The objectives of the experiment were the following:

• Evaluation of continuous ground motion from satellite data (European Ground Motion Service for the period 2019-2023 and IRIDE for 2024);
• Analysis of different types of landslides (active landslides, dormant landslides, landslide-prone areas, subsidence);
• Identification of elements for the Emergency Limit Condition (CLE) analysis near areas affected by specific ground motions derived from satellite data in the period 2019-2024
• Identification of buildings in the municipality of Perugia near areas affected by specific ground motions derived from satellite data in the period 2019-2024
• Submission of the results to all regional offices that authorize, evaluate, design, or schedule interventions on the territory and to the regional civil protection agency.

QGIS version 3.42 software was used for the experiment.

The following databases were imported into QGIS:

• European Ground Motion Service satellite data.
• IRIDE satellite data – Cross Monitoring of Ground Motion and "Hot Spots" of Cover Change.
• PAI geomorphological landslide hazard maps.
• Local Seismic Hazard Map of the Umbria Region.
• Umbria Region Emergency Limit Condition Analysis (CLE) maps.
• Geological Map of the Umbria Region.
• Building database of the Umbria Region's land registry system for the Municipality of Perugia.
• Administrative Boundaries of the Umbria Region and base maps such as the Regional Technical Map and Google Satellite.

Spatial analyses were performed using the GIS on the collected data to homogenize and select specific information useful for subsequent processing.

Multiple analyses werw performed for 2 specific case studies.

All objectives were achieved: assessment of continuous ground motion from satellite data (two different databases: IRIDE for 2024 and EGMS 2019-2023); subsequent analysis using different types of landslides (active landslides, dormant landslides, landslide-prone areas, subsidence); subsequent assessment using the CLE (Emergency Limit Condition) elements; subsequent assessment using buildings in the Municipality of Perugia.

How to cite: Motti, A. and Natali, N.: Experimentation with the use of EGMS and IRIDE satellite data, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3065, https://doi.org/10.5194/egusphere-egu26-3065, 2026.

Landslides have become one of the most destructive geological hazards in the Himalayan region, exhibiting a significant increase in both occurrence and intensity in recent decades. This increasing trend poses serious threats to human life, infrastructure, and essential public assets, underscoring the need for comprehensive risk evaluation in these highly vulnerable mountainous terrains. The present study offers an extensive assessment of landslide hazard, vulnerability, and associated risk in the Darma Valley of the Kumaun Himalaya, India. Landslide susceptibility was modelled using a Multilayer Perceptron (MLP) neural network, and the model’s predictive performance was validated through ROC–AUC analysis. Vulnerability was quantified by integrating land-use/land-cover categories with their respective economic valuations. Furthermore, rainfall and seismic intensity maps were combined with the susceptibility outputs to derive a detailed landslide hazard map. The results indicate that roads are the most vulnerable elements, followed by settlements and dam infrastructures, largely due to their substantial reconstruction costs and higher exposure levels. The final risk map, produced by integrating hazard and vulnerability layers, reveals that approximately 9% of the study area falls within high to very high risk zones, 22% within moderate risk, 26% within low risk, and 43% within very low risk zones. These findings offer essential guidance for promoting sustainable development and supporting land-use planning that accounts for environmental risks. They also contribute to more informed and effective decision-making aimed at strengthening the resilience of the fragile and sensitive Himalayan landscape.

How to cite: Shawez, M., Kumar, S., Gupta, V., Kumar, P., and Rawat, G.: Landslide Hazard, Vulnerability, and Risk Analysis (HVRA) Using Machine Learning and AI: A Case Study of the Darma Valley, Kumaun Himalaya, India, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18022, https://doi.org/10.5194/egusphere-egu26-18022, 2026.

Slow-moving, deep-seated landslides represent a significantly underestimated geologic hazard, incurring huge economic loss and persistent long-term risk to communities annually. Further, they have the potential to evolve into catastrophic events, which necessitates continuous monitoring to better understand their dynamics, minimize potential losses, and implement appropriate mitigation measures. The present study aims at understanding the dynamics of the slow-moving slopes housing villages such as Bhatwari, Raithal, and Barsu in the Bhagirathi Valley, Uttarakhand Himalaya, by means of PS-InSAR techniques. A total of 129 ascending-pass and 114 descending-pass scenes of Sentinel-1, from January-2021 up to March-2025, have been utilized to estimate slope velocities along the radar line-of-sight (LOS) for each pass, using open-source tools such as ISCE and StaMPS.  Further, these LOS velocities were decomposed to obtain vertical (up-down) and horizontal (east-west) velocities. The results reveal that Raithal (elevation ~2150 m), on middle of the slope, is subsiding at ~3 mm/year with an eastward movement of ~5 mm/year. Bhatwari (1650 m), on the lower slope, shows eastward creep at ~4 mm/year and upliftment at ~2 mm/year, suggesting rotational landslide activity. Barsu (2262 m), situated at a slope ~3 km upstream, exhibits eastward movement at ~6 mm/year and subsidence at ~3 mm/year. Field investigations corroborate these findings, revealing features such as scarps, cracks, tilted structures, disrupted roads, and longitudinal and transverse ponds. The persistent creeping suggests the potential for sudden slope failure during heavy rainfall or earthquakes, which may dam the Bhagirathi River, and the impoundment may further trigger cascading downstream hazards. Therefore, there is a need for a comprehensive investigation integrating the PS results with the slope stability analysis that assesses the role of geology, rainfall, and earthquakes. This integration shall assist in estimating the risk posed by the failure and further help in mitigation planning.

How to cite: Gupta, A. K., Luirei, K., Gupta, V., and Shawez, M.: PS-InSAR based Slope Deformation Monitoring in the Bhagirathi Valley, Uttarakhand Himalaya, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18221, https://doi.org/10.5194/egusphere-egu26-18221, 2026.

Chișinău, the capital of the Republic of Moldova, faces significant geohazard challenges due to its unique geological setting on loess-covered plateaus dissected by river valleys and ravines. Urban expansion and infrastructure development have intensified landslide susceptibility in this region, threatening residential areas, transportation networks, and critical infrastructure. This study presents a comprehensive analysis of urban landslides in Chișinău using Persistent Scatterer Interferometric Synthetic Aperture Radar (PS-InSAR) technique applied to Sentinel-1 satellite data spanning the last five years (2019-2025).

The PS-InSAR methodology provides millimeter-level precision in detecting and monitoring ground deformation over time, making it particularly suitable for identifying slow-moving landslides and ground subsidence in urban environments. We processed ascending and descending Sentinel-1 SAR imagery to generate time-series deformation maps and identify persistent scatterers across the Chișinău metropolitan area. The analysis revealed multiple zones of significant ground displacement, with deformation rates ranging from -15 to +25 mm/year, concentrated primarily in areas with steep terrain, proximity to water courses, and urban development on historically unstable slopes.

The susceptibility map derived from our analysis indicates high-risk zones in the northern and western sectors of Chișinău, particularly around suburb localities Vatra, Ghidighici, and Durlești, where loesslike deposits on valley slopes are subjected to both natural erosion processes and anthropogenic pressures. The southeastern areas near locality Bubuieci also show elevated landslide susceptibility, correlating with urban expansion into previously undeveloped terrain. Integration of PS-InSAR results with geological maps, digital elevation models, and land-use data enabled the development of a comprehensive landslide susceptibility assessment framework.

Key findings reveal that ground deformation patterns in Chișinău exhibit strong seasonal variations, with accelerated movement during spring months corresponding to snowmelt and precipitation events. Urban infrastructure, including roads, buildings, and utilities, located within identified high-risk zones, shows structural damage consistent with slow-moving landslide activity. The study identifies critical infrastructure corridors, including major transportation routes (E583, E581) traversing the study area, that require enhanced monitoring and mitigation measures.

Acknowledgements: This work was supported by a grant of the Ministry of Research, Innovation and Digitization, CNCS – UEFISCDI, project number 40PCBROMD within PNCDI IV.

How to cite: Sandric, I., Nicoara, I., Spian, C., Tambur, A., Ilinca, V., Jeleapov, V., Irimia, R., Daia-Creinicean, T., and Alexandru, N.: Urban Landslide Monitoring Using PS-InSAR Sentinel-1 Data in Chișinău, Republic of Moldova (2019-2025), EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19054, https://doi.org/10.5194/egusphere-egu26-19054, 2026.

Land subsidence poses growing risks to urban infrastructure, water resources, and long-term resilience, requiring assessment frameworks that link present-day observations with planning-relevant forecasts. This study develops an integrated approach for land subsidence susceptibility mapping and trend forecasting over multi-year horizons. The analysis uses SBAS-InSAR deformation time series derived from Sentinel-1 observations from 2017 to 2025 to characterize subsidence patterns across East Baton Rouge Parish, Louisiana. Subsidence susceptibility is modeled using an ensemble machine-learning framework that combines Extra Trees and Random Forest regressors and incorporates geological, topographic, hydrological, land use, infrastructure, and climatic conditioning factors. The susceptibility results highlight the dominant influence of land use, elevation, proximity to faults and rivers, and terrain-hydrology interactions on subsidence patterns. To extend assessment beyond observation periods, a physics-informed long short-term memory (LSTM) ensemble is introduced for forecasting. The model integrates data-driven learning with physically motivated constraints to ensure stable and realistic deformation trajectories. The forecasts preserve observed spatial patterns while exhibiting physically consistent temporal evolution and quantified uncertainty. The results demonstrate that combining InSAR observations with physics-informed deep learning enables robust, planning-scale subsidence assessment and forecasting. The proposed framework is transferable to other urban settings where long-term subsidence poses increasing societal risk.

How to cite: Kangah, D. and Abdalla, A.: Near-Decadal Land Subsidence Susceptibility and Trends Using Physics-Informed LSTM, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16451, https://doi.org/10.5194/egusphere-egu26-16451, 2026.

Remote sensing enables spatially continuous and timely monitoring of hydro-climatological extremes by capturing key land–atmosphere variables across large regions, including for data-scarce areas. The rising frequency of heat extremes across India in recent decades underscores the need for effective monitoring, especially in data-scarce regions. This study evaluates the potential of monitoring heat extremes over the Indian sub-continent using satellite based observations and data driven approaches. For this, MODIS land surface temperature (LST) along with NDVI, land use/land cover and elevation information is used with traditional machine learning models namely Random Forest (RF) and XGBoost. Subsequently, the performance of the two ML models in estimating maximum temperatures across the Indian subcontinent was evaluated and validated using in situ temperature observations from the Indian Meteorological Department. Heat extremes were identified using both absolute temperature percentile thresholds and Standardized Temperature Index based heat stress categories. The performance of ML models was evaluated using station‑wise categorical verification metrics such as hit rate, false alarm ratio, and critical success index. Results show that the ML models exhibit higher accuracy in predicting mean temperatures compared to extremes, and XGBoost outperforms the RF model with lower RMSE and higher R². The results further reveals that ML model prediction skill exhibits considerable geographic variability across the sub-continent, with reduced performance over mountainous areas. This study demonstrates that integrating satellite-based data with machine learning provides an effective approach for monitoring heat extremes across the Indian subcontinent, particularly in data-scarce environments.

How to cite: Remesh Ancy, A. and Mitra, S.: Monitoring Heat Extremes over India Using Earth Observations and Data Driven Approaches, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16372, https://doi.org/10.5194/egusphere-egu26-16372, 2026.

A remote sensing index is often used to identify meteorological and agricultural droughts. Google Earth Engine analyzes CHIRPS data from 2015 to 2024 and Landsat-8/Sentinel-2 data from 2020 to 2024. The Vegetation Condition Index (VCI), Temperature Condition Index (TCI), composite Vegetation Health Index (VHI), and Standardized Precipitation Index (SPI) were calculated for four seasons using NDVI, EVI, LST, and CHIRPS precipitation data to explain specific spatiotemporal trends. Meteorological and agricultural droughts include precipitation deficits and vegetation stress. From the study, pre-monsoon analysis reveals significant intra-seasonal correlations between VCI and VHI (0.84) and TCI and VHI (0.75), indicating that moisture reserves and thermal stress influence vegetation health during arid periods. The VCI-VHI correlation (0.91) predominates during the monsoon season, indicating plant growth amidst substantial precipitation. As the season nears peak aridity, the correlations between post-monsoon and winter TCI-VHI increase (0.81 and 0.83), signifying thermal stress. A weak correlation (≤ 0.50) between SPI and vegetation indices across the seasons indicates that current precipitation does not succeed in reliably predicting vegetation stress, since vegetation depends on accumulated soil moisture rather than instantaneous rainfall. Vegetation indices exhibit substantial temporal persistence: Pre-monsoon VCI conditions are strong predictors of winter VCI (0.98), VHI forecasts winter VHI (0.92), and TCI predicts winter TCI (0.87), thereby enabling nine-month drought forecasting. The findings demonstrate that vegetation indices serve as drought indicators for seasonal water resource planning and agricultural vulnerability assessment in monsoon-affected nations.

How to cite: Thounaojam, L. and Oinam, B.: Spatio-temporal dynamics of meteorological and agricultural droughts: A multi-seasonal analysis of Vegetation Health and Climate Indices Using Google Earth Engine, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2205, https://doi.org/10.5194/egusphere-egu26-2205, 2026.

The Himalayan river Basins frequently experience devastating floods. These river basins require accurate predictions and timely warnings to support effective flood risk management. While accurate prediction is crucial for saving lives, disaster managers often face a difficult trade-off between computational cost and warning lead time. High-fidelity physics-based models are precise but are computationally expensive for rapid decision-making, whereas low-fidelity geo-spatial models often lack accuracy in data-scarce regions. Our proposal is a framework to improve the flood inundation prediction in the Himalayan basin by combining the reliability of hydrodynamic modeling with the speed of machine learning.

In this study, we developed a 2D HEC-RAS model using a Rain-on-Grid approach to simulate the historical floods. We utilize the developed hydrodynamic model to generate a dataset of flood inundations that captures the basin's flow dynamics. These datasets will serve as the foundation for training advanced machine learning algorithms, including a Random Forest Regressor (RF) and a Convolutional Neural Network (CNN), to identify and predict flood patterns. Our model will integrate critical landscape features, including elevation, slope, land-use characteristics, the Normalized Difference Vegetation Index (NDVI), and satellite-derived rainfall data, to approximate the complex physical processes embedded in the hydrodynamic model. This allows the machine learning approach to achieve comparable predictive accuracy while reducing computational time. Through comprehensive validation against established benchmarks and real-world flood events, our research aims to deliver a scalable, computationally efficient, and highly accurate flood prediction tool. This framework has the potential to transform disaster preparedness and response capabilities in the Himalayan region by enabling timely, data-driven policy planning and proactive risk mitigation strategies.

How to cite: Duwal, S., Pradhan, P. M., Liu, D., and Bhattarai, Y.: Coupling Hydrodynamic Modeling with Machine Learning for Flood Risk Assessment in the Himalayan River Basin, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7810, https://doi.org/10.5194/egusphere-egu26-7810, 2026.

Earthquake clusters can be broadly classified into two types: swarm-like sequences and mainshock–aftershock sequences. The spatial organization of the two types provides important insights into underlying tectonic processes and fluid migration in earthquake source regions. In this study, we apply the nearest-neighbor distance approach on the Southern California focal-mechanism earthquake catalog (the CNN_SoCal catalog) and introduce two new statistical indicators-skewness and kurtosis to distinguish between these two classes of earthquake clusters. We find that the square root of kurtosis and skewness provide effective and interpretable indicators for clusters classification. In the kurtosis–skewness diagram, swarm-like sequences and mainshock–aftershock sequences tend to occupy distinct regions, enabling a practical distinction between the two sequence types without relying on subjective inspection of individual clusters. Overall, the proposed approach offers an efficient way to differentiate swarm-like and mainshock–aftershock seismicity in large catalogs. The method is computationally light, easy to implement, and suitable for rapid screening of earthquake sequence types in high-resolution regional datasets.

How to cite: Fan, Y. and Hu, F.: A New Statistical Method to distinguish Different Earthquake Cluster Types, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4745, https://doi.org/10.5194/egusphere-egu26-4745, 2026.

Earthquake catalogue declustering is a critical preprocessing step in time-dependent seismicity analyses (Gardner and Knopoff, 1974; Reasenberg, 1985), yet its systematic influence on conditional earthquake probability estimates remains insufficiently quantified, particularly in tectonically complex continental collision zones such as the Himalayas (Bungum et al., 2017). Renewal-based recurrence models typically assume that declustered catalogues isolate tectonically driven mainshock recurrence by removing dependent events. However, recent advances in declustering theory demonstrate that methodological choices, ranging from fixed spatio-temporal windows to adaptive and stochastic approaches, can substantially modify inter-event time statistics and inferred recurrence memory (Zaliapin et al., 2008; Zaliapin & Ben-Zion, 2020; Teng & Baker, 2019). Despite these developments, the implications of declustering-induced variability for time-dependent conditional probabilities remain underexplored in active orogenic belts.

In this study, we explicitly quantify how alternative declustering strategies influence time-dependent recurrence behavior and conditional rupture probabilities across selected Himalayan seismic source zones. Inter-event time series were constructed for moderate-to-large earthquakes (M ≥ 4.0) using both raw (non-declustered) and declustered catalogues derived from regional earthquake compilations. Declustering was performed using commonly applied fixed-window and adaptive approaches to capture epistemic variability associated with catalogue preprocessing. The resulting inter-event times were analyzed within renewal process models, including Brownian Passage Time (BPT), Lognormal, Weibull, and Gamma distributions, to estimate conditional probabilities as functions of elapsed time since the most recent major event.

Results show that declustered catalogues consistently yield smoother initial probability gradients and delayed probability peaks relative to raw catalogues, reflecting reduced short-term temporal clustering in inter-event time distributions. These shifts correspond to systematic changes in inferred renewal memory parameters, with declustering suppressing short-term contagion effects while largely preserving long-term mean recurrence intervals. In the Himalayas, collision-driven aftershock swarms and spatially heterogeneous fault interactions amplify these effects, introducing substantial epistemic uncertainty in early-time conditional probabilities, which can locally exceed factors of two to three depending on the declustering strategy employed. In contrast, long-term probability remains comparatively robust across declustering scenarios, consistent with steady-state tectonic strain accumulation.

These findings identify catalogue declustering as a dominant and often underappreciated source of uncertainty in time-dependent seismic probability modelling, reinforcing recent calls for ensemble-based and transparent pre-processing strategies in probabilistic seismic hazard workflows. This study advances a methodological framework for interpreting renewal-based conditional probabilities in clustered tectonic regimes. The Himalayas emerge as a natural laboratory where combined raw and declustered analyses can yield more resilient probabilistic interpretations.

How to cite: Pratap, B. and Sharma, M. L.:  Sensitivity of Time-Dependent Earthquake Conditional Probabilities to Catalogue Declustering in the Himalayas , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17842, https://doi.org/10.5194/egusphere-egu26-17842, 2026.

The Arabian Peninsula is among the most water-stressed regions globally, where limited precipitation, high evapotranspiration and rapid socio-economic development exacerbate vulnerability to drought. Emerging evidence indicates a significant intensification of drought conditions in recent decades, driven by climate variability and long-term warming trends posing serious challenges to water security, ecosystem stability and socio-economic resilience. Therefore, understanding historical drought dynamics, together with reliable drought prediction, is essential for strengthening drought monitoring and mitigation strategies in arid environments and for reducing drought-related risks. However, accurate drought prediction at fine resolution scale remains challenging due to the sparse distribution of meteorological stations. This study investigates the performance of the Standardized Precipitation Index (SPI) and the Standardized Precipitation Evapotranspiration Index (SPEI) at 3-, 6- and 12-month timescales using precipitation data from the Climate Hazards Group InfraRed Precipitation with Station data (CHIRPS) and potential evapotranspiration derived from the TerraClimate dataset, respectively, for pixel-level drought assessment over the period 1992-2024. The historical dynamics were studied using Mann-Kendall trend, Sen’s slope and hotspot analysis. Random Forest (RF) was employed to assess its applicability for drought prediction in arid environments using satellite data, owing to its widespread adoption in global drought-prediction studies. The analysis demonstrates that the RF model exhibits high predictive performance under the studied conditions, with robust performance for SPEI-6 (R² = 0.92, RMSE = 0.12, NSE = 0.92) and satisfactory results for SPEI-12 (R² = 0.77, RMSE = 0.22, NSE = 0.77). These findings confirm enhanced predictability of seasonal to long-term drought variability across the Arabian Peninsula using a satellite-driven RF framework. The results showed the dominance of antecedent SPEI variables (>90%) indicating that cumulative moisture deficits and rising atmospheric evaporative demand primarily govern seasonal to long-term drought evolution over the Arabian Peninsula. In contrast, the consistently low contribution of SPI based indices (<3%) underscores the limited standalone role of precipitation variability in sustaining drought conditions in this arid region. Consistent with these predictive results, spatial trend analysis reveals pronounced heterogeneity in drought evolution across the Arabian Peninsula, with SPI exhibiting mixed and weak precipitation-driven signals, whereas SPEI shows widespread and statistically significant drying, particularly at 6- and 12-month timescales. This divergence further confirms that increasing evaporative demand and regional warming are the primary drivers of long-term drought intensification, reinforcing the dominant role of evapotranspiration processes identified by the machine-learning models. Therefore, the integration of satellite-derived pixel-level datasets with the RF model provides an effective framework for drought prediction across the Arabian Peninsula, offering valuable insights for water resource managers and policymakers to support the development of robust early warning systems and targeted mitigation strategies.

How to cite: Mushtaq, F. and Alhems, L. M.: Machine learning based prediction of long-term drought persistence over the Arabian Peninsula, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2262, https://doi.org/10.5194/egusphere-egu26-2262, 2026.

Forecasting seismic hazard on complex fault systems remains a global challenge, particularly where ruptures can cascade across structural transitions. Aotearoa–New Zealand’s (A–NZ) central transition zone exemplifies this, where the Alpine Fault (AF) and Marlborough Fault System (MFS) connect the Puysegur and Hikurangi subduction zones and pose a major seismic risk to A–NZ communities. The Alpine Fault is late in its interseismic cycle, with a 75% probability of rupture on its central segment within the next 50 years, and a high likelihood of this cascading into a Mw>8 multi-fault rupture onto the MFS. Understanding the behaviour of past earthquake sequences in this region is therefore a national priority to better estimate the extent and dynamics of future shaking. Instrumental records only span a fraction of an earthquake cycle, leaving critical gaps in recurrence patterns and rupture behaviour, which paleo-seismic archives can help to resolve.

We address this gap by integrating lake-sediment paleo-shaking records with calibrated ground-motion modelling and empirical source inversion. Using South Island lakes as binary seismometers, we reconstruct rupture scenarios for historical earthquakes in the central A–NZ transition zone. For each event, we define the probable fault planes and forward-model potential peak ground velocities at each lake site using a suite of ground-motion models that have been extensively tested and adopted in the New Zealand National Seismic Hazard Model. These modelled ground motions are then compared with age-dated mass-transport deposits, which record earthquake-induced shaking and allow calibration of the sequence and timing of events at each site. Finally, a source-inversion technique is used to identify rupture extents and magnitudes that satisfy both rupture-scaling constraints and the binary shaking evidence preserved in the sedimentary record.

In this presentation, we will demonstrate how our integrated approach constrains the magnitudes, rupture locations, and recurrence histories of eight historical earthquakes in central Aotearoa–New Zealand at unprecedented spatial and temporal resolution. The methodology reduces epistemic uncertainty associated with conventional intensity-based methods and is transferable to other complex fault systems, including subduction zones. Crucially, our research provides essential empirical inputs for time-dependent seismic hazard models in Aotearoa–New Zealand.

How to cite: OKane, A., Howarth, J., Fitzsimons, S., Moody, A., and Clark, K.: Reconstructing rupture dynamics of historical Alpine–Marlborough Fault earthquakes, Aotearoa–New Zealand, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-905, https://doi.org/10.5194/egusphere-egu26-905, 2026.

Extreme windstorms pose significant risks to interconnected critical infrastructures such as power, transportation, and telecommunication systems. Wind-induced damage to physical assets, including overhead lines and roadside vegetation, can trigger cascading failures across interdependent networks, leading to widespread service disruptions and societal impacts. Anticipating these cascading effects under uncertain and evolving windstorm conditions remains a major challenge for emergency and crisis management.

An AI-powered Digital Twin (DT) framework for windstorm emergency management is introduced in this presentation, focusing on interconnected critical infrastructures exposed to extreme wind hazards. The framework integrates physics-based windstorm simulation with cascading impact analysis within a unified digital environment, enabling systematic assessment of the interconnected infrastructure performance across a wide range of plausible windstorm scenarios. Rather than relying solely on historical events, physically informed models are used to generate synthetic windstorm scenarios that support preparedness planning and stress-testing under future extreme conditions.

Building on ensembles of simulated windstorm scenarios, the framework can incorporate Generative AI (GenAI) techniques as a post-simulation analytical layer for vulnerability and risk analysis. GenAI operates on the outputs of physics-based simulations, learning asset-level and system-level operational behaviors and vulnerability patterns from simulated impacts, rather than replacing the underlying hazard or infrastructure models. In this role, GenAI captures complex and nonlinear relationships between wind event characteristics and cascading infrastructure failures, enabling efficient synthesis and generalization across large scenario ensembles. This hybrid physics–AI approach supports rapid and accurate identification of vulnerable assets across interconnected infrastructures, spatial hotspots of risk, and conditions that may lead to cascading disruptions under future windstorm scenarios, while preserving the physical consistency of the Digital Twin.

The applicability of the proposed framework is demonstrated through representative case studies involving national-scale interconnected power, telecommunication, and transportation infrastructures in Cyprus, serving as an example implementation. The results illustrate how the AI-powered Digital Twin can support emergency and crisis management at a national level by enabling stress-testing of infrastructure systems, identification of highly vulnerable and critical assets in the Cyprus interconnected infrastructure, improving situational awareness on critical wind-induced cascading risks, and informing response and recovery strategies under severe windstorm conditions.

Overall, this work highlights the potential of hybrid physics-based and AI-enhanced Digital Twins as decision-support tools for windstorm emergency management in interconnected critical infrastructures, providing a flexible and extensible foundation for improving resilience to climate-driven hazards.

How to cite: Venkatasubramanian, B. V., Laoudias, C., and Panteli, M.: AI-Powered Digital Twin Framework for Windstorm Emergency Management in Interconnected Critical Infrastructures, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6783, https://doi.org/10.5194/egusphere-egu26-6783, 2026.