ALBERT

Description: Ice-wedges are common permafrost features formed over hundreds to thousands of years of repeated frost cracking and ice vein growth. We used field and remote sensing observations to assess changes in areas dominated by ice-wedges, and we simulated the effects of those changes on snow accumulation and runoff. We show that top melting of ice-wedges and subsequent ground subsidence has occurred at multiple sites in the North American and Russian Arctic. At most sites, melting ice-wedges have initially resulted in increased wetness contrast across the landscape, evident as increased surface water in the ice-wedge polygon troughs and somewhat drier polygon centers. Most areas are becoming more heterogeneous with wetter troughs, more small ponds (themokarst pits forming initially at ice-wedge intersections and then spreading along the troughs) and drier polygon centers. Some areas with initial good drainage, such as near creeks, lake margins, and in hilly terrain, highcentered polygons form an overall landscape drying due to a drying of both polygon centers and troughs. Unlike the multi-decadal warming observed in permafrost temperatures, the ice-wedge melting that we observed appeared as a sub-decadal response, even at locations with low mean annual permafrost temperatures (down to -14 °C). Gradual long-term air and permafrost warming combined with anomalously warm summers or deep snow winters preceded the onset of the ice-wedge melting. To assess hydrological impacts of ice-wedge melting, we simulated tundra water balance before and after melting. Our coupled hydrological and thermal model experiments applied over hypothetical polygon surfaces suggest that 1. ice-wedge melting that produces a connected trough-network reduces inundation and increases runoff, and that 2. changing patterns of snow distribution due to differential ground subsidence has a major control on ice-wedge polygon tundra water balance despite an identical snow water equivalent at the landscape-scale. These decimeter-scale geomorphic changes are expected to continue in permafrost regions dominated by ice-wedge polygons, with implications for landatmosphere and land-ocean fluxes of water, carbon, and energy.