Holocene tephra record of Lake Veliko jezero, Croatia: implications for the central Mediterranean tephrostratigraphy and sea level rise

Abstract

Core M1‐A from Mljet Island, Croatia, was retrieved from a submerged sinkhole to investigate tephra and reconstruct past sea levels. Eleven tephra layers were found, out of which six are macroscopically visible, while five are cryptotephra. For two of the tephra layers, glass shard concentrations were below the critical amount necessary for reliable analysis, while two more originated from a stratigraphical interval likely disturbed by drilling operations. Major and trace element compositions of glass shards were determined by wavelength dispersive spectroscopy (WDS) using an electron microprobe analyser (EMPA) and by laser ablation inductively coupled plasma mass spectrometry (LA‐ICP‐MS). The tephra discrimination relied on a novel approach based on a multivariate analysis of element selection and the use of log‐ratio scatterplots and common bivariate plots. The results enabled correlation of five tephra to known eruptions originating from Somma‐Vesuvius and Campi Flegrei. Specifically, we identified Avellino Pumice, Mercato, Agnano Monte Spina and Neapolitan Yellow Tuff, extending known distributions for Avellino and Agnano Monte Spina. Moreover, our findings possibly support an earlier proposition that Agnano Monte Spina tephra originated from two eruptions with a pause of a few decades in between. Based on the tephra correlations and radiocarbon dating, an age‐depth model was compiled that provided chronological constraints for the sea level during the formation of the lake (10.7 cal. ka BP, 49 m b.s.l.) and sea intrusion (2.3 cal. ka BP, 2.5 m b.s.l.).

Past climate variations recorded in needle-like aragonites correlate with organic carbon burial efficiency as revealed by lake sediments in Croatia

“…The lake is part of the karst depression system, which is now a submerged sinkhole due to Holocene sea level rise. At the early Holocene, 10.8 cal ka BP 25 , because of the sea level rise and the presence of permeable karst, a wetland formed that transitioned into a brackish lake and finally ended as a marine lake because of sea flooding through the Soline channel (2.5 m deep) 2.3 cal ka BP 25 . The lake was perennial, with an almost constant salinity of ~25‰ during the studied interval 33 (8.3–2.6 cal ka BP).…”

“…We compared the Sr/Ca data and ratios between the two investigated cores (M1-A and M2) to confirm the robustness of our record and exclude any potential analytical artifacts. M1-A core chronology is based on four tephra layers and four C-14 dates 25 . Briefly, the studied interval chronology is very well constrained since in ~1.25 m, there are three tephra layers, which serve as depth-age model anchor points: two charcoal C-14 dates free of reservoir effects and one marine shell C-14 date with known reservoir effects (Fig.…”

“…Three additional data points were included in the age-depth model of this core. Two of these ages are based on visible tephra layers correlated to the known volcanic eruptions previously recognized in Core M1-A 25 . The third date corresponds to a time of marine intrusion into the lake, which is marked by a sharp boundary between the top and underlying lithological units (Fig.…”

“…2 ) in both cores. It was radiometrically dated in Core M1-A 25 . Individual radiocarbon dates were calibrated using the R package rbacon 34 (the age-depth model for core M2 is presented in the Supplementary file, Figure S4).…”

“…An age-depth model of Core M1-A (Fig. 2 ) was previously described 25 . The significance of the Sr/Ca ratio as a paleotemperature proxy was also confirmed by correlation with other previously published studies, including the main Holocene climatic events from the wider Mediterranean region 26 – 31 .…”