Landscape surfaces and soils are known to evolve in non-linear and complex ways over time. Due to changing environmental conditions, soil erosion and denudation processes also change significantly. Although, soil erosion is a growing concern wold-wide, also in Europe (see figure; European Soil Bureau, 2000; Raab et al., 2018), our current knowledge in this field remains fragmented and incomplete. Especially the fact that soil erosion is discontinuous over time is mostly neglected, because usually only average erosion or denudation rates are estimated (i.e. catchment-wide approaches). We therefore strived after an archive that could capture soil mass fluctuations over a continues time-frame, preferable for many millennia.
We investigated large residual rocks, that are still attached to the bedrock, so called "Tors", along vertical rock profiles.The conceptual idea is comparatively illustrated (on the left) with an ice on a stick. The stick represents the tor, and th ice the soil. Denudation (or ice melting) reduces the surrounding soil (ice) depth and volume. This can be quantified using modern dating techniques.We anticipated to gain information of their, possible multi-step, evolutionary exhumation path by using cosmogenic nuclide technique (10Be), aka, surface exposure dating.
We hypothesised that the exhumation path would be a reflection of the occurred soil erosion that led to its exhumation in the first place. We subsequently would be able to calculate long-term soil erosion and denudation rates. To link our long-term data to the more recent past we used another isotope technique for comparison. The fallout radionuclides 239+240Pu have gained recent attention (Alewell et al., 2017) through their suitability to trace soil erosion over last few decades (e.g. ~1964-today). We applied those methods in an upland in Calabria, southern Italy – The Sila Massif.
We chose the upland of the Sila Massif because of its ideal geomorphological and geological setting. The massif has formed through the active subduction of the Ionian Basin following several uplifting phases during the Miocene and Pleistocene. Due to long-term deep weathering processes, and subsequent erosion, granitic boulder field and tor landforms have arisen. The surface is generally characterised by wide, flat to gently-rolling plaeosurfaces ranging from 1000 to 1700 m a.s.l. Our main results can be found in Raab et al. (2017), Raab et al. (2018) & Raab et al., (2019).
Raab et al., (2017) contains local soil analytics and age estimations using radiocarbon dating, semi-quantitative and relative dating with chemical weathering indices and geochemical fingerprinting of volcanic ash (glass) deposits provided a basic geochronological understanding of the local soils. Raab et al., (2018) comprises of fallout radionuclide derived soil erosion rates for the last ~50-60 years which are found to be significantly higher than rates of the last 100 ka derived from the TEA (Tor Exhumation Approach). In both Raab et al., (2018) and Raab et al., (2019) the TEA enabled the capturing of past surface denudation and soil erosion variations over multi-millennia (Holocene–Pleistocene) in contrast to catchment-wide derived rates (Olivetti et al. (2012)) which only provide average rates. In addition, the erosion differences of various angled surfaces (e.g. slopes and planes) could be captured. The combination of climate and vegetation data (Allen et al., 1999; Pelle et al., 2012) resulted in a local soil evolution model. The synthesis proposes the following evolutionary stages: Early soil genesis and soil deepening prior to 70 ka was followed by a balanced pedogenesis which was disrupted around 15 ka by an increase of physical erosion. Overall the, TEA contributed greatly to fill the current gap for in-situ investigation techniques for time-scales of a few thousand to >100,000 years. The remaining gaps are still in need of adequate investigation techniques. Full results in Raab (2019).