Orals: Thu, 7 May, 16:15–18:00 | Room 2.17
Extreme heat poses a major and growing risk to human health, yet accurately predicting its impacts on mortality remains challenging. In this study, we compared established nonlinear statistical models - including the epidemiological standard distributed lag non-linear model (DLNM) - with machine learning (ML) approaches both for predicting excess mortality and quantifying the heat-attributable mortality across Europe. We evaluated random forest regressions (RFs) and neural networks (NNs) trained on pooled European data, contrasting their performance with two-stage DLNMs and locally fitted generalized additive models. In each model, we included the lagged effect of temperature and additionally explored the inclusion of multiple environmental exposure variables (such as air pollution and humidity).
We assessed each model’s out-of-sample skill for predicting excess mortality, with our preliminary findings suggesting that the ML frameworks tend to improve skill across Europe. Notably, we find evidence that pooled ML models improve predictive performance for countries with fewer observations, suggesting that they are better able to learn from shared, diverse regional information. We also compared the spatial patterns and magnitudes of heat-attributable mortality estimated by the ML models with those from the DLNM, providing a benchmark. Together, our findings highlight the potential for ML-based frameworks to inform future heat-health impact assessments.
How to cite: Wilson Kemsley, S., Fromentin, J., Pastine, B., Dong, X., Guo, Y., Matthews, T., Meissner, K., Perkins-Kirkpatrick, S., and Slater, L.: Comparing machine learning and statistical models for quantification of heat-attributable mortality across Europe, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-5812, https://doi.org/10.5194/egusphere-egu26-5812, 2026.
Karachi, Pakistan’s largest coastal city with a population of 25 million, falls within the hottest zone in the world when the combination of heat and humidity (lethal heat) is considered. In 2015, Karachi underwent a fatal heatwave, resulting in 1300 deaths. While in 2024, Karachi suffered from another devastating spell of extreme heat, where the official number of fatalities stayed around 55. However, alternative evidence suggests that the actual number of deaths was far more than officially reported, strengthening the issue of ‘invisible deaths’ in developing countries as suggested in the earlier literature. This is a critical issue in efforts to address the impacts of climate change, considering that ‘you cannot cure the disease unless you know its intensity’.
We compared the weather conditions of 2015 and 2024 heatwaves at hourly temporal resolution based on a seminal physiological approach for assessing human livability to conduct sustained levels of work. Our results indicate that the lethal heat conditions, for multiple hours of each heat spell day, went beyond the levels where sustained basic activities for older adults (above 65 years) at a very light intensity, such as slow-based walking and house chores, were not possible. Considering that such fatal heat episodes are accompanied by disruptions in the power supply system, such conditions prove to be fatal for older adults. Similarly, the outdoor conditions also reached the levels for both the heatwaves for younger adults (18-40 years) where sustained livelihood generating activities such as lifting, fishing, street hocking, activities in agriculture and building sector etc were not possible for multiple hours of each day, putting serious limitations especially for the daily wagers responsible for putting the bread on the table. Our analysis further reveals that weather conditions during the 2024 heatwave were more severe than those during the 2015 heatwave, strengthening the findings of studies that suggest deaths during the 2024 heatwave were far more than the ones officially reported.
We further carried out impact attribution analysis to underscore the role of climate change in exacerbating the 2024 Karachi heatwave. Using station data and ERA-Land for observations, and the data of 10 CMIP6 climate models at sub-daily (6-hourly) temporal resolution, application of probabilistic attribution methods shows that climate change has a discernible role in amplifying the impacts of the 2024 Karachi heatwave based on physiological thresholds. Climate Change decreases the livability limit for older adults and young adults by approximately 0.3-0.5 MET (Maximum Metabolic Rate) in indoor and outdoor settings, respectively.
Our study presents a novel approach to advance the field of heat impact climate attribution. Our results are also useful for the policy makers, stakeholders, and implementers working in the fields of climate litigation, loss and damage, weather forecasting, and disaster risks, among others.
How to cite: Saeed, F., Ullah, A., Sadad, A., Saleh Khan, M., and Guerra, M.: From Lethal Heat to Invisible Deaths: Physiological Impact Attribution of Extreme Humid Heat in Karachi, Pakistan, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19826, https://doi.org/10.5194/egusphere-egu26-19826, 2026.
The impact of extreme heat stress on mortality has received growing attention in recent years. Historically, South America has been characterized by a relatively high number of days per year with combined extreme temperature and humidity posing a risk to human health. Future projections suggest that, under global warming, the continent will be one of the regions worldwide where humid-heat extremes are expected to intensify the most1. Temperature (T) and specific humidity (q) are key variables for determining heat stress, as human thermoregulation relies on heat dissipation through cutaneous vasodilation, sweating and evaporative cooling2. Wet-bulb temperature (WBT), which integrates the effects of T and q, has been widely used as a proxy to quantify human exposure to heat and the physiological capacity to cool down through sweat evaporation.
In water-limited conditions, soil moisture plays a critical role in the land surface energy partitioning, influencing sensible and latent heat fluxes, cloud cover, downward long-wave radiation and boundary layer height, thus modulating T, q and, ultimately, WBT3. Decreased soil moisture typically enhances T via sensible heat increases, while potentially reducing q by constraining latent heat. Hence, the overall impact of soil moisture on WBT—and potentially on heat-related mortality—is not straightforward and depends on the relative contributions of these competing processes.
Using daily mortality records for the 2000–2023 period, this study aims to unravel the dual effect of soil moisture on WBT extremes for several metropolitan regions in South America. Preliminary results show that the sensitivity of mortality to summer WBT is stronger in subtropical urban areas that are more water-limited, with the local influence of soil moisture varying significantly in nature and intensity. In São Paulo and Rio de Janeiro, high mortality rates linked to WBT extremes are mainly explained by increasing values of T associated with substantial reductions in the evaporative fraction and enhanced sensible heat flux. On the other hand, in Porto Alegre, the local impact of soil moisture manifests through exceptional values of both sensible and latent heat fluxes leading to WBT and mortality extremes leveraged by enhanced T and q. These results highlight the role of soil moisture as a key modulator of heat stress, shaping wet-bulb temperature extremes through competing thermodynamic pathways.
1. IPCC, 2023: Summary for Policymakers. In: Climate Change 2023: Synthesis Report. Contribution of Working Groups I, II and III to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change [Core Writing Team, H. Lee and J. Romero (eds.)]. IPCC, Geneva, Switzerland, pp. 1-34, doi: 10.59327/IPCC/AR6-9789291691647.001
2. Armstrong B., Sera F., Vicedo-Cabrera A. M., et al. (2019). The role of humidity in associations of high temperature with mortality: a multicountry, multicity study. Environ. Health Perspect. 127, 097007. https://doi.org/10.1289/EHP5430.
3. Chagnaud G., Taylor CM., Jackson, L. S., Birch, C. E., Marsham, J. H., & Klein, C. (2025). Wet-bulb temperature extremes locally amplified by wet soils. Geophysical Research Letters, 52, e2024GL112467. https://doi.org/10.1029/2024GL112467
How to cite: Geirinhas, J. L., Miralles, D. G., Hagan, D. F. T., Libonati, R., and dos Santos, D. M.: The influence of soil moisture on wet-bulb temperature extremes and excess mortality in South America, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10094, https://doi.org/10.5194/egusphere-egu26-10094, 2026.
Exposure to non-optimal temperatures poses significant risks to human health, yet evidence on mental health outcomes remains limited. This study examined associations between daily temperature exposure and hospital admissions for mental disorders in Hong Kong from 2006 to 2019. Daily hospitalization data, including both emergency and non-emergency admissions, were obtained from the Hong Kong Hospital Authority, encompassing all public hospitals across 18 districts. Generalized Additive Models (GAM) combined with Distributed Lag Non-linear Models (DLNM) were employed to investigate temperature effects. Results indicated a significant adverse effect of cold temperatures exclusively on persistent mental disorders due to other diseases (physical illness related), while a protective effect was observed for schizophrenia, mood disorders, other non-organic psychoses, and adjustment reactions. Notably, moderate-hot day exposure (27–32 °C) emerged as an important risk factor, particularly during prolonged heat events. Additionally, female patients demonstrated higher vulnerability compared to males. Our findings highlight differential effects of temperature exposure on mental health disorders, emphasizing the necessity for targeted interventions and adaptive strategies to mitigate adverse mental health impacts, particularly among females and during sustained moderate heat exposures.
How to cite: Huang, T.: Warm-Day Matters: Associations Between Non-Optimal Temperature Exposure and Mental Health Hospitalizations in Hong Kong, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11309, https://doi.org/10.5194/egusphere-egu26-11309, 2026.
Heatwave phenomena are associated with excess mortality. Their occurrence and frequency are projected to increase in future years. Therefore, it becomes urgent to develop tools and methods to better understand the association between heatwaves and their impacts on human health. This association relies greatly on the chosen definition of heatwave and the specific health condition under investigation.
Heatwave days are often defined using temperature-based quantile thresholds. However, other meteorological factors play a crucial role in shaping the physiological response of the human body during extreme heat. Multivariate model-based indicators such as the Universal Thermal Climate Index (UTCI) can be used to accurately estimate heat stress under specific hypotheses. Yet even these indices are frequently applied only to days pre‑selected by a temperature‑only criterion. This raises questions on the robustness of the results to changes in the heatwave definition, and it is unclear to what extent high heat stress days coincide with heatwave days.
The mismatch between extreme heat stress events according to UTCI and those identified by conventional definitions based on the simpler apparent temperature (AT) and two‑metre temperature (T2) is presented and discussed. The in-depth analysis has been based on available UTCI datasets (ERA5-Heat¹ and HiGTS²) for the period 2000-2023 over Europe and the Mediterranean region.
It has been found that around one quarter of the extreme UTCI days are not covered by extreme T2. In addition, from the study it emerges the presence of several strong heat stress hotspots in Southern Europe, e.g., in Spain and in the Po Valley. These have been identified as areas where each of the tested conventional heatwave definitions misses more than 10% of strong heat stress days. The use of AT in place of T2 mitigates this disagreement and may offer a low-cost alternative when UTCI is not available.
The days with the strongest physiological impact do not necessarily correspond to the hottest days in a season. This may impact current studies in the areas most affected by the disagreement, leading to an underestimation of the impacts. These findings also support the need to revisit extreme‑heat alerts, as well as the criteria that trigger financing mechanisms for heat-related losses.
¹ https://doi.org/10.1002/gdj3.102
² https://doi.org/10.1038/s41597-024-03966-x
How to cite: Bentivoglio, G., Ruggieri, P., and Di Sabatino, S.: Quantifying Missed Heat Stress by Temperature-Only Definitions over Europe, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11872, https://doi.org/10.5194/egusphere-egu26-11872, 2026.
Heat waves have emerged as a health hazard over Europe in recent decades with severe episodes of morbidity and mortality reported in the recent past during major heat waves. Several studies have investigated the temperature-mortality associations across Europe establishing the impact of rising temperature on increasing human health risk. In addition to these associations, the influence of changing characteristics (frequency, intensity, and duration) and dry and humid heat wave trends at different spatial and temporal scales would enhance the understanding of the rising risk to human health in the region. In the present study, the impact of intensity, frequency, and duration of heat wave events (dry and humid) was analyzed for Northern, Southern, Eastern and Western European regions using weekly all-cause mortality record from 976 contiguous NUTS 3 regions from 2001–2024 (obtained from EUROSTAT) and meteorological data comprising of daily mean temperature and relative humidity (derived from hourly ERA5 Land hourly dataset).
A two-stage modeling framework was employed: i) quasi-Poisson time series regression models used to estimate temperature-mortality associations for each region ii) mixed-effects meta-regression models were applied to derive pooled estimates of heat-related mortality across different regions of Europe, incorporating between-region heterogeneity heatwave effect across Europe. The results indicate the southern European region to be most affected by heat related mortality however, the inconsistency in the health data constraints adding limitations to derive robust spatio-temporal patterns based on the present long-term records. The study observed a consistent increase in heat attributed to deaths in the past decade which rises with increasing age and varies by gender reflecting rising vulnerability to extreme heat. The findings suggest the need for immediate targeted adaptation measures to protect the most at-risk populations and future risk associated with heat extremes.
How to cite: Singh, S., Plavcová, E., Lhotka, O., and Urban, A.: Evolution of Intense Heat Wave Hazard and Heat-Related Mortality in Europe, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15042, https://doi.org/10.5194/egusphere-egu26-15042, 2026.
Rising global temperatures increasingly expose populations to extreme heat, yet real-world evidence of how heat-humidity conditions affect cognitive function remains limited. Here, we build a unique data set using chess tournament outcomes as a proxy for cognitive performance. We analyse over 250,000 chess tournaments worldwide spanning 2003–2025 that include performance measures as specified by the international chess federations ELO rating system. By linking geolocated tournament results to climate reanalysis data, we quantify performance impacts across multiple heat stress metrics including daily maximum temperature, heat index, and wet-bulb globe temperature using panel-econometric methods that allow for causal interpretation. We focus on identifying heterogeneities of performance deviations by various dimensions such as player age, nationality, or skill. These heterogeneities reveal differential vulnerabilities that traditional laboratory studies cannot capture. Our results provide rare empirical evidence of heat's cognitive toll in naturalistic settings and establish a scalable framework for estimating productivity losses in service sectors, where cognitive work predominates but physiological heat thresholds applicable to manual labour are less relevant. As heatwaves intensify, understanding these cognitive impacts becomes critical for adaptation planning.
How to cite: Quante, L. and Stechemesser, A.: Heterogenous effects of Heat-Humidity Events on cognitive performance, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-20590, https://doi.org/10.5194/egusphere-egu26-20590, 2026.
Faiths play a central role in the lives of billions of people across the globe. Many religious rites and celebrations are performed at fixed times and locations. With the ongoing rise in global temperatures, climate change is now directly affecting how these faith-based activities are carried out. In particular, extreme heat and humidity are making large religious gatherings increasingly difficult and risky.
Muslim Pilgrimage (Hajj) to Makkah is one of the five pillars of Islam and is mandatory for every Muslim who is physically and financially able to perform it at least once in their lifetime. The pilgrimage is one of the largest religious gatherings in the world, bringing millions of pilgrims from across the globe to Makkah each year, making it one of the largest recurring human gatherings globally. Makkah is a hot region and already faces significant heat-related challenges. Previous studies suggest that in a 2.0 °C warmer world, the risk of heat stroke could increase by up to ten times, whereas limiting warming to 1.5 °C could reduce this increase to approximately five times.
Pilgrimage is an intensive five-day event involving physically demanding activities, including circling the Kaaba (Tawaf) multiple times, walking between the hills of Safa and Marwa, standing in prayer at Mount Arafat, spending nights in Mina and Muzdalifah, and stoning the pillars. During the 2024 Hajj, approximately 1,300 fatalities were reported amid extreme humid heat conditions. The Government of the Kingdom of Saudi Arabia plans to increase the number of pilgrims in the future, raising serious concerns about increased exposure to extreme heat and humidity.
In this study, we analyzed the 2024 pilgrimage in terms of human physiological limits using temperature and humidity sub daily station-based data. Our results show that survivability limits were exceeded during several hours on each day of the pilgrimage even for the younger adult group (18-40). Although the Pilgrimage will occur during relatively cooler seasons over the next 20–30 years, it is expected to shift back to hotter periods by around 2050. We therefore further utilized sub-daily CORDEX climate model outputs to investigate survivability-limit exceedances during future June pilgrimages. The results indicate that survivability limits will be breached more frequently and rapidly in the future, highlighting an urgent need for adaptation measures and, critically, mitigation efforts to reduce climate-change-related risks to pilgrims.
While adaptation strategies by the Government of Saudi Arabia may reduce some risks, the essence and traditional practice of the pilgrimage could still be compromised under extreme heat conditions. Therefore, mitigation remains essential to limit global warming and safeguard the future of the Pilgrimage.
How to cite: Ullah, A., Sadad, A., Saleh Khan, M., and Saeed, F.: When Faith Meets Heat: Climate Change Risks During the Hajj Pilgrimage, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21325, https://doi.org/10.5194/egusphere-egu26-21325, 2026.
While urban humid heat is a major concern, adaptation strategies often overlook the surrounding rural land management. How rural land cover changes modulate urban humid heat by altering regional thermal-humidity dynamics, and whether these effects differ across climates, remains unclear. Here, we conduct regional‑scale simulations of summer heatwaves under three rural land cover scenarios—bare land, grassland, and forest—for three Chinese cities (humid Guangzhou, semi‑humid Beijing, and arid Lanzhou). We analyze changes in near-surface wet-bulb temperature (TW) and decompose them into temperature- and humidity-driven components. We find that the impact of rural land cover on urban humid heat shifts with background hydroclimate. Counterintuitively, expanding rural bare land amplifies urban TW in humid climates (+0.19 ℃) but reduces it in semi-humid and arid climates (-0.36 ℃ and -0.43 ℃). This contrast is driven by water availability: moisture-abundant humid climates can satisfy the increased evaporative demand from enhanced rural sensible heating, adding humidity and reinforcing TW; whereas water-limited climates cannot, generating a strong drying effect that outweighs warming. Rural greening yields divergent outcomes. Conversion of rural land to high‑evapotranspiration grassland intensifies humid heat stress, particularly in arid climates, where the cooling effect is largely outweighed by pronounced humidification. In contrast, rural afforestation mitigates humid‑heat stress in the humid and semi‑humid climates through a drying effect driven by physiological water retention that suppresses evapotranspiration, while offering little benefit and slightly increasing TW in the arid city. Our results establish rural hydroclimate as a critical factor in urban humid heat adaptation, demanding climate-specific strategies that account for the trade-off between thermal cooling and humidity accumulation.
How to cite: Xu, Y., Du, Y., Hang, J., Liao, J., and Luo, Z.: Rural land cover management reverses urban humid heat effects across climates, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15756, https://doi.org/10.5194/egusphere-egu26-15756, 2026.
Weather and climate extremes such as extreme heat events are crucial, and increasingly lethal, climate hazards to people and communities worldwide. In any region, climate change may alter the characteristics of such events in complex ways so that a rigorous and holistic quantification of their extremity remains a challenge. This impedes hazard data users concerned with impact, attribution and litigation and likewise research, policy and practice users in the field of human health who are involved with reducing vulnerability and inequality, improving early warning systems and strengthening adaptation and resilience in severely-stressed regions.
Here we use a new holistic class of threshold-exceedance-amount metrics to globally track the extremity and amplification of extreme heat and human heat stress over land regions worldwide. We recently introduced these “TEA metrics” as a rigorous methodology and demonstrated their use through tracking heat amplification over Europe since the 1960s, revealing an over ten-fold increase of extreme heat over more than half of continental Europe (Kirchengast et al., Weather Clim. Extremes, https://doi.org/10.1016/j.wace.2026.100855). The metrics consistently track changes in event frequency, duration, magnitude, area, and timing aspects like daily exposure and seasonal shift—as separate metrics, partially compound like as average event severity in a region, and as total events extremity.
For the worldwide tracking on land regions and per country we use daily maximum temperature as key variable for extreme heat (key thresholds TX99p, TX30, TX35) and the daily maximum universal thermal climate index (UTCI) for human thermal stress (thresholds TXutci99p local-region’s 99th percentile in 1961-1990, TXutci38 very strong heat stress, TXutci46 extreme heat stress). State-of-the-art datasets are used over 1961 to 2025 (reanalyses ERA5, ERA5-HEAT; 0.25° x 0.25° grid) and core metrics results are provided online at the ClimateTracer.Earth data portal (“Extremes”; https://climatetracer.earth/ewm). Comparing the recent period 2011-2025 to the reference climate period 1961-1990, we reveal the most severely affected hot spots of heat stress, showing over thirty-fold amplification of total events extremity, further exacerbated if we also account for daily exposure time (using hourly key variable input data).
We discuss these results, and the prospective use of CMIP6 climate model data for extending the records through Shared-Socioeconomic-Pathways-based scenarios to 2100, and in particular discuss their significance and utility for downstream uses by scientific and practice users in the human health sector.
How to cite: Kirchengast, G., Fuchsberger, J., Haas, S. J., and Pichler, M.: Revealing lethal spots: global tracking of extreme heat and human thermal stress using a new holistic class of climate hazard metrics, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13617, https://doi.org/10.5194/egusphere-egu26-13617, 2026.
Introduction: Heat exposure poses an increasing threat to human health, particularly in African low- and middle-income countries, where rapid urbanization, limited adaptation infrastructure, and climate change vulnerability merge. However, fine-scale meteorological data in this region are scarce, limiting heat exposure assessments. Moreover, compound humid heat remains largely unexplored in this context. This study aims to assess humid heat exposure in Basse Santa Su, The Gambia, a region highly vulnerable to humid heat, by generating spatial predictions informed by high-resolution microclimate measurements.
Methods: Over one year, a fixed network of low-cost measurement devices mounted across Basse Santa Su (approximately 11km2) collects time-resolved meteorological parameters (temperature, humidity, solar radiation, atmospheric pressure, wind speed and direction). Multilinear land-use regression (LUR) models will estimate spatial and temporal patterns of heat, humidity, and heat-stress distribution across the study area. Model predictors will include climate variables from ERA5-Land reanalysis and global high-resolution remote sensing data on relevant characteristics such as land-use, vegetation, topography and urban surface geometry.
Results: In 2025, the fixed measurement network mounted at 12 locations in Basse Santa Su recorded an average daily temperature of 29.6ºC with 38.6% relative humidity (RH) in the dry season (November–May), and an average daily temperature of 28.8ºC with 78.6% RH in the rainy season (June–October). The predictive models will estimate high-resolution, daily and hourly single weather variables (ambient temperature and relative humidity) and combined heat stress indices (e.g. Wet Bulb Globe Temperature (WBGT), physiological equivalent temperature (PET)) over the whole study area. These maps will assess the spatial and temporal variability in humid heat and identify high-risk neighbourhoods, contextual variables, and periods or seasons associated with higher or lower exposure.
Conclusion: This study evaluates the feasibility of combining a low-cost microclimate measurement network with a land-use regression modeling approach to characterize fine-scale spatial variability in temperature, humidity, and heat stress in a data-scarce, extreme climate setting. The resulting high-resolution humid heat exposure estimates will provide a critical foundation for further applications, such as heat-health impact assessments and targeted adaptation strategies and interventions in Basse Santa Su and comparable settings.
How to cite: Tadiri, E., Saucy, A., Bonell, A., Burger, M., Gubler, M., and Vicedo-Cabrera, A. M.: Microclimate analysis in Basse Santa Su, The Gambia: modeling temperature, humidity, and heat stress in an extreme climate, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21517, https://doi.org/10.5194/egusphere-egu26-21517, 2026.
Anthropogenic warming has significantly exacerbated heatwaves (HWs) globally, posing severe threats to public health. In light of the insufficiency of using solely ambient temperature to assess human heat stress, previous studies identified human-perceived HWs (HPHWs) by considering the synergistic effects of temperature and humidity. However, the limited attention given to the influence of local antecedent heat conditions and human acclimatization hampers the comprehensive evaluation of HPHW changes. Through a systematic comparison of three HPHW definitions, this study employs the Excess Heat Factor (EHF) to examine the long-term spatiotemporal variations in HPHWs across three seasons (excluding winter) in China, as well as the associated extreme heat exposure. The results show that most HPHW metrics exhibit opposite directional changes between the periods of 1961−1984 and 1985−2022. Regionally averaged, South and Southwest China experience more substantial rises in HPHW occurrence, duration and frequency. The most pronounced intensification of HPHW events is found in Northeast China, and the onset of the first yearly HPHW advances most significantly in North China. At both national and sub-regional scales, the population-weighted HPHW frequency increases at a faster rate than its area-weighted counterpart, indicating the disproportionate effect of HPHW occurrence on populated areas. Jianghuai and South China generally undergo the most notable increases in both mean and maximum population/land area affected by extreme heat. Our findings contribute to a better understanding of HPHW changes across China and highlight the urgent need for adaptation strategies to mitigate escalating dangers of heat stress in a warming climate.
How to cite: Chen, X., Huang, C., Fan, H., Liu, Y., and Jiang, D.: Escalating population and land exposure to human-perceived heatwaves in China under a warming climate, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7335, https://doi.org/10.5194/egusphere-egu26-7335, 2026.
Extreme heat has a complex and delayed effect on human mortality operating across sub-daily to weekly timescales. Many large-scale mortality datasets are reported at weekly resolution, with stratified age brackets. However, temporal aggregation in prediction can obscure short-lived lethal heat episodes and lead to underestimation of heat-related mortality. Methods for estimating temperature–mortality relationships from temporally aggregated data have been explored within statistical frameworks, which remain the standard approach in environmental epidemiology, but these approaches constrain the form of the risk function and limit the flexibility of predictor representations.
We propose a machine-learning framework that enables daily mortality prediction with no restriction on the temporal resolution of the training mortality dataset. The method learns a heat-related risk function conditioned on lagged sequences of recent daily meteorological conditions and regional socio-environmental characteristics. Weekly expected deaths are decomposed into daily estimates which are multiplied with the learned risk factors to get the model’s daily predicted deaths. The daily death predictions for a week are summed to a weekly total to match the available temporal resolution of the death dataset. The gradients of the learned risk factor propagate through this aggregation step, allowing the model to learn temporally resolved mortality responses without requiring daily death labels.
The framework is trained using weekly NUTS-3 Eurostat mortality data with five-year age stratification, together with high-resolution MSWX daily reanalysis meteorology. Validation is performed using a French 2019 individual-level daily mortality dataset, which reports spatial, age, and sex information for all registered deaths in France, enabling direct evaluation of predicted daily deaths aggregated to consistent spatial and age resolutions.
We expect this approach to recover intra-week variability in mortality associated with short-duration temperature signals, outperform uniform or heuristic temporal disaggregation methods, and improve attribution of lethal heat events. By linking daily climate exposure to weekly mortality records without requiring more fine-grained data collection, this method expands the analytical value of existing mortality datasets and supports more timely assessment of lethal heat risk.
How to cite: Fromentin, J. L., Wilson Kemsley, S., Dong, X., and Slater, L.: Inferring daily lethal heat mortality from weekly death records using machine learning, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7881, https://doi.org/10.5194/egusphere-egu26-7881, 2026.
Humid-heat stress is rising rapidly across the Arabian Peninsula (AP), where sea-surface temperatures (SSTs) strongly modulate both the magnitude and spatial expression of extreme humid-heat stress. Although local SSTs in adjacent basins are known to intensify boundary-layer moisture and elevate coastal humid-heat, the degree to which SST anomalies—both locally and remotely forced—independently influence temperature and humidity remains poorly understood. In particular, it is not yet understood how large-scale teleconnections such as the El Niño–Southern Oscillation (ENSO) and the Indian Ocean Dipole (IOD) shape the occurrence and severity of humid-heat extremes in the AP, nor how these modes interact with local SST forcing. Here, we quantify how local and teleconnected SST anomalies independently and jointly influence present and future humid-heat extremes, characterized by wet-bulb temperature, heat index, and the humidity–temperature partitioning metric stickiness, across five major AP coastal cities (Doha, Dubai, Jeddah, Aden, and Muscat). We employ a hierarchical Bayesian peak-over-threshold (POT) framework using a generalized Pareto likelihood applied to the 95th-percentile threshold of daily humid-heat metrics. This structure enables us to:
- isolate the sensitivity of extreme humid-heat to ENSO and IOD phases;
- assess whether ENSO–IOD combinations amplify or dampen AP humid-heat risk; and
- separate humidity-driven vs. temperature-driven contributions to extremes.
After establishing the observed relationships, we perturb the model with scenarios of increased local SSTs (+1°C, +2°C, +3°C, +4°C) in each adjacent basin to evaluate how direct ocean warming may alter extreme humid-heat distributions in coming decades. These experiments provide a mechanistic basis for attributing humid-heat amplification to specific SST pathways and for estimating the compound impacts of global teleconnections and regional warming on future coastal risk. Expected findings include (i) strong city-specific variability in ENSO and IOD influence, (ii) robust humidity-driven amplification under positive ENSO/IOD phases, and (iii) nonlinear increases in extreme humid-heat under uniform local SST warming. Together, these results establish a unified Bayesian framework for attributing and projecting SST-driven humid-heat risk across one of the world’s fastest-warming coastal regions.
How to cite: Bose, D., Tuholske, C., Raymond, C., Nazemi, N., and Cowherd, M.: Local and Remote Sea-Surface Temperature Forcing of Extreme Humid-Heat in the Coastal Arabian Peninsula, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8134, https://doi.org/10.5194/egusphere-egu26-8134, 2026.
Posters virtual: Mon, 4 May, 14:00–18:00 | vPoster spot A
EGU26-10525 | ECS | Posters virtual | VPS31
Clocking the Heat: Projected Diurnal Patterns of Thermal Discomfort Across Saudi Arabia Under Future Climate ScenariosMon, 04 May, 15:03–15:06 (CEST) vPoster spot A
Rising extreme temperatures driven by climate change are expected to significantly degrade outdoor thermal conditions, stretching the day of extreme heat and leaving fewer hours for comfortable and safe outdoor activity, while increasing the health risks associated with outdoor exposure. This study investigates the impact of climate change on thermal discomfort across the Kingdom of Saudi Arabia. Projections from two CMIP6 models, at 6-hour temporal resolution, were used to compute the Discomfort Index (DI) based on dry-bulb temperature and relative humidity, and to assess diurnal variations in thermal stress at 03 UTC (06:00 AST), 09 UTC (12:00 AST), 15 UTC (18:00 AST), and 21 UTC (00:00 AST) under SSP1-2.6, SSP2-4.5, and SSP5-8.5 scenarios. Changes were evaluated for the near (2021–2040), mid (2041–2060), and far future (2081–2100). Thermal discomfort across Saudi Arabia intensifies progressively from the historical period to the far future, exhibiting pronounced spatial and diurnal variability. Historically, daytime discomfort (09–15 UTC) had a mean DI value of ~25.4 °C, corresponding to uncomfortable conditions across most regions, with some areas, particularly in the southeast and coastal regions, reaching very uncomfortable conditions. Early morning and evening hours (03–21 UTC) were slightly lower, with mean DI values around 22.8–23.4 °C, corresponding to slightly uncomfortable conditions. Future projections indicate a substantial increase in discomfort magnitude, particularly in coastal and southeastern areas. In the near-future (2021–2040), mean DI values increase to ~25–26 °C during daytime and ~23 °C during early morning and evening hours. By the mid-future (2041–2060), at 09 UTC (12:00 AST), the southeast and coastal regions are very uncomfortable and can reach extremely uncomfortable conditions under SSP5-8.5, reflecting peak thermal stress during the day. In the far-future period (2081–2100), at 09 UTC (12:00 AST), mean DI values reach ~27–28.6 °C under SSP2-4.5 and SSP5-8.5 scenarios, with maximum values exceeding 32 °C in the southeast region under SSP5-8.5, corresponding to dangerous conditions, highlighting the severity of midday thermal stress and its potential impacts on outdoor activities and urban livability. Evening and early morning mean DI values also rise substantially compared to historical conditions, reaching ~25–27 °C (uncomfortable), with some regions, particularly in the southeast, reaching up to ~30.5 °C (extremely uncomfortable), indicating that nighttime relief is markedly reduced and thermal discomfort persists even outside peak daytime hours. These findings emphasize the necessity of adaptive strategies to ensure the resilience, safety, and comfort of outdoor environments under increasing heat stress.
How to cite: Abuwaer, N., Vinodhkumar, B., and Al-Ghamdi, S.: Clocking the Heat: Projected Diurnal Patterns of Thermal Discomfort Across Saudi Arabia Under Future Climate Scenarios, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10525, https://doi.org/10.5194/egusphere-egu26-10525, 2026.
