ALBERT

Description: A large amount of organic carbon stored in permafrost soils across the high latitudes is vulnerable to thaw, decomposition and release to the atmosphere as a result of climate warming. This process is anticipated to be a significant positive feedback on future radiative forcing from terrestrial ecosystems to the Earth’s climate system. Here, we describe the development of a geospatial framework designed to characterize permafrost carbon vulnerability in the northern hemisphere. The broadly-defined regional classification is based on a Pan-Arctic spatial representation of the major environmental controls on a) the rate and extent of permafrost degradation and thaw, b) the quantity and quality of soil organic matter stocks, and c) the form of permafrost carbon emissions as CO2 and CH4. The framework was developed by integrating existing spatial data layers describing permafrost and ground ice conditions, bioclimatic zones, and topographic and geographic attributes. The resulting Permafrost Regionalization Map (PeRM) can be used for synthesis studies on permafrost carbon vulnerability, including data representativeness and gap analysis, model-data integration and model benchmarking. The utility of the PeRM framework is demonstrated here through areal density analysis and spatial summaries of existing data collections describing the fundamental components of permafrost carbon vulnerability. We use this framework to describe the spatial representativeness and variability in measurements within and across PeRM regions using observational data sets describing active layer thickness, soil pedons and carbon storage, longterm incubations for carbon turnover rates, and site-level monitoring of CO2 and CH4 fluxes from arctic tundra and boreal forest ecosystems. We then use these regional summaries of the observational data to benchmark the results of a process-based biogeochemical model for its skill in representing the magnitudes and spatial variability in these key indicators. Finally, we are using this framework as a basis for higher-resolution mapping of key regions of particular vulnerability to both press (active layer thickening) and pulse (thermokarst development) disturbances, which is guiding on-going research toward characterizing permafrost degradation and associated vegetation changes through multi-scale remote sensing. Overall, this work provides a critical bridge between the abundant but disordered observational and experimental data collections and the development of higher-complexity process representation of the permafrost carbon feedback in geospatial modeling frameworks.

Description: Permafrost soils were characterized first in the field by numerous investigators according to standard soil descriptions and were sampled by depth increment within soil horizon boundaries. Measures of bulk density, C, N, and pH were used to further characterize C and N storage for soil horizons and profiles. Field attributes for organic (Oi, Oe, Oa or L, F, H) horizons, mineral (A, E, B, C) horizons, cryoturbated (jj subscripts with mixtures of organic and mineral matrices), and gleying (subscript g with reduced colors), ice-rich layers (e.g., Wf/Cgfjj, Wf/Oafjj) were examined for differences in C, N, and bulk density. Using the Community Climate System Model (CCSM4) we calculated cumulative distributions of active layer thickness (ALT) under current and future climates. We then superposed physical state over soil horizons to explore how chemical attributes are exposed by progressive thaw. Thawing will likely expose 147 Pg of C with 10 Pg of N by 2050 (representative concentration pathway RCP scenario 4.5) and as much as 436 PgC with 29 PgN by 2100 (RCP 8.5). This represents about 30% and 80% of circumarctic permafrost carbon for yr 2050 (RCP 4.5) and yr 2100 (RCP 8.5) scenarios, respectively. Organic horizons will likely contribute the earliest pulse of CO2 via combustion and decomposition. These changes have the potential for strong additional loading to our atmosphere, water resources, and ecosystems.

Description: Water bodies are a highly abundant feature of Arctic permafrost ecosystems and strongly influence their hydrology, ecology and biogeochemical cycling. While very high resolution satellite images enable detailed mapping of these water bodies, the increasing availability and abundance of this imagery calls for fast, reliable and automatized monitoring. This technical work presents a largely automated and scalable workflow that removes image noise, detects water bodies, removes potential misclassifications from infrastructural features, derives lake shoreline geometries and retrieves their movement rate and direction on the basis of ortho-ready very high resolution satellite imagery from Arctic permafrost lowlands. We applied this workflow to typical Arctic lake areas on the Alaska North Slope and achieved a successful and fast detection of water bodies. We derived representative values for shoreline movement rates ranging from 0.40–0.56 myr^-1 for lake sizes of 0.10 ha–23.04 ha. The approach also gives an insight into seasonal water level changes. Based on an extensive quantification of error sources, we discuss how the results of the automated workflow can be further enhanced by incorporating additional information on weather conditions and image metadata and by improving the input database. The workflow is suitable for the seasonal to annual monitoring of lake changes on a sub-meter scale in the study areas in northern Alaska and can readily be scaled for application across larger regions within certain accuracy limitations.

Description: In response to increasing Arctic temperatures, ice-rich permafrost landscapes are undergoing rapid changes. In permafrost lowlands, polygonal ice wedges are especially prone to degradation. Melting of ice wedges results in deepening troughs and the transition from low-centered to high-centered ice-wedge polygons. This process has important implications for surface hydrology, as the connectivity of such troughs determines the rate of drainage for these lowland landscapes. In this study, we present a comprehensive, modular, and highly automated workflow to extract, to represent, and to analyze remotely sensed ice-wedge polygonal trough networks as a graph (i.e., network structure). With computer vision methods, we efficiently extract the trough locations as well as their geomorphometric information on trough depth and width from high-resolution digital elevation models and link these data within the graph. Further, we present and discuss the benefits of graph analysis algorithms for characterizing the erosional development of such thaw-affected landscapes. Based on our graph analysis, we show how thaw subsidence has progressed between 2009 and 2019 following burning at the Anaktuvuk River fire scar in northern Alaska, USA. We observed a considerable increase in the number of discernible troughs within the study area, while simultaneously the number of disconnected networks decreased from 54 small networks in 2009 to only six considerably larger disconnected networks in 2019. On average, the width of the troughs has increased by 13.86%, while the average depth has slightly decreased by 10.31%. Overall, our new automated approach allows for monitoring ice-wedge dynamics in unprecedented spatial detail, while simultaneously reducing the data to quantifiable geometric measures and spatial relationships.

Description: Lake formation and drainage are pervasive phenomena in permafrost regions. Drained lake basins (DLBs) are often the most common landforms in lowland permafrost regions in the Arctic (50% to 75% of the landscape). However, detailed assessments of DLB distribution and abundance are limited. In this study, we present a novel and scalable remote sensing-based approach to identifying DLBs in lowland permafrost regions, using the North Slope of Alaska as a case study. We validated this first North Slope-wide DLB data product against several previously published sub-regional scale datasets and manually classified points. The study area covered 〉71,000 km2, including a 〉39,000 km2 area not previously covered in existing DLB datasets. Our approach used Landsat-8 multispectral imagery and ArcticDEM data to derive a pixel-by-pixel statistical assessment of likelihood of DLB occurrence in sub-regions with different permafrost and periglacial landscape conditions, as well as to quantify aerial coverage of DLBs on the North Slope of Alaska. The results were consistent with previously published regional DLB datasets (up to 87% agreement) and showed high agreement with manually classified random points (64.4–95.5% for DLB and 83.2–95.4% for non-DLB areas). Validation of the remote sensing-based statistical approach on the North Slope of Alaska indicated that it may be possible to extend this methodology to conduct a comprehensive assessment of DLBs in pan-Arctic lowland permafrost regions. Better resolution of the spatial distribution of DLBs in lowland permafrost regions is important for quantitative studies on landscape diversity, wildlife habitat, permafrost, hydrology, geotechnical conditions, and high-latitude carbon cycling.

Description: Ground subsidence and erosion processes caused by permafrost thaw pose a high risk to infrastructure in the Arctic. Climate warming is increasingly accelerating the thawing of permafrost, emphasizing the need for thorough monitoring to detect damages and hazards at an early stage. The use of unoccupied aerial vehicles (UAVs) allows a fast and uncomplicated analysis of sub-meter changes across larger areas compared to manual surveys in the field. In our study, we investigated the potential of photogrammetry products derived from imagery acquired with off-the-shelf UAVs in order to provide a low-cost assessment of the risks of permafrost degradation along critical infrastructure. We tested a minimal drone setup without ground control points to derive high-resolution 3D point clouds via structure from motion (SfM) at a site affected by thermal erosion along the Dalton Highway on the North Slope of Alaska. For the sub-meter change analysis, we used a multiscale point cloud comparison which we improved by applying (i) denoising filters and (ii) alignment procedures to correct for horizontal and vertical offsets. Our results show a successful reduction in outliers and a thorough correction of the horizontal and vertical point cloud offset by a factor of 6 and 10, respectively. In a defined point cloud subset of an erosion feature, we derive a median land surface displacement of (Formula presented.) m from 2018 to 2019. Projecting the development of the erosion feature, we observe an expansion to NNE, following the ice-wedge polygon network. With a land surface displacement of (Formula presented.) m and an alignment root mean square error of (Formula presented.) m, we find our workflow is best suitable for detecting and quantifying rapid land surface changes. For a future improvement of the workflow, we recommend using alternate flight patterns and an enhancement of the point cloud comparison algorithm.

Description: In a warming Arctic, permafrost-related disturbances, such as retrogressive thaw slumps (RTS), are becoming more abundant and dynamic, with serious implications for permafrost stability and bio-geochemical cycles on local to regional scales. Despite recent advances in the field of earth observation, many of these have remained undetected as RTS are highly dynamic, small, and scattered across the remote permafrost region. Here, we assessed the potential strengths and limitations of using deep learning for the automatic segmentation of RTS using PlanetScope satellite imagery, ArcticDEM and auxiliary datasets. We analyzed the transferability and potential for pan- Arctic upscaling and regional cross-validation, with independent training and validation regions, in six different thaw slump-affected regions in Canada and Russia. We further tested state-ofthe- art model architectures (UNet, UNet++, DeepLabv3) and encoder networks to find optimal model configurations for potential upscaling to continental scales. The best deep learning models achieved mixed results from good to very good agreement in four of the six regions (maxIoU: 0.39 to 0.58; Lena River, Horton Delta, Herschel Island, Kolguev Island), while they failed in two regions (Banks Island, Tuktoyaktuk). Of the tested architectures, UNet++ performed the best. The large variance in regional performance highlights the requirement for a sufficient quantity, quality and spatial variability in the training data used for segmenting RTS across diverse permafrost landscapes, in varying environmental conditions. With our highly automated and configurable workflow, we see great potential for the transfer to active RTS clusters (e.g., Peel Plateau) and upscaling to much larger regions.

Description: Thermokarst lakes are widespread in Arctic lowlands. Under a warming climate, landscapes with highly ice-rich Yedoma Ice Complex (IC) deposits are particularly vulnerable, and thermokarst lake area dynamics serve as an indicator for their response to climate change. We conducted lake change trend analysis for a 44,500 km2 region of the Kolyma Lowland using Landsat imagery in conjunction with TanDEM-X digital elevation model and Quaternary Geology map data. We delineated yedoma–alas relief types with different yedoma fractions, serving as a base for geospatial analysis of lake area dynamics. We quantified lake changes over the 1999–2018 period using machine-learning-based classification of robust trends of multi-spectral indices of Landsat data and object-based long-term lake detection. We analyzed the lake area dynamics separately for 1999–2013 and 1999–2018 periods, including the most recent five years that were characterized by very high precipitation. Comparison of drained lake basin area with thermokarst lake extents reveal the overall limnicity decrease by 80% during the Holocene. Current climate warming and wetting in the region led to a lake area increase by 0.89% for the 1999–2013 period and an increase by 4.15% for the 1999–2018 period. We analyzed geomorphological factors impacting modern lake area changes for both periods such as lake size, elevation, and yedoma–alas relief type. We detected a lake area expansion trend in high yedoma fraction areas indicating ongoing Yedoma IC degradation by lake thermokarst. Our concept of differentiating yedoma–alas relief types helps to characterize landscape-scale lake area changes and could potentially be applied for refined assessments of greenhouse gas emissions in Yedoma regions. Comprehensive geomorphological inventories of Yedoma regions using geospatial data provide a better understanding of the extent of thermokarst processes during the Holocene and the pre-conditioning of modern thermokarst lake area dynamics.