Retrogressive thaw slumps (RTSs) are spectacular landforms that occur due to the thawing of ice-rich permafrost or melting of massive ground ice, often in hillslope terrain. RTSs occur in the Arctic, the subarctic, and high mountain (Qinghai–Tibet Plateau) permafrost regions and are observed to expand in size and number due to climate warming. As the observation of RTSs is receiving more and more attention due to their important role in permafrost thaw; impacts on topography; mobilization of sediment, carbon, nutrients, and contaminants; and their effects on downstream hydrology and water quality, the thematic breadth of studies increases and scientists from different scientific backgrounds and perspectives contribute to new RTS research. At this point, a wide range of terminologies originating from different scientific schools is used, and we identified the need to provide an overview of variable characteristics of RTSs to clarify terminologies and ease the understanding of the literature related to RTS processes, dynamics, and feedbacks. We review the theoretical geomorphological background of RTS formation and landform characteristics to provide an up-to-date understanding of the current views on terminology and underlying processes. The presented overview can be used not only by the international permafrost community but also by scientists working on ecological, hydrological, and biogeochemical consequences of RTS occurrence and by remote-sensing specialists developing automated methods for mapping RTS dynamics. The review will foster a better understanding of the nature and diversity of RTS phenomena and provide a useful base for experts in the field but also ease the introduction to the topic of RTSs for scientists who are new to it.

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.

Stable isotopes (δ¹⁸O, δD) of wedge ice hold potential to reconstruct past winter climate conditions. Here, we present records of the marine isotope stages (MIS) 3 and 2 including the last Glacial maximum (LGM) from Bol’shoy Lyakhovsky Island (NE Siberia). MIS 3 wedge ice dated from 52 to 40 Kyr b2k varies between −32 and −29‰ in δ¹⁸O. Colder LGM conditions are implied by δ¹⁸O of −37‰ around 25 Kyr b2k. Similar Deuterium excess values indicate comparable moisture sources during MIS 3 and MIS 2. Regional LGM climate reconstructions depend on the seasonal resolution of the proxies and model simulations. Our wedge‐ice record reflects coldest winters during global minima in atmospheric CO₂ and sea level. The extreme LGM winter cooling is not represented in model projections of global LGM climate where West Beringia shows noticeably little cooling or even warming in mean annual temperatures compared to the late Holocene.