Earth Surface Dynamics最新文献

{"title":"Pliocene shorelines and the epeirogenic motion of continental margins: a target dataset for dynamic topography models","authors":"A. Hollyday, M. Raymo, J. Austermann, Fred D Richards, M. Hoggard, Alessio Rovere","doi":"10.5194/esurf-12-883-2024","DOIUrl":"https://doi.org/10.5194/esurf-12-883-2024","url":null,"abstract":"Abstract. Global mean sea level during the mid-Pliocene epoch (∼3 Ma), when CO2 and temperatures were above present levels, was notably higher than today due to reduced global ice sheet coverage. Nevertheless, the extent to which ice sheets responded to Pliocene warmth remains in question owing to high levels of uncertainty in proxy-based sea level reconstructions as well as solid Earth dynamic models that have been used to evaluate a limited number of data constraints. Here, we present a global dataset of 10 wave-cut scarps that formed by successive Pliocene sea level oscillations and which are observed today at elevations ranging from ∼6 to 109 m above sea level. The present-day elevations of these features have been identified using a combination of high-resolution digital elevation models and field mapping. Using the MATLAB interface TerraceM, we extrapolate the cliff and platform surfaces to determine the elevation of the scarp toe, which in most settings is buried under meters of talus. We correct the scarp-toe elevations for glacial isostatic adjustment and find that this process alone cannot explain observed differences in Pliocene paleo-shoreline elevations around the globe. We next determine the signal associated with mantle dynamic topography by back-advecting the present-day three-dimensional buoyancy structure of the mantle and calculating the difference in radial surface stresses over the last 3 Myr using the convection code ASPECT. We include a wide range of present-day mantle structures (buoyancy and viscosity) constrained by seismic tomography models, geodynamic observations, and rock mechanics laboratory experiments. Finally, we identify preferred dynamic topography change predictions based on their agreement with scarp elevations and use our most confident result to estimate a Pliocene global mean sea level based on one scarp from De Hoop, South Africa. This inference (11.6 ± 5.2 m) is a downward revision and may imply that ice sheets were relatively resistant to warm Pliocene climate conditions. We also conclude, however, that more targeted model development is needed to more reliably infer mid-Pliocene global mean sea level based on all scarps mapped in this study.\u0000","PeriodicalId":48749,"journal":{"name":"Earth Surface Dynamics","volume":null,"pages":null},"PeriodicalIF":2.8,"publicationDate":"2024-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141806591","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Decadal-scale decay of landslide-derived fluvial suspended sediment after Typhoon Morakot","authors":"G. Ruetenik, K. Ferrier, O. Marc","doi":"10.5194/esurf-12-863-2024","DOIUrl":"https://doi.org/10.5194/esurf-12-863-2024","url":null,"abstract":"Abstract. Landslides influence fluvial suspended sediment transport by changing sediment supply and grain size, which alter suspended sediment concentrations and fluxes for a period of time after landsliding. To investigate the duration and scale of altered suspended sediment transport due to landsliding, we analyzed suspended sediment concentration and water discharge measurements at 87 gauging stations across Taiwan over an 11-year period after Typhoon Morakot, which generated nearly 20 000 landslides in 2009. At each gauging station, we computed annual rating curves to quantify changes over time in the sensitivity of suspended sediment concentrations to water discharge. Among the 40 stations in basins that were impacted by landsliding, the discharge-normalized rating curve coefficient ã was higher than that before Morakot by a factor of 5.1±1.1 (mean ± standard error) in 2010, the first year after Morakot. The rating curve exponent b did not decrease at most stations until a year later (2011), when the average b value was lower than that before Morakot by 0.25±0.05. Across the compilation of gauging stations, post-Morakot changes in discharge-normalized sediment concentration (ã) were positively correlated with landslide intensity for 7 years after Morakot, while post-Morakot changes in the exponent of the discharge–concentration relationship (b) were negatively correlated with landslide intensity from 2011 to 2014. This reflects a tendency for larger changes in ã and b to occur in basins with more intense landsliding. At 26 of these 40 stations, elevated values of ã declined after the initial post-Morakot peak, consistent with a gradual return to pre-Morakot suspended sediment transport conditions. Exponential regressions to these ã values reveal a median characteristic decay time of 8.8 years (interquartile range: 5.7–14.8 years). Values of ã increased more and declined faster in basins with more intense landsliding, with a mean characteristic decay time of 6 years in the basins hit hardest by landsliding. Furthermore, changes in ã and b tended to be larger in basins with more intense landsliding. At stations that were not impacted or only minimally impacted by landsliding, neither ã nor b exhibited systematic responses to Morakot. To quantify the effect of landsliding on sediment discharge, we compared the measured sediment discharges after Morakot to the hypothetical sediment discharges that would have occurred if Morakot had induced no landslides, calculated by applying each station's pre-Morakot rating curve to its post-Morakot water discharge history. This analysis suggests that Morakot-induced landsliding increased sediment discharge by as much as > 10-fold in some basins in the 1–2 years after Morakot. Together, these results indicate that the influence of Morakot-induced landsliding on rating curves was large shortly after Morakot but diminished in less than a decade in most of the study rivers and will be imperceptible in an","PeriodicalId":48749,"journal":{"name":"Earth Surface Dynamics","volume":null,"pages":null},"PeriodicalIF":2.8,"publicationDate":"2024-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141831058","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Exotic tree plantations in the Chilean Coastal Range: balancing the effects of discrete disturbances, connectivity, and a persistent drought on catchment erosion","authors":"Violeta Tolorza, Christian H. Mohr, Mauricio Zambrano-Bigiarini, Benjamín Sotomayor, Dagoberto Poblete-Caballero, Sebastien Carretier, Mauricio Galleguillos, Oscar Seguel","doi":"10.5194/esurf-12-841-2024","DOIUrl":"https://doi.org/10.5194/esurf-12-841-2024","url":null,"abstract":"Abstract. The Chilean Coastal Range, located in the Mediterranean segment of Chile, is a soil-mantled landscape with the potential to store valuable freshwater supplies and support a biodiverse native forest. Nevertheless, human intervention has been increasing soil erosion for ∼ 200 years, culminating in the intense management of exotic tree plantations throughout the last ∼ 45 years. At the same time, this landscape has been severely affected by a prolonged megadrought. As a result, this combination of stressors complicates disentangling the effects of anthropogenic disturbances and hydroclimatic trends on sediment fluxes at the catchment scale. In this study, we calculate decennial catchment erosion rates from suspended-sediment loads and compare them with a millennial catchment denudation rate estimated from detrital 10Be. We then contrast both of these rates with the effects of discrete anthropogenic-disturbance events and hydroclimatic trends. Erosion and denudation rates are similar in magnitude on decennial and millennial timescales, i.e., 0.018 ± 0.005 and 0.024 ± 0.004 mm yr−1, respectively. Recent human-made disturbances include logging operations throughout all seasons and a dense network of forestry roads, thereby increasing structural sediment connectivity. Further disturbances include two widespread wildfires (2015 and 2017) and an earthquake with an Mw value of 8.8 in 2010. We observe decreased suspended-sediment loads during the wet seasons for the period 1986–2018, coinciding with declining streamflow, baseflow, and rainfall. The low millennial denudation rate aligns with a landscape dominated by slow diffusive soil creep. However, the low decennial erosion rate and the decrease in suspended sediment disagree with the expected effect of intense anthropogenic disturbances and increased structural (sediment) connectivity. Such a paradox suggests that suspended-sediment loads, and thus respective catchment erosion, are underestimated and that decennial sediment detachment and transport have been masked by decreasing rainfall and streamflow (i.e., weakened hydroclimatic drivers). Our findings indicate that human-made disturbances and hydrologic trends may result in opposite, partially offsetting effects on recent erosion, yet both contribute to landscape degradation.","PeriodicalId":48749,"journal":{"name":"Earth Surface Dynamics","volume":null,"pages":null},"PeriodicalIF":3.4,"publicationDate":"2024-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141573883","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Role of the forcing sources in morphodynamic modelling of an embayed beach","authors":"Nil Carrion-Bertran, Albert Falqués, Francesca Ribas, Daniel Calvete, Rinse de Swart, Ruth Durán, Candela Marco-Peretó, Marta Marcos, Angel Amores, Tim Toomey, Àngels Fernández-Mora, Jorge Guillén","doi":"10.5194/esurf-12-819-2024","DOIUrl":"https://doi.org/10.5194/esurf-12-819-2024","url":null,"abstract":"Abstract. The sensitivity of a 2DH coastal area (XBeach) and a reduced-complexity (Q2Dmorfo) morphodynamic model to using different forcing sources is studied. The models are tested by simulating the morphodynamic response of an embayed beach in the NW Mediterranean over a 6-month period. Wave and sea-level forcing from in situ data, propagated buoy measurements, and hindcasts, as well as combinations of these different data sources, are used, and the outputs are compared to in situ bathymetric measurements. Results show that when the two models are calibrated with in situ measurements, they accurately reproduce the morphodynamic evolution with a “good” Brier skill score (BSS). The calibration process reduces the errors by 65 %–85 % compared with the default setting. The wave data propagated from the buoy also produce reliable morphodynamic simulations but with a slight decrease in the BSS. Conversely, when the models are forced with hindcast wave data, the mismatch between the modelled and observed beach evolution increases. This is attributed to a large extent to biased mean directions in hindcast waves. Interestingly, in this small tide site, the accuracy of the simulations hardly depends on the sea-level data source, and using filtered or non-filtered tides also yields similar results. These results have implications for long-term morphodynamic studies, like those needed to validate models for climate change projections, emphasizing the need to use accurate forcing sources such as those obtained by propagating buoy data.","PeriodicalId":48749,"journal":{"name":"Earth Surface Dynamics","volume":null,"pages":null},"PeriodicalIF":3.4,"publicationDate":"2024-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141509574","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Equilibrium distance from long-range dune interactions","authors":"Jean Vérité, Clément Narteau, Olivier Rozier, Jeanne Alkalla, Laurie Barrier, Sylvain Courrech du Pont","doi":"10.5194/egusphere-2024-1634","DOIUrl":"https://doi.org/10.5194/egusphere-2024-1634","url":null,"abstract":"<strong>Abstract.</strong> Flow perturbations induced by dune topography affect sediment transport locally, but can also be felt over long distances altering the dynamics of isolated neighbouring dunes downstream. In order to work under optimal conditions that eliminate transverse flow components, collisions and mass exchange between dunes, we study here these long-range interactions using a 2D numerical model where two equal-sized dunes lying on a non-erodible bed are exposed to a symmetric reversing flow. Depending on the initial spacing, dunes either attract or repel each other, to eventually converge towards a steady-state spacing. This equilibrium distance decreases with flow strength and increases with period of flow reorientation and dune size. It is mainly controlled by the reversing dune shape and the structure of the turbulent wake it generates, which continuously modulates the mean shear stress on the downstream dune. Under multi-directional wind regimes, these long-range flow perturbations offer an alternative mechanism for wavelength selection in dune fields with non-erodible interdune areas. Within these dune fields, estimates of mean shear stress could be used to assess the relative migration rate and the state of attraction/repulsion between neighbouring dunes.","PeriodicalId":48749,"journal":{"name":"Earth Surface Dynamics","volume":null,"pages":null},"PeriodicalIF":3.4,"publicationDate":"2024-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141509643","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An empirically-derived hydraulic head model controlling water storage and outflow over a decade in degraded permafrost rock slopes (Zugspitze, D/A)","authors":"Riccardo Scandroglio, Samuel Weber, Till Rehm, Michael Krautblatter","doi":"10.5194/egusphere-2024-1512","DOIUrl":"https://doi.org/10.5194/egusphere-2024-1512","url":null,"abstract":"<strong>Abstract.</strong> While recent permafrost degradation in Alpine peri- and paraglacial slopes has been documented in several studies, only restricted information is available on the respective hydrology. Water boosts permafrost degradation by advective heat transport and destabilizes periglacial mountain slopes. Even if multiple recent rock slope failures indicate the presence of water, only a few studies provide evidence of water availability and related hydrostatic pressures at bigger depths, showing a significant research gap. This study combines a unique decennial data set of meteorological data, snowmelt modeling, and discharge measurements from two rock fractures in a tunnel located ≈ 55 m under the permafrost-affected N-S facing Zugspitze Ridge (2815–2962 m asl). To decipher the hydrological properties of fractures, we analyze inputs, i.e., snowmelt and rainfall, and outputs, i.e., discharge from fractures, baseflow, and no-flow events, detecting flow anomalies. For summer precipitation events, we developed i) a uniform recession curve, ii) an empirical water storage model, and iii) an approximate hydraulic water pressure model according to Darcy’s falling-head law. Extreme events with up to 800 l/d and 58 l/h are likely to fully saturate the observed fractures with corresponding hydraulic heads of up to 40 ± 10 m and to increase fracture interconnectivity. The average daily discharge during snowmelt, 10 l/h, can lead to hydraulic heads up to 27 ± 6 m. Water dynamics suggest hydraulic conductivities in the range of 10−4 m/s, with variations according to the fracture’s saturation. E.g., no-flow and baseflow events indicate unsaturated and partially saturated conditions. Here, we show an empirical fluid flow approximation model of hydrostatic pressure regimes in high-alpine deep-bedrock fractures. Pressures from water accumulation in bedrock reach levels that can weaken or even destabilize rock slopes. This process can easily outpace thermal conductive warming of active layers in the foreseeable future, provide positive feedback on water infiltration, and is crucial for the stability of the rapidly warming alpine permafrost environments.","PeriodicalId":48749,"journal":{"name":"Earth Surface Dynamics","volume":null,"pages":null},"PeriodicalIF":3.4,"publicationDate":"2024-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141509644","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Geomorphic imprint of high mountain floods: Insight from the 2022 hydrological extreme across the Upper Indus terrain in NW Himalayas","authors":"Abhishek Kashyap, Kristen Cook, Mukunda Dev Behera","doi":"10.5194/egusphere-2024-1618","DOIUrl":"https://doi.org/10.5194/egusphere-2024-1618","url":null,"abstract":"<strong>Abstract.</strong> The interaction of tectonics, surface processes, and climate extremes impacts how the landscape responds to extreme hydrological events. An anomalous precipitation event in 2022 occurred during the monsoon season along the lower reaches of the Upper Indus River, resulting in short-lived high-magnitude flooding and socioeconomic disruption downstream. To understand the spatial relationship between the geomorphic response and climatic controls of this flood event, as well as their primary triggers, we performed a landscape analysis using topographic metrics and quantified the causal association between hydro-climatic variables. Temperature anomalies in upstream glaciated sub-catchments had a considerable impact on snow cover distribution, based on our observations. As snow cover changed, glacial melt runoff rose, contributing to increased fluvial stream power after traversing higher-order reaches. The higher-order reaches of the Upper Indus River received an anomalously high amount of precipitation, which, when combined with substantial glacial and melt discharge, contributed to an extreme flood across the high-relief steep gradient channels. The flood-affected regions had a high mean basin ksn and SL-index, including numerous spikes in their magnitudes along their channel profiles downstream. To determine how the lower reaches of the Upper Indus River responded to this flood event, we employed the Enhanced Vegetation Index (EVI) and Normalized Difference Water Index (NDWI) as change indicator metrics. We observed an inverse causal influence of NDWI on EVI and a statistically significant relationship between anomalous stream power and relative EVI, suggesting that downstream channel morphology changed rapidly during this episodic event and highlighting EVI as a useful indicator of geomorphic change. We suggest that this extreme flood event is a result of the interaction of anomalous glacial melt and anomalous precipitation over a high-relief landscape, with a certain causal connection with anomalous temperature over the event duration. The synoptic observations suggest that this meteorological condition involves the interaction of the Indian Summer Monsoon (ISM) and Western Disturbance (WD) moisture fluxes. However, the geomorphic consequences of such anomalous monsoon periods, as well as their influence on long-term landscape change, are still unclear.","PeriodicalId":48749,"journal":{"name":"Earth Surface Dynamics","volume":null,"pages":null},"PeriodicalIF":3.4,"publicationDate":"2024-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141509645","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A machine learning approach to the geomorphometric detection of ribbed moraines in Norway","authors":"Thomas J. Barnes, T. V. Schuler, S. Filhol, K. S. Lilleøren","doi":"10.5194/esurf-12-801-2024","DOIUrl":"https://doi.org/10.5194/esurf-12-801-2024","url":null,"abstract":"Abstract. Machine learning is a powerful yet underutilised tool in geomorphology, commonly used for image-based pattern recognition. Analysing new high-resolution (1–10 m) elevation datasets, we investigate its usefulness for detecting discrete geomorphological features. This study develops a machine-learning-based method for identifying ribbed moraines in digital elevation data and progresses to test its performance versus time-consuming, manual methods. Ribbed moraines share geomorphometric characteristics with other glacial landforms, hence representing a valuable test of our new methodology in terms of differentiating between similar features, and for detecting landforms with similar characteristics. Furthermore, mapping ribbed moraines may provide valuable indications of their origin, a topic of debate within glacial geomorphology. To automatically detect ribbed moraines, we extract simple morphometrics from high-resolution digital elevation model data and mask regions where ribbed moraines are unlikely to form. We then test several machine learning algorithms before examining the best performer (K-means clustering) for three study areas of 15 km2 in Norway. Our results demonstrate a balanced accuracy of 65 %–75 % when validating versus ground-truthing. The performance depends on the availability of high-resolution elevation data in Norway that are needed to resolve the spatial scale of the target (10–100 m). We find the method effective at detecting both fields of ribbed moraines, as well as individual ribbed moraines. We propose pathways for the future implementation of this method on a large scale and for increasing the detail of information gained about detected landforms. In conclusion, we demonstrate K-means clustering as a promising method for detecting ribbed moraines, with great potential to reduce the time needed to produce landform maps.\u0000","PeriodicalId":48749,"journal":{"name":"Earth Surface Dynamics","volume":null,"pages":null},"PeriodicalIF":3.4,"publicationDate":"2024-06-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141366105","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Geomorphic indicators of continental-scale landscape transience in the Hengduan Mountains, SE Tibet, China","authors":"Katrina Gelwick, Sean D. Willett, Rong Yang","doi":"10.5194/esurf-12-783-2024","DOIUrl":"https://doi.org/10.5194/esurf-12-783-2024","url":null,"abstract":"Abstract. Landscapes are sculpted by a complex response of surface processes to external forcings, such as climate and tectonics. Several major river captures have been documented in the Hengduan Mountains, leading to the hypothesis that the region experiences exceptionally high rates of drainage reorganization driven by horizontal shortening and propagating uplift. Here we determine the prevalence, intensity, and spatial patterns of ongoing drainage reorganization in the Hengduan Mountains and evaluate the relative timescales of this transience by comparing drainage divide asymmetry for four geomorphic metrics that operate at different spatial and temporal scales. Specifically, we calculate the migration direction and the divide asymmetry index (DAI) for drainage divides using catchment-restricted topographic relief (CRR), hillslope gradient (HSG), normalized channel steepness (ksn), and normalized channel distance (χ). ksn and χ are both precipitation-corrected to account for the strong precipitation gradient across the region. The different spatial scales of these geomorphic metrics allow us to establish the relative timescales of observed landscape transience in the Hengduan Mountains, where local-scale metrics measure short-term change and integral quantities measure long-term disequilibrium. We find a high incidence of strongly asymmetric divides in all metrics across the Hengduan Mountain region. Although the magnitude of asymmetry varies significantly between metrics, possibly due to a combination of metric-specific thresholds and varying proxy relationships with erosion rate, a majority of divides agree on divide migration direction. Agreement in divide migration direction indicates an actively responding landscape when asymmetry is high and a state of quasi-equilibrium when asymmetry is low. Disagreements between the integral quantity, χ, and the other geomorphic metrics can be explained by different timescales of the underlying geomorphic processes, with χ reflecting a long-term response and CRR, HSG, and ksn capturing short-term perturbations to catchment structure. These perturbations include various transient mechanisms, such as differential tectonic uplift or erodibility, glacial alteration, and river captures. Our work confirms the high incidence of drainage reorganization across the Hengduan Mountains and highlights both transient and stable areas in the landscape with high resolution. We also offer valuable insights into the application of geomorphic metrics that can be generalized and applied to the study of landscape transience and drainage divide asymmetry in other settings.\u0000","PeriodicalId":48749,"journal":{"name":"Earth Surface Dynamics","volume":null,"pages":null},"PeriodicalIF":3.4,"publicationDate":"2024-05-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141108460","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evaluating the accuracy of binary classifiers for geomorphic applications","authors":"M. Rossi","doi":"10.5194/esurf-12-765-2024","DOIUrl":"https://doi.org/10.5194/esurf-12-765-2024","url":null,"abstract":"Abstract. Increased access to high-resolution topography has revolutionized our ability to map out fine-scale topographic features at watershed to landscape scales. As our “vision” of the land surface has improved, so has the need for more robust quantification of the accuracy of the geomorphic maps we derive from these data. One broad class of mapping challenges is that of binary classification whereby remote sensing data are used to identify the presence or absence of a given feature. Fortunately, there is a large suite of metrics developed in the data sciences well suited to quantifying the pixel-level accuracy of binary classifiers. This analysis focuses on how these metrics perform when there is a need to quantify how the number and extent of landforms are expected to vary as a function of the environmental forcing (e.g., due to climate, ecology, material property, erosion rate). Results from a suite of synthetic surfaces show how the most widely used pixel-level accuracy metric, the F1 score, is particularly poorly suited to quantifying accuracy for this kind of application. Well-known biases to imbalanced data are exacerbated by methodological strategies that calibrate and validate classifiers across settings where feature abundances vary. The Matthews correlation coefficient largely removes this bias over a wide range of feature abundances such that the sensitivity of accuracy scores to geomorphic setting instead embeds information about the size and shape of features and the type of error. If error is random, the Matthews correlation coefficient is insensitive to feature size and shape, though preferential modification of the dominant class can limit the domain over which scores can be compared. If the error is systematic (e.g., due to co-registration error between remote sensing datasets), this metric shows strong sensitivity to feature size and shape such that smaller features with more complex boundaries induce more classification error. Future studies should build on this analysis by interrogating how pixel-level accuracy metrics respond to different kinds of feature distributions indicative of different types of surface processes.\u0000","PeriodicalId":48749,"journal":{"name":"Earth Surface Dynamics","volume":null,"pages":null},"PeriodicalIF":3.4,"publicationDate":"2024-05-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140966164","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}