Mapping and Characterizing Rock Glaciers in the Arid Western Kunlun Mountains Supported by InSAR and Deep Learning

Abstract

Rock glaciers manifest the creep of mountain permafrost occurring in the past or at present. Their presence and dynamics are indicators of permafrost distribution and changes in response to climate forcing. There is a complete lack of knowledge about rock glaciers in the Western Kunlun Mountains, one of the driest mountain ranges in Asia, where extensive permafrost is rapidly warming. In this study, we first mapped and quantified the kinematics of active rock glaciers based on satellite Interferometric Synthetic Aperture Radar (InSAR) and Google Earth images. Then we trained DeepLabv3+, a deep learning network for semantic image segmentation, to automate the mapping task. The well‐trained model was applied for a region‐wide, extensive delineation of rock glaciers from Sentinel‐2 images to map the landforms that were previously missed due to the limitations of the InSAR‐based identification. Finally, we mapped 413 rock glaciers across the Western Kunlun Mountains: 290 of them were active rock glaciers mapped manually based on InSAR and 123 of them were newly identified and outlined by deep learning. The rock glaciers are categorized by their spatial connection to the upslope geomorphic units. All the rock glaciers are located at altitudes between 3,390 m and 5,540 m with an average size of 0.26 km2 and a mean slope angle of 17°. Characteristics of the inventoried rock glaciers provided insights into permafrost distribution in the Western Kunlun Mountains. The median and maximum surface downslope velocities of the active ones are 17±1 cm yr‐1 and 127±6 cm yr‐1, respectively.

Rock Glacier Inventory of the Southwestern Pamirs Supported by InSAR Kinematics

“…Notably, the western Nyainqêntanglha and western Kunlun Mountains, similar to our study area, are characterized by arid to semi-arid climates, hosting RGs with the same order of velocity magnitudes. Investigating the upslope connection units of RGs, glacier-related types (G-RGs and GF-RGs) typically demonstrate higher velocities than talus-related ones (T-RGs and DMS-RGs) (Figure 5a), a finding similar to that in Hu et al [29] for the western Kunlun Mountains. This velocity discrepancy is likely due to acceleration caused by glacial meltwater during spring and summer [45,52], suggesting a correlation between RG movement and precipitation or meltwater [51,54,55,57,58], supported by their spatial distribution (Figure 1c).…”

“…Conversely, RGs in the northern Tien Shan (Ile Alatau and Kungöy Ala-Too ranges) are notably active, with 45 out of 550 exhibiting velocities exceeding 100 cm/yr [24]. Similar velocity patterns are observed in the northern Tien Shan of China [23], the western Nyainqêntanglha Range [28], and the western Kunlun Mountains [29], aligning with findings from the southwestern Pamirs. Notably, the western Nyainqêntanglha and western Kunlun Mountains, similar to our study area, are characterized by arid to semi-arid climates, hosting RGs with the same order of velocity magnitudes.…”

“…Despite the potential of InSAR, it also presents certain limitations. Firstly, the displacement measurements can be obscured by artifacts resulting from uncorrected atmospheric delays [29,34,46] and decorrelation or phase shifts due to changes in surface physical properties, such as vegetation, snow, and soil moisture [25]. To mitigate these limitations, it is crucial to exclude areas of decorrelation and rely on a stack of several interferograms from different periods.…”

“…This area is also notably susceptible to climate change impacts [18,19]. In-depth RG inventories have been developed for several parts of the HMA region, including the Himalayas [8,12,15,[20][21][22], the Karakoram [5], the Tien Shan [23][24][25], the Gangdise Mountains [26], the southeastern Tibetan Plateau [16,27,28] and the western Kunlun Mountains [29]. However, the southwestern part of the Pamir region, recognized as a significant solid water storage area, still lacks a systematic and detailed RG inventory.…”

“…Recognizing the limitations of subjective assessments, the International Permafrost Association (IPA) advocates adopting more objective and precise methods, such as Interferometric Synthetic Aperture Radar (InSAR). This technology has been effectively applied in the HMA region to extract the velocity of RGs, demonstrating its feasibility and reliability in enhancing the precision of kinematic studies [23][24][25]28,29,[31][32][33][34]. The shift toward InSAR in RG research marks a significant advancement in permafrost studies, ensuring more standardized assessments of RG dynamics, which are essential for understanding permafrost behavior and its environmental implications.…”