upstream-Aware Terrain Representations and Rainfall Adapter for Generalizable Pluvial Flood Prediction
Deep learning-based surrogate models can substantially accelerate pluvial flood simulations. However, most existing approaches struggle to generalize across heterogeneous terrains or modified topographies, limiting their applicability for real-world flood risk assessment unless retrained for each specific terrain.In this work, we investigate how deep learning flood surrogates can be designed to achieve robust spatial generalization across previously unseen terrain configurations without retraining. We train models on diverse terrain and introduce upstream-aware input maps that propagate terrain properties into downstream regions, allowing the model to capture long-range spatial dependencies and supporting the learning of more generalizable spatial features. In addition, we add architectural design choices to improve the model’s capacity to generalize beyond the training domain. We further propose a two-stage training strategy with an adapter-like rainfall encoder that enables efficient adaptation to previously unseen rainfall intensities while preserving spatial generalizability.The resulting model predicts maximum water depths and is evaluated on unseen catchments in Austria and Denmark, as well as on artificially modified terrains such as dikes. Our results demonstrate that combining a large training dataset with upstream-aware terrain representations and a generalization-oriented architectural design enables accurate flood predictions across unseen terrains and rainfall scenarios, reaching a RMSE below $0.05$ and CSI$_{0.3}$ of up to $0.91$ in test cases, outperforming previous models in both prediction accuracy and Critical Success Index.