3 January 2022 in The Cryosphere
Arctic ice-rich permafrost is becoming increasingly vulnerable to terrain-altering thermokarst, and among the most rapid and dramatic of these changes are retrogressive thaw slumps (RTSs). They initiate when ice-rich soils are exposed and thaw, leading to the formation of a steep headwall which retreats during the summer months. The impacts and the distribution and scaling laws governing RTS changes within and between regions are unknown. Using TanDEM-X-derived digital elevation models, we estimated RTS volume and area changes over a 5-year time period from winter 2011/12 to winter 2016/17 and used for the first time probability density functions to describe their distributions. We found that over this time period all 1853 RTSs mobilized a combined volume of 17×106 m3 yr−1, corresponding to a volumetric change density of 77 m3 yr−1 km−2. Our remote sensing data reveal inter-regional differences in mobilized volumes, scaling laws, and terrain controls. The distributions of RTS area and volumetric change rates follow an inverse gamma function with a distinct peak and an exponential decrease for the largest RTSs. We found that the distributions in the high Arctic are shifted towards larger values than at other study sites We observed that the area-to-volume scaling was well described by a power law with an exponent of 1.15 across all study sites; however the individual sites had scaling exponents ranging from 1.05 to 1.37, indicating that regional characteristics need to be taken into account when estimating RTS volumetric changes from area changes. Among the terrain controls on RTS distributions that we examined, which included slope, adjacency to waterbodies, and aspect, the latter showed the greatest but regionally variable association with RTS occurrence. Accounting for the observed regional differences in volumetric change distributions, scaling relations, and terrain controls may enhance the modelling and monitoring of Arctic carbon, nutrient, and sediment cycles.
6 May 2021 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Great Aletsch Glacier (Grosser Aletschgletscher), the largest glacier in the European Alps, contains 20% of the entire Swiss ice mass. Therefore, it has been selected as a super-testsite for the TanDEM-X satellite mission. Dense time series with a repeat interval down to 11 days were acquired between 2011 and 2019 using two polarizations (HH, VV) and across-track baselines of 0-1 km. To evaluate the use of interferometric single-pass SAR mission for glaciological applications we implemented a complete processing pipeline in IDL and computed 130 digital elevation models (DEMs) from bistatic radar interferograms. We present a method to circumvent a common pitfall during orthorectification of radar DEM differences. The observed surface height loss is compare with model predictions from literature. With patch-based incoherent offset tracking we obtained a complete, high resolution velocity map. Velocities reach up to 0.8 m d-1, show a seasonal variability of 5cm d-1 and agree within 0.04m d-1(RMSE) with field measurements. Copolar phase differences are used to approximate snow accumulation. In spring 2013, we observed X-Band penetration depths of 42 m in the accumulation area. For strongly crevassed areas the coherence drops already to 0.5 for across-track baselines B > 200 m. The glacier shows a height loss of up to 8 m a-1 on the tongue and an average height loss of 1.5 m a-1 over its whole area. We demonstrate how TanDEM-X interferograms can be used for orthorectification of the measured backscatter signal which allows for the decoupling of velocity estimates from height changes.
8 June 2020 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Vast areas of the Arctic host ice-rich permafrost, which is becoming increasingly vulnerable to terrain-altering thermokarst in a warming climate. Among the most rapid and dramatic changes are retrogressive thaw slumps. These slumps evolve by a retreat of the slump headwall during the summer months, making them detectable by comparing digital elevation models over time using the volumetric change as an indicator. Here, we present and assess a method to detect and monitor thaw slumps using time series of elevation models applied on two contrasting study areas in Northern Canada. Our two-step method is tailored to single-pass InSAR observations from the TanDEM-X satellite pair, which have been acquired since 2011. For each acquisition, we derive a digital elevation model and uncertainty estimates. In the first step, we difference digital elevation models and detect the significant elevation changes using a blob-detection algorithm. In the second step, we classify the detections into those due to thaw slumps and other causes using a simple thresholding method (accuracy: 78%), a random forest classifier (87%), and a support vector machine (86%). When our method is applied to other areas, the classifiers should be trained with data from part of the study area or with data obtained from similar areas in terms of topography, vegetation, and thaw slump characteristics to achieve the best performance. The obtained locations of thaw slumps can be used as a starting point to extract important slump properties, such as the headwall height and the volumetric change, which are currently not available on regional scales.
15 July 2022 in The Cryosphere
With climate change, Arctic hillslopes above ice-rich permafrost are vulnerable to enhanced mass wasting and organic carbon mobilization. In this study we use TanDEM-X-derived (TerraSAR-X add-on for Digital Elevation Measurement; synthetic-aperture radar) digital elevation models to document an approximately 43-fold increase in thaw slumping and concomitant 28-fold increase in carbon mobilization on the northern Taymyr Peninsula from 2010 to 2021. The available observations allowed us to compare two time periods, from 2010/11 to 2016/17 and from 2017/18 to 2020/21, and contrast retrogressive thaw slump (RTS) activity between these periods. We find that all quantities describing RTS activity increased in the observed period. The total volumetric change per year increased from about 0.17×106 to , a 43-fold increase. The observed surge in RTS activity is mainly driven by the initiation of new RTS, indicated by the 17-fold increase in active RTS numbers from 82 to 1404 and the relatively low average volumetric change rate per RTS increase of 2.3. In annual Sentinel-2 imagery, the number of detected RTSs in a subregion increased 10-fold in 2020. This coincides with a severe heatwave that occurred in northern Siberia in 2020. The area-to-volume scaling of the RTSs varied only slightly over time, despite the 2020 heatwave, indicating a robustness of the relationship to such an event. To estimate the slump-mobilized organic carbon, we intersected the elevation changes with a soil organic carbon (SOC) map, with contrasting assumptions about the deep carbon pool and massive-ice content. We estimated that the SOC mobilization rate increases 28-fold. The normalization of the SOC mobilization rate to our study region yields values of with a confidence interval of 5 to . A comparison to an independent estimate of the net ecosystem exchange of illustrates the importance of RTS activity to the carbon cycle. These results underscore that mass wasting is an important but commonly neglected component of the Arctic carbon cycle and particularly sensitive to extreme events.
20 October 2024 in Permafrost and Periglacial Processes
Deep‐learning (DL) models have become increasingly beneficial for the detection of retrogressive thaw slumps (RTS) in the permafrost domain. However, comparing accuracy metrics is challenging due to unstandardized labeling guidelines. To address this, we conducted an experiment with 12 international domain experts from a broad range of scientific backgrounds. Using 3 m PlanetScope multispectral imagery, they digitized RTS footprints in two sites. We evaluated label uncertainty by comparing manually outlined RTS labels using Intersection‐over‐Union (IoU) and F1 metrics. At the Canadian Peel Plateau site, we see good agreement, particularly in the active parts of RTS. Differences were observed in the interpretation of the debris tongue and the stable vegetated sections of RTS. At the Russian Bykovsky site, we observed a larger mismatch. Here, the same differences were documented, but several participants mistakenly identified non‐RTS features. This emphasizes the importance of site‐specific knowledge for reliable label creation. The experiment highlights the need for standardized labeling procedures and definition of their scientific purpose. The most similar expert labels outperformed the accuracy metrics reported in the literature, highlighting human labeling capabilities with proper training, site knowledge, and clear guidelines. These findings lay the groundwork for DL‐based RTS monitoring in the pan‐Arctic.
22 February 2025 in International Journal of Applied Earth Observation and Geoinformation