The Carbon Story at the Shale Hills Critical Zone Observatory.

Forest ecosystems play a leading role in the global terrestrial carbon (C) cycle owing to their huge C pool and high productivity. However, while much research has focused on the dynamics of C, the factors that control C export are still debatable. The objectives of this 2-yr study conducted at the forested Shale Hills Critical Zone Observatory in central PA were to (1) evaluate the vertical profiles of soil organic C (SOC) and dissolved organic C (DOC) along two different hydropedological functional units (HFUs) (swale and planar hillslopes) and (2) examine the spatiotemporal patterns of DOC in soil pore waters and stream water to understand export controls on DOC. Results showed that SOC was not significantly different between the two HFUs and decreased with increasing soil depth. The largest decrease (~ 60%) was across the A - B horizon interface. DOC concentrations along the swale were consistently higher than concentrations along the planar hillslope with concentrations also decreasing with soil depth for both HFUs. The two HFUs showed a decreasing trend for DOC concentrations: valley floor ≥ ridge-top > mid-slope. The main factors controlling the concentration of DOC along the HFUs were soil horizon interfaces, particularly the B_w-B_t and the soil-bedrock interfaces and soil moisture content. At these interfaces, DOC concentrations and soil moisture content were consistently higher compared to other soil depths indicating that concentrations are higher along lateral preferential flowpaths compared to the soil matrix; making soil horizon interfaces major hot spots for DOC. Streamwater DOC varied from 0.6 and 28.6 mg/L based on daily sampling at the catchment outlet and DOC export was calculated as 40 kg/ha/yr.  Monthly DOC concentrations and export increased during snowmelt and late summer/early fall wet-up (hot moments for DOC), with variations in discharge explaining 63% of the variability in DOC concentrations. Streamwater DOC was also significantly (p < 0.05) correlated to streamwater temperature (R² = 0.25) when data was aggregated monthly. These results reflect a combined control of biological activity (related to temperature) as well as flushing (linked to discharge).