Slope sediment yield in and lowland continuous permafrost environments, Canadian Arctic Archipelago

Surface wash erosion was measured at runoff plots on low to moderate slopes in clayey and sandy silts underlain by continuous permafrost on the Fosheim Peninsula, Ellesmere Island. Due to snow redistribution in winter, total precipitation on the plots varied from 34 to 150 mm, with corresponding surface runoff values of 0 to 102 mm. Where runoff occurred, at least 80% of it was derived from snowmelt. Suspended sediment removal was < 75 g m(-2) a(-1) at relatively well-vegetated sites but averaged more than 1200 g m(-2) a(-1) at a plot where the vegetation had been removed by landsliding. Niveo-aeolian deposition was greater than suspended sediment removal at some plots, indicating net accumulation. Solute removal ranged up to 80 g m(-2) a(-1) and exceeded elastic sediment transport at one vegetated site. Elevated rates of erosion at the sites of detachment slides that pre-date 1950 demonstrated that terrain disturbance in permafrost environments can affect slopewash processes for at least several decades. Plot data (precipitation, vegetation and surface grain-size) from the Fosheim Peninsula and Banks Island were used to develop a statistical model of suspended sediment removal by surface wash on undisturbed slopes. For any given grain-size, the model predicts a rise in erosion from zero precipitation (because of an absence of runoff) to a peak at about 50 mm, a decline as precipitation increases to 300 mm and a further increase in erosion beyond this inflection point. This non-linear response is due to the complex interaction of moisture (primarily snow) and vegetation cover. Erosion at any given precipitation value varies through three orders of magnitude depending on surface grain-size. The maximum erosion predicted is 1 kg m(-2) a(-1) for a runoff plot with 1100 mm of precipitation, a corresponding vegetation cover of 77% and a median surface grain-size of phi. (C) 2002 Elsevier Science B.V. All rights reserved.