focalinterest.net

Tag: Cryosphere & Permafrost

  • Sentinel-2 classifies Yellow River winter ice types with high accuracy

    What the study found

    The study found that high-resolution Sentinel-2 optical imagery, combined with a support vector machine (an automated classification algorithm), can classify river ice types in the Inner Mongolia reach of the Yellow River with 94.91% overall accuracy. It also reported changes in the winter 2023–2024 proportions of juxtaposed ice, consolidated ice, and open water.

    Why the authors say this matters

    The authors say the findings provide technical support for faster interpretation of ice conditions in the Yellow River. They also state that the work offers a scientific basis for precise monitoring and disaster prevention and management related to river ice phenomena.

    What the researchers tested

    The researchers developed an optimized classification model for river ice types using Sentinel-2 imagery. The model used multi-band spectral features and multi-spectral fusion indices, including the normalized difference snow index (NDSI) and the normalized difference frozen surface index (NDFSI), as feature vectors, with support vector machine classification.

    What worked and what didn't

    The classification approach achieved an overall accuracy of 94.91%. In winter 2023–2024, the proportion of juxtaposed ice changed from 45% to 55%, consolidated ice changed from 30% to 40%, and open water changed from 9% to 19%.

    What to keep in mind

    The abstract does not describe specific limitations, error sources, or validation details beyond the reported overall accuracy. The summary is limited to the Inner Mongolia section of the Yellow River and to the winter 2023–2024 period.

    • Sentinel-2 imagery was used to classify winter river ice types in the Inner Mongolia reach of the Yellow River.
    • The model combined support vector machine classification with spectral features, NDSI, and NDFSI.
    • The reported overall classification accuracy was 94.91%.
    • The winter 2023–2024 proportions of juxtaposed ice, consolidated ice, and open water all changed.
    • The authors say the work supports faster ice-condition interpretation and river-ice disaster management.

  • Bias-corrected Greenland accumulation maps reduce model errors

    What the study found

    The study found that a statistical-semi-empirical bias-adjustment model can substantially reduce systematic errors in Greenland ice-sheet snow accumulation estimates. It also found improved agreement among several regional climate models and a reanalysis product after adjustment.

    Why the authors say this matters

    The authors state that more accurate accumulation estimates are essential for reliable sea-level rise projections. The study suggests that better integration of observational data could improve inputs to ice-sheet models and help reduce uncertainty in future sea-level rise projections.

    What the researchers tested

    The researchers developed a statistical-semi-empirical model that uses Empirical Orthogonal Function analysis, a method for separating dominant spatial patterns and their time evolution, to bias-correct gridded accumulation output. They fitted the model with SUMup observations and applied it to monthly accumulation from HIRHAM5, MAR3.14, RACMO 2.4p1, and the Copernicus Arctic Regional Reanalysis (CARRA) across different time periods.

    What worked and what didn't

    Initial mean point-wise biases of −7.4% for HIRHAM, −0.5% for MAR, 0.0% for RACMO, and +10.1% for CARRA were reduced to ±0.3% after adjustment. Bias-corrected mean annual accumulation rates over the ice sheet were estimated at 469 mm yr−1, 412 mm yr−1, 435 mm yr−1, and 408 mm yr−1 for HIRHAM, MAR, RACMO, and CARRA, respectively, between 1991 and 2022. Inter-model agreement improved by 68% in the observation-rich accumulation zone but worsened by 27% in the sparsely sampled ablation zone.

    What to keep in mind

    The abstract says that the method performs better where observations are more abundant and worse where observations are sparse. It also notes that the largest statistically significant bias contributions come from the southern ice sheet, and that additional observational constraints are needed in the ablation zone.

    • A bias-adjustment model reduced mean point-wise errors for Greenland accumulation maps to ±0.3%.
    • The method was applied to HIRHAM5, MAR3.14, RACMO 2.4p1, and CARRA output.
    • Inter-model agreement improved by 68% in the accumulation zone.
    • Agreement worsened by 27% in the sparsely sampled ablation zone.
    • The authors say more observational data are needed in the ablation zone.

  • Freeze–thaw cycling alters concrete–soil interface shear behavior

    What the study found

    The study found that freeze–thaw cycling changes the shear behavior of concrete–crushed rock–soil interfaces and that a four-parameter modified Duncan–Chang model can describe the observed deformation better than traditional hyperbolic models. The authors also report that interface strength depends on both normal stress and freeze–thaw history.

    Why the authors say this matters

    The authors say interfacial strength characterization is important for evaluating the durability of permafrost infrastructure. They also conclude that better constitutive modeling may help with numerical simulation and theoretical calculation of structures in frozen ground.

    What the researchers tested

    The researchers carried out laboratory direct shear tests on concrete–crushed rock soil interfaces under freeze–thaw cycles. They analyzed shear stress–displacement behavior across cycle numbers, used nuclear magnetic resonance to measure interfacial pore structure evolution, and developed a modified Duncan–Chang constitutive model with freeze–thaw damage effects.

    What worked and what didn't

    The proposed four-parameter model, with cubic–hyperbolic cyclic damage corrections, achieved R2 greater than 0.95 in nonlinear least-squares validation. It was reported to capture plastic hardening and incipient strain softening, while traditional one- or two-parameter hyperbolic models were described as less able to represent the freeze–thaw-damaged interface behavior.

    What to keep in mind

    The abstract does not describe detailed experimental limits beyond the tested interface type and freeze–thaw conditions. It also does not provide information about field validation outside the laboratory setting.

    • Freeze–thaw cycling changes shear behavior at concrete–crushed rock–soil interfaces.
    • A modified four-parameter Duncan–Chang model fit the data with R2 greater than 0.95.
    • Interfacial shear strength depended on both normal stress and freeze–thaw history.
    • Nuclear magnetic resonance showed progressive pore-structure evolution during cycling.
    • The authors report a two-stage energy decoupling mechanism for interfacial debonding and friction.

  • Global glacial lake volume increased since 1990

    What the study found

    The study found that about 71,000 glacial lakes worldwide stored an estimated 2,048 km³ of water in 2020, which is a 12.7% increase compared with 1990. The authors also found that lake size and location strongly affect how long this stored meltwater may remain available.

    Why the authors say this matters

    The authors conclude that these lakes are increasingly important freshwater reservoirs. They say the findings help define a window of opportunity in which growing water demands must be balanced with hazard mitigation and protection of rapidly changing high-mountain ecosystems.

    What the researchers tested

    The researchers estimated the volumes and sediment storage capacities of roughly 71,000 glacial lakes globally as of 2020. They also compared the 2020 lake water volume with 1990 and examined how lake longevity varies by size and region.

    What worked and what didn't

    Half of the 2020 glacial lake water volume was located within 63 km of a coastline and below 200 m above sea level, mostly in sparsely populated, high-latitude regions such as Greenland, Arctic Canada, Patagonia, and Alaska. The smallest lakes, which make up about 80% of all lakes and are smaller than 0.1 km², could lose 10% of their storage capacity within a century because of sedimentation, while the 40 largest lakes, which hold half of the global volume, could last for tens of thousands of years.

    What to keep in mind

    The abstract describes estimates and projected lifespans, so the values are not direct measurements for every lake. It also notes that demand for freshwater remains limited in some high-latitude regions, and that unstable dams in High Mountain Asia could rapidly reduce some of the available capacity.

    • About 71,000 glacial lakes worldwide stored an estimated 2,048 km³ of water in 2020.
    • This amount was 12.7% higher than the 1990 estimate.
    • Half of the global lake water volume lies near coasts and below 200 m elevation, mostly in sparsely populated high-latitude regions.
    • Small lakes under 0.1 km² may lose 10% of storage capacity within a century from sedimentation.
    • The 40 largest lakes hold half of the global volume and could persist for tens of thousands of years.

  • Freshwater and organic carbon exports increased across northern Alaska estuaries

    What the study found

    The study found increasing freshwater discharge and dissolved organic carbon export from northern Alaska to coastal waters over more than four decades. It also found that surface, supra-permafrost, and subsurface flows all increased, with a larger summer and autumn share coming from subsurface flow.

    Why the authors say this matters

    The authors say these changes could substantially affect salinity and trophic conditions, meaning the balance of salt and the food web, along Alaska's Beaufort Sea coast. They also conclude that model-data syntheses are valuable where observations are sparse and can provide export metrics useful to researchers, resource managers, and other stakeholders.

    What the researchers tested

    The researchers used an updated numerical process model to estimate freshwater discharge and dissolved organic carbon export from the North Slope of Alaska to the Beaufort Sea. The model was applied at 1 resolution across the 166,483 square kilometer domain from 1980 to 2023, with watershed inputs routed along river networks to 1,039 outlets at the land-sea boundary.

    What worked and what didn't

    The model was used to quantify spatial and temporal patterns of export and to examine climate-linked changes. It showed that exports to specific lagoons, bays, and sounds varied with landscape composition, terrestrial drainage basin size, and estuary area, and that freshwater discharge and dissolved organic carbon export increased over time alongside changes in precipitation and permafrost thaw.

    What to keep in mind

    The abstract says measured data are sparse across the region, which is why a model was needed. It does not describe specific limitations of the model beyond that scope constraint.

    • Freshwater discharge and dissolved organic carbon export increased from the North Slope of Alaska to the Beaufort Sea over 1980 to 2023.
    • The increases were associated with changes in precipitation and permafrost thaw.
    • Surface, supra-permafrost subsurface, and subsurface fluxes all increased.
    • Subsurface flow made a larger proportional contribution in summer and autumn.
    • Exports differed among lagoons, bays, and sounds depending on landscape composition, drainage basin size, and estuary area.