Deep Soil Carbon: Suprises We Find When We Keep Digging

At the surface, most soil organic C starts in plant roots, with belowground inputs five times more likely to persist than aboveground plant biomass. Microbes swarm these roots, driving blooms of activity and predation and releasing necromass, enzymes, and protein- and polysaccharide-rich extracellular polymeric substances (EPS) that aggregate soil and lock in C. But in deeper soil horizons, C is acted upon via distinct metabolic pathways. Indeed, deep soils are the cradle of soil formation--at the bedrock-soil interface (“regolith”), mineral weathering changes soil pH and releases nutrients and secondary mineral forming ions (Si, Al, Fe), impacting soil structure, microbial composition and activity, and the turnover time of soil organic matter. While deep soil horizons (>30 cm) are often considered biologically quiescent, deep soil C is highly sensitive to environmental change and comprises the majority of the global soil C pool. Our deep soils research has identified many active microorganisms at depth: “dark autotrophs” with genes for non-photosynthetic CO2 fixation, archaeal ammonia oxidizers, symbiotrophic fungi, and evidence of mineral weathering that forges secondary minerals, setting the stage for long-lasting mineral-associated organic matter (MAOM). We have also tested the durability of deep soil carbon. Roots of hardy perennial grasses that penetrate down to a meter (or more) introduce a net flux of radiocarbon-young recently fixed carbohydrates, but these new resources do not seem to accelerate decomposition (priming), and the added carbon is often short-lived. Persistence of deep root carbon does not appear to correlate with common soil properties and environmental factors (silt+clay, cation exchange capacity, pH, precipitation, temperature). However, certain soil organic matter components do have markedly distinct turnover times. When we used compound specific 14C analyses along a soil depth profile, we found significant differences in the 14C age of distinct organic molecules (acid insoluble>bulk soil carbon> chloroform-extracted microbial biomass, total lipids and amino acids>phospholipids and respired CO2). These findings suggest that in deep soils, organic matter consumers appear to preferentially use young C transported from the surface (perhaps as roots or dissolved organic carbon) rather than recycling older soil organic carbon. These data also suggest that molecular structure may play a role in soil organic matter stability in the oligotrophic habitat of deep soils.