The marine radiocarbon reservoir effect is an offset in 14 C age between contemporaneous organisms from the terrestrial environment and organisms that derive their carbon from the marine environment. Quantification of this effect is of crucial importance for correct calibration of the 14 C ages of marine-influenced samples to the calendrical time-scale. This is fundamental to the construction of archaeological and palaeoenvironmental chronologies when such samples are employed in 14 C analysis.Quantitative measurements of temporal variations in regional marine reservoir ages also have the potential to be used as a measure of process changes within Earth surface systems, due to their link with climatic and oceanic changes. The various approaches to quantification of the marine radiocarbon reservoir effect are assessed, focusing particularly on the North Atlantic Ocean. Currently, the global average marine reservoir age of surface waters, R(t), is c.400 radiocarbon years, however, regional values deviate from this as a function of climate and oceanic circulation systems. These local deviations from R(t) are expressed as ΔR values. Hence, polar waters exhibit greater reservoir ages (ΔR = c. +400 to +800 14 C y) than equatorial waters (ΔR = c.0 14 C y). 2Observed temporal variations in ΔR appear to reflect climatic and oceanographic changes. We assess three approaches to quantification of marine reservoir effects using known age samples (from museum collections), tephra isochrones (present onshore/offshore), and paired marine/terrestrial samples (from the same context in, for example, archaeological sites). The strengths and limitations of these approaches are evaluated using examples from the North Atlantic region. It is proposed that, with a suitable protocol, accelerator mass spectrometry (AMS) measurements on paired, short lived, single entity marine and terrestrial samples from archaeological deposits is the most promising approach to constraining changes over at least the last 5 ky BP.

Mountain rivers play a key role in the delivery of particulate organic carbon (POC) to large river systems and the ocean. Due to the extent of its drainage area and runoff, the Amazon River is one of Earth's most important biogeochemical systems. However, the source of POC eroded from the humid region of the Eastern Andes and the input of fossil POC from sedimentary rocks (POCfossil) remains poorly constrained. Here we collected suspended sediments from the Kosñipata River during flood events to better characterize Andean POC, measuring the nitrogen to organic carbon ratio (N/C), stable carbon isotopes (δ13Corg) and radiocarbon (Δ14Corg). Δ14Corg values ranged from −711‰ to −15‰, and significant linear trends between Δ14Corg, N/C and δ13Corg suggested that this reflects the mixing of POCfossil with very young organic matter (Δ14Corg ~ 50‰) from the terrestrial biosphere (POCnon‐fossil). Using N/C and Δ14Corg in an end‐member mixing analysis, we quantify the fraction of POCfossil (to within 0.1) and find that it contributes a constant proportion of the suspended sediment mass (0.37 ± 0.03%) and up to 80% of total POC. In contrast, the relative contribution of POCnon‐fossil was variable, being most important during the rising limb and peak discharges of flood events. The new data shed light on published measurements of “old” POC (low Δ14Corg) in Andean‐fed tributaries of the Amazon River, with their Δ14Corg and δ13Corg values consistent with variable addition of POCfossil. The findings suggest a greater persistence of Andean POC in the lowland Amazon than previously recognized.

No abstract