Earth Surface Processes and Landforms最新文献

Detecting river centrelines and estimating river water surface widths are valuable for measuring planform geometry. Extracting the river centreline and water surface width estimation from satellite images enables a better understanding of the river network dynamics and can be useful in predicting future changes. This study introduces a novel approach to detect river centrelines and water surface widths that leverages the powerful feature extraction capabilities of DeepLabV3, a state-of-the-art semantic segmentation model and integrates them with the geometric analysis strengths of the Medial Axis Transform (MAT). Our MAT approach identifies the midpoints of the river to detect the centreline. It calculates the distance from the centreline to the edge of the river and estimates the water surface width of the river. The effectiveness of this approach was validated through case studies of the Sipsey, Coosa, Tennessee and Mississippi Rivers. This approach utilizes Sentinel-1A Synthetic Aperture Radar (SAR) imagery, enabling data acquisition independent of the weather conditions. We validated this approach using the National Hydrography Dataset Plus (NHDPlus) and in situ water surface width measurements from the HYDRoacoustic dataset in support of the Surface Water Oceanographic Topography (HYDRoSWOT) and cross-section width from the United States Geological Survey (USGS). The MAT approach accurately extracted centrelines and estimated water surface widths for the straight and meandering river sections. Quantitatively, the fine-tuned DeepLabV3 model achieved a 0.933 F1-score for water mask extraction, while the resulting centreline RMSEs against NHDPlus ranged from 0.55 m (Sipsey River) to 3.10 m (Mississippi River), and water surface width estimations generally varied by 2–15% from in-situ measurements. The accuracy of the method is high for straight and meandering rivers; however, errors, primarily caused by complex river morphology, increased in the Mississippi River because its braided channel system challenged the ability of MAT to define a consistent centreline and water surface width. However, it exhibited reduced accuracy and significant spatial deviations when applied to complex braided sections of the Mississippi River. This approach will significantly advance the field of planform geometry measurement by providing researchers with reliable, scalable and practical methodologies that can be used to develop robust and efficient tools.

Due to the particular morphology and complex environments in the Three Gorges Reservoir Area (TGRA), the dynamic failure process of rocky reservoir banks is difficult to quantify, especially the transition relationship between progressive deformation and sudden failure. In this context, the Guanmuling dangerous rock mass (GDRM) in the TGRA was taken as an example to study the evolution model of rocky reservoir banks. Combined with the indoor tests and field surveys, the damage evolution and the critical state of GDRM were determined through theoretical deduction. Afterwards, the mechanical parameters of the critical state were extracted, and the numerical calculation was used to analyse the dynamic collapse of the rocky reservoir banks. It can be found that the disintegration of the rocky reservoir banks was caused by the combined action of the bias pressure under the self-weight and the deterioration of the base rock. Moreover, the statistical analysis of the force bond fracture was used to characterize the evolution process of rocky reservoir banks, which can effectively reflect the nonlinear transformation stages. The related contents can provide a novel approach to quantifying the collapse process of unstable rocks in a complex mechanical environment.

Alluvial fans in coastal regions serve as valuable archives of past climate variability, but the potential insights that could be derived from their aggradational histories remain largely unexplored. In this study, the depositional history of the ~60-m-thick Iwatahara terrace deposits along the coastal Tenryu River, central Japan, was reconstructed based on post-IR IRSL dating of K-feldspar grains. Multi-grain measurements of both pIRIR_50/150 and pIRIR_50/225 signals, along with single-grain measurements of pIRIR_50/225, were conducted to estimate depositional ages. Fading corrections were applied using different models, and their validity was evaluated based on the corrected ages of lagoonal muds independently dated to MIS 7c. Among the various combinations of fading correction models and luminescence signals, the close agreement between the pIRIR_50/225 ages corrected using one of the models and the expected age range for MIS 7c led to the selection of this model–signal combination as a reasonable approach for constraining depositional age. The results revealed three distinct phases of fan aggradation during MIS 8–6: 255–245 ka, 220–210 ka, and a major episode at 180–160 ka. Notably, the 180–160 ka deposits overlie those attributed to the MIS 7 sea-level highstand (~215 ka), despite an overall sea-level fall after ~200 ka. This stratigraphic relationship may indicate that the effects of fluvial incision driven by sea-level fall were outpaced by a substantially increased sediment supply from upstream during this period. The significant aggradation at 180–160 ka may have been driven primarily by enhanced sediment supply, resulting from intensified East Asian summer monsoon precipitation and widespread slope instability linked to a lowered treeline and reduced vegetation cover during this glacial period. These findings underscore the importance of thick alluvial fan deposits as sensitive recorders of climatic fluctuations.

The Atlantic Coast of south-western Namibia is renowned for its Pliocene to Holocene diamondiferous deposits, which contributed more than 63 million carats of diamonds to the global market. A continuum of fossil shoreline deposits associated with barriers to linear and pocket beaches stretches over circa 110 km in length. Of these, the most highly economic are the fossil linear and pocket beach deposits that are preserved over highly competent Proterozoic bedrock. Potholes and gullies, the prominent bedrock-trapsite features, have controlled diamond concentration. By contrast, the barrier shorelines that developed on unconsolidated fluvial and marine sediments at a palaeo-river mouth are less productive. Here, trapsites are absent and diamond accumulation and retention occurred via interstitial trapping in a coarse gravel framework under highly energetic littoral settings. It therefore follows that diamond enrichment is governed by the sedimentological characteristics of the gravel allied to the dynamics of their depositional environments. We examine the relationship of diamond enrichment with the littoral processes at an environmental and subenvironmental level of a latest Pliocene/earliest Pliocene gravel barrier beach complex (BBC). Results show variability in both diamond concentrations and sizes across the barrier that commensurate with a diversity of morpho-sedimentary dynamics.

The study, a first of its kind, provides insights into the diamond enrichment of a BBC. It demonstrates how detailed sedimentological reconstruction of the barrier environments can improve the exploration efficacy of such diamond-bearing littoral placers.

Pulses of sediment finer than existing channel bed material can modify sediment transport dynamics, but they can also pass without a lasting effect. How a channel adjusts to a sudden influx of finer grained sediment is even more uncertain when sediment pulses are punctuated by extended periods of low sediment supply conditions. Such oscillations in sediment supply can be a byproduct of flood-control dams that only temporarily maintain reservoirs, sending pulses of finer grained sediment each time the reservoir is drained. Using a seasonally operated flood-control dam as a natural experiment, we quantified the entrainment thresholds of cobble-sized sediment through seasonal cycles in sediment supply conditions. The channel bed ranged from being without finer grains when dam gates were closed to being completely buried by them after dam gates opened. We paired tracer clasts embedded with accelerometers with a hydraulic flow model to determine the stresses acting on coarse grains at the time of motion across bed conditions. The influence of the sediment pulses was most evident in reduced variability in Shields numbers at the onset of motion relative to the expectations for a sand-free cobble bed. The observed variance in Shields numbers at the onset of motion best matched the expectation for cobbles moving over a bed of sand, likely reflecting the finer grains from the sediment pulses limiting hiding and protrusion effects on the mobility of coarser grains. While cobbles downstream of the flood-control dam were periodically disturbed, the pulses of finer grains were not able to disrupt the established armour layer.

Rainfall erosivity, expressed in kinetic energy, is determined by intensity, velocity and drop size distribution. In natural precipitation, these properties vary and can be substantially altered by the vegetation before reaching the ground. The splash effect of impacting raindrops on the soil surface can initiate soil erosion. While research in this regard has focussed on relating plant characteristics to erosion processes, there is a lack of studies that attempt to reverse this by predicting throughfall kinetic energy from plant properties. This has been attempted by the Vegetation Splash Factor (VSF), a model solely based on the three-dimensional distribution of vegetation surfaces derived from forest lidar forest data. We conducted a pilot study using the VSF model to validate it with in situ measurements and confirm its general suitability. We found a significant correlation between the observed and predicted effect of vegetation on the kinetic energy of rainfall, which demonstrates the suitability of the VSF, despite being solely based on structural traits. The observed effect of vegetation on rainfall kinetic energy exceeded literature reports, leading to systematic underestimation by the model. Our results showed that the VSF can be used to spatially continuously predict the effect of vegetation on the erosivity of rainfall from high-resolution lidar data. These findings open new possibilities for research on splash erosion under vegetation, shifting the perspective from point-based studies towards area-wide approaches. The simplicity of the approach facilitates adaptation for wider use. The first application of the VSF in a field study has proved that the concept is functional and can illustrate zones of increased potential for soil loss under full vegetation cover. This adds to the methodological tool box for erosion studies and can support decision-makers in forestry and agriculture in the future.

Opedes, H., Fuchs, L.F., Baartman, J.E.M., Mücher, C.A., Kessler, A. & Ritsema, C.J. (2025) Modelling the impact of trenches on soil erosion control using OpenLISEM on Mount Elgon, Uganda. Earth Surface Processes and Landforms, 50(6) e70074. Available from: https://doi.org/10.1002/esp.70074

An error appears in Equation (8) in paragraph 5 of Section 3.5 (Sub-catchment simulation, calibration and scenario analysis) on page 7/17. The hydraulic radius was reported as “R = A×P”. The correct formula is R = A/P. Note that this was a typological error; the correct formula was used in all calculations.

We apologize for this error.

Low-lying coral reef islands are presumed to be highly vulnerable landforms to the effects of climate change. Rising sea levels, changes in wave regimes and reef degradation are all considered key threats to their future persistence and habitability. While a number of studies have examined morphological changes on islands over multidecadal timescales, there is a paucity of high-frequency data from recent years that discern variability in shoreline change trends at the local scale. In this study, we used frequently sampled high-resolution satellite imagery covering the past two decades and analysed the morphological evolution and dynamics of 22 reef islands of the Spermonde Archipelago at the southwest coast of Sulawesi, Indonesia - a location deemed as a climate change hotspot with sea-level rise rates higher than the global average, and anthropogenically affected reef ecosystems. Analysis of 4,329 transects cast across 192 recorded shorelines revealed a balance in erosional and accretionary response. Specifically, 32% of transects were characterized by statistically significant accretion, 29% by erosion and the remaining exhibited no significant change. The magnitude of shoreline changes showed high spatial variability across the archipelago, with marked differences between islands perched on patch reefs on the outer shelf and those in the mid-shelf and nearshore. Archipelago-wide, irrespective of a net gain or loss in land area on islands, accretion was predominant on the western margins, while the eastern margins experienced relatively high degrees of erosion, leading to a westward migration of 55% of the islands on their reef platforms. Collectively, this study provides the first high-resolution shoreline change record for the archipelago, explores contemporary patterns of island morphological change and highlights the importance of high-frequency sampling in reef island studies for understanding projections of island change and efforts towards developing robust adaptation strategies and decision-making.

The mechanism by which loess collapses serves as a critical basis for researching disasters in loess regions. This paper employed the employs element method (DEM) to link the internal mechanisms of collapse with its macroscopic manifestations. We first proposed that the softening of clay particles upon water contact is the direct cause of loess collapse. Additionally, an Enhanced Virtual Clay Method (EVCM) was developed, which integrates the finite element method (FEM) and DEM to construct a numerical model for characterizing macroscopic collapse of loess based on microscopic clay particle behaviours. The results revealed that the collapse settlement of the numerical model was highly consistent with theoretical values. Primarily, the collapse settlement of loess originated from the pore space generated by particle displacement, with secondary contributions from the reduction in the effective volume of loess particles caused by clay softening. The numerical model of loess collapse proposed in this study associates microscopic clay particle behaviours with macroscopic collapse behaviours, opening a new pathway for researching the relationship between the macro–micro-properties of loess.