Explore the vast range of titles available.

Among the most fascinating, recent discoveries in seismology are the phenomena of dynamically triggered fault slip, including earthquakes, tremor, slow and silent slip—during which little seismic energy is radiated—and low frequency earthquakes. Dynamic triggering refers to the initiation of fault slip by a transient deformation perturbation, most often in the form of passing seismic waves. Determining the frictional constitutive laws and the physical mechanism(s) governing triggered faulting is extremely challenging because slip nucleation depths for tectonic faults cannot be probed directly. Of the spectrum of slip behaviors, triggered slow slip is particularly difficult to characterize due to the absence of significant seismic radiation, implying mechanical conditions different from triggered earthquakes. Slow slip is often accompanied by nonvolcanic tremor in close spatial and temporal proximity. The causal relationship between them has implications for the properties and physics governing the fault slip behavior. We are characterizing the physical controls of triggered slow slip via laboratory experiments using sheared granular media to simulate fault gouge. Granular rock and glass beads are sheared under constant normal stress, while subjected to transient stress perturbation by acoustic waves. Here we describe experiments with glass beads, showing that slow and silent slip can be dynamically triggered on laboratory faults by ultrasonic waves. The laboratory triggering may take place during stable sliding (constant friction and slip velocity) and/or early in the slip cycle, during unstable sliding (stick‐slip). Experimental evidence indicates that the nonlinear‐dynamical response of the gouge material is responsible for the triggered slow slip. Key Points Laboratory triggered slow slip is due to nonlinear dynamics Triggered slow slip occurs only at modest applied pressure The work places constraints on effective pressure under which slow slip occurs

Although kinematic earthquake source inversions show dominantly pulse‐like subshear rupture behavior, seismological observations, laboratory experiments and theoretical models indicate that earthquakes can operate with different rupture styles: either as pulses or cracks, that propagate at subshear or supershear speeds. The determination of rupture style and speed has important implications for ground motions and may inform about the state of stress and strength of active fault zones. We conduct 2D in‐plane dynamic rupture simulations with a spectral element method to investigate the diversity of rupture styles on faults governed by velocity‐and‐state‐dependent friction with dramatic velocity‐weakening at high slip rate. Our rupture models are governed by uniform initial stresses, and are artificially initiated. We identify the conditions that lead to different rupture styles by investigating the transitions between decaying, steady state and growing pulses, cracks, sub‐shear and super‐shear ruptures as a function of background stress, nucleation size and characteristic velocity at the onset of severe weakening. Our models show that small changes of background stress or nucleation size may lead to dramatic changes of rupture style. We characterize the asymptotic properties of steady state and self‐similar pulses as a function of background stress. We show that an earthquake may not be restricted to a single rupture style, but that complex rupture patterns may emerge that consist of multiple rupture fronts, possibly involving different styles and back‐propagating fronts. We also demonstrate the possibility of a super‐shear transition for pulse‐like ruptures. Finally, we draw connections between our findings and recent seismological observations. Key Points Earthquakes may not be restricted to a single rupture style Small changes of stress, friction, nucleation may lead to rupture style change Results may explain complex seismological observations including repeated slip

Despite continuous improvements, precipitation forecasts are still not as accurate and reliable as those of other meteorological variables. A major contributing factor to this is that several key processes affecting precipitation distribution and intensity occur below the resolved scale of global weather models. Generative adversarial networks (GANs) have been demonstrated by the computer vision community to be successful at super‐resolution problems, that is, learning to add fine‐scale structure to coarse images. Leinonen et al. (2020, https://doi.org/10.1109/TGRS.2020.3032790) previously applied a GAN to produce ensembles of reconstructed high‐resolution atmospheric fields, given coarsened input data. In this paper, we demonstrate this approach can be extended to the more challenging problem of increasing the accuracy and resolution of comparatively low‐resolution input from a weather forecasting model, using high‐resolution radar measurements as a “ground truth.” The neural network must learn to add resolution and structure whilst accounting for non‐negligible forecast error. We show that GANs and VAE‐GANs can match the statistical properties of state‐of‐the‐art pointwise post‐processing methods whilst creating high‐resolution, spatially coherent precipitation maps. Our model compares favorably to the best existing downscaling methods in both pixel‐wise and pooled CRPS scores, power spectrum information and rank histograms (used to assess calibration). We test our models and show that they perform in a range of scenarios, including heavy rainfall. Plain Language Summary The processes that lead to precipitation (rainfall) happen on a very small scale. Weather forecast computer models work on much larger scales, so rainfall is often poorly predicted. In this paper, we develop a method that enhances the resolution of rainfall forecasts by a factor of 10, and makes the forecasts more accurate. We generate many samples of what the rainfall pattern could be, which gives an idea of the uncertainty in the forecast. Our method is based on machine learning and neural networks, which means that we use many past examples of weather forecasts, together with the rainfall that actually happened, and our method “automatically” learns how the forecasts can be improved. We use an existing idea called “Generative Adversarial Networks,” which has been used very successfully in image‐related tasks, such as producing realistic higher‐resolution images from low‐resolution ones. Our task is similar to producing a high‐resolution image from a low‐resolution one, hence this approach is promising. Our method outperforms a variety of existing approaches, and even produces good predictions for the most extreme rainfall situations in our data set. These are the scenarios that cause the most real‐world disruption, the most useful events to produce good forecasts for. Key Points We use generative adversarial neural networks to post‐process global weather forecast model output over the UK We produce more realistic precipitation forecasts than the input forecast data, at 10X resolution, with excellent statistical properties We match or outperform a state‐of‐the‐art pointwise downscaling scheme, while also producing spatially coherent images

The 11 April 2012 Mw 8.6 earthquake offshore Sumatra is the largest of the rare great intraplate earthquakes of the instrumental era. This major strike‐slip event occurred in the diffuse zone of deformation that accommodates differential rotation between Indian and Australian plates. We perform a back projection analysis – calibrated with well‐located aftershocks – of short‐period teleseismic P‐waves recorded by the European array to image the rupture process during the mainshock. In complement, a Love wave analysis is conducted for tracking azimuthal change in the apparent global source duration due to the source spatio‐temporal extent. The combined analysis reveals a complex rupture pattern, characterized by three main episodes of energy release, the latest being located 370 km west of the epicenter, on the Ninety East Ridge, with a delay of 120 s. We interpret the 11 April 2012 Mw 8.6 offshore Sumatra earthquake as a complex westward‐propagating sequence of dynamically triggered strike‐slip fault ruptures, associated to the reactivation of the inherited NNE–striking sea floor fabric. The dynamic triggering mechanism could result from the interaction between transient surface wave stress perturbations and fluids. Key Points Source mechanism of the great strike‐slip intraplate earthquake Sequential rupture: triggering, reactivation of seafloor fossil structures Backprojection avoiding biases by a‐priori rupture parameterization

The 6 April 2009 Mw 6.3 L'Aquila destructive earthquake was successfully recorded by closely spaced 10 Hz and 1 Hz recording GPS receivers and strong motion accelerometers located above or close to the 50° dipping activated fault. We retrieved both static and dynamic displacements from very high rate GPS (VHRGPS) recordings by using Precise Point Positioning kinematic analysis. We compared the GPS positions' time series with the closest displacement time series obtained by doubly integrating strong motion data, first, to assess the GPS capability to detect the first seismic arrivals (P waves) and, second, to evaluate the accelerometers' capability to detect coseismic offsets up to ∼45 s after the earthquake occurrence. By comparing seismic and VHRGPS frequency contents, we inferred that GPS sampling rates greater than 2.5 Hz (i.e., 5 or 10 Hz) are required in the near field of moderate‐magnitude events to provide “alias‐free” solutions of coseismic dynamic displacements. Finally, we assessed the consistency of the dynamic VHRGPS results as a constraint on the kinematic rupture history of the main shock. These results suggested that the high‐rate sampling GPS sites in the near field can be as useful as strong motion stations for earthquake source studies.

We investigated changes in the b value of the Gutenberg–Richter’s law in and around the focal areas of earthquakes on March 20 and on May 1, 2021, with magnitude (M) 6.9 and 6.8, respectively, which occurred off the Pacific coast of Miyagi prefecture, northeastern Japan. We showed that the b value in these focal areas had been noticeably small, especially within a few years before the occurrence of the M6.9 earthquake in its vicinity, indicating that differential stress had been high in the focal areas. The coseismic slip of the 2011 Tohoku earthquake seems to have stopped just short of the east side of the focus of the M6.9 earthquake. Furthermore, the afterslip of the 2011 Tohoku earthquake was relatively small in the focal areas of the M6.9 and M6.8 earthquakes, compared to the surrounding regions. In addition, the focus of the M6.9 earthquake was situated close to the border point where the interplate slip in the period from 2012 through 2021 has been considerably larger on the northern side than on the southern side. The high-stress state inferred by the b-value analysis is concordant with those characteristics of interplate slip events. We found that the M6.8 earthquake on May 1 occurred near an area where the b value remained small, even after the M6.9 quake. The ruptured areas by the two earthquakes now seem to almost coincide with the small-b-value region that had existed before their occurrence. The b value on the east side of the focal areas of the M6.9 and M6.8 earthquakes which corresponds to the eastern part of the source region of the 1978 off-Miyagi prefecture earthquake was consistently large, while the seismicity enhanced by the two earthquakes also shows a large b value, implying that stress in the region has not been very high.

Subseasonal rainfall forecast skill is critical to support preparedness for hydrometeorological extremes. We assess how a process‐informed evaluation, which subsamples forecasting model members based on their ability to represent potential predictors of rainfall, can improve monthly rainfall forecasts within Central America in the following month, using Costa Rica and Guatemala as test cases. We generate a constrained ensemble mean by subsampling 130 members from five dynamic forecasting models in the C3S multimodel ensemble based on their representation of both (a) zonal wind direction and (b) Pacific and Atlantic sea surface temperatures (SSTs), at the time of initialization. Our results show in multiple months and locations increased mean squared error skill by 0.4 and improved detection rates of rainfall extremes. This method is transferrable to other regions driven by slowly‐changing processes. Process‐informed subsampling is successful because it identifies members that fail to represent the entire rainfall distribution when wind/SST error increases. Plain Language Summary Subseasonal rainfall forecasts provide alerts multiple weeks ahead. These forecasts present an opportunity to facilitate anticipatory actions yet are often unreliable to use when preparing for extreme weather. We develop a method to optimize rainfall forecasts by selecting individual members from a large ensemble of dynamic forecasting model outputs based on their ability to represent potential predictors of rainfall. We test our method on monthly rainfall forecasts within Central America in the following month, using Costa Rica and Guatemala as key test cases. We select members from five contributing models of the C3S multimodel ensemble using regional predictors, including wind direction and sea surface temperatures (SSTs). Our results show improvements in the detection of low and high rainfall extremes. This method is transferrable to other regions driven by slowly‐changing processes like SSTs and is beneficial for operational forecasters who can leverage regional expertise of relevant rainfall‐generating processes to subsample better performing ensemble members for their regions. Key Points Subsampling members using sea surface temperatures and zonal wind improves subseasonal ensemble rainfall forecasts in Central America In multiple months and locations mean squared error skill increases by 0.4 and extreme rainfall skill improves by 0.5 (Heidke skill) Process‐informed subsampling is useful because the models' representation of rainfall degrades as process error increases

Large and shallow earthquakes can trigger seismicity from long-distance ranges, ideally along significant plate boundaries and in active geothermal/volcanic regions. The present study aims to garner evidence for dynamically triggered events in the intraplate Surendranagar and Talala regions of Saurashtra Horst, Northwestern India, which are the premier sites of mining- and monsoon-induced activities, respectively. A routine catalogue analysis did not reveal any apparent dynamic triggering of earthquakes in the Saurashtra region. To investigate the possibility of triggered earthquake signatures in the waveform data, we applied the Matched Filter Technique (MFT) to the waveform data of 31 teleseismic earthquakes with Peak dynamic stresses ≥ 1 kPa, that occurred between 2007 and 2017. Results reveal that one (2017 M w 7.9 Papua New Guinea) event triggered seismicity in Surendranagar and four (2007 M w 7.9 Sumatra; 2009 M w 7.6 Sumatra; 2010 M w 8.8 Chile, and 2012 M w 7.6 Costa Rica) in the Talala region. β-statistics further confirm the triggering. Application of the MFT revealed 81 hitherto unrecognized local events in a 20-hour duration around the triggering mainshocks. Only ∼ 16% of the examined remote mainshocks produced dynamic triggering in the Saurashtra region. The other recent earthquakes, i.e., the 2011 M w 9.1 Tohoku-Oki and the 2012 M w 8.6 Indian Ocean earthquakes, did not trigger any seismicity despite having a significant value of peak dynamic stress. This suggests that the amplitude of surface waves is not a necessary condition for dynamic triggering. A detailed investigation revealed an intriguing observation that the identified seismicity in the study region is more likely to be triggered after the Indian monsoon when the faults are critically stressed. Given the crustal fluids in the region, their presence and/or sub-critical crack growth model may be plausible mechanisms for triggering. The study suggests that mining- and monsoon-related activities may perturb the subsurface stress conditions, making the region more susceptible to dynamic triggering.

Recurrent and large forest fires negatively impact ecosystem, air quality, and human health. Moderate Resolution Imaging Spectroradiometer fire product is used to identify forest fires over central India domain, an extremely fire prone region. The study finds that from 2001 to 2020, ∼70% of yearly forest fires over the region occurred during March (1,857.5 counts/month) and April (922.8 counts/month). Some years such as 2009, 2012, and 2017 show anomalously high forest fires. The role of persistent warmer temperatures and multiple climate extremes in increasing forest fire activity over central India is comprehensively investigated. Warmer period from 2006 to 2020 showed doubling and tripling of forest fire activity during forest fire (February–June; FMAMJ) and non‐fire (July–January; JASONDJ) seasons, respectively. From 2015 JASONDJ to 2018 FMAMJ, central India experienced a severe heatwave, a rare drought and an extremely strong El Niño, the combined effect of which is linked to increased forest fires. Further, the study assesses quinquennial spatiotemporal changes in forest fire characteristics such as fire count density and average fire intensity. Deciduous forests of Jagdalpur‐Gadchiroli Range and Indravati National Park in Chhattisgarh state are particularly fire prone (>61 fire counts/grid) during FMAMJ and many forest fires are of high intensity (>45 MW). Statistical associations link high near surface air temperature and low precipitation during FMAMJ to significantly high soil temperature, low soil moisture content, low evapotranspiration and low normalized difference vegetation index. This creates a significantly drier environment, conducive for high forest fire activity in the region. Plain Language Summary Forest fire activity is strongly related to three factors‐availability of combustible fuels, climate and weather forcing, and ignition agents (natural or anthropogenic). Under a warming climate, many studies suggest a significant increase in forest fires. Forest fire activity is already found to be significantly increasing in many regions for example, California and the Arctic. In India, most studies focus on either the forest fires in Himalayas or Jhum‐cultivation led forest fire activity. However, central India which has a high forest fire activity goes neglected. The present study investigates the forest fire activity changes in central India domain over a period of 20 years using satellite observations. Compared to 2001–2005, forest fire activity during 2006–2020 doubled in the forest fire season and tripled in the non‐fire season. In central India domain, forest fires are also found to increase under periods of persistent warmer temperatures and under simultaneous multiple climate extremes for example, severe drought, strong El Niño and intense heatwave. Further, some dense deciduous forests of the region are found to be extremely fire prone, and are identified. Significantly high soil temperature, low soil moisture content, low evapotranspiration and low vegetation index dries out the environment, and high forest fire activity occurs. Key Points Compared to 2001–2005, central India domain forest fire activity during 2006–2020 doubled in forest fire season and tripled in non‐fire season Role of persistent warmer temperatures and multiple climate extremes in increasing forest fire activity over central India is highlighted Deciduous forests of Jagdalpur‐Gadchiroli Range and Indravati National Park are extremely fire prone and fires are of high intensity