New insights into source and dispersal of Mediterranean S1 tephra, an early Holocene marker horizon erupted at Mt. Erciyes (Turkey)

Abstract Deposition of early Holocene Eastern Mediterranean S1 tephra and a Black Sea cryptotephra coincides with cultural transitions in the Fertile Crescent termed the Neolithic Revolution as well as sapropel formation during climate variability of the African humid period, classifying them as paramount regional marker horizons for archaeology as well as paleoclimatology. Their correlations with specific eruptions of the Mt. Erciyes stratovolcanic complex (Central Anatolia) remained inconclusive though. Here, we use zircon double-dating by (U–Th)/He and U–Th disequilibrium methods, major and trace element tephra glass geochemistry, and probabilistic modeling of tephra dispersal in an attempt to characterize all major late Quaternary proximal tephras of Mt. Erciyes, and to correlate them with distal deposits. Furthermore, we discuss contrasting proximal and distal tephra dispersal. Three nearly-coeval rhyolitic satellite domes (Dikkartin, Perikartin, and Karagullu) erupted at Mt. Erciyes in the early Holocene, and their dome extrusions were all preceded by explosive phases producing pyroclastic material that formed tephra fall and pyroclastic flow deposits. The new eruption age of 9.03 ± 0.55 ka (1σ uncertainty here and elsewhere) for proximal Dikkartin pumice is consistent with 14C-based S1 tephra chronologies in distal locations averaging 8.92 ± 0.03 cal ka BP. Perikartin pyroclastic flow deposits predate S1 tephra by ca. 0.8 ka according to a pair of published 14C ages, and stratigraphically overlie Karagullu fall-out, here dated to 8.2 ± 1.8 ka. Previously undated proximal tephras of Mt. Erciyes erupted in the Late (85.2 ± 4.9 ka) and Middle Pleistocene (154.5 ± 5.3 ka). S1 tephra glass is chemically similar to that of Dikkartin fall-out, but also indistinguishable from that of Perikartin fall-out. Karagullu pumice is characterized by a distinct glass chemical composition, which correlates with that of unnamed cryptotephra reported for the southeastern Black Sea instead, where these results call for a re-evaluation of existing age models. Maximum lithic clast size isopleths for proximal Dikkartin fall-out indicate eastward dispersal of a 20 ± 5 km high eruption plume by stratospheric winds, in agreement with results of probabilistic tephra dispersal modeling. This azimuth contrasts with the known distribution of S1 tephra at distal locations that are all south of Mt. Erciyes. Significant tephra occurrences at up to 1300 km distance and orthogonal to prevalent stratospheric wind directions either result from very atypical wind conditions (probability ≪10 %), or are caused by tephra transport by prevailing low altitude winds. Two scenarios are proposed for low altitude transport: eolian reworking of primary fall-out (more likely from the more widespread Dikkartin deposits), or co-ignimbrite ash cloud dispersal (more likely from the Perikartin eruption which predominantly produced pyroclastic flows). Because S1 tephra is chemically indistinguishable from both Dikkartin and Perikartin by major and trace element glass compositions, its exact source and dispersal mechanism remain ambiguous, although existing 14C ages for Perikartin predating those for S1 tephra favor Dikkartin as its source.