Peloponnesian Stalagmites and Soda Straw Stalactites As Climate Archives

Institutionen för naturgeografi

Peloponnesian Stalagmites and Soda Straw Stalactites as Climate Archives

Stable Isotopes in New Speleothem Material from Kapsia Cave, Peloponnese, Greece

Linn Haking

Examensarbete grundnivå NG 63 Naturgeografi, 15 hp 2017

Förord

Denna uppsats utgör Linn Hakings examensarbete i Naturgeografi på grundnivå vid Institutionen för naturgeografi, Stockholms universitet. Examensarbetet omfattar 15 högskolepoäng (ca 10 veckors heltidsstudier).

Handledare har varit Martin Finné, Institutionen för arkeologi och antik historia, Uppsala universitet. Examinator för examensarbetet har varit Steffen Holzkämper, Institutionen för naturgeografi, Stockholms universitet.

Författaren är ensam ansvarig för uppsatsens innehåll.

Stockholm, den 4 januari 2018

Steffen Holzkämper Chefstudierektor

Abstract This study presents results from stable isotope analyses of a modern stalagmite and three soda straw stalactites from Kapsia Cave, the Peloponnese, Greece. The resulting values from the stalagmite are put into context of local meteorological data, as well as previous research from Kapsia Cave. The potential for using soda straw stalactites as complementary climate archives on shorter time scales on the Peloponnese is also explored. The isotopic values in the stalagmite confirm a strong link to the amount effect on an annual scale. On a seasonal scale, variations in the isotopic signal can be detected as a result of i.e. increased cave air temperature in summer. The stable isotope values in the soda straw stalactites largely correspond to previous isotopic measurements in Kapsia Cave. The trend of the isotopic carbon signal in two of the straws also strengthens earlier theories suggesting a link to CO2 concentrations in the external atmosphere. Soda straws are, thus, encouraged for use in future climate studies, although the sampling method should be further explored. The results of this study contribute to an increased understanding of Peloponnesian speleothems in relation to environmental processes and new insights are suggested into the use of soda straw stalactites as climate archives.

Keywords

Stable Isotopes; Stalagmite; Soda Straw Stalactites; Climate Archive; Climate Variability; The Amount Effect; Isotopic Fractionation; Kapsia Cave; Peloponnese; Greece; Mediterranean

Table of content

1. Introduction 1 2. Aim and research questions 2 3. Background 3 3.1 Speleothem formation 3 3.2 Speleothems as climate archives 4 3.2.1 Soda straw stalactites 5 3.2.2 Stalagmites 6 3.3 Speleothems and stable isotopes of carbon and oxygen 7 3.4 Isotopic fractionation and equilibrium conditions 9 3.5 Setting 10 3.5.1 Kapsia Cave 10 3.5.2 Local climate 12 3.5.3 Previous research in Kapsia Cave 13 4. Material and method 14 4.1 Fieldwork 14 4.2 Lab-work 15 5. Results 16

5.1 Stalagmite GK02MODERN 16 5.2 Soda straw stalactites KS2, KS3 and KS4 18 6. Discussion 19

6.1 Stalagmite GK02MODERN 19 6.2 Soda straw stalactites KS2, KS3 and KS4 22 7. Conclusion 25 Acknowledgements 26 8. References 26 8.1 Electronic references 31

1. Introduction Speleothems are calcareous cave deposits that can form and be preserved over long periods of time making them interesting as natural climate archives. They offer the possibility to study climate variability through several proxies, for example, oxygen- and carbon isotopes, trace elements (e.g. Mg, Sr, Ba), and petrographic changes in the speleothem laminea, which can be used in combination with precise radiometric dating with Uranium-Thorium. Proxy records from speleothems can be of different resolutions from sub-annual to millennial and sometimes cover extensive time series (up to 105 of years) (Fairchild et al. 2006). Speleothem studies mostly cover tropical- to sub-tropical regions of the world (e.g. Williams et al. 1999; Baker et al. 2007; Verheyden et al. 2008; Jo et al. 2010; Boyd 2015; Bergel et al. 2017; Scroxton et al. 2017) and are important complements as terrestrial archives from low latitudes to studies on e.g. ice cores and ocean sediments cores (Baker et al. 2014). Several speleothem climate studies have been conducted around the Mediterranean (e.g. Bar-Matthews et al. 1996; Bard et al. 2002; Bar-Matthews et al. 2003; Frisia et al. 2003; Baker et al. 2007; Verheyden et al. 2008; Fleitmann et al. 2009; Zanchetta et al. 2014; Surić et al. 2017). However, they are sparse on the Greek peninsula Peloponnese, where speleothem material was first published only a few years ago (Finné et al. 2014). Before 2014 this area constituted a gap in the speleothem climate research. Further work from Finné (Finné 2014; Finné et al. 2015) as well as Boyd (2015), has contributed with valuable data for the region, and ongoing work (e.g. Finné et al. in press) will not only continue to contribute to the overall climate history of the Peloponnese and the Mediterranean in general, but also to local archaeological research (e.g. Kordatzaki et al. 2016; Weiberg et al. 2016; Ntinou & Tsartsidou 2017). The archaeology of the Peloponnese is extensive and complex societies started to form as early as 8750BP (6800BC) at the beginning of the Neolithic revolution (Demoule & Perlès 1993). Current archaeological research of the Peloponnese is focusing on elucidating how past societies from 8750BP (6800BC) to 1650BP (AD300) have developed in relation to climate variability and aims to increase our possibilities to interpret human-environmental interactions (e.g. Weiberg et al. 2016; DoLP project). The overall objective of this study is to contribute to the speleothem climate record of the Peloponnese and, hence, increase the knowledge about climate variability in this region. In particular, new modern speleothem material consisting of a stalagmite

1 (forming within the last 5 years) from Kapsia Cave, central Peloponnese, is analysed for stable oxygen- and carbon isotopes in order to investigate how isotopic signals in speleothems correlate with the external climate of today. In combination with a short review on soda straw stalactite studies, three soda straws from Kapsia Cave are also analysed for stable isotopes to explore whether this kind of speleothem has the potential to be used for future climate studies on the Peloponnese. The results of this study increase and strengthen the knowledge and potential of the Peloponnesian speleothem climate record. It also brings some new insights into how soda straw stalactites may be used for future climate studies in general, and further contributes to our understanding of how meteorological data and isotopic signals in speleothems correlate in Kapsia Cave in particular.

2. Aim and research questions This study aims to contribute to the Peloponnesian speleothem climate record through 1) stable oxygen- and carbon isotope analysis of new speleothem material from Kapsia Cave, as well as by 2) exploring possibilities in using soda straw stalactites for Peloponnesian climate studies. The study further aims to relate the stable isotope results from the stalagmite to local meteorological data and examine the relationship to previous isotopic measurements, with the ambition to confirm and strengthen existing data and bring more insight into climate variability.

Research questions: - How do the isotopic values analysed in the speleothem material compare to previous isotopic measurements from Kapsia Cave? - In what ways does the isotopic signal in the stalagmite respond to external climate variability and kinetic fractionation in the cave? - What potential exists in using soda straw stalactites as complementary climate archives?

2 3. Background 3.1 Speleothem formation Speleothems are calcite deposits created from drip water in karstic caves, cavities formed through dissolution of carbonate rocks such as limestone (CaCO3) and dolomite

(CaMg(CO3)2). The process of speleothem formation (fig. 1) starts as surface water percolates through soil rich in carbon dioxide due to respiration of plants and decay of organic matter, creating acidic water (carbonic acid (H2CO3)). As the acidic water reaches the epikarst (a zone of fissured bedrock underlying the soil zone) it dissolves the carbonate bedrock, which in turn makes the calcium concentration rise in the solution. The calcium-saturated water continues through fractures in the porous rock and descends into the cave. As the carbon dioxide pressure (pCO2) is lower in the cave than in the soil- and epikarst zone, degassing of carbon dioxide occurs as the solution reaches the cave. This causes calcite deposition, initiating the formation of speleothems (Ford & Williams 2007; Fairchild & Baker 2012).

Figure 1. The process of speleothem formation (created by author after Fairchild et al. 2006).

3 Speleothems occur in several sizes and shapes (fig. 1), for example as flowstones, stalactites and stalagmites. Flowstones form as water flows down the cave walls and on the floor creating layers of calcite film, sometimes also hanging like curtains from the ceiling and wall. Stalactites start as soda straws (thin tubular speleothems) hanging from the ceiling, which with time grows in length and thickness, while stalagmites build up from the cave floor. Speleothems growing from the ceiling and the floor may eventually meet and form a column (Ford & Williams 2007).

3.2 Speleothems as climate archives Even though speleothems have been studied for over 40 years it is during the last 20 years that their use as climate archives has increased rapidly. This increase can mainly be traced to improvements of the radiometric Uranium-Thorium dating method (U-Th dating). By calculating the disequilibrium between the parent isotope 234U and its daughter isotope 230Th, this technique has the strength to provide precise dating to the speleothem laminae, up to 500,000 years back in time (Dorale et al. 2004; Fairchild & Baker 2012). Today speleothems are widely used as climate archives thanks to the potential of providing detailed time series using U-Th dating as well as counting of regularly forming laminae. In addition, speleothems have the capability to grow continuously for hundreds of thousands of years, thanks to the sheltered cave environment, and preserve a climate signal within their robust structure. As a terrestrial climate archive, speleothems as well as ice sheets provide climate data comparable to other archives like ocean sediment cores, covering extensive time periods. The possibility of precise independent dating, however, is what makes speleothems stand out as a climate archive (McDermott 2004; Fairchild et al. 2006; Fairchild & Baker 2012; Baker et al. 2014). Proxies that are commonly studied in speleothems are stable isotopes of carbon (δ13C) (e.g. Breecker 2016) and oxygen (δ18O) (e.g. McDermott 2004; Lachniet 2009), trace elements such as Magnesium (Mg), Strontium (Sr) and Barium (Ba) (e.g. Desmarchelier et al. 2006), as well as growth rates and thickness of laminae (e.g. Frisia et al. 2003; Tan et al. 2006). Though these proxies may be studied to provide an external climate signal, variations in the depositional environment, such as temperature, CO2 concentrations, humidity and air circulation in the cave, may affect this connection (see section 3.3 and 3.4 for further details) (Fairchild et al. 2006).

4 3.2.1 Soda straw stalactites Soda straw stalactites are the very first state in the formation of the more robust conical stalactites (St Pierre et al. 2009). The name soda straw originates from the tubular shape of this speleothem resembling a drinking straw (Desmarchelier et al. 2006). Like all speleothems, soda straw stalactites form from supersaturated drip water. As the water enters the cave and degassing of CO2 occurs, calcite precipitates from the drop, which is hanging from the cave ceiling. Due to gravity, the drop eventually falls, leaving a ring of calcite from where it Figure 2. Formation and structure of a banded soda straw stalactite with a suggested sample span for was hanging. As the straw in this fragments vs. powder (created by author after manner builds towards the cave floor, Paul et al. 2013, original published in International Journal of Speleology). the drip water will move in the centre of the forming tube, increasing in supersaturation, depositing a smaller amount of calcite in its inner canal and a greater amount at the tip as a ring. Because of this calcite deposition, soda straws many times demonstrate horizontal banding from 0.05 – 0.5mm thick, interpreted as annual laminae (fig. 2) (Moore 1962; Huang et al. 2001; Paul et al. 2013). The average diameter of soda straws normally measures around 5mm. The thickness of the tube walls is usually 100-300μm at the tip and increases towards the root, decreasing the width of its inner canal. Calcite growth becomes lateral when the central canal of the straw gets blocked and, thus, begins to form a conical stalactite (Baldini 2001; Fairchild & Baker 2012). Although it is not uncommon to find straws over a metre in length they are usually not longer than a few decimetres owing to their fragility (Desmarchelier et al. 2006; Fairchild & Baker 2012). It is commonly believed that soda straws form and remain intact for a relatively limited time period estimated to range at a few decades but possibly up to a century. However, growth at the same position may continue after

5 breakage (Williams et al. 1999; St Pierre et al. 2009; Fairchild & Baker 2012). Growth rates vary greatly, but can be as fast as 40mm per year (Jo et al. 2010). Because of the fragile nature of soda straws, they have many times been avoided for paleoclimate research (Desmarchelier et al. 2006) and, thus, the supply of such studies is limited. Still, many relevant areas are covered: dating (e.g. St Pierre et al 2009; St Pierre et al 2012), growth rate and structure (e.g. Moore 1962; Baldini 2001; Perrette & Jaillet 2010; Paul et al. 2013), trace elements (e.g. Huang et al. 2001; Desmarchelier et al. 2006), and stable isotopes (e.g. Baskaran & Krishnamurthy 1993; Williams et al. 1999; Woo et al. 2005). Studies on short-term climate variability and climate change over the past century are an important step in understanding the correlation between stable isotopes in speleothems and the external climate. This is a necessary complement to paleoclimate research and because analyses for both isotopes and dating today allow small sample size, soda straws stand a potential candidate for this kind of research (McDermott 2004; Woo et al. 2005; Fairchild et al. 2006; Paul et al. 2013).

3.2.2 Stalagmites Stalagmites are used to a great extent in paleoclimate research. Among the climate proxies normally studied in stalagmites, morphology and petrography can be a physical indication to climate variability (Baker et al. 2014). Laminae thickness and colour may, for example, provide an insight into changes in external temperature (Frisia et al. 2003). Changes in drip location and in drip rates, largely controlled by rainfall, can be seen in the growth geometry and porosities of the stalagmite (Genty & Quinif 1996; Frisia et al. 2003; McDermott 2004; Fairchild et al. 2006). Tan et al. (2006) also demonstrated the possibility of tracing long-term climate changes from the growth structure and thickness of stalagmite lamination. Although they emphasise that stalagmite laminae and its sensitivity to the external climate may differ greatly between individual specimens, the usage of stalagmites before other speleothems is still encouraged (Tan et al. 2006).

6 3.3 Speleothems and stable isotopes of carbon and oxygen Naturally occurring stable isotopes are variations of the same element, but the number of neutrons in the atom’s nuclei differs and consequently constitutes a different mass weight. This weight difference may cause physical separation of the different isotopes, so called isotopic fractionation, for example when elements are transferred between the reservoirs of the Earth system driven by solar energy. The relative abundance of heavy and light isotopes in each reservoir is dependent on the favoured transfer of one isotope over the other as a result of isotopic mass as well as temperature. This creates a ratio of heavy to light isotope that can be measured and is reported in relation to a set standard (v-SMOW: Vienna Standard Mean Ocean Water) (Sharp 2017). The ratio of stable oxygen isotopes in carbonates is defined by the formula:

18 18 16 δ O‰ V-PDB (Vienna Pee Dee Belemnite)=(Rsample/Rstandard -1)×1000 where R = O/ O. Similarly, the ratio of stable carbon isotopes in carbonates is expressed as:

13 13 12 δ C‰ V-PDB (Vienna Pee Dee Belemnite)=(Rsample / Rstandard -1)×1000 where R = C/ C (e.g. Rozanski et al. 1993). The most common stable isotopes that are analysed in speleothems for paleoclimatological purposes are oxygen and carbon. In nature, oxygen is normally found with 8 protons and 8 neutrons in its nuclei (16O – abundance: 99.76%), while the most common form of naturally occurring carbon contains 6 protons and 6 neutrons (12C – abundance: 98.89%). The heavier isotopes of these elements occur with one or two more neutrons forming: 17O (abundance: 0.04%), 18O (abundance: 0.2%), 13C (abundance: 1.11%) and 14C (radioactive). Among these, 18O and 13C are used for stable isotope analysis in carbonates (Sharp 2017). The isotopic signal in the cave drip water, captured within the growth laminae of speleothems, can under equilibrium conditions originate from meteoric water and, thus, climate-regulated processes on the surface control the 18O abundance in relation to the set standard. As a simplified example, a warmer climate with high solar energy increases ocean water evaporation allowing more of the heavier oxygen isotopes (18O) to get into the atmosphere. In turn, heavier isotopes are favoured during condensation, causing isotopic enrichment with increasing condensation temperature (the condensation effect). Consequently, meteoric precipitation will hold a high concentration of the heavier isotope 18O, eventually causing groundwater to be enriched in 18O and, thus, to display relatively enriched δ18O values. This relationship between

7 temperature at the ocean water source and 18O is known as the ocean source effect (Dansgaard 1964; Williams et al. 1999; Sharp 2017). Examples of further alterations of 18O may occur according to the amount effect, where δ18O of precipitation is inversely related to precipitation amount, or the seasonal effect, where δ18O of precipitation is more depleted in winter. The distance between precipitation and its source, the continental effect, also has an impact on δ18O, as the abundance of 18O will decrease further from the source (Rozanski et al. 1993; Williams et al. 1999). The amount of 13C that gets transported into the groundwater is partly controlled by CO2 concentrations in the atmosphere. In an article from Baskaran & Krishnamurthy (1993), soda straw stalactites, conical stalactites and stalagmites were analysed for stable isotopes to determine the influence of atmospheric CO2 in calcite cave deposits. They argued that soda straws dated to the last century recorded more enriched δ13C values closer to preindustrial times (an average around −5.9‰ (v-PDB) in AD1920) and more depleted values towards the tip of the straw deposited in recent times (an average around −8.2‰ (v-PDB) in AD1990), which was interpreted as a result of global industrial emissions of CO2. Williams et al. (1999) as well as Woo et al. (2005) confirm the correlation between more enriched δ13C values with distance from the tip in modern soda straws. Consequently, higher concentration of 13C and more enriched δ13C values in drip water are generated when CO2 concentrations are low in the atmosphere, for instance, on longer time scales during periods of colder climate (glacial conditions) when

13 vegetation is forced to absorb even the heavier CO2 molecules ( CO2). Other factors that

13 affect δ C in speleothems are photosynthesis in C3 versus C4 plants, microbiological processes and human land use (Woo et al. 2005; Ford & Williams 2007; Sharp 2017). Hence, the δ18O signal in cave drip water can be interpreted as an indicator for changes in the hydrological cycle and properties of precipitation, such as amount and

13 temperature, while the δ C signal is related to atmospheric CO2, vegetation cover and biological activity in the soil zone (Williams et al. 1999; Woo et al. 2005; Ford & Williams 2007; Sharp 2017).

8 3.4 Isotopic fractionation and equilibrium conditions Even though the oxygen and carbon isotope ratio in cave drip water originates from the external environment, this climate signal may be compromised if the speleothem does not deposit in, or near, equilibrium. Equilibrium refers to the state where HCO-3 and CO2 of the cave drip water remains unfractionated during calcite precipitation (Hendy 1971; Mickler et al. 2004; Sharp 2017). Since the work of Hendy (1971), the processes that impact isotopic fractionation have remained an ongoing debate. In his article, Hendy discusses possible kinetic effects on the drip water as it enters the cave, causing calcite to precipitate in disequilibrium with its parent water, possibly compromising any climate signal. Processes that can lead to enriched isotopic values compared to equilibrium are, for example, (1) rapid degassing of CO2 if pCO2 in the drip water solution is much higher than pCO2 in the cave atmosphere, enriching both δ18O and δ13C, as well as (2) evaporation of the drip water if the relative humidity of the cave is low, causing δ18O in the solution to become more enriched (Hendy 1971). Subsequent studies have further explored this concept (e.g. Bar-Matthews et al. 1996; Mickler et al. 2004; Dietzel et al. 2009; Lachniet 2009; Tremaine et al. 2011; Stoll et al. 2015). Mickler et al. (2004) examined kinetic fractionation and equilibrium in modern speleothems in relation to contemporary climatology and groundwater chemistry. Although most speleothems at the cave site were expected to deposit near equilibrium because of a stable depositional environment and high humidity, their results showed that especially δ18O values were far from what they predicted. They argued that several fractionating processes must have acted on the drip water since isotopic values were both enriched and depleted in different places in the cave. Mickler et al. (2004) concluded that enriched isotopic values were seen close to the cave entrance where stronger air ventilation lowered the CO2 concentrations and caused increased degassing of CO2 in the drip water, as well as increasing the evaporation due to a more varied humidity. Later on, the study of Tremaine et al. (2011) also showed that a linear enrichment in calcite δ13C could be detected with closer distance to the cave entrance. Mickler et al. (2004) further argued that rapid or large variability in rates of calcite precipitation could have a depleting effect on δ13C, but also enrich δ18O. Alteration of the δ18O may also occur according to the cave temperature effect, where fractionation of the drip water occurs as an effect of the cave air temperature by

9 −0.24‰ per °C increase in cooler temperatures (around 10°C), and −0.22‰ per °C increase in temperatures around 20°C (Friedman & O’Neil 1977; Williams et al. 1999). It has been suggested, however, that the cave temperature effect is largely counterbalanced by the condensation effect in the atmosphere causing enrichment of δ18O by 0.2-0.3‰ per °C. This is especially evident on longer time scales since a warming of the atmosphere also will lead to warming of the cave air over time (Williams et al. 1999; Finné et al. 2014). Williams et al. (1999) argued that the cave temperature effect is likely to become more dominant with increased distance from the moisture source. It is generally believed that paleoclimate records from speleothems should only be considered reliable if modern calcite precipitation occurs in or near equilibrium, which is to be determined by site-specific monitoring (e.g. Mickler et al. 2004; Dietzel et al. 2009; Lachniet 2009; Tremaine et al. 2011). Nevertheless, it is acknowledged that the many different fractionating processes rarely offer perfect equilibrium deposition of calcite, and that the existent of such a state is highly questionable. Although this complicates the possibility of interpreting an isotopic signal in speleothems, external climate signals may still be retrieved (Sharp 2017). If, however, kinetic fractionation occurs, complementary studies are needed to further investigate the processes behind the observed disequilibrium in order to obtain a reliable interpretation of the isotope signal (e.g. Lachniet 2009; McDermott et al. 2011).

3.5 Setting 3.5.1 Kapsia Cave

Kapsia Cave (N37.623°, E22.354°), central Peloponnese, Greece, (fig. 3) lies beneath 20- 30m of Triassic-Eocene Gavrovo-Tripolitza zone limestone bedrock and is stretching for about 200m. It is located on the Mantinea Plain at the base of the Mainalo Mountains. Above the cave, patches of soil, exposed bedrock and dead juniper trunks, show that the vegetation is sometimes subject to burning (last noted in AD1997). Further, the vegetation is comprised of 2m high oak shrubs, grass species, Lamiaceae and Euphorbia. Kapsia Cave has a natural entrance and an artificial entrance, constructed in 2004, both about 700m a.s.l. Since 2010 parts of the cave are open to tourists. The mean annual temperature measured over the last 3.5 years is 11.9°C ±0.52°C and mean annual

10 relative humidity is ≥95%. An unusually low relative humidity of 89% was measured during the winter of 2012 (Finné et al. 2014; Finné 2014). The many varied and impressive speleothems in Kapsia Cave (stalactites, stalagmites, flowstones, curtains etc.) are partly covered by a layer of clay, indicating a high-water mark from flooding events occurring both in the distant past (e.g. 500BC, 70BC, and AD450) and more recently (last noted in AD2001). Low-lying parts of the cave have meter-thick deposits of clay suggesting repeated flooding. This is also evident from a previous speleothem study in Kapsia Cave by Finné (2014), where a sliced stalagmite showed clay horizons in its laminae. These floodings have occurred as the drainage capacity of 5 sinkholes (one adjacent to the natural entrance of the cave) on the Mantinea Plain has been exceeded by the surface water input (Finné et al. 2014; Finné 2014). Though no systematic excavations have been made so far, traces from many archaeological periods have been identified in Kapsia Cave. Undated human remains from 50 individuals of all ages are scattered in the cave, as well as traces of human made fires, which date to the Neolithic (6500-3000BC). It is still discussed whether the cave was used as a burial ground or if these people were victims of a flooding event (Finné et al. 2014 and references therein). Further, artefacts from the Hellenistic period (323- 31BC) and the 4th-6th century AD have been found deep in Kapsia, among the things some bronze coins and fibulae from the 2nd half of the 6th century AD (Finné et al. 2014).

Figure 3. Location of Kapsia Cave on the Greek peninsula Peloponnese (after Finné 2014).

11 3.5.2 Local climate Meteorological data from AD2010 to 2017 (fig. 4) has been collected from a meteorological station in Tripoli (N37.509°, E22.418°, 10 km south of Kapsia, 646m elevation). Previous meteorological data, also displayed in figure 4, was presented in the study by Finné (2014), which likewise was collected from a station in Tripoli, although a different one. The meteorological data indicates a characteristic Mediterranean climate where summers are hot and dry and winters are mild and wet. The mean annual temperature between AD1980-2004 and AD2010-2016 combined is 13.7°C ±0.4°C with a slightly colder mean temperature in the most recent period. The mean total precipitation for the wet season (ONDJFMA) between AD1980 and 2017 is around 580mm. During the wet season, 70-80% of the yearly amount falls (Dotsika et al. 2010). The limited amount of rainfall in the dry season is commonly occurring as heavy rain showers, which is clear when going through the monthly meteorological reports during the last seven years. In earlier studies (e.g. Finné et al. 2014; Finné 2014), the contribution of dry season (MJJAS) precipitation to the karstic aquifer has been considered to be negligible since these rain showers cause fast runoff and no significant percolation of rainwater into deeper soils occur. Therefore, the cave drip water in Kapsia is assumed to primarily originate from wet season precipitation (Finné 2014); hence this study is based on the same premises.

Figure 4. Annual wet season precipitation data compiled from Finné (2014) and Tripoli Meteosearch.

12 3.5.3 Previous research in Kapsia Cave Even though speleothems from several countries around the Mediterranean have been analysed for paleoclimate research since many years back (e.g. Bar-Matthews et al. 1996; Bard et al. 2002; Bar-Matthews et al. 2003; Frisia et al. 2003; Baker et al. 2007; Verheyden et al. 2008; Fleitmann et al. 2009), no extensive studies had been conducted on the Peloponnese until 2009 when Kapsia Cave and Glyfada Cave first started to be monitored for a PhD project by Martin Finné (2014). By establishing the first stalagmite-based record for the Peloponnese, this project, in 2014, could present new data on changes in the local environment covering a period from 2900 to 1120BP (950BC to AD830), as well as regional climate changes. Periods with wetter climate than average were identified 2800, 2650, 2450, 2350-2050, 1790-1650, and 1180BP through stable oxygen isotope variations (δ18O), which contributed to the overall picture of the paleoclimate of the Mediterranean as well as to archaeological research in the region (e.g. Weiberg et al. 2016; DoLP project). Variations in stable carbon isotopes (δ13C) could further be associated with biological activity and local human land use (Finné et al. 2014; Finné 2014). Stable isotope analyses were also conducted for the laminated modern top of a stalagmite (GK02) from Kapsia Cave. Lamina counting suggests that the top formed from the 1980’s until collection in 2009. The top was analysed for stable isotopes at sub-annual resolution. These high-resolution δ18O values were binned into individual years and an annual average was calculated. These annual average δ18O values were then correlated to the total amount of wet season rainfall (Finné 2014). This data as well as monitoring of Kapsia Cave from September 2009 to March 2013 has revealed that drip water is supersaturated all year around and a strong seasonal variability in drip rate. There is also a relatively strong correlation between precipitation amount and variations in δ18O, i.e. the amount effect, when considering a lag of 1-2 years, i.e. the time it takes for water to pass through the aquifer. This is seen as depleted δ18O values when rainfall increases during wet season (Finné 2014). A correlation between δ13C and rainfall has not been detected in the modern laminae of GK02, nor is there a correlation in δ18O and δ13C to surface temperature (Finné 2014). Although the risk of kinetic fractionation has been noted to increase since the opening of the artificial entrance in 2004 (Finné 2014), in accordance with the study of Mickler et al. (2004) and Tremaine et al. (2011), calculations indicate that speleothems

13 in Kapsia Cave are likely to deposit near equilibrium (Finné et al. 2014; Finné 2014). Further climate studies for the Peloponnese and Kapsia Cave, using both past and recent time series, can help to provide more knowledge on the correlation of the external climate and isotopic signals captured within speleothems, as well as knowledge of kinetic effects. In this way, correlations so far observed can hopefully be confirmed and strengthened and, thus, provide a stronger paleoclimate record (Finné 2014).

4. Material and method 4.1 Fieldwork For this study, fieldwork was carried out during one day on the 25th of September 2017, in Kapsia Cave, central Peloponnese. The initial objective was to clean out previously installed monitoring equipment and to collect an actively growing stalagmite

(GK02MODERN, a continuation of stalagmite GK02 collected in 2009) that was depositing on a plastic film on top of a drip sensor placed in a bucket since March 2013 (fig. 5a & 5b). Unfortunately, the installation had fallen and, hence, the only present calcite had formed on the artificial base beneath the fallen sensor and bucket. The stalagmite was, nevertheless, collected for further analysis. However, the start date of its formation is unknown.

Figure 5. a) Sample site for stalagmite GK02MODERN and location of previously sampled soda straws (KS2, KS3 and KS4) in Kapsia Cave (after Finné et al. 2014). b) Installation for the formation of GK02MODERN (Finné 2014).

14 In addition, the positions of three previously collected soda straws were analysed (KS2, KS3 and KS4). They had been noted to actively drip at the tip and seemingly depositing calcite underneath upon collection. This could also be noted in the field, and so these three straws were selected for further analyses in this study. Point measurements of temperature (12.9°C) and relative humidity (96%) during the fieldwork falls within the range of earlier measurements from monitoring of the cave between 2009-2013.

4.2 Lab-work

Both the stalagmite (GK02MODERN) and the three soda straw stalactites (KS2, KS3 and KS4) were prepared for stable isotope analysis in the lab of the Physical Geography Department, Stockholm University.

Stalagmite GK02MODERN was cut in half, with a diamond-coated wire, across the centre where it was thickest. The height of the stalagmite was measured to roughly 1.5mm in cross-section. Three layers could be distinguished when looking at the specimen in microscope (fig. 6a). It is not clear whether these have been deposited annually.

Four samples were extracted from GK02MODERN by scraping with a handheld drill (Dremmel) with a diamond-coated drill bit (fig. 6a & 6b). Sample 1 was extracted from the uppermost half of layer 1, while sample 2 was taken from the lowest half of layer 1. Sample 3 represents layer 2 and sample 4 represents layer 3.

Figure 6. a) Samples in GK02MODERN indicated by numbers with approximate span (picture by author). b) GK02MODERN after sampling (picture by author).

15 The soda straw stalactites KS2, KS3 and KS4 were studied under a microscope in order to identify any banding. Since no unambiguous banding or structure could be distinguished, the straws were divided into 5mm intervals, starting from the tip, from which samples were extracted. The straws measured 45 to 60mm in length (KS2 ≈ 45mm, KS3 ≈ 60-70mm, and KS4 ≈ 55-60mm). Straw KS2 and KS3 were similar in structure, being more fragile in the first half from the tip and becoming more solid towards the root of the straw. KS4 was more fragile and the thickness of its walls did not increase with distance from the tip along the part removed from the cave, despite being about the same length as the other two straws. Samples were extracted from the 5mm intervals starting from the tip of the straws, excluding the root of KS2 and KS3. This was done by scraping the surface using a handheld drill (Dremmel) with a diamond-coated drill bit. In the more fragile parts (closer to the tip), fragments were instead collected and later ground to a powder with an agate mortar. Straw KS2 yielded 8 samples, KS3 yielded 10 samples and KS4 yielded 11 samples. However, sample KS4:7 did not yield enough material for the stable isotope analysis (yielded less than 0.2mg) and was consequently excluded. The material was sent to the stable isotope laboratory in Geo Zentrum Nordbayern at the University of Erlangen-Nuremberg, Germany, where a Gasbench II connected to a ThermoFisher Delta V Plus mass spectrometer was used. Reproducibility for δ13C and δ18O was ±0.04 and ±0.05 (1 standard deviation), respectively.

5. Results

5.1 Stalagmite GK02MODERN The stable oxygen isotope values in the four samples extracted from stalagmite

GK02MODERN range from −4.35‰ (v-PDB) to −4.82‰ (v-PDB), with increased depletion towards the top (fig. 7).

16 Figure 7. Stable oxygen isotope values from GK02MODERN with indication of sample spans in relation to depth.

The stable carbon isotope values in GK02MODERN, ranging from −9.13‰ (v-PDB) to −9.97‰ (v-PDB), shows a similar, but not the exact same pattern as δ18O, due to a slight enrichment from sample 4 to sample 3, and then moves towards depletion (fig. 8).

Figure 8. Stable carbon isotope values from GK02MODERN with indication of sample spans in relation to depth.

17 5.2 Soda straw stalactites KS2, KS3 and KS4 The three soda straw stalactites KS2, KS3 and KS4 yielded stable oxygen isotope values ranging from −4.03‰ (v-PDB) to −5.30‰ (v-PDB). A couple of sample points are within the measured uncertainty of one standard deviation (±0.05‰) of each other (fig. 9).

Figure 9. Stable oxygen isotope values from the soda straw stalactites KS2, KS3 and KS4, with indication of one standard deviation. Sample type displayed with the letters P (powder) and F (fragment). The missing sample KS4:7 is seen as a gap in KS4.

As for the stable carbon isotope values, KS2 and KS3 are close to parallel with a few sample points within or near the measured uncertainty of one standard deviation (±0.04‰). However, KS4 yielded more depleted stable carbon values than any sample point of the other straws and also has a smoother isotopic curve (fig. 10). Values range from −6.16‰ (v-PDB) to −11.13‰ (v-PDB).

Figure 10. (Next page) Stable carbon isotope values from the soda straw stalactites KS2, KS3 and KS4, with indication of one standard deviation (not visible on this scale). Sample type displayed with the letters P (powder) and F (fragment). The missing sample KS4:7 is seen as a gap in KS4.

6. Discussion

6.1 Stalagmite GK02MODERN

Stalagmite GK02MODERN that has been analysed in this study has been growing in the same place as a stalagmite (GK02) collected in Kapsia Cave in 2009 for paleoclimate research of the Peloponnese (Finné et al. 2014; Finné 2014). The maximum possible time of formation for GK02MODERN is 4.5 years (March 2013 to September 2017), but since the installation on which this stalagmite grew had tipped over causing the stalagmite to restart its growth, the actual time period is shorter, but unknown. The calculated growth rate in the fossil part (2900 to 1120BP) of stalagmite GK02 was 0.15mm/year (Finné et al. 2014), while the most recent laminae (AD1996 to 2008) indicate a growth rate of 0.9mm/year in average (calculated after Finné 2014).

Calculating the growth rate in GK02MODERN based on its thickness and the maximum time period of formation results in a growth rate of 0.33mm/year, which is closer to fossil rates rather than modern rates. If instead the recent growth rate of 0.9mm/year is applied to the stalagmite, considering the modern and active deposition of calcite, the time period of formation is calculated to 1 year and 8 months. Although there are some uncertainties concerning the growth rate, the modern rate has been considered to be a more likely estimate and is, thus, used as the age model in this study.

19 With an assumed growth rate of 0.9mm/year the start of deposition in

GK02MODERN is modelled to January 2016. The constructed age model suggests that the top sample consists of calcite deposited during the summer months of 2017 (June - September). The material in sample 2 has been deposited between February – May 2017, while sample 3 and 4 cover a longer time span; material in sample 3 was deposited July 2016 – January 2017, and sample 4 was deposited January 2016 – June 2016. Even though the isotopic values of these samples are well within the range of previous measurements, both fossil and recent, from Kapsia Cave (mean −4.9‰ (v- PDB) for oxygen and −9.3‰ (v-PDB) for carbon for the last 30 years) (fig. 11) (Finné 2014 for further details), the top sample clearly shows more depleted oxygen- and carbon values than the other samples in GK02MODERN. In previous research (Finné 2014) the amount effect has proven to be relatively strong (R2=0.30) on an annual scale (AD1989-2009). An approximate annual average of the sample values from this study, yielding δ18O −4.62‰ (v-PDB) and δ13C −9.65‰ (v- PDB) for 2017, and δ18O −4.36‰ (v-PDB) and δ13C −9.15‰ (v-PDB) for 2016, suggest a climate trend towards wetter conditions with richer vegetation and increased biological activity, i.e. towards depletion of δ18O and δ13C. The correlation between δ18O and precipitation amount does, however, slightly decrease (R2=0.28) when adding the values from this study to the earlier regression analysis (fig. 12). Even though the monitoring of Kapsia Cave has shown that the environment remains relatively stable throughout the year, an increase of the cave air temperature has been recorded in summer. Since the top sample in GK02MODERN is suggested as a summer depositional signal it could have been affected by the cave temperature effect, driving the oxygen values towards depletion with increased cave air temperature. The reason for the depleted carbon value is not as clear, but could possibly be related to higher CO2 concentrations in the cave in summer. In winter, isotopic values may instead become slightly enriched due to lower CO2 concentrations in the cave atmosphere

(Finné 2014). Hence, if the top sample in GK02MODERN also would include winter deposition of calcite like the other samples, the isotopic value would be less depleted. Since no extreme variations in the environment have been observed in Kapsia, these fractionating processes are probably acting on a smaller scale that cannot be observed when examining isotopic values at multi-annual resolution. Instead, the

20 amount effect is controlling the general trend. Although the annual average δ18O values calculated for GK02MODERN does not show a strong correlation to the wet season precipitation with the response time of 1-2 year that has been estimated for Kapsia Cave (Finné 2014), it is likely that the average isotopic value for 2017 will become less depleted, since the fall/winter (OND) deposition signal is not included. As the potential cave temperature effect is reduced, leading to more enriched δ18O values, the imprint of the amount effect on the isotopic signal will become more pronounced. Adding more data points to this regression analysis would be valuable in the future. In accordance with the study of Finné (2014), this study neither indicates that external temperature strongly correlates to δ18O and δ13C, nor that a strong correlation exists between δ13C and rainfall.

Figure 11. Stable isotope values from GK02 (AD1989-2009) (after Finné 2014) and GK02MODERN (after age model, AD2016-2017)

21 Figure 12. Correlation between precipitation amount and stable oxygen isotopes in GK02 (after Finné 2014) and GK02MODERN.

6.2 Soda straw stalactites KS2, KS3 and KS4 The three soda straws were of unknown age and had not been monitored in the cave before being collected at the same time. Even though the time period covered by soda straws normally do not exceed a century, approaching this issue by, for example, correlating the oxygen isotope signal to precipitation from the last hundred years to find a dating is not reliable in this case. First, the risk of kinetic fractionation may compromise the δ18O correlation to the amount effect. Second, even under near equilibrium deposition and an expected relationship with the amount effect, the growth rates are unknown and also have been shown to vary considerably in straws. Hence, the oxygen isotope curve may be responding to precipitation on a monthly, annual, decadal or centennial basis. With these uncertainties, the isotopic signal could possibly fit at any

22 place in the last couple of hundred years. Consequently, theories on dating of the soda straws KS2, KS3 and KS4 have been deliberately avoided in this study. It is still possible to discuss whether the three straws are coeval, regardless of the exact date. Although straw KS4 measured around the same length (~55-60mm) as KS2 and KS3 (~45-65mm), it clearly differed in appearance. KS2 and KS3 had fragile tips and increasing thickness of the tube walls towards the root. Despite that the very root of KS4 was never collected it exhibited a fragile structure along its whole length without increasing solidity and thickness of the tube walls. It could, for example, be argued that the whole collected length of KS4 deposited faster than KS2 and KS3 and is coeval with the tips of the other two straws. Still, there is not any additional evidence for this conclusion and even though KS4 was dripping, like the other two straws, when it was collected, it may not have been depositing at the time of KS2 and KS3 if the water was not supersaturated with calcite. The stable isotopes analysed may further reveal the relationship between the three straws. KS2 and KS3 seem to follow a general trend towards more depleted isotopic values towards the tips, more pronounced in carbon, which is not seen in KS4 (fig. 9 & 10). Despite that only a few sample points in KS2 and KS3 are within the measured uncertainty of one standard deviation (±0.05‰ for oxygen and ±0.04‰ for carbon), their common trend possibly indicates that KS2 and KS3 are coeval. This is further strengthened by the similar appearance in the two straws as well as the active dripping when they were collected. The δ18O values in all three straws are within the range of earlier measurements on stalagmites (Finné 2014 for further details). Likewise, this is true for the δ13C values in KS2 and KS3. Their general trend in δ13C also strengthen the theory presented by

Baskaran & Krishnamurthy (1993), indicating a correlation between increasing CO2 concentrations in the external atmosphere and depletion of carbon values in soda straws. Such analyses are encouraged for further studies on material from Kapsia Cave and the Peloponnese, as well as for soda straw studies in general. Unlike straw KS2 and KS3, KS4 shows δ13C values not only more depleted than those in the other two straws, but also more depleted values than in any of the previously presented speleothems from Kapsia Cave, both in modern- and fossil laminae (Finné 2014 for further details). The reasons for the significantly depleted values in straw KS4 could be many. Fractionating processes such as rapid or varied drip

23 rates could lead to δ13C depletion. With the opening of the artificial entrance in Kapsia Cave it is also probable that air ventilation became stronger, increasing degassing of the drip water as CO2 concentrations get lower, as well as altering the humidity in the cave and, thus, increasing evaporation. However, these processes have previously been associated with isotopic enrichment and cannot explain the depletion of carbon in KS4. Another possibility is that KS4 formed a few years after the latest wild fire above the cave (AD1997). At first, isotopic carbon values would become enriched due to the loss of vegetation. A few years later as vegetation is established again, the remains from the fire may have a fertilising effect increasing the biological activity and depleting isotopic carbon values. Still, this strong depletion in carbon has not been observed in any of the previously analysed speleothems covering this period. The reason for the depleted δ13C in KS4 is too vague to argue and will be left unanswered in this study. In the case of the soda straw stalactites it is important to highlight the way samples were extracted since a standard technique has not been clearly presented in previous studies and this kind of speleothem has not been used for climate studies on the Peloponnese. As for the fragile parts of the straws it was not possible to obtain samples by scraping off powder from the surface, but had to be collected as fragments, which were later ground to powder. These samples represent the whole thickness of the tube wall, unlike powder samples that only represent the outer surface. Since some calcite also deposits inside the central canal during formation, several years may be included in one sample (fig. 2). This could possibly result in a greater smoothing of the isotopic signal, if, for example, there is a mixing of winter and summer deposition like in

GK02MODERN. It is also possible that calcite deposited inside the canal yield different values due to e.g. a less exposed depositional environment than at the tip of the straw. A smoother isotopic curve, both for oxygen and carbon in all three straws, is present where samples were collected as fragments in contrast to powder (fig. 9 & 10). This could possibly explain some existing differences between KS2 and KS3, if they happen to be coeval, as well as the remarkably smooth carbon curve for KS4. Nevertheless, this phenomenon should be further examined in future studies to increase the possibility of interpreting the isotopic signal in soda straw stalactites. Dating of soda straws from Kapsia Cave and the Peloponnese is also encouraged in combination with stable isotope analyses and analyses of trace elements, to further explore the possibilities of using these speleothems as a high-resolution climate archive.

24 7. Conclusions This study contributes to an increased and strengthened understanding of how stable isotopes in Peloponnesian speleothems correlate to climate variability and their depositional cave environment. The study further elucidates how soda straw stalactites can be used in climate research, locally as well as in general. The following conclusions have been established on the basis of the research questions of this study: - The assumed annual averages as well as the measured individual sample values

in GK02MODERN are well within the range of previously measured isotopic values in Kapsia Cave.

- The new data points added from GK02MODERN support the relatively strong link between isotopic depletion and wet season precipitation (linked to the amount effect) that has been previously established for Kapsia Cave. It is likely that an average oxygen isotope value assumed for 2017 will become enriched, as parts of the wet season (OND) have not been included. This enrichment would lead to a stronger correlation between rainfall and δ18O in speleothems, possibly than what previously has been shown. - The top sample from the stalagmite should be regarded as a summer depositional signal, likely deposited during the summer of 2017. It is suggested that higher temperatures during summer cause this sample to be more depleted. This is not observable in the same way in the other samples, which are less depleted due to the mix of summer and winter deposition. Thus, on an annual scale calcite δ18O is strongly linked to the amount effect. - The results from the stable isotope analysis of the soda straw stalactites show potential in using this speleothem as a climate archive, as they yielded values that are within the range of previous measurements, with the exception of the δ13C in straw KS4. The δ13C trend in KS2 and KS3 is also in accordance with what has been concluded in earlier studies on soda straws, suggesting that the carbon

values correspond to external atmospheric CO2. - The sampling technique in soda straw stalactites may affect the isotopic values, where sampling fragments from straws in contrast to powder from the surface yields a smoothed isotopic signal. If this relationship is further explored, soda straws may provide a complementary high-resolution climate archive on a shorter time scale on the Peloponnese and in general.

25 Acknowledgement First of all, special thanks to my supervisor Martin Finné for letting me take part in his research and for interesting discussions, an inspiring field trip and supportive and dedicated supervision. I also thank Steffen Holzkämper, Head Director of Studies and lecturer at the Department of Physical Geography at Stockholm University, for providing me with the opportunity to work on this project with Martin as supervisor. Further thanks to my amazing friends Resa and Krister Davén for proof-reading my English. Many thanks to my other friends and family for encouragement and support, and last, but not least, to Alex who always manages to make me happy from any part of the world.

8. References

Baker, A., Asrat, A., Fairchild, I.J., Leng, M.J., Wynn, P.M., Bryant, C., Genty, D., and Umer, M. 2007. Analysis of the climate signal contained within d18O and growth rate parameters in two Ethiopian stalagmites. Geochimica et Cosmochimica Acta 71 (2007), 2975–2988, doi:10.1016/j.gca.2007.03.029.

Baker, A., Mattey, D.P., and Baldini, J.U.L. 2014. Reconstructing modern stalagmite growth from cave monitoring, local meteorology, and experimental measurements of dripwater films. Earth and Planetary Science Letters 392 (2014), 239-249, doi:10.1016/j.epsl.2014.02.036.

Baldini, J.U.L., 2001. Morphology And Dimensional Linkage between Recently Deposited Speleothems and Drip Water from Browns Folly Mine, Wiltshire, England. Journal of Cave and Karst Studies 63(3), 83-90.

Bar-Matthews, M., Ayalon, A., Matthews, A., Sass, E., and Halicz, L. 1996. Carbon and oxygen isotope study of the active water-carbonate system in a karstic Mediterranean cave: Implications for paleoclimate research in semiarid regions. Geochimica et Cosmochimica Acta 60 (2), 337-347.

Bar-Matthews, M., Ayalon, A., Gilmour, M., Matthews, A., and Hawkesworth, C.J. 2003. Sea–land oxygen isotopic relationships from planktonic foraminifera and speleothems in the Eastern Mediterranean region and their implication for paleorainfall during interglacial intervals. Geochimica et Cosmochimica Acta 67 (17), 3181–3199.

Bard, E., Antonioli, F., and Silenzi, S. 2002. Sea-Level during the Penultimate Interglacial Period Based on a Submerged Stalagmite from Argentarola Cave (Italy). Earth and Planetary Science Letters 196 (3-4), 135–146.

Baskaran, M., and Krishnamurthy, R.V. 1993. Speleothems as proxy for the carbon isotope composition of atmospheric CO2. Geophysical Research Letters 20 (24), 2905- 2908.

26 Bergel, S.J., Carlson, P.E., Larson, T.E., Wood, C.T., Johnson, K.R., Banner, J.L., and

Breecker, D.O. 2017. Constraining the subsoil carbon source to cave-air CO2 and speleothem calcite in central Texas. Geochimica et Cosmochimica Acta 217 (2017), 112- 127, doi:10.1016/j.gca.2017.08.017.

Boyd, M. 2015. Speleothems from Warm Climates: Holocene Records from the Caribbean and Mediterranean Regions. Doctoral dissertation, Stockholm University: Department of Physical Geography.

Breecker, D.O. 2016. Atmospheric pCO2 control on speleothem stable carbon isotope compositions. Earth and Planetary Science Letters 458 (2017), 58-68, doi:10.1016/j.epsl.2016.10.042.

Dansgaard, W. 1964. Stable isotopes in precipitation. Tellus XVI (4), 436-468.

Demoule, J-P., and Perlès, C. 1993. The Greek Neolithic: A New Review. Journal of World Prehistory 7 (4), 355-416.

Desmarchelier, J.M., Hellstrom, J.C., and McCulloch, M.T. 2006. Rapid trace element analysis of speleothems by ELA-ICP-MS. Chemical Geology 231 (1–2), 102-117, doi:10.1016/j.chemgeo.2006.01.002.

Dietzel, M., Tang, J., Leis, A., and Köhler, S.J. 2009. Oxygen isotopic fractionation during inorganic calcite precipitation ― Effects of temperature, precipitation rate and pH. Chemical Geology 268 (2009), 107–115, doi:10.1016/j.chemgeo.2009.07.015.

Dorale, J.A., Edwards, R.L., Alexander, E.C., Shen, C.-C., Richards, D.A., and Cheng, H. 2004. Uranium-series dating of speleothems: current techniques, limits and applications. In: Sasowsky, I.D., and Mylroie, J. (Eds.), Studies of Cave Sediments. Physical and Chemical Records of Palaeoclimate. Kluwer Academic, New York, pp. 177–197.

Dotsika, E., Lykoudis, S., and Poutoukis, D. 2010. Spatial distribution of the isotopic composition of precipitation and spring water in Greece. Global and Planetary Change 71 (2010), 141–149, doi:10.1016/j.gloplacha.2009.10.007.

Fairchild, I.J., Smith, C.L., Baker, A., Fuller, L., Spötl, C., Mattey, D., McDermott, F., and Edinburgh Ion Microprobe Facility, Institute of Earth Science, School of Geosciences, University of Edinburgh. 2006. Modification and preservation of environmental signals in speleothems. Earth-Science Reviews 75 (2006), 105– 153, doi:10.1016/j.earscirev.2005.08.003.

Fairchild, I.J., and Baker, A. 2012. Speleothem Science: From Process to Past Environments. Wiley-Blackwell, West Sussex, 432 p.

27 Finné, M. 2014. Climate in the eastern Mediterranean during the Holocene and beyond: A Peloponnesian perspective. Doctoral dissertation, Stockholm University: Department of Physical Geography and Quaternary Geology.

Finné, M., Bar-Matthews, M., Holmgren, K., Sundqvist, H.S., Liakopoulos, I., and Zhang, Q. 2014. Speleothem evidence for late Holocene climate variability and floods in Southern Greece. Quaternary Research 81 (2014), 213–227, doi:10.1016/j.yqres.2013.12.009.

Finné, M., Kylander, M., Boyd, M., Sundqvist H.S., and Löwemark, L. 2015. Can XRF scanning of speleothems be used as a non-destructive method to identify paleoflood events in caves? International Journal of Speleology 44 (1), 17-23, doi:10.5038/1827- 806X.44.1.2.

Finné, M., Holmgren, K., Shen, C.C., Hu, H.M., Boyd, M., and Stocker, S. in press. Late Bronze Age climate change and the destruction of the Mycenaean Palace of Nestor at Pylos. PLOS ONE.

Fleitmann, D., Cheng, H., Badertscher, S., Edwards, R.L., Mudelsee, M., G.ktürk, O.M., Fankhauser, A., Pickering, R., Raible, C.C., Matter, A., Kramers, J., and Tüysüz, O., 2009. Timing and climatic impact of Greenland interstadials recorded in stalagmites from northern Turkey. Geophysical Research Letters 36 (19), art. no. L19707, doi:10.1029/2009GL040050.

Ford, D., and Williams, P. 2007. Karst Hydrogeology and Geomorphology. John Wiley & Sons, Ltd, West Sussex, 554 p.

Friedman, I., and O'Neil, J.R. 1977. Compilation of stable isotope fractionation factors of geochemical interest. In: Fleischer, M. (Ed.), Data of geochemistry, 6th edition. Geological Survey, Washington, pp. KK1-K110.

Frisia, S., Borsato, A., Preto, N., and McDermott, F. 2003. Late Holocene annual growth in three Alpine stalagmites records the influence of solar activity and the North Atlantic Oscillation on winter climate. Earth and Planetary Science Letters 216 (3), 411-424, doi:10.1016/S0012-821X(03)00515-6.

Genty, D., and Quinif, Y. 1996. Annually laminated sequences in the internal structure of some Belgian stalagmites – Importance for paleoclimatology. Journal of Sedimentary Research 66, 275–288.

Hendy, C.H. 1971. The isotopic geochemistry of speleothems – I. The calculation of the effects of different modes of formation on the isotopic composition of speleothems and their applicability as paleoclimatic indicators. Geochimica et Cosmochimica Acta 35, 801–824.

28 Huang, Y., Fairchild, I.J., Borsato, A., Frisia, S., Cassidy, N.J., McDermott, F., and Hawkesworth, C.J. 2001. Seasonal variations in Sr, Mg and P in modern speleothems (Grotta di Ernesto, Italy). Chemical Geology 175 (2001), 429–448.

Jo, K.N., Woo, K.S., Hong, G.H., Kim, S.H., and Suk, B.C. 2010. Rainfall and hydrological controls on speleothem geochemistry during climatic events (droughts and typhoons): An example from Seopdong Cave, Republic of Korea. Earth and Planetary Science Letters 295 (2010), 441–450, doi:10.1016/j.epsl.2010.04.024.

Kordatzaki, Kiriatzi, E., Müller, N.S., Voyatzis, M., Romano, D., Petrakis, S., Forsén, J., Nordquist, G., Rodriguez-Alvarez, E., and Linn, S. 2016. A diachronic investigation of ‘local’ pottery production and supply at the sanctuary of Zeus, Mount Lykaion, Arcadia, Peloponnese. Journal of Archaeological Science: Reports 7 (2016), 526-529, doi:10.1016/j.jasrep.2015.12.017.

Lachniet, M.S. 2009. Climatic and environmental controls on speleothem oxygen-isotope values. Quaternary Science Reviews 28 (5–6), 412-432, doi:10.1016/j.quascirev.2008.10.021.

McDermott, F. 2004. Palaeo-climate reconstruction from stable isotope variations in speleothems: a review. Quaternary Science Reviews 23 (7–8), 901-918, doi:10.1016/j.quascirev.2003.06.021.

Mickler, P.J., Banner, J.L., Stern, L., Asmerom, Y., Edwards, R.L., and Ito, E. 2004. Stable isotope variations in modern tropical speleothems: evaluating equilibrium vs. kinetic isotope effects. Geochimica et Cosmochimica Acta 68, 4381–4393.

Moore, G.W. 1962. The Growth of Stalactites. Bulletin of the National Speleological Society 24 (2), 95-106.

Ntinou, M., and Tsartsidou, G. 2017. Domestic and ritual use of plants and fuels in the neolithic cave of Alepotrypa, southern Peloponnese, Greece: The wood charcoal and phytolith evidence. Quaternary International 457 (2017), 211-227, doi:10.1016/j.quaint.2016.11.028.

Paul, B., Drysdale, R., Green, H., Woodhead, J., Hellstrom, J., and Eberhard, R. 2013. A model for the formation of layered soda-straw stalactites. International Journal of Speleology 42 (2), 155-160, doi:10.5038/1827-806X.42.2.8.

Perrette, Y., and Jaillet, S. 2010. Spatial distribution of soda straws growth rates of the Coufin Cave (Vercors, France). International Journal of Speleology 39 (2), 61-70, doi:10.5038/1827-806X.39.2.2.

29 Rozanski, K., Araguás-Araguás, L., and Gonfiantini, R. 1993. Isotopic patterns in modern global precipitation. In: Swart, P.K. (Ed.), Climate Change in Continental Isotopic Records. Geophysical Monograph 78, AGU, Washington, pp. 1–36. doi:10.1029/GM078p0001.

Scroxton, N., Burns, S.J., McGee, D., Hardt, B., Godfrey, L.R., Ranivoharimanana, L., and Faina, P. 2017. Hemispherically in-phase precipitation variability over the last 1700 years in a Madagascar speleothem record. Quaternary Science Reviews 164 (2017), 25- 36, doi:10.1016/j.quascirev.2017.03.017.

Sharp, Z,D. 2017. Principles of Stable Isotope Geochemistry, 2nd edition, 384 p. doi:10.5072/FK2GB24S9F.

Stoll, H., Mendez-Vicente, A., Gonzalez-Lemos, S., Moreno, A., Cacho, I., Cheng, H., and Edwards, R.L. 2015. Interpretation of orbital scale variability in mid-latitude speleothem δ18O: Significance of growth rate controlled kinetic fractionation effects. Quaternary Science Reviews 127 (2015), 215-228, doi:10.1016/j.quascirev.2015.08.025.

St Pierre, E., Zhao, J.X., and Reed, E. 2009. Expanding the utility of Uranium-series dating of speleothems for archaeological and palaeontological applications. Journal of Archaeological Science 36 (2009), 1416–1423, doi:10.1016/j.jas.2009.02.004.

St Pierre, E., Zhao, J.X., Feng, Y.X., and Reed, E. 2012. U-series dating of soda straw stalactites from excavated deposits: method development and application to Blanche Cave, Naracoorte, South Australia. Journal of Archaeological Science 39 (2012), 922-930, doi:10.1016/j.jas.2011.10.027.

Surić, M., Lončarić, R., Bočić, N., Lončar, N., and Buzjak, N. 2017. Monitoring of selected caves as a prerequisite for the speleothem-based reconstruction of the Quaternary environment in Croatia. Quaternary International, in press (2017), 1-12, doi:10.1016/j.quaint.2017.06.042.

Tan, M., Baker, A., Genty, D., Smith, C., Esper, J., and Cai, B. 2006. Applications of stalagmite laminae to paleoclimate reconstructions: Comparison with dendrochronology/climatology. Quaternary Science Reviews 25 (2006), 2103–2117, doi:10.1016/j.quascirev.2006.01.034.

Tremaine, D.M., Froelich, P., and Wang, Y. 2011. Speleothem calcite farmed in situ: Modern calibration of δ18O and d13C paleoclimate proxies in a continuously-monitored natural cave system. Geochimica et Cosmochimica Acta 75 (2011), 4929–4950, doi:10.1016/j.gca.2011.06.005.

30 Verheyden, S. Nader, F.H., Cheng, H.J., Edwards, L.R., and Swenne, R. 2008. Paleoclimate Reconstruction in the Levant Region from the Geochemistry of a Holocene Stalagmite from the Jeita Cave, Lebanon. Quaternary Research 70 (2008), 368–381, doi:10.1016/j.yqres.2008.05.004.

Weiberg, E., Unkel, I., Kouli, K., Holmgren, K., Avramidis, P., Bonnier, A., Dibble, F., Finné, M., Izdebski, A., Katrantsiotis, C., Stocker, S.R., Andwinge, M., Baika, K., Boyd, M., and Heymann, C. 2016. The socio-environmental history of the Peloponnese during the Holocene: Towards an integrated understanding of the past. Quaternary Science Reviews 136, 40-65, doi:10.1016/j.quascirev.2015.10.042.

Williams, P.W., Marshall, A., Ford, D.C., and Jenskinson, A.V. 1999. Palaeoclimatic interpretation of stable isotope data from Holocene speleothems of the Waitomo district, North Island, New Zealand. The Holocene 9, 649-657.

Woo, K.S., Hong, G.H., Choi, D.W., Jo, K.N., Baskaran, M., and Lee, H.M. 2005. A reconnaissance on the use of the speleothems in Korean limestone caves to retrospective study on the regional climate change for the recent and geologic past. Geosciences Journal 9 (3), 243 – 247.

Zanchetta, G., Bar-Matthews, M., Drysdale, R.N., Lionello, P., Ayalon, A., Hellstrom, J.C., Isola, I., and Regattieri, E. 2014. 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. Global and Planetary Change 122 (2014), 130–139, doi:10.1016/j.gloplacha.2014.07.013.

8.1 Electronic references

DoLP (Domesticated Landscapes of the Peloponnese, Uppsala University), project plan: http://www.arkeologi.uu.se/Research/Projects/domesticated-landscapes-en/ (6th of December 2017)

Tripoli Meteosearch, meteorological data for Tripoli, Greece: http://meteosearch.meteo.gr/stationInfo.asp (6th of November 2017)