Description: During the last deglaciation (18–8 kyr BP), shelf flooding and warming presumably led to a large-scale decomposition of permafrost soils in the mid-to-high latitudes of the Northern Hemisphere. Microbial degradation of old organic matter released from the decomposing permafrost potentially contributed to the deglacial rise in atmospheric CO2 and also to the declining atmospheric radiocarbon contents (Δ14C). The significance of permafrost for the atmospheric carbon pool is not well understood as the timing of the carbon activation is poorly constrained by proxy data. Here, we trace the mobilization of organic matter from permafrost in the Pacific sector of Beringia over the last 22 kyr using mass-accumulation rates and radiocarbon signatures of terrigenous biomarkers in four sediment cores from the Bering Sea and the Northwest Pacific. We find that pronounced reworking and thus the vulnerability of old organic carbon to remineralization commenced during the early deglaciation (~16.8 kyr BP) when meltwater runoff in the Yukon River intensified riverbank erosion of permafrost soils and fluvial discharge. Regional deglaciation in Alaska additionally mobilized significant fractions of fossil, petrogenic organic matter at this time. Permafrost decomposition across Beringia's Pacific sector occurred in two major pulses that match the Bølling-Allerød and Preboreal warm spells and rapidly initiated within centuries. The carbon mobilization likely resulted from massive shelf flooding during meltwater pulses 1A (~14.6 kyr BP) and 1B (~11.5 kyr BP) followed by permafrost thaw in the hinterland. Our findings emphasize that coastal erosion was a major control to rapidly mobilize permafrost carbon along Beringia's Pacific coast at ~14.6 and ~11.5 kyr BP implying that shelf flooding in Beringia may partly explain the centennial-scale rises in atmospheric CO2 at these times. Around 16.5 kyr BP, the mobilization of old terrigenous organic matter caused by meltwater-floods may have additionally contributed to increasing CO2 levels.

Description: The western tropical Atlantic plays an important role in the interhemispheric redistribution of heat during millennial-scale changes in the strength of the Atlantic Meridional Overturning Circulation (AMOC). The proper evaluation of this role depends on a clear understanding of sea surface temperature (SST) variations during AMOC slowdown periods like Heinrich Stadials (HS) in the western tropical Atlantic. However, published SST records from the western tropical Atlantic between ca. 4°S and 7°N show inconsistencies that are apparently related to the employed temperature proxy (i.e., Mg/Ca versus alkenone unsaturation index U37k′). In general, while Mg/Ca values indicate warming during Heinrich Stadials, U37k′ values show cooling. To assess this issue, we sampled core GeoB16224-1 retrieved off French Guiana (i.e., 6°39.38′N) and reconstructed water temperatures at high resolution using Mg/Ca on the foraminifera species Globigerinoides ruber, U37k′, TEX86 and modern analogue technique (MAT) transfer functions using planktonic foraminifera assemblages calibrated for 50 m water depth. Our results show that Mg/Ca and TEX86 values recorded an increase in SST related to AMOC slowdown. Conversely, U37k′ and MAT values registered a decrease in temperatures during HS3 and HS1. Our U37k′ and Mg/Ca results thus confirm the previously reported inconsistency for the period between 48-13 cal ka BP. We suggest that several non-thermal physiological effects probably imparted a negative temperature bias on the U37k′ temperatures during Heinrich Stadials. However, MAT-based temperatures show similar variability with U37k′-based temperatures. Hence, we also suggest that during severe slowdown periods of the AMOC, a steeper meridional temperature gradient together with a southward shift of the Intertropical Convergent Zone produced not only an increase in SST but also a stronger upper water column stratification and a shoaling of the thermocline, decreasing subsurface temperatures. Our new high resolution temperature records allow a better characterization of the thermal response of the upper water column in the tropical western Atlantic to slowdown events of the AMOC, reconciling previously discrepant records.

Description: This is the temperature lipid proxy (GDGT) data measured in sinking particles collected using a sediment trap mooring system in the eastern Fram Strait and in the Antarctic Polar Front of the Atlantic sector. In the eastern Fram Strait (FEVI16), particles were collected at 1296 m water depth. In the Antarctic Polar Front (PF3), particles were collected at 614 m and 3196 m water depth. Using this dataset, TEX86-derived temperatures and OH-GDGTs-derived temperatures can be calculated.

Description: The dataset comprises contents, mass-accumulation rates and compound-specific radiocarbon data of several terrigenous biomarkers in sediment cores from the Northwest Pacific (core SO201-2-12KL) and the Bering Sea (cores SO201-2-114KL, SO202/1_18-6 and SO202/1_18-3). The dataset includes contents and mass accumulation rates of high molecular weight n-alkanoic acids, high molecular weight n-alkanes and branched glycerol dialkyl glycerol tetraethers. Moreover, contents and fractional abundances of homohopane isomers are given. Contents are calculated relative to total organic carbon contents and to the dry weight of the sediment. The dataset also contains Paq values and the carbon preference indices of n-alkanoic acids and n-alkanes. Compound-specific radiocarbon data have been obtained from high molecular weight n-alkanoic acids (n-C26:0 alkanoic acid) and high molecular weight n-alkanes (n-C23, n-C27 and n-C29 alkanes). The F14C has been corrected for procedural blanks associated with the isolation of the biomarkers. The data compilation contains the measured (AMS-results) as well as the blank-corrected F14C values. The 14C-data are reported with the 1sigma uncertainty.

Description: Compound-specific radiocarbon data have been obtained from terrigenous biomarkers, i.e. high molecular weight n-alkanoic acids (n-C26:0 alkanoic acid) and high molecular weight n-alkanes (n-C23, n-C27 and n-C29 alkanes). The F14C has been corrected for procedural blanks associated with the isolation of the biomarkers. The data compilation contains the measured (AMS-results) as well as the blank-corrected F14C values. The 14C-data are reported with the 1sigma uncertainty.

Description: Compound-specific radiocarbon data have been obtained from terrigenous biomarkers, i.e. high molecular weight n-alkanoic acids (n-C26:0 alkanoic acid) and high molecular weight n-alkanes (n-C23, n-C27 and n-C29 alkanes). The F14C has been corrected for procedural blanks associated with the isolation of the biomarkers. The data compilation contains the measured (AMS-results) as well as the blank-corrected F14C values. The 14C-data are reported with the 1sigma uncertainty.

Description: Compound-specific radiocarbon data have been obtained from terrigenous biomarkers, i.e. high molecular weight n-alkanoic acids (n-C26:0 alkanoic acid) and high molecular weight n-alkanes (n-C23, n-C27 and n-C29 alkanes). The F14C has been corrected for procedural blanks associated with the isolation of the biomarkers. The data compilation contains the measured (AMS-results) as well as the blank-corrected F14C values. The 14C-data are reported with the 1sigma uncertainty.

Description: Compound-specific radiocarbon data have been obtained from terrigenous biomarkers, i.e. high molecular weight n-alkanoic acids (n-C26:0 alkanoic acid) and high molecular weight n-alkanes (n-C23, n-C27 and n-C29 alkanes). The F14C has been corrected for procedural blanks associated with the isolation of the biomarkers. The data compilation contains the measured (AMS-results) as well as the blank-corrected F14C values. The 14C-data are reported with the 1sigma uncertainty.