日本海洋地球科学和技术机构

Grain size is an important control on landslide mobility, particularly when mass entrainment is considered. However, field-scale discrete element method (DEM) simulations cannot explicitly resolve individual soil grains, requiring much larger equivalent particles as coarse-grained representations of slope materials. How the equivalent particle size influences the mobility of cohesive, entraining landslides over realistic terrain remains insufficiently understood. To address this gap, we use the 2016 Aso Bridge landslide as a real-terrain reference case and represent the entire slope as a cohesively bonded granular medium, allowing initially stable slope material to fail, become entrained, and contribute to the moving mass. Based on this erodible-slope framework, we conduct controlled virtual landslide experiments with three equivalent particle sizes. The simulations demonstrate that models with smaller equivalent particle sizes produce longer runout, broader spreading, and substantial mass entrainment; using a 2-m travel-distance criterion, the entrained-to-released mass ratio in the smallest-particle model exceeds 45%. The enhanced mobility is observed in both the initially released and entrained masses and is accompanied by higher transient energy-conversion ratios and broader particle travel-distance distributions. The observed equivalent-particle-size dependence is qualitatively consistent with the grain-size-dependent mobility reported for idealized cohesionless granular flows, extending the understanding of such size-dependent mobility to cohesive, entraining landslide systems represented on erodible real terrain. These findings suggest that equivalent DEM particle size is a fundamental modeling parameter that strongly influences simulated landslide mobility, mass entrainment, and energy conversion.

General circulation models of the Coupled Model Intercomparison Project Phase 6 (CMIP6) are examined with respect to their ability to simulate the mean state and variability of the tropical Atlantic and its linkage to the tropical Pacific. While, on average, mean state biases have improved little, relative to the previous intercomparison (CMIP5), there are now a few models with very small biases. In particular the equatorial Atlantic warm SST and westerly wind biases are mostly eliminated in these models. Furthermore, interannual variability in the equatorial and subtropical Atlantic is quite realistic in a number of CMIP6 models, which suggests that they should be useful tools for understanding and predicting variability patterns. The evolution of equatorial Atlantic biases follows the same pattern as in previous model generations, with westerly wind biases during boreal spring preceding warm sea-surface temperature (SST) biases in the east during boreal summer. A substantial portion of the westerly wind bias exists already in atmosphere-only simulations forced with observed SST, suggesting an atmospheric origin. While variability is relatively realistic in many models, SSTs seem less responsive to wind forcing than observed, both on the equator and in the subtropics, possibly due to an excessively deep mixed layer originating in the oceanic component. Thus models with realistic SST amplitude tend to have excessive wind amplitude. The models with the smallest mean state biases all have relatively high resolution but there are also a few low-resolution models that perform similarly well, indicating that resolution is not the only way toward reducing tropical Atlantic biases. The results also show a relatively weak link between mean state biases and the quality of the simulated variability. The linkage to the tropical Pacific shows a wide range of behaviors across models, indicating the need for further model improvement.

Abstract A meter-thick deposit preserved in five piston cores along the southern Kuril Trench provides evidence of an extensive turbidite, interpreted to result from a giant megathrust earthquake in the thirteenth century. The emplacement age of this thick turbidite is determined to be 719–672 years BP based on high-resolution age–depth models constructed from paleomagnetic secular variation data and anchored by a regionally correlated tephra layer and calibrated radiocarbon dates. When accounting for the uncertainty introduced by the smoothing of paleomagnetic data, the modeled age range expands to 752–620 years BP. Nevertheless, both age intervals closely match that of widespread seismogenic tsunami deposits along the Hokkaido coast, supporting the interpretation of a shared trigger of the 13th-century megathrust earthquake. Lateral correlations and volume estimates indicate the deposit extends 100–150 km along the trench axis, with a total volume exceeding 1,700 million m3. Magnetic fabric patterns, along with spatial variations in sedimentary structures and textures, indicate that the main turbidity current entered the trench via the Kushiro Canyon and propagated bilaterally along the trench axis. This study presents the first offshore evidence of a margin-scale turbidite from the > 7000 m-deep Kuril Trench, strengthening land–sea correlations for the 13th-century earthquake and demonstrating the broader applicability of PSV-based chronostratigraphy in ultra-deep trench environments globally.

Connectivity among isolated habitat patches via planktonic larval dispersal is crucial for maintaining the regional diversity of hydrothermal vents. Despite increasing sophistication of techniques for simulating dispersal, a lack of information on biological and behavioural traits of vent-associated species limits the applicability of these methods for inferring connectivity. Here we focus on the role of periodic reproduction on dispersal among hydrothermal vents, as periodic and seasonal spawning has increasingly been observed in a variety of taxa. For generalisability, we simulate the dispersal of larvae under treatments of monthly and continuous release timing at various depths, with consistent behavioural traits. Our results show a highly variable effect of periodicity on the characteristics and distribution of dispersal, which are heavily modified by the dispersal depth and source location. The capacity for reproductive periodicity to impact the among-site dispersal warrants further investigation into its prevalence and timing among vent-associated fauna.

High-temperature corrosion in ammonia-containing environments is an important issue for materials used in ammonia-based energy systems. However, the presence of ammonia introduces complex corrosion behavior due to the coupled effects of temperature, gas composition, and alloy composition, making systematic understanding challenging. In this study, the corrosion behavior of Fe and Fe–Cr alloys was investigated under controlled atmospheres by systematically varying temperature and ammonia (NH3) concentration, combined with a data-driven approach to capture the resulting nonlinear behavior. Pure Fe exhibited the largest mass gain in NH3-free environments, whereas the addition of NH3 suppressed oxidation and significantly reduced the overall corrosion rate. In contrast, Fe–Cr alloys showed a strong dependence on temperature and NH3 concentration, with the dominant reactions shifting between oxidation and nitridation depending on Cr content. To capture this intricate corrosion behavior, Gaussian process regression (GPR) was employed to model corrosion mass gain as a function of temperature, NH3 concentration, and alloy composition. The model successfully reproduced the nonlinear response of corrosion behavior across these variables (R2 > 0.95). Furthermore, Bayesian optimization was applied to propose promising experimental conditions and alloy compositions for improved corrosion resistance. This data-efficient framework enables accurate mapping of corrosion behavior over a wide parameter space with a limited number of experiments and provides practical guidance for material selection in high-temperature ammonia environments.