Alaskan lake cores show hillslope erosion running 2 to 3 times higher on patchy permafrost

"Discontinuous" is a loose category. In such a landscape only 50% to 90% of the ground is underlain by permafrost ^[2], so the two study areas differ by degree: Toolik in the north has more continuous permafrost, and Delta Junction in the south has more discontinuous ^[5]. Across that gradient, the average intensity of diffusive erosion, the process that mainly moves sediment on hillslopes ^[3], was 2 to 3 times greater in the discontinuous areas ^[9]. That is a difference of 100% to 200% ^[13]. In both areas, diffusive processes were statistically more significant than advective ones ^[8]. The Eos summary does not say how the 12 lakes divide between the regions or how large the diffusive-over-advective gap is.

The time window is the design choice to look at. The cores reach back to the middle Holocene, a period when climate and permafrost conditions at both areas resembled today's ^[6]. They therefore describe landscapes near their present state, and the 2 to 3 times figure compares two of them. It does not follow a slope through thaw. The forecast that hillslope erosion could increase significantly as warming shifts landscapes toward discontinuous permafrost ^[11] is an extrapolation across that gap. We think it is a reasonable hypothesis and not yet a measured response.

The authors warn that the gap in erosion intensity could be peculiar to the sites they studied, and they call for comparable work in other catchments ^[10]. The base is 12 lakes in two parts of Alaska ^[5].

The stated payoff is better models of how the Arctic will respond to climate change ^[12]. In our view the 2 to 3 times figure works as a calibration for landscape evolution models, the kind the authors constrained with their core-derived rates ^[7]. Carrying it into infrastructure planning would stretch it. The cores average catchment-scale erosion from the middle Holocene to the present, and a road or a pipeline sits on one slope.