As bedrock weathers to regolith – defined here as weathered rock, saprolite, and soil – porosity grows, guides fluid flow, and liberates nutrients from minerals. Though vital to terrestrial life, the processes that transform bedrock into soil are poorly understood, especially in deep regolith, where direct observations are difficult. A 65-m-deep borehole in the Calhoun Critical Zone Observatory, South Carolina, provides unusual access to a complete weathering profile in an Appalachian granitoid. Co-located geophysical and geochemical datasets in the borehole show a remarkably consistent picture of linked chemical and physical weathering processes, acting over a 38-m-thick regolith divided into three layers: soil; porous, highly weathered saprolite; and weathered, fractured bedrock. The data document that major minerals (plagioclase and biotite) commence to weather at 38 m depth, 20 m below the base of saprolite, in deep, weathered rock where physical, chemical and optical properties abruptly change. The transition from saprolite to weathered bedrock is more gradational, over a depth range of 11–18 m. Chemical weathering increases steadily upward in the weathered bedrock, with intervals of more intense weathering along fractures, documenting the combined influence of time, reactive fluid transport, and the opening of fractures as rock is exhumed and transformed near Earth’s surface.

We couple meteoric 10 Be measurements with mass balance analysis of 9 Be to estimate the soil residence time (SRT) of a biogeomorphically stable Ultisol in the Southern Piedmont physiographic region of the southeastern United States. We estimate SRT after correcting the meteoric 10 Be inventory to account for observed 9 Be losses, which indicate that more than half of the 9 Be weathered from primary minerals has been leached from the upper 18.3 m of the Ultisol. Our estimates of minimum SRT range between 1.3-1.4 Ma and between 2.6-3.1 Ma under high and low (2.0 and 1.3 × 10 6 atoms cm -2 yr -1 , respectively), estimates of 10 Be delivery. Denudation rates of the physiographic region corroborate our estimates. We redefi ne pedogenic time constraints in the Southern Piedmont, and demonstrate that the assumption of complete meteoric 10 Be retention in acidic soil systems cannot always be made; the latter has far-reaching consequences for soil, sediment, river, and ocean research using meteoric 10 Be.

Reforestation of formerly cultivated land is widely understood to accumulate above- and belowground detrital organic matter pools, including soil organic matter. However, during 40 years of study of reforestation in the subtropical southeastern USA, repeated observations of above- and belowground carbon documented that significant gains in soil organic matter (SOM) in surface soils (0-7.5 cm) were offset by significant SOM losses in subsoils (35-60 cm). Here, we extended the observation period in this long-term experiment by an additional decade, and used soil fractionation and stable isotopes and radioisotopes to explore changes in soil organic carbon and soil nitrogen that accompanied nearly 50 years of loblolly pine secondary forest development. We observed that accumulations of mineral soil C and N from 0 to 7.5 cm were almost entirely due to accumulations of light-fraction SOM. Meanwhile, losses of soil C and N from mineral soils at 35 to 60 cm were from SOM associated with silt and clay-sized particles. Isotopic signatures showed relatively large accumulations of forest-derived carbon in surface soils, and little to no accumulation of forest-derived carbon in subsoils. We argue that the land use change from old field to secondary forest drove biogeochemical and hydrological changes throughout the soil profile that enhanced microbial activity and SOM decomposition in subsoils. However, when the pine stands aged and began to transition to mixed pines and hardwoods, demands on soil organic matter for nutrients to support aboveground growth eased due to pine mortality, and subsoil organic matter levels stabilized. This study emphasizes the importance of long-term experiments and deep measurements when characterizing soil C and N responses to land use change and the remarkable paucity of such long-term soil data deeper than 30 cm.

No abstract