Both ecosystems and humans are dependent on soil for nutrient and water cycling as well as food, making the loss of soil a major issue facing humanity. However, the rate at which soil forms has not been well quantified. To investigate rates of soil formation as a function of climate, a latitudinal climosequence of forested sites has been established in North America and Wales. The climosequence is bounded by a
cold/wet end member in Wales and a warm/wet end member in Puerto Rico. In between, temperature and rainfall increase as sites extend south through New York, Pennsylvania, Virginia, Tennessee and Alabama. All sites, except Puerto Rico, are underlain by an organic-poor, iron-rich (Silurianage) shale, providing a constant parent material from which soil is forming. Puerto Rico is located on chemically similar,
but younger, shale. Soil sampling and geochemical analyses were completed similarly at all sites to allow direct comparisons and modelling of shale weathering. Independent 10Be estimates of erosion rates for a few locations along the transect are used to estimate residence times. Initial results show soil depth increases as a function of temperature, with shallow (~30 cm) profiles in Wales and Pennsylvania varying up to 630 cm deep in Puerto Rico. Depletion profiles of Na, a proxy for feldspar dissolution, are less than 20 % depleted at the surface in Wales and Pennsylvania, 50-60% depleted in Virginia and Tennessee, and 100% depleted at the surface in Puerto Rico. Using estimated soil residence times, apparent activation energies for Na depletion were calculated using different assumptions to range from 15-19 kcal mol-1, values that are slightly higher than those reported for Na plagioclase dissolution in the laboratory. Overall, data collected from soils across the transect will promote a better understanding of how climate changes can impact soil formation rates.
Brantley, S.L., Dere A.L. and White T.S. (2011): Quantifying rates and mechanisms of shale weathering across a continental-scale climosequence. Goldschmidt.
This Paper/Book acknowledges NSF CZO grant support.