Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Late Bronze Age climate change and the destruction of the Mycenaean Palace of Nestor at Pylos

  • Martin Finné ,

    Roles Conceptualization, Formal analysis, Funding acquisition, Investigation, Visualization, Writing – original draft, Writing – review & editing

    martin.finne@antiken.uu.se

    Affiliations Department of Archaeology and Ancient History, Uppsala University, Uppsala, Sweden, Department of Physical Geography, Stockholm University, Stockholm, Sweden, Navarino Environmental Observatory, Navarino Dunes, Costa Navarino, Messinia, Greece

  • Karin Holmgren,

    Roles Conceptualization, Funding acquisition, Investigation, Writing – review & editing

    Affiliations Department of Physical Geography, Stockholm University, Stockholm, Sweden, Navarino Environmental Observatory, Navarino Dunes, Costa Navarino, Messinia, Greece, Swedish University of Agricultural Sciences, Uppsala, Sweden

  • Chuan-Chou Shen,

    Roles Funding acquisition, Investigation, Resources, Validation, Writing – review & editing

    Affiliation High-Precision Mass Spectrometry and Environment Change Laboratory (HISPEC), Department of Geosciences, National Taiwan University, Taipei, Taiwan ROC

  • Hsun-Ming Hu,

    Roles Investigation, Writing – review & editing

    Affiliation High-Precision Mass Spectrometry and Environment Change Laboratory (HISPEC), Department of Geosciences, National Taiwan University, Taipei, Taiwan ROC

  • Meighan Boyd,

    Roles Validation, Writing – review & editing

    Affiliations Department of Physical Geography, Stockholm University, Stockholm, Sweden, Navarino Environmental Observatory, Navarino Dunes, Costa Navarino, Messinia, Greece, Department of Earth Sciences, Royal Holloway, University of London, Egham, Surrey, United Kingdom

  • Sharon Stocker

    Roles Resources, Validation, Writing – review & editing

    Affiliation Department of Classics, University of Cincinnati, 410 Blegen Library, Cincinnati, OH, United States of America

Abstract

This paper offers new high-resolution oxygen and carbon isotope data from Stalagmite S1 from Mavri Trypa Cave, SW Peloponnese. Our data provide the climate background to the destruction of the nearby Mycenaean Palace of Nestor at Pylos at the transition from Late Helladic (LH) IIIB to LH IIIC, ~3150–3130 years before present (before AD 1950, hereafter yrs BP) and the subsequent period. S1 is dated by 24 U-Th dates with an averaged precision of ±26 yrs (2σ), providing one of the most robust paleoclimate records from the eastern Mediterranean for the end of the Late Bronze Age (LBA). The δ18O record shows generally wetter conditions at the time when the Palace of Nestor at Pylos was destroyed, but a brief period of drier conditions around 3200 yrs BP may have disrupted the Mycenaean agricultural system that at the time was likely operating close to its limit. Gradually developing aridity after 3150 yrs BP, i.e. subsequent to the destruction, probably reduced crop yields and helped to erode the basis for the reinstitution of a central authority and the Palace itself.

1. Introduction

The impact of past climate variability and abrupt climate change on ancient human societies is an ongoing debate. This debate often focuses around 1) certain time periods, commonly around so-called climate events, i.e. times when the climate rapidly changed, e.g. at 8200, 4200 and 3200 years before present (before AD 1950, hereafter yrs BP), and 2) archaeologically rich areas, such as SE Mexico (Yucatan Peninsula), the Indus Valley and the eastern Mediterranean [16]. In the eastern Mediterranean, there has been intense discussion about the impact of climate change on the fall of the Akkadian Empire and the end of the Late Bronze Age (LBA) occurring at ~4200 and ~3200 yrs BP respectively [718].

The debate about the causes of the rapid demise of many societies in the eastern Mediterranean at the end of the LBA includes a number of factors such as climate change, earthquakes, famine, political instability and/or invasions by the infamous Sea Peoples [17,1923]. Lately the number of studies investigating the role played by climate at the end of the LBA in the eastern Mediterranean has increased and a number of them suggest aridity as a major factor [9,10,12,14,15,18]. Paleoclimate data, primarily from Cyprus and the Levant, suggest that a 300-year period of arid conditions that began around 3200 yrs BP led to reductions in agricultural productivity and subsequently contributed to a general socioeconomic crisis in the eastern Mediterranean [10,14,15,18]. However, it was recently pointed out that many of these datasets do not have sufficient chronological resolution to reliably tie climate information to archaeological data [20]. Despite chronological uncertainties and the fact that attributing sociopolitical changes to drought can be seen as overly simplistic and deterministic [21,24,25], the idea of widespread aridity has recently gained a prominent position in discussions about LBA societal change in the eastern Mediterranean as well as on the Greek mainland. Even though direct climate evidence from mainland Greece has not been available for this period, it has been suggested that the destruction of the Mycenaean Palatial centers towards the end of the LBA should be viewed in light of the severe aridity recorded for this period elsewhere see e.g. [10]. In this paper, we present new high-resolution paleoclimate data extracted from a cave located just off the coast of the Greek mainland, in close proximity to one of the major Mycenaean Palatial centers.

During the LBA the Mycenaean culture made a strong imprint on a large part of the Aegean world and in particular on the Peloponnesian Peninsula in southern Greece, where a number of palaces functioned as administrative and economic centers [26,27]. One of these was the Mycenaean Palace of Nestor at Pylos, located in present day Messinia in the SW Peloponnese, which controlled large areas of land in that region. The Mycenaean culture reached its zenith between approximately 3350 and 3150 yrs BP. The destruction of the Mycenaean palaces throughout the Peloponnese occurred at the transition between the Late Helladic (LH) IIIB (~3280 to ~3150 yrs BP) and LH IIIC (~3150 to ~3020 yrs BP). It was followed by a period of abrupt decline, during which the use of writing (Linear B) and complex forms of political, economic and social organization disappeared, settlement patterns changed, and the size and number of sites were drastically reduced [2632]. However, some Mycenaean cultural traits continued throughout the LH IIIC period [26,29]. In addition, there is also evidence for rebuilding at the Mycenaean palace at Tiryns, some 120 km to the NE of Pylos [33]. By the Protogeometric period (~3020 yrs BP), new social frameworks were established displacing Mycenaean traditions [26,29]. The destruction of the palaces has often been seen as the prime marker for societal change on the mainland. However, the broader question is not why the palaces were destroyed, but rather why they were not rebuilt [12]. Several factors, among them rapid climate change, have been discussed as triggers for the sudden destruction of the palaces and the inability of Mycenaean society to recover [26,29,30].

The Mycenaean Palace of Nestor at Pylos provides a rare case study where textual and archaeological evidence can be combined to offer a more complete picture of the local economy (S1 File). The Palace functioned as the central administrative center of ancient Messinia and played a crucial role in the economy, but it did not maintain control over every aspect [34,35]. Mycenaean palaces were redistributive centers, although in a more nuanced manner than traditionally thought, based upon their ability to mobilize resources [3638]. The economy was organized around the production, acquisition and distribution of raw materials and prestige goods, such as textiles, i.e., linen and wool, and perfumed olive oil [34,35]. The Palace did not, however, maintain control over the production of all staple goods; for example Linear B tablets indicate that there were local, independent small-scale producers of cereal crops [34,35]. The agricultural economy of the area was largely rainfed and land use would have ranged from large-scale flax fields and olive groves, possibly grazed by sheep, to small-scale units of land devoted to cereal production [34,35]. The importance of land in the Pylian economy is evident from textual evidence, much of which is concerned with landholdings and crops, and some of the land near the Palace is measured in seed-grain [36,38]. Although the Mycenaeans were familiar with hydrological engineering, there was a strong dependence on winter and spring precipitation for crop yields, a dependence that carries through to modern times [39,40].

Throughout the LH IIIB period the increase in the size and number of sites in the area surrounding the Palace is a good indicator of population growth in Messinia (S1 File). Such a demographic increase likely added additional stresses to the agricultural system and forced the cultivation of marginal lands. Agricultural products sustained the majority of the population of the kingdom of Pylos, and a dynamic relationship existed between the agricultural and social systems [41]. During the height of the LH IIIB period, the economy was a finely tuned system that depended on the proper functioning of all its components: agriculture, redistribution and tax collection [41].

The richness of the archaeological and textual evidence from the Peloponnese has, up to now, not been matched by the quality of the paleoclimate information from the area [42]. This means that inferences about socio-environmental links rely on paleoclimate data from other areas of the eastern Mediterranean, which, considering local variability, is problematic [11,42]. Here, we present a high-resolution, stable isotope based climate record from stalagmite S1 that formed from 4687 to 1297 yrs BP, in three discrete growth periods. The record covers large parts of the Greek Bronze Age, during which the Peloponnese saw the development of interconnected and complex societies and the intensification of agriculture, although with strong regional variability [4345]. The record further covers the period extending from late Hellenistic times through the transition to the Byzantine period. The S1 proxy record sheds new light on the effects of climate on, the large scale social reorganization that occurred ~4200 yrs BP, the expansion in the area embraced by Mycenaean civilization ~3400 yrs BP, the subsequent destruction of the Mycenaean palaces ~3200 yrs BP, and the expansion in the number of rural settlements in the late Roman period. Here, however, we evaluate and investigate principally the chronological fit between variability in climate and the destruction of the Mycenaean Palace at Pylos and the subsequent sociopolitical change that took place at the end of the palatial period on the Greek mainland.

2. Methods and results

Candle-shaped stalagmite S1, which is 230 mm in length (Fig 1), was collected in Mavri Trypa (Black Hole) Cave, located in the central part of Schiza Island, ~4 km off the SW coast of the Peloponnese (N36.7360° E21.7596°) (Fig 2; S2 File and S3 File). Permission for visiting and sampling the cave was issued by the Ephorate of Palaeoanthropology and Speleology, Athens, Greece. The cave formed in bedded Paleocene-Eocene limestones at 70 m above sea level. Twenty-four subsamples, 50–100 mg each, for U-Th dating [46,47] were milled along growth layers (S3 File). For carbon and oxygen stable isotopic measurements, subsamples were milled along a profile parallel to the central growth axis at 0.3-2-mm intervals (S3 File). Results are reported relative to the Vienna Pee Dee Belemnite (VPDB) standard.

thumbnail
Fig 1. Image of stalagmite S1.

Visible are the two sides of the central slab. Image on left shows the location of the samples milled for U-Th dating together with the results. Image on right shows holes from conventional drilling for samples for stable isotope analysis, together with tracks from sub-millimeter micromilling (highlighted by green lines). Red lines in both images indicate the position of inferred depositional hiatuses.

https://doi.org/10.1371/journal.pone.0189447.g001

thumbnail
Fig 2. Location map.

The location of Mavri Trypa Cave in relation to the Mycenaean Palace of Nestor at Pylos and other paleoclimate records mentioned in the text. Map made with Natural Earth.

https://doi.org/10.1371/journal.pone.0189447.g002

The corrected U-Th ages for S1 are precise with an average uncertainty <±1%. Twenty of 24 samples fall in stratigraphical order within uncertainties. High [230Th/232Th] values in all 24 samples indicate that only minimal age corrections are needed [48] (Fig 3; S1 Table, S4 File).

thumbnail
Fig 3. Age-depth model for S1 with uncertainties.

Upper: Age uncertainties for each corrected U-Th age as percentages. Note logarithmic scale. Lower: Age-depth model for stalagmite S1 (green line) with corresponding 95% confidence limits (red lines). Black circles indicate individual U-Th ages, error bars show 2σ uncertainties. #1 to #3 denotes growth periods separated by hiatuses.

https://doi.org/10.1371/journal.pone.0189447.g003

S1 mainly consists of open columnar calcite. Micrite and new crystal nucleation indicate growth interruptions at 47.7 and 197.5 mm from the stalagmite top (S1 Fig, S4 File). Large chronological shifts in adjacent U-Th samples at these levels indicate that these interruptions are associated with extended growth hiatuses. Age-depth modeling using StalAge [49] (v. 1.0) suggests that stalagmite S1 formed between 4687±68 and 1297±103 yrs BP in three individual growth periods: from 4687±68 to 4182±33 yrs BP (#1), from 3813±370 to 2953±63 yrs BP (#2), and from 2067±27 to 1297±103 yrs BP (#3) (Fig 3; S4 File).

δ18O and δ13C range from -3.74 to -5.99% and from -6.15 to -11.07%, respectively (S2 Fig, S3 File, S1 Dataset). There is a positive correlation between δ18O and δ13C along the growth axis in all three growth periods: #1 r = 0.80 (n = 113), #2 r = 0.84 (n = 146), and #3 r = 0.66 (n = 96), all significant at the 95% level.

3. Interpretation and discussion

Speleothem δ18O can, under certain conditions, be a good proxy for hydro-climatic change. Drip water δ18O, which reflects the isotopic composition of infiltrating meteoric water, in combination with processes occurring during percolation, controls speleothem δ18O [50]. A number of processes that relate to the climate-system control precipitation δ18O, including precipitation amount and seasonality, moisture source and transport distance, as well as condensation temperature [51]. Further processes in the atmosphere, such as the proportions of low-intensity stratiform vs high-intensity convective precipitation, altitudinal transportation of raindrops within rainfall systems, and evaporation below the cloud base, have also been shown to affect δ18O in precipitation, highlighting the complex nature of δ18O values in meteoric waters [52,53].

Previous studies of δ18O in speleothems from the eastern Mediterranean demonstrate that the precipitation amount often is a main control when sea surface conditions remain close to constant [5466]. In the Peloponnese, precipitation δ18O values are more depleted during winter because of lower surface air temperature. However, in an annual cycle, there is a negative correlation between precipitation δ18O and rainfall amount, i.e. depletion increases with the precipitation amount, especially when the monthly average is below 100 mm [67]. A series of stable oxygen and hydrogen measurements on precipitation samples collected ~40 km north of Mavri Trypa Cave between January 2012 and September 2015 confirm the inverse relationship between precipitation amount and δ18O depletion [68]. A comparison between meteorological data and δ18O values from a modern stalagmite from Kapsia Cave in the central Peloponnese reveals a strong connection between δ18O depletion and an increase in the precipitation amount during the period of positive water excess (ONDJFMA) [69]. Considering the thickness of the bedrock above Mavri Trypa Cave, which is around 40 m, and that the catchment area for the karstic aquifer is relatively confined, given its location on an island, the climate signal from Mavri Trypa is not affected by aquifer mixing processes to a large extent. This could, however, also result in a relatively quick drying out of the aquifer under dry climate conditions. The two growth hiatuses and the inferred termination of deposition in Stalagmite S1 are likely a result of the aquifer drying out and they occur after longer periods of enriched δ18O and δ13C values.

Speleothem δ13C remains more complex to interpret than δ18O. Recent studies e.g. [70] have taken the interpretation beyond the influence of vegetation shifts from C3 to C4 [71,72] to a better understanding of the role of prior calcite precipitation (PCP) [73], the dead carbon component from the host rock, atmospheric CO2, the CO2 of the soil zone, and CO2 resulting from the decay of organic material found within the vadose zone [74] on speleothem δ13C. In the Mediterranean region and other semi-arid areas, depleted δ13C in speleothems has been linked to increased biological activity, including soil microbial activity, above the cave [54,56,7577]. Biological activity, in turn, is linked to precipitation in many areas of the Mediterranean, including our study area. Thus positive covariation between δ13C values and hydro-climate variability, as recorded by δ18O, is to be expected unless human activities above the cave, such as grazing or deforestation, have affected the vegetation [54,56,64,75,78].

The strong correlation between δ18O and δ13C values in the three growth periods indicates that Stalagmite S1 did not form in isotopic equilibrium with its parent drip water. Few stalagmites form under true equilibrium conditions [79], especially in semi-arid environments, and kinetic effects should be considered [77,80]. Kinetic fractionation is mainly driven by evaporation of drip water emerging in the cave due to low relative humidity (RH) in the cave air, and by the degassing of CO2 from the parent water due to differences in CO2 partial pressure (pCO2) between water and cave air. During drier climate conditions, δ18O will be enriched by increased evaporation in the epikarst and in the cave due to low RH, and δ13C will be enriched by increased degassing connected to reduced drip rates [81], enhancing any signal of drought in the stalagmite. Additionally, PCP may enrich δ13C values and since this factor typically is enhanced during drier climate conditions, it also acts to enhance the enrichment signature of δ13C during more arid phases. In short, a strong positive correlation between δ18O and δ13C is expected since both proxies respond to similar environmental drivers and similarly to kinetic effects. In addition, the closed nature of the cave and the high RH in the cave air (S2 File) should act to significantly reduce the effects of kinetic fractionation.

Based on the above discussion we suggest that the δ18O in the stalagmite from Mavri Trypa Cave should be interpreted as a signal for moisture, although there may be an influence of kinetic fractionation, with more negative δ18O values indicating wetter conditions and vice versa. The δ13C signal from Mavri Trypa may tentatively be used as a proxy for biological activity above the cave linked to moisture availability. However, in this paper we favor the use of δ18O over δ13C for making interpretations about past hydro-climatic variability, although we use the δ13C signal to better understand the δ18O.

3.1 Climate during the Bronze Age and from the Hellenistic to the Byzantine period

The Mavri Trypa δ18O record is supported by other stable isotope records from the Mediterranean region indicating a large-scale control on the isotopic signal [58,65,75,76,8285] (S3 Fig).

The δ18O and δ13C results from Mavri Trypa indicate relatively wet conditions from 4700 yrs BP to 4500 yrs BP, followed by a transitional period with large isotopic fluctuations from 4500 yrs BP to 4300 yrs BP leading towards drier conditions (Fig 4; S2 Fig). A diatom record from the nearby Agios Floros wetland also shows a development of aridity from 4500 yrs BP [86]. Stable arid conditions in Mavri Trypa occur from 4300 yrs BP until stalagmite S1 stops growing at 4200 yrs BP. There is widespread evidence from the eastern Mediterranean region in general for more arid conditions around 4200 yrs BP [42,58].

thumbnail
Fig 4. Stable oxygen isotopes as a proxy for precipitation showing wetter and drier periods through time.

A: Full sequence of stable oxygen isotopes from S1 plotted against time. Individual U-Th ages presented as black circles, error bars show 2σ uncertainties. B: Stable oxygen isotope results covering the Late Bronze Age (LBA) and Early Iron Age (EIA). The sub-periods of the LBA (Late Helladic (LH) I to LH IIIC) and the Protogeometric (PG) are shown together with the suggested period when the Palace of Nestor at Pylos was destroyed based on information from [27,29,31].

https://doi.org/10.1371/journal.pone.0189447.g004

Growth in stalagmite S1 resumes around 3800 yrs BP with δ18O values indicating wetter conditions (Fig 4). The resumption concurs with a period of wetter conditions suggested by other proxy data from the Peloponnese [45]. The generally wetter conditions indicated by our record last until 3150 yrs BP, although interrupted by two periods of drier conditions. The first drier phase develops rapidly at 3550 yrs BP and lasts until 3400 yrs BP. This phase is roughly divided in half by a brief return to wetter conditions at 3450 yrs BP. The second dry phase also develops rapidly and is centered around 3200 yrs BP (Fig 4) (see extended discussion below). The drier periods are also characterized by enriched δ13C values interpreted as indicating reduced biological activity above the cave, supporting the δ18O signal (S2 Fig). In Mavri Trypa a transition towards drier conditions starts ~3150 yrs BP, marking the gradual end of the wetter period. The transitional period is one of high-amplitude isotopic fluctuations on a decadal scale, superimposed on a centennial scale trend towards less negative δ18O values, indicating reduced precipitation. Overall more arid conditions are evident from 3100 yrs BP until 2950 yrs BP when growth in S1 terminates again. At this time, there is regional and local evidence from the eastern and central Mediterranean for drier conditions e.g. [42,45]. High-resolution stable isotope records from Jeita Cave [65], Renella Cave [58], Sofular Cave [75,76], and Nar Gölü [85] all show that aridity starts to develop around 3200 years BP but that the most arid conditions only occur after 3200 yrs BP (S3 Fig). From a chronological perspective, these records are some of the most robust paleoclimate records for the period around 3200 yrs BP (S4 Fig).

The third growth period in the Mavri Trypa stalagmite starts at 2050 yrs BP and coincides with the Roman Warm Period. The overall enrichment in the δ18O values that occurred during the first 200 years of this period likely results from these samples being milled towards the flank of the stalagmite in this section and is not related to climate (S2 Fig). Other paleoclimate records from the Peloponnese indicate wetter conditions from ~2550 yrs BP [45,64,86,87]. It is possible that the duration of the second hiatus in the Mavri Trypa stalagmite is not only a result of drier conditions, but also relates to the rerouting of water through the bedrock or human intervention in or near the cave. Following a brief interval of arid conditions just before 1900 yrs BP, a wetter period between 1850 and 1300 yrs BP, which peaks at 1675 yrs BP, is evident (Fig 4). After 1300 yrs BP there is a rapid change to drier conditions and the stalagmite stops growing 1297±27 yrs BP. Around this time many records from the eastern and central Mediterranean indicate drier conditions, and speleothem formation also ceased in Kapsia Cave in the central Peloponnese and in nearby Alepotrypa Cave, possibly indicating both local and regional aridity (S3 Fig) [54,64,65,75,76,84,88].

3.2 Climate, the destruction of the Mycenaean Palace of Nestor at Pylos and the end of the Late Bronze Age

The sampling resolution in the Mavri Trypa record for the period 3350–3000 yrs BP (i.e. much of the LBA) is on average 5 years and the uncertainty in the age-depth model is on average ±31.5 years, making it one of the most precise paleoclimate records for the LBA in the eastern Mediterranean (Fig 3, S4 Fig). Thus, the record approaches the criteria Knapp and Manning [20] argue are necessary in order to reliably compare archaeological events to climate data. Considering the proximity between Mavri Trypa and the Palace of Nestor, the paleoclimate information from this cave is pertinent for evaluating the potential impact of climate on the destruction of that Mycenaean palace and social processes during the subsequent periods. The archaeological chronology around 3200 yrs BP in the Peloponnese is generally well accepted, although some uncertainties remain. It is based on the cross-dating of ceramic material with “fixed points” in Egyptian chronology and contextually secure radiocarbon dates [20,8991]. We are thus able to rather precisely compare our paleoclimate data with the timing of the destruction of the Palace of Nestor and the end of the LBA in the Peloponnese.

The destruction of the Palace of Nestor is thought to have occurred ~3150–3130 yrs BP [27,29,31] (Fig 4). There is no unequivocal evidence for a prolonged period of drier conditions in the Mavri Trypa record at that time. Instead, it appears that the period around 3150 yrs BP is one of generally wetter, albeit fluctuating, climate conditions marking the beginning of a transitional period that leads into a distinct period of drier conditions that exists from ~3100 yrs BP (Fig 4). At the time of the destruction of the Palace there is a very short period of enriched δ18O values (mainly defined by one measurement point); this fluctuation, however, is not evident in the δ13C (S2 Fig). Considering the expected contribution of local noise to the δ18O signal and the lack of response in the δ13C, we suggest this enrichment cannot not be linked to drought, at least not of the magnitude seen around 3200 yrs BP and after 3100 yrs BP.

Before the Palace is destroyed, the Mavri Trypa record shows evidence for a drier period around 3200 yrs BP that lasted ~20 years. This dry period can be firmly placed in the LH IIIB period and, given the new data from Mavri Trypa, occurs, taking the age uncertainties in to account, two to eight decades before the palace is destroyed. This dry interval is followed by a return to wetter conditions, before the transitional period leading towards drier conditions that begins at 3150 yrs BP. It seems from the Mavri Trypa record that the dry phase around 3200 yrs BP was minor compared to what would come ~100 years later. The record from Mavri Trypa suggests that it was in the postpalatial period that arid conditions developed, and that it is only after the Bronze Age, in the Protogeometric period, that very arid conditions were established (Fig 4). The dry conditions recorded from 3100 yrs BP firmly belong to the LH IIIC period and likely contributed to the inability of the Mycenaean palaces in the Peloponnese to reassert their power.

Other paleoclimate records from the Peloponnese do not offer as much detail as the record from Mavri Trypa for the LH IIIB/LH IIIC transition [45], although a sedimentary record from the nearby Asea Valley indicates gradually cooling conditions from 3200 to 2700 yrs BP [87].

The stable isotope record from Mavri Trypa, together with other stable isotope records from the region, suggests that the end of the LBA (LH IIIC) is marked by increasing aridity but that it only reached its peak after ~3000 yrs BP.

Where does the new paleoclimate data from Mavri Trypa, with its current age-depth model and uncertainties, leave us in relation to the possible influence of climate on the destruction of the Mycenaean Palace of Nestor at Pylos and to the broader question of why it was not rebuilt in the LH IIIC period? The drought recorded at 3200 yrs BP clearly precedes the destruction of the Palace. Evidently the centralized administrative system controlled by the Palace could survive such a relatively short-term dry period and remain in control. Some fifty years later, however, when the pronounced period of aridity started to develop, the system would crumble. Further east, it has been proposed that triggers such as social unrest linked to drought induced food shortages could have been instrumental in LBA change [14,18] and there are few reasons to think that the mainly rainfed agricultural system of Messinia was less susceptible to those stresses. In this area, which naturally suffers from erratic rainfall, one of the principal ways in which the Palace central authority made themselves essential to small-scale producers was by providing food security in the case of crop failure or shortfalls during drought [34,39]. Although the record from Mavri Trypa does not offer any clear evidence for altered climate conditions that could have acted as a trigger for the destruction event, the period of drought around 3200 yrs BP could have contributed to the destabilization of the political and economic order. Increased aridity could have led to reduced agricultural output affecting the finely tuned economic system of a society that was close to, or already, over-extended, rattling the very foundations of the fragile palatial economy (S5 File). Although the palatial society at Pylos survived the short-term drought around 3200 yrs BP, it may have destabilized, or at least challenged, the system, which produced archaeologically and textually discernible responses by the Mycenaean elite. The suggestion of social turbulence and larger scale socioeconomic problems is hinted at both in the Linear B tablets and other evidence from the Palace; storage was increased and access to the Palace was restricted shortly before the destruction (S5 File) [36,92]. The new climate evidence from the Greek mainland, while not directly supporting a climate explanation for the destruction of the Palace, suggests that drier local conditions was one of several factors contributing to its demise. Rather than viewing the evidence of climate change as a cause of the collapse, we view it as part of the process of destabilization that contributed to the palatial administration's inability to reconstruct social hierarchies after the destruction. It has been suggested that the largely synchronous abandonment of the palatial centers across the Peloponnese at the end of the LH IIIB was, rather than an event, a process that took decades, and a short-term downturn in climate can be seen as one of many drivers [12].

The new data from Mavri Trypa also provides an opportunity to investigate the climate backdrop to the question of why the Mycenaean elite did not re-form and why the Palace was not rebuilt. Many signs of political and social collapse are visible in Pylos and Messinia as a whole after the destruction of the Palace. For instance, there was no urban reconstruction or subsequent cultural regeneration on the acropolis or in the adjacent lower town at Pylos and a pronounced depopulation of Messinia is evident from survey results [36]. The clear trend toward aridity from 3150 yrs BP probably meant a gradual increase in the number of years of drought, leading to failed crops or strongly reduced yields, and, more importantly, that agricultural productivity in normal and good years was reduced [93]. This in turn meant that it became increasingly difficult for farmers in Messinia to produce a ‘normal surplus’ that could be stored and taxed, either directly or socially, which has been argued to be an important mechanism behind the creation of social elites [39]. In an environment with developing aridity and reduced crop yields it would have become increasingly difficult to produce the ‘natural surplus’ that would enable a central authority to reassert itself by providing food relief, or trigger the formation of new social elites. There is, however, evidence for the continuation of small-scale subsistence farming in Messinia in LH IIIC and through the Early Iron Age (EIA), albeit at a much reduced scale [34]. The new evidence from Mavri Trypa makes it possible for the first time to situate the trajectory of events following the destruction of the Mycenaean Palace of Nestor at Pylos within a period of developing aridity throughout LH IIIC and the early part of the EIA.

4. Conclusions

The influence of climate on the Mycenaean world and the destruction of the Palace of Nestor at Pylos can, for the first time, be assessed through the investigation of a local high-resolution δ18O record that is the most precisely dated paleoclimate record from the eastern Mediterranean for the end of the LBA.

The Mavri Trypa δ18O record shows little or no unequivocal evidence for drier conditions when the palace in Pylos is destroyed ~3150–3130 yrs BP, at the transition from LH IIIB to LH IIIC. While the new paleoclimate evidence from the Greek mainland does not support a clear chronological synchronism between the destruction of the Mycenaean Palace at Pylos and drier conditions, as has been suggested previously, it does offer an insight into difficulties that existed several decades before the collapse of the palatial system.

There is evidence for a dry phase extending for approximately two decades around 3200±30 yrs BP, which can be firmly placed in the LH IIIB period, i.e. before the destruction of the palace. Evidently the centralized administrative system at Pylos managed to survive that period of drought, although evidence suggests that cracks were beginning to emerge in the period immediately preceding the palace’s destruction. This dry period was slight in comparison to what would come 100 years later, both in terms of magnitude and duration, but it nevertheless would have been felt in the agriculturally dominant palace economy. With the evidence in hand, the precise reasons for the destruction should be sought beyond climate explanations, although the effect of the climate should be considered as a contributing factor.

Following the destruction of the Mycenaean Palace at Pylos, there is strong evidence that climate conditions became progressively more arid during the LH IIIC period, and pronounced aridity is evident at the very end of the LBA and in the Protogeometric period, before the Mavri Trypa stalagmite stopped growing at 2953±63 yrs BP. We suggest this clear trend towards drier conditions caused reduced agricultural output, hampering the restoration of a central authority or the formation of new social elites. Small-scale subsistence agriculture however, persisted in the area.

For the first time, there are indications that climate may be one mechanism behind the process that led to the failure of the Mycenaean way of life in Pylos and there is strong evidence that developing aridity following the destruction of the Palace made it difficult for social elites to re-form and for the palatial system to be re-established. One cannot, however, attribute the collapse of the Mycenaean way of life to a single monolithic cause or event. Instead one should look for a suite of factors that contributed to the inability of the palatial elite to reconstitute the political, economic, and social organization that existed at the end of LH IIIB. Climate change is certainly a critical component in the equation.

Supporting information

S1 Fig. Plate of petrographic thin sections from stalagmite S1.

A-D show micrite and new crystal nucleation (red arrows) at 197.5 mm depth from the top indicating a growth interruption. A and B show the same slide in crossed polar light and plane-polarized light respectively. C and D show the same slide in crossed polar light and plane-polarized light respectively. E and F show new crystal nucleation at 47.7 mm depth from the top in crossed polar light and plane-polarized light respectively, indicating a growth interruption. At this depth, there is also an almost perpendicular change in the direction of the growth axis (see Fig 1). G shows a segment of the area between 55 and 145 mm depth from the top that has a more irregular and fibrous fabric toward the center compared to the flanks. Green arrows indicate direction of growth.

https://doi.org/10.1371/journal.pone.0189447.s001

(TIF)

S2 Fig. Stable oxygen (δ18O) and carbon (δ13C) isotopes from stalagmite S1 plotted vs. age.

Thicker black line represents 5-point moving average. Note inverted y-axes.

https://doi.org/10.1371/journal.pone.0189447.s002

(PDF)

S3 Fig. Selected high-resolution stable isotope records from eastern and central Mediterranean compared the δ18O record from Mavri Trypa (this study).

Bold black lines for each record represent a running average. The running average was selected for each individual record to filter the average resolution to ~30 years, the lowest resolution in any record in the figure, in order to enhance comparability between records. Wetter climate conditions are up and drier down. Records are organized from west to east. For the δ18O record from Soreq Cave no running average was calculated for the period of low resolution (i.e. between 3600 and 2000 yrs BP).

Similarly, colored bars indicate possible parallel periods of wetter and drier conditions in other records from central and eastern Mediterranean. Question marks indicate less certain matching with the record from Mavri Trypa.

https://doi.org/10.1371/journal.pone.0189447.s003

(EPS)

S4 Fig. Dating points and uncertainties from selected paleoclimate records from the central and eastern Mediterranean around 3200 yrs BP.

Renella Cave [58]; Shkodra Lake [84]; Mavri Trypa (this study); Sofular Cave [75,76], and Jeita Cave [65]. The δ18O record from Nar Gölü [85] is not represented in this graph because the chronology is based on the counting of annual lamina. There is only one U-Th age (3770±310 [83]) from Soreq Cave within the span of the figure. Records are organized from west to east.

https://doi.org/10.1371/journal.pone.0189447.s004

(EPS)

S5 Fig. Comparison of δ18O and δ13C results from parallel tracks in stalagmite S1.

Figure showing the similar isotopic signal in different but parallel tracks in stalagmite S1 indicating the stability of the signal. Slight offset on x-axis caused by imperfect matching between the results.

https://doi.org/10.1371/journal.pone.0189447.s005

(TIF)

S1 Table. U-Th dating chemistry and results table.

Uranium and thorium isotopic compositions and 230Th ages for stalagmite S1 by MC-ICPMS, Thermo Electron Neptune, at HISPEC, NTU.

https://doi.org/10.1371/journal.pone.0189447.s006

(DOCX)

S1 File. Archaeological background to the Palace of Nestor at Pylos and the area around Pylos.

https://doi.org/10.1371/journal.pone.0189447.s007

(DOCX)

S2 File. More detailed information about the setting of the cave where speleothem S1 was collected.

https://doi.org/10.1371/journal.pone.0189447.s008

(DOCX)

S3 File. More detailed information about the material and methods.

https://doi.org/10.1371/journal.pone.0189447.s009

(DOCX)

S4 File. More detailed information about the results of the analyses presented in this paper.

https://doi.org/10.1371/journal.pone.0189447.s010

(DOCX)

S5 File. Extended discussion: Potential impacts of climate variability on Mycenaean society.

https://doi.org/10.1371/journal.pone.0189447.s011

(DOCX)

S1 Dataset. Stable oxygen and carbon isotope data for stalagmite S1.

https://doi.org/10.1371/journal.pone.0189447.s012

(TXT)

Acknowledgments

This work was carried out within the framework of the Navarino Environmental Observatory (NEO), Messinia, Greece, a cooperation between Stockholm University, the Academy of Athens and TEMES S.A. NEO is dedicated to research and education on the climate and environment of the Mediterranean region. We thank the Ephorate of Palaeoanthropology and Speleology for granting permission to visit and take samples in Mavri Trypa Cave. We thank Erika Weiberg and two anonymous reviewers for helpful comments on an earlier version of the manuscript. Assistance during fieldwork and sampling was provided by Ilias Liakopoulos, Giorgos Maneas, Nikos Kalivitis, Herman Sundberg, Ingmar Borgström and Martina Hättestrand.

References

  1. 1. Staubwasser M, Sirocko F, Grootes PM, Segl M. Climate change at the 4.2 ka BP termination of the Indus valley civilization and Holocene south Asian monsoon variability: SOUTH ASIAN HOLOCENE CLIMATE CHANGE. Geophys Res Lett. 2003;30.
  2. 2. Hodell DA, Brenner M, Curtis JH. Terminal Classic drought in the northern Maya lowlands inferred from multiple sediment cores in Lake Chichancanab (Mexico). Quat Sci Rev. 2005;24: 1413–1427.
  3. 3. MacDonald G. Potential influence of the Pacific Ocean on the Indian summer monsoon and Harappan decline. Quat Int. 2011;229: 140–148.
  4. 4. Giosan L, Clift PD, Macklin MG, Fuller DQ, Constantinescu S, Durcan JA, et al. Fluvial landscapes of the Harappan civilization. Proc Natl Acad Sci. 2012;109: E1688–E1694. pmid:22645375
  5. 5. Kennett DJ, Breitenbach SFM, Aquino VV, Asmerom Y, Awe J, Baldini JUL, et al. Development and Disintegration of Maya Political Systems in Response to Climate Change. Science. 2012;338: 788–791. pmid:23139330
  6. 6. Douglas PMJ, Pagani M, Canuto MA, Brenner M, Hodell DA, Eglinton TI, et al. Drought, agricultural adaptation, and sociopolitical collapse in the Maya Lowlands. Proc Natl Acad Sci. 2015;112: 5607–5612. pmid:25902508
  7. 7. Weiss H, Courty M-A, Wetterstrom W, Guichard F, Senior L, Meadow R, et al. The Genesis and Collapse of Third Millennium North Mesopotamian Civilization. Science. 1993;261: 995–1004. pmid:17739617
  8. 8. Cullen HM, deMenocal PB, Hemming S, Hemming G, Brown FH, Guilderson T, et al. Climate change and the collapse of the Akkadian empire: Evidence from the deep sea. Geology. 2000;28: 379–382.
  9. 9. Kaniewski D, Campo EV, Guiot J, Burel SL, Otto T, Baeteman C. Environmental Roots of the Late Bronze Age Crisis. PLOS ONE. 2013;8: e71004. pmid:23967146
  10. 10. Kaniewski D, Guiot J, Van Campo E. Drought and societal collapse 3200 years ago in the Eastern Mediterranean: a review. Wiley Interdiscip Rev Clim Change. 2015;6: 369–382.
  11. 11. Roberts N, Eastwood WJ, Kuzucuoglu C, Fiorentino G, Caracuta V. Climatic, vegetation and cultural change in the eastern Mediterranean during the mid-Holocene environmental transition. The Holocene. 2011;21: 147–162.
  12. 12. Drake BL. The influence of climatic change on the Late Bronze Age Collapse and the Greek Dark Ages. J Archaeol Sci. 2012;39: 1862–1870.
  13. 13. Weiberg E, Finné M. Mind or matter? People-environment interactions and the demise of Early Helladic II society in the northeastern Peloponnese. Am J Archaeol. 2013;117: 1–31.
  14. 14. Langgut D, Finkelstein I, Litt T. Climate and the Late Bronze Collapse: New Evidence from the Southern Levant. Tel Aviv. 2013;40: 149–175.
  15. 15. Langgut D, Finkelstein I, Litt T, Harald Neumann F, Stein M. Vegetation and Climate Changes during the Bronze and Iron Ages (∼3600–600 BCE) in the Southern Levant Based on Palynological Records. Radiocarbon. 2015;57: 217–235.
  16. 16. Wiener MH. “Minding the Gap”: Gaps, Destructions, and Migrations in the Early Bronze Age Aegean. Causes and Consequences. Am J Archaeol. 2013;117: 581–592.
  17. 17. Cline EH. 1177 B.C.: the year civilization collapsed. Princeton: Princeton University Press; 2014.
  18. 18. Kaniewski D, Paulissen E, Van Campo E, Weiss H, Otto T, Bretschneider J, et al. Late second–early first millennium BC abrupt climate changes in coastal Syria and their possible significance for the history of the Eastern Mediterranean. Quat Res. 2010;74: 207–215.
  19. 19. Wiener M. The Interaction of Climate Change and Agency in the Collapse of Civilizations ca. 2300–2000 BC. Radiocarbon. 2014;56: S1–S16.
  20. 20. Knapp AB, Manning SW. Crisis in Context: The End of the Late Bronze Age in the Eastern Mediterranean. Am J Archaeol. 2016;120: 99.
  21. 21. Middleton GD. Nothing Lasts Forever: Environmental Discourses on the Collapse of Past Societies. J Archaeol Res. 2012;20: 257–307.
  22. 22. Nur A, Cline EH. Poseidon’s Horses: Plate Tectonics and Earthquake Storms in the Late Bronze Age Aegean and Eastern Mediterranean. J Archaeol Sci. 2000;27: 43–63.
  23. 23. Nur A, Burgess D. Apocalypse: earthquakes, archaeology, and the wrath of God. Princeton: Princeton University Press; 2008.
  24. 24. Akers PD, Brook GA, Railsback LB, Liang F, Iannone G, Webster JW, et al. An extended and higher-resolution record of climate and land use from stalagmite MC01 from Macal Chasm, Belize, revealing connections between major dry events, overall climate variability, and Maya sociopolitical changes. Palaeogeogr Palaeoclimatol Palaeoecol. 2016;459: 268–288.
  25. 25. Hoflmayer F, Manning SW, editors. Comments on Climate, Intra-regional Variations, Chronology, the 2200 B.C. Horizon of Change in the East Mediterranean Region, and Socio-political Change on Crete. The early/middle bronze age transition in the ancient near east: chronology, c14, and climate change. Chicago, IL: Oriental Institute of the University of Chicago; 2017. pp. 451–490.
  26. 26. Bintliff J. The Complete Archaeology of Greece: From Hunter-Gatherers to the 20th Century A.D. Chichester, UK: John Wiley & Sons; 2012.
  27. 27. Shelmerdine C. The Palatial Bronze Age of the Southern and central Greek Mainland. Addendum. In: Cullen T, editor. Aegean prehistory: a review. Boston: Archaeological Institute of America; 2001. pp. 329–381.
  28. 28. Harrison AB, Spencer N. After the Palace: The Early “History” of Messenia. In: Davis JL, editor. Sandy Pylos: An Archaeological History from Nestor to Navarino. 2nd ed. Austin: Texas University Press; 1998. pp. 147–162.
  29. 29. Deger-Jalkotzy S. Decline, destruction, and aftermath. In: Shelmerdine CW, editor. The Cambridge companion to the Aegean Bronze Age. New York: Cambridge University Press; 2008. pp. 387–415.
  30. 30. Maran J. The crisis years?: reflections on signs of instability in the last decades of the Mycenaean palaces. Sci DellAntichità. 2009;15: 241–262.
  31. 31. Davis JL. Pylos. In: Cline EH, editor. The Oxford handbook of the Bronze Age Aegean (ca 3000–1000 BC). Oxford: Oxford University Press; 2010. pp. 680–689.
  32. 32. Jazwa CS, Jazwa KA. Settlement ecology in Bronze Age Messenia. J Anthropol Archaeol. 2017;45: 157–169.
  33. 33. Maran J. Tiryns. In: Cline EH, editor. The Oxford handbook of the Bronze Age Aegean (ca 3000–1000 BC). Oxford: Oxford University Press; 2010. pp. 722–734.
  34. 34. Foxhall L. Bronze to iron: Agricultural systems and political structures in late Bronze Age and early Iron Age Greece. Annu Br Sch Athens. 1995;90: 239–250.
  35. 35. Shelmerdine CW. THE INDIVIDUAL AND THE STATE IN MYCENAEAN GREECE. Bull Inst Class Stud. 2011;54: 19–28.
  36. 36. Bennet J. The Aegean Bronze Age. In: Scheidel W, Morris I, Saller RP, editors. The Cambridge economic history of the Greco-Roman world. Cambridge, UK; New York: Cambridge University Press; 2007.
  37. 37. Killen J. The Linear B tablets and the Mycenaean economy. In: Davies AM, Duhoux Y, editors. Linear B, a 1984 survey: proceedings of the Mycenaean Colloquium of the VIIIth Congress of the International Federation of the Societies of Classical Studies (Dublin, 27 August-1st September 1984). Louvain-la-Neuve: Cabay; 1985. pp. 241–305.
  38. 38. Killen J. A view from the Tablets. In: Galaty ML, Parkinson WA, editors. Rethinking Mycenaean palaces II. Rev. and expanded 2nd ed. Los Angeles: Cotsen Institute of Archaeology, University of California; 2007. pp. 114–117.
  39. 39. Halstead P. The economy has a normal surplus: economic stability and social change among early farming communities of Thessaly, Greece. In: Halstead P, O’Shea , editors. Bad year economics: cultural responses to risk and uncertainty. Cambridge [England]; New York: Cambridge University Press; 1989.
  40. 40. Knauss J. Arkadian and Boiotian Orchomenos, centres of Mycenaean hydraulic engineering. Irrig Drain Syst. 1991;5: 363–381.
  41. 41. Halstead P. Toward a model of Mycenaean palatial mobilization. In: Galaty ML, Parkinson WA, editors. Rethinking Mycenaean palaces II. Rev. and expanded 2nd ed. Los Angeles: Cotsen Institute of Archaeology, University of California; 2007. pp. 66–73.
  42. 42. Finné M, Holmgren K, Sundqvist HS, Weiberg E, Lindblom M. Climate in the eastern Mediterranean, and adjacent regions, during the past 6000 years–A review. J Archaeol Sci. 2011;38: 3153–3173.
  43. 43. Cullen T. Aegean prehistory: a review. Boston: Archaeological Institute of America; 2001.
  44. 44. Cline EH, editor. The Oxford handbook of the Bronze Age Aegean (ca. 3000–1000 BC). Oxford; 2010.
  45. 45. Weiberg E, Unkel I, Kouli K, Holmgren K, Avramidis P, Bonnier A, et al. The socio-environmental history of the Peloponnese during the Holocene: Towards an integrated understanding of the past. Quat Sci Rev. 2016;136: 40–65.
  46. 46. Shen C-C, Wu C-C, Cheng H, Lawrence Edwards R, Hsieh Y-T, Gallet S, et al. High-precision and high-resolution carbonate 230Th dating by MC-ICP-MS with SEM protocols. Geochim Cosmochim Acta. 2012;99: 71–86.
  47. 47. Cheng H, Lawrence Edwards R, Shen C-C, Polyak VJ, Asmerom Y, Woodhead J, et al. Improvements in 230Th dating, 230Th and 234U half-life values, and U–Th isotopic measurements by multi-collector inductively coupled plasma mass spectrometry. Earth Planet Sci Lett. 2013;371–372: 82–91.
  48. 48. Hellstrom J. U–Th dating of speleothems with high initial 230Th using stratigraphical constraint. Quat Geochronol. 2006;1: 289–295.
  49. 49. Scholz D, Hoffmann DL. StalAge–An algorithm designed for construction of speleothem age models. Quat Geochronol. 2011;6: 369–382.
  50. 50. Lachniet MS. Climatic and environmental controls on speleothem oxygen-isotope values. Quat Sci Rev. 2009;28: 412–432.
  51. 51. Rozanski K, Araguás-Araguás L, Gonfiantini R. Isotopic Patterns in Modern Global Precipitation. In: Swart PK, Lohmann KC, Mckenzie J, Savin S, editors. Geophysical Monograph Series. Washington, D. C.: American Geophysical Union; 1993. pp. 1–36. Available: http://doi.wiley.com/10.1029/GM078p0001
  52. 52. Aggarwal PK, Romatschke U, Araguas-Araguas L, Belachew D, Longstaffe FJ, Berg P, et al. Proportions of convective and stratiform precipitation revealed in water isotope ratios. Nat Geosci. 2016;9: 624–629.
  53. 53. Narayana Rao T, Radhakrishna B, Srivastava R, Mohan Satyanarayana T, Narayana Rao D, Ramesh R. Inferring microphysical processes occurring in mesoscale convective systems from radar measurements and isotopic analysis. Geophys Res Lett. 2008;35.
  54. 54. Bar-Matthews M, Ayalon A, Gilmour M, Matthews A, Hawkesworth CJ. Sea–land oxygen isotopic relationships from planktonic foraminifera and speleothems in the Eastern Mediterranean region and their implication for paleorainfall during interglacial intervals. Geochim Cosmochim Acta. 2003;67: 3181–3199.
  55. 55. Drysdale R, Zanchetta G, Hellstrom J, Maas R, Fallick A, Pickett M, et al. Late Holocene drought responsible for the collapse of Old World civilizations is recorded in an Italian cave flowstone. Geology. 2006;34: 101.
  56. 56. Frisia S, Borsato A, Mangini A, Spötl C, Madonia G, Sauro U. Holocene Climate Variability in Sicily from a Discontinuous Stalagmite Record and the Mesolithic to Neolithic Transition. Quat Res. 2006;66: 388–400.
  57. 57. Zanchetta G, Drysdale RN, Hellstrom JC, Fallick AE, Isola I, Gagan MK, et al. Enhanced rainfall in the Western Mediterranean during deposition of sapropel S1: stalagmite evidence from Corchia cave (Central Italy). Quat Sci Rev. 2007;26: 279–286.
  58. 58. Zanchetta G, Regattieri E, Isola I, Drysdale RN, Bini M, Baneschi I, et al. The so-called “4.2 event” in the central mediterranean and its climatic teleconnections. Alp Mediterr Quat. 2016;29: 5–17.
  59. 59. Verheyden S, Nader FH, Cheng HJ, Edwards LR, Swennen R. Paleoclimate reconstruction in the Levant region from the geochemistry of a Holocene stalagmite from the Jeita cave, Lebanon. Quat Res. 2008;70: 368–381.
  60. 60. Orland IJ, Bar-Matthews M, Kita NT, Ayalon A, Matthews A, Valley JW. Climate deterioration in the Eastern Mediterranean as revealed by ion microprobe analysis of a speleothem that grew from 2.2 to 0.9 ka in Soreq Cave, Israel. Quat Res. 2009;71: 27–35.
  61. 61. Jex CN, Baker A, Eden JM, Eastwood WJ, Fairchild IJ, Leng MJ, et al. A 500 yr speleothem-derived reconstruction of late autumn-winter precipitation, northeast Turkey. Quat Res. 2011;75: 399–405.
  62. 62. Jex CN, Baker A, Fairchild IJ, Eastwood WJ, Leng MJ, Sloane HJ, et al. Calibration of speleothem δ18O with instrumental climate records from Turkey. Glob Planet Change. 2010;71: 207–217.
  63. 63. Rowe PJ, Mason JE, Andrews JE, Marca AD, Thomas L, van Calsteren P, et al. Speleothem isotopic evidence of winter rainfall variability in northeast Turkey between 77 and 6 ka. Quat Sci Rev. 2012;45: 60–72.
  64. 64. Finné M, Bar-Matthews M, Holmgren K, Sundqvist HS, Liakopoulos I, Zhang Q. Speleothem evidence for late Holocene climate variability and floods in Southern Greece. Quat Res. 2014;81: 213–227.
  65. 65. Cheng H, Sinha A, Verheyden S, Nader FH, Li XL, Zhang PZ, et al. The climate variability in northern Levant over the past 20,000 years: CLIMATE VARIABILITY IN NORTHERN LEVANT. Geophys Res Lett. 2015;42: 8641–8650.
  66. 66. Zanchetta G, Bar-Matthews M, Drysdale RN, Lionello P, Ayalon A, Hellstrom JC, et al. Coeval dry events in the central and eastern Mediterranean basin at 5.2 and 5.6ka recorded in Corchia (Italy) and Soreq caves (Israel) speleothems. Glob Planet Change. 2014;122: 130–139.
  67. 67. Argiriou AA, Lykoudis S. Isotopic composition of precipitation in Greece. J Hydrol. 2006;327: 486–495.
  68. 68. Bolin Centre Database [Internet]. [cited 17 Aug 2017]. Available: http://bolin.su.se/data/Norstrom-2017
  69. 69. Finné M. Climate in the eastern Mediterranean during the Holocene and beyond—a Peloponnesian perspective. Department of Physical Geography and Quaternary Geology, Stockholm University. 2014.
  70. 70. Bajo P, Borsato A, Drysdale R, Hua Q, Frisia S, Zanchetta G, et al. Stalagmite carbon isotopes and dead carbon proportion (DCP) in a near-closed-system situation: An interplay between sulphuric and carbonic acid dissolution. Geochim Cosmochim Acta. 2017;210: 208–227.
  71. 71. Dorale JA, Gonzalez LA, Reagan MK, Pickett DA, Murrell MT, Baker RG. A High-Resolution Record of Holocene Climate Change in Speleothem Calcite from Cold Water Cave, Northeast Iowa. Science. 1992;258: 1626–1630. pmid:17742531
  72. 72. Bar-Matthews M, Ayalon A, Kaufman A. Late Quaternary Paleoclimate in the Eastern Mediterranean Region from Stable Isotope Analysis of Speleothems at Soreq Cave, Israel. Quat Res. 1997;47: 155–168.
  73. 73. Fairchild IJ, Smith CL, Baker A, Fuller L, Spötl C, Mattey D, et al. Modification and preservation of environmental signals in speleothems. Earth-Sci Rev. 2006;75: 105–153.
  74. 74. Mattey DP, Atkinson TC, Barker JA, Fisher R, Latin J-P, Durrell R, et al. Carbon dioxide, ground air and carbon cycling in Gibraltar karst. Geochim Cosmochim Acta. 2016;184: 88–113.
  75. 75. Fleitmann D, Cheng H, Badertscher S, Edwards RL, Mudelsee M, Göktürk OM, et al. Timing and climatic impact of Greenland interstadials recorded in stalagmites from northern Turkey. Geophys Res Lett. 2009;36.
  76. 76. Göktürk OM, Fleitmann D, Badertscher S, Cheng H, Edwards RL, Leuenberger M, et al. Climate on the southern Black Sea coast during the Holocene: implications from the Sofular Cave record. Quat Sci Rev. 2011;30: 2433–2445.
  77. 77. Flohr P, Fleitmann D, Zorita E, Sadekov A, Cheng H, Bosomworth M, et al. Late Holocene droughts in the Fertile Crescent recorded in a speleothem from northern Iraq: Late Holocene Droughts. Geophys Res Lett. 2017;44: 1528–1536.
  78. 78. Frumkin A, Carmi I, Gopher A, Ford DC, Schwarcz HP, Tsuk T. A Holocene millennial-scale climatic cycle from a speleothem in Nahal Qanah Cave, Israel. The Holocene. 1999;9: 677–682.
  79. 79. Fairchild IJ, Baker A. Speleothem science: from process to past environments. Hoboken, N.J: Wiley; 2012.
  80. 80. Fleitmann D, Burns SJ, Neff U, Mudelsee M, Mangini A, Matter A. Palaeoclimatic interpretation of high-resolution oxygen isotope profiles derived from annually laminated speleothems from Southern Oman. Quat Sci Rev. 2004;23: 935–945.
  81. 81. McDermott F. Palaeo-climate reconstruction from stable isotope variations in speleothems: a review. Quat Sci Rev. 2004;23: 901–918.
  82. 82. Jones MD, Roberts CN, Leng MJ, Türkeş M. A high-resolution late Holocene lake isotope record from Turkey and links to North Atlantic and monsoon climate. Geology. 2006;34: 361–364.
  83. 83. Bar-Matthews M, Ayalon A. Mid-Holocene climate variations revealed by high-resolution speleothem records from Soreq Cave, Israel and their correlation with cultural changes. The Holocene. 2011;21: 163–171.
  84. 84. Zanchetta G, Van Welden A, Baneschi I, Drysdale R, Sadori L, Roberts N, et al. Multiproxy record for the last 4500 years from Lake Shkodra (Albania/Montenegro). J Quat Sci. 2012;27: 780–789.
  85. 85. Dean JR, Jones MD, Leng MJ, Noble SR, Metcalfe SE, Sloane HJ, et al. Eastern Mediterranean hydroclimate over the late glacial and Holocene, reconstructed from the sediments of Nar lake, central Turkey, using stable isotopes and carbonate mineralogy. Quat Sci Rev. 2015;124: 162–174.
  86. 86. Katrantsiotis C, Norström E, Holmgren K, Risberg J, Skelton A. High-resolution environmental reconstruction in SW Peloponnese, Greece, covering the last c. 6000 years: Evidence from Agios Floros fen, Messenian plain. The Holocene. 2016;26: 188–204.
  87. 87. Unkel I, Schimmelmann A, Shriner C, Forsén J, Heymann C, Brückner H. The environmental history of the last 6500 years in the Asea Valley (Peloponnese, Greece) and its linkage to the local archaeological record. Z Für Geomorphol Suppl Issues. 2014;58: 89–107.
  88. 88. Boyd M. Speleothems from warm climates: holocene records from the Caribbean and Mediterranean regions. Department of Physical Geography, Stockholm University. 2015.
  89. 89. Manning SW, Weninger B. A light in the dark: archaeological wiggle matching and the absolute chronology of the close of the Aegean Late Bronze Age. Antiquity. 1992;66: 636–663.
  90. 90. Manning SW. Chronology and terminology. In: Cline EH, editor. The Oxford handbook of the Bronze Age Aegean (ca 3000–1000 BC). Oxford; 2010. pp. 11–28.
  91. 91. Wiener MH. Oh, No–Not Another Chronology! In: Oppenheim A, Goelet O, editors. The Art and Culture of Ancient Egypt: Studies in Honor of Dorthea Arnold. New York: Egyptological Seminar of New York; 2015. pp. 649–663.
  92. 92. Wright J. Changes in form and function of the palace at Pylos. In: Shelmerdine CW, Palaima TG, editors. Pylos comes alive: industry and administration in a Mycenaean palace: papers of a symposium. New York: Archaeological Institute of America. Fordham University; 1984.
  93. 93. Roberts N. “Living with a Moving Target”: Long-term Climatic Variability and Environmental Risk in Dryland Regions. In: Miller NF, Moore KM, Ryan K, editors. Sustainable lifeways: cultural persistence in an ever-changing environment; [Symposium entitled “Forces of Nature: Risk and Resilience as Factors of Long-Term Cultural Change.” Philadelphia, Pa: Univ. of Pennsylvania Press; 2011.