Preservation of organic carbon (OC) in marine and terrestrial deposits is enhanced by bonding with reactive iron (Fe_R). Association of OC with Fe_R (OC‐Fe_R) provides physical protection and hinders microbiological degradation. Roughly 20% of all OC stored in unconsolidated marine sediments and 40% of all OC present in Quaternary terrestrial deposits is preserved as OC‐Fe_R, but this value varies from 10% to 80% across global depositional environments. Here, we provide a new assessment of global OC‐Fe_R burial rates in both marine and terrestrial environments, using published estimates of OC associated with Fe_R, carbon burial, and probabilistic modeling. We estimate the marine OC‐Fe_R sink between 31 and 70 Mt C yr⁻¹ (average 52 Mt C yr⁻¹), and the terrestrial OC‐Fe_R sink at between 146 and 917 Mt C yr⁻¹ (average 446 Mt C yr⁻¹). In marine environments, continental shelves (average 17 Mt C yr⁻¹) and deltaic/estuarine environments (average 11 Mg C yr⁻¹) are the primary settings of OC‐Fe_R burial. On land, croplands (279 Mt C yr⁻¹) and grasslands (121 Mt C yr⁻¹) dominate the OC‐Fe_R burial budget. Changes in the Earth system through geological time impact the OC‐Fe_R pools, particularly in marine settings. For example, periods of intense explosive volcanism may lead to increased net OC‐Fe_R burial in marine sediments. Our work highlights the importance of OC‐Fe_R in marine carbon burial and demonstrates how OC‐Fe_R burial rates may be an order of magnitude greater in terrestrial environments, but here OC‐Fe_R stocks are most sensitive to the anthropogenic impacts of climatic change.

Lead (Pb) isotopes provide valuable insights into the origin of Pb within a sample, typically allowing for reliable fingerprinting of their source. This is useful for a variety of applications, from tracing sources of pollution-related Pb, to the origins of Pb in archaeological artefacts. However, current approaches investigate source proportions via graphical means, or simple mixing models. As such, an approach, which quantitatively assesses source proportions and fingerprints the signature of analysed Pb, especially for larger numbers of sources, would be valuable. Here we use an advanced Bayesian isotope mixing model for three such applications: tracing dust sources in pre-anthropogenic environmental samples, tracking changing ore exploitation during the Roman period, and identifying the source of Pb in a Roman-age mining artefact. These examples indicate this approach can understand changing Pb sources deposited during both pre-anthropogenic times, when natural cycling of Pb dominated, and the Roman period, one marked by significant anthropogenic pollution. Our archaeometric investigation indicates clear input of Pb from Romanian ores previously speculated, but not proven, to have been the Pb source. Our approach can be applied to a range of disciplines, providing a new method for robustly tracing sources of Pb observed within a variety of environments.

Significance: A detailed record of historical lead (Pb) pollution from a peat bog in Serbia provides a unique view on the extent and timing of Balkan mining and metallurgy. Evidence of the earliest European environmental pollution is followed by large-scale and sustained increases in the amount of anthropogenically derived Pb after 600 BCE, through the Roman/Byzantine periods, and into the medieval period. Occasional evidence of drops in pollution output reflects the disruptive socioeconomic impact of periods of turmoil. Our data show a trend significantly different to records in western Europe, where Pb pollution decreases dramatically after the collapse of the Roman Empire. These results suggest metal-rich southeastern Europe should be considered a more major player in environmental metal pollution through time.

Reconstructions of dust flux have been used to produce valuable global records of changes in atmospheric circulation and aridity. These studies have highlighted the importance of atmospheric dust in marine and terrestrial biogeochemistry and nutrient cycling. By investigating a 10 800-year-long paleoclimate archive from the Eastern Carpathians (Romania) we present the first peat record of changing dust deposition over the Holocene for the Carpathian–Balkan region. Using qualitative (X-ray fluorescence (XRF) core scanning) and quantitative inductively coupled plasma optical emission spectrometer(ICP-OES) measurements of lithogenic (K, Si, Ti) elements, we identify 10 periods of major dust deposition between 9500–9200, 8400–8100, 7720–7250, 6350–5950, 5450–5050, 4130–3770, 3450–2850, 2000–1450, 800–620, and 60 cal yr BP to present. In addition, we used testate amoeba assemblages preserved within the peat to infer local palaeohydroclimatic conditions. Our record highlights several discrepancies between eastern and western European dust depositional records and the impact of highly complex hydrological regimes in the Carpathian region. Since 6100 cal yr BP, we find that the geochemical indicators of dust flux have become uncoupled from the local hydrology. This coincides with the appearance of millennial-scale cycles in the dust input and changes in geochemical composition of dust. We suggest that this is indicative of a shift in dust provenance from local–regional (likely loess-related) to distal (Saharan) sources, which coincide with the end of the African Humid Period and the onset of Saharan desertification.