A model for archaeologically relevant Holocene climate impacts in the Aegean-Levantine region (easternmost Mediterranean)

Abstract

A repeating pattern of multi-centennial-scale Holocene climate events has been widely (globally) documented, and they were termed Rapid Climate Change (RCC) events. Non-seasalt potassium ion (K+) series in Greenland ice cores provide well-constrained timings for the events, and a direct timing relationship has been inferred between these events and the frequency of northerly cold polar/continental air outbreaks over the eastern Mediterranean Sea through gaps in the mountain ranges along the northern margin of the basin. There also appears to be a remarkable timing agreement with major archaeological turnover events in the Aegean/Levantine region. Yet no physically consistent assessment exists for understanding the regional climatic impacts of the events around this critical region. We present a simple 2-dimensional Lagrangian model, which yields a broad suite of physically coherent simulations of the impacts of frequency changes in winter-time northerly air outbreaks over the Aegean/Levantine region. We validate this with existing reconstructions from palaeoclimate proxy data, with emphasis on well-validated sea-surface temperature reconstructions and a highly resolved cave speleothem stable oxygen isotope record from Lebanon. Given that the RCCs were clearly marked by negative sea surface temperature anomalies in the region, we find that the predominant climatic impacts of this winter-time mechanism were “cold and wet,” in contrast with intercalated “warmer and more arid” conditions of non-RCC periods. More specifically, the RCCs are found to be periods of highly variable conditions, with an overall tendency toward cold and wet conditions with potential for flash flooding and for episodic snow-cover at low altitudes, at least in the lower-altitude (lower 1–1.5 km) regions of Crete and the Levant. The modelled winter-anomaly process cannot address underlying longer-term, astronomically forced trends, or the relatively warm and arid anomalies in between RCCs. The latter require further study, for example with respect to potential (summer-time?) extension of evaporative subtropical conditions over the region. Finally, our results imply that the “amount effect” observed in Levantine cave δ18O (and precipitation or drip-water δ18O) may not reflect the conventional concept related to temperature-dependent fractionation and Rayleigh distillation. Instead, it appears to arise from a complex and somewhat counter-intuitive mixing, in shifting proportionalities, between advected (external) and evaporated (Mediterranean) moisture.