Maïté Van Rampelbergh

Faculteit WetenschappenAnalytical, Environmental & Geo-ChemistryAMGC

Speleothems as tools to reconstruct paleoclimates in tem-

perate (Belgium) and semi-arid (Socotra, Yemen) regions

during the Mid- to Late Holocene

Ph.D. dissertation presented to obtain the degree of Doctor of Science from the Vrije Uni-versiteit Brussel by

Maïté Van Rampelbergh

Promotor: Prof. Dr. Philippe ClaeysProf. Dr. Eddy Keppens

Co-Promotor: Dr. Sophie Verheyden

November 2014

i

Acknowledgements

This thesis is the result of four years of reading, learning, travelling, doing field work, meeting great people and working hard, which would not have been possible without the great team that surrounded me. First of all I want to thank my two amazing advisors, Prof. Dr. Eddy Keppens and Prof. Dr. Philippe Claeys, who, from the first day we met at the VUB, triggered my passion and huge interest in Geology and past climates. The scientist and Geologist I have become today is the result of their education filled with passion and patience. Thank you for believing in me and giving me the chance to do this PhD. You both put a lot of time and effort in this project and I really appreciated the way you supported me. Thank you also for feeding me with the occasional cheesecakes (Philippe) and home made cakes (Eddy), which softened the rougher scientific periods. I really enjoyed working with both of you and will certainly miss it. Second I would like to thank Dr. Sophie Verheyden, the co-‐advisor and brain behind this whole project. Sophie, without you I could not have brought this PhD to this level. Thank you for teaching me how to reconstruct climate from speleothems with such a huge passion and motivation. I cannot thank you enough for the many hours you have put in discussing my results and proofreading my papers. I really enjoyed working with you. I would like to say a big ‘thank you’ to Mr. Van Dierendonck, not only for the financial support, but also for his enormous interest in my work and his friendly visits at our department. Thank you to Michael Korntheuer, David Verstraeten, Claire Mourgues and Luc Deriemaker for the help and the measurements in the lab and for keeping the infamous “Kiel Device” running. You always made time for me when needed and were always ready to answer my questions. My thanks also go to the ‘Domaine des Grottes de Han’ and Guy Evrard for allowing me to core the Proserpine and letting me sample the drip site every two weeks. A special thank you goes to Etienne Lannoy, guide at the Han-‐sur-‐Lesse cave, who during the 4 years of my project, sampled rainwater and drip water every two weeks and always showed a big interest in my work. It was always a pleasure to come and pick up my samples at your house.

ii

Many thanks to my colleagues for the support, advice and the good times during the coffee breaks and BBQ’s: Dr. Steven Goderis, Dr. Virginie Renson, Kevin Debondt, David De Vleeschouwer, Joke Belza, Christina Makarona, Aurélie Sorel, Janos Kodolanyi, Claudio Ventura Bordenca, Matthew Hubert, Lidia Pittarello, Sean McKibbin, Harry Zekollari, Niels De Winter and many others. I will miss you guys! I would like to thank my parents for giving me the opportunity to study Geology and for always supporting me in my decisions. I could not have reached this far without your love. Also a big ‘thank you’ to my great family and friends for their encouragements and drinks when I needed them the most: Maud, Vincent, Ellen, Tim, Hanneke, Philippe, Yannick, Nick, Marc, Aline, Lana, Kim, Eyra, Wouter, Thomas, Bart, Daan, Arnaud, Monia, Wiet, Justine, Laurens and many others. Finally I want to thank Robbert, the man and recent husband of my life and dreams. This PhD is the results of your never-‐ending support and love. You always believed in me and filled me with the confidence needed to proceed. Thank you for sharing my joy during the happy moments but also for getting me through the harder times, I could not have done this without you. The way you supported me the last months really impressed me, I know I am safe with you and that you will keep making me laugh even in the hardest times. I love you.

iii

Table of contents Acknowledgements i Summary v Samenvatting vii Chapter 1 -‐ General Introduction 1 Chapter 2 -‐ Speleothems and Climate 11 2.1 Speleothems 11 2.2 Dating of speleothems 14

2.2.1 Dating techniques 15 2.2.2 Establishing an age-‐depth model 18

2.3 δ18O and δ13C in Speleothems 19 2.3.1 Factors that determine the δ18O of the drip water 19 2.3.2 Factors that determine the δ13C of the drip water 22 2.3.3 Equilibrium fractionation factors 24 2.3.4 Disequilibrium fractionation, new insights 26 2.3.5 Disequilibrium fractionation, effects and tests 30 2.3.6 Prior Calcite Precipitation (= PCP) 32 2.3.7 Summary δ18O and δ13C signals in speleothems 33

2.4 Mg and Sr in speleothems 36 2.4.1 The Mg and Sr distribution coefficients 36 2.4.2 The Mg/Ca and Sr/Ca ratio in the precipitating solution 39

2.5 Climate reconstructions from speleothems 42 2.5.1 Quantitative paleotemperature estimates 42 2.5.2 Semi-‐empirical climate relationships 44

2.5.3 Quantification of the isotope effects in the meteorological cycle 45

2.5.4 Speleothems as tools to reconstruct continental climates 46 Chapter 3 -‐ Socotran speleothems reveal monsoon changes during the Mid-‐ to Late-‐Holocene 65 Chapter 4 -‐ Monitoring of the Proserpine stalagmite 83 Chapter 5 -‐ A 500-‐year speleothem multiproxy record from

the Han-‐sur-‐Lesse cave, Belgium 103 Chapter 6 -‐ General Conclusions and Perspectives 131

v

Summary In the past 60 years, speleothems have successfully contributed to paleoclimate reconstruction and provided important insights in climate teleconnections. Although large-‐amplitude long-‐term global variations are well documented, better insights in the smaller-‐scaled and shorter-‐term variations such as during the Holocene and the last millennium are still needed. To allow such reconstructions, the behavior of proxy signals measured in speleothems needs to be understood in more details. In this PhD, speleothems from two contrasting climate regions, being temperate (Belgium) and semi-‐arid (Socotra, Yemen) systems, are studied for multiple proxies at high resolution going up to seasonal scales to investigate their potential as tool to reconstruct climate variations during respectively, the most recent 500 years and the Mid-‐ to Late Holocene. Speleothems from Socotra Island, located in the northern Indian Ocean, indicate detailed variations of the less-‐known winter subsystem of the Indian Ocean Monsoon over the last 6 000 years. Monitoring results of a fast growing (up to 2 mm/y) stalagmite (called ‘Proserpine’) in the Han-‐sur-‐Lesse cave in Belgium provide new insights on how δ18O, δ13C, growth rate and calcite fabric link to climate variations at a seasonal scale. Applying these findings on the Proserpine speleothem allows reconstructing climate variations during the last 500 years up to seasonal scale in terms of wetter and warmer or colder and dryer winters. Chapter 1 highlights the important position of speleothems within the broad climate science area and describes the need for more detailed high-‐resolution reconstructions to allow refining the climate models to more regional scales. Chapter 2 provides an elaborated state-‐of-‐the art of the processes and the theory behind speleothem based climate reconstructions. At the end of this chapter, the history of speleothem science with the milestone papers are discussed together with important recent methodological insights, which will further reinforce the position of speleothems within the field of climate science. Chapter 3 reports the first part of this PhD research that focuses on using speleothems to reconstruct monsoon variations in the northern Indian Ocean. Four speleothems from the eastern side of Socotra Island (Yemen) are studied for their δ18O, δ13C, Mg/Ca and Sr/Ca composition to investigate how the Indian Ocean Monsoon (IOM) evolved in that area over the last 6 000 years. The 4 records provide unique new insights in the northeast winter IOM subsystem, which is difficultly recorded in archives since it is often overwritten by the stronger summer IOM variations. The results show that the northeast winter IOM subsystem evolves differently than its southwest summer counterpart over the last 6 000 years with no links to the North Atlantic climate records. Chapter 4 and Chapter 5 report the second part of this PhD research where the Proserpine stalagmite from the Han-‐sur-‐

vi

Lesse cave is studied to investigate how speleothem proxies record climate variations at seasonal resolution in temperate climates such as Belgium. First, a one year (= 2013), biweekly cave monitoring of the Proserpine growth site is carried out in Chapter 4 to investigate how δ18O, δ13C, layer thickness and calcite fabric changes can be used to reconstruct the paleoclimate at seasonal scale. The most important conclusions of this work are that seasonal climate variations, i.e. seasonal variations in the climate parameters (in this thesis, effective precipitation and temperature), are quickly transferred into the cave and that they are successfully recorded in the speleothem calcite. Seasonal δ18O variations reflect temperature variations, whereas seasonal δ13C, layer thickness and growth rate variations reflect changes in effective precipitation. These acquired insights are used in Chapter 5 to interpret seasonally resolved δ18O, δ13C, layer thickness and calcite fabric changes in the Proserpine stalagmite over the last 500 years to investigate climate variations during the last part of the Little Ice Age (LIA, ± 1300-‐1850) and the anthropogenic period (most recent 150 years). Decadal and multi-‐decadal changes in the measured proxies reconstruct winter precipitation intensities (and temperatures) and indicate the occurrence of different climatic events that correspond with known European variations. Seasonal variations are well recorded in the stalagmite and provide new knowledge on seasonal temperature and effective precipitation over the studied period in Northern Continental Europe. This work illustrates that, provided a good knowledge of the cave system combined with a multiproxy approach; speleothems from temperate climates successfully record lower-‐amplitude and seasonal variations.

vii

Samenvatting Tijdens de jongste 60 jaar hebben speleothemen bewezen succesvol te kunnen worden ingezet bij het bestuderen van klimaatveranderingen uit het verleden en hebben ze vooral bijgedragen tot het verwerven van nieuwe inzichten in de teleconnecties tussen verschillende klimaatsystemen. Hoewel de klimaatgemeenschap reeds een grondige kennis bezit op het vlak van globale, sterk uitgesproken klimaatveranderingen, is er nog steeds nood aan het beter begrijpen van de minder uitgesproken veranderingen op regionale schaal. Het bestuderen van dergelijke ‘zwakkere’ klimaatvariaties op een hoge resolutie kan gebeuren aan de hand van speleothemen maar vereist een grondige kennis van de relatie tussen de veranderingen in het klimaat en de veranderingen in de meetbare proxies. In deze doctoraatsthesis wordt, door het meten van meerdere proxies op een hoge tijdsresolutie (tot seizoenaal) in speleothemen van een gematigde (België) en een semi-‐aride (Socotra, Jemen) klimaatregio, het potentieel van deze archieven geëvalueerd voor het reconstrueren van klimaatveranderingen tijdens respectievelijk de jongste 500 jaar en het midden-‐ tot laat-‐Holoceen. De resultaten van de speleothemen van het eiland Socotra, gelegen in het noorden van de Indische Oceaan, geven gedetailleerde variaties weer van het minder gekende winter-‐subsysteem van de Indische Oceaan Moesson (IOM) tijdens de afgelopen 6 000 jaar. De resultaten van de monitoring van een snel groeiende (tot 2 mm/jaar) stalagmiet (de ‘Proserpine’), in de grot van Han-‐sur-‐Lesse in België, bieden nieuwe inzichten over hoe δ18O en δ13C-‐signalen, groeisnelheid en calcietstructuur relateren met klimaatveranderingen tot op seizoenale schaal in gematigde klimaatregio’s. Het toepassen van deze resultaten op de proxy-‐reeksen gemeten in de Proserpine, over de jongste 500 jaar, laat toe klimaatveranderingen tot op seizoenale schaal te reconstrueren in termen van veranderingen in winterneerslag en –temperatuur. Hoofdstuk 1 schetst de belangrijke positie van klimaatreconstructies aan de hand van speleothemen binnen het bredere domein van de klimaatwetenschappen en beschrijft de behoefte aan meer gedetailleerde hoge-‐resolutie klimaatreconstructies die het verfijnen van klimaatmodellen naar regionale schaal mogelijk maken. Hoofdstuk 2 geeft een uitgebreide ‘state-‐of-‐the-‐art’ van de theorie en van de belangrijke processen die spelen bij het gebruik van speleothemen in het bestuderen en reconstrueren van het klimaat. Het laatste deel van dit hoofdstuk beschrijft de geschiedenis van de speleotheemwetenschap, de mijlpaalstudies en de recente methodologische ontwikkelingen die mits verdere ontwikkeling zullen bijdragen tot betere en meer gedetailleerde klimaatreconstructies aan de hand van speleothemen.

viii

Hoofdstuk 3 beschrijft het eerste luik van dit doctoraatsonderzoek dat zich richt op het gebruik van speleothemen om variaties van de IOM te reconstrueren in het noordelijk deel van de Indische Oceaan. Vier speleothemen, bemonsterd op de oostelijke kant van het eiland Socotra (Jemen) worden onderzocht naar hun δ18O, δ13C, Mg/Ca en Sr/Ca-‐samenstelling om na te gaan hoe de IOM, tijdens de afgelopen 6 000 jaar, is geëvolueerd in dit gebied. De 4 stalagmieten bieden unieke nieuwe inzichten in het noordoostelijk winter IOM subsysteem, dat moeilijk te bestuderen is omdat het in klimaatarchieven vaak wordt overschreven door de sterkere zomer IOM variaties. De resultaten tonen aan dat, over de jongste 6 000 jaar, het noordoostelijk winter IOM subsysteem op een andere manier evolueerde dan zijn zuidwesten zomer IOM tegenhanger. Bovendien vertoont het noordoosten winter IOM subsysteem geen link met het Noord-‐Atlantisch klimaat. Hoofdstuk 4 en 5 omvatten het tweede luik van dit doctoraatsonderzoek waarin de Proserpine stalagmiet uit de grot van Han-‐sur-‐Lesse wordt bestudeerd om te achterhalen hoe speleotheem-‐proxies seizoenale klimaatveranderingen, i.e. seizoenale veranderingen van klimaatparameters (= in deze studie effectieve neerslag en temperatuur), opnemen in gematigde klimaten zoals België. Hoofdstuk 4 beschrijft de eerste fase van dit luik waarin een tweewekelijkse monitoring van de Proserpine-‐groeisite gedurende een periode van één jaar wordt uitgevoerd om te onderzoeken hoe variaties in δ18O, δ13C, laagdikte en calcietstructuur veranderen in functie van de seizoenen. Resultaten tonen dat seizoenale klimaatveranderingen snel worden overgebracht naar de grot en dat ze op betrouwbare wijze worden opgenomen in het speleotheemcalciet. Seizoenale δ18O variaties weerspiegelen temperatuurschommelingen, terwijl seizoenale δ13C, laagdikte-‐ en groeisnelheidvariaties, veranderingen in effectieve neerslag weerspiegelen. In Hoofdstuk 5 worden deze verworven inzichten gebruikt om veranderingen in δ18O, δ13C, laagdikte en calcietstructuur, gemeten op seizoenale schaal, in de Proserpine over de afgelopen 500 jaar te interpreteren en zo een beter inzicht te krijgen in seizoenale klimaatveranderingen tijdens het laatste deel van de Kleine IJstijd (± 1300-‐ 1850) en de antropogene periode (laatste 150 jaar). Tienjarige en honderdjarige variaties in de gemeten proxies weerspiegelen veranderingen in neerslagintensiteit (en temperatuur) tijdens de winters en stemmen overeen met gekende variaties in het Europese klimaat. Seizoenale klimaatveranderingen in de stalagmiet verschaffen nieuwe inzichten in seizoenale variaties in temperatuur en effectieve neerslag tijdens de bestudeerde periode. Het werk uit luik 2 van dit doctoraat illustreert dat een goede kennis van het grotsysteem, in combinatie met een multiproxy-‐aanpak, het mogelijk maakt om speleothemen uit een gematigd klimaat succesvol te gebruiken bij de reconstructie van lagere amplitude en seizoenale klimaatvariaties.

Chapter 1: General Introduction

1

Chapter 1

General Introduction “The bad weather of last month is the consequence of climate change” a statement often heard on television, read in newspapers or made by friends and family. Are these changes really due to climate change? Is the extremely warm summer of 2003 in Europe the consequence of climate change or is it a normal consequence of the combination of different climate factors such as the North Atlantic Oscillation (NAO), the El Niño Southern Oscillation (ENSO), solar insolation, etc. Will the ongoing climate change cause a general warming in Europe, or will it lead to more extreme seasons with warmer summers and colder winters? Or will winters become warmer and wetter? To answer such questions, climate studies are carried out to better understand the factors that influence climate variations for different climate regions. The results from such studies are used to establish and refine climate models, which are used to simulate past but also future climate parameters such as temperature or precipitation intensity. Figure 1 illustrates the most recent temperature reconstruction based on a combination of such climate models and is published in the IPCC’s Fifth Assessment Report (AR5) (IPCC, 2013). The results clearly indicate the future global temperature increase that is related to the increase in greenhouse gases in the atmosphere, a process already suggested by Arrhenius in 1896 (Arrhenius, 1896). To optimize and refine such climate models, more detailed climate studies from the continents are necessary. Polar and ocean regions are well-‐covered through the ice core (Dansgaard et al., 1993; Thompson et al., 2013) and oceanic sediment core records (Heinrich, 1988; Shacklet.N, 1968; Shackleton and Vincent, 1978; Zachos et al., 2001). Lake records can be used to reconstruct continental climate evolutions, but they do not provide the high-‐resolution and good dating needed to refine the climate models to the yearly or seasonal scale (except for the varved lake sediments that allow yearly resolved time series). Other archives such as tree-‐rings and corals can deliver the needed high-‐resolution climate data but they are often seasonally biased.


2

Figure 1. Combined results of multiple climate models for the period between 1950 and 2100 for the change in global annual mean surface temperature relative to 1986–2005 (adapted after IPCC, 2013). The results are given for different scenarios or RPC’s used during the modeling. Speleothems such as stalagmites have the huge potential to provide the missing high-‐resolution paleoclimate continental information needed to refine the climate models to regional scales. Caves containing speleothems are found in almost every continent (Fig. 2) delivering a good coverage of the continental area to reconstruct the climate. Through the well-‐established U-‐series dating technique (Cheng et al., 2000; Edwards et al., 1987), speleothems can be very precisely be dated back to 600 kyr BP. In contrast to 14C-‐dating, U/Th-‐ages are absolute and do not need to be calibrated and consequently deliver ages with small uncertainty intervals. Furthermore, more recent speleothems that display seasonal or annual laminations can provide age models based on layer counting that delivers exact timing of the measured events (Fairchild et al., 2001; Mattey et al., 2008; Treble et al., 2003; 2005; Van Rampelbergh et al., in review). Precisely dated speleothems have already played key roles in dating large climatic transitions such as the Dansgaard-‐Oescher cycles (Genty et al., 2003), short-‐lived abrupt millennial scale events such as the 8.2 kyr cold event (Cheng et al., 2009b), glacial inceptions and terminations (Fig. 3) (Cheng et al., 2009a; Fleitmann et al., 2009; Yuan et al., 2004), important geological events such as the formation of the Grand Canyon (Polyak et al., 2008) or delivered long, continuous and well-‐dated records such as the Devils Hole Cave, Nevada, record covering the last 560 kyr (Winograd and Ludwig, 1996).


3

Figure 2. Karst regions of the world indicated in pink on the world map (www.circleofblue.org). 12.5 % of the continents (excluding Antarctica and large parts of Greenland and Iceland) are covered by carbonate-‐rocks outcrops and indicate that speleothems form an important archive to reconstruct the terrestrial climate. Another advantage of speleothems as paleoclimatic archive is that they often display continuous growth over long time intervals (10 to 100 000 years) making it possible to provide continuous paleoclimate reconstructions covering different glacial-‐interglacial cycles. Furthermore, speleothems are physically and chemically robust due to their relatively protective cave environment. Any calcite alteration that may have occurred is easily detectable usually by clear macroscopic calcite fabric changes. Finally, with the recent analytical and technical advances, different possible geochemical or physical parameters such as stable isotopes of oxygen and carbon, trace elements or layers thickness can be measured at decadal to sub-‐seasonal resolution. These parameters contain information about the climate above the cave during their deposition such as temperature, rainfall amount, type of vegetation, soil productivity and glacier extend. A good understanding how these parameters reflect, through their transfer functions, climate variations at the surface allows detailed reconstructions. Hendy and Wilson suggested in 1968 that the δ18O composition of speleothems deposited in equilibrium with their drip water could be used to reconstruct the paleo-‐temperature. This was the onset of the growing success of speleothems in reconstructing the paleoclimate as illustrated by the exponential increase of the amount of published papers since then (see ISI Web of Science). However, the early promise of reconstructing temperature curves from the δ18O composition, have proven to be difficult to fulfill due to the fact that most speleothems do not


4

reflect perfect isotopic equilibrium conditions (Dorale and Liu, 2009; Fairchild and Baker, 2012). However, even when affected by disequilibrium effects, different studies have proven that when carefully studied and correctly interpreted the signals can deliver important climate information such as on shifts in the intensity of the Monsoons (Cai et al., 2012; Fleitmann et al., 2007; Yuan et al., 2004) or variations of the NAO variation cycles (Proctor et al., 2000; Trouet et al., 2009). The advantage of the increasing amount of speleothem records is that continental and inter-‐continental comparison between the proxy-‐time series (McDermott et al., 2011) and with other paleoclimate archives (Bar-‐Matthews et al., 2003; Genty et al., 2003; Van Rampelbergh et al., 2013; Wang et al., 2001) become possible. Their increased temporal resolution and reliable absolute chronology also allow the records to be compared to climate forcing such as solar insolation that is necessary to understand the climate mechanisms (Neff et al., 2001; Wang et al., 2008). With the advances in analytical techniques such as laser absorption spectroscopy on speleothem fluid inclusions (Affolter et al., 2014) or the clumped isotope techniques (Kluge et al., 2008), the speleothem community is getting closer to extract temperature signals from speleothems such as first suggested by Hendy and Wilson in 1968. Although speleothems may seem the perfect archive to reconstruct the climate, the increasing amount of speleothem studies indicates that a variety of climatic, environmental and hydrological parameters influence the geochemical and physical properties of speleothems making their interpretation as a simple climate proxy not straightforward. More and more, the speleothem community realizes that each cave has its own unique geological and environmental setting that needs to be well understood before stable isotopes, trace elements, crystallographic changes, layer thickness records or other proxies can be used to reconstruct the climate. This clearly emphasizes that a profound understanding is needed of how the measured proxy in the selected cave is reflecting the climate. This can be achieved with long-‐term monitoring programs with frequent cave visits. However, such programs are logistically, technically and financially demanding making it difficult for research groups to carry them out. When cave monitoring programs are not feasible due to the remote location of the cave or due to logistic difficulties with installing the measurement material in the field, a multiproxy or if possible a multiple stalagmite analysis of the selected cave is necessary to get a good understanding of the cave system and how to interpret the measured proxies in terms of climate.


5

Figure 3. Comparison of different speleothem records with the NGRIP ice record summarized in Fleitmann et al. (2009). The curves indicate the value of speleothems not only for the absolute dating of transitions, but also to indicate how transitions or shorter climate events such as Heinrich events affected the terrestrial climate. In this PhD thesis, speleothems from two different climatic settings are studied, the semi-‐arid tropical region of the northern Indian Ocean and the more temperate region of the Belgium. The first part of the research focuses on analyzing four speleothems from the island of Socotra in Yemen to investigate how the Indian Ocean Monsoon has evolved in the Holocene. Because frequent cave visits were not possible due to the remoteness of the caves, no monitoring program was carried out. When deposited in equilibrium, the δ18O signal in speleothems from tropical and subtropical areas, such as Socotra, is mostly influenced by the “amount effect”, describing the inverse relationship between the amount of precipitation and its oxygen isotopic composition (Dansgaard, 1964; Rozanski et al., 1992) facilitating the interpretation of the measured


6

signals to a climate signal. However to be 100% certain that the amount-‐effect is responsible for the variations in the sampled speleothems, a multiproxy approach was done for each sample. Different speleothems covering the same time periods were sampled in the same cave to test the reproducibility of the results. Speleothems were sampled in two different caves to obtain a good understanding of the local climate evolution of Socotra. The multi-‐proxy, multi-‐speleothem and inter-‐cave comparison leads to a profound understanding of how the proxies reflect climate. Comparing the obtained Holocene climate variations derived from the studied speleothems with other records in the area such as ocean and lake cores or speleothem records from Southern Oman delivers a more profound knowledge of the mechanisms of the Indian Ocean Monsoon around the Northern Indian Ocean. In the second part of this PhD, a cave monitoring is combined with a seasonally resolved speleothem record from the Han-‐sur-‐Lesse cave, Belgium. The problem with most proxies used in speleothems from mid-‐latitudes is that they cannot be interpreted in terms of a single climate parameter such as temperature and/or precipitation due to the multiple potential influences (e.g. source, temperature, seasonal changes in rainfall, groundwater residence time, etc.) on the speleothem geochemistry and/or morphology (Baker et al., 2011). Therefore the approach to reconstruct a climate signal from a Belgian speleothem is different compared to the speleothem climate reconstruction in the sub-‐tropics. To investigate how the climate parameters such as temperature, precipitation or soil activity are linked to the cave parameters, a biweekly cave-‐monitoring program was carried out for a period of one year. In a second phase, the most recent 500 years of a 2 000 year old seasonally laminated stalagmite, called the Proserpine stalagmite, from the Han-‐sur-‐Lesse cave was analyzed for its δ18O and δ13C composition, layer thickness and calcite fabric changes at a seasonal scale. Together with the cave monitoring results the different analyzed proxies deliver information on how the climate evolved up to seasonal scale during the most recent 500 years. The results obtained from this PhD provide information that can be used to refine climate models around the northern Indian Ocean and around Western Europe. For the monsoon regions our work showed that not only the summer-‐monsoon affects the continental regions around the northern Indian Ocean, but that a winter monsoon also has to be taken into account and that they both evolve differently over time. The work in Belgium elaborated the general knowledge of how climate variations are linked to variations in cave parameters. These findings can be used to translate speleothem proxies recovered from mid-‐latitude locations to climate signals.


7

References Affolter, S., Fleitmann, D., Leuenberger, M., 2014. New online method for water isotope analysis of speleothem fluid inclusions using laser absorption spectroscopy (WS-‐CRDS). Climate of the Past 10, 1291-‐1304.

Arrhenius, S., 1896. On the Influence of Carbonic Acid in the Air upon the Temperature of the Ground Philosophical Magazine and Journal of Science Series 5, 237-‐276.

Baker, A., Wilson, R., Fairchild, I.J., Franke, J., Spoetl, C., Mattey, D., Trouet, V., Fuller, L., 2011. High resolution delta O-‐18 and delta C-‐13 records from an annually laminated Scottish stalagmite and relationship with last millennium climate. Glob. Planet. Change 79, 303-‐311.

Bar-‐Matthews, M., Ayalon, A., Gilmour, M., Matthews, A., 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, 3181-‐3199.

Cai, Y., Zhang, H., Cheng, H., An, Z., Edwards, R.L., Wang, X., Tan, L., Liang, F., Wang, J., Kelly, M., 2012. The Holocene Indian monsoon variability over the southern Tibetan Plateau and its teleconnections. Earth and Planetary Science Letters 335, 135-‐144.

Cheng, H., Edwards, R.L., Broecker, W.S., Denton, G.H., Kong, X., Wang, Y., Zhang, R., Wang, X., 2009a. Ice Age Terminations. Science 326, 248-‐252.

Cheng, H., Edwards, R.L., Hoff, J., Gallup, C.D., Richards, D.A., Asmerom, Y., 2000. The half-‐lives of uranium-‐234 and thorium-‐230. Chemical Geology 169, 17-‐33.

Cheng, H., Fleitmann, D., Edwards, R.L., Wang, X.F., Cruz, F.W., Auler, A.S., Mangini, A., Wang, Y.J., Kong, X.G., Burns, S.J., Matter, A., 2009b. Timing and structure of the 8.2 kyr BP event inferred from delta O-‐18 records of stalagmites from China, Oman, and Brazil. Geology 37, 1007-‐1010.

Dansgaard, W., 1964. Stable Isotopes In Precipitation. Tellus 16, 436-‐468.

Dansgaard, W., Johnsen, S.J., Clausen, H.B., Dahljensen, D., Gundestrup, N.S., Hammer, C.U., Hvidberg, C.S., Steffensen, J.P., Sveinbjornsdottir, A.E., Jouzel, J., Bond, G., 1993. Evidence for general instability of past climate from a 250-‐Kyr Ice-‐Core record Nature 364, 218-‐220.

Dorale, J.A., Liu, Z., 2009. Limitations of the Hendy test criterai in judging the paleoclimatic suitability of speleothems and the need for replication. Journal of Cave and Karst Studies 71, 73-‐80.

Edwards, R.L., Chen, J.H., Wasserburg, G.J., 1987. 238U-‐234U-‐230Th-‐232Th systematics and the precise measurement of time over the past 500 000 years. Earth and Planetary Science Letters 81, 175-‐192.


8

Fairchild, I.J., Baker, A., 2012. Speleothem Science: From process to past environment. Blackwell Publishing ltd.

Fairchild, I.J., Baker, A., Borsato, A., Frisia, S., Hinton, R.W., McDermott, F., Tooth, A.F., 2001. Annual to sub-‐annual resolution of multiple trace-‐element trends in speleothems. Journal of the Geological Society 158, 831-‐841.

Fleitmann, D., Burns, S.J., Mangini, A., Mudelsee, M., Kramers, J., Villa, I., Neff, U., Al-‐Subbary, A.A., Buettner, A., Hippler, D., Matter, A., 2007. Holocene ITCZ and Indian monsoon dynamics recorded in stalagmites from Oman and Yemen (Socotra). Quaternary Science Reviews 26, 170-‐188.

Fleitmann, D., Cheng, H., Badertscher, S., Edwards, R.L., Mudelsee, M., Goektuerk, O.M., Fankhauser, A., Pickering, R., Raible, C.C., Matter, A., Kramers, J., Tuysuz, O., 2009. Timing and climatic impact of Greenland interstadials recorded in stalagmites from northern Turkey. Geophysical Research Letters 36.

Genty, D., Blamart, D., Ouahdi, R., Gilmour, M., Baker, A., Jouzel, J., Van-‐Exter, S., 2003. Precise dating of Dansgaard-‐Oeschger climate oscillations in western Europe from stalagmite data. Nature 421, 833-‐837.

Heinrich, H., 1988. Origin and consequences of cyclic ice rafting in the northeast atlantic-‐ocean during the past 130 000 years. . Quaternary Research 29, 142-‐152.

IPCC, 2013. IPCC 2013: Summary for policy makers., In: Stocker, T.F., Qin, D., Plattner, G.K., Tingor, M., Allen, S.K., Boschung, J., Nauels, A., Xia, Y., Bex, V., Midgley, P.M. (Eds.), Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fift Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom and New York, USA.

Kluge, T., Marx, T., Scholz, D., Niggemann, S., Mangini, A., Aeschbach-‐Hertig, W., 2008. A new tool for palaeoclimate reconstruction: Noble gas temperatures from fluid inclusions in speleothems. Earth and Planetary Science Letters 269, 407-‐414.

Mattey, D., Lowry, D., Duffet, J., Fisher, R., Hodge, E., Frisia, S., 2008. A 53 year seasonally resolved oxygen and carbon isotope record from a modem Gibraltar speleothem: Reconstructed drip water and relationship to local precipitation. Earth and Planetary Science Letters 269, 80-‐95.

McDermott, F., Atkinson, T.C., Fairchild, I.J., Baldini, L.M., Mattey, D.P., 2011. A first evaluation of the spatial gradients in delta O-‐18 recorded by European Holocene speleothems. Glob. Planet. Change 79, 275-‐287.

Neff, U., Burns, S.J., Mangini, A., Mudelsee, M., Fleitmann, D., Matter, A., 2001. Strong coherence between solar variability and the monsoon in Oman between 9 and 6 kyr ago. Nature 411, 290-‐293.

Polyak, V., Hill, C., Asmerom, Y., 2008. Age and evolution of the Grand Canyon revealed by U-‐Pb dating of water table-‐type speleothems. Science 319, 1377-‐1380.


9

Proctor, C.J., Baker, A., Barnes, W.L., Gilmour, R.A., 2000. A thousand year speleothem proxy record of North Atlantic climate from Scotland. Climate Dynamics 16, 815-‐820.

Rozanski, K., Araguasaraguas, L., Gonfiantini, R., 1992. Relationship between long-‐term trends of 18O isotope composition of precipitation and climate Science 258, 981-‐985.

Shacklet.N, 1968. Depth of pelagic foraminifera and isotopic changes in pleistocene oceans Nature 218, 79-‐&.

Shackleton, N.J., Vincent, E., 1978. Oxygen and carbon isotope studies in recent foraminifera from Southwest Indian Ocean Marine Micropaleontology 3, 1-‐13.

Thompson, L.G., Mosley-‐Thompson, E., Davis, M.E., Zagorodnov, V.S., Howat, I.M., Mikhatenko, V.N., Lin, P.N., 2013. Annually Resolved Ice Core Records of Tropical Climate Variability over the Past similar to 1800 Years. Science 340, 945-‐950.

Treble, P., Shelley, J.M.G., Chappell, J., 2003. Comparison of high resolution sub-‐annual records of trace elements in a modern (1911-‐1992) speleothem with instrumental climate data from southwest Australia. Earth and Planetary Science Letters 216, 141-‐153.

Treble, P.C., Chappell, J., Shelley, J.M.G., 2005. Complex speleothem growth processes revealed by trace element mapping and scanning electron microscopy of annual layers. Geochimica Et Cosmochimica Acta 69, 4855-‐4863.

Trouet, V., Esper, J., Graham, N.E., Baker, A., Scourse, J.D., Frank, D.C., 2009. Persistent Positive North Atlantic Oscillation Mode Dominated the Medieval Climate Anomaly. Science 324, 78-‐80.

Van Rampelbergh, M., Fleitmann, D., Verheyden, S., Cheng, H., Edwards, L., De Geest, P., De Vleeschouwer, D., Burns, S.J., Matter, A., Claeys, P., Keppens, E., 2013. Mid-‐ to late Holocene Indian Ocean Monsoon variability recorded in four speleothems from Socotra Island, Yemen. Quaternary Science Reviews 65, 129-‐142.

Van Rampelbergh, M., Verheyden, S., Allan, M., Quinif, Y., Keppens, E., Cheng, H., Edwards, L.R., Claeys, P., in review. A 500-‐year seasonally resolved δ18O and δ13C and layer thickness record from a speleothem of the Han-‐sur-‐Lesse cave, Belgium. Climate of the Past.

Wang, Y., Cheng, H., Edwards, R.L., Kong, X., Shao, X., Chen, S., Wu, J., Jiang, X., Wang, X., An, Z., 2008. Millennial-‐ and orbital-‐scale changes in the East Asian monsoon over the past 224,000 years. Nature 451, 1090-‐1093.

Wang, Y.J., Cheng, H., Edwards, R.L., An, Z.S., Wu, J.Y., Shen, C.C., Dorale, J.A., 2001. A high-‐resolution absolute-‐dated Late Pleistocene monsoon record from Hulu Cave, China. Science 294, 2345-‐2348.


10

Winograd, I.J., Ludwig, K.R., 1996. High resolution paleotemperature proxy record for the last interglaciation based on Norwegian speleothems -‐ Comment. Quaternary Research 45, 102-‐102.

Yuan, D., Cheng, H., Edwards, R.L., Dykoski, C.A., Kelly, M.J., Meiliang, Z., Jiaming, Q., Yushi, L., Yongjin, W., Jiangyin, W., Dorale, J.A., Zhisheng, A., Yanjun, C., 2004. Timing, duration, and transitions of the last interglacial Asian monsoon. Science 304, 575-‐578578.

Zachos, J., Pagani, M., Sloan, L., Thomas, E., Billups, K., 2001. Trends, rhythms, and aberrations in global climate 65 Ma to present. Science 292, 686-‐693.

Chapter 2: Speleothems and climate

11

Chapter 2

Speleothems and Climate

2.1 Speleothems Speleothems such as stalagmites, stalactites or flowstones are natural cave deposits that mostly consist of calcium carbonate (CaCO3). The development and growth of speleothems in karst regions have been described extensively in several publications (Dreybrodt, 1988; Fairchild et al., 2006; Lachniet, 2009; Fairchild and Baker, 2012). When rainwater falls only a fraction will effectively infiltrate into the soil due to evaporation and plant uptake. In the soil, the higher pCO2 (typically 0.1 atm) compared to the water pCO2 (typically 10-‐4 atm) acidifies the water according to reaction [1] (Fig. 1): CO2 + H2O à H2CO3 [1] until equilibrium with respect to the partial CO2 pressure in the soil is attained. After passing through the soil zone, the water enters the epikarst. The epikarst is the upper surface of the bedrock characterized by solution features in the vadose zone where water may be stored and mixed (Williams, 2008). In the epikarst, the corrosive water comes in contact with the carbonate bedrock (mostly marine limestone) and dissolves it to become enriched in calcium carbonate according to reaction [2] (Fig. 1): CaCO3 + H2CO3 à Ca2+ + 2HCO3-‐ [2]


12

This zone is referred as the ‘dissolution’ zone of the epikarst (Fig. 1) (Fairchild et al., 2006). The amount of dissolved calcite depends on the amount of dissolved CO2 in the solution (the acidity of the water) and whether dissolution occurs in an open or closed system (see 2.3.2) (Dreybrodt, 1988).

Figure 1. The dissolution and precipitation regimes of the karst environment. Water gets enriched in CO2 in the soil zone, becomes more acidic and dissolves the calcite bedrock (dissolution zone). In the cave, the lower cave air pCO2 compared to the water pCO2 forces the drip water to degas. The water becomes supersaturated with respect to calcite and CaCO3 is deposited to form speleothems (adapted after Fairchild et al. 2006). When vadose water enters the cave, where pCO2 levels are lower than in the limestone crack system, CO2 degasses from the water, the drip water becomes supersaturated with respect to calcite, and calcite is precipitated in the form of speleothems according to the reaction [3] (Fig. 1): Ca2+ + 2HCO3-‐ à CaCO3 + H2O +CO2 [3] The CO2 degassing of the water occurs immediately within 10 seconds after entering the cave atmosphere through molecular diffusion (Dreybrodt and Scholz, 2011). When the drip water falls on top of a stalagmite, further CO2 is lost due to diffusion within a few seconds (Dreybrodt, 1980) and supersaturation with respect to calcite increases even further. It is likely that a portion of the drip water is lost during the impact of the drop on the surface of the stalagmite


13

due to splashing effects. The zone, where speleothems are formed is referred as the ‘precipitation’ zone (Fig. 1) (Fairchild et al., 2006). Around the drip hanging on the ceiling, calcite deposits in the shape of a rim. The consecutive rims at the ceiling will lead to a straw shaped calcite form referred as “soda straws” (Fig. 2). When too long, these straws become clogged and the drip starts to flow along the surface of the straw. Over time, when more calcite is deposited, the soda straw evolves into a stalactite. This explains why paleoclimate reconstructions cannot be based on stalactites. The center of the stalactite is hollow due to the soda straw and the consecutively deposited calcite layers are not nicely following each other. However in stalagmites, the consecutive falling drips will cause the layers to be deposited on top of each other with the youngest layers at the top of the stalagmite (Fig.2). Stalagmites are consequently better suited for paleoclimate reconstructions.

Figure 2. Soda straws (left) and the different location of speleothem deposits in caves (after Fairchild et al. 2006). The speleothem growth rate depends on different factors such as (i) drip water calcium ion concentration, (ii) discharge amounts, and (iii) the pCO2 gradient between the cave air and drip water rendering the estimation of an average rate difficult (Genty and Quinif, 1996; Baker et al., 1998; Genty et al., 2001b; Frisia et al., 2003). The main factor driving growth rate changes is (i) the drip water calcium ion concentration (Genty et al., 2001b), which can increase due to longer residence times of water in the epikarst, when soil pCO2 values are higher. The drip water calcium ion concentration may decrease when calcite precipitates from the water before entering the cave (Prior Calcite Precipitation, see 2.3.6 for more details about this process), a process mostly occurring during drier periods when aerated zones in the epikarst become more important (e.g. Van Rampelbergh et al., in review) . Discharge changes (ii) may also affect growth rate, with higher discharge causing faster calcite deposition. This effect is more pronounced for slow dripping sites, where the amount of water provided to precipitate a stalagmite is low. A stronger pCO2 gradient between the drip water


14

and the cave air (iii) will cause stronger CO2 degassing of the drip water when entering the cave, which causes higher calcite supersaturation of the drip water. Generally, stalagmites are considered to grow on average at 10-‐100 μm/y in cool temperate climates and at 300-‐500 μm/y in sub-‐tropical climates (Fairchild et al., 2006). However, in reality a large variety of growth rates are observed. For example, growth rates up to 2 mm/y may occur such as is observed in the Proserpine stalagmite from the Han-‐sur-‐Lesse cave, Belgium (Van Rampelbergh et al., in review). In such fast growing speleothems annual or seasonal layering may be detected if growth parameters such as calcium ion concentration or drip discharge vary annually or seasonally (Genty and Quinif, 1996). Layering in stalagmites mostly consists of alternating dark and more compact layers with whiter and more porous layers and is mostly an indicator of the seasonal changes in cave climate and chemistry (Treble et al., 2005b; Mattey et al., 2008; Van Rampelbergh et al., 2014). Fluorescent annual lamina, rich in organic matter, may also occur in speleothems and are related to changes in organic matter flux from the surface (Frisia et al., 2000; Proctor et al., 2000; Fairchild et al., 2001; Genty et al., 2001b), or to seasonal flushing of soil-‐derived elements and particles (Frisia et al., 2000). The time needed for the rainwater to reach the cave, referred to as the transition time, depends on the ‘flow-‐type’ of the epikarst. The flow-‐type of karst groundwater encountered in caves vary from very slow dripping seepage flow to rather high-‐discharge types, including shaft flow and subcutaneous flow (Smart and Friedrich, 1987). Drip sites feeding actively forming stalagmites are typically of the seepage flow type (i.e. low discharge, low variability), the seasonal-‐drip type (i.e. low discharge, but seasonal variability), or the vadose flow type (higher discharge, commonly giving rise to flowstone deposits). Summarized, three major sources are important for the formation of speleothems: rainwater, soil activity and the presence of a carbonate bedrock. The calcite layer that is formed consequently reflects the geochemical and chemical composition of the water, soil and bedrock at the time of its formation. 2.2 Dating of speleothems The development of techniques for precise and accurate dating of speleothems on small amounts of samples has allowed speleothem science to become so prominent in the recent years (Fig. 3) (Fairchild and Baker, 2012). Speleothems can now be absolutely dated with very small error bars, down to 0.1 %, for the ages making them valuable archives in dating large climatic events.


15

Figure 3. a) An overview of the different speleothem dating techniques and their corresponding timescales (y-‐axis). When visible, layering can be used to date young speleothems. 14C can be used in young speleothems to detect the 1964 ‘bomb peak’ as anchor point for the age model. The most commonly used dating technique in speleothems is the U/Th-‐dating technique that delivers absolute ages with very small error ranges. The recently developed U/Pb-‐dating technique can be used to determine speleothem samples older than the U/Th-‐age limit of 600 kyr. b) Methodological advances in dating techniques have increased the precision of ages obtained in speleothems. The most used speleothem dating technique is based on ICP-‐MS (Inductively Coupled Plasma Mass Spectrometer) determined U/Th-‐ages. Picture from Fairchild and Baker (2012). 2.2.1. Dating techniques U-‐series or U/Th-‐dating is the most commonly used technique when working with speleothems. The U/Th method allows precise dating of the samples back to 600 kyr (Fig. 3a) (Edwards et al., 1987; Cheng et al., 2000; Cheng et al., 2009a; 2009b). The technique is based on the decay of the parent isotope 238U to 230Th. Uranium has the property to be soluble in water while thorium is not. Both


16

elements are present in the carbonate bedrock. The drip water will dissolve uranium, while Th will link to bigger colloids or organic molecules. When calcite is deposited from the drip water in the cave, the radioactive 238U gets enclosed in the crystal structure at concentrations typically varying between 0.05 and 0.5 ppm. From there, 238U decays radioactively according to its series until stable 206Pb is produced. For U/Th-‐ages, only a part of the decay series is used, being the decay of 234U to 230Th. This dating method is different from other dating techniques since it does not use the stable end of the decay chain but rather looks at the balance between mother 234U and the produced radioactive daughter 230Th. The amount of 230Th measured in the speleothem calcite is thus an indicator of the age of the calcite. However, ‘detrital 230Th’ can be incorporated in the deposited calcite together with impurities causing artificial older U-‐Th ages. The measured U/Th-‐age can be corrected by measuring the amount of 232Th that is used as proxy for the detrital 230Th in the sample. Together with an estimation of the 230Th/232Th ratio of the detrital phase, the measured U/Th-‐age can be corrected for its amount of ‘detrital 230Th’. The 230Th/232Th ratio can be estimated using the isochron techniques or better by measuring the ratio directly in the chemically separated detrital phase (Dorale et al., 1998). The main problem when using the isochron technique is that only few estimates of the 230Th/232Th ratio exists and that it is high likely that this ratio varies from site to site and through time at a given site, such that the 232Th-‐corrected ages can be associated with extremely large errors. This is especially the case in young speleothems or with samples with low U-‐concentrations where little 230Th is present from decay of uranium. Dating results with too high 232Th content are mostly not used to establish the age model of the stalagmite. Although the problem of ‘detrital 230Th’ may be present in speleothems, most studied speleothem samples are not contaminated and can be U/Th datet with high precision. The development of the U/Th-‐dating technique started in the 1970s with α-‐spectrometry and has been further developed using TIMS (Thermal Ionization Mass Spectrometer) and finally ICP-‐MS (Inductively Coupled Plasma Mass Spectrometer) based measurements (Fig. 3b) that allow to determine the ages of small samples between 100 and 400 mg depending on the U-‐content (only ~2.5ng of U is needed). Precision goes up to 5 % and much worse for detrital contaminated samples.

Seasonal layers, if visible, can be used to establish the age model for speleothems covering the last millennia by counting the layers back in time (Fig. 3a). This technique has the large advantage to be non-‐destructive for the sample. Due to small changes in cave climatology between seasons, layers deposited in stalagmites are often seasonal with one darker and one lighter layer deposited every year (Genty and Quinif, 1996). Clear seasonal layering mostly occurs in speleothems with significant high growth rates (>1 mm/y) and has proven to be very efficient in providing absolute time chronologies for the measured proxies


17

(Mattey et al., 2008; Van Rampelbergh et al., in review). Not only visible layers can be used for counted chronologies, trace elemental cycles are often observed in speleothems with growth rates higher than 0.05 mm/y and can also be used to provide counting ages (Fairchild et al., 2001; Treble et al., 2003; 2005b; Smith et al., 2009). However, not all stalagmites display layering and if present, the layering is often interrupted by hiatuses or parts where the layering is hard to define, making it difficult to estimate when to restart the counting. Therefore counted age models always have to be calibrated with U-‐series ages. Radiocarbon or 14C-‐dating can also be applied on speleothems and was the first dating technique used to date speleothem samples in the 1960’s (Broecker et al., 1960). However, the problem with this technique is that the amount of ‘dead carbon’ coming from dissolution of the host rock or old organic matter remains in the epikarst needs to be estimated to correct the systematically too old measured 14C-‐ages. This dead carbon proportion typically varies around 15%, but needs to be estimated for each sample by comparing the 14C-‐ages with U/Th-‐ages (Genty and Massault, 1997). For example, 14C measurements in Scottish stalagmites suggest a dead carbon proportion between 22 and 38 % that is the results of the ageing of the soil organic matter related to peat bog development above the cave (Genty et al., 2001a). Even when the ‘dead carbon’ proportion of the analyzed sample is known, 14C ages have to be calibrated against the dendrochronologic calibration curve causing the obtained ages to have larger uncertainties compared to the U/Th or layer counting ages. The measured 14C value and its 2σ uncertainty range are projected against the dendrochonologic calibration curve. Because the calibration curve is not a straight line it is possible to have many intercepts on the calendar year axis, each with its own probability range. Ages are reported as an interval with their own probability range, which depends strongly on the measured 14C activity. The 14C ages obtained on speleothems are thus flawed and mostly not used to establish speleothem age models. The 14C-‐ages can be useful to date young samples where no layering is visible and too little uranium or too high amounts of detrital 230Th are present. For such samples the speleothem 14C activity can be compared with the 1964 ‘bomb test’ peak delivering absolute ages (Genty and Massault, 1997). This has for example been done to determine the age of a recent speleothem from St Michael’s cave, Gibraltar, where the ‘bomb peak’ was used as spike and the age model was established counting annual geochemical cycles (Mattey et al., 2008). In 1998, the age limit to date speleothems back to 600 kyr was broken by the publication of Richards et al. (1998) where they described that older speleothems could be dated with the U/Pb technique, which allows dating back to > 400 Ma with a precision of 1 to 5 %. However, more than a decade was necessary to develop suitable methods to determine these ratios (Walker et al., 2006; Woodhead et al., 2006). The main problem with U/Pb ages is the very low


18

levels of Pb in speleothems and the difficulty in obtaining a range of parent/daughter isotope ratios for isochron reconstructions. The development of MC-‐ICPMS formed a breakthrough is this discipline making it possible to determine isotope ratios with very low concentrations. However, this dating technique for speleothems is still under development due to problems with chemical opening of the system in old sampled where U may be lost and Th and Pb may be injected in the system. Bajo et al. (2012) successfully used the U/Pb dating technique to date an Early to Middle Pleistocene (± 800 ka) speleothem of Corcia Cave (Italy). Vaks et al. (2013) used U/Pb dated speleothem depositions in an Israeli cave to reconstruct climate variations during the Pliocene and Pleistocene (± 3 Ma). Water table speleothem U/Th-‐ages provided age estimations for the incision of the Grand Canyon (Polyak et al., 2008). 2.2.2. Establishing an age-‐depth model Since the measured climate proxies have a higher resolution (mm to µm) than the age points (mm to cm), ages need to be interpolated in an age-‐depth model based on the obtained ages. Small growth rate changes or hiatuses between two consecutive dating points can induce significant chronological errors into the model. Therefore is it important to sample the dating points at strategic places in the stalagmite where growth rate changes or hiatuses are expected. Such changes or hiatuses are often indicated by changes in the crystallography of the sample or in the measured proxy records. The most common approach to obtain an age model is to linearly interpolate the ages between two dating points. However, the age model is only based on two adjacent data points, is not smooth at data points and usually has no quantification of the age error between data points. Least-‐squares polynomial fits have the tendency to display too much curving and can create overshoots (Scholz et al., 2012). Splines can also be used (Hodge et al., 2008), but they require too much modification of the parameters making the age model too subjective. Different authors have tried to establish models that provide the best possible ages for speleothems such as the MODAGE by Hercman and Pawlak (2012) or COPRA by Breitenbach (2012), but these model are mostly complex and difficult to use. The StalAge age model established by Scholz and Hoffmann (2011) delivers the best age-‐depth model for speleothems currently available that is easy to use. However, the weakness of the age model is that it tries to fit straight lines through three adjacent age-‐points averaging out possible small growth rate changes between points. The best results are obtained by comparing the age-‐depth relation obtained by StalAge with crystallographic and climate proxy variations in the speleothem. Models to establish age-‐depth relationships in speleothems are useful but still need to be interpreted critically by the author. The location of the age points is still a very important factor in the correctness of the age model and needs to be well thought before sampling


19

2.3 δ18O and δ13C in Speleothems 2.3.1 Factors that determine the δ18O of the drip water 1. δ18O VARIATIONS IN THE RAINWATER The water dripping from a cave ceiling originates from the part of the rainwater that is infiltrated in the soil, which is the difference between the amount of precipitation and the amount of water lost by evaporation and plant transpiration (evapo-‐transpiration). The δ18O composition of rainwater is influenced by different isotope effects being the latitude effect, the continental effect, the altitude effect, the temperature effect and amount effect (Dansgaard, 1964; Rozanski et al., 1992). Derived effects are the seasonal effect due to seasonal variations of temperature and/or amount, and the source effect by geographical location of the source of rainwater.

Figure 4. Schematic oxygen (and hydrogen) isotope fractionation of water in the atmosphere (adapted after Hoefs (1997)). The latitude effect relates to the gradual depletion of the δ18O (and also δD) in the atmospheric water during its successive rainouts (Fig. 4) on its way to the poles (Rozanski et al., 1992). The evolution of the rainwater δ18O signal due to the successive rainouts of the clouds follows the Rayleigh distillation model (Rayleigh, 1896) and roughly equals -‐0.18 ‰ per latitudinal degree of the Northern Hemisphere (Rozanski et al., 1992). The same occurs when a cloud moves inland from the coast. This effect is referred as the continental effect and


20

describes the gradual depletion in δ18O (and also δD) of the rainwater further away from its water source. Recently, McDermott et al. (2011) demonstrated that the continental effect observed in the European precipitation is preserved by speleothems δ18O values. The altitude effect relates to the fact that the higher the altitudinal location of the rain, the more the water becomes depleted with generally -‐0.2 to -‐0.3 ‰ per increase of 100 m (Lachniet and Patterson, 2006). For one geographical location the latitude, altitude and the distance in-‐land or continentality determine the rainwater isotopic composition and can be estimated constant during relatively short geological periods of time. The rainwater isotopic composition at a certain geographical location will vary over time due to the temperature effect, the amount effect or the source effect. The temperature effect relates to the influence of temperature on the δ18O of precipitation through time. During warmer periods, more heavy isotopes will evaporate from the source of the rainwater (mostly the ocean) and cause increased rainwater δ18O values (Rozanski et al., 1992). The temperature effect on the rainwater δ18O signal can vary between +0.17 and +0.9 ‰/1°C, depending on the geographical location (Dansgaard, 1964; Rozanski et al., 1992; Mook, 2000; Schmidt et al., 2007). Schmidt et al. (2007) suggested a dependence of +0.3 ‰/1°C for central Europe. When colder and warmer seasons are present the temperature effect will cause a seasonal effect on the rainwater isotopic composition with isotopically lighter rainwater during the colder season. The seasonal effect is the smallest at the equator, where temperature remains constant throughout the year but is more pronounced at the poles, where temperature can vary more strongly between winter and summer. The amount effect implies that more rain causes the water to be more depleted in isotopic composition (Rozanski et al., 1992). During heavy rainfalls, the water drop only partially exchanges its oxygen isotopic composition with surrounding more positive air preserving its more negative δ18O composition from in the cloud (Gat, 1996). In tropical and sub-‐tropical areas, where precipitation intensities display large variations through monsoonal periods the amount effect will have the largest influence on the isotopic composition of the rainwater. For one geographical location, rainwater can also originate from different sources with different isotopic compositions inducing a source effect leading to abrupt changes in the rainwater isotopic composition. 2. δ18O VARIATIONS IN THE SOIL ZONE The δ18O of the water in the soil zone is expected to represent the amount-‐weighted mean of the δ18O of infiltrating rainwater that may be further modified by evaporation. In (semi-‐)arid regions the soil-‐ and groundwater δ18O can increase compared to the weighed mean of the rainwater δ18O due to substantial


21

evaporation of water from surface and vadose zone. Also, when a part of the rainwater evaporates before entering the soil, such as is often the case, the δ18O signal of the drip water can increase. The magnitude of δ18O increase will be related to the relative humidity in the soil pores and the evaporated water volume (Lachniet, 2009). However, the most intense rainstorms that commonly have low δ18O values are likely to dominate recharge into the soil zone, and would tend to counteract the isotopic enrichment associated with evaporation (Dansgaard, 1964; Rozanski et al., 1992). Plant respiration does not affect the δ18O composition of water (Gat, 1996). 3. δ18O VARIATIONS IN THE EPIKARST The timing and amount of recharge to the epikarst is an important control on the resulting drip water δ18O. For most western European caves such as the Han-‐sur-‐Lesse cave (Belgium, Bonniver, 2011) or the Bunker Cave (western-‐Germany, Fohlmeister et al. 2012) , recharge occurs in winter causing the δ18O variations in the speleothem to reflect winter precipitation changes (Fohlmeister et al., 2012; Van Rampelbergh et al., in review). In tropical and sub-‐tropical regions such as Oman, heavy monsoon rains recharge the epikarst and only a small amount of water evaporates causing the drip water δ18O to reflect the rainwater annual mean δ18O composition (Fleitmann et al., 2004). If evaporation processes are more important, the δ18O composition of the water in the epikarst can increase compared to the annual mean rainwater δ18O (BarMatthews et al., 1996). The transition time of water in the vadose zone determines how rainwater signals are transferred to the cave. This transition time can be estimated by lag times of δ18O values (using them as an built-‐in tracer) in drip waters relative to rainfall values (Baldini et al., 2006; Cobb et al., 2007), chemical variations in karst waters (Genty and Deflandre, 1998), delay response in precipitation events (Cobb et al., 2007; Van Rampelbergh et al., 2014) from fluorescent tracers (Bonniver, 2011), fluorescent organic matter in the drip water (Hartland et al., 2010) or by tritium (3H) tracing (Kluge et al., 2010). Depending on the epikarst thickness and epikarst flow systems, the transfer time of the drip water from the soil to the cave varies from months to several years such as observed in the shallow (-‐15 to -‐30m) Ernesto Cave, Italy (Miorandi et al., 2010), Bunker cave, Germany (Kluge et al., 2010) and Han-‐sur-‐Lesse Cave, Belgium (Van Rampelbergh et al., 2014). For very deep cave systems, such as Monte Corchia cave in central Italy (-‐1187m) drip water residence times of decades can be observed (Piccini et al., 2008). Transit time is expected to be most rapid for the conduits, and slowest for the diffuse seepage flow. In a general sense and all of the factors being equal, the thicker the overlying limestone, the longer the potential transit time and the more complete the groundwater mixing. Systems


22

with shorter transit times will be more suitable for capturing rapid, high-‐frequency climate events (McDonald et al., 2007), whereas very slow transit time with substantial mixing will be more suited for constraining long-‐term climate change. The saturation state of the drip water may vary over time, thus influencing the timing of calcite deposition (Treble et al., 2005a; Baldini et al., 2006). Only those recharge waters, with a given isotopic composition that are saturated with CaCO3 will participate to the deposition of speleothem CaCO3. It is this isotopically effective recharge, i.e. the recharge water with a certain isotopic signature that is saturated with CaCO3, that is relevant to the interpretation of speleothem δ18O time series (Lachniet, 2009). The timing of isotopically effective recharge may be forced by seasonal variations in drip water and cave air pCO2, which influences drip water degassing rates, and can impart a seasonal bias to the speleothem record if certain months produce more calcite than other ones (Baldini et al., 2008; Mattey et al., 2008). Ideally, comprehensive studies of infiltration and drip water geochemical measurements (δ18O, pH, pCO2, calcite saturation indices) over the course of several years should be completed to understand the saturation variations and the timing of drip and of the transfer of isotopic signals (Baldini et al., 2006; Mattey et al., 2008). The δ18O value of the drip water is a function of seasonality of recharge and modification within the soil and epikarst. Typically drip waters δ18O variability is attenuated relative to precipitation δ18O due to mixing in the soil zone and the epikarst (Mattey et al., 2008; Genty et al., 2014; Van Rampelbergh et al., 2014) 2.3.2 Factors that determine the δ13C of the drip water The carbon isotopic composition of the drip water is mainly determined by the δ13C of soil CO2 and limestone and, to a smaller part, by the dissolution intensity of the host rock, atmospheric CO2 or pCO2 in the cave atmosphere (Dreybrodt and Scholz, 2011). Rainwater equilibrates with the atmospheric pCO2 and with its δ13C composition. The increased combustion of isotopically light organic carbon over the past nearly two centuries has caused the δ13C composition of the atmospheric CO2 to decrease due to the ‘Suess-‐effect’ (Suess, 1955) from ∼ -‐6.4 ‰ between 1000 and 1800 AD to about -‐8.2 ‰ in 2011 (Cuntz, 2011). The major sources of CO2 in the soil zone are the vegetation and the decay of organic matter (Salomons and Mook, 1986). The soil CO2 composition is mainly driven by the intensity and type of vegetation (C3, C4 or CAM plants) above the


23

cave. The δ13C values of soil CO2 around -‐26 ‰ reflect a C3-‐type of vegetation that is typical for humid climates while δ13C values of soil CO2 around -‐13 ‰ reflect a C4-‐type of vegetation that is typical for water stressed environments (Salomons and Mook, 1986; McDermott et al., 2005). During periods of decreased plant CO2 input in the total soil CO2 reservoir, atmospheric CO2 that infiltrates into the soil can influence the soil δ13C signature by several ‰ (Baker et al., 1997). Within the soil, the high soil pCO2 (0.1 atm) compared to the water pCO2 (0.001 atm) forces the dissolution of soil CO2 in the water. This process goes on until the soil CO2 partial pressure is attained in the water. Consequently, higher soil pCO2 values will cause more CO2 to be dissolved in the water. In the epikarst, the corrosive water dissolves the carbonate bedrock. Two end-‐member models, being an open or a closed system, describe how this dissolution process occurs (Hendy, 1971; Salomons and Mook, 1986). In an open system model, continuous equilibration occurs between the seepage water and the ‘infinite’ reservoir of soil CO2 during the dissolution. Under these conditions, the δ13C value of the water reflects the δ13C composition of the soil CO2 (mainly from vegetation). Under closed conditions, the percolating water becomes isolated from soil CO2 as soon as the carbonate dissolution starts. The δ13C value of the drip water is influenced by the δ13C value of the host rock typically varying around 0 ‰ (Salomons and Mook, 1986), the host rock nearly always being a marine limestone. Most natural cave systems are likely to be partially open causing the δ13C value of the drip water to be mainly determined by the isotopic composition of the soil CO2 (Dreybrodt and Scholz, 2011). Genty and Massault (1997) estimated a ca. 15 % contamination of the soil water δ13C by limestone δ13C. However, this % can vary according to the amount of dissolved CaCO3, which relates to the degree of the open/closed dissolution system and of the pCO2 in the soil. At a given pCO2 value, under a more open dissolution system, more CaCO3 will be dissolved, because more CO2 is available during the dissolution in comparison to a closed system state (Hendy, 1971). Under a given dissolution system, the higher the soil pCO2 value, the more corrosive the water and the more CaCO3 can be dissolved in the water. More intense dissolution of isotopically heavy CaCO3 will cause an increase of the drip water δ13C. In the cave, the δ13C composition of the drip water can also be influenced by equilibration with the δ13C values of the cave air. Dreybrodt and Scholz (2011) studied the possible effect of equilibration between the pCO2 of the dripwater and the pCO2 of the cave atmosphere. Isotopic exchange between the carbon isotopes in the drip water and the carbon isotopes of the cave air CO2 drives the δ13C of the drip water to the δ13C value of the cave air CO2. Equilibration between


24

the δ13C of the drip water and δ13C of the cave air needs about 1 h. This process will, however, not affect the drip water δ13C if the interaction time is 4 times shorter than the time needed for full equilibration (= 1h) (Dreybrodt and Scholz, 2011). Consequently, only drip water, which falls on the stalagmite within less than 15 minutes after entering the cave atmosphere, will have a δ13C composition that is not affected by the δ13C composition of the cave air (Dreybrodt and Scholz, 2011). 2.3.3 Equilibrium fractionation factors For a given δ18O or δ13C composition of the drip water, the isotopic composition of the calcite formed in equilibrium will be determined by its equilibrium fractionation factor. The oxygen-‐isotope equilibrium fractionation factor between water and the precipitated calcite is temperature dependent and results in the preferential incorporation of 18O in the solid phase. The equilibrium fractionation factor of the δ13C also leads to the preferential incorporation of 13C in the calcite, but is much less sensitive to temperature than in the case of oxygen. Different studies and authors have tried to determine the δ18O water-‐calcite fractionation factor (Table 1). The mostly used laboratory established fractionation factors remain the ones by O'Neil et al. (1969) later modified by Friedman and O'Neil (1977), the relationship of Kim and O'Neil (1997) later modified by Kim et al. (2007), the results of Tarutani et al. (1969) and Jimenez-‐Lopez et al. (2001) (Table 1). A second approach to determine fractionation factors is established by using theoretical models as the ones from Horita and Clayton (2007) and Chacko and Deines (2008). A third approach consists of using cave-‐monitoring data to make an average of the in-‐cave observed fractionation factors (Demeny et al., 2010; Tremaine et al., 2011). Tremaine et al. (2011) established such a 'cave calcite' relationship by doing a best fit through the data on a large number of modern caves at different latitudes, altitudes and temperatures. The different studies display different reactions for the fractionation factor and different temperature dependencies (Table 1). However, they only suggest small differences for the enrichment values around 10°C and they all indicate a δ18O increase of about 0.2 ‰ in the deposited calcite per 1°C decrease of the depositional temperature. This is the value, used in this study.


25

Author Method 1000*lnα dα/dT (‰/°C)

O'Neil et al., 1969 modified by Friedman & O'Neil 1977

Laboratory 2.78(106T-‐2)-‐2.89 -‐0.24

Kim & O'Neil 1997 modified by Kim et al., 2007

Laboratory 18.03(103T-‐1)-‐31.17 -‐0.22

Chacko & Deines 2008 constructed relation by Tremaine et al., 2011

Theoretical calculation 2.573(106T-‐2)-‐0.869 -‐0.22

Horita & Clayton 2007

Calculations compared with experimental results

0.952(106T-‐2)+11.59(103T-‐1)-‐21.56 -‐0.23

Tremaine et al., 2011

Linear best-‐fit through large number of cave studies

(16.01±0.65)*(103T-‐1)-‐(24.6±2.2) -‐0.17

Demeny et al., 2010 Cave monitoring results Hungarian cave

17500*T-‐1-‐29.89 -‐0.22

Table 1. A selection of the most commonly used water-‐calcite oxygen fractionation factors. Laboratory and theoretical approaches differ from the relationships found in cave settings. The δ13C of calcite deposited in equilibrium with the δ13C signal of the dissolved inorganic carbon (DIC) in the water is determined by the δ13C of the DIC and the fractionation factor between the DIC and the calcite. Different studies have tried to determine an equilibrium fractionation factor for the δ13C between DIC and calcite. A selection is listed in Table 2. We used the results of Emrich et al. (1970) and Dulinski and Rozanski (1990) because experimental conditions are close to those in caves and because they are based on a compilation of data. In contrast to the temperature dependent oxygen fractionation factor, the carbon fractionation factor between the dissolved inorganic carbon (DIC) in the water and precipitated calcite is much less temperature dependent (ca. -‐0.06 ‰/1°C between 0 and 25 °C, Table 2).


26

Author Method ε dα/dT (‰/°C)

Emrich et al., 1970 Compilation of data 10.17 ± 0.18 (20°C) -‐0.063 ±0.008 (20°C)

Dulinski and Rozanski 1990 and references

Calcite precipitated from a carbonate solution by removal of CO2

10.14 (10°C) 9.47 (15°C) -‐0.07 (5-‐15°C)

Romanek et al, 1992

Calcite precipitation from a NaHCO3, CaCl2 NaCl solution by removal of CO2

11.98 (±0.13) -‐0.12(±0.01)*T(°C)

-‐0.12 ± 0.1 (10-‐40°C)

Table 2. Different values for the carbon fractionation factor between DIC and calcite reported in enrichment values (ε) at the experiment temperature with their temperature dependency. 2.3.4 Disequilibrium fractionation, new insights In a pioneering publication, Hendy (1971) examined the effect of different modes of calcite deposition on speleothem δ18O and δ13C signals and concluded that if the loss of CO2 from the solution is slow, the precipitated calcite will be in isotopic equilibrium with the solution. In this case, speleothem δ18O and δ13C values will only depend on the isotopic composition of the drip water and the corresponding temperature dependent fractionation factors. If however, degassing of CO2 is rapid, kinetic fractionation will occur and δ18O and δ13C values will show a simultaneous enrichment along individual growth layers. The ‘‘Hendy test” consists of analyzing δ18O and δ13C values in a minimum number of ~ 10 samples taken within a single growth layer and at increasing distance on both sides from the growth axis. If the δ18O and δ13C values follow a co-‐varying increasing trend, this observation is considered to be the result of kinetic fractionation during calcite deposition in the studied growth layer. Apart from the equilibrium and kinetic fractionation, the ‘Hendy’-‐theory also describes disequilibrium fractionation, being intermediate between kinetic fractionation and equilibrium fractionation (Hendy, 1971). A longer residence time of the water on the surface of the stalagmite leaves more time for CO2 degassing. More CO2 degassing will lead to a gradual enrichment in the δ13C signal of the water due to the preferential removal of light carbon isotopes. The δ18O signal, however, remains practically unaffected because the oxygen reservoir in the water film is several magnitudes (104) vaster than the carbon reservoir (DIC), and the CO2 escape is faster than the evaporation of H2O in the cave environment (Hendy, 1971).


27

This disequilibrium theory has later been proved incorrect for the evolution of the δ18O of the DIC in the water film. A recent study by Beck et al. (2005) showed that the commonly used exchange time between the oxygen isotopes of the water and the oxygen isotopes of the DIC was too low by one order of magnitude. Dreybrodt (2008) and Dreybrodt and Scholz (2011) used this new longer exchange time to model the disequilibrium effects on the δ18O and δ13C signals in the water film during calcite deposition and delivered important new insights. The new used time constants by Dreybrodt and Scholz (2011) and Dreybrodt (2011) were later confirmed in experimental setups studying the chemical and isotopic processes in the water covering the stalagmite (Hansen et al., 2013). The new insights are:

1) There is no such thing as ‘fast’ or ‘enhanced’ CO2-‐degassing. Calcite deposition from water that enters the cave can be divided into three steps (Fig. 5) (Dreybrodt and Scholz, 2011; Hansen et al., 2013). The first step is the CO2-‐outgassing of the solution by molecular diffusion that occurs immediately (mostly within 10s) after the drip water (at pH ≈ 7) enters the cave (Hansen et al., 2013). In a second step, the solution equilibrates with the low pCO2 and the pH increases to a value around 8. This step is one order of magnitude longer compared to the CO2 outgassing and takes several 100s (Zeebe et al., 1999). In a third step, CaCO3 precipitates from the solution. This step is the longest and is in the order of several 1000s. Precipitation of calcite from the solution during the third step converts HCO3-‐ into CO2 that degasses from the solution. CO2 degassing from on the water film covering the stalagmite is caused by precipitation of the calcite (Dreybrodt and Scholz, 2011; Hansen et al., 2013). To conclude, first CO2 outgassing occurs due to a pressure difference between the drip water and the cave. The oversaturated solution starts to deposit calcite to restore the chemical equilibrium. The precipitation of calcite from the solution further converts HCO3-‐ into CO2. This produced “new” CO2 will then degas from the solution. Degassing is thus caused by the precipitation of calcite. A longer residence time of the solution on the surface of the stalagmite allows more calcite to be deposited and thus a larger production of CO2. It is the degassing of this newly produced CO2 that gradually increases the δ13C signal of the DIC in the solution.


28

Figure 5. Evolution of the concentration of the [CO2] and the [Ca] in the solution together with the pH and timing of the three different steps from the entrance of the drip in the cave atmosphere to the deposition of calcite. (Hansen et al., 2013, adapted by Dreybrodt, presentation at the S4 Speleothem Summer School, Heidelberg, 2013).

2) After CO2 degassing, the δ18O of the DIC does not have the time to re-‐equilibrate isotopically with the δ18O in the water before calcite is deposited.

When the drip enters the cave, degassing, driven by the pCO2 difference between the drip water and the cave air, causes preferential removal of both the light 12C and 16O isotopes from the solution. This occurs within seconds (Dreybrodt and Scholz, 2011; Hansen et al., 2013). Removal of light CO2 molecules causes the δ13C value of the bicarbonate in the solution to increase. The increase can be described as a Rayleigh distillation process (Salomons and Mook, 1986) or as a more recently developed model referred as the ‘kinetic approach’ (Dreybrodt, 2008). The removal of light 16O will also cause the δ18O of the DIC in the solution to increase. However, buffering effects due to the isotopic oxygen exchange between the water in the solution and the bicarbonate in the solution (hydration and hydroxylation) will restore the δ18O signal of the HCO3-‐ to its initial composition. This process of re-‐equilibration in the thin water film on the surface of the stalagmite is longer as first expected and takes between 6 000 and


29

65 000 seconds, depending on the temperature (Dreybrodt and Scholz, 2011). Full equilibrium between the δ18O in the water and the δ18O of the DIC is reached after 4 times the exchange time (general rule handled by Dreybrodt and Scholz, 2011). Restoring the oxygen isotope equilibrium between the DIC and the H2O in the solution is consequently a long process. Calcite deposition, starts about ± 1000 s after degassing, being long before the equilibrium can be reached. CO2 degassing will thus always cause increased δ18O values (Dreybrodt and Scholz, 2011).

3) The δ18O signal of the rainwater will only be reflected in the calcite if the residence time of the water in the epikarst is long (i.e several days).

Speleothems reflect the δ18O composition of the DIC in the water from which they are precipitated. Because the amount of water molecules is much higher (105 times) compared to the amount of DIC molecules, the δ18O signal of the DIC was commonly expected to have equilibrated with the water δ18O signal (Scholz et al., 2009). However, at a temperature of 10°C, equilibrium between the water and the DIC is only established after ± 1.6 days (Dreybrodt and Scholz, 2011). Consequently, the δ18O of the DIC will only reflect the rainwater δ18O if the residence time in the epikarst is sufficiently long (i.e., several days) to equilibrate the isotopes between the water and the dissolved carbon species (95 ‰ HCO3-‐ at pH = 8) (Dreybrodt and Scholz, 2011). If equilibration is not completed, the δ18O of the DIC and consequently the speleothem, will not reflect the δ18O of the water (Dreybrodt and Scholz, 2011). This new insight will also have implications for the process of Prior Calcite Precipitation (see section 2.3.6).

4) H2O evaporation of the water film covering the stalagmite needs to be significantly high before having an effect on the δ18O of the deposited calcite

Evaporation of the thin water film covering the surface of the stalagmite tends to be high when relative humidity is low, the drip water has a long residence time on the surface of the stalagmite or stalactite tip and/or the cave is windy and well ventilated. After evaporation, the isotopic composition in the water film covering the stalagmite is enriched in 18O. The δ18O value of the DIC, mainly consisting of HCO3-‐ at pH ≈ 8, remains initially unaffected by evaporation. However, since 105 times more water molecules are present in the solution, HCO3-‐ that is in disequilibrium with the water will start to re-‐equilibrate its oxygen isotopic composition through hydration and hydroxylation processes. However, these equilibration processes are slow and can take up to several days (Beck et al., 2005; Dreybrodt and Scholz, 2011). Since precipitation occurs within


30

±1000s after CO2 degassing, the HCO3-‐ that determines the isotopic signal of the deposited calcite will not have the time to equilibrate with the light water isotopes of the H2O. The formed calcite thus reflects the isotopic composition of the water before evaporation. However, under certain boundary conditions evaporation processes may cause increased δ18O values in the formed calcite (Deininger et al., 2012). If the relative humidity inside the cave is below 85 % and the wind velocity above the solution is higher than 0.2 m/s, evaporation influences the δ18O. Since most caves have a relative humidity above 95 %, evaporative effects might be of subordinated order (Deininger et al., 2012). 2.3.5 Disequilibrium fractionation, effects and tests Dependence of calcite δ18O and δ13C on drip interval Both the δ18O and δ13C signals are similarly affected by the drip interval. Under a slower drip, the residence time of the water in the thin film on the surface of the stalagmite is longer and calcite deposition has more time. The CO2 produced during the calcite deposition will degas from the solution and increase both the δ18O and δ13C in the solution. The δ18O and δ13C of the deposited calcite will therefore increase with longer drip intervals. Dependence of calcite δ18O and δ13C on the pCO2 gradient The rate of CO2 degassing is a function of the CO2 gradient between the drip water, which is largely determined by the degree of biological respiration in the soil, and the cave atmosphere, which is controlled by (seasonal) cave ventilation (Mühlinghaus et al., 2007; Baldini et al., 2008; 2009; Scholz et al., 2009). Large pCO2 gradients between drip water and cave atmosphere, increase the amount of CO2 degassing from the solution. However, the time constant for degassing of excess CO2 from the solution does not depend on the difference between the pCO2 of the solution and cave air (Dreybrodt and Scholz, 2011). Degassing is always fast and occurs within ± 10s after entering the cave atmosphere (Hansen et al., 2013). More CO2 degassing causes the δ18O and also the δ13C composition of the DIC to increase, as lighter CO2 is preferentially lost from the solution. Calcite deposition onsets before CO2 hydration and hydroxylation processes can re-‐equilibrate the decreased δ18O value of the HCO3-‐ in the solution. CO2 degassing has a similar effect on the δ18O and δ13C signals. Stronger degassing, due to a larger pCO2 gradient, will cause heavier δ18O and δ13C signals in the solution and consequently in the deposited calcite. Drip intervals approaching zero (i.e. very fast drip rates) represent an exception by showing no relation between the calcite δ18O and δ13C and drip interval or pCO2 gradient variations. The isotope ratio of the solution on the surface of the stalagmite is close to the isotope ratio of the impinging drops (Mühlinghaus et al.,


31

2009). Calcite deposited under such dip sites represents fractionation under conditions of isotopic equilibrium (Mühlinghaus et al., 2009). In such cases the δ18O and δ13C values of the precipitated calcite are determined by their isotope fractionation factor (which is temperature dependent for the δ18O). Different tests can be carried out to investigate the presence of disequilibrium processes. A Hendy-‐test, where several samples are drilled along a single growth layer, may indicate disequilibrium processes. Better than the Hendy-‐test is to calculate the expected δ18O and δ13C values based on the theoretical “equilibrium” water-‐calcite fractionation factors for C and O (Kim and O'Neil, 1997), the present-‐day cave temperature and the isotopic compositions of seepage waters and to compare these results with the measured δ18O and δ13C values. However, the problem with this test is that cave conditions may have changed in the past. Present-‐day equilibrium deposition of the calcite in the cave is certainly not a guarantee for a similar condition in the past. Speleothems growing in most natural cave environments often display δ18O and δ13C values that are higher than predicted by calculations starting from the drip water δ18O and δ13CDIC composition (Mickler et al., 2006; Tremaine et al., 2011). Only a number of cave environments such as the fast growing Proserpine stalagmite at the Han-‐sur-‐Lesse cave (Van Rampelbergh et al., 2014; Van Rampelbergh et al., in review), drip sites in Bunker cave in western Germany (Riechelmann et al., 2013), Soreq Cave in Israel (BarMatthews et al., 1996) or in Scotland (Fuller et al., 2008) have been shown to deposit speleothem calcite according to the equilibrium fractionation rules. The difficult match between the calculated values and the values measured in natural cave environments result in a large debate on whether speleothems can reflect isotopic equilibrium. The used fractionation factors to estimate if calcite is deposited in isotopic equilibrium are mostly determined by artificial (laboratory or theoretical) setups and may not be absolutely similar in natural cave environments. Tremaine et al. (2009) estimated a ‘speleothem δ18O fractionation factor’ based on a large collection of speleothem data. Their suggested fractionation factor is smaller than the fractionation factors derived from theoretical and laboratory results (such as O’ Neil et al. 1969 or Horita & Clayton, 2007) indicating that the theoretical and mathematical results tend to overestimate the natural conditions. A larger amount of detailed cave monitoring studies is needed to provide better insights in the water-‐calcite fractionation factors for natural cave environments. Other than the discussion on the fractionation factor, the different processes and timing of these processes affecting the drip water DIC, which largely determine the speleothem isotopic composition, are still not fully understood. Recent findings on the longer isotopic exchange times between HCO3-‐ and H2O and their relation with speleothem calcite formation (Dreybrodt and Scholz, 2011;


32

Deininger et al., 2012; Hansen et al., 2013) brought new insights in the correct impact of disequilibrium processes caused by longer residence times of the water on the speleothem surface or pCO2 gradients. More detailed studies on the isotopic processes in the solution layer from which calcite is formed and on the timing of these processes will provide better insights how disequilibrium processes may affect the isotopic composition of the drip water DIC and consequently the deposited calcite. 2.3.6 Prior Calcite Precipitation (= PCP) Prior calcite precipitation (PCP) is a common process occurring in karst aquifers (Fairchild et al., 2000; Fairchild et al., 2006; Verheyden et al., 2008b; Riechelmann et al., 2011; Tremaine and Froelich, 2013; Rutlidge et al., 2014). When downward percolating water encounters a zone with lower pCO2, degassing occurs and calcite can precipitate within the epikarst before reaching the cave. During drier periods, PCP is enhanced as aerated zones increase in the aquifer and residence time of the water becomes longer (Fairchild et al., 2000). The CO2-‐degassing from the solution within the epikarst will cause the isotopic composition of the HCO3-‐ in the solution to increase due to the removal of light oxygen and carbon isotopes. However, buffering effects of hydration and hydroxylation will re-‐equilibrate the increased δ18O of the DIC in the solution with the unaffected δ18O of the water. This re-‐equilibration process can take up to several days (Dreybrodt and Scholz, 2011). Only water that stayed in the epikarst for several days after being affected by PCP will display δ18O values that are in equilibrium with the initial rainwater signal. PCP will thus always cause increased δ13C values of the DIC in the drip water while the effect on the δ18O depends on the residence time of the water after PCP. If the residence time after PCP is longer than several days, the drip water δ13C will increase while the δ18O signal represents the initial water composition before PCP. Comparing the isotopic signal of soil and drip water can indicate the occurrence of PCP. Seasonal variations in PCP can cause seasonal variations in the δ13C and under short residence times of the water also in the δ18O signals of the drip water and the formed speleothem (Van Rampelbergh et al., 2014).


33

The most visible expression of PCP is the formation of a stalactite above a stalagmite. During this form of PCP, the drip water will first deposit stalactite calcite before forming a stalagmite. Deposition of stalactite calcite increases both the δ18O and δ13C values of the DIC in the remaining solution. This isotopically heavier solution will fall on the top of the stalagmite and start to deposit stalagmite calcite. The formation of the stalagmite calcite will start long before the increased δ18O of the DIC in the solution can re-‐equilibrate with the large reservoir of unaffected (=lighter) δ18O of the H2O in the solution (Dreybrodt and Scholz, 2011). During stalactite-‐PCP, both the δ13C and the δ18O value of the DIC in the solution and consequently the deposited stalagmite calcite are increased compared to the composition of the vadose water. PCP can thus be detected by increased δ13C signals and unaffected δ18O for long residence times or by both increased δ18O and δ13C signals under short residence times of the water in the epikarst. However, since multiple other processes than PCP can cause increased δ13C (and δ18O), combination of stable isotope records with trace elemental records (mostly Mg and Sr, but also Ba can be used) is advised to indicate the process of PCP. Co-‐varying increased δ13C and trace elemental ratios with distribution coefficients smaller than 1 (e.g. Mg and Sr) are the mostly used tool to identify PCP (see 2.4.1). 2.3.7 Summary δ18O and δ13C signals in speleothems As discussed in the previous sections, numerous effects, which act independently or influence each other in a positive or negative manner, determine the final δ18O and δ13C composition of the deposited speleothem calcite. No two caves are similar; therefore interpreting the measured speleothem δ18O and δ13C signals requires a deep understanding of the studied cave system to link the obtained data to climate parameters. Furthermore, the higher the temporal resolution of the δ18O and δ13C time series, the more complex the interactions become. The required knowledge of the cave system is obtained by cave monitoring programs, together with calcite production experiments.


34

Figure 6. Oxygen isotopes in speleothems Figure 6 summarizes the different factors determining the speleothem δ18O signal. For semi-‐arid tropical regions such as Socotra Island (see Chapter 3), the rainwater δ18O is driven principally by the amount effect. Speleothems calcite growing in such regions, -‐ if not disturbed by other factors such as for example disequilibrium effects or mixing of water in the epikarst -‐, is expected to reflect changes in wetter versus drier conditions (Fairchild and Baker, 2012). In semi-‐arid tropical regions, speleothems deposited in equilibrium with their drip water should roughly vary around an average of -‐4 ‰. In temperate regions, such as in Belgium (Chapter 4 and 5), rainwater δ18O averages -‐7 ‰ and is influenced both by the temperature and by the amount effect. In such climate regions speleothem calcite deposited in equilibrium with the drip water at an average cave temperature of 12°C is expected to be ∼ 1 ‰ heavier compared to the drip water δ18O. Belgian speleothems are thus expected to display values around -‐6 ‰.


35

Figure 7. Carbon isotopes in speleothems Figure 7 summarizes the different factors to be considered when reconstructing climate based on δ13C signals in speleothems. The carbon fractionation factor is only slightly temperature dependent and provided that deposition occurs in isotopic equilibrium, speleothem calcite mainly reflect the δ13C of soil CO2, which is controlled by the vegetation above the cave. For dominant C3 vegetation (δ13C between -‐32 ‰ and -‐25 ‰), soil CO2 has a δ13C value between -‐28 ‰ and -‐21 ‰. Considering the enrichment factor between CO2 in the soil and DIC in the water (95% HCO3-‐ at pH≈ 7) of ∼ 10 ‰ and the enrichment factor between DIC in the water and the CaCO3 in the speleothem of 0.2 ‰ to 0.9 ‰ (for 10 °C and 25 °C resp.), speleothem calcite -‐ without bedrock carbon contamination-‐ displays δ13C values between -‐18 ‰ and -‐11 ‰. For C4 vegetation (δ13C between -‐14‰ and -‐10 ‰), speleothem calcite displays a δ13C of between 0 ‰ and +4 ‰. Dissolution of bedrock (δ13C ≈ 0 ‰) increases the epikarst water δ13CDIC composition and consequently the speleothem δ13C. For speleothems growing under a C3-‐vegetation covered soil and with a 15 % bedrock input, the speleothem δ13C will vary between -‐15 ‰ (-‐18 ‰ * 85% + 0‰*15%) and -‐9 ‰. Under a C4-‐vegetation covered soil and a 15 % bedrock input, the speleothem δ13C will vary between 0 ‰ and 5 ‰.


36

2.4 Mg and Sr in speleothems Trace elements represent a large portion of the measurable proxies in speleothems and have increasingly provided useful information on speleothem based paleoclimate reconstructions. They can be used either in chronology, where trace elements vary rhythmically (Smith et al., 2009) or as paleoclimate proxy (Borsato et al., 2007; Fairchild and Treble, 2009; Tremaine and Froelich, 2013; Van Rampelbergh et al., 2013; Orland et al., 2014; Rutlidge et al., 2014; Verheyden et al., 2014). Most studies of speleothems focus on elements that form divalent cations in solution and which substitute for Ca in the carbonate crystal lattice, particularly Mg and Sr (Fairchild and Treble, 2009). In this study, we focused on the two trace elemental ratios of Mg/Ca and Sr/Ca. For these species, the Mg and Sr content measured in the stalagmite depend on their ratio in the precipitating solution and their distribution coefficient. 2.4.1 The Mg and Sr distribution coefficients As a mineral grows from an aqueous solution and if equilibrium is maintained, trace elements will partition between the two phases in a characteristic manner. The fundamental law that controls the distribution of a trace element between coexisting phases is usually referred as the Berthelot-‐Nernst distribution law (McIntire, 1963). According to this law and at equilibrium, the ratio of the concentration of the trace component in the solid to its concentration in the liquid is a constant. A more convenient form of the distribution law refers to the ratio of the trace element (Me) to the major element (Ca in the case of calcite): D (distribution coefficient) = (cMe/Ca)solid / (cMe/Ca)solution with cMe/Ca being the concentration ratios of the trace element (in this study Sr and Mg) to calcium on either a weight or molar basis. An overview of the average distribution coefficients for Mg and Sr between water and calcite derived from laboratory experiments or cave observations are listed in Table 3. Laboratory-‐based inorganic calcite precipitation experiments similar to natural cave precipitates have been performed under well-‐controlled solution chemistry conditions to determine trace element distribution into calcite (Katz et al., 1972; Pingitore and Eastman, 1986; Huang et al., 2001), and others). Huang and Fairchild (2001) examined trace elements under variable but controlled [Ca2+], [HCO3-‐], and pCO2 conditions that closely mimicked natural


37

cave systematics. Distribution coefficients determined by modern natural speleothem calcite farmed under active drips in natural cave systems are starting to increase (Holland et al., 1964; Gascoyne, 1983; Huang et al., 2001; Stern et al., 2005; Tremaine and Froelich, 2013) but often display different values (Tremaine and Froelich, 2013). Generally, large differences between the distribution coefficients of the different used approaches and cave sites are observed. All results clearly show partition coefficients for water-‐calcite solutions smaller than 1. This implies that the Mg/Ca and Sr/Ca ratios of the calcite are smaller than those of the precipitating solution. Inversely, when a solution precipitates calcite, its Mg/Ca and Sr/Ca ratios increase (this is of major importance for the process of PCP discussed in 2.3.6). A number of laboratory and cave studies have demonstrated a temperature dependence on DMg, which can be of interest for paleo-‐temperature reconstructions (Katz, 1973; Gascoyne, 1983; Huang et al., 2001). At higher temperatures, more Mg is incorporated in the calcite. Gascoyne (1983, 1992) suggested that, if the Mg/Ca of the precipitating solution remained stable, Mg/Ca ratios could be used a temperature proxy. Huang and Fairchild (2001) performed an experimental investigation in cave like setups to try to determine the temperature dependence of the DMg value. They obtained a rather small temperature dependence of about 0.006/°C in the range of 7 to 15 °C. However, up until now no speleothem Mg/Ca ratios have successfully been able to reconstruct temperature variations. The reason is that cave temperature variations are often too small to induce significant (= detectable) Mg/Ca variations in stalagmites. Indeed, the temperature signals are typically masked by noise from other factors with much larger variation, such as changes in the Mg/Ca–ratio of the precipitation solutions. The use of Mg/Ca in speleothem studies has consequently largely shifted from temperature reconstructions to interpreting variations in terms of hydrological changes (Fairchild and Treble, 2009).


38

Table 3. Average distribution coefficients for Mg (DMg) and Sr (DSr) from the literature for inorganic calcites determine by laboratory experiments and cave observation results (adapted from Tremaine et al., 2013 and references herein). Laboratory and speleothems studies indicate a growth rate dependency of the DSr (Pingitore and Eastman, 1986; Huang et al., 2001; Treble et al., 2003; 2005b; Gabitov and Watson, 2006). However, the growth rate needs to be higher than 0.5 mm/y to affect the DSr (Gabitov and Watson, 2006). The fact that DSr is only influenced for growth rates higher than 0.5 mm/y relates to the fact that Sr2+ can be incorporated in the lattice sites to replace Ca2+ but also in crystal defect site (non-‐lattice sites) (Pingitore and Eastmann, (1986)). Such non-‐lattice sites


39

become more frequent under high growth rates. However, increased Sr-‐concentrations have been observed together with increased speleothem growth rates that were smaller than 0.5 mm/y (Frisia et al., 2003; Treble et al., 2003). Fairchild and Treble (2009) suggested an alternative, and better suited theory for the relation between Sr-‐concentration and growth rates, which explains why Sr-‐concentration can covary with growth rates lower than the previously suggested 0.5 mm/year limit by Gabitov and Watson (2006). Fairchild and Treble (2009) suggest that the good correspondence between Sr-‐concentrations (and Ba, Na concentrations) and growth rate is related to growth kinetics rather than to growth rate. Growth kinetics are better suited to refer to changes in crystal morphology and/or growth mechanisms, usually associated with growth rate changes. Increased kinetics cause augmentation of crystal defects and leads to more Sr2+ being incorporated in the calcite. However, growth kinetics is not the only factor that determines the Sr-‐concentration in the calcite. The presence of other competing elements in the water such as Ba and Na can lower the amount of Sr that is introduced in the crystal system since these elements also occupy the non-‐lattice sites. To determine which factor causes the Sr-‐concentration variations, Sr needs to be measured together with the ‘competing’ elements Ba and Na. 2.4.2 The Mg/Ca and Sr/Ca ratio in the precipitating solution Trace elements in speleothems can originate from a variety of sources such as aeolian particles, dry and wet atmospheric deposition, bedrock, superficial sediment deposits and inorganic soil constituents, and elements recycled via soil biota (Fairchild and Treble, 2009). Atmosphere is only a subordinate source for trace elements. However, Sr-‐isotope signals have been related to the supply of aeolian dust in speleothems (Goede et al., 1998; Bar-‐Matthews et al., 1999; Zhou et al., 2009; Hori et al., 2013). Sea-‐salt aerosols in wet atmospheric deposition can be an important source for certain trace species in karst waters, sometimes including Mg and Sr where bedrock supply is limited (Bar-‐Matthews et al., 1999; Fairchild et al., 2000). In arid environments a large input of soil derived Mg from aeolian input or evaporational salts formed during the dry season can also form a major Mg source in the drip water (Rutlidge et al., 2014). Despite exceptions, the primary source of calcium Mg and Sr is the carbonate bedrock (calcite, dolomite) and overlying soil, including bedrock fragments in the soil (Fairchild and Treble, 2009). However, exceptions occur such as in NW Scotland where the high Sr-‐concentrations in calcite speleothems arise from localized veins associated with a thin igneous sill in the overlying Sr-‐poor


40

dolomite (Roberts et al., 1998). When clay minerals are present, Sr is usually supplied from both carbonates and silicates. In the case of a calcite and dolomite composed epikarst, Mg/Ca and Sr/Ca ratios will arise from their dissolution. The slow dissolution of dolomite (main source of Mg) will cause the water Mg/Ca ratio to be smaller than the bulk ratio in the bedrock. Raised Mg/Ca ratios compared with bulk carbonate were first thought to be linked to incongruent dissolution (Fairchild et al., 2000). A higher residence time of the water in the epikarst due to slow water flow or drier conditions were suggested to lead to higher Mg/Ca ratios by enhanced dolomite dissolution (Roberts et al., 1998; Fairchild et al., 2000; Tooth and Fairchild, 2003; Musgrove and Banner, 2004). However, this mechanism requires dissolution of dolomite into solutions already saturated for calcite. The process of measurably enhancing Mg contents is very slow and is likely to require long residence times of months to years. Fairchild et al. (2006) suggested a more straightforward hydrological routing model for which there is strong evidence in Mesozoic aquifers. In this model, the residence time effect is related to an enhanced contribution of Mg rich waters from low-‐permeability parts of the aquifer at low flow (Fairchild et al., 2006). Enhanced Mg and/or Sr enrichment during annual low flows can be related to higher contents of dolomite (for Mg) or aragonite (for Sr) associated with clay (Fairchild et al., 2006). High Mg/Ca and Sr/Ca ratios can arise from preferential leaching form newly created calcite surfaces (McGillen and Fairchild, 2005). Selective leaching is used for the preferential removal of Mg and/or Sr from an homogenous bedrock phase (Tremaine and Froelich, 2013). However, selective leaching of calcite does not occur in nature in older limestone beds (Palmer and Edmond, 1992) indicating that ‘recent’ processes must create new leaching surfaces. Winter freezing creating fresh surfaces can generate such new leaching surfaces and is probably the case for the variations in the Ernesto Cave (Fairchild et al., 2000). The process of selective leaching can also arise where storage occurs during the dry season as sulphate or chloride salts, where there is fresh supply of clay minerals with other exchangeable ions, or where other more soluble phases such as aragonite are present (Fairchild et al., 2006). Rutlidge et al. (2014) illustrated the importance of such salt depositions in the soil during the dry season in arid environments. They pointed out that the Mg-‐input by soil salts can be as large as Mg-‐input by bedrock dissolution. The process mostly used to interpret increased Mg/Ca and Sr/Ca values is Prior Calcite Precipitation or PCP (see 2.3.6). This process was first suggested by Holland et al. (1964) and further illustrated in many later studies to be the cause of increased Mg/Ca and Sr/Ca ratios (Tooth and Fairchild, 2003; McDonald et al., 2004; Fairchild et al., 2006; Karmann et al., 2007; Tremaine and Froelich, 2013;


41

Van Rampelbergh et al., 2013; Rutlidge et al., 2014). The effect of calcite precipitation is to remove cations from the water in the proportion in which they are incorporated into calcite (Holland et al 1964). PCP causes trace elements with partition coefficients <1, such as those for Mg, Sr and Ba to become enriched in the solution compared to calcite. Trace elements with partition coefficients >1, such as is the case for Zn, will become depleted in the solution compared to calcite (Fairchild and Treble, 2009). As discussed in 2.3.6, PCP most often causes increased drip water δ13C values under long residence times of the water in the epikarst. The main evidence for PCP is the co-‐variation of all three Mg/Ca, Sr/Ca and δ13C in the record (McMillan et al., 2005; Johnson et al., 2006). However, in monsoon regions (such as Socotra), δ18O signals reflecting the amount effect can also co-‐vary with increased Mg/Ca and Sr/Ca ratios (and δ13C) (Cruz et al., 2007; Van Rampelbergh et al., 2013). During periods of increased rainfall, the δ18O value of the drip water lowers and more water is pushed through the epikarst reducing the occurrence of PCP. Consequently lower Mg/Ca and Sr/Ca (and δ13C) correlating with lower δ18O values indicate periods of higher water excess. Tremaine et al. (2013) indicated that co-‐varying Mg/Ca and Sr/Ca ratios are directly linked to water excess (precipitation minus evapotranspiration); higher net precipitation causes PCP to decrease. The Mg/Ca and Sr/Ca ratios in combination with δ18O signals can also be used to determine changed rainfall sources (Cruz et al., 2007; Tremaine and Froelich, 2013). A change in rainfall source (source effect) induces a change in δ18O signal but no changes in the Mg/Ca and Sr/Ca ratios. Large shifts in δ18O values not visible in trace elemental ratios are consequently suggesting source effects on rainfall δ18O values (Tremaine and Froelich, 2013). However, given the potential multitude of factors affecting Mg and Sr concentrations including aeolian input, selective leaching, hydrological routing causing mixing with other fluids or bedrock with different Mg and Sr concentration and prior calcite precipitation, caution should be exercised when interpreting Mg/Ca and Sr/Ca ratios in terms of climate. This is particularly the case when no long-‐term drip water studies have been carried out at the cave site or when the results of such studies may not be relevant due to strongly different climate regimes between the present day and past times (Fairchild and Treble, 2009). Best speleothem samples for using trace elemental ratios to reconstruct climate parameters are stalagmites growing under seepage flow systems (Fairchild and Treble, 2009) where mixing of water with different ages is not expected to overshadow the signal of effective rainfall and PCP. By contrast, drip water through crystalline, fractured limestones are likely to be dominated by


42

hydrological routing (Fairchild and Treble, 2009). Epikarst thickness may also play a role for the rate at which climate signals are brought to the cave by trace elements. This effect is similar to the effects on stable isotopes (last paragraph of 2.3.1). Recent studies (Galy et al., 2002; Buhl et al., 2007) have indicated that Mg-‐isotopes (δ26Mg) could be a tool for understanding whether source effects, hydrologic routing or residence time are dominating at a particular site. Sr-‐isotopes (87Sr/86Sr) can be used to determine the source of Sr not originating from the epikarst. Bar-‐Matthews et al. (1999) interpreted higher 87Sr/86Sr in a speleothem record from Soreq cave (Israel) to reflect increased input of sea spray droplets and dust particles through the last glacial period. More recent studies have also indicated the importance of 87Sr/86Sr ratios in speleothems for paleoclimate reconstructions (Zhou et al., 2009; Hori et al., 2013). 2.5 Climate reconstructions from speleothems 2.5.1 Quantitative paleotemperature estimates An early goal of speleothems was to reconstruct absolute changes in mean annual temperature based on the temperature dependency of the water-‐calcite oxygen fractionation factor (Hendy and Wilson, 1968; Thompson et al., 1974). However, few reliable estimates have been published because of the considerable complexity of δ18O in the atmosphere, hydrosphere, and cave environment. To estimate quantitatively past temperatures from speleothem δ18O data, the δ18O value of the drip water and the calcite at time of calcite formation are required. Since only the δ18O of the calcite can be measured, estimating the calcite temperature formation remains difficult. Different techniques have been and are still being developed to reconstruct the paleo-‐dripwater δ18O signal either by measuring old groundwater with the same age or better by measuring the δ18O value of water entrapped in fluid inclusions in the speleothem calcite. The δ18O value of the drip water can be determined by groundwater of known age and isotopic composition from near the studied cave. For example, a stalagmite from South Africa and the δ18O values from a dated water aquifer were used to determine the relative temperature difference between the late glacial and today (Talma and Vogel, 1992). Fluid inclusions are considered a better technique to estimate paleotemperatures (Matthews et al., 2000; Dennis et al., 2001; Zhang et al., 2008). Although fluid inclusions seemed promising, difficulties associated with the measuring technique impede their application. A first problem is that the δ18O of the water in the fluid inclusion may be expected to have equilibrated with the δ18O value of the calcite. However, δD values, cannot be affected by this process and are used to estimate the original δ18O


43

values of the water using the meteoric water line. Analytical difficulties in measuring the δD values mainly due to fractionation have been reported (Matthews et al., 2000). Also, the meteoric water line of today, used to calculate the δ18O, may be different from the one in the past, which may result in significant errors in deduced paleotemperatures. Different studies have focused on optimizing the measuring techniques (Vonhof et al., 2006; Verheyden et al., 2008a; De Cisneros et al., 2011). Recent developments allow to measure water isotopes by cavity ring down spectroscopy, which leads to better δ18O and δD measurements in fluid inclusions (Arienzo et al., 2013) and seems promising for future work on the speleothems. Estimation of past drip water δ18O with better confidence has been done using noble gases in fluid inclusion (Kluge et al., 2008; Scheidegger et al., 2010). The solubility of noble gases in water depends on the temperature of the water. By measuring noble gases content in speleothem fluid inclusions, the paleotemperature of the calcite formation can be determined. However, a major problem with this technique is the often high contribution of noble gases from air inclusions that mask the temperature information present in the noble gases dissolved in water-‐filled inclusions (Kluge et al., 2008). The clumped isotope technique seems the most promising in speleothem based paleotemperature reconstructions (Eiler, 2007). It is based on the temperature dependent preference for heavy nuclides to bond to each other, rather than to a light isotope. For carbonates, the clumped isotope paleo-‐thermometer is based on the abundance of 13C-‐18O bonds in the carbonate lattice relative to that expected at a random distribution of isotopes among all isotopologues. The CO2-‐mass 47, which is dominantly 13C18O16O, is used as parameter to determine the abundance of the 13C-‐18O bonds. By using a calibration curve (Kim and O'Neil, 1997), the measured Delta(47) provides a paleo-‐temperature estimation for the formation of calcite. However, results form clumped isotope measurements on speleothems have shown that kinetic effects during calcite formation (mainly driven by CO2 degassing) cause large offsets in the Delta(47) (Affek et al., 2008; Daeron et al., 2011). Affek et al. (2008) found that a modern speleothem from Soreq Cave (Israel) yielded formation temperatures 8°C too high, which they attributed to kinetic effects. However, by assuming that this offset is constant for older samples from the same cave, they arrived at a set of results that are largely consistent with previous independent assessments. Although this is a good results for a first-‐of-‐a-‐kind study, a much more stringent test for the degree of kinetic modification is needed for this method to become routine.


44

2.5.2 Semi-‐empirical climate relationships Because of the difficulties associated with quantitative paleotemperature reconstructions from speleothem δ18O signals, alternative approaches have tried to estimate paleotemperature variations in a semi-‐empirical way. Such paleotemperature estimations are based on the combined relationship between (i) the temperature and the δ18O value of rainwater (dδ18Op/dT) (δ18Op being the δ18O of the meteoric precipitation) and (ii) the effect of cave temperature on the equilibrium fractionation associated with calcite precipitation (dδ18Oct/dT) (δ18Oc being the δ18O of the speleothem calcite). The dδ18Op/dT can vary between +0.17 and +0.9 ‰/1°C, depending on the geographical location (Dansgaard, 1964; Rozanski et al., 1992; Mook, 2000; Schmidt et al., 2007). Schmidt et al. (2007) suggested a dependence of +0.3 ‰/1°C for central Europe. The dδ18Oct/dT has to be estimated for the studied speleothem and can vary between -‐0.18‰/°C to -‐0.25‰/°C (Lachniet, 2009). Depending on the value of both factors the net effect can be a positive or a negative link between the speleothem δ18O and the external temperature. Lauritzen an Lundberg (1999) calculated the dδ18Op/dT of a Norwegian speleothem using independently derived estimated mean annual temperatures of today and of the Little Ice Age (1300-‐1800 AD). Another successful temperature reconstruction has been carried out by Mangini et al. (2005) using a speleothem form the Austrian Alps. They suggested a net effect of -‐0.22‰/°C between the speleothem δ18O and the external temperature. However, such estimations remain difficult and fragile since the dδ18Op/dT may vary over time. Also, other effects (such as source effect, amount effect or disequilibrium effects, discussed in section 2.3) can cause variations in the speleothem δ18O that are not related to temperature. A similar approach can be carried out to reconstruct changes in precipitation intensity from speleothem δ18O values that reflect the amount effect, and that are often growing in tropical or sub-‐tropical environments. To do this, the precipitation dependency of the δ18O (dδ18Op/dP) needs to be known (dP being the variation in amount of precipitation). However, only a few quantitative applications of a dδ18Op/dP relation to speleothem δ18O variations have been published such as for the eastern Mediterranean region (Bar-‐Matthews et al., 2003) or the Peruvian Amazonia (van Breukelen et al., 2008). Such calibrations have been rarely exploited for the very simple reason that the dδ18Op/dP relationship may not have been constant over time (Lachniet, 2009). Further, the smoothing effect in an aquifer due to mixing of infiltrated waters of varying ages would produce a dampened δ18O signal in a speleothem, and would underestimate true rainfall variation.


45

2.5.3 Quantification of the isotope effects in the meteoric water cycle. To date, little work has explored the spatial variations caused by the so-‐called isotope ‘effects’ on the δ18O of meteoric waters (such as latitude effects, altitude effect, continentality or source effect) as they are preserved in the speleothem δ18O. As speleothem records are becoming more widely used, such attempts are now feasible. For example, an altitude effect of -‐2 ‰/km in soda-‐straw stalactite δ18O (covering the last 100 years) was reconstructed from New Zealand (Williams et al., 1999), from stalagmites from China (Kong et al., 2005) and in the European Alps (McDermott et al., 1999). A latitude effect of 0.3‰/°latitude was found by Williams et al. (1999) for the late Holocene that is larger than the 0.27‰/°latitude for modern precipitation. In southern Brazil, a δ18O latitudinal gradient of 0.81‰/°latitude was calculated during the Holocene near the Atlantic coast (Cruz et al., 2006). A recent study of McDermott et al. (2011) provided a first evolution of the spatial and temporal gradients in δ18O during the Holocene in Europe (Fig. 8).

Figure 8. Map showing locations of cave sites used to investigate the gradients in δ18O of European Holocene speleothems. Contours are based on δ18O values for recently deposited (zero-‐age) calcite at these cave sites. The altitudinal effect on speleothem δ18O in the region surrounding the Alps is clearly visible (after McDermott et al. 2011).


46

The results indicate two major trends. Trend 1 (Fig. 8) shows decreasing speleothem δ18O values from the oceanic western European margins (SW-‐Ireland) to continental Europe reflecting a continental effect. Trend 2 (Fig. 8) indicates decreased speleothem δ18O values from Scotland northward to Scandinavia reflecting a latitudinal effect. During the Early Holocene speleothem δ18O values have show that trend 1 displayed a steeper gradient compared to today. The decreased δ18O values of speleothems from the Alps indicate an altitudinal effect. 2.5.4 Speleothems as tools to reconstruct continental climates Important insights in the evolution of past climates have been retrieved from ice cores and ocean drilling cores. Ice cores are limited to the Polar Regions and provide no information on climate variations closer to the equator. Ocean cores are geographically well distributed, but they are often homogenized and thus not able to reflect more regional climate variations, they do not allow absolute dating nor reconstructing recent climate variations at high resolution, such as for example during the last 1000 years. Although both very useful, ice cores and ocean cores do not allow reconstructing past climate variations in temperate, tropical and subtropical continental regions. Furthermore, reconstructing continental climates is largely challenging due to the numerous regional or even local effects. Climate variations on the continents form a large puzzle with different pieces that are connected and interact. To reconstruct such climate variations, well-‐spread and sensitive archives are necessary. Speleothems fulfill these conditions necessary to reconstruct the continental climate successfully. Furthermore they allow absolute dating of the time series up to the most recent 500 000 years. The best-‐known long-‐term speleothem δ18O time series comes from subaqeous vein-‐filling calcite at Devils Hole, a tectonic fracture in the Basin and Range region of southern Nevada, continuously covering the last 500 ka (Winograd et al., 1992) (Fig. 9). Surprisingly, the timing of the glacial to interglacial cycles suggested by the speleothem, differed significantly from that predicted by Milankovitch solar insolation variations (Winograd et al., 1992). In particular, most attention has focused on the differences in timing of Termination II, which is predated by some 12 ka compared to the timing in ice core or deep-‐sea records. Although the scientific debate on the significance and forcing of the Devils Hole record continues (Herbert et al., 2002; Winograd, 2002; Yang et al., 2005; Winograd et al., 2006), the time series has proven to capture long-‐term and well-‐dated paleoclimatic variations.


47

Large amplitude North Atlantic climate variations over the last 200 kyr such as the rapid warming associated with the Bølling Allerød (B-‐A), the cooling associated with the Younger Dryas (YD), millennial variability associated with Dansgaard-‐Oescher (D/O) events (1 to 2 kyr cyclicity) (Dansgaard et al., 1993) and iceberg discharge events known as Heinrich Events (Bond et al., 1992) are recorded in speleothems. Comparing speleothem time series with ice core and deep-‐sea records provides absolute ages on these events and indicate how these events are manifested on the continents in the different climate regions. Stalagmites from Hulu Cave (China), covering the last 160 kyr, located in the Asian Monsoon region were the first to demonstrate a millennia-‐scale monsoon response associated with the D/O events (Wang et al., 2001)(Fig. 9). The teleconnection between the North Atlantic and the Asian Monsoon implies that warmer periods in Greenland correspond with lower speleothem δ18O values, which relates to both the amount effect and the seasonal variations in winter versus summer rainfall. A similar effect on the monsoon rainfall is also recorded in the δ18O values of stalagmites collected from Moomi cave on Socotra Island, Yemen (Burns et al., 2003; Shakun et al., 2007) confirming the warm/wet and cold/dry teleconnection between the North Atlantic and the Asian Monsoon region. The impact of D/O events and the Bølling-‐Allerød is also recorded in the δ18O signals of speleothems from the European Alps (Spötl and Mangini, 2002; Spötl et al., 2002) and Lebanon (Bar-‐Matthews et al., 1997; Kaufman et al., 1998; 1999; 2000; Ayalon et al., 2002; 2003) and in both the δ18O and δ13C signals of western European speleothems (Genty et al., 2003) and Turkey (Fleitmann et al., 2009). The study of Genty et al. (2003) offers perhaps the best available chronological control on the timing and occurrence of such events, and provides an improved chronological framework for the GRIP and GISP ice-‐cores (McDermott, 2004). Particularly impressive is the well-‐dated composite δ18O record covering the past 185 ka based on 21 speleothems from Soreq Cave (Israel), which allowed to establish a link between the oceanic realm and continental climate in the Levant region. (Bar-‐Matthews et al., 1997; Kaufman et al., 1998; 1999; 2000; Ayalon et al., 2002; 2003) (Fig. 9). The Soreq δ18O record appears to reflect dominantly 2 effects: (i) changes in the δ18O of the oceanic vapor source and (ii) the amount effect. Increased δ18O values, are interpreted to reflect cooling forced by Heinrich events and during the last glacial maximum (Bar-‐Matthews et al., 1999), although some of the Heinrich events in the North Atlantic do not appear to have caused δ18O events in the Soreq Cave speleothems. The timing of low δ18O values coincided with regional wet periods over the past 180 ka during which the distinctive organic-‐rich sapropels were deposited in the Mediterranean Sea (Bar-‐Matthews et al., 2000; Ayalon et al., 2002; Bar-‐Matthews et al., 2003).


48

Figure 9. A selection of speleothem δ18O and ice core records indicate the global climate teleconnections. Speleothem records from the northern hemisphere (e.g. Moomi Cave, Hulu and Dongge Caves, Devils Hole, and Soreq Cave) have a strong imprint of high-‐latitude climates (Greenland and Antarctica), whereas low latitude records also contain an imprint of precessional variations (dashed thin lines), e.g. in Brazil (Botuvera Cave), and China (Hulu and Dongge Caves). These climate records demonstrate a robust response of atmospheric circulation to ocean–atmosphere–cryosphere reorganizations. For all records, up is warm and/or wet. There is a clear antiphasing between northern and southern hemisphere monsoon records (Brazil and China) that reflects hemispherically anti-‐symmetric precessional-‐scale insolation variations. The raw dataseries are shown by thin grey lines with a 5-‐pt running average (thick black line), except for the Devils Hole record, which is the raw data, and the older Moomi Cave record which is a running average. See text for references (after Lachniet et al. 2009).


49

Speleothems also provided evidence that monsoon and climate dynamics in low-‐latitude regions are partly controlled by solar insolation on precessional time scales. Records from the Indian Ocean and South American monsoon regions support the connection between stronger summer insolation and a more intense monsoon (Fig. 9). In southern Oman, the intensity of the monsoon appears to be strongly linked to summer insolation over the Holocene period (Neff et al., 2001; Fleitmann et al., 2003), as do changes in the strength of the Asian Monsoon (Dykoski et al., 2005) where decreased monsoon rainfall is linked to lower northern hemisphere summer insolation. In the southern South America, subtropical stalagmites show a pronounced Holocene increase in monsoon intensity that coincides with an increase in southern hemisphere summer insolation (Cruz et al., 2005; Wang et al., 2006; van Breukelen et al., 2008). Brazilian stalagmite δ18O show a strong inverse relation with the Asian Monsoon intensity (Fig. 9), which is interpreted to reflect an interhemispheric antiphasing of tropical rainfall attributed to variations in the north-‐south position of the ITCZ (Wang et al., 2006). Shorter-‐scaled events such as the 8.2 kyr event, which caused global cooling due to the drainage of the glacial Lake Agassiz in North America are also recorded in speleothems. European speleothems registered this period as generally cold (Verheyden et al., 2014), while monsoonal speleothems recorded a decrease in monsoon rainfall (Fleitmann et al., 2003; Dykoski et al., 2005). The 8.2 kyr event is well documented in a laminated speleothem from Monsoonal East Asia (Liu et al., 2013) and indicates strongly decreased monsoonal precipitation despite the relatively warm conditions that caused the event. Through layer counting Liu et al. (2013) indicated that the 8.2 kyr event lasted for 150 year with peak drought over a period of 68 years. They suggest that the Northern Hemisphere-‐Monsoonal Asia cold/dry and warm/wet teleconnection reacts rapidly to climate variations and are mostly driven by an atmospheric teleconnection. Cold conditions increase the snow cover in the Northern Hemisphere causing stronger reflection and thus lower heating of the northern hemisphere landmass, which shifts the ITCZ southward and causes drier conditions in Monsoonal Asia (Liu et al., 2013). “For paleoclimate, the last two decades have been the age of the ice core. The next two may be the age of the speleothem”, Gideon Henderson, Science, 2006. The numerous published speleothem studies in the last decade confirm the statement made by G. Henderson. Speleothems are valuable archives and allow reconstructing climate variations on the continents at high resolutions. Most important, the absolute ages that can be measured on speleothems allow estimating the timing and duration of events such as recently done for the duration of the 8.2 kyr event.


50

References Affek, H. P., Bar-‐Matthews, M., Ayalon, A., Matthews, A., and Eiler, J. M.: Glacial/interglacial temperature variations in Soreq cave speleothems as recorded by 'clumped isotope' thermometry, Geochimica Et Cosmochimica Acta, 72, 5351-‐5360, 2008.

Arienzo, M. M., Swart, P. K., and Vonhof, H. B.: Measurement of delta O-‐18 and delta H-‐2 values of fluid inclusion water in speleothems using cavity ring-‐down spectroscopy compared with isotope ratio mass spectrometry, Rapid Communications in Mass Spectrometry, 27, 2616-‐2624, 2013.

Ayalon, A., Bar-‐Matthews, M., and Kaufman, A.: Climatic conditions in the Eastern Mediterranean region during glacial marine isotopic stage 6, Geochimica Et Cosmochimica Acta, 66, A39-‐A39, 2002.

Bajo, P., Drysdale, R., Woodhead, J., Hellstrom, J., and Zanchetta, G.: High-‐resolution U-‐Pb dating of an Early Pleistocene stalagmite from Corchia Cave (central Italy), Quaternary Geochronology, 14, 5-‐17, 2012.

Baker, A., Genty, D., Dreybrodt, W., Barnes, W. L., Mockler, N. J., and Grapes, J.: Testing theoretically predicted stalagmite growth rate with Recent annually laminated samples: Implications for past stalagmite deposition, Geochimica Et Cosmochimica Acta, 62, 393-‐404, 1998.

Baker, A., Ito, E., Smart, P. L., and McEwan, R. F.: Elevated and variable values of C-‐13 in speleothems in a British cave system, Chemical Geology, 136, 263-‐270, 1997.

Baldini, J. U. L., McDermott, F., and Fairchild, I. J.: Spatial variability in cave drip water hydrochemistry: Implications for stalagmite paleoclimate records, Chemical Geology, 235, 390-‐404, 2006.

Baldini, J. U. L., McDermott, F., Hoffmann, D. L., Richards, D. A., and Clipson, N.: Very high-‐frequency and seasonal cave atmosphere P-‐CO2 variability: Implications for stalagmite growth and oxygen isotope-‐based paleoclimate records, Earth and Planetary Science Letters, 272, 118-‐129, 2008.

Bar-‐Matthews, M., Ayalon, A., Gilmour, M., Matthews, A., and Hawkesworth, C. J.: 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, 3181-‐3199, 2003.

Bar-‐Matthews, M., Ayalon, A., and Kaufman, A.: Late quaternary paleoclimate in the eastern Mediterranean region from stable isotope analysis of speleothems at Soreq Cave, Israel, Quaternary Research, 47, 155-‐168, 1997.


51

Bar-‐Matthews, M., Ayalon, A., and Kaufman, A.: Timing and hydrological conditions of Sapropel events in the Eastern Mediterranean, as evident from speleothems, Soreq cave, Israel, Chemical Geology, 169, 145-‐156, 2000.

Bar-‐Matthews, M., Ayalon, A., Kaufman, A., and Wasserburg, G. J.: The Eastern Mediterranean paleoclimate as a reflection of regional events: Soreq cave, Israel, Earth and Planetary Science Letters, 166, 85-‐95, 1999.

BarMatthews, M., Ayalon, A., Matthews, A., Sass, E., and Halicz, L.: 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, 337-‐347, 1996.

Beck, W. C., Grossman, E. L., and Morse, J. W.: Experimental studies of oxygen isotope fractionation in the carbonic acid system at 15 degrees, 25 degrees, and 40 degrees C, Geochimica Et Cosmochimica Acta, 69, 3493-‐3503, 2005.

Bond, G., Heinrich, H., Broecker, W., Labeyrie, L., McManus, J., Andrews, J., Huon, S., Jantschik, R., Clasen, S., Simet, C., Tedesco, K., Klas, M., Bonani, G., and Ivy, S.: Evidence for massive discharge of icebergs into the North-‐Atlantic ocean during the Last Glacial period Nature, 360, 245-‐249, 1992.

Bonniver, I.: Etude Hyrogeologique et dimmensionnement par modelisation du "systeme-‐tracage" du reseau karstique the Han-‐sur-‐Lesse (Massif de Boine, Belgique), 2011. Geologie, FUNDP Namur, Namur, p 93 to 97 pp., 2011.

Borsato, A., Frisia, S., Fairchild, I. J., Somogyi, A., and Susini, J.: Trace element distribution in annual stalagmite laminae mapped by micrometer-‐resolution X-‐ray fluorescence: Implications for incorporation of environmentally significant species, Geochimica Et Cosmochimica Acta, 71, 1494-‐1512, 2007.

Breitenbach, S. F. M., Rehfeld, K., Goswami, B., Baldini, J. U. L., Ridley, H. E., Kennett, D. J., Prufer, K. M., Aquino, V. V., Asmerom, Y., Polyak, V. J., Cheng, H., Kurths, J., and Marwan, N.: COnstructing Proxy Records from Age models (COPRA), Climate of the Past, 8, 1765-‐1779, 2012.

Broecker, W. S., Olson, E. A., and Orp, P. C.: Radiacarbon measurements and annual rings in cave formations, Nature, 185, 93-‐94, 1960.

Buhl, D., Immenhauser, A., Smeulders, G., Kabiri, L., and Richter, D. K.: Time series delta Mg-‐26 analysis in speleothem calcite: Kinetic versus equilibrium fractionation, comparison with other proxies and implications for palaeoclimate research, Chemical Geology, 244, 715-‐729, 2007.

Burns, S. J., Fleitmann, D., Matter, A., Kramers, J., and Al-‐Subbary, A. A.: Indian Ocean climate and an absolute chronology over Dansgaard/Oeschger events 9 to 13, Science, 301, 1365-‐1367, 2003.

Chacko, T. and Deines, P.: Theoretical calculation of oxygen isotope fractionation factors in carbonate systems, Geochimica Et Cosmochimica Acta, 72, 3642-‐3660, 2008.


52

Cheng, H., Edwards, R. L., Broecker, W. S., Denton, G. H., Kong, X., Wang, Y., Zhang, R., and Wang, X.: Ice Age Terminations, Science, 326, 248-‐252, 2009a.

Cheng, H., Edwards, R. L., Hoff, J., Gallup, C. D., Richards, D. A., and Asmerom, Y.: The half-‐lives of uranium-‐234 and thorium-‐230, Chemical Geology, 169, 17-‐33, 2000.

Cheng, H., Fleitmann, D., Edwards, R. L., Wang, X. F., Cruz, F. W., Auler, A. S., Mangini, A., Wang, Y. J., Kong, X. G., Burns, S. J., and Matter, A.: Timing and structure of the 8.2 kyr BP event inferred from delta O-‐18 records of stalagmites from China, Oman, and Brazil, Geology, 37, 1007-‐1010, 2009b.

Cobb, K. M., Adkins, J. F., Partin, J. W., and Clark, B.: Regional-‐scale climate influences on temporal variations of rainwater and cave dripwater oxygen isotopes in northern Borneo, Earth and Planetary Science Letters, 263, 207-‐220, 2007.

Cruz, F. W., Burns, S. J., Jercinovic, M., Karmann, I., Sharp, W. D., and Vuille, M.: Evidence of rainfall variations in Southern Brazil from trace element ratios (Mg/Ca and Sr/Ca) in a Late Pleistocene stalagmite, Geochimica Et Cosmochimica Acta, 71, 2250-‐2263, 2007.

Cruz, F. W., Burns, S. J., Karmann, I., Sharp, W. D., and Vuille, M.: Reconstruction of regional atmospheric circulation features during the late Pleistocene in subtropical Brazil from oxygen isotope composition of speleothems, Earth and Planetary Science Letters, 248, 495-‐507, 2006.

Cruz, F. W., Burns, S. J., Karmann, I., Sharp, W. D., Vuille, M., Cardoso, A. O., Ferrari, J. A., Dias, P. L. S., and Viana, O.: Insolation-‐driven changes in atmospheric circulation over the past 116,000 years in subtropical Brazil, Nature, 434, 63-‐66, 2005.

Cuntz, M.: CARBON CYCLE A dent in carbon's gold standard, Nature, 477, 547-‐548, 2011.

Daeron, M., Guo, W., Eiler, J., Genty, D., Blamart, D., Boch, R., Drysdale, R., Maire, R., Wainer, K., and Zanchetta, G.: (CO)-‐C-‐13-‐O-‐18 clumping in speleothems: Observations from natural caves and precipitation experiments, Geochimica Et Cosmochimica Acta, 75, 3303-‐3317, 2011.

Dansgaard, W.: Stable Isotopes In Precipitation, Tellus, 16, 436-‐468, 1964.

Dansgaard, W., Johnsen, S. J., Clausen, H. B., Dahljensen, D., Gundestrup, N. S., Hammer, C. U., Hvidberg, C. S., Steffensen, J. P., Sveinbjornsdottir, A. E., Jouzel, J., and Bond, G.: Evidence for general instability of past climate from a 250-‐Kyr Ice-‐Core record Nature, 364, 218-‐220, 1993.

De Cisneros, C. J., Caballero, E., Vera, J. A., and Andreo, B.: An optimized thermal extraction system for preparation of water from fluid inclusions in speleothems, Geologica Acta, 9, 149-‐158, 2011.


53

Deininger, M., Fohlmeister, J., Scholz, D., and Mangini, A.: Isotope disequilibrium effects: The influence of evaporation and ventilation effects on the carbon and oxygen isotope composition of speleothems -‐ A model approach, Geochimica Et Cosmochimica Acta, 96, 57-‐79, 2012.

Demeny, A., Kele, S., and Siklosy, Z.: Empirical equations for the temperature dependence of calcite-‐water oxygen isotope fractionation from 10 to 70 degrees C, Rapid Communications in Mass Spectrometry, 24, 3521-‐3526, 2010.

Dennis, P. F., Rowe, P. J., and Atkinson, T. C.: The recovery and isotopic measurement of water from fluid inclusions in speleothems, Geochimica Et Cosmochimica Acta, 65, 871-‐884, 2001.

Dorale, J. A., Edwards, R. L., Ito, E., and Gonzalez, L. A.: Climate and vegetation history of the midcontinent from 75 to 25 ka: A speleothem record from Crevice Cave, Missouri, USA, Science, 282, 1871-‐1874, 1998.

Dreybrodt, W.: Comments on processes contributing to the isotope composition of 13C and 18O in calcite depostied in speleothems, Acta Carsologica, 40, 2011.

Dreybrodt, W.: Deposition of calcite from thin water films of natural calcareous solutions and the growth of speleothems Chemical Geology, 29, 89-‐105, 1980.

Dreybrodt, W.: Evolution of the isotopic composition of carbon in a calcite precipitating H2O-‐CO2-‐CaCO3 solution and the related isotopic composition of calcite in stalagmites, Geochimica Et Cosmochimica Acta, 72, 4712-‐4724, 2008.

Dreybrodt, W.: Processes in karts systems. Physics, Chemistry and Geology, Belrin/Heidelberg/ New York, 1988.

Dreybrodt, W. and Scholz, D.: Climatic dependence of stable carbon and oxygen isotope signals recorded in speleothems: From soil water to speleothem calcite, Geochimica Et Cosmochimica Acta, 75, 734-‐752, 2011.

Dulinski, M. and Rozanski, K.: Formation of 12C/13C isotope ratios in spsleothems: a semi-‐dynamic model, Radiocarbon, 32, 7-‐16, 1990.

Dykoski, C. A., Edwards, R. L., Cheng, H., Yuan, D. X., Cai, Y. J., Zhang, M. L., Lin, Y. S., Qing, J. M., An, Z. S., and Revenaugh, J.: A high-‐resolution, absolute-‐dated Holocene and deglacial Asian monsoon record from Dongge Cave, China, Earth and Planetary Science Letters, 233, 71-‐86, 2005.

Edwards, R. L., Chen, J. H., and Wasserburg, G. J.: 238U-‐234U-‐230Th-‐232Th systematics and the precise measurement of time over the past 500 000 years, Earth and Planetary Science Letters, 81, 175-‐192, 1987.

Eiler, J. M.: "Clumped-‐isotope" geochemistry -‐ The study of naturally-‐occurring, multiply-‐substituted isotopologues, Earth and Planetary Science Letters, 262, 309-‐327, 2007.


54

Emrich, K., Ehhalt, D. H., and Volgel, J. C.: Carbon isotope fractionation during the precipitation of calciumcarbonates, Earth and Planetary Science Letters, 8, 363-‐371, 1970.

Fairchild, I. J. and Baker, A.: Speleothem Science: From process to past environment, Blackwell Publishing ltd., 2012.

Fairchild, I. J., Baker, A., Borsato, A., Frisia, S., Hinton, R. W., McDermott, F., and Tooth, A. F.: Annual to sub-‐annual resolution of multiple trace-‐element trends in speleothems, Journal of the Geological Society, 158, 831-‐841, 2001.

Fairchild, I. J., Borsato, A., Tooth, A. F., Frisia, S., Hawkesworth, C. J., Huang, Y. M., McDermott, F., and Spiro, B.: Controls on trace element (Sr-‐Mg) compositions of carbonate cave waters: implications for speleothem climatic records, Chemical Geology, 166, 255-‐269, 2000.

Fairchild, I. J., Smith, C. L., Baker, A., Fuller, L., Spötl, C., Mattey, D., McDermott, F., and Eimp: Modification and preservation of environmental signals in speleothems, Earth-‐Science Reviews, 75, 105-‐153, 2006.

Fairchild, I. J. and Treble, P. C.: Trace elements in speleothems as recorders of environmental change, Quaternary Science Reviews, 28, 449-‐468, 2009.

Fleitmann, D., Burns, S. J., Mudelsee, M., Neff, U., Kramers, J., Mangini, A., and Matter, A.: Holocene forcing of the Indian monsoon recorded in a stalagmite from Southern Oman, Science, 300, 1737-‐1739, 2003.

Fleitmann, D., Burns, S. J., Neff, U., Mudelsee, M., Mangini, A., and Matter, A.: Palaeoclimatic interpretation of high-‐resolution oxygen isotope profiles derived from annually laminated speleothems from Southern Oman, Quaternary Science Reviews, 23, 935-‐945, 2004.

Fleitmann, D., Cheng, H., Badertscher, S., Edwards, R. L., Mudelsee, M., Gokturk, O. M., Fankhauser, A., Pickering, R., Raible, C. C., Matter, A., Kramers, J., and Tuysuz, O.: Timing and climatic impact of Greenland interstadials recorded in stalagmites from northern Turkey, Geophysical Research Letters, 36, 2009.

Fohlmeister, J., Schroder-‐Ritzrau, A., Scholz, D., Spötl, C., Riechelmann, D. F. C., Mudelsee, M., Wackerbarth, A., Gerdes, A., Riechelmann, S., Immenhauser, A., Richter, D. K., and Mangini, A.: Bunker Cave stalagmites: an archive for central European Holocene climate variability, Climate of the Past, 8, 1751-‐1764, 2012.

Friedman, I. and O'Neil, J. R.: Compilation of stable isotope fractionation factors of geochemical interest, Geolog. Surv. Prof. Paper 440-‐KK, 1977. 117, 1977.

Frisia, S., Borsato, A., Fairchild, I. J., and McDermott, F.: Calcite fabrics, growth mechanisms, and environments of formation in speleothems from the Italian Alps and southwestern Ireland, Journal of Sedimentary Research, 70, 1183-‐1196, 2000.


55

Frisia, S., Borsato, A., Preto, N., and McDermott, F.: 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, 411-‐424, 2003.

Fuller, L., Baker, A., Fairchild, I. J., Spötl, C., Marca-‐Bell, A., Rowe, P., and Dennis, P. F.: Isotope hydrology of dripwaters in a Scottish cave and implications for stalagmite palaeoclimate research, Hydrology and Earth System Sciences, 12, 1065-‐1074, 2008.

Gabitov, R. I. and Watson, E. B.: Partitioning of strontium between calcite and fluid, Geochemistry Geophysics Geosystems, 7, 2006.

Galy, A., Bar-‐Matthews, M., Halicz, L., and O'Nions, R. K.: Mg isotopic composition of carbonate: insight from speleothem formation, Earth and Planetary Science Letters, 201, 105-‐115, 2002.

Gascoyne, M.: Paleioclimatic determination from cave calcite deposits Quaternary Science Reviews, 11, 609-‐632, 1992.

Gascoyne, M.: Trace element partition coefficients in the calcite water system and their paleoclimatic significance in cave studies Journal of Hydrology, 61, 213-‐222, 1983.

Gat, J. R.: Oxygen and hydrogen isotopes in the hydrologic cycle, Annual Review of Earth and Planetary Sciences, 24, 225-‐262, 1996.

Genty, D., Baker, A., Massault, M., Proctor, C., Gilmour, M., Pons-‐Branchu, E., and Hamelin, B.: Dead carbon in stalagmites: Carbonate bedrock paleodissolution vs. ageing of soil organic matter. Implications for C-‐13 variations in speleothems, Geochimica Et Cosmochimica Acta, 65, 3443-‐3457, 2001a.

Genty, D., Baker, A., and Vokal, B.: Intra-‐ and inter-‐annual growth rate of modern stalagmites, Chemical Geology, 176, 191-‐212, 2001b.

Genty, D., Blamart, D., Ouahdi, R., Gilmour, M., Baker, A., Jouzel, J., and Van-‐Exter, S.: Precise dating of Dansgaard-‐Oeschger climate oscillations in western Europe from stalagmite data, Nature, 421, 833-‐837, 2003.

Genty, D. and Deflandre, G.: Drip flow variations under a stalactite of the Pere Noel cave (Belgium). Evidence of seasonal variations and air pressure constraints, Journal of Hydrology, 211, 208-‐232, 1998.

Genty, D., Labuhn, I., Hoffmann, G., Danis, P. A., Mestre, O., Bourges, F., Wainer, K., Massault, M., Van Exter, S., Regnier, E., Orengo, P., Falourd, S., and Minster, B.: Rainfall and cave water isotopic relationships in two South-‐France sites, Geochimica Et Cosmochimica Acta, 131, 323-‐343, 2014.

Genty, D. and Massault, M.: Bomb C-‐14 recorded in laminated speleothems: Calculation of dead carbon proportion, Radiocarbon, 39, 33-‐48, 1997.


56

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

Goede, A., McCulloch, M., McDermott, F., and Hawkesworth, C.: Aeolian contribution to strontium and strontium isotope variations in a Tasmanian speleothem, Chemical Geology, 149, 37-‐50, 1998.

Hansen, M., Dreybrodt, W., and Scholz, D.: Chemical evolution of dissolved inorganic carbon species flowing in thin water films and its implications for (rapid) degassing of CO2 during speleothem growth, Geochimica Et Cosmochimica Acta, 107, 242-‐251, 2013.

Hartland, A., Fairchild, I. J., Lead, J. R., and Baker, A.: Fluorescent properties of organic carbon in cave dripwaters: Effects of filtration, temperature and pH, Science of the Total Environment, 408, 5940-‐5950, 2010.

Henderson, G. M.: Climate -‐ Caving in to new chronologies, Science, 313, 620-‐622, 2006.

Hendy, C. H.: Isotopic geochemistry of speleothems: 1. Calculations of effects on different modes of formation on isotopic composition of speleothems and their applicability as paleoclimatic indicators, Geochimica Et Cosmochimica Acta, 35, 801-‐824, 1971.

Hendy, C. H. and Wilson, A. T.: Palaeoclimatic data from speleothems Nature, 219, 48-‐&, 1968.

Herbert, T. D., Schuffert, J. D., Andreasen, D., Heusser, L., Lyle, M., Mix, A., Ravelo, A. C., Stott, L. D., and Heguera, J. C.: The California Current, devils hole, and Pleistocene climate -‐ Response, Science, 296, 2002.

Hercman, H. and Pawlak, J.: MOD-‐AGE: An age-‐depth model construction algorithm, Quaternary Geochronology, 12, 1-‐10, 2012.

Hodge, E. J., Richards, D. A., Smart, P. L., Gines, A., and Mattey, D. P.: Sub-‐millennial climate shifts in the western Mediterranean during the last glacial period recorded in a speleothem from Mallorca, Spain, Journal of Quaternary Science, 23, 713-‐718, 2008.

Hoefs, J.: Stable isotopes geochemistry Berlin, 1997.

Holland, H. D., Kirsipu, T. V., Huebner, J. S., and Oxburgh, U. M.: On some aspects of the chemical evolution of cave waters Journal of Geology, 72, 36-‐67, 1964.

Hori, M., Ishikawa, T., Nagaishi, K., Lin, K., Wang, B. S., You, C. F., Shen, C. C., and Kano, A.: Prior calcite precipitation and source mixing process influence Sr/Ca, Ba/Ca and Sr-‐87/Sr-‐86 of a stalagmite developed in southwestern Japan during 18.0 4.5 ka, Chemical Geology, 347, 190-‐198, 2013.


57

Horita, J. and Clayton, R. N.: Comment on the studies of oxygen isotope fractionation between calcium carbonates and water at low temperatures by Zhou and Zheng (2003;2005), Geochimica Et Cosmochimica Acta, 71, 3131-‐3135, 2007.

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

Jimenez-‐Lopez, C., Caballero, E., Huertas, F. J., and Romanek, C. S.: Chemical, mineralogical and isotope behavior, and phase transformation during the precipitation of calcium carbonate minerals from intermediate ionic solution at 25 degrees C, Geochimica Et Cosmochimica Acta, 65, 3219-‐3231, 2001.

Karmann, I., Cruz, F. W., Viana, O., and Burns, S. J.: Climate influence on geochemistry parameters of waters from Santana-‐Perolas cave system, Brazil, Chemical Geology, 244, 232-‐247, 2007.

Katz, A.: Interaction of Mg with Calcite during crystal-‐growth at 25-‐90 degrees C and 1 atmophere, Geochimica Et Cosmochimica Acta, 37, 1563-‐&, 1973.

Katz, A., Starinsk.A, Sass, E., and Holland, H. D.: Strontium behaviour in aragonite-‐calcite transformation-‐ experimental study at 40-‐98 degrees C Geochimica Et Cosmochimica Acta, 36, 481-‐&, 1972.

Kaufman, A., Wasserburg, G. J., Porcelli, D., Bar-‐Matthews, M., Ayalon, A., and Halicz, L.: U-‐Th isotope systematics from the Soreq cave, Israel and climatic correlations, Earth and Planetary Science Letters, 156, 141-‐155, 1998.

Kim, S.-‐T., Mucci, A., and Taylor, B. E.: Phosphoric acid fractionation factors for calcite and aragonite between 25 and 75 degrees C: Revisited, Chemical Geology, 246, 135-‐146, 2007.

Kim, S. T. and O'Neil, J. R.: Equilibrium and nonequilibrium oxygen isotope effects in synthetic carbonates, Geochimica Et Cosmochimica Acta, 61, 3461-‐3475, 1997.

Kluge, T., Marx, T., Scholz, D., Niggemann, S., Mangini, A., and Aeschbach-‐Hertig, W.: A new tool for palaeoclimate reconstruction: Noble gas temperatures from fluid inclusions in speleothems, Earth and Planetary Science Letters, 269, 407-‐414, 2008.

Kluge, T., Riechelmann, D. F. C., Wieser, M., Spötl, C., Sultenfuss, J., Schroder-‐Ritzrau, A., Niggemann, S., and Aeschbach-‐Hertig, W.: Dating cave drip water by tritium, Journal of Hydrology, 394, 396-‐406, 2010.

Kong, X. G., Wang, Y. J., Edwards, L. R., Cheng, H., and Wang, X. F.: The altitude effect on oxygen isotope composition of stalagmite, Abstracts with programs, Geological Society of America, 37, 13, 2005.

Lachniet, M. S.: Climatic and environmental controls on speleothem oxygen-‐isotope values, Quaternary Science Reviews, 28, 412-‐432, 2009.


58

Lachniet, M. S. and Patterson, W. P.: Use of correlation and stepwise regression to evaluate physical controls on the stable isotope values of Panamanian rain and surface waters, Journal of Hydrology, 324, 115-‐140, 2006.

Lauritzen, S. E. and Lundberg, J.: Calibration of the speleothem delta function: an absolute temperature record for the Holocene in northern Norway, Holocene, 9, 659-‐669, 1999.

Liu, Y. H., Henderson, G. M., Hu, C. Y., Mason, A. J., Charnley, N., Johnson, K. R., and Xie, S. C.: Links between the East Asian monsoon and North Atlantic climate during the 8,200 year event, Nature Geoscience, 6, 117-‐120, 2013.

Mangini, A., Spötl, C., and Verdes, P.: Reconstruction of temperature in the Central Alps during the past 2000 yr from a delta O-‐18 stalagmite record, Earth and Planetary Science Letters, 235, 741-‐751, 2005.

Mattey, D., Lowry, D., Duffet, J., Fisher, R., Hodge, E., and Frisia, S.: A 53 year seasonally resolved oxygen and carbon isotope record from a modem Gibraltar speleothem: Reconstructed drip water and relationship to local precipitation, Earth and Planetary Science Letters, 269, 80-‐95, 2008.

Matthews, A., Ayalon, A., and Bar-‐Matthews, M.: D/H ratios of fluid inclusions of Soreq cave (Israel) speleothems as a guide to the Eastern Mediterranean Meteoric Line relationships in the last 120 ky, Chemical Geology, 166, 183-‐191, 2000.

McDermott, F.: Palaeo-‐climate reconstruction from stable isotope variations in speleothems: a review, Quaternary Science Reviews, 23, 901-‐918, 2004.

McDermott, F., Atkinson, T. C., Fairchild, I. J., Baldini, L. M., and Mattey, D. P.: A first evaluation of the spatial gradients in delta O-‐18 recorded by European Holocene speleothems, Glob. Planet. Change, 79, 275-‐287, 2011.

McDermott, F., Frisia, S., Huang, Y. M., Longinelli, A., Spiro, B., Heaton, T. H. E., Hawkesworth, C. J., Borsato, A., Keppens, E., Fairchild, I. J., van der Borg, K., Verheyden, S., and Selmo, E.: Holocene climate variability in Europe: Evidence from delta O-‐18, textural and extension-‐rate variations in three speleothems, Quaternary Science Reviews, 18, 1021-‐1038, 1999.

McDermott, F., Schwarcz, H. P., and Rowe, P. J.: Isotopes in speleothems, In Isotopes in paleoenvironmental research (ed. M.J. Leng), 2005. 185-‐226, 2005.

McDonald, J., Drysdale, R., and Hill, D.: The 2002-‐2003 El Nino recorded in Australian cave drip waters: Implications for reconstructing rainfall histories using stalagmites, Geophysical Research Letters, 31, 2004.

McDonald, J., Drysdale, R., Hill, D., Chisari, R., and Wong, H.: The hydrochemical response of cave drip waters to sub-‐annual and inter-‐annual climate variability, Wombeyan Caves, SE Australia, Chemical Geology, 244, 605-‐623, 2007.


59

McGillen, M. R. and Fairchild, I. J.: An experimental study of incongruent dissolution of CaCO3 under analogue glacial conditions, Journal of Glaciology, 51, 383-‐390, 2005.

McIntire, W. L.: Trace element partition coeficients-‐ a review of theory and applications to Geology, Geochimica Et Cosmochimica Acta, 27, 1209-‐1264, 1963.

Mickler, P. J., Stern, L. A., and Banner, J. L.: Large kinetic isotope effects in modern speleothems, Geological Society of America Bulletin, 118, 65-‐81, 2006.

Miorandi, R., Borsato, A., Frisia, S., Fairchild, I. J., and Richter, D. K.: Epikarst hydrology and implications for stalagmite capture of climate changes at Grotta di Ernesto (NE Italy): results from long-‐term monitoring, Hydrological Processes, 24, 3101-‐3114, 2010.

Mook, W. G.: Volume 1:Introduction-‐theory, methods, review. In:, Environmental Isotopes in the Hydrological Cycle. Priciples and Applications., 2000. 2000.

Mühlinghaus, C., Scholz, D., and Mangini, A.: Modelling fractionation of stable isotopes in stalagmites, Geochimica Et Cosmochimica Acta, doi: 10.1016/j.gca.2009.09.010, 2009. 7275-‐7289, 2009.

Mühlinghaus, C., Scholz, D., and Mangini, A.: Modelling stalagmite growth and delta C-‐13 as a function of drip interval and temperature, Geochimica Et Cosmochimica Acta, 71, 2780-‐2790, 2007.

Musgrove, M. and Banner, J. L.: Controls on the spatial and temporal variability of vadose dripwater geochemistry: Edwards Aquifer, central Texas, Geochimica Et Cosmochimica Acta, 68, 1007-‐1020, 2004.

Neff, U., Burns, S. J., Mangini, A., Mudelsee, M., Fleitmann, D., and Matter, A.: Strong coherence between solar variability and the monsoon in Oman between 9 and 6 kyr ago, Nature, 411, 290-‐293, 2001.

O'Neil, J. R., Clayton, R. N., and Mayeda, T. K.: Oxygen isotope fractionation in divalent metal carbonates Journal of Chemical Physics, 51, 5547-‐&, 1969.

Orland, I. J., Burstyn, Y., Bar-‐Matthews, M., Kozdon, R., Ayalon, A., Matthews, A., and Valley, J. W.: Seasonal climate signals (1990-‐2008) in a modern Soreq Cave stalagmite as revealed by high-‐resolution geochemical analysis, Chemical Geology, 363, 322-‐333, 2014.

Palmer, M. R. and Edmond, J. M.: Controls over Strontium isotope composition of river water Geochimica Et Cosmochimica Acta, 56, 2099-‐2111, 1992.

Piccini, L., Zanchetta, G., Drysdale, R. N., Hellstrom, J., Isola, I., Fallick, A. E., Leone, G., Doveri, M., Mussi, M., Mantelli, F., Molli, G., Lotti, L., Roncioni, A., Regattieri, E., Meccheri, M., and Vaselli, L.: The environmental features of the Monte Corchia cave system (Apuan Alps, central Italy) and their effects on speleothem growth, International Journal of Speleology, 37, 153-‐172, 2008.


60

Pingitore, N. E. and Eastman, M. P.: The co-‐precipitation of Sr2+ with calcite at 25 degrees C and 1 atm, Geochimica Et Cosmochimica Acta, 50, 2195-‐2203, 1986.

Polyak, V., Hill, C., and Asmerom, Y.: Age and evolution of the Grand Canyon revealed by U-‐Pb dating of water table-‐type speleothems, Science, 319, 1377-‐1380, 2008.

Proctor, C. J., Baker, A., Barnes, W. L., and Gilmour, R. A.: A thousand year speleothem proxy record of North Atlantic climate from Scotland, Climate Dynamics, 16, 815-‐820, 2000.

Rayleigh, J. W. S.: Theoritical considerations respecting the separation of gasses by diffusion and similar processes, Philos Magasine, 12, 493p, 1896.

Richards, D. A., Bottrell, S. H., Cliff, R. A., Strohle, K., and Rowe, P. J.: U-‐Pb dating of a speleothem of Quaternary age, Geochimica Et Cosmochimica Acta, 62, 3683-‐3688, 1998.

Riechelmann, D. F. C., Deininger, M., Scholz, D., Riechelmann, S., Schroeder-‐Ritzrau, A., Spoetl, C., Richter, D. K., Mangini, A., and Immenhauser, A.: Disequilibrium carbon and oxygen isotope fractionation in recent cave calcite: Comparison of cave precipitates and model data, Geochimica Et Cosmochimica Acta, 103, 232-‐244, 2013.

Riechelmann, D. F. C., Schroeder-‐Ritzrau, A., Scholz, D., Fohlmeister, J., Spoetl, C., Richter, D. K., and Mangini, A.: Monitoring Bunker Cave (NW Germany): A prerequisite to interpret geochemical proxy data of speleothems from this site, Journal of Hydrology, 409, 682-‐695, 2011.

Roberts, M. S., Smart, P. L., and Baker, A.: Annual trace element variations in a Holocene speleothem, Earth and Planetary Science Letters, 154, 237-‐246, 1998.

Rozanski, K., Araguasaraguas, L., and Gonfiantini, R.: Relationship between long-‐term trends of 18O isotope composition of precipitation and climate Science, 258, 981-‐985, 1992.

Rutlidge, H., Baker, A., Marjo, C. E., Andersen, M. S., Graham, P. W., Cuthbert, M. O., Rau, G. C., Roshan, H., Markowska, M., Mariethoz, G., and Jex, C. N.: Dripwater organic matter and trace element geochemistry in a semi-‐arid karst environment: Implications for speleothem paleoclimatology, Geochimica Et Cosmochimica Acta, 135, 217-‐230, 2014.

Salomons, W. and Mook, W. G.: Isotope Geochemistry in carbonates in the weathering zone, 1986.

Scheidegger, Y., Baur, H., Brennwald, M. S., Fleitmann, D., Wieler, R., and Kipfer, R.: Accurate analysis of noble gas concentrations in small water samples and its application to fluid inclusions in stalagmites, Chemical Geology, 272, 31-‐39, 2010.


61

Schmidt, G. A., LeGrande, A. N., and Hoffmann, G.: Water isotope expressions of intrinsic and forced variability in a coupled ocean-‐atmosphere model, Journal of Geophysical Research-‐Atmospheres, 112, 2007.

Scholz, D. and Hoffmann, D. L.: StalAge -‐ An algorithm designed for construction of speleothem age models, Quaternary Geochronology, 6, 369-‐382, 2011.

Scholz, D., Hoffmann, D. L., Hellstrom, J., and Ramsey, C. B.: A comparison of different methods for speleothem age modelling, Quaternary Geochronology, 14, 94-‐104, 2012.

Scholz, D., Muehlinghaus, C., and Mangini, A.: Modelling delta C-‐13 and delta O-‐18 in the solution layer on stalagmite surfaces, Geochimica Et Cosmochimica Acta, 73, 2592-‐2602, 2009.

Shakun, J. D., Burns, S. J., Fleitmann, D., Kramers, J., Matter, A., and Al-‐Subary, A.: A high-‐resolution, absolute-‐dated deglacial speleothem record of Indian Ocean climate from Socotra Island, Yemen, Earth and Planetary Science Letters, 259, 442-‐456, 2007.

Smart, P. L. and Friedrich, H.: Water movement and storage in the unsaturated zone of amaturely karstified aquifer, Mendip Hills, England, Proceedings of the Conference on Environ-‐mental Problems in Karst Terrains and Teir Solution, Bowling Green, Kentucky,National Water Well Association, 1987. 57-‐87, 1987.

Smith, C. L., Fairchild, I. J., Spötl, C., Frisia, S., Borsato, A., Moreton, S. G., and Wynn, P. M.: Chronology building using objective identification of annual signals in trace element profiles of stalagmites, Quaternary Geochronology, 4, 11-‐21, 2009.

Spötl, C. and Mangini, A.: Stalagmite from the Austrian Alps reveals Dansgaard-‐Oeschger events during isotope stage 3: Implications for the absolute chronology of Greenland ice cores, Earth and Planetary Science Letters, 203, 507-‐518, 2002.

Spötl, C., Unterwurzacher, M., Mangini, A., and Longstaffe, F. J.: Carbonate speleothems in the dry, inneralpine Vinschgau Valley, northernmost Italy: Witnesses of changes in climate and hydrology since the last glacial maximum, Journal of Sedimentary Research, 72, 793-‐808, 2002.

Stern, L., Banner, J. L., Cowan, B., Copeland, E., Mickler, P. J., Guifoyle, A., James, E., Musgrove, M., and Mack, L.: Trace element variations in speleothem calcite: influence of non-‐environmental factors, Geol. Soc. Am. Abstr. Prog., 193, 435, 2005.

Suess, H.: Radiocarbon concentration in modern wood, Science, 122, 415-‐417, 1955.

Talma, A. S. and Vogel, J. C.: Late Quaternary paleotemperatures derived from a speleothem from Cango Caves, Cape province, South Africa Quaternary Research, 37, 203-‐213, 1992.


62

Tarutani, T., Clayton, R. N., and Mayeda, T. K.: Effect of polymorphism and magnesium substitution on oxygen isotope fractionnation between calcium carbonate and water, Geochimica Et Cosmochimica Acta, 33, 987-‐&, 1969.

Thompson, P., Schwarcz, H. P., and Ford, D. C.: Continental Pleistocene climatic variations from speleothen age and isotopic data Science, 184, 893-‐895, 1974.

Tooth, A. F. and Fairchild, I. J.: Soil and karst aquifer hydrological controls on the geochemical evolution of speleothem-‐forming drip waters, Crag Cave, southwest Ireland, Journal of Hydrology, 273, 51-‐68, 2003.

Treble, P., Shelley, J. M. G., and Chappell, J.: Comparison of high resolution sub-‐annual records of trace elements in a modern (1911-‐1992) speleothem with instrumental climate data from southwest Australia, Earth and Planetary Science Letters, 216, 141-‐153, 2003.

Treble, P. C., Chappell, J., Gagan, M. K., McKeegan, K. D., and Harrison, T. M.: In situ measurement of seasonal delta O-‐18 variations and analysis of isotopic trends in a modem speleothem from southwest Australia, Earth and Planetary Science Letters, 233, 17-‐32, 2005a.

Treble, P. C., Chappell, J., and Shelley, J. M. G.: Complex speleothem growth processes revealed by trace element mapping and scanning electron microscopy of annual layers, Geochimica Et Cosmochimica Acta, 69, 4855-‐4863, 2005b.

Tremaine, D. M. and Froelich, P. N.: Speleothem trace element signatures: A hydrologic geochemical study of modern cave dripwaters and farmed calcite, Geochimica Et Cosmochimica Acta, 121, 522-‐545, 2013.

Tremaine, D. M., Froelich, P. N., and Wang, Y.: Speleothem calcite farmed in situ: Modern calibration of delta O-‐18 and delta C-‐13 paleoclimate proxies in a continuously-‐monitored natural cave system, Geochimica Et Cosmochimica Acta, 75, 4929-‐4950, 2011.

Vaks, A., Woodhead, J., Bar-‐Matthews, M., Ayalon, A., Cliff, R. A., Zilberman, T., Matthews, A., and Frumkin, A.: Pliocene-‐Pleistocene climate of the northern margin of Saharan-‐Arabian Desert recorded in speleothems from the Negev Desert, Israel, Earth and Planetary Science Letters, 368, 88-‐100, 2013.

van Breukelen, M. R., Vonhof, H. B., Hellstrom, J. C., Wester, W. C. G., and Kroon, D.: Fossil dripwater in stalagmites reveals Holocene temperature and rainfall variation in Amazonia, Earth and Planetary Science Letters, 275, 54-‐60, 2008.

Van Rampelbergh, M., Fleitmann, D., Verheyden, S., Cheng, H., Edwards, L., De Geest, P., De Vleeschouwer, D., Burns, S. J., Matter, A., Claeys, P., and Keppens, E.: Mid-‐ to late Holocene Indian Ocean Monsoon variability recorded in four speleothems from Socotra Island, Yemen, Quaternary Science Reviews, 65, 129-‐142, 2013.

Van Rampelbergh, M., Verheyden, S., Allan, M., Quinif, Y., Keppens, E., Cheng, H., Edwards, L. R., and Claeys, P.: A 500-‐year seasonally resolved δ18O and δ13C and


63

layer thickness record from a speleothem of the Han-‐sur-‐Lesse cave, Belgium, Climate of the Past, in review. in review.

Van Rampelbergh, M., Verheyden, S., Allan, M., Quinif, Y., Keppens, E., and Claeys, P.: Seasonal variations recorded in cave monitoring results and a 10-‐year monthly resolved speleothem δ18O and δ13C record from the Han-‐sur-‐Lesse cave, Belgium., Climate of the Past, 10, 1-‐15, 2014.

Verheyden, S., Genty, D., Cattani, O., and van Breukelen, M. R.: Water release patterns of heated speleothem calcite and hydrogen isotope composition of fluid inclusions, Chemical Geology, 247, 266-‐281, 2008a.

Verheyden, S., Genty, D., Deflandre, G., Quinif, Y., and Keppens, E.: Monitoring climatological, hydrological and geochemical parameters in the Pere Noel cave (Belgium): implication for the interpretation of speleothem isotopic and geochemical time-‐series, International Journal of Speleology, 37, 221-‐234, 2008b.

Verheyden, S., Keppens, E., Quinif, Y., Cheng, H. J., and Edwards, L. R.: Late-‐glacial and Holocene climate reconstruction as inferred from a stalagmite -‐ Grotte du Pere Noel, Han-‐sur-‐Lesse, Belgium, Geologica Belgica, 17, 83-‐89, 2014.

Vonhof, H. B., van Breukelen, M. R., Postma, O., Rowe, P. J., Atkinson, T. C., and Kroon, D.: A continuous-‐flow crushing device for on-‐line delta(2)H analysis of fluid inclusion water in speleothems, Rapid Communications in Mass Spectrometry, 20, 2553-‐2558, 2006.

Walker, J., Cliff, R. A., and Latham, A. G.: U-‐Pb isotopic age of the StW 573 hominid from Sterkfontein, South Africa, Science, 314, 1592-‐1594, 2006.

Wang, X. F., Auler, A. S., Edwards, R. L., Cheng, H., Ito, E., and Solheid, M.: Interhemispheric anti-‐phasing of rainfall during the last glacial period, Quaternary Science Reviews, 25, 3391-‐3403, 2006.

Wang, Y. J., Cheng, H., Edwards, R. L., An, Z. S., Wu, J. Y., Shen, C. C., and Dorale, J. A.: A high-‐resolution absolute-‐dated Late Pleistocene monsoon record from Hulu Cave, China, Science, 294, 2345-‐2348, 2001.

Williams, P.: Th erole of the epikarst in karst and cave hydrology: a review, International Journal of Speleology, 37, 1-‐10, 2008.

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

Winograd, I. J.: Evidence from uranium-‐series-‐dated speleothems for the timing of the penultimate deglaciation of northwestern Europe, Quaternary Research, 58, 60-‐61, 2002.


64

Winograd, I. J., Coplen, T. B., Landwehr, J. M., Riggs, A. C., Ludwig, K. R., Szabo, B. J., Kolesar, P. T., and Revesz, K. M.: Continuous 500,000-‐year climate record from vein calcite in Devils-‐Hole, Nevada, Science, 258, 255-‐260, 1992.

Winograd, I. J., Landwehr, J. M., Coplen, T. B., Sharp, W. D., Riggs, A. C., Ludwig, K. R., and Kolesar, P. T.: Devils Hole, Nevada, delta O-‐18 record extended to the mid-‐Holocene, Quaternary Research, 66, 202-‐212, 2006.

Woodhead, J., Hellstrom, J., Maas, R., Drysdale, R., Zanchetta, G., Devine, P., and Taylor, E.: U-‐Pb geochronology of speleothems by MC-‐ICPMS, Quaternary Geochronology, 1, 208-‐221, 2006.

Yang, W. B., Lowenstein, T. K., Krouse, H. R., Spencer, R. J., and Ku, T. L.: A 200,000-‐year delta O-‐18 record of closed-‐basin lacustrine calcite, Death Valley, California, Chemical Geology, 216, 99-‐111, 2005.

Zeebe, R. E., Wolf-‐Gladrow, D. A., and Jansen, H.: On the time required to establish chemical and isotopic equilibrium in the carbon dioxide system in seawater, Marine Chemistry, 65, 135-‐153, 1999.

Zhang, R., Schwarcz, H. P., Ford, D. C., Schroeder, F. S., and Beddows, P. A.: An absolute paleotemperature record from 10 to 6 Ka inferred from fluid inclusion D/H ratios of a stalagmite from Vancouver Island, British Columbia, Canada, Geochimica Et Cosmochimica Acta, 72, 1014-‐1026, 2008.

Zhou, H. Y., Feng, Y. X., Zhao, J. X., Shen, C. C., You, C. F., and Lin, Y.: Deglacial variations of Sr and Sr-‐87/Sr-‐86 ratio recorded by a stalagmite from Central China and their association with past climate and environment, Chemical Geology, 268, 233-‐247, 2009.

Chapter 3: Speleothems from Socotra

65

Chapter 3

Socotran speleothems reveal monsoon changes during the Mid-‐ to Late-‐Holocene This chapter elaborates on the current knowledge on the behavior of the Indian Ocean Monsoon (IOM) during the Mid-‐ to Late-‐Holocene for the northern Indian Ocean. Numerous speleothem studies already provide information on the IOM variations and its warm/wet and cold/dry teleconnection with the Northern Atlantic climate (Burns et al., 1998; Burns et al., 2001; Neff et al., 2001; Burns et al., 2003; Fleitmann et al., 2004; Fleitmann et al., 2007; Shakun et al., 2007). The North Atlantic and the IOM are linked through the boreal summer position of the Intertropical Convergence Zone (ITCZ) (Fig. 1). During warmer North Atlantic periods, stronger northern hemisphere insolation causes the summer position of the ITCZ to move further up north, which causes the precipitation amounts brought by the southwest IOM to increase. Since the Mid-‐Holocene, the summer position of the ITCZ is gradually moving south due to the diminishing boreal summer insolation (Wanner et al., 2006). This southern retreat of the summer ITCZ causes a decrease in summer IOM precipitation since 8 ka. However, no records describing the evolution of the other IOM subsystem, being the northeast winter monsoon, have been established yet.


66

Figure 1. The present-‐day location of the Intertropical Convergence zone (ITCZ). During boreal summer, increased northern hemisphere insolation causes the intensification of low pressure cells, which pull the ITCZ to the north until it reaches its northernmost position in July. During boreal winter, the pressure gradients reverse and the ITCZ moves south until it reaches its southernmost position in December. Socotra Island lies in the upper part of the oscillation belt (red star) (adapted from www.meteoweb.eu). The Island of Socotra, located in the northern Indian Ocean (Fig.1), has the valuable advantage to separate the rains from both subsystems of the IOM due to the ‘barrier’ action of the Haggeher Mountains. Northeast winter monsoon rains mainly affect the northern and eastern side of the island between September and December, while southwestern summer monsoon rains cause wetter conditions between May and June only on the southwestern part of the island. Four stalagmites from the eastern side of Socotra cover the last 6 000 years and indicated a different evolution of the northeast monsoon compared to the southwest monsoon. While, the southwest monsoon displays a gradual precipitation decrease, the northeast monsoon precipitation intensity weakens between 6.0 and 3.8 ka, and remains constant afterwards with two superimposed drier periods, between 0 and 0.6 ka and from 2.2 to 3.8 ka. No link can be established between the winter IOM variations and the Greenland ice cores. More high-‐resolution records from this region are required to understand the exact forcing behind the northeast monsoon in this area.


67

REFERENCES Burns, S. J., Fleitmann, D., Matter, A., Kramers, J., and Al-‐Subbary, A. A.: Indian Ocean climate and an absolute chronology over Dansgaard/Oeschger events 9 to 13, Science, 301, 1365-‐1367, 2003.

Burns, S. J., Fleitmann, D., Matter, A., Neff, U., and Mangini, A.: Speleothem evidence from Oman for continental pluvial events during interglacial periods, Geology, 29, 623-‐626, 2001.

Burns, S. J., Matter, A., Frank, N., and Mangini, A.: Speleothem-‐based paleoclimate record from northern Oman, Geology, 26, 499-‐502, 1998.

Fleitmann, D., Burns, S. J., Mangini, A., Mudelsee, M., Kramers, J., Villa, I., Neff, U., Al-‐Subbary, A. A., Buettner, A., Hippler, D., and Matter, A.: Holocene ITCZ and Indian monsoon dynamics recorded in stalagmites from Oman and Yemen (Socotra), Quaternary Science Reviews, 26, 170-‐188, 2007.

Fleitmann, D., Burns, S. J., Neff, U., Mudelsee, M., Mangini, A., and Matter, A.: Palaeoclimatic interpretation of high-‐resolution oxygen isotope profiles derived from annually laminated speleothems from Southern Oman, Quaternary Science Reviews, 23, 935-‐945, 2004.

Neff, U., Burns, S. J., Mangini, A., Mudelsee, M., Fleitmann, D., and Matter, A.: Strong coherence between solar variability and the monsoon in Oman between 9 and 6 kyr ago, Nature, 411, 290-‐293, 2001.

Shakun, J. D., Burns, S. J., Fleitmann, D., Kramers, J., Matter, A., and Al-‐Subary, A.: A high-‐resolution, absolute-‐dated deglacial speleothem record of Indian Ocean climate from Socotra Island, Yemen, Earth and Planetary Science Letters, 259, 442-‐456, 2007.


68

Mid- to late Holocene Indian Ocean Monsoon variability recorded infour speleothems from Socotra Island, Yemen

Maïté Van Rampelbergh a,*, Dominik Fleitmann b,c, Sophie Verheyden a,d, Hai Cheng e,f,Lawrence Edwards f, Peter De Geest g, David De Vleeschouwer a, Stephen J. Burns h, Albert Matter b,Philippe Claeys a, Eddy Keppens a

a Earth System Sciences Department, Vrije Universiteit Brussel (VUB), Pleinlaan 2, B-1050 Brussels, Belgiumb Institute of Geosciences, University of Bern, Baltzerstrasse 1-3, CH-3012 Bern, SwitzerlandcOeschger Centre for Climate Change Research, University of Bern, Zähringerstrasse 25, CH-3012 Bern, SwitzerlanddGeological Survey of Belgium, Royal Belgian Institute of Natural Sciences, Jennerstraat 13, B-1000 Brussels, Belgiume Institute of Global Environmental Change, Xi’an Jiaotong University, Xi’an 710049, ChinafDepartment of Geological Sciences, University of Minnesota, 100 Union Street SE, Minneapolis, MN 55455, USAgBijlokestraat 57, B-9070 Destelbergen, BelgiumhDepartment of Geosciences, University of Massachusetts, Morrill Science Center 233, Amherst, MA 01003, USA

a r t i c l e i n f o

Article history:Received 10 May 2012Received in revised form11 January 2013Accepted 17 January 2013Available online

Keywords:SpeleothemsIndian Ocean MonsoonSocotraPaleoclimateStable isotopesTrace elements

a b s t r a c t

Four stalagmites covering the last 7.0 ka were sampled on Socotra, an island in the northern Indian Oceanto investigate the evolution of the northeast Indian Ocean Monsoon (IOM) since the mid Holocene. OnSocotra, rain is delivered at the start of the southwest IOM in MayeJune and at the start of the northeastIOM from September to December. The Haggeher Mountains act as a barrier forcing precipitationbrought by the northeast winds to fall preferentially on the eastern side of the island, where the studiedcaves are located. d18O and d13C and Mg/Ca and Sr/Ca signals in the stalagmites reflect precipitationamounts brought by the northeast winds. For stalagmite STM6, this amount effect is amplified by kineticeffects during calcite deposition. Combined interpretation of the stalagmites’ signals suggest a weaken-ing of the northeast precipitation between 6.0 and 3.8 ka. After 3.8 ka precipitation intensities remainconstant with two superimposed drier periods, between 0 and 0.6 ka and from 2.2 to 3.8 ka. No link canbe established with Greenland ice cores and with the summer IOM variability.

In contrast to the stable northeast rainy season suggested by the records in this study, speleothemrecords from western Socotra indicate a wettening of the southwest rainy season on Socotra after 4.4 ka.The local wettening of western Socotra could relate to a more southerly path (more over the IndianOcean) taken by the southwest winds. Stalagmite STM5, sampled at the fringe between both rain areasdisplays intermediate d18O values. After 6.2 ka, similar precipitation changes are seen between easternSocotra and northern Oman indicating that both regions are affected similarly by the monsoon. Differentpalaeoclimatologic records from the Arabian Peninsula currently located outside the ITCZ migrationpathway display an abrupt drying around 6 ka due to their disconnection from the southwest rain in-fluence. Records that are nowadays still receiving rain by the southwest winds, suggest a more gradualdrying reflecting the weakening of the southwest monsoon.

! 2013 Elsevier Ltd. All rights reserved.

1. Introduction

The seasonal migration of the Intertropical Convergence Zone(ITCZ) and coupled monsoon systems influences more than half of

the world population. For the countries around the Arabian Sea, atthe northern limit of the ITCZ-pathway, a good understanding ofthe Indian Ocean Monsoon (IOM) and its summer and wintersubsystems is of major importance, especially considering thepredicted further drying (Fleitmann et al., 2007; Kropelin et al.,2008).

Around thenorthern IndianOcean, informationonchanges in theIOM wind direction and strength during the late-Pleistocene and

* Corresponding author. Tel.: !32 2 6293397; fax: !32 2 6293391.E-mail address: [email protected] (M. Van Rampelbergh).

Contents lists available at SciVerse ScienceDirect

Quaternary Science Reviews

journal homepage: www.elsevier .com/locate/quascirev

0277-3791/$ e see front matter ! 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.quascirev.2013.01.016

Quaternary Science Reviews 65 (2013) 129e142

Maite Van Rampelbergh

69




Chapter 3: Spelothems from Socotra

Delta:1_surname

Delta:1_surname

Delta:1_surname

Delta:1_surname

Delta:1_surname

mailto:[email protected]

www.sciencedirect.com/science/journal/02773791

http://www.elsevier.com/locate/quascirev

http://dx.doi.org/10.1016/j.quascirev.2013.01.016



Holocene have mostly been extracted from Indian Ocean sedimentcores (Sirocko et al., 1993; Gupta et al., 2003; Ivanochko et al., 2005)and fromdunedeposits in the southof theArabianPeninsula (Radieset al., 2005; Parker et al., 2006; Lezine et al., 2010). However, in thesearchives, radiocarbon ages with relative high uncertainties impedethe precise determination of the timing and duration of IOM varia-tions. Tree rings, allowing higher resolution dates, have proveduseful around the Indian Ocean, but they are limited to the last 1000years (Cook et al., 2010). The lacking information on hydrologicalpatterns can be retrieved from speleothems (Neff et al., 2001;Wanget al., 2001; Fleitmann et al., 2003a; Shakun et al., 2007). Speleo-thems have a relatively fast and often continuous growth over longtime periods allowing the elaboration of long-term records at highresolution (McDermott, 2004; Fairchild et al., 2006).Most important,speleothems can be precisely dated both by counting annual layers(if present) and/or by using the U/Th-dating method, which makesthem powerful archives to date important climatic, historical orcultural eventsprecisely (Wanget al., 2001;Henderson, 2006; Zhanget al., 2008; Cheng et al., 2009a).

Most of available Holocene palaeoclimatological records forthe area reflect variations in the summer IOM subsystem becausethe highest amounts of precipitation are associated with thesummer IOM, also known as the Asian summer monsoon. Con-sequently, the precipitation signals brought by the winter ornortheast IOM subsystem are often overwritten by the southwestsignal.

In this study, we present four new high-resolution stalagmiterecords sampled on the eastern side of Socotra Island (Yemen)(Fig. 1). Rainfall on the eastern side of the island mainly consists ofnortheast winter monsoon precipitation due to orographic effects(Scholte and De Geest, 2010). Eastern Socotra constitutes thereforean ideal location to study the changes in the northeast or winterIOM subsystem (source, directions, amounts). Different studiesaround the Arabian Sea have already shown that speleothem d18Ovalues reflect changes in palaeo-precipitation intensities(Fleitmann et al., 2003b, 2004a, 2007; Shakun et al., 2007). In thisstudy, several proxies (d18O, d13C, Mg/Ca and Sr/Ca ratios) aremeasured on selected stalagmites to complete the current under-standing of the northeast IOM subsystem around the NorthernIndian Ocean during the mid- to late Holocene (7 kae0 ka). A betterinsight in the mid- to late Holocene palaeoclimatic and environ-mental evolution of Socotra constitutes an important step to un-derstand regional climate dynamics.

2. Regional setting

2.1. Site location and present climate

Socotra Island lies in the northwestern part of the Indian Ocean,between the horn of Africa and the Arabic Peninsula (Fig. 1). Meanannual rainfall and temperature measured by a network of 11meteorological stations from 2002 to 2006, are 216mm and 28.9 "Crespectively, referring to a semi-arid tropical climate (Scholte andDe Geest, 2010) (see also Fig. 1). The present-day climate onSocotra Island is governed by the seasonal migration of the ITCZand episodic passages of tropical cyclones as well as oceaneatmosphere interactions, such as the Indian Ocean Dipole (IOD)and the El Nino-Southern Oscillation (ENSO) (Cheung et al., 2006).From the western Pacific to the Indian Ocean, the ITCZ differs fromits traditional notion as a narrowwell-defined cloud band. The ITCZabove the Indian Ocean is broader in latitude with significantlymore horizontal spatial variation (Waliser and Gautier, 1993).Because of its atypical configuration, discussion can arise on theexact use of the term ITCZ in region around the northern IndianOcean. In this study, the ITCZ forms part of the ascending branch ofthe Hadley cell and is defined as the convergence zone betweensouthwest and northeast winds where precipitation is fairly high.This definition is similar to the ITCZ definition used in Fleitmannet al. (2003a, 2007).

In boreal summer, the low-pressure cell centred above thesouthern foothills of the Tibetan plateau pulls the ITCZ north until itreaches its northernmost position in July. During boreal autumn,the pressure gradients reverse and the ITCZ retreats southwarduntil it reaches its southernmost position in January (Fleitmannet al., 2004b). This annual migration of the ITCZ creates two windpatterns on Socotra, known as the southwest summer and thenortheast winter monsoon period interrupted by two short tran-sition periods whenwinds from all directions ensue. The southwestmonsoon season begins in early May (Fig. 2a) with a stablesouthwesterly and moisture loaden wind (course w230") comingfrom the Indian Ocean. In July, the ITCZ reaches its northernmostposition, wind speed increases and the airflow takes a morewesterly path (w240") across the semi-desert of Somalia. Strongand dry southwest monsoonal winds blow over the island. InSeptember at the end of the summer monsoon season, wind di-rection swings back to w230", wind speed decreases and windsfilled with moisture from the Indian Ocean deliver precipitation

Fig. 1. Location of the island of Socotra in the northern Indian Ocean. Black dots represent the location of the caves. The numbers represent the location of the 11 meteorologicalstations studied in Scholte and De Geest (2010). The watershed is represented by the dotted line.

M. Van Rampelbergh et al. / Quaternary Science Reviews 65 (2013) 129e142130



70






(Culek et al., 2006). After an autumn transition period starting inOctober, the northeast monsoon starts in the first half of November(Fig. 2b). The northeast monsoon onset, however, is not as suddenas that of the southwest summer monsoon due to the lower

pressure gradient between the high-pressure cell above the Tibetanplateau and the high-pressure cell near Madagascar. During thisperiod, rainfall reaches a maximum averaging of 120 mm (or 42% ofthe mean annual rainfall) as winds transport moisture originating

Fig. 2. a) The black star indicates the location of Socotra. Wind patterns and ITCZ location for the northern Indian Ocean during (1) northeast winter monsoon in January, (2)southwest rainy season in May, (3) southwest summer monsoon in July and (4) northeast rainy season in November (adapted after Fleitmann et al., 2004a). b) Socotra’s monsoonand precipitation intensities for every month derived from NASA’s Tropical Rainfall Measuring Mission. Purple bars indicate percentages of the mean annual rainfall of 100.3 mm.The black and dotted lines indicate the percentage of monsoon strength for both monsoon periods. Grey bars indicate the transition periods.

M. Van Rampelbergh et al. / Quaternary Science Reviews 65 (2013) 129e142 131



71





from the warm Arabian Sea towards Socotra Island (Scholte and DeGeest, 2010). In early February, the ITCZ starts tomigrate back northand the northeast monsoon weakens. The spring transition occursat the end of March with winds blowing from both northwesternand southwestern directions. At the beginning of April, southwestwinds become dominant again, cumulus clouds form and bring rain(Culek et al., 2006). In summary, Socotra experiences two distinctrainy seasons (typically from April to June and from November toDecember) interrupted by two dry seasons (typically from Januaryto March and June to October) (Fig. 2b). Since the ITCZ is defined asthe convergence zone between northeast and southwest winds, thetwo Socotran rainy seasons are associated with the passage of theITCZ over the island.

Based on detailed analyses of cloud cover satellite images forSocotra Island (Fig. 1), the northern and southern plateaus experi-ence different precipitation regimes (Scholte and De Geest, 2010).The 400e600 m high limestone cliffs at the northern and southerncoast and elevated plateaus around the Haggeher Mountains causeorographic uplift and induce two distinct rain areas on both sides ofthemountain range. The northern and eastern parts receive most oftheir rain during the northeast rainy season in November (Fig. 1)whereas the western and southern coasts receive almost equalamounts of rain during both rainy seasons (Scholte and De Geest,2010). Sporadically, during strong southwest spring rainy years,the northern regions, which are normally located in the rainshadow of the Haggeher Mountains, can be influenced by south-west rains (Culek et al., 2006).

Based on the geology and geomorphology, Socotra Island issubdivided into three zones: the Quaternary alluvial coastal andinland plains, the PalaeoceneeEocene reef-limestone plateaus andthe Precambrian granitic Haggeher Mountains (Cheung et al.,2006). The limestone plateaus, covering approximately half ofSocotra’s surface, are strongly karstified and harbour numerouslarge cave systems (Cheung et al., 2006). The main karst areas arethe Momi karst plateau in the east, the Diksam/Sibehon karst pla-teau in the centre and the Ma’alah karst plateau in the north-western part of the island (Fig. 1).

Three stalagmites were sampled in Hoq Cave (12"35011.900N;54"21015.4400E, elevation 335 masl), located on the Momi karstplateau, 5 km from the northeast coast (Fig. 1). The entrance, fullyfacing the seaside, is around 45 m wide and 30 m high. The firststalagmite (specimen Hq1) was sampled in 2000, 200 m from theentrance in a chamber that often experienced strong ventilationand varying humidity, as observed during fieldwork. Two coevalstalagmites, STM1 and STM6, were sampled 1 m next to each otherin 2003 and 2006 respectively, approximately 2 km from the onlyknown entrance, where ventilation is minimal. At this site, cave airtemperature remained constant throughout the year at 25 # 0.5 "C(continuously monitored between January and December 2003).Relative humidity measured during six visits between 2003 and2006 was always higher than 98%.

Stalagmite STM5 was collected from Casecas Cave(12"33020.0200Ne54"18033.3400E, elevation 542 masl) in January2004, 6 km southwest from Hoq Cave (Fig. 1), also on the Momikarst, in an upper gallery where ventilation is expected to beminimal. Temperature and humidity measured during samplingwere 29 # 0.5 "C and above 95%.

2.2. Palaeoclimate of the region

Changes in palaeorainfall in the areas located at the northernfringe of the Indian Ocean Monsoon domain are generallyexplained by variations in the strength of the IOM and the asso-ciated boreal summer position of the ITCZ (Fleitmann et al., 2003a,2007). Various Late Pleistocene palaeoclimatic studies using

different archives from northeast Africa (Gasse, 2000), southernArabia (Burns et al., 1998, 2001; Preusser et al., 2002; Fleitmannet al., 2003a, 2003b, 2004a; Parker et al., 2006; Fleitmann et al.,2007; Fuchs and Buerkert, 2008; Lezine et al., 2010), the ArabianSea (Gupta et al., 2003) and China (Dykoski et al., 2005; Wang et al.,2005), demonstrate that changes in the position of the ITCZ aregoing along with variations in IOM rainfall intensity. During theGlacial to Lateglacial, the ITCZwas considered to be located south ofthe Arabian Peninsula and IOM rainfall did not reach southernArabia as indicated by the absence of speleothem growth (Burnset al., 2001; Fleitmann et al., 2007), lack of lacustrine deposits(Lezine et al., 2010) and homogenous sedimentation rates of aeo-lian sediments (Fuchs and Buerkert, 2008). A northward shift in themean latitudinal position of the summer ITCZ and orbitally forcedintensification of the IOM during the early Holocene led to theonset of a humid period. This Holocene wet optimum is welldocumented in northeast Africa and southern Arabia and charac-terized by widespread formation of lakes (Gasse, 2000) andenhanced speleothems deposition (high effective moisture) (Burnset al., 1998, 2001; Fleitmann et al., 2003a, 2007). The termination ofthe Holocene wet optimum is associated with a southward dis-placement of the ITCZ to its present-day position along the coast ofsouthern Arabia (Fleitmann et al., 2007). Discrepancies in thetiming and duration of this humid period remain. Speleothemsfrom Oman and Yemen suggest an onset at 10.5 ka whereas datafrom Yemen (Lezine et al., 2007) suggest an onset at 12 ka or in theUAE an even later onset at 8.5 ka (Parker et al., 2006). However,most of the studies from southern Arabia confirm a period ofmaximal rainfall at 8 ka. According to sediment records fromnorthern Oman (Fuchs and Buerkert, 2008), speleothems fromChina (Dykoski et al., 2005; Wang et al., 2005), Oman (Fleitmannet al., 2003a, 2007) and the findings of Lézine et al. (2010) inYemen, a gradual long term decrease in precipitation starts shortafter 8 ka. In contrast, Burns et al. (2001) identified a reduction inrainfall at ca 6 ka in northern Oman as well as in the UAE whereParker et al. (2006) placed the Holocene wet period termination at6 ka. These partially contrary results from various studies might bedue to differences in the sensitivity of the investigated archives andto a more complex pattern of atmospheric conditions at a localscale, as will also be shown further in this study.

Contrary tothe long termdecrease inprecipitationon theArabianPeninsula, a speleothem from western Socotra (Dimarshim Cave;Fig. 1) shows increasing humid conditions since 4.4 ka (Fleitmannet al., 2007). This anticorrelation between western Socotra and theArabian Peninsula is explained by Fleitmann et al. (2007) as a resultof a progressing southward displacement in the mean latitudinalsummer position of the ITCZ, implying a decrease in precipitation inthe areas located at the northern fringe of the IOM, but an increase inareas closer to the equator (Fleitmann et al., 2007).

3. Materials and methods

For analyses, a slab of 1 cm was cut from the middle of eachstalagmite parallel to its growth axis using a diamond saw. Theslabs were polished using carbide powder and finished with Al2O3-powder. Twenty-six (13 for STM1, 7 for STM6 and 6 for STM6) U-series age determinations were carried out at the University ofMinnesota (USA), using the procedures for uranium and thoriumchemical separation and purification described in Edwards et al.(1987) and Cheng et al. (2000, 2009a, 2009b). Age determinationson Hq1 were carried out at the University of Bern. Details on ana-lytical methods are given in Fleitmann et al. (2007). Age models areestablished using the StalAge algorithm (Scholz and Hoffmann,2011) and are reported with their uncertainty. All ages areexpressed in a B2011.



72



Samples for d13C and d18O measurements in STM1, STM5 andSTM6 were drilled along the stalagmites central axis with a Mer-chantec Micromill. Ethanol was used to clean the speleothem sur-face and drill bit prior to sampling. Between every sampling, thedrill bit and sampling surface were blown clean with compressedair. Sample resolution in STM1 was 500 mm for the upper 32 mm,from 48 to 51 mm and from 177 to 248 mm. The remaining partswere sampled at a 1 mm resolution. Stalagmites STM5 and STM6were sampled every 500 mm. Stable isotope sampling in the Hq1stalagmite was carried out in the upper 305mm every 600 mm. OneHendy test was carried out on STM1 and three on STM5 (Hendy,1971) by drilling samples along an individual growth layer. Toobtain samples of modern precipitated calcite, 6 glass slabs wereplaced in Hoq Cave for approximately one year between January2003 and May 2005; 2 slabs rested on top of STM6 and 4 at the endof the cave. Stable isotope analyses on STM1, STM5, STM6 and onfresh calcite from the glass slabs were carried out at the VrijeUniversiteit Brussel with a Kiel-III-device coupled on a ThermoDelta plus XL. Analytical uncertainties (1s) were $0.06& for d13Cand $0.08& for d18O.

For isotopic analyses on stalagmite Hq1 approximately 5 mg ofpowder was drilled from the sample and analysed with an on-line,automated, carbonate preparation system linked to a VG Prism IIisotope ratio mass spectrometer at the University of Bern. Repro-ducibility of standardmaterials is 0.08& (1s). All isotopic values arereported in per mille (&) relative to Vienna Pee Dee Belemnite(VPDB).

Samples for elemental concentration determination of Ca, Mgand Sr were taken every 5 mm along the growth axis of STM1 andSTM6 using a carbide dental drill (1 mm diameter). A total of 111samples (ca 15 mg) for STM1 were analysed by Atomic AbsorptionSpectrometry at the Vrije Universiteit Brussel with analytical un-certainties (2s) less than 5%. 32 samples (ca 5 mg) for STM6 weremeasured on an Element 2 HR-ICP-MS at the Royal Museum forCentral Africa (Brussels, Belgium) with analytical uncertainties (2s)less than 5%.

Three seepagewater samples from Casecas Cave and 21 samplesfrom Hoq were collected for d18O measurements. The water sam-ples were prepared using the CO2/H2O-equilibration methoddescribed by Epstein and Mayeda (1953). Measurements wereperformed on a Finnigan Delta E mass spectrometer at the VrijeUniversiteit Brussel. All values are reported in per mill (&) relativeto Standard Mean Ocean Water (SMOW). Analytical uncertainties(2s) were less than 0.10&.

4. Results

4.1. Hoq Cave

The stable isotopic compositions of calcite deposited on glassslabs in the deepest parts of the cave (near STM6) display largevariations between %2.15& and %4.22& for d18O and between%3.18& and %7.86& for d13C (Table 1). No link can be establishedbetween the measured isotopic values and the slabs location in thecave. A similar large range in d18O values is also observed in the 21seepage waters ranging between %1.36& and %4.26& (Fig. 3).

Stalagmite Hq1, sampled only 200 m from the wide entrance, isa 381 mm tall stalagmite that was still actively dripping whencollected in 2000. Nine U-series dates carried out on the upper290mm (Table 2a) indicate continuous growth at an average rate of32 mm/yr. According the results given by StalAge, large age un-certainties occur after 6.9 ka (Fig. 4). Because the timing of events islargely insecure after 6.9 ka, no climatic interpretations are basedon that part. The d18O values vary between %3.06& and %0.70&around an average of%1.69& (Fig. 5). The d13C values vary between

%8.69& and 2.41& around an average of %0.50& and correlatesignificantlywith the d18O values (r& 0.83 and p& 3.5758' 10%129).Two periods of very positive isotopic values occur from 0.7 to 1.5 kaand from 3.9 until 6.9 ka.

Stalagmites STM1 (length 520 mm, sampled January 2003) andSTM6 (length 160 mm, sampled January 2006) were activelydripping when sampled at the end of the cave. Calcite fabric in bothstalagmites varies between dark compact and white porous parts,with lamination only visible in the white porous parts. In STM1, at10 cm from the base, a shift in growth axis occurs (Fig. 4). Based on13 U-series dates (Table 2b) and the StalAge age model STM1 grewconstantly since 5.6 ka at a rate of around 87 mm/yr. In STM6, 7 U/Th-ages (Table 2b) show that the stalagmite grew constantly from4.5 ka (Fig. 4) with an average growth rate of 34 mm/yr, which issimilar to the growth rate of stalagmite Hq1 located about 2 kmaway, and three times slower than stalagmite STM1 only 1 m away.

The STM1 d18O values vary between %4.24 and %0.84&, arounda mean of %2.97& (Fig. 5). Similar d18O values are found in STM6varying between %5.06& and %0.60& (average at %2.27&). Thed13C values of STM1 vary between%12.11& and%3.64&, average at%8.41&. For STM6, the d13C values are slightly more positive,varying between %10.58& and %3.54& around an average at%5.88&. A good correlation can be established between the d18Oand d13C profiles in STM1 (r & 0.70 and p & 1.7988 ' 10%93) and inSTM6 (r & 0.55 and p & 8.6530 ' 10%26). Compared to stalagmiteHq1, sampled close to the cave entrance, STM1 and STM6 displaygenerally more negative d18O and d13C values. From the start of therecords until 3.8 ka, the STM1 and STM6 isotopic records display

Table 1Location of the 6 glass slabs in the cave, period of calcite deposition and d18O andd13C values of the fresh present-day calcite.

Fresh calcite from glass slabs

Location Period d13C(% VPDB)

d18O(& VPDB)

Start Stop

STM6 January 2003 January 2004 %7.7 %2.5STM6 November 2004 January 2006 %4.5 %2.1END October 2003 November 2004 %7.2 %3.4END October 2003 November 2004 %3.2 %2.2END November 2004 May 2005 %7.9 %4.2END November 2004 May 2005 %7.3 %4.1

Fig. 3. d18O composition of the 21 Hoq Cave (black dots) and 3 Casecas Cave (whitedots) seepage waters. Most negative values occur in November, when Socotra receivesmost of its rain, suggesting that the drip water d18O values are influenced by the‘amount effect’.




73






decreasing values with similar centennial andmillennial variations.After 3.8 ka, the d18O signal of both stalagmites varies around%2.5& without significant trend. Between 0 and 0.6 ka and from2.2 to 3.8 ka the STM6 record shifts to more positive values.

The Mg/Ca ratios ('103) of STM1 range from 7.03 to 13.83, withan average of 9.40 (Fig. 5). The STM1 Sr/Ca ratios ('103) vary be-tween 0.20 and 0.34 averaging 0.27 and correlate successfully withtheMg/Ca ratios (r& 0.65 and p& 6.5197'10%15) (Fig. 5). For STM6,similar conclusions can be established; the Mg/Ca ('103) valuesvary between 7.74 and 13.03 and average at 10.69. The Sr/Ca ('103)values vary between 0.21 and 0.31 and average at 0.26 and correlatewell with the Mg/Ca ('103) profile (r & 0.72, p & 6.6556 ' 10%6).

For all three stalagmites, calcite colour, growth rate and isotopicsignals appear to correlate, with darker compact calcite (indicative

of slowgrowth) coincidingwithmore positive d18O and d13C values.Lighter calcite is formed during faster growth periods and is char-acterized by more negative d18O and d13C values. In the coevalstalagmites STM1 and STM6, higher trace elemental concentrationscorrespond to darker calcite, slower growth rate and more positiveisotopic values. Due to the low-resolution trace elemental mea-surements in STM6, the covariationwith the d18O and d13C values isless clear.

4.2. Casecas Cave

Three seepage water samples were collected from Casecas Caveduring the dry winter monsoon season in January 2003 and 2004.Their d18O values vary between%1.97& and%3.01& (Fig. 3) and areslightly more negative than the January d18O values of Hoq Cave.STM5 was continuously deposited between January 2004 (date ofsampling) and 0.7 ka as is shown by the six 230Th ages (Table 2b).

Table 2aU/Th measurements for stalagmite Hq1 (University of Bern). All ages were converted to before 2011

Samplenumber

Depth(mm)

c (U) (ppb) c (Th) (ppb) 234U/238U 230Th/232Th 230Th/234U Age (ka)a Age (ka B2011)

Hq1-1 13.1 20,321.1 # 52.51 0.3160 # 0.0061 1.0000 # 0.0006 349.6907 # 7.4303 0.0018 # 0.0000 0.198 # 0.02 0.207 # 0.02Hq1-2 38.9 22,029.6 # 58.77 0.4610 # 0.0222 0.9931 # 0.0007 808.1328 # 39.1271 0.0056 # 0.0000 0.610 # 0.02 0.619 # 0.02Hq1-3 67.1 10,921.3 # 28.55 0.3595 # 0.0027 1.0136 # 0.0009 1341.2558 # 14.8675 0.0143 # 0.0001 1.572 # 0.03 1.581 # 0.03Hq1-4 110.3 22,383.2 # 58.84 2.4010 # 0.0133 1.0166 # 0.0009 602.1049 # 6.4128 0.0210 # 0.0002 2.319 # 0.03 2.328 # 0.03Hq1-5 157.1 13,513.6 # 35.45 6.5158 # 0.0354 1.0284 # 0.0009 195.7649 # 1.4870 0.0304 # 0.0002 3.357 # 0.03 3.366 # 0.03Hq1-6 191.9 8725.9 # 22.46 6.0375 # 0.0308 1.0007 # 0.0009 167.6249 # 1.2385 0.0384 # 0.0002 4.227 # 0.05 4.236 # 0.05Hq1-7 220.l 8048.1 # 21.26 6.1969 # 0.0445 1.0123 # 0.0010 196.0517 # 1.7976 0.0494 # 0.0003 5.524 # 0.06 5.533 # 0.06Hq1-8 250.1 12,223.9 # 31.18 4.5306 # 0.0248 1.0234 # 0.0006 514.2779 # 4.0313 0.0615 # 0.0004 6.927 # 0.06 6.936 # 0.06Hq1-9 290.3 18,368.8 # 47.14 1.0082 # 0.0060 1.0216 # 0.0007 3933.7280 # 34.9869 0.0686 # 0.0005 7.760 # 0.09 7.769 # 0.09

a Ages relative to AD 2002.

Fig. 4. Age versus depth plots and average growth rate of the 4 studied stalagmites.Black line represent the StalAge (Scholz and Hoffmann, 2011) age models. Grey linesindicate the uncertainties as modelled by StalAge.

Fig. 5. d18O (black line) and d13C (grey line) values in & VPDB for the studied sta-lagmites plotted against age in ka before 2011. Dots with error bars mark the U/Th ageswith their uncertainty. For STM1 and STM6, the Mg/Ca ' 103 record is indicated inblack and the Sr/Ca ' 103 record is indicated in grey.



74




STM5’s growth rate averages 123 mm/yr making this stalagmite thefastest growing speleothem of our four samples (Fig. 4). The large U/Th-error bars, especially in U/Th-samples STM5-1 and STM5-3, aredue to the high amounts of detrital 232Th, leading to large un-certainties in the age model (Fig. 4). Therefore, this record will onlybe used to discuss multi-millennial variations. STM 5’s d18O values,averaging%3.24&, range from%4.24& to%1.41& (Fig. 5). As for theseepage waters, these d18O values in STM5 are more negativecompared to the d18O values found in the Hoq Cave stalagmites. Thed13C values of STM5 range from %7.40& to %2.33& and average at%5.26&. They are more positive compared to the d13C values of theHoq Cave speleothems. A good correlation can be established be-tween the d18O and the d13C signal (r& 0.76 and p& 4.9724'10%34).

5. Discussion

5.1. Low d18O and d13C values indicate wetter conditions

Provided that calcite formed under conditions of isotopic equi-librium, the d18O of speleothem calcite is governed by the waterecalcite fractionation factor (0.25& decrease per 1 "C increase),and by the d18O of the cave-seepage water that in turn is deter-mined by the d18O of rainwater (Lachniet, 2009). In tropical andsubtropical areas, such as Socotra, the “amount effect”, describingthe inverse relationship between the amount of precipitation andits oxygen isotopic composition, is mainly responsible for changesin rainwater d18O (Dansgaard, 1964; Rozanski et al., 1992) andconsequently for the d18O composition of cave seepage waters. InSocotra, changes in the d18O of stalagmites deposited in or close to

isotopic equilibrium with their seepage waters, reflect fluctuationsin the amount of precipitation. In this study, the presence of the“amount effect” is clearly demonstrated by more negative d18Ovalues of the seepagewaters in November (Fig. 3), whenmost of therain falls on the island. The rest of the year, rainwater d18O valuesdisplay less negative values, independent of source of the rainindicating the absence of a strong ‘source effect’. Other studiesusing Socotran speleothems also interpreted the changes in d18Ocomposition as reflecting the “amount effect” with more negatived18O values occurring during wetter conditions (Burns et al., 2003;Fleitmann et al., 2007; Shakun et al., 2007).

The d13C of speleothems deposited in equilibrium is mainlydetermined by the isotopic composition of soil-CO2, which reachesthe cavewith the seepagewater and normally constitutes themajorcarbon source (Genty et al., 2001). Variations in the d13C compo-sition of soil-CO2 are mainly controlled by changes in the type ofvegetation cover in terms of C3/C4/CAM-plants above the cave(Smith and Epstein, 1970; Frumkin et al., 2000). If no major vege-tation changes occurred over the studied period, as is most likelythe case in this study, variations in the soil-CO2 d13C are primarilyrelated to the intensity of soil activity with heavier d13C valuesduring drier periods (Genty et al., 2003).

If the stalagmites were not deposited in full isotopic equilibrium,additional intra-cave mechanisms may have a distinct influence ond18O and d13C calcite values. A first way to investigate the equilib-rium conditions of speleothem calcite can be done by a Hendy-testwhere several samples are drilled along a single growth layer acrossthe stalagmite. If in one layer (1) a simultaneous enrichment in d18Oand d13C occurs away from the growth axis and (2) a good

Table 2bU/Th measurements for stalagmites STM1, STM5 and STM6 (University of Minnesota). All ages were converted to before 2011.

Samplenumber

Depth(mm)

238U (ppb) 232Th (ppt) 230/Th232Th(atomic ' 10%6)

d234Ua

(measured)

230/Th238U(activity)

230Th age (y)(uncorrected)

230Th age (y)b

(corrected)d234UInitial

c

(corrected)

230Th age(ka B2011)d

(corrected)

STM1 Hoq CaveSTM1-1 6 213.6 # 0.4 42 # 22 18 # 30 99.8 # 2.9 0.00041 # 0.00036 41 # 36 36 # 36 99.8 # 2.9 0.044 # 0.036STM1-2 50 161.4 # 0.3 86 # 24 140 # 44 92.6 # 2.9 0.00451 # 0.00059 450 # 60 440 # 60 92.7 # 2.9 0.448 # 0.060STM1-3 80 289.3 # 0.6 322 # 24 152 # 13 94.1 # 2.7 0.01025 # 0.00047 1027 # 47 998 # 49 94.4 # 2.7 1.006 # 0.049STM1-4 166 202.7 # 0.4 35 # 13 1590 # 570 100.2 # 2.1 0.01678 # 0.00054 1678 # 54 1673 # 54 100.7 # 2.1 1.681 # 0.054STM1-5 201 307.7 # 0.5 204 # 13 454 # 31 98.8 # 1.8 0.01821 # 0.00037 1824 # 38 1807 # 39 99.3 # 1.8 1.815 # 0.039STM1-6 219 192.0 # 0.4 38 # 17 1816 # 800 91.4 # 2.8 0.02164 # 0.00068 2186 # 70 2180 # 70 92.0 # 2.9 2.188 # 0.070STM1-7 302 185.3 # 0.3 58 # 13 1632 # 378 93.9 # 2.1 0.03074 # 0.00058 3112 # 60 3104 # 60 94.7 # 2.1 3.112 # 0.060STM1-8 352 197.1 # 0.3 33 # 14 3877 # 1590 98.2 # 2.2 0.03946 # 0.00070 3993 # 70 3990 # 72 99.3 # 2.2 3.998 # 0.072STM1-9 386 239.8 # 0.4 119 # 20 1420 # 240 106.6 # 2.6 0.04259 # 0.00081 4283 # 83 4270 # 84 107.9 # 2.6 4.278 # 0.084STM1-10 422 227.0 # 0.4 132 # 16 1342 # 168 100.5 # 2.5 0.04723 # 0.00080 4787 # 83 4772 # 84 101.8 # 2.5 4.780 # 0.084STM1-11 428 197.9 # 0.3 806 # 16 210 # 6 101.1 # 2.2 0.05189 # 0.00091 5268 # 95 5160 # 109 102.5 # 2.2 5.168 # 0.109STM1-12 487 180.3 # 0.4 197 # 21 816 # 88 91.3 # 3.1 0.0540 # 0.0011 5540 # 120 5510 # 120 92.7 # 3.1 5.518 # 0.120STM1-13 548 148.8 # 0.3 253 # 19 558 # 45 94.9 # 3.0 0.0568 # 0.0014 5816 # 150 5770 # 150 96.4 # 3.1 5.778 # 0.150STM6 Hoq CaveSTM6-1 17 122.6 # 0.1 46 # 1 376 # 19 105.6 # 1.9 0.0086 # 0.0004 848 # 38 838 # 38 106 # 2 0.838 # 0.038STM6-2 42 275.8 # 0.3 29 # 1 2140 # 74 100.1 # 1.7 0.0138 # 0.0002 1372 # 18 1369 # 18 100 # 2 1.369 # 0.018STM6-3 50 201.2 # 0.4 324 # 13 184 # 8 97.4 # 2.3 0.0180 # 0.0003 1803 # 28 1761 # 36 97.9 # 2.3 1.766 # 0.036STM6-4 82 111.3 # 0.1 46 # 1 1082 # 31 94.4 # 1.6 0.0272 # 0.0004 2747 # 41 2736 # 41 95 # 12 2.736 # 0.041STM6-5 100 225.4 # 0.4 249 # 12 527 # 26 93.7 # 2.2 0.03528 # 0.00036 3579 # 38 3549 # 40 94.6 # 2.2 3.554 # 0.038STM6-6 166 185.4 # 0.5 506 # 17 281 # 10 94.2 # 3.0 0.0465 # 0.0005 4737 # 53 4664 # 64 95.4 # 3.0 4.669 # 0.053STM5 Casecas CaveSTM5-1 12 790 # 2 7237 # 30 5.5 # 0.3 173.7 # 1.6 0.00315 # 0.00018 293 # 16 66 # 115 173.7 # 1.6 0.073 # 0.115STM5-2 15 741 # 1 884 # 18 33 # 2 231.7 # 2.2 0.0024 # 0.0001 212 # 9 184 # 22 232 # 2 0.184 # 0.022STM5-3 31 871 # 2 7333 # 147 10 # 1 203.2 # 2.4 0.0053 # 0.0001 484 # 10 280 # 145 203 # 2 0.280 # 0.145STM5-4 52 689 # 1 3073 # 35 23.4 # 1.1 165.4 # 1.9 0.00631 # 0.00029 593 # 28 481 # 62 165.6 # 1.9 0.488 # 0.062STM5-5 64 595 # 1 579 # 12 104 # 3 185.9 # 2.4 0.0061 # 0.0001 566 # 9 542 # 19 186 # 2 0.542 # 0.019STM5-6 88 617 # 1 4490 # 28 24.6 # 0.7 70.7 # 1.6 0.01083 # 0.00032 1110 # 33 910 # 110 70.8 # 1.6 0.917 # 0.110

230Th dating results. The error is 2s. l230 & 9.1577 ' 10%6 y%1, l234 & 2.8263 ' 10%6 y%1, l238 & 1.55125 ' 10%10 y%1.Corrected 230Th ages assume the initial 230Th/232Th atomic ratio of 14.4 # 2.2 ' 10%6.Those are the values for a material at secular equilibrium, with the crustal 232Th/238U value of 3.8. The errors are arbitrarily assumed to be 50%.

a d234U & [234U/238U]activity ' 1000.b Ages relative to AD 2003 for STM1, AD 2006 for STM6 and AD 2004 for STM5. Samples STM6-1, STM6-2, STM6-4, STM5-2, STM5-3 and STM5-5 are relative to AD 2011.c d234Uinitial was calculated based on 230Th age (T), i.e., d234Uinitial & d234Umeasured ' el234'T.d Ages before AD 2011.



75




correlation between both stable isotopic signals can be established,calcite precipitation was affected by kinetic fractionation (Hendy,1971). One Hendy-test was carried out on STM1 and three onSTM5 (Fig. 6). The Hendy test of STM1 displays constant d18O andd13C signals within 20 mm of the stalagmites apex and an increasefurther away. The three Hendy tests in STM5 are very different fromeach otherwhich is partially due to the difficulty of samplingwithinthe very thin layers. STM5(75mm) fails theHendy test by displayingincreasing d18O and d13C values away from the apex. For the Hendytests STM5(9 mm) and STM5(90 mm), d18O and d13C values remainconstant within 10mmof the stalagmites apex. Only the d18O signalin STM5 (90 mm) on the right side of the apex displays decreasingvalues before displaying an increase after 30 mm from the apex.Although, STM1 and STM5 theoretically fail the Hendy-test, therather constant d18O and d13C values close to the stalagmites’ apexand the weak correlation suggest fragile equilibrium near the cen-tral axis. However, different studies have pointed out that even ifstalagmites pass the Hendy test, speleothem calcite could still havebeen deposited out of isotopic equilibrium (Dorale and Liu, 2009;Mühlinghaus et al., 2009). Better is to look at the enrichment of d13Cvalues along an individual growth layer as indicator of dis-equilibrium (Mühlinghaus et al., 2009). As discussed in the Hendytest, the d13C values remain rather constant within 20 mm of theapex for STM1 andwithin 10mm of the apex for STM5 and increasetowards the sides of the stalagmite suggesting fragile equilibriumconditions of the deposited calcite. The fragile equilibrium condi-tions andpresence of kinetic effects in the studied stalagmites is alsosuggested by a strong correlation between the d18O and the d13Csignals in each stalagmite, indicating that similar processes, mostlikelybykinetic effects, influencebothproxies. A third test providinginformation on the degree of isotopic equilibrium is to calculate theexpected d18O and d13C values based on the theoretical “equilib-rium” waterecalcite fractionation factors for C and O (Kim andO’Neil, 1997), the present-day cave temperature and the isotopiccompositions of seepage waters and to compare these results withthe measured d18O and d13C values. For Hoq Cave, the large range(2.89&) in d18O values measured for the seepage waters (Fig. 3)hampers any meaningful modelling of this kind. Moreover, freshcalcite deposited on glass slabs also displays large variations inisotopic composition (Table 1). Such large variations do suggest thatsite specific differences such as different groundwater flow-pathsand the water residence time in the epikarst (storage water versusevent water) can partially be responsible for these large variations.

Also strong locally varying intra-cave factors such as degree of hu-midity and in particular ventilation may be responsible for theselarge variations in isotopic composition of the seepage waters andthe present-day calcite within one cave. Also for Casecas Cave, nosignificant modelling of this kind is possible because the seepagewaters were sampled in January only, which is the dry season andwill consequently lead to too heavy calculated d18O values for theexpected speleothem calcite.

Taken the evidences together, stalagmites in Hoq Cave andCasecas Cave were deposited under fragile equilibrium conditionsand the isotopic signals may partially be influenced by kineticeffects.

To summarize, any above described mechanism can/will influ-ence the d18O and d13C signal of the studied stalagmites in the samedirection; higher d18O and d13C values will always occur duringdrier conditions. Similar conclusions have been established inprevious work on speleothems in Yemen and Oman (Burns et al.,2001; Fleitmann et al., 2003a, 2004a, 2007; Shakun et al., 2007).

5.2. High Mg/Ca and Sr/Ca ratios indicate drier conditions

Mg/Ca ratios and to a lesser extend Sr/Ca ratios can be used ashydrological proxies (Fairchild and Treble, 2009). In arid and semi-arid areas, prior calcite precipitation (PCP) is considered to bea main reason for variable Mg/Ca and Sr/Ca ratios in speleothems(Fairchild et al., 2000, 2006). When downward percolating waterencounters a zone with lower pCO2, degassing occurs and calcitecan precipitate. Consequently, Mg and Sr become enriched com-pared to Ca in the residual water. During drier periods PCP isenhanced as aerated zones increase in the aquifer and residencetime of the water becomes longer (Fairchild et al., 2000). The mainevidence for PCP is covarying Mg/Ca and Sr/Ca ratios (McMillanet al., 2005; Johnson et al., 2006). Because of calcite precipitationin the epikarst during PCP, d13C calcite values increase in tandemwith Mg/Ca and Sr/Ca ratios. The strong similarities between theMg/Ca and Sr/Ca profiles in stalagmites STM1 and STM6 suggestthat PCP is a primary control for trace elemental ratios in bothstalagmites (Fig. 5). This assumption is further validated by thesignificant correlation between d13C and Mg/Ca (r & 0.33,p & 5.8112 ' 10%4) ratios in STM1. Furthermore, d18O values alsoshow a significant correlation with Mg/Ca (r & 0.38,p & 6.3180 ' 10%5) and Sr/Ca (r & 0.42, p & 8.6029 ' 10%6) ratios instalagmite STM1. As mentioned before, the low resolution of the

Fig. 6. d18O (black line) and d13C (grey line) values in & VPDB for the Hendy tests carried out on stalagmites STM1 (b) and STM5 (a and d). Speleothem calcite of STM1 is depositednear equilibrium conditions within 20 mm from the stalagmites apex. Due to very thin layers in STM5, sampling was difficult explaining the less successful results. STM5(75 mm)fails the Hendy test while STM5 (9 mm) and STM5(90 mm), displays rather constant values within 10 mm of the stalagmites apex.



76





STM6 Mg/Ca and Sr/Ca time series hampers a meaningful correla-tion with the d-signals. Nevertheless lower Mg/Ca and Sr/Ca ratiosappear to correspond roughly to more negative isotope values.Based on our observations, we suggest that Mg/Ca and Sr/Ca ratiosare sensitive hydrological proxies with higher ratios during drierperiods when PCP is enhanced.

5.3. Kinetic effects as amplifier of the rainfall signal

As indicated by the different equilibrium tests and by the strongcorrelation between the trace elements and the stable isotopesignals, calcite deposition of the studied stalagmites is affected bykinetic fractionation. As discussed in Dreybrodt and Scholz (2011)and Dreybrodt (2011), the degree of kinetic isotopic enrichmentof the deposited calcite is not caused by rapid CO2 degassing butmainly depends on drip rates and calcite precipitation rates, whichin turn depends on the calcite supersaturation. Since the isotopicprofiles of stalagmites STM1 and STM6 sampled at the end of thecave differ strongly from the isotopic signals of stalagmite Hq1,sampled near the cave entrance, we expect that different kineticeffects affect calcite deposition. At the end of 2 km-long Hoq Cave,changes in the d18O and d13C signals of the coeval stalagmites STM1and STM6 are interpreted to be controlled by variations in effectivemoisture, with lower isotopic values indicating higher net precip-itation. However, despite their close proximity (w1 m), the isotopicprofiles of stalagmites STM1 and STM6 are not identical. The moststriking difference occurs from 0 to 0.6 ka and from 2.2 ka to 3.8 kawhen the STM6 isotopic profiles shift to positive values differing by1.5& for d18O and 3& for d13C with the STM1 isotopic record(Fig. 7). A similar isotopic variation range is also observed in freshcalcite deposited on glass slabs (Table 1), in this case 2& for d18Oand 4.5& for d13C. This suggests that despite the high humidity andreduced ventilation at the end of the cave, strong differences inisotopic composition occur between the different drip sites. Duringdry periods drip rates will decrease at the fast growing (STM1) and

slow growing (STM6) sites. However, calcite precipitation at theSTM6 site will be more strongly affected by kinetic effects, as itsgrowth rate is considerably lower compared to the one at the STM1site. The effect of drip rate on the degree of kinetic isotopicenrichment of the deposited calcite is confirmed by differentmodelling (Dreybrodt, 2011; Dreybrodt and Scholz, 2011; Deiningeret al., 2012) and laboratory experiments with synthetic carbonates(Polag et al., 2010; Day and Henderson, 2011). Stalagmite STM6 isthusmore sensitive to small reduction in precipitation and effectivemoisture. Because of its faster growth rate, STM1 keeps growingduring these slightly drier periods and calcite precipitation stilloccurs closer to isotopic equilibrium. Deep in the cave, the slowerthe growth rate, the stronger calcite deposition will be sensitive tosmall changes in kinetic effects and the more the rainfall signalsinduced by the amount effect will be amplified.

Compared to the coeval stalagmites STM1 and STM6, the iso-topic signals of Hq1 vary around much more positive values. Also,the isotopic records of Hq1 display no similarities onmillennial andcentennial scale with the records from the coeval stalagmites STM1and STM6 (Fig. 7). These observations suggest very strong isotopicdisequilibrium deposition of the Hq1 calcite that is further con-firmed by its very strong correlated d18O and d13C signals. Com-pared to the coeval stalagmites sampled at the end of Hoq Cave,stalagmite Hq1 was sampled only 200 m from the entrance whererelative humidity is low and ventilation effects are considerablystronger. Thewide Hoq Cave entrance is located on the face of a cliffwith a large opening towards the downhill seaside allowing strongair circulation in the first parts of the cave. The presence of aircirculation in chambers near the entrance of Hoq Cave is validatedby the preferential growth direction of helictitese a small variety ofstalactites that are twisted and contorted with no apparent regardfor gravity. The stronger the air circulation in the chambers near theentrance, the stronger the evaporation effects on the stalagmitesurface and the further out of equilibrium the calcite will precipi-tate. The relationship between enhanced evaporation and dis-equilibrium deposition of the speleothem calcite has quantitativelybeen confirmed by Deininger et al. (2012). They demonstrated thatloss of water on the solution layer due to evaporation increases theCa2! leading to higher precipitation rates and consequently largerkinetic fractionation effects. Higher isotopic values in stalagmiteHq1 therefore most probably reflect stronger ventilation near thecave entrance.

To summarize, kinetic effects affecting the d18O and d13C signalsin the Hoq Cave stalagmites are related to their location in the cave.For stalagmites STM1 and STM6, sampled deep in the cave whereventilation is minimal, the d18O and d13C signals are affected bychanges in precipitation and effective moisture. Furthermore, theslower the stalagmite’s growth rate, the stronger the amplificationof the precipitation signal by the kinetic effects. For stalagmite Hq1,sampled near the entrance, kinetic effects are related to enhancedventilation.

The Hq1 record provides information on the intensity of ven-tilation near the cave entrance. For reconstructing the easternSocotra climate variability, the best records are given by the coevalSTM1 and STM6 stalagmites. Stalagmite STM6 is even more sensi-tive to small climatic variations because kinetic effects amplify itsclimate signal.

5.4. Evolution of the IOM and its northeast subsystem since 7 ka

Combined interpretation of the stalagmites STM1 and STM6provides the best precipitation reconstruction for eastern Socotra.Both stalagmites reflect variations in precipitation brought by thenortheast winds. From 6.0 ka until 3.8 ka, the STM1 and STM6 re-cords suggest a gradual decrease in precipitation brought by the

Fig. 7. Superimposed d18O and d13C values, both in & VPDB, of the Hoq Cave stalag-mites STM1 (black) STM6 (grey) and Hq1 (dotted black) together with their U/Thpoints (large dots). Two drier northeast monsoon periods with lower cave ventilationoccur from 0 to 0.6 ka and from 2.2 until 3.8 ka (indicated in grey), when stalagmitesSTM1 and STM6 break-up and stalagmite Hq1 shifts to more negative values.



77




northeast monsoon (Fig. 7). After 3.8 ka, no long-term trend isvisible and precipitation intensities brought by the northeast windsvary around a constant value. As suggested by the more sensitiveSTM6, two drier periods occur between 0 and 0.6 ka and from 2.2 to3.8 ka. Superimposed shorter-term millennial and centennial var-iations are similar in both stalagmites but can differ up to 100 yearsdue to age uncertainties in the agemodels. The exact forcing behindthe variations in precipitation brought by northeast winds remainsunclear. Its characterization would require comparison with othernortheast precipitation records in the region. The first limitationthat hampers such comparison is that records affected by onlywinter monsoon precipitation are difficult to find in the area. Formost of the sites in Monsoonal Asia, the amounts of precipitationbrought by the southwest monsoon are higher compared to thosebrought by the northeast winds. Consequently, the northeast signalis often overwritten by the southwest signal. Furthermore, mostexisting records such as ocean cores or sedimentary records do notprovide the high resolution needed to compare centennial or mil-lennial scale variations.

The only record allowing comparison with our northeast pre-cipitation records is a d18O stalagmite record from Hoti Cave,northern Oman (Fleitmann et al., 2007). This speleothem shows theevolution of rain brought by northeast winds for northern Oman atmillennial and centennial scale covering the period between 6 kauntil 5.2 ka and from 2.5 ka until present (Fig. 8). As for the coevalstalagmites STM1 and STM6, the d18O values in the northern Oman

stalagmite vary around%2.3& and display no significant long-termtrends. The presence of an important hiatus in the Hoti Cave record,makes a long-term comparison with STM1 and STM6 difficult.However, similar millennial scale variations can be found betweenboth regions (see also dotted lines in Fig. 8), confirming thatnorthern Oman and eastern Socotra are affected by similar mon-soon dynamics since 6.2 ka. Indeed, after 6.2 ka the southwardretreating ITCZ shifted south of northern Oman, making winterprecipitation brought by northeast winds the only moisture source(Fleitmann et al., 2007).

Comparison of the records reflecting northeast rain variationswith the Greenland ice core records (Rasmussen et al., 2006;Vinther et al., 2006) is difficult. On a long term, the Greenland icecores records display a gradual decrease that cannot be found in ourrecords (Fig. 8). This suggests that the long-term decrease in highlatitude temperatures doesn’t influence the long-term evolution ofthe northeast monsoon precipitation. Also on a shorter time scale,millennial and centennial variations don’t correlate between ourrecords and the Greenland ice core records. Also no links existswith the Bond events (Bond et al., 1997).

In contrast to the evolution of the northeast monsoon, recordsreflecting variations in southwest-monsoon precipitation showsimilarities with Greenland ice core records. Southern Oman sta-lagmites (Fleitmann et al., 2003a, 2007) display a long termweakening of the monsoon and shorter-term variations that cor-relate with Greenland ice core variations (Fig. 8). Colder Northernhemisphere periods correspond to weaker and consequently driersouthwest monsoon periods. Since our records do not matchthose of the Greenland ice core, we expect no similarities withprecipitation variations of the southwest monsoon. Indeed, com-parison between different southwest records (Sirocko et al., 1993;Cullen et al., 2000; Fleitmann et al., 2003a; Gupta et al., 2003) andthe northeast records from this study display no similarities(Figs. 8 and 10).

A speleothem d18O record fromwestern Socotra is interpreted toreflect variations in the southwest summer monsoon since 4.4 ka(Fleitmann et al., 2007). Comparison with our records from theeastern side of the island, show that both sides of Socotra havea different long-term evolution (Fig. 9a). Whereas western Socotragradually evolves towards wetter conditions (Fleitmann et al.,2007), eastern Socotra (this study) has a stable long-term precipi-tation trend since 3.8 ka. This emphasizes the important role of theHaggeher Mountains as a watershed creating two different pre-cipitation areas on the island. The presence of two precipitationareas on Socotra is confirmed by the intermediate d18O values ofstalagmite STM5 from Casecas Cave (Fig. 9a). The latter is located atthe fringe between the northeast rain area in the east and themixed southwest northeast rain area in the west (Fig. 1).

Despite the different long-term evolution of both sides ofSocotra Island, the shorter-term (millennial and centennial scale)variations between eastern and western Socotra display similarchanges. Due to age uncertainties, 200-year offsets occur betweenthe variations of western and eastern Socotra, causing an unsuc-cessful statistical correlation between the records. Such an offset isclearly visible for the positive peak around 2.6 ka where the STM6record lags approximately 200 years (thus within age un-certainties) behind the peak in the western Socotran record(Fig. 9a). The observation that shorter-termvariations are similar onboth sides of the island can be explained in two ways. The south-west rains also affects the eastern part of Socotra or the northeastrains also affect western Socotra. Nowadays, western Socotra re-ceives equal amounts of rain during the southwest and the north-east rainy season. For eastern Socotra, the rainfall amounts broughtby the northeast rains are three times the precipitation amountsbrought by the southwest rains (Scholte and De Geest, 2010).

Fig. 8. After 6 ka, the northern Oman record (Hoti Cave (3), Fleitmann et al., 2007)displays similar variations as the eastern Socotran records (STM1 and STM6 (2), thisstudy). Values vary around a similar average of %2.3& (dotted line) with similarmillennial scale variations. No similarities with the Greenland ice core record ((1),Rasmussen et al., 2006; Vinther et al., 2006) indicate that the northeast rains innorthern Oman and eastern Socotra are not sensitive to Northern Hemisphere tem-perature variations. No similarities can be seen between the northeast monsoon sig-nals from this study and a southwest monsoon precipitation signal ((4), Qunf Cave,Fleitmann et al., 2007) suggesting that both monsoons have different mechanisms.Values are reporten in & SMOW for (1) and in & VPDB for (2, 3 and 4).



78




Therefore, we conclude that the most plausible hypothesisexplaining the similar millennial and centennial scale variationsbetween east and west Socotra is that the northeast rains reach thewestern side of Socotra. Consequently, the millennial and cen-tennial scale variations in the d18O record fromwestern Socotra arecreated by variations in the northeast rainy season.

The long-term trend of the western Socotra record displays anevolution towards wetter conditions that cannot be seen in easternSocotra (Fig. 9a). Since the northeast monsoon does not display anincrease in precipitation after 3.8 ka, the long-term trend onwestern Socotra is most likely linked to increased southwestmonsoonal rainfall. This is apparently surprising considering thatfor Monsoonal Asia, the southwest monsoon is weakening from thebeginning of the Holocene (Sirocko et al., 1993; Neff et al., 2001;Fleitmann et al., 2003a, 2004a, 2007; Gupta et al., 2003;Wang et al.,2005). Modern wind direction measurements carried out on thewestern side of Socotra indicate that when the southwest monsoonreaches its maximal intensity in July, the southwest winds gradu-ally change direction from the south to southwest (Culek et al.,2006). The southwest winds are first passing over dry Somaliabefore hitting Socotra. This means that during the summer mon-soon, thus when the ITCZ reaches its northernmost position, airreaching Socotra is drier than during other periods when air isdirectly coming from over the Indian Ocean. The more northern theITCZ, the more the southwest winds are forced into awesterly path,and the drier the southwest rainy season on Socotra. Based on theseobservations, the following hypothesis for the increasing wetteningof western Socotra can be established. After 8 ka, the mean lat-itudinal position of the summer ITCZmoved southward in response

to the decreasing boreal summer insolation (Fleitmann et al.,2003a). As a consequence, the southwest winds that were origi-nally coming from over dry Somalia are forced into a more south-erly path over the Indian Ocean (Fig. 9b). The resulting winds willtherefore contain more moisture and lead to a wetter summermonsoon and thus an increase of precipitation on the western sideof Socotra. This hypothesis suggests that the increasingly wettersouthwest summer monsoon over Socotra reflects a local effect andis therefore not representative for the whole summer monsoonregion. More research and comparison with currently other highresolution regional late Holocene precipitation reconstructionsfrom around Socotra are necessary to confirm this hypothesis.However, so far such records are lacking.

5.5. End of the Holocene wet period

The similar forcing behind the northern Oman and the easternSocotra records sheds new light on the timing and characteristic ofthe termination of the Holocene wet period in southern Arabia.During the early to middle Holocene (6e10.5 ka), southern Arabiaexperienced clearly wetter conditions than today (Burns et al.,2001). Different proxy records such as from speleothems, lakeand dune deposits and marine sediments from the Arabian Sea donot seem to agree on the timing and characteristics of the end of theHolocene wet period. Roughly, these records can be subdivided intwo groups. The ones that suggest a gradual decrease in precipi-tation since 8 ka (Fig. 10, records 8e10) and those advocating anabrupt decrease around 6 ka (Fig. 10, records 1e7). The first group(Fig. 10, records 8e10) suggesting a gradual decrease since 8 ka

Fig. 9. (a) Whereas western Socotra becomes wetter over the last 4.4 ka (black line, Dimarshim cave, Fleitmann et al., 2007), conditions remain stable on eastern socotra after 3.8 kaas indicated by STM1 (light grey) and STM6 (dark grey) from Hoq Cave (this study). STM5 from Casecas Cave (dotted grey, this study) is influenced by both rainy seasons and hasintermediate d18O values between east and west Socotra. Superimposed millennial variations are similar for all stalagmites but can display offsets of 200 years (within the ageuncertainty ranges). (b) Possible hypothesis: southwest winds hitting western Socotra gradually come from more over the Indian Ocean due to the southward displacement of thesummer ITCZ causing a wetter southwest rainy season.



79




includes the speleothems from southern Oman (Fleitmann et al.,2007) and sedimentary cores from the Arabian Sea (Sirocko et al.,1993; Gupta et al., 2003). To the second group (Fig. 10, records 1e7) belong speleothems from northern Oman (Burns et al., 2001;Fleitmann et al., 2007), sediment records from the UAE (Parkeret al., 2006) and the Wahiba sands in northern Oman (Radieset al., 2005), all of them place the termination of Holocene wetperiod at around 6 ka. All locations showing an abrupt end of theHolocene wet period at 6 ka are currently located outside the ITCZmigration pathway and receive their precipitation only once a yearduring the winter from the northeast winds. Around 6 ka the ITCZshifted south of the UAE and northern Oman, disconnecting theseareas from the southwest rains and explaining the clearly markedend of the wet period. Winter and spring precipitation brought by

northeast winds is the only moisture source for that region. Ar-chives still located within the ITCZ migration pathway such assouthern Oman, the Yemenite coast and the northern Indian Oceanshow a rather gradual decrease in precipitation since 8 ka. Theseareas experience a gradual decrease in precipitation due to thegradual southward retreat of the summer ITCZ. In this matter, thetwo apparently opposite views can be reconciled.

Two lake records from the Yemenite lowlands and the Yem-enite Highlands in the west (Lezine et al., 2010) seem to form anexception to this hypothesis by showing an abrupt end of Holo-cene wet period at 7.2 ka (Fig. 10, record 4) and 5 ka (Fig. 10, record5). A more gradual decrease would be expected for this area sincethese lakes receive no rain from northeast winds. These abruptshifts correspond perhaps to a different timing of the threshold

Fig. 10. Compilation of mid Holocene records from the southern Arabian Peninsula and the northern Indian Ocean. Location of all records is shown on the map on top. The Hoq CaveSTM6 record (7) correlates well with the northern Oman speleothem record (6) suggesting that over the last 6.0 ka northern Oman received rain only by northeast winds. Recordscurrently located out of the ITCZ influence (1e4 and 6) display an abrupt end of the mid Holocene wet period around 6.0 ka (indicated by the dotted line) due to their disconnectionof the influence of the southwest monsoonal rains. Records still located within the ITCZ migration pathway (8e10) display a gradual decrease in precipitation due to the southwestmonsoon weakening. Sediment records from northwestern Yemen (5) still received rain after 6.0 ka due to orographic effects explaining the longer wet conditions for that area. TheHolocene boreal summer insolation curve is established using Analyseries (Paillard et al., 1996). 1. Awafi, Parker et al., 2006; 2. Hajar Mountains, Fuchs and Buerkert, 2008; 3. WahibaSands, Radies et al., 2005; 4. Al-Hawa, Lezine et al., 2010; 5. Rada and Saada, Lezine et al., 2010; 6. N-Oman, Fleitmann et al., 2007; 7. E-Socotra, this study; 8. S-Oman, Fleitmannet al., 2007; 9. Arabian Sea, Gupta et al., 2003; 10. Arabian Sea, Sirocko et al., 1993.



80




point for lake survival. At 7.2 ka BP rain brought by the southwardmigrating ITCZ was too weak to feed the Yemenite lowland lakeslocated in the rain shadow of the Yemenite Highlands, explainingthe abrupt drying. Due to orographic effects, lakes located in theYemen Highlands still received enough precipitation until thethreshold value was reached at 5 ka, when the amount of rainbrought by the southward migrating ITCZ also became insufficientto maintain the lake. A similar situation occurs with a playa-likesediment record from the Hajar Mountains (Northern Oman,Fuchs and Buerkert, 2008) (see also Fig. 10, record 2) that show anabrupt decrease in sedimentation rate at 8 ka while an abruptdecrease at 6 ka should be expected for this region. Althoughprecipitation decreased gradually until the abrupt shift at 6 ka innorthern Oman, the threshold point for these lakes was alreadyreached at 8 ka.

In summary, the end of the Holocene wet period at 8 ka insouthern Arabia is related to the continuous southward retreat ofthe summer ITCZ in response to the decreasing Holocene borealsummer insolation (Fleitmann et al., 2003a). Currently, areas stillreached by the summer ITCZ display a gradual end of the Holocenewet period starting around 8 ka. The areas that are not affectedanymore by the ITCZ and currently receive rain only once a yearonly from northeast winds display an abrupt end at 6.2 ka. Around6 ka, the summer ITCZ was located south of northern Oman andthus winter precipitation delivered by frontal depression systemsfrom the Mediterranean Sea became the dominant source ofmoisture. Climate in the western parts of Socotra evolved graduallytowards wetter conditions since 4.4 ka due to the trajectory of thesouthwest summer monsoon winds passing gradually more overthe Indian Ocean. However, since the currently available westernSocotra records do not cover the mid Holocene period, no robustconclusion can be established about the presence or absence ofa mid Holocene wet period on Socotra.

6. Conclusions

1. Strong differences in the isotopic composition of the seepagewater, modern calcite and between the isotopic profiles of theHoq Cave stalagmites demonstrate that kinetic effects are sitespecific and affect the isotopic composition of contempora-neously deposited stalagmites significantly. A detailed under-standing of the cave dynamics and waterecarbonateinteraction by cave monitoring, combined with a multi-proxyand multi-stalagmite approach is necessary to derive solidclimate reconstruction in semi-arid environments. At the endof Hoq Cave, kinetic effects are caused by variations in growthand drip rate whereas near the entrance, evaporation effectsare responsible for out of equilibrium deposition of the calcite.

2. Understanding the present local climate is necessary to cor-rectly interpret the obtained records. For Socotra, due to thewatershed action of the Haggeher Mountains, records obtainedfrom the southern and western parts reflect the long-termchanges of the southwest rainy season with superimposedsmaller scale variations of the northeast monsoon. Recordsobtained from the eastern and northeast parts documentchanges in northeast rains only.

3. The northeast winter monsoon displays a drying from 6.0 kauntil 3.8 ka and remains stable after 3.8 ka. Two superimposedweaker northeast monsoon periods occur between 0 and 0.6 kaand from 2.2 until 3.8 ka. No correlation can be establishedwith variations in the southwest monsoon and with theNorthern Hemisphere climatic variations. More high resolutionrecords are required to understand the exact forcing behind thenortheast monsoon for this area.

4. The long term wettening of the southwest monsoon on west-ern Socotra since 4.4 ka (Fleitmann et al., 2007) is a local effectthat relates to a changing wind path. A possible hypothesiscould be that in response to the weakening of the southwestmonsoon, the southwest winds are forced into a more south-erly path over the Indian Ocean consequently causing a wettersouthwest monsoon rainy season on western Socotra.

5. After 6.2 ka, similar precipitation signals can be found betweeneastern Socotra and northern Oman suggesting both regionsare similarly affected by the northeast winter monsoon fromthen on. Areas on the Arabian Peninsula such as northernOman currently receiving rain only once a year from northeastwinds display an abrupt end of the Holocene wet optimumaround 6 ka due to their disconnection from the southwestwinds. In contrast, records from the southern Arabian Pen-insula still located within the ITCZ migration pathway andreceiving rain during both monsoon seasons display a gradualdrying after the Holocene wet optimum due to the weakeningof the southwest monsoon after 8 ka.

Acknowledgements

We thank the Yemen Ministry of Water and Environment andthe Environment Protection Authority (EPA e Socotra Branch), themembers of the Socotra Karst Project and of the Friends of Socotragroup for their help during the fieldwork. Maïté Van Rampelberghalso want to thank Mr. R. Van Dierendonck for his interest andsupport, Kay Van Damme for sharing his knowledge on Socotra’svegetation cover and Dirk Van Dorpe for providing additional in-formation on the Hoq Cave dimensions. This work is support by theHercules Foundation to Philippe Claeys, and Research FoundationFlanders (FWO) through project G-0422-10 to Philippe Claeys.

References

Bond, G., Showers, W., Cheseby, M., Lotti, R., Almasi, P., deMenocal, P., Priore, P.,Cullen, H., Hajdas, I., Bonani, G., 1997. A pervasive millennial-scale cycle in NorthAtlantic Holocene and glacial climates. Science 278, 1257e1266.

Burns, S.J., Matter, A., Frank, N., Mangini, A., 1998. Speleothem-based paleoclimaterecord from northern Oman. Geology 26, 499e502.

Burns, S.J., Fleitmann, D., Matter, A., Neff, U., Mangini, A., 2001. Speleothem evidencefrom Oman for continental pluvial events during interglacial periods. Geology29, 623e626.

Burns, S.J., Fleitmann, D., Matter, A., Kramers, J., Al-Subbary, A.A., 2003. IndianOcean climate and an absolute chronology over Dansgaard/Oeschger events 9to 13. Science 301, 1365e1367.

Cheng, H., Edwards, R.L., Hoff, J., Gallup, C.D., Richards, D.A., Asmerom, Y., 2000. Thehalf-lives of uranium-234 and thorium-230. Chemical Geology 169, 17e33.

Cheng, H., Edwards, R.L., Broecker, W.S., Denton, G.H., Kong, X., Wang, Y., Zhang, R.,Wang, X., 2009a. Ice Age terminations. Science 326, 248e252.

Cheng, H., Fleitmann, D., Edwards, R.L., Wang, X.F., Cruz, F.W., Auler, A.S.,Mangini, A., Wang, Y.J., Kong, X.G., Burns, S.J., Matter, A., 2009b. Timing andstructure of the 8.2 kyr BP event inferred from delta O-18 records of stalagmitesfrom China, Oman, and Brazil. Geology 37, 1007e1010.

Cheung, C., De Vantier, L., Van Damme, K., 2006. Socotra, A Natural History of theIslands and Their People. Odyssey Books and Guides, Airphoto InternationalLtd., Hong Kong, pp. 408.

Cook, E.R., Anchukaitis, K.J., Buckley, B.M., D’Arrigo, R.D., Jacoby, G.C., Wright, W.E.,2010. Asian monsoon failure and megadrought during the last millennium.Science, 486e489.

Culek, M., Kral, K., Habrova, H., Aldolt, R., Pavlis, J., Madera, P., 2006. Socotra’s annualweather pattern. In: Cheung, C., De Vantier, L., Van Damme, K. (Eds.), Socotra, ANatural History of the Islands and Their People. Odyssey Books and Guides,Airphoto International Ltd., Hong Kong, pp. 42e46.

Cullen, H.M., deMenocal, P.B., Hemming, S., Hemming, G., Brown, F.H.,Guilderson, T., Sirocko, F., 2000. Climate change and the collapse of the Akka-dian empire: evidence from the deep sea. Geology 28, 379e382.

Dansgaard, W., 1964. Stable isotopes in precipitation. Tellus 16, 436e468.Day, C.C., Henderson, G.M., 2011. Oxygen isotopes in calcite grown under cave-

analogue conditions. Geochimica Et Cosmochimica Acta 75, 3956e3972.Deininger, M., Fohlmeister, J., Scholz, D., Mangini, A., 2012. Isotope disequilibrium

effects: the influence of evaporation and ventilation effects on the carbon andoxygen isotope composition of speleothem e a model approach. Geochimica EtCosmochimica Acta 96, 57e79.



81




Dorale, J.A., Liu, Z., 2009. Limitations of Hendy test criteria in judging the paleo-climatic suitability of speleothems and the need for replication. Journal of Caveand Karst Studies 71, 73e80.

Dreybrodt, W., Scholz, D., 2011. Climatic dependence of stable carbon and oxygenisotope signals recorded in speleothems: from soil water to speleothem calcite.Geochimica Et Cosmochimica Acta 75, 734e752.

Dreybrodt, W., 2011. Comments on processes contributing to the isotope compo-sition of 13C and 18O in calcite deposited in speleothems. Acta Carsologica 40.

Dykoski, C.A., Edwards, R.L., Cheng, H., Yuan, D.X., Cai, Y.J., Zhang, M.L., Lin, Y.S.,Qing, J.M., An, Z.S., Revenaugh, J., 2005. A high-resolution, absolute-dated Ho-locene and deglacial Asian monsoon record from Dongge Cave, China. Earth andPlanetary Science Letters 233, 71e86.

Edwards, R.L., Chen, J.H., Wasserburg, G.J., 1987. 238U-234U-230Th-232Th systematicsand the precise measurement of time over the past 500 000 years. Earth andPlanetary Science Letters 81, 175e192.

Epstein, S., Mayeda, 1953. Variations of the 18O/16O ratio in natural waters. Geo-chimica Et Cosmochimica Acta 4, 213e224.

Fairchild, I.J., Treble, P.C., 2009. Trace elements in speleothems as recorders ofenvironmental change. Quaternary Science Reviews 28, 449e468.

Fairchild, I.J., Borsato, A., Tooth, A.F., Frisia, S., Hawkesworth, C.J., Huang, Y.M.,McDermott, F., Spiro, B., 2000. Controls on trace element (Sr-Mg) compositionsof carbonate cave waters: implications for speleothem climatic records.Chemical Geology 166, 255e269.

Fairchild, I.J., Smith, C.L., Baker, A., Fuller, L., Spotl, C., Mattey, D., McDermott, F.,Eimp, 2006. Modification and preservation of environmental signals in spe-leothems. Earth-science Reviews 75, 105e153.

Fleitmann, D., Burns, S.J., Mudelsee, M., Neff, U., Kramers, J., Mangini, A., Matter, A.,2003a. Holocene forcing of the Indian monsoon recorded in a stalagmite fromSouthern Oman. Science 300, 1737e1739.

Fleitmann, D., Burns, S.J., Neff, U., Mangini, A., Matter, A., 2003b. Changing moisturesources over the last 330,000 years in Northern Oman from fluid-inclusionevidence in speleothems. Quaternary Research 60, 223e232.

Fleitmann, D., Burns, S.J., Neff, U., Mudelsee, M., Mangini, A., Matter, A., 2004a.Palaeoclimatic interpretation of high-resolution oxygen isotope profiles derivedfrom annually laminated speleothems from Southern Oman. Quaternary Sci-ence Reviews 23, 935e945.

Fleitmann, D., Matter, A., Burns, S.J., Al-Subbary, A., Al-Aowah, M.A., 2004b. Geologyand Quaternary climate history of Socotra. Fauna of Arabia 20, 27e44.

Fleitmann, D., Burns, S.J., Mangini, A., Mudelsee, M., Kramers, J., Villa, I., Neff, U., Al-Subbary, A.A., Buettner, A., Hippler, D., Matter, A., 2007. Holocene ITCZ andIndian monsoon dynamics recorded in stalagmites from Oman and Yemen(Socotra). Quaternary Science Reviews 26, 170e188.

Frumkin, A., Ford, D.C., Schwarcz, H.P., 2000. Paleoclimate and vegetation of the lastglacial cycles in Jerusalem from a speleothem record. Global BiogeochemicalCycles 14, 863e870.

Fuchs,M., Buerkert, A., 2008. A 20 ka sediment record from theHajarMountain rangein N-Oman, and its implication for detecting arid-humid periods on the south-eastern Arabian Peninsula. Earth and Planetary Science Letters 265, 546e558.

Gasse, F., 2000. Hydrological changes in the African tropics since the Last GlacialMaximum. Quaternary Science Reviews 19, 189e211.

Genty, D., Baker, A., Massault, M., Proctor, C., Gilmour, M., Pons-Branchu, E.,Hamelin, B., 2001. Dead carbon in stalagmites: carbonate bedrock paleo-dissolution vs. ageing of soil organic matter. Implications for C-13 variations inspeleothems. Geochimica Et Cosmochimica Acta 65, 3443e3457.

Genty, D., Blamart, D., Ouahdi, R., Gilmour, M., Baker, A., Jouzel, J., Van-Exter, S.,2003. Precise dating of Dansgaard-Oeschger climate oscillations in westernEurope from stalagmite data. Nature 421, 833e837.

Gupta, A.K., Anderson, D.M., Overpeck, J.T., 2003. Abrupt changes in the Asiansouthwest monsoon during the Holocene and their links to the North AtlanticOcean. Nature 421, 354e357.

Henderson, G.M., 2006. Climate e caving in to new chronologies. Science 313, 620e622.

Hendy, C.H., 1971. Isotopic geochemistry of speleothems: 1. Calculations of effectson different modes of formation on isotopic composition of speleothems andtheir applicability as paleoclimatic indicators. Geochimica Et CosmochimicaActa 35, 801.

Ivanochko, T.S., Ganeshram, R.S., Brummer, G.J.A., Ganssen, G., Jung, S.J.A.,Moreton, S.G., Kroon, D., 2005. Variations in tropical convection as an amplifierof global climate change at the millennial scale. Earth and Planetary ScienceLetters 235, 302e314.

Johnson, K.R., Hu, C.Y., Belshaw, N.S., Henderson, G.M., 2006. Seasonal trace-element and stable-isotope variations in a Chinese speleothem: the potentialfor high-resolution paleomonsoon reconstruction. Earth and Planetary ScienceLetters 244, 394e407.

Kim, S.T., O’Neil, J.R., 1997. Equilibrium and nonequilibrium oxygen isotope effects insynthetic carbonates. Geochimica Et Cosmochimica Acta 61, 3461e3475.

Kropelin, S., Verschuren, D., Lezine, A.M., 2008. Response to comment on “Climate-driven ecosystem succession in the Sahara: the past 6000 years”. Science 322.

Lachniet, M.S., 2009. Climatic and environmental controls on speleothem oxygen-isotope values. Quaternary Science Reviews 28, 412e432.

Lezine, A.M., Tiercelin, J.J., Robert, C., Saliege, J.F., Cleuziou, S., Inizan, M.L.,Braemer, F., 2007. Centennial to millennial-scale variability of the Indianmonsoon during the early Holocene from a sediment, pollen and isotope recordfrom the desert of Yemen. Palaeogeography, Palaeoclimatology, Palaeoecology243, 235e249.

Lezine, A.M., Robert, C., Cleuziou, S., Inizan, M.L., Braemer, F., Saliege, J.F.,Sylvestre, F., Tiercelin, J.J., Crassard, R., Mery, S., Charpentier, V., Steimer-Herbet, T., 2010. Climate change and human occupation in the Southern Ara-bian lowlands during the last deglaciation and the Holocene. Global andPlanetary Change 72, 412e428.

McDermott, F., 2004. Palaeo-climate reconstruction from stable isotope variationsin speleothems: a review. Quaternary Science Reviews 23, 901e918.

McMillan, E.A., Fairchild, I.J., Frisia, S., Borsato, A., McDermott, F., 2005. Annual traceelement cycles in calcite-aragonite speleothems: evidence of drought in thewestern Mediterranean 1200-1100 yr BP. Journal of Quaternary Science 20,423e433.

Mühlinghaus, C., Scholz, D., Mangini, A., 2009. Modelling fractionation of stableisotopes in stalagmites. Geochimica Et Cosmochimica Acta, 7275e7289.

Neff, U., Burns, S.J., Mangini, A., Mudelsee, M., Fleitmann, D., Matter, A., 2001. Strongcoherence between solar variability and the monsoon in Oman between 9 and6 kyr ago. Nature 411, 290e293.

Paillard, D., Labeyrie, L., Yiou, P., 1996. Macintosh program performs time-seriesanalysis. Eos, Transactions American Geophysical Union 77, 379.

Parker, A.G., Goudie, A.S., Stokes, S., White, K., Hodson, M.J., Manning, M., Kennet, D.,2006. A record of Holocene climate change from lake geochemical analyses insoutheastern Arabia. Quaternary Research 66, 465e476.

Polag, D., Scholz, D., Muehlinghaus, C., Spoetl, C., Schroeder-Ritzrau, A., Segl, M.,Mangini, A., 2010. Stable isotope fractionation in speleothems: laboratory ex-periments. Chemical Geology 279, 31e39.

Preusser, F., Radies, D., Matter, A., 2002. A 160,000-year record of dune devel-opment and atmospheric circulation in southern Arabia. Science 296, 2018e2020.

Radies, D., Hasiots, S.T., Preusser, F., Neubert, E., Matter, A., 2005. Paleoclimaticsignificance of Early Holocene faunal assemblages in wet interdune deposits ofthe Wahiba Sand Sea, sultanate of Oman. Journal of Arid Environments 62, 109e125.

Rasmussen, S.O., Andersen, K.K., Svensson, A.M., Steffensen, J.P., Vinther, B.M.,Clausen, H.B., Siggaard-Andersen, M.L., Johnsen, S.J., Larsen, L.B., Dahl-Jensen, D., Bigler, M., Rothlisberger, R., Fischer, H., Goto-Azuma, K.,Hansson, M.E., Ruth, U., 2006. A new Greenland ice core chronology for the lastglacial termination. Journal of Geophysical Research-atmospheres 111.

Rozanski, K., Araguasaraguas, L., Gonfiantini, R., 1992. Relationship between long-term trends of 18O isotope composition of precipitation and climate. Science258, 981e985.

Scholte, P., De Geest, P., 2010. The climate of Socotra Island (Yemen): a first-timeassessment of the timing of the monsoon wind reversal and its influence onprecipitation and vegetation patterns. Journal of Arid Environments 74, 1507e1515.

Scholz, D., Hoffmann, D.L., 2011. StalAge e an algorithm designed for construction ofspeleothem age models. Quaternary Geochronology 6, 369e382.

Shakun, J.D., Burns, S.J., Fleitmann, D., Kramers, J., Matter, A., Al-Subary, A., 2007.A high-resolution, absolute-dated deglacial speleothem record of Indian Oceanclimate from Socotra Island, Yemen. Earth and Planetary Science Letters 259,442e456.

Sirocko, F., Sarnthein, M., Erlenkeuser, H., Lange, H., Arnold, M., Duplessy, J.C., 1993.Century-scale events in monsoonal climate over the past 24,000 years. Nature364, 322e324.

Smith, B., Epstein, S., 1970. Two categories of 13C/12C ratios for higher plants. PlantPhysiology 47, 380e384.

Vinther, B.M., Clausen, H.B., Johnsen, S.J., Rasmussen, S.O., Andersen, S.S.,Buchardt, S.L., Dahl-Jensen, D., Seierstad, I.K., Siggaard-Andersen, M.L.,Steffensen, J.P., Svensson, A., Olsen, J., Heinemeier, J., 2006. A synchronizeddating of three Greenland ice cores throughout the Holocene. Journal of Geo-physical Research-part D-atmospheres 111, 11.

Waliser, D.E., Gautier, C., 1993. A satellite-derived climatology of the ITCZ. Journal ofClimate 6, 2162e2174.

Wang, Y.J., Cheng, H., Edwards, R.L., An, Z.S., Wu, J.Y., Shen, C.C., Dorale, J.A., 2001.A high-resolution absolute-dated Late Pleistocene monsoon record from HuluCave, China. Science 294, 2345e2348.

Wang, Y.J., Cheng, H., Edwards, R.L., He, Y.Q., Kong, X.G., An, Z.S., Wu, J.Y., Kelly, M.J.,Dykoski, C.A., Li, X.D., 2005. The Holocene Asian monsoon: links to solarchanges and North Atlantic climate. Science 308, 854e857.

Zhang, P., Cheng, H., Edwards, R.L., Chen, F., Wang, Y., Yang, X., Liu, J., Tan, M.,Wang, X., Liu, J., An, C., Dai, Z., Zhou, J., Zhang, D., Jia, J., Jin, L., Johnson, K.R.,2008. A test of climate, sun, and culture relationships from an 1810-year Chi-nese cave record. Science 322, 940e942.



82




Chapter 4: Cave Monitoring

83

Chapter 4

Monitoring of the Proserpine stalagmite “The present is the key to the past” (Charles Lyell, 1830) Understanding how the present-‐day cave system works may be of large interest when interpreting speleothem proxy variations in the recent past. As discussed in Chapter 2, a multitude of factors can affect speleothem δ18O and δ13C values, chemical composition, growth rate, layering, layer thickness and calcite fabric, to name only the most important (Baker et al., 1998; Dreybrodt, 1999, 2008; Mühlinghaus et al., 2009; Scholz et al., 2009; Dreybrodt, 2011; Deininger et al., 2012; Dreybrodt, 2012). In particular, speleothem records from mid-‐latitude temperate regions have shown to be difficult to interpret in terms of a single climate proxy (Baker et al., 2011). Cave monitoring data can reveal important links that can be used to understand how proxies record climate (Frisia et al., 2000; Spötl et al., 2005; Mattey et al., 2008; Riechelmann et al., 2011). The Proserpine stalagmite grows in the well-‐known and easily accessible Han-‐sur-‐Lesse showcave, located the south of Belgium, and has been suggested to reflect climate variations up to seasonal scales as indicated by its clear seasonal layering (Verheyden et al., 2006). To correctly interpret variations in δ18O and δ13C values, layer thickness and calcite fabric (see Chapter 5), a cave monitoring campaign was set up to investigate the possible environmental factors affecting these proxies. Between October 2012 and January 2014, the Proserpine drip site was monitored on a biweekly basis. Different cave parameters were measured such as cave air and drip water temperature, the water discharge amount, pCO2 and δ13CCO2 values of cave air, rainwater δ18O and δD values, drip water pH, δ18O, δD and δ13CDIC values, and the δ18O and δ13C values of freshly farmed cave calcite. Results indicate that all cave parameters vary seasonally between a ‘summer mode’ from June to December and a ‘winter mode’ from December to June. Of major importance for the interpretation of the proxies measured in Chapter 5 is that the δ18O and δ13C values of present-‐day calcite are deposited in isotopic


84

equilibrium with the drip water. The δ18O values vary seasonally and become more negative in summer (June, July and August) due to the increased cave air and drip water temperature. The δ13C values vary seasonally in response to the seasonal changes in prior calcite precipitation (PCP). In summer, the absence of water recharge increases PCP, which causes higher δ13CDIC values of the drip water and of the calcite deposited from the drip water. δ18O and δ13C values were measured at high resolution in a small part of the Proserpine core to investigate how they vary according to the layering. Results showed that dark layers display low δ18O values and high δ13C values. In the clearly visible yearly layer couplets, darker calcite is interpreted to reflect summer conditions and slower growth while the whiter, more porous layers are formed in winter when calcite growth rate increases. REFERNCES Baker, A., Genty, D., Dreybrodt, W., Barnes, W. L., Mockler, N. J., and Grapes, J.: Testing theoretically predicted stalagmite growth rate with Recent annually laminated samples: Implications for past stalagmite deposition, Geochimica Et Cosmochimica Acta, 62, 393-‐404, 1998.

Baker, A., Wilson, R., Fairchild, I. J., Franke, J., Spoetl, C., Mattey, D., Trouet, V., and Fuller, L.: High resolution delta O-‐18 and delta C-‐13 records from an annually laminated Scottish stalagmite and relationship with last millennium climate, Glob. Planet. Change, 79, 303-‐311, 2011.


Dreybrodt, W.: Chemical kinetics, speleothem growth and climate, Boreas, 28, 347-‐356, 1999.

Dreybrodt, W.: Comment on "Oxygen isotopes in calcite grown under cave-‐analogue conditions" by C.C. Day and G.M. Henderson, Geochimica Et Cosmochimica Acta, 85, 383-‐387, 2012.

Dreybrodt, W.: Comments on processes contributing to the isotope composition of 13C and 18O in calcite depostied in speleothems, Acta Carsologica, 40, 2011.

Dreybrodt, W.: Evolution of the isotopic composition of carbon in a calcite precipitating H2O-‐CO2-‐CaCO3 solution and the related isotopic composition of calcite in stalagmites, Geochimica Et Cosmochimica Acta, 72, 4712-‐4724, 2008.

Frisia, S., Borsato, A., Fairchild, I. J., and McDermott, F.: Calcite fabrics, growth mechanisms, and environments of formation in speleothems from the Italian Alps and southwestern Ireland, Journal of Sedimentary Research, 70, 1183-‐1196, 2000.


85



Riechelmann, D. F. C., Schroeder-‐Ritzrau, A., Scholz, D., Fohlmeister, J., Spoetl, C., Richter, D. K., and Mangini, A.: Monitoring Bunker Cave (NW Germany): A prerequisite to interpret geochemical proxy data of speleothems from this site, Journal of Hydrology, 409, 682-‐695, 2011.


Spötl, C., Fairchild, I. J., and Tooth, A. F.: Cave air control on dripwater geochemistry, Obir Caves (Austria): Implications for speleothem deposition in dynamically ventilated caves, Geochimica Et Cosmochimica Acta, 69, 2451-‐2468, 2005.

Verheyden, S., Baele, J.-‐M., Keppens, E., Genty, D., Cattani, O., Hai, C., Edwards, L., Hucai, Z., Van Strijdonck, M., and Quinif, Y.: The proserpine stalagmite (Han-‐sur-‐Lesse cave, Belgium): Preliminary environmental interpretation of the last 1000 years as recorded in a layered speleothem, Geologica Belgica, 9, 245-‐256, 2006.


86

Clim. Past, 10, 1–15, 2014www.clim-past.net/10/1/2014/doi:10.5194/cp-10-1-2014© Author(s) 2014. CC Attribution 3.0 License.

Monitoring of a fast-growing speleothem site from theHan-sur-Lesse cave, Belgium, indicates equilibriumdeposition of the seasonal �18O and �13C signals in the calcite

M. Van Rampelbergh1, S. Verheyden1,2, M Allan3, Y. Quinif4, E. Keppens1, and P. Claeys1

1Earth System Sciences, Vrije Universiteit Brussel (VUB), Pleinlaan, 1050, Brussels, Belgium2Royal Belgian Institute of Natural Sciences, Geological Survey, Direction Earth and History of Life, Jennerstraat 13,1000, Brussels, Belgium3AGEs, Départment de Géologie, Université de Liège, Allée du 6 Août, B18 Sart-Tilman, 4000, Liège, Belgium4Faculté Polytechnique, Université de Mons, Rue de Houdain 9, 7000, Mons, Belgium

Correspondence to: M. Van Rampelbergh ([email protected])

Received: 21 March 2014 – Published in Clim. Past Discuss.: 22 April 2014Revised: 15 September 2014 – Accepted: 16 September 2014 – Published:

Abstract. Speleothems provide paleoclimate information onmultimillennial to decadal scales in the Holocene. However,seasonal or even monthly resolved records remain scarce.Such records require fast-growing stalagmites and a goodunderstanding of the proxy system on very short timescales.The Proserpine stalagmite from the Han-sur-Less cave (Bel-gium) displays well-defined/clearly visible darker and lighterseasonal layers of 0.5 to 2 mm thickness per single layer,which allows a measuring resolution at a monthly scale.Through a regular cave monitoring, we acquired a good un-derstanding of how �18O and �13C signals in modern calcitereflect climate variations on the seasonal scale. From Decem-ber to June, outside temperatures are cold, inducing low caveair and water temperature, and bio-productivity in the soil islimited, leading to lower pCO2 and higher �13C values of theCO2 in the cave air. From June to December, the measuredfactors display an opposite behavior.

The absence of epikarst water recharge between May andOctober increases prior calcite precipitation (PCP) in the va-dose zone, causing drip water to display increasing pH and�13C values over the summer months. Water recharge of theepikarst in winter diminishes the effect of PCP and as a resultthe pH and �13C of the drip water gradually decrease. The�18O and �13C signals of fresh calcite precipitated on glassslabs also vary seasonally and are both reflecting equilibriumconditions. Lowest �18O values occur during the summer,when the �13C values are high. The �18O values of the cal-

cite display seasonal variations due to changes in the cave airand water temperature. The �13C values reflect the seasonalvariation of the �13CDIC of the drip water, which is affectedby the intensity of PCP. This same anticorrelation of the�18O versus the �13C signals is seen in the monthly resolvedspeleothem record that covers the period between 1976 and1985 AD. Dark layers display lower �18O and higher �13Cvalues. The cave system varies seasonally in response to theactivity of the vegetation cover and outside air temperaturebetween a “summer mode” lasting from June to Decemberand a “winter mode” from December to June. The low �18Oand high �13C values of the darker speleothem layers indi-cate that they are formed during summer, while light layersare formed during winter. The darker the color of a layer,the more compact its calcite structure is, and the more nega-tive its �18O signal and the more positive its �13C signal are.Darker layers deposited from summer drip water affected byPCP are suggested to contain lower Ca2+ concentration. Ifindeed the calcite saturation represents the main factor driv-ing the Proserpine growth rate, the dark layers should growslower than the white layers.

Published by Copernicus Publications on behalf of the European Geosciences Union.




87



2 M. Van Rampelbergh: Monitoring of a fast-growing speleothem site

1 Introduction

In the past 25 years, speleothem records have provided im-portant information on past climate variations on multimil-lennial to decadal scales (e.g., Genty et al., 2003; McDer-mott, 2005; Verheyden et al., 2008b; Wang et al., 2008; VanRampelbergh et al., 2013). With the increasing number ofstudies on cave calcite deposition dynamics (e.g., Dreybrodt,1999, 2008; Verheyden et al., 2008a; Lachniet, 2009; Scholzet al., 2009; Ruan and Hu, 2010; Oster et al., 2012) and withthe help of modern analytical tools (Fairchild et al., 2006;Spötl and Mattey, 2006; Jochum et al., 2012), progress hasbeen made to measure resolution at sub-seasonal and evenbi-monthly scales. However, only a few studies, so far, havereached such high temporal resolution for the measured prox-ies (Treble et al., 2003; Mattey et al., 2008) mainly due tofollowing two limitations.

The first limitation to study the paleoclimate at the sea-sonal scale from stalagmites is that their growth rate needsto be significantly high (around 1 mm yr�1) to deposit thicklayers allowing monthly resolved time series. Speleothemgrowth rates vary according to different factors, such as dripwater rate and calcium ion concentration (Baker et al., 1998;Dreybrodt, 1999), rendering the estimation of an average ratedifficult. Generally, stalagmites increase at 10–100 µm yr�1

in cool temperate climates and at 300–500 µm yr�1 in sub-tropical climates (Fairchild et al., 2006), clearly showing thatfast-growing (more than 1 mm yr�1) speleothems are trulyexceptional.

A second limitation is that a good understanding of thecave system is needed to understand what the measured prox-ies are reflecting. The interaction between the climate param-eters, the soil, the host rock and the cave environment needsto be well understood for the studied cave and time frame.On classical multimillennial and centennial timescales, theprocesses influencing the stable isotopes of oxygen and car-bon are well established (e.g., Fairchild et al., 2006; Bakeret al., 2007). However, local cave-specific effects affect sea-sonally or even monthly resolved �18O and �13C signals. Forstudies at seasonal scales, a detailed study of the cave dy-namics is required (Mattey et al., 2008) in order to under-stand which factors drive the isotopic signals, and at whichintensity. In the last few years, different authors have tried tomodel how �18O and �13C signals are affected by tempera-ture, drip rate, amount of CO2 degassing or residence time ofthe water film on the surface of the stalagmite (Mühlinghauset al., 2007, 2009; Dreybrodt and Scholz, 2011; Deininger etal., 2012). Results of these studies have provided importantprogress in the understanding of the isotope system in caveenvironments. However, cave-monitoring programs remainof crucial importance to test in real cave environments the hy-potheses derived from the models. Different cave-monitoringstudies have been set up all over the world to better un-derstand these seasonal and sub-seasonal processes (Gentyand Deflandre, 1998; Spötl et al., 2005; Mattey et al., 2008;

Riechelmann et al., 2011). Only few of them have providedanswers on the isotope fractionation processes occurring be-tween the drip water and recent precipitated calcite due tothe complexity of the system and the variety of the differentspecific environments (Verheyden et al., 2008a; Tremaine etal., 2011; Riechelmann et al., 2013).

Previous studies of the Han-sur-Lesse karst system showthat the cave responds seasonally to external climate fac-tors and that it is well suited for high-resolution speleothem-based climate reconstructions (Genty and Quinif, 1996; Ver-heyden et al., 2006, 2008a). In the Han-sur-Lesse cave, thehigh growth rate (up to 2.1 mm yr�1) and clear seasonalbanding of the “Proserpine” stalagmite make it possible to re-construct climate variations at the seasonal scale (Verheydenet al., 2006). In this study, we report results of a cave environ-ment monitoring carried out once every 2 weeks for 1 year(2013) that shows how oxygen and carbon isotope signals ob-tained from the Proserpine banding reflect climate variationsat the seasonal scale. The results are then compared to high-resolution �18O and �13C signals measured on the 10 thickestlayers from the upper 10 cm of the Proserpine, which coverthe period from 1976 to 1985 AD. This approach improvesthe knowledge and accuracy for the use of �18O and �13Csignals in speleothems at the seasonal scale.

2 Study area and hydrological setting

The Han-sur-Lesse cave is located within Givetian lime-stones of the Dinant synclinorium and is the largest andbest-developed karst system in Belgium (Delvaux De Fenffe,1985). The Lesse river formed the cave within a hill calledthe “Massif de Boine” entering the karst system at the “Gouf-fre de Belvaux” and exiting approximately 24 h later throughthe “Trou de Han” (Fig. 1). The cave has been exploited sincethe mid-19th century as a touristic attraction and is character-ized by large chambers and well-developed speleothem for-mations. The cave monitoring and speleothem sampling forthis study is carried out in the “Salle-Du-Dôme” chamber.This 150 m wide and 60 m high chamber formed by collapseand is the largest of the whole cave system. The Proserpinestalagmite is easily reachable following the tourist path intothe cave for approximately 700 m from the cave’s exit at theTrou de Han (Fig. 1). It has a large tabular shape with a rela-tively horizontal slightly undulating surface of about 1.5 m2

and is fed by a continuous high drip water flow that dripson mainly four sites. The epikarst thickness above the cavechamber is estimated to be 40 m (Quinif, 1988). Two pas-sages connect the Salle-Du-Dôme to the neighboring cham-bers, and the Lesse river flows at the bottom of the chamber.

The mean annual precipitation at the nearest meteorolog-ical station of Han-sur-Lesse is 844 mm yr�1, and the meanannual air temperature is 10.3 �C (Royal Meteorological In-stitute Belgium). While the temperature displays a well-marked seasonality with cool summers and mild winters, the

Clim. Past, 10, 1–15, 2014 www.clim-past.net/10/1/2014/





88


M. Van Rampelbergh: Monitoring of a fast-growing speleothem site 3

Figure 1. The Han-sur-Lesse karst system, with the Han-sur-Lesse cave as the northern cave system and the Père Noël cave (Verheyden etal., 2008a) as the southern cave system. The Lesse river enters the cave system at the “Gouffre de Belvaux” and exits 24 h later at the “Troude Han”. The studied speleothem is located in the “Salle-du-Dôme” and grows on a pile of debris. Figure adapted after Quinif (1988).

rainfall is spread all over the entire year. According to theKöppen and Geiger classification, the climate above the Han-sur-Lesse cave is considered as a warm temperate, fully hu-mid climate with cool summers (Kottek et al., 2006). For thestudied period lasting between November 2012 and January2014, air temperature was at its lowest between Decemberand March and highest between July and September. Thecoldest temperature of �4.2 �C was reached on 15 March2013. Such an unusually cold March is an exception and doesnot represent the average weather conditions for the studiedregion. The plant coverage above the cave consists of C3-type vegetation with oaks, beech and hazel trees. The soilis approximately 40 cm tick and consists of silty stony soilwith more than 50 % limestone fragments (Belgian Geolog-ical Survey map). The area above the cave is part of a pro-tected natural reserve, preserved from direct human influencefor more than 50 years.

3 Methods

Between October 2012 and January 2014, the cave was vis-ited every 2 weeks to record environmental parameters. Tomake sure that the measured parameters are closely reflect-ing the natural conditions and to guarantee a consistent mea-surement campaign, the cave parameters were always mea-sured around 09:00 a.m. before the first visitor enters thecave. To investigate the visitors’ possible influence on themeasured cave parameters, a test was carried out by mea-suring these parameters before visitors were allowed into thecave and after the passage of different groups. Cave air andwater temperature were measured with a HANNA HI955501thermometer with a precision of 0.2 �C. Air temperature wastaken directly above the stalagmite. The drip water tempera-ture and pH were determined in a small natural pool (6 cm

wide and 3 cm deep) formed on the stalagmite’s surface,where drip water is continuously falling in. The extremelyshort residence time of the water in this small pool guaran-tees that the temperature and pH suffer minimal alteration.The pH of the drip water was measured with a HANNAHI991300 sensor (precision of 0.01 pH). The concentrationof CO2 in the cave air was obtained using an ACCURO640 000 manual Dräger pump with a standard deviation of 10to 15 %. Three times per visit, pCO2 values were measuredat the same spot, right above the surface of the speleothem,and reported as an average of the three values. The drip waterdischarge (volumetric flow rate, here given in volume (mL)of water reaching the speleothem surface per minute) abovethe Proserpine stalagmite was measured in a graded cylinderafter collecting the drip water during 10 min in an inflatablesoft plastic swimming pool (1.77 m2) which was placed onthe stalagmite’s surface to collect the drips from the four dripsites of the Proserpine. Drip water samples for �18O and �Dmeasurements were collected using a container placed at thesurface of the stalagmite, where the drip fell from the stalac-tite approximately 30 m above the stalagmite. Water sampleswere stored in fully filled glass bottles in a cool and darkenvironment. Rainwater samples for �18O and �D measure-ments were collected in a garden close to the cave using athermos bottle and sampled every 15 days between Novem-ber 2012 and January 2014. To avoid evaporation processes,the rainwater was collected using a funnel with a raised edgeand connected to a tube reaching the bottom of the thermosbottle. The funnel was attached to the bottle through a her-metic cap. Glass bottles were fully filled with the collectedrainwater and stored in a dark fridge until being analyzed.

The �18O and �D composition of the waters were mea-sured using a PICARRO L2130-i Cavity Ring-Down Spec-trometer at the Vrije Universiteit Brussel (VUB). For ev-

www.clim-past.net/10/1/2014/ Clim. Past, 10, 1–15, 2014





89



ery sample 1.4 mL of water was used for the measure-ments. Every sample was injected and analyzed 10 consec-utive times. The measured values were then corrected us-ing house standards with a strongly different isotopic com-position. The first house standard, called DO1, has a �18Ovalue of �0.79 ± 0.04 ‰ and a �D value of �26.2 ± 0.2 ‰and was made by collecting the damp of boiling water. Thesecond house standard, called DO2, is Milli-Q water witha �18O composition of �7.38 ± 0.04 ‰ and a �D value of�48.8 ± 0.2 ‰. The third house standard, called DO3, is wa-ter from Antarctic glacier ice that was filtered and has a �18Ocomposition of �14.77 ± 0.04 ‰ and a �D composition of�105.1 ± 0.2 ‰. The most positive (DO1) and the most neg-ative (DO3) house standards are used to obtain a two-pointcalibration line. The DO2 house standard, with intermedi-ate values, serves as a “target” or “control” point. By usingthese three standards, we can correct the measured valuesfor a lateral difference as well as for a stretch that can occurin the measurement range. All three working standards weremade in the lab and calibrated against the international stan-dards VSMOW2 (Vienna Standard Mean Ocean Water 2),GISP (Greenland Ice Sheet Precipitation) and SLAP2 (Stan-dard Light Antarctic Precipitation 2). These three interna-tional standards were used to correct the house standards inthe same way the DO1, DO2 and DO3 standards are usedto correct the measurements. The calibration curve was ob-tained using VSMOW2 and SLAP2, and the GISP standardwas used as target. Every house standard was measured 55times on the PICARRO L2130-i. The collected rain and dripwater samples were analyzed two times in different order.The reported values are the average of the two measurementsand reported in per mill VSMOW. Analytical uncertainties(2� ) equal 0.07 ‰ for the measured �18O values and 0.5 ‰for the measured �D values.

The isotopic composition of the cave air CO2 was mea-sured from samples collected by filling vacuum 2 L glasscontainers. To avoid “human” contamination, these sampleswere taken at the beginning of every cave visit. The CO2 wasextracted from the container using a manual extraction lineat the VUB. The extracted CO2 was then analyzed for itsisotopic composition on a Thermo Delta plus XL mass spec-trometer in dual-inlet mode. The standard deviation of thethree measurements reports the error on the measured �13Cvalue. The 2� values average 0.6 ‰. All values are reportedin per mill VPDB (Vienna Pee Dee Belemnite).

Samples for the analyses of the �13C composition of thedissolved inorganic carbon (DIC) in the water were collectedby filling 12 mL gastight glass tubes all the way to the top toavoid air CO2 contamination. A drop of HgCl2 was immedi-ately added and the bottle hermetically closed and stored ina dark and cool environment until being analyzed. The daybefore the analysis, a headspace was created in the bottle bytaking out 3 mL of water, while bubbling He through the sep-tum. Once the headspace was formed, H3PO4 was added andthe sample shaken overnight to convert all DIC species into

CO2. The CO2 gas was then extracted from the bottle andmeasured for its �13C composition. Samples were duplicatedand measured immediately after sampling and 1 month laterto test whether degassing processes affect the DIC composi-tion. This sampling and storing method was tested against themethod described by Spötl et al. (2005) and delivers similarresults within the analytical uncertainties. The �13C compo-sition of the DIC was measured on a Flash EA 1112 deviceconnected to a Delta V plus mass spectrometer. The injectedCO2 is measured against a house standard that consists ofCO2 gas with a �13C composition of �34.07 ‰. The mea-surement series starts with five house standard injections. Af-ter a series of five samples a new house standard injectionis measured to correct the drift. The house standard is cal-ibrated against two international standards, IAEA-CH6 (su-crose) and IAEA-CH7 (polyethylene). All measurements arereported in per mill VPDB with an analytical uncertainty of0.4 ‰ (2� ).

During every visit, three glass slabs were placed on thesurface of the stalagmite to study current calcite depositionconditions on the surface of the stalagmite. All three slabswere positioned very near to where the speleothem core wasdrilled and collected during the next visit. Each time, all threecollected slabs were always completely covered with calcite.The freshly precipitated calcite was then scraped off from theentire slab. Five aliquots per slab were taken from the col-lected powder and measured for their �18O and �13C compo-sition. The reported value per cave visit is the average of theresults of all 15 aliquots of the three slabs collected duringeach visit.

In January 2011, the Proserpine was sampled by drillinga 1 m long core in the middle of the large stalagmite. Aslab was cut from the middle of the core and polished withAl2O3 powder. Layer counting established the age model ofthe upper 16 mm of the core, knowing that one dark and onelight layer are deposited every year (Verheyden et al., 2006).Layer counting was carried out on high-resolution scans us-ing Adobe Photoshop, by counting on the slab itself and byusing the Merchantek MicroMill microscope. Samples to testthe evolution of the isotopic composition of the individuallayers were milled in nine consecutive layer couplets, wherethe dark and white layers were the largest. Samples weredrilled every 40 µm over a length of 4 mm with a MerchantekMicroMill, giving a temporal resolution of approximately 1sample a month. The glass slab and speleothem calcite pow-ders were measured for their �18O and �13C composition us-ing a Kiel III device coupled on a Thermo Delta plus XL. Allvalues were corrected using the international calcite powderstandard NBS-19 and reported in per mill VPDB. Analyticaluncertainties (2� ) were 0.12 ‰ for �13C and 0.16 ‰ for�18O.





90



4 Results

The drip water feeding the Proserpine falls from a smalldrapery-shaped stalactite, indicating that only a small partof the dissolved calcite precipitates from the drip water whenhanging on the ceiling of the cave. The drip falls approxi-mately 30 m before reaching the surface of the Proserpine.The Proserpine grows under a “flow” or continuous “rain”that falls on the surface of the stalagmite at four points. Thecore used in this study was sampled in the center of thespeleothem, where water flows on the stalagmite during thewhole year.

Figure 2 presents the results of the cave-monitoring cam-paign from November 2012 to January 2014 (15 months)together with precipitation amounts and air temperatures atthe RMI station of Han-sur-Lesse (Fig. 2a and b, respec-tively) for the same period. No difference in the measuredcave parameters is observed before the start of the visitsand after a large number of visitors entered the chamber.The cave air temperature (Fig. 2c) in the Salle-Du-Dômevaries seasonally over a range of ca. 4 �C, with highest val-ues reaching up to 14.5 �C in August and lowest value of10.5 �C in February. After the warm month of August, tem-perature values decrease gradually until December. From De-cember through the end of May, temperatures remain sta-ble and vary around an average of 11 �C. In July, the tem-perature increases quickly to reach the warmest values inAugust.Drip water temperatures (Fig. 2d) follow a similartrend as the cave air temperature (Fig. 2c) but are on average0.5 �C colder.

The drip flow of the Proserpine was always measured forthe whole stalagmite, thus including the four drip sites, tohave an idea of how much water is dripping on the wholesurface of the speleothem. The water flow above the Proser-pine stalagmite (Fig. 2e) averages 161 mL min�1. It sharplyincreases during early winter (December) to an average valueabove 200 mL min�1to remain high throughout winter andspring until early summer (June), when it decreases again toaround 100 mL min�1. With a discharge of 3 ⇥ 10�3 L s�1

and a coefficient of variation of approximately 2, the dripsite can be characterized as a “percolation stream” accord-ing to the classification of Smart and Friederich (1987).In early June, discharge record shows a short maximum to280 mL min�1, which is most probably related to the heavy-rainfall period at the end of May 2013 (Fig. 2a). Superim-posed on this seasonal cycle, very short events of increaseddrip flow are observed within 24 h following a heavy-rainfallevent.

The drip water pH (Fig. 2f) varies between 8.4 and 7.9,decreasing in spring (sharply in May) and gradually increas-ing back at the end of summer and throughout autumn (fromSeptember through January). The heavy-rainfall period at theend of May 2013 (Fig. 2a) seems to correspond with a pHdecrease below 8.0. The pCO2 values (Fig. 2g) remain rela-tively stable around 500 ppm throughout much of the year ex-

cept for a marked increase in the summer, reaching 1000 ppmin July and August. The �13C signature (Fig. 2h) of thecave air varies around an average of �19.5 ‰, displayingan anticorrelation to that of the pCO2 concentrations. The�13C of the DIC (Fig. 2i) varies between �12.2 ± 0.3 ‰ and�11.0 ± 0.3 ‰, increases between June and November anddecreases between December and May.

Rainwater �18O values (Fig. 3c) average �8.18 ± 0.07 ‰,and the �D values (Fig. 3d) �55.52 ± 0.5 ‰. The �18O and�D signals increase by 3 ‰ and 30 ‰, respectively, duringthe summer months, presumably due to temperature effect.One larger drop (red arrow in Fig. 3c and d) of 9 ‰ for the�18O and of 90 ‰ for the �D signal occurs at the beginningof March.

The drip water �18O and �D values (Fig. 3e and f)weakly vary around an average of �7.65 ± 0.07 ‰ and�50.1 ± 0.5 ‰, respectively. These values appear slightlyhigher compared to the yearly average �18O and �D valuesof the rainwater. The drip water isotopic records of oxygenand hydrogen are well correlated and remain stable through-out the year with the exception of one small but meaningfulnegative excursion in July and August of 0.06 ‰ for �18Oand of 0.5 ‰ for �D (red arrow in Fig. 3e and f). The rangeof these shifts is on the order of the analytical uncertainties(0.07 ‰ for �18O and 0.5 ‰ for �D), but they are recorded byat least four consecutive measurements, suggesting that theyare significant.

The �18O signal of the calcite recovered from glass slabsplaced on top of the stalagmite (Fig. 2l) remains stable at�6.5 ± 0.16 ‰ most of the year but decreases to more neg-ative values of �7.1 ± 0.16 ‰ during summer (JJA) (red ar-row in Fig. 2l). The slabs’ calcite �13C signal (Fig. 2m) re-mains relatively constant at �10 ± 0.12 ‰ except for a bulgefrom August through January, with maximal �13C values of�9.0 ± 0.12 ‰ at the end of October (blue arrow in Fig. 2m).The two isotopic signatures are decoupled, suggesting thatdifferent forcing factors affect these signals.

The individual layers of the Proserpine stalagmite also dis-play an anticorrelation pattern between the oxygen and car-bon isotopic signals. The �18O composition oscillates aroundan average �6.5 ± 0.16 ‰ over a range of 0.9 ‰. The �13Cvaries around an average �8.4 ± 0.12 ‰ and over a range of2.4 ‰. Both oxygen and carbon isotopic signals measured inthe stalagmite correspond to the values measured on the glassslabs. At the end of a dark layer (dotted lines in Fig. 4) �18Ovalues reach their minimum while the �13C values reach theirmaximum, illustrating the anticorrelation pattern between the�18O and �13C values at seasonal level.






91



Figure 2. Precipitation intensities (a) and air temperatures (b) aremeasured at the Han-sur-Lesse station of the Royal MeteorologicalInstitute. All measured cave parameters – the cave air and drip wa-ter temperature (c and d), drip flow (e), drip water pH (f), cave airpCO2 (g), �13CCAVE AIR CO2 (h), �13CDIC (i), �18ORAINWATER(j), �18ODRIP WATER (k), and the �18O (l) and �13C (m) of freshlydeposited calcite on glass slabs – were measured during the cave-monitoring campaign from November 2012 through January 2014.Cave-monitoring results show that the cave conditions vary season-ally between a “summer mode” (red shadow) lasting from June toDecember and a “winter mode” (blue shadow) lasting from Decem-ber to June. The water �18O values (j and k) are reported in per millVSMOW. �13CCAVE AIR CO2 (h), �13CDIC (i), and �13C and �18Oof freshly deposited calcite on the glass slabs (l and m) are reportedin per mill VPDB.

Figure 3. Precipitation (a) and air temperatures (b) are measuredat the Han-sur-Lesse station of the Royal Meteorological Institute.The �18O and �D of the rainwater (c and d) and the �18O and �D ofthe drip water (e and f) are all reported in per mill VSMOW. At theend of March a cold temperature peak and prolonged snowfall causethe rainwater �18O to display a sharp drop. This negative spike inthe 18Orain water in March can be found back in the 18Odrip water inAugust, indicating that at least part of the infiltrating water reachesthe cave 5 to 6 months later.

5 Discussion

5.1 Forcing of the rain and drip water �18O and�D variations

Generally, rainwater �18O values at a specific location varydue to temperature changes, variation in the amount of rain-fall, fluctuations in the source of the rainwater or cloud track(Rozanski et al., 1992). The rainwater �18O signal increasesby a few per mill during the summer months, when air tem-perature is higher (Fig. 3a and c). A single larger drop indrip water �18O of about 0.1 ‰ occurs in March 2013 andis indicated by a red arrow in Fig. 3. Although it is only ofthe size of the analytical uncertainty, we consider it mean-ingful because several points support it and it is the onlyexcursion of this magnitude. This larger drop does not cor-respond with a decrease in temperature or with an increasein rainfall amount. A modification in rainwater source couldbe a plausible explanation but is not supported by changesin wind direction during that month based on RMI data. Onthe other hand, in March 2013 an unusually late snow layercovered the area for several weeks (RMI data). The observeddecrease in �18O is then most probably related to the collec-tion of isotopically light snow in the sampling bottle.

The average �18O and �D compositions of the drip wa-ter are slightly higher compared to the average �18O and �D






92



compositions of the rainwater. This slight increase in � valuesis possibly due to evaporation of precipitated meteoric waterfrom the surface before percolating into the epikarst, leadingto slightly increased �18O and �D values in the vadose wa-ter and in the drip water. Effective precipitation calculationsusing the Thornthwaite (1948) equation on the Han-sur-LessRMI data set for the years of 2011 and 2012 indicate thatJanuary is the only month where no evapotranspiration af-fects the isotopic composition of the water (Bonniver, 2011).Higher evaporation of a part of the precipitating water dur-ing the summer half year is most probably responsible forthe slightly more positive �18O and �D values in the drip wa-ter compared to the average rainwater �18O and �D values.

The �18O and �D compositions of the drip water dis-play almost no variations throughout the year, indicating thatthe water residence time is sufficiently long to homogenizeits isotopic composition (Fig. 2j). No detailed hydrologi-cal study of the Han-sur-Lesse cave system was carried outduring this study. However, the water residence time in theepikarst is supposed to be on the order of months to years.In general, depending on the epikarst thickness and epikarstflow systems, the transfer time of the drip water from thesurface, via the soil and epikarst, to the cave can vary frommonths to several decades. The epikarst thickness above theHan-sut-Lesse cave (40 m) is similar to that observed in shal-low caves such as Ernesto Cave in Italy or Bunker Cave inGermany with epikarst thicknesses between 15 and 30 m.Drip water residence times in such caves are suggested tovary from months to years (Kluge et al., 2010; Miorandi etal., 2010). Considering similar flow patterns, the Han-sur-Lesse residence time of the water in the epikarst may also beon the order of months to years such as Ernesto and Bunkercaves rather than on the order of decades such as observed indeep caves such as Monte Corchia (�1187 m) (Piccini et al.,2008).

The Proserpine drip water �18O displays a small negativeexcursion of 0.1 ‰ in July and August (Fig. 3e). This smalldrop in the drip water composition is most probably relatedto the strong decrease in isotopic composition of the rainwa-ter in March due to a single brief last snowfall. However,the intense mixing of the percolating vadose water in theepikarst reduces the �18O shift of about 8 ‰ in the meteoricto a hardly detectable one of about 0.1 ‰ in the drip water.The presence of this �18O minimum suggests that possiblya small part of the infiltrating water reaches the cave within5 to 6 months, albeit strongly mixed with the larger epikarstreservoir. A similar functioning was suggested in the PèreNoël cave, which also forms part of the Han-sur-Lesse karstsystem (Verheyden et al., 2008a). Uranine-tracing tests of thePère Noël epikarst showed first tracer occurrence 200 h afterinjection, with long restitution time of the curve going upto > 600 h containing superimposed smaller uranine concen-tration peaks (Bonniver, 2011). They described the epikarstas a large reservoir where a part of the infiltrating waterhas a fast flow to the cave while another part is stored for

longer time spans. A similar situation may be present in theepikarst above the Proserpine. However, a more detailed un-derstanding of the epikarst system with additional tracer testsis needed. So far, the hydrological system above the Proser-pine is interpreted as a piston flow system with the first wa-ters coming through after half a year.

5.2 Seasonal variations of the cave atmosphere anddrip flow

A large number of visitors enter the cave and the Salle-du-Dôme chamber every day. Such a large number of visitorscan induce an artificial increase in temperature, pCO2 levelsor a decrease in humidity (Baker and Genty, 1998). The stud-ied Salle-du-Dôme chamber has a height of 60 m and a widthof 150 m, for a total volume of 124 000 m3, and a river flow-ing at the bottom of the chamber causing good air mixing.Due to the large size and the good ventilation of the cham-ber, the effect of the visitors on the measured parameters isexpected to be negligible. This was confirmed by a series oftests where similar cave parameter values were measured be-fore and after groups visited the chamber. For the Salle-du-Dôme chamber, we consider the measured values to reflectthe natural conditions of the cave atmosphere. No cave en-trances connected to the studied Salle-du-Dôme have beenartificially enlarged.

Based on our observations, the temperatures of cave air,drip water and outside air all follow the same seasonal cycle.However, the air temperature, which varies between 20 �C insummer and 0 �C in winter outside the cave, only varies be-tween 14 �C and 11 �C inside the cave, respectively. Apartfrom the influence of the external temperature, the temper-ature regulation of the Salle-du-Dôme chamber is probablyalso influenced by the Lesse river flowing at the bottom of thechamber. Due to the water flow, a good mixing of the cave airis induced, causing the outside air to enter the cave chambermore easily. In addition, the water of the river transports en-ergy from the outside (gained before it entered the cave) tothe inner cave. This heat can then be released in the cave withmore impact in the warm summer months than in the wintermonths. Since no temperature record of the river is available,the effect of the Lesse river on the cave air temperature can-not be quantified.

Compared to other Belgian caves where values up to15 000 ppmv are measured for cave air CO2 (Verheyden,2001; Ek and Godissart, 2014), the CO2 content in the Salle-du-Dôme chamber, which hardly ever reaches 1000 ppmv, in-dicates that exchanges between cave air and external air mustbe relatively important. The Lesse river and the connectionsof the Salle-du-Dôme with neighboring chambers are mostprobably influencing the cave ventilation. In the Salle-du-Dôme, CO2 values fall close to outside air pCO2, and valuesof 400 to 600 ppmv are measured during much of the year(Fig. 2g). Only during summer, higher pCO2 values up to1000 ppmv are measured in the chamber, which corresponds






93



to the often observed summer increase in cave air pCO2(Baker and Genty, 1998; Spötl et al., 2005; Riechelmann etal., 2011). In general, even higher amplitude variations areobserved in most seasonal dynamically ventilated caves, suchas St Michael’s cave in Gibraltar with values varying between500 and 6000 ppmv (Mattey et al., 2008), Obir Cave in Aus-tria with values varying between 400 and 1500 ppmv (Spötlet al., 2005) Crag Cave (Ireland) with values between 1100and 8000 ppmv (Baldini et al., 2008). In the Han-sur-Lessecave, lower pCO2 values with smaller variation ranges areobserved, similar to what is observed in year-round well-ventilated caves such as Bunker Cave (Riechelmann et al.,2011). This confirms the former suggestion that the Han-sur-Lesse cave air is most probably well mixed year-round dueto good ventilation, due to the three known entrances and theLesse river, which continuously flows through it. The dif-ference between seasonally ventilated caves and year-roundventilated caves is also visible in the �13C composition of thecave air (Riechelmann et al., 2011). The �13C composition ofthe cave air in seasonally dynamically ventilated caves variesover a larger range around 10 ‰ (Spötl et al., 2005; Matteyet al., 2008) compared to year-round ventilated caves, suchas Bunker Cave, where a variation range of only 3 ‰ is ob-served (Riechelmann et al., 2011). The Han-sur-Lesse cavedisplays a �13C variation of 4 ‰, corresponding with a levelthat is expected for a year-round well-ventilated cave. Al-though the cave is well ventilated, a more negative �13C sig-nature of the CO2 air down to �21.9 ‰ is measured in theSalle-du-Dôme in spring and summer (Fig. 2h), suggestingenhanced input of C3-plant-derived soil CO2 into the caveatmosphere.

Seasonal variations are also seen in the flow rates abovethe stalagmite, with less water dripping on the stalagmiteduring spring and summer (Fig. 2e). Drip rate monitoringin the Père Noël cave, which is part of the Han-sur-Lessecave system, also demonstrated seasonal drip rate variations,with higher discharge amounts in winter compared to sum-mer (Genty and Deflandre, 1998). This seasonal variationin discharge above the Han-sur-Lesse cave is mostly relatedto seasonal variations in evapotranspiration. This lowers thequantity of water feeding the epikarst during summer, reduc-ing the amount of water that can reach the cave. In winter,the situation reverses: the activity of the vegetation cover di-minishes and air temperatures lowers, reducing evapotran-spiration, all of which allows more water to enter the soiland the epikarst. The pressure on the piston flow system israised and more water is pushed into the cave, increasing thedischarge above the stalagmite. Effective precipitation calcu-lation of the Han-sur-Lesse karst aquifer indicates that fromJune to September the net evaporation or evapotranspirationis larger than the amounts of precipitation. The karst aquiferis thus only recharged during winter and spring: from Oc-tober to May (Bonniver, 2011). This seasonal recharge isclearly visible in the evolution of the drip rate. A gradual

decrease in drip rate occurs from June to November, whiledrip rate clearly increases from December to May (Fig. 2e).

5.3 Seasonal variations in the pH and �13C of thedrip water

The drip water pH and �13CDIC display similar seasonal vari-ations with an increase during summer and a decrease duringwinter. Synchronous seasonal variations in pH and �13CDICdriven by cave air pCO2 variations are commonly observedin caves (Spötl et al., 2005; Mattey et al., 2008). The de-gree of CO2 degassing can influence the pH and �13CDIC ofthe drip water; removal of CO2 from the water during de-gassing increases its pH, and removal of light 12C isotopesfrom the drip water during degassing increases its �13CDIC.The degree of CO2 degassing of the drip water is mainlydriven by the pCO2 difference between the cave air (withlow pCO2) and the drip water (with high pCO2) (Mühling-haus et al., 2009; Dreybrodt and Scholz, 2011; Deininger etal., 2012). This process is most pronounced in caves witha large seasonal variation in cave air pCO2 such as in theseasonally ventilated St Michael’s Cave in Gibraltar (Mat-tey et al., 2008) or Obir Cave in Austria (Spötl et al., 2005).The monitoring data indicate that the similar seasonal varia-tions of the drip water pH and �13CDIC are not a response tothe seasonal variations of cave air pCO2. The Han-sur-Lessecave is a year-round well-ventilated cave with a much smallerpCO2 seasonal variation range, roughly between 400 and1000 ppmv (Fig. 2g). If present in the Han-sur-Lesse cave,the effect of degree of CO2 degassing on the drip water pHand �13CDIC is expected to be rather small. More importantis that, if the degree of CO2 degassing is driving the pH and�13CDIC variations at the Proserpine drip site, they must an-ticorrelate with the pH and �13CDIC of the drip water. Suchanticorrelation is not visible in the results (Fig. 2f, g and i);the pCO2 peaks in summer, while the pH and �13CDIC dis-play a more gradual decrease. Furthermore, modeling resultshave shown that the effect of degree of CO2 degassing on the�13C and pH of the drip water is negligible for fast drip sitessuch as the Proserpine (Mühlinghaus et al., 2009; Dreybrodtand Scholz, 2011; Deininger et al., 2012). The effect of de-gassing due to seasonal pCO2 variations in the cave air isthus most probably not driving the variations observed in thepH and the �13CDIC of the drip water.

Another factor that can seasonally increase both the pHand �13CDIC of the drip water (and of the deposited calcite)is prior calcite precipitation (PCP). PCP is a common pro-cess occurring in karst aquifers (e.g., Fairchild et al., 2000,2006; Verheyden et al., 2008a; Riechelmann et al., 2011).When downward-percolating water encounters a zone withlower pCO2, degassing occurs and calcite can precipitate.This process is enhanced during drier periods and lower dis-charge as aerated zones increase in the epikarst and residencetime of the water becomes longer (Fairchild et al., 2000).The CO2 degassing in the epikarts causes both the �18O and






94


�13C signals in the remaining water to increase. However, ifthe transition time of the solution until reaching the cave islonger than several days, the oxygen isotope equilibrium ofthe CO2–H2O–CaCO3 system in the water is re-establishedagain (Dreybrodt and Scholz, 2011). The net effect is thus anincrease in �13C and pH of the drip water, while no variationis visible in the �18O of the drip water. Furthermore, effectiveprecipitation data show that no water recharges the Han-sur-Lesse epikarst between June and the end of September (Bon-niver, 2011). This lower epikarst recharge is also visible ina decrease of drip flow between June and October (Fig. 2e).The decreased recharge of the epikarst is suggested to in-crease the aerated zone in the epikarst and thus enhance PCPgradually from June throughout September. The increase ofrecharge in winter decreases the effect of PCP, and the dripwater pH and �13C display a gradual decreasing trend. PCPis thus strongly suggested to cause the seasonal variations inpH and �13C of the Proserpine drip water.

The occurrence of PCP can be indicated by an increaseddrip water �13C composition compared to the soil water �13Cand/or by a negative relationship between the Mg / Ca ratioor Sr / Ca of the drip water and the Ca2+ concentration anddrip rate (Fairchild and Treble, 2009). However, no data onthe evolution of the Ca, Mg or Sr concentration in the Pros-erpine drip water or the �13C composition of the soil waterare available, making it difficult to estimate the importance ofthe process. Another argument in support of the PCP effect isthe formation of a stalactite above the stalagmite. However,during this formation of PCP, and due to the high drip rateof the Proserpine, the transition time is not long enough tore-establish the increased �18O signal in the water and boththe �13C and the �18O values of the drip water are increased(Dreybrodt and Scholz, 2011). Since the �18O and �13C sig-nals of the Proserpine drip water evolve differently, stalac-tite growth can increase both signals but cannot be respon-sible for the seasonal variations. Furthermore, the draperyabove the Proserpine is estimated to be about 0.5 to 1 m longand is thus rather small compared to the 2 to 3 m height ofthe Proserpine stalagmite, which displays very high growthrates up to 2 mm yr�1 (Verheyden et al., 2006). It is thus notcertain whether the stalactite is actively growing in presenttimes. Also, due to the continuous water flow, calcite willmost probably not have the time to precipitate stalactite cal-cite before falling on the stalagmite. Dating and analysis ofthe stalactite above the Proserpine would provide the missinginformation. However, its location 30 m above the Proserpinemakes it difficult to be sampled.

In summary, the cave system is subdivided into a “win-ter mode” lasting from December to June and a “summermode” from June to December (Fig. 2). During the wintermode, cave air and drip water temperature and pCO2 are lowand �13C values of the cave air CO2 are high. The plant cov-erage above the cave inactively facilitates water recharge ofthe epikarst reservoir, leading to a decrease of PCP, whichgradually decreases the pH and the �13CDIC of the drip wa-

ter. During the summer mode, cave air and drip water tem-peratures increase, the pCO2 increases and the �13C valuesof the cave air CO2 decrease. The plant coverage above thecave reactivates, leading to lower water recharge, which en-hances the PCP effect, leading to increased pH and �13CDICvalues in the drip water.

5.4 The �18O of the precipitated calcite reflectstemperature variations

The �18O of calcite deposited on the glass slabs varies sea-sonally with more negative values during summer (Fig. 2l).If the calcite is deposited in isotopic equilibrium with its dripwater, these variations can be caused by changes in the �18Oof the drip water and/or by changes in temperature that af-fect the fractionation factor between the drip water and theprecipitating calcite (Dreybrodt and Scholz, 2011). If not de-posited in equilibrium, the seasonal �18O variations on theglass slabs would be due to disequilibrium effects that re-quire further investigation.

A first step in understanding the �18O system of the Salle-du-Dôme demands determining whether the calcite is de-posited in equilibrium or not. In speleothems, this is tra-ditionally done by applying the Hendy test (Hendy, 1971),which compares the �18O and �13C values in the center ofthe stalagmite with those on the sides within a single growthlayer. However, since we work on a drill core taken “in themiddle” of a 1–2 m wide speleothem, the Hendy test cannotbe applied. As an alternative, we calculated equilibrium con-ditions of the deposited calcite using the calcite–water frac-tionation factor. Different authors have proposed fractiona-tion factors based on three different approaches, summarizedin Table 1.

The first approach, using laboratory experiments, has beentested in different studies, each giving another value for awater–calcite fractionation factor. The most-used laboratory-established fractionation factors remain the ones by O’Neilet al. (1969) later modified by Friedman and O’Neil (1977),the relationship of Kim and O’Neil (1997) later modified byKim et al. (2007), and the results of Tarutani et al. (1969)and of Jimenez-Lopez et al. (2001) (Table 1). A second ap-proach to determine fractionation factors is established by us-ing theoretical models such as the ones from Horita and Clay-ton (2007) and from Chacko and Deines (2008). A third ap-proach consists of using cave-monitoring data to make an av-erage of the in-cave-observed fractionation factors (Demenyet al., 2010; Tremaine et al., 2011). Tremaine et al. (2011)established such a “cave calcite” relationship by doing a bestfit through the data on a large number of modern caves atdifferent latitudes, altitudes and temperatures.

Applying the different fractionation factors to our datashows that the measured �18O signals of the glass-slab cal-cite correspond within 1 ‰ with the calculated values, sug-gesting that it is deposited close to equilibrium with the dripwater (Table 2). However, variations occur between the re-






95



Table 1. A selection of the most commonly used water–calcite oxygen fractionation factors. Laboratory and theoretical approaches differfrom the relationships found in cave settings.

Author Method 1000 ⇥ ln↵ d↵/dT (‰ �C�1)

O’Neil et al. (1969) modified Laboratory 2.78(106T �2) � 2.89 �0.24by Friedman and O’Neil (1977)Kim and O’Neil (1997) modified Laboratory 18.03(106T �2) � 31.17 �0.22by Kim et al. (2007)Chacko and Deines (2008) constructed Theoretical calculation 2.57333(106T �2) � 0.869 �0.22relation by Tremaine et al. (2011)Horita and Clayton (2007) Calculations compared with 0.9521(106T �2) + 11.59(103T �1) � 21.56 �0.23

experimental resultsTremaine et al. (2011) Linear best fit through large (16.01 ± 0.65) ⇥ (103T �1) � (24.6 ± 2.2) �0.18

number of cave studiesDemeny et al. (2010) Cave-monitoring results 17500 ⇥ T �1 � 29.89 �0.22

Hungarian cave

sults derived from different methods. Among the laboratory-established relationships, that of O’Neil et al. (1969) mod-ified by Friedmann and O’Neil (1977) best corresponds toour observations with on average 0.25 ‰ difference with thevalues measured on the glass slabs. This is also confirmed inother studies where this experimental fractionation factor isconsidered to give the best approximation for in-cave obser-vations (McDermott et al., 2005; Riechelmann et al., 2013).The theoretical values (Horita and Clayton, 2007; Chackoand Deines, 2008) suggest a more negative �18O compositionfor the deposited calcites. This is also the case for other stud-ies where theoretical results seem to overestimate the frac-tionation factor (Demeny et al., 2010; Tremaine et al., 2011).For our results the best agreement is found by applying theTremaine et al. (2011) fractionation factor, which is not sur-prising since the latter is based on experimental studies oncalcite formed in caves. However, the good match betweenthe measured values and those calculated from the Tremaineet al. (2011) relationship does not constitute a proof of equi-librium condition. The fractionation factor from Tremaine etal. (2011) is derived from the natural system, and it is ques-tionable whether real equilibrium does exist in nature.

The small difference between the measured data and theO’Neil et al. (1969) relationship modified by Friedmann andO’Neil (1977) indicates that the deposition of calcite occursnear oxygen isotopic equilibrium in the Salle-du-Dôme. Sea-sonal variations of �18O observed in calcite are likely causedby variations in the drip water �18O composition and/or in thetemperature-dependent fractionation factor (Fairchild et al.,2006). However, within analytical uncertainty, the �18O com-position of the drip water remains constant throughout theyear. Consequently, variations in the fractionation factor dueto temperature changes in the cave air likely explain the sea-sonal pattern seen in the �18O composition of the glass slab.If the most commonly accepted temperature dependence ofthe water–calcite fractionation factor for the oxygen isotopesof 0.247 ‰ 1 �C�1 (O’Neil et al., 1969) is used, our mea-

sured net difference of the �18O values of the glass-slab cal-cite (i.e., �18O range of the glass-slab calcite minus the �18Orange of the drip water) of 0.58 ‰ would result from a 2.3 �Cvariation in the drip water temperature. This temperature cor-responds well with the 2–2.5 �C temperature range measuredin the drip waters. This correspondence constitutes a strongconfirmation of both the isotopic equilibrium and the temper-ature dependence of the calcite �18O.

To summarize, the �18O composition of the glass-slab cal-cite (Fig. 2l) is deposited very close to equilibrium with itsdrip water. Seasonal variations in �18O composition of thecalcite on the glass slabs are caused by the very similar sea-sonal temperature variation of the cave air and the drip water(Fig. 2c and d). A temperature increase of 1 �C correspondswith a decrease of 0.20 ‰ in isotopic composition of the de-posited calcite. The warmer the cave air, the more negativethe �18O composition of the formed calcite. No disequilib-rium processes are active, as they would shift the isotopiccomposition to heavier values with increasing temperature.

5.5 The calcite �13C reflects equilibrium conditionsand is driven by PCP

The carbon isotopic composition of calcite deposited in equi-librium with its drip water depends on (i) the �13C of the DICin the drip water and (ii) the temperature-dependent fraction-ation factor between the DIC and the deposited calcite fora pH range around 7. The average �13C values of the DICin the Han-sur-Lesse drip water (Fig. 2i) display a season-ality between �12.2 ‰ at the end of winter and �11 ‰ atthe end of summer. The �13C of the deposited glass-slab cal-cite are on average 1 ‰ heavier than the �13C of the dripwater and display similar seasonal variations (r = 0.62 andp = 0.0017). The fractionation factor between the DIC andthe deposited calcite as determined by Emrich et al. (1970)and Dulinski and Rozanki (1990) are estimated to correspondbest with the natural speleothem depositional conditions, and






96



Table 2. Calculated �18O values for calcite deposited in equilibrium with its drip water for different authors. Values were calculated usingthe average drip water composition of �7.65 ‰ for the lowest, highest and average measured temperatures at the Salle-du-Dôme. The secondpart of the table shows the difference between the calculated �18O values and the �18O values measured on the glass slabs. Best results aregiven by the “in-cave” relationship of Tremaine et al. (2011).

�18ODripWater = �7.65 ‰ Predicted �18O Difference with glass-slab valuesTmin Tmid Tmax �18Omin �18Omid �18Omax

Author 10.5 �C 12 �C 14.5 �C �6.25 �6.55 �6.98

O’Neil et al. (1969) �6.44 �6.80 �7.39 0.19 0.25 0.41modified by Friedmann and O’Neil (1977)Kim and O’Neil (1997) �5.71 �6.04 �6.59 �0.54 �0.51 �0.39modified by Kim et al. (2007)Horita and Clayton (2007) �6.96 �7.30 �7.85 0.71 0.75 0.87Chacko and Deines (2008) �6.98 �7.32 �7.86 0.73 0.77 0.88Demeny et al. (2010) �6.30 �6.62 �7.15 0.05 0.07 0.17Tremaine et al. (2011) �6.26 ± 0.09 �6.55 ± 0.08 �7.04 ± 0.06 0.01 ± 0.09 0 ± 0.08 0.06 ± 0.06

both suggest " values around 1 ‰ for temperatures between5 and 20 �C. The 1 ‰ enrichment of the �13C calcite on theglass slabs compared to the drip water �13C indicates that the�13C is deposited in equilibrium with the drip water.

Our hypothesis that the deposition of the calcite occurs inisotopic equilibrium is also confirmed by the different evolu-tion of the �18O (Fig. 2l) and the �13C (Fig. 2m) values of theglass-slab calcite, which indicates that both proxies evolveindependently under the influence of different factors. Em-rich et al. (1970) and Dulinski and Rozanki (1990) both sug-gest a small temperature dependence of the fractionation fac-tor of about 0.07 ‰/ �C. The ⇠ 2.5 �C seasonal temperaturevariation in the Salle-Du-Dôme causes a change of hardly0.17 ‰ in the calcite composition. The seasonal temperaturevariations are thus too small to significantly influence the car-bon fractionation factor and thus the isotopic composition ofthe deposited calcite. Furthermore typical disequilibrium ef-fects such as increased drip interval and stronger pCO2 de-gassing of the drip water due to longer residence times havebeen shown not to affect the �13C of calcite deposited un-der fast-flow sites (Dreybrodt and Scholz, 2011; Deininger etal., 2012). The �13C composition of the deposited Proserpinecalcite is thus considered deposited in equilibrium with thedrip water. The seasonal variations in calcite �13C are causedby seasonal variations in the drip water �13C, which relatesprobably to the intensity of PCP.

5.6 Variations in the �18O and �13C of the stalagmitereflect seasonal variations

The Proserpine stalagmite displays clear lamination formedby alternating dark, compact layers and white, more porouslayers (Verheyden et al., 2006). The seasonal character of thelayering in the Proserpine stalagmite, with one dark and onewhite layer deposited every year, was already demonstratedby Verheyden et al. (2006). However, these authors were notable to determine the correspondence between layer type and

season. In our studied core, 10 additional layer couplets (=dark + white layer) can be counted, compared to the Proser-pine core of Verheyden et al. (2006) drilled in 2001. The 10additional layer couplets, counted over a period of 10 years,confirm that one layer couplet is deposited every year. Layercouplets establish the age model of the laminated part of thestalagmite. Twenty-six layer couplets are counted from thetop of the stalagmite to the start of the isotope sampling, in-dicating that the youngest analyzed layer formed in 1985 AD.The isotopic measurements were conducted on nine consec-utive layer couplets and consequently run from 1985 to 1976AD (Fig. 5).

A first conclusion from the isotopic analyses of the indi-vidual layers is that the �13C and �18O signals display anopposite behavior within one layer (Fig. 4). When the �13Csignal reaches its maximum value at the end of a dark layer,the �18O value arrives at its minimum value. This anticorre-lation is also seen in the cave-monitoring results where �13Cvalues increase through summer while the �18O values arelow. This indicates that dark layers are most probably de-posited in summer, while light layers are formed in winter.Furthermore, a link between the variation range of the �13Cand �18O signals and the color intensity of a layer (darkerdark layers or lighter white layers) may be possible. The lay-ers around 1977 AD display a stronger visual color contrastbetween dark and white layers compared to the layers around1985 AD (Fig. 4). In parallel, a larger difference in the iso-topic composition of two consecutive layers is observed forthat year. More compact and darker layers have more neg-ative �18O and more positive �13C values while whiter andmore porous layers have more negative �13C and more posi-tive �18O values.

The difference in calcite between white and dark may pos-sibly be related to seasonal changes in growth rate of the cal-cite. The mean annual growth rate of speleothems primarilydepends on the drip water calcium ion concentration (Gentyet al., 2001). In addition, there is a correlation between the






97


Figure 4. The �18O and �13C signals, reported in per mill VPDB,from the Proserpine stalagmite anticorrelate at monthly scale. Notethat, in the scanned figure of the Proserpine slab, dark and com-pact layers become whiter due to the translucent light of the scanwhile the white and porous layers become dark. At the end of adark compact layer (dotted line), the �18O values reach their min-imum values, while the �13C values reach their maximum. Darklayers are formed during summer, when cave temperatures are high(leading the low �18O) and PCP increases (leading to high �13C).The clearer the lamination, the larger the amplitude of the variationsin the isotopic composition.

measured growth rate and the temperature due to the inter-relationship between calcium ion concentration, soil pCO2and surface temperature. However, the latter link is not al-ways true since PCP and/or lack of soil cover can decreasethe calcium ion concentration of the water and still repre-sent high temperatures (Genty et al., 2001). Discharge alsoaffects growth rate but is more pronounced for slowly drip-ping sites. For faster-drip sites, such as the Proserpine, cal-cite saturation index is the main factor driving the growthrate changes (Genty et al., 2001). Unfortunately, no dataon the seasonal evolution of the calcite saturation index atthe Proserpine grow site are available. However, seasonalvariations in the drip water �13C values of the Proserpinehave shown that PCP affects its geochemistry in summer.PCP lowers the calcite saturation index (Genty et al., 2001;Riechelmann et al., 2013). The increase of the PCP effect insummer strongly suggests that the Proserpine summer watercontains less dissolved Ca2+, which may be responsible fora decreased growth rate in summer.

The observations gained from combined monitoring ob-servation and stable isotopic analyses answer the remain-ing question in Verheyden et al. (2006). Darker layers re-flect summer cave conditions, and white layers reflect win-ter cave conditions. In addition, in the summer, the proba-bly lower calcite saturation index of the drip water due to

Figure 5. Detail of the upper 16 mm of the Proserpine core drilled in2011. Ten additional dark layers are counted compared to the coreof Verheyden et al. (2006), confirming the seasonal character of thelayering. Ages are based on layer counting, with one dark and onewhite layer deposited every year. The frame indicates the locationof the monthly resolved �18O and �13C measurements.

enhanced PCP causes a decreased in the calcite growth ratein the speleothem. Darker and more compact calcite is sug-gested to reflect slower growth rate conditions. Further in-vestigation in the seasonal cycle of the carbonate-dissolvedion concentration of the Proserpine drip water is necessaryto confirm this hypothesis.

6 Conclusions

Through a biweekly cave monitoring and a high-resolutionstable isotopic record of a recent finely laminated part ofa growing speleothem, the seasonal variation of the caveenvironment and how these parameters are recorded in thespeleothem proxies are documented.

1. The temperature effect clearly influences the �18O and�D composition of the rainwater with increasing valuesduring the summer months. The �18O and �D compo-sition of the drip water remains constant, indicating along residence time of the water in the epikarst and awell-mixed aquifer. A residence time of more than oneyear is assumed with the first waters percolating intothe cave 5 to 6 months after entering the soil. Such longwater residence time in the epikarst combined with a re-action time of 24 h to a heavy-rainfall event indicates apiston flow hydrology in the epikarst.

2. The records of the different measured factors in andoutside the cave suggest that the physico-chemical be-havior of the Han-sur-Lesse cave closely responds tothe seasonally varying external climate. The cave inte-rior climatology varies between a summer mode lastingfrom June to December and a winter mode from De-cember to June. During the summer mode, the cave airand drip water temperatures are higher, the drip flow






98


decreases the cave air pCO2 increases. During the win-ter mode the temperature cools down and the vegetationcover becomes less active, leading to an opposite be-havior of the measured factors compared to the summermode.

3. Seasonal variation in drip water pH and �13C valuemost probably relate to the seasonality in PCP intensity,driven by seasonal changes in water discharge. Duringsummer, lower recharge will cause stronger PCP, whichgradually increases the pH and �13C composition of thedrip water between June and September. In winter, thevadose zone gets recharged with water and PCP de-creases. The pH and �13C composition of the drip waterdecreases gradually from October trough May.

4. Calcite precipitated on glass slabs indicates more neg-ative �18O and more positive �13C values in summer.Both isotopic signals are deposited in equilibrium withtheir drip water. The seasonal variations of the �18Ovalues of precipitated calcite are caused by the sea-sonal temperature variations inside the cave. The sea-sonal variations of the �13C values are mainly controlledby the seasonal variation in drip water �13C caused byseasonally varying PCP.

5. The studied part of the Proserpine stalagmite displaysseasonal layering with one dark and one white layer de-posited every year. The opposite seasonal behavior ofthe �18O and the �13C signals on the glass slabs is alsovisible in the isotopic signals of the individual layersin the stalagmite. Dark layers have low �18O and high�13C values, while white layers have high �18O and low�13C values. Dark layers are formed during the cave’ssummer mode, while white layers grow calcite duringthe cave’s winter mode. The clearer the lamination andcoloring of the layers, the larger the amplitudes of thevariations in the isotopic composition.

Acknowledgements. We thank the “Domaine des Grottes de Han”for allowing us to sample the stalagmite and visit the cave on abiweekly basis, Etienne Lannoy for sampling the rain and dripwater during the monitoring period, Claire Mourgues and DavidVerstraeten for the help in the lab and Mr. Van Dierendonck for hisinterest and support. P. Claeys thanks the Hercules Foundation andResearch Foundation Flanders (FWO, project G-0422-10).

Edited by: J. Guiot

References

Baker, A. and Genty, A.: Environmental pressures on conservingcave speleothems: effects of changing surface land use and in-creased cave tourism, J. Environ. Manage., 53, 165–175, 1998.

Baker, A., Genty, D., Dreybrodt, W., Barnes, W. L., Mockler, N. J.,and Grapes, J.: Testing theoretically predicted stalagmite growthrate with Recent annually laminated samples: implications forpast stalagmite deposition, Geochim. Cosmochim. Ac., 62, 393–404, 1998.

Baker, A., Asrat, A., Fairchild, I. J., Leng, M. J., Wynn, P. M.,Bryant, C., Genty, D., and Umer, M.: Analysis of the climatesignal contained within delta O-18 and growth rate parametersin two Ethiopian stalagmites, Geochim. Cosmochim. Ac., 71,2975–2988, 2007.

Baldini, J. U. L., McDermott, F., Hoffmann, D. L., Richards, D. A.,and Clipson, N.: Very high-frequency and seasonal cave atmo-sphere pCO2 variability: Implications for stalagmite growth andoxygen isotope-based paleoclimate records, Earth Planet. Sci.Lett., 272, 118–129, 2008.

Bonniver, I.: Etude Hyrogeologique et dimmensionnement parmodelisation du “systeme-tracage” du reseau karstique the Han-sur-Lesse (Massif de Boine, Belgique), 2011, Geologie, FUNDPNamur, Namur, 93–97, 2011.

Chacko, T. and Deines, P.: Theoretical calculation of oxygen iso-tope fractionation factors in carbonate systems, Geochim. Cos-mochim. Ac., 72, 3642–3660, 2008.

Deininger, M., Fohlmeister, J., Scholz, D., and Mangini, A.: Isotopedisequilibrium effects: The influence of evaporation and ventila-tion effects on the carbon and oxygen isotope composition ofspeleothems – A model approach, Geochim. Cosmochim. Ac.,96, 57–79, 2012.

Delvaux De Fenffe, D.: Geologie et tectonique du Parc de Lesse etLomme au bord sud du Bassin de Dinant (Rochefort, Belgique),Bulletins de la Societe Belge de Geologie, 94, 81–95, 1985.

Demeny, A., Kele, S., and Siklosy, Z.: Empirical equations for thetemperature dependence of calcite-water oxygen isotope frac-tionation from 10 to 70 �C, Rapid Commun. Mass Sp., 24, 3521–3526, 2010.

Dreybrodt, W.: Chemical kinetics, speleothem growth and climate,Boreas, 28, 347–356, 1999.

Dreybrodt, W.: Evolution of the isotopic composition of carbon ina calcite precipitating H2O-CO2-CaCO3 solution and the relatedisotopic composition of calcite in stalagmites, Geochim. Cos-mochim. Ac., 72, 4712–4724, 2008.

Dreybrodt, W. and Scholz, D.: Climatic dependence of stable car-bon and oxygen isotope signals recorded in speleothems: Fromsoil water to speleothem calcite, Geochim. Cosmochim. Ac., 75,734–752, 2011.

Dulinski, M. and Rozanski, K.: Formation of 12C/13C isotope ratiosin spsleothems: a semi-dynamic model, Radiocarbon, 32, 7–16,1990.

Ek, C. and Godissart, J.: Carbon dioxide in cave air and soil airin some karstic areas of Belgium, A prospective view, Geolog.Belgica, 17, 102–106, 2014.

Emrich, K., Ehhalt, D. H., and Volgel, J. C.: Carbon isotope frac-tionation during the precipitation of calciumcarbonates, EarthPlanet. Sci. Lett., 8, 363–371, 1970.

Fairchild, I. J., Borsato, A., Tooth, A. F., Frisia, S., Hawkesworth,C. J., Huang, Y. M., McDermott, F., and Spiro, B.: Controls on







99



trace element (Sr-Mg) compositions of carbonate cave waters:implications for speleothem climatic records, Chem. Geol., 166,255–269, 2000.

Fairchild, I. J., Smith, C. L., Baker, A., Fuller, L., Spotl, C., Mat-tey, D., McDermott, F., and Eimp: Modification and preserva-tion of environmental signals in speleothems, Earth-Sci. Rev., 75,105–153, 2006.

Fairchild, I. J. and Treble, P. C.: Trace elements in speleothems asrecorders of environmental change, Quat. Sci. Rev., 28, 449–468,2009.

Friedman, I. and O’Neil, J. R.: Compilation of stable isotope frac-tionation factors of geochemical interest, Geolog. Surv. Prof. Pa-per 440-KK, 177, 1977.

Genty, D., Baker, A., and Vokal, B.: Intra- and inter-annual growthrate of modern stalagmites, Chem. Geol., 176, 191–212, 2001.

Genty, D., Blamart, D., Ouahdi, R., Gilmour, M., Baker, A.,Jouzel, J., and Van-Exter, S.: Precise dating of Dansgaard–Oeschger climate oscillations in western Europe from stalagmitedata, Nature, 421, 833–837, 2003.

Genty, D. and Deflandre, G.: Drip flow variations under a stalactiteof the Pere Noel cave (Belgium). Evidence of seasonal variationsand air pressure constraints, J. Hydrol., 211, 208–232, 1998.

Genty, D. and Quinif, Y.: Annually laminated sequences in the in-ternal structure of some Belgian stalagmites – Importance for pa-leoclimatology, J. Sediment. Res., 66, 275–288, 1996.

Hendy, C. H.: Isotopic geochemistry of speleothems: 1. Calcula-tions of effects on different modes of formation on isotopic com-position of speleothems and their applicability as paleoclimaticindicators, Geochim. Cosmochim. Ac., 35, 801–824, 1971.

Horita, J. and Clayton, R. N.: Comment on the studies of oxygenisotope fractionation between calcium carbonates and water atlow temperatures by Zhou and Zheng (2003, 2005), Geochim.Cosmochim. Ac., 71, 3131–3135, 2007.

Jimenez-Lopez, C., Caballero, E., Huertas, F. J., and Ro-manek, C. S.: Chemical, mineralogical and isotope behavior, andphase transformation during the precipitation of calcium carbon-ate minerals from intermediate ionic solution at 25 �C, Geochim.Cosmochim. Ac., 65, 3219–3231, 2001.

Jochum, K. P., Scholz, D., Stoll, B., Weis, U., Wilson, S. A., Yang,Q., Schwalb, A., Boerner, N., Jacob, D. E., and Andreae, M.O.: Accurate trace element analysis of speleothems and biogeniccalcium carbonates by LA-ICP-MS, Chem. Geol., 318, 31–44,2012.

Kim, S.-T., Mucci, A., and Taylor, B. E.: Phosphoric acid fraction-ation factors for calcite and aragonite between 25 and 75 �C: re-visited, Chem. Geol., 246, 135–146, 2007.

Kim, S. T. and O’Neil, J. R.: Equilibrium and nonequilibriumoxygen isotope effects in synthetic carbonates, Geochim. Cos-mochim. Ac., 61, 3461–3475, 1997.

Kluge, T., Riechelmann, D. F. C., Wieser, M., Spötl, C., Sultenfuss,J., Schroder-Ritzrau, A., Niggemann, S., and Aeschbach-Hertig,W.: Dating cave drip water by tritium, J. Hydrol., 394, 396–406,2010.

Kottek, M., Grieser, J., Beck, C., Rudolf, B., and Rubel, F.: Worldmap of the Koppen-Geiger climate classification updated, Mete-orologische Zeitschrift, 15, 259–263, 2006.

Lachniet, M. S.: Climatic and environmental controls onspeleothem oxygen-isotope values, Quat. Sci. Rev., 28, 412-432,2009.

Mattey, D., Lowry, D., Duffet, J., Fisher, R., Hodge, E., and Frisia,S.: A 53 year seasonally resolved oxygen and carbon isotoperecord from a modem Gibraltar speleothem: Reconstructed dripwater and relationship to local precipitation, Earth Planet. Sci.Lett., 269, 80–95, 2008.

McDermott, F.: Centennial-scale Holocene climate variability re-vealed by a high-resolution speleothem �18O record from SWIreland (9 November, pg 1328, 2001), Science, 309, 1816–1816,2005.

McDermott, F., Schwarcz, H. P., and Rowe, P. J.: Isotopesin speleothems, in: Isotopes in Paleoenvironmental Research,edited by: Leng, M. J., Springer Netherlands, 185–226, 2005.

Miorandi, R., Borsato, A., Frisia, S., Fairchild, I. J., and Richter, D.K.: Epikarst hydrology and implications for stalagmite captureof climate changes at Grotta di Ernesto (NE Italy): results fromlong-term monitoring, Hydrol. Proc., 24, 3101–3114, 2010.

Mühlinghaus, C., Scholz, D., and Mangini, A.: Modelling stalag-mite growth and �13C as a function of drip interval and tempera-ture, Geochim. Cosmochim. Ac., 71, 2780–2790, 2007.

Mühlinghaus, C., Scholz, D., and Mangini, A.: Modelling fraction-ation of stable isotopes in stalagmites, Geochim. Cosmochim.Ac., 73, 7275–7289, doi:10.1016/j.gca.2009.09.010, 2009.

O’Neil, J. R., Clayton, R. N., and Mayeda, T. K.: Oxygen isotopefractionation in divalent metal carbonates J. Chem. Phys., 51,5547–5558, 1969.

Oster, J. L., Montanez, I. P., and Kelley, N. P.: Response of a moderncave system to large seasonal precipitation variability, Geochim.Cosmochim. Ac., 91, 92–108, 2012.

Piccini, L., Zanchetta, G., Drysdale, R. N., Hellstrom, J., Isola,I., Fallick, A. E., Leone, G., Doveri, M., Mussi, M., Mantelli,F., Molli, G., Lotti, L., Roncioni, A., Regattieri, E., Meccheri,M., and Vaselli, L.: The environmental features of the MonteCorchia cave system (Apuan Alps, central Italy) and their effectson speleothem growth, International Journal of Speleology, 37,153–172, 2008.

Quinif, Y.: Une nouvelle topographie de la Grotte de Han, Lapiazhors serie “Special Han”, 15–18, 1988.

Riechelmann, D. F. C., Deininger, M., Scholz, D., Riechelmann, S.,Schroeder-Ritzrau, A., Spoetl, C., Richter, D. K., Mangini, A.,and Immenhauser, A.: Disequilibrium carbon and oxygen isotopefractionation in recent cave calcite: Comparison of cave precipi-tates and model data, Geochim. Cosmochim. Ac., 103, 232–244,2013.

Riechelmann, D. F. C., Schroeder-Ritzrau, A., Scholz, D.,Fohlmeister, J., Spoetl, C., Richter, D. K., and Mangini, A.: Mon-itoring Bunker Cave (NW Germany): A prerequisite to interpretgeochemical proxy data of speleothems from this site, J. Hydrol.,409, 682–695, 2011.

Rozanski, K., Araguasaraguas, L., and Gonfiantini, R.: Relationshipbetween long-term trends of 18O isotope composition of precip-itation and climate, Science, 258, 981–985, 1992.

Ruan, J. and Hu, C.: Seasonal variations and environmental controlson stalagmite calcite crystal growth in Heshang Cave, CentralChina, Chinese Sci. Bull., 55, 3929–3935, 2010.

Scholz, D., Muehlinghaus, C., and Mangini, A.: Modelling �13Cand �18O in the solution layer on stalagmite surfaces, Geochim.Cosmochim. Ac., 73, 2592–2602, 2009.

Smart, P. L. and Friedrich, H.: Water movement and storage in theunsaturated zone of amaturely karstified aquifer, Mendip Hills,


http://dx.doi.org/10.1016/j.gca.2009.09.010





100



England, Proceedings of the Conference on Environ-mentalProblems in Karst Terrains and Teir Solution, Bowling Green,Kentucky, National Water Well Association, 57–87, 1987.

Spötl, C., Fairchild, I. J., and Tooth, A. F.: Cave air controlon dripwater geochemistry, Obir Caves (Austria): Implicationsfor speleothem deposition in dynamically ventilated caves,Geochim. Cosmochim. Ac., 69, 2451–2468, 2005.

Spötl, C. and Mattey, D.: Stable isotope microsampling ofspeleothems for palaeoenvironmental studies: A comparison ofmicrodrill, micromill and laser ablation techniques, Chem. Geol.,235, 48–58, 2006.

Tarutani, T., Clayton, R. N., and Mayeda, T. K.: Effect of poly-morphism and magnesium substitution on oxygen isotope frac-tionnation between calcium carbonate and water, Geochim. Cos-mochim. Ac., 33, 987–996, 1969.

Thornthwaite, C.: An approaxh toward a rational classification ofclimate, Geographical Review, 38, 55–94, 1948.

Treble, P., Shelley, J. M. G., and Chappell, J.: Comparison of highresolution sub-annual records of trace elements in a modern(1911–1992) speleothem with instrumental climate data fromsouthwest Australia, Earth Planet. Sci. Lett., 216, 141–153, 2003.

Tremaine, D. M., Froelich, P. N., and Wang, Y.: Speleothem calcitefarmed in situ: Modern calibration of delta �18O and �13C paleo-climate proxies in a continuously-monitored natural cave system,Geochim. Cosmochim. Ac., 75, 4929–4950, 2011.

Van Rampelbergh, M., Fleitmann, D., Verheyden, S., Cheng, H.,Edwards, L., De Geest, P., De Vleeschouwer, D., Burns, S. J.,Matter, A., Claeys, P., and Keppens, E.: Mid- to late HoloceneIndian Ocean Monsoon variability recorded in four speleothemsfrom Socotra Island, Yemen, Quat. Sci. Rev., 65, 129–142, 2013.

Verheyden, S.: Speleothems as palaeoclimatic archives, Isotopicand geochemical study of the cave environment and its Late Qua-ternary records, Unpubl. PhD Thesis, Vrije Universiteit Brussel,Belgium, 132 p., 2001.

Verheyden, S., Baele, J.-M., Keppens, E., Genty, D., Cattani, O.,Hai, C., Edwards, L., Hucai, Z., Van Strijdonck, M., andQuinif, Y.: The proserpine stalagmite (Han-sur-Lesse cave,Belgium): preliminary environmental interpretation of the last1000 years as recorded in a layered speleothem, Geol. Belg., 9,245–256, 2006.

Verheyden, S., Genty, D., Deflandre, G., Quinif, Y., and Kep-pens, E.: Monitoring climatological, hydrological and geochem-ical parameters in the Pere Noel cave (Belgium): implication forthe interpretation of speleothem isotopic and geochemical time-series, Int. J. Speleol., 37, 221–234, 2008a.

Verheyden, S., Nader, F. H., Cheng, H. J., Edwards, L. R., andSwennen, R.: Paleoclimate reconstruction in the Levant regionfrom the geochemistry of a Holocene stalagmite from the Jeitacave, Lebanon, Quaternary Res., 70, 368–381, 2008b.

Wang, Y., Cheng, H., Edwards, R. L., Kong, X., Shao, X., Chen, S.,Wu, J., Jiang, X., Wang, X., and An, Z.: Millennial- andorbital-scale changes in the East Asian monsoon over the past224 000 years, Nature, 451, 1090–1093, 2008.






101



102

Chapter 5: A 500 year speleothem record from Belgium

103

Chapter 5

A 500-‐year speleothem multiproxy record from the Han-‐sur-‐Lesse cave, Belgium Different studies have indicated that speleothem proxies successfully record climate variations in Western Europe (Frisia et al., 2003; Mangini et al., 2005; Verheyden et al., 2006; Mattey et al., 2008; Baker et al., 2011; Fohlmeister et al., 2012). In Belgium, the Proserpine stalagmite from the Han-‐sur-‐Lesse cave, displays extremely high-‐growth rates (up to 2 mm/y) and seasonal layering allowing to measure different proxies at seasonal scale back to 2000 BP (Verheyden et al., 2006). In this PhD, we focused on the clearly layered upper 500 years of the Proserpine, containing a large part of Little Ice Age (LIA, ±1300 to 1850 AD) (Jones and Mann, 2004) as well as the ongoing anthropogenic climate warming of the last 150 years (Mann et al., 1998, 1999; 2003; 2008). The results of the cave monitoring, discussed in Chapter 4, are used to link the measured variations in the δ18O and δ13C values, layer thickness and calcite color with climate. The observed variations are then compared with historical, instrumental records and other temperature reconstructions to investigate how recognized regional climate variations are recorded in the (local) speleothem proxies. Conclusions are that the δ18O and δ13C signals, layer thickness and calcite fabric of the Proserpine can successfully be used to investigate climate variations over the last 500 years. The recorded variations compare well with known regional climate changes. Moreover, the seasonally resolved proxies provide additional information on the amplitude of the seasonal variations during the LIA. The Proserpine has demonstrated to successfully record climate at seasonal scale for the most recent 500 years. Older sections can most probably be used in further studies to investigate in more detail for example the Medieval Warm Period (±950-‐1250 AD).


104

REFERENCES Baker, A., Wilson, R., Fairchild, I. J., Franke, J., Spoetl, C., Mattey, D., Trouet, V., and Fuller, L.: High resolution delta O-‐18 and delta C-‐13 records from an annually laminated Scottish stalagmite and relationship with last millennium climate, Glob. Planet. Change, 79, 303-‐311, 2011.

Cook, E. R., D'Arrigo, R. D., and Mann, M. E.: A well-‐verified, multiproxy reconstruction of the winter North Atlantic Oscillation index since AD 1400, Journal of Climate, 15, 1754-‐1764, 2002.

Esper, J., Frank, D., Buntgen, U., Verstege, A., and Luterbacher, J.: Long-‐term drought severity variations in Morocco, Geophysical Research Letters, 34, 2007.

Fohlmeister, J., Schroder-‐Ritzrau, A., Scholz, D., Spotl, C., Riechelmann, D. F. C., Mudelsee, M., Wackerbarth, A., Gerdes, A., Riechelmann, S., Immenhauser, A., Richter, D. K., and Mangini, A.: Bunker Cave stalagmites: an archive for central European Holocene climate variability, Climate of the Past, 8, 1751-‐1764, 2012.


Hurrell, J. W.: Decadal trends in the North Atlantic Osscilation-‐ Regional temperatures and precipitation Science, 269, 676-‐679, 1995.

Jones, P. D. and Mann, M. E.: Climate over past millennia, Reviews of Geophysics, 42, 2004.

Luterbacher, J., Xoplaki, E., Dietrich, D., Jones, P. D., Davies, T. D., Portis, D., Gonzalez-‐Rouco, J. F., von Storch, H., Gyalistras, D., Casty, C., and Wanner, H.: Extending North Atlantic Oscillation reconstructions back to 1500, Atmospheric Science Letters, 2, 114-‐124, 2001.

Mangini, A., Spotl, C., and Verdes, P.: Reconstruction of temperature in the Central Alps during the past 2000 yr from a delta O-‐18 stalagmite record, Earth and Planetary Science Letters, 235, 741-‐751, 2005.

Mann, M. E., Bradley, R. S., and Hughes, M. K.: Global-‐scale temperature patterns and climate forcing over the past six centuries, Nature, 392, 779-‐787, 1998.

Mann, M. E., Bradley, R. S., and Hughes, M. K.: Northern hemisphere temperatures during the past millennium: Inferences, uncertainties, and limitations, Geophysical Research Letters, 26, 759-‐762, 1999.

Mann, M. E. and Jones, P. D.: Global surface temperatures over the past two millennia, Geophysical Research Letters, 30, 2003.

Mann, M. E., Zhang, Z., Hughes, M. K., Bradley, R. S., Miller, S. K., Rutherford, S., and Ni, F.: Proxy-‐based reconstructions of hemispheric and global surface


105

temperature variations over the past two millennia, Proceedings of the National Academy of Sciences of the United States of America, 105, 13252-‐13257, 2008.


Olsen, J., Anderson, N. J., and Knudsen, M. F.: Variability of the North Atlantic Oscillation over the past 5,200 years, Nature Geoscience, 5, 808-‐812, 2012.

Proctor, C. J., Baker, A., Barnes, W. L., and Gilmour, R. A.: A thousand year speleothem proxy record of North Atlantic climate from Scotland, Climate Dynamics, 16, 815-‐820, 2000.

Trouet, V., Esper, J., Graham, N. E., Baker, A., Scourse, J. D., and Frank, D. C.: Persistent Positive North Atlantic Oscillation Mode Dominated the Medieval Climate Anomaly, Science, 324, 78-‐80, 2009.


Wanner, H., Bronnimann, S., Casty, C., Gyalistras, D., Luterbacher, J., Schmutz, C., Stephenson, D. B., and Xoplaki, E.: North Atlantic Oscillation -‐ Concepts and studies, Surveys in Geophysics, 22, 321-‐382, 2001.


106


Paper in review in Climate of the Past 107

A 500-‐year seasonally resolved δ18O and δ13C, layer thickness and calcite fabric record from a speleothem deposited in equilibrium of the Han-‐sur-‐Lesse cave, Belgium. Maïté Van Rampelbergh1, Sophie Verheyden1,2, Mohammed Allan3, Yves Quinif4, Hai Cheng5,6, Lawrence Edwards6, Edward Keppens1 and Philippe Claeys1 1Earth System Sciences, Vrije Universiteit Brussel (VUB), Pleinlaan, B-‐1050, Brussels, Belgium 2Royal Belgian Institute of Natural Sciences, Geological Survey, Direction Earth and History of Life, Jennerstraat 13, B-‐1000, Brussels, Belgium 3AGEs, Départment de Géologie, Université de Liège, Allée du 6 Août, B18 Sart-‐Tilman, B-‐4000, Liège, Belgium 4Faculté Polytechnique, Université de Mons, Rue de Houdain 9, B-‐7000, Mons, Belgium 5 Institute of Global Environmental Change, Xi'an Jiaotong University, Xi'an 710049, China 6 Department of Geological Sciences, University of Minnesota, 100 Union Street SE, Minneapolis MN 55455, USA Abstract Speleothem δ18O and δ13C signals have already proven to enable climate reconstructions at high resolution. However, decadal and seasonally resolved speleothem records are still scarce and often difficult to interpret in terms of climate due to the multitude of factors that can affect the proxy signals. In this paper, a fast growing (up to 2 mm/y) seasonally laminated speleothem from the Han-‐sur-‐Lesse cave (Belgium) is analyzed for its δ18O and δ13C values, layer thickness and changes in calcite fabric. The studied part of the speleothem covers the period between 2001 and 1479 AD as indicated by layer counting and confirmed by 20 U/Th-‐ages. Recharge mainly occurs during winter and lesser during spring and fall causing the Proserpine proxies to be seasonally biased. On decadal and multi-‐decadal scale, increased δ18O values in the Proserpine are interpreted to reflect drier (and colder) winters. Higher δ13C signals are interpreted to reflect increased prior calcite precipitation (PCP) and lower soil activity during drier (and colder) winters. Thinner layers and darker calcite relate to slower growth and both occur during drier (and colder) winter periods. Exceptionally dry (and cold) winter periods induce simultaneous large-‐amplitude shifts in the four proxies. Such anomalies occur from 1565 to 1610, at 1730, from 1770 to 1800, from 1810 to 1860 and from 1880 to 1895 and correspond with exceptionally cold periods in historical and instrumental records as well as in European winter temperature reconstructions. More relative climate variations, during which the four measured proxies vary independently and display lower amplitude variations, occur between 1479 and 1565, between 1610 and 1730 and between 1730 and 1770. The winters during the two periods between 1479 and 1565 and between 1730 and 1770 are interpreted as relatively wetter (and warmer) and correspond with warmer periods in historical data and in winter temperature reconstructions in Europe. The winters in the period between 1610 and 1730 are interpreted as relatively


Paper in review in Climate of the Past

108

drier (and cooler) and correspond with generally colder conditions in Europe. Interpretation of the seasonal variations in δ18O and δ13C signals differs from the interpretation on decadal and multi-‐decadal scale. Seasonal δ18O variations reflect cave air temperature variations and suggest a 2.5 °C seasonality in cave air temperature during the two relatively wetter (and warmer) winter periods (1479-‐1565 and 1730-‐1770), which corresponds to the cave air temperature seasonality observed today. Between 1610 and 1730, the δ18O values suggest a 1.5 °C seasonality in cave air temperature suggesting colder summer temperatures during this drier (and cooler) interval. The δ13C seasonality is driven by PCP and suggests generally lower PCP seasonal effects between 1479 and 1810, compared to the PCP seasonal effect observed today. A short interval of increased PCP-‐seasonality occurs between 1600 and 1660, and reflects increased PCP in summer due to decreased winter recharge. 1. Introduction In the studied western European region, high-‐resolution climate records covering the last 500 years are scarce. Most climate information at seasonal or yearly scale is retrieved from historical data such as the price of flour or grapes (Van Engelen et al., 2001; Le Roy Ladurie, 2004) which may induce biases in the climate record. Therefore it is necessary to confront information from different archives, based on different approaches. Speleothems have already often proven to enable climate reconstruction at high-‐resolution in Europe (Genty et al., 2003; Baker et al., 2011; McDermott et al., 2011; Fohlmeister et al., 2012; Verheyden et al., 2014). On millennial and centennial scales, the δ18O and δ13C variations can often be related to a single climate proxy such as temperature or vegetation cover (Spötl and Mangini, 2002; Genty et al., 2003; McDermott, 2005). However, on decadal and seasonal scale, a larger range of factors can influence the δ18O, δ13C, layer thickness or calcite fabric of a speleothem making an interpretation in terms of climate more difficult. To allow reconstructing the climate up to seasonal variation using mid-‐latitude speleothems, a detailed analysis of each used proxy must be compared with a multiproxy approach. Different European records have enabled to reconstruct climate successfully by using this approach (e.g. Frisia et al., 2003; Niggemann et al., 2003; Mangini et al., 2005; Mattey et al., 2008; Fohlmeister et al., 2012). Belgian speleothems have the valuable advantage to often display a clear internal layered structure reflecting seasonal variations (Genty and Quinif, 1996). The link between layer thickness and water excess in Belgian stalagmites for the Late Glacial and Holocene period has clearly been demonstrated by Genty and Quinif (1996). The δ18O and δ13C signals from a speleothem sampled in the Père Noël cave were interpreted as due to variations in cave humidity and drip rate inducing changes in the kinetics of the calcite deposition occurring closer or less close to isotopic equilibrium. More negative δ18O and δ13C values occur during periods of higher cave water recharge, when calcite deposition occurs closer to isotopic equilibrium (Verheyden et al., 2008). In this speleothem, the isotopic (δ18O and δ13C) and geochemical (Mg/Ca and Sr/Ca) proxies vary



similarly and record the climate in terms of wetter and drier phases (Verheyden et al., 2014). The studied Proserpine stalagmite is a large tabular shaped speleothem, growing in the Han-‐sur-‐Lesse cave, which is part of the same cave system as the Père Noël cave. A former study of the stalagmite (Verheyden et al., 2006) revealed deposition from 200 AD to 2001 AD, indicating an exceptionally high average growth rate of ± 1 mm/y. The upper 56 cm, which covers the last 522 years is clearly layered. The similar variability of the δ18O and δ13C signals and the layer thickness was linked to changes in effective precipitation (rainfall minus evapo-‐transpiration). These proxies therefore have the potential to be used to reconstruct climate in terms of wetter and drier phases. In this paper we study this potential more in detail and up to a seasonally resolved timescale. An absolute age model is established by combining layer-‐counting ages with measured U/Th-‐ages. A comparison of variations in layer thickness, calcite fabric, δ18O and δ13C signals in the light of former studies (Genty and Quinif, 1996; Verheyden, 2001; Genty et al., 2003; Mühlinghaus et al., 2007; Wackerbarth et al., 2010; Fohlmeister et al., 2012; Verheyden et al., 2014) and monitoring of the same stalagmite location (Van Rampelbergh et al., 2014) leads to a better understanding of how these proxies are related among them and how they reflect climate variations. Comparing the Proserpine climate signal with winter temperature reconstructions in Europe (Le Roy Ladurie, 2004; Luterbacher et al., 2004; Dobrovolny et al., 2010) further verifies the proposed climate interpretation. 2. Study area The Proserpine stalagmite is sampled in the Salle-‐du-‐Dôme chamber in the Han-‐sur-‐Lesse cave, southern Belgium (Fig. 1). The Han-‐sur-‐Lesse cave is a meander cutting of the Lesse-‐river, which still flows through the cave. The large rooms, the multiple entrances and the presence of the river make it a well-‐ventilated cave. Part of the cave, including the Salle-‐du-‐Dôme, is a show cave since the mid 19th century. The Salle-‐du-‐Dôme, being the largest chamber of the cave system (150 m wide and 60 m high), is located under ca. 40 m of Givetian limestone (Quinif, 1988) with a C3-‐type vegetation covered soil. The Proserpine stalagmite is a 2 m high stalagmite with a large tabular shape (with a horizontal 70 cm by 150 cm to surface) that was actively growing when cored in 2001. A rain of seepage water throughout the year feeds the stalagmite. Such fast growing ‘tam-‐tam’ shaped stalagmites have the property to record climate signals and environmental information at high resolution (Perette, 2000). The mean annual precipitation at the meteorological station of Han-‐sur-‐Lesse is 844 mm/y and the mean annual air temperature averages 10.3°C (Royal Meteorological Institute Belgium, hereafter named RMI) characterizing a warm temperate, fully humid climate with cool summers (Kottek et al., 2006). While the temperature displays a well-‐marked seasonality with cool summers and mild winters, the rainfall is spread all over the entire year. The external seasonality in temperature causes a subdued temperature variation within the Salle-‐du-‐Dôme of 2 to 2.5 °C between summer and winter (Van Rampelbergh et al., 2014). Present-‐day calcite is deposited in isotopic equilibrium with its drip water (Van



110

Rampelbergh et al., 2014). The δ18O signal of freshly formed calcite collected on top of the Proserpine varies seasonally due the changes in cave air temperature. The δ13C signal varies seasonally due to changes in prior calcite precipitation (PCP) intensity, driven by changes in effective precipitation. At a seasonal scale the δ18O and δ13C signals display an opposite behavior with more negative δ18O values in summer, when the δ13C values are less negative (Van Rampelbergh et al., 2014).

Figure 1 a) The Han-‐sur-‐Lesse cave system is located in the southern part of Belgium. The Proserpine stalagmite was sampled in the Salle-‐du-‐Dôme chamber (white square) located 500 m from the cave exit. b) The Proserpine stalagmite with the location of the 2 m long core that was drilled in 2001 at the spot where most of the drip water falls. 3. Methods The Proserpine stalagmite was sampled in January 2001, by drilling a 2 m core in the tabular shaped stalagmite. The precise location was on the side with the highest drip rate but far enough away from the edge to avoid disturbance of the expected horizontal layering of the growth increments (Fig. 1b). The core was cut in half and a slab of 1 cm was cut from the center. The slabs were polished by hand with carbide powder and finished with Al2O3. The upper 56 cm, was further studied and cut in seven parts, numbered I to VII (Fig. 2), to allow easy handling in the laboratory. Layers were counted per part under the Mercantec Micromill microscope and on high-‐resolution scans using Adobe Illustrator. To increase the reliability of the layer counting, layers were counted by different authors, on different days and with different zooms when counted on computer screen. The reported layer amount is given by the average of 10 layer counting rounds per part. The thickness of each layer was measured using the measurement tool of the Merchantec Micromill microscope with an uncertainty of 0.1 μm. Samples for δ18O and δ13C measurements were taken with a drill bit of 0.3 mm diameter mounted on a Merchantec Micromill. Ethanol was used to clean the speleothem surface and drill bit prior to sampling. Between samplings, drill bit and speleothem surface were cleaned with compressed air. Samples were drilled every 0.5 mm in part I and in every layer for the other parts, in total 867 samples. Stable isotope measurements were carried out using a Kiel-‐III-‐device coupled on a Thermo Delta plus XL with analytical uncertainties ≤ 0.12‰ for



δ13C and ≤ 0.16‰ for δ18O. A total of 20 U-‐series age, among which 8 from a former study (Verheyden et al., 2006) were measured at the University of Minnesota (USA), using the procedures for uranium and thorium as described in Edwards et al. (1987) and Cheng et al. (2000; 2009a; 2009b). StalAge (Scholz and Hoffmann, 2011) was used to interpolate the ages between the U/Th-‐age points. The seasonal character of the layering (Verheyden et al., 2006; Van Rampelbergh et al., 2014) in the Proserpine allows using layer counting to establish an age model. The amount of counted layer couplets per part represents the amount of years for that part. The amount of years obtained by layer counting is then compared with the amount of years suggested by the U/Th-‐ages per part. Results of both independent dating methods are combined to provide the final age model. The uncertainties (2σ) on all reported values correspond with a 95% confidence interval and are calculated according to the following relation: 𝑥 − 𝑡!.!",!!! ∙ 𝑠 𝑛 ≤ 𝑥 ≤ 𝑥 + 𝑡!.!",!!! ∙ 𝑠 𝑛

where 𝑥 is the arithmetic mean of the results, n the number of replicates, t the student distribution function and s the standard deviation on the results. If n ≥ 30, t approximates a normal distribution and is roughly equal to 2. 4. Results Layering is present in the studied upper 56 cm of the Proserpine core and is formed by alternating dark more compact and white more porous layers. The seasonal character of the layering in the Proserpine stalagmite, with one dark and one white layer deposited every year is suggested by Verheyden et al. (2006) and further confirmed by monitoring results of the Proserpine growth site (Van Rampelbergh et al., 2014). The Proserpine stalagmite displays a clear sedimentological perturbation between 9 cm and 10 cm (Fig. 2). During this perturbation, calcite deposition is heavily disturbed with straw pieces embedded in the calcite, which might be relics from fires lit on the paleo-‐surface of the stalagmite to illuminate the Salle-‐du-‐Dôme (Verheyden et al., 2006). Four proxies were measured on the Proserpine stalagmite: calcite fabric, layer thickness, δ18O and δ13C values. Layer thickness varies between 0.05 and 1.7 mm/layer (Fig. 3) and dark layers are on average 0.05 mm thinner than white layers. The δ18O values average -‐6.9 ± 0.16 ‰ and the δ13C values average -‐10 ± 0.12 ‰. Four intervals characterized by large amplitude variations of the four measured proxies occur between 7 and 8 cm, between 10.5 and 12.4 cm, between 18 and 20 cm and between 34 and 36 cm (blue lines Fig. 3). Between 7 and 8 cm and between 34 and 36 cm, calcite fabric is dark compact with almost no visible layering. During these two intervals layer thickness decreases to 0.2 mm/layer and the δ18O and δ13C values increase to values around -‐6.0 ± 0.16 ‰ and -‐8.0 ± 0.12 ‰ respectively. Between 10.5 and 12.4 cm, calcite is heavily altered and more matte and whiter compared to the generally more translucent calcite fabric of the Proserpine. The heat of the fires made on the surface of the stalagmite during the perturbation period may have altered the calcite in this part. In this interval, layer thickness decreases to 0.2 mm/layer and the δ18O and δ13C values increase to values around -‐6.0 ± 0.16 ‰ and -‐6.5 ± 0.12 ‰



112

respectively. From 18 to 20 cm, layering is heavily undulating with vertically orientated layers in some parts, which may reflect small basin or rimstone structures. In this interval, layer thickness decreases to 0.4 mm/layer and the δ18O and δ13C values increase sharply to -‐6.2 ± 0.16 ‰ and -‐7.0 ± 0.12 ‰ respectively. With the exception of the four intervals characterized by simultaneous large amplitude variations of the four measured proxies, the time-‐series can be subdivided in two parts. For the part above the perturbation (part I), calcite fabric is generally darker and more compact. The δ18O values average -‐6.6 ± 0.16 ‰ and δ13C values average -‐10 ± 0.12 ‰. Both display a good correlation as indicated by a Spearman’s correlation coefficient of ρ= 0.811 (p= 8.86 x 10-‐44). Layer thickness in part I averages 0.3 mm/layer and displays similar variations as the isotopes with thicker layers corresponding with more negative isotopic values. The parts below the perturbation (parts II to VII) display more negative δ18O values at -‐7.0 ± 0.12 ‰ while the δ13C values vary around the same mean of -‐10 ± 0.12 ‰. A lower Spearman’s correlation coefficient between the δ18O and δ13C signals is calculated for these parts (parts II to VII) (ρ= 0.37, p= 9.54 x 10-‐24). Below the perturbation, layer thickness varies between 0.5 and 1 mm/layer and displays similar variations as the δ18O values. In lower part II and the upper part III (14 -‐ 18.5 cm) and for the most of part V, part VI and VII (38 -‐ 56 cm), the δ18O signal is generally more negative (-‐7.5 ± 0.16 ‰) and the layer thickness increases to 0.8 mm/layer (Fig. 3). In the lower part III and part IV (18.5 and 38 cm), the δ18O values increase to -‐6.6 ± 0.16 ‰ and the layer thickness decreases to 0.5 mm/layer, while no general particular change is observed for the δ13C values. Sampling for the stable isotopes was done layer per layer in the parts II to VII and reflects seasonal variations in the δ18O and δ13C signals. The δ13C seasonality evolves differently from the δ18O seasonality. A larger δ18O-‐seasonality of 0.5 ‰ occurs in the lower part II and upper part III (14 -‐ 18.5 cm) and for the most of part V, part VI and VII (38 -‐ 56 cm), while in lower part III to IV (18.5 -‐ 32 cm), the δ18O seasonality lowers to 0.25 ‰. For δ13C, the overall seasonality averages at 0.7 ‰. An increase in δ13C seasonality to 1.5 ‰ occurs at 32 cm and is followed by a gradual decrease until 27 cm when the seasonality returns to 0.7 ‰.



Figure 2. The upper laminated 56 cm of the Proserpine core with the description of the calcite fabric. The blue bars indicate intervals during which calcite deposition is disturbed or calcite fabric is very dark compact or white matte.

Figure 3. The δ18O and δ13C signals (‰ VPDB) and layer thickness of the Proserpine core plotted against distance from top. Blue bars indicate intervals during which the calcite fabric, δ18O and δ13C signals and layer thickness all display simultaneous large amplitude variations.



114

Eight U/Th-‐ages that were previously published by some of us (Verheyden et al., (2006) are used and numbered 1, 2, 7, 8, 15, 17, 18 and 19, and marked in light grey in Table 1. Twelve new U/Th ages measured in this study are listed in black in Table 1 and correspond well with the previously measured ages. Layer counting ages were carried out per part (i.e. part I to part VII) and are listed in Table 2 (column 5) together with their 2σ uncertainty range. To compare the two independent age methods (layer counting method and U/Th-‐age method), the U/Th-‐age points have to be interpolated to obtain an age for the top and bottom of each part. The interpolation of the measured U/Th-‐ages was carried out using StalAge and top and bottom ages of each part are listed in Table 2 (column 3). The difference between the top and the bottom age of each part provides the number of years of that part (Table 2, column 4). The amount of years per part derived from the U/Th-‐ages display larger 2σ uncertainties for the parts I, II and III (∼ 70) compared to the parts IV to VII where uncertainties are smaller (∼ 30). The amount of years per part derived from the layer counting display 2σ uncertainties of ∼ 7, being smaller than the uncertainties on the U/Th-‐ages. The obtained amount of layers per part correspond for the two methods in the parts I, II, II, V and VII. Note that, the U/Th-‐age method suggests much smaller amount of years (Table 2, columns 4 and 5) in the parts IV and VI.

Table 1. U/Th measurements (University of Minnesota) of the Proserpine stalagmite. All ages are converted to before 2013. Ages number 1, 2, 7, 8, 15, 17, 18 and 19, marked in light grey are the U/Th-‐ages from Verheyden et al., 2006. The growth rates per part derived from the U/Th-‐ages are listed in Table 2, column 6. The growth rates per part derived from the layer counting ages are listed in Table 2, column 7. The growth rates per part based on layer counting increase in two steps: they are low at 0.6 mm/y in part I, higher around 1 mm/y in part II, III and IV, and very high at 2 mm/year in the parts V, VI, and VII. The growth rates per part derived from the U/Th-‐ages display much larger variations between the different parts, with exceptionally high growth rates of 5.6 mm/y for the part IV and of 6.5 mm/y for part VI.



Table 2. Comparison between the layer counting ages and U/Th-‐ages per part together with their growth rates. The interpolated U/Th-‐ages for the top and the bottom of each part were obtained using StalAge. The interpolated U/Th ages and amount of years per part are reported with their 2σ uncertainty range. 5. Discussion 5.1 Speleothem Age model Two independent geochronological methods are used to establish the age model of the Proserpine: StalAge based on 20 U/Th-‐ages and layer counting. Due to the interruption in calcite deposition between 9 and 10 cm, the layer counting ages cannot be used to count the years back from present until 56 cm. To compare the U/Th-‐ages and the layer counting ages, the amount of years must be determined for each part (Table 2, columns 4 and 5). Results show that the layer counting method displays smaller uncertainties. Both independent geochronological methods deliver similar ages with the exception of parts IV and VI, where the U/Th-‐ages suggest a lower number of years. The U/Th-‐ages indicate that Part IV was deposited in 19 ± 33 years while the layer counting indicates a total of 105 ± 7 years (Table 2). The U/Th-‐ages suggest that Part VI was deposited in 13 ± 27 years while the layer counting indicates a total of 42 ± 10 years (Table 2). The number of years obtained by layer counting in the two parts IV and VI is considered more probable compared to the number of years obtained by U/Th ages. Based on in-‐situ monitoring of the Proserpine drip site demonstrating the seasonal character of the layering and the good agreement of the layer counting and the U/Th ages in most of the other parts, the layer counting model is seen as the most accurate to establish the chronology. Furthermore, the U/Th ages give improbable high growth rates (∼ 6 mm/y) for the parts IV and VI (Table 2). Using the layer counting ages, the Proserpine age model is subdivided in two parts: the part above the perturbation and the part below the perturbation. The age of part I above the perturbation can be obtained by simply counting the layers back from 2011. This leads to an age of 1857 ± 6 AD for the end of the perturbation (Fig. 4). Below the perturbation (at 10 cm), the age of the onset of the perturbation has to be estimated in order to restart the layer counting downwards. This is carried out by counting the layers back upward from the U/Th-‐age located closest below the perturbation (=1798 ± 45 AD). By doing this, a total of 12 ± 2 layer-‐couplets are obtained, indicating that the age of the onset of the perturbation is estimated at 1810 ± 45 AD (Fig. 4). The good estimation of



116

this age is confirmed by the fact that StalAge suggests an age of 1810 ± 48 AD for the onset of the perturbation. Furthermore, a 14C-‐date on a straw piece embedded in the perturbed calcite indicates an age interval of 1760 to 1810 (probability of 95.4 %) (Verheyden et al., 2006) also suggesting a similar time window for the perturbation. The age of 1810 ± 45 AD is consequently considered a good estimation of the onset of the perturbation. This age is used to restart layer counting downwards. Since the uncertainties on the layer counting ages are determined per part, the uncertainty on the age model increases per older part according to the propagation of uncertainties on a sum (Table 3). The age obtained for the bottom of the laminated part of the Proserpine stalagmite at 56 cm is 1479 ± 48 AD (Fig. 4).

Figure 4. Age-‐depth model of the Proserpine based on layer counting ages reported with their 2σ uncertainty. The onset of the perturbation is estimated by counting the layers back up from the U/Th-‐age located closest below the perturbation. U/Th-‐ages are plotted in light grey in the age-‐depth graph with their 2σ uncertainty. Location of U/Th samples on the Proserpine core is indicated by the black dots. All ages are reported in years AD.



Table 3. Uncertainties on the counted layers per part below the perturbation (II to VII) together with the uncertainties on the obtained ages (AD) per part using the age of 1810 ± 45 AD as starting point for the age model counting. 5.2 Factors driving decadal and multi-‐decadal changes in the measured proxies Variations in δ18O values of speleothems deposited in equilibrium with their drip water relate mainly to changes in air temperature, rainfall amount and/or source of the rainfall (Fairchild et al., 2006). Rainfall sources often imply δ18O shifts in the order of several ‰ (Fleitmann et al., 2007) while the δ13C values and layer thickness values remain unchanged. The large-‐scale δ18O variations in the Proserpine are in the order of 1 to 2 ‰ and always occur together with large-‐scale δ13C variations of the same order and a decrease in layer thickness indicating that the source effect is most probably not responsible for these δ18O variations. In temperate regions speleothem δ18O values often display a difficult link with surface air temperature due to the inverse effect of temperature on the rainwater δ18O compared to the calcite δ18O. The relation between surface air temperature and rainwater δ18O varies between ∼ 0.1 and 0.3 ‰/1 °C for Central Europe (Schmidt et al., 2007). The temperature dependent fractionation during calcite formation within the cave acts in the opposite direction, and is around -‐0.2 ‰/1 °C for the Proserpine drip site as suggested by monitoring results (Van Rampelbergh et al., 2014). The net effect of air temperature changes on the Proserpine δ18O signal may thus vary between ∼ -‐0.1 and 0.1 ‰/1 °C considering that the temperature dependence of the rainwater of ∼ 0.1 and 0.3 ‰/1 °C is also valid for Belgium. Consequently, the temperature effect most probably only has a minor influence on the decadal and multi-‐decadal variations in the Proserpine δ18O signal. In the studied region, heavier δ18O values have been observed to correspond to drier periods and thus reflecting the amount effect (Verheyden, 2001). Variations in the Proserpine δ18O may thus possibly relate to changes in wetter or drier conditions. If recharge is seasonally biased, the decadal and multi-‐decadal δ18O variations may be caused by variations in air temperature and/or by rainfall amount during a certain season. Hydrological studies of the Han-‐sur-‐Lesse epikarst show that recharge mostly occurs between spring and fall with largest amounts of recharge



118

in winter (Bonniver, 2011). Rainfall δ18O data shows that winter rainfall has a lower isotopic composition compared to the rainfall from other seasons (Van Rampelbergh et al., 2014). During a period of lower winter recharge, less isotopically light (winter) water is added to the epikarst reservoir compared to the heavier spring and fall water and the total δ18O composition of the epikarst water increases, causing increased δ18O values in the speleothem. Periods of increased δ18O values in the Proserpine record may thus be reflecting drier winter periods and vice versa. The relation between lower drip water δ18O and higher winter recharge amounts can be illustrated by drip water monitoring data over several years. Although no such data is available, winter recharge is considered the main factor determining the δ18O values of the Proserpine. More positive δ18O values are interpreted to reflect drier winter periods and vice versa. Furthermore, a good Spearman correlation can be established between lower winter precipitation intensities (DJF) and lower winter temperatures (DJF) measured by the RMI since 1833 (ρ = 0.47 and p = 3.99 x 10-‐11) suggesting that drier winters correspond to colder winters. More negative δ18O values in the Proserpine may thus possibly reflect drier winter conditions that are most probably also colder. A similar interpretation is used for the decadal and centennial δ18O variations measured of a German speleothem with similar yearly temperature and yearly precipitation amounts as the Proserpine growth site (Wackerbarth et al., 2010; Fohlmeister et al., 2012). Since no major vegetation changes (mainly C3-‐vegetation) occurred above the cave for studied period and site, changes in δ13C values might relate to changes in soil activity (Genty et al., 2003; Fohlmeister et al., 2012) and/or Prior Calcite Precipitation (PCP) (Fairchild et al., 2000). Plant-‐CO2 has a lower isotopic signature compared to atmospheric CO2 (δ13C of C3-‐vegetation is between -‐20 and -‐25‰, while in atmospheric CO2 it evolved from –7 ‰ to –8 ‰ during the studied period). A reduced plant-‐CO2 input in the soil due to lower soil activity will increase the δ13C of the soil-‐CO2 reservoir and consequently the dissolved inorganic carbon (DIC) in the epikarst water. During PCP, calcite is deposited from the percolating epikarst water before entering the cave as drip water. This process mostly occurs during drier periods when aerated zones become more important in the epikarst. PCP causes a simultaneous increase in the δ13C and in the Mg/Ca and Sr/Ca composition of the drip water and speleothem calcite (Fairchild et al., 2000). Although no Mg/Ca and Sr/Ca ratios are measured in the Proserpine, which makes it difficult to evaluate the process of PCP, monitoring results have clearly demonstrated that PCP is an important process in the Han-‐sur-‐Lesse epikarst (Van Rampelbergh et al., 2014). Both effects, being soil activity and PCP act in the same direction and both cause the δ13C values to increase during drier periods. Since drier periods in the cave are caused by lower winter recharge periods, increased δ13C values are interpreted to reflect drier and most probably also colder winter periods. Disequilibrium processes due to a stronger pCO2 gradient between the cave air and drip water and/or due to longer drip intervals may cause simultaneously increased δ18O and δ13C values (Mühlinghaus et al., 2009; Scholz et al., 2009; Deininger et al., 2012). Under the present-‐day conditions, pCO2 levels of the cave



air in the Salle-‐du-‐Dôme are low year-‐round and equal the outside air values. pCO2 levels may change over time due to changes in ventilation patterns, which may change over time due to new cave openings. No such new openings that may have affected the Salle-‐du-‐Dôme ventilation occurred in the last 500 years. The effect of changing pCO2 gradient on the drip water δ18O and δ13C values over the studied period is thus unlikely. Longer drip intervals due to decreased drip flow may be possible. However, under the present-‐day conditions, a continuous drip water flow feeds the stalagmite, which inhibits disequilibrium effects related to longer drip interval (Mühlinghaus et al., 2009). The drip discharge consequently needs to be sufficiently decreased, beneath a certain threshold value, to allow disequilibrium processes to be present. Since recharge occurs in winter (Bonniver, 2011), a decreased drip discharge is expected to relate with significantly drier winters, that are also colder. Furthermore, during periods of lower drip discharge, PCP will occur and further increase the δ13C signal. Decreased drip discharge due to significantly drier (and colder) winters will consequently cause increased correlating δ18O and δ13C values with a larger increase in δ13C values compared to the δ18O values, the latter being not affected by PCP. Layer thickness and calcite fabric in the Proserpine are expected to relate to growth rate, with thinner layers and darker calcite formed under slower growth. Growth rate is primarily dependent on two factors; the discharge amount, which is expected to lower during drier (and colder) winter periods and the cave seepage water calcium ion concentration (Genty et al., 2001). The cave seepage water calcium ion concentration depends on mainly two factors. The first factor, being the soil pCO2 is expected to increases during warmer and wetter periods. Higher soil pCO2 increase the amount of CO2 dissolved in the soil water. Water containing higher CO2 amounts more easily dissolves CaCO3, which increases its calcium ion concentration. The second factor determining seepage water calcium ion concentration is the intensity of PCP. PCP mostly occurs during dry periods and decreases the Ca2+ concentration of the drip water due to precipitation of calcite in the epikarst. Cave monitoring results show that PCP is an important process in the Han-‐sur-‐Lesse epikarst that becomes more intense during the drier summer season (Van Rampelbergh et al., 2014). During drier periods, most probably cased by drier (and colder) winter periods, soil activity will decrease and PCP will increase, both causing lower calcium ion concentration of the drip water. A lower calcium ion concentration and a lower drip discharge during drier (and colder) winters will both cause slower growth of the calcite and consequently thinner layers and darker calcite. To conclude, decadal and centennial changes in the proxies (δ18O and δ13C signals, layer thickness and calcite color) reflect changes in drier (and colder) versus wetter (and warmer) winters. Exceptionally dry (and cold) winters shift the drip discharge below a certain threshold value, which causes the proxies to display simultaneous large amplitude shifts. During such exceptionally dry (and cold) winter periods, the δ18O and δ13C values increase, layer thickness will decrease and calcite fabric will become darker and/or disturbed. When the



120

discharge threshold is not reached, calcite is deposited close to equilibrium and the four proxies may vary independently. 5.3 Anomalies in the proxy records Proserpine calcite deposited in equilibrium with its drip water has a δ18O value of -‐6.7 ± 0.16 ‰ and a δ13C value of -‐10 ± 0.12 ‰ (Van Rampelbergh et al., 2014). Four periods where the δ18O and δ13C values abruptly increase away from the present-‐day equilibrium occur in the Proserpine from 1565 to 1610, at 1730, from 1770 to 1800 and from 1880 to 1895 and are interpreted as anomalies in the record (blue bars Fig. 5). During these anomalies layer thickness decreases below 0.2 mm/layer and calcite fabric is disturbed or very dark and compact. As indicated by the detailed analysis of the climatic factors affecting the different used proxies, as soon as a certain threshold value is reached, the four proxies display simultaneous large-‐amplitude changes and reflect exceptionally dry (and cold) winter periods. No calcite was deposited between 1810 and 1860, which strongly suggests that too little water was dripping on the Proserpine during that period. Therefore, this period is also interpreted as an anomaly reflecting exceptionally dry (and cold) winters. A total of five anomalies are suggested by the Proserpine proxies and last between 1565 and 1610, at 1730, between 1770 and 1800, between 1810 and 1860 and between 1880 and 1895 (blue bars Fig. 5). The five anomalies suggesting exceptionally dry (and cold) winter conditions correspond with known cold and/or dry periods in historical and instrumental archives and in winter temperature reconstructions from Europe and Central Europe (Fig. 5):

• Between 1565 and 1610 winter temperatures in Europe (Luterbacher et al., 2004) and Central Europe (Dobrovolny et al., 2010) were low (Fig. 5, f and g). Historical data of France, Belgium and the Netherlands indicate icy cold winters, harsh famines, low numbers of child births and weddings, and the outbreak of the plague with its worst years from 1562 to 1570 (Le Roy Ladurie, 2004). The shift to cold and dry conditions at 1565 AD is interpreted as the onset of the second pulse of the Little Ice Age (LIA, ±1300-‐1850) (Le Roy Ladurie, 2004) and is nicely recorded in the Proserpine proxies as a shift to drier (and colder) winters. Between 1590 and 1600, the Proserpine proxies suggest a shorter wetter (and warmer) interval as indicated by the more negative δ18O and δ13C values and thicker layers (Fig. 5 a, b and c). A similar decade of warmer conditions between 1590 and 1600 is also reported in winter temperature reconstructions from Europe (Luterbacher et al., 2004), Central Europe (Dobrovolny et al., 2010) and from historical archives (Le Roy Ladurie, 2004).

• At 1730, the abrupt shift in the measured proxies suggests a short but exceptionally dry (and cold) winter period. Considering the age uncertainty of ± 45 years for this period (Fig. 5), the dry (and cold) conditions suggested by the Proserpine at 1730 ± 45 AD, most probably relate to the exceptionally cold and dry decade between 1690 and 1700 AD recorded in historical archives (Le Roy Ladurie, 2004) and by



extremely low winter temperatures in Europe (Luterbacher et al., 2004) and Central Europe (Dobrovolny et al., 2010) (Fig. 5, f and g).

• Between 1770 and 1800, the Proserpine proxies suggest a dry (and cold) winter period that corresponds to a known period of colder winters in Europe (Fig. 5, f and g) (Le Roy Ladurie, 2004; Luterbacher et al., 2004; Dobrovolny et al., 2010).

• The exceptionally dry (and cold) winter conditions between 1810 and 1860, as suggested by the Proserpine, correspond nicely with decreased winter temperatures in Europe (Luterbacher et al., 2004) and Central Europe (Dobrovolny et al., 2010) (Fig. 5, f and g). Historical climate data from France, Belgium and the Netherlands indicate that this interval corresponds with the third and last cold pulse of the LIA and is characterized by exceptionally cold winters and warm summers (Le Roy Ladurie, 2004).

• The most recent dry (and cold) period recorded in the Proserpine (1880 and 1895) corresponds with colder winter temperatures and lower winter precipitation amounts as measured by the RMI in Belgium since 1833 (Fig. 5, d and e). The temperature drop is clearly visible in the winter temperature reconstruction from Europe (Luterbacher et al., 2004) (Fig. 5, f). A decrease in precipitation has also been recorded in the England and Wales precipitation record, where this period is known as very dry with peak dry years at 1884, 1887 and 1893 (Nicholas and Glasspoole, 1931).

The exact forcing behind these five dry (and cold) winter periods is still a matter of discussion. The most trivial forcing of the western European climate is the variation in winter North Atlantic Oscillation (NAO) (Trouet et al., 2009). During a negative winter NAO phase, westerlies winds are forced over southern Europe, which may cause drier and colder winter conditions over Belgium. However, the five dry (and cold) winter periods observed in the Proserpine do not always correspond with negative winter NAO phases (Trouet et al., 2009). Other than negative NAO phases, lower solar irradiance combined with the input of volcanic gasses in the atmosphere may also be responsible for decreased temperatures in Europe. Such is probably the case for the cold and dry period between 1810 and 1860 (third pulse of the LIA). In this period, solar insolation decreased during the Dalton Minimum (1790-‐1810, Mann, 2002) and the Tamborra volcano (Indonesia) erupted in 1815. Combination of negative NAO conditions (Luterbacher et al., 2001), the eruption of the Krakatoa volcano (Indonesia) in 1883 and lower sunspot activity (Lassen and Friischristensen, 1995) are most probably responsible for the exceptionally dry (and cold) winter period between 1880 and 1895.



122

Figure 5. The (a) δ18O and (b) δ13C values (‰ VPDB) and (c) layer thickness plotted against (d) the instrumental winter temperature (DJF) and (e) winter precipitation (DJF) record of the Belgian Royal Meteorological Institute (RMI) measured in Brussels since 1833 (f) the winter temperature reconstruction based on multiple proxies in Europe (Luterbacher et al., 2004) and (g) the winter temperature reconstruction derived from documentary and instrumental evidence in Central Europe (Dobrovolny et al., 2010). Five exceptionally dry (and cold) winter periods suggested by the Proserpine are indicated by blue bars and correspond with clear cold periods in instrumental records and winter temperature reconstructions in Europe and Central Europe. Two periods of relatively wetter (and warmer) winters occur from 1479 and 1565 and from 1730 to 1770 and corresponds with known warmer intervals. Between 1610 and 1730 the Proserpine suggests relatively drier (and colder) winter periods, which correspond with colder winter conditions in Europe and Central Europe.



5.4 More relative warmer and wetter and colder and drier periods In contrast to the five anomaly periods where the four proxies suggest exceptionally dry (and cold) winter conditions, the remaining parts of the Proserpine stalagmite display more relative variations. Between 2001 and 1860, above the perturbation, the δ18O and δ13C values display a bulge with most negative values around 1930. Layer thickness follows the same evolution with the thickest layers around 1930 indicating an evolution to wetter (and warmer) winters up to 1930 followed by an evolution to drier (and colder) winters to 2001. This observation in the Proserpine proxies does not correspond with instrumental winter precipitation and temperature data measured by the RMI since 1833 nor with European winter temperature reconstructions (Luterbacher et al., 2004) (Fig. 5). Calcite is darker in this part due to the incorporation of soot from torches used to illuminate the chamber during cave visits (Verheyden et al., 2006). Soot incorporation in the calcite structure may hamper the calcite deposition and overprint lower-‐amplitude climate variations. However, large-‐amplitude variations such as the dry (and cold) winter anomaly between 1880 and 1895 are still visible within this part, indicating that the climate signal is not fully overprinted. The possible effects of soot on δ18O and δ13C values and layer thickness need further investigation to allow deriving low-‐amplitude climate variations in the part above the perturbation. Below the perturbation, and with exception of the anomaly periods, the measured proxy signals can be subdivided in three periods; between 1479 and 1565, between 1610 and 1730 and between 1730 and 1770 (Fig 5 a, b and c). Between 1479 and 1565 and between 1730 and 1770, more negative δ18O values and thicker layers indicate relatively wetter (and warmer) winter conditions. In between the two latter periods (1610-‐1730), the δ18O values become less negative and layers become thinner indicating relatively drier (and cooler) winter conditions. During the three above described periods (1479-‐1565, 1610-‐1730, 1730-‐1770), the δ13C values display no variations indicating no major changes in soil activity or PCP intensity. Only during the relatively drier (and colder) winter period between 1610 and 1730, the δ13C values display a weak gradual increase from 1700 to 1730. The relatively dry (and cool) conditions in the period between 1610 and 1730 may have caused lower soil activity and a gradual increase in prior calcite precipitation, which gradually augment the δ13C signal. The two periods with relatively wetter (and warmer) winters (1479-‐1565 and 1730-‐1770) interrupted by a period with drier (and cooler) winters (1610-‐1730) observed in the Proserpine are also recorded in the winter temperatures reconstructions of Europe (Luterbacher et al., 2004) and Central Europe (Dobrovolny et al., 2010)(Fig. 5) and in historical archives (Le Roy Ladurie, 2004). The relatively drier (and cooler) winter period between 1610 and 1730 corresponds to colder winter conditions in Europe and Central Europe and is referred as the second pulse of the LIA (Le Roy Ladurie, 2004). This relatively cooler interval may relate to the Maunder Minimum, being a period of decreased solar activity between 1640 and 1714. However, lower solar irradiance alone



124

cannot be responsible for the cooler conditions between 1610 and 1730. The exact forcing of this second pulse of the LIA is still a matter of discussions. 5.5 Seasonality in δ18O and δ13C values The δ18O and δ13C values were measure at a seasonal scale between 1479 and 1810 and clearly display seasonal variations (Fig. 6). Interpretation of the δ18O and δ13C variations on a seasonal scale strongly differs from the interpretation of these proxies on decadal and multi-‐decadal scale. Whereas the decadal and multi-‐decadal variations in δ18O and δ13C vary in phase and reflect changes in drier (and colder) versus wetter (and warmer) winters, the seasonal δ18O and δ13C values vary in anti-‐phase. Seasonal δ18O variations are driven by seasonal cave air temperature changes with a temperature dependence of -‐0.2 ‰/1 °C (Van Rampelbergh et al., 2014). Higher cave air temperatures in summer lead to lower δ18O values of the formed calcite. The seasonal variation in δ13C values is driven by the seasonal change in PCP intensity, with stronger PCP in summer leading to increased calcite δ13C values (Van Rampelbergh et al., 2014). The seasonality in δ18O measured during the two wetter (and warmer) winter periods (1479-‐1565 and 1730-‐1770), equals 0.5 ‰, which is similar to the present-‐day conditions (Van Rampelbergh et al., 2014) and corresponds with a 2 to 2.5 °C seasonality in cave air temperature. Between 1610 and 1730, winters are relatively drier (and cooler), and the δ18O seasonality lowers to 0.25 ‰ corresponding with a 1 to 1.5 °C cave air temperature seasonality. Lower summer temperatures during this cold LIA period are most probably responsible for the lower cave air seasonality. The δ13C signal mostly displays a seasonality of 0.7 ‰ being smaller than the 1 ‰ seasonality in δ13C values observed under the present-‐day conditions (Van Rampelbergh et al., 2014). At 1600, the δ13C seasonality increases to 1.5 ‰ and displays a gradual decreasing trend back to 0.7 ‰ at 1660. The increase in δ13C seasonality between 1600 and 1660 also corresponds with an interval where layers are thinner (∼ 0.4 mm/layer) but clearly alternating between dark compact and white porous layers. This suggests well-‐expressed wet winter conditions and dry summer conditions in the cave. The relatively drier (and colder) winter conditions in the period between 1610 and 1730 cause the yearly water recharge (occurring mostly in winter) to be lower compared to the two periods with wetter (and warmer) winters (1479-‐1565 and 1730-‐1770). A lower recharge during winter will consequently lead to drier cave conditions in summer, and increase the effect of PCP. Increased PCP in summer due to lower winter recharge is interpreted to be responsible for the increased δ13C seasonality and the clear layering between 1600 and 1660.



Figure 6. A decrease in δ18O seasonality in the drier (and cooler) period between 1610 and 1730 (blue) indicates lower cave air temperature seasonality than during the wetter (and warmer) periods (1479-‐1565 and 1730-‐1770) (red). Seasonality in the δ13C signal is higher between 1600 and 1660 and indicates more intense PCP during summer (green), which is caused by decreased winter recharge. All values are in ‰ VPDB. 6. Conclusions

1. A multiproxy approach using δ18O and δ13C values, layer thickness and calcite fabric of the Proserpine stalagmite from the Han-‐sur-‐Less cave, Belgium, successfully reconstructs the climate over the last 522 years in terms of drier (and colder) versus wetter (and warmer) winters.

2. Thinner layers and darker calcite correspond to periods with decreased growth rate, driven by lower recharge and stronger PCP effects during drier (and colder) winters. Higher δ18O values are interpreted to reflect drier (and colder) winters, due to the decreased input of winter recharge water with light isotopic composition. Higher δ13C values reflect lower soil activity and increased PCP during drier (and colder) winter periods.

3. Anomalies in the measured proxies occur when discharge drops under a certain threshold value. During these anomalies, the δ18O and δ13C values increase away from isotopic equilibrium, layers become thin and the calcite becomes very dark or disturbed. Such periods occur between 1565 and 1610, around 1730, between 1770 and 1800, between 1810 and 1860 and between 1880 and 1895 and are interpreted as reflecting exceptionally dry (and cold) winter conditions. The exceptionally dry (and cold) periods found in the Proserpine speleothem correspond well with known dry and cold periods in historical, instrumental and/or temperature reconstruction records from Europe.

4. Less exceptional variations occur between 1479 and 1565 and between 1730 and 1770, with more negative δ18O values and thicker layers reflecting two relatively wetter (and warmer) winters. Less negative δ18O values, still reflecting equilibrium conditions, and thinner layers between 1610 and 1730 are interpreted to reflect a period of relatively drier (and cooler) winters. The two relatively wetter (and warmer) winter periods correspond with warmer periods in European winter temperature reconstructions and historical data from Belgium, the Netherland and France. The drier (and cooler) winter period between 1610 and 1730



126

corresponds with relatively colder conditions in winter temperature reconstructions and historical data.

5. Seasonally resolved isotopic signals successfully record seasonal changes in cave air temperature and PCP. The δ18O signals suggest a 2 to 2.5 °C cave air temperature seasonality between 1479 and 1565 and between 1730 and 1770, which is similar to the seasonality in cave air temperature observed today. Between 1610 and 1730, corresponding with a period with drier (and cooler) winters, the seasonality in cave air temperature decreases to 1 to 1.5°C. The δ13C seasonal changes suggest that the seasonality in discharge was lower than observed today with a short interval of increased seasonality between 1600 and 1660 reflecting stronger summer PCP-‐effects due to decreased winter recharge.

7. Acknowledgements We thank the ‘Domaine des Grottes de Han’ for allowing us to sample the stalagmite and to carry out other fieldwork, Mr Van Dierendonck for his interest and support. PC thanks the Hercules Foundation and Research Foundation Flanders (FWO, project G-‐0422-‐10). 8. References Baker, A., Wilson, R., Fairchild, I. J., Franke, J., Spoetl, C., Mattey, D., Trouet, V., and Fuller, L.: High resolution delta O-‐18 and delta C-‐13 records from an annually laminated Scottish stalagmite and relationship with last millennium climate, Glob. Planet. Change, 79, 303-‐311, 2011.

Bonniver, I.: Etude Hyrogeologique et dimmensionnement par modelisation du "systeme-‐tracage" du reseau karstique the Han-‐sur-‐Lesse (Massif de Boine, Belgique), 2011. Geologie, FUNDP Namur, Namur, p 93 to 97 pp., 2011.

Cheng, H., Edwards, R. L., Broecker, W. S., Denton, G. H., Kong, X., Wang, Y., Zhang, R., and Wang, X.: Ice Age Terminations, Science, 326, 248-‐252, 2009a.

Cheng, H., Edwards, R. L., Hoff, J., Gallup, C. D., Richards, D. A., and Asmerom, Y.: The half-‐lives of uranium-‐234 and thorium-‐230, Chemical Geology, 169, 17-‐33, 2000.

Cheng, H., Fleitmann, D., Edwards, R. L., Wang, X. F., Cruz, F. W., Auler, A. S., Mangini, A., Wang, Y. J., Kong, X. G., Burns, S. J., and Matter, A.: Timing and structure of the 8.2 kyr BP event inferred from delta O-‐18 records of stalagmites from China, Oman, and Brazil, Geology, 37, 1007-‐1010, 2009b.


Dobrovolny, P., Moberg, A., Brazdil, R., Pfister, C., Glaser, R., Wilson, R., van Engelen, A., Limanowka, D., Kiss, A., Halickova, M., Mackova, J., Riemann, D., Luterbacher, J., and Boehm, R.: Monthly, seasonal and annual temperature



reconstructions for Central Europe derived from documentary evidence and instrumental records since AD 1500, Climatic Change, 101, 69-‐107, 2010.

Edwards, R. L., Chen, J. H., and Wasserburg, G. J.: 238U-‐234U-‐230Th-‐232Th systematics and the precise measurement of time over the past 500 000 years, Earth and Planetary Science Letters, 81, 175-‐192, 1987.

Fairchild, I. J., Borsato, A., Tooth, A. F., Frisia, S., Hawkesworth, C. J., Huang, Y. M., McDermott, F., and Spiro, B.: Controls on trace element (Sr-‐Mg) compositions of carbonate cave waters: implications for speleothem climatic records, Chemical Geology, 166, 255-‐269, 2000.

Fairchild, I. J., Smith, C. L., Baker, A., Fuller, L., Spötl, C., Mattey, D., McDermott, F., and Eimp: Modification and preservation of environmental signals in speleothems, Earth-‐Science Reviews, 75, 105-‐153, 2006.


Fohlmeister, J., Schroder-‐Ritzrau, A., Scholz, D., Spötl, C., Riechelmann, D. F. C., Mudelsee, M., Wackerbarth, A., Gerdes, A., Riechelmann, S., Immenhauser, A., Richter, D. K., and Mangini, A.: Bunker Cave stalagmites: an archive for central European Holocene climate variability, Climate of the Past, 8, 1751-‐1764, 2012.


Genty, D., Baker, A., and Vokal, B.: Intra-‐ and inter-‐annual growth rate of modern stalagmites, Chemical Geology, 176, 191-‐212, 2001.

Genty, D., Blamart, D., Ouahdi, R., Gilmour, M., Baker, A., Jouzel, J., and Van-‐Exter, S.: Precise dating of Dansgaard-‐Oeschger climate oscillations in western Europe from stalagmite data, Nature, 421, 833-‐837, 2003.

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

Kottek, M., Grieser, J., Beck, C., Rudolf, B., and Rubel, F.: World map of the Koppen-‐Geiger climate classification updated, Meteorologische Zeitschrift, 15, 259-‐263, 2006.

Lassen, K. and Friischristensen, E.: Variability of Solar-‐Cycle length during the past 5 centuries and the apparent association with terrestrial climate Journal of Atmospheric and Terrestrial Physics, 57, 835-‐845, 1995.



128

Le Roy Ladurie, E.: Histoire humaine et comparée du climat, Canucules et Glaciers XIII et XVIII sciècles, 2004.

Luterbacher, J., Dietrich, D., Xoplaki, E., Grosjean, M., and Wanner, H.: European seasonal and annual temperature variability, trends, and extremes since 1500, Science, 303, 1499-‐1503, 2004.

Luterbacher, J., Xoplaki, E., Dietrich, D., Jones, P. D., Davies, T. D., Portis, D., Gonzalez-‐Rouco, J. F., von Storch, H., Gyalistras, D., Casty, C., and Wanner, H.: Extending North Atlantic Oscillation reconstructions back to 1500, Atmospheric Science Letters, 2, 114-‐124, 2001.

Mangini, A., Spötl, C., and Verdes, P.: Reconstruction of temperature in the Central Alps during the past 2000 yr from a delta O-‐18 stalagmite record, Earth and Planetary Science Letters, 235, 741-‐751, 2005.

Mann, M. E.: The Earth system: physical and chamical dimensions of global environmental change. In: Encyclopedia of Global Environmental Change, MacCracken, M. C. and Perry, J. S. (Eds.), John Wiley & Sons, Chichester, 2002.


McDermott, F.: Centennial-‐scale Holocene climate variability revealed by a high-‐resolution speleothem delta O-‐18 record from SW Ireland (9 Nov, pg 1328, 2001), Science, 309, 1816-‐1816, 2005.



Mühlinghaus, C., Scholz, D., and Mangini, A.: Modelling stalagmite growth and delta C-‐13 as a function of drip interval and temperature, Geochimica Et Cosmochimica Acta, 71, 2780-‐2790, 2007.

Nicholas, F. J. and Glasspoole, J.: General monthly rainfall over England and Wales, 1727 to 1931, British Rainfall, 1931. 299-‐306, 1931.

Niggemann, S., Mangini, A., Mudelsee, M., Richter, D. K., and Wurth, G.: Sub-‐Milankovitch climatic cycles in Holocene stalagmites from Sauerland, Germany, Earth and Planetary Science Letters, 216, 539-‐547, 2003.

Perette, Y.: Etude de la structure interne des stalagmites: contribution a la connaisance geographique des evolutions environnementales du Vercors (France), 2000. Geographie, Univ. de Savoie, France, 277 pp., 2000.



Quinif, Y.: Une nouvelle topographie de la Grotte de Han, Lapiaz hors serie "Special Han", 1988. 15-‐18, 1988.

Schmidt, G. A., LeGrande, A. N., and Hoffmann, G.: Water isotope expressions of intrinsic and forced variability in a coupled ocean-‐atmosphere model, Journal of Geophysical Research-‐Atmospheres, 112, 2007.

Scholz, D. and Hoffmann, D. L.: StalAge -‐ An algorithm designed for construction of speleothem age models, Quaternary Geochronology, 6, 369-‐382, 2011.


Spötl, C. and Mangini, A.: Stalagmite from the Austrian Alps reveals Dansgaard-‐Oeschger events during isotope stage 3: Implications for the absolute chronology of Greenland ice cores, Earth and Planetary Science Letters, 203, 507-‐518, 2002.

Trouet, V., Esper, J., Graham, N. E., Baker, A., Scourse, J. D., and Frank, D. C.: Persistent Positive North Atlantic Oscillation Mode Dominated the Medieval Climate Anomaly, Science, 324, 78-‐80, 2009.

Van Engelen, A. F. V., Buisman, J., and Ijnsen, F.: A millennium of weather, winds and water in the Low Countries. In: History and Climate, memories of the future?, Kluwer Academics, 2001. 2001.

Van Rampelbergh, M., Verheyden, S., Allan, M., Quinif, Y., Keppens, E., and Claeys, P.: Seasonal variations recorded in cave monitoring results and a 10-‐year monthly resolved speleothem δ18O and δ13C record from the Han-‐sur-‐Lesse cave, Belgium., Climate of the Past, 10, 1-‐15, 2014.

Verheyden, S.: Speleothems as palaeoclimatic archives. Isotopic and geochemical study of the cave environment and its Late Quaternary records., Unpubl. PhD Thesis, Vrije Universiteit Brussel, Belgium, 132p., 2001.


Verheyden, S., Genty, D., Deflandre, G., Quinif, Y., and Keppens, E.: Monitoring climatological, hydrological and geochemical parameters in the Pere Noel cave (Belgium): implication for the interpretation of speleothem isotopic and geochemical time-‐series, International Journal of Speleology, 37, 221-‐234, 2008.

Verheyden, S., Keppens, E., Quinif, Y., Cheng, H. J., and Edwards, L. R.: Late-‐glacial and Holocene climate reconstruction as inferred from a stalagmite -‐ Grotte du Pere Noel, Han-‐sur-‐Lesse, Belgium, Geologica Belgica, 17, 83-‐89, 2014.



130

Wackerbarth, A., Scholz, D., Fohlmeister, J., and Mangini, A.: Modelling the delta O-‐18 value of cave drip water and speleothem calcite, Earth and Planetary Science Letters, 299, 387-‐397, 2010.

Chapter 6: General Conclusions and Perspectives

131

Chapter 6

General Conclusions and Perspectives In this thesis, the use of speleothems as tools to reconstruct the paleoclimate during the Holocene is evaluated for Socotra (Yemen) and Belgium. The results from this work provided better insights in the evolution of oceanic and atmospheric phenomena such as the Indian Ocean Monsoon (IOM) and the Intertropical Convergence Zone (ITCZ) around the northern Indian Ocean. Results of both study sites clarified how speleothem proxies relate to climate variations in semi-‐arid (Socotra) and temperate (Belgium) climates, with resolutions up to seasonal scale. In Chapter 3 a multi-‐cave, multi-‐speleothem and multi-‐proxy (δ18O, δ13C, Mg/Ca and Sr/Ca) approach was applied to investigate the IOM system in the northern Indian Ocean, which extends the knowledge already established by previous studies in that region (Burns et al., 2003; Gupta et al., 2003; Fleitmann et al., 2007; Lezine et al., 2010). The speleothems collected on the eastern side of Socotra Island, provided the first successful record of northeast winter monsoon precipitation variations since 6 000 a BP. A major conclusion of the study is the different evolution of the northeast winter monsoon compared to the southwest summer monsoon, with no link between the winter monsoon variations and the North Atlantic climate variations. To understand the exact factors driving the northeast winter monsoon, better geographically spread, higher-‐resolution and longer time records are necessary. Of major interest would be to obtain an Eastern Socotran speleothem record that covers the last 15 ka to investigate how larger amplitude climate variations such as the Last Glacial Maximum, Bølling Allerød and Younger Dryas influenced the northeast winter monsoon. A longer record should also contain the Holocene humid period, lasting between ±10 and ±6 ka, and would help clarifying its timing and intensity in the northwestern Indian Ocean (Lezine et al., 2014). High-‐resolution speleothem data from regions, that have not been studied, could provide the necessary links to help understand the IOM winter monsoon forcing (Fig. 1). Speleothem data from such


132

region like Iraq, Iran and Pakistan could indicate the link between IOM variations and Mediterranean climate variations such as observed in Israel (Bar-‐Matthews et al., 1997; 1999; 2000; 2003; 2011) and Lebanon (Verheyden et al., 2008). The eastern African coast, a region that has hardly been studied, could possibly house numerous caves, the study of which could help to connect the variations observed between the northern and southern Indian Ocean.

Figure 1. Wind pattern of the winter monsoon above the Indian Ocean in January. The southern edge of the ITCZ migration path is located around the latitude of Madagascar. The red question marks indicate the regions where currently no speleothem based paleoclimate reconstructions are available. High-‐resolution long-‐term speleothem records from these regions could provide important links between climate variations and highlight their connection (adapted after Van Rampelbergh et al. 2013). Of particular interest is the island of Madagascar that displays two different precipitation regimes in the north and the south of the island separated by a mountain range (similar to the situation on Socotra Island). Speleothem records from both sides of the Island could bring new insights in the evolution of the summer and winter monsoons at the southern limit of the ITCZ. A recent ocean core study from the Mauritian lowlands (east of Madagascar) indicated that Mauritian rainfall and the Indian and Asian summer monsoons are linked since they both receive moisture from the southern equatorial Indian Ocean (de Boer et al., 2014). This study highlights the complex climate system in the Southern Indian Ocean where speleothems could help providing very valuable complementary information. Apart from refining the IOM mechanics, the trace elemental records in the Socotran stalagmites also helped to gain insights in how


133

trace elements are inked to climate in semi-‐arid regions (Rutlidge et al., 2014). Analyzing other trace elements, such as Ba, Na, Fe, Pb, U, Cd, etc. in the Socotran speleothems may further provide insights in how they can be used to reconstruct climate in semi-‐arid regions. In Chapter 4 and 5, the potential of a Belgian fast growing and visibly layered speleothem (Proserpine) from the Han-‐sur-‐Lesse cave in recording the climate at seasonal scale was evaluated. Results from the cave monitoring (Chapter 4) indicated that the cave parameters (cave air and drip water temperature, the water discharge amount, pCO2 and δ13CCO2 values of cave air, drip water pH, δ18O, δD and δ13CDIC values) displayed seasonal variations in response to the external climate. Freshly farmed calcite and detailed measurements in the speleothem confirmed that the seasonal patterns are recorded in the calcite fabric and in its δ18O and δ13C values. A 500-‐year time series, which is seasonally resolved between 1810 and 1479 AD, indicated that the calcite fabric, layer thickness, and δ18O and δ13C values can successfully be used to reconstruct variations in the past climate (Chapter 5). Due to the multitude of factors that can affect the different proxies, a multiproxy approach is necessary to identify the factors influencing the proxy variations in Belgian speleothems. This multiproxy approach revealed that the last part of the LIA was generally drier (and colder) with two relatively wetter (and warmer) periods between 1479 and 1565 and between 1730 and 1770 AD. The seasonally resolved series revealed a decrease in temperature seasonality of 1°C between 1610 and 1730, when conditions were generally drier and colder. The results from the work on the Proserpine indicate that the measured proxies (δ18O, δ13C, layer thickness and calcite fabric) successfully record climate variations up to seasonal scale. In a following phase, the measured proxies can be extended further back in time, up to 2000 AD as is indicated by the bottom U/Th-‐age of the Proserpine core (Verheyden et al., 2006), and cover the Medieval Warm Period (±950-‐1250 AD) and its transition into the Little Ice Age (±1300-‐1850). Studying the MWP and LIA climatic periods up to seasonal level has not been done yet and may throw a light on how the seasonal variations evolved during these two climatic periods. Apart from extending the time series of the measured proxies, other proxies such as elemental concentrations of Mg, Sr, Ca, U, Ba, Pb, Cd, Fe, Zn,… can be measured and used to investigate their link with the environment. Monitoring how these elements are transported from the atmosphere, the soil and the epikarst into the drip water and finally in the calcite will allow understanding which factors cause them to vary over time. These conclusions can then be applied on trace element time series measured in the core and be used as additional proxies to retrieve past climatic and environmental changes.


134

The conclusions and perspective from this work are based on the currently available methods to extract climate information from speleothems. However, all studied samples and cave sites most probably contain more and other information about past climates and environments, which cannot be extracted due to methodological constrains. Higher resolution sampling and better analytical methods and tools may not only allow to increase the time resolution of the known proxies but will also help in developing new proxies. One of the most promising perspectives is the U/Pb-‐dating technique, which will allow the speleothem community to reconstruct climates past the U/Th-‐age limit of 600 ka. Also, further work on the clumped isotope technique or fluid inclusion measurements form an important perspective for deriving absolute temperature variations from speleothems. Finally, the large collection of speleothem samples and records allows comparing them to retrieve patterns and gradients such as has been done by McDermott et al. (2011). The development of a global speleothem database and network should become a priority for speleothem-‐based paleoclimate research. I would like to conclude this thesis with a quote from one of my favorite authors, Roald Dahl, in his book ‘The Minpins’. These words perfectly describe the science of speleothems, as I have experienced it. At first sight, no one would expect the dark caves and their calcite formations to be such great environments to record climate variations. Since the start of the speleothem science, together, we have been able to read these climate archives and understand how the past terrestrial climates evolved. Even if speleothems already taught us a lot, we have to keep watching because they still contain a lot of more secrets, which we are not able to see yet. “And above all, watch with glittering eyes the whole world around you because the greatest secrets are always hidden in the most unlikely places.”

Roald Dahl


135

REFERENCES Bar-‐Matthews, M. and Ayalon, A.: Mid-‐Holocene climate variations revealed by high-‐resolution speleothem records from Soreq Cave, Israel and their correlation with cultural changes, Holocene, 21, 163-‐171, 2011.

Bar-‐Matthews, M., Ayalon, A., Gilmour, M., Matthews, A., and Hawkesworth, C. J.: 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, 3181-‐3199, 2003.

Bar-‐Matthews, M., Ayalon, A., and Kaufman, A.: Late quaternary paleoclimate in the eastern Mediterranean region from stable isotope analysis of speleothems at Soreq Cave, Israel, Quaternary Research, 47, 155-‐168, 1997.

Bar-‐Matthews, M., Ayalon, A., and Kaufman, A.: Timing and hydrological conditions of Sapropel events in the Eastern Mediterranean, as evident from speleothems, Soreq cave, Israel, Chemical Geology, 169, 145-‐156, 2000.

Bar-‐Matthews, M., Ayalon, A., Kaufman, A., and Wasserburg, G. J.: The Eastern Mediterranean paleoclimate as a reflection of regional events: Soreq cave, Israel, Earth and Planetary Science Letters, 166, 85-‐95, 1999.

Burns, S. J., Fleitmann, D., Matter, A., Kramers, J., and Al-‐Subbary, A. A.: Indian Ocean climate and an absolute chronology over Dansgaard/Oeschger events 9 to 13, Science, 301, 1365-‐1367, 2003.

de Boer, E. J., Tjallingii, R., Velez, M. I., Rijsdijk, K. F., Vlug, A., Reichart, G. J., Prendergast, A. L., de Louw, P. G. B., Florens, F. B. V., Baider, C., and Hooghiemstra, H.: Climate variability in the SW Indian Ocean from an 8000-‐yr long multi-‐proxy record in the Mauritian lowlands shows a middle to late Holocene shift from negative IOD-‐state to ENSO-‐state, Quaternary Science Reviews, 86, 175-‐189, 2014.


Gupta, A. K., Anderson, D. M., and Overpeck, J. T.: Abrupt changes in the Asian southwest monsoon during the Holocene and their links to the North Atlantic Ocean, Nature, 421, 354-‐357, 2003.

Lezine, A.-‐M., Bassinot, F., and Peterschmitt, J.-‐Y.: Orbitally-‐induced changes of the Atlantic and Indian monsoons over the past 20,000 years: New insights based on the comparison of continental and marine records, Bulletin De La Societe Geologique De France, 185, 3-‐12, 2014.


136

Lezine, A. M., Robert, C., Cleuziou, S., Inizan, M. L., Braemer, F., Saliege, J. F., Sylvestre, F., Tiercelin, J. J., Crassard, R., Mery, S., Charpentier, V., and Steimer-‐Herbet, T.: Climate change and human occupation in the Southern Arabian lowlands during the last deglaciation and the Holocene, Glob. Planet. Change, 72, 412-‐428, 2010.


Rutlidge, H., Baker, A., Marjo, C. E., Andersen, M. S., Graham, P. W., Cuthbert, M. O., Rau, G. C., Roshan, H., Markowska, M., Mariethoz, G., and Jex, C. N.: Dripwater organic matter and trace element geochemistry in a semi-‐arid karst environment: Implications for speleothem paleoclimatology, Geochimica Et Cosmochimica Acta, 135, 217-‐230, 2014.

Van Rampelbergh, M., Fleitmann, D., Verheyden, S., Cheng, H., Edwards, L., De Geest, P., De Vleeschouwer, D., Burns, S. J., Matter, A., Claeys, P., and Keppens, E.: Mid-‐ to late Holocene Indian Ocean Monsoon variability recorded in four speleothems from Socotra Island, Yemen, Quaternary Science Reviews, 65, 129-‐142, 2013.


Verheyden, S., Nader, F. H., Cheng, H. J., Edwards, L. R., and Swennen, R.: Paleoclimate reconstruction in the Levant region from the geochemistry of a Holocene stalagmite from the Jeita cave, Lebanon, Quaternary Research, 70, 368-‐381, 2008.

Maïté Van Rampelbergh

Documents

Transcript of Maïté Van Rampelbergh