Short- and long-term thermo-erosion of ice-rich permafrost coasts in the Laptev Sea region

Abstract

Permafrost coasts in the Arctic are susceptible to a variety of changing environmental factors all of which currently point to increasing coastal erosion rates and mass fluxes of sediment and carbon to the shallow arctic shelf seas. Rapid erosion along high yedoma coasts composed of Ice Complex permafrost deposits creates impressive coastal ice cliffs and inspired research for designing and implementing change detection studies for a long time, but continuous quantitative monitoring and a qualitative inventory of coastal thermo-erosion for large coastline segments is still lacking. Our goal is to use observations of thermo-erosion along the mainland coast of the Laptev Sea, in eastern Siberia, to understand how it depends on coastal geomorphology and the relative contributions of water level and atmospheric drivers. We compared multi-temporal sets of orthorectified satellite imagery from 1965 to 2011 for three segments of coastline ranging in length from 73 to 95 km and analyzed thermo-denudation (TD) along the cliff top and thermo-abrasion (TA) along the cliff bottom for two nested time periods: long-term rates (the past 39–43 yr) and short-term rates (the past 1–4 yr). The Normalized Difference Thermo-erosion Index (NDTI) was used as a proxy to qualitatively describe the relative proportions of TD and TA. Mean annual erosion rates at all three sites were higher in recent years (−5.3 ± 1.3 m a⁻¹) than over the long-term mean (−2.2 ± 0.1 m a⁻¹). The Mamontov Klyk coast exhibits primarily spatial variations of thermo-erosion, while intrasite-specific variations caused by local relief were strongest at the Buor Khaya coast, where the slowest long-term rates of around −0.5 ± 0.1 m a⁻¹ were observed. The Oyogos Yar coast showed continuously rapid erosion up to −6.5 ± 0.2 m a⁻¹. In general, variable characteristics of coastal thermo-erosion were observed not only between study sites and over time, but also within single coastal transects along the cliff profile. Varying intensities of cliff bottom and top erosion are leading to diverse qualities of coastal erosion that have different impacts on coastal mass fluxes. The different extents of Ice Complex permafrost degradation within our study sites turned out to influence not only the degree of coupling between TD and TA, and the magnitude of effectively eroded volumes, but also the quantity of organic carbon released to the shallow Laptev Sea from coastal erosion, which ranged on a long-term from 88 ± 21 to 800 ± 61 t per km coastline per year and will correspond to considerably higher amounts, if recently observed more rapid coastal erosion rates prove to be persistent

Observing Muostakh disappear: permafrost thaw subsidence and erosion of a ground-ice-rich island in response to arctic summer warming and sea ice reduction

“…This proportion is consistent with observations made by Günther et al (2013a), who find recent rates are at least 1.6 times more rapid than historical TD and TA along Ice Complex coastlines throughout the Laptev Sea. They report that for a subset of sites TA accelerated from long-term −3.3 ± 1.2 m a −1 to −5.7 ± 1.2 m a −1 over the past few years.…”

“…Günther et al (2013a) found that TD and TA along Ice Complex coasts are interconnected: TA is the limiting component for coastal thermo-erosion intensity on the longterm scale. TA also better reflects the overall land loss of the base area of the island.…”

“…Cliff top line positions were manually mapped in orthorectified imagery at different points in time. Unlike other studies on coastal thermo-erosion using the transect method (Günther et al, 2013a), we derived vector data of areal land loss and subsequently calculated seasonal variations of TD. This approach is also used by Aguirre et al (2008) and Tweedie et al (2012) for monitoring changes along an arctic coastline but on the basis of DGPS (differential GPS) measurements.…”

“…The zero-order polynomial turned out to be the most stable and best possible solution of the block, yielding a submetre accuracy within the entire block of 0.81 m RMSE (Table 1). According to Günther et al (2013a), the mutual RMSE of each data set (Table 1) is then considered to be the relative georeferencing uncertainty in the determination of the cumulative uncertainty in coastline position, which results out of the combination of this error, the ground resolution, and the additional 2-D positional error introduced by the DEM used for orthorectification, depending on the incidence angle. Uncertainties in change rate calculation over six periods between 15 June 2010 and 17 July 2013 are also applied following Günther et al (2013a) and were in the range of 0.47-0.65 m.…”

“…During sum-mer, this coast is no longer protected by sea ice and retreats at erosion rates of −2 to −6 m a −1 (Grigoriev et al, 2006). Large areas of up to −3400 m 2 km −1 of coastline are lost annually and currently this rate is more than twice as rapid as historical erosion (Günther et al, 2013a). The distinguishing feature of polar coasts is the presence of a variety of ice types on and ground ice below the earth surface (Forbes and Hansom, 2011).…”