Other language confidence: 0.8922026236111672
Trace element contents in microg/g measured on the <2 microns, 2-20 microns size fractions and bulk samples from LGM European loess sequences. Samples were crushed in an agate mortar and trace element concentrations were measured following Chauvel et al. (2011). Reproducibility for trace element analyses is better than 5% based on repeat measurements, and the accuracy is also better than 5%, based on the analyses of international rock standards (JSD-1, JSD-3 and LKSD-1.
Sr-Pb isotopic ratios for the <2 microns, 2-20 microns size fractions and bulk samples from LGM European loess sequences. Samples were crushed in an agate mortar and Sr and Pb isotopes were measured following Chauvel et al. (2011). Lead and strontium were analyzed after dissolution of the powder and without leaching procedure. Blanks run in parallel with the samples during the course of this work were less than 80 pg of Pb and 250 pg of Sr for the entire chemical separation procedure. These amounts are negligible relative to the amount of Pb and Sr present in the beakers after isolation of pure Pb and Sr (typically, 600 ng of Pb and 4 microg of Sr).
In a previous analysis of bulk sediment from Last Glacial Maximum loess deposits over Europe, a local to regional origin of the deposited material has been demonstrated (Rousseau et al., 2014). In a refined study, the same European sequences have been analyzed studying the trace elements and the strontium and lead isotopic ratios from different grain-size fractions: <2 micron, 2-20 micron and bulk. The later was reanalyzed to compare with the previous study. First the new bulk values are similar to the previously published ones. Furthermore, the results show that the <2 micron fraction is different than the 2-20 micron or bulk ones when comparing the trace element. Second, the 87Sr/86Sr vs 208Pb/204Pb diagram shows that the sequences analyzed keep their regional to local signature, mainly along the 87Sr/86Sr axis for the 2-20 micron and bulk samples. On the contrary, the <2 micron samples are all grouped in a well identified cluster clearly indicating a different source and even a different transport. The comparison of the results of the geochemical analysis with climate modeling of dust transport at LGM supports this interpretation. The datasets correspond to i) the code and geographical coordinates, ii) the trace element contents for the <2 micron, 2-20 micron and bulk fractions, iii) the Sr-Pb isotopic ratios for the <2 micron, 2-20 micron and bulk fractions, and iv) the grain size composition of the studied LGM European loess samples.
Grain size composition of loess samples from LGM European loess sequences. Loess samples of about 200 g were prepared to extract the grain size fractions studied. Grain size separations were performed on at least 10 g of dry sample. First, the entire sample was sieved with demineralized water on 63 microns and 20 microns sieves. The rejects were collected, dried and weighed. The clay fraction was obtained by decanting the fraction below 20 microns. The rest of the sample was mixed and left to settle for 1 hour. This procedure is repeated until a transparent supernatant is obtained. The two fractions thus obtained are dried and weighed. The size of the different fractions was then checked by laser granulometry.
The dataset contains long‐term means of δ18Oprecipitation values and temperatures from 84 European sites (GNIP database; IAEA/WMO, 2023), which were used to estimate mean annual (palaeo-) temperatures from the measured oxygen isotope data on horse tooth enamel phosphate. Mean temperatures of the warmest (July/August) and coldest (December/January) periods were considered representative for summer and winter temperatures, respectively, using the peak and trough values of the modelled phosphate data for their calculation.
Paleo±Dust is an updated compilation of bulk and <10-µm paleo-dust deposition rate with quantitative 1-σ uncertainties that are inter-comparable among archive types (lake sediment cores, marine sediment cores, polar ice cores, peat bog cores, loess samples). Paleo±Dust incorporates a total of 285 pre-industrial Holocene (pi-HOL) and 209 Last Glacial Maximum (LGM) dust flux constraints from studies published until December 2022. We also recalculate previously published dust fluxes to exclude data from the last deglaciation and thus obtain more representative constraints for the last pre-industrial interglacial and glacial end-member climate states. Metadata include all components necessary to derive dust deposition rate, including: age range, thickness, density, eolian content. We also include 1-sigma uncertainties on each of these components, and on the final bulk and <10-µm dust deposition rates. Specific notes for each site and a list of references are also included.
Our study focuses on European loess sequences, particularly the eolian intervals in between the observed pedogenic units. The classical concept of soil formation from parent material is reformulated to estimate of the duration and the associated sedimentation rate (SR) and mass accumulation rate (MAR) of these paleodust intervals. We show that the Greenland Stadial (GS) duration in European loess deposits includes the thickness of the overlying pedogenic unit, which in fact developed downward into the upper part of the eolian unit. The lower stratigraphical limit of the eolian unit overlying the pedogenic unit corresponds to the restart of the dust sedimentation of the younger GS. We illustrate this interpretation first by computing both SRs and MARs first for the Nussloch key sequence, the most complete European series. The correlation between Nussloch and other European loess sequences, located along a 1,800 km longitudinal transect, allows computation of SR and MAR for several identified GS events. Comparing GS from marine and ice core records, our study shows that the two last Bond cycles are preserved in every European eolian record. Bulk SR and MAR are estimated and compared for these two Bond cycles, showing the highest SRs and MARs in western Europe. These indices also indicate that the last stadials, embedding an Henrich event, were not the dustiest in every Bond cycle. Our estimated MAR also differ from previously published computations, which did not take into account the various pedogenic units present in the studied loess sequences. The bulk SR and MAR estimates computed for the two last Bond Cycles from Chinese sequences from the Loess Plateau indicate lower atmospheric dust than in Europe during the Last Glacial Maximum. SR and MAR estimates computed from the fine-grained material for European records fit with Earth System model reconstructions.
| Organisation | Count |
|---|---|
| Wissenschaft | 15 |
| Type | Count |
|---|---|
| Daten und Messstellen | 15 |
| License | Count |
|---|---|
| Offen | 15 |
| Language | Count |
|---|---|
| Englisch | 15 |
| Resource type | Count |
|---|---|
| Archiv | 3 |
| Datei | 12 |
| Topic | Count |
|---|---|
| Boden | 14 |
| Lebewesen und Lebensräume | 8 |
| Luft | 5 |
| Mensch und Umwelt | 11 |
| Wasser | 7 |
| Weitere | 15 |