LONG TERM ROCK WEATHERING AND EROSION DRIVEN BY SUBCRITICAL CRACKING

We set out to test the hypothesis that virtually all rock mechanical weathering – and thus mechanical rock erosion – within the critical zone, at or within ~100 m of Earth’s surface, occurs due to the subcritical growth of cracks. This approach to understanding mechanical weathering and erosion is different from most such studies in that it addresses 1) the cracking processes rather than the stressing processes and 2) subcritical cracking, cracking dictated by rock parameters that are distinct from those of rock strength, compressive or tensile, which define the large stresses necessary for a rock to critically fail. Furthermore, it has been hypothesized that most mechanical weathering occurs by subcritical cracking during most of geologic time, because the vast majority of Earth surface stresses are in the subcritical range (Eppes and Keanini, 2017). If these hypotheses are true, then rock erosion rates should correlate with rock subcritical cracking parameters when other variables (climate, tectonics, erosion process) are held constant. This correlation should occur because the subcritical cracking rate will dictate how fast cracks can coalesce such that rock pieces are detached.

We therefore present 10Be-derived erosion rates for 20 ridgeline outcrops from a small region of the Blue Ridge Mountains in Shenandoah National Park, Virginia. The outcrops vary in lithology – quartzite, granite, fine and coarse meta sandstones - but not environment. We also measured subcritical cracking parameters for representative samples from the same outcrops. We find strong correlations between subcritical parameters and erosion rates (R²>0.86; p-values<0.05), but not between erosion rates and compressive strength as measured with a Schmidt Hammer. We also find a correlation between the Subcritical Cracking Index and our field measurements of outcrop crack length. This correlation is not widely field-tested but is nevertheless predicted by existing numerical models of subcritical fracture propagation (e.g. Olson, 2004), further supporting our hypothesis. Overall, our dataset, though small, points to a promising new avenue of research by which we may better quantify mechanical weathering and rock erosion on Earth.