SBSeth Bryant

Seth Bryant

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Seth Bryant has a total of 39 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Data-driven flood classification to assess climate change effects on riverine compound floods in pre-Alpine catchments

    by Henrique M. D. Goulart, Lily-Belle Sweet, Axel Deijns, Seth Bryant, Benjamin Poschlod

    Flood events are some of the most impactful natural hazards globally, and in most cases, they result from combinations of different hydrometeorological drivers, defined as riverine compound floods. Climate change is projected to alter the frequency and interplay of these drivers, posing a challenge for flood risk assessment. In this study, we develop an unsupervised machine learning approach to identify distinct riverine compound flood types and to classify individual flood events based on hydrometeorological features. These classifications are then used to assess climate change effects on flood risk. We use hydrological large-ensemble simulations in four neighbouring Bavarian pre-Alpine catchments, encompassing historical and future periods. Results project systematic shifts in the frequency and magnitude of flood regimes under warming. The frequency of rainfall-driven and rain-on-snow floods increases with warming, rising by 22% and 8%, respectively, between +1 K and +4 K. Conversely, snowmelt-driven flood frequency declines substantially (approximately 42% between +1 K and +4 K), consistent with a diminishing role of snow in pre-Alpine flood generation. Frequencies of saturation excess floods, characterised by antecedent wet conditions, show declines of about 10%. For 100-year return levels, rainfall-driven events show the largest increases in flood magnitude, though with strong variability between catchments (1% to 48% between +1 K and +4 K). Conversely, the magnitude of 100-year return-level snowmelt-driven events decreases 10–13%. Seasonally, pre-Alpine flood risk shifts from a spring, snowmelt-dominated regime towards rainfall-driven flooding throughout the year, with implications for seasonal preparedness, infrastructure design, and forecasting systems. We compare our unsupervised machine learning approach with an established threshold-based classification method, with the two classifications converging for 64% of events. By identifying which flood types increase or decrease under warming scenarios, our compound flood classification framework supports the development of process-specific adaptation strategies.

  • Open Access

    Building-level exposure asset value modelling for Germany: an Ahrweiler case study

    by Aaron Buhrmann, Nivedita Sairam, Cecilia I. Nievas, James E. Daniell, Heidi Kreibich, Seth Bryant

    Verified, harmonised, open-access, object-level exposure data are essential for next-generation risk assessments, risk management, and impact-based forecasting. However, this object-level information is often proprietary, protected by regulation, poorly documented, and fragmented because data on building usage, structural type, or replacement costs are not readily available or not compiled in one dataset. To address this gap, we present an evaluation of exposure model workflows employing various disaggregation approaches and source data from cadastre-derived, crowd-sourced, national accounts, and fit-for-purpose datasets. Using information collected from one flood-affected region in Germany and a weighted scoring model, we evaluate the ability of each workflow to assign a building's economic sector and asset value against our hand-labelled benchmark dataset. Ultimately, we find an exposure model workflow disaggregating national accounts onto cadastre-derived building footprints slightly outperforms the other workflows owing mainly to its transparency and adaptability. However, we conclude that all but the land-use-derived workflow are defensible for object-level exposure modelling – when validated. While these findings are limited to one specific region, the workflows developed here provide the basis for broader evaluation across regions with different building stocks, economic structures, and data quality. Workflows like these enable the transparent, reproducible, and maintainable multi-sector object-level exposure modelling necessary for the next generation of risk analysis and impact forecasting.

  • Open Access

    Reply on RC1

    by Seth Bryant

    This study addresses the challenges of exposure modelling at the building or object-level in Germany, motivated by the need for harmonized, open-access data in next generation risk assessments. While aggregated exposure data suffice for many applications, detailed object-level data are increasingly essential for tasks such as local risk management and impact forecasting. However, this object-level information is often proprietary, protected by regulation, poorly documented, and fragmented because data on building usage, structural type, or replacement costs is often not readily available or not compiled in one dataset. To address this gap, we present an evaluation of potential exposure modelling frameworks utilizing various disaggregation approaches and source data from cadastre-derived, crowd-sourced, national accounts, and fit-for-purpose datasets. Using information collected from an area recently affected by a flood disaster and a weighted scoring model, we evaluate the ability of candidates to assign a building’s economic sector and asset value against our hand-labelled benchmark dataset. Ultimately, we find an exposure modelling framework disaggregating national-accounts onto cadastre-derived building footprints slightly out-performs other candidates owing mainly to its transparency and adaptability. However, we conclude that all but the land-use derived candidate are defensible exposure modelling frameworks — so long as some relevant validation is performed. The frameworks presented here enable the transparent, reproducible, and maintainable multi-sector object-level exposure modelling necessary for the next generation of risk analysis and impact forecasting.

  • Open Access

    Publisher Correction: Unveiling hidden risks in healthcare from flood-induced transportation disruption in Germany

    by Jonas Wassmer, Seth Bryant, Paul Schimansky, Lindsay T. Keegan, Maria Pregnolato, Jürgen Kurths, Norbert Marwan, Bruno Merz

  • Open Access

    Multi-sensor satellite AI database of observed flood extents 2001–2024

    by Seth Bryant, Saurabh Kaushik, Jonathan Sullivan, Beth Tellman

    High-resolution, event-scale flood footprints remain scarce, constraining flood hazard and risk mapping and limiting calibration and validation of hydrological and hydraulic models. We introduce a standardized, event-resolved, multi-sensor flood extent database that converts disparate Earth observation acquisitions into analysis-ready inundation footprints at national scale. The workflow begins by filtering 120,000 on-the-ground U.S. flood reports to retain only those with sufficient satellite coverage in space and time to support event-scale inundation mapping. For each event, we co-register and harmonize available MODIS, Landsat 8 and 9, Sentinel-2 MSI, Sentinel-1 SAR, and PlanetScope imagery, combining optical and radar observations to reduce cloud-driven gaps and improve spatial completeness. A vision-transformer-based geo-foundational model then converts each stacked scene into 10m water-segmentation masks, blending long-range attention with convolutional decoding. These sensor-level masks are then merged into maximum-inundation footprints, yielding roughly 30,000 high-quality flood event maps across the United States. The database and companion code are being released openly to support flood dynamics research, including benchmarking ML flood mapping, validating and calibrating hydrodynamic models, and providing assimilation-ready inundation observations. This talk will summarize the methods, show example flood maps, and highlight potential applications.

  • Open Access

    Comment on egusphere-2025-5172

    by Aaron Buhrmann, Cecilia I. Nievas, Nivedita Sairam, James E. Daniell, Heidi Kreibich, Seth Bryant

    This study addresses the challenges of exposure modelling at the building or object-level in Germany, motivated by the need for harmonized, open-access data in next generation risk assessments. While aggregated exposure data suffice for many applications, detailed object-level data are increasingly essential for tasks such as local risk management and impact forecasting. However, this object-level information is often proprietary, protected by regulation, poorly documented, and fragmented because data on building usage, structural type, or replacement costs is often not readily available or not compiled in one dataset. To address this gap, we present an evaluation of potential exposure modelling frameworks utilizing various disaggregation approaches and source data from cadastre-derived, crowd-sourced, national accounts, and fit-for-purpose datasets. Using information collected from an area recently affected by a flood disaster and a weighted scoring model, we evaluate the ability of candidates to assign a building’s economic sector and asset value against our hand-labelled benchmark dataset. Ultimately, we find an exposure modelling framework disaggregating national-accounts onto cadastre-derived building footprints slightly out-performs other candidates owing mainly to its transparency and adaptability. However, we conclude that all but the land-use derived candidate are defensible exposure modelling frameworks — so long as some relevant validation is performed. The frameworks presented here enable the transparent, reproducible, and maintainable multi-sector object-level exposure modelling necessary for the next generation of risk analysis and impact forecasting.

  • Open Access

    Comment on egusphere-2025-5172

    by Aaron Buhrmann, Cecilia I. Nievas, Nivedita Sairam, James E. Daniell, Heidi Kreibich, Seth Bryant

    This study addresses the challenges of exposure modelling at the building or object-level in Germany, motivated by the need for harmonized, open-access data in next generation risk assessments. While aggregated exposure data suffice for many applications, detailed object-level data are increasingly essential for tasks such as local risk management and impact forecasting. However, this object-level information is often proprietary, protected by regulation, poorly documented, and fragmented because data on building usage, structural type, or replacement costs is often not readily available or not compiled in one dataset. To address this gap, we present an evaluation of potential exposure modelling frameworks utilizing various disaggregation approaches and source data from cadastre-derived, crowd-sourced, national accounts, and fit-for-purpose datasets. Using information collected from an area recently affected by a flood disaster and a weighted scoring model, we evaluate the ability of candidates to assign a building’s economic sector and asset value against our hand-labelled benchmark dataset. Ultimately, we find an exposure modelling framework disaggregating national-accounts onto cadastre-derived building footprints slightly out-performs other candidates owing mainly to its transparency and adaptability. However, we conclude that all but the land-use derived candidate are defensible exposure modelling frameworks — so long as some relevant validation is performed. The frameworks presented here enable the transparent, reproducible, and maintainable multi-sector object-level exposure modelling necessary for the next generation of risk analysis and impact forecasting.

  • Open Access

    Comment on egusphere-2025-5172

    by Aaron Buhrmann, Cecilia I. Nievas, Nivedita Sairam, James E. Daniell, Heidi Kreibich, Seth Bryant

    This study addresses the challenges of exposure modelling at the building or object-level in Germany, motivated by the need for harmonized, open-access data in next generation risk assessments. While aggregated exposure data suffice for many applications, detailed object-level data are increasingly essential for tasks such as local risk management and impact forecasting. However, this object-level information is often proprietary, protected by regulation, poorly documented, and fragmented because data on building usage, structural type, or replacement costs is often not readily available or not compiled in one dataset. To address this gap, we present an evaluation of potential exposure modelling frameworks utilizing various disaggregation approaches and source data from cadastre-derived, crowd-sourced, national accounts, and fit-for-purpose datasets. Using information collected from an area recently affected by a flood disaster and a weighted scoring model, we evaluate the ability of candidates to assign a building’s economic sector and asset value against our hand-labelled benchmark dataset. Ultimately, we find an exposure modelling framework disaggregating national-accounts onto cadastre-derived building footprints slightly out-performs other candidates owing mainly to its transparency and adaptability. However, we conclude that all but the land-use derived candidate are defensible exposure modelling frameworks — so long as some relevant validation is performed. The frameworks presented here enable the transparent, reproducible, and maintainable multi-sector object-level exposure modelling necessary for the next generation of risk analysis and impact forecasting.

  • Open Access

    Building-level exposed asset value modelling for Germany

    by Aaron Buhrmann, Cecilia I. Nievas, Nivedita Sairam, James E. Daniell, Heidi Kreibich, Seth Bryant

    This study addresses the challenges of exposure modelling at the building or object-level in Germany, motivated by the need for harmonized, open-access data in next generation risk assessments. While aggregated exposure data suffice for many applications, detailed object-level data are increasingly essential for tasks such as local risk management and impact forecasting. However, this object-level information is often proprietary, protected by regulation, poorly documented, and fragmented because data on building usage, structural type, or replacement costs is often not readily available or not compiled in one dataset. To address this gap, we present an evaluation of potential exposure modelling frameworks utilizing various disaggregation approaches and source data from cadastre-derived, crowd-sourced, national accounts, and fit-for-purpose datasets. Using information collected from an area recently affected by a flood disaster and a weighted scoring model, we evaluate the ability of candidates to assign a building’s economic sector and asset value against our hand-labelled benchmark dataset. Ultimately, we find an exposure modelling framework disaggregating national-accounts onto cadastre-derived building footprints slightly out-performs other candidates owing mainly to its transparency and adaptability. However, we conclude that all but the land-use derived candidate are defensible exposure modelling frameworks — so long as some relevant validation is performed. The frameworks presented here enable the transparent, reproducible, and maintainable multi-sector object-level exposure modelling necessary for the next generation of risk analysis and impact forecasting.

  • Open Access

    Supplementary material to "Building-level exposed asset value modelling for Germany"

    by Aaron Buhrmann, Cecilia I. Nievas, Nivedita Sairam, James E. Daniell, Heidi Kreibich, Seth Bryant