Automatic Identifcation of Thaw Slumps Based on Neural Network Methods and Thaw Slumping Susceptibility

Thaw slumping is a periglacial process that occurs on slopes in cold environments,where the ground becomes unstable and the surface slides downhill due to saturation with water during thawing.In this study,GaoFen-1 remote sensing and fused multi-source feature data were used to automatically map thaw slumping landforms in the Beilu River Basin of the Qinghai–Tibet Plateau.The bi-directional cascade network structure was used to extract edges at diferent scales,where an individual layer was supervised by labeled edges at its specifc scale,rather than directly applying the same supervision to all convolutional neural network outputs.Additionally,we conducted a 5-year multi-scale feature analysis of small baseline subset interferometric synthetic aperture radar deformation,normalized diference vegetation index,and slope,among other features.Our study analyzed the performance and accuracy of three methods based on edge object supervised learning and three preconfgured neural networks,ResNet101,VGG16,and ResNet152.Through verifcation using site surveys and multi-data fusion results,we obtained the best ResNet101 model score of intersection over union of 0.85(overall accuracy of 84.59%).The value of intersection over union of the VGG and ResNet152 are 0.569 and 0.773,respectively.This work provides a new insight for the potential feasibility of applying the designed edge detection method to map diverse thaw slumping landforms in larger areas with high-resolution images.

Impact of a retrogressive thaw slump on surrounding vegetation communities in the Fenghuoshan mountains,Qinghai-Tibet Plateau

Under global warming,permafrost around the world is experiencing degradation which is especially so on the Third Pole,the Qinghai-Tibet Plateau(QTP),China.Retrogressive thaw slump(RTS)is one of the thermokarst features caused by rapid degradation of ice rich permafrost,which transforms landforms and threatens infrastructures,and even affects the terrestrial carbon cycle.In this work,vegetation communities surrounding a RTS in the Fenghuoshan Mountains of the interior portion of the Qinghai-Tibet Plateau have been investigated to examine the impact from RTS.This investigation indicates that the occurrence of RTS influences the vegetation community by altering their habitats,especially the soil water content,which forces the vegetation community to evolve in order to adapt to the alterations.In the interior part of RTS where it has been disturbed tremendously,alterations have produced a wider niche and richer plant species.This favors species of a wet environment in a habitat where it was a relatively dry environment of alpine steppe prior to the occurrence of RTS.This study adds to limited observations regarding the impact of RTS to vegetation community on the QTP and helps us to reach a broader understanding of the effects of permafrost degradation as well as global warming.

High-resolution assessment of retrogressive thaw slump susceptibility in the Qinghai-Tibet Engineering Corridor

Under the rapidly warming climate in the Arctic and high mountain areas,permafrost is thawing,leading to various hazards at a global scale.One common permafrost hazard termed retrogressive thaw slump(RTS)occurs extensively in ice-rich permafrost areas.Understanding the spatial and temporal distributive features of RTSs in a changing climate is crucial to assessing the damage to infrastructure and decision-making.To this end,we used a machine learning-based model to investigate the environmental factors that could lead to RTS occurrence and create a susceptibility map for RTS along the Qinghai-Tibet Engineering Corridor(QTEC)at a local scale.The results indicate that extreme summer climate events(e.g.,maximum air temperature and rainfall)contributes the most to the RTS occurrence over the flat areas with fine-grained soils.The model predicts that 13%(ca.22,948 km^(2))of the QTEC falls into high to very high susceptibility categories under the current climate over the permafrost areas with mean annual ground temperature at 10 m depth ranging from-3 to-1℃.This study provides insights into the impacts of permafrost thaw on the stability of landscape,carbon stock,and infrastructure,and the results are of value for engineering planning and maintenance.

Machine learning-based predictions of current and future susceptibility to retrogressive thaw slumps across the Northern Hemisphere

Retrogressive thaw slumps(RTSs)caused by the thawing of ground ice on permafrost slopes have dramatically increased and become a common permafrost hazard across the Northern Hemisphere during previous decades.However,a gap remains in our comprehensive understanding of the spatial controlling factors,including the climate and terrain,that are conducive to these RTSs at a global scale.Using machine learning methodologies,we mapped the current and future RTSs susceptibility distributions by incorporating a range of environmental factors and RTSs inventories.We identified freezing-degree days and maximum summer rainfall as the primary environmental factors affecting RTSs susceptibility.The final ensemble susceptibility map suggests that regions with high to very high susceptibility could constitute(11.6±0.78)%of the Northern Hemisphere's permafrost region.When juxtaposed with the current(2000-2020)RTSs susceptibility map,the total area with high to very high susceptibility could witness an increase ranging from(31.7±0.65)%(SSP585)to(51.9±0.73)%(SSP126)by the 2041-2060.The insights gleaned from this study not only offer valuable implications for engineering applications across the Northern Hemisphere,but also provide a long-term insight into the potential change of RTSs in permafrost regions in response to climate change.

Three‑Dimensional Numerical Modeling of Ground Ice Ablation in a Retrogressive Thaw Slump and Its Hydrological Ecosystem Response on the Qinghai‑Tibet Plateau,China

Retrogressive thaw slumps(RTSs),which frequently occur in permafrost regions of the Qinghai-Tibet Plateau(QTP),China,can cause signifcant damage to the local surface,resulting in material losses and posing a threat to infrastructure and ecosystems in the region.However,quantitative assessment of ground ice ablation and hydrological ecosystem response was limited due to a lack of understanding of the complex hydro-thermal process during RTS development.In this study,we developed a three-dimensional hydro-thermal coupled numerical model of a RTS in the permafrost terrain at the Beilu River Basin of the QTP,including ice–water phase transitions,heat exchange,mass transport,and the parameterized exchange of heat between the active layer and air.Based on the calibrated hydro-thermal model and combined with the electrical resistivity tomography survey and sample analysis results,a method for estimating the melting of ground ice was proposed.Simulation results indicate that the model efectively refects the factual hydro-thermal regime of the RTS and can evaluate the ground ice ablation and total suspended sediment variation,represented by turbidity.Between 2011 and 2021,the maximum simulated ground ice ablation was in 2016 within the slump region,amounting to a total of 492 m^(3),and it induced the reciprocal evolution,especially in the headwall of the RTS.High ponding depression water turbidity values of 28 and 49 occurred in the thawing season in 2021.The simulated ground ice ablation and turbidity events were highly correlated with climatic warming and wetting.The results ofer a valuable approach to assessing the efects of RTS on infrastructure and the environment,especially in the context of a changing climate.

Heterogeneity of Surface Heat Exchange of Slopes and Potential Drivers of the Initiation of Thaw Slump,Qinghai‑Tibet Plateau

In the mountainous permafrost area,most thaw slumps are distributed in north or northeast-facing shady slope areas.It is commonly known that there is a heterogeneity in permafrost between diferent slope aspects,but there has been a lack of detailed measured data to quantitatively evaluate their relationships,and in-depth understandings on how the slope aspects are linked to the distribution of thaw slumps.This study examined the heterogenous thermal regime,soil moisture content,and surface radiation at two slope sites with opposing aspects in a warming permafrost region on the Qinghai-Tibet Plateau(QTP).The results indicate that similar air temperatures(T_(a))were monitored on the two slopes,but there were signifcant diferences in ground temperature and moisture content in the active layer from 2016 to 2021.The sunny slope exhibited a higher mean annual ground surface temperature(T_(s)),and over the fve years the mean annual temperature at the top of permafrost was 1.3–1.4℃warmer on the sunny slope than the shady slope.On the contrary,the near-surface soil moisture content was about 10–13%lower on the sunny slope(~22–27%)than the shady slope(~35–38%)during the thawing season(June–September).Radiation data indicate that signifcantly higher shortwave downward radiation(DR)appeared at the sunny slope site.However,due to the greater surface albedo,the net radiation(Rn)was lower on the sunny slope.Slope aspect also afects the ground ice content due to its infuence on ground temperature,freeze-thaw cycles,and soil moisture.Shady slopes have a shallower burial of ice-rich permafrost compared to sunny slopes.The results highlight greatly diferent near-surface ground thermal conditions at the two slope sites with diferent aspects in a mountainous permafrost region.This helps identify the slope-related causes of increasing thaw slumps and provides a basis for predicting their future development.

Potential of Multi‑temporal InSAR for Detecting Retrogressive Thaw Slumps: A Case of the Beiluhe Region of the Tibetan Plateau

Permafrost degradation due to climate warming is severely reducing slope stability by increasing soil pore water pressure and decreasing shear strength.Retrogressive thaw slumps(RTSs)are among the most dynamic landforms in permafrost areas,which can result in the instability of landscape and ecosystem.However,the spatiotemporal characteristics of surface deformation of RTSs are still unclear,and the potentials of deformation properties in mapping large-scale RTSs need to be further assessed.In this study,we applied a multi-temporal Interferometric Synthetic Aperture Radar(MT-InSAR)method to map the spatiotemporal variations in surface deformation of RTSs in the Beiluhe region of the Tibetan Plateau by using 112 scenes of Sentinel-1 SAR data acquired from 2017 to 2021.The deformation rates of RTSs ranged from−35 to 20 mm/year,and three typical motion stages were inferred by analyzing the deformation variation trend of the headwall of RTSs:stable,abrupt thaw,and linear subsidence.A total of 375 RTSs were identifed in the Mati Hill region by combining InSAR-based deformation results with visual interpretation of optical remote sensing images.Among them,76 RTSs were newly developed,and 26%more than the inventory derived from the optical images alone.This study demonstrated that the combination of InSAR-derived deformation with optical images has signifcant potential for detecting RTSs with high accuracy and efciency at the regional scale.

Enhanced detection of freeze‒thaw induced landslides in Zhidoi county(Tibetan Plateau,China)with Google Earth Engine and image fusion

Freeze‒thaw induced landslides(FTILs)in grasslands on the Tibetan Plateau are a geological disaster leading to soil erosion.These landslides reduce biodiversity and intensify landscape fragmentation,which in turn are strengthen by the persistent climate change and increased anthropogenic activities.However,conventional techniques for mapping FTILs on a regional scale are impractical due to their labor-intensive,costly,and time-consuming nature.This study focuses on improving FTILs detection by implementing image fusion-based Google Earth Engine(GEE)and a random forest algorithm.Integration of multiple data sources,including texture features,index features,spectral features,slope,and vertical‒vertical polarization data,allow automatic detection of the spatial distribution characteristics of FTILs in Zhidoi county,which is located within the Qinghai‒Tibet Engineering Corridor(QTEC).We employed statistical techniques to elucidate the mechanisms influencing FTILs occurrence.The enhanced method identifies two schemes that achieve high accuracy using a smaller training sample(scheme A:94.1%;scheme D:94.5%)compared to other methods(scheme B:50.0%;scheme C:95.8%).This methodology is effective in generating accurate results using only~10%of the training sample size necessitated by other methods.The spatial distribution patterns of FTILs generated for 2021 are similar to those obtained using various other training sample sources,with a primary concentration observed along the central region traversed by the QTEC.The results highlight the slope as the most crucial feature in the fusion images,accounting for 93%of FTILs occurring on gentle slopes ranging from 0°to 14°.This study provides a theoretical framework and technological reference for the identification,monitoring,prevention and control of FTILs in grasslands.Such developments hold the potential to benefit the management of grassland ecosystem,reduce economic losses,and promote grassland sustainability.