Search

As the world's glaciers recede in response to a warming atmosphere, a change in the magnitude and frequency of related hazards is expected. Among the most destructive hazards are glacier lake outburst floods (GLOFs), and their future evolution is concerning for local populations and sustainable development policy. Central to this is a better understanding of triggers. There is a long‐standing assumption that earthquakes are a major GLOF trigger, and seismic activity is consistently included as a key hazard assessment criterion. Here, we provide the first empirical evidence that this assumption is largely incorrect. Focusing on the Tropical Andes, we show that, of 59 earthquakes (1900–2021) the effects of which intersect with known glacier lakes, only one has triggered GLOFs. We argue that, to help develop climate resilient protocols, the focus for future assessments should be on understanding other key GLOF drivers, such as thawing permafrost and underlying structural geology. Plain Language Summary: Climate change is increasing glacier melt in high mountains and increasing the size and number of glacial lakes. Over time, these lakes may drain catastrophically to generate glacier lake outburst floods (GLOFs), which can devastate downstream communities and destroy valuable infrastructure such as roads, bridges, and hydroelectric power facilities. As a result, many risk assessments have been carried out to understand what the triggers of GLOFs might be, in order to better predict their occurrence. One of the main triggers has been assumed to be earthquakes, but this association has not been properly tested. In this paper, we use earthquake and GLOF data from the Peruvian and Bolivian Andes to test this and find that there is little association between earthquakes and GLOFs. We conclude by arguing that focus needs to be on other GLOF triggers, and earthquake activity should be used as a secondary – not primary – indicator in GLOF susceptibility studies. Key Points: Glacier lake outburst floods (GLOFs) are a major risk concern, with earthquakes being commonly considered effective triggers in hazard assessmentsOf 59 earthquakes (1900–2021), the effects of which intersect lakes in the Tropical Andes, only one triggered GLOFsWe suggest that earthquake activity should be used as a secondary – not primary – susceptibility indicator for GLOFs [ABSTRACT FROM AUTHOR]

This study presents the observation and evaluation of a meteotsunami in the Indian Ocean triggered by the Hunga‐Tonga volcanic eruption. The event was detected through tide gauges and bottom‐pressure recordings across the Indian Ocean, with an amplitude of 10–15 cm, lasting for a few days. A numerical model was used to understand the ocean's response to meteotsunami and evaluate the dynamics behind it. The model results show that the sea‐level oscillations result from the ocean waves generated by a propagating Lamb wave. In addition to interaction with bathymetry, refracted and reflected waves also determine the sea‐level variability. Our analysis shows that bathymetric slope plays a vital role in near‐shore processes. The spectral and spatial characteristics of the meteotsunami were reminiscent of seismic tsunamis. Our research on this rare event elucidates the unresolved issues and eventually leads to designing a blueprint for future observation and modeling of meteotsunamis and seismic tsunamis. Plain Language Summary: This study evaluates the observation of a rare meteotsunami event that occurred in the Indian Ocean. An atmospheric pressure wave generated by the eruption of the Hunga‐Tonga volcano in the Pacific Ocean caused the meteotsunami. The signatures of this event were captured by tide gauges installed along the coast and pressure recorders moored at the ocean floor. We demonstrate using a numerical model that fast‐moving atmospheric pressure waves, Lamb waves, caused the sea‐level oscillations. Interactions with bathymetry further influenced sea‐level variability in the basin, reflected waves, and refracted waves. The vulnerable areas showed a striking resemblance to the area impacted by the tsunami that hit the Indian Ocean in 2004. This research eventually helps to identify and incorporate observational and modeling issues associated with seismic tsunamis and meteotsunami. Key Points: We present the first comprehensive evaluation of meteotsunami in the Indian Ocean using observation and modelingA forced wave to free wave transformation happens on the continental slope, and the bathymetry slope is crucial to near‐shore processesThe meteotsunami exhibits a spectral and spatial similarity to seismic tsunamis that previously occurred in the Indian Ocean [ABSTRACT FROM AUTHOR]

Abstract As the world’s glaciers recede in response to a warming atmosphere, a change in the magnitude and frequency of related hazards is expected. Among the most destructive hazards are glacier lake outburst floods (GLOFs), and their future evolution is concerning for local populations and sustainable development policy. Central to this is a better understanding of triggers. There is a long‐standing assumption that earthquakes are a major GLOF trigger, and seismic activity is consistently included as a key hazard assessment criterion. Here, we provide the first empirical evidence that this assumption is largely incorrect. Focusing on the Tropical Andes, we show that, of 59 earthquakes (1900–2021) the effects of which intersect with known glacier lakes, only one has triggered GLOFs. We argue that, to help develop climate resilient protocols, the focus for future assessments should be on understanding other key GLOF drivers, such as thawing permafrost and underlying structural geology.

Abstract This study presents the observation and evaluation of a meteotsunami in the Indian Ocean triggered by the Hunga‐Tonga volcanic eruption. The event was detected through tide gauges and bottom‐pressure recordings across the Indian Ocean, with an amplitude of 10–15 cm, lasting for a few days. A numerical model was used to understand the ocean's response to meteotsunami and evaluate the dynamics behind it. The model results show that the sea‐level oscillations result from the ocean waves generated by a propagating Lamb wave. In addition to interaction with bathymetry, refracted and reflected waves also determine the sea‐level variability. Our analysis shows that bathymetric slope plays a vital role in near‐shore processes. The spectral and spatial characteristics of the meteotsunami were reminiscent of seismic tsunamis. Our research on this rare event elucidates the unresolved issues and eventually leads to designing a blueprint for future observation and modeling of meteotsunamis and seismic tsunamis.

Seismic wave arrival time measurements form the basis for numerous downstream applications. State‐of‐the‐art approaches for phase picking use deep neural networks to annotate seismograms at each station independently, yet human experts annotate seismic data by examining the whole network jointly. Here, we introduce a general‐purpose network‐wide phase picking algorithm based on a recently developed machine learning paradigm called Neural Operator. Our model, called Phase Neural Operator, leverages the spatio‐temporal contextual information to pick phases simultaneously for any seismic network geometry. This results in superior performance over leading baseline algorithms by detecting many more earthquakes, picking more phase arrivals, while also greatly improving measurement accuracy. Following similar trends being seen across the domains of artificial intelligence, our approach provides but a glimpse of the potential gains from fully‐utilizing the massive seismic data sets being collected worldwide. Plain Language Summary: Earthquake monitoring often involves measuring arrival times of P‐ and S‐waves of earthquakes from continuous seismic data. With the advancement of artificial intelligence, state‐of‐the‐art phase picking methods use deep neural networks to examine seismic data from each station independently; this is in stark contrast to the way that human experts annotate seismic data, in which waveforms from the whole network containing multiple stations are examined simultaneously. With the performance gains of single‐station algorithms approaching saturation, it is clear that meaningful future advances will require algorithms that can naturally examine data for entire networks at once. Here we introduce a multi‐station phase picking algorithm based on a recently developed machine learning paradigm called Neural Operator. Our algorithm, called Phase Neural Operator, leverages the spatial‐temporal information of earthquake signals from an input seismic network with arbitrary geometry. This results in superior performance over leading baseline algorithms by detecting many more earthquakes, picking many more seismic wave arrivals, yet also greatly improving measurement accuracy. Key Points: We introduce a multi‐station phase picking algorithm, Phase Neural Operator (PhaseNO), that is based on a new machine learning paradigm called Neural OperatorPhaseNO can use data from any number of stations arranged in any arbitrary geometry to pick phases across the input stations simultaneouslyBy leveraging the spatial and temporal contextual information, PhaseNO achieves superior performance over leading baseline algorithms [ABSTRACT FROM AUTHOR]

Abstract Seismic wave arrival time measurements form the basis for numerous downstream applications. State‐of‐the‐art approaches for phase picking use deep neural networks to annotate seismograms at each station independently, yet human experts annotate seismic data by examining the whole network jointly. Here, we introduce a general‐purpose network‐wide phase picking algorithm based on a recently developed machine learning paradigm called Neural Operator. Our model, called Phase Neural Operator, leverages the spatio‐temporal contextual information to pick phases simultaneously for any seismic network geometry. This results in superior performance over leading baseline algorithms by detecting many more earthquakes, picking more phase arrivals, while also greatly improving measurement accuracy. Following similar trends being seen across the domains of artificial intelligence, our approach provides but a glimpse of the potential gains from fully‐utilizing the massive seismic data sets being collected worldwide.

Monsoon rainfall triggers hundreds of landslides across Nepal every year, causing significant hazard and mass wasting. Annual inventories of these landslides have been mapped using multi‐spectral satellite images, but these images are obscured by cloud cover during the monsoon, making it impossible to use them to constrain landslide timing. We employ recently developed techniques to derive individual timings from Sentinel‐1 for 579 landslides in 2015, 2017, 2018, and 2019 in Nepal. We use this new timing information alongside satellite rainfall data to identify spatio‐temporal clusters of landslides and associate these with periods of particularly intense rainfall. We also observed that during the 2015 monsoon, many landslides failed earlier and in dryer conditions than in 2017–2019. We use physical models to demonstrate how this requires a temporary loss of hillslope strength following the Mw 7.8 Gorkha Earthquake sequence and suggest a modeled cohesion loss in the range 1–3 kPa. Plain Language Summary: Earthquakes, storms and long periods of rain during the monsoon all contribute to landsliding in Nepal. Understanding these different triggers requires information on where and when landslides happen every year. We applied a new landslide timing method based on Sentinel‐1 satellite images alongside satellite measurements of rainfall to model the amount of rain needed to trigger individual landslides during the monsoon. This allowed us to connect landslides with particular storms and to observe the influence of a large earthquake in 2015 on landslide timing. Landslides occurred in unusually dry conditions following the earthquake, which can be interpreted as a temporary weakening of the landscape. Key Points: We use Sentinel‐1 time series to constrain the timings of hundreds of monsoon‐triggered landslides across Nepal in 2015, 2017, 2018, and 2019We identify spatio‐temporal clusters of landslides and associate them with particular periods of intense rainfallWe use landslide timing information to calibrate numerical models of soil saturation and infer post‐seismic hillslope weakening in 2015 [ABSTRACT FROM AUTHOR]

Forecasting hanging glacier instabilities remain challenging as sensing technology focusing on the ice surface fails to detect englacial damage leading to large‐scale failure. Here, we combine icequake cluster analysis with coda wave interferometry constraining damage growth on Switzerland's Eiger hanging glacier before a 15,000 m3 break‐off event. The method focuses on icequake migration within clusters rather than previously proposed "event counting." Results show that one cluster originated from the glacier front and migrated by 13.9(±1.2) m within 5 weeks before the break‐off event. The corresponding crevasse extension separates unstable and stable ice masses. We use the measured source displacement for damage parametrization and find a 90% agreement between an analytical model based on damage mechanics and frontal flow velocities measured with an interferometric radar. Our analysis provides observational constraints for damage growth, which to date is primarily a theoretical concept for modeling englacial fractures. Plain Language Summary: Predicting the development of ice mass breaking off from unstable glaciers is challenging. Such glaciers are often located in remote places with steep terrain, making it difficult to gather observations. Because of that, our current view on ice damage development on unstable glaciers is incomplete. Here, we propose a new approach to tackle this problem using seismic observations from the Eiger hanging glacier in the Swiss Alps before a moderate 15,000 m3 break‐off event. We first group seismic signals according to their similarity. We then use waves scattered by fractures within the ice and at the glacier‐rock boundaries to track displacement of ice damage. Our results suggest that a group of seismic signals is generated by crevasse propagation separating the unstable ice mass and the stable ice uphill. Combined with a simple analytical model, these observations indicate that seismic source displacement can be associated with damage propagation within unstable glacier ice. This is an important step toward a better understanding of unstable ice flow and forecasting glacier instabilities. Key Points: In the months leading up to a break‐off event, we find thousands of recurring icequakes on the Eiger hanging glacier in SwitzerlandCoda wave interferometry resolves displacements of icequake multipletsOne multiplet displacement represents englacial damage corresponding to crevasse extension between unstable and stable ice masses [ABSTRACT FROM AUTHOR]

Abstract Monsoon rainfall triggers hundreds of landslides across Nepal every year, causing significant hazard and mass wasting. Annual inventories of these landslides have been mapped using multi‐spectral satellite images, but these images are obscured by cloud cover during the monsoon, making it impossible to use them to constrain landslide timing. We employ recently developed techniques to derive individual timings from Sentinel‐1 for 579 landslides in 2015, 2017, 2018, and 2019 in Nepal. We use this new timing information alongside satellite rainfall data to identify spatio‐temporal clusters of landslides and associate these with periods of particularly intense rainfall. We also observed that during the 2015 monsoon, many landslides failed earlier and in dryer conditions than in 2017–2019. We use physical models to demonstrate how this requires a temporary loss of hillslope strength following the Mw 7.8 Gorkha Earthquake sequence and suggest a modeled cohesion loss in the range 1–3 kPa.

Abstract Forecasting hanging glacier instabilities remain challenging as sensing technology focusing on the ice surface fails to detect englacial damage leading to large‐scale failure. Here, we combine icequake cluster analysis with coda wave interferometry constraining damage growth on Switzerland's Eiger hanging glacier before a 15,000 m3 break‐off event. The method focuses on icequake migration within clusters rather than previously proposed “event counting.” Results show that one cluster originated from the glacier front and migrated by 13.9(±1.2) m within 5 weeks before the break‐off event. The corresponding crevasse extension separates unstable and stable ice masses. We use the measured source displacement for damage parametrization and find a 90% agreement between an analytical model based on damage mechanics and frontal flow velocities measured with an interferometric radar. Our analysis provides observational constraints for damage growth, which to date is primarily a theoretical concept for modeling englacial fractures.

Surging debris flows are among the most destructive natural hazards, and elucidating the interaction between coarse‐grained fronts and the trailing liquefied slurry is key to understanding these flows. Here, we describe the application of high‐resolution and high‐frequency 3D LiDAR data to explore the dynamics of a debris flow at Illgraben, Switzerland. The LiDAR measurements facilitate automated detection of features on the flow surface, and construction of the 3D flow depth and velocity fields through time. Measured surface velocities (2–3 m s−1) are faster than front velocities (0.8–2 m s−1), illustrating the mechanism whereby the flow front is maintained along the channel. Further, we interpret the relative velocity of different particles to infer that the vertical velocity profile varies between plug flow and one that features internal shear. Our measurements provide unique insights into debris‐flow motion, and provide the foundation for a more detailed understanding of these hazardous events. Plain Language Summary: Debris flows are surging flows of soil, wood, and water that can impact people and infrastructure far downstream of their initiation zone. Work by others has identified that debris flows tend to develop a distinct segregation between large particles concentrated at the front of the flow and small particles at the tail; however, the formation process and implications of this for debris‐flow motion have remained vague. In this work, we present measurements from laser scanners, originally developed for autonomous vehicles, that provide insight into this process. The scanners provide 10 scans per second, which can be used to measure the velocity of objects (rocks and woody debris) on the surface of the flow. We show that different objects in the flow move at different speeds, which results in many destructive features of debris flows. These measurements, and the resulting process understanding, are important for predicting debris‐flow hazard and reducing the associated risk. Key Points: High‐resolution 3D LiDAR scans at subsecond intervals demonstrate a novel method for exploring debris‐flow dynamicsLiDAR‐derived front and surface velocities allow identification of processes forming and maintaining the debris‐flow frontObservations of individual particle motion place constraints on the vertical velocity profile and temporal variation in flow regimes [ABSTRACT FROM AUTHOR]

Kinematic backward trajectories are used to globally quantify the contributions of temperature advection, adiabatic compression and diabatic processes to near‐surface temperature anomalies (hereafter T′ ${T}^{\prime }$) during the coldest day of each year (TN1day events) based on ERA5. Diabatic cooling dominates TN1day anomalies in the climatologically coldest regions, while advection forms TN1day anomalies over most ocean regions. Over most extratropical land masses, TN1day anomalies arise from a combination of both processes. The mean age and formation distance of TN1day anomalies vary strongly in space, from one to 8 days, and 500–5,500 km, respectively. Five distinct types of TN1day events are identified from these physical and spatio‐temporal characteristics, and their geographical occurrence is investigated. Furthermore, advective, adiabatic and diabatic contributions typically cancel each other partially, but less so for the most intense TN1day events, which occur when the atmosphere's ability to dampen near‐surface temperature anomalies is limited. Plain Language Summary: This study globally quantifies the relative importance of the two physical processes that lead to large negative temperature anomalies during cold extremes at the Earth's surface. These are, first, cold air advection, that is, the transport of air from climatologically cold to warmer regions and, second, so‐called diabatic cooling which refers to cooling of the air by radiation, turbulence and processes related to phase changes of water in the air. Here, their relative importance is quantified by tracing roughly 250'000'000 air parcels contributing to cold extremes backwards in time. Cold air advection is globally the dominant process for cold extreme formation, in particular over ocean regions. Over land regions, however, the importance of diabatic cooling gradually increases toward the climatologically coldest regions. The temporal and spatial scales over which these anomalies form vary from less than a day (in tropical Africa) to more than 8 days (in Siberia), and from less than 500 km (in tropical Africa) to more than 5,000 km (in subtropical regions), respectively. The results of this study for the first time provide a global picture of how cold extremes form and will enable us to better evaluate climate models with regard to cold extreme formation. Key Points: Using kinematic backward trajectories, the physical processes contributing to near‐surface cold extremes are quantified globallyAdvection dominates over oceans, and over land the importance of diabatic cooling increases toward the climatologically coldest regionsThe most intense cold extremes form in situations with a limited ability of the atmosphere to dampen negative temperature anomalies [ABSTRACT FROM AUTHOR]

Abstract Kinematic backward trajectories are used to globally quantify the contributions of temperature advection, adiabatic compression and diabatic processes to near‐surface temperature anomalies (hereafter T′) during the coldest day of each year (TN1day events) based on ERA5. Diabatic cooling dominates TN1day anomalies in the climatologically coldest regions, while advection forms TN1day anomalies over most ocean regions. Over most extratropical land masses, TN1day anomalies arise from a combination of both processes. The mean age and formation distance of TN1day anomalies vary strongly in space, from one to 8 days, and 500–5,500 km, respectively. Five distinct types of TN1day events are identified from these physical and spatio‐temporal characteristics, and their geographical occurrence is investigated. Furthermore, advective, adiabatic and diabatic contributions typically cancel each other partially, but less so for the most intense TN1day events, which occur when the atmosphere's ability to dampen near‐surface temperature anomalies is limited.

Abstract Surging debris flows are among the most destructive natural hazards, and elucidating the interaction between coarse‐grained fronts and the trailing liquefied slurry is key to understanding these flows. Here, we describe the application of high‐resolution and high‐frequency 3D LiDAR data to explore the dynamics of a debris flow at Illgraben, Switzerland. The LiDAR measurements facilitate automated detection of features on the flow surface, and construction of the 3D flow depth and velocity fields through time. Measured surface velocities (2–3 m s−1) are faster than front velocities (0.8–2 m s−1), illustrating the mechanism whereby the flow front is maintained along the channel. Further, we interpret the relative velocity of different particles to infer that the vertical velocity profile varies between plug flow and one that features internal shear. Our measurements provide unique insights into debris‐flow motion, and provide the foundation for a more detailed understanding of these hazardous events.

A coincidence in the timing of floods and their drivers can be used as a proxy for the causality of flood generation. Here, we investigate the relationship between the seasonality of floods, maximum annual rainfall, and maximum annual soil moisture data of 886 basins in Brazil for 1980–2015 to shed light on process controls of flood generation. Floods tend to occur at the same time of year as soil moisture peaks and lag behind rainfall peaks by 3 weeks. In Amazonia, central and northern Brazil, flood timing is more correlated with the timing of soil moisture peaks than with that of rainfall peaks, which is interpreted as resulting from high subsurface water storage capacities. In southern and southeastern Brazil, on the other hand, flood timing is highly correlated with both soil moisture and rainfall because of low subsurface water storage capacities. These findings can support flood forecasting and climate impact studies. Plain Language Summary: In warm regions, floods are usually generated by a combination of intense rainfall and wet soils. In this paper, we analyze the average timing within the year of floods, extreme rainfall, and soil moisture to elucidate how floods come about in the main Brazilian rivers. We find that in some regions, such as Amazonia and central Brazil, floods tend to occur when soils are wet. In other regions, such as southern Brazil, floods tend to occur when rainfall is most extreme. We believe that these differences are related to differences in the soil water storage capacity. The understanding of the regional importance of each of these components helps increase the efficiency of flood prevention measures and climate change adaptation. Key Points: Flood peaks tend to occur at the same time of year as annual soil moisture peaks and lag behind annual rainfall peaks by 3 weeksFlood seasonality is linked mainly with soil moisture peaks in Amazonia and central Brazil, where soil storage capacity is highFlood timing is highly correlated with rainfall and soil moisture peaks in the south and southeast, where soil storage capacity is low [ABSTRACT FROM AUTHOR]

Automatic identification of debris flow signals in continuous seismic records remains a challenge. To tackle this problem, we use machine learning, which can be applied to continuous real‐time data. We show that a machine learning model based on the random forest algorithm recognizes different stages of debris flow formation and propagation at the Illgraben torrent, Switzerland, with an accuracy exceeding 90 %. In contrast to typical debris flow detection requiring instrumentation installed in the torrent, our approach provides a significant gain in warning times of tens of minutes to hours. For real‐time data from 2020, our detector raises alarms for all 13 independently confirmed Illgraben events, giving no false alarms. We suggest that our seismic machine‐learning detector is a critical step toward the next generation of debris‐flow warning, which increases warning times using simpler instrumentation compared to existing operational systems. Plain Language Summary: Debris flows are fast‐moving masses of mud, soil, fragmented rock, and water transporting large volume of material in mountainous areas. They pose a significant danger to human life, property, and infrastructure. It is crucial to detect debris flows early enough to send an alarm message to local communities. We propose a novel approach for debris‐flow detection using recorded ground vibrations generated by 22 debris flows at the Illgraben torrent, Switzerland. We use a machine‐learning algorithm that learns to distinguish between debris flow generated ground vibrations and other seismic signals. At the Illgraben torrent, this increases warning times by at least 20 min compared to existing detection systems. Key Points: A novel debris‐flow detector is developed using a machine‐learning model and seismic data from a Swiss torrent channelSignals of 22 debris‐flow events recorded by six seismic stations are used to train and test the machine‐learning modelA detector ran on continuous real‐time 2020 data stream detecting all 13 observed hazardous flows in 3 months and raising no false alarms [ABSTRACT FROM AUTHOR]

Abstract Dike propagation is an intrinsically multiphase problem, where deformation and fluid flow are intricately coupled in a fracture process. Here we perform the first fully coupled simulations of dike propagation in two dimensions, accounting for depressurization of a circular magma chamber, dynamic fluid flow, fracture formation, and elastic deformation. Despite the complexity of the governing equations, we observe that the lengthening is well explained by a simple model a(t)=c1log(1+t/c2), where a is the dike length, t is time, and c1 and c2 are constants. We compare the model to seismic data from eight dikes in Iceland and Ethiopia, and, in spite of the assumption of plane strain, we find good agreement between the data and the model. In addition, we derive an approximate model for the depressurization of the chamber with the dike length. These models may help forecast the growth of lateral dikes and magma chamber depressurization.

Prescribed burning is used globally to mitigate the risks of wildfires, with severe wildfires increasing in frequency in recent decades. Despite their importance in wildfire management, the nature of future changes to prescribed burn windows under global warming remains uncertain. We use a regional climate projection ensemble to provide a robust spatiotemporal quantification of statistically significant future changes in prescribed burn windows for southeastern Australia. There are significant decreases during months presently used for prescribed burning, that is, in March to May in 2060–2079 versus 1990–2009 across several temperate regions. Conversely, burn windows show widespread significant increases in June to August, that is, months when burns have rarely occurred historically, and also in spring (September–October). Overall, projected changes in temperature and fuel moisture show the most widespread and largest decreases (or increases) in the number of days within their respective ranges suitable for conducting burns. These results support wildfire risk mitigation planning. Plain Language Summary: The frequency of destructive wildfires is increasing in many fire‐prone regions, threatening lives, property, and damaging the environment. To reduce the risk of wildfires occurring, agencies conduct planned, controlled, and preemptive burns of vegetation to reduce the available fuel load. Currently, in southeastern Australia, these "prescribed burns" are conducted during autumn (March–May) and early spring (September) but rarely during winter (June–August). Given climate change, we need to understand how opportunities for conducting these burns might change in future. This is particularly important for forest and woodland areas adjacent to populated urban areas. We use regional climate projection data to assess future changes to the suitability for prescribed burning. We demonstrate that over many regions, windows are projected to decrease during months currently used for conducting burns in 2060–2079 compared to 1990–2009. In contrast, they generally increase in the winter months and further into spring (September–October). This change in seasonality means that these periods could be more suitable for conducting burns in future. Future changes in burn windows are most strongly associated with changes in temperature and vegetation moisture content, though there are many other contributing factors to these changes. These findings are relevant to strategic wildfire prevention and regional risk mitigation. Key Points: Significant decreases in wildfire mitigation windows occur in March–May in 2060–2079 versus 1990–2009 but increase in June–August in 2060–2079Projected temperature and fuel moisture changes have the largest contribution to the change in the wildfire mitigation windowLocations in southeast Australia where changes in wildfire mitigation windows are projected to occur are identified on a monthly basis [ABSTRACT FROM AUTHOR]

Dike propagation is an intrinsically multiphase problem, where deformation and fluid flow are intricately coupled in a fracture process. Here we perform the first fully coupled simulations of dike propagation in two dimensions, accounting for depressurization of a circular magma chamber, dynamic fluid flow, fracture formation, and elastic deformation. Despite the complexity of the governing equations, we observe that the lengthening is well explained by a simple model a(t)=c1log(1+t/c2), where a is the dike length, t is time, and c1 and c2 are constants. We compare the model to seismic data from eight dikes in Iceland and Ethiopia, and, in spite of the assumption of plane strain, we find good agreement between the data and the model. In addition, we derive an approximate model for the depressurization of the chamber with the dike length. These models may help forecast the growth of lateral dikes and magma chamber depressurization. Plain Language Summary: Volcanic dike intrusions, propagating magma‐filled fractures, precede most eruptions. Dike propagation has been studied for decades through simplified analytical and numerical models. To date, no study has fully addressed how the fluid magma, host rock, and the magma chamber all interact at the same time and drive the dike forward. We present such simulations for a two‐dimensional configuration and deduce that a simple formula can explain how the dike lengthens with time. We suggest that this simple formula may be used to forecast dike growth. Key Points: Fully coupled simulations of dike growth and magma chamber depressurization are performedA simple model for dike length with time is identified and compared to seismic observationsA simple model of chamber pressure versus dike length is derived and compared to simulations [ABSTRACT FROM AUTHOR]

Pyrocumulonimbus (pyroCb) wildfires cause devastation in many regions globally. Given that fire‐atmosphere coupling is associated with pyroCbs, future changes in coincident high index values of atmospheric instability and dryness (C‐Haines) and near‐surface fire weather are assessed for southeastern Australia using a regional climate projection ensemble. We show that observed pyroCb events occur predominantly on forested, rugged landscapes during extreme C‐Haines conditions, but over a wide range of surface fire weather conditions. Statistically significant increases in the number of days where both C‐Haines and near‐surface fire weather values are conducive to pyroCb development are projected across southeastern Australia, predominantly for November (spring), and less strongly for December (summer) in 2060‐2079 versus 1990‐2009, with future C‐Haines increases linked to increased 850‐hPa dewpoint depression. The increased future occurrence of conditions conducive to pyroCb development and their extension into spring have implications for mitigating these dangerous wildfires and urbanizing fire‐prone landscapes. Plain Language Summary: Pyrocumulonimbus wildfires (pyroCb) are extreme wildfires. They involve a coupling between the fire and the atmosphere, which drives dangerous fire conditions that result in fatalities and considerable damage. Climate change could amplify the conditions associated with pyroCb development. We use high‐resolution regional climate projection data to assess future changes in the risk of pyroCb occurrence for southeastern Australia. We demonstrate that atmospheric instability and dryness is a better indicator of pyroCb development than surface fire weather conditions, with topographic roughness and vegetation type also playing roles. Coincidences of high index values of atmospheric conditions (tropospheric instability and dryness) and surface fire weather conditions (temperature, humidity, wind speed, and precipitation) associated with pyroCb development are projected to increase in frequency predominantly for November (spring) in 2060‐2079 versus 1990‐2009, but less so for summer. The extension of the season conducive to pyroCbs into spring is important because Australian pyroCbs are typically summer phenomena. A change in seasonality has implications for resource allocations by fire agencies. Projected changes in the conditions conducive to pyroCbs are primarily associated with increased dryness at 1.5‐km altitude. These findings are relevant to the mitigation of pyroCbs, and they have implications for continued urban expansion on fire‐prone landscapes. Key Points: Significant increases in the frequency of conditions conducive to pyrocumulonimbus fires occur in November (Spring) 2060‐2079 versus 1990‐2009Atmospheric instability and dryness are better indicators of pyrocumulonimbus fire development than surface fire weatherLocations in southeast Australia where these changes are projected to occur are identified [ABSTRACT FROM AUTHOR]

We present a 250‐year (1766–2015) inventory of European meteorological, hydrological, and agricultural droughts derived from ensemble simulations of the mesoscale Hydrological Model (mHM). The inventory of droughts takes into account an ensemble of 100 simulations from the hydrological model, allowing for assessment of how different meteorological forcing and model parameterizations affect a drought ranking. For the most extreme droughts, the variability in the ranking of drought events is low, while for the years with moderate precipitation deficits, the variability increases. Despite the underlying uncertainties, our drought inventory shows increased recurrence of soil moisture droughts in the Mediterranean and declining spatial extent of hydrological drought in Central Europe over the last three decades. Plain Language Summary: Drought is one of the most costly natural hazards. Complete understanding of this phenomenon evades us because drought needs a long period of time to evolve and when it does, it might propagate through different components of the hydrological cycle. For example, meteorological drought, which is a lack of rainfall, might result in hydrological drought (low discharge/lake levels), agricultural drought (low soil moisture), or both. This is why it is important to record past occurrences of individual droughts to better understand under what conditions they propagated. To address this, herein we produce a 250‐year long European drought inventory based on hydrological modeling that presents the interaction of the three types of drought, namely, meteorological, hydrological, and agricultural, as well as their properties in a single, comprehensive framework. Key Points: A 250‐year inventory depicts the major meteorological, hydrological, and agricultural droughts across EuropeVisualization of ensemble allows for an effective comparison among major drought years (severity and spatial extent)Increased recurrence of soil moisture droughts across the Mediterranean in the recent decades [ABSTRACT FROM AUTHOR]

Abstract: We compiled damages to hydropower projects (HPPs) in the wake of Nepal's 2015 Gorkha earthquake and its main aftershock. Based on 41 records of HPP losses, we identify the main mechanisms determining the degree of loss. We show that, among these mechanisms, earthquake‐triggered landslides had the largest share. Landslides clustered where peak ground acceleration and river steepness Mχ, a metric frequently used in tectonic geomorphology, attain high values. A Bayesian logistic regression reveals that both metrics are credible predictors for HPP damage states. Our study underscores the vulnerability of Himalayan hydropower to landsliding. We estimate that ~25% of existing, currently constructed, and planned HPPs and >10% of potential HPP sites along Himalayan rivers have high probabilities of moderate to severe damage during future earthquake. Our study highlights that HPP hazard assessments need to consider both seismic and the geomorphic setting of HPP sites conjunctively. [ABSTRACT FROM AUTHOR]