Weathering in the cold: Granite hillslopes in Osborn Mountain, WY and Bodmin Moor, UK

Low temperatures generally limit rates of chemical weathering, and hence might be expected to limit development and evolution of mobile regolith in cold climates. Chemical and physical processes operate to release material from bedrock into the mobile regolith. Rock is weakened by chemical weathering, while physical breakdown produces particles susceptible to transport. We examine two hillslopes mantled with mobile regolith in granitic terrains that formed in cold climates where the rates of chemical processes are expected to be low, while at least some physical processes are expected to be enhanced. These end-member environments provide insight into mobile regolith development. Our study sites are gentle parabolic hillslopes at Osborn Mtn, WY, USA (elevation: 3600 m, MAT: -5°C) and Bodmin Moor, Cornwall, UK (elevation: 370 m, MAT: 9°C) developed on granitic bedrock. Neither site experienced direct glaciation during the LGM, but both were in periglacial zones. At present, the sites are populated by neither trees nor burrowing mammals. Biotite, which might spawn rock breakdown by hydration/oxidation cracking, is rare. Although there is evidence that trees have grown in Bodmin Moor in the past, frost cracking rather than tree throw has likely been the dominant mechanism releasing intact material into the mobile soil column at both field sites over the Quaternary. Present day ground surface temperatures at Osborn Mtn are below 0°C for ~9 months of the year, as measured by temperature sensors, leaving only 3 months for chemical processes to operate. During winter, temperatures are cold enough for frost cracking to occur down to significant depths. In contrast, we calculate from a numerical model of the annual thermal cycle that the immediate sub-surface of the Bodmin Moor hillslope is not currently cold enough for frost cracking at any time of the year, meaning that chemical processes can operate continuously. However, given temperature drops of order 5-10°C during LGM, we would expect Bodmin Moor to have experienced periglacial conditions at that time. Total denudation rates determined from in situ 10Be of saprolite from the bottom of soil pits are 14 m/My at Osborn Mtn (Small et al., 1999) and 15 m/My at Bodmin Moor. Mean soil thickness of 100 cm at Osborn Mtn yields a soil residence time of 71,000 years, while the 60 cm mean soil thickness at Bodmin Moor yields soil residence time of 41,000 years. We use a mass balance approach to quantify chemical losses at both field sites by calculating δ, the mass gain or loss of an element per volume relative to parent material. We integrate δ over the soil profile and divide by the residence time to determine the total chemical erosion rate on each hillslope. The erosion rate calculated from chemical weathering at Osborn Mtn is 0.002 m/My whereas at Bodmin Moor it is 0.3 m/My. Comparison of the chemical erosion rates to the 10Be-derived total erosion rates shows that while the chemical erosion at Bodmin Moor is two orders of magnitude higher than at Osborn Mtn, likely reflecting both wetter and warmer conditions, the production of regolith at both sites is dominated by physical erosion.