Journal of Geophysical Research: Earth Surface最新文献

The advent of machine learning has significantly improved the accuracy of identifying mass movements through the seismic waves they generate, making it possible to implement real-time early warning systems for debris flows. However, we lack a profound understanding of the effective seismic features and the limitations of different machine learning models. In this work, we investigate eighty seismic features and three machine learning models for single-station-based binary debris flow classification and multi-station-based warning tasks. These seismic features, derived from physical and statistical knowledge of impact sources, are grouped into five sets: Benford's law, waveform, spectra, spectrogram, and network. The machine learning models belong to two families: two ensemble models, Random Forest and eXtreme Gradient Boosting (XGBoost); one recurrent neural network model, Long Short-Term Memory (LSTM). We analyzed feature importance from the ensemble models and found that the number and even the types of seismic features are not critical for training an effective binary classifier for debris flow. When using models designed to capture patterns in sequential data rather than focusing on information only in one given window, using the LSTM does not significantly improve the performance of binary debris flow classification task over Random Forest and XGBoost. For the multi-station-based debris flow warning task, the LSTM model predicts debris flow probability more consistently and provides longer warning times. Our proposed framework simplifies machine learning-driven debris flow classification and lays the foundation for affordable seismic signal-driven early warning using a sparse seismic network.

Debris-ice avalanches often exhibit higher mobility than debris avalanches, known as hypermobility. Given the increasing destructive power and frequency of such hazards, understanding their rheology could provide insights into their mobility and flow dynamics. Using 3 mm-sized glass and polystyrene beads instead of debris and ice materials, a series of rheological tests was conducted in a vane rheometer at seven different vane speeds on dry specimens with six different polystyrene volumetric contents. The flow dynamics, including the flow curves and the volume change, were evaluated to deduce governing rheological laws. Results reveal that the rheological behavior follows a non-monotonic friction law with a transitional solid-liquid-like behavior and a monotonic dynamic dilatancy law. A density-induced segregation process has been observed, which interplays with rheology. Further analysis of the results demonstrates that, besides the roles of density and the critical state friction parameter, the dynamic dilatancy and segregation process also contribute to enhance the flow mobility of these bi-density mixtures. The proposed rheological laws can be applied to glacier mountain hazard assessment.

Dunes in the presence of vegetation exhibit a variety of shapes, from barchan to parabolic forms. Given the rapid fluctuations in environmental conditions across space and time, it is challenging to ascertain whether these dune shapes are merely transient or indicative of a dynamic equilibrium between sediment transport and vegetation growth. In this study, plant cover is introduced into a 3D cellular automaton dune model to numerically investigate the influence of vegetation on dune morphodynamics. Numerical simulations show that isolated parabolic dunes are unstable, increasing or decreasing in size according to the volume of vegetated sediment they remobilize downstream or deposit upstream in their horns. When specific boundary conditions are applied, dunes converge on steady states that accurately capture the barchan-parabolic transition. Most of the isolated dune shapes observed in nature are reproduced at steady state in the model by increasing the intensity of vegetation processes, from the typical migrating barchans to the fully stabilized parabolic dunes. As the impact of vegetation increases, the steepness of the steady-state dune slopes changes, and the crest curvature reversal occurs. This ensures that all the longitudinal sections of the dune migrate at the same rate by reorienting the transverse sediment fluxes. In the model parameter space, sudden jumps in steady-state properties are associated with the stabilization of upstream horns or crest curvature reversals. These results illustrate why transitional dune shapes between barchans and parabolic dunes are less common in natural environments where environmental conditions are heterogeneous and variable in both space and time.

Glacial lake outburst floods can transport large volumes of sediment. Where these floods reach the coastline, much of the particulate matter is delivered directly to the marine environment. It has been suggested that offshore deposits, specifically in fjord settings, may provide a faithful record of past outburst flood events. However, a lack of observations means that the mechanics and the timing of sediment transport offshore following a glacial lake outburst event remain poorly constrained. Here, we document the changes in sea surface sediment dynamics following the 28 November 2020 Elliot Lake outburst flood in British Columbia, which transported ∼4.3 × 10⁶ m³ of sediment into an adjacent fjord (Bute Inlet) as a deep nepheloid layer directly following the event. However, analysis of sea surface turbidity using in situ measurements and satellite-derived estimates reveals that changes in fjord-head surface turbidity in the months following the major flood were surprisingly small. The highest measured sea surface turbidity instead occurred 5 months after the initial outburst flood. This delayed increase in seaward sediment flux coincided with the onset of the spring freshet, when the discharge of the rivers feeding Bute Inlet increases each year. We suggest that large quantities of sediment were temporarily stored within the river catchment and were only remobilized when river discharge exceeded a threshold level following seasonal snowmelt. Our results reveal a temporal disconnect, where onshore to offshore transfer of sediment is stepped following a glacial lake outburst flood, which could complicate the sedimentology of subsequent deposits.

Barrier islands are dynamic coastal landforms that can migrate landward from the press of sea-level rise and the pulse of storms. Previous work on barriers largely focuses on landward sediment mobilization, particularly through overwash, while the role of outwash—where sediment is transported seaward—remains underexamined. There exists a lack of direct comparisons between the processes that restore sediment volume and the timescales of recovery following outwash and overwash events. Here, we used high-resolution mapping and in situ and modeled water levels to quantify morphologic change and its relation to inundation at three contrasting sites. Our results demonstrate that outwash can remain a net erosive scar for years after formation, while overwash magnitude, frequency, and thus persistence vary largely depending on the width and elevational resistance of the barrier. When elevational resistance to overtopping is low, we show that intermediate high-water events can contribute as much sediment to island overwash as larger named storms and that these processes are key for outwash recovery. We find that modeled total water level correlates positively with volume change, while discrepancies between modeled and observed water levels implicate runup overwash as the dominant mode of transport. Together, we use these data to suggest a differentiation between overwash and outwash processes and their resulting morphologies in studies that aim to predict the impact of storms on barrier island transgression rates and broader ecological function.

Landslides are geomorphic hazards in mountainous terrains across the globe, driven by a complex interplay of static and dynamic controls. Data-driven approaches have been employed to assess landslide occurrence at regional scales by analyzing the spatial aspects and time-varying conditions separately. However, the joint assessment of landslides in space and time remains challenging. This study aims to predict the occurrence of precipitation-induced shallow landslides in space and time within the Italian province of South Tyrol (7,400 km²). We introduce a functional predictor framework where precipitation is represented as a continuous time series, in contrast to conventional approaches that treat precipitation as a scalar predictor. Using hourly precipitation data and past landslide occurrences from 2012 to 2021, we implemented a functional generalized additive model to derive statistical relationships between landslide occurrence, various static scalar factors, and the preceding hourly precipitation as a functional predictor. We evaluated the resulting predictions through several cross-validation routines, yielding performance scores frequently exceeding 0.90. To demonstrate the model predictive capabilities, we performed a hindcast for a storm event in the Passeier Valley on 4–5 August 2016, capturing the observed landslide locations and illustrating the hourly evolution of the predicted probabilities. Compared to standard early warning approaches, this framework eliminates the need to predefine fixed time windows for precipitation aggregation while inherently accounting for lagged effects. By integrating static and dynamic controls, this research advances the prediction of landslides in space and time for large areas, addressing seasonal effects and underlying data limitations.

Floodplains can significantly impact the routing of flood waves across the landscape, however, their representation in broad-scale water resource and flood prediction models is limited. To identify hydraulically relevant floodplains at scale, we developed a workflow to automatically extract reach-averaged topographic features from high resolution (1-m) LiDAR-derived topographic data. These features were identified from departures in the relationship between hydraulic geometry and flood stage and hypothesized to define and characterize a zone within the floodplain that disproportionately dissipates energy and attenuates floodwaters, called the Energy Dissipation Zone. We applied the workflow in the topographically diverse Lake Champlain Basin in Vermont, USA, and used a K-medoids analysis to cluster reaches into distinct feature-based types that were expected to uniquely route hydrographs. In total, we identified eight clusters of reach types: two that were pre-sorted because of the presence of a waterbody or limited floodplain access and six that reflected variability in reach-averaged mesoscale floodplain features that describe the size and shape of the Energy Dissipation Zone. Reach types had distinct impacts on the attenuation of synthetically derived hydrographs, evaluated using the Muskingum-Cunge method. From these clusters, we propose a Hydraulic Floodplain Classification, which is comparable to other geomorphically defined systems but novel in its focus on the landscape potential to influence flood routing. The automated workflow is repeatable and has the potential to improve the functionality of continental floodplain mapping efforts. Identification of hydraulically effective zones has implications for improved watershed management to meet flood resiliency goals and to improve flood predictions and warnings.

Understanding landscape evolution history and sedimentary dynamics in high mountainous regions is tampered by rapid erosion of the sedimentary archives. Naturally dammed lakes provide unique snapshots of these processes and enable evaluating these processes under climatic conditions different from the present. Marpha Lake, in the Himalayan rain-shadow of the upper Kali Gandaki, central Nepal, with its ∼450 m thick lacustrine sequence provides a rare opportunity to study these processes. Optically Stimulated Luminescence (OSL) of quartz and feldspars was used to date the full sequence of filling, breaching and sediment evacuation of the lake. The results show that the lake initiated at ∼120 ka and sediment accumulated until ∼80 ka, corresponding to the intense monsoon period of Marine Isotope Stage (MIS) 5. The calculated minimum catchment erosion rate during the lake filling is typical of modern erosion rates of the Himalayan rain shadow (∼150 mm/ka). The lake was breached at ∼30 ka and the majority of sediments were evacuated within 10 kyr. Between 80 and 30 ka, there was little sedimentation, corresponding to the Last Glacial period (MIS 2–4) associated with weaker Indian monsoon and possible ice coverage of the lake's drainage basin down to the elevation of the lake. Breaching of the dam may have been the result of ice pressure from the lake and/or ice build-up in the pores within the dam. Thus, the sediments of Marpha Lake provide a fascinating archive for understanding how the interplay between mass movement and climate shaped the Himalayan rain shadow morphology during the Late Quaternary.

Changes in waves and wind caused by climate change would induce changes in the cross-shore distribution of the longshore sediment transport rate, which would lead to morphological changes on the updrift and downdrift sides of coastal structures. Therefore, the impacts of climate change on the cross-shore distributions of the longshore sediment transport rate and the longshore current velocity, which induces sediment transport, were examined at a sandy beach in Japan using a one-dimensional numerical model and 9-year wave and wind data simulated at 2-hr intervals for the present and future climates. Both the present-climate distributions had northward and southward predominant values near the shore and offshore, respectively, as a result of the combination of the southerly and northerly waves. Under the RCP8.5 scenario, the distributions shifted southward in the nearshore region, even though the mean wave direction did not change. This occurred because the significant wave height of the southerly waves decreased more than that of the northerly waves under this scenario. In the offshore region, northward longshore sediment transport became predominant because the number of large southerly waves increased. The results obtained using the peak wave directions differed from those obtained using the mean wave directions. There was a significant shift in the distributions to the south, and the bimodal distributions became unimodal. Future changes in the distributions can be estimated using 1-day interval data instead of 2-hr interval data with an error of 30% in the nearshore region.

Jakobshavn Isbræ (Sermeq Kujalleq) has been the largest single contributor to mass loss from the Greenland Ice Sheet over the past three decades. Previous research emphasizes the dominant role of oceanic forcing, with the recent advance, deceleration and thickening of Jakobshavn attributed to reduced ocean temperatures. Here, we use satellite imagery and remotely sensed data sets of ice surface velocity, ice surface elevation and ice discharge to extend observations of ice dynamics at Jakobshavn Isbræ between 2018 and 2023. We then use in situ oceanic and meteorological data, in combination with modeled estimates of surface runoff, to explore the potential role of climatic forcing over this 5-year period. Our results show that Jakobshavn began to re-accelerate in 2018, with mean annual near-terminus velocity increasing by 49% between 2018 and 2021. The onset of re-acceleration occurred prior to the arrival of warmer water, and was likely facilitated by the near-terminus being close to flotation and thus highly sensitive to reductions in effective pressure. Such reductions likely resulted from ice surface lowering, driven by both negative surface mass balance and dynamic thinning. During winter 2020/2021, ice velocities remained elevated, with sustained thinning and iceberg calving observed. This unusual behavior corresponded with a significant decrease in rigid mélange extent, likely driven by increased ocean temperatures observed in Disko Bay and Ilulissat Icefjord. This study thus further emphasizes the complexity of climatic forcing at the ice-ocean interface, highlighting that both oceanic and atmospheric forcing must be considered when projecting the future behavior of marine-terminating outlet glaciers.