Articles tagged with Earthquake Forecasting
Researchers found the Tonga tsunami reached 90 meters in height, outperforming previous tsunamis like 2011 Japan tsunami. The event emphasized the need for improved detection systems, as volcano-based tsunamis are currently 30 years behind earthquake-based event monitoring tools.
Researchers at the University of Illinois used single calcite crystals with varying surface roughness to simplify the physics of fault movement. The study found that friction can increase or decrease with sliding velocity depending on mineral types and environment, providing a fundamental understanding of rate-and-state equations.
A new approach based on deep learning AI detects weak gravitational signals, or PEGS, generated by large-mass motion in megaquakes. This allows for real-time tracking of earthquake growth after a magnitude 8 event.
Low-frequency tectonic tremors in Alaska are linked to high levels of dehydration in the Yakutat terrane, a subducting oceanic plateau. The study suggests that this dehydration reaction is caused by temperature and pressure conditions during plate subduction.
Aftershocks form spatial clusters around mainshocks, with rates decreasing further from mapped faults and increasing with strain rate. Hardebeck's study improves models but falls short of capturing clustering density. Future research may incorporate crustal properties for better forecasts.
Geophysicists and computer scientists collaborate to better understand the dynamics of earthquakes and tsunamis. The team has identified three major characteristics that play a significant role in determining an earthquake's potential to stoke a tsunami, including stress along the fault line, rock rigidity, and sediment layer strength.
Scientists discovered that earthquakes influence tectonic plate movement, altering frequency and patterns of quakes. This finding suggests improved earthquake risk models can be developed by incorporating feedback mechanisms after an earthquake.
A new study finds that cell phone GNSS networks can accurately track crustal deformation, offering a more comprehensive view of seismic activity. By combining private and public sector networks, researchers aim to improve fault models and enhance disaster prevention.
Researchers analyzed the 2018-2019 Bungo Channel slow slip event to gain insight into megathrust earthquake behavior. Despite its short duration, this event was larger in terms of slippage amount and slip velocity compared to past events, providing valuable information for predicting future earthquakes.
Researchers at the University of Massachusetts Amherst developed a physical model that yields unprecedented high-resolution look at slip rates of faults, determining likelihood of earthquakes. The study reveals fault lines in kitchen sinks can predict earthquake-causing forces.
A research team analyzed seismic noise recorded over the last decade to better understand volcanic processes at Campi Flegrei. They found that directionality loss in noise data indicates the migration of deep fluids towards the eastern caldera, which triggers earthquakes.
Scientists aim to develop computer models that can forecast earthquake chances and impact, like weather forecasting. The project will also train students and researchers from diverse backgrounds to work on computational geoscience.
A University of Delaware study of ocean rocks has informed earthquake science by understanding the properties of underwater faults and their impact on seismic activity. Researchers have found that seawater infiltration in these faults weakens the rock, allowing it to flow faster and potentially reducing the risk of large earthquakes.
A multidisciplinary team of scientists developed a method to dispose wastewater safely, reducing the danger of triggering earthquakes. The approach was tested in Italy's largest onshore oil field and found to be sustainable.
Researchers identified eight California faults with higher rates of magnitude 4 earthquakes before a magnitude 6.7 quake, predicting their likelihood for future large earthquakes. So far, only one such event has occurred since the forecast was made in 2017.
Researchers found that Cascadia intraslab earthquakes produce fewer aftershocks, with rates lower by more than half the global average. The study suggests a 'clock-advance' model, where mainshock causes tectonically loaded fault patches to slip earlier.
A machine-learning competition to predict earthquake simulations attracted over 4,500 teams, yielding insights into fault processes. The winning teams employed novel computational approaches, suggesting the value of engaging the machine-learning community in predicting significant scientific problems.
A new study published in Geology has shed light on the mechanisms driving induced subsidence and seismicity in gas-producing sandstone reservoirs. Researchers analyzed drill core samples from the Groningen field, finding evidence of elastic strain plus inelastic compression of weak clay films within grain contacts.
Passelègue's groundbreaking study sheds light on the dynamics of faults, finding that initial strain plays a crucial role in determining rupture speed and energy release. His model shows that higher strains trigger faster ruptures while lower strains result in slower ones.
Scientists at Imperial College London have developed a method to estimate the likelihood of future earthquakes in high-risk areas, improving precision by up to 49%. By analyzing rare atoms in precarious balance rocks (PBRs), they created a new technique to validate earthquake hazard estimates.
Researchers from Northwestern University used a crowd-sourced platform to analyze seismic recordings and found that citizen scientists can classify earthquakes with 85% accuracy, outperforming machine learning algorithms. Citizen scientists also successfully identified tectonic tremors, which AI could not do previously.
Seismological studies revealed that weak rocks formed over 500 million years ago controlled the pathways of the 2016 magnitude 6.0 Petermann earthquake. The unusually long and smooth rupture was guided by these zones of weaker rocks, which can help forecast future earthquakes.
Researchers found that large earthquake sequences are 'burstier' and more difficult to predict than expected, with irregular gaps between event bursts. This finding could impact seismic hazard assessment and the way we evaluate an event's likelihood of repeating soon after a large earthquake.
A new study found that underwater mountains pulled into subduction zones can set the stage for powerful quakes and create conditions that end up dampening them. Researchers used a computer model to simulate the effects of seamounts on surrounding rock and sediment, finding that the brittle rock ahead of the seamount creates powerful ea...
Engineers at Duke University have developed a model predicting mechanical behaviors and origins of earthquakes in various rocks, providing insights into unobservable phenomena deep beneath the Earth's surface. The model accurately reproduces how friction decreases as rock speed increases, shedding light on earthquake mechanisms.
Researchers developed a method to forecast probability and severity of aftershocks, analyzing seismic patterns and incorporating statistical methods. This tool provides a useful approach for mitigating earthquake hazard, enabling timely updates based on new data.
A new algorithm developed by University of California, Riverside researchers can spot patterns in massive datasets quickly, improving earthquake detection and monitoring insect vectors. The SCAMP algorithm has been used to detect 16 times more earthquakes than previously known and can also analyze the behavior of chickens.
Researchers at the University of South Florida have successfully developed and tested a new seafloor geodesy system that can detect small movements in the Earth's seafloor, potentially improving forecasting of earthquakes and tsunamis. The buoy, anchored on the sea floor with precision GPS, has been producing data on three-dimensional ...
A team of researchers created a model to predict induced earthquake activity from wastewater injection, accounting for various variables such as subsurface hydrology parameters and regional stress on faults. The model can help oil operators restrict injection control and prevent large earthquakes in Oklahoma.
Researchers discovered a connection between changes in internal gravity wave parameters five days before an earthquake, enabling the development of short-term forecast methods. The study's results show that IGWs can be used to identify seismically active regions and make predictions about upcoming seismic events.
A recent study analyzed 100,000 localized seismic events to search for patterns in the data. Researchers found that earthquakes of differing magnitudes share more similarities than previously thought, suggesting predictable characteristics may aid forecasting.
Researchers found nearly three-fourths of foreshocks preceded mainshocks by days to weeks, significantly higher than previously understood. Advanced signal processing techniques and computing capabilities enabled the detection of small foreshocks with magnitudes less than 1.
Researchers developed a model forecasting earthquake hazards in Oklahoma due to fluid injection, highlighting the importance of pore pressure diffusion and poroelastic stresses. The study found that mandatory reduction in injection volumes substantially reduced earthquake probability in western Oklahoma but not central Oklahoma.
A study published in Nature Communications suggests that historical earthquake stresses can predict future seismic activity. Researchers analyzed centuries-old written records of damage to reveal 97% of earthquakes occurred on positively stressed faults.
Researchers have found that faults in the Imperial Valley are mechanically linked, forming a continuous fault structure. This could increase the likelihood of larger earthquakes in Los Angeles and surrounding counties.
Researchers characterized two massive deep earthquakes in the Tonga-Fiji region, discovering complex geological processes and dual mechanism propagation patterns. The study suggests that these events can trigger subsequent large earthquakes and highlights the need to better understand deep-Earth processes.
Researchers have simulated surface roughness creation and fractal characteristic emergence on multiple scales. The study proposes that subsurface crack propagation drives the self-affine nature of rough surfaces, with implications for earthquake prediction and material durability.
Researchers are using machine learning to analyze seismic data, identify earthquake centers and distinguish seismic activity from noise. Machine learning methods can also preserve analog records of past earthquakes by categorizing images.
Researchers from the University of Bristol used satellite imagery to detect uplift on the northern flank of Agung volcano, indicating magma movement sideways as well as vertically. This study provides the first geophysical evidence of a connected plumbing system between Agung and Batur volcanoes.
Researchers at the University of Texas at Austin developed a new computer modeling approach to investigate the connection between tiny tremors and devastating megathrust earthquakes. The study shows that changes in crustal stress state occur before major earthquakes, providing valuable insights into the forces driving these events.
Researchers analyzed historical data to determine the sources of destructive Indonesian earthquakes, finding that intraslab earthquakes were responsible for many damaging quakes. The study suggests that Indonesia's 2010 and 2017 seismic hazard assessments perform well in predicting ground motion in key Javanese cities.
Researchers developed a model to forecast man-made earthquake activity in Oklahoma and Kansas, incorporating earthquake physics and wastewater injection data. The model predicts a 32% probability of potentially damaging earthquakes in 2018, decreasing to 19% by 2020.
Researchers at UMass Amherst found that small earthquakes near the San Andreas and San Jacinto faults may be due to deep creep 10 km below the surface. This finding should support more refined assessments of fault loading and earthquake rupture risk in the region.
Researchers present a new continuous model describing self-organized criticality, integrating areas such as economics and developmental biology. The model uses tropical geometry to describe the dynamics of critical systems, providing a universal solution for phenomena like earthquakes and sandpiles.
Using deep learning algorithms, researchers have developed a system that forecasts aftershocks significantly better than random assignment. By analyzing earthquake data and physics-based models, they identified the second invariant of the deviatoric stress tensor as an important factor in predicting aftershock locations.
A joint British-Italian team has shown that the clustering of three deadly quakes in Italy's Apennine mountains may have been caused by a cross-cutting network of underground faults. The research found that smaller faults prevented a single massive earthquake from occurring and also triggered later earthquakes.
A new model developed by Indiana University researchers estimates the likelihood of landslides triggered by earthquakes, offering potentially life-saving information to those affected. The model uses data from past earthquakes and landslides to provide fast, regional estimates of landslide occurrence.
ASU scientists found that slow earthquakes on the central section of the San Andreas Fault release energy over months, triggering large destructive quakes in their surroundings. The research suggests that seismic hazard in California varies over time and is higher than previously thought.
A global collaboration, CSEP, has been studying earthquake forecast models to assess their performance and predict future seismic activity. The experiments have shed light on the predictability of earthquakes, with some models proving useful for real-time warnings during major earthquakes.
Researchers have made a groundbreaking discovery of accelerating activity before a 2016 earthquake in central Alaska. The study found evidence for very low-frequency earthquakes, which do not typically exhibit the usual P and S waves associated with typical earthquakes.
A new study by Virginia Tech researchers found that efforts to curb earthquakes triggered by oilfield wastewater injections are not targeting the most dangerous tremblers. To mitigate large magnitude earthquakes, a larger reduction in injection volumes is needed, according to the study.
Researchers found a correlation between saltwater disposal and earthquake occurrence up to 125 km, but reducing small faults didn't affect larger earthquakes. To reduce large fluid-triggered earthquakes, injecting less water appears to be the solution.
Scientists analyzed seismic data from Japan's Hi-net network to identify changes in subsurface conditions before and after the 2016 Kumamoto earthquake. These changes, including slowed seismic velocity, may signal impending earthquakes or volcanic activity.
The new earthquake forecast, UCERF3, provides self-consistent rupture probabilities from short-term to long-term, including increased likelihood of powerful aftershocks and revised earthquake frequencies. The model also assesses short-term changes in seismic hazard based on earthquake clustering and aftershock excitations.
A team of researchers successfully predicted earthquakes using machine learning techniques in a laboratory setting. By analyzing acoustic signals from faults, they identified a pattern that can be used to estimate the stress on the fault and predict when an earthquake will occur.
Researchers have discovered a potential method to predict nearby strong earthquakes by analyzing deep tremors. The study, published in the Journal of Geophysical Research: Solid Earth, found that changes in deep tremor patterns can signal an impending earthquake.
Researchers validated a new earthquake forecasting model based on the Amatrice-Norcia sequence, showing improved accuracy compared to existing models. The 'brick-by-brick' approach of the operational earthquake forecasting system may revolutionize seismologists' ability to forecast earthquakes.
A new analysis published in Science Advances found that the probability of moderate earthquakes in Oklahoma is two times higher than initially predicted, contradicting earlier predictions of reduced seismic activity. Moderate earthquakes continue to pose a significant hazard in central and western Oklahoma.
Scientists have found that human-induced and natural earthquakes in the central US share similar ground motions, allowing existing prediction frameworks to be applied to both types. This simplifies hazard assessment and could lead to improved building design and infrastructure resilience.
Researchers found that human-induced and naturally occurring earthquakes in the central US share the same shaking potential, causing similar damage. The study used seismic data from 39 earthquakes, concluding that stress drop values of induced and natural earthquakes are identical.