@article{RamageIrrgangHerzschuhetal.2017,
  author    = {Ramage, Justine Lucille and Irrgang, Anna Maria and Herzschuh, Ulrike and Morgenstern, Anne and Couture, Nicole and Lantuit, Hugues},
  title     = {Terrain controls on the occurrence of coastal retrogressive thaw slumps along the Yukon Coast, Canada},
  series = {Journal of geophysical research : Earth surface},
  volume    = {122},
  journal   = {Journal of geophysical research : Earth surface},
  publisher = {American Geophysical Union},
  address   = {Washington},
  issn      = {2169-9003},
  doi       = {10.1002/2017JF004231},
  pages     = {1619 -- 1634},
  year      = {2017},
  abstract  = {Retrogressive thaw slumps (RTSs) are among the most active landforms in the Arctic; their number has increased significantly over the past decades. While processes initiating discrete RTSs are well identified, the major terrain controls on the development of coastal RTSs at a regional scale are not yet defined. Our research reveals the main geomorphic factors that determine the development of RTSs along a 238km segment of the Yukon Coast, Canada. We (1) show the current extent of RTSs, (2) ascertain the factors controlling their activity and initiation, and (3) explain the spatial differences in the density and areal coverage of RTSs. We mapped and classified 287 RTSs using high-resolution satellite images acquired in 2011. We highlighted the main terrain controls over their development using univariate regression trees model. Coastal geomorphology influenced both the activity and initiation of RTSs: active RTSs and RTSs initiated after 1972 occurred primarily on terrains with slope angles greater than 3.9 degrees and 5.9 degrees, respectively. The density and areal coverage of RTSs were constrained by the volume and thickness of massive ice bodies. Differences in rates of coastal change along the coast did not affect the model. We infer that rates of coastal change averaged over a 39year period are unable to reflect the complex relationship between RTSs and coastline dynamics. We emphasize the need for large-scale studies of RTSs to evaluate their impact on the ecosystem and to measure their contribution to the global carbon budget. Plain Language Summary Retrogressive thaw slumps, henceforth slumps are a type of landslides that occur when permafrost thaws. Slumps are active landforms: they develop quickly and extend over several hectares. Satellite imagery allows to map such slumps over large areas. Our research shows where slumps develop along a 238 km segment of the Yukon Coast in Canada and explains which environments are most suitable for slump occurrence. We found that active and newly developed slumps were triggered where coastal slopes were greater than 3.9 degrees and 5.9 degrees, respectively. We explain that coastal erosion influences the development of slumps by modifying coastal slopes. We found that the highest density of slumps as well as the largest slumps occurred on terrains with high amounts of ice bodies in the ground. This study provides tools to better identify areas in the Arctic that are prone to slump development.},
  language  = {en}
}
@article{PestryakovaHerzschuhGorodnichevetal.2018,
  author    = {Pestryakova, Luidmila Agafyevna and Herzschuh, Ulrike and Gorodnichev, Ruslan and Wetterich, Sebastian},
  title     = {The sensitivity of diatom taxa from Yakutian lakes (north-eastern Siberia) to electrical conductivity and other environmental variables},
  series = {Polar research : a Norwegian journal of Polar research},
  volume    = {37},
  journal   = {Polar research : a Norwegian journal of Polar research},
  publisher = {Routledge, Taylor \& Francis Group},
  address   = {Abingdon},
  issn      = {0800-0395},
  doi       = {10.1080/17518369.2018.1485625},
  pages     = {16},
  year      = {2018},
  abstract  = {Relative abundances of 157 diatom taxa from Yakutian lake surface-sediments were investigated for their potential to indicate certain environmental conditions. Data from 206 sites from Arctic, sub-Arctic and boreal environments were included. Redundancy analyses were performed to assess the explanatory power of mean July temperature (T-July), conductivity, pH, dissolved silica concentration, phosphate concentration, lake depth and vegetation type on diatom species composition. Boosted regression tree analyses were performed to infer the most relevant environmental variables for abundances of individual taxa and weighted average regression was applied to infer their respective optimum and tolerance. Electrical conductivity was best indicated by diatom taxa. In contrast, only few taxa were indicative of Si and water depth. Few taxa were related to specific pH values. Although T-July, explained the highest proportion of variance in the diatom spectra and was, after conductivity, the second-most selected splitting variable, we a priori decided not to present indicator taxa because of the poorly understood relationship between diatom occurrences and T-July. In total, 92 diatom taxa were reliable indicators of a certain vegetation type or a combination of several types. The high numbers of indicative species for open vegetation sites and for forested sites suggest that the principal turnover is the transition from forest-tundra to northern taiga. Overall, our results reveal that preference ranges of diatom taxa for environmental variables are mostly broad, and the use of indicator taxa for the purposes of environmental reconstruction or environmental monitoring is therefore restricted to marked rather than subtle environmental transitions.},
  language  = {en}
}