Presentation type:
NH – Natural Hazards

EGU24-4270 | Orals | NH2.1 | Highlight | Plinius Medal Lecture

Hazard forecasting: is it a matter of time? 

Jacopo Selva

Hazard models aim at making explicit our forecasting capability about future potentially adverse natural events. Hazard events are typically rare and not deterministically predictable, forcing hazard models to speak the language of certainty and uncertainty, that is, of probability. This is valid for any forecasting time window, from years to days/hours in the future (long- to short-term hazard), to the evaluation of the potential impact of an ongoing event in the next seconds/minutes/hours (warning/now-casting to urgent computing). Even though the definition of the target time window is driven by the users of the forecast (e.g. civil protections) and is not a scientific matter, the quantification of existing uncertainty given the time frame is certainly a scientific matter. Probabilistic hazard is commonly discussed mainly for long-term hazards, where large uncertainty dominates. In shorter-term forecasts, uncertainty may deacrease and practitioners are often tempted by simplified approaches that neglect uncertainty, like for eruption forecasting during volcanic crises, or for tsunami warning models after seismic or volcanic events. Nevertheless, uncertainty may still exist, and a rational scientific approach should let the results to speak about existing uncertainty, rather than to neglect it by definition. Is it possible to define a unified approach to probabilistic hazard entailing all time scales? The long-term integral hazard integrating all potential sources and generation/propagation conditions can be adapted to the different forecasting time windows, generating a unified framework in which the different time scales may feed to each other, producing homogeneous and easy-to-interpret results. This unified vision of hazard models, embracing long- to short-term hazard as well as warning and urgent computing models, is here discussed based on the recent advancements in models for volcanic, seismic and tsunami hazard and warning.

How to cite: Selva, J.: Hazard forecasting: is it a matter of time?, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-4270,, 2024.

The intensification of extreme precipitation in a warming climate has been shown in observations and climate models to follow approximately theoretical Clausius-Clapeyron scaling. However, larger changes have been indicated in events of short-duration which frequently trigger flash floods or landslides, causing loss of life. Global analyses of continental-scale convection-permitting climate models (CPCMs) and new observational datasets will be presented that provide the state-of-the-art in understanding changes to extreme weather (rainfall, wind, hail, lightning) and their compounding effects with global warming. These analyses suggest that not only warming, but dynamical circulation changes, are important in the manifestation of change to some types of extreme weather, which must be addressed in the design of new CPCM ensembles. We use our projections to provide the first analyses of impacts on infrastructure systems using a new consequence forecasting framework and show the implications for adaptation. It will be argued that a shift in focus is needed towards examining extreme weather events in the context of their ‘ingredients’ through their evolution in time and space. Coupled with exploration of their causal pathways, sequencing, and compounding effects – ‘storylines’ –, this can be used to improve both early warning systems and projections of extreme weather events for climate adaptation.

How to cite: Fowler, H.: Rapidly intensifying extreme weather events in a warming world: how important are large-scale dynamics in generating extreme floods?, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-22472,, 2024.

EGU24-10624 | ECS | Orals | NH10.1 | Highlight | NH Division Outstanding ECS Award Lecture

Advancing multi-(hazard)risk science: embracing complexity and cross-disciplinary collaborations 

Marleen de Ruiter

Recent disasters have demonstrated the growing challenges faced by society as a result of multi-hazards and compound events. The impacts of such disasters differ significantly from those caused by single hazard disasters: often the impacts of a multi-hazard disaster exceed those of the sum of the impacts of the individual hazards. Recognizing this complexity, the scientific community and international organizations, such as the UNDRR, have been advocating for a more integrated approach in multi-(hazard)risk research. This requires bridging across individual hazard types, but also learning from methodological advances made in neighbouring research fields such as the compound events community.

This talk aims to highlight recent advances in assessing the complexities of multi-(hazard)risk and discusses opportunities for further enhancing our modeling capabilities through multidisciplinary collaboration. A crucial challenge of modelling compound and multi-hazard risk, is that of the spatiotemporal dynamics of risk. This includes for example, an improved understanding of post-disaster recovery after multi-hazard disasters and the role of (changing) local contexts within which disasters take place such as the dynamics of socioeconomic vulnerability and the likelihood of post-disaster disease outbreaks. Embracing these challenges and opportunities can support more comprehensive and effective disaster risk management strategies in the future.

How to cite: de Ruiter, M.: Advancing multi-(hazard)risk science: embracing complexity and cross-disciplinary collaborations, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-10624,, 2024.

NH1 – Hydro-Meteorological Hazards

Introduction: The increasing number of heat events of summer 2003, 2010 and 2015 induced a rising impact on heat health morbidity and mortality in Germany. A considerable proportion of urban population is affected by thermal discomfort due to the urban heat island burden during summer, particularly vulnerable people at risk. As contribution to the National Adaptation Strategy to Climate Change a federal expert working group prepared and published ‘Recommendations for Action for the Preparation of Heat Action Plans to Protect Human Health’ in 2017, on the basis of the 2008 WHO Heat Health Guidance. A first country-wide project had been conducted in Germany between 2019 and 2023 which investigated the status and impact of current or planned Heat Health Action Plans (HHAP), and adaptation measures appropriate to protect and prevent human health. One aim of this study was to conduct an online survey on the awareness and degree of use of the 2017 recommendations and the development and implementation of HHAP.


Methods: The online survey questionnaire referred to climate change, heat and health aspects. The survey had been conducted in May/June 2020 to assess the current state of affairs as well as factors of success and barriers in the development and implementation of HHAP, with a particular focus on municipal administrations. Therefore, various networks of national, regional and local environment and health administrations, as well as stakeholders had been invited via an Email, web and social media campaign to participate anonymously. 


Results: The study fell in the period of increasing incidence of COVID19 pandemic, which influenced and limited the responses from the public health sector country-wide. Nonetheless, the online survey had been conducted in May/June 2020. In total 116 questionnaires had been responded, mainly by participants from the environment sector (53%), and 32% from the public health sector. More than half of the respondents (n=67) indicated to be aware of the 2017 recommendations (very well-known at the federal state level: 90%, at county level known by 43%). The recommended health-related adaptation measures were appreciated very or helpful by 81%. Respondents from large cities (> 100 000 inhabitants) were main contributors to the study (41 of 81 large cities replied, while 34 rural counties responded). So far, no federal state responded, but four municipalities and one county indicated to start planning activities on HHAP.


Conclusion: Recent heat events clearly indicate the demand to protect public health against heat extremes in Germany. The results of this first country-wide survey show on the one hand that currently HHAP are being developed rarely. On the other hand, results reflect that first HHAP actions turned into force after 2017 recommendation had been launched. Federal states and municipalities feel motivated and act responsible to take health-related adaptation measures in advance, such as HHAP, to prevent heat-related illnesses and deaths sustainable. Policy and decision makers must develop structures and regulations that anchor HHAP as a nationwide instrument to be established timely. 

How to cite: Mücke, H.-G.: Climate change induced heat health challenges – a study on the development of health-related adaption measures in Germany, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-1224,, 2024.

EGU24-1262 | ECS | Posters on site | NH1.1

A systematic review on climate change, trending outcomes for the care of older adults, and financial expenditure 

Keriin Katsaros and Jo-Ting Huang-Lachmann

As global populations age, life expectancy increases, and the impacts of climate change intensify, there is a corresponding rise in financial expenditure associated with both extreme climate events and a growing demand for long-term healthcare. There is a gap in research showing how climate change, the care of aging populations, and financial expenditure are interconnected.  A systematic review of literature is carried out to identify the interrelationships and to explore existing trends and evidence on senior care during the time of climate change. This work is important to identify barriers and priority action areas for adaptation, mitigation, and future planning to increase health gain and achieve positive economic outcomes.

The systematic review of existing peer-reviewed publications is carried out by following PRISMA guidelines.  The methodology is guided by recommendations from Cochrane and in the WHO Guidance on Research Methods for Health Emergency and Disaster Risk Management. Three electronic databases have been searched (PubMed, Web of Science, and Scopus) to assess available literature on climate change, the care of older persons, and financial expenditure. A risk of bias assessment is completed using CASP checklists.

Initial results of this review have identified significant adverse impacts for older society members including decreased physical and mental health during extreme weather events resulting in increased medical and care costs. Preliminary results also highlight a need for energy-efficient built environments, clean and affordable energy sources to overcome energy poverty, and a new way of rethinking how we care for senior society members, including increased support from families and communities.

This research aims to contribute to increasing transdisciplinary knowledge from the fields of health and care, energy, and climate change to create societies for older people that are friendly, affordable, and resilient to the adverse effects of climate change. The results aim to advance the transdisciplinary knowledge of climate services, health, and energy economics; co-creating synergies and actionable solutions; and working with societal actors for implementation, transfer, and upscaling of research.

How to cite: Katsaros, K. and Huang-Lachmann, J.-T.: A systematic review on climate change, trending outcomes for the care of older adults, and financial expenditure, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-1262,, 2024.

EGU24-1762 | ECS | Orals | NH1.1

Assessing the impact of mobility on heat exposure 

Guo-Shiuan Lin and Gabriele Manoli

Higher temperatures are linked to elevated mortality risk and reduced economic productivity. In urban centers, human mobility and the urban heat island effect generally result in higher population density and increased temperatures. Yet, existing studies on urban heat exposure rely on static residential population data, thus neglecting population dynamics and their covariation with the spatial distribution of urban temperatures. Here, we evaluate how seasonal and daily mobility modify heat exposure and risk across 80 European cities by combining monthly daytime and nighttime population estimates with high-resolution urban climate simulations. Our findings reveal that, on a daily scale, mobility increases population and summer heat exposure in most cities due to daily commuting behavior. Conversely, commuting to warmer city centers (from colder rural areas) may be advantageous in winter. On seasonal timescales, summer populations decrease in most cities as urban residents travel outward for holidays but heat exposure increases significantly in touristic destinations where population peaks during the June-August period. These results emphasize the significant impact of mobility on the space-time distribution of heat exposure in cities and offer valuable insights for mitigating temperature-related risks in diverse climatic and urban contexts.

How to cite: Lin, G.-S. and Manoli, G.: Assessing the impact of mobility on heat exposure, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-1762,, 2024.

EGU24-1899 | ECS | Posters on site | NH1.1

Heatwave social vulnerability index validated on mortality data in Europe 

Benedetta Sestito, Lena Reimann, Maurizio Mazzoleni, Wouter Botzen, and Jeroen Aerts

Climate change projections underscore an imminent temperature escalation of 1.1°C to 6.4°C above the 1990 baseline by 2100, leading to heatwaves of higher frequency, intensity, and duration. In the last decades, Europe has experienced severe heatwave events, such as the catastrophic one in summer 2003 that claimed over 70,000 lives. Thus, it is of pivotal importance to better understand the drivers of heatwave impacts to promote effective adaptation and mitigation strategies. While many studies have been carried out to explore the social vulnerability to heat-related impacts at local and regional scales, a large-scale continental analysis is still missing. This study aims at exploring the temporal and spatial dynamics of social vulnerability to heatwaves in Europe. This will be achieved by developing a dynamic spatial and temporal social vulnerability index for heatwaves in Europe. The index is validated against impact metrics such as mortality, which remains underexplored in large-scale assessments. In particular, a regression analysis between heatwave mortality and socio-economic and demographic factors, including population changes, income variations, and alterations in social infrastructure is carried out to assess the degree to which these factors are associated with heatwave mortality. The regression coefficients serve as the weights in the composite social vulnerability index. Such validation enhances the credibility, accuracy, and applicability of social vulnerability indices, bridging the divide between theoretical assessments and real-world consequences. Our empirical results indicate that diverse socio-economic and demographic variables exhibit distinct correlations with heatwave mortality. Consequently, an index incorporating ad hoc weighting of its constituent terms more effectively captures the social vulnerability component to heatwaves. This research provided new insights to better understand social vulnerability to heatwaves and allow better-informed decision-making to enable the development of resilient communities. Moreover, our findings advanced the understanding of heat risks in the broader context of escalating climate change challenges across Europe.

How to cite: Sestito, B., Reimann, L., Mazzoleni, M., Botzen, W., and Aerts, J.: Heatwave social vulnerability index validated on mortality data in Europe, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-1899,, 2024.

EGU24-2014 | ECS | Posters on site | NH1.1

Developing a Multivariate System for Predicting and Mitigating the Health Effects ofHeat waves in Niterói, Rio de Janeiro 

Vitor Luiz Galves, Marcio Cataldi, and Joan Souza

This study aims to develop a heat wave forecasting system using a new multivariate index that encompasses hydration-related mitigation measures. Heatwaves have increasingly occurred with greater frequency and intensity in various regions worldwide, particularly in Europe and Asia since 1990, although they are not exclusive to these areas. The principal health effects of heatwaves on populations include heat-related illnesses and fatalities, cardiovascular and kidney diseases, as well as adverse reproductive effects. These detrimental impacts are widespread and commonly affect individuals aged 65 and above. Many nations have established metrics to assess the prevalence of this occurrence within their borders. These metrics typically utilize specific thresholds and/or temperature ranges at a height of 2 meters, which denote extreme percentiles of values from past records. While some of these metrics consider the persistence of the phenomenon, few take into account the relative humidity. It is noteworthy that, in most instances, the temperature thresholds lead to a linear escalation in conditions posing a risk to the population. This can result in a biased perception of the actual level of risk involved. To thoroughly evaluate the health hazards associated with heatwaves, it is essential to acknowledge the considerable variability in global climate, as well as the diverse responses of living organisms to extreme temperature and humidity conditions. Furthermore, factors such as individuals' gender, race, age, pre-existing medical conditions, and geographical location should be taken into account.This study is divided into several components to reach a comprehensive solution. The first step involves determining the monthly distribution curve of accumulated daily maximum temperatures for each grid point of the ERA 5 data. After completing this process, machine learning models must be developed to calibrate the temperature values to the percentile of the cumulative distribution. Subsequently, the temperature value exceeding 95% of the distribution will be applied to this coefficient Coef = (eTpe*Ur)/1000, where Tpe is the value of the distribution that exceeds 95% and Ur is the relative humidity. These adjusted values will then be used to compute the normalized index I=(Coef-0.022)/9.7, accounting for the exponential temperature increase and providing weightage to the relative humidity. Upon establishment of these functions, a time series of the index value will be generated. This value will be multiplied by the hours of the day during which the index deviates from zero, facilitating the evaluation of its correlation with hospitalization and mortality data related to diseases such as thrombosis, which may be linked to heat waves. The results of this phase will be presented at the Niterói region in Rio de Janeiro, Brazil, during the upcoming congress. Moreover, according to previous analyzes, since 2010 the frequency and intensity of heat waves have increased, being apparently modulated by Enso events and also by indices developed at LAMMOC/UFF related to anomalies of sea surface temperature of the Equatorial Atlantic Ocean and the South Atlantic Convergence Zone. Furthermore, the index data will subsequently undergo validation based on body water loss rates and their impact on blood viscosity fluctuations.

How to cite: Galves, V. L., Cataldi, M., and Souza, J.: Developing a Multivariate System for Predicting and Mitigating the Health Effects ofHeat waves in Niterói, Rio de Janeiro, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-2014,, 2024.

EGU24-3061 | ECS | Posters on site | NH1.1

Links between weather variability and Dengue outbreaks in Sao Paulo, Brazil 

Falak Naz, Júlia Araújo, Sheila Oliveira, Seyma Celina, Aleš Urban, and Jiří Černý

Mosquito-borne diseases are among the most dangerous threats for all people living in tropical areas. Previous research has shown that the highest incidence of mosquito-borne diseases is associated with a particular type of weather (usually wet and hot) as mosquitos’ activity and development are highly dependent on meteorological conditions. However, short-term associations (on the scale of days up to a few weeks) have been less understood.

In this study, we collected weekly data on the incidence of Dengue on a municipality level in the state of Sao Paulo, Brazil, 2016–2022, and matched it with ERA5-based weather variables (ambient temperature, relative humidity, wind speed and precipitation). We employed a multilevel meta-regression analysis to i) analyse the links between Dengue incidence and weather variability in, and ii) develop a model to predict a Dengue fever outbreak based on actual weather conditions and socioeconomic variables.

Our preliminary results suggest a significant association of a Dengue outbreak with above-average daily mean temperature and humidity, heavy rainfalls, and calm conditions in previous 2-6 weeks. Further analysis is needed to identify spatial differences in these patterns based on socioeconomic conditions.

How to cite: Naz, F., Araújo, J., Oliveira, S., Celina, S., Urban, A., and Černý, J.: Links between weather variability and Dengue outbreaks in Sao Paulo, Brazil, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-3061,, 2024.

EGU24-3081 | Orals | NH1.1

Developing a Multivariate System for Predicting and Mitigating the Health Impacts of Heatwaves 

Marcio Cataldi, Vitor Galvez, Victoria Gallardo Fernandez, Juan Pedro Montávez, Pedro Jiménez-Guerrero, Guillermo Felipe Lopez Sanchez, and Christian Jair Martínez Urrutia

The main purpose of this study is to develop a heatwave impact-based forecasting system using a new multivariate index, that also encompasses a mitigation action plan with hydration-related measures. Since 1990, heatwaves have become more frequent and intense in various regions worldwide, particularly in Europe and Asia. The principal health effects of heatwaves include organs' strain and damage, complications of cardiovascular and kidney diseases, as well as adverse reproductive effects. These detrimental impacts are widespread and commonly affect individuals aged 65 and above. Many nations have established metrics to assess the prevalence of this occurrence within their borders. These metrics typically use specific threshold values and/or ranges of the near-surface (2 m) air temperature, usually denoted by the extreme values from past records. To the best of our knowledge, only some of these metrics take into account the persistence of the phenomenon and few consider the relative humidity. It is noteworthy that in most of these metrics the temperature thresholds lead to a linear escalation of the conditions posing a risk to the population, which may lead to a misperception of the actual level of risk involved. To thoroughly evaluate the health hazards associated with heatwaves, it is essential to consider the climate variability and change at regional and local scales, as well as the diverse responses of living organisms to extreme (and long-lasting) temperature and humidity conditions. Factors such as individuals' sex, ancestry, age, pre-existing medical conditions, and geographical location should be considered too. The first step of this study consisted of the characterization of the monthly Cumulative Distribution Function of the daily maximum near-surface air temperature (TX) in summer, in recent climate. We used the ERA5-Land reanalysis dataset and performed the analysis for each grid point, considering 1960-1990 as baseline period.  Subsequently, in order to compute the index, the temperature values exceeding the 95th percentile (TX95p) were subjected to a normalized scaling function whose values grow exponentially with the magnitude of the temperature and also depend on the ambient relative humidity. The resulting index values range from 0 to 1, only being greater than zero when the temperature exceeds TX95p. To calibrate the index, we considered the hours of the day during which the index deviates from zero and its correlation with hospitalization and mortality data, mainly related to cardiovascular diseases such as thrombosis. The preliminary work concerned the Region of Murcia, in Spain. The index was validated in the period 2000-2022. Results show the sensitivity of the index, which displays its largest values in the summer of 2022, coinciding with the high number of heat-related deaths observed that year in Spain. Future research will be focused on index calibration and validation in other regions which are also subjected to extreme heat conditions.

How to cite: Cataldi, M., Galvez, V., Gallardo Fernandez, V., Montávez, J. P., Jiménez-Guerrero, P., Lopez Sanchez, G. F., and Martínez Urrutia, C. J.: Developing a Multivariate System for Predicting and Mitigating the Health Impacts of Heatwaves, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-3081,, 2024.

EGU24-3446 | Orals | NH1.1

Drivers of the time-varying heat-cold-mortality association in Spain 

Hicham Achebak, Grégoire Rey, Simon Lloyd, Marcos Quijal-Zamorano, Raúl Fernando Méndez-Turrubiates, and Joan Ballester

Background: A number of studies have reported reductions in mortality risk due to heat and cold over time. However, questions remain about the drivers of these adaptation processes to ambient temperatures. We aimed to analyse the demographic and socioeconomic drivers of the downward trends in vulnerability to heat- and cold-related mortality observed in Spain during recent decades (1980-2018).

Methods: We collected data on all-cause mortality, temperature and relevant contextual indicators for 48 provinces in mainland Spain and the Balearic Islands between Jan 1, 1980, and Dec 31, 2018. Fourteen contextual indicators were analysed representing ageing, isolation, urbanicity, heating, air conditioning (AC), house antiquity and ownership, education, life expectancy, macroeconomics, socioeconomics, and health investment. The statistical analysis was separately performed for the range of months mostly causing heat- (June-September) and cold- (October-May) related mortality. We first applied a quasi-Poisson generalised linear regression in combination with distributed lag non-linear models (DLNM) to estimate province-specific temperature-mortality associations for different periods, and then we fitted univariable and multivariable multilevel spatiotemporal meta-regression models to evaluate the effect modification of the contextual characteristics on heat- and cold-related mortality risks over time.

Findings: The average annual mean temperature has risen at an average rate of 0·36°C per decade in Spain over 1980-2012, although the increase in temperature has been more pronounced in summer (0·40°C per decade in June-September) than during the rest of the year (0·33°C per decade). This warming has been observed, however, in parallel with a progressive reduction in the mortality risk associated to both hot and cold temperatures. We found independent associations for AC with heat-related mortality, and heating with cold-related mortality. AC was responsible for about 28·6% (31·5%) of the decrease in deaths due to heat (extreme heat) between 1989-1993 and 2009-2013, and heating for about 38·3% (50·8%) of the reductions in deaths due to cold (extreme cold) temperatures. Ageing (ie, proportion of population over 64 years) attenuated the decrease in cold-related mortality.

Conclusion: AC and heating are effective societal adaptive measures to heat and cold temperatures. This evidence holds important implications for climate change health adaptation policies, and for the projections of climate change impacts on human health.

How to cite: Achebak, H., Rey, G., Lloyd, S., Quijal-Zamorano, M., Méndez-Turrubiates, R. F., and Ballester, J.: Drivers of the time-varying heat-cold-mortality association in Spain, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-3446,, 2024.

Humid heatwaves, defined as extreme temperatures combined with high relative humidity, can affect large populations and result in crop damage, causing public health emergencies and threatening food security. The consequences for society are even more severe when extreme rainfall follows a humid heatwave. There is evidence of the increasing occurrence of humid heatwaves—extreme rainfall compound events in several parts of the globe. Their compound impact depends on the response time, the statistical interdependency between the two interacting causal drivers, and their severity. While the correlation between temperature and rainfall tends to be negative at daily or monthly time scales, the correlation between short-duration rainfall extremes and high temperatures is often positive at shorter time scales during the summer. A positive (negative) correlation can result in a higher (lower) risk of compound heatwave-extreme rainfall events. Across the coasts, the dependence strengths between these two variables are often elusive due to the influence of large-scale atmospheric teleconnection. On a global scale, the statistical coupling between humid heatwaves and extreme rainfall events has not been investigated across the coasts. To fill this knowledge gap, this study provides an observational assessment of the compound interactions of summer heatwave amplitude (i.e., the peak temperature of the hottest day during the heatwave episode) and extreme rainfall (described by > 90th percentile threshold of daily rainfall magnitude) across 29 major coastal cities in the tropics (23.5°N - 23.5°S), subtropics (23.5°N - 35°N and 23.5°S - 35°S) and mid-latitudes (35°N – 60°N and 35°S - 60°S). It finds a significant (P < 0.05) increase in the frequency of compound humid heatwaves-extreme rainfall events in the past few decades, with a more robust increase over the northern hemisphere compared to the southern hemisphere. The mean response times between the heatwave amplitude and the peak rainfall tend to be shorter for the southern sub-tropics than the northern hemisphere sites, indicating a swift transition between two extremes in these regions. Using a multivariate probabilistic framework, we further demonstrate that a modest to substantial increase in heatwave amplitude in summer can enhance the rainfall extremes by 80%, with the most significant increase occurring in the sub-tropics. The findings reveal a strong coupling between humid heatwaves and extreme rainfall in sub-tropical climate regimes, contrasted by a relatively weak coupling across the tropics. Understanding the interactions between humid heatwaves and extreme precipitation across coastal megacities will help decision-makers and stakeholders to adapt and mitigate these compound hazards in densely populated settlements.

How to cite: Ganguli, P. and Merz, B.: A Global Assessment of Compound Humid Heatwaves-Extreme Rainfall in Major Coastal Cities, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-4088,, 2024.

EGU24-4858 | Posters on site | NH1.1

Significant reduction of unequal population exposure to climate extremes by achieving the carbon neutrality 

Jung Choi, Seok-Geun Oh, Min-Jee Kang, Sujong Jeong, Seung-Ki Min, Sang-Wook Yeh, Yeon-Hee Kim, and Seok-Woo Son

Climate extremes, such as hot temperature and heavy precipitation events, have devastating effects on human societies. As the planet warms, they have become more intense and more frequent. To avoid irreversible damage from climate extremes, many countries have committed to achieving net-zero anthropogenic carbon emissions, or carbon neutrality, by 2050s. Here, we quantify the impact of carbon neutrality on population exposure to climate extremes using multi-model projections from the Coupled Model Intercomparison Project Phase 6 (CMIP6) based on the Shared Socioeconomic Pathway (SSP)1-1.9 and SSP3-7.0 scenarios. It is found that the increasing exposure of the population to hot-temperature and heavy-precipitation extremes can be substantially reduced by 87–98% in the late 21st century by achieving carbon neutrality. The benefits of carbon neutrality are particularly pronounced in Africa and Asia. The potential benefits of carbon neutrality are also significant in North America, Europe, and Oceania, where a reduction in climate extremes is more than twice as important as population decline in reducing population exposure to climate extremes. These results provide important scientific support for ongoing efforts to achieve net-zero carbon emissions by 2050s to reduce potential climate risk and its inequity across continents.

How to cite: Choi, J., Oh, S.-G., Kang, M.-J., Jeong, S., Min, S.-K., Yeh, S.-W., Kim, Y.-H., and Son, S.-W.: Significant reduction of unequal population exposure to climate extremes by achieving the carbon neutrality, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-4858,, 2024.

EGU24-4898 | ECS | Orals | NH1.1

Assessment of humidity-inclusive heat index models over tropical megacities: Indonesia case 

Muhammad Rezza Ferdiansyah, Alberth Nahas, and Ardhasena Sopaheluwakan

One of the impacts on humans in megacities experiencing rapid urbanization is the increase in heat risk, primarily due to the urban heat island (UHI) phenomenon. One of the reasons for the intensification of UHI can be attributed to changes in land use and population growth. Additionally, global warming and climate change conditions that are currently occurring exacerbate this issue. For megacities located in tropical regions, such as Indonesia, there is limited available data regarding the impact of heat stress. Therefore, it is essential to develop a heat index suitable for tropical climates, characterized by high temperatures and humidity levels. Temperature and humidity are two crucial factors in quantifying the heat index to mitigate the risk of heat-related disasters. Consequently, when modeling the heat index for megacities in tropical regions, it is necessary to incorporate humidity. This study aims to compare different models for humidity-inclusive temperature indices, specifically Apparent Temperature (AT), Heat Index (HI), and Wet Bulb Globe Temperature (WBGT). These indices are computed using standard weather measurement data and re-analysis data to obtain spatial distribution. The current results demonstrate that these indices effectively captured the unusual heat conditions in Indonesia during the September 2023 period.

How to cite: Ferdiansyah, M. R., Nahas, A., and Sopaheluwakan, A.: Assessment of humidity-inclusive heat index models over tropical megacities: Indonesia case, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-4898,, 2024.

EGU24-6555 | ECS | Posters on site | NH1.1

Mapping the vulnerability to heat: an application in the city of Bern 

Sujung Lee, Moritz Burger, Moritz Gubler, Stefan Brönnimann, and Ana Maria Vicedo-Cabrera

Background: Heat is widely acknowledged as one of the most hazardous climate-related risk factors affecting human health. Increasing urban development has led to an amplification of its health impacts due to the Urban Heat Island (UHI) effect. However, our understanding of neighbourhood-level vulnerability to the UHI effect remains limited. This information can be crucial for policymakers to identify high-risk areas in cities and develop more targeted public health interventions. Thus, we propose a comprehensive approach to map the vulnerability to UHI in the city of Bern (Switzerland) by (1) assessing the demographic and socio-economic factors contributing to increased UHI exposure and (2) analysing the spatial distribution of vulnerability to the UHI effect.

Methods: We collected population and household statistics at the individual level from 2012 to 2021 from the Federal Statistical Office of Switzerland. Firstly, we calculated the intensity of UHI (representing the temperature difference between the inner city and the rural surroundings) in each district of Bern using high-resolution (50mx50m) modelled urban temperature data. Next, we performed univariate logistic regression models to estimate the association between UHI exposure and population characteristics, reporting odds ratio (OR) and 95% confidence intervals (CI). We defined UHI exposure as individuals being exposed to UHI intensity exceeding the city-mean for the corresponding census year. Subsequently, we established the Heat Vulnerability Index (HVI) by selecting key determinants: 1) the elderly population (aged ≥65 years), 2) females, and 3) individuals with low socio-economic status. The overall percentile ranks for districts were calculated by summing variable rankings.

Result: First, our study identified several factors contributing to increased UHI exposure, in particular, single individuals had 60% higher odds of UHI exposure (OR:1.60; CI:1.59-1.62) compared to married individuals, and individuals aged 26-44 (1.71; 1.70-1.74) compared to those aged 0-17. Also, wealthier individuals appeared to have higher odds of UHI exposure (medium: 2.32; 2.30-2.35, high: 1.66; 1.64-1.67) compared to the lowest group. In the context of the work environment, individuals in large-size companies (≥250 employees) had an increased risk (1.85; 0.77-6.05) of UHI exposure compared to those in micro-size companies (<10 employees) and employees of public companies (1.17; 0.88-1.62) compared to their counterparts in private companies. Our results highlighted varying vulnerability patterns in different districts. In the city centre, despite a medium HVI, UHI intensity surpassed other areas, intensifying vulnerability to heat. The western part showed lower UHI but had high HVI due to a concentration of individuals with the lowest socio-economic status.

Conclusion: Our preliminary results emphasize the importance of considering demographic and socioeconomic characteristics when assessing the impact of UHI exposure on population health. Building upon these findings, we plan to develop a heat vulnerability map of the city of Bern by applying a more advanced epidemiological analysis using Bayesian methods to assess the spatial distribution of the UHI mortality risk. This investigation will provide valuable evidence and methods to improve our understanding of the impact of UHI on health and aid in developing targeted interventions to protect at-risk communities.

How to cite: Lee, S., Burger, M., Gubler, M., Brönnimann, S., and Vicedo-Cabrera, A. M.: Mapping the vulnerability to heat: an application in the city of Bern, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-6555,, 2024.

EGU24-7213 | Posters on site | NH1.1

Assessing Present-Day and Future Perspectives of Climate Impact on Thermal Stress Risks in Korea from 1km High Resolution Scenarios 

Jae-Hee Lee, Hyun Min Sung, Jin-Uk Kim, Sungbo Shim, Chu-Yong Chung, and Young-Hwa Byun

Among the various thermal stress indices, apparent temperature (AT) is closely related to public health indicators, and consequently is widely used by weather agencies around the world. Therefore, in this study we estimate the changes in AT and contributing components in Korea as a whole and in five major cities (Seoul, Gwanju, Daegu, Daejeon, and Busan) using national standard climate scenarios based on the coupled model inter-comparison project (CMIP6). In the present day, high AT occurs in major cities due to high temperature (TAS) and relative humidity (RH). Our findings reveal that even when TAS is relatively low, large AT occurs with higher humidity. Notably, in future warmer climate conditions, high AT may first appear in the five major cities and then extend to the surrounding areas. An increase in TAS and RH during the pre-hot season (March to June) may lead to earlier occurrence of thermal risks in future warmer climate conditions and more frequent occurrence of high thermal stress events. Our study can serve as a reference for future information on thermal risk changes in Korea. Considering those who have not adapted to high temperature environments, our findings imply that thermal risks will become more serious and that heat adaptation strategies will be needed during the pre-hot season under future warmer climate conditions.

How to cite: Lee, J.-H., Sung, H. M., Kim, J.-U., Shim, S., Chung, C.-Y., and Byun, Y.-H.: Assessing Present-Day and Future Perspectives of Climate Impact on Thermal Stress Risks in Korea from 1km High Resolution Scenarios, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-7213,, 2024.

EGU24-8032 | ECS | Posters on site | NH1.1

The interconnections between household energy, health, and climate change: a comprehensive umbrella review 

Clemens Marggraf and Jo-Ting Huang-Lachmann

Household energy is becoming increasingly important for the maintenance of good health, especially as people spend 90% of their lifetime indoors. However, there are a number of health and climate impacts associated with the generation and use of household energy. To date, there is no holistic picture in the literature describing the interlinkages between household energy, health outcomes and climate change. In order to systematically synthesize the fragmented evidence, an umbrella review will be conducted.

Therefore, a systematic review of peer-reviewed literature was carried out following PRISMA guidelines. Five electronic databases were searched (PubMed, Web of Science, Google Scholar, Cochrane and Scopus) to assess available literature on climate change, health and household energy, from January 1, 1900 through to June 5, 2023.

Preliminary findings highlight the dynamic interactions between the three issues, e.g., the impact of climate change on energy use/production and health outcomes, as well as the impact of different methods of energy use/production on climate change and health outcomes. In addition, the lack of consideration of the current literature on climate change  in the context of health and energy is a further finding of the umbrella review. Furthermore, the literature reviewed tends to ignore inclusion criteria (e.g., gender, socioeconomic or spatial aspects), which are also essential for a just transition to a more climate-friendly society in the future.

The goal of the umbrella review is to help policymakers understand the complex interrelationships between the three topics, both now and in the future, as climate change progresses and humanity is forced to adopt different mitigation and adaptation methods that affect energy use and production, as well as health outcomes.

How to cite: Marggraf, C. and Huang-Lachmann, J.-T.: The interconnections between household energy, health, and climate change: a comprehensive umbrella review, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-8032,, 2024.

EGU24-8234 * | Orals | NH1.1 | Highlight

Urgency of Climate Change through the lens of COVID-19 Pandemic: the case of heat-related mortality 

Fulden Batibeniz, Sonia Isabelle Seneviratne, Srinidhi Jha, Andreia Ribeiro, Laura Suarez Gutierrez, Christoph C. Raible, Ben Armstrong, Michelle L. Bell, Eric Lavigne, Antonio Gasparrini, Yuming Guo, Masahiro Hashizume, Pierre Masselot, Susana Pereira da Silva, Dominic Royé, Francesco Sera, Shilu Tong, Aleš Urban, and Ana M. Vicedo-Cabrera and the Multi-Country Multi-City Collaborative Research Network

The COVID-19 pandemic and climate change are both urgent global health concerns. However, their impact on human lives has not been compared on the same scale. In this study, we compared mortality due to heat with COVID-19 in 38 cities worldwide, considering different levels of global warming (+1°C, +1.5°C, +2°C, and +3°C). Our findings reveal that even at a global warming level of +1.0ºC, 6 cities are already at a point where heat-related deaths could equal COVID-19 deaths within 15 years. Regardless of high or low COVID-19 mortality in the cities, the number of years to reach the level of COVID-19 mortality decreases with higher global warming levels. In 18.4% to 47.4% of the cities, heat-related mortality is projected to equal COVID-19 mortality within 15 years, ranging from +1.0ºC to +3.0ºC of global warming. The vulnerability to climate change varies among regions, with European, Mediterranean, and North American cities experiencing a significant rise in heat-related mortality with higher global warming levels. It is important to note that the given number of years represents the time required to reach COVID-19 mortality. However, unlike the peak and decline of COVID-19, climate change-driven heat-related deaths will persistently worsen unless substantial adaptation measures are taken. This emphasizes the crucial need to integrate climate change into public health discourse and policy.

How to cite: Batibeniz, F., Seneviratne, S. I., Jha, S., Ribeiro, A., Gutierrez, L. S., Raible, C. C., Armstrong, B., Bell, M. L., Lavigne, E., Gasparrini, A., Guo, Y., Hashizume, M., Masselot, P., Pereira da Silva, S., Royé, D., Sera, F., Tong, S., Urban, A., and Vicedo-Cabrera, A. M. and the Multi-Country Multi-City Collaborative Research Network: Urgency of Climate Change through the lens of COVID-19 Pandemic: the case of heat-related mortality, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-8234,, 2024.

EGU24-8342 * | Orals | NH1.1 | Highlight

Global assessment of atmospheric and land surface drivers of heatwaves 

Yigit Uckan, Melissa Ruiz-Vasquez, Kelley De Polt, and Rene Orth

Heatwaves are extreme weather events characterized by exceptionally high temperatures that have severe impacts on society and ecosystems. Their magnitude and frequency are increasing with climate change in many regions. They are driven by both atmospheric and land surface processes such as advection or reduced evaporative cooling. The contributions of these individual drivers to the formation of heatwaves have been analyzed in case studies for major past events with model experiments. At the same time, the global relevance of heatwave drivers remains unclear.

We perform a global analysis with reanalysis data to determine the relation of heatwave temperatures to (i) atmospheric variables such as wind, pressure, and pressure differences, each at different geopotential heights, as well as (ii) land surface variables such as evaporative fraction, enhanced vegetation index, and surface net radiation. First, we identify the hottest day in each grid cell during the period 2001-2020. We also determine the values of the driver variables on this day. Then, for each driver variable, we select five days from the entire study period where the variables’ value most closely matched the hottest day value (=analogues). Next, we compare the averaged temperature anomalies of these analogues to those of the hottest day. The more similar the analogue temperature anomalies are (=hotter), the more relevant the driver variable is deemed. This is done for the three hottest days in each grid cell, ensuring that they are at least 15 days apart from each other to belong to separate heatwave events. 

The results show that pressure at the 500 hPa level is the most relevant driver of heatwaves in the mid-latitudes, while in the tropics a combination of variables plays a more important role than individual variables. Radiation is the second most relevant driver in many regions, particularly in tropical areas. In most cases, several drivers seem to contribute to the heatwave events such that only their aggregated analogue temperature anomalies can match the observed anomalies. These findings confirm previous case studies which highlighted the relevance of atmospheric circulation patterns such as blocking as well as reduced evaporation related to vegetation water stress. For the first time we identify the relevance of these processes across the globe, and with observation-based data. This can contribute to a better management and potential mitigation of heatwave temperatures and their impacts.

How to cite: Uckan, Y., Ruiz-Vasquez, M., De Polt, K., and Orth, R.: Global assessment of atmospheric and land surface drivers of heatwaves, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-8342,, 2024.

EGU24-8448 | ECS | Posters on site | NH1.1 | Highlight

 Beyond Ambient Temperature: A Comparative Analysis of Heat Metrics for Detecting Heatwaves on a Country Level 

Tobias Monthaler, Katharina Wieser, and Chloe Brimicombe

Ambient temperature is the standard metric for the detection of heatwaves. However, when considering the impacts on humans the severity in terms of our wellbeing may be underestimated. The field of detecting human-perceived heatwaves is new and fast growing. It is important to recognise the difference in measuring heatwaves using heat metrics in comparison to the standard of ambient temperature. For this approach we develop an algorithm to track heatwaves on a country level through time for the last 30 years. Maximum daily temperature data is used in comparison to the heat metrics of WBGT and UTCI. Through all this, a substantiated knowledge basis should be established of how well the stated heat metrics are detecting heatwave impacts. The findings contribute to the development of standardized and robust approaches for utilizing reanalysis data in conjunction with heat indices, ultimately improving heatwave detection, forecasting and policy development.

How to cite: Monthaler, T., Wieser, K., and Brimicombe, C.:  Beyond Ambient Temperature: A Comparative Analysis of Heat Metrics for Detecting Heatwaves on a Country Level, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-8448,, 2024.

EGU24-8713 | ECS | Orals | NH1.1

Future Exposure to Extreme Heat in Southeast Asia 

Sonali Manimaran, Dennis Wagenaar, Christine Nam, Ludwig Lierhammer, Laurens Bouwer, and David Lallemant

Southeast Asia has seen an increasing trend in extreme temperatures, with records of the highest surface air temperature being set in recent years. Heat waves are also increasing in frequency, severity and duration, with higher maximum temperatures being recorded both in daytime and nighttime. These trends are expected to intensify in the coming decades with climate change, with many implications for human health. For Southeast Asia, however, the number of studies quantifying future heat hazards and exposure are severely limited, but first analyses show that several countries in the region could see dramatic shifts in risk due to climate change. Therefore, it is imperative to quantify the heat stress that populations will be exposed to due to future extremes in the region. To this end, this study uses projections from Regional Climate Models to compute the Wet Bulb Globe Temperature (WBGT) and Universal Thermal Climate Index (UTCI) as measures of extreme heat. WBGT has been widely used in studies of extreme heat and impacts, particularly in the case of outdoor physical work, and has been shown to be representative of heat stress in hot and humid environments, such as Southeast Asia. On the other hand, the UTCI is a newer metric which has been demonstrated to be applicable in a range of climatic conditions and representative of the human physiological response. Using the novel CORDEX-CORE dataset for Asia, WBGT and UTCI metrics for Southeast Asia are developed at a 22x22 km resolution up till 2100 using a wide range of ensemble simulations. Future population exposure to extreme heat is then computed for each country in the region, across a range of Shared Socioeconomic Pathway (SSP) scenarios. Our first results show that nearly the entire region will experience strong heat stress by mid-century and some countries will experience strong heat stress for most of the year by the end of the century.

How to cite: Manimaran, S., Wagenaar, D., Nam, C., Lierhammer, L., Bouwer, L., and Lallemant, D.: Future Exposure to Extreme Heat in Southeast Asia, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-8713,, 2024.

EGU24-9026 | ECS | Orals | NH1.1

The compound effect of acute respiratory infections and temperatures on mortality in the Czech Republic, 1982–2019 

Ekaterina Borisova, Aleš Urban, Hana Hanzlíková, Eva Plavcová, Jan Kyselý, Jan Kynčl, and Joan Ballester

Numerous studies have thoroughly documented the contribution of non-optimal temperatures and acute respiratory infections (ARIs) to increased mortality. However, there is still a gap in understanding how these factors interact together to affect human mortality during the cold season, and how this impacts population susceptibility to heat waves in the summer.

In this study we conduct an analysis over a period spanning 38 years (1982–2019), utilizing: a) daily all-cause mortality counts across the Czech Republic, b) daily proxies of acute respiratory infections (ARIs) incidence, interpolated from weekly healthcare surveillance data, and distinguished regarding three dominant influenza viruses (A/H3N2, A/H1N1, and B), and c) a suite of weather variables, sourced from E-OBS gridded data, including daily mean, maximum, and minimum temperatures, daily precipitation, daily mean sea level pressure, daily mean wind speed, daily mean relative humidity, and radiation level.

To investigate the complex associations between mortality rates, ARI incidence, and weather variability, we employ a distributed lag non-linear model (DLNM) with multiple cross-bases. This approach facilitates the adjustment for confounding meteorological variables and provides a better understanding of their impact as fluctuating confounders. From these refined models, we derived the fraction of mortality attributable to ARIs and low temperatures, offering a quantification of their impact on excess mortality in the cold season. Additionally, we analyse changes in seasonal patterns of mortality according to the meteorological and epidemiological characteristics and assess temporal associations between air temperature and mortality in summer considering factors like intensity of ARI outbreaks and the mean winter temperature in the previous cold season. Our results contribute to better understanding of the links between temperature variability, respiratory infection dynamics and the seasonal variations in mortality.

How to cite: Borisova, E., Urban, A., Hanzlíková, H., Plavcová, E., Kyselý, J., Kynčl, J., and Ballester, J.: The compound effect of acute respiratory infections and temperatures on mortality in the Czech Republic, 1982–2019, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-9026,, 2024.

In recent years, the escalation of climate change has led to a discernible increase in occurrences of extreme heat events, subjecting individuals to heightened and more recurrent heat stress. This phenomenon is particularly pronounced among socioeconomically disadvantaged populations, the elderly, and other vulnerable groups. Prolonged exposure to elevated temperatures has been identified as a significant factor contributing to adverse health outcomes within these demographics. Concurrently, alterations in land use and land cover (LULC) exert a notable influence on the thermal environment. While rural areas experience relatively modest changes in land use compared to urban counterparts, the expansion of non-vegetative zones in these regions still engenders temperature fluctuations. 

This study endeavors to ascertain the spatial-temporal characteristics of the thermal environment and heat-related comfort levels in aging rural areas, a facet that has been largely overlooked in prior research. The focal area of investigation is Yunlin County in Taiwan, a pivotal agricultural region characterized by a noteworthy aging demographic. The research methodology involves an analysis of the spatial distribution of meteorological parameters to discern the thermal landscape in Yunlin. Additionally, a thermal comfort index, Physiological Equivalent Temperature (PET), is employed to gauge the impact of spatial characteristics on human comfortability within this aging rural enclave. LULC data are acquired through supervised classification of remote sensing images. The computation of PET and associated parameters in the Yunlin region is facilitated by the Python package, Pythermalcomfort. 

The analysis reveals thermal comfort and environmental conditions for the elderly in Yunlin, identifying specific hotspots and periods characterized by elevated PET. Notably, the influence of LULC on thermal conditions is discerned, with built-up and bare soil areas exhibiting elevated temperatures. Furthermore, projections of future scenarios indicate an escalating trend of discomfort for the elderly, manifesting in a heightened frequency of PET exceeding 42°C. The study also undertakes an exposure analysis to identify individuals susceptible to heat injuries under diverse scenarios. Ultimately, a set of preventive measures and recommendations are delineated, encompassing the augmentation of green spaces, provision of adequate shading, identification of cooler areas during heatwaves, and scheduling physically demanding activities during cooler hours (between 4:00 and 7:00). In summation, this research utilizes PET to pinpoint high-risk periods and locations for aging rural areas in Yunlin during the summer, with the overarching aim of furnishing valuable insights into heat-related risks.

How to cite: Juang, J.-Y.: Examining the temporospatial patterns of thermal risk for the elderly in an aging rural area in Taiwan, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-9262,, 2024.

EGU24-11600 | ECS | Orals | NH1.1

Assessment of heat exposure under a high-end climate change scenario and projected population scenario around Africa’s Lake Victoria region. 

Delphine Ramon, Clare Heaviside, Oscar Brousse, Charles Simpson, Irene Amuron, Eddie Jjemba, Jonas Van de Walle, Wim Thiery, and Nicole P.M. van Lipzig

Recent global temperature increases and extreme heat events have raised concerns about their impact on health. The area surrounding Lake Victoria, accommodating over 45 million people and ranking among Africa's most densely populated regions, faces rapid population growth and urbanization, set to double its population by 2050 compared to 2022 for most countries in the region. Global-scale projections indicate a potential amplification in heat stress, reaching levels up to 200 times the current rates under high-end scenarios, with the Democratic Republic of the Congo and Uganda facing the most pronounced impacts. Children born in this area in 2020 may face about 1.4 times more heatwaves than their counterparts elsewhere. The combination of population growth and intensifying heat renders the region around the Lake Victoria particularly susceptible to future heat stress.

This research investigates the impacts of climate and population changes on heat exposure and heat stress in the region surrounding Lake Victoria. We analyze how dangerous heat stress by the end-of-the century could change under the ensemble mean climate change signal of the high-end SSP5-8.5 climate change scenario compared to the recent past. Furthermore, we evaluate to what extent and where the population could be affected by dangerous heat stress by these changes. Climate model simulations performed with the COSMO-CLM regional climate model at 0.025° are used, forced with ERA-5 data, applying a pseudo global warming approach for the end-of-century run. Dangerous heat stress is assessed based on categories of heat index and humidex heat stress metrics.

Results indicate a substantial rise in dangerous heat stress across the region. By the 2080s, up to 122 million people (i.e. 44% of the projected population) may experience dangerous heat stress for over 5% of the year (i.e. ~18 days), in contrast to an estimated 1 million people (i.e. 1% of the population) in 2010. Moreover, around 28% of the population (i.e. ~78 million people) might face such dangerous heat for 15% of the time (i.e. ~55 days) by the 2080s. The inhabitants most exposed to dangerous heat stress are notably clustered along the northern shores of Lake Victoria and the southern region, including their urban areas. These findings emphasize the urgent need to address the escalating threat of dangerous heat stress in this region.

How to cite: Ramon, D., Heaviside, C., Brousse, O., Simpson, C., Amuron, I., Jjemba, E., Van de Walle, J., Thiery, W., and van Lipzig, N. P. M.: Assessment of heat exposure under a high-end climate change scenario and projected population scenario around Africa’s Lake Victoria region., EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-11600,, 2024.

EGU24-14155 * | ECS | Orals | NH1.1 | Highlight

Mortality burden in 35 European countries attributed to anthropogenic warming during the record-breaking summer of 2022  

Thessa M Beck, Dominik L Schumacher, Ana M Vicedo Cabrera, Sonia I Seneviratne, Hicham Achebak, and Joan Ballester

More than 61,000 heat-related deaths were associated with the record-breaking temperatures in Europe during the summer of 2022. In this study, we quantify the number of heat-related deaths that would have been avoided in the absence of anthropogenic warming.

For this study, we utilize epidemiological models calibrated for the period 2015–2019 to estimate the heat-related mortality burden in the summer of 2022 for the factual and counterfactual scenario. We derive a counterfactual scenario by removing the regional summer mean warming that arises in response to rising global mean temperatures from the factual temperatures. We use ERA5-Land temperature data and mortality counts from the Eurostat database to estimate the heat-related deaths across 823 distinct administrative regions spanning 35 European countries. 

At 1.15 °C of global warming since pre-industrial times, we obtain a population-weighted median increase over all regions in Europe of more than 2 °C in summer mean temperatures, with the Mediterranean regions being most affected by the increase. By comparing the factual and counterfactual heat-related mortality, we estimate that approx. 70% [95th CI 53.33%– 82.17%] of the total heat-related deaths would not have occurred without anthropogenic warming. Southern Europe has been the most affected by dangerous heat and consequently features the highest number of heat-related deaths attributable to climate change [64.19% of the climate change-attributable deaths]. In relative terms, however, the impact of anthropogenic warming is strongest in Central Europe where approx. 78% of the heat-related deaths are attributable to anthropogenic warming.

How to cite: Beck, T. M., Schumacher, D. L., Vicedo Cabrera, A. M., Seneviratne, S. I., Achebak, H., and Ballester, J.: Mortality burden in 35 European countries attributed to anthropogenic warming during the record-breaking summer of 2022 , EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-14155,, 2024.

EGU24-14273 | ECS | Posters on site | NH1.1

Heat, humidity and health impacts: How causal diagrams can help tell the complex story 

Sidharth Sivaraj, Jonathan Buzan, Olivia Romppainen-Martius, and Ana M. Vicedo-Cabrera

The global health burden associated with exposure to heat is a grave concern and is projected to further increase under global warming. While physiological studies have demonstrated the role of humidity alongside temperature in exacerbating heat stress for humans, epidemiological findings remain conflicted to date. Understanding the intricate relationships between heat, humidity, and health outcomes are crucial for future adaptation and mitigation. This project introduces 'directed acyclic graphs' (DAGs) as causal models to elucidate the analytical complexity in observational epidemiological studies focusing on humid heat related health impacts. DAGs are employed to delineate implicit assumptions often overlooked in such studies, depicting humidity as a confounder, a mediator, or an effect modifier. The complexities arising from using composite heat-stress indices such as wet-bulb temperature, emphasizing the limitations induced in extracting individual effects of humidity are also portrayed through DAGs. Theoretical generalisations for regression models corresponding to each of the causal assumptions are also discussed. The goal of the study is not to prioritize one causal model, but to explicitly discuss the potential causal models suitable for representing associations between heat, humidity, and related health impacts. In the process, we highlight the implications of selecting one model over another. The project aims to inspire further quantitative studies on the topic and motivate researchers to explicitly characterize the assumptions underlying the analytical models with DAGs, facilitating accurate interpretations of the findings. This extends beyond analysing the role of humidity in heat-related health impacts, encompassing similarly complex research questions associated with other compound events.

How to cite: Sivaraj, S., Buzan, J., Romppainen-Martius, O., and Vicedo-Cabrera, A. M.: Heat, humidity and health impacts: How causal diagrams can help tell the complex story, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-14273,, 2024.

EGU24-15322 | Orals | NH1.1 | Highlight

The Role of Rainfall in Humid Heat Extremes across the Global Tropics 

Lawrence Jackson, Cathryn Birch, Guillaume Chagnoud, John Marsham, and Christopher Taylor

Extreme humid heat poses a serious risk to human health, reducing the body’s ability to cool itself through sweating. The impact on humans will increase under climate change, particularly in tropical regions, such as the Indian subcontinent, that are highly populated and already hot and humid. Whilst there is a growing body of research on dry-bulb temperature extremes, there is limited understanding of the meteorological drivers of humid heat extremes, particularly the role of moisture transport, rainfall, and evaporation of moisture from the Earth’s surface.

In this study, we use ERA5 data to identify and analyse extreme humid heat events in the global tropics during 1993-2022. In particular, we focus on the relationship between rainfall and the occurrence of humid heat and use extremes in wet-bulb temperatures to define the humid heat events.

We find that rainfall is a key driver of humid heat extremes across much of the global tropics. In monsoon regions, dry-bulb temperature extremes typically occur in the pre-monsoon period whereas wet-bulb extremes occur more frequently during the monsoon season. The role of rainfall varies between humid heat events characterised by extremes in dry-bulb temperature versus those characterised by extremes in humidity. In much of the global tropics, rainfall followed by a few days of dry clear weather primes the surface and boundary layer climates for the initiation of humid heat events. These events typically have extremes in dry-bulb temperatures accompanied by what we characterise as a sufficiently high level of humidity. In arid regions, away from irrigated areas, rainfall is critical for the initiation of humid heat and frequently occurs locally on the first day of humid heat events. These events typically have extremes in humidity whereas dry-bulb temperatures are less likely to be extreme.

These findings are a step towards greater understanding of the meteorological drivers of humid heat extremes at the regional scale. They will be valuable in the evaluation of weather and climate models, will aid the use and interpretation of climate model projections, and ultimately inform the design of much needed early warning systems for humid heat extremes.

How to cite: Jackson, L., Birch, C., Chagnoud, G., Marsham, J., and Taylor, C.: The Role of Rainfall in Humid Heat Extremes across the Global Tropics, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-15322,, 2024.

EGU24-15748 | ECS | Posters virtual | NH1.1

Building a heat wave anticipatory action plan for the Sahelian city of Ouagadougou 

Rachidate Wend-layolsda D. Somdakouma, Bakari Sankara, Ilyassa Sawadogo, Emmanuel Poan, Dorothy Heinrich, and Kiswendsida Guigma

The West African Sahel is one of the hottest regions of the globe, but as in other African regions, heat waves largely remain a neglected hazard from both a preparedness and a research perspective. They are deadly and underestimated. Yet, there is a large potential to decrease their impacts through early warning and anticipatory action, particularly because they are so predictable. In this research, we present a case study of the development of a heat wave early action protocol for the Sahelian city of Ouagadougou, Burkina Faso. The work was led by the Burkina Faso Red Cross (BFRC), with support from the Red Cross Red Crescent Climate Centre and partners. Two research questions guided the process: (i) are heat waves a concern to various actors in the city of Ouagadougou and if so, why? (ii) What are the impacts of heatwaves in Ouagadougou and what can be done to mitigate them?

In collaboration with the National Meteorological Agency (ANAM), the first ever definition for heat waves in the city of Ouagadougou was developed based on statistical analyses of meteorological records and reports of historical severe events. Heat waves were thus defined as spells of three or more days where the daytime and/or night-time temperature exceeds the 90th percentile of the distribution of the hottest month of the year i.e. April in Ouagadougou. This definition has now been incorporated by ANAM into its early warning platform and will automatically alert BFRC when needed.

Faced with the lack of quantitative impact data about heatwave impacts in the Sahel region, a qualitative cross-sectional study based on focus group discussions, and key informant interviews, and a review of grey literature (especially media reports) were used. The target audience population for this study was experts/practitioners from various disaster management sectors including health, water, energy and municipal officers, vulnerable social groups and vulnerable communities living in slums.

Experts, practitioners, vulnerable social groups and communities all stressed that extreme heat is a major concern in Ouagadougou, has become more severe in recent years and should be better tackled at the individual, community and national levels. The elderly, children under the age of five and people suffering from chronic diseases such as albinism, leprosy and other conditions were frequently mentioned by interviewees as the most vulnerable to extreme heat. From a geographical perspective,

slums, which are generally located in the periphery of the city, were identified as the most vulnerable neighbourhoods, mainly because of the poverty rates and the lack of infrastructures. The most recurrent impacts found across the study were around insufficient water and power availability, increase in some diseases, thermal discomfort and subsequent social and economic impacts. Among the suggested solutions, BFRC and their partners have decided to prioritise in the Early Action Protocol: early warning dissemination, potable water distribution, medical monitoring of chronically ill people and cash distribution.

Keywords: heat waves, Ouagadougou, slums, anticipatory action, Red Cross

How to cite: Somdakouma, R. W.-D., Sankara, B., Sawadogo, I., Poan, E., Heinrich, D., and Guigma, K.: Building a heat wave anticipatory action plan for the Sahelian city of Ouagadougou, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-15748,, 2024.

EGU24-17242 | ECS | Posters on site | NH1.1 | Highlight

Limits to adaptation strategies for heat impacts on rural labors filtered through stabilized climate mitigation scenarios.  

Jonathan Buzan, Yona Silvy, Fabrice Larcoix, Friedrich Burger, Jens Terhaar, Thomas Frölicher, Édouard Davin, and Fortunat Joos

With global warming, increased heat stress will substantially impact the rural labor force. Understanding and quantifying this impact is difficult, especially due to regional differences: Does the temperature increase? Is there more solar exposure? Does humidity respond non-linearly with respect to temperature changes? Furthermore, humans are resourceful, and local environments could provide adaptation methods to decrease heat impacts.


A policy-relevant assessment in the context of the Paris Agreement is even more difficult with existing CMIP-type simulations with prescribed greenhouse gas trajectories that lead to a different and often non-stable warming for each model. To resolve the impacts climate mitigation and adaptation on heat stress on warming levels with specific relevance for the Paris Agreement, we use the Community Earth System Model (CESM2) driven by emissions from the Adaptive Emissions Reduction Approach (AERA) to generate climate mitigation scenarios stabilized at 1.5°C, 2.0°C and 3.0°C of global warming.


One form of adaptation to heat stress impacts is to use the local environment for cooling. Within CESM2, we compare the direct and indirect exposure to solar radiation within the vegetated canopy as an inexpensive form adaptation. To diagnose the heat stress conditions we use the International Organization for Standardization (ISO) 7243, the Wet Bulb Globe Temperature (WBGT), realized by first principles representation of the globe, dry bulb, and natural wet bulb thermometers utilizing CESM2’s temperature, humidity, winds, and radiation. The WBGT values are transformed into labor capacity using standardized algorithms (e.g. NIOSH or Lancet) and the above canopy (no adaptation) and below canopy (with adaptation) labor capacity are directly compared to each other.


We show that the potential to adapt by using the local environment for cooling is not uniform across regions. For example, evaluating the hottest seasonal period (defined as a local summer), at the 3.0°C mitigation scenario in equatorial Southeast Asia, adaptation can save up to 50% of total labor capacity losses. However, in northern South Asia, adaptation saves only 10% of the seasonal labor capacity losses. These results demonstrate that rural laborers in some locations may have limited capacity to adapt to differing global mitigation strategies and may require mechanical cooling or other expensive forms of adaptation.

How to cite: Buzan, J., Silvy, Y., Larcoix, F., Burger, F., Terhaar, J., Frölicher, T., Davin, É., and Joos, F.: Limits to adaptation strategies for heat impacts on rural labors filtered through stabilized climate mitigation scenarios. , EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-17242,, 2024.

EGU24-17545 | Orals | NH1.1 | Highlight

Development of a deep learning based system for heatwave detection using seasonal forecast data 

Fatemeh Heidari, Qing lin, Yanet Díaz Esteban, Edgar Espitia Sarmiento1, and Elena Xoplaki

Heatwaves have been widely studied in recent years because of their major impact on human health, mortality, ecosystems, agriculture, and the economy. Globally, heatwaves are becoming more severe, longer, and recurrent with global temperature rise. Therefore, the study of heat waves and the development of an early warning system for prediction of regional heatwaves help climate preparedness and decision-making. In this research, we propose a heatwave prediction algorithm based on a deep learning model, a convolutional neural network (CNN). This CNN model is trained with reanalysis data ERA5 and real heatwave events from EMO observation data for years from 1993 to 2021. We illustrate the relationship between the patterns in geopotential height at 500 hpa (GPH), sea surface temperature (SST), and the real heatwaves that happened in the last 20 years. This study employs the hindcast data from SEAS5.1 with 25 ensemble members, available at C3S. GPH and SST from observation data are input to the model and the heatwave magnitude at every single grid point is the output. The heatwave is defined as a period of three or more consecutive hot days and nights when the daily maximum and minimum temperature (TX/TN) exceeds the long‐term (1993–2022) daily 90th percentile. For estimating the heat wave magnitude we accumulated TX exceedance the local 90th percentile for all heat wave days over a user-defined interval (monthly, seasonal, etc.) as in Zampieri et al. (2017), Toreti et al. (2019). The results show the CNN model using atmospheric circulation fields (SST and GPH) with adjusted parameters is able to forecast extreme events in Europe, and it can potentially enhance the AI-based early warning systems for extreme weather.

Zampieri, M., Ceglar, A., Dentener, F., and Toreti, A. (2017). Wheat yield loss attributable to heat waves, drought and water excess at the global, national and subnational scales. Environmental Research Letters, 12 (6), 064008. doi:10.1088/1748-9326/aa723b

Toreti, A., Cronie, O., and Zampieri, M. (2019). Concurrent climate extremes in the key wheat producing regions of the world. Scientific Reports, 9(1), 5493. doi:10.1038/s41598-019-41932-5

How to cite: Heidari, F., lin, Q., Díaz Esteban, Y., Espitia Sarmiento1, E., and Xoplaki, E.: Development of a deep learning based system for heatwave detection using seasonal forecast data, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-17545,, 2024.

EGU24-18354 | ECS | Orals | NH1.1 | Highlight

Projecting heat-related excess mortality under climate change scenarios in Europe: A multi-domain analysis 

Hélder Relvas, Silvia Coelho, Vera Rodrigues, Ana Isabel Miranda, Myriam Lopes, Daniel Graça, Bruno Augusto, João Basso, and Joana Ferreira

Recent studies have highlighted the diverse health consequences associated with climate change. However, a comprehensive evaluation of the specific susceptibilities of individuals and cities to these changes remains lacking. Addressing this gap, our study offers insights into potential excess mortality risks attributable to heat-related events under various climate change scenarios across diverse European regions, within the framework of the European Project DISTENDER.

The primary objective of this research is to assess the potential impact of climate change scenarios, specifically Shared Socioeconomic Pathways (SSPs) SSP1-2.6, SSP2-4.5, SSP3-7.0, and SSP5-8.5, on heat-related excess mortality in a range of European locations. Focusing on Austria, the EURAF domain (encompassing parts of Portugal and Spain), a region in the Netherlands, the metropolitan area of Turin (Italy), and the urban area of Guimarães (Portugal), our investigation spans varied socio-geographic domains.

By employing local-specific relative risk functions and daily average temperature data spanning 2015 to 2049, derived through statistical downscaling from CanESM5, EC-EARTH3, and MPI-ESM1-2-HR global climate models, our analysis encompasses resolutions ranging from 9000 to 500 meters, depending on the specific domain. This multidomain approach allows for capturing localized variations in climate impacts with high spatial resolution.

The significance of our findings lies in their contribution to informing adaptive strategies, public health policies, and urban planning efforts aimed at mitigating the effects of climate change on vulnerable populations. Preliminary results indicate a distinct upward trend in heat-related excess mortality over the years, with the highest values observed for SSP5-8.5. Furthermore, there exists considerable variability among climate models.


The authors would also like to acknowledge the support of CESAM (UIDP/50017/2020 + UIDB/50017/2020 + LA/P/0094/2020) and C2TN (UIDB/04349/2020). Thanks are due to the DISTENDER Project (Grant agreement ID: 101056836). Thanks are due to FCT/MCTES for the contract granted to Helder Relvas (10.54499/2021.00185.CEECIND/CP1659/CT0026).

How to cite: Relvas, H., Coelho, S., Rodrigues, V., Isabel Miranda, A., Lopes, M., Graça, D., Augusto, B., Basso, J., and Ferreira, J.: Projecting heat-related excess mortality under climate change scenarios in Europe: A multi-domain analysis, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-18354,, 2024.

EGU24-18980 | Posters on site | NH1.1

So, will you live an unprecedented life? 

Wim Thiery

Under a continued increase in global warming, extreme events such as heatwaves will further rise in frequency, intensity, and duration over the next decades. Climate change impact studies routinely assess hazard and exposure change across discrete time windows, but thereby ignore how vulnerability and climate risk evolve across a person’s lifetime. Conversely, demographic research has a long tradition of assessing population processes and vulnerability from a cohort perspective, but generally neglects climate impacts. 

In the soon-to-start ERC CoG project LACRIMA (LAgrangian Climate Risk and Impact Attribution), we will develop novel concepts and methodologies to express climate change impacts and risk from a cohort perspective. More specifically, the project pursues to (i) to reconstruct two iconic climate change impacts on people around the world using machine learning (heat-related mortality and burned area), (ii) to uncover age-specific vulnerability to climate extremes including heatwaves, wildfires, river floods, droughts, tropical cyclones, and crop failures, (iii) to detect and attribute changes in lifetime extreme event exposure and climate impacts on mortality across generations and regions, (iv) to quantify how these attributable cohort impacts change country-level life expectancy around the world under a range of warming scenarios, and finally (v) to project how lifetime exposure to extreme events including compound events may trigger irreversible impacts under scenarios of temporary overshooting of long-term warming targets.

By bridging physical climate science, demography, and planetary health, LACRIMA will comprehensively identify whether and where people will live an unprecedented life in terms of climate impacts, and how mitigation choices can alter the climate change burden on current young generations around the world

How to cite: Thiery, W.: So, will you live an unprecedented life?, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-18980,, 2024.

EGU24-19866 | ECS | Orals | NH1.1 | Highlight

Developing a proof-of-concept within Destination Earth for heat-stress adaptation under climate change scenarios, with a special focus on public health management 

Joshua Lizundia-Loiola, Nieves Peña, Efren Feliu, Jorge Paz Jiménez, Niels Souverijns, Dirk Lauwaet, and Filip Lefebre

According to the World Health Organization, between 2000 and 2016, the number of people exposed to heat waves increased by around 125 million. In Europe in 2003, 70.000 people died as a result of the June-August events and in Spain, that summer produced a total of 6.534 deaths.  2022 summer was even worst and its record-breaking heat caused 61.672 deaths in Europe, 11.324 of them in Spain, according to Ballester et al. (2023) [1].

The most worrying thing is not that all these episodes confirm the global warming, but that what is coming is even worse. If greenhouse gas emissions are not significantly curtailed, extreme temperatures and specifically heatwaves will become more frequent, more intense and longer every year that coupled with urban population growth and the trend towards an ageing population will produce devastating impacts on human health especially on vulnerable populations (see Impacts will depend on local factors related with exposure, vulnerability to climate-related stresses and the capacity to cope with, so well-prepared health systems and well-suited adaptation measures at different levels are essential to protect populations, limit adverse impacts of heat and therefore reduce the magnitude of their risks.

Current information and heat simulation models are limited in time and space due to the high computational costs, so sometimes make it too complex for multiple stakeholders to have a good understanding of heat-stress assessment. This situation leads to ambiguity among stakeholders when implementing adaptation measures.  The idea of this ongoing work is to use Destination Earth’s Climate Adaptation digital twin to develop an interactive tool that will support decision-makers in the assessment of different adaptation options for heat stress adaptation under climate change scenarios, with a special focus on public health management. The tool will allow to have a better understanding of heat-stress assessment by simulating different type of climate change scenarios and, hence, identifying hotspot areas, and high-risk populations and locating opportunities to incorporate solutions to reduce impacts.

The architecture of DestinE provides a unique opportunity to develop an operational environment in which different but interconnected components give guidance and advice to decision makers in their process of designing adaptation pathways. From the data provided by Destination Earth Data Lake, it will be developed a proof-of-concept over the Basque region based on data-driven statistical models, physically based simulations, cost-impact analysis and algorithms. The demonstrator will also include a web-based graphical interface providing easy and accessible dashboards to support decision-makers in the assessment of adaptation options for heat stress adaptation under different climate change scenarios.
Key words: Destination Earth, Climate Adaptation digital Twin, heat-stress, adaptation, health, Copernicus, Europe. 

1. Ballester, J., Quijal-Zamorano, M., Méndez Turrubiates, R.F. et al. Heat-related mortality in Europe during the summer of 2022. Nat Med 29, 1857–1866 (2023).

How to cite: Lizundia-Loiola, J., Peña, N., Feliu, E., Paz Jiménez, J., Souverijns, N., Lauwaet, D., and Lefebre, F.: Developing a proof-of-concept within Destination Earth for heat-stress adaptation under climate change scenarios, with a special focus on public health management, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-19866,, 2024.

EGU24-20179 | Orals | NH1.1

Heat waves and health impacts in the northern part of Senegal: implementation of an early warning system to support Health National Adaptation Plan (HNAP) 

Ibrahima Sy, Birane Cissé, Babacar Ndao, Mory Touré, Abdoul Aziz Diouf, Mamadou Adama Sarr, and Ousmane Ndiaye

The Sahelian zone of Senegal experienced heat waves in the previous decades, such as 2013, 2016 and 2018 that were characterised by temperatures exceeding 45°C for up to 3 successive days. The health impacts of these heat waves are not yet analysed in Senegal although their negative efects have been shown in many countries. This study analyses the health impacts of observed extreme temperatures in the Sahelian zone of the country, focusing on morbidity and mortality by com[1]bining data from station observation, climate model projections, and household survey to investigate heat wave detection, occurrence of climate-sensitive diseases and risk factors for exposure. To do this, a set of climatic (temperatures) and health (morbidity, mortality) data were collected for the months of April, May and June from 2009 to 2019. These data have been completed with 1246 households’ surveys on risk factor exposure. Statistical methods were used to carry out univariate and bivariate analyses while cartographic techniques allowed mapping of the main climatic and health indicators. The results show an increase in temperatures compared to seasonal normal for the 1971–2000 reference period with threshold exceed[1]ances of the 90th percentiles (42°C) for the maxima and (27°C) the minima and higher temperatures during the months of May and June. From health perspective, it was noted an increase in cases of consultation in health facilities as well as a rise in declared morbidity by households especially in the departments of Kanel (17.7%), Ranérou (16.1 %), Matam (13.7%) and Bakel (13.7%). The heat waves of May 2013 were also associated with cases of death with a reported mortality (observed by medical staf) of 12.4% unequally distributed according to the departments with a higher number of deaths in Matam (25, 2%) and in Bakel (23.5%) than in Podor (8.4%) and Kanel (0.8%). The morbidity and mortality distribution according to gender shows that women (57%) were more afected than men (43%). These health risks have been associated with a number of factors including age, access to drinkable water, type of fuel, type of housing and construction materials, existence of fan and an air conditioner, and health history.The heat wave recurrence has led to a frequency in certain diseases sensitive to rising temperatures, which is increasingly a public health issue in the Sahelian zone of Senegal. The main scientific evidence and findings generated from this research initiative support the adaptation options of health national adaptation plan (HNAP) with the implementation of an early warning system for local communities and health system workers.

Keywords: Climate · Temperature · Heat waves · Diseases · Health risks, adaptation, Senegal

How to cite: Sy, I., Cissé, B., Ndao, B., Touré, M., Diouf, A. A., Sarr, M. A., and Ndiaye, O.: Heat waves and health impacts in the northern part of Senegal: implementation of an early warning system to support Health National Adaptation Plan (HNAP), EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-20179,, 2024.

Employing seasonal forecasting in the domain of impact and risk assessment is particularly beneficial. It facilitates early warning systems and proactive adaptation strategies, which are essential for minimizing the adverse effects of heat waves. This proactive approach is crucial for public health, urban planning, and disaster management, where timely information can significantly alter response strategies and mitigation measures.

This study, we underscore the value of integrating high-quality climate data with impact assessment models. It demonstrates how bridging the gap between climate science and practical risk management can lead to more effective and informed decision-making processes in the face of climate change challenges.

In our study, we integrate Copernicus seasonal forecasting data with the CLIMADA platform, adopting a forward-looking approach to assess the potential impacts of heat waves on populations. This integration involved developing a bespoke pipeline to seamlessly bridge the gap between raw forecasting data from Copernicus and the analytical capabilities of CLIMADA, an ETH Zurich-developed tool for climate impact and risk assessment. The focus is not only on facilitating data integration but also on automating the processing and communication of results.

One significant aspect of this work is managing extensive datasets containing multiple simulations. To efficiently handle this, we implemented an automated system for data extraction, transformation, and loading. This is crucial in maintaining the integrity and usability of the data within CLIMADA's impact modeling framework. Part of this process also entailed resolving spatial and temporal alignment issues, a step essential to ensuring the ability of the seasonal forecasting data to reflect the potential heat wave impacts. Our approach aim to streamline the complexities of large-scale climate data, enhancing the precision and effectiveness of our assessments.

Building a pipeline that links these probabilistic forecasting with impact assessment tools has multiple benefits. It enhances the capability to identify critical data needs and model improvements, thus fostering a feedback loop that drives data and model refinement. Furthermore, it contributes to laying the groundwork for the effective use of the next generation of seasonal forecast data, potentially transforming how we prepare for and adapt to climate risks.

How to cite: Araya, D., Kropf, C. M., and Bresch, D. N.: Integrating Copernicus Seasonal Forecasting Data with CLIMADA for Heat Wave Impact Analysis: Challenges and Solutions in Pipeline Development, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-20478,, 2024.

EGU24-108 | Orals | NH1.2

Reconstructing Historical Flood Events: A Monte Carlo-Based Uncertainty Approach 

Ramtin Sabeti, Thomas R. Kjeldsen, and Ioanna Stamataki

Reconstructing historical flood events can offer critical insight into past hydrological responses to extreme weather, informing contemporary flood risk management and infrastructure design. This study employs reverse engineering, based on historical data such as recorded rainfall, flood marks, visual records, and eyewitness accounts to reconstruct a flood event. Historical data was collected by the team during a workshop with the local community. The approach involves hydrological (HEC-HMS) and hydraulic (HEC-RAS) models to simulate the flood event. The July 1968 UK storm, remarkable for record rainfall reaching 175 millimetres within 18 hours, caused extensive devastation in south-west England. This study focuses on reconstructing the 1968 flash flood on the River Chew, notably the peak hydrograph in the village of Pensford. A Monte Carlo simulation approach is used in conjunction with the HEC-HMS and HEC-RAS models to produce a range of potential input hydrographs with uncertainty input parameters (primarily event rainfall and Manning’s roughness) that match the historical evidence.  In particular, the Monte Carlo approach is implemented using a series of Python scripts enabling multiple HEC-RAS simulations to be conducted and the results synthesised in the form of an uncertainty analysis of key parameters such as peak flow. 

How to cite: Sabeti, R., R. Kjeldsen, T., and Stamataki, I.: Reconstructing Historical Flood Events: A Monte Carlo-Based Uncertainty Approach, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-108,, 2024.

EGU24-395 | ECS | Orals | NH1.2

Beyond Extreme Temperature: Spatiotemporal Analysis of Humid Heat Stress  

Jency Maria Sojan and Jayaraman Srinivasan

Extreme humid heat stress presents significant challenges to human health and productivity. Traditional heat action plans formulated to tackle dry heat stress are insufficient to address the complexities associated with humid heat stress. Furthermore, there is limited quantitative evidence on the evolving patterns of humid heat stress under changing climate. This study investigates the spatiotemporal trends of extreme heat stress across the Global South from 1964 to 2023, distinguishing between dry and humid heat, using high-resolution ERA5 reanalysis hourly data and the Heat Index (HI).

Notably, South Asia and the Middle East experience the highest frequency of extremely humid heat stress. Specific regions in peninsular South Asia have extremely humid heat stress hours from May to June due to persistent high humidity levels. In contrast, western regions of South Asia encounter extreme dry heat stress preceding the monsoon season, followed by a transition to humid heat stress immediately after the onset of the monsoon. The temporal analysis reveals a more rapid increase in the occurrence of extremely humid heat stress compared to that of dry heat stress from May to July over the past 60 years. This underscores the evolving nature of heat stress and the intensification of humid conditions compared to dry ones.

In conclusion, this study advocates for a shift from exclusively addressing dry stress to a comprehensive approach that accounts for the diverse impacts of humid heat stress, particularly on vulnerable populations. This understanding is critical for policymakers to formulate adaptive strategies tailored to the changing landscape of extreme heat stress. 

How to cite: Sojan, J. M. and Srinivasan, J.: Beyond Extreme Temperature: Spatiotemporal Analysis of Humid Heat Stress , EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-395,, 2024.

EGU24-737 | ECS | Posters on site | NH1.2

A database for the outer sizes of tropical cyclones over the Middle Americas 

Adolfo Perez Estrada and Christian Domínguez Sarmiento

Tropical cyclones (TCs) pose a constant threat to populations residing within tropical and subtropical regions. The direct impacts of TCs, such as intense surface winds, storm surge, and heavy precipitation near the center, are well known. However, the indirect effects (e.g, disruption of the upper-level mean wind flow resulting in continental convection, and precipitation associated with cloud bands away from the cyclone's center), are often underestimated.

It is crucial to comprehensively characterize the size of TCs, taking into account both direct and indirect effects, as this new size definition  could improve early warning systems. While various studies employ different parameterizations to describe cyclone size, many of them overlook precipitation. To address this gap, the ROCLOUD technique was developed using  a Python-based algorithm. This algorithm utilizes information on the TC’s position, the extent of cloud bands, and the size of the wind field to define an outersize for TCs located over the oceanic basins in the Middle Americas. In addition to ROCLOUD, we also developed a technique that uses the spatial distribution of TC rainfall to define the outer TC size, named as RPB algorithm. This technique  utilizes a threshold of 2.5 mm in the precipitation satellite products for depicting TC rainfall. Our dual approach provides a comprehensive understanding of TC  sizes, considering the presence of rainfall that can lead to disasters.

Our database shows  external sizes and positions of TCs (recorded every 6 hours) over the North Atlantic (NA) and Eastern Pacific (EP) Oceans during the 2000-2020 period. We got 191 and 336 positions  from the NA and EP basins, respectively. Statistical analysis reveals the coverage of oceanic basins and highlights their differences. We conclude that ROCLOUD offers an operational approximation of the external size of TCs, especially in situations where storms pose a threat to continental regions. The study discusses the utility of both versions of ROCLOUD and RPB for  the Tropical Cyclone Early Warning System over Mexico (EWS-TC), shedding light on the impact of TC sizes that can lead to disasters.

How to cite: Perez Estrada, A. and Domínguez Sarmiento, C.: A database for the outer sizes of tropical cyclones over the Middle Americas, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-737,, 2024.

EGU24-1525 | ECS | Orals | NH1.2

Objective Identification of Tropical Cyclones with Severe Storm Surge Potential for the North-west Pacific 

Xiaoqi Zhang, Gregor C Leckebusch, and Kelvin S Ng

Storm surges caused by tropical cyclones can significantly impact on coastal areas in East Asia, including megacities e.g., in China. To inform effective adaptation and mitigation planning, a robust storm surge hazard assessment is essential. Unfortunately, the real frequency-intensity distribution of relevant storm-surge levels can only be estimated with large uncertainly based on limited historical observations.

This study demonstrates the successful development of a two-step, objective and automated identification and selection approach of storm-surge relevant TCs for large model data sets where no ground truth verification is possible. In our approach, we combine for the first time two established identification and tracking tools originally developed for extra-tropical cyclones and storms and apply these to identify tropical cyclones. In the first step, we adapted the widely used Murray & Simmonds (1991) University of Melbourne tracking scheme (MS-Track) to the specific conditions of TC tracking in the North-west Pacific. In the second step, we apply the windstorm tracking tool WiTRACK to TC-induced severe wind fields to provide and attach the potential storm-surge relevant information in addition to just the core track provided by the MS-Track.

By validating our results with ERA5 reanalysis data and IBTrACS, we show that our method is simple yet has a well comparable performance in detecting and assessing relevant TC events than more complex tracking approaches. Based on this performance this approach is well-designed and specifically intended to specific applications in CAT modelling approaches, e.g. for the creation of physically consistent event sets for storm surges.

How to cite: Zhang, X., Leckebusch, G. C., and Ng, K. S.: Objective Identification of Tropical Cyclones with Severe Storm Surge Potential for the North-west Pacific, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-1525,, 2024.

The physical mechanisms underlying extreme precipitation events linked to atmospheric moisture transport (IVT-P) are investigated in this study. It investigates changes in synoptic-scale weather patterns over the Indian subcontinent and identifies regions where IVT influences extreme precipitation. The study discovers a strong relationship between daily IVT and precipitation over the core monsoon region and the complex terrains of the Western Ghats and Himalayas. Event Coincidence Analysis reveals that extreme IVT can be used to forecast extreme precipitation in these regions. The dynamic component of moisture transport has a strong influence on daily and extreme precipitation over complex terrains. In contrast, the thermodynamic component has an influence on precipitation over regions with an abundance of water vapor and weak horizontal winds. The study also identifies synoptic features and moisture transport ahead of IVT-P events and finds intense low-pressure anomalies, the transition from ridge to trough patterns, and intense 700 hPa relative humidity in the specified regions. Overall, the study provides insights into the physical mechanisms underlying IVT-linked extreme precipitation events.

How to cite: Raghuvanshi, A. S. and Agarwal, A.: Deciphering the connections between extreme precipitation events, atmospheric moisture transport, and associated synoptic features over India, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-1625,, 2024.

EGU24-1631 | ECS | Orals | NH1.2

Shift of soil moisture-temperature coupling exacerbated 2022 compound hot-dry event in eastern China 

Yueyang Ni, Bo Qiu, Xin Miao, Lingfeng Li, Jiuyi Chen, Xiaohui Tian, Siwen Zhao, and Weidong Guo

Compound hot-dry events (CHDEs) are among the deadliest climate hazards and are occurring with increasing frequency under global warming. The Yangtze River Basin in China experienced a record-breaking CHDE in the summer of 2022, causing severe damage to human societies and ecosystems. Recent studies have emphasized the role of atmospheric circulation anomalies in driving this event. However, the contribution of land–atmosphere feedback to the development of this event remains unclear. Here, we investigated the impacts of soil moisture-temperature coupling on the development of this concurrent heatwave and drought. We showed that large amounts of surface net radiation were partitioned to sensible heat instead of latent heat as the soil moisture-temperature coupling pattern shifted from energy-limited to water-limited under low soil moisture conditions, forming positive land–atmosphere feedback and leading to unprecedented hot extremes in August. The spatial heterogeneity of hot extremes was also largely modulated by the land–atmosphere coupling strength. Furthermore, enhanced land–atmosphere feedback has played an important role in intensifying CHDEs in this traditional humid region. This study improves the understanding of the development of CHDEs from three aspects, including timing, intensity, and spatial distribution, and enables more effective early warning of CHDEs.

How to cite: Ni, Y., Qiu, B., Miao, X., Li, L., Chen, J., Tian, X., Zhao, S., and Guo, W.: Shift of soil moisture-temperature coupling exacerbated 2022 compound hot-dry event in eastern China, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-1631,, 2024.

EGU24-2445 | ECS | Posters virtual | NH1.2

A method of dynamic diagnosis for regional drought degree 

Ruxin Zhao, Hongquan Sun, and Lisong Xing

In view of the difficulty in quantifying the severity of regional drought, this study proposes a method that can quantify and dynamically diagnose the severity of regional drought events, and consider the cumulative superposition effect of drought in the process of spatial and temporal development and evolution. Starting from the site drought index, the method firstly establishes a regional drought index to determine whether drought occurs in the study area in the same month; secondly, it constructs a two-dimensional Copula joint probability model by counting the cumulative duration and cumulative severity of droughts; and finally, it uses the percentile method to classify the joint probability of two-dimensional cumulative drought characteristics into four degree levels of regional drought: light, moderate, severe, and extreme. In the study, the SPEI drought index from 1961 to 2022 was used as the basic data, and the drought centers of China, such as North China Plain, Yangtze River Basin, and Yunnan Province, were selected as the case validation zones, and the results showed that this method can effectively identify the historical drought events in the study area and dynamically diagnose the development process of severe and extreme droughts therein. 

How to cite: Zhao, R., Sun, H., and Xing, L.: A method of dynamic diagnosis for regional drought degree, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-2445,, 2024.

        Using multi-source global station and grid monitoring data, FY-2H satellite, and ERA5 reanalysis data, the life history and precipitation characteristics of tropical cyclone "Freddy" as well as the causes of heavy precipitation in southern Mozambique were analyzed. The results show that "Freddy" had a lifespan of 35.5 days which made it the longest lived tropical cyclone in the world, as well as the widest latitude-crossing TC in the southern hemisphere. The extreme long life cycle of "Freddy" was related to favorable large-scale circulation conditions. The strong and sustained subtropical high pressure system made "Freddy" moving westward over the Southern Indian Ocean stably, without the opportunity to combine with the mid latitude trough or cold air which may cause the path turning, intensity weakening, or transformation. After the generation of "Freddy", more than 70% of its life time was over the sea, and the surrounding SST was generally abnormally high, which provided favorable conditions for the development or maintenance of TC intensity. Especially, the SST within the Mozambique Strait remained above 28 ℃, providing excellent underlying conditions for the enhancement of TC intensity, allowing "Freddy" to develop and strengthen rapidly twice after experiencing intensity weakening caused by landfall. The combined influence of favorable circulation conditions and warm sea surface temperature led to the extreme long life of "Freddy".

        "Freddy" made three landfall, bringing sustained heavy precipitation and severe floods to countries in Southeastern Africa. Especially in the southern part of Mozambique, precipitation had characteristics such as long duration, concentrated areas, and large accumulated amount. After landing in Mozambique, "Freddy" was located in a saddle field, leading to weakened steering airflow. Combined with high-level divergence and sustained transportation of warm and humid air by low-level jet, the large-scale circulation system provided favorable background conditions for the slow movement and maintenance of tropical cyclone. The development of low-level convergence and vorticity bands in lower troposphere, as well as strong and sustained water vapor transport, led to the persistence of heavy rainfall in Mozambique. The invasion of cold air induced the formation of a pseudo equivalent potential temperature high-gradient zone in southern Mozambique, and the cold air in the middle layer enhanced atmospheric instability, which was conducive to the development of convection. The southern part of Mozambique was continuously affected by several mesoscale convective systems (MCSs), which not only improved precipitation efficiency but also prolonged the duration of precipitation. The evolution of MCSs had obvious diurnal variation characteristics, with its rapid development and maturity stages almost concentrating in the afternoon to the earlier evening of local time. The increase in low-level wind speed promoted the enhancement of both water vapor and energy, and under the above conditions, the convergence of tropical cyclone wind direction and wind speed triggered the generation of MCSs continuously.

How to cite: Yang, S.: Analysis on the Characteristics of Extreme Long Life Cycle Tropical Cyclone "Freddy" and the Causes of Heavy Rainfall, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-2724,, 2024.

Extreme temperature changes from one day to another, either associated with warming or cooling, can have a significant impact on health, environment, and society. Previous studies have quantified that such day-to-day temperature (DTDT) variations and extremes are typically more pronounced in mid-and high latitudes compared to tropical zones. However, the underlying physical processes and the relationship between extreme events and large-scale atmospheric circulation remain poorly understood. Here, such processes are investigated for different locations around the globe based on Observation, ERA5 reanalysis data, and Lagrangian backward trajectory calculations. In the extratropics, extreme DTDT changes are generally associated with changes in air mass transport, in particular shifts from warm to cold air advection or vice versa, linked to regionally specific synoptic-scale circulation anomalies (ridge or through patterns). Lagrangian temperature changes in the advected air masses are due to adiabatic warming, which is dominant in the local winter season, and diabatic warming, most importantly in summer. In contrast, for extreme DTDT changes in the tropics, local processes are more important than changes in advection. For instance, the strongest DTDT decreases over central South America in December-February are linked to a transition from mostly cloud-free to cloudy conditions, indicating an important role of radiative heating. The mechanistic insights into extreme DTDT changes obtained in this study can be helpful for improving the prediction of such events and anticipating future changes in their occurrence frequency and intensity.


How to cite: Hamal, K.: Quantification of the Physical Process Leading to Extreme Day-to-Day Temperature Changes, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-3564,, 2024.

EGU24-3613 | ECS | Orals | NH1.2

Assessing windstorm hazard emerging from multiple types of storms 

Nasrin Fathollahzadeh Attar, Francesco Marra, and Antonio Canale

In the context of global climate change, windstorms pose significant environmental, ecological, and socioeconomic challenges. Mountainous and forested regions of Europe, including the Veneto region in northern Italy, have been devastated by unprecedented events such as the storms in July 2023 and Vaia in October 2018, raising the question whether such events may occur more frequently in the future. The probability of observing such extremes in present-day climate can be quantified using cumulative distribution functions of annual maxima wind speeds, obtained from extreme value analysis methods. Once these are derived, however, is it near to impossible to project future changes in these distributions as extreme wind speeds are caused by storms driven by diverse synoptic conditions, the characteristics and occurrence frequency of which may change differently in response to climate change.

This study introduces a method to derive cumulative distribution functions of annual maximum wind speeds explicitly considering the intensity and occurrence frequency of multiple types of storms. Independent windstorms are separated and their maximum hourly wind speed is isolated. Storms are then organized into types based on their local wind direction using a clustering technique. We then use a multi-type Simplified Metastatistical Extreme Value distribution (SMEV) to estimate the cumulative distribution function of annual maximum wind speed for the location of interest. The study focuses on mountainous areas, seeking a simpler relation between typical wind directions and synoptic conditions.

A thorough leave-one-out evaluation with benchmark models, including the traditional Generalized Extreme Value distribution (GEV) and a single-type SMEV, is conducted on 22 mountain stations in the Veneto region (northern Italy). We show that, overall, the proposed multi-type method provides estimates of extreme return levels that are comparable with the ones of single-type SMEV and GEV. Our results demonstrate that it is possible to derive cumulative distribution functions of annual maximum wind speeds explicitly considering storms emerging from different marginal processes. This paves the way to the use of projections of large-scale atmospheric dynamics from climate models to improve our prediction of future extreme wind speeds.


Keywords: Windstorm; Extreme events; Wind direction classification; Multiple types; Simplified Metastatistical Extreme Value (SMEV); Mountainous areas.

How to cite: Fathollahzadeh Attar, N., Marra, F., and Canale, A.: Assessing windstorm hazard emerging from multiple types of storms, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-3613,, 2024.

EGU24-3992 | Posters on site | NH1.2

Synoptic and Mesoscale Conditions of Deep Moist Convection during the Cold Season in Croatia 

Maja Telišman Prtenjak, Domagoj Dolički, Petra Mikuš Jurković, and Damjan Jelić

In this study, thunderstorm activity during the cold part of the year was analyzed based on Thunderstorm Intensity Index (TSII) data on a pre-defined grid with a resolution of 3 km x 3 km in Croatia. The study covered a five-year period from 2016 to 2020, focusing on the months from October to March. The goal of the research was to conduct a spatial and temporal analysis of thunderstorm activity and determine the synoptic and thermodynamic conditions under which it occurs. The analysis aimed to provide an overview of the fundamental characteristics, thereby improving the understanding of deep moist convection in the cold part of the year, which poses a significant challenge in operational forecasting due to its lower frequency and more difficult intensity assessment. The occurrence of surface frontal disturbances was detected based on surface and upper-level synoptic charts, and the flow regime at the 500 hPa level was determined. Thermodynamic and kinematic parameters were calculated from radiosonde profiles from stations in San Pietro Capofiume, Brindisi, Pratica di Mare, Zagreb, and Zadar, using the thundeR free software package.

        A total of 290 convective days were selected for analysis from the observed period. The results indicate that synoptic forcing plays a significantly greater role in the development of convection during the cold part of the year compared to the warm part, while the dominant upper-level flow regime is southwest. The obtained values of CAPE (Convective Available Potential Energy) in the cold part of the year are much lower than those in the warmer part, with a significant contribution from the considerably lower amount of solar surface heating. Additionally, most thunderstorms developed under conditions of strong vertical wind shear, indicating that the atmospheric environment conducive to winter thunderstorms is predominantly a High Shear-Low CAPE (HSLC) environment.

How to cite: Telišman Prtenjak, M., Dolički, D., Mikuš Jurković, P., and Jelić, D.: Synoptic and Mesoscale Conditions of Deep Moist Convection during the Cold Season in Croatia, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-3992,, 2024.

EGU24-4050 | ECS | Posters on site | NH1.2

Compound dry and hot extreme events in the Mediterranean region 

André Correia Lourenço, Ana Russo, Virgílio A. Bento, and João Lucas Geirinhas

Over the last few decades, the frequency, duration, magnitude of heatwaves in Europe have increased considerably, with major natural and socioeconomic impacts (Basarin et al., 2020; K.P. Tripathy et al., 2022). In climate change scenarios, these events are expected to present an increasing trend (Zscheischler et al., 2018) due to variations in dynamic and thermodynamic mechanisms, triggering unusually longer and more intense periods of drought and causing a reduction in agricultural production and the supply of water reservoirs. The Mediterranean region is a climate change hotspot and therefore a region susceptible to the development and intensification of single or compound hot and dry events (Giorgio and Linello, 2008).

This work aims at studying single and compound heatwaves and droughts based on ERA5 and ERA5-Land databases for Southern Europe on a 0.25º x 0.25º and 0.1º x 0.1º spatial resolution, respectively.

The results show positive trends for the duration and intensity of heatwaves and droughts and, conversely, negative trends for soil moisture. Most of the study area shows statistically significant negative trends when aggregating spatially. On the other hand, the annual temperature means tends to migrate towards higher values and precipitation means show a small decrease. Furthermore, the relation between large scale climatic patterns such as the North Atlantic Oscillation (NAO) and compound drought and heatwaves are studied here.

It is expected that compound hot and dry events will have a positive trend in their frequency, duration and intensity, as a consequence of climatic phenomena, such as the synoptic systems, or even due to previous dry characteristics of the soil. Our findings highlight the intricate interplay between different mechanisms in the occurrence of extreme events in Mediterranean Europe, putting into evidence the need for better representation this interplay in climate models.

A.L., A.R., V.B. and J.G. have been supported by the Fundação para a Ciência e a Tecnologia (FCT) I.P./MCTES through national funds (PIDDAC, grant no. UIDB/50019/2020,, and LA/P/0068/2020,, to Instituto Dom Luiz; project DHEFEUS, J.G. acknowledges Fundação para a Ciência e a Tecnologia (FCT) for the PhD Grant 2020.05198.BD.



Basarin, Biljana, Tin Lukić, and Andreas Matzarakis. 2020. "Review of Biometeorology of Heatwaves and Warm Extremes in Europe" Atmosphere 11, no. 12: 1276.

Giorgi, F., Lionello, P., 2008. Climate change projections for the Mediterranean region. Global Planet. Change 63 (2), 90–104.

Tripathy, K. P., & Mishra, A. K. (2023). How unusual is the 2022 European compound drought and heatwave event? Geophysical Research Letters, 50, e2023GL105453.

Zscheischler, J., Westra, S., van den Hurk, B. J. J. M., Seneviratne, S. I., Ward, P. J., Pitman, A., et al. (2018). Future climate risk from compound events. Nat. Clim. Change 8, 469–477. doi: 10.1038/s41558-018-0156-3.

How to cite: Lourenço, A. C., Russo, A., Bento, V. A., and Geirinhas, J. L.: Compound dry and hot extreme events in the Mediterranean region, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-4050,, 2024.

Rainfall return levels are guiding hazard protection, insurance models, infrastructure design, construction, and planning of cities. When deriving information about the frequency and intensity of extremes by fitting extreme value models to pointwise observations, the regionalization of these models is challenging. Rain gauges are distributed unevenly, where some regions suffer from data scarcity in space and time. To address this, topographical and/or climatological covariates are often used for the spatial interpolation. On the other hand, high-resolution climate simulations are available to provide spatial information on rainfall extremes. However, these simulations are still governed by model biases, where the bias adjustment of extremes at ungauged locations is also inducing uncertainty.   

In this study, we propose a combination of observations and a high-resolution convection-permitting climate model simulation in the framework of smooth spatial Generalized Extreme Value (GEV) models in order to estimate spatial rainfall return levels. We choose a study area over southern Germany with complex terrain, which is densely monitored with 1132 rain gauges providing more than 30-year daily rainfall observations. There, a 30-year simulation of the Weather and Forecasting Research (WRF) model is available at 1.5 km resolution driven by ERA5 reanalysis data. We combine observations and covariates from the WRF simulation in the spatial GEV and refer to this approach as sGEV-WRF.

We want to answer three research questions to assess the added value of the proposed framework:

  • Is it worth the effort? Does the sGEV-WRF improve the generation of rainfall return levels compared to the WRF alone?
  • Does the WRF simulation as covariate add value? Can the sGEV-WRF outperform a topography-only spatial GEV?
  • Does the dynamical downscaling at high resolution add value? Can sGEV-WRF outperform a spatial GEV based on observations and covariates from the coarser resolution ERA5?

By evaluating the percentage bias, mean absolute error, and root-mean-square error, we show that the combination of observations and WRF can improve the representation of 10-year and 100-year return levels of daily rainfall.

In addition, we aim to assess the performance of this framework under data-scarce conditions. Therefore, we devise an extensive cross-validation study. We select 5%, 10%, 20%, 50%, 80%, 90%, and 95% of all 1132 rain gauges to re-build the spatial GEV models with 1000 random folds each. We show that the performance is robust under these conditions, highlighting the potential for the application in data-scarce regions. Furthermore, in a non-stationary setup with climate model future projections, it can serve as a reliable tool to assess climate change effects on heavy rainfall.     

How to cite: Poschlod, B. and Koh, J.: Combining observations and a high-resolution climate model for the generation of spatial rainfall return levels, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-5310,, 2024.

EGU24-5441 | ECS | Orals | NH1.2

Compounding drivers amplify the severity of river floods  

Shijie Jiang, Larisa Tarasova, Guo Yu, and Jakob Zscheischler

Estimating the risk of river flooding under global warming is challenging, mainly due to the compound nature of the various drivers, which is not yet fully understood. Our study aims to quantitatively unravel the complex dynamics of multiple factors that interact and influence river flooding. Using interpretable machine learning techniques, we analyzed thousands of global catchments to identify the role of compounding drivers in river flooding. Our results indicate that these compounding drivers have played a significant role in increasing the magnitude of river floods over the past four decades. In particular, the influence of the interaction effects between these drivers becomes more pronounced with increasing flood magnitude, and the degree is modulated by specific physioclimatic conditions. Importantly, traditional flood analysis will underestimate the magnitude of extreme floods due to insufficient consideration of the varying compounding effects in flood generation. Overall, our results emphasize the need to more carefully incorporate compounding factors to improve extreme flood estimates.

How to cite: Jiang, S., Tarasova, L., Yu, G., and Zscheischler, J.: Compounding drivers amplify the severity of river floods , EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-5441,, 2024.

EGU24-5446 | ECS | Posters on site | NH1.2

Enhancing flood forecasting and prevention: The multidisciplinary approach of Flood2Now project and its innovative solutions 

Carlo Guzzon, Raül Marcos, Maria Carmen Llasat, Montserrat Llasat-Botija, Dimitri Marinelli, Albert Diaz Guilera, Luis Mediero, Luis Garrote, Alicia Cabañas Ibañez, Javier Arbaizar Gonzalez, and Olga Varela

Spain and the Mediterranean coast are largely affected by flash floods, which are generated by intense, localized storms within smaller basins, typically less than 100 km2 (Gaume et al., 2016). Predicting these events remains challenging as they are frequently triggered by convective systems operating at scales below the resolution of conventional meteorological models. In Spain, floods are the country's primary recurring natural disaster, accounting for nearly 70% of the compensation amount issued by the Consorcio de Compensación de Seguros (CCS, 2011). 

In this hydrogeological risk context, the ultimate goal of the Flood2Now project is to support the population and mitigate the risk associated with this natural hazard, through the implementation of an automatic real-time warning system in two basins (Francolí and Arga) located in the north-east part of the Iberian peninsula. Multidisciplinarity plays a pivotal role in defining this system, integrating various disciplines and information sources, ranging from complex systems physics and hydrometeorological data to citizen science and socio-economic statistics.

Flood2Now embodies a collaborative effort between universities, companies, and social foundations, to explore the following technical aspects: (i) establishing a comprehensive digital database spanning four decades of flood occurrences; (ii) exploring complex systems methodologies to discern interrelationships among various factors influencing flood impacts; (iii) studying weather patterns associated with diverse flood events, accounting for their impact; (iv) implementing analogous methodologies to enhance flood risk forecasting; and (v) integrating this knowledge to enhance operational systems aiding flood-related decision-making.

This research extends its impact on society by implementing citizen science methodologies to gather supplementary data for flood risk management, enhancing early warning systems' precision, and raising community awareness of flood risks and climate change. Innovative approaches include integrating historical and citizen-collected data into decision-making, employing ensemble prediction systems, and implementing advanced hydrological modeling techniques for streamflow prediction and decision support.

This contribution shows the selected basins and case studies, the proposed applied hydrometeorological chain to forecast flash flood impacts, and the improvements that citizen science can provide, on the one hand, in obtaining flow data and the state of rivers, especially in ungauged basins, and, on the other, in increasing risk awareness.

This research has been done in the framework of the Flood2Now project, Grant PLEC2022-009403 funded by MCIN/AEI/10.13039/501100011033 and by the European UnionNextGenerationEU/PRTR. 



Gaume, E., Llasat M.C., et al., 2016. Mediterranean extreme floods and flash floods. Into Hydro-meteorological extremes, chapter 3, The Mediterranean Region under Climate Change. A Scientific Update (coordinated byAllEnvi).133-144. ISBN : 978-2-7099-2219-7.

CCS, 2021, Estadística riesgos extraordinarios. Serie 1971-2020. Available at:



How to cite: Guzzon, C., Marcos, R., Llasat, M. C., Llasat-Botija, M., Marinelli, D., Diaz Guilera, A., Mediero, L., Garrote, L., Cabañas Ibañez, A., Arbaizar Gonzalez, J., and Varela, O.: Enhancing flood forecasting and prevention: The multidisciplinary approach of Flood2Now project and its innovative solutions, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-5446,, 2024.

Hail is by far the greatest contributor worldwide to insured losses from severe convective storms on an annual basis. Individual outbreaks can cause losses well above EUR 1 bn. In Italy, severe convective storm losses have been dominating the market in the last 5-7 years, with a record of EUR 1.4 bn in 2019 prior to year 2023. On 18-25 July 2023 an unprecedented outbreak brought large hail and strong winds to Lombardy, Veneto, Friuli-Venezia Giulia, Piedmont and Emilia-Romagna, with affected cities including Parma, Turin, Milan and Venice. There were many reports of large hailstones, causing significant damage to property and motor vehicle. The European hail record was breached too. Twice. On 19 July, a hailstone measuring 16 cm in diameter was recorded in Carmignano di Brenta, and broke the previous largest hail record in Europe, which was held by a 15 cm stone found in Romania in 2016. Just five days later, a new record was set, when a 19 cm hailstone was found in the town of Azzano Decimo. This is very close to the all-time largest hail recorded of 20.3 cm, found in 2010 in South Dakota, US. Total loss estimates, of which hail was by far the largest contributor, exceeds EUR 3 bn, of which 70-80% in the property sector (residential and commercial buildings), and the remaining 20-30% in the motor vehicle sector. These were the largest hail events in Italy in recorded history, and the costliest cat event in the third quarter of 2023 for the global insurance market.

Following in the footsteps of the severe convective storm outbreak that impacted France in June 2022, these storms came after a record-hot air mass that languished over Southern Europe much of the week prior. Persistent meteorological conditions conducive to rotating supercell thunderstorms were observed for several consecutive days. These compounded with local conditions favorable for the development of severe hail over the Po Valley. In this study we present a reconstruction of these events based on event reports from European Severe Weather Database. We analyze the synoptic configurations and pre-convective environments that characterized them, with focus on those properties and features that are peculiar to severe hail-forming thunderstorms. We look at different formulations of CAPE and vertical wind shear, as well as composite parameters such as the Significant Hail Parameter and the Supercell Composite Parameter. We make use of Gallagher Re’s Severe Convective Storm Index to contextualize these events historically, and to discuss climate change trends over Northern Italy. Finally, we discuss the implications that such events and their expected frequency under climate change have on the (re)insurance market.

How to cite: Panosetti, D. and Tomassetti, U.: The July 2023 Northern Italy hailstorms from a climatological and (re)insurance market perspective, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-5962,, 2024.

The consensus of climatic research indicates that the likelihoods of extreme precipitation events are going to change significantly, but specific trends depend on the type of dominant weather system, and regional landscape and climate details. Despite the effort of increasingly accurate and converging global circulation models, the uncertainties in the ensemble of CMIP6 models, and the often course spatial resolution make translation of climate models to actionable information of flood forecasts complex and uncertain. We carried out an analysis of a large ensemble of Global Circulation Models (GCMs) of the CMIP6 ensemble that were downscaled statistically as part of the NASA NEX-GDDP-CMIP6 dataset. The analysis looked at segmented windowed return period analysis using the method of l moments to fit general extreme value distributions to global climate models. With analysis of 1, 3 and 7 day duration, median, 15 and 85 percent quantiles, between 5 and 100 year return period, and global spatial coverage, the results show variations in how precipitation events of various return periods and durational are predicted to change in GCMs, and what the associated uncertainty is for various regions of the world. Intermediate analysis outputs show artifacts in yearly extreme precipitation due to the applied statistical downscaling, but relative factors to be used in precipitation scaling under climate change resolves these. Average increases in precipitation extremes of percent are observed globally (+5.1%), with many local outliers for the SSP585 scenario in 2050 (e.g. regions such as the Himalayan region (+23.4 percent median), the Sahel region(+21.6%) or South-Western Spain (-3.9%)). The other SSP scenarios change the global average factors to +3.75% and +4.32% for SPP245 and SSP370 Respectively. Very low variability in the changes is observed for return periods, indicating that the intensity probability curves shift uniformly in the model output. Precipitation events duration does more significantly alter the analysis outputs, and various areas show differences here that correlate with flash and fluvial flood susceptibility. Finally, we open-source the analysis code and link the output as a built-in dataset in the rapid flood simulation platform. Here, automatically derived extreme precipitation events from era5 datasets can be rescaled under climate change conditions by applying the scaling factors derived in this work.

How to cite: van den Bout, B.: Global changes in extreme precipitation linked with rapid flood simulation tools, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-7617,, 2024.

EGU24-8116 | ECS | Posters on site | NH1.2

IMERG-E and IMERG-L: A Comprehensive Evaluation of the Medicane Daniel in Thessaly, Greece 

Evangelos Leivadiotis, Silvia Kohnová, and Aris Psilovikos

On 4 September 2023, the area of Thessaly (Greece) experienced a catastrophic flood as a result of the Daniel hurricane sequence. This severe phenomenon is characterized by extreme rainfall records ranging from 305 mm to 1096 mm between 4 and 7 September, causing severe damage to infrastructure, agriculture and buildings. Seventeen casaulties were recorded. The aim of the study is to complete the integrated multi-satellite harvesting of the Global Precipitation Measurement Mission (IMERG) using 10 precipitation stations distributed in the Thessaly region. Specifically, two precipitation products (IMERG-E and IMERG-L) were used to evaluate the early and late extreme precipitation events of IMERG version 7. In order to obtain the rainfall data needed for the research, a time period of 4 September 2023 (0000UTC) to 7 September 2023 (2330UTC) was chosen. This window corresponds to the approximate time at which Daniel's storm convective zone was on the area of interest. The National Meteorological Agency collected six of the ten precipitation stations and four of the Public Electricity Agency. The evaluation process was divided into two parts: the first part aimed at estimating the total rainfall of IMERG-E and IMERG-L, and the second part aimed at estimating the total daily rainfall of both products. Two statistical assessment indicators were used: the Pearson correlation coefficient and the root mean square error (RMSE) to quantitatively assess the performance of satellite precipitation products using rain-gauge data. Firstly, the correlation coefficient between IMERG-E, IMERG-L and total precipitation at IMERG-E, IMERG-L and IMERG-L is -0.03 and 0.27, respectively. Early products did not correlate with ground data, but later versions showed weak positive linear relationships. The RMSE values are 0.8 and 0.52, respectively. The daily analyses of IMERG-E showed moderate negative correlations on September 4 (-0.29), September 5 (-0.15), and September 7 (-0.25), and moderate positive correlations on September 6 (0.37). In terms of daily performance, the correlation coefficients suggest weak positive correlations (0.22 in 4 September, 0.13 in 5 September, 0.23 in 7 September), with the exception of -0.3 in 6 September. RMSE values remain low (0.31 on 4 September, 0.34 on 5 September, 0.20 on 7 September), except for 6, September, where values (0.95) indicate high levels of error. Overall, the late version is more efficient than the early version, but there are rooms for improvements when the IMERG final version will be available.

How to cite: Leivadiotis, E., Kohnová, S., and Psilovikos, A.: IMERG-E and IMERG-L: A Comprehensive Evaluation of the Medicane Daniel in Thessaly, Greece, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-8116,, 2024.

EGU24-9412 | Orals | NH1.2

Extratropical intrusions and their role in tropical flood events: A South Pacific perspective 

Romain Pilon, Andries de Vries, and Daniela Domeisen

Extratropical Rossby waves are a potential source of instability for driving convective disturbances in the tropics. In the South Pacific, island nations are subject to flooding associated with such convective disturbances, yet these have not been conclusively linked to any large-scale processes. Using an object-based approach, this study specifically explores in particular how Rossby waves propagating into the tropics can contribute to the formation of extratropical-tropical cloud bands, which can cause flooding events. These cloud bands are associated with substantial precipitation events and serve as easily detectable proxies to identify when such intrusions occur. Building upon this foundation we use ERA5 reanalysis along with a detection analysis for tropical-extratropical cloud bands and potential vorticity streamers and cutoffs to establish a climatology of such intrusions and cloud bands. This allows us to demonstrate the statistical association of extratropical intrusions with intensified deep convection, in particular over the tropical central South Pacific. We find that these intrusions contribute significantly to the occurrence of floods in the Polynesian islands. In summary, this study allows us to connect the interaction between the extratropics and the tropics with flood events in the South Pacific.

How to cite: Pilon, R., de Vries, A., and Domeisen, D.: Extratropical intrusions and their role in tropical flood events: A South Pacific perspective, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-9412,, 2024.

EGU24-10058 | ECS | Orals | NH1.2

Detection of past extreme precipitation events and connection to recorded impacts: a multi-data and multi-method assessment over the Central-Eastern Alps 

Katharina Enigl, Alice Crespi, Sebastian Lehner, Klaus Haslinger, and Massimiliano Pittore

Extreme hydro-meteorological events are increasingly observed in southern Europe and especially in the European Alps, where they threaten ecological and socio-economic systems. To detect such events and analyse the changes in their occurrence, a proper definition of an extreme event is needed. Statistically, we define extremes from the tails of the probability distributions. However, these events are not necessarily extreme in terms of impact, and impact-related thresholds may vary spatially and temporally, i.e., single absolute thresholds do not necessarily reflect the extremes at all locations, in all time periods and all seasons. Moreover, the availability of harmonized and consistent datasets is crucial for investigating extremes in a transnational context. In this study, we focus on the identification and characterisation of extreme hydro-meteorological events affecting a transboundary Alpine region between Austria and Italy from 2003 to 2021 based on different definitions of extreme events considering spatiotemporal aspects and multiple datasets. Daily accumulated precipitation is used as the main proxy parameter to describe the potential for severe consequences, as it as it is the most broadly available quantity across different datasets compared to e.g., sub-daily precipitation sums. Moreover, its role as a triggering factor for various hazards (e.g., landslides, debris flows, pluvial and fluvial floods) is widely recognised. We analyse three different statistical methods for the detection of extreme events: (i) the identification of the highest daily precipitation amounts on a regional scale, (ii) the detection of daily precipitation values of high intensity on a local scale and (iii) the identification of exceptional daily precipitation records not in absolute terms but with respect to average conditions associated to a specific period of the year. All detection algorithms are applied to four gridded precipitation datasets, including both observation and reanalysis products, with different technical specifications. Subsequently, identified events for each method-dataset combination are blended with existing records of gravitational mass movements and fluvial floods in the Austrian-Italian border region to analyse the suitability of each combination to detect actual occurred impacts. First results indicate that most detected precipitation extremes relate to actual observed impacts (e.g., 74% for regional scale identification with reanalysis data). However, different method-dataset combinations have different strengths and weaknesses, which reflect inherent characteristics of the dataset and/or of the statistical method employed. Furthermore, some combinations show lower performance in detecting impactful events, because the dataset and method applied conflict with each other (e.g., a coarse-resolution dataset not resolving local-scale features conflicts with a statistical method searching for locally high intensities). The findings could contribute to better inform civil protection authorities about risks related to extreme hydrometeorological events, possibly affected by climate change.

How to cite: Enigl, K., Crespi, A., Lehner, S., Haslinger, K., and Pittore, M.: Detection of past extreme precipitation events and connection to recorded impacts: a multi-data and multi-method assessment over the Central-Eastern Alps, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-10058,, 2024.

EGU24-10451 | Posters on site | NH1.2

Influence of Design Storm Profiles on Flood Peak Discharge in a Small River Catchment 

Kazimierz Banasik, Leszek Hejduk, Adam Krajewski, Donald E. Woodward, Andrzej Wałęga, and Beniamin Więzik

Estimations of flood peak discharges of low probability of exceedance are required for designing and maintaining hydraulic and road structures (reservoirs, weirs, water intakes, bridges, culverts) as well as for flood protection, including assessment of the risk of flooding. Rainfall-runoff models are usually the only alternative for such estimations in case of small catchments, as there is a lack of sufficient, good quality historic data to be used for applying the traditional i.e. statistical methods. The aim of this study was to check responses of a small agro-forested, lowland catchment located in center of Poland to rainfall of assumed probability of exceedance and of three profiles of intensity (i.e. a/ constant intensity, b/ asymmetric one with highest intensity between 0.3 and 0.5 its duration, c/ symmetric one with single peaked intensity) and various storm duration.

A regional formula, developed by state hydrological service, on relationship of intensity-duration-frequency, applicable also for region of center of Poland, has been used to find rainfall depths of the events with probability of exceedance of 1% (return period of 100 years) and various duration (i.e. D = 6, 12, 18, 24, 30, 36, 42, 48, 60 and 72 h), as input data for runoff hydrograph simulation. As the catchment, which area is 82.4 km2, has long term monitoring history, the model parameters, as Curve Number of NRCS (Natural Resources Conservation Service), used for extracting the effective rainfall (direct runoff) from total rainfall depth and parameters of Nash model, used for transformation of effective rainfall into direct runoff hydrograph, were estimated from recorded rainfall-runoff events. Over 50-year-continuous discharge record allowed us to estimate the 100 year flood, by applying statistical method for the investigated catchment, as 25,6 m3/s which form a base for comparison of the results of application of the rainfall-runoff model.

Results of modelling of the of rainfall-runoff process indicate: a/ that critical rainfall duration (producing highest peak discharges) of the three storm profiles were between 24 and 60 hours, and b/ higher peak discharges at critical rainfall durations of the three storm profiles than one of statistical method. The differences (overestimates) were from 1.6% for the constant intensity to 30.0% for the symmetric single peaked intensity.

How to cite: Banasik, K., Hejduk, L., Krajewski, A., Woodward, D. E., Wałęga, A., and Więzik, B.: Influence of Design Storm Profiles on Flood Peak Discharge in a Small River Catchment, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-10451,, 2024.

EGU24-10531 | Orals | NH1.2

Evaluating Standard Precipitation Index (SPI) using MIROC6 historicclimate simulations and ERA 5 reanalysis data as a tool to map theimpacts of climate change in rainfall regime in Brazil 

Gean Paulo Michel, Aimée Guida Barroso, Franciele Zanandrea, Márcio Vinicius Aguiar Soares, Gabriel Ferreira Subtil de Almeida, Marcio Cataldi, Priscila Esposte Coutinho, Lívia Sancho, and Vitor Luiz Galves

Rising global average temperatures, as a consequence of climate change, have worsened the occurrences of extreme weather events, causing disruptions in rainfall patterns around the world. In Brazil, such effects are already observed with the increase of heat waves, floods, droughts, and wildfires. The correlation between disruptions in precipitation patterns and fires is complex, nevertheless, the intensity, frequency, and duration of drought events have significant impacts on fuel flammability and fire behavior. Drought monitoring is particularly relevant in Brazil, where the vast majority of forest fires have an anthropogenic ignition and prolonged dry periods favor such fires to spread out of control. The Standardized Precipitation Index (SPI) is one of the most important tools used to evaluate precipitation variability, offering simple yet robust statistical information on the distribution, duration, and frequency of rainfalls and, consequently, droughts. The SPI uses precipitation as input data to standardize the deviation of cumulated rainfall from the mean of historical precipitation, detecting water deficit (negative values) or water surplus (positive values) for a given location. In doing so, this index allows direct spatial comparability between arid and humid regions. This is an advantageous characteristic when drought analysis is applied to a country with different regional rainfall regimes, such as Brazil. The applicability of SPI as a source of drought prediction was investigated by observing its performance with historical climate simulations of the 6th phase of the Model for Interdisciplinary Research on Climate (MIROC6) and the fifth generation ECMWF atmospheric reanalysis of the global climate, ERA5. The direct comparison of the SPI data, employing the climatology extending from 1980-2014 in Brazil, derived both from the climate simulation model and the reanalysis data - which combines observations and models – has provided valuable insights. Preliminary results show an overall consistency in the calculated indexes from both sources, which are in line with seasonal regional rainfall patterns in Brazil. On average, the SPI indexes recognize water deficits for the North-east, north of the South-east and central regions of Brazil. During the months of winter, both indexes detect droughts in these regions, with ERA-5 SPI index registering severe droughts in central Brazil. These results suggest that the SPI index calculated using the reanalysis data seems to register droughts with greater severity and longer duration, identifying more precisely periods with little to no rainfall, whilst the SPI derived from the MIROC6 simulation data, although able to acceptably identify and delimitate droughts, records less severity for the same period. These findings are important to recognize the MIROC6-derived SPI index as a valuable tool in drought prediction. However, they also highlight the necessity of acknowledging the limitations of the model regarding the severity of droughts. The understanding and prediction of precipitation anomalies is fundamental to coping with the impacts of climate change on water resources, agriculture, and biodiversity, guiding mitigation and adaptation strategies in Brazil.

How to cite: Michel, G. P., Guida Barroso, A., Zanandrea, F., Aguiar Soares, M. V., Ferreira Subtil de Almeida, G., Cataldi, M., Esposte Coutinho, P., Sancho, L., and Galves, V. L.: Evaluating Standard Precipitation Index (SPI) using MIROC6 historicclimate simulations and ERA 5 reanalysis data as a tool to map theimpacts of climate change in rainfall regime in Brazil, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-10531,, 2024.

EGU24-10737 | ECS | Orals | NH1.2

Projecting Extreme Rainfall in Sicily: Integrating Simple Scaling and Hourly Projections into Depth-Duration-Frequency Analysis 

Gaetano Buonacera, David J. Peres, Nunziarita Palazzolo, and Antonino Cancelliere

In this present work, we propose a robust methodology for the derivation of future rainfall depth-duration-frequency curves (DDFs), utilizing hourly projections, the assumption of simple scaling of precipitation, and the application of the method of moments for parameter estimation in dimensionless precipitation height distributions. The methodology introduced herein involves the application of change factors derived from climate projections to precipitation averages across various durations (1, 3, 6, 12, and 24 hours) and to the dimensionless moments of the precipitation series. To implement this methodology, we leverage regional scale models (RCM) from the EURO-CORDEX initiative, characterized by hourly temporal resolution. The direct utilization of hourly projection data allows to bypass the necessity for temporal disaggregation techniques. Change factors are calculated through an analysis of annual maxima derived from both future and control series (1971-2000) generated via RCMs. We consider two distinct emission scenarios, namely RCP (Representative Concentration Pathways) 4.5 and 8.5, spanning three future periods: near future (2021-2050), middle future (2051-2070), and far future (2071-2100). Our methodology is applied to multiple rain gauges located across the Sicily region. The outcomes of our investigation underscore an upward trend in future DDFs, particularly pronounced in the RCP 4.5 scenario and during the far future period. This trend is attributed to an observed intensification in the variability of rainfall events. Depending on the specific geographic location, chosen emission scenario, and future time period, future Depth-Duration-Frequency (DDF) curves may correspond to return periods that more than double those observed in the control climate. The methodology, given the easy availability of the exploited data, can turn useful for updating hydrological design criteria for flood mitigation.  


How to cite: Buonacera, G., Peres, D. J., Palazzolo, N., and Cancelliere, A.: Projecting Extreme Rainfall in Sicily: Integrating Simple Scaling and Hourly Projections into Depth-Duration-Frequency Analysis, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-10737,, 2024.

EGU24-10848 | ECS | Orals | NH1.2

Extreme precipitation – temperature scaling: disentangling causality and covariation 

Sarosh Alam Ghausi, Erwin Zehe, Subimal Ghosh, Yinglin Tian, and Axel Kleidon

Warmer temperatures are expected to cause more intense rainfall, primarily due to the rise in atmospheric moisture at the rate of 7%/K, as indicated by the Clausius-Clapeyron (CC) equation. To evaluate this effect, studies use a statistical approach known as precipitation-temperature scaling that involves fitting an exponential regression between observations of extreme rainfall events and local temperatures, resembling how saturation-vapor pressure scales with temperature. However, the estimated sensitivities (also called scaling rates), exhibit notable deviations from the CC scaling (7%/K). These rates remain mostly negative in the tropics as the rainfall extremes exhibit a general monotonic decrease with temperature and “hook-shape” structures in most parts of tropics and mid-latitudes.

Here we show that most of the variability in the observed scaling rates arises from the confounding radiative effect of clouds associated with rainfall events. Clouds substantially reduce the net radiative heating of the surface during the storms by up to 100 W/m2 in the tropics, leading to the cooling of surface temperatures by up to 8K. This cloud-induced cooling results in a covariation between precipitation and local temperature, inducing a two-way causality in the observed scaling rates. To isolate this cooling effect, we used a thermodynamically constrained surface energy balance model and force it with radiative fluxes under both "clear" and "cloudy" sky conditions. We then quantified the changes in surface temperatures due to clouds and remove it from temperature observations during rainy days. After removing this effect, we found positive scaling across the global land areas, closely aligning with CC rates of 7%/K. We demonstrate that cloud radiative effects alone can explain the observed negative and hook-shaped relationships found in precipitation-temperature scaling.

Our findings imply that projected intensification of rainfall extremes with temperature by climate models is consistent with observations after the cloud-cooling effect is corrected for. Our results emphasize on making a clear distinction between causality and covariation by explicitly separating the temperatures before the rainfall event that are shaped by less clouds from temperature during the rainfall event which include clouds. This adds a crucial effect to the debate of interpreting observed precipitation - temperature scaling rates. Furthermore, our methodology of removing cloud effects on temperatures can be extended to estimate climate sensitivities from observations beyond precipitation extremes.

How to cite: Ghausi, S. A., Zehe, E., Ghosh, S., Tian, Y., and Kleidon, A.: Extreme precipitation – temperature scaling: disentangling causality and covariation, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-10848,, 2024.

EGU24-11716 | ECS | Posters on site | NH1.2

Tropical Cyclone Rainfall Asymmetries Inferred from GPM-IMERG: A Focus on Lesser Antilles 

Catherine Nabukulu, Janneke Ettema, Victor Jetten, and Bastian van den Bout


This study utilizes GPM-IMERG satellite rainfall estimates to assess the asymmetric rainfall patterns in 27 tropical cyclones (TCs) across the Lesser Antilles region from 2000 to 2020. The aim is to evaluate whether there is a persistent relationship between precipitation and wind characteristics, which could support improved TC-related flood risk assessment for these islands. With a focus on hurricane and tropical storm categories, the 30-minute precipitation variability was assessed within a radius of 500 km from the TC’s eye during its path in the study area. In addition, TC’s forward speed and wind characteristics, like  TC’s category and the extent of 34-knot winds (R34), are included. The analysis reveals temporal trends, indicating increased TC rainfall events in the study area during the second decade. Correlations show positive relationships between rainfall total (RT), rainfall area (RA), and rainfall intensity at the 90th percentile (RI0.9), with RT and RI0.9 showing the strongest link in the majority of the observations. Contrary to conventional assumptions, this research challenges the idea that highest category TCs in the wind intensity always produce higher rainfall, as we see that higher-category hurricanes such as H4 (209-251km/hr) and H5 (>=252km/hr) were often associated with lower rainfall values in RT and RI0.9 compared to tropical storms (63 - 118 km/hr). Tropical storms, like higher-category hurricanes, can display large rainfall areas. In addition,  quadrant analysis of rainfall zones around the TC eye highlights that the NE and SE quadrants in TC have significantly more rainfall impact. However, it also reveals the danger posed by weaker quadrants in wind characteristics such as SW and NW, as they can exhibit high rainfall values in RA and RT. The study indicates complex, non-linear relationships between TC’s wind and precipitation characteristics in the Lesser Antilles region. Incorporating the rainfall variability observed in TC dynamics into early warning systems and risk assessment is essential for a more effective emergency response and mitigation planning.

General methodology

The satellite rainfall estimates were obtained within a defined buffer of a diameter of 500km around the TC eye while following the TC trajectory. The buffer was further dissected into quadrant spatial zones  (NE, SE, SW and NW) to provide a detailed perspective on rainfall distribution in different parts of the TC impact area. For each eye position, rainfall characteristics (RT, RA and  RI0.9) were computed for the whole buffer and later individual quadrant partitions. The computed rainfall characteristics were then investigated for potential correlation relationships with the TC wind intensity. In addition, quadrant rainfall patterns were analyzed for persistence throughout the TC duration.


How to cite: Nabukulu, C., Ettema, J., Jetten, V., and van den Bout, B.: Tropical Cyclone Rainfall Asymmetries Inferred from GPM-IMERG: A Focus on Lesser Antilles, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-11716,, 2024.

EGU24-12191 | ECS | Posters on site | NH1.2

Prediction and predictability of drought events in the Spree region 

Clara Hauke, Uwe Ulbrich, and Henning Rust

The predictability of drought events in the Spree region is analyzed, aiming at developing hydrological extreme events forecast and warning systems and long-term solutions regarding sustainable, interdisciplinary and integrated water resources management in the project SpreeWasser:N.

Predictors acting as potential indicators of imminent drought risk are inferred from statistical analyses, modeling and literature. Connections between certain states of the atmosphere (large-scale weather patterns) and local drought events are drawn, focussing mainly on agriculture as a user group. Special attention is paid to the succession of certain weather patterns and their impact on precipitation.

A drought forecast based on k-nearest neighbor regression is being developed using an algorithm which automatically selects the meteorological variables and regions yielding the largest forecast skill as input predictor variables during a hindcast period. This machine learning approach supports the discovery of underlying physical links in atmospheric phenomena.

The analysis and software development is based on ECMWF ERA5 reanalysis data and the objective weather type classification by the German Weather Service (DWD), spanning the years 1980 to 2021.

How to cite: Hauke, C., Ulbrich, U., and Rust, H.: Prediction and predictability of drought events in the Spree region, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-12191,, 2024.

Tornadoes represent major meteorological hazards, in terms of damages to buildings, vehicles and structures and casualties. Because of their small space scale (order of 1km or less), duration (order of 1000s), strongly nonlinear and chaotic dynamics, tornadoes cannot be reproduced in operational weather prediction and climate models. It is important to develop approaches overcoming this limitation and capable of delivering reliable early warnings by civil protection services and estimating whether frequency and strength of tornadoes will change because of anthropogenic climate change. Recently, a probabilistic approach has been developed that resulted in analytical expressions of the probability of tornadoes occurrence based on meteorological parameters that can be extracted from weather prediction and climate models, such as WMAX (updraft maximum parcel vertical velocity, linked to the Convective Available Potential Energy CAPE), WS700 (the wind shear in the lower troposphere), LCL (the lifting condensation level), SRH900 (low-level storm relative helicity). An example is the formula log10(P)=-6.6+WMAX/(3.1+5.2 · WMAX/WS700), which is meant to describe dependence of probability P of occurrence of a tornadoes  on the surrounding environmental conditions and to distinguish among conditions with low and high probability. In this study this and similar formulas are applied to hindcasting the probability of tornadoes using ERA5 data. The purpose is to assess the skill of the method for operational prediction and explore its validity for climate change studies.

The methodology supporting this formula is extensively described in Ingrosso, R., Lionello, P., Miglietta, M. M., and Salvadori, G.: Brief communication: Towards a universal formula for the probability of tornadoes, Nat. Hazards Earth Syst. Sci., 23, 2443–2448,, 2023.

How to cite: Lionello, P. and Muhammadi, A.: Testing the skill of an analytical expression for the probability of occurrence of tornadoes, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-12741,, 2024.

EGU24-15012 | ECS | Orals | NH1.2

Drought projections and associated uncertainties over the Arabian Peninsula from CMIP6 models 

Md Saquib Saharwardi, Hari Prasad Dasari, Waqar Ul Hassan, Harikishan Gandham, Raju Pathak, Karumuri Ashok, and Ibrahim Hoteit

Drought frequency and severity have increased over the water-stressed Arid regions. This research employs multiple CMIP6 global climate models (GCMs) for projecting droughts over the Arabian Peninsula (AP) until the end of the 21st century. We utilized the standardized precipitation index (SPI) and standardized precipitation evapotranspiration index (SPEI) to generate projected future statistics of droughts along with uncertainties assessment from inter-model spread, scenarios, timescale, and methods therein.

For this purpose, after a meticulous analysis, we first identify the most suitable GCMs for better representation of AP's drought spatiotemporal pattern over the historical period (1985-2014). Our results indicate an increase in potential evapotranspiration (PET), which dominates simulated drought statistics relative to the precipitation. The projected evolution of the SPEI, which is derived from both precipitation and PET, indicates droughts  consistently increasing from low to high emission scenarios, In contrast, the SPI, owing to relatively-weaker amplification of the precipitation shows a moderately increasing wetness, except for a few northern regions where both indices evolve in agreement The fidelity of the simulated precipitation by many models over the historical period is also relatively poor compared to the PET, which may also be potentially adding to the uncertainties. In general, the principal sources of uncertainty in drought projections evolve from the choices of index, followed by scenarios, and inter-model variability, whereas methods and timescale mostly impact the magnitude of the trend in drought statistics.  

How to cite: Saharwardi, M. S., Dasari, H. P., Hassan, W. U., Gandham, H., Pathak, R., Ashok, K., and Hoteit, I.: Drought projections and associated uncertainties over the Arabian Peninsula from CMIP6 models, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-15012,, 2024.

EGU24-15295 | Orals | NH1.2

Creation of an automatic workflow for a National Flood assessment in Aotearoa New Zealand 

Alice Harang, Emily Lane, Cyprien Bosserelle, Rose Pearson, Celine Cattoën-Gilbert, Trevor Carey-Smith, Hisako Shiona, Sam Dean, Raghav Srinivasan, Graeme Smart, and Matt Wilkins

To manage current flood hazard and help develop climate change adaptation strategies, the government-funded project “Mā te haumaru ō ngā puna wai: Increasing flood resilience across Aotearoa” aims to better understand flood hazard and risk across all Aotearoa New Zealand, now and in the future. A crucial part of this project is the generation of nationally consistent flood maps across the whole country for the current climate and future climate projections.

First, the workflow requires as input the identification of independent floodplains. Each floodplain will be associated to its catchment and be considered a computational unit. For each domain, a design storm is generated for a given scenario (Annual Exceedance Probability, climate projection, antecedent conditions) or an historical storm is used for validation purposes. The runoff and flow routing of streams and rivers on the steep part of the catchment are simulated with the NIWA TopNET model (McMillan et al. 2016). Used uncalibrated, this hydrological model was modified to include a physically realistic soil conductivity and provide a consistent response between gauge and ungauged catchments. The model is spun up to an average base flow with consistent soil and ground water antecedent conditions. The design storm is then run through the model to provide realistic flow boundary conditions to the hydrodynamic model in the populated lower catchment. Before the inundation modelling, the spatial maps are generated, using the GeoFabrics suite (Pearson et al. 2023), across the lower catchment, based on the latest LiDAR data available and complementary databases such as OpenStreetMap for infrastructure. This process produces a hydrologically conditioned DEM (Digital Elevation Model), including waterways opening and a basic riverbed estimation, associated to a roughness length map. Finally, the flood is simulated using the hydrodynamic model BG_Flood (Bosserelle et al. 2022). The model is a GPU-enabled inundation model using a modern shock-capturing St Venant solver. The model uses a quadtree type mesh that is well suited for GPU computation and allows iterative refinement of the mesh. A first coarse resolution run is used to define the expected flood extent. This flood extent and external data such as stop bank locations, is then used to produce a refinement map defining areas where higher resolution is needed. The model is then run a second time using the variably refined mesh.

Figure 1: Scheme of the cascade of model used to develop consistent flood maps in Aotearoa New Zealand.

This workflow has been validated on several historic flood events including a fluvial flood in Westport, ANZ (56h duration, 60-year flood), a fluvial and pluvial flood in Waikanae, ANZ (12h duration, 80-year flood) and the floods in the Hawkes Bay and Tairāwhiti regions (ANZ) following the Tropical Cyclone Gabrielle in February 2023 (over 100-year flood in some areas).

This workflow is based on open-sources tools; it is modular and automated for continual improvement, to enable data update and to facilitate the creation of new scenarios.

How to cite: Harang, A., Lane, E., Bosserelle, C., Pearson, R., Cattoën-Gilbert, C., Carey-Smith, T., Shiona, H., Dean, S., Srinivasan, R., Smart, G., and Wilkins, M.: Creation of an automatic workflow for a National Flood assessment in Aotearoa New Zealand, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-15295,, 2024.

EGU24-15755 | ECS | Orals | NH1.2

The September 2023 flood in Derna, Libya: an extreme weather event or man-made disaster? 

Elad Dente, Moshe Armon, and Yuval Shmilovitz

Storm Daniel, the deadliest recorded Mediterranean tropical-like (medicane) storm, led to severe floods in large parts of the eastern-central Mediterranean, including Greece and northern Libya. Extreme rainfall, reaching more than 400 mm day-1, triggered a flash flood in Wadi Derna (Libya)– an ephemeral river with a drainage area of 575 km2 that crosses the city of Derna at its outlet to the Mediterranean Sea. Historical measures to mitigate flood risks included dam construction in the Wadi Dernah basin since the 1970s. However, during Storm Daniel, at least two of the dams were breached, resulting in a devastating flood that inundated much of the city of Derna, with over 4,000 casualties, 8,000 missing persons, and the displacement of tens of thousands. The devastating event was the focus of media coverage for a long time, but questions regarding the role of dams and their collapse remain open, and are relevant for other dammed regions as well: How extreme was the storm? How extreme the flood would have been if the dams had not been breached? What would the outcomes of the flood look like if dams were not built in the first place?

To analyze the characteristics of the storm over Wadi Derna, the catchment’s hydrological response, and the impact of the flood on the city of Derna, we integrate various datasets and models. Satellite-based precipitation estimations (IMERG) were used to quantify spatiotemporal storm properties and the catchment-scale rainfall, which were fed into the KINEROS2 hydrological model to quantify surface runoff upstream of the collapsed dams. The modeled flood hydrograph is then fed into a 2D hydraulic model (HEC-RAS) to test three end-member scenarios: (a) dam filling, overflow, and collapse, (b) dam overflow but no collapse, and (c) no dams exist in the wadi. This combination of methods reveals that the peak discharge during the flood was ~1,400 m3 s-1, just below the expected maximum extreme flood for this region. In the dam-collapse scenario, the populated flooded area is 40% larger than the no-dam scenario. These results emphasize the anthropogenic influence of damming natural streams on flood impacts. Given the high variability of precipitation in arid and semi-arid areas and the projected increase in extreme precipitation intensity with climate change, the Wadi Derna flood should serve as a warning sign for other populated areas downstream of a man-made dam in similar environments.

How to cite: Dente, E., Armon, M., and Shmilovitz, Y.: The September 2023 flood in Derna, Libya: an extreme weather event or man-made disaster?, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-15755,, 2024.

EGU24-16370 | ECS | Posters on site | NH1.2

Analysis of historical flood events in Denmark with information from digital news media 

Jonas Wied Pedersen, Peter Steen Mikkelsen, and Michael Brian Butts

Reliable information on historical flood events is critical for flood risk analysis, climate change adaptation, verification of forecast models, etc. Unfortunately, such information is often difficult to find, due to e.g. lack of monitoring equipment at the location of a flood. In Denmark, management of water has traditionally been the responsibility of local authorities, which means there is a limited national overview of historical events and their consequences. Previous studies have employed different strategies for compiling a flood event inventory, including mining information from (1) insurance data, (2) social media data, and (3) newspaper archives. The aim of this study is to exploit a comprehensive digital news media archive to compile an inventory of Danish flood events in the period 2007-2020 with information on the time and location of the event, to classify the type of flood, and note any available information on local consequences and damages.

We have gained access to the company Infomedia’s large digital media archive, which consists of digitized articles from news sources ranging from major national newspapers to small, local outlets. The archive contains more than 75 million news articles with the earliest articles dating back to 1990. The archive is searchable through calls to an API with a custom search language that combine user-specified keywords. A hydrologist has read all articles that match the keywords, noting all the relevant information.

1,118 distinct flooded locations where identified over the 14-year period of 2007-2020. Results show that there is large year-to-year variability in the different types of floods. Urban pluvial floods are experienced somewhere in Denmark every single year, while the number of both fluvial and storm surge floods are very low (or entirely missing) in some years. Urban pluvial floods occur throughout the year but are highly concentrated in the summer months with a mean date of occurrence in late July, while storm surges are observed only between September and March with a mean date in mid-December. Fluvial floods are the least concentrated type of floods and occur throughout the year with a slight overweight in winter months (mean date in early January). The spatial distribution of floods is uneven with four out the 10 municipalities that experience the highest number of floods being located in Eastern Jutland (Vejle, Horsens, Kolding, Aarhus) and another four located in the Northern half of Zealand (Copenhagen, Roskilde, Gribskov, Holbæk).

Storm surge events occur over large geographical areas and we therefore speculate that they are more likely to be reported in news media than urban pluvial floods, which are often local events due to the small-scale nature of convective rainfall cells. Ongoing work is trying to quantify these aspects and validate the individual flood events in the inventory using additional data sources.

How to cite: Pedersen, J. W., Mikkelsen, P. S., and Butts, M. B.: Analysis of historical flood events in Denmark with information from digital news media, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-16370,, 2024.

EGU24-16614 | Posters on site | NH1.2

A 172-year Drought Atlas for Romania  

Mihai-Gabriel Cotos, Monica Ionita, Catalin-Constantin Roibu, Adrian-Bogdan Antonescu, Petru-Cosmin Vaideanu, and Viorica Nagavciuc

In this study, we have created a 172-year historic drought catalogue for Romania by applying both the Standardized Precipitation Index (SPI) and the Standardized Precipitation Evapotranspiration Index (SPEI) to 16 long-term meteorological records/stations, covering the period 1852 – 2023. The long-term meteorological records together with documentary sources (e.g., newspapers, meteorological archives) spanning the last 172 years, are used to analyze the spatio-temporal patterns of variability, trends, and potential drivers of drought conditions, thus contributing to a nuanced understanding of Romania's hydroclimatic conditions over time. The results based on the SPEI point to the fact that the southern and eastern parts of Romania are becoming drier due to an increase in the potential evapotranspiration and mean air temperature, especially after the 1990’s. By contrast, the SPI drought index does not reveal these changes in the drought variability, mainly due to the fact that the precipitation does not exhibit a significant change. Five major drought-rich periods, in terms of duration and severity, were identified at the country level from 1852–2023, based on SPEI: 1866 – 1867, 1918 – 1920, 1947 – 1948, 2000 – 2001, and 2019 – 2022, respectively. The most pronounced drought event occurred during 2019 – 2022, followed by the 1866 – 1867 event. When analyzing the SPI-based events, similar results are found over the period 1852 – 1980, but the drought event from 2019 – 2022 is not captured by the SPI index. The most pronounced drought event, based on SPI, is the 1866 – 1867 event, followed by the 1919 – 1920 event. Nevertheless, due to the influence of the Carpathian Mountains, there are also strong regional differences in the drought events and their magnitude, with the southern and eastern parts of Romania being more affected by long-lasting drought events compared to the north-western part. Highlighting the above, a Drought Atlas for Romania (1852 – 2023) was developed using long-term meteorological data, which can provide comprehensive information on drought occurrence, magnitude and impacts over a period that goes beyond the currently available products.

How to cite: Cotos, M.-G., Ionita, M., Roibu, C.-C., Antonescu, A.-B., Vaideanu, P.-C., and Nagavciuc, V.: A 172-year Drought Atlas for Romania , EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-16614,, 2024.

EGU24-17041 | ECS | Orals | NH1.2

Evaluation of the performance of hydrological model LISFLOOD using the ECMWF seasonal meteorological forecast at 1arcmin-1day spatiotemporal resolution over German catchments 

Edgar Fabian Espitia Espitia, Yanet Diaz Esteban, Fatemeh Heidari, Qing Lin, and Elena Xoplaki

Floods and their devastating effects on society and economy have increased dramatically in Germany, and Europe in recent years. At the end of 2023, rivers and streams across Germany burst their banks due to heavy rainfall, affecting property, transport and power supplies and necessitating rescue operations and evacuations to protect human lives. One measure to deal with flooding and safeguard lives and property is the implementation of early warning systems, such as the European Flood Awareness System (EFAS), which provides short-term hydrological forecasts in real time. However, preparedness is essential along the responders value chain and longer term forecasts are important to anticipate, take precautions, raise awareness and generally mitigate the effects of flooding. The objective of this study is to evaluate the performance of hydrological forecasting using the seasonal meteorological forecast at a spatio-temporal resolution of 1 arcmin and day over Germany including all transboundary catchments for the period from 1990 to 2020. The hydrological model used was LISFLOOD. In the first step, LISFLOOD was calibrated using the meteorological observations, the EMO 1arcmin dataset and the discharge data from the transnational hydrological portal for all federal states and neighboring countries. The characteristics of land use, land cover, soil, groundwater, and human activity referred to as surface fields for global environmental modelling, were provided by EFAS. The second step, downscaling of the seasonal (long-term) forecast meteorological forcing to 1arcmin, is performed using a Deep Residual Neural Network (DRNN), and a bilinear interpolation approach over the seasonal forecast information of atmospheric conditions up to seven months into the future provided by the European Center for Medium-Range Weather Forecasts (ECMWF), 25 ensemble members in total. In the third step, the discharge is simulated by feeding the LISFLOOD model with two meteorological forcing scenarios, the DRNN downscaled and the bilinear approach of the seasonal meteorological forecast, to finally compare the performance with the observed runoff using the modified Kling-Gupta efficiency criteria (KGE'). The calibrated and validated LISFLOOD parameters showed a good and acceptable performance in all catchments, KGE' between 0.6 and 0.9. The DRNN downscaling technique shows a promising result, providing a good agreement between downscaled and observed dataset. Finally, the hydrological performance, KGE', is expected to be improved by 0.05 to 1 in the hydrological stations with good and poor performance, respectively, by using the DRNN downscaled seasonal forecast.

How to cite: Espitia Espitia, E. F., Diaz Esteban, Y., Heidari, F., Lin, Q., and Xoplaki, E.: Evaluation of the performance of hydrological model LISFLOOD using the ECMWF seasonal meteorological forecast at 1arcmin-1day spatiotemporal resolution over German catchments, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-17041,, 2024.

EGU24-18684 | ECS | Orals | NH1.2

Modeling Uncertainty of Copula-based Joint Return Period of Flood Events under Climate Change 

Ankita Manekar and Meenu Ramadas

Modeling the joint behavior of flood characteristics under climate change is necessary for understanding the potential changes in associated flood risk and hazards. In this study, we assessed the changes in flood duration, peak, and volume between historical and future periods through copula-based flood frequency analysis, employing the Soil and Water Assessment Tool (SWAT) hydrological model for modeling flood risk in a tropical watershed (Govindpur) lying in eastern India. Observed streamflow at the watershed outlet is obtained for the baseline period (1990-2014) for flood analysis. A suitable copula model is selected for bivariate flood frequency analysis while assuming copula parameters vary between baseline and future periods under climate change. In this study, high-resolution (12-km) climate reanalysis dataset from the Indian Monsoon Data Assimilation and Analysis (IMDAA) and future climate projections from general circulation models (BCC-CSM2-MR, MPI-ESM1-2-HR) after downscaling and bias correction, are used for simulating flood events using SWAT. The use of high-resolution climate data for hydrological modeling and flood frequency analysis is a novel aspect of the presented study. Uncertainty in the estimation of joint return periods of flood events under climate change due to climate model selection and assumption of stationarity is also quantified in this study for the near future (2041-2070) period under the shared socio-economic pathway (SSP585) scenario. Among the GCMs used, BCC-CSM2-MR performed relatively better in simulating baseline period streamflow in the study watershed. In this study, the Clayton copula is obtained as the most suitable based on its lowest Akaike information criterion (AIC) value, and joint return periods are then derived with the help of a conditional copula. It is found that flood events are projected to become more severe in the near future; the flood peak value increased by more than 90%, while the duration is projected to decrease. Flood volume may likely double in the future, as per our analysis, suggesting the need for mitigation and precautionary measures to reduce flood risk in the watershed. Based on the analysis, uncertainty in flood return period estimation under changed future climate is to be accounted for extreme event studies, and that can aid in managing and minimizing the flood-associated risks.

Keywords: Climate Change, Flood Frequency Analysis, Soil and Water Assessment Tool, Copula, General Circulation Model, Uncertainty Analysis

How to cite: Manekar, A. and Ramadas, M.: Modeling Uncertainty of Copula-based Joint Return Period of Flood Events under Climate Change, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-18684,, 2024.

The flood events in Germany during the summer of 2021 have once again brought to the forefront the challenges in translating scientific knowledge into effective disaster risk management practices. This paper examines the critical gap between the scientific understanding of flood risks and the practical needs of those who manage these risks. We delve into the limitations of current scientific approaches, such as flood risk and hazard mapping, in fully addressing the complexities and nuances required for practical disaster risk management, especially in the face of uncertain climate change impacts. We examine the dynamics of how flood risk information, inclusive of uncertainties, is perceived and acted upon, highlighting the psychological factors influencing these processes. The paper discusses the challenges and opportunities in translating scientific risk assessments and forecasts into practical, actionable strategies for communities and stakeholders. By highlighting the disconnects and potential areas for improvement in the science-practice interface, this paper seeks to foster a more coherent and comprehensive approach to disaster risk management. Within the framework of the Safe Development Paradox, the importance of communicating uncertainties and evaluating their potential impacts on planning and emergency responses is discussed. This paper addresses uncertainties at multiple levels and for different stakeholders, highlighting the integration of uncertainty information as a vital step in preparing for surprises and ambiguities in the context of extreme meteorological and hydrological events induced by severe weather and climate change.

How to cite: Höllermann, B.: Navigating Uncertainty in Flood Risk Perception in the Context of Climate-Induced Extremes , EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-18692,, 2024.

EGU24-19701 | Orals | NH1.2

Counterfactual floods: What if the storm track would have taken a different path? 

Bruno Merz, Viet Dung Nguyen, Guse Björn, Li Han, Xiaoxiang Guan, Oldrich Rakovec, Luis Samaniego, Bodo Ahrens, and Sergiy Vorogushyn

When a flood disaster occurs, there is an opportunity for affected individuals and decision-makers to learn from the experience. However, this learning tends to be narrowly focused on the specific event, missing the chance to discuss and prepare for even more severe or different events. For instance, regions that have been spared from havoc might feel safe and underestimate the risk. We suggest spatial counterfactual floods to encourage society to engage in discussions about exceptional events and appropriate risk management strategies. We create a series of floods across Germany by spatially shifting the rainfall fields of the 10 most expensive floods, arguing that past storm tracks could have occurred several tens of kilometers away from their actual paths. The set of spatial counterfactual floods generated includes events that are more than twice as severe as the most devastating flood in Germany since 1950. Our approach obtains peak flows that exceed the current flood-of-record at more than 70% of the gauges (369 out of 516). Spatial counterfactuals are proposed as an easy-to-understand approach to overcome society's unwillingness to consider and prepare for exceptional floods, which are expected to occur more frequently in a warmer world.

How to cite: Merz, B., Nguyen, V. D., Björn, G., Han, L., Guan, X., Rakovec, O., Samaniego, L., Ahrens, B., and Vorogushyn, S.: Counterfactual floods: What if the storm track would have taken a different path?, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-19701,, 2024.

EGU24-19897 | Orals | NH1.2

Long-Term Trends and Drivers of Hailstorms in Switzerland 

Lena Wilhelm, Olivia Martius, Katharina Schröer, and Cornelia Schwierz

Climate change affects the severity and frequency of extreme meteorological events, including hailstorms. In this regard, it is imperative to understand the factors driving the intra- and interannual variability of hailstorms. In Switzerland, this remains insufficiently understood. To address this knowledge gap, our study conducts a long-term analysis to identify potential drivers and precursors of Swiss hailstorm variability. Due to the lack of long-term data on Swiss hailstorms, we developed statistical models reconstructing hail days from 1959 to 2022, utilizing radar-based hail observations and environmental data from ERA-5. Our hailday time series shows a statistically significant positive trend in yearly hail days in both southern and northern Switzerland. This trend is mainly attributed to heightened atmospheric instability and moisture content evident in recent decades' ERA-5 data. Noteworthy natural variability is observed in both regions. To delve into the large-scale mechanisms influencing Swiss hail activity, our study uses composites to explore potential drivers and precursors. Those include soil moisture conditions, sea surface temperature anomalies, large-scale variability patterns (Piper and Kunz 2017), central European weather types (e.g., Rohrer et al. 2018), cold fronts (Schemm et al. 2015, 2016), and atmospheric blocks (e.g. Barras et al. 2021). 

How to cite: Wilhelm, L., Martius, O., Schröer, K., and Schwierz, C.: Long-Term Trends and Drivers of Hailstorms in Switzerland, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-19897,, 2024.

EGU24-19970 | ECS | Posters on site | NH1.2

Freva for ClimXtreme: helping to systematize holistic analysis of extreme events 

Etor E. Lucio-Eceiza, Christopher Kadow, Martin Bergemann, Andrej Fast, and Thomas Ludwig

Climate change is responsible for more extreme weather situations with damaging consequences. Public interest projects such as ClimXtreme [1, 2] were conceived to improve our knowledge on extreme events, the role of climate change, and their impacts. Focusing on an integrated approach, ClimXtreme evaluates the physical processes behind the extremes, their statistical assessment and their societal impact. On its second phase ClimXtreme [3] aims to open up its findings to a wider stakeholder base of different kinds.

Frameworks such as Freva (Free Evaluation System Framework [4, 5]) offer an efficient solution to handle customisable evaluation systems of large research projects, institutes or universities in the Earth system community [6-8] via the HPC environment and in a centralised manner. Mainly written in Python, Freva offers:

  • Centralised access. Freva can be accessed via command line interface, web, and a Python module with similar functionality.
  • Standardised data search. Freva allows quick and intuitive integration and searching of multiple, centrally stored data sets.
  • Flexible analysis. Freva provides a common interface for user-defined data analysis routines to be plugged into the system, regardless of the programming language. These plugins are able to search from and integrate their own results back into Freva. This environment enables an ecosystem of plugins that promotes the exchange of results and ideas between researchers, and facilitates the portability to any other research project using a Freva instance.
  • Transparent and reproducible results. Every analysis run through Freva (including parameter configuration and plugin version information) is stored in a central database and can be viewed, shared, modified and re-run by anyone within the project. Freva optimises the use of computing and storage resources and paves the way for traceability in line with the FAIR data principles [9].

The Freva instance of ClimXtreme (XCES [7]), hosted at DKRZ, provides fast access to more than 10 million data files from models (e.g. CMIP, CORDEX), observations (e.g. ERA5, HYRAS, stations) and plugin outputs. The ClimXtreme community has actively contributed plugins to XCES, its biggest asset, with nearly 20 plugins of different disciplines available to all within the project.

We would like to show a practical application of the capabilities of XCES by using it to systematise the characterisation (e.g. return periods, severity, co-occurrence...) of several past extreme events extracted from the ClimXtreme Phase 1 catalogue. Such an application can be extended to create workflows focused, for example, on the rapid assessment of the analysis of currently occurring events, allowing a quicker response to stakeholders or the public in general.














How to cite: Lucio-Eceiza, E. E., Kadow, C., Bergemann, M., Fast, A., and Ludwig, T.: Freva for ClimXtreme: helping to systematize holistic analysis of extreme events, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-19970,, 2024.

EGU24-20455 | Posters virtual | NH1.2

Investigating the effects of initial soil moisture and the uncertainty of Manning friction coefficient on flood hazard estimation and mapping. 

Athanasios Loukas, Anastasios Katsiolas, and George Papaioannou

Floods are among the most devastating water-related hazards and are primarily responsible for the loss of human life and destruction of the natural and man-made environment. This study addresses the estimation and mapping of flood hazard in small mountain watersheds with urban areas at the lowlands and the related uncertainty. Specifically, this research studies the flood hazard for the Metropolitan city of Volos in Central Greece, which is frequently affected by intense storms that cause flash floods. The above study area is crossed by three (3) streams.The methodology used in the study is divided into three stages. At first the 24-hour design storm hydrographs were constructed for the three sub-basins of the study area with using the mean IDF parameters and the relevant confidence limits. The Alternating Block Method was used for the design hyetographs for return periods, T = 50-year, T=100-year and T=1000-year (worst-case scenario). The second stage concerns the hydrological analysis using a rainfall-runoff model. Firstly, the net rainfall was estimated by using the U.S. Soil Conservation Service (SCS-CN) method for three (3) soil's Antecedent Moisture Conditions (AMC) for dry-average-wet conditions. Then, the net rainfall was transformed by using the Instantaneous Unit Clark hydrograph into discharge and the flood hydrographs for each return period were estimated. At the final stage, the flood hydrograph estimated for each watershed was routed through the hydrographic network using the HEC-RAS 2D hydraulic-hydrodynamic simulation (2D) model.  For the flow routing, Manning’s n was estimated for various cross sections by visual inspection and corresponding values reported in international reports. The “upper” and “lower” boundaries of Manning’s n were estimated as the -50% and +50% of the average Manning’s n values, respectively. In this simulation approach, flood hazard maps for three return periods, T=50, T=100 and T=1000 years considering three different soil moisture conditions and three different values of Manning’s n have been estimated. The values of Manning’s n in the flood plain were estimated by using land cover/land use data.  The flow routing with in the urban areas was simulated by the block rising method. In total twenty-seven (27) flood scenarios have been simulated for each watershed. The results were validated with the flooded areas during a specific historical flood event using the Critical Success Index (CSI) method and reports and photographs of the historical flood event. The results of hydrological analysis and hydraulic simulation were also compared with the results of the Greek Flood Hazard Management Plans.

How to cite: Loukas, A., Katsiolas, A., and Papaioannou, G.: Investigating the effects of initial soil moisture and the uncertainty of Manning friction coefficient on flood hazard estimation and mapping., EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-20455,, 2024.

EGU24-22160 | Posters on site | NH1.2

Novel approach to quantifying long-term rainfall distribution variation: the region of Europe 

Andrew Barnes and Ioanna Stamataki

Climate change is changing rainfall and flood regimes across the world with severe and widespread impacts on society. Rainfall extremes are intensifying in frequency and magnitude due to the effects of climate change, and thus in this research, we introduce a new, novel framework for understanding how rainfall distributions are changing through time, enabling more accurate flood risk analysis. The framework offers two approaches to comparing rainfall distributions, the first of these utilises a stagnant benchmark distribution and the second highlights a moving benchmark approach. When combined the framework enables the identification of significant sudden and gradual changes in the distributions without the need to fit statistical distributions to the data.

 The region of Europe is selected as the case study and analysed in the four UN regions of Europe: Northern Europe, Eastern Europe, Southern Europe, and Western Europe. Using daily precipitation data generated using the ERA5 Reanalysis hourly data from the ECMWF’s Copernicus data store, the case study is used to highlight the capability of both frameworks to capture different forms of rainfall distribution shift.

 Comparing the frameworks presented revealed similar long term changes in the rainfall variation. The stagnant comparison showed that rainfall distributions have intensified since 1940 with a clear increase across all four regions of Europe regarding the percentage of days with rainfall, averaging at 2.75% across Europe. The largest changes seen are in the last comparison period for Eastern Europe (1960-1975) at 3.07% and in the latest comparison period (2005-2020) for Northern Europe (2.64%). The moving comparison method unveiled the strongest changes between the periods 1940-1960 and 1960-1980 with an average of 2.09% of rainfall days being intensified across all Europe. The most considerable shifts in rainfall variability occurred in Eastern (2.39%) and Western Europe (2.72%) during the 1960-1980 period.

 By applying it over the European region, this paper demonstrated how this novel approach can be used to identify long-term rainfall variation in the 20th century. The suggested frameworks do not rely on fitting statistical distributions and thus enable both long and short term change identification, providing flood risk managers a new solution to understanding local, regional and global rainfall variability and quantification. The analysis of the changing dynamics of precipitation patterns and the increase of the intensity of precipitation events, offers considerable potential for further investigations in the mitigation strategies of a resilient future.

How to cite: Barnes, A. and Stamataki, I.: Novel approach to quantifying long-term rainfall distribution variation: the region of Europe, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-22160,, 2024.

The intensification of extreme precipitation in a warming climate has been shown in observations and climate models to follow approximately theoretical Clausius-Clapeyron scaling. However, larger changes have been indicated in events of short-duration which frequently trigger flash floods or landslides, causing loss of life. Global analyses of continental-scale convection-permitting climate models (CPCMs) and new observational datasets will be presented that provide the state-of-the-art in understanding changes to extreme weather (rainfall, wind, hail, lightning) and their compounding effects with global warming. These analyses suggest that not only warming, but dynamical circulation changes, are important in the manifestation of change to some types of extreme weather, which must be addressed in the design of new CPCM ensembles. We use our projections to provide the first analyses of impacts on infrastructure systems using a new consequence forecasting framework and show the implications for adaptation. It will be argued that a shift in focus is needed towards examining extreme weather events in the context of their ‘ingredients’ through their evolution in time and space. Coupled with exploration of their causal pathways, sequencing, and compounding effects – ‘storylines’ –, this can be used to improve both early warning systems and projections of extreme weather events for climate adaptation.

How to cite: Fowler, H.: Rapidly intensifying extreme weather events in a warming world: how important are large-scale dynamics in generating extreme floods?, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-22472,, 2024.

EGU24-532 | ECS | Orals | NH1.3 | Highlight

Rapid unsupervised economic assessment of urban flood damage using SAR images 

jeremy Eudaric, Andres Camero, Kasra Rafiezadeh Shahi, Heidi Kreibich, Sandro Martinis, and Xiao Xiang Zhu

Climate change projections for 2030 indicate a concerning increase in the frequency of floods, which is expected to result in significant economic damages and losses on a global scale. The growth of urbanization has indeed increased flood risk, highlighting the need for a prompt evaluation of economic losses to facilitate rapid response and effective reconstruction. However, providing timely and accurate economic damage assessment immediately after a flood event is difficult and associated with high uncertainty. Remote sensing  data can support this task, but challenges such as cloud cover, infrequent return times from satellites, and the lack of ground truth data make supervised approaches challenging. To address these challenges, we propose a new economic damage assessment approach based on the analysis of multi-temporal and multi-source, Synthetic Aperture Radar (SAR) images before and after the flood peak with an unsupervised change detection method. This method utilizes computer vision techniques, specifically a pixel-based approach with SAR data (Sentinel-1 and TerraSAR-X/TanDEM-X) to monitor changes in buildings and the flood extension. It employs various threshold techniques and parameters to determine the optimal threshold values for highlighting changes and the presence of water. By using this method, our aim is to obtain an economic model based on pixels, which represents the volume of water surrounding or on each building and the flood extension. The purpose of this study is to support governments in decision-making processes and enable insurers to efficiently assess and compensate for damages caused by flood events. 

How to cite: Eudaric, J., Camero, A., Rafiezadeh Shahi, K., Kreibich, H., Martinis, S., and Zhu, X. X.: Rapid unsupervised economic assessment of urban flood damage using SAR images, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-532,, 2024.

EGU24-957 | ECS | Posters on site | NH1.3

Flood Mapping Using High-Resolution Topography and Crowdsourced Data with the Geomorphic HAND Approach in Rural Plains 

Hassan Sabeh, Marie-George Tournoud, Nanée Chahinian, Chadi Abdallah, Roger Moussa, and Rouya Hdeib

Flood mapping is essential for risk management and emergency response. The most common approach is hydraulic modelling, a method that is still challenging and demanding in terms of data and computation. Low complexity models are an increasingly adopted alternative that are capable of achieving good results while using minimal data input and low calculation time. Yet, the reliability and effectiveness of such approaches remain unclear in flat and engineered plains. In this study we aim to optimize flood hazard mapping based on the Height Above Nearest Drainage (HAND) geomorphic approach by utilizing a high-resolution digital elevation model (15 cm) with crowdsourced data. The approach is tested on the Ostouane river basin (144 km2) in Lebanon, and validated using crowdsourced data of the January 2019 flood, which was the most intense flood within the past decade. The workflow begins by developing a database of spatial and topographic information, including the digital elevation model, bathymetry, land use and crowdsourced flood depths. Five scenarios representing different terrain configurations with varying levels of hydro-conditioning and feature inclusion (e.g. bathymetry, canals and levees) are simulated. The model’s thresholding is then optimized by integrating rating curves produced by 1D HEC-RAS hydraulic model to assess and correct HAND based synthetic rating curves (SRC). Results shows that extensive hydro-conditioning is necessary to improve the inundation extents within the floodplains. Correcting synthetic rating curves is essential to overcome errors produced by terrain conditioning. Overall, the model is able to yield high accuracy of flood extent when ensuring hydrologic connectivity between the river and floodplain and within the floodplain itself. Our findings indicate that leveraging high-resolution topography and crowdsourced inputs can enhance the accuracy of flood mapping results. However, achieving this precision necessitates a meticulous optimization procedure.

How to cite: Sabeh, H., Tournoud, M.-G., Chahinian, N., Abdallah, C., Moussa, R., and Hdeib, R.: Flood Mapping Using High-Resolution Topography and Crowdsourced Data with the Geomorphic HAND Approach in Rural Plains, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-957,, 2024.

EGU24-1081 | ECS | Orals | NH1.3

CFD modelling to investigate hydrodynamic forces on bridges in case of submergence and material deposition  

Natasha Petruccelli, Diego Panici, Alessio Domeneghetti, and Armando Brath

Increasingly frequent and intense flood events, combined with the remarkable industrialization process of cities, are placing transportation networks under stress. The loads that roads and railways must resist are nowadays often greater than those considered for their design; furthermore, their state of ageing is such that any disturbance (flood, earthquake, landslide) could cause a total or partial interruption of traffic resulting in socio-economic losses.
Bridges represent the most vulnerable component of a transport system and their failure can compromise the functionality of the entire network, as well as causing loss of life. During floods, bridges can be partially or completely submerged, having to withstand higher hydrodynamic loads which can lead to the collapse of the structure itself. Furthermore, accumulations of large wood and scour total around the bridge piers can reduce the load-bearing capacity of the structure and therefore its structural integrity.
In this study, we investigated the hydrodynamic actions and the 3-dimensional flow field at a model bridge (comprising deck and pier) using CFD (Computational Fluid Dynamics) modelling. Drag and lift forces acting on the rectangular-shaped deck were estimated for different submergence values to evaluate the structure's maximum permissible load. In particular, drag and lift coefficients were calculated by simulating various flow conditions (Froude number varying between  0.16 and 0.50) and adopting three different turbulence models (RNG, k-ε, k-ω).
In addition, the effect on the drag coefficient of the accumulation of large wood around the pier was also examined, considering different geometries. Numerical simulations, performed for both fixed and live river bed conditions, were validated using experimental data. However, the trends of the synthetic curves constructed so far have presented characteristics similar to those present in the literature, with all positive values ​​for the drag coefficient and negative ​​for the lift coefficient. 
The emerging evaluations allow us to provide useful indications to designers to evaluate the possible state of stresses on existing bridges and improve knowledge for designing new ones.

How to cite: Petruccelli, N., Panici, D., Domeneghetti, A., and Brath, A.: CFD modelling to investigate hydrodynamic forces on bridges in case of submergence and material deposition , EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-1081,, 2024.

EGU24-1594 | ECS | Orals | NH1.3

Drivers of coastal real estate demand under flood model predictions in the UK 

Avidesh Seenath, Scott Mark Romeo Mahadeo, Matthew Blackett, and Jade Catterson

Flood model predictions are becoming increasingly available online through open access flood risk maps and communications. While these predictions are important for flood management, their inherent uncertainty presents a considerable risk for real estate markets, a leading indicator of macroeconomic performance.  We, therefore, need to understand the factors influencing real estate demand in an era of open access flood model predictions. Here, we investigate the role of gender, education, employment, place of residence, caring responsibilities, income, insurance, location preferences, level of risk aversion, and flood experience and awareness on coastal real estate demand decisions in the UK in response to flood model predictions. Here, our objective is test whether access to flood predictions is a leading driver of real estate demand decisions or whether alternative factors influence how people perceive such predictions. We achieve this by applying an inter-disciplinary approach, involving numerical flood modelling, a novel experimental willingness-to-pay real estate survey of UK residents in response to flood model outputs, statistical and geospatial modelling, and thematic analysis. Our preliminary findings indicate that access to flood model predictions is the primary factor influencing real estate demand decisions, whereas alternative factors considered have negligible impact. Such preliminary findings suggest that we need to re-think how flood model predictions are communicated in order to minimise real estate risks.  

How to cite: Seenath, A., Mahadeo, S. M. R., Blackett, M., and Catterson, J.: Drivers of coastal real estate demand under flood model predictions in the UK, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-1594,, 2024.

EGU24-1692 | Orals | NH1.3 | Highlight

A Fresh Start for Flood Estimation in Ungauged Catchments 

Ross Woods, Yiming Yin, Giulia Evangelista, Pierluigi Claps, Giulia Giani, Yanchen Zheng, Gemma Coxon, Roberto Quaglia, Dawei Han, and Miguel Rico-Ramirez

Flood estimation in ungauged basins is important for flood design, and for improving our understanding of the sensitivity of flood magnitude to changes in climate and land cover. Flood estimates in ungauged basins by current methods (e.g. statistical regression, unit hydrograph) have high uncertainty, even in places with dense observing networks (e.g. +/- 50-100% in the UK). Reductions in this uncertainty are being sought by using alternative methods, such as continuous simulation using hydrological models (spatially-distributed or lumped), and event-scale derived distribution approaches. The very significant challenges for reliable application of continuous simulation models in ungauged catchments are well known. So far there has been only limited application of machine learning techniques to this problem, but it seems an obvious route to try, but to exploit the big-data strengths of this approach, the problem must be recast to extract information from many more events at each site than just annual maximum events.

The event-scale derived distribution approach also has challenges, which we explore below. The derived distribution approach at the event scale typically combines the following elements: a stochastic rainfall model, an event-scale rainfall-runoff model (including “losses” and a “baseflow” component), and a runoff routing model. In principle, every element of this approach may be considered as a (seasonally varying) random variable. The flood peak distribution is obtained by integrating over joint distributions of the model elements. After giving an overview of our approach, I will focus on challenges regarding the catchment response time associated with flood events.

How should we define catchment response time? Why do we need this quantity and how will it be used? What are the relative merits of empirical and model/theory-based approaches? Specifically, I will discuss the empirical DMCA method for catchment response time of Giani et al, How is it relevant for ungauged catchments? What does DMCA really measure? How do we assign hydrological meaning to this empirical response time? How does this response time vary between events and catchments?

How to cite: Woods, R., Yin, Y., Evangelista, G., Claps, P., Giani, G., Zheng, Y., Coxon, G., Quaglia, R., Han, D., and Rico-Ramirez, M.: A Fresh Start for Flood Estimation in Ungauged Catchments, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-1692,, 2024.

EGU24-3629 | ECS | Orals | NH1.3

Flood plain inundation modeling with explicit description of land surface macrotopography 

Simone Pizzileo, Giovanni Moretti, and Stefano Orlandini

Land surface topography plays an essential role in flood plain inundation modeling. High-resolution digital surface models (DSMs) based on LiDAR surveys have become increasingly accessible in various geographical areas. Nevertheless, common practice involves filtering out land surface macrostructures, such as trees and buildings, by using obtained digital terrain models (DTMs) to represent the land surface hydraulic geometry. This is done by letting resistance coefficients represent the effects of both micro and macrostructures on surface flow propagation. In addition, significant information loss is observed when digital terrain models are coarsened for computational efficiency.

In the present study, physically meaningful unstructured meshes are automatically extracted from high-resolution digital surface models to explicitly describe land surface macrostructures. This is achieved by extracting relevant ridges at a selected level of representation without applying any coarsening or depression filling pre-processing. The effects of these macrostructures on floodwater propagation are evaluated by comparing simulations obtained by using digital terrain models and related Manning coefficients, simulations obtained by using digital surface models representing land surface macrostructures and related Manning coefficients, and observations for a real flood inundation event occurred after a levee failure in the lowlands adjoining the Panaro River in Northern Italy in 2020.

The explicit description of land surface macrostructures based on a 1-m digital surface model is found to yield a 42% improvement in the prediction of flooded area extent, a 36% improvement in the prediction of flooded areal position, and a 24% improvement in the prediction of flood plain inundation travel time with respect to the case in which resistance coefficients representing both land surface micro and macrostructures are used. Unstructured meshing of land surface macrostructures based on extracted ridge networks is essential for achieving a detailed description of land surface hydraulic geometry without altering the original topographic data, while also preserving computational efficiency. The obtained results highlight the role of natural and human-made macrotopographic structures in delineating flood plain inundation models and generating flood hazard mapping. These tools represent valuable assets in the context of Emergency Action Planning (EAP) and prevention strategies.

How to cite: Pizzileo, S., Moretti, G., and Orlandini, S.: Flood plain inundation modeling with explicit description of land surface macrotopography, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-3629,, 2024.

EGU24-3747 | Posters on site | NH1.3

Developing a flood image detection model using deep learning algorithms 

Cheng-Lin Yang

As the impact of climate change intensifies, the frequency of short-duration heavy rainfall events gradually increases, posing a serious challenge to urban infrastructure and underground drainage systems. Assessing flood-prone areas and disaster extents relies heavily on manual surveys, lacking real-time and effective methodologies. Our study uses Mask R-CNN deep learning and closed-circuit television (CCTV) flood images to develop a real-time and effective flood detection model. The results of our study demonstrate that the proposed flood image recognition model achieves a precision of 60.6%, a recall rate of 92.2%, and an F1 score of 73.1 for the flood category. These results signify the model's exceptional capability of the model in flood detection. Additionally, through on-site measurements of road dimensions and binary matrix-based area estimation, the average error is only 1.6%. This model can be applied effectively and serves as a reference for authorities to promptly determine the occurrence of flooding and the extent of the disaster, thus facilitating the formulation of more effective disaster response measures. The developed model exhibits promising potential for real-time flood detection in urban disaster management, providing a valuable tool for authorities to respond promptly to the dynamic challenges posed by climate change.

How to cite: Yang, C.-L.: Developing a flood image detection model using deep learning algorithms, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-3747,, 2024.

EGU24-4842 | ECS | Posters on site | NH1.3

Substantial increase in future fluvial flood risk projected in China's major urban agglomerations 

Ruijie Jiang, Hui Lu, Kun Yang, Deliang Chen, Jiayue Zhou, Dai Yamazaki, Ming Pan, Wenyu Li, Nan Xu, Yuan Yang, Dabo Guan, and Fuqiang Tian

Floods are one of the most destructive natural disasters and projecting future flood risk is essential for protecting lives and livelihoods. China is in the process of rapid urbanization, and most of the urban agglomerations are distributed on floodplains, facing high fluvial flood risk. The effect of urban spatial expansion, instead of densification of assets within existing urban cells, on flood risk has rarely been reported. Here, based on the latest projected urban land data and bias-corrected CMIP6 outputs, we project the future flood risk of seven urban agglomerations in China, home to over 750 million people. The inundated urban land areas in the future are projected to be 4 to 19 times that at present, with southern China facing the greatest increase. Although climate change is the main driver for this strong projected rise in flood risk, the inundated urban land areas will be underestimated by 10-50% if the urban spatial expansion is not considered. Urban land is more likely to be inundated than non-urban land, and the newly-developed urban land will be inundated more easily than the historical urban land due to the marginal expansion of urban land. The results demonstrate the urgency of integrating climate change mitigation, reasonable urban land expansion, and increased flood protection levels to minimize the flood risk in urban land.

How to cite: Jiang, R., Lu, H., Yang, K., Chen, D., Zhou, J., Yamazaki, D., Pan, M., Li, W., Xu, N., Yang, Y., Guan, D., and Tian, F.: Substantial increase in future fluvial flood risk projected in China's major urban agglomerations, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-4842,, 2024.

EGU24-5935 | ECS | Orals | NH1.3

Assessing the flooding hazard through a probabilistic approach including earthen levees vulnerability estimate     

Bianca Bonaccorsi, Silvia Barbetta, and Giuseppe Tito Aronica

Levees collapse cause huge financial and social losses, especially in highly developed areas of many countries. Since that 2563 floods occurred in Europe between 1980 and 2010 (EEA, 2018), the European Parliament issued the Floods Directive, approved in 2009, in which EU Member States are invited to minimise this risk of failure, improving methods and finding simple solutions for large-scale application. For this reason, the scientific community is gradually performing stochastic approaches allow a large number of simulations runs in a Monte Carlo framework, providing the basis for a probabilistic risk assessment considering also the influence of levee breaches on the flood risk (Apel et al., 2006, Castellarin et al., 2011). Indeed, in many studies, seepage analyses account only the hydraulic boundary conditions, i.e.  the water head upstream of the embankment (Tracy et al., 2016, 2020).  

In this context, the present work is focused on the evaluation of the residual flood risk through the analysis of earthen levees’ seepage vulnerability. In particular, the levee fragility curves determined with the use of simplified and expeditious approaches and those assessed by using geotechnical finite element models (i.e. PLAXIS 2D) are compared. Furthermore, the goal of this study is to find the relation between the frequency of levee’s failure due to hydraulic and geotechnical conditions, to aim of define the conditional probability of the residual flood risk.



Apel, H., Annegret, H. T., Bruno, M., & Günter, B. (2006). A probabilistic modelling system for assessing flood risks. Natural Hazard, 38, 79-100.

Castellarin, A., Di Baldassare, G., & Brath, A. (2011). Floodplain management strategies for flood attenuation in the River Po. River Research and Applications, 27(8), 1037 –1047.

EEA, European Environment Agency. (2018). European past floods [Online]. Copenhagen, Denmark: Author. Retrieved from .

Tracy, F.T., Brandon, T. L., Corcoran, M.K. (2016). Transient seepage analyses in levee engineering practice, Technical Report TR-16-8, U.S. Army Engineer Research and Development Center, Vicksburg, MS,

Tracy, F.T., Ryder, J.L., Schultz, M.T., Ellithy, G.S., Breland, B.R., Massey, T.C., Corcoran, M.K. (2020). Monte Carlo Simulations of Coupled Transient Seepage Flow and Soil Deformation in Levees. Scalable Computing Practice and Experience 21(1):147-156.

How to cite: Bonaccorsi, B., Barbetta, S., and Aronica, G. T.: Assessing the flooding hazard through a probabilistic approach including earthen levees vulnerability estimate    , EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-5935,, 2024.

EGU24-6509 | Orals | NH1.3

Dam operating strategies and hydrodynamic modeling to mitigate floods in the lower Tagus valley similar to the 1979 catastrophic event  

Diego Fernández-Nóvoa, Alexandre M. Ramos, José González-Cao, Orlando García-Feal, Cristina Catita, Moncho Gómez-Gesteira, and Ricardo M. Trigo

The lower valley of the Tagus River, one of the most important rivers in the Iberian Peninsula, is a particularly relevant and vulnerable area in terms of flood impact. This valley is characterized by a flattened and large alluvial plain, which implies that floods can affect large areas of territory, causing significant damage and affecting a large number of people. Although several floods have occurred in the lower Tagus valley, the one in February 1979 stands out, since the vast flooded area affected around 10,000 people, many of whom were evacuated or made homeless. The Tagus River flow in its lower valley is controlled, to a large extent, by the functioning of the Alcántara dam, which has the largest water storage in the Tagus basin. In this context, this study aims to develop strategies to take advantage of this infrastructure to effectively mitigate floods in the lower Tagus valley. For that, dam operating strategies, focused on flood mitigation, are developed sustained on a sequence of logical principles, such as avoiding inducing man-made floods or maintaining average water storage similar to the actual one. The effectiveness of the proposed strategies, in terms of flood mitigation, is analyzed by applying the Iber+ hydrodynamic model. For this, the numerical model is validated in the lower Tagus valley by evaluating its ability to reproduce the outstanding flood of 1979. Additionally, several Digital Elevation Models (DEMs) are also analyzed to determine which is the most accurate for the area under scope. The results show that Iber+ model, coupled with Copernicus DEM, is able to provide an efficient reproduction of this flood. In particular, the simulation shows good agreement with some descriptions and watermarks available for the 1979 event. This also allows the analysis of this historical event from a hydrologic-hydraulic perspective, which contributes to improving knowledge and understanding of how floods occur and develop in the lower Tagus valley.

Regarding flood mitigation, results indicate that, since 1970, when data is available, the frequency of floods is reduced by more than 80%, compared to the natural flow regime, with the application of the proposed strategies. In addition, the mitigation of the most extreme floods that occurred during the analyzed period, is also achieved. In particular, peak river flows are reduced for the most extreme events. This implies that flood extension is reduced by around 5-10% in the lower Tagus valley. A more efficient mitigation is achieved for flood indicators closely linked to the damage caused by these events. Thus, water depth is reduced by around 25% and water velocity by around 25-30%, in the flooded areas, for the most extreme events. This corroborates the effectiveness of the proposed dam operating strategies to mitigate floods in the lower Tagus valley through an adequate dam functioning.

The developed proposal provides an affordable approach to flood mitigation in comparison with the construction of additional structural measures, which could also be applicable to other areas vulnerable to floods affected by dam-regulated rivers.

How to cite: Fernández-Nóvoa, D., Ramos, A. M., González-Cao, J., García-Feal, O., Catita, C., Gómez-Gesteira, M., and Trigo, R. M.: Dam operating strategies and hydrodynamic modeling to mitigate floods in the lower Tagus valley similar to the 1979 catastrophic event , EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-6509,, 2024.

Developing reliable and efficient flood modelling systems on a large scale is crucial for addressing errors and inconsistencies in both observations and modelling. However, the computational demands of hydrodynamic models have constrained their widespread application to coarse resolutions (30m-1km), compromising accuracy by neglecting the local and small-scale features that may significantly influence flooding, especially in urban areas. Furthermore, traditional models struggle to effectively incorporate river bathymetry, especially given the significant flood volume conveyed by the river channel during floods. These models often rely on surveyed cross-sections for river channel representation, leading to missing topography between cross-sections and hindering the resolution of complex floodplain flow paths. To resolve small-scale effects in limited areas while simulating large domains, grid adaptation methodologies are implemented in this project to locally adjust the resolution of the computation in a static or a dynamic way. A hybrid 1D-2D flood model is developed, incorporating the static/dynamic adaptive mesh generation and an integrated sub-/super grid channel model. The sub-/super grid channel is applied to accommodate situations where river channel width exceeds or fall below the grid resolution. Parallelized with GPU architecture, the performance of hybrid 1D-2D with either static or dynamics nonuniform structured grid was thoroughly evaluated, benchmarked with the full resolution CPU solver, shedding light on their effectiveness in enhancing flood modelling approach.

How to cite: Rong, Y., Bates, P., and Neal, J.: Towards a large-scale locally relevant flood modelling using adaptive mesh generation and an integrated sub-/super grid channel solver, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-6787,, 2024.

Recent climate change has led to extreme floods surpassing levee design criteria, posing a threat to safety. Consequently, there is a demand for the development of technologies capable of handling such severe floods. In this study, a method assessing failure probabilities, represented by fragility curves, was developed for the levee slope under rapid drawdown. The time-dependent probabilistic stability assessment of the levee slope due to a water level drop was explored. Integrating seepage analysis results from finite element analysis with slope stability analysis, Monte Carlo simulations were conducted to scrutinize the time-dependent behavior of the levee slope under rapid drawdown conditions. The probability of failure was calculated to develop fragility curves for the levee slope. The developed fragility curves were significantly influenced by the drawdown rate. Since the drawdown rate is determined through hydraulic analysis based on flood scenarios, the stability of the water-side slope of the embankment due to a water level drop will be greatly affected by climate change. The fragility curves obtained using the proposed methods are valuable for risk assessment, offering information to evaluate the performance of the levee under various water level drawdown conditions.

How to cite: Cho, S. E.: Fragility assessment of levee based on time-dependent reliability analysis under rapid drawdown, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-7672,, 2024.

Levees are linear structures that can be thousands of kilometers long and play a very important role in flood protection. They are usually monitored by traditional direct survey techniques, such as CPTU or coring, or piezometers, which provide high accuracy, but are localized and performed in predetermined locations.

As a result, long distances between investigated sections limit the detailed analysis of the entire structure. In addition, predetermined locations may not cover areas of actual potential weakness. Recently, new survey technologies from aerial media (drones) have been successfully applied to obtain a first level of levee investigation in order to identify the location of possible weak areas or potential locations of levee failure, so as to plan further local investigations in those areas.

Usually, levee failures are localized in the presence of:

(i) concrete or other materials structures passing the levee;

(ii) large trees, which can be dangerous because their roots are a preferred route for water infiltration. In addition, at higher erosion levels of the river bank, large trees can promote bank collapse due to their weight (i.e. cantilever failure);

(iii) sections where unfavorable conditions of the levee body, such as soils with high permeability or the presence of animal burrows crossing the levee or obstructed drains, prevent proper drainage and bring the phreatic surface close to the levee surface.

From previous experience, we have noticed that several times levee failures have occurred at sections previously vegetated by reeds. Reed canes usually grow on sandy soils and, in addition, are characterized by very deep and large roots, possible routes of localized infiltration through the body of the levee. From these observations comes the idea of using reedbeds as indicators of sandy soils and possible weak levee sections;

Thus, we performed two UAV-supported surveys on the same test area aimed at identifying the position and extension of the reeds vegetated areas, in combination with local on-site surveys with soil sampling along levee transversal sections, to compare and combine the obtained results. The RGB orthophotos obtained by the two surveys have been elaborated to determine the DSM and the vegetation cover map of the embankment, to compare them in different seasons. The obtained data have been calibrated with on-site surveys conducted by vegetation experts. To facilitate the identification of reedbeds, the first campaign has been carried out in winter, when reedbeds are yellowish in color, unlike short grass. In areas identified as reedbed vegetated, the soil has been sampled by coring and fully classified in the geotechnical laboratory to check if reedbed can effectively be an indicator of sandy soils. Similarly, other samples have been taken from sections not covered by reeds for comparison.

The final aim is to test the possibility of using vegetation maps as an indicator of weak sections of the embankment, thus to develop an innovative method of low-cost aerial monitoring of levee structures that can provide an initial state of information and identify areas in need of further direct investigation in order to define the necessary maintenance works, decreasing associated risks.

How to cite: Dalla Santa, G. and Simonini, P.: Reeds influence of levee hazards: detection through UAV survey and soil geotechnical analysis, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-8071,, 2024.

EGU24-9050 | ECS | Orals | NH1.3 | Highlight

Is 3D modelling necessary for simulating long-duration urban flooding? 

Shuyue Yu, Xuefang Li, Vasileios Kitsikoudis, Guilhem Dellinger, and Benjamin Dewals

Urban flood modeling within complex urban environments demands sophisticated methodologies. While 2D computational models have historically served as foundational tools, their inherent limitations in capturing the intricate three-dimensional dynamics necessitate further exploration. Our research endeavors to expand this understanding by delving into 3D computational simulations, providing a more holistic perspective on urban flood dynamics.

In the current research, we conducted 3D simulations to replicate urban flood processes, drawing comparisons with earlier 2D modeling results and experimental observations. The simulations were executed considering various urban layouts and turbulence closure models. The urban layouts include two groups, totaling 13 architectural models. These models feature varying numbers or positions of openings on their exterior walls to represent architectural elements such as doors and windows that could allow floodwaters to enter in the interior of the buildings. As for the turbulence equations, k-omega SST and k-epsilon were considered. By analyzing the surface velocity, flow depth, and flowrate distribution, preliminary findings indicate that 3D simulations offer enhanced accuracy in capturing intricate flow patterns within urban settings compared to their 2D counterparts. Moreover, the tested simulations from various turbulence models influence the 2D and 3D simulations in different ways. This direct comparison allowed us to dissect and understand the influence of turbulence modeling on the accuracy of 3D simulations, thereby enhancing the robustness of our findings.

After obtaining the relevant results, we applied them to flood risk analysis. Compared to traditional 2D analyses, we derived some new insights to guide informed decision-making, enhancing the applicability of our approach. By integrating sophisticated modeling techniques and risk evaluations, this study paves the way for more resilient and adaptive urban planning strategies, ensuring safer and more sustainable urban environments in the face of increasing flood challenges.

How to cite: Yu, S., Li, X., Kitsikoudis, V., Dellinger, G., and Dewals, B.: Is 3D modelling necessary for simulating long-duration urban flooding?, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-9050,, 2024.

EGU24-9555 | ECS | Orals | NH1.3

A tracer-aided criterion to discretize pluvial and fluvial flood hazard maps in catchment scale shallow water models 

Pasquale Perrini, Luis Cea, Andrea Gioia, and Vito Iacobellis

Knowing the nature of the flood hazard is a crucial factor for improving the resilience of the urban areas, since awareness, preparedness and early warning systems are based on the scientific tools such as the 2D depth-averaged shallow water models. Inland flood hazard primarily stems from pluvial and fluvial inundations, typically modeled separately respecting the pertaining spatial domains of the assessment, namely the urban areas and the riverine floodplains. Considering the high computational power and efficiency of both hardware and codes, the catchment scale hydrological-hydrodynamic modeling is becoming an increasingly adopted approach in flood hazard assessments. Since a complete rainfall-induced routing is preserved, these simulators determine fluvial, pluvial and compound inundations caused by heavy storm events within the entire watershed.

However, this approach leads to flood extent maps in which the inundations such as those resulting from pluvial and fluvial processes, are usually not differentiated, even if significant disparity in the space-time scales and volumes of water are involved. Indeed, these two hazards follow distinct normative and regulatory flood risk management rules among different countries. 

With such a rationale we established a tracer-aided criterion to systematically categorize and map pluvial and fluvial hazard in a catchment scale shallow water model, exploiting the advection process of a conservative tracer. The physically based methodology, implemented in the GPU-parallelized Iber+ software and its water-quality module (IberWQ+), is applied in a small urban catchment for multiple probabilistic scenarios. The results demonstrate the effectiveness of nesting transport and shallow water equations, univocally discretizing the two inundation sources in function of the computational cells reached by the tracer. This enables to define the spatial domains of the pluvial and fluvial processes, providing valuable insights for holistic catchment-scale flood risk management. Additionally, the advancements achieved by the proposed method are showcased in comparison to commonly employed modeling techniques for mapping fluvial inundations. As the tracers continue to improve our understanding of catchment sciences, we conceptualized them role through an abstraction that can aid surface hydrodynamic modelling to identify pluvial and fluvial sources of hazard.

How to cite: Perrini, P., Cea, L., Gioia, A., and Iacobellis, V.: A tracer-aided criterion to discretize pluvial and fluvial flood hazard maps in catchment scale shallow water models, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-9555,, 2024.

EGU24-11106 | ECS | Posters virtual | NH1.3

Impact of vegetation on urban open-channel flow: Practical experiment with 2D IBER hydraulic simulations in Monterrey, Mexico. 

Jacob Jesús Nieto Butrón, Nelly Lucero Ramírez Serrato, Selene Barco Coyote, Fabiola Doracely Yépez Rincon, and Mariana Patricia Jácome Paz

Flooding is a constant danger in many cities. To prevent and mitigate their impacts, mathematical modeling is carried out to simulate the behavior of the flow within the environment and define the possible flood zones. Incorporating vegetation in hydraulic models is pivotal for understanding its impact on flow characteristics, sediment transport, and channel morphology.

The Santa Catarina River in Monterrey, Nuevo León, Mexico, grapples with irregular water flows. During dry seasons, minimal water levels promote unchecked vegetation growth along its banks and bed, potentially obstructing normal flow. Conversely, extreme weather events like hurricanes lead to rapid surges, sweeping away vegetation and debris. Balancing this fluctuation—from sparse to intense flows—presents challenges in managing the river's vegetation, necessitating strategies that reconcile environmental preservation with urban infrastructure resilience.

For this purpose, This study utilized two hydraulic models through IBER to assess vegetation's impact on flood simulations. One model employed a Digital Elevation Model (DEM), portraying terrain topography. The second model used a Digital Surface Model (DSM) integrating manually digitized vegetation from (2020) Google Earth imagery. Assigned heights of 3m for shrubs and 15m for trees emulated their impact on water flow. Both the DEM and DSM, with a 5-meter resolution, were obtained via LiDAR techniques from the INEGI government web platform. the models also utilized a land use classification obtained from a Sentinel-2 satellite image (from 2023). Hydrological data for both models were derived from the cumulative rainfall during Hurricane Alex in 2010.

The findings highlight significant changes in flood patterns attributed to vegetation. Its presence alters the flow, shifting the flood zone towards a southwest residential-commercial area. In this integrated model, the maximum depth reaches 16.78 meters, compared to 10.70 meters in the DEM-based hydraulic model. Additionally, the consistently affected area deepens from 2 meters to 4.37 meters when considering the vegetation-inclusive DSM-based approach.

These findings underscore the crucial role of vegetation in shaping flood pathways within urban environments, emphasizing the need to consider both natural and human-introduced elements in flood risk management strategies. Future research directions could explore the evolving impact on populations across varied flood zones and conduct comprehensive cost evaluations regarding risk mitigation, recovery efforts, and infrastructure fortification. These avenues present promising trajectories for further studies, offering insights into the socio-economic and financial implications of diverse flooding patterns in urban settings.

How to cite: Nieto Butrón, J. J., Ramírez Serrato, N. L., Barco Coyote, S., Yépez Rincon, F. D., and Jácome Paz, M. P.: Impact of vegetation on urban open-channel flow: Practical experiment with 2D IBER hydraulic simulations in Monterrey, Mexico., EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-11106,, 2024.

EGU24-11521 | ECS | Orals | NH1.3

Assessment of sewer network data quality on urban pluvial flood modeling with a 2D/1D dual drainage model 

Carlos Montalvo, Paolo Tamagnone, and Luis Cea

Urban pluvial floods are one of the most common water-related hazards and are going to become more frequent and severe looking at the upsetting climate projections. These events mainly occur due to intense and short precipitation events, leading to the overload of the sewer network and resulting in physical, economic, and even human losses. To address this hazard, effective methods are needed to estimate the scale and impact of pluvial flood events and to develop mitigation strategies. In this context, 2D/1D dual drainage models have become one of the most useful tools for these purposes, being able to simulate all hydraulic phenomena occurring on and beneath the surface. However, these models require detailed information about the topography and geometrical specifications of the sewer network, which are not always readily accessible or, when available, are often incomplete or of poor quality, particularly in large urban environments.

In this work, considering that pluvial flood studies are becoming more popular and several numerical tools are available, we wanted to address a recurrent question raised by the flood modeler community: is the effort/level of complexity of implementing a detailed dual drainage model worth it? To answer this question, we assess the influence of sewer network data quality on the results of water depth and velocity obtained with a 2D/1D dual drainage model applied to urban flood modelling. For this purpose, an ad-hoc 2D/1D hydraulic model was implemented to simulate the complex network system of the city of Differdange (LU) exploiting the recently developed Iber-SWMM. This city was chosen as study case because it has experienced several flooding events in recent years, such as those recorded in 2021, and it has an extensive dataset of detailed geospatial data available, enabling the setup of a high-resolution resolution and fully coupled 2D/1D dual drainage model.

Sewer network links were classified based on their physical properties, such as diameter and length. The sewer network layout was gradually simplified, starting from the minor links to the more complex segments of the network, obtaining new simplified versions of the network that could represent incomplete or poor-quality scenarios. These simplified versions were successively implemented in the 2D/1D model. The comparison between the results of the complete and comprehensive model and the simplified scenarios reveals the impacts of the quality of the sewer network information on pluvial flood modeling.

How to cite: Montalvo, C., Tamagnone, P., and Cea, L.: Assessment of sewer network data quality on urban pluvial flood modeling with a 2D/1D dual drainage model, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-11521,, 2024.

EGU24-11904 | ECS | Posters on site | NH1.3

The role of predisposing factors in determining the rainfall intensity necessary to cause flash floods in Portugal 

Caio Vidaurre Nassif Villaça, José Luís Zêzere, and Pedro Pinto Santos

Flash floods are often responsible for deaths and damage to infrastructure. The general objective of this work is to create data-based models to understand how the predisposing factors influence the triggering factor (precipitation) in the case of flash floods in the continental area of Portugal. Flash floods occurrences were extracted from the DISASTER database, which contains the location and date of historical flood events in the study region. Historical daily rainfall data was collected automatically from the Copernicus database. We extracted the accumulated precipitation for 3 days preceding each event and calculated the rainfall intensity. The predisposing factors were extracted considering the whole basin that corresponds to each flood event. The  analyzed predisposing factors were: accumulated flow, average slope, average elevation, predominant slope aspect, predominant lithology and soil properties (percentage of clay, coarse sand and coarse elements and field capacity). Elevation can often define different climatic and vegetation zones, while slope influences both the concentration and the infiltration capacity. The slope aspect can influence the amount and intensity of rainfall that affects the hillslope, as well as the amount and intensity of solar radiation. Lithology represents the properties of bedrock and the soils properties influence water infiltration and percolation. The Random Forest algorithm and the Leave-One-Out cross-validation technique were used to evaluate the model's performance and create a final model that identifies the relationship between the predisposing factors and the different rainfall intensities related to each flash flood occurrence. The final model obtained a root mean square error (RMSE) value of 3, an acceptable value for the objectives of the work. The percentage of coarse elements in the soil, average slope and field capacity were defined as the most important factors in the model for defining the amount of rainfall needed for flash floods to occur in mainland Portugal. The model developed can help to predict flash flood occurrence and future work involves combining the susceptibility model with the model created in this project to create a warning system that can be updated in real time, taking into account rainfall forecasts.

Acknowledgements: This work was financed by national funds through the FCT – Fundação Portuguesa para a Ciência e Tecnologia, I.P., under the grant to support the completion of the doctoral dissertation with the reference 2022.14473.BD.

How to cite: Vidaurre Nassif Villaça, C., Luís Zêzere, J., and Pinto Santos, P.: The role of predisposing factors in determining the rainfall intensity necessary to cause flash floods in Portugal, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-11904,, 2024.

EGU24-12195 | ECS | Orals | NH1.3

Modeling flood fatalities in the Italian context: an empirical approach 

Mina Yazdani, Christian N. Gencarelli, Paola Salvati, and Daniela Molinari

Floods are among the most frequent and damaging natural hazards, affecting millions of people worldwide, and the risk of catastrophic losses due to flooding is expected to increase as a result of climate change. The possibility of predicting and estimating the expected fatalities in flood-prone regions is among the top priorities of decision-makers in flood risk management. Thus, predicting the conditions leading to loss of life is crucial for assessing the risk to the population. Here we focus on the Po River District in Northern Italy which covers the largest Italian hydrographic basin. We demonstrate that the occurrence of flood-related fatalities can be estimated by utilizing a random forest (RF) algorithm applied to a dataset of fatalities that occurred in this area from 1970 to 2019. This method relies on nine explanatory variables that describe the hazard intensity, and the environmental and sociodemographic conditions leading to fatalities. The proposed model is a primary attempt to estimate the probability of flood-related fatalities in the Italian context, and it provides a proxy for the quantitative estimation of flood risk to the population.

How to cite: Yazdani, M., N. Gencarelli, C., Salvati, P., and Molinari, D.: Modeling flood fatalities in the Italian context: an empirical approach, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-12195,, 2024.

EGU24-12935 | ECS | Orals | NH1.3

Estimating flood exceedance probabilities for several levee breach scenarios for an urban riverine environment in Toronto, Canada 

Florence Mainguenaud, Usman Khan, Laurent Peyras, Claudio Carvajal, Jitendra Sharma, and Bruno Beullac

Assessing flood risk requires the combination of flood hazard, exposure, and vulnerability.  Hence, flood hazard is a key component of flood risk assessments. As flood propagation is impacted by hydraulic structures built along the river, flood defense such as levees have gained attention as they are rarely included in large scaled flood risk assessments. However, flood events such as hurricane Katrina showcased the impact that levee failure has on flow depth, velocity, and flood extent. Therefore, its consideration should be regularly implemented in flood risk assessments. However, with current flood risk assessment methods, considering different levee failure scenarios results in numerous flood scenarios, simulations, and hazard maps. The multiplication of simulations and maps increases the complexity of flood risk management. We propose to improve flood hazard assessments by considering a single probabilistic flood map accounting for several flood events and levee breaching scenarios. For flood events enabling the performance assessment of the levee (i.e. levee breaching), we assessed levee failure probabilities, associating each levee segment to a fragility curve. Then, we defined breaching and non-breaching scenarios and ran flood simulations using HEC-RAS and its integrated parametric levee breaching model. We propose a new method to compute flood scenario probabilities and flood exceedance probabilities. The cumulative flood exceedance probability provides a curve for every location of the flooded area. Using GIS, we applied this method to the entire flooded area, resulting in an interactive flood hazard map. An application to the Etobicoke Creek located in the Greater Toronto Area showed that this new approach provides an operational levee breaching flood hazard method that can be used in integrated flood risk assessments.

How to cite: Mainguenaud, F., Khan, U., Peyras, L., Carvajal, C., Sharma, J., and Beullac, B.: Estimating flood exceedance probabilities for several levee breach scenarios for an urban riverine environment in Toronto, Canada, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-12935,, 2024.

EGU24-15306 | ECS | Posters on site | NH1.3

Numerical modelling of flood hazard mitigation strategies: the case of the Baganza River after the 2014 inundation of the city of Parma 

Alessia Ferrari, Renato Vacondio, and Paolo Mignosa

Urban flood risk mitigation is a paramount priority given the increasing frequency of flood events, which have become predominant natural disasters in recent decades. Over half of the global populace now resides in urbanized areas, amplifying the vulnerability to such events that is further accentuated by climatic shifts and rapid urban sprawl. In addressing these challenges, sophisticated flood risk management strategies often integrate advanced numerical models for precise hydrological assessments. These models can support e.g. urban planning, emergency response preparedness and the design of structural measures. In the present work, the Baganza River in the city of Parma (Northern Italy) is investigated with particular emphasis on recent modifications that have been designed with outcomes deriving from a computationally efficient parallel 2D numerical model solving the Shallow Water Equations (SWEs).

On October 2014, a severe flood event occurring on the Baganza River caused the inundation of the southwestern part of the city of Parma. Since the urban river reach showed limitations in the propagation of the flood wave, a comprehensive re-evaluation of the river's hydraulic conveyance capacities was required. Thus, in 2015, hydraulic authorities started designing and realizing several modifications along this river reach, including levee modification and removal, in order to increase its conveyance. With the aim of assessing the effectiveness of these strategies, the PARFLOOD numerical model, which solves the 2D-SWEs on a finite volume scheme and ensures high computational efficiency due to its parallel implementation on GPU, was adopted. The model was initially calibrated and adopted to simulate the 2014 flood event. Thereafter, leveraging a refined spatial resolution and incorporating detailed urban topographies, the model delineated residual flood hazard maps, facilitating evidence-based mitigation strategy refinements.

Once the most promising strategies were outlined and implemented over these last ten years, a new high-resolution Digital Terrain Model (DTM) deriving from a LiDAR survey was provided in 2023. By simulating the same synthetic discharge hydrograph, e.g. with a return period of 100 and 200 years, using both the 2014 DTM and the 2023 one, it clearly emerged that the current asset strongly reduces the residual flood hazard in these districts of the city of Parma, both in terms of flood extent and magnitude.

How to cite: Ferrari, A., Vacondio, R., and Mignosa, P.: Numerical modelling of flood hazard mitigation strategies: the case of the Baganza River after the 2014 inundation of the city of Parma, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-15306,, 2024.

With the worsening of climate change, extreme weather events are on the rise, leading to more frequent occurrences of climate-related disasters. Analyzing people's perceptions and attitudes towards disasters after they occur can help determine the spatial pattern of the disaster intensity and the post-disaster needs of different populations. The implication is to provide references for disaster assessment and post-disaster relief needs analysis.


Starting from July 29, 2023, due to the influence of Typhoon Dusrayi and Typhoon Canu, the Beijing-Tianjin-Hebei region in China suffered from catastrophic rainfall, resulting in severe flooding in multiple areas. This study utilized web crawlers to collect relevant Weibo data during the disaster, applied machine learning models to conduct public opinion analysis on the flooding disaster, developed the evolutionary patterns of public opinions on the disaster, and obtained heat maps and sentiment indicators for different cities. The results will contribute to the rapid assessment of post-disaster losses and guide the resource allocation in the initial emergency rescue process after the disaster.

How to cite: Zhang, X.:  A rapid disaster intensity assessment method using social media data: a case study of the flood disaster in the Beijing-Tianjin-Hebei region, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-15415,, 2024.

EGU24-15605 | ECS | Orals | NH1.3

Establishing a Germany-wide Standardized Indication Map Representing the Flood Situation Caused by Heavy Rainfall 

Lukas Wimmer, Michael Hovenbitzer, and Patrick Merita

Recent studies on climate change show an increasing trend in the frequency of extreme weather events (IPCC, 2021; Tradowsky et al., 2023). These include storms with high-intensity precipitation, known as heavy rainfall. During those events the amounts of precipitation can be so high in a very short period of time that catastrophic flooding can also develop far away from rivers and lakes. Heavy rainfall events have occurred more frequently in Germany in recent years, resulting in severe damage and therefore focusing attention on risk management and prevention.

Contributing to an optimal preparation for the consequences of heavy rainfall events the Federal Agency for Cartography and Geodesy (BKG) is working with federal and state authorities to develop a Germany-wide indication map representing simulated flood situations after heavy rainfall events based on standardized guidelines. Once the mapping has been completed within the first half of 2024, it will be freely available as OpenData to politicians, the public administration and the general public for damage prevention and civil protection.

Geodata of the federal and state governments are essential for the hydronumerical two-dimensional modelling. A digital terrain model with a grid width of one meter forms the basis. Road culverts with corresponding dimensions, 3D building models, pumping stations as well as land cover data representing the surface roughness are integrated into this model in order to achieve a hydrologically effective modification and thus a realistic discharge.

The heavy rainfall indication map shows realistic simulation events for possible flooding scenarios that follow the heavy rainfall index according to Schmitt et al., 2018. The index describes the hazardous character of heavy rainfall events based on the return period and is commonly used in heavy rainfall risk communication by German federal and state authorities. Two scenarios are simulated: First, a 100-year event based on KOSTRA data from the German Weather Service (DWD), a dataset including regionalized precipitation heights as a function of precipitation duration and annularity. Second, an extreme heavy rainfall event with a precipitation of 100 mm/h. For both scenarios flood depths, flow velocities and flow directions are simulated.

The indication map for heavy rainfall provides an initial assessment of the risk potential, which, in combination with existing local expertise, should considerably simplify the planning of measures. It serves as an important tool for identifying areas at risk from heavy rainfall. This enables local authorities, planners and emergency services throughout Germany to derive appropriate measures, both preventively and in the event of an actual disaster.



Intergovernmental Panel on Climate Change (Ed.): Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change (IPCC). Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, <doi:10.1017/9781009157896>.

Schmitt, T. G. et al.: Einheitliches Konzept zur Bewertung von Starkregenereignissen mittels Starkregenindex, KA Korrespondenz Abwasser, Abfall, 2018(65), Nr. 2.

Tradowsky, J.S., Philip, S.Y., Kreienkamp, F. et al.: Attribution of the heavy rainfall events leading to severe flooding in Western Europe during July 2021. Climatic Change 176, 90 (2023). <>.

How to cite: Wimmer, L., Hovenbitzer, M., and Merita, P.: Establishing a Germany-wide Standardized Indication Map Representing the Flood Situation Caused by Heavy Rainfall, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-15605,, 2024.

The subject of the study is the considerable uncertainty in determining flood risk when long-term climate change projections are developed. Risk management decision-making involves comparing options based on their benefits and costs. The purpose of the analysis is to reveal the uncertainty robustness of alternative flood protection measures. The treatment of different sources of uncertainty is done by using probabilistic net present value (NPV) analysis as well as by using Information-gap decision theory (IGDT). The case study is a settlement in northern Bulgaria with a record of severe flooding in the past, for which different climate change projections are generated under RCP 4.5 and RCP 8.5 scenarios. The behaviour of three civil protection options under these uncertainty conditions is investigated for an extended 30-year time period to 2050. A probabilistic analysis with NPV performance criterion is performed sequentially, followed by Info-gap decision theory analysis.

After discussing the results, the advantages and disadvantages of the two methods are compared. Some limitations and advantages of the Information gap theory are discussed. Finally, it is highlighted that when making decisions about long-term flood protection, it is recommended to use multiple methods that differ in data and assumptions, necessarily taking into account the hydrological uncertainty arising from climate change, which can radically change our choices.

How to cite: Mavrova-Guirguinova, M.: Impact of Uncertainty on the Choice of Long-Term Flood Protection Option under Climate Change, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-16798,, 2024.

EGU24-17663 | ECS | Posters on site | NH1.3

Flood Risk Mapping Through Advanced Machine Learning Techniques and Geomorphic Data Integration 

Jorge Saavedra Navarro, Ruodan Zhuang, Cinzia Albertini, and Salvatore Manfreda

Flood events rank among the most destructive natural hazards, necessitating comprehensive risk management strategies to mitigate their impact on human health, the environment, cultural heritage, and economic activities. In this context, various approaches have been developed for identifying flood-prone areas, but there is still a need to enhance their capabilities due to dynamic changes in landscape and infrastructure.

In recent years, there has been a proliferation of remote sensing observations that can support dynamic and continuous mapping of flood-prone areas by integrating the most updated information. This study explores the potential of machine learning (ML) techniques, including Random Forest, Support Vector Machine, and Navies Bayer model, utilizing geomorphic information such as slope, elevation, precipitation, land use/land cover, elevation difference to the nearest river, and others as predictor variables. The best model and set of variables were explored by adopting approximately 30 variables spanning types, hydrologic, topographic, and categorical categories. Careful consideration was given to avoiding high correlations between variables in test subsets, ensuring relevance, and avoiding redundancy. Calibration and validation of the model employ Copernicus Emergency Management Service maps from Sentinel-2 satellite coupled with regional maps of past flood events.

Results highlight that the best ML technique is represented by the Random Forest, adopting a range of 5 to 8 variables for effective delineation of flood-prone areas. Among the selected variables, the most relevant ones include Rainfall, Geomorphic Flood index - GFI, Lithology, and others. The study demonstrates that a minimal amount of information (between 0.1% and 10%) suffices for optimal model performance (AUC greater than 0.8).

The study covered the entire territory of Italy, resulting in a flood-prone map at a 90m resolution, validated with flood maps provided by national agencies and obtained through traditional hydraulic models.

Keywords: satellite images, flood-prone areas, Machine Learning, GFI, flood risk.

How to cite: Saavedra Navarro, J., Zhuang, R., Albertini, C., and Manfreda, S.: Flood Risk Mapping Through Advanced Machine Learning Techniques and Geomorphic Data Integration, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-17663,, 2024.

EGU24-18219 | ECS | Posters on site | NH1.3

Meta-heuristic Algorithms Applied to Urban Flood Evacuation Routes: A case study in Beijing, China 

Chuannan Li, Changbo Jiang, Reza Ahmadian, Man Yue Lam, and Jie Chen

Abstract: Climate change and urbanization have increased the occurrence of natural disasters, including floods, tsunamis and hurricanes. Among these disasters, floods occur with high frequency, impact a large number of people, cause high economic losses, and lead to high toll of deaths. Examples include floods in Indonesia in 2021 and Pakistan in 2022. The flood in Indonesia affected about 1 million people. The flood in Pakistan affected 33 million and killed 1,739 people and costed US$15 billion in economic damage. Flood risk assessment and evacuation are effective mitigation measures to create flood-resilient cities. Previous studies have focused on flood modelling and risk assessment, yet it is recently recognized that optimal evacuation routes are necessary and critical for social adaptation to flood risks. To date, there are limited research on evacuation route optimisation problem.

There are two approaches for evacuation route optimisation: namely exact methods and meta-heuristic methods. The exact methods such as linear programming, weighted summation, and mixed integer programming have been widely applied. Nevertheless, meta-heuristic algorithms are gaining attention as flexible, non-problem-specific, and computationally efficient optimisation methods. The principle of Meta-heuristic algorithms is based on simulating the optimisations that occur naturally in biological or physicochemical processes. For example, there is a commonality between an animal herd searching for routes and a population searching for routes in a flood disaster. Commonly applied meta-heuristic algorithms are Genetic Algorithms, Ant Colony Algorithms, Particle Swarm Algorithms, and Sparrow’s Algorithms. This is because these algorithms have simple structures and high adaptability, desirable local and global convergence properties and require few parameters.

In this study, the flood in Beijing, China, in late July and early August 2023 will be simulated. The flood claimed at least 33 lives, damaged 209,000 homes and more than 15,000 hectares of cropland and caused 127 thousand people to evacuate. The flood extent, water depth and flow velocity will be obtained from a two-dimensional hydrodynamic flood model. The flood risk for pedestrians or vehicles will be estimated with the hydrodynamic model result and a mechanic-based stability method. Optimal evacuation routes will be obtained with Genetic Algorithm, Ant Colony Algorithm, Particle Swarm Algorithm, and Sparrow’s Algorithm. The performance of the optimisation algorithms will be compared and evaluated.  This study contributes to the scientific planning of urban flood evacuation routes and provides insight for urban planners and managers to enhance urban resilience.

Keywords: Urban floods, evacuation routes, Meta-heuristic algorithms.

How to cite: Li, C., Jiang, C., Ahmadian, R., Lam, M. Y., and Chen, J.: Meta-heuristic Algorithms Applied to Urban Flood Evacuation Routes: A case study in Beijing, China, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-18219,, 2024.

EGU24-18287 | ECS | Orals | NH1.3

Sensitivity Analysis of Image Augmentation Methods to Improve Flooded Area Detection Performance 

Seon Woo Kim, Soon Ho Kwon, Sanghoon Jun, and Donghwi Jung

Recently, detecting flooded areas in CCTV images was performed based on semantic segmentation models (e.g., U-Net, FCN, etc.). However, these flooded area detection techniques are based on large-scale manually annotated images, which consume manpower and time. Image augmentation is one of the ways to overcome the limitations mentioned above. Some previous studies have used image augmentation to improve the performance of flooded area detection by combining two or more methods. However, there has been no study quantifying which augmentation methods are reasonable. This study aims to verify which image augmentation method is reasonable to improve the performance of urban flooded area detection techniques. First, this study develops a flood area detection technology corresponding to training images augmented with five different methods (Brightness, Blur, Contrast, Rotation, Crop). Subsequently, the performance changes for each technique were quantified, and characteristics related to the performance variations of each method were examined.

How to cite: Kim, S. W., Kwon, S. H., Jun, S., and Jung, D.: Sensitivity Analysis of Image Augmentation Methods to Improve Flooded Area Detection Performance, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-18287,, 2024.

EGU24-22491 | ECS | Posters virtual | NH1.3

Application of Advanced Deep Learning Models for Flood Image Processing and Semantic Segmentation 

Sai Praneeth Dulam, Vidya Samadi, and Carlos Toxtli-Hernández

In developing Version 2.0 of our Flood Image Classifier, we underscore the significant role of Convolutional Neural Networks (CNNs), mainly Faster R-CNN and YOLOv3, in detecting and segmenting flood-related labels in images. Additionally, our research delves into the potential of Vision Transformers (ViT) for advanced object detection and image classification for flood-related images extracted for the USGS river cameras. Transformer methods offer improved predictions of flood depth and inundation areas, marking a substantial step forward in flood vision technology. The integration of advanced image processing techniques, the enhancement of CNN capabilities, and the incorporation of cutting-edge detection and classification models are pivotal in developing a comprehensive, real-time flood monitoring system. This system is designed to equip frontline decision-makers and emergency responders with essential insights into flooding conditions, thereby significantly contributing to disaster management and response through the innovative use of our flood image classifier, Version 2.0.

How to cite: Dulam, S. P., Samadi, V., and Toxtli-Hernández, C.: Application of Advanced Deep Learning Models for Flood Image Processing and Semantic Segmentation, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-22491,, 2024.

EGU24-529 | ECS | Posters on site | NH1.5

The Attachment Process of Negative Connecting Leader to the Lateral Surface of Downward Positive Leader in a +CG Lightning Flash 

Qi Qi, Bin Wu, Weitao Lyu, Ying Ma, Lyuwen Chen, Fanchao Lyu, and Yan Gao

In the lightning attachment process, the leader connecting behavior is an interesting topic. In the attachment process of a negative cloud-to-ground lightning flash, the “Tip to the lateral surface” connection type has been widely observed, and researchers have carried out a series of studies and discussions on the characteristics and the physical mechanisms of the leader connecting behavior. However, is there also a “Tip to the lateral surface” connecting behavior in the attachment process of the positive cloud-to-ground lightning flash? In this study, using high-speed video cameras operating with framing rates of 20 and 50 kiloframes per second, we captured an attachment process during a positive cloud-to-ground flash, which demonstrates the connection of the negative connecting leader (NCL) to the lateral surface of the downward positive leader (DPL) for the first time. When the NCL was initiated, the tip of the DPL had passed the initiation position of the NCL for about 50 m. A common streamer zone (CSZ) was observed when the three-dimensional distance between the NCL tip and the lateral surface of DPL was about 30 m. It is remarkable to note that a luminous segment (space stem/leader) with a length of about 7 m was captured within the CSZ during the attachment process. The connection between the NCL tip and the lateral surface of the DPL was caused by the development of the CSZ and its inner space leader.

How to cite: Qi, Q., Wu, B., Lyu, W., Ma, Y., Chen, L., Lyu, F., and Gao, Y.: The Attachment Process of Negative Connecting Leader to the Lateral Surface of Downward Positive Leader in a +CG Lightning Flash, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-529,, 2024.

EGU24-1238 | ECS | Posters on site | NH1.5

Optical observations of needles evolving into negative leaders in a positive cloud-to-ground lightning flash 

Bin Wu, Qi Qi, Weitao Lyu, Ying Ma, Lyuwen Chen, and Vladimir Rakov

High-speed video records of a single-stroke positive cloud-to-ground (+CG) flash were used to examine the evolution of eight needles developing more or less radially from the +CG channel. All these eight needles occurred during the later return-stroke stage and the following continuing current stage. Six needles, after their initial extension from the lateral surface of the parent channel core, elongated via bidirectional recoil events, which are responsible for flickering, and two of them evolved into negative stepped leaders. For the latter two, the mean extension speed decreased from 5.3 × 10^6 to 3.4 × 10^5 and then to 1.3 × 10^5 m/s during the initial, recoil-event, and stepping stages, respectively. The initial needle extension ranged from 70 to 320 m (N = 8), extension via recoil events from 50 to 210 m (N = 6), and extension via stepping from 810 to 1,870 m (N = 2). Compared with needles developing from leader channels, the different behavior of needle flickering, the longer length, the faster extension speed, and the higher flickering rate observed in this work may be attributed to a considerably higher current (rate of charge supply) during the return-stroke and early continuing-current stages of +CG flashes.

How to cite: Wu, B., Qi, Q., Lyu, W., Ma, Y., Chen, L., and Rakov, V.: Optical observations of needles evolving into negative leaders in a positive cloud-to-ground lightning flash, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-1238,, 2024.

EGU24-1639 | Orals | NH1.5

Modelling the collision of streamers using the AMReX framework 

Christoph Köhn, Angel Ricardo Jara, Morten Jung Westermann, Mathias Gammelmark, and Elloise Fangel-Lloyd

Streamers, precursors of the hot, long lightning leaders, are small filamentary discharges with high electric fields at their tips. Experiments of laboratory discharges have shown that streamers in their corona can approach each other and it has been suggested that such collisions enhance the electric field in-between beyond the thermal runaway electric field accelerating electrons to the runaway regime thus generating X-rays. Streamer collision also plays a role in the interaction of wind turbine blades with lightning when streamers locally incept from the surface of blades and attract the downward moving lightning leader. Despite the relevance of streamer collisions in the runaway process or their role in the interaction of lightning with wind turbine blades, there have only been a few numerical studies due to computational limitations. We have therefore developed a novel 3D fluid model for streamer propagation implemented in the AMREX framework. AMREX allows us to solve drift-diffusion and Poisson equation using parallelization and GPU support to accelerate the block structured adaptive mesh refinement. We will present details of the implementation as well as a parameter study on typical streamer parameters (electron density, electric field, tip width and velocity,…) during streamer collision in various ambient fields and for various initial electron densities. We will also study various geometries with different displacements of the initial electrons perpendicular to the ambient electric field. Finally, we will interpret our results with respect to the runaway process and wind turbine-lightning interaction.

How to cite: Köhn, C., Jara, A. R., Westermann, M. J., Gammelmark, M., and Fangel-Lloyd, E.: Modelling the collision of streamers using the AMReX framework, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-1639,, 2024.

The ground-level potential gradient (PG) or the atmospheric electric field, the air-Earth current density as well as the main Global Electric Circuit (GEC) parameters such as the ionospheric potential, global resistance and the total current, can be obtained from the EGATEC engineering model of the GEC (Odzimek et al. 2010) at the resolution of 3 hours. The model input data based on satellite cloud and lightning observation datasets from the period 1998-2006 for evaluating the activity of the GEC cloud generators, and the summer/winter and low/high solar activity conductivity model of Tinsley and Zhou (2006) allow calculating the GEC parameters in the summers and winters of the period. In this work we compare the modelling results to observations from the Stanislaw Kalinowski Geophysical Observatory in Świder, Poland (52°07' N, 21°14' E) of the ground-level potential gradient and conduction current density calculated from the newly digitised PG and positive conductivity data from 1965-2005. We also look for connections in the time variations of the model meteorological input and atmospheric electricity observational data. The work is supported by the Polish National Science Centre grant no 2021/41/B/ST10/04448.

How to cite: Odzimek, A., Tacza, J., Pawlak, I., and Kępski, D.: Analysis of time variations in the Global Electric Circuit parameters from the EGATEC model and Świder atmospheric electricity data, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-1658,, 2024.

The different morphologies of lightning channels are caused by different electrical environments within the cloud, the charge distribution determines the lightning channel morphology, and the lightning morphology can reflect the charge structure to some extent. The distribution of charges is mainly determined by the dynamics and microphysical conditions in clouds, and turbulence plays a significant role in the distribution of charges. Due to the dependence of lightning morphology on the distribution of thunderstorm charges, which is regulated by thunderstorm dynamic effects, a relationship can be established between lightning morphology and thunderstorm dynamic effects.

In this study, the lightning channel was obtained from three-dimensional radiation source localization data from the Lightning Mapping Array at the Langmuir Laboratory of the New Mexico Institute of Mining and Technology. The fractal dimension was used to characterize the complexity of lightning channels, which was calculated by the box-counting method. The S-band dual-polarization Doppler radar data was used to estimate the cube root of the eddy dissipation rate (EDR, the EDR was estimated using the Python Turbulence Detection Algorithm). The EDR and radar radial velocity were used to represent the thunderstorm dynamic characteristics.

Superimposing EDR and radar radial velocities with LMA radiation sources, our analysis shows that the overall morphology and detailed morphology of the lightning channel correspond to different EDR characteristics. Lightning with complex channel morphology has a larger average FD and occurs in regions with large EDRs. In single lightning events, channels that extend directly within a certain height range without significant bifurcation and turning tend to propagate in the direction of decreasing EDRs, while channel bifurcations and turns usually occur in regions with large radial velocity gradients and large EDRs. This study shows the relationship between channel morphology and thunderstorm dynamics and provides a new method for the direct application of channel-level localization data to understand thunderstorm dynamics characteristics.

How to cite: Li, Y., Zhang, Y., Zhang, Y., and Krehbiel, P. R.: Analysis of the Relationship between the Morphological Characteristics of Lightning Channels and Turbulent Dynamics Based on the Localization of VHF Radiation Sources, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-2144,, 2024.

EGU24-2536 | Posters on site | NH1.5

Lightning Activities near the Red Sea: Effects of Aerosols Morphology and Local Meteorology 

Ashraf Farahat and Maher Dayeh

Lightning activity is one of the global natural hazards that pose significant risks to human life and numerous aspects of society's technological infrastructure. Understanding the linkage between aerosols present in the atmosphere and lightning activity is important to further advance our knowledge of the global lightning activity cycle.

Saudi Arabia and Yemen host one of the world’s largest desert areas namely the Empty Quarter (al-Rubea Al-Khali). Moreover, Saudi Arabia is one of the world’s largest oil exporters with many water desalination, petrochemical, and cement industrial plants, while large cities in both Saudi Arabia and Yemen have large construction projects and vehicle emissions. This increases both natural and anthropogenic aerosol loading in both countries.  Meanwhile, the inland regions close to the Red Sea are one of the 500 hottest lightning regions in the world. This work identifies a possible correlation between lightning activity and aerosol loading.

Using data of individual lightning strokes from the Global Lightning Detection Network (GLD360), in conjunction with remote sensing measurements of the aerosol optical depth (AOD) obtained at 500 nm from the Moderate Resolution Imaging Spectroradiometer (MODIS) instrument onboard the Terra and Aqua satellites during active lightning days, we examine the evolution of lightning activity in two geographically and topologically different regions over Saudi Arabia and Yemen. One region extends inland to the desert (R1) and the other is in the southwest mountainous region that is close to the Red Sea (R2). In both regions, results from thunder days only indicate that lightning is strongly and positively correlated with the AOD loading, up to AOD ~ 0.8, after which the trend flattens or reverses direction. Results suggest the two opposite effects that aerosols could indirectly have on lightning activity are at play. The mountainous region exhibits a much stronger linear relation compared to the inland region. Furthermore, both regions exhibit seasonal and asynchronous lightning activity and AOD loading. The year 2018 in R1 shows very high lightning activity, likely linked to the 2018 intense dust storms in the region.

How to cite: Farahat, A. and Dayeh, M.: Lightning Activities near the Red Sea: Effects of Aerosols Morphology and Local Meteorology, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-2536,, 2024.

EGU24-3358 | Orals | NH1.5

Investigation of the Electric Fields Related to Elves Simulations 

Petr Kaspar, Ivana Kolmasova, Ondrej Santolik, and Martin Popek

Elves are transient luminous events occurring above thunderclouds. They appear as an expanding ring of light at altitudes of 85 – 95 km with diameters of more than 200 km and lasting less than 1 ms. The elves are produced by electromagnetic pulses emitted by underlying high-peak current lightning discharges, which excite nitrogen molecules at the bottom of the ionosphere. We develop an electromagnetic model of elves, which consists of two steps. As the first step, we compute the horizontal part of the electric field at a height of 15 km from transmission line return stroke (RS) models without damping, with linear, and/or exponential damping of the current wave. Subsequently, we solve Maxwell’s equations self consistently for altitudes from 15 km to 95 km, including finite neutral and electron densities, and nonlinearities related to heating, ionization, and attachment of free electrons caused by the RS transient electric field. We show computed electric fields and optical emission rates at the heights of the development of elves. This procedure allows us to distinguish between the electrostatic, induction, and radiation part of the electric field and to investigate their role in the evolution of elves in the full wave simulations.

How to cite: Kaspar, P., Kolmasova, I., Santolik, O., and Popek, M.: Investigation of the Electric Fields Related to Elves Simulations, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-3358,, 2024.

EGU24-3628 | Orals | NH1.5

Glow-terminating terrestrial gamma-ray flashes observed during the ALOFT Campaign 

Steven Cummer, Yunjiao Pu, Andrew Mezentsev, Marni Pazos, Morris Cohen, Nikolai Ostgaard, Mark Stanley, Timothy Lang, Martino Marisaldi, J. Eric Grove, Mason Quick, Hugh Christian, Christopher Schultz, Richard Blakeslee, Ian Adams, Phillip Bitzer, Martin Fullekrug, Bilal Qureshi, Bendik Husa, and Gerald Heymsfield and the additional members of ALOFT team

The ALOFT campaign targeted aircraft measurements of terrestrial gamma-ray flashes (TGFs) through NASA ER-2 overflights of strong thunderstorms.  We report here the analysis of glow-terminating TGFs (GT-TGFs) that occur at the end of some gamma-ray glows.  GT-TGFs were generated by most of the observed storms during the campaign and were prolifically generated by two specific storms that were particularly active in gamma ray production.  One unique feature of GT-TGFs is that they always occur within several tens of microseconds of a narrow bipolar event (NBE).  The characteristics of GT-TGFs and the associated NBE radio emissions will be described in detail.

How to cite: Cummer, S., Pu, Y., Mezentsev, A., Pazos, M., Cohen, M., Ostgaard, N., Stanley, M., Lang, T., Marisaldi, M., Grove, J. E., Quick, M., Christian, H., Schultz, C., Blakeslee, R., Adams, I., Bitzer, P., Fullekrug, M., Qureshi, B., Husa, B., and Heymsfield, G. and the additional members of ALOFT team: Glow-terminating terrestrial gamma-ray flashes observed during the ALOFT Campaign, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-3628,, 2024.

EGU24-3670 | ECS | Orals | NH1.5

Using meteorological reanalysis to identify weather conditions for classifying atmospheric electricity data  

Hripsime Mkrtchyan, Giles Harrison, and Keri Nicoll

Atmospheric electricity Potential Gradient (PG) data has typically been classified by local weather conditions, such as by identifying data recorded during “fair weather” (FW) or in the absence of rainfall “no hydrometeors” (NH), to try and obtain globally representative values. In general, this approach is essential in obtaining global atmospheric circuit (GEC) signals. The weather information needed to do this is, however, only available from some of the sites providing atmospheric electricity measurements. For other sites, meteorological reanalysis – of which there are many products available, spanning different times and scales - may provide a data source for such classification of PG data. This study investigates the integration of ERA5 meteorological reanalysis data to identify FW and NH conditions and improves the quality of data used in long-term atmospheric electricity studies.  

Initial findings investigating the meteorological quantities show a strong correlation between wind speed, total cloud coverage and total precipitation from ERA5 and observed ground-based measurements at the Eskdalemuir and Lerwick sites. This is to be applied to classifying past atmospheric electricity data, specifically of the hourly potential gradient (PG), which were obtained at the Lerwick observatory from 1925 to 1984, and Eskdalemuir observatory, which made atmospheric electricity measurements from 1911-1981 (Harrison & Riddick, 2022; Märcz & Harrison, 2003). 

Identified criteria from ERA5 which best match for FW and NH conditions are implemented in historical data from the Lerwick and Eskdalemuir observatories, enhancing the reliability of past studies which is important for atmospheric electricity analyses. This supports the potential of ERA5 data for providing information to identify FW and NH conditions. From this, we are evaluating a range of methods to use the meteorological reanalysis, with the aim of recovering representative FW data at sites lacking meteorological measurements. 

How to cite: Mkrtchyan, H., Harrison, G., and Nicoll, K.: Using meteorological reanalysis to identify weather conditions for classifying atmospheric electricity data , EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-3670,, 2024.

EGU24-4116 | Orals | NH1.5

Stream Machine Learning for Lightning Nowcasting - Harnessing the Power of Continuously Updated Data 

Cesar Beneti, Luis Pavam, Luiz Oliveira, Marco Alves, Leonardo Calvetti, and Fernanda Verdelho

Uninterrupted access to electricity is a fundamental feature of civilization. In its absence, an all-embracing cessation of activities occurs, ranging from essential services to more frivolous activities. The maintenance of the energy supply is critical for society's day-to-day functions. The Brazilian state of Paraná (PR) is home to the world's second-largest hydropower plant, Itaipu, which, in conjunction with other power plants in the state, provides almost one-third of the power energy production in Brazil. The transmission lines that pervade PR are essential to Brazil's power distribution system, for hydropower generation is typically made far away from the regions that most demand it, being transported by transmission lines in an interconnected power grid. This type of asset mainly depends on the forecast of Cloud-to-Ground (CG) lightning, as it is one of the leading weather-related causes of power outages. Lightning and wind gusts are the two leading weather-related causes of disruptions, representing at least 23% of the known causes of energy disruption, as declared by the local power distribution company. Our study of lightning incidence and power outages from 2017-2021 indicates a correlation of 0.98 between these events, denoting that more outages must be lightning-related. Reliable CG lightning forecasts are crucial for proactive hazard mitigation. This work expounds on developing a Machine Learning (ML) model for CG lightning forecasting for PR. Our ML model predicts the occurrence or lack of CG lightning near power company assets in PR, defining a binary classification task. The model makes its predictions based on the past spatio-temporal conditions of lightning occurrences, requiring only past lightning data to forecast lightning. We chose to use a stream ML method, i.e., the model is continuously trained as new data arrives. Using a stream ML, we intend to harness the machine's capacity to continuously learn the patterns of lightning occurrence and power outages in real-time -- thus constructing an ever-updating model capable of adapting to transient weather conditions. Given its rapid training time and aptitude for classification tasks, the chosen algorithm was a Very Fast Decision Tree. The stream ML classifier outperforms a classic static ML model by 30% regarding the ROC AUC metric (stream: 71.80%, static: 40.85%) and 50% considering the Micro-f1 score (stream: 91.05%, static: 40.91%). These results arise from the highly dynamic nature of lightning, defining an ideal phenomenon for prediction based on a constantly updated stream of data. An automatic system for CG lightning forecasting for power company assets is helpful for risk management and operational planning. Future steps include increasing the lead time from ten min. to up to one hour, allowing for more time to prepare and anticipate hazards, preventing power outages, and optimizing personnel allocation.

How to cite: Beneti, C., Pavam, L., Oliveira, L., Alves, M., Calvetti, L., and Verdelho, F.: Stream Machine Learning for Lightning Nowcasting - Harnessing the Power of Continuously Updated Data, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-4116,, 2024.

EGU24-4214 | Posters on site | NH1.5

Bottom-heavy charge structure and lightning discharges in Tibetan Plateau thunderstorms 

Xiushu Qie, Zhuling Sun, Fengquan Li, Lei Wei, Chunfa Sun, Kexin Zhu, Shanfneg Yuan, Dongxia Liu, and Rubin Jiang

The main charge region in thunderstorms over Lhasa city with an elevation of 3700 m is investigated by using a VHF interferometer, incorporating with fast antenna, weather radar and cloud-to-ground lightning location. The evolution of charge structure and its effects on lightning discharges were discussed in a bottom-heavy thunderstorm. During the early developing stage, the thunderstorm exhibited an inverted dipolar charge structure with negative charge center over the positive, and lower negative intracloud (IC) lightning occurred in between. Then an upper positive charge region appeared as the convection intensifying, and the charge structure exhibited obvious tripolar pattern and with large lower positive charge center (LPCC), and fewer positive IC discharges occurred in the upper dipole but lower negative IC lightning still dominated. As the thunderstorm entered the later mature stage, both negative IC between the lower dipole and positive IC between the upper dipole observed simultaneously. With gradually depleting of the positive charge carriers by precipitation, the LPCC weakened, the positive IC lightning between the upper dipole dominated, and two negative CG flashes were able to occur. In the later stage, positive IC dominated, although not much.  The study further confirms the previous conclusion (Qie et al., GRL, 2005) that weak thunderstorms are characterized by a bottom-heavy charge structure, and in the vigorous stage of thunderstorm, it may exhibit tripolar charge structure with a large LPCC, which has a significant impact on lightning types.

How to cite: Qie, X., Sun, Z., Li, F., Wei, L., Sun, C., Zhu, K., Yuan, S., Liu, D., and Jiang, R.: Bottom-heavy charge structure and lightning discharges in Tibetan Plateau thunderstorms, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-4214,, 2024.

The evolution of charge structure plays a crucial role in thunderstorm electrification. In this paper, the signatures related to the upper charge regions consisting of charged ice crystals are analyzed in an isolated thunderstorm, observed by an X-band dual-polarized phased array weather radar (DP-PAWR), which operates in its normal operational mode which performs a volume scan with 110 elevations at a temporal resolution of 30 seconds. The radar data quality control is applied to polarized parameters of DP-PAWR, including the horizontal reflectivity ZH, differential propagation phase shift, and specific differential phase. The lightning data was obtained by a lightning detection system called LIDEN (LIghtning DEtection Network system) operated by the JMA. A flash group algorithm is employed to group lightning discharges into flash branches according to a spatial range, azimuth interval, and time criterion.


To explore the mean structure of upper charge regions in the convective part of the thunderstorms, an expanded quasi-vertical profile method is applied to examine the temporal evolution of microphysical processes of upper charge regions. The convective part in the isolated thunderstorm is defined as one separated from nearby storms by an area of composite ZH larger than 40 dBZ at and above -10℃ layer, and a criterion of correlation coefficient ρHV greater than 0.8 is used to remove poor quality radar data. Meanwhile, only the lightning flashes within the given volume are used to calculate the IC lighting flash rate and explore the signatures with the upper charge regions.


The results indicate that during the different stages from the early developing stage of isolated thunderstorms to the end of the mature stage, the upper charge regions above the -10 ℃ layer experienced an evolution process from initiation to development accompanied by the rise of the charge region in the updraft and the enhancement of charge concentration. In the mature stage of thunderstorm, the upper charge regions extended from the -30℃ layer to the cloud top, followed by a decay process in the upper charge region at the end of the mature stage, in which the IC lightning flash rate is larger than 60 flashes/min. At the same time, the mean structure evolution of the upper charge regions exhibited a good relationship with the in-cloud lightning flash rate.

How to cite: Wang, S.: Analysis of the Signatures Related to the Upper Charge Regions in an Isolated Thunderstorm Observed by Dual-Polarized Phased Array Weather Radar, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-4254,, 2024.

Winter thunderstorms often exhibit compact vertical dimensions and lower heights of the major charge centers and are often accompanied by strong wind shear, with a propensity for positive cloud-to-ground strokes that can produce mesospheric transient luminous events (e.g. sprites, haloes, elves and jets). There are many optical observations confirming this over the Sea of Japan and the Mediterranean Sea, which are known to be the most convectively active regions during Northern Hemisphere winter.

We use a 3D quasi-electrostatic model (Haspel et al., 2022) with wintertime thunderstorm charge configurations to evaluate sprite inception regions in the mesosphere under various conditions typical of the Eastern Mediterranean. This is a is a relatively new, numerically robust model based on an analytical solution to Poisson’s equation that was developed specifically to handle non-symmetric charge configurations in a large 3D domain.  We address several key questions related to the onset of sprites in winter: (a) the minimum charge that enables sprite inception under the compact thunderstorm structures, (b) the effect of wind shear (lateral offsets of 3-5 km between the cloud charge centers) on the electric field and the location of the area of possible sprite inception, and (c) how the time difference between consecutive strokes in adjacent cumulonimbus clouds affects the size and location of the area of possible sprite inception. Additionally, we will present results of sensitivity studies on the discharge time and profile, showing how the area of possible sprite inception depends on this factor.



Haspel, C., G. Kurtser and Y. Yair (2022). The feasibility of a 3D time-dependent model for predicting the area of possible sprite inception in the mesosphere based on an analytical solution to Poisson's equation. Jour. Atmos. Sol. Terr. Phys.,230, 105853, doi:10.1016/j.jastp.2022.105853.

How to cite: Haspel, C. and Yair, Y.: Numerical simulations of the mesospheric region for sprite inception in winter thunderstorms over the Eastern Mediterranean, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-4618,, 2024.

EGU24-4634 | ECS | Orals | NH1.5

Regional differences in thunderstorm intensity driven by monsoon and westerlies over the Tibetan Plateau 

Lei Wei, Xiushu Qie, Zhuling Sun, and Chen Xu

Thunderstorms are weak but frequent, and exhibit unique charge structures over the Tibetan Plateau (TP) where the average elevation is higher than 4 km. In this study, all detected thunderstorms over the TP between 1998 and 2013 by TRMM were divided into four intensity categories: weak, median, severe and extreme. This classification was based on the 75%, 90%, and 99% values of flash rate, maximum 40 dBZ height, minimum 85 GHz polarization-corrected temperature (PCT), and minimum 37 GHz PCT, respectively. The monthly distributions of thunderstorm intensity show that all categories mostly occur in summer over most regions of the TP, and in spring near the Himalayas. Although the peaks of thunderstorms occur during 1300-1600 LT, the thunderstorms occurring in the early morning and evening have a high probability of developing into severe and extreme thunderstorms. This is distinct from the thunderstorms over the Sichuan Basin, the surrounding areas, and the middle and lower reaches of the Yangtze River at the same latitude. On the basis of westerlies- and monsoon-dominated regions, as well as the altitude, the TP was divided into four regions: the eastern, northern, southern and western regions of the TP (namely ETP, NTP, STP and WTP, respectively). The ETP and STP are primarily influenced by the monsoon, with the ETP at a lower altitude than the STP. Conversely, the WTP and NTP are affected by the westerlies, with the WTP situated at a higher altitude than the NTP. Thunderstorms over the ETP are more likely to be severe and extreme than those over the NTP. The percentage of weak thunderstorms is highest over the WTP. It is found that the maximum top height, development depth, horizontal development area, and development volume at 20 dBZ, 30 dBZ, and 40 dBZ echoes are largest over the ETP, followed by the NTP and STP, while being smallest over the WTP. The results imply that thunderstorms influenced by the monsoon are larger and more likely to be severe and extreme than those influenced by the westerlies.

How to cite: Wei, L., Qie, X., Sun, Z., and Xu, C.: Regional differences in thunderstorm intensity driven by monsoon and westerlies over the Tibetan Plateau, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-4634,, 2024.

EGU24-5346 | Posters on site | NH1.5

Spectral Analysis of High-Energy Radiation Events Observed during the ALOFT 2023 Campaign 

David Sarria, Nikolai Østgaard, Martino Marisaldi, Timothy Lang, Eric Grove, Mason Quick, Hugh Christian, Chris Schultz, Richard Blakeslee, Ian Adams, Rachael Kroodsma, Gerald Heymsfield, Andrey Mezentsev, Ingrid Bjørg Engeland, Anders Fuglestad, Nikolai Lehtinen, Kjetil Ullaland, Shiming Yang, Bilal Hasan Quresh, and Jens Søndergaard and the ALOFT Team

The Airborne Lighting Observatory for FEGS and TGFs (ALOFT) is equipped with a comprehensive set of instruments on-board a NASA ER-2 research aircraft for observing Terrestrial Gamma-ray Flashes (TGFs) and gamma-ray glows from thunderclouds. The ER-2 research aircraft flew at about 20 km altitude, above thunderstorms, from July 1st to July 30th, 2023, for a total flight time of about 60 hours.  The onboard instrument suite comprised several X/gamma-ray detectors, which spanned a dynamic range of four orders of magnitude in flux and covered the entire energy spectrum associated with the gamma-ray transients.

    During the campaign, we observed over 130 short gamma-ray transients, along with hundreds of gamma-ray glows. Several of these detections consisted of thousands of photon counts, allowing precise and unprecedented spectral analyses.

    In this study, we present a comprehensive spectral analysis of various events using a forward modeling technique and Monte-Carlo simulations. This approach enables us to constrain the source characteristics of these events, including their source energy spectrum, production altitude and offset, spatial extension, and the brightness (fluence) of the source RREA electrons.

How to cite: Sarria, D., Østgaard, N., Marisaldi, M., Lang, T., Grove, E., Quick, M., Christian, H., Schultz, C., Blakeslee, R., Adams, I., Kroodsma, R., Heymsfield, G., Mezentsev, A., Bjørg Engeland, I., Fuglestad, A., Lehtinen, N., Ullaland, K., Yang, S., Hasan Quresh, B., and Søndergaard, J. and the ALOFT Team: Spectral Analysis of High-Energy Radiation Events Observed during the ALOFT 2023 Campaign, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-5346,, 2024.

Lightning now has designated as an Essential Climate Variable in the Global Climate Observing System to understand the climate change. Lightning detection from geostationary satellites enables continuous monitoring of lightning activity. The satellite-borne lightning imagers take advantage of optical imaging technology combined with multiple filtering methods to extract the weak signals of lightning from very strong background signals and eventually clustering to reconstruct the original lightning flashes. By using the observation data of Fengyun-4A Lightning Mapper Imager (LMI), the first geostationary satellite-borne lightning imager developed in China, the lightning activity and the optical characteristics of lightning flashes in China were analyzed. The lightning activity observed by LMI exhibits obvious regional, seasonal and diurnal variation properties. The flashes are mainly concentrated in the southeastern coastal region in China and the southwestern China. During the pre-monsoon period (March-May), LMI detected lightning outbreaks in southwestern China and its surrounding areas, while during the monsoon period (June-September), both eastern southwestern China and southeastern coastal region in China show a significant dense distribution of lightning flashes. The climatic characteristics of lightning activity and the simultaneous observations of Lightning Imaging Sensor (LIS) on the International Space Station (ISS) confirm the LMI observations. However, there is a difference between the absolute amounts of the LMI and LIS observations. The overall number of lightning flashes observed by LMI is relatively lower than that observed by LIS. In addition, the detection capability of LMI is higher at low latitudes compared to mid-latitudes, and is higher during daytime hours than that during nighttime hours. As for the flash properties, which mainly refer to the optical radiance, area, and duration of lightning flashes, there are also regional differences for these properties observed by LMI. The high values of flash properties are concentrated in southern China. The LMI observations are related to the radiometric response of its detector and the difference in spatial resolution within the large field of view of geostationary orbit observations.

How to cite: Hui, W. and Zhang, W.: Lightning Activity in China and Its Optical Characteristics Observed by Geostationary Satellite, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-5363,, 2024.

EGU24-5400 | ECS | Posters on site | NH1.5

The intensity distribution of Terrestrial Gamma-ray Flashes from the ALOFT flight campaign 

Anders Fuglestad and the ALOFT team

In July 2023, the Airborne Lightning Observatory for FEGS and TGFs (ALOFT) flight campaign took place using a NASA ER-2 research aircraft flying over the Gulf of Mexico and the Caribbean Sea. The campaign consisted of about 60 flight hours at a cruise altitude of 20 km, using live telemetry to target gamma-ray glowing thunderclouds.

The payload consisted of several instruments including gamma-ray detectors with a dynamic range spanning four orders of magnitude in flux, an imaging array of optical photometers, electric field change meters, radiometers, and radar systems. In addition to several ground stations measuring very low frequency, low frequency, and very high frequency radio signals.

96 TGFs were detected by ALOFT. For 44 of these events, it was possible to get an estimate of the location of the source using both correlated optical pulses and lightning detection networks.

With the estimate of the source location and the gamma-ray observation from ALOFT. Monte Carlo simulations were used to get an estimate of the source intensity of the TGFs.

Based on the results it was determined that the vast majority of the 44 TGFs investigated have source intensities below the threshold needed to be observed from current satellite instruments, which indicates a large population of low intensity TGFs that has gone previously undetected. These results contribute to the open debate on the rarity of TGFs.

How to cite: Fuglestad, A. and the ALOFT team: The intensity distribution of Terrestrial Gamma-ray Flashes from the ALOFT flight campaign, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-5400,, 2024.

EGU24-6398 | ECS | Orals | NH1.5

Side discharges on positively charged lightning leaders 

Shanfeng Yuan, Xiushu Qie, Rubin Jiang, and Dongfang Wang

Recent observations unveiled two types of side discharges associated with positive leaders: needle discharges and nearby bidirectional leaders. The formation mechanism and connections of two phenomena remained unclear due to the lack of synchronous optical detection and radio mapping data. Here we present the first high-speed video and low-frequency lightning mapping results. Negative branches of nearby bidirectional leaders can propagate after connecting to the parent positive channel, and needle discharges act as positive connecting leaders. Our research shows that positive leaders exhibit unconventional channel extensions, maintained by frequent recoil leaders, sharing characteristics with streamer discharges. Notably, when two approaching positive leaders develop in this manner, they can eventually collide. These findings significantly advance our understanding of side discharges on positive leaders, offering fresh insights into these intriguing phenomena.

How to cite: Yuan, S., Qie, X., Jiang, R., and Wang, D.: Side discharges on positively charged lightning leaders, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-6398,, 2024.

EGU24-6468 | Posters on site | NH1.5

On the impact of thunder on cloud droplets and ice crystals  

Konstantinos Kourtidis and Stavros Stathopoulos

In the lightning channel pressures can be of the order of 100 atm and hence in the produced thunder, sound pressure levels (SPL) can be very high. Additionally, the thunder frequency spectra have peaks for peal and claps at around 100 Hz and around 50 Hz for rumble sounds, with intracloud lightning having peaks at even fewer Hz. These low frequencies are ideal for acoustically induced orthokinetic agglomeration of droplets. Thunder occurs in cloud environments where not only large numbers of droplets are present, but additionally the shockwave front expands at supersonic velocities and hence could cause near the lightning channel modulations of droplet size distributions and increase ice crystals numbers through e.g. vibrational breakup. We present calculations for the two mechanisms above (orthokinetic agglomeration and vibrational breakup) for typical cloud droplet sizes and concentrations, including also clouds containing desert dust. In thunderstorm conditions, it is found that acoustic orthokinetic agglomeration of droplets can be very effective and can produce very rapidly changes in the mean cloud droplet diameter. Also, it is found that the critical flow velocities, over which breakup occurs, is easily exceeded near the lightning channel and will lead to droplet and ice crystal breakup. We note that all models of ice crystal generation in clouds substantially underestimate the observed ice crystal numbers, and the mechanism presented here may be responsible for the discrepancy. We also note that these processes need further study to assess how they could interfere with the lightning generation process itself, through both charge redistribution in the modified droplet size distribution spectra, as well as the increase in vertical and turbulent transport velocities of the smaller ice crystals resulting from breakup. 

How to cite: Kourtidis, K. and Stathopoulos, S.: On the impact of thunder on cloud droplets and ice crystals , EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-6468,, 2024.

EGU24-6523 | Posters on site | NH1.5

ESTHER: a small project to investigate gamma-ray emissions in thunderstorms and volcanic lightning 

Alessandro Ursi and Danilo Reitano

Detecting terrestrial gamma-ray flashes (TGFs) from the ground is a relatively new frontier in atmospheric science and has opened up new avenues for research. Also, the recent detection of a TGF produced during the massive Hunga Tonga–Hunga Ha'apai eruption, pointed out the possibility that even volcanic lightning might produce gamma-ray emissions at MeV energies.

In this context, we present the Experiment to Study Thunderstorm High-Energy Radiation (ESTHER), a small project funded by the Italian National Institute for Astrophysics (INAF), aimed at monitoring from the ground gamma-ray emissions produced during thunderstorms and, possibly, by volcanic lightning. The ESTHER set-up consists of a gamma-ray detection system and a VLF radio receiver, to be installed on the top of the Etna volcano (Italy). The selected installation site is the Etnean Observatory of the Italian National Institute of Geophysics and Volcanology (INGV), located at 2,818 m altitude and laying less than 2.7 km from the main volcano craters.

An extensive analysis of the flash rate recorded at Mt. Etna in the last eight years pointed out that the mountain top is interested by strong lightning activity in the summer months, making it a suitable location for the investigation of lightning and associated high-energy phenomena. In particular, the largest fraction of discharges turned out to cluster nearby the mountain peak and right above the main volcano craters, where the frequent presence of volcanic ashes possibly increases the electrical conductivity, under conditions of humid air typical of thunderstorms, making the region above the volcano's top a natural trigger for lightning. Moreover, as for other volcanoes around the world, Etna has been documented to produce volcanic lightning (last times in 2015 and 2022). As a consequence, given the proximity of the Etnean Observatory to the main craters, ESTHER will enjoy a privileged location for investigating potential gamma-ray emissions produced either by thunderstorms and volcanic lightning. In conditions of clear sky, ESTHER will also provide an as much as possible continuous monitoring of the environmental gamma-ray background, allowing to point out potential variations of it before, during, or after volcanic eruptions. The ESTHER set-up will be installed and start its first data acquisitions in spring 2024.

How to cite: Ursi, A. and Reitano, D.: ESTHER: a small project to investigate gamma-ray emissions in thunderstorms and volcanic lightning, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-6523,, 2024.

EGU24-7273 | Orals | NH1.5

Observation of positive Narrow Bipolar Events in the Mediterranean region 

Ivana Kolmašová, Ondřej Santolík, Serge Soula, Eric Defer, Yanan Zhu, Radek Lán, Stéphane Pedeboy, and Andrea Kolínská

Narrow Bipolar Events (NBEs) are brief intracloud (IC) discharge processes that generate powerful radiation in the HF and VHF radio bands. NBEs typically occur in isolation, but they have also been identified as initial events in IC lightning flashes. Their incidence is statistically correlated with the strength of convection. NBEs can exhibit both polarities and usually occur in the upper regions of the thundercloud.

We present, for the first time, properties of NBEs detected in the Mediterranean region. The dataset comprises 37 events recorded by broadband magnetic loops located at two sites in France. The events were identified using the list of NBEs from 2022 provided by the Earth Network. The frequency range of our broadband sensors enabled us to obtain detailed shapes of NBE pulses. We calculated rise times, full width at half maximum times, and zero-crossing times of NBE pulses to facilitate comparisons with observations of NBEs in other parts of the world. The majority of NBE pulses observed in the Mediterranean region were isolated events occurring above the land and displaying a simple bipolar waveform with an overshoot peak of the opposite polarity. For two events, we supplemented our observation with the data from the SAETTA (Suivi de l’Activité Electrique Tridimensionnelle Totale de l’Atmosphère) lightning mapping array. Additionally, we estimated the altitude of the NBE events and placed our observations in the meteorological contexts to determine why NBE occurrences in the Mediterranean region have been overlooked until now.


How to cite: Kolmašová, I., Santolík, O., Soula, S., Defer, E., Zhu, Y., Lán, R., Pedeboy, S., and Kolínská, A.: Observation of positive Narrow Bipolar Events in the Mediterranean region, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-7273,, 2024.

EGU24-7900 | Orals | NH1.5

TGF and gamma-ray glow highlights from the ALOFT 2023 flight campaign 

Nikolai Ostgaard, Timothy Lang, Martino Marisaldi, Eric Grove, Mason Quick, Hugh Christian, Cristopher Schultz, Richard Blakeslee, Ian Adams, Rachael Kroodsma, Gerald Heymsfield, Andrey Mezentsev, David Sarria, Ingrid Bjorg Engeland, Anders Fuglestad, Nikolai Lehtinen, Kjetil Ullaland, Shiming Yang, Bilal Hasan Qureshi, and Jens Sondergaard and the ALOFT team

During the summer of 2023 the  Airborne Lighting Observatory for FEGS and TGFs (ALOFT) field campaign was performed. With a NASA ER-2 research aircraft, flying at 20 km altitude, ALOFT was searching for Terrestrial Gamma ray Flashes (TGF) and gamma-glowing thunderclouds in Central America and Caribbean. The ALOFT payload included a comprehensive number of instruments:

1) Several gamma-ray detectors covering four orders of magnitude dynamic range in flux as well as the full energy range for TGF/gamma-ray glow detection (UIB-BGO and ISTORM).

2) Fly’s Eye GLM Simulator (FEGS), an imaging array of photometers sensitive to different wavelengths, and electric field change meters.

3) Lightning Instrument Package (LIP), giving three component electric field measurements.

4) a suite of microwave radiometers and radars for cloud characterization: the Advanced Microwave Precipitation Radiometer (AMPR), Configurable Scanning Submillimeter-wave Instrument/Radiometer (CoSSIR), Cloud Radar System (CRS), and X-band Radar (EXRAD)


5) An extensive set of ground-based radio observations.


For all the 10 flights, 60 hours total, realtime gamma-ray detections were downlinked. Due to this simple but novel mission concept, we knew in real time if the aircraft was passing a gamma-glowing cloud and the pilot was instructed to return to the same thundercloud as long as the cloud was glowing. During the campaign ALOFT observed a total of 130 transient gamma-ray events and hundreds of gamma-ray glows. With the richness of the ALOFT observations we learned that thundercloud can glow for much longer than minute scale and over much larger areas than previously reported. We also learned that transient gamma-ray events come in a large variety and new types of events were discovered.  In this presentation we will give an overview of the main results and discoveries by the ALOFT campaign


How to cite: Ostgaard, N., Lang, T., Marisaldi, M., Grove, E., Quick, M., Christian, H., Schultz, C., Blakeslee, R., Adams, I., Kroodsma, R., Heymsfield, G., Mezentsev, A., Sarria, D., Bjorg Engeland, I., Fuglestad, A., Lehtinen, N., Ullaland, K., Yang, S., Hasan Qureshi, B., and Sondergaard, J. and the ALOFT team: TGF and gamma-ray glow highlights from the ALOFT 2023 flight campaign, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-7900,, 2024.

EGU24-7927 | Orals | NH1.5

A novel view of gamma-ray glows from the ALOFT 2023 flight campaign 

Martino Marisaldi, Nikolai Østgaard, Timothy J. Lang, J. Eric Grove, Mason Quick, Hugh Christian, Christopher J. Schultz, Richard Blakeslee, Ian S. Adams, Rachael A. Kroodsma, Gerald M. Heymsfield, Andrey Mezentsev, David Sarria, Ingrid Bjørge-Engeland, Anders Fuglestad, Nikolai Lehtinen, Kjetil Ullaland, Shiming Yang, Bilal Hasan Qureshi, and Jens Søndergaard and the ALOFT team

The Airborne Lightning Observatory for FEGS and TGFs (ALOFT) was a field campaign targeted at Terrestrial Gamma-ray Flashes (TGFs) and gamma-ray glows from thunderclouds. The campaign was successfully carried out during July 2023, for a total of 60 flight hours in the Gulf of Mexico and the Caribbean. The scientific payload was flown on a NASA ER-2 research aircraft, capable to fly at 20 km altitude above thunderclouds. The payload included a suite of gamma-ray detectors spanning four orders of magnitude dynamic range in flux, and a complete suite of instruments for the characterisation of the electrical and optical activity, and the thundercloud environment. A key asset of the mission was the real-time downlink of gamma-ray count rates, which enabled the immediate identification of gamma-ray glowing regions. The pilot was then instructed to turn and pass over the same glowing region to explore its spatial extension and duration.

ALOFT resulted in the detection of hundreds of gamma-ray glows, anticipating a revolution in our understanding of the phenomenon. Thunderclouds were observed to glow for hours and over several thousands of square kilometers, making glows a much more pervasive phenomenon than previously reported. Glows show significant time variability from seconds down to millisecond time scale, suggesting a relation to short transients such as TGFs more complex than previously thought. Glows are observed in association with the overpass of active convective cores, 20-25 km in size, yet their time variability and intensity modulation suggest a more complex spatial structure.

These observations challenge the current view of glows as quasi-stationary phenomena related to relatively stable electrification conditions. The observed glows show highly dynamic temporal and spatial structures and are closely related to the development phases of active thunderclouds. These observations call for a rethinking of the assumptions at the basis of current modeling efforts.

How to cite: Marisaldi, M., Østgaard, N., Lang, T. J., Grove, J. E., Quick, M., Christian, H., Schultz, C. J., Blakeslee, R., Adams, I. S., Kroodsma, R. A., Heymsfield, G. M., Mezentsev, A., Sarria, D., Bjørge-Engeland, I., Fuglestad, A., Lehtinen, N., Ullaland, K., Yang, S., Qureshi, B. H., and Søndergaard, J. and the ALOFT team: A novel view of gamma-ray glows from the ALOFT 2023 flight campaign, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-7927,, 2024.

EGU24-7940 | Orals | NH1.5

LOFAR Observations of the Initial Stage of IC Dart Leaders 

Brian Hare, Olaf Scholten, Paulina Ťureková, Steven Cummer, Joseph Dwyer, Ningyu Liu, Chris Sterpka, and Sander ter Veen

In previous work we have found that dart leaders quench needle activity; where dart leaders are charge pulses that re-trace previously established lightning leader channels, and needles are small repeating negative discharges that propagate away from positive lightning channels. We hypothesized that dart leaders could be quenching needles by carrying negative charge away from the region of needle activity. Therefore, in order to further explore the interactions between dart leaders and needles, we are investigating the beginnings of different dart leaders with the LOFAR radio telescope, which uses hundreds of antennas in northern Netherlands to image lightning in the 30-80 MHz band with meter and nanosecond level accuracy. We have found that, consistent with previous work, dart leaders start slow with weak radio emission and then accelerate over a period roughly around 50 µs in duration until they reach a maximum speed and radio intensity. However, we also observe that the power of the radio emissions from the dart leaders exhibits large, randomly-timed, variations. These variations do not appear to be a form of leader stepping. The time-differences between individual peaks in the time trace is significantly longer than the width of each peak (or pulse) that is dominated by the antenna function, (FWHM ~ 50 ns). One possible explanation could be that the power fluctuations are consistent with Poisson statistical variations of radio sources (possibly streamers), which would imply that at any point in time the radio emission is dominated by a small number of strong emitters, as opposed to millions of small streamers. A second possible explanation is that the fluctuations could be due to small-scale structural variations along the previously established plasma channel, which we have observed in previous work.

How to cite: Hare, B., Scholten, O., Ťureková, P., Cummer, S., Dwyer, J., Liu, N., Sterpka, C., and ter Veen, S.: LOFAR Observations of the Initial Stage of IC Dart Leaders, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-7940,, 2024.

EGU24-7982 | ECS | Posters virtual | NH1.5

Enhancement of Catastrophic Positive Cloud to Ground Lightning in recent years over Maharashtra (India): Role of Dust Aerosols 

Abhijeet Gangane, Sunil Pawar, Prajna Priyadarshini, and Venkatachalam Gopalakrishnan

Many studies have shown that aerosols can influence microphysical processes inside thunderclouds that could affect charge-generation processes. Cloud to Ground (CG) lightning data from Ground-based observations (IITM-LLN) over the State of Maharashtra, India, from 2014 to 2023, have been analyzed here to study the percentage and physical mechanism associated with the enhancement of catastrophic Positive CG in total CG lightning. Our analysis shows that the average positive CG percentage remains above 25% during the monsoon (July-September) and post-monsoon (October-November). This increased percentage of positive CG is attributed to elevated dust aerosol concentration over the study region during the monsoon and post-monsoon periods. An enormous amount of dust can be seen during the Indian Summer Monsoon (ISM) over the Arabian Desert and neighborhood extending up to the western Indian (Maharashtra) region. Dust aerosol intrusion into the thunderstorm acts as Ice nuclei (IN) as well as Cloud Condensation Nuclei (CCN) and can influence charge separation processes inside the cloud. In recent years, we observed an enhancement of Dust AOT over Maharashtra state, indicating that the increasing trend in Positive CG lightning is closely linked to the transport of desert dust from the Middle East and elevated aerosol content during the post-monsoon season. Here, we propose that these high concentrations of dust aerosols near the cloud base acting as IN produce a high concentration of ice crystals in the lower portion of the cloud, which can form a strong positive charge region in the lower part of the mixed-phase region by non-inductive charging mechanism. This strong positive charge region in the lower portion of the mixed phase region may be responsible for the observed increased percentage of positive CG over the study region.

How to cite: Gangane, A., Pawar, S., Priyadarshini, P., and Gopalakrishnan, V.: Enhancement of Catastrophic Positive Cloud to Ground Lightning in recent years over Maharashtra (India): Role of Dust Aerosols, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-7982,, 2024.

EGU24-7996 | ECS | Posters on site | NH1.5

Monte Carlo Error Analysis of Lightning Interferometry with LOFAR 

Paulina Turekova, Brian Hare, Olaf Scholten, Steven Cummer, Joseph Dwyer, Ningyu Liu, Chris Sterpka, and Sander ter Veen
The LOFAR radio telescope works on a principle of radio interferometric imaging. It coherently sums the signal of hundreds of antennas in northern Netherlands, covering the 30-80 MHz window of the very high frequency (VHF) band of 30-300 MHz. We are using the TRI-D algorithm to extract 3-D polarization data of a lightning flash observed by LOFAR. TRI-D functions by coherently summing recorded voltages, accounting for the antenna function, polarization, and geometric time delay for each voxel. The result is split into time slices. A coherent intensity is calculated for each time slice, and the maximum of this value is set as a source location. The outcome is a reconstructed source location and polarization as seen by the LOFAR antennas. We are now exploring the accuracy of TRI-D in response to realistic parameters. In this work, we perform a Monte Carlo error analysis which simulates the voltages on each antenna from an assumed dipole emitter, adds normally distributed noise, and then reconstructs the source properties with TRI-D. The difference between the simulated input and the reconstruction gives us an estimate of the resulting error bars. We will show a detailed account of the interferometry technique that produces our data, the Monte Carlo simulation that tests the accuracy of our model and finally, our polarization results.

How to cite: Turekova, P., Hare, B., Scholten, O., Cummer, S., Dwyer, J., Liu, N., Sterpka, C., and ter Veen, S.: Monte Carlo Error Analysis of Lightning Interferometry with LOFAR, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-7996,, 2024.

EGU24-8002 | ECS | Posters on site | NH1.5

Measuring evaporation-condensation charging of individual aerosol particles 

Andrea Stoellner, Isaac Christopher David Lenton, Caroline Muller, and Scott Russell Waitukaitis

Although cloud electrification has been studied for hundreds of years, it is still not fully understood [1]. The most promising charging mechanism – ice crystal-graupel collision charging – answers some of our questions, but leaves us with others. Why do ice crystals and graupel charge on collision in the first place? And why do they reverse their charging behavior below a certain temperature? To get some insights we take a step back and look at the charging behavior of individual aerosol particles in a humid environment. Shavlov et al. [2] suggest that the hydroxide and hydronium ions formed by the autodissociation of water are sufficient to cause charging during evaporation and condensation of water droplets or surface-adsorbed water on solid particles. This small amount of charge could be a precursor to bigger charge exchange during collision.

            We aim to test this hypothesis by levitating individual aerosol particles in an optical trap and measuring their charge while varying humidity. Our setup allows for trapping of different types of solid and liquid particles in the micrometer size range, like water droplets and silica microspheres. In the future we also hope to study ice crystals. Figure 1 shows an illustration of the measurement principle. The particle’s charge is measured by applying a sinusoidal electric field and observing the resulting particle motion. The Mie scattering pattern of the particle furthermore gives information about the particle’s size and refractive index, both at equilibrium and during evaporation/condensation. The experiment allows us to control the relative humidity, pressure and air ion concentration around as well as air flow across the particle.

Ultimately we hope to contribute to a better understanding of the microphysical processes involved in thundercloud electrification and adjacent electrical phenomena in the atmosphere. 

FIGURE 1. Optical tweezers (wavelength λ = 532 nm) holding a solid or liquid aerosol particle. A sinusoidal electric field is applied between the two electrodes and the resulting particle motion as well as the particle’s Mie scattering pattern are recorded.


This project has received funding from the European Research Council (ERC) under the European Union’s Starting Grant (A. Stoellner, I.C.D. Lenton & S.R. Waitukaitis received funding from ERC No. 949120, C. Muller received funding from ERC No. 805041).



  • Berdeklis, P. and List, R. (2001) J Aerosol Sci. 58(18) 2751–2770.
  • Shavlov A. et al. (2018) J Aerosol Sci. 123 17-26.

How to cite: Stoellner, A., Lenton, I. C. D., Muller, C., and Waitukaitis, S. R.: Measuring evaporation-condensation charging of individual aerosol particles, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-8002,, 2024.

EGU24-9355 | Posters on site | NH1.5

Dynamics of global lightning activity on different time scales as indicated by Schumann resonance frequency variations 

Gabriella Sátori, Tamás Bozóki, Earle Williams, Ernő Prácser, Raidiel Puig, and Rachel Albrecht

The electromagnetic waves in the Schumann resonance (SR) frequency range (<100 Hz) radiated by natural “lightning antennas” excite the Earth-ionosphere cavity confined between the Earth’s surface and the ionospheric D-region of ~100 km height. This contribution provides observational evidence for the relationships between the variations of peak frequencies of the first three modes and the global/regional lightning dynamics based on SR observations of the vertical electric field component, EZ, at Nagycenk (NCK), Hungary, Central Europe. Lightning source-observer distance-dependent frequency variations are considered on the annual, seasonal and diurnal time scale as well as during specific events when squall-line formation of lightning activity in South America moves toward NCK. The observations are interpreted with model calculations. The distance-dependent frequency variation has important applications to climate issues as well.

How to cite: Sátori, G., Bozóki, T., Williams, E., Prácser, E., Puig, R., and Albrecht, R.: Dynamics of global lightning activity on different time scales as indicated by Schumann resonance frequency variations, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-9355,, 2024.

EGU24-9526 | Orals | NH1.5

On the radio wave polarization of Saturn lightning 

Georg Fischer, Ulrich Taubenschuss, David Pisa, and Masafumi Imai

The radio waves with Saturn lightning origin have been studied since the first detection by Voyager 1, but their wave polarization has rarely been explored. Fischer et al. (2007, JGR 112, A12308) examined lightning from a storm located at 35° south latitude and found its radio emissions below 2 MHz to be highly polarized (80%) in a right-handed circular sense with respect to the wave propagation direction. They explained this by absorption of the extraordinary mode in Saturn's ionosphere and the dominance of the ordinary mode emission, as the radio waves are propagating against a direction of the magnetic field when coming from a source in the southern hemisphere. A limited examination of Saturn lightning from the so-called Great White Spot at 35° north latitude by Fischer et al. (2011, Nature 475, 75-77) revealed radio wave polarization in the left-handed sense. In this presentation we will show the radio wave polarization of lightning from various other storms in Saturn's atmosphere, which have not been examined until today. In this way we want to corroborate the hypothesis that the sense of the circular radio wave polarization of Saturn lightning depends on the hemispherical location of the storm.

How to cite: Fischer, G., Taubenschuss, U., Pisa, D., and Imai, M.: On the radio wave polarization of Saturn lightning, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-9526,, 2024.

EGU24-9986 | ECS | Orals | NH1.5 | Highlight

Potential gradient as a predictor of fog 

Caleb Miller, Keri Nicoll, Chris Westbrook, and R. Giles Harrison

Although fog is an important weather phenomenon, it remains difficult to predict using traditional methods. This could be improved by new observations-based nowcasting systems. It has long been understood that fog affects measurements of atmospheric electricity. However, there has been disagreement in the literature on whether these changes contain information which is valuable for fog prediction beyond other commonly used methods. Here, results are presented which show that the potential gradient (PG), a measure of atmospheric electricity, could be used as an additional diagnostic in predicting fog for timescales of several hours. A much larger dataset of fog and PG is examined than has been previously possible, which allows for a more robust understanding of the behaviour of the PG during radiation fog. It is found to increase by a median of 58 V/m by the start of the event. In addition, this increase is found to begin over two hours in advance of the fog, 30% of the time. This shows that PG may contain useful fog nowcasting information. A number of individual fog case studies are presented and the applicability of the general results to these specific cases is discussed. 

How to cite: Miller, C., Nicoll, K., Westbrook, C., and Harrison, R. G.: Potential gradient as a predictor of fog, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-9986,, 2024.

EGU24-10352 | Orals | NH1.5

Cloud Microphysical Characteristics Associated with Blue Corona Discharges at thundercloud tops 

Dongshuai Li, Alejandro Luque, Torsten Neubert, Olivier Chanrion, Yanan Zhu, Jeff Lapierre, Nikolai Østgaard, and Víctor Reglero

Blue corona discharges are bursts of streamer discharges often observed at the top of thunderclouds, but the conditions in the clouds that generate them are not well understood.

The cloud microphysical parameters related to them are important for future empirical studies and for theoretical models and simulations. Previous studies modeled the scattering and absorption emissions from blue corona discharges by assuming mean particle radius of 10–20 μm and densities of 1–2.5 × 10^8 m^−3, resulting in photon mean free paths of 1–20 m.

Here we present the first-ever estimate of important microphysical parameters related to blue corona discharges based on data measurements from the CALIPSO lidar. The results showed that most blue corona discharges were associated with ice particles with a radius of ∼50 μm and a number density of ∼ 2 × 10^7 m^−3, resulting in a photon mean free path of ∼3 m.

Around 20% of the blue corona discharges coincide with Narrow Bipolar Events (NBEs) indentified from the Earth Networks Total Lightning Network.The altitudes of blue corona discharges that were identified as NBEs are derived from both the optical and radio bands. It revealed that in six out of nine cases, the R^2 value was greater than 0.85, indicating a good agreement between the two methods and supporting our estimate of the photon mean free path as 3 m. However, in the shallowest and deepest cases, there was some discrepancy between the altitudes determined by the two methods, suggesting more complex cloud microphysical parameters. Possible reasons for the discrepancy, such as the homogeneous approximation for the cloud's microphysical parameters and the simplification of the source length, will be discussed.

How to cite: Li, D., Luque, A., Neubert, T., Chanrion, O., Zhu, Y., Lapierre, J., Østgaard, N., and Reglero, V.: Cloud Microphysical Characteristics Associated with Blue Corona Discharges at thundercloud tops, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-10352,, 2024.

EGU24-11101 | Orals | NH1.5

Mapping out lightning processes in both the VHF and VLF using LOFAR and the Met Office’s lightning detection system, LEELA 

Graeme Marlton, Brian Hare, Olaf Scholten, Mike Protts, Ed Stone, Sue Twelves, and Francesco Devoto

Lightning is one of the most destructive meteorological phenomena being a hazard to people and objects on the ground as well as aircraft. In addition to the strong currents and optical emission from a lightning stroke broadband radio emissions are also produced from the VLF to VHF. The LOw Frequency ARray (LOFAR) telescope centred in the Netherlands consists of a large array of VHF (30-300 MHz) receivers which can be configured to image a lightning strike in the 30-80 MHz bandwidth. The Met Office Lightning Electromagnetic Emission Location using Arrival time differencing LEELA system operates in the VLF (3-30 kHz). It also archives the raw incoming VLF data allowing the individual VLF waveforms to be analysed. From a lightning flash recorded in June 2021 over the Netherlands, 8 distinct events were detected by both systems. Here we present an analysis of these 8 events which include dart leaders, negative leaders, an intensely radiating negative leader and a cloud to ground strike. Initial results show that while both systems co-locate the events they are sensitive to different processes within the lightning strike process. VHF emission from a lightning strike is observed for periods of 30-40 ms and captures the development of the lightning channel. However, VLF emission is observed for much shorter periods of a few ms likely corresponding to the rapid vertical movement of charge during the strikes.

How to cite: Marlton, G., Hare, B., Scholten, O., Protts, M., Stone, E., Twelves, S., and Devoto, F.: Mapping out lightning processes in both the VHF and VLF using LOFAR and the Met Office’s lightning detection system, LEELA, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-11101,, 2024.

EGU24-11257 | Orals | NH1.5

Locating charged regions in extensive layer cloud 

R.Giles Harrison and Keri Nicoll

Extensive layer clouds are common in Earth’s atmosphere. They acquire charge at their upper and lower boundaries, from the vertical current flowing in the global atmospheric electric circuit. The quantity of charge collected is related to the current, the transition distance from clear air to cloudy air at the cloud boundary, and the background cosmic ray ionisation. The transition distance is the region in which a change in conductivity occurs, which determines the charge acquisition. This differs between cloud top and cloud base. At cloud top, the boundary transition distance is closely related to the temperature inversion, which can be less than the transition distance at cloud base. At cloud base, the transition distance depends on droplet growth rate and updraft speed. The combined effects of the local ionisation, current flow and conductivity gradient leads to droplet charging.

Using instrumentation carried on enhanced meteorological radiosondes, the extent of the charged region in extensive layer clouds has been observed with specially developed cloud sensors operating at multiple optical wavelengths, simultaneously with the in situ electrical measurements. (Further, in some situations, ceilometer measurements of backscatter are also available). These soundings are compared with modelled profiles of droplet properties and layer cloud charges, for situations characteristic of mid-latitude and polar clouds. Effects of the droplet size distribution on the layer cloud electrification are also investigated, and responses to variations in cosmic ray ion production.

Charging is known to affect some aspects of the microphysical behaviour of droplets, such as their evaporation and growth rates. This may in turn influence properties of layer clouds in the climate system.

How to cite: Harrison, R. G. and Nicoll, K.: Locating charged regions in extensive layer cloud, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-11257,, 2024.

EGU24-11482 | Orals | NH1.5 | Highlight

Thundercloud high-energy radiation production by long streamers 

Nikolai Lehtinen, David Sarria, Martino Marisaldi, Andrey Mezentsev, Nikolai Østgaard, Steven Cummer, and Yunjiao Pu

The novel Streamer Parameter Model (SPM) [Lehtinen, 2021, doi:10.1007/s11141-021-10108-5] allows to quickly calculate the shape, velocity, and electric field of an electric streamer in air, without resorting to lengthy hydrodynamic simulations. A streamer propagates faster as its length grows. When the streamer length exceeds several meters, the velocity may become comparable to the speed of light, which necessitates correcting the model for relativistic effects. Such long streamers may describe the experimentally observed fast positive and negative breakdown. We propose that they may produce large quantities of relativistic runaway electrons, and therefore x-rays. This is facilitated by several conditions: (1) electric fields at the streamer tip may be sufficiently close to the so-called thermal runaway threshold (~30 MV/m), at which free electrons may accelerate from thermal energies up to relativistic energies; (2) in negative streamers, the energetic electrons are synchronized in velocity with the streamer front; (3) the streamer tip radius may exceed tens of centimeters, providing a large volume of the high field where the thermal runaway acceleration may take place.

We apply SPM to long streamer propagation inside a thundercloud and calculate the relativistic runaway electron production, as well as radio, optical and x-ray radiation. The calculations are compared to the observations of Narrow Bipolar Events (NBE), Terrestrial Gamma Flashes (TGF), and luminous phenomena obtained during the recent ALOFT campaign.

How to cite: Lehtinen, N., Sarria, D., Marisaldi, M., Mezentsev, A., Østgaard, N., Cummer, S., and Pu, Y.: Thundercloud high-energy radiation production by long streamers, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-11482,, 2024.

EGU24-11937 | Orals | NH1.5

Feedback Effects in Positive Corona and Relativistic Runaway Discharges 

Victor Pasko, Sebastien Celestin, Anne Bourdon, Reza Janalizadeh, and Jaroslav Jansky

We discuss characteristic scales and direct physical analogy between the photoionization feedback in conventional positive corona discharges in air and the photoelectric feedback in discharges driven by relativistic runaway electrons in air. In a positive corona system the avalanche of electrons in bulk of discharge volume is initiated by specific distribution of photoionization far away from the electrode.  Under inception conditions in positive corona each electron arriving at the anode creates on average just enough seed electrons in discharge volume through photoionization to replicate itself. Under these self-sustained steady state conditions, photoionization feedback produces just enough secondary electrons upstream of the avalanche to maintain the system in steady state. Analogically, in case of relativistic electron avalanches a feedback process is realized when X-rays emitted by these electrons travel backwards with respect to the electron motion and generate new relativistic electron seeds due to the photoelectric absorption in air. It is demonstrated that terrestrial gamma-ray flashes are produced by growth of long bidirectional lightning leader system consisting of positive and stepping negative leaders. The spatial extent of streamer zones of a typical lightning leader with tip potential exceeding several tens of megavolts is on the order of 10–100 m. The photoelectric absorption of bremsstrahlung radiation generated by avalanching relativistic runaway electrons occurs efficiently on the same spatial scales. The intense multiplication of these electrons is triggered when the size of the negative leader streamer zone crosses a threshold of approximately 100 m (for sea-level air pressure conditions) allowing self-replication of these avalanches due to the upstream relativistic electron seeds generated by the photoelectric absorption.

Pasko et al., GRL, 50, e2022GL102710, 2023,
Pasko et al., PSST, 32, 075014, 2023,

How to cite: Pasko, V., Celestin, S., Bourdon, A., Janalizadeh, R., and Jansky, J.: Feedback Effects in Positive Corona and Relativistic Runaway Discharges, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-11937,, 2024.

Terrestrial gamma-ray flashes (TGFs), powerful bursts of gamma-rays produced within our atmosphere, often occur in association with lightning. However, the mechanisms for generating the large number of runaway electrons required to account for the TGF luminosities remain uncertain. For example, TGFs might be produced by cold-runaway electron production from streamer heads and/or leader tips in the high-field regions near lightning, or TGFs might be produced by the self-sustained production of runaway electrons by relativistic feedback involving backward propagating runaway positrons and backscattered x-rays. Because both mechanisms could possibly occur in the presence of lightning leaders, it has been challenging to test which TGF production mechanisms are important. In this work, detailed simulations are used to test whether TGFs may be produced by thunderstorm electrification alone, without the presence if lightning. It is found that rapid thunderstorm charging may first produce strong gamma-ray glows, followed by large pulses of gamma-rays, followed by multi-pulsed TGFs similar to the TGFs first observed by CGRO/BATSE. Furthermore, the ionization produced by the high-energy particles partially discharges the electric field in some regions while amplifying the field in other regions, potentially allowing for the initiation of narrow bipolar events (NBEs) and/or lightning. If confirmed, such sequence of events would be strong evidence for the relativistic feedback mechanism.

How to cite: Dwyer, J. and Liu, N.: Gamma-ray glows and terrestrial gamma-ray flashes produced by thunderstorm electrification without lightning , EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-12103,, 2024.

EGU24-12247 | ECS | Orals | NH1.5

Investigating Storm Charge Distribution Trends with Intracloud Lightning Polarity Data 

Elizabeth DiGangi, Jeff Lapierre, and Yanan Zhu

There are, at present, two accepted primary paradigms for thunderstorm charge distribution using a simple tripole model: “normal” polarity storms, which are characterized by a central negative charge region, an upper positive charge region, and sometimes a lower positive charge region; and “inverted” polarity storms, which are characterized by a central positive charge region, an upper negative charge region, and sometimes a lower negative charge region. The real distribution of thunderstorm charge is known to be more complex than the tripole model can represent, but the normal/inverted paradigm is still widely used in the field. Characterizing storms as having a normal or inverted polarity has been a subject of interest in lightning research since discovering that inverted storms produce a larger-than-average fraction of positive amplitude cloud-to-ground (CG) lightning compared with normal storms. +CG lightning is understood to have generally higher peak currents and a much greater probability of producing continuing current than -CGs, which is relevant for research into subjects like lightning-initiated wildfires and transient luminous events. Thunderstorm charge distribution is also directly related to storm microphysics and thermodynamics, which, in turn, links it to the meteorological characteristics of storms and storm environments.

Most published research on storm polarity has either investigated large-scale trends in +CG versus -CG frequency from long-range lightning detection systems (LDSs), or has used LDSs which map lightning in 3D to infer storm polarity directly from intracloud (IC) lightning leader propagation patterns. Data on IC lightning from long-range LDSs is a resource which, to our knowledge, has not yet been used to study bulk storm charge structures. It stands to reason that if inverted storms favor the production of more +CGs than normal storms, then they would also favor the production of more -ICs. The goal of this study is therefore to interrogate several years of lightning data from the Earth Networks Total Lightning Network (ENTLN) to determine whether or not IC peak current information can be used to study storm charge structure and the geographic distributions of inverted and normal polarity storms.

How to cite: DiGangi, E., Lapierre, J., and Zhu, Y.: Investigating Storm Charge Distribution Trends with Intracloud Lightning Polarity Data, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-12247,, 2024.

EGU24-12606 | Posters on site | NH1.5

CubeSpark: Space-based 3-D Lightning Mapping using a Constellation of Radio Frequency Sensors 

Sonja Behnke, Kim Katko, Harald Edens, Patrick Gatlin, Timothy Lang, William Haynes, Paul Snow, Jeremiah Rushton, Joellen Renck, Charley Weaver, Larry Bronisz, Jacob Pratt, Steven Dobson, Nikhil Pailoor, Jackson Remington, and Sarah Stough

CubeSpark is a new concept for a constellation of CubeSats that combines bi-spectral optical lightning imaging with radio frequency (RF) sensing to provide a 3-D lightning detection capability with global coverage from low-Earth Orbit. The development of CubeSpark is a collaboration between Los Alamos National Laboratory and NASA Marshall Space Flight Center. CubeSpark innovates over current ground and space-based global lightning capabilities by determining the altitude of lightning radiation sources, enabling new science in thunderstorm processes and the impact of lightning on climate. The key to determining the altitude of lightning is using a constellation of RF sensors to make coordinated measurements of impulsive RF radiation sources, similar to the approach of a ground-based lightning mapping array. The RF measurements will be enhanced with bi-spectral optical sensors to improve overall lightning detection efficiency and provide additional, complementary information about lightning processes.

This presentation introduces the CubeSpark mission concept and science applications with a focus on the RF hardware under development. Two challenges of space-based RF lightning detection are ionospheric effects and RF noise from both the satellite bus and anthropogenic sources from Earth. While the process of removing ionospheric dispersion from broadband waveforms for time-of-arrival (TOA) estimation is well established, CubeSpark further reduces ionospheric impacts on TOA by using a circularly polarized antenna, which suppresses one of the birefringent wave modes. For noise reduction, the CubeSpark receiver leverages programmable high- and low-pass filters to allow for on-orbit modifications of its passband. A benchtop demonstration of the RF hardware has been completed.

How to cite: Behnke, S., Katko, K., Edens, H., Gatlin, P., Lang, T., Haynes, W., Snow, P., Rushton, J., Renck, J., Weaver, C., Bronisz, L., Pratt, J., Dobson, S., Pailoor, N., Remington, J., and Stough, S.: CubeSpark: Space-based 3-D Lightning Mapping using a Constellation of Radio Frequency Sensors, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-12606,, 2024.

EGU24-12658 | Posters on site | NH1.5

GLM lightning flashes observed during ASIM triggers over Tropical South America 

Carlos Morales, Joan Montanyà, Jesus Lopéz, Oscar Van Der Velde, Nicolai Østgaard, Torsten Neubert, and Víctor Reglero

The Atmosphere-Space Interactions Monitor (ASIM) on board the International Space Station (ISS) is collecting data of lightning and Terrestrial Gamma Flashes (TGF) over the globe since April 2018 by means of two suites: i) modular multispectral imaging array (MMIA); and ii) modular X and gamma-ray sensors (MXGS). MMIA responds to lightning flashes, while high energy detector (HED) and low energy detector (LED) of MXGS are employed to estimate TGF spectra and source. Based on these features, ASIM is providing a large dataset of MMIA, LED and HED triggers that are used identify potential TGF events that require an extra imaging analysis to depict the exact location and validation. Upon such measurements, this study employs coincident ASIM and GLM lightning flashes over Tropical South America (90-30W and 20S-10N) to inspect if the electrically active thunderstorms present unequivocal features associated with each ASIM trigger, i.e., MMIA, LED, HED and TGF. Electrically active thunderstorms were identified as contiguous GLM lightning flashes clustered at 0. 1 x 0.1 degrees on ± 30 minutes of ASIM trigger time following Barnes et al. (2015) and Morales et al. (2021) procedures. During the period of 2018 and 2021, we were able to find 30,417 active thunderstorms that have lightning flashes within ± 3 seconds of trigger time (19,546 during the night and 10,871 during the day). Of those thunderstorms, 343 (1,745) were identified with HED, 278 (1,752) with LED, 12,858 (27811) with MMIA and 49 (116) with TGF within 0-200 ms (200ms-3 sec) of the trigger time. The spatial distribution of those thunderstorms do not show any lightning hot spot. MMIA thunderstorms coincide with the location of HED and LED thunderstorms, except HED thunderstorms over the Peruvian Andes mountain range. Moreover, we did not find any TGF thunderstorms along the mountain regions, especially in Peru and Ecuador. The 60 minutes lightning activity (# flashes/per minute) reveal that TGF thunderstorms show higher lightning flash rates than the MMIA, HED and LED triggered thunderstorms, in addition of a sudden lightning flash rate increase prior to the TGF trigger and sustained high lightning activity for the following 10 minutes. HED and LED show similar lightning temporal evolution (flash rate increase before the trigger and decay afterwards), but LED triggered thunderstorms have higher flash rates over the entire 60 minutes time period. MMIA triggered thunderstorms show the lowest flash rates and almost steady lightning activity during the entire 60 minutes. Based on 90% confidence level of T-Student test, we found that TGF and MMIA thunderstorms are statistical different during the entire 60 minute time period, meaning that not all MMIA thunderstorms produce TGFs. In another hand, we can state that HED and LED triggers are good indicators of TGF emissions, since they are not statistically different, meaning that these parameters could be used as triggers to identify TGF occurrences.

How to cite: Morales, C., Montanyà, J., Lopéz, J., Van Der Velde, O., Østgaard, N., Neubert, T., and Reglero, V.: GLM lightning flashes observed during ASIM triggers over Tropical South America, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-12658,, 2024.

EGU24-13074 | ECS | Posters on site | NH1.5 | Highlight

A Deep Learning Approach to Lightning Nowcasting and Forecasting 

Randall Jones, Joel Thornton, Dale Durran, Lyatt Jaeglé, Christopher Wright, and Robert Holzworth

Lightning plays a fundamental role in Earth’s climate system and is a frequently occurring natural hazard. However, lightning remains a relatively unpredictable area of meteorology, especially in terms of lightning frequency per convective event, with limited ability for nowcasting and forecasting of lightning occurrence. The goal of this study is to develop a deep learning algorithm able to replicate lightning stroke density on a climatological average, as well as on a convective feature basis. We use a convolutional neural network (CNN) containing combinations of the following variables at 0.5-degree by 0.5-degree spatial resolution and a 3-hourly temporal-resolution over a domain that encompasses most of the Western Hemisphere: lightning from the World-Wide Lightning Location Network (WWLLN), precipitation rate from NASA’s Integrated Multi-satellite Retrievals for GPM (IMERG) and convective available potential energy (CAPE), cloud base height (CBH), two-meter temperature (T2M) and zero degree level (ZDL) from the European Centre for Medium-Range Weather Forecasting (ECMWF). We train the CNN on the years from 2010 to 2018, and tested on the years 2019 to 2022. Model performance was evaluated on a four-year average through changes to the initial seed used to train the model, the loss function used, transformations to the lightning dataset, and changing the spatial and temporal resolution of the input datasets. We further examined the value of 11 input variable combinations, from single variables to all five variables used in training. Preliminary results show that changing the initial seed, as well as changing the loss function from mean squared error to mean-squared logarithmic error, does not greatly impact model performance when running the model with more than one input variable. Results vary amongst the variable combinations, but amongst the different initial seeds and loss functions, the r-squared values remain above 0.75 for every model configuration over both land and ocean. Model performance is improved when using higher time resolution training set but not necessarily a higher spatial resolution. For example, a 1-degree by 1-degree spatial resolution and a 3-hourly time resolution resulted in an r-squared between predicted and observed lightning frequency 0.1 higher than that using 0.5-degree by 0.5-degree spatial resolution and a daily time resolution. The model is able to reproduce the approximate evolution of lightning stroke density of individual convective events, but tends to overestimate the stroke density on a 3-hourly basis. Future work will include a steeper penalty for overestimating lightning occurrence during training. These results show that larger-scale weather forecasting and earth system models could significantly improve lightning stroke density parameterizations by incorporating deep learning results.

How to cite: Jones, R., Thornton, J., Durran, D., Jaeglé, L., Wright, C., and Holzworth, R.: A Deep Learning Approach to Lightning Nowcasting and Forecasting, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-13074,, 2024.

EGU24-13383 | ECS | Posters on site | NH1.5

Peak currents of terminating flashes in thunderstorm ground enhancements around Mt Aragats, Armenia 

Gayane Karapetyan, Earle Williams, Hripsime Mkrtchyan, and Reik V. Donner

Thunderstorm ground enhancements (TGEs) are high-energy particle fluxes detected at the ground level during thunderstorms. It has been observed that some TGEs experience abrupt termination by lightning strikes (Chilingarian 2015, Tsuchiya 2013, Williams et al., 2022) often accompanied by simultaneous reductions in flux. Understanding the origin and parameters of terminating lightning can provide insights into the distribution of electric fields and potential within thunderclouds. 

Thundercloud potential is a key factor in determining the maximum peak current of lightning. One expects a linear relationship between peak current and cloud potential because the charge that is deposited on the leader channel is proportional to the leader potential (e.g. Chronis et. al. 2015). 

This study evaluates peak currents in terminating flashes documented in TGEs observed around Mt Aragats (Armenia) using a ground-based VLF lightning detection network, GLD360. A total of 71 terminating flashes have been identified over a period of 6 years (2017-2022). The events documented at Aragats were detected by particle detectors that showed the abrupt decrease in flux associated with lightning. These events were accurately timed using an EFM100 electric field mill (resolution of 2Hz). Thereafter, correlations between these events and the corresponding GLD360 lightning events were established, using millisecond precision times of GLD360 and electric field mill.

Our findings show that the mean peak current of this collection of terminating flashes (45 kA) is 3.4 times higher than that of the general population of lightning flashes measured in the same location (13.6 kA) over a similar period of time. However, it is difficult to define the relationship between the change in electric field during TGE or lightning and the peak currents. It appears that lightning with smaller peak currents tends to have larger values of the change of electric field, while lightning with larger peak currents is characterized by an average change in the electric field.

This research provides insights into peak currents of terminating lightning flashes with general parameters of the TGEs, such as duration and flash rate. Additionally, it shows that flashes with extremely high peak currents occur during thunderstorms with smaller flash rates and are located within 10 km distance from the particle detectors. Furthermore, flash rates of thunderstorms with terminating lightning are larger compared to general thunderstorms without TGEs.

How to cite: Karapetyan, G., Williams, E., Mkrtchyan, H., and Donner, R. V.: Peak currents of terminating flashes in thunderstorm ground enhancements around Mt Aragats, Armenia, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-13383,, 2024.

EGU24-13989 | Posters on site | NH1.5

C3IEL, the Cluster for Cloud evolution ClImatE and Lightning mission to study convective clouds at high spatial and temporal resolutions 

Eric Defer, Celine Cornet, Daniel Rosenfeld, Cecile Cheymol, Adrien Deschamps, Alex Frid, Laurene Gillot, Vadim Holodovsky, Avner Kaidar, Raphael Peroni, Colin Price, Didier Ricard, Antoine Rimboud, Yoav Schechner, Aviad Shmaryahu, and Yoav Yair

The French-Israeli space-borne C3IEL (Cluster for Cloud evolution, ClImate and Lightning) mission aims at providing new insights on convective clouds, at high spatial and temporal resolutions, close to the scales of the individual convective eddies. The mission will simultaneously characterize the convective cloud dynamics, the interactions of clouds with