Levees are critical adaptation measures for mitigating the escalating flood risks posed by intensifying climatic extremes and rapid urban expansion into flood-prone areas. However, the implementation of these measures is often constrained by administrative boundaries and financial limitations, which confine adaptation efforts to predefined jurisdictions. These constraints result in adaptation gaps that disproportionately affect communities beyond protected zones and exacerbate inequalities in flood risk distribution. Our study integrates hydrodynamic and economic modeling to evaluate the magnitude, spatial distribution, and economic losses associated with levee-based flood protection strategies in Surat, a coastal city frequently exposed to severe riverine flooding. The findings indicate that levees designed to safeguard administrative boundaries can inadvertently intensify flood risks for unprotected communities and infrastructure by altering hydrodynamic conditions. Specifically, 100-year flood damages and extents increase significantly due to levee-induced changes. Our preliminary results highlight the spatially widespread nature of flood impacts, unaccounted costs, and the potential for increased socioeconomic inequities. These findings emphasize the need for a systems-based approach to flood management that considers the interconnectedness of river systems and promotes equitable sharing of flood risks across jurisdictions.

How to cite: Kumar, A. and Bhatia, U.: Flood risk redistribution due to gaps and constraints in adaptation strategies, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-15089, https://doi.org/10.5194/egusphere-egu25-15089, 2025.

Feature selection is an essential step in the development of empirical flood damage models based on machine learning techniques. So far, most models of this type were developed using data from a single region or country, and few of them utilize harmonized transboundary datasets. Here, we have harmonized 38 variables present in the datasets of three flood damage surveys conducted in Germany (n = 516), the Netherlands (n = 409) and Belgium (n = 320) after the 2021 mega-floods in Europe. After performing data imputation and multicollinearity check, we used linear and non-linear machine learning algorithms to assess permutation importance and identify features most influencing flood damage. The results of the four models suggest that besides the hazard variables such as water depth and human stability, the location of the heating system (in the basement or at a higher floor) appears among the topmost important features for both building and contents damage.

Subsequently, we did an analysis for a low and high range of water depths using the median value (0.6 m) as splitting criteria. In the lower range, for both types of damage, water depth appears to be the dominating driver, and specifically for the building damage, it exceeds by far the importance of any other variable. In contrast, for water depths above 0.6 m other factors outweigh water depth. In the case of content damage, building footprint area becomes the most important factor across all the models. For the building damage some hazard (e.g. human stability), exposure (e.g. building size) and vulnerability (e.g. hazard knowledge) variables have a comparable importance with that of water depth. Hence, our results show that multivariable models appear particularly necessary for modelling flood damage induced by high and extreme hazard conditions.

How to cite: Avila, M. P., Rodriguez Castro, D., Endendijk, T., Lisa, D., Ravikumar, G., Erpicum, S., Thieken, A., Aerts, J., Heidi, K., and Benjamin, D.: Flood damage in the residential sector: on the value of transnational datasets for robust feature selection, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-18240, https://doi.org/10.5194/egusphere-egu25-18240, 2025.

The Indian Himalayan region, distinguished by its ecological sensitivity and dynamic topography, suffers significant losses annually due to frequent natural disasters like landslides, earthquakes, and floods. Estimating loss and damage (L&D) is one of the most important tools for disaster risk management. It provides information on post-disaster recovery, resource allocation, redevelopment/rehabilitation project prioritization, and compensations to the affected communities. Calculating the impact of a disaster and developing long-term recovery plans for the Himalayan community specific to the region's unique urban and rural contexts require evaluating and prioritizing the indicators of economic loss and damage (ELD) and non-economic loss and damage (NELD). This study focuses on Udham Singh Nagar and Nainital districts of Uttarakhand, with a structured approach to identify, prioritize, and validate relevant indicators for multi-hazard loss and damage (MH L&D) calculation. Comprehensive datasets from the Uttarakhand Disaster Risk Atlas and government reports are used, including socioeconomic, environmental, infrastructure, and hazard-specific information for earthquakes, landslides, and floods. ELD and NELD indicators are processed and prioritized using a mixed statistical approach that includes principal component analysis (PCA) and the covariance matrix. This method's successful reduction of data dimensionality while maintaining important information made identifying high-priority indicators possible. To direct focused actions in the rural and urban settlements of the Himalayan region, these indicators were then rated. The HAZUS model—a standardized instrument for hazard loss estimation— guarantees the prioritized indicators' validation. Due to varying socioeconomic dynamics, exposure levels, and vulnerabilities, the study found notable differences in priority indicators across rural and urban locations. The findings underscore the importance of region-specific, hazard-sensitive prioritization frameworks for effective loss and damage assessment and disaster risk reduction (DRR). By highlighting the interplay between ELD and NELD indicators across multiple hazards, this study provides a valuable tool for policymakers, planners, and disaster management agencies to target investments in the required sector of the community for their rapid post-disaster long-term recovery. The validated indicators can serve as a baseline for future MH L&D assessments in similar geographies and Gram Panchayat Development Plans (GPDP).

How to cite: Goyal, S. and Mukherjee, M.: ELD and NELD Prioritization for Multi-Hazard Loss and Damage in Rural and Urban Areas of Uttarakhand’s Himalayan Districts, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-6259, https://doi.org/10.5194/egusphere-egu25-6259, 2025.

With our work, we’re proposing an approach to damage modeling oriented to the potential damage evaluation, with the aim to use it to address the prevention strategies in a more efficient way.

The methodology will be presented on a case study developed in the Italian region of Tuscany. For this application, our data collection on perceived damage, made up of claims compiled by citizens in the aftermath of relevant flood events, has been enriched with new data to cover the whole set of national state of emergency for Tuscany in the period 2013/2023.

Claims have been geolocalized and extracted on the plane areas of the region. We came up with more than 10800 points, providing a picture of where damage occurred, the declared economic losses and the areas affected by more than one event.

This dataset has been later adopted to train a machine learning model which combines the occurred damages, the characteristics of the territory (obtained from digital terrain model and other open data) and the communities’ social variables primarly derived from national census. Instead of superpose hazard and exposure, we have been working combining data sources which are different for origine, scale and semantic area in a big database to be provided to the algorithms.

We are here presenting the results of our work as well as the lesson learned in the modeling procedure.

How to cite: Zambrini, F., Nava, E. M., and Mendui, G.: Damage susceptibility in Italy: a case study for Tuscany, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-19909, https://doi.org/10.5194/egusphere-egu25-19909, 2025.

The desiccation of Utah's Great Salt Lake (GSL) poses significant health risks, particularly for vulnerable populations. This study examines how the diminishing GSL, exacerbated by anthropogenic changes, affects community mental health. Reduced water inflow has exposed the lakebed, increasing airborne particulate matter and dust storms, which impact air quality. By integrating diverse datasets spanning from 1980 to present—including in-situ measurements, satellite imagery, and reanalysis products—this study synthesizes hydrological, atmospheric, and epidemiological variables to comprehensively track the extent of the GSL’s surface water, local air quality fluctuations, and their effects on community mental health. The findings indicate a clear relationship between higher pollution days and more severe depressive symptoms. Specifically, individuals exposed to ~ 22 days with PM2.5 levels above the World Health Organization's 24-hour guideline of 15 μg/m³ were more likely to experience severe depressive symptoms. Our results also suggest that people experiencing more severe depression not only face a higher number of high-pollution days but also encounter such days more frequently. The study highlights the interconnectedness of poor air quality, environmental degradation and mental health emphasizing the need for more sustainable economic growth in the region.

How to cite: Neelam, M., Bhui, K., and Freitag, B.: Diminishing Waters: The Great Salt Lake's Desiccation and Its Mental Health Consequences, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-20664, https://doi.org/10.5194/egusphere-egu25-20664, 2025.

Climate change poses a major challenge to the insurance industry, highlighting the need for sustainable crop insurance programs to protect food production in developing countries amid increasing climate risks. Insurance plays a key role in advancing SDGs 1 (no poverty), 2 (zero hunger), and 13 (climate action). However, the short- and long-term impacts of climate-driven extreme weather events remain insufficiently understood. This study maps major climate threats to crop production in Brazil and examines the influence of extreme weather on price adjustments and insurance uptake. The research used a database of the Brazilian Program of Subsidies for Rural Insurance Premium (PSA) with 1.5 million policies and claims from 2006 to 2023 and meteorological daily data from the Brazilian Daily Weather Gridded Data (BR-DWGD) from 1991 to 2024, both aggregated at municipality level. Four perennials and non-perennials crops were observed as the most insured: soybeans (47%), maize second cycle (13.5%), wheat (8.7%), and grapes (8.9%). Moreover, the most critical hazards were droughts (43.1%), hail (34.4%), frost (10.3%), excessive rainfall (8.2%), floods (1.3%), cold winds (1.0%), and temperature variation (0.3%). In 2021, claim payments reached a historic high, totaling nearly 12 billion BRL, which was unprecedented when compared with the baseline annual values ranging from 0.043 to 2 billion BRL. Two critical periods significantly impacted crop production in 2021. The first occurred between March and April (Austral fall), with severe droughts and frost events affecting maize second cycle in southern and central-western states and wheat in the south. The second critical moment was between August and October (end of Austral spring and beginning of Austral summer), in which major droughts affected soybean production in the states of south, southeast, central west and northeast. The impact on only three crops explain the expressive increase in claim payments. From 2019 to 2023, soybean prices revealed significant evidence of weather shock impacts, as a major driver of premium rates increase. Rates increased by 38%, growing from 344.33 BRL/ha in 2021 to 562.12 BRL/ha in 2022. The insurance uptake increased 11% in 2022, growing from 188,179 to 212,839 policies and had a dramatic decrease of 72% in the following year. The analysis of insurance and weather data highlights significant impacts of increased climate-related stress on Brazil's insurance industry. Initially, unprecedented extreme events tend to drive an increase in insurance uptake. In response, insurance companies often raise premium rates to maintain a balanced ratio between revenue and payouts. However, this adjustment can lead to a subsequent decline in insurance uptake. Climate shocks may have prolonged effects on insurance, potentially undermining the financial sustainability of farmers and their capacity to recover from economic losses. Thus, these dynamics underscore the need for adaptive strategies to ensure resilience in both the insurance sector and agricultural systems.

How to cite: Benso, M. R., Silva, G. M. E., Silva, P. G. G. D., and Mendiondo, E. M.: Multidimensional indicators of sustainability of crop insurance under increased climate-stress, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-13435, https://doi.org/10.5194/egusphere-egu25-13435, 2025.

Integrated flood risk management requires an extension from hazard to risk analysis and an involvement of various stakeholders including the general public. Since no standard protocols for collecting data about flood-affected societies are in place, post-disaster surveys have been initiated to gain information from affected residents and companies. Using the most damaging flood events that have occurred in Germany since 2000 as examples, the lecture will address how data collected from flood-affected people have been used a) to develop and improve loss models, b) to better understand how and why people adapt to flood risk, c) to evaluate how people respond to warnings, d) to provide insights into flood-related health impacts and e) to comprehend how people recover from flood impacts. Since flood processes in Germany between 2002 to 2024 differed considerably, it will be addressed how much the flood type – in particular slow-onset river flooding, flash floods and pluvial floods – influence impacts and coping mechanisms. Research outcomes have informed flood early warning systems, risk communication and recovery programs in Germany and beyond. However, surveying or interviewing flood-affected people might also put an additional burden on them. Hence, the lecture will discuss some ethical considerations about collecting data in (highly) affected areas as well as some pros and cons of cross-sectional versus longitudinal survey designs. Finally, transfer to other regions and hazards will be highlighted.

How to cite: Thieken, A.: More than two decades of post-disaster household surveys to improve flood risk management, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-6429, https://doi.org/10.5194/egusphere-egu25-6429, 2025.

Reliable road infrastructure is vital for daily commuters and economic activities in the UK, yet it faces growing flood risks due to climate change. Effective flood risk management requires an integrated approach that includes pre-disaster traffic flow modelling, direct damage estimation, disruption and recovery analysis to quantify systemic failure impacts. Indirect costs from traffic disruptions are frequently oversimplified, often estimated as multipliers of direct damages. While traffic flow rerouting models are applied in current research, they often overlook critical factors, such as traffic flow constraints and road capacity limitations, instead assigning origin-destination flows to least-cost paths without accounting for congestion. Moreover, the recovery process, which is critical for understanding how restored road and bridge capacities reduced isolated flows and indirect damages, is rarely modelled.

To address these gaps, we developed an open-source modelling framework for Great Britain that integrates a process-based flow model with a stress-testing model to assess road flood damages. Our framework starts with simulating passenger-to-work flows at a national scale by modelling the lifeline connections between physical road networks and demographic factors (e.g., population and economic activities). The flow model employs an iterative approach to simulated congested equilibrium flow assignments, dynamically accounting for road capacities and flow speeds until all traffic is accommodated without causing overflow.

We stress-tested the road networks using 18 historical UK flood events and one synthetic flood event. To model flood-induced disruptions, we developed a speed-flood depth function that restricts maximum flow speeds on flooded roads based on floodwater depth. We applied 30cm and 60cm separately in disruption analysis as threshold for road closure for uncertainty analysis. In each scenario, flood impacts on traffic flows were evaluated by comparing edge flows under floods with those under base flow condition. Direct damages were calculated using generalised damage curves (i.e., function to estimate damage fractions based on floodwater depth), and cost functions (i.e., function to estimate unite asset cost, million £/length or area) for different road types (e.g., bridges, tunnels, ordinary roads) and flood types (e.g., surface floods, river floods). Indirect damages were quantified by calculating rerouting costs due to road closures, including additional fuel costs, tolls, and time-equivalent costs.

We introduced a novel recovery analysis to dynamically evaluate indirect damages by designing various road capacity recovery rates, accounting for road types and damage levels. The recovery process identifies disrupted flows resulting from missing routes or reduced speeds, and reallocates these flows as road capacities are restored on a daily basis. The analysis captures the evolving number of isolated flows, rerouting costs and asset repair costs, offering a more realistic representation of dynamic indirect damages.

Overall, this research advances large-scale flow modelling by integrating capacity constraints, disruption dynamics, and recover processes. It provides actionable insights to enhance the resilience of the UK’s road infrastructure. The framework can be adapted to contexts beyond UK, different spatial scales, multi-modal transport systems, and multi-hazard scenarios, supporting more comprehensive risk assessments and decision-making.

How to cite: Li, Y., Pant, R., Russell, T., Thomas, F., Hall, J., and Oldham, P.: A Process-based Flow Model for Assessing Direct and Indirect Damages to Flooded Roads in Great Britain, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-9365, https://doi.org/10.5194/egusphere-egu25-9365, 2025.

The increasing frequency and intensity of tropical cyclones, driven by climate change, pose significant risks to global supply chains, amplifying economic vulnerabilities. This study explores how interactions between multiple extreme events influence the propagation of indirect economic costs, focusing on the compounding effects that arise within interconnected systems. Leveraging a combination of the CLIMADA risk modeling platform and the ARIO  indirect impact economic model, we generate synthetic ensembles of tropical cyclones. These simulations allow us to analyze direct and indirect economic impacts at global and regional scales.

Our results reveal that compounding events can, in some cases, mitigate indirect losses. This effect arises from the accumulation of reconstruction demand, which stimulates production across sectors, particularly those heavily involved in rebuilding, such as construction and manufacturing. The interplay between reconstruction demand and overproduction mechanisms creates a virtuous cycle, accelerating recovery and offsetting consequent losses.

However, the observed mitigation is highly dependent on the underlying modeling assumptions and sectoral resolution of the modeled economy. Indeed, some adverse indirect economic consequences only emerge when employing economic data with a higher granularity of sectors. When such higher granularity of sectors is combined with less optimistic assumptions on adaptation capacity, not only does the mitigation effect disappear, but observed outcomes show significantly aggravated indirect losses.

This study underscores the complexity of modeling compounding risks and highlights the importance of carefully chosen parameters and granularity of economic data, as qualitatively different results can emerge. In this context, ARIO serves as an effective tool for exploring the drivers of indirect economic impacts from extreme events, providing valuable insights to guide and enhance more advanced modeling approaches.

How to cite: Juhel, S., Stalhandske, Z., Viguié, V., and Bresch, D. N.: Understanding the interactions of tropical cyclones in global supply chains, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-8509, https://doi.org/10.5194/egusphere-egu25-8509, 2025.

With globally interconnected economies through supply chains, the economic impacts of flooding—one of the most devastating natural disasters—pose significant concerns for both direct flood-affected countries and their trade partners. This underscores the need for a global assessment of these direct economic impacts and their potential propagation to develop flood-resilient supply chains on a global scale. Here, to assess the generic global flood risks, we evaluate direct economic losses across different sectors and propose indicators for assessing the indirect impacts of flood propagation through international trade. We demonstrate that the estimated global annual economic loss across agricultural, industrial, and service sectors is US$194 billion. China, India, the USA, Indonesia, and Egypt are significant sources of flood-related risks due to their considerable direct economic losses and diverse export partners, collectively accounting for more than 50% of the global direct economic loss. Meanwhile, emerging and developing countries in Asia and Africa and some developed countries with concentrated imports from high-risk-giving countries show significant potential to be affected by flood impacts indirectly; the relevance of indirect risk to these countries differs from the sector. These findings highlight the importance of a sector-wise assessment of flood economic impacts and their potential propagation via trade. Therefore, the assessment methods and indicators developed in this work will help inform policy and investment decisions for building flood-resilient supply chains and supporting business continuity plans.

How to cite: Mtibaa, S., Maeno, K., Islam, K., and Motoshita, M.: Assessing the global economic impacts of floods and their potential propagation through international trade, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-1874, https://doi.org/10.5194/egusphere-egu25-1874, 2025.

Natural hazards such as floods pose significant threats to communities worldwide, impacting lives, livelihoods, and infrastructure. Effective flood modelling in combination with accurate impact assessments are crucial for enabling timely interventions, effective emergency management, reducing disaster-related losses and enhancing societal resilience.

This work presents recent advancements in a comprehensive toolset for worldwide application, integrating advanced flood and impact modelling in a flexible way. Our tools quantify critical aspects of emergency planning and management, including the estimation of the number and location of (socially vulnerable) people affected by floods, the identification of individuals that should be warned and evacuated, the impact on populations in terms of accessibility, the identification of potential and available evacuation and provisioning routes before, during and after hazards, and the resultant time and distance for populations to the nearest shelter. We integrate social vulnerability and socio-economic characteristics into our analyses to prioritize socially vulnerable areas.

Our dynamic modeling of flood depths, extents, and durations utilizes the open-source flood model 'SFINCS'. The open-source impact model ‘RA2CE’ quantifies disruptions to road infrastructure due to any natural hazard, equipping road operators, spatial planners, and emergency managers with actionable information. The advancements were recently applied to a real-world case study in Mozambique in a participatory workshop with local stakeholders from the cities of Beira and Quelimane. 

This work provides applicable solutions for prevention, planning, early warning, response and recovery, extending across most disaster risk phases and significantly contributing to the Sendai Framework's goal of reducing disaster risk and losses in lives and livelihoods.

How to cite: Meijer, L., de Goede, R., Costa de Barros, E., Gunaratne, C., Hauth, M., van Marle, M., Nobre, G., and van Dongeren, A.: Comprehensive Flood and Impact Modeling: Advanced Quantitative Tools for Emergency Management and Societal Resilience, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-17951, https://doi.org/10.5194/egusphere-egu25-17951, 2025.