Journal of Geophysical Research: Earth Surface最新文献

Noise from calving icebergs, cracking ice, and melting ice dominates the underwater soundscape of glacierized fjords creating one of the loudest recorded ambient ocean environments. While progress has been made toward identifying and describing individual sound sources—including the automatic detection of calving and quantification of ice-mass loss—the relative contributions of multiple, simultaneous processes, and how these contributions evolve over time, remain underexplored, limiting robust interpretation of ice-ocean interactions. Here, we show that unsupervised machine learning separates a series of recordings captured over 8 months into five dominant sound profiles related to glacier activity. We deployed an array of hydrophones approximately 400 m from the terminus of Xeitl Sít’ (LeConte Glacier) in Southeast Alaska and recorded sound regularly between October 2016 and May 2017. Using the k-means clustering algorithm, we cluster spectral shapes of 10,440 background acoustic spectra, defined as the 25th-percentile spectral level of each recording. We identify five distinct acoustic clusters and relate their temporal occurrence to environmental time series including ice movement, meteorology, and oceanographic data. We further link spectral shapes to known glacier sources such as calving and ice melt. Our analysis reveals that these clusters correspond more closely with glacier and ice-mélange activity than with other environmental variables, confirming the dominance of glacier behavior on fjord soundscapes. This research demonstrates the effectiveness of clustering passive acoustic data and provides a framework for analyzing large, complex acoustic data sets of undersampled environments—such as glacierized fjords—to guide interpretation and track changes in dominant environmental processes.

Collisional and frictional shape evolution of coastal and fluvial pebbles has long been at the focus of geomorphological research. Interestingly, the well-known rounded pebble shapes show remarkable similarities all around the world. The almost universal axis ratios observed in naturally occurring pebbles suggest the existence of a stable shape toward which pebbles converge during abrasion. However, no widely accepted and robust explanation for this phenomenon exists to date. The aim of the present work is to provide a novel perspective on the shape evolution of rounded pebbles. The investigation focuses on dominant motions that depend on the shape and the abrasion processes that are expected to be induced by these motions. Motivated by the big picture of shape-dependent motions of pebbles and the corresponding predicted abrasion, a highly intuitive heuristic model is constructed, in which a motion-dependent, selective curvature-driven abrasion reveals a self-exciting process that may occur during the long-term motion. Unlike previous models, the introduced approach suggests an unstable ellipsoidal shape near the axis ratios characterizing natural pebbles. In this state, changes in axis ratios are slower because of the statistical variety of expected motions, whereas for shapes that differ significantly, the self-exciting effect accelerates shape change as a dominant mode of a motion emerges. Experiments were also conducted to validate the most critical predicted behavior of the model.

Changes in precipitation rate (a climate proxy) affect erosion rates in mountain belts, driving river profile adjustment and leading to fluctuations in sediment flux and downstream sediment thickness. Here, we use a numerical model to investigate the response of the mountain-basin system to cyclic climate variations. Model results show that rivers exhibit an erosion saturation effect when the precipitation forcing period $��P�$ exceeds the system response time $��\tau �$, corresponding to low-frequency climate variations. In this state, maximum erosion occurs before the precipitation peak, and the response amplitude is significantly lower than for . Consequently, there exists a specific $��P�/�\tau �$ range that maximizes the amplitude of the sediment flux response before the system reaches erosion saturation. Our modeling also shows that sedimentation amplifies the sediment flux response and accelerates erosion saturation, while parameters such as uplift rate and bedrock erodibility seem to have no influence. Compared to the sediment flux response in the mountain belt, variations in basin sediment thickness consistently lag behind precipitation changes, with a smaller response amplitude when . As the forcing period $��P�$ increases, this lag gradually decreases and eventually approaches zero, while the response amplitude progressively increases. Our modeling suggests that the mountain belt and basin should be analyzed as a coupled system, rather than as independent source and sink regions, because they reach their maximum responses at different forcing periods.

Loess tablelands, plateau-like landforms built of wind-blown dust, store large volumes of sediment on its path from bedrock sources to sedimentary basins. Long-term storage of organic carbon also occurs in buried soils beneath loess tableland summits. The evolution of loess tablelands can be conceptualized as a process of relief generation through loess accumulation, raising tableland summits above local base level, followed by stream dissection driven by the increased local relief. We investigated how these landforms of highly erodible material have persisted over 10,000–100,000 years timescales in the central Great Plains, USA, using stratigraphic data, field measurements of soil strength, geospatial analysis, and numerical modeling. Where subsurface data are adequate, it appears that loess tablelands developed on pre-existing bedrock tablelands. The presence of many closed depressions on tableland summits is the key factor limiting the rate of tableland dissection, rather than an erosion-resistant cap as in bedrock tablelands. These depressions capture most runoff on tableland summits, limiting the drainage area of channels eroding the steep tableland margins. Numerical modeling of loess tableland evolution with the Landlab toolkit produces tablelands similar to those of our study area; tableland summits with many closed depressions persisted substantially longer than those without depressions. Modeling also indicates that depression-breaching by gullies is an important process in tableland dissection, and there is clear evidence of this process in tablelands of our study area.

Badlands are sensitive components of the Earth's surface where weathering, erosion and transport processes can be observed on human timescales. Within the Draix-Bléone Critical Zone Observatory (CZO) in SE France, both water and sediment fluxes and their climatic drivers have been recorded for >35 yrs, making it an ideal natural laboratory to develop a landscape-evolution model (LEM) for badland evolution calibrated with field data. The aim of this study is to develop an LEM that reproduces the intra-annual sediment-flux variability observed in the Draix-Bléone CZO, in particular the transition from transport-limited to supply-limited conditions that occurs during summer, and the associated hysteresis between rainfall and sediment fluxes. The model predicts soil thickness and sediment export at monthly timescales, providing a potential link between “classical” LEMs that pertain to long (≫1 yr) timescales and event-scale models. Sediment supply is limited in the model by winter sediment production induced by frost-cracking. We include the impact of rainfall intensity, identified as the main trigger of sediment mobility both on hillslopes and in streams, to reproduce the dynamics of the transport-limited regime. Parameter calibration is performed using average annual sediment-export and seasonal soil-depth data in specific compartments of the catchment. The progressive increase in model complexity leads to the identification of a minimum number of process laws needed to reproduce the observed badland dynamics. Our model constitutes an important first step toward modeling observed annual morphologic changes in badlands and more accurately predicting badland evolution in the context of anthropogenic climate change.

Rainfall characteristics such as intensity, duration, and frequency are key determinants of the hydrogeomorphic response of a catchment. The presence of non-linear and threshold effects makes the relationship between rainfall variability and geomorphological dynamics difficult to quantify. Yet, this is particularly relevant under predicted exacerbated erosion induced by an intensification of hydroclimatic extremes and ensuing erosional processes. In this study, we evaluate the effects of rainfall temporal variability on catchment morphology and sediment erosion, transport, and deposition across diverse grain sizes, catchment shapes, and climates. Specifically, we simulate multiple rainfall realizations using the modified Bartlett-Lewis rectangular pulse model and assess catchment geomorphic response through the CAESAR-Lisflood landscape evolution model. Virtual catchments are used in the numerical experiments and simulations are conducted over centennial timescales. Simulation results show that higher rainfall temporal variability increases net sediment discharge, domain erosion, and deposition volumes. Particularly, more arid regions respond more actively to rainfall variations and coarser grain size configurations amplify the hydrogeomorphic response. We derive a power-law function linking standardized changes in catchment net sediment discharge and fluctuations in rainfall temporal variability, with a consistent exponent across simulations and supported by long-term observational data. Such quantification of the effects of predicted changes in rainfall patterns on catchment hydromorphological response is crucial to forecasting the implications of expected changes in rainfall patterns for downstream sediment delivery. Results further highlight the need to explicitly account for local variability in rainfall intensification when estimating potential changes in soil erosion fluxes.

Carbonate stable isotope paleoaltimetry, including clumped (Δ₄₇) and oxygen (δ¹⁸O_w) isotopes, is extensively applied across the Tibetan Plateau. However, traditional archives such as lacustrine and soil carbonates typically require extensive sedimentary basins, which are not always available, especially in erosion-dominated regions. This highlights the critical need for more accessible archives to diversify the options available for paleoelevation reconstruction. Widespread terrestrial carbonates such as tufas and speleothems offer a promising alternative, but their utility is often complicated by disequilibrium isotope effects from rapid CO₂ degassing. Therefore, a systematic evaluation of these non-equilibrium archives is critically needed before they can be reliably applied to paleoelevation reconstruction. This study presents δ¹³C, δ¹⁸O, and Δ₄₇ data from modern tufas and speleothems along a 0–3,500 m elevation transect in the Yunnan region, the southeastern margin of the Tibetan Plateau. We find that despite being influenced by disequilibrium effects, site-averaged Δ₄₇ values exhibit a robust positive correlation with elevation, yielding a temperature lapse rate (−5.2 ± 0.4°C/km) consistent with the regional modern air temperature lapse rate (−5.1°C/km), confirming temperature as the primary control on Δ₄₇ variations. Furthermore, carbonate-derived δ¹⁸O_w values show a strong negative correlation with elevation above 1,100 m, reflecting Rayleigh fractionation of oxygen isotopes during precipitation. The broader applicability of these relationships was further evaluated by reconstructing the elevations of Holocene carbonate coatings from the southeastern and central Tibetan Plateau. This study validates the potential of tufas and speleothems as alternative paleoelevation archives and establishes a foundational framework for their application in the Tibetan Plateau.

Longshore sediment transport (LST) is influenced by storms, background wave climate, and shoreline orientation. While shifts in Pacific wave climate have been shown to reverse the LST direction, studies investigating these links in the Atlantic are few despite reported changes in storms and wave climate. Here, a novel technique is applied along 50 km of the Southeastern Atlantic coast of North America where satellite-derived shoreline orientations and wave hindcasts are coupled to model changes in LST from 1984 to 2022. This method assesses how changes in wave climate and storminess influence coherent changes in LST patterns in the Southeast Atlantic and examines the relative influence of storm and non-storm wave conditions on LST. Our analyses reveal that storms drive sporadic shifts in LST indicative of stochastic processes, while non-storm intervals are characterized by coherent shifts to more northern-directed LST from an increasingly southern wave climate and increased wave energy. These results indicate that non-storm waves drive an increasing portion of LST relative to storms over the past ∼40 years and are becoming a dominant driver of shoreline change in the Southeast Atlantic. Furthermore, this work demonstrates that the shift to more southerly non-storm waves is statistically significantly correlated with an equatorial migration of mid-Atlantic basin sea level pressure (SLP) highs. This work is among the first to show that changing wave conditions and LST dynamics in the Atlantic are likely driven by latitudinal shifts in wind-producing SLP gradients, which may significantly alter sediment transport regimes along the southeastern US Atlantic coast.

Coastal depositional environments, particularly salt marshes, are among the morphologies most vulnerable to relative sea-level rise (RSLR). Their ability to withstand changing conditions is closely related to their capacity to accrete sediment, gaining elevation at a rate comparable to RSLR. Sedimentation, decomposition of organic remains, and autocompaction are among the main processes that govern the long-term evolution of marsh elevation, with soil compressibility that controls the salt marsh thickening. In fact, autocompaction may significantly offset elevation gain from sediment deposition. Autocompaction is especially apparent in the shallowest decimeter depth where sediments have the highest porosity and are most compressible. However, the geomechanical characterization of these landforms is uncommon worldwide. Laboratory geotechnical testing faces challenges in terms of (a) reliability due to sampling disturbance of these soft soils, and (b) representativeness of in situ conditions because of the dense root network of halophytes in the upper 0.1–0.3 m depth. Here we quantify for the first time the compressibility of the shallowest decimeters of salt marsh soils by novel field-scale loading tests carried out in representative salt marshes of the Venice Lagoon, Italy. The data collected are analyzed using an advanced 3D coupled flow-consolidation simulator with nonlinear constitutive models. Our analysis reveals that salt marshes exhibit considerable variability in soil compressibility within the upper 1–2 m, with differences of up to 2 orders of magnitude. This variability is attributed to several factors, including lithology, depositional environment, and soil age. Moreover, they are characterized by a preconsolidation stress ranging from 2 to 8 kPa, linked to the specific depositional environment, which plays a major role in their mechanical response. The quantification of these parameters will allow a reliable prediction of the salt marshes fate in the Venice Lagoon following global environmental changes and local human interventions.