Earth Surface Processes and Landforms最新文献

{"title":"Interacting Effects of Wood and Channel Geometry on Pool Depth in Gravel-Bed Rivers","authors":"Shayla Triantafillou,&nbsp;Ellen Wohl,&nbsp;Daniel N. Scott,&nbsp;Anna Marshall","doi":"10.1002/esp.70424","DOIUrl":"https://doi.org/10.1002/esp.70424","url":null,"abstract":"<p>The presence of large wood (LW) and logjams as flow obstructions in rivers results in, among other processes and forms, backwater and scour pools. General trends from several regions have shown that as the amount of wood in a river increases pools become more closely spaced for a given section of channel. Despite the clear evidence that wood influences patterns of pool formation, relationships between individual wood characteristics and the associated pool depth are poorly constrained in the field yet may be of most interest in river management and restoration that reintroduces large wood and engineered logjams to create and maintain aquatic habitat. Until the ability to numerically model the interactions between large wood and pool formation improves substantially, understanding and predicting pool dimensions in relation to LW will continue to depend on empirical data from diverse regions. We use data from step-pool (high-gradient) and pool-riffle (low-gradient) settings to explore the relationships between LW and resulting pool depth. Wood exerted a greater influence on pool depth in the pool-riffle site relative to the step-pool site where channel geometry most influenced the bed configuration. Even at the pool-riffle site, LW piece and logjam characteristics explained a minimal amount of pool depth variability, but when paired with channel geometry characteristics, the relationships were stronger. We conclude that bed morphology reflects the interaction between channel geometry and wood characteristics. This suggests that the most effective restoration to create pool habitat will consider how wood will interact with channel geometry.</p>","PeriodicalId":11408,"journal":{"name":"Earth Surface Processes and Landforms","volume":"51 10","pages":""},"PeriodicalIF":2.8,"publicationDate":"2026-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/esp.70424","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149145822","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Geomorphic Imprints of Coseismic Uplift: Knickpoints and Landslides on the Noto Peninsula, Japan","authors":"Haruka Tsunetaka,&nbsp;Wataru Murakami","doi":"10.1002/esp.70434","DOIUrl":"https://doi.org/10.1002/esp.70434","url":null,"abstract":"<p>Coseismic uplift can profoundly reshape landscapes, but the extent to which its spatial pattern is preserved in river profiles and hillslope morphology remains uncertain. Knickpoints and slow-moving landslides may provide complementary geomorphic information regarding episodic vertical and horizontal ground deformation. We examined the relationships between these landforms and the surface-displacement field of the 2024 Noto Peninsula earthquake (Mw 7.5), Japan, using high-resolution topographic and displacement datasets. Knickpoint density tended to increase in watersheds characterised by relatively large and spatially uniform vertical uplift during the 2024 earthquake. River-profile morphology, along with marine-terrace chronology, also suggested long-term fluvial adjustment to repeated coastal uplift and relative base-level lowering. Both slow-moving landslides triggered by the 2024 earthquake and historical slow-moving landslides were concentrated in areas of relatively large coseismic horizontal displacement during the 2024 event. This spatial overlap raises the possibility that repeated earthquakes have contributed to the reactivation and continued deformation of pre-existing landslides in these areas. Taken together, knickpoints and slow-moving landslides provide complementary geomorphic information regarding the vertical and horizontal components of repeated coseismic deformation. Linking these persistent landforms with the well-resolved deformation field from the 2024 earthquake provides a framework for connecting individual seismic events with cumulative landscape adjustment in tectonically active coastal regions.</p>","PeriodicalId":11408,"journal":{"name":"Earth Surface Processes and Landforms","volume":"51 10","pages":""},"PeriodicalIF":2.8,"publicationDate":"2026-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/esp.70434","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149145800","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evolution Process and Failure Mechanism of a Multi-Factor Coupled Loess Landslide: A Case Study in Xiangning County, China","authors":"Penghui Ma,&nbsp;Yifan Xue,&nbsp;Jingye Wang,&nbsp;Yaming Tang,&nbsp;Bo Hong,&nbsp;Luqing Zhao,&nbsp;Zekun Li,&nbsp;Lisen Chen,&nbsp;Tao Ding,&nbsp;Jianbing Peng","doi":"10.1002/esp.70427","DOIUrl":"https://doi.org/10.1002/esp.70427","url":null,"abstract":"<div>\u0000 \u0000 <p>Loess landslides are major geohazards on the Chinese Loess Plateau, where rainfall and engineering activities commonly trigger slope failures that also drive slope-to-gully evolution. This study investigates the ZQ landslide (23 April 2023) in Xiangning County, Shanxi Province, using field surveys, UAV photogrammetry, SBAS-InSAR, dynamic triaxial tests and finite-element simulations to clarify deformation, material behaviour, geomorphic evolution and failure mechanisms. The landslide (200 m long, 308 m wide; 6.0 × 10⁴ m²) developed in a Quaternary loess–palaeosol sequence and showed episodic retrogression, producing three rear scarps. The sliding mass moved from one side of the gully toward the valley bottom, where the deposit partially occupied the gully and induced frontal bulging, showing local adjustment of the gully. SBAS-InSAR revealed central displacement accumulation with peak line-of-sight displacement of −43.3 mm and a mean rate of −41 mm/year; deformation acceleration coincided with rainfall, especially the 71.6 mm total in April 2023. Dynamic triaxial tests show that saturated loess is sensitive to cyclic loading. Higher CSR and loading frequency accelerated plastic-strain accumulation and dynamic strength degradation, providing laboratory evidence for cyclic-load–induced weakening. Scenario-based finite-element modelling indicated progressive weakening under sustained rainfall. Under the coupled rainfall–vibration scenario, deformation became more toe-controlled, with a maximum horizontal displacement of 4.28 m, toe uplift of 4.10 m and a factor of safety of 0.773. The development of a connected plastic zone indicates progressive failure and potential reactivation. The results suggest that rainfall-driven infiltration softening within the loess–palaeosol stratigraphic structure was the main hydrological trigger, whereas mining-related disturbance may have contributed to failure by promoting crack development, deformation accumulation and cyclic weakening of saturated loess. The landslide evolved through creep and cracking, crack propagation, shear localization, retrogressive sliding, frontal bulging, partial gully blockage and postfailure creep, linking slope failure with local gully evolution.</p>\u0000 </div>","PeriodicalId":11408,"journal":{"name":"Earth Surface Processes and Landforms","volume":"51 10","pages":""},"PeriodicalIF":2.8,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149145600","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Estimation of Error Margin in the Quantification of Large Wood Volumes in River From In Situ and UAV Measurements","authors":"Maxime Gillet,&nbsp;Maxime Boivin,&nbsp;Maxime Maltais,&nbsp;Félix Lachapelle,&nbsp;Thomas Buffin-Bélanger","doi":"10.1002/esp.70429","DOIUrl":"https://doi.org/10.1002/esp.70429","url":null,"abstract":"<p>Large wood (LW) affects worldwide rivers by adding complexity to their flow structure, their sediment transport dynamics and by contributing to the spatial and temporal occurrences of fluvial morphologies. As a result, LW also promotes biodiversity by increasing the variability and quantity of fluvial habitats. To fully address the role of LW on fluvial habitat and dynamics, it is a common practice to measure the LW volumes in a river corridor and to produce a LW budget at various spatial and temporal scales. The quantification of LW volume and LW budget remains, however, a difficult task for practical and theoretical reasons. In this paper, we aim at providing insights on the uncertainty of LW volume measurements. We tested two methods (unmanned aerial vehicle [UAV] imageries and in situ measurements and duplicate measurements) to estimate margins of error in the quantification of LW volumes and budget on two rivers of the Gaspé Peninsula (30 000 <span></span><math>\u0000 <semantics>\u0000 <mrow>\u0000 <msup>\u0000 <mrow>\u0000 <mtext>km</mtext>\u0000 </mrow>\u0000 <mrow>\u0000 <mn>2</mn>\u0000 </mrow>\u0000 </msup>\u0000 </mrow>\u0000 <annotation>$$ {mathrm{km}}^2 $$</annotation>\u0000 </semantics></math>). A particular situation in 2012 led to the duplication of LW in situ volume measurements accumulated in the Saint-Jean River. Thus, we were able to compare the volume measurements on 40 identical LW jams (LWJs) twice with only one month of interval. No flood with LW transport capacity was observed between the two field campaigns, which assures that the 40 measurements were made on identical LWJs. In order to improve the data set, four other campaigns of this type were carried out, each time with at least three operators taking the same measurements. Two were done by in situ measurement and two by UAV measurement. The results show high variability in the measurements due to operator bias. This high variability, identified mainly in the in situ measurements (59% for in situ values and 25% for UAV values on the wood budget), was then reduced by applying a new methodology for in situ measurements, which significantly reduced the variability. The estimation of errors in measured volumes is crucial to produce sound and robust LW budget. Our results shed light on the uncertainty of LWJ volume measurement and allow us to propose an equation to estimate the uncertainty of LW budget at various spatial and temporal scales.</p>","PeriodicalId":11408,"journal":{"name":"Earth Surface Processes and Landforms","volume":"51 10","pages":""},"PeriodicalIF":2.8,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/esp.70429","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149145599","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Riverbank Erosion Susceptibility Under Varying Sedimentological and Freeze–Thaw Conditions in Cold Climate Regions","authors":"Tuure Takala,&nbsp;Pauli Kolisoja,&nbsp;Nuutti Vuorimies,&nbsp;Erik van Rooijen,&nbsp;Eliisa Lotsari","doi":"10.1002/esp.70418","DOIUrl":"https://doi.org/10.1002/esp.70418","url":null,"abstract":"<p>High-latitude rivers are undergoing rapid environmental change as climate warming alters seasonal freezing, hydrology and sediment dynamics. This study investigates the erosion susceptibility of riverbank sediments in meandering reaches of three northern rivers: Pulmanki (subarctic Finland), Sävar (subarctic Sweden) and Koita (boreal Finland), focusing on how geotechnical and sedimentological properties and freeze–thaw processes influence erosion. Field sampling included grain size, bulk density and moisture measurements, with selected samples analysed using triaxial tests. At Pulmanki, continuous SoilScout monitoring provided high-resolution data on soil temperature, moisture and movement during freeze–thaw cycles, which were linked to seasonal bank erosion derived from aerial surveys. Additionally, observed Accumulated Freezing Degree Days (AFDDs) from 2021–2024 were compared with mid-century climate projections (2040–2099) from four global climate models under three SSP emission scenarios. Fine-grained sediments dominated all sites, but differences in moisture content, grain size and soil structure strongly influenced shear strength and erosion susceptibility. Sävar sediments were the most erosion-prone, whereas Pulmanki sediments were denser and more resistant, and Koita exhibited pronounced local variability. Water content emerged as the primary control on shear strength, while coarser grain sizes increased resistance. Integration of DEM of Difference (DoD) analyses and in situ monitoring demonstrated that erosion is continuous and spatially variable, with sediment movement occurring during both spring thaw and autumn freezing. These findings highlight the key role of freeze–thaw processes in controlling the timing and mechanisms of bank erosion in cold-region rivers. Climate projections indicate a substantial decline in AFDD under high-emissions scenarios, suggesting shorter and milder winters. This is expected to alter freeze–thaw dynamics and extend periods during which riverbanks remain unfrozen and vulnerable to erosion, potentially increasing erosion frequency in fine-grained, moisture-rich sediments. Collectively, the results show that erosion in seasonally frozen rivers is governed by the combined effects of sediment properties, moisture conditions and freeze–thaw processes. The integrated approach, combining geotechnical testing, in situ monitoring and climate projections, provides a transferable framework for assessing erosion risk in cold-region river systems and supports improved riverbank management under a warming climate.</p>","PeriodicalId":11408,"journal":{"name":"Earth Surface Processes and Landforms","volume":"51 10","pages":""},"PeriodicalIF":2.8,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/esp.70418","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149145741","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Unravelling GLOF Hazards in the Sikkim Himalaya: Simulating Multi-Lake Flood Scenarios Under Extreme Conditions","authors":"Vikas Singh Jadoun,&nbsp;Deepali Gaikwad,&nbsp;Reet Kamal Tiwari,&nbsp; Abhishek","doi":"10.1002/esp.70421","DOIUrl":"https://doi.org/10.1002/esp.70421","url":null,"abstract":"<div>\u0000 \u0000 <p>In recent decades, the Himalayan region has witnessed accelerated glacier retreat, driven by climate change, leading to the formation and expansion of glacial lakes. These lakes pose significant threats in the form of glacial lake outburst floods (GLOFs), which can be triggered by external events such as rockfalls, ice avalanches, intense precipitation and seismic activity. Previous Himalayan GLOF studies largely focused on single-lake breaches, whereas the cascading or simultaneous failure of multiple glacial lakes under regionally correlated triggers remains insufficiently explored. Given that many high-hazard lakes are located within similar hydro-climatic and seismic zones, understanding worst-case GLOF scenarios involving simultaneous breaches from multiple lakes requires detailed investigation. This study addresses this critical gap by rigorously quantifying the impacts of multi-lake GLOF events, considered among the most severe hazards threatening downstream communities. Using two-dimensional hydrodynamic modelling in HEC-RAS, we simulate the hydraulic behaviour of GLOFs originating from three highly vulnerable lakes: South Lhonak Lake, Gurudongmar Lake and Shako Cho Lake. The analysis encompasses seven distinct scenarios (S1–S7), ranging from individual lake outbursts to simultaneous breaches of all three lakes. Our findings reveal maximum flow depths ranging from 5 to 20 m, with flow velocities between 1 and 4 m/s. The worst-case scenario produced a peak discharge of approximately 9400 m³/s at Chungthang, a critical downstream settlement. These findings highlight the potentially catastrophic consequences of multi-lake GLOF events, emphasizing the urgent need for integrated hazard assessment and proactive mitigation planning in vulnerable Himalayan river basins.</p>\u0000 </div>","PeriodicalId":11408,"journal":{"name":"Earth Surface Processes and Landforms","volume":"51 10","pages":""},"PeriodicalIF":2.8,"publicationDate":"2026-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149146906","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Thermokarst Lake Extraction and Dynamics With scSE-U-Net and Polarimetric Sentinel-1 Images: A Case Study in the Yellow River Source Region","authors":"Minjie Yang,&nbsp;Zhengjia Zhang,&nbsp;Chenrui Ni,&nbsp;Mengmeng Wang,&nbsp;Peifeng Ma,&nbsp;Fulong Chen","doi":"10.1002/esp.70423","DOIUrl":"https://doi.org/10.1002/esp.70423","url":null,"abstract":"<div><p>Thermokarst lakes are sensitive surface expressions of permafrost degradation and hydrological reorganization in alpine permafrost regions, but their small size and complex surrounding environments make accurate mapping difficult. Accurate extraction of these lake is fundamental for analysing their spatiotemporal dynamics. We develop a semantic segmentation model based on scSE-U-Net (U-Net with Concurrent Spatial and Channel Squeeze and Excitation), utilizing polarimetric Sentinel-1 images. VGG16 was integrated into the U-Net architecture, and a scSE attention mechanism was introduced to improve feature representation for small water bodies and complex backgrounds. Additionally, this study investigates the variations in lake water and their associations with environmental factors in the Yellow River Source Region (YRSR), with particular attention to thermokarst lakes. Compared with the baseline U-Net, scSE-U-Net achieved moderate improvements, with mIoU increasing by 0.60% for large lakes and 0.09% for small lakes, and mPA increasing by 0.25% for both lake groups. The fused VV+VH+PCA feature scheme produced better lake extraction results, with mIoU and mPA values of 98.33% and 99.23% for large lakes, and 94.16% and 96.01% for small lakes, respectively. Our analysis reveals that the lake areas in the YRSR from 2014 to 2023 initially diminished, then gradually expanded and later exhibited slight fluctuations. Changes in lake number were mainly driven by thermokarst lakes, whereas the number of large lakes remained relatively stable. Lake-area variations were positively correlated with precipitation and temperature, suggesting that climate warming and changes in water supply conditions may jointly influence thermokarst lake dynamics through permafrost degradation, seasonal snowmelt and surface-water redistribution.</p> </div>","PeriodicalId":11408,"journal":{"name":"Earth Surface Processes and Landforms","volume":"51 10","pages":""},"PeriodicalIF":2.8,"publicationDate":"2026-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149146782","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The Influence of Context in the Reach-Scale Distribution of Logjams","authors":"Bridget Livers,&nbsp;Ellen Wohl,&nbsp;Katie Larkin","doi":"10.1002/esp.70426","DOIUrl":"https://doi.org/10.1002/esp.70426","url":null,"abstract":"<div>\u0000 \u0000 <p>We analysed the longitudinal distribution of logjams along 17 subalpine stream reaches in the Southern Rocky Mountains of Colorado, USA. Valley length in individual study reaches ranged from 210 to 700 m. Study reaches included old-growth and younger forest and single-thread and multithread channel planforms. We hypothesized that multithread reaches contain more abundant logjams per 100 m of channel length and 100 m of valley length. We examined three scenarios of large wood recruitment dominated by individual treefall, downstream transport and mass wood introduction, respectively. We calculated the logjam distribution density, or number of logjams per length of stream, and statistically evaluated the spatial distribution of logjams to determine whether they exhibit complete spatial randomness of location. Our hypotheses were not supported: we found no significant differences in logjam distribution densities by channel length or valley length (1.4 jams/100 m single-thread, 1.2 jams/100 m multithread) or valley length (1.6 jams/100 m single-thread, 3 jams/100 m multithread). Total wood load per 100 m of valley length was significantly greater in multithread channels (6.1 m³ vs. 3.1 m³ for single-thread channels). Only two multithread reaches deviate from complete spatial randomness, although most reaches exhibit clustering of logjams at some length scale. Much of the scatter within logjam distribution densities for multithread reaches can be explained by site-specific features such as mass recruitment during blowdowns or removal of logjams during a large flood shortly before our field study. Each of the three scenarios of wood recruitment operates to some degree in the study reaches. The relative importance of each recruitment mechanism, interacting with valley-floor geometry within and upstream from each reach, provides insight into observed logjam distributions.</p>\u0000 </div>","PeriodicalId":11408,"journal":{"name":"Earth Surface Processes and Landforms","volume":"51 10","pages":""},"PeriodicalIF":2.8,"publicationDate":"2026-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149146820","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Debris Flow Susceptibility Mapping Using Explainable Deep Learning in Eastern Hindukush Pakistan","authors":"Kashif Ullah,&nbsp;Chen Xi,&nbsp;Liu Tie","doi":"10.1002/esp.70419","DOIUrl":"https://doi.org/10.1002/esp.70419","url":null,"abstract":"<div>\u0000 \u0000 <p>Debris flows are among the most destructive geomorphological hazards in mountainous regions, causing damage to human settlements, infrastructure and ecosystems. The Eastern Hindu Kush region of Pakistan is particularly vulnerable due to steep topography, intense monsoonal precipitation, active tectonics and increasing anthropogenic disturbance. This study develops an integrated debris flow susceptibility assessment framework by combining convolutional neural networks (CNN) with SHapley Additive exPlanations (SHAP). A total of 13 conditioning factors representing topographic, hydrological, geological, environmental and anthropogenic characteristics were prepared using GIS and remote sensing datasets at a spatial resolution of 30 × 30 m. The CNN model demonstrated strong predictive performance, achieving an area under the curve (AUC) of 0.94, overall accuracy (ACC) of 0.86, precision of 0.84, recall of 0.88, an F1-score of 0.86 and a Kappa coefficient of 0.71. The resulting susceptibility map identified approximately 26.3% of the study area as high to very high susceptibility zones, predominantly concentrated along steep river valleys and highly dissected mountainous terrain. The susceptibility map was further validated by debris flow density analysis, which showed hazard density increasing from 0 events/km² in very low zones to 0.048 events/km² in very high zones, with most historical debris flow events concentrated within the very high susceptibility class. SHAP analysis identified slope, NDVI, rainfall, distance to roads and distance to streams as the most influential conditioning factors controlling debris flow occurrence. The proposed CNN–SHAP framework provides a robust and interpretable approach for regional-scale debris flow susceptibility assessment, with applicability to similar mountainous environments. These findings provide valuable information for hazard mitigation, infrastructure planning and land-use management in data-scarce mountain regions.</p>\u0000 </div>","PeriodicalId":11408,"journal":{"name":"Earth Surface Processes and Landforms","volume":"51 10","pages":""},"PeriodicalIF":2.8,"publicationDate":"2026-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149068485","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deep Learning Models for Mapping Surficial Geology in Selected Physiographic Regions of New York","authors":"Joshua C. Woda,&nbsp;Jason Finkelstein,&nbsp;William E. Odom,&nbsp;John H. Williams","doi":"10.1002/esp.70411","DOIUrl":"https://doi.org/10.1002/esp.70411","url":null,"abstract":"<p>Surficial geologic mapping is required for decisions including infrastructure and water-resource development and management. Deep learning, a type of machine learning that uses training data to self-learn and perform tasks, is explored as a tool to help increase efficiency of the labour- and time-intensive mapping process. Deep learning models were trained, and their potential to aid in mapping surficial geology was explored for two physiographic regions in New York: the high-relief Allegheny Plateau and the low-relief Erie-Ontario Lowlands. Key to the development of deep learning models is the availability of highly detailed surficial geologic maps in each of these regions from which the models can learn. Through exploring different groupings of surficial deposits and their corresponding spatial data, deep learning models were iteratively developed to reproduce published mapping for greater than 79% of the training areas, and with similar accuracy in test areas within the same physiographic region. These models were trained using only two inputs: high-resolution lidar data and previously published surficial geologic maps. This straightforward approach is intended to make the methods and models created easily reproducible using widely accessible datasets. This research describes the strengths and limitations of lidar-derived surficial geologic models and how broadly grouping surficial types, characteristics of different physiographic regions, and testing models in physiographic regions outside of their training areas affect model creation and performance. In addition, the models created could be used as a tool for mapping surficial geology in similar physiographic regions, in addition to establishing a framework for creating similar models elsewhere.</p>","PeriodicalId":11408,"journal":{"name":"Earth Surface Processes and Landforms","volume":"51 9","pages":""},"PeriodicalIF":2.8,"publicationDate":"2026-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/esp.70411","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148994144","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}