Rapid climate warming at northern high latitudes is driving geomorphic changes across the permafrost zone. In the Yamal and Gydan peninsulas in western Siberia, subterranean accumulation of methane beneath or within ice-rich permafrost can create mounds at the land surface. Once over-pressurized by methane, these mounds can explode and eject frozen ground, forming a gas emission crater (GEC). While GECs pose a hazard to human populations and infrastructure, only a small number have been identified in the Yamal and Gydan peninsulas, where the regional distribution and frequency of GECs and other types of land surface change are relatively unconstrained. To understand the distribution of landscape change within 327,000 km2 of the Yamal-Gydan region, we developed a semi-automated multivariate change detection algorithm using satellite-derived surface reflectance, elevation, and water extent in the Google Earth Engine cloud computing platform. We found that 5% of the landscape changed from 1984 to 2017. The algorithm detected all seven GECs reported in the scientific literature and three new GEC-like features, and further revealed that retrogressive thaw slumps were more abundant than GECs. Our methodology can be refined to detect and better understand diverse types of land surface change and potentially mitigate risks across the northern permafrost zone.

No abstract

<p>Retrogressive thaw slumps (RTS) are thermokarst features in ice-rich hillslope permafrost terrain and can cause dynamic changes to the landscape. Their occurrence in the Arctic has become increasingly frequent. RTS can significantly impact permafrost stability and generate substantial carbon emissions. Understanding the spatial distribution of RTS is critical to understanding and modelling global warming factors from permafrost thaw. Mapping RTS using conventional Earth observation approaches is challenging due to the highly dynamic nature and often small scale of RTS in the Arctic. In this study, we trained deep neural network models to map RTS across several landscapes in Siberia and Canada. Convolutional neural networks were trained with 965 RTS features, where 509 were from the Yamal and Gydan peninsulas in Siberia, and 456 from six other pan-Arctic regions including Canada and Northeastern Siberia. We used 4-m Maxar commercial imagery as the base map, 10-m NDVI derived from Sentinel-2 as the vegetation feature and 2-m ArcticDEM as the elevation feature. The best-performing model reached a validation Intersection over Union (IoU) score of 0.74 and a test IoU score of 0.71. Compared to past efforts to map RTS features, this represents one of the best-performing models and generalises well for mapping RTS in different permafrost regions, representing a critical step towards pan-Arctic deployment. Our experiments shed light on the impact of within-class and between-class variances of RTS in different regions on the model performance and provided critical implications for our follow-up study. We propose this method as an effective, accurate and computationally undemanding approach for RTS mapping.</p>