Thermokarst Landforms Research Articles

The Response of Soil Respiration to Temperature and Humidity in the Thermokarst Depression Zone of the Headwater Wetlands of Qinghai Lake.

As the climate warms, the thickening of the active layer of permafrost has led to permafrost melting and surface collapse, forming thermokarst landforms. These changes significantly impact regional vegetation, soil physicochemical properties, and hydrological processes, thereby exacerbating regional carbon cycling. This study analyzed the relationship between soil respiration rate (Rs), soil temperature (T), and volumetric water content (VWC) in the thermokarst depression zone of the headwater wetlands of Qinghai Lake, revealing their influence on these soil parameters. Results showed a significant positive correlation between soil temperature and Rs (p < 0.001), and a significant negative correlation between VWC and Rs (p < 0.001). The inhibitory effect of VWC on Rs in the thermokarst depression zone was stronger than under natural conditions (p < 0.05). Single-factor models indicated that the temperature-driven model had higher explanatory power for Rs variation in both the thermokarst depression zone (R2 = 0.509) and under natural conditions (R2 = 0.414), while the humidity-driven model had lower explanatory power. Dual-factor models further improved explanatory power, slightly more so in the thermokarst depression zone. This indicates that temperature and humidity jointly drive Rs. Additionally, during the daytime, temperature had a more significant impact on Rs under natural conditions, while increased VWC inhibited Rs. At night, the positive correlation between Rs and temperature in the thermokarst depression zone increased significantly. The temperature sensitivity (Q10) values of Rs were 3.32 and 1.80 for the thermokarst depression zone and natural conditions, respectively, indicating higher sensitivity to temperature changes at night in the thermokarst depression zone. This study highlights the complexity of soil respiration responses to temperature and humidity in the thermokarst depression zone of Qinghai Lake's headwater wetlands, contributing to understanding carbon cycling in wetland ecosystems and predicting wetland carbon emissions under climate change.

The first 10 m resolution thermokarst lake and pond dataset for the Lena Basin in the 2020 thawing season

ABSTRACT Climate warming rates in the Arctic are far greater than the global average, exerting stronger impacts on permafrost degradation and thermokarst landform development. Thermokarst lakes and ponds (TLPs), which are widely distributed in the Lena Basin in the Russian Arctic, play a vital role in altering local ecosystem. However, the detailed distribution of TLPs in the Lena Basin still remains poorly known. In this study, we built the first 10 m resolution TLP dataset for the Lena Basin in the 2020 thawing season by utilizing 4902 Sentinel-2 images. A robust mapping workflow was developed and implemented in the Google Earth Engine (GEE) platform. The accuracy assessment demonstrates a satisfactory accuracy (93.63%), and our results exhibit a better consistency with real TLPs than global water body products. A total of 380,477 TLPs (~0.53% of the total surface area of the Lena Basin) were identified, showing an uneven distribution in the five sub-basins. The TLPs were found to be mainly located within plain areas, with an active layer thickness in the range of 80–100 cm. The higher ground ice content and mean annual ground temperature were favorable for TLP development. This dataset will be valuable for investigating the complex interaction between TLPs and permafrost. It will also serve as a baseline product for better incorporating thermokarst processes into permafrost-climate models.

Observation of spatial and temporal patterns of seasonal ground deformation in central Yakutia using time series InSAR data in the freezing season

The seasonal freeze/thaw variations of the active layer cause seasonal ground deformation, such as thaw subsidence in summer and frost heave in winter due to the cyclic phase changes of soil water. The Synthetic Aperture Radar (SAR) remote sensing, especially interferometric SAR (InSAR) techniques, has been proven to be a very useful tool for observing surface displacements of permafrost environments. Most previous studies on seasonal InSAR displacement have been concentrated in several study sites in the Arctic tundra regions. The objective of this study is to extend the InSAR analysis of seasonal ground deformations to boreal forest regions, where monitoring ground deformation is particularly important in terms of better understanding active layer dynamics under various ecological and geological conditions. In this study, we presented a strategy to analyze the time series SAR data in the freezing season to estimate the differential phase for seasonal ground deformation while overcoming the problem of coherence losses due to the rapid growth of vegetation during the short growing season. We applied the SBAS-based InSAR processing method to Sentinel-1 SAR in the central Yakutian lowlands, where boreal forests and thermokarst landforms are widely distributed, and retrieved frost-related displacement signals successfully. In addition to the magnitude of ground deformation in the freezing season, we further examined the temporal deformation pattern related to the progression of the freezing front in the soil during active freezing. Based on the Stefan solution, we proposed a piecewise linear model for the time series deformation. The proposed model divides the time series displacement according to the progression of freezing into three linear segments, and the slope parameter of each linear model can be interpreted as indicating the hydrological characteristics of the upper, middle, and lower parts of the soil. The results for the Yakutian lowlands illustrated the potential of SAR remote sensing to observe and quantify spatial details of the thermal and hydrological properties inside the active layer soil even in highly vegetated subarctic boreal forest areas.

Evidence for preservation of organic carbon interacting with iron in material displaced from retrogressive thaw slumps: Case study in Peel Plateau, western Canadian Arctic

In northern high latitudes, rapid warming is set to amplify carbon-climate feedbacks by enhancing permafrost thaw and biogeochemical transformation of large amounts of soil organic carbon. However, between 30 % and 80 % of permafrost soil organic carbon is considered to be stabilized by geochemical interactions with the soil mineral pool and thus less susceptible to be emitted as greenhouse gases. Quantification of the nature of and controls on mineral-organic carbon interactions is needed to better constrain permafrost-carbon-climate feedbacks, particularly in ice-rich environments resulting in rapid thaw and development of thermokarst landforms. On sloping terrain, mass wasting features called retrogressive thaw slumps are amongst the most dynamic forms of thermokarst. These multi-decadal disturbances grow due to ablation of an ice-rich headwall, and their enlargement due to warming of the Arctic is mobilizing vast stores of previously frozen materials. Here, we investigate headwall profiles of seven retrogressive thaw slumps and sediments displaced from these mass wasting features from the Peel Plateau, western Canadian Arctic. The disturbances varied in their headwall height (2 to 25 m) and affected land surface area (<1 to > 30 ha). We present total and water extractable mineral element concentrations, mineralogy, and mineral-organic carbon interactions in the headwall layers (active layer, permafrost materials above an early Holocene thaw unconformity, and Pleistocene-aged permafrost tills) and in displaced material (suspended sediments in runoff and material accumulated on the debris tongue). Our data show that the main mechanism of organic carbon stabilization through mineral-organic carbon interactions within the headwall is the complexation with metals (mainly iron), which stabilizes 30 ± 15 % of the total organic carbon pool with higher concentrations in near-surface layers compared to deep permafrost. In the displaced material, this proportion drops to 18 ± 5 %. In addition, we estimate that up to 12 ± 5% of the total organic carbon is stabilized by associations to poorly crystalline iron oxides, with no significant difference between near-surface layers, deep permafrost and displaced material. Our findings suggest that the organic carbon interacting with the sediment mineral pool in slump headwalls is preserved in the material mobilized by slumping and displaced as debris. Overall, up to 32 ± 6 % of the total organic carbon displaced by retrogressive thaw slumps is stabilized by organo-mineral interactions in this region. This indicates that organo-mineral interactions play a significant role in the preservation of organic carbon in the material displaced from retrogressive thaw slumps over years to decades after their development resulting in decadal to centennial scale sequestration of this retrogressive thaw slump-mobilized organic carbon interacting with the soil mineral pool.

A transfer learning approach for automatic mapping of retrogressive thaw slumps (RTSs) in the western Canadian Arctic

ABSTRACT Retrogressive thaw slumps (RTSs) are thermokarst landforms that develop on slopes in permafrost regions when thawing permafrost causes the land surface to collapse. RTSs are an indicator of climate change and pose a threat to human infrastructure and ecosystems in the affected areas. As the availability of ready-to-use high-resolution satellite imagery increases, automatic RTS mapping is being explored with deep learning methods. We employed a pre-trained Mask-RCNN model to automatically map RTSs on Banks Island and Victoria Island in the western Canadian Arctic, where there is extensive RTS activity. We tested the model with different settings, including image band combinations, backbones, and backbone trainable layers, and performed hyper-parameter tuning and determined the optimal learning rate, momentum, and decay rate for each of the model settings. Our final model successfully mapped most of the RTSs in our test sites, with F1 scores ranging from 0.61 to 0.79. Our study demonstrates that transfer learning from a pre-trained Mask-RCNN model is an effective deep learning model that has the potential to be applied for RTS mapping across the Canadian Arctic.

Patterns and drivers of anaerobic nitrogen transformations in sediments of thermokarst lakes.

Significant attention has been given to the way in which the soil nitrogen (N) cycle responds to permafrost thaw in recent years, yet little is known about anaerobic N transformations in thermokarst lakes, which account for more than one-third of thermokarst landforms across permafrost regions. Based on the N isotope dilution and tracing technique, combined with qPCR and high-throughput sequencing, we presented large-scale measurements of anaerobic N transformations of sediments across 30 thermokarst lakes over the Tibetan alpine permafrost region. Our results showed that gross N mineralization, ammonium immobilization, and dissimilatory nitrate reduction rates in thermokarst lakes were higher in the eastern part of our study area than in the west. Denitrification dominated in the dissimilatory nitrate reduction processes, being two and one orders of magnitude higher than anaerobic ammonium oxidation (anammox) and dissimilatory nitrate reduction to ammonium (DNRA), respectively. The abundances of the dissimilatory nitrate reduction genes (nirK, nirS, hzsB, and nrfA) exhibited patterns consistent with sediment N transformation rates, while α diversity did not. The inter-lake variability in gross N mineralization and ammonium immobilization was dominantly driven by microbial biomass, while the variability in anammox and DNRA was driven by substrate supply and organic carbon content, respectively. Denitrification was jointly affected by nirS abundance and organic carbon content. Overall, the patterns and drivers of anaerobic N transformation rates detected in this study provide a new perspective on potential N release, retention, and removal upon the formation and development of thermokarst lakes.

Ecosystems of Alas Landscapes—The Basis for the Development of Cattle Breeding in the Harsh Natural and Climatic Conditions of the Permafrost Zone

Alas landscapes are unique ecosystems, which are dynamic, geochemically closed thermokarst landforms of the permafrost zone. Alases have a limited capacity in their active layer, and specific conditions for soil, flora and fauna formation. A comprehensive study of alas landscape functionswas carried out in Central Yakutia from 1988 to the present time using conventional methods of geobotany, zoology, entomology, etc. This paper presents long-term observations of lake fluctuation cycles and changes in the spatial structure of meadow spaces. The dynamics of the spatial structure lead to significant fluctuations in the productivity of alas phytocenoses. It was revealed that wet and normal alas meadows have the highest vegetation productivity. The long-term course of their productivity tends to decrease, which shows the influence of anthropogenic pressure since the main haymaking areas are located in these meadows. With sharp fluctuations in interannual weather conditions, which determine the microclimatic and soil characteristics of grass growth, the productivity of the edge phytocenoses tends to increase. The productivity of the steppe phytocenoses of the alas remains practically at the same level. Over the years of observation, the economic capacity of alas pastures and hayfields was calculated. Additionally, the paper presents the important role of fauna within closed alas ecosystems, which directly affects the functioning of alas landscapes and is directly involved in soil formation and the circulation of matter and energy.

Interaction of permafrost degradation and thermokarst lakes in the Qinghai–Tibet Plateau

The degradation of ice-rich permafrost can form thermokarst landforms such as thermokarst lakes. There are a large number of thermokarst lakes distributed in the ice-rich permafrost region of the Qinghai–Tibet Plateau. However, the effect of the spatial distribution of thermokarst lakes on permafrost degradation at a regional-scale in the Qinghai–Tibet Plateau is not fully understood. Time-series interferometric synthetic aperture radar is an established remotely-sensing tool to measure surface deformation in permafrost regions, and has been widely used in permafrost degradation investigation. In this study, we characterized spatial-temporal variations of surface deformation in the Beiluhe Basin in the Qinghai–Tibet Plateau with time series InSAR analysis, and extracted thermokarst lakes from Landset-8 images. Additionally, MODIS land surface temperature data and Stefan model were used to estimate the active layer thickness of permafrost in the basin. Results showed that the linear subsidence rate in the permafrost region within the basin generally exceeded 5 mm/yr, with a seasonal deformation amplitude of >10 mm, and the active layer thickness ranged from 2.1 m to 3.5 m. We found that permafrost degradation was concentrated in the areas with a high density of thermokarst lakes. Specifically, the density of thermokarst lakes was positively correlated with the linear subsidence rate and the seasonal deformation amplitude, and the larger active layer thickness clustered significantly in the area of thermokarst lakes. We also found that accelerated thermokarst collapse of the lake shore intensified the development of thermokarst lakes, and the lake expansion further aggravated the permafrost deformation. In addition, there is a strongly negative correlation (r = −0.78) between active layer thickness and Normalized Difference Vegetative Index in the basin, inferring that the degradation of permafrost may lead to the transformation from alpine meadow to alpine grassland and alpine desert.

MT-InSAR Unveils Dynamic Permafrost Disturbances in Hoh Xil (Kekexili) on the Tibetan Plateau Hinterland

Hoh Xil is an uninhabited extremity secluded on the Tibetan Plateau hinterland. A complete mapping of ground motion variation in Hoh Xil is essential for in-depth understanding of terrain’s responses to climate change on the Tibetan Plateau. However, the inaccessibility and extremely harsh environment impeded extensive field investigations on landform alteration and its formative process. Such difficulty can be resolved by Interferometric Synthetic Aperture Radar (InSAR), which enables a broad detection of subtle permafrost motions at millimeter precision. This study, for the first time, accomplished a Multi-temporal InSAR (MT-InSAR) deformation mapping from 2015 to 2020 in Hoh Xil, with a wide coverage of about 200,000 km <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> . 1,592 Sentinel-1 images were processed based on the small baseline subset (SBAS) technique. The results show that Hoh Xil was experiencing dynamic permafrost disturbances. Thawing permafrost with both the linear subsidence rate higher than 2 mm/yr and the periodic amplitude over 2 mm was primarily detected in areas of flat or gentle slopes. The InSAR cumulative deformation is highly correlated with permafrost thawing depth. Significant lag times were identified between seasonal oscillation of InSAR deformation and land surface temperature (LST). Thermokarst landforms of retrogressive thaw slumps and thermokarst lakes broadly formed and dynamically evolved as a consequence of permafrost degradation. Particularly, widespread thawing permafrost characterized by the spatial clustering of thermokarst lakes appeared to occur in areas adjacent to large lakes. The discovered dynamic permafrost disturbances in Hoh Xil manifested even the secluded Tibetan Plateau hinterland was facing the threat of climate change.

Towards precise drone-based measurement of elevation change in permafrost terrain experiencing thaw and thermokarst

Measuring ground elevation changes plays a crucial role in several environmental applications. For instance, permafrost soils undergo seasonal active layer freezing and thawing that causes cyclic elevation changes. Permafrost thaw can result in unidirectional ground subsidence, which may be gradual and uniform, or rapid and irregular in the case of thermokarst landforms such as slumps and degrading ice-wedges. Photogrammetric drone surveys have effectively characterized large (&gt; 0.1 m) ground elevation changes resulting from thermokarst, yet many permafrost processes of interest lead to more subtle elevation changes. In this study, we assessed various drone-based surveying strategies for their precision to measure smaller (&lt; 0.1 m) ground elevation changes to better characterize permafrost-driven surface dynamics. The strategies were compared by examining the short-term reproducibility of modeled elevation for 76 bare ground targets, derived from six repeat drone surveys captured under variable illumination. We found that the Phantom 4 RTK drone using direct georeferencing, combined with one fixed ground control point, could reproduce elevations with a mean absolute deviation of 0.6 cm, suggesting a minimum level of change detection of 1.4 cm at 95% confidence. Drone-based methods for measuring permafrost elevation changes should be complementary to in situ and satellite-based (e.g. differential interferometric Synthetic Aperture Radar) approaches.

Sentinel‐Based Inventory of Thermokarst Lakes and Ponds Across Permafrost Landscapes on the Qinghai‐Tibet Plateau

AbstractThermokarst lakes and ponds (hereafter referred to as thaw lakes) play an important role in the permafrost regions by regulating hydrology, ecology, and biogeochemistry. However, detailed quantitative information on thaw lake extent and distribution remains poorly resolved across the entire permafrost regions on the Qinghai‐Tibet Plateau (QTP). Here, we applied the random forest (RF) model and manual visual vectorization methods to extract thaw lake boundaries on the QTP based on Sentinel‐2 images. Accuracy assessment was comprehensively demonstrated regarding the inherent spatial resolution of imagery and RF model performance. The results showed that the accumulated uncertainty of the total thaw lake area was ±5.75 km2, and the mean accuracy (91.9%) from field‐measured boundaries of 132 thaw lakes supported the accuracy of this inventory. A total of ∼161,300 thaw lakes with sizes ranging from 500 m2 to 3 km2 were detected, with a total area of ∼2,825.45 ± 5.75 km2. Most thaw lakes were detected in the continuous permafrost type (94.1%) and within the elevations of 4,500–5,000 m (68.4%). The small thaw lakes (&lt;10,000 m2) predominated the total lake number (78.9%) but contributed to a small portion of the total lake area (12.7%). Spatial distributions of thaw lakes in terms of different climatic and environmental conditions were also comprehensively explored, including temperature, precipitation, ground thermal stability, active layer thickness, vegetation, soil properties, and underground ice content. This inventory is expected to be incorporated into Earth system models for a more comprehensive projection of the large‐scale biogeochemical feedback of thermokarst landforms on the QTP under continued global warming.

Seasonal deformation monitoring over thermokarst landforms using terrestrial laser scanning in Northeastern Qinghai-Tibetan Plateau

Climate warming has accelerated permafrost degradation over the Qinghai-Tibet Plateau (QTP) over the past several decades. The development of thermokarst landforms is a key indicator of permafrost degradation, while it lacks quantified measurements and comprehensive research over the QTP. The aim of this study is to investigate the development of thermokarst terrains through repeated ground-based elevation observations by using terrestrial laser scanning (TLS) in the northern QTP from April 11, 2016 through June 16, 2018. TLS Time series analysis reveals that the margin of the thermokarst landforms undergoes significant ground subsidence, side materials collapse, and/or uplift and deposit, especially in the southeastern, northwestern part of thermokarst landforms during the middle through the late thaw season. The vertical deformation and headwall retreat of thermokarst landforms reached −3.364 m and 10.66 m from April 11, 2016 to June 16, 2018, respectively. We also generated high-resolution orthophotos based on aerial photos acquired by the built-in 4 K RGB (red, green, blue) camera of DJI Phantom 3 Professional unmanned aerial vehicle (UAV) in April and October of 2016. The UAV images confirmed the TLS observations during the same period and presented the severely deformed area. This study reveals that the seasonal vertical deformation and headwall retreat of the thermokarst landforms in the study site are consistent with seasonal ground temperature change during the observed period. Extreme precipitation event as a key factor triggered the severe deforming of the case study thermokarst. Ground ice, peat layer, and human activity also contributed to the thermokarst landforms formation. The results also illustrate that TLS is an effective method for studying thermokarst development quantitatively.

Investigation of permafrost engineering geological environment with electrical resistivity tomography: A case study along the China-Russia crude oil pipelines

The electrical resistivity tomography and vegetation surveys were conducted for investigating the distribution of frozen ground and other engineering geological characteristics at four representative sites (BHP, XFB, WLG, and XAZ) along the China-Russia Crude Oil Pipelines (CRCOPs) I and II. The comprehensive analysis shows that CRCOPs play an important role in forming and connecting supra-permafrost subaerial taliks and in facilitating thermokarst landforms, further accelerating permafrost degradation. The survey results indicate that on 15–29 August 2019, the permafrost base under the CRCOP at the XFB site was 21.5 m in depth, that of WLG site was 21.5 m (32.8 m at the undisturbed sites), and that of XAZ site, 37.3 m; the permafrost table of XFB site was 6.72 m (4.97 m on 20 m away from the pipeline), that of WLG site was 8.64 m (1.94 m at the undisturbed sites), and that of XAZ site was 6.72 and 3.38 m on the southern and northern slopes; the hydrothermal influences of CRCOPs extended horizontally to about 60 m away from the pipelines, but the vertical hydrothermal impacts were largely limited to about 10–15 m in depth at the BHP site in the zone of deep (>1.5 m) seasonal frost. The extent of engineering influences from the pipelines and associated engineering activities in patchy and island permafrost are evidently larger than those in seasonal frost. The sites with substantial changes are found in pipeline foundation soils underlain by warm (> − 1 °C; thermally unstable) and ice-rich permafrost. This study can timely help decision makers mitigate frost hazards in a proactive manner.

Application of an Unmanned Aerial Vehicle for Large-Scale Mapping of Thermokarst Landforms

thermokarst refers to cryogenic processes, as a result of which characteristic relief forms are formed, as a result of melting of icy frozen soils or an ice complex. The study of the dynamics and distribution of modern thermokarst forms is important from the point of view of using them as one of the main indicators of the degradation of permafrost soils. In recent years, unmanned aerial vehicles, which can create orthorectified images and digital terrain models (DTM) with very high spatial resolution, have been actively used to study thermokarst. This article provides an example of the use of UAVs in the study of the development of thermokarst in s. Uncure. Based on landscape-geomorphological studies and decoding of the created orthophotomap, the areas of thermokarst development were identified. The total land area covered by thermokarst is 0.66 km2, which is 19.5% of the territory of the settlement. According to the classification of P.A. Solovyov, two types of thermokarst relief were distinguished here: bylary and alasy. The largest area is covered by bylars, which are the initial stage of the formation of Alas. They are formed as a result of melting of the upper part of revein ice and form a polygonal-lowland microrelief. More than 6,000 elementary polygons with a total area of 0.4 km2 were discovered here. The landfill area varies from 6 m2 to 223 m2, and the average area is 57 m2. They are formed by local areas throughout the area of the village.

Increase in beaver dams controls surface water and thermokarst dynamics in an Arctic tundra region, Baldwin Peninsula, northwestern Alaska

Beavers are starting to colonize low arctic tundra regions in Alaska and Canada, which has implications for surface water changes and ice-rich permafrost degradation. In this study, we assessed the spatial and temporal dynamics of beaver dam building in relation to surface water dynamics and thermokarst landforms using sub-meter resolution satellite imagery acquired between 2002 and 2019 for two tundra areas in northwestern Alaska. In a 100 km2 study area near Kotzebue, the number of dams increased markedly from 2 to 98 between 2002 and 2019. In a 430 km2 study area encompassing the entire northern Baldwin Peninsula, the number of dams increased from 94 to 409 between 2010 and 2019, indicating a regional trend. Correlating data on beaver dam numbers with surface water area mapped for 12 individual years between 2002 and 2019 for the Kotzebue study area showed a significant positive correlation (R2 = 0.61; p < .003). Beaver-influenced waterbodies accounted for two-thirds of the 8.3% increase in total surface water area in the Kotzebue study area during the 17 year period. Beavers specifically targeted thermokarst landforms in their dam building activities. Flooding of drained thermokarst lake basins accounted for 68% of beaver-influenced surface water increases, damming of lake outlets accounted for 26%, and damming of beaded streams accounted for 6%. Surface water increases resulting from beaver dam building likely exacerbated permafrost degradation in the region, but dam failure also factored into the drainage of several thermokarst lakes in the northern Baldwin Peninsula study region, which could promote local permafrost aggradation in freshly exposed lake sediments. Our findings highlight that beaver-driven ecosystem engineering must be carefully considered when accounting for changes occurring in some permafrost regions, and in particular, regional surface water dynamics in low Arctic and Boreal landscapes.

Pathways of ice-wedge degradation in polygonal tundra under different hydrological conditions

Abstract. Ice-wedge polygons are common features of lowland tundra in the continuous permafrost zone and prone to rapid degradation through melting of ground ice. There are many interrelated processes involved in ice-wedge thermokarst and it is a major challenge to quantify their influence on the stability of the permafrost underlying the landscape. In this study we used a numerical modelling approach to investigate the degradation of ice wedges with a focus on the influence of hydrological conditions. Our study area was Samoylov Island in the Lena River delta of northern Siberia, for which we had in situ measurements to evaluate the model. The tailored version of the CryoGrid 3 land surface model was capable of simulating the changing microtopography of polygonal tundra and also regarded lateral fluxes of heat, water, and snow. We demonstrated that the approach is capable of simulating ice-wedge degradation and the associated transition from a low-centred to a high-centred polygonal microtopography. The model simulations showed ice-wedge degradation under recent climatic conditions of the study area, irrespective of hydrological conditions. However, we found that wetter conditions lead to an earlier onset of degradation and cause more rapid ground subsidence. We set our findings in correspondence to observed types of ice-wedge polygons in the study area and hypothesized on remaining discrepancies between modelled and observed ice-wedge thermokarst activity. Our quantitative approach provides a valuable complement to previous, more qualitative and conceptual, descriptions of the possible pathways of ice-wedge polygon evolution. We concluded that our study is a blueprint for investigating thermokarst landforms and marks a step forward in understanding the complex interrelationships between various processes shaping ice-rich permafrost landscapes.

Automatic Mapping of Thermokarst Landforms from Remote Sensing Images Using Deep Learning: A Case Study in the Northeastern Tibetan Plateau

Thawing of ice-rich permafrost causes thermokarst landforms on the ground surface. Obtaining the distribution of thermokarst landforms is a prerequisite for understanding permafrost degradation and carbon exchange at local and regional scales. However, because of their diverse types and characteristics, it is challenging to map thermokarst landforms from remote sensing images. We conducted a case study towards automatically mapping a type of thermokarst landforms (i.e., thermo-erosion gullies) in a local area in the northeastern Tibetan Plateau from high-resolution images by the use of deep learning. In particular, we applied the DeepLab algorithm (based on Convolutional Neural Networks) to a 0.15-m-resolution Digital Orthophoto Map (created using aerial photographs taken by an Unmanned Aerial Vehicle). Here, we document the detailed processing flow with key steps including preparing training data, fine-tuning, inference, and post-processing. Validating against the field measurements and manual digitizing results, we obtained an F1 score of 0.74 (precision is 0.59 and recall is 1.0), showing that the proposed method can effectively map small and irregular thermokarst landforms. It is potentially viable to apply the designed method to mapping diverse thermokarst landforms in a larger area where high-resolution images and training data are available.

Thermokarst Development Detected from High-Definition Topographic Data in Central Yakutia

Eastern Siberia is characterized by widespread permafrost thawing and subsequent thermokarst development. Estimation of the impacts of the predicted rise in precipitation and air temperatures under climate change requires quantitative knowledge about the spatial distribution of thermokarst development. In the last few years, unmanned aerial systems (UAS) and structure-from-motion multi-view stereo (SfM-MVS) photogrammetry attracted a tremendous amount of interest for acquiring high-definition topographic data. This study detected characteristics of thermokarst landforms using UAS and SfM-MVS photogrammetry at a disused airfield (3.0 ha) and for arable land that was previously used for farming (6.3 ha) in the Churapcha area, located on the right bank of the Lena River in central Yakutia. Orthorectified photographs and digital terrain models with spatial resolutions of 4.0 cm and 8.0 cm, respectively, were obtained for this study. At the disused airfield site and the abandoned arable land, 174 and 867 high-centered polygons that developed after the 1990s were detected, respectively. The data showed that the average diameter and average area of the polygons at the disused airfield site were 11.6 m and 111.2 m2, respectively, while those of the polygons in the abandoned arable land were 7.4 m and 46.8 m2, respectively. The abandoned arable land was characterized by smaller polygons and a higher polygon density. The differences in polygon size for the abandoned arable land and the disused airfield site indicate a difference in the ice wedge distributions and thermokarst developments. The subsidence rate was estimated as 2.1 cm/year for the disused airfield site and 3.9 cm/year for the abandoned arable land.

Effects of permafrost collapse on soil bacterial communities in a wet meadow on the northern Qinghai-Tibetan Plateau

BackgroundPermafrost degradation may develop thermokarst landforms, which substantially change physico–chemical characteristics in the soil as well as the soil carbon stock. However, little is known about changes of bacterial community among the microfeatures within thermokarst area.ResultsWe investigated bacterial communities using the Illumina sequencing method and examined their relationships with soil parameters in a thermokarst feature on the northern Qinghai-Tibetan Plateau. We categorized the ground surface into three different micro-relief patches based on the type and extent of permafrost collapse (control, collapsing and subsided areas). Permafrost collapse significantly decreased the soil carbon density and moisture content in the upper 10 cm samples in the collapsing areas. The highest loading factors for the first principal component (PC) extracted from the soil parameters were soil carbon and nitrogen contents, while soil moisture content and C:N ratios were the highest loading factors for the second PC. The relative abundance of Acidobacteria decreased with depth. Bacterial diversity in subsided areas was higher than that in control areas.ConclusionsBacterial community structure was significantly affected by pH and depth. The relative abundance of Gemmatimonadetes and Firmicutes were significantly correlated with the first and second PCs extracted from multiple soil parameters, suggesting these phyla could be used as indicators for the soil parameters in the thermokarst terrain.

Carbon and nitrogen pools in thermokarst-affected permafrost landscapes in Arctic Siberia

Abstract. Ice-rich yedoma-dominated landscapes store considerable amounts of organic carbon (C) and nitrogen (N) and are vulnerable to degradation under climate warming. We investigate the C and N pools in two thermokarst-affected yedoma landscapes – on Sobo-Sise Island and on Bykovsky Peninsula in the north of eastern Siberia. Soil cores up to 3 m depth were collected along geomorphic gradients and analysed for organic C and N contents. A high vertical sampling density in the profiles allowed the calculation of C and N stocks for short soil column intervals and enhanced understanding of within-core parameter variability. Profile-level C and N stocks were scaled to the landscape level based on landform classifications from 5 m resolution, multispectral RapidEye satellite imagery. Mean landscape C and N storage in the first metre of soil for Sobo-Sise Island is estimated to be 20.2 kg C m−2 and 1.8 kg N m−2 and for Bykovsky Peninsula 25.9 kg C m−2 and 2.2 kg N m−2. Radiocarbon dating demonstrates the Holocene age of thermokarst basin deposits but also suggests the presence of thick Holocene-age cover layers which can reach up to 2 m on top of intact yedoma landforms. Reconstructed sedimentation rates of 0.10–0.57 mm yr−1 suggest sustained mineral soil accumulation across all investigated landforms. Both yedoma and thermokarst landforms are characterized by limited accumulation of organic soil layers (peat). We further estimate that an active layer deepening of about 100 cm will increase organic C availability in a seasonally thawed state in the two study areas by ∼ 5.8 Tg (13.2 kg C m−2). Our study demonstrates the importance of increasing the number of C and N storage inventories in ice-rich yedoma and thermokarst environments in order to account for high variability of permafrost and thermokarst environments in pan-permafrost soil C and N pool estimates.