Where fast weathering creates thin regolith and slow weathering creates thick regolith Ekaterina Bazilevskaya, 1 * Marina Lebedeva, 1 Milan Pavich, 2 Gernot Rother, 3 Dilworth Y. Parkinson, 4 David Cole 5 and Susan L. Brantley 1 1 Earth and Environmental Systems Institute, Penn State University, University Park, PA, USA 2 US Geological Survey, Eastern Geology and Paleoclimate Science Center, Reston, VA, USA 3 Geochemistry and Interfacial Sciences Group, Chemical Sciences Division, Oak Ridge National Laboratory, Oak Ridge, TN, USA 4 Advanced Light Source, Lawrence Berkeley National Laboratory, Berkeley, CA, USA 5 School of Earth Science, The Ohio State University, Columbus, OH, USA Received 12 June 2012; Revised 23 October 2012; Accepted 30 October 2012 *Correspondence to: Ekaterina Bazilevskaya, Earth and Environmental Systems Institute, Penn State University, University Park, PA 16802, USA. E-mail: eab204@psu.edu ABSTRACT: Weathering disaggregates rock into regolith – the fractured or granular earth material that sustains life on the continental land surface. Here, we investigate what controls the depth of regolith formed on ridges of two rock compositions with similar initial porosities in Virginia (USA). A priori, we predicted that the regolith on diabase would be thicker than on granite because the dominant mineral (feldspar) in the diabase weathers faster than its granitic counterpart. However, weathering advanced 20 deeper into the granite than the diabase. The 20  -thicker regolith is attributed mainly to connected micron-sized pores, microfractures formed around oxidizing biotite at 20 m depth, and the lower iron (Fe) content in the felsic rock. Such porosity allows pervasive advection and deep oxidation in the granite. These observations may explain why regolith worldwide is thicker on felsic compared to mafic rock under similar conditions. To understand regolith formation will require better understanding of such deep oxidation reactions and how they impact fluid flow during weathering. Copyright © 2012 John Wiley & Sons, Ltd. KEYWORDS: regolith thickness; fluid transport; neutron scattering; geomorphology; nanoporosity Introduction Formation of regolith – weathered material that lies above unaltered bedrock – influences riverine and ocean chemistry, shapes landscapes, and controls the supply of nutrients. The rate at which regolith forms can be estimated from chemical and isotopic analysis of regolith versus depth (Pavich, 1986; Pavich et al., 1989; Anderson et al., 2002; Brantley and White, 2009; Hilley et al., 2010; Brantley and Lebedeva, 2011). For each mineral–water reaction, the thickness of the reaction front, l, is defined as the depth interval across which the mineral weathers. In actively weathering rocks, each reaction consumes unaltered mineral at the reaction front at the rate of weathering advance, o. The thickness of the reaction front l and the magnitude of the weathering advance rate o are influ- enced by bedrock properties that include porosity, permeabil- ity, grain size, composition, and mineral reactivity. Currently, geologists lack models to predict a priori the depth of regolith. Our inability to estimate rates of regolith formation are particularly important because humans are destroying the top layer of regolith – soil – at rates that are ~ 30 times faster than pre-human rates (Hooke, 2000; Wilkinson and McElroy, 2007). This inability even extends to predictions of regolith depth on the two main crystalline rock types, basalt and granite. A strictly geochemical view would suggest that a rock that weathers quickly to clay should produce a thicker mantle of regolith than a slow-weathering counterpart. For example, feldspars in basaltic rocks weather faster than their granitic counterparts in both the laboratory and field because they are more calcium (Ca)-rich (Colman, 1986; Meybeck, 1987; Drever and Clow, 1995; Wilson, 2004; Bandstra et al., 2008; Hausrath et al., 2009; Heckman and Rasmussen, 2011). However, interro- gation of regolith formation in sites in Virginia indicates that thicker regolith on the more mafic rock, diabase, is not observed. The thickness of regolith is 20m on a granitic ridgetop but only 1 m on a ridgetop composed of the basaltic composition, diabase, despite similar exposure times and geomorphological regimes (Pavich et al., 1985; Pavich, 1986). This contrary behavior, also observed elsewhere around the globe where precipitation exceeds evapotranspiration (Figure 1, Table I), indicates that factors besides mineral composition govern regolith formation. We investigated these factors by analyzing cores drilled in Piedmont ridgetops and reported by Pavich et al. (1989). The Virginia Piedmont (USA) is a forested upland plain that lies at 90–200 m above sea level (m a.s.l.). The diabase we investi- gated is Jurassic (200–145 Myr age) whereas the Occoquan granite is Paleozoic in age (450 Myr age). Both have been exposed to weathering for ≥ 100 Myr. Erosion rates within the Piedmont have been constrained in Virginia and North Carolina by beryllium-10 ( 10 Be) measurements to lie within the range 2  2–8 m Myr –1 (Pavich et al., 1985; Price et al., 2008; Bacon et al., 2012). Given the long exposure and similar elevations of the granite and diabase hilltops (30%), the erosion rates, E, of the ridges are identical within 30% (Pavich et al., 1989). The EARTH SURFACE PROCESSES AND LANDFORMS Earth Surf. Process. Landforms 38, 847–858 (2013) Copyright © 2012 John Wiley & Sons, Ltd. Published online 28 December 2012 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/esp.3369