Wednesday, 4 November 2026: 3 new papers, 3 cross-lists and 1 replacement in Geophysics (physics.geo-ph), in arXiv's order, with full abstracts. The same papers as arXiv's own listing, set as a newspaper; every title links to arXiv.
Accurately characterizing near-fault ground motions, particularly permanent tectonic displacements and velocity pulses associated with rupture directivity, is essential for seismic hazard assessment and infrastructure design. However, conventional ground-motion processing can distort or remove these characteristics. This study extends existing baseline correction methods by synchronously processing all three components of strong-motion recordings to preserve near-fault effects. The proposed methodology is applied to 59 recordings from the 2022 M 6.5 Guanshan foreshock and M 6.9 Chishang mainshock in Taiwan. Comparison with nearby Global Navigation Satellite System (GNSS) observations demonstrates that the method reliably captures the amplitude and orientation of permanent tectonic displacements. Analysis of the processed recordings reveals pronounced velocity pulses and permanent displacements, with the largest peak-to-peak velocities concentrated along the direction of rupture propagation. A classification framework is also introduced to distinguish fling-step pulses from dynamic velocity pulses. The results indicate that fling-step pulses typically occur early in the ground shaking, whereas dynamic pulses span a broader temporal range. Furthermore, the spatial distribution and orientation of near-fault effects are consistent with bridge damage documented during post-earthquake reconnaissance. The resulting dataset provides ground-motion time histories that preserve near-fault characteristics for seismic hazard assessment and engineering applications.
Lunar seismic records are characterized by long-lasting spindle-shaped coda, which has long been considered to be caused by strong velocity contrasts in a continuous-medium lunar crust; however, simulated waveforms could not well match the observation even with unrealistically strong velocity contrasts, demonstrating its mechanism still unsolved. Here we propose a novel model of coda wave generation with random fractures in the lunar crust, where vacuum fractures prevent seismic wave propagation as insulating boundaries, causing strong persistent reverberation in terms of total reflection among fractures. Our model can easily reproduce the typical characteristics of observed Apollo data even using a limited number of simplified fractures in homogeneous matrix, with envelope correlations over 0.9, much higher than previous studies. Numerical simulations suggest that fractures in the uppermost ~1 km of the crust may dominate lunar seismic scattering. This model enables us to investigate the shallow lunar fracture system using seismic data and to suppress seismic coda for detecting critical phases reflected from deep interiors.
Learning reliable surrogates for geodynamic models requires benchmarks that resolve coupled fields and temporal stability. We present GeoSim, a benchmark of high-resolution two-dimensional mantle-plume simulations generated with I2VIS, comprising 237 trajectories across two parameter regimes, with six coupled fields, physical timestamps, and trajectory-level splits. GeoSim defines short-window prediction and autonomous rollout, and evaluates representative neural operators field by field. Across models, we observe a pronounced scalar--velocity gap: scalar fields are reconstructed accurately while velocity errors remain large, and short-window accuracy does not imply stable rollout. We also evaluate MFGNO (Multi-Field Graph Neural Operator), a joint multi-field graph baseline designed for this benchmark; on GeoSim-Plume1, under a fixed per-field objective, it reduces horizontal and vertical velocity errors by 58.4% and 38.5% relative to an independent-field graph baseline, while an explicit field-difference input shows no measurable benefit at this scale. GeoSim provides a reproducible benchmark for field-wise diagnosis and long-horizon evaluation of geodynamic surrogates.
Groundwater inversion is commonly built on Gaussian assumptions for both network weights and output errors. The assumptions are convenient, but it is unclear which one should be relaxed when hydraulic conductivity is heavy tailed. Here, Gaussian and q-Gaussian distributions were assigned independently to variational weights and the conditional output likelihood in a $2\times2$ Bayesian neural design. Hydraulic conductivity, porosity, and dispersion were inferred jointly from sparse head and transport observations. Four q regimes ($q=1.0$, 1.2, 1.5, and 1.8) were evaluated with five independently generated data sets and matched training seeds, giving 130 held-out cases per regime and 500 posterior draws per case. At $q=1.5$, q-Gaussian weights with a Gaussian likelihood reduced log-conductivity CRPS by 2.85\% (95\% paired CI 0.82--4.88) and calibration error by 3.72 percentage points. Conductivity RMSE was unchanged, while 95\% interval width increased by 9.86\%. At $q=1.8$, assigning q-Gaussian tails to both components produced no stable accuracy gain and increased interval width by 138.46\%. The Gaussian-generated porosity and dispersion controls showed negligible accuracy changes. Moderate weight-side tails can thus improve distributional prediction, whereas an overly heavy output likelihood trades sharpness for coverage. Separating these two roles provides a clearer basis for uncertainty modeling in groundwater inversion.
Mechanical disturbances such as seismic shaking can set a granular slope in motion. Understanding the resulting transport before arrest is important for predicting topographic change, assessing landslide hazards and controlling powder flows in industrial processes. Experiments show that smaller disturbances can trigger avalanches and larger areas are mobilized as slopes approach their stability limit. Yet how much transport varies between events under the same macroscopic conditions and forcing remains unclear. Here we use numerical simulations to show that transport variability is amplified near the critical angle. We apply identical basal pulses to static granular layers with the same macroscopic conditions but different initial particle configurations. Although the mean transport increases approximately in inverse proportion to the angular distance from the critical angle, the standard deviation increases even faster and can exceed the mean. Thus, identical forcing can produce widely different transport amounts. Deterministic constitutive laws that assign a unique response to given macroscopic conditions cannot capture this variability. We introduce fluctuations into the energy balance to develop a stochastic model that captures the main features of the broadening transport distribution. These results show that assessing transport near the critical angle requires considering its distribution, not just its mean, and motivate constitutive descriptions that incorporate fluctuations.
Stable stratification reorganizes turbulence, suppressing vertical motion and altering the balance between vortical and strain-dominated flow structures. How these changes affect the temporal asymmetry of kinetic-energy exchange along fluid-particle trajectories remains incompletely understood. Using direct numerical simulations of homogeneous Boussinesq turbulence, we investigate Lagrangian power statistics across four local velocity-gradient topology classes and three nonzero stratification strengths, with an unstratified reference case. With increasing stratification, the global power variance decreases while the standardized skewness becomes more negative, indicating enhanced asymmetry despite weaker fluctuations. Conditioning on local topology reveals distinct responses: axial-strain-dominated regions exhibit the most negative total-power skewness at all stratification strengths, whereas biaxial-strain-dominated regions change sign from positive to negative under stratification. Horizontal and vertical component powers also respond differently, revealing directional dependence. Among the dynamical contributions to power, the pressure contribution, nearly symmetric in the unstratified case, becomes more negatively skewed with increasing stratification. Energy-conditioned pressure statistics further indicate altered pressure-mediated energy redistribution. Finally, tracer persistence-time distributions exhibit exponential tails, with time scales that differ across topology classes and respond differently to stratification. Increasing stratification is thus accompanied by three concurrent changes: enhanced asymmetry of Lagrangian power, modified energy redistribution by pressure, and altered temporal sampling of local flow structures by tracers. These co-occurring signatures characterize the response to stratification, though their causal relations remain to be established.