Conference Agenda

Posters #1
Monday, 18/July/2022:
5:00pm - 7:00pm

Location: Foyer & Online

CCB, Innrain 80, 6020 Innsbruck


Daily resolution electrical conductivity of drip water measured in slow-drip sites - implementation in Stara Jametina Cave (Croatia)

Kristina Krklec1, Mirna Švob1, Josip Šušnjar1, Željka Zgorelec1, David Domínguez-Villar2

1University of Zagreb, Faculty of Agriculture, Croatia; 2University of Salamanca, Faculty of Sciences, Spain

Paleoclimate studies for long-term records are often obtained from speleothems fed with very slow drips that have a diffusional flow and consequently a limited amount of water available for monitoring studies. Using a HOBO U24-001 data logger, we have implemented a methodology for measuring electrical conductivity (EC) in drip water samples with volume <2 ml. After applying a filter, we provide daily resolution drip water EC data during the period 2019-2020 in a 60 m deep hall within the cave Stara Jametina (Croatia). EC data was calibrated with data obtained from aggregated samples measured monthly using a Myron L Ultrameter 6P that is calibrated with standard solution prior to every use. Both datasets show same variability, allowing us to correct the continuous dataset and obtain accurate daily values of drip water EC and to record high-frequency events that were not observed during monthly monitoring campaigns.

The cave of Stara Jametina is characterized by Mediterranean climate (Csa) with dry and hot summers. Here, the mean annual temperature recorded during the 2019-2020 period was 14.7 °C, while the mean annual amount of precipitation was 1050 mm. Together with EC measurements, temperature, drip rate, concentration of cave air CO2 and drip water hydrochemistry were monitored. Results show that Ca2+ and dissolved inorganic carbon are mayor chemical species affecting EC of drip water. The outstanding seasonal variability of drip water hydrochemistry is the result of prior calcite precipitation (PCP), a process driven by enhanced cave ventilation. PCP occurs mostly during the “winter season”, when cave air temperature is higher than the external temperature, resulting in low values of cave air CO2 concentration. Furthermore, saturation index in relation to calcite and the Mg/Ca ratio of the drip waters are highly dependent of the ventilation mode of the cave, favouring seasonal variability of calcite precipitation.

The hydrochemistry shows that drip water is mostly controlled by well mixed vadose water, although the high-resolution EC record shows the existence of rare high-frequency events with no clear connection to any of the parameters measured. Those occasional events represent only 1% of the studied period and could be attributed to occasional preferential water flows. Due to their short duration, these events are unlikely to leave a significant fingerprint in the speleothem record. Thus, the studied speleothem is potentially a very good archive of long-term paleoclimate in the region with an outstanding seasonal character on the calcite precipitation that could result in annual laminations.

Acknowledgements: This research is a part of the research project “Inter-comparison of karst denudation measurement methods” (KADEME) (IP-2018-01-7080) and “Young Researchers’ Career Development Project – Training New Doctoral Students” projects (DOK-2018-09-5748 and DOK-2021-02-1788) financed by Croatian Science Foundation.

A potential issue in speleothem-based climate reconstructions: Uncoupled surface and cave temperatures

Yuri Dublyansky1, Olga Kadebskaya2

1Innsbruck University, Austria; 2Mining Institute SB RAS, Perm, Russia

Recent advancements in speleothem research, for example laser-assisted fluid inclusion micro-thermometry, allow reconstruction of the temperature of speleothem growth with accuracy approaching 1C (Krüger et al., 2011). From 1970-ies, the paradigm has been that the temperature in a large cave which lacks natural ventilation corresponds to mean annual surface atmosphere temperature at the cave’s elevation (e.g. Wigley & Brown, 1976; Moore & Sullivan, 1978). It follows that speleothem-based temperatures can be used directly to characterize temperature outside the cave.

It is also understood that positive and negative deviations from this relationship may be caused by air and water advection in the cave. Therefore, constant temperatures in the cave are considered as an important criterium indicating suitability of the given stalagmite for paleoclimate reconstruction. Although many examples of caves with temperatures similar to those recorded on the surface are reported in the literature, we posit that the above-mentioned paradigm may not be universally valid.

In 2016–2021 we performed cave temperature monitoring (no less than 1 year of hourly measurements) in 28 caves in Northern Eurasia (51–62N; 57–117E). When evaluating the data, as a first step, we screened out records in which the temperature variation (1 standard deviation of the mean) over the whole period of observations exceeded 0.05C. The set of screened-out cave locations included small caves (less than 500 m in length), caves with geometry promoting inflows of cold winter air (caves extending downward from entrance; cold trap), and one cave passage hosting a stream prone to snowmelt flooding.

The retained set of “stable” temperatures included 17 caves. Among those, we identified three records from two caves, where stable temperatures were abnormally warm as a result of ascending air flow (warm trap). These records were also screened out. The temperature in the remaining 15 caves is dominated by the heat transferred by conduction. These temperatures turned to be invariably higher, by 2.7±0.7C (), than the temperature at the surface above caves. It follows that paleo temperatures reconstructed from speleothems formed in these caves would not be directly attributable to the surface air temperatures.


Krüger, Y., Marti, D., Staub, R.H., Fleitman, D., Frenz, M. (2011) Liquid–vapour homogenisation of fluid inclusions in stalagmites: Evaluation of a new thermometer for palaeoclimate research. Chem. Geol. 289: 39-47. doi:10.1016/j.chemgeo.2011.07.009

Moore, G.W. & Sullivan, G.N. (1978) Speleology The study of Caves. Zephyrus Press, Teaneck (NY).

Wigley, T.M.L. & Brown, M.C. (1976) The Physics of Caves. In: Ford, T.D. & Cullingford, C.H.D. (eds) The Science of Speleology, Academic Press, London, pp. 329–358.

Characterizing the variation of fluorescent organic matter in drip waters and active stalagmites through a one-year monitoring study in La Vallina cave, NW Spain

Laura Endres1, Céline Jacquin2, Jacqueline Traber2, Saul González-Lemos3, Laura Rodriquez-Rodriquez3, Jakub Sliwinski4, Nikita Kaushal1, Oliver Kost1, Heather M. Stoll1

1ETH Zürich, Switzerland; 2EAWAG, CH-8600 Dübendorf, Switzerland; 3Departamento de Geología, Universidad de Oviedo, Spain; 4School of Earth and Environmental Sciences, University of St. Andrews; Queen’s Terrace, St. Andrews KY16 9TS

Organic matter in drip water and speleothems can originate from different sources such as vegetation and soil above the cave, ancient organic compounds within the bedrock, and from microbial processes within the karst and cave. Thus, determining the organic signal in cave drip waters potentially gives valuable information about overlying vegetation and its changes, but also about the conditions within the epikarst and cave. Furthermore, because organic matter has been shown to alter speleothem fabrics or inhibit precipitation, researching the underlying processes is of critical interest for speleothem paleoclimate applications. Natural organic matter in drip water can be rapidly characterized by measuring excitation-emission matrices (EEMs) with fluorescence spectroscopy. Although not all organic matter fluoresces, it has been shown previously that fluorescence generally correlates with total organic carbon in drip water. Moreover, layers of fluorescent organic matter have been found within several speleothems from mid-latitude caves by confocal laser scanning microscopy (CLSM) and have become a promising source of information to enhance paleo-seasonal reconstructions, improve age models via layer counting and, consequently, constrain rates of past climate changes. It has been suggested that these layers form due to enhanced seasonal rainfall or due to changes in cave conditions. However, the mechanisms leading to the formation of fluorescent banding are still under debate. Here, we present results from a one-year monitoring study of fluorescence properties in drip water, sampled from May 2020 to May 2021 in a quasi-monthly resolution at 3-7 locations within the cave system La Vallina in Northwestern Spain. We have measured EEMs and absorbance spectra and have compared them to drip water geochemistry, fluorescence of active speleothems at the same site and vegetation type above the cave. Our results indicate rather soft intensities up to 0.4 raman units (RU), but rich fluorescent spectra containing several peaks. Across the cave system, we have observed strong gradients of fluorescent properties. Results of PARAFAC analysis suggest that humic-like fluorescence, which likely originates from the overlying soil and vegetation, is rather constant per drip site and not significantly higher during increased (autumn) rainfall. While contrasting prior assumptions, this result is also confirmed by the absence of correlation of humic-like fluorescence with colloid-related trace elements such as Cu and Y in drip water. In contrast, protein-like fluorescence is more likely to appear during periods with less rainfall, which potentially suggests higher microbial activity. Overall, our results indicate that in a cave system like La Vallina, fluorescent layer formation in speleothems is not controlled solely by rainfall amount, but instead by changes in local conditions such as soil redox potential, cave ventilation, crystal fabric type and drip water pH.


François Bourges1, Dominique Genty2, Bruno Lartiges3, Frédéric Perrier4, Frédéric Girault4, Rémi Losno4, Alexandre François5, Johann Leplat5, Stephanie Touron5, Faisl Bousta5, Stéphane Tocino6

11) Géologie Environnement Conseil, 30 rue de la République, F-09200 Saint-Girons, France; 22) Environnements et Paléoenvironnements Océaniques et Continentaux (EPOC), UMR 5805, Université de Bordeaux, Allée Geoffroy Saint-Hilaire, 33615 Pessac Cedex, France; 33) University of Toulouse III (Paul Sabatier), Geosciences Environnement Toulouse (UMR 7 CNRS-UPS-IRD), 14 Av. E. Belin, 31400 Toulouse, France; 44) Institut de Physique du Globe de Paris, Université Paris Diderot, 1 rue Jussieu, F-75005 Paris, France; 55) Laboratoire de Recherches sur les Monuments Historiques, Ministère de la Culture, 29 rue de Paris, F-77420 Champs-sur-Marne, France; 6Site de l’Aven d’Orgnac F-07150 Vallon-Pont d’Arc, France

There are growing concerns about CO2 concentration upward drifts in cave atmospheres as many sites are affected by worrying levels of CO2 concentration for human safety and underground patrimony preservation. In some cases, speleological and touristic activities and some scientific operations have been restricted and could be prevented by further increase in CO2 concentration. Unprecedented situations occurred in southern France at two neighboring sites monitored for CO2 since 1997: In June 2016, CO2 concentrations in the Aven d’Orgnac reached the safety level of 3% Vol., whereas in September 2017, after 5 years of rise, the CO2 topped at 4.4% in the Chauvet-Pont d’Arc cave rear room.

In the Aven d’Orgnac tourist cave, CO2% atmosphere variations are controlled by ventilation regimes. During the hot season, some air is drained from the karst porosity (2.5 to 3 %Vol. CO2 in the visited rooms, and 3 to 5% in remote networks) and fills the cave. During the cold season, ventilation switches to a density driven regime: mixing with outside air falling from the upper entrance dilutes the CO2 concentration (<0.5%).

In the Chauvet cave, permanent confinement has preserved the 360 centuries old rock art. Two stable compartments with contrasted compositions (an average 2.2% CO2 concentration in the main volume, and an average 3.4% CO2 concentration in the rear room) are subject to smooth, 1.6 to 2% in amplitude, annual sinusoidal variations. Recent work showed that high CO2 concentration is associated with low water excess periods and that infiltration in the upper karst control the gas fluxes in the cave volume. However, the long-term upward drift can also be attributed to increasing carbon stock in soil and epikarst due to growing vegetation density in earlier decades.

The remediation in the Orgnac visited sector has consisted in an artificial ventilation circulation that mimics the winter regime. The set point value of 0.5% CO2 for diluted extracted air was optimized with a 3 000 to 5 000 m3/h flux. Until now, the monitoring has revealed that the original thermo-hygrometric conditions are preserved. In Chauvet cave, neither ventilation nor hydrological conditions can possibly be modified. Critical situations could be triggered by long lasting droughts expected in the Mediterranean climate foreseeable evolution. Only the vegetal cover could be a factor of control pending confirmation of the preliminary results.


Alexandre Honiat1, François Bourges2, Bruno Lartiges1, Dominique Genty3, Nicolas Ferrer4, Stéphane Tocine5

1University of Toulouse III (Paul Sabatier), Geosciences Environnement Toulouse (UMR 7 CNRS-UPS-IRD), 14 Av. E. Belin, 31400 Toulouse, France.; 2Géologie Environnement Conseil, 30 rue de la République, F-09200 Saint-Girons, France; 3Environnements et Paléoenvironnements Océaniques et Continentaux (EPOC), UMR 5805, Université de Bordeaux, Allée Geoffroy Saint-Hilaire, 33615 Pessac Cedex, France; 4Grotte de Gargas, Nestploria, RD 261, 81, route de Gargas, 65660 Aventignan, France; 5Site de l’Aven d’Orgnac F-07150 Vallon-Pont d’Arc, France

Most walls of karstic cavities are coated with a thin film of water involved in the thermo-hydric and gaseous exchanges between rock walls and cave atmosphere. In painted caves, thickness and stability of the water film are critical to the conservation conditions of paintings. Indeed, the water film can either run off, be involved in condensation/corrosion processes, or induce concretionning by degassing or evaporation. The thickness measurement of the water film on the cave wall and its temporal evolution are essential parameters for the conservation of rock art heritage that have not been measured until now. They are also key parameters for the modelling of cave concretions growth. Those parameters are sensitive to the internal areological conditions of the cavity and could be impacted by global warming.

A contactless optical sensor, commercialized by Micro-epsilon with resolution levels of 1 μm and a measuring range of 0 to 300 μm, was selected as the most suitable measuring device for water film thickness. When the light is reflected by a surface, a single wavelength is focused at the object interface (two wavelengths for thickness measurement), and is then selected by the confocal configuration. The system is passive (no moving part) and does not generate any significant heat as light is transported from the generator through an optical fiber. The measuring device can be powered by 12V rechargeable batteries, and can then be used in caves with no access to the electrical network.

Preliminary laboratory tests performed on natural limestone surfaces, gave a range of measurable thicknesses from 25 to 220 μm (thickness sensor range begin at 15 μm; uncertainties of <±0.3 μm). Water film thickness measurements were then conducted in various french caves (Orgnac, Beideilhac, Gargas, Niaux...). The in-situ thickness measurements on vertical cave walls give a common value of 30-50 µm associated with a good stability in time. Visual contrasts on painted walls are consistent with the presence/absence of measurable thin water films (Gargas cave). Thicker films 80-100 µm are measured on active speleothems. In the latter case, we were able to observe short time dynamic variations on the film thickness due to water flows on walls and calcite.


Ana Moreno1, Miguel Bartolomé1, Carlos Sancho2, Ánchel Belmonte3, Antonio Delgado-Huertas4, Isabel Cacho5, Christoph Spötl6, Heather Stoll7

1Pyrenean Institute of Ecology, CSIC, Spain; 2University of Zaragoza, Spain; 3Geoparque del Sobrarbe, UNESCO, Spain; 4Earth Science Institute of Andalusia, CSIC, Spain; 5University of Barcelona, Spain; 6Innsbruck University, Austria; 7ETH-Zurich, Switzerland

Unravelling the paleoclimate signal stored in speleothems requires a thorough understanding of the modern processes that drive the transfer of the isotopic signal from the rainfall to present-day carbonate precipitates in a cave, including all possible modifications in the soil and the epikarst. A long-term monitoring survey has been carried out in Seso cave since 2011, a small cave in the Central Pyrenees with just 2-3 m of overlying rock thickness and a very poorly developed soil with patches of vegetation, which is slightly better developed above the inner sector of the main cave gallery. The main goal of this study is to determine the influence of the vegetation on the timing and patterns of the isotopic transfer from rainfall to present-day carbonates precipitating in the cave comparing both inner and outer sectors of this cave.

We characterized rainfall isotopic variability (n= 380 events) and assessed the principal influencing factors on daily and monthly scales, signaling the important role played by the type of rainfall (convective shower vs continuous stratiform precipitation) and the trajectory path (Moreno et al., 2021). Dripwaters monitored (temperature, drip rate, hydrochemistry and isotopes) at 12 sites in the cave show a rapid response in terms of temperature variability, slightly delayed in the innermost sites as a consequence of the attenuation by the vegetation cover. Similarly, we observe an immediate response of drip rate to precipitation amount outside the cave and a clear seasonal cycle in dripwater18O in all dripwater sites. Interestingly, 18O of dripwater in the inner sector shows an attenuated variation, which is especially evident from years 2018 to 2020 when the isotopic signal does not follow the seasonality, indicating the effect of soil and vegetation. Carbonate precipitation occurs throughout the year, and the amount of carbonate correlates with rainfall on a seasonal scale. Isotope data show that the seasonal cycle is smaller in the carbonates than in the drip and soil water but still significant, specially in the outer sector of the cave where the amplitude of 13C and 18O in carbonates is much larger (7‰ and 3.8‰, respectively) than in the inner sector.

Calculations show that about 50% of the isotopic variability of modern carbonate precipitates can be explained by air temperature and amount of rainfall. An ongoing study of speleothems from Seso cave suggests a significant correlation between 18O of calcite and regional air temperature for the instrumental period, thus opening the door to calibrate the temperature signal in the speleothems from Seso cave using monitoring results and recent speleothems.

Multi-Pool Monitoring of Organic and Inorganic Carbon at Milandre Cave, Switzerland – Implications for Future Paleoecosystem Proxies.

Sarah Rowan1, Marc Luetscher2, Sönke Szidat1, Thomas Laemmel1, Oliver Kost3, Franziska Lechleitner1

1Department of Chemistry, Biochemistry and Pharmaceutical Sciences & Oeschger Centre for Climate Change Research, University of Bern, Freiestrasse 3, 3012 Bern, Switzerland; 2Swiss Institute for Speleology and Karst Studies. Rue de la Serre 68, 2300 La Chaux-de-Fonds, Switzerland; 3Department of Earth Sciences, ETH Zürich, Sonneggstrasse 5, 8092 Zürich, Switzerland

Terrestrial ecosystems are important reservoirs of carbon (IPCC, 2000). However, the complex response of terrestrial ecosystems to climate change remains poorly constrained. Over the past 20 years, interest in the organic matter (OM) fraction of speleothems, typically comprising 0.01-0.3% of the total carbon (Blyth et al., 2016), has increased due to its potential to offer information about past ecosystems. The provenances of speleothem OM are not well understood, though are speculated to be dominated by contributions from overlying vegetation and soil whereby dissolved OM is transported into the karst system. Other potential sources include microbial activity within the karst system, cave fauna, and fossil carbon sourced from the carbonate host rock (Blyth et al., 2016). The isotopic characterisation (𝛿13C and 14C) of stalagmite OM has the potential to give information about past ecological and climatic state of the surrounding region (Blyth et al., 2016).

Here we present the first results of a high-resolution process study of the organic and inorganic carbon fluxes in Milandre cave (Switzerland), whereby the main carbon source reservoirs will be monitored for two years. Our preliminary data includes 1) cave drip water dissolved inorganic carbon (DIC) 𝛿13C, collected before significant degassing could take place, and 2) atmospheric, cave, soil, and well CO2 𝛿13C and 14C. We find that DIC 𝛿13C is isotopically light, implying that the cave is an open system with a substantial contribution of biologically respired CO2 feeding carbonate growth. The cave gas samples are more depleted in 14C than soil and well gas samples. This suggests either an additional fossil reservoir of CO2 contributing to the cave air or substantial influence from degassing of carbonate-derived CO2 from drips in the cave. Ultimately, this information will be used to constrain the source of speleothem OC and allow the assessment of its suitability as a proxy for ecosystem change.

Blyth, A., Hartland, A. and Baker, A., 2016. Organic proxies in speleothems – New developments, advantages and limitations. Quaternary Science Reviews, 149, pp.1-17.

Blyth, A., Smith, C., and Drysdale, R., 2013. A new perspective on the 13C signal preserved in speleothems using LC-IRMS analysis of bulk organic matter and compound specific stable isotope analysis. Quaternary Science Reviews, 75, pp. 143-149.

IPCC, 2000: Land Use, Land-Use Change and Forestry. (R.Watson, I.Noble, B Bolin, N.H. Ravindranath,D.J. Verardo and D.J. Dokken (eds.)). Cambridge University Press, UK, pp.375.

Stalactite drip rate variations in French Caves : causes and interest for karstic reservoir knowledge

Dominique Genty1, Bourges François2, Jossoud Olivier3, Labhun Inga3, Zhang Jian1, Devaux Ludovic1, Bujan Stephane1, Perrier Frédéric4, Girault Frédéric4, Lartiges Bruno5, Bonnet Stéphane5, Regard Vincent5, Minster Bénédicte3, Regnier Edouard3

1Environnement et Paléoenvironnements Océaniques et Continentaux (EPOC), UMR 5805, Université de Bordeaux, CNRS, France; 2Géologie Environnement Conseil, 30 rue de la République, F-09200 Saint-Girons, France; 3Laboratoire des Sciences du Climat et de l’Environnement (LSCE), LSCE/IPSL, CEA-CNRS-UVSQ, Université Paris-Saclay; 4Institut de Physique du Globe de Paris, Université de Paris, 1 rue Jussieu, F-75005 Paris, France; 5Université de Toulouse III Paul Sabatier, Géosciences Environnement-Toulouse, 14 av. Edouard Belin, F-31400 Toulouse, France

Stalactite dripping time series, originally set up in order to better understand the stalagmite growth environment in relationships with paleoclimatologic studies, appear to be a source of valuable information on infiltration and recharge processes in micro-fissured karstic terrains. The synthesis of several years long monitoring (drip number/15 min) from 25 drippings stations in ten different caves have been compared with local meteorological data, revealing common causes of dripping pacing at different time scale: rainfall (RR) and evapotranspiration (ETP) constrain seasonal and monthly-weekly drip rate variations while atmospheric air pressure controls daily-hours scale drip variations. More precisely, we observed the following :

1) RR minus ETP baseline displays a well-marked seasonality that is the precursor of the dripping yearly seasonality;

2) water excess (or efficient rainfall) explains monthly-weekly drip rate variations during humid period with more or less delay of the order of 1-2 days;

3) large atmospheric air pressure changes linked to atmospheric circulation controls daily scale drip variations;

4) small atmospheric (barometric) tides with 8H, 12H/S2, 24H/S1 periodicities, are the causes of short terms drip changes which sensitivity is controlled by the karst saturation. Mechanism of the influence of air pressure wiggles on the dripping rate is put in relation with the humid/dry season of each site, leading to a conceptual physical model. The barometric effect, quantified in several dripping sites, is compared with the physical characters and to the geology of the sites.

Near Chauvet and Aven d’Orgnac sites (Ardèche, S-France), an associated borehole monitoring demonstrates the close but complex relationship between piezometric level, stalactite drip rates and water excess, giving detailed evidence of interconnections at various spatial scales in karst aquifer. Finally, we observe and describe, for the first time, the strange graphic patterns that appear when comparing two or three synchronized dripping time series each other in the same cave. These enigmatic figures, which suggest non-linear couplings with bifurcation dynamics, may be related to reservoir interconnections or disconnections above the stations opening a potential window to the infiltration processes and micro-fissure geometry.

The Impacts of Fire on Cave Climate

Alexandra Repetto1, Andy Baker1, KSS Cave Studies Team2

1UNSW, Australia; 2Kempsey Speleological Society

Cave air temperature is critical in controlling the climate within caves. Cave climate influences subterranean biodiversity, surface cave calibrations for paleoclimate research, and the passive storage of greenhouse gases. While there is prior research on seasonal heat transfer mechanisms into caves, this research investigated how these processes were altered during a fire event.

Daylight Cave is a short, shallow cave, located in the Macleay region of NSW, Australia. The subtropical regional climate experiences a seasonal summer rainfall bias, with an annual average precipitation of 1200mm. Thermal and hydrological responses of Daylight Cave were monitored between July 2016 to December 2021 to determine its cave climate characteristics, and the climatic response to a fire event in October 2019. Daylight Cave was observed to have three clear seasons:

Minimal Advection (spring /early summer) – Dominated by conduction as temperatures gradually increase. Ventilation occurs only at night. The impact and duration of ventilation decreases throughout the season as surface temperatures increase. There is no advection by water due to limited recharge events.

Hydrological Warming (late summer / early-autumn) – Characterised by heat advection by water. Very limited ventilation. Hydrological warming is controlled by large thermal fluxes when water enters the cave. Elevated cave temperatures from recharge events last for multiple days while the cave is hydrologically active, and water remains in the cave.

Ventilation (late-autumn / winter) - Characterised by frequent and regular ventilation, the ventilation season starts due to an elevated cave temperature from previous hydrological activity as surface temperatures start to decrease in mid-autumn.

With respect to the October 2019 fire, Daylight Cave was too deep to measure a conductive fire signal. It was determined that the fire initiated a one-day response in suppressed ventilation and caused an ongoing 0.4 °C to 0.7 °C increase in cave air temperature. The long-term effects can be attributed to the decrease in vegetation, which increased direct solar insolation heating the soil above the cave. During the first significant precipitation events, the warmer surface conducted heat energy to the resultant recharge water, which advected additional heat into the cave compared to before the fire. The warmer cave temperatures changed the timing of seasonal cave ventilation, with ventilation occurring earlier in winter, and a longer ventilation period in the year following. Cave temperatures remained elevated relative to pre-fire temperatures for the duration of post-fire monitoring. Understanding these long-term feedback mechanisms has potential to assist with paleoclimate reconstructions, conservation of subterranean biodiversity, and understanding the function of cave systems in the greater context of earth’s climate.

The influence of hydrogeological conditions and carbon sources on speleothem δ13C based on decade monitoring data in Furong Cave, southwest China

Jun-Yun Li1,2,3, Ting-Yong Li4, Hong-Chun Li5, Christoph Spötl3, Chao-Jun Chen1, Jian Zhang6, Yao Wu7, Min Zhao8

1Chongqing Key Laboratory of Karst Environment, School of Geographical Sciences, Southwest University, Chongqing 400715, China; 2Key Laboratory of Karst Dynamics, MNR & Guangxi, Institute of Karst Geology, CAGS, Guilin 541004, China; 3Institute of Geology, University of Innsbruck, 6020 Innsbruck, Austria; 4Yunnan Key laboratory of Plateau Geographical Processes & Environmental Changes, Faculty of Geography, Yunnan Normal University, Kunming 650500, China; 5Department of Geosciences, National Taiwan University, Taipei 10617, Taiwan, ROC; 6Environnements et Paléoenvironments Océaniques et Continentaux (EPOC), UMR CNRS, 5805, Université de Bordeaux, Pessac, France; 7Institute of Environmental Physics, Heidelberg University , Heidelberg, 69120, Germany; 8State Key Laboratory of Environmental Geochemistry, Institute of Geochemistry, CAS, Guiyang 550081, China

This study analyzes the influences of δ13C in active speleothem (abbreviation: AS) from cave environment, dripping rate, growth rate and the carbon sources basing on the monitoring data from six monitoring sites (MP1-MP5 and MP9) in Furong Cave, Chongqing, southwest China, between 2009 to 2019.

The results show that 1) the temperature and relatively humidity of MP1-MP3 and MP9 which are located in “Great Hall” with a large space were relatively stable. However, those of MP4 and MP5 located in “Galax Passage” were higher and more outliers attributing to the smaller space and the influence from tourists. 2) Cave pCO2 showed characteristics of significant seasonal variation with higher in summer-autumn and lower in winter-spring, which resulted in the average values of δ13CAS of MP1-MP3 and MP9 being heavier in winter-spring than in summer-autumn. But there was no seasonal variation features of average δ13CAS values of MP4 and MP5. 3) The variation trends of δ13CDIC values of six dripping water were similar, which was lighter in summer-autumn and heavier in winter-spring because of the seasonal characters of cave pCO2. 4) The growth rates of active speleothem of MP4 decreased and the δ13CAS values of MP4 became heavier with decreasing drip rate. The δ13CAS values of others remained within a certain range, and neither the deposition rate of active speleothem nor the δ13CAS values were affected by the drip rate. 5) The 14C values of dripping water and active speleothem of MP2 and MP9 were higher than that of MP3. It is explained that amount of groundwater was supplied to MP2 and MP9 to keep a closed system in the bedrock, which inhibited carbon isotope exchange between the karst water dissolved inorganic carbon and soil CO2, and leaded to a great contribution of dead carbon from the bedrock. Ultimately, both δ13CDIC and δ13CAS of MP2 and MP9 with faster drip rate and growth rate were more positive than those of MP3 with slower drip rate and growth rate. 6) The average 14C value of active speleothem of MP4 and MP5 was 292.58±38.60‰ and 254.30±48.08‰, respectively, which was more than the mean 14C value of MP9 (185.54±46.60‰). However, both of the average δ13CAS value of MP4 and MP5 (-10.48±1.29‰ and -10.72±0.44‰, respectively) were lower than that of MP9 (-9.40±0.55‰). It is interpreted that the “dead carbon” of MP4 and MP5 is mainly provided by CO2 with lighter δ13C released from decomposition aged soil organic matter, while the dead carbon provided by the bedrock is relatively less. The “dead carbon” of MP9 may mainly come from the bedrock.

This study demonstrates that the radiocarbon isotope (14C) of dripping water and active speleothem is a useful indicator for further understanding the difference carbon sources in the stalagmites and the effect of the composition on δ13C. It can also be used to infer the relationship between the hydrogeological conditions overlying the cave and the changes of the surface environment, and provide more basic data for reconstructing the paleoclimate and paleoenvironment using δ13C of the stalagmite.

Variability of trace-elements and δ18O in drip water from Gruta Rey Marcos, Northern Guatemala

Patricia Piacsek1, Juan Pablo Bernal1, Mario Cu-Xi2, Fernando Revolorio2, R. López Martinez3, Patricia Beddows4, Matthew. S. Lachniet5, Giuseppe Lucia5, Korynthia López-Aguiar6, Osmin Vazquez2

1Centro de Geociencias, Universidad Nacional Autónoma de México, Campus UNAM Juriquilla, Querértaro, Mexico; 2Carrera de Geología, Centro Universitario del Norte, Universidad de San Carlos Guatemala, Guatemala; 3Instituto de Geología, Universidad Nacional Autónoma de México, Ciudad Universitaria, Mexico City, Mexico; 4Department of Earth & Planetary Sciences, Northwestern University, USA; 5Department of Geoscience, University of Nevada, Las Vegas, USA; 6Facultad de Química, Universidad Nacional Autónoma de México, Unidad Académica de Ciencias y Tecnología, Parque Científico y Tecnológico de Yucatán Tablaje Catastral, Mexico

Cave monitoring is one of the most important protocols to identify and characterize the environmental and geochemical processes occurring in the epikarst. The Epikarst can modulate the isotopic and elemental composition of drip-waters and neo-formed calcite, providing a geochemical context for the observed isotopic and elemental variability in stalagmites. Whilst most paleohydroclimatic reconstructions are based on the variability of carbonate δ18O (Lachniet, 2009), there is increasing interest in the use of other geochemical proxies, namely trace elements, to extract further environmental information from stalagmites. However, trace-element (TE) systematics in karstic aquifers can be significantly more complex than those for δ18O, due to a myriad of local geochemical and environmental factors that impinge a degree of complexity not observed for δ18O and hinders the wide-spread use of TE variability in stalagmites as a proxy for different environmental processes.

The understanding of the geochemical and environmental processes that might control the abundance or variability of TE in drip-water, hence stalagmites, has improved significantly during the last decades (Borsato et al., 2007; Fairchild and Treble, 2009; Hartland et al., 2012; Huang et al., 2001; Huang and Fairchild, 2001; Wassenburg et al., 2020; Wassenburg et al., 2016). However, the general recognition of epikarstic complexity, and its influence on the drip-water composition, requires careful monitoring of its seasonal and annual variability for each location, with a special focus on drip sites feeding stalagmites with potential for collection (Treble et al., 2015). The prolonged monitoring allows to identify the seasonal signals, characterize the drip-water flow and carbonate precipitation regimes, identify the effects of atmospheric and environmental conditions onto drip-water chemistry, and the chemical and isotopic composition of the calcite or aragonite precipitated. Consequently, based upon the key assumption that such processes occurred during the period of stalagmite growth, cave and drip water monitoring enhances the robustness of the corresponding paleohydrological records by providing information on the processes that modulate TE and δ18O variability in the corresponding stalagmite.

Here, we report two-year monitoring of drip-sites at Gruta Rey Marcos, in Coban, Alta Verapaz Guatemala, a touristic cave hosted in Cretaceous limestone from the Cobán formation. The area is key to understanding the development of the Meso-American and North American monsoon systems (Winter et al., 2020), as well as the trade wind dynamics as a consequence of the Intertropical Convergence Zone (ITCZ) displacements (Desplazes et al., 2013). Additionally, robust paleohydrological records from the area can provide crucial climatological context to the rich cultural history of the area. The long-term monitoring generates information that subsidies the identification of the main processes modulating the variability of trace elements and δ18O in stalagmites in two drip-sites under different dripping regimes. The results shed light on the interpretation of records of TE variability in the corresponding stalagmites.

A comparative study of the micrometeorology in two sag-type ice caves in the Austrian Alps

Maria Wind1, Friedrich Obleitner1, Christoph Spötl2

1Department of Atmospheric and Cryospheric Sciences, University of Innsbruck, Austria; 2Institute of Geology, University of Innsbruck, Austria

Assessing and understanding the micrometeorological conditions inside ice caves and their coupling to the outside environment is integral for a robust interpretation of the perennial ice deposits stored inside these caves. To gain better knowledge of the cave’s thermal and dynamic characteristics, extensive meteorological measurements were done in two sag-type ice caves in the Austrian Alps between 2019 and 2021: Hundsalm ice cave and Guffert ice shaft. While Hundsalm ice cave is operated as a show cave and hence partly manipulated by human activity, Guffert ice shaft is a purely natural system with no previous meteorological records.

This work aims to give an overview of the micrometeorological characteristics of these two caves, therefore extending the knowledge on Hundsalm ice cave beyond temperature as well as presenting the first-ever measurements of Guffert ice cave. Simultaneous measurements inside the caves as well as outside near the cave entrances enable studying the interaction of the external and cave atmosphere.

Besides thermodynamic variables (temperature, humidity and pressure) we also give an overview on the air flow conditions with a special focus on cold air intrusions as they cause stronger dynamics in the cave compared to its stable summer conditions. The data furthermore allows us to look at the different components of the energy budget at the cave ice surface and related ice developments which, at both caves, show a continuous decrease within the observation period.

Parallel measurements at the two cave sites from November 2019 to July 2020 enable us to do an intercomparison, discussing similarities and differences between the two ice caves. Since they are located in more or less the same atmospheric regime, but with different specific geometries, these measurements further allow addressing the impact of the individual cave morphology on the microclimatic conditions in the caves and their response to the outside atmosphere. With a much wider entrance, Guffert ice cave shows a more direct reaction to the outside forcing compared to Hundsalm ice cave.


Magdalena Jelonek1, Michał Gradziński1, Ján Zelinka2, Jacek Madeja3

1Institute of Geological Sciences, Jagiellonian University, Gronostajowa 3A, 30-387 Kraków, Poland; 2Slovak Caves Administration, Hodžova 11, 031 01 Liptovský Mikuláš, Slovakia; 3Institute of Botany, Jagiellonian University, Gronostajowa 3, 30-387 Kraków, Poland

Dobšiná Ice Cave (in Slovak: Dobšinská ľadová jaskyňa) is one of the most significant ice caves in the world. It was formed in the Middle Triassic Steinalm and Wetterstein limestones as a result of the underground flow of the palaeo-stream Hnilec. The cave is located in the southern part of the Slovak Paradise National Park, in the Spiš-Gemer Karst and is a part of the Stratena Cave system. The collapse of the ceiling in the Quaternary (Duča Collapse) caused the separation of these caves. It had a significant impact on the circulation conditions and resulted in the ice formation in Dobšiná Ice Cave (Bella & Zelinka, 2018). The volume of cave ice is estimated at over 110,000 m3. The ice is a source of valuable information about the past, because it records the environmental and climate changes of the late Holocene. This record is unique and changes over time because the ice mass increases from above, while from below it slowly melts.

The research is aimed at interpreting the age of ice, ice mass increase and reconstruction of vegetation based on the analysis of pollen grains preserved in individual layers of cave ice. An 11 meters ice section was selected for the research. The ice profile is characterized by clearly marked layers. Ten samples of bats remains were collected from the ice profile. They were dated by 14C method (AMS technique). The data showed that the oldest organic remains was 1385 ± 30 years BP (605–676 AD). It confirms earlier interpretations that ice in the cave has been forming at the turn of the Dark Ages Period and the Medieval Warm Period (Gradziński et al., 2016). Properly prepared material for pollen analysis was analyzed quantitatively and qualitatively. These microscopic studies have confirmed the presence of pollen grains in cave ice. There have been identified pollen grains of many trees and shrubs, such as Alnus, Betula, Pinus, Quercus, Salix, Abies, Corylus and also herbaceous plants such as Artemisia, Plantago, Poaceae, Apiaceae, Asteraceae, Cichorioideae. The diversity of pollen grains in individual layers was observed, which may indicate a change in environmental conditions during the formation of the ice massif. Fossil pollen grains from the ice profile were compared to the modern pollen spectrum from samples taken at several places in the cave.

It can be concluded that pollen analysis of ice layers reflect the vegetation of the region at that time and thus may prove to be a useful tool for reconstructing climate change in the past millennium.

Scientific research was financed by the National Science Center (Poland) as part of the project No. 2017/27/N/ST10/02900.


Bella, P. & Zelinka, J., 2018. Ice caves in Slovakia. In: Ice caves (ed. A. Persoiu and S.-E. Lauritzen): 657-689. Elsevier, Amsterdam.
Gradziński, M., Hercman, H., Pereswiet-Soltan, A., Zelinka, J. & Jelonek, M., 2016. Radiocarbon dating of fossil bats from Dobsina Ice Cave (Slovakia) and potential palaeoclimatic implications. Annales Societatis Geologorum Poloniae, 86: 341-350.

Coarse cryogenic cave carbonate formation over last 16,000 years in Devaux cave (Central Pyrenees)

Miguel Bartolomé1, Marc Luetscher2, Christoph Spötl3, Gerard Cazenave4, María Cinta Osácar5, Ánchel Belmonte6, Hai Cheng7, R.L Edwards8, Carlos Sancho5, Ana Moreno1

1Instituto Pirenaico de Ecología, Spain; 2Swiss Institute for Speleology and Karst Studies, Switzerland; 3Institute of Geology, University of Innsbruck, Austria; 4Société de Spéléologie et de Préhistoire des Pyrénées Occidentales, France; 5Earth Science Department, University of Zaragoza, Spain; 6Geoparque Mundial UNESCO Sobrarbe-Pirineos; 7Institute of Global Environmental Change, Xi'an Jiaotong University, China; 8Department of Earth Sciences, University of Minnesota, USA

Coarse cryogenic cave carbonates (CCCcoarse) are rare crystals that form in caves when water freezes in a semi-closed system. Freezing provokes the loss of CO2 and the segregation of solutes, leading to the precipitation of CCCs. This work focuses on Devaux cave (2820 m a.s.l.) in the Central Pyrenees. A river runs through part of the cave fed by water from a polje and a ponor over the cave. During late winter and spring, the main cave outlet is blocked by seasonal ice formation. As a consequence, part of the cave is flooded, forming a seasonal lake that partially freezes. Dripping points are markedly absent in the first ~400 m of the cave, and a constant rock temperature of -1.2 ºC indicate permafrost conditions. Ice bodies inside passages are characterized by transparent non-layered ice, with huge ice crystals (30-40 cm) suggesting slow ice congelation in a flooded cave. CCCcoarse were found in a chamber located ~350 m from the cave entrance. The CCC were lying on blocks underneath an ice body covering the cave ceiling. CCCcoarse were also found trapped within the ice. Six different CCC types were identified based on size, shape and colour and individual crystals were dated by U-Th. Preliminary results show that the oldest CCC generation formed during the Late Glacial (15912±2780 yr BP), while several later episodes of CCC formation occurred during the early to late Holocene (9210±1230 to 2702±72 yr BP). The isotopic composition ranges from -21‰ to -12.2‰ and from 0.4‰ to 5.4‰ for 18O and 13C, respectively. Comparing crystal morphologies and isotopic compositions the six generations of CCCs fall on a single isotopic evolution line (r2=0.95). However, the isotopic composition of one specific generation is similar to other crystal generations of very different ages, indicating a different water composition. Moreover, some crystals show a trend towards more enriched 18O values along their growth axis. This could be related i) to partial dissolution of previous crystals formed in the same pool, due to CO2 increase (pH decrease) during the second pulse of congelation, or ii) to calcite crystal formation taking place during cave flooding while new water entered during congelation (semi-closed conditions). Overall, the presence of 5956±734 y old CCC trapped in cave ice rules out any recent influx of seepage water from the karst surface, strongly supporting the presence of uninterrupted mountain permafrost.

Ice caves as key archives for plant ancient DNA studies

Maria Leunda1,2, Christoph Sperisen2, Penélope González-Sampériz3, Graciela Gil-Romera3, Miguel Bartolomé3, Ánchel Belmonte4, M. Begoña García González5, Daniel Gómez-García5, Willy Tinner1, Nadir Alvarez6, Christoph Schwörer1

1Institute of Plant Sciences, University of Bern, Switzerland; 2Swiss Federal Institute for Forest, Snow and Landscape Research, Birmensdorf, Switzerland; 3Department of Geoenvironmental Processes and Global Change, Pyrenean Institute of Ecology, Zaragoza, Spain; 4Sobrarbe-Pirineos Unesco Global Geopark, Spain; 5Biodiversity and Restauration Department, Pyrenean Institute of Ecology, Zaragoza, Spain; 6Natural History Museum Geneva, Geneva, Switzerland

Perennial ice deposits in karstic caves represent unique paleoenvironmental archives to study past climate, environmental and ecological changes. In the last decade, ice caves have become an increasingly important target for the scientific community, however, they have hardly been studied from a paleoecological perspective. Ice caves act as natural traps for pollen and plant macrofossils, providing evidence of past vegetation changes. Due to the excellent preservation of plant macrofossils in frozen environments, ice deposits host ideal conditions for the conservation of their endogenous ancient DNA (aDNA), and thus, to conduct paleogenetic studies. Recent advances in molecular techniques allow the analysis of aDNA still present in macrofossils, providing insights into demographic processes and their genetic consequences across several hundreds to thousands of years. However, due to technical difficulties only few studies have successfully exploited aDNA from plant macrofossils.

The Pyrenees host a large number of ice caves with great potential for paleoenvironmental analyses, such as the Armeña A-294 Ice Cave (2238 m a.s.l; Spanish Central Pyrenees). This ice deposit contains several detrital layers with plenty of well-preserved plant macrofossils such as Pinus uncinata needles and Dryas octopetala leaves. The period recorded spans the mid to late Holocene, with ages ranging from 5700 to 2200 cal yr BP, making it the oldest known firn ice cave worldwide. The analyses of pollen and plant macrofossils carried out in a previous study revealed the impact of the Neoglacial cooling period (NGC; starting at ca. 4600 cal yr BP) on the vegetation, resulting in a progressive lowering of the treeline and the consequent establishment and expansion of alpine meadows dominated by D. octopetala in the surroundings of the cave.

In this study, we aim to evaluate the aDNA preservation of D. octopetala leaves in order to study the impact of the NGC on the genetic diversity of this species. Here we provide preliminary results of endogenous aDNA successfully isolated from D. octopetala leaves that date back to ca. 4800 to 2300 cal yr BP. Signs of chemical damage (cytosine deamination) indicate that the DNA we isolated is ancient. On the basis of our novel results we discuss the potential of ice caves for aDNA studies, providing valuable information about the preservation, extraction and amplification of aDNA from plant macrofossils.

Ice in lava tubes: paleoclimatic history in the southwest US

Laura Calabrò1, Bogdan Petroniu Onac1, Steve Baumann2, Eric Weaver2

1University of South Florida, United States of America; 2National Park Service, El Malpais and El Morro National Monuments, Grants, NM 87020, USA

The decadal and centennial climate variability of the southwest of the United States over the last 3000 years has been previously determined using speleothems and tree-ring studies. Here, we present a preliminary climate reconstruction from an under-investigated continental archive: cave ice deposits inside lava tubes in El Malpais National Monument. The area, located in the west-central part of New Mexico, is highly affected by distinct large-scale atmospheric circulation patterns (e.g., North American Monsoon System, El Niño-Southern Oscillation). Ice accumulates inside lava tubes as a result of seeping water from rainfall and snowmelt and retains the precipitation isotopic signature. The main objectives of this work are to establish the frequency and the causes of droughts, the source of moisture over the past 3000 years based on the isotopic composition of ice and to determine how these climate events affected the Native Americans and their activities inside and nearby lava tubes. Eight ice cores were cored from six lava tubes during two field campaigns and charcoal samples were collected from within the core and from the proximity of the ice deposits. Water stable isotope analysis results show values that span from –13.3‰ to –4.6‰ for oxygen and from –89.7‰ to –10.5‰ for hydrogen maintaining a different footprint based on whether the moisture source is the Pacific (more negative) or the Gulf of Mexico (less negative). The radiocarbon-dated charcoal recovered from the ice indicate that this variability occurred between AD 150 and 1400. The ice deposits are threatened by current warming trends, and the volume loss is significant and visible especially over the past decade, jeopardizing the paleoclimate record preserved within these archives.


Lorenzo Sánchez1, Antonio Garralon1, Maria Jesús Turrero1, Ana Isabel Ortega2, Paloma Gómez1, Miguel Ángel Martín-Merino3, Javier Martín-Chivelet4

1Unidad de Hidrogeociencias Ambientales, Dpto. de Medioambiente, CIEMAT, Madrid, 28040, Spain; 2Fundación Atapuerca-CENIEH, Paseo Sierra de Atapuerca, 3, Burgos, 09002, Spain; 3Grupo Espeleológico Edelweiss, GEE, Burgos, Spain; 4Dpto. de Geodinámica, Estratigrafía y Paleontología. Facultad de CC. Geológicas, Universidad Complutense de Madrid, Madrid, 28040, Spain

Torca La Grajera y Torca La Len, both located in Montes de Valnera, Cantabrian Range (NW Spain), are two potholes in which over time the snow precipitation events has become layered ice. Torca La Grajera is -185 m deep, and had around 15 m of accumulated ice, and Torca La Len is -55 m deep, and had around 28 m of accumulated ice; both are natural open spaces directly connected to the surface. Ice sample collection was made in three campaigns performed in September 2015, 2016 and 2017. The Grupo Espeleológico Edelweiss (GEE from its Spanish initials) was responsible for arranging all the equipment to descent into the hole and taking samples when possible from a quite narrow space left between the rock wall and the ice cone, which was the only accessible part to sample (Martín-Merino, 2017; Alegre, 2018). Unfortunately, there has been a dramatic loss of ice in last years and currently the presence of ice is only testimonial.

Ice sample chronologies become difficult due to impossibility of having a core and a clear picture for annual layer counting. However, enough organic remains could be recovered from some ice samples for radiocarbon dating. Lead isotope ratio measurements (Pb-IRM) have been used for tracing lead sources and environmental changes by using an HR-ICP-MS sector field ELEMENT™ 2 of ThermoFisher® mass spectrometer.

The data indicate that ice accumulation is 340  30 years old. The distribution of the ages allowed us to set the lead variability recorded in the ice samples. The results indicate that influence of industrial activity from the beginning of the industrial revolution can be observed, according the determined lead isotopes variations.

Martín-Merino, M.A., 2017. Prosiguen los trabajos en la Torca de La Grajera (Castro Valnera, Espinosa de los Monteros, Burgos). Cubía, 21: 18-21.

Alegre Rincón, R. 2018. Investigación paleoclimática en la Torca la Len (Cubada Grande). Nueva campaña de muestreos de hielo fósil en los Montes de Valnera (Espinosa de los Monteros). Cubía, 22: 16-19.

Contribution to Projects CGL2013-43257-R and CGL2017-83287-R (MICINN, Spain) and to research group ‘‘Sedimentary Geology, Paleoclimate and Environmental Change’’ (UCM, Spain).We thank Junta de Castilla y León for permissions and Grupo Espeleológico Edelweiss (GEE) for speleological work and taking ice samples.

Permafrost Evolution on the British Isles during the Last Deglaciation.

Paul Töchterle1, R. Lawrence Edwards2, John Gunn3, Tim Atkinson4, Julian B. Murton5, Marc Luetscher6, Gina E. Moseley1

1Inst. of Geology, University of Innsbruck, Austria; 2Dept. Earth Sciences, University of Minnesota, USA; 3School of Geography, Earth and Environmental Sciences, University of Birmingham, UK; 4Dept. Earth Sciences, University College London, UK; 5Dept. Geography, University of Sussex, Brighton, UK; 6Swiss Inst. for Speleology and Karst Studies (SISKA), La Chaux-de-Fonds, Switzerland

Globally, near-surface permafrost is likely to warm, thin or disappear in many areas subject to future climate warming and wetting, creating a positive atmospheric feedback where the permafrost is rich in carbon. Unfortunately, substantial uncertainty exists as to the extent and timing of thawing in response to atmospheric forcing. Cryogenic cave carbonates (CCCs), a recently described type of speleothem, precipitate when cave ice forms and thus provide opportunities to constrain periods when permafrost was present at a given cave site. Here, we present a unique dataset comprising 38 230Th/U ages of CCCs from two caves in the Mendips, southwest England (51°N), and two caves in the Peak District, central England (53°N), all of which are currently ice-free. Whilst many ages are clean, reliable and high precision, the accuracy of those containing initial 230This improved greatly by constructing isochrons and applying further statistical methods.

The ages of CCCs reveal two distinct periods of isothermal permafrost conditions, peaking during i) the early Bølling–Allerød interstadial at approximately 14,463 ± 145 yBP* and ii) the late Younger Dryas around 11,719 ± 229 yBP. Such isothermal conditions (i.e., where values of mean annual ground temperature are commonly a fraction of a degree below 0°C and exist through much of the depth profile of permafrost) are thought to represent the later stages of permafrost warming prior to its disappearance. We attribute this isothermal, disequilibrium permafrost evolution during the last deglaciation of the British–Irish Ice Sheet to climatic variations linked to North Atlantic sea-ice extent and seasonality.

*years before 1950

Winter precipitation records from ice caves of the Eastern Alps (Austria): stable-isotopic variability over the Common Era.

Tanguy Michel François Racine, Christoph Spötl

Universität Innsbruck, Austria

Stable isotopes of water are a famous proxy used ubiquitously in polar or glacier ice core studies to reconstruct past climate. Less known underground ice accumulations found in certain Alpine caves can reach millennial ages locally. Winter snow falling such single entrance descending caves (sag-type) undergoes firnification underground, with liquid water percolation and freezing contributing to the transformation of snow into ice. Stratified cave-ice deposits are thus archives with tangible links to past winter climate variability, but also to changes in the relative contribution of ice-forming processes: diagenesis and freezing.

Well-dated, replicated records from sag-type ice caves in the Alps are currently scarce, often lacking robust chronological control due to complex ice stratigraphy, discontinuous exposures or lack of datable material. The replicated analysis of several ice exposures spanning similar time periods, coupled with the monitoring of modern values in precipitation is thus needed to further develop the cave-ice stable-isotope proxy.

Using a 100 sample radiocarbon dataset to place the ice-accumulation in absolute chronological context, we present the stable isotopic record hosted in six well-dated underground ice records of the caves in the Austrian Alps spanning the last two millennia. We review the impact of fractionation processes which alter the original link to past precipitation by comparing modern drip water values and that of snow recently deposited in the cave. Comparing stable isotopic values of ancient ice with modern measurements carried out at stations of the Austrian Network of Isotopes in Precipitation, we find that ancient firn-derived ice stable isotopic values plot along the modern respective local meteoric water lines, suggesting that the ice cave records preserve past precipitation. In most highly resolved part of the proposed record, we report several multi-decadal to multi-centennial trends. In the period spanning 1000 CE to 1800 CE, we find several 100- to 200-year-long trends of enrichment or depletion ranging between 2-4‰ in δ18O values, and 15-30‰ in δD. We interpret these trends as indicators of past winter climate variability in the Alps, indicating chiefly past changes in moisture source, with a variable dominance of Mediterranean or North Atlantic precipitation.


Ros Fatihah Muhammad1, Chiang Hong Wei2, Lu Yanbin3, Wang Xianfeng4, Lim Tze Tshen5

1Universiti Malaya, Malaysia; 2National Taiwan University, Taiwan; 3Xi'an Jiaotong University, China; 4Nanyang Technological University, Singapore; 5Universiti Malaya

The variations and characteristics of the Indian summer monsoon (ISM) on different time scales are of crucial scientific and societal importance. The current understanding is that the ISM closely follows the Northern Hemisphere summer insolation on orbital timescales [1, 2], and responds to ocean circulation changes in the North Atlantic on millennial timescales [e.g., 3]. The solar activity is suggested to have dominant influence on ISM variability on shorter timescales during the Holocene [4, 5]. However, highly resolved proxy data are relatively sparse and scattered in the Southeast Asia, one of the critical zones for the ISM. Here, a fast-growing stalagmite (~1 mm/yr) was collected from Padang Kawad cave in the central of Peninsular Malaysia. The stalagmite contains high uranium contents of 3-5 parts per million. We have obtained an oxygen isotope (δ18O) record on the sample, which spans the past millennium at a temporal resolution of ~5 years. The δ18O data show strong multi-decadal to centennial oscillations which feature the ISM variability in this time period, and the δ18O mean value (~ -8‰) is consistent with the ones from northern Laos [6] and central and northeast India [7]. In addition, their δ18O and δ13C profiles share similar patterns and imply a potential correlation between large-scale monsoon circulation and local wet/dry conditions. We will further increase the record temporal resolution and also test the record reproducibility with another stalagmite from the same cave.

Rapid Mediterranean storm track reorganisation and cooling during the mid-Last Interglacial at Peqi'in (Israel)

Elan J. Levy1, Hubert B. Vonhof1, Miryam Bar-Matthews2, Alfredo Martínez-García1, Avner Ayalon2, Alan Matthews3, Vered Silverman4, Mareike Schmitt1, Gerald H. Haug1,5

1Department of Climate Geochemistry, Max-Planck Institute for Chemistry, Mainz, Germany; 2The Geological Survey of Israel, Jerusalem, Israel; 3The Fredy & Nadine Herrmann Institute of Earth Sciences, The Hebrew University of Jerusalem, Jerusalem, Israel; 4Department of Geophysical, Atmospheric and Planetary Sciences, Weizmann Institute of Science, Rehovot, Israel; 5Department of Earth Sciences, ETH Zürich, Zürich, Switzerland

The warm last interglacial (LIG, 129 to 116 thousand years ago) was punctuated by reductions of North Atlantic Deep Water (NADW) ventilation and North Atlantic Sea Surface Temperature (SST) cooling. In central and southern Europe speleothem proxy records exhibit evidence of abrupt mid-LIG cooling and climate change. In this study we reconstructed a high resolution TEX86 temperature record from a Peqi’in Cave speleothem (northern Israel). The record shows a climate optimum at ca. 128 - 127 ka followed by a significant irregular cooling of up to 4°C ca. 126 - 123 ka and then an increase in temperature from ca. 123 - 118 ka; providing for the first-time evidence of mid LIG cooling in the Eastern Mediterranean. Although the atmosphere is the link between the marine and terrestrial mid-LIG events there is a lack of atmospheric reconstructions from proxy records. Using a novel approach we reconstructed LIG atmospheric configurations in the Eastern Mediterranean from speleothem fluid inclusion δ18O and δD. Speleothem samples were crushed in a heated humidified carrier-gas line connected to a CRDS (Picarro L2140-i) at the Max Planck Institute for Chemistry (de Graaf et al., 2020). The data provide an unprecedented robust LIG fluid inclusion d-excess (δD – 8×δ18O) record at Peqi’in Cave, decreasing from a peak of 35‰ at ca. 128 - 127 ka to below 20‰ at ca. 126 – 122.5 ka (minimum 16.5‰) and then increasing after ca. 122 ka to 26 ‰ due to shifts in the paleo- Eastern Mediterranean Meteoric Water Line (EMMWL). To explain the LIG fluid inclusion d-excess trends we determined the long-term contributing factors for modern rainfall d-excess in Israel and applied them to the LIG. Using the ERA5 reanalysis and Lagrangian tracking tool (LAGRANTO), near-surface parameters (relative humidity, air temperature, SST) along air-trajectories with high ocean-moisture uptake were reconstructed for rainfall events at Soreq Cave between 1995 and 2021. The results argue for a wintertime storm track reorganization from meridional northerly winds over the Eastern Mediterranean at ca. 128 - 127 ka, to westerly air trajectories with increased relative moisture uptake along a W-E Mediterranean Sea transect between ca. 126 - 122.5 ka. The mid-LIG zonal Mediterranean wintertime westerlies would have resulted from a deepened low atmospheric pressure region over the mid-latitude North Atlantic, thus allowing North Atlantic climate to facilitate Eastern Mediterranean cooling.

S. De Graaf, H. B. Vonhof, T. Weissbach, J. A. Wassenburg, E. J. Levy, T. Kluge, G. H. Haug. A comparison of isotope ratio mass spectrometry and cavity ring‐down spectroscopy techniques for isotope analysis of fluid inclusion water. Rapid Communications in Mass Spectrometry, 34, p.e8837.6. (2020)

Abrupt climate change in the circum-Indian Ocean basin at the 4.2 kyr event? A multiple stalagmite, statistical approach to event detection

Nick Scroxton1, Stephen J Burns2, David McGee3, Laurie R Godfrey4, Benjamin Tiger3, Lovasoa Ranivoharimanana5, Peterson Faina5

1Irish Climate and Analysis Research Units, Maynooth University, Ireland; 2Department of Geosciences, University of Massachusetts Amherst, USA; 3Department of Earth, Atmospheric and Planetary Sciences, Massachusetts Institute of Technology, USA; 4Department of Anthropology, University of Massachusetts Amherst, USA; 5Mention Bassins sédimentaires, Evolution, Conservation (BEC), Université d’Antananarivo, Madagascar

Understanding the spatial pattern of abrupt climate anomalies is crucial to determine the mechanisms of change and to distinguish between unforced and forced climate variability. For example, it is not yet clear whether the 4.2 kyr event is globally coherent and synchronous, or a term given to a series of unrelated asynchronous climate changes occurring over a few hundred years. Additional insight into abrupt climate change can therefore be gained from a big data, statistical approach, synthesising numerous high-quality paleoclimate records and incorporating age model uncertainty. Speleothem records, thanks to their high precision U-Th dates, high sample resolution, and high data availability via the SISAL database, can be at the forefront of these new analyses.

In our study we investigate the mid to late Holocene transition (i.e., the 4.2 kyr event) in the tropical Indian Ocean using ten high-resolution precisely dated paleohydroclimate records. This includes six previously published speleothem records, one new speleothem record from Madagascar and three high resolution sediment cores. We conducted age uncertainty resolved Monte-Carlo principal component analysis on these records. The dominant mode of variability (PC1) was drying between 3.97 kyr BP (±0.10 kyr standard error) and 3.71 kyr BP (±0.09 kyr standard error). Dry conditions lasted for at least 300 years in some records but resulted in a permanent change in others. We interpret PC1 and the drying event as a proxy of summer monsoon variability, which fits with a previously recognised tropic wide change in hydroclimate at 4.0 kyr BP. An abrupt event from 4.2 to 3.9 kyr BP is seen locally in individual records but lacks regional coherence.

The lack of an apparent 4.2 kyr event in tropical Indian Ocean hydroclimate has ramifications for climate variability in the Indus valley and for our understanding of the collapse/deurbanization of the Harappan civilization. Further investigation comparing the seasonal ‘upstream source regions’ of the summer rainfall dominated tropical Indian Ocean and the winter rainfall dominated Middle East Winter Westerlies allows us to formulate the ‘Double Drought hypothesis’. We hypothesise that a winter rainfall drought between 4.2 to 3.9 kyr BP was followed by a summer rainfall drought between 3.97 kyr BP and at least 3.4 kyr BP. The Double Drought hypothesis provides more detailed climatic context for the Harappan civilization and resolves outstanding archaeological questions regarding the cropping paradox and the spatial-temporal pattern of urban abandonment.


Juan Luis Bernal Wormull1,2, Ana Moreno1, Miguel Bartolomé1, Arantza Aranburu3, Martin Arriolabengoa3, Eneko Iriarte4, Christoph Spötl5, Hai Cheng6, Carlos Pérez-Mejías6

1Pyrinean institute of ecology - CSIC, Spain; 2Earth Sciences Department, University of Zaragoza, Zaragoza, Spain; 3Department of Mineralogy and Petrology, University of the Basque Country, Leioa, Bizkaia; 4Department of Historical Science and Geography, University of Burgos, Burgos, Spain; 5Department of Geology, University of Innsbruck, Innsbruck, Austria; 6Institute for Global Environmental Change, Xi’an Jiaotong University, Xi’an, China

The Iberian Peninsula, due to its location in southern Europe close to the Atlantic Ocean, is a particularly sensitive area to major modes of atmospheric and oceanic circulation. Currently, in northern Iberia several speleothem records have become available to reconstruct Holocene paleoclimate (review by Baldini et al., 2019). However, the information is still scarce and insufficient considering the high spatial variability of this area and the complexity of that particular time period. In this work, new isotopic and trace element data from four overlapping stalagmites from Mendukilo cave (42º58’25’’ N, 1º53’45’’ W, 780 m a.s.l.) are provided. Our main objective is to obtain a high-resolution paleoclimate record to contribute to the understanding of the main oscillations during the Holocene at a regional scale.

The results indicate a continuous record for the last 12.7 kyr BP combining four overlapping speleothems (MEN-4 [0-3.0 ka BP]; MEN-3 [2.2-6.0 ka BP]; MEN-5 [0-8.9 ka BP]; MEN-2 [6.3-12.7 ka BP]) using iscam software and thus covering the entire Holocene and the Greenland Stadial 1 (GS-1) with decadal resolution. The isotopic values (δ18O and δ13C) of this composite record range between -4.31 and -6.33 ‰ in δ18O and between -4.43 and -10.41 ‰ in δ13C, pointing to important environmental changes in temperature and precipitation during the current interglacial period. A detailed monitoring survey in the cave is helping with the interpretation of main factors controlling isotopic variability.

During GS-1, the MEN δ13C dataset shows heavier values (-5.35‰) in relation to the averages of the Greenlandian period (-6.62‰), mean values of the MEN δ18O record are slightly lower than at the warmer beginning of the Holocene and presents the lowest speleothem growth rates of the entire record (29-37 μm/yr). All this allows to elucidate that drier and cold conditions dominated during GS-1. The beginning of the Holocene is characterized by heavier δ18O and δ13C values that progressively change to lighter ones, indicating drier conditions until ca. 10 ka BP when the vegetation cover got denser under more humid conditions in agreement with published lake records (Morellón et al., 2018). The end of the optimal humidity conditions and the beginning of the neoglacial cooling in the MEN composite record at 7.8 kyr BP is represented by a δ13C trend towards heavier values. This progressive passage to arid conditions intensifies at 5 kyr BP. On the other hand, the values of δ18O allow to identify, from the comparison with other speleothem records of the region, different abrupt events at a centennial scale (e.g. “8.2 event”) throughout the Holocene.

The MEN isotope record uninterruptedly follows the paleoenvironmental changes of the last 12,700 years and allows a good and chronologically continuous identification of the drastic events that are often seen more discreetly in other continental records of the region.

A Holocene speleothem record of paleoclimate in the Northern French Alps - Is d18O a proxy for karst recharge seasonality?

Isabelle Couchoud1,2, Russell Drysdale2, Yves Perrette1, John Hellstrom2, Marine Quiers3

1EDYTEM, Université Savoie Mont Blanc, France; 2School of Geography, Earth and Atmospheric Sciences, The University of Melbourne, Australia; 3SpecSolE, EDYTEM, Université Savoie Mont Blanc, France

The French Alps have been inhabited since Neolithic times. Local paleo-environments have been studied for the context of these early occupations, the evolution of landscapes, and interactions between the societies and their environments. However, these studies are mostly based on lake deposits, pollen and tree rings, with no speleothem studies thus far to bring complementary climatic information and precise chronologies.

Here we present the first reconstruction of Holocene paleoclimate evolution in the French Alps based on proxy records from four stalagmites from Garde-Cavale cave in the Bauges massif (45.64°N, 5.98°E). The stalagmites formed under the Revard plateau located at an average altitude of 1400 m asl. This is the highest and one of the first orographic obstacles on the trajectory of the westerlies, which carry most of the moisture to the site from the North Atlantic Ocean. Stable isotope records were reconstructed for each stalagmite at a sampling resolution of 1 mm. The chronology was constrained by MC-ICPMS U-Th dating and finite growth-rate age modelling. The average resolution of the obtained isotopic records ranges between ~9 and ~34 years depending on the stalagmite, and the stacked record covers the last 12 kyr.

Forty years of monthly data from the closest GNIP station (Thonon) were analysed to investigate the potential control of temperature and rainfall amount on the delta18Op. The d18Op values show a clear positive correlation with air temperature, but no obvious relationship to precipitation amount.

When the record is viewed over the entire Holocene, the d18O signal mainly follows an opposite trend to the 60N summer insolation intensity curve. We suggest that the evolution of insolation intensity through the Holocene influenced the pattern of the d18O signal recorded by the speleothems at this sub-alpine site via both its impact on the temperature of precipitation and on the seasonality of groundwater recharge. The relative proportion of summer or winter effective precipitation at this elevation, whose associated d18O values depend strongly on temperature, controls the overall d18O of the groundwater recharge that feeds the stalagmite growth. In addition, we find several short-term (e.g. 7.2 ka, 8.2 ka, 2.0 ka) and multi-centennial (4.2-3.2 ka, 0.9-0 ka) events superimposed on the long-term trend, which are associated with regional abrupt climatic events or possible anthropic activities, such as deforestation.

The 8.2 ka event in the North-Western French Alps from multiple speleothems

Isabelle Couchoud1,2, Russell Drysdale2, Yves Perrette1, John Hellstrom2, Hege Kilhavn1,2, Henri Wong3

1EDYTEM, Université Savoie Mont Blanc, France; 2School of Geography, Earth and Atmospheric Sciences, The University of Melbourne, Australia; 3ANSTO, Australia

The so-called 8.2 ka event represents one of the most prominent cold climate anomalies of the Holocene warm period. This short-lived event was forced by the release of freshwater to the North Atlantic Ocean during the final retreat phase of the last-glacial Laurentide ice sheet. The sudden input of freshwater slowed down the Atlantic meridional overturning circulation (AMOC), resulting in a pronounced regional cooling in the North Atlantic realm.

In Europe, several proxy-based studies have proposed reconstructions of the structure, timing, duration, amplitude of the event, and its climatic or environmental impacts. However, differences appear in these reconstructions and a lot of uncertainties remain over the interpretations, which makes it difficult to test climate-model simulations for the event. Differences might be associated with the complexity of proxy-based reconstructions (i.e. equivocal interpretations of proxies, various proxy/archive sensitivity), local threshold or site effects, and/or insufficient temporal resolution or dating precision.

Here we present a new study of the 8.2 ka event signature in speleothems from the French Alps. Speleothems suffer from the limitations evoked above. Hence, we try here to overcome some of these limitations using a multi-site approach in the same region, and using the same proxies. Delta18O, trace elements and organic fluorescence time-series are compared for speleothems growing in subalpine to montane settings (~500 to 1500 m a.s.l.) in two massifs on the western edge of the Alps, which are among the first obstacles that intercept the trajectories of the moisture-bearing westerlies from the North Atlantic Ocean. We exploit the differences between these records to refine interpretations of the climatic impacts of the 8.2 ka event in this region.

Middle Holocene ITCZ shifts: Evidence from Central Ecuador

Danny Vargas, Marjan Temovski, Elemér László, Gabriella Ilona Kiss, László Palcsu

Isotope Climatology and Environmental Research Centre, Institute for Nuclear Research, ELKH, Bem tér 18/c, 4026, Debrecen, Hungary

Speleothem-based paleoclimate studies have thrived in areas directly modulated by the South American Monsoon System (SAMS). However, equatorial-tropical records (low latitudes <10°) are scarce. The current understanding of the climatology and evidence from a few archives (e.g., tree rings) show that the Intertropical Convergence Zone (ITCZ) is the chief rainfall system controlling the hydrology at these latitudes. The SAMS only reaches its mature phase during austral summer (DJF), whereas the ITCZ is a full-year mode. Nevertheless, their dynamics are interconnected through the propagation of convection. Their extension and links in their features are observed in the modern climatology, but discussion about their past dynamics in northern South America is still lacking. Therefore, we show a new, U-Th dated, speleothem record from Garganta del Dino cave (Dino1 stalagmite), central Ecuador aiming to bridge the gap between the available interhemispheric records during the Middle Holocene (~5-7 ka). The source cave has a ponor morphology located within the Napo Cretaceous limestone formation at 1200 m a.s.l. at the eastern foothill of the Andes cordillera. It presents mostly vadose to epiphreatic passages, rich in various vadose speleothems. Ongoing cave monitoring started in February 2019 and comprises recording air temperatures on an hourly rate inside and outside the cave. Since 2021, dripping water has been collected monthly for δ18O and δ2H analyses and calcite farming was initiated at two selected sites. The monitoring program of local precipitation was also initiated, and it includes a collection of composite monthly rainfall samples for isotopic analyses (δ18O, δ2H, 3H).

Climate change event at 8.5 ka detected from Oman to the Carpathian Basin: teleconnections with the Indian ocean

Attila Demény1, Dániel Topál1, Gergely Surányi2, György Czuppon1, Ágnes Berentés1, Mihály Molnár3, Szabolcs Leél-Őssy4, Richárd Kovács5

1Institute for Geological and Geochemical Research, Research Centre for Astronomy and Earth Sciences, Hungary; 2MTA-ELTE Geological, Geophysical and Space Sciences Research Group, HUngary; 3Institute for Nuclear Research, Hungary; 4Department of Physical and Applied Geology, Eötvös Loránd University, Hungary; 5Duna-Ipoly National Park Directorate, Hungary

The Pilis caves (35 km NW from Budapest) have been investigated in recent years to determine if their speleothems can reliably record paleoclimate conditions. Drip water monitoring was conducted for two years that revealed varying water migration routes and winter/summer precipitation contributions (Czuppon et al., 2019). Investigations on calcite deposits on glass plates as well as carbonate samples obtained by scraping speleothem surfaces helped to distinguish ventilated and closed cave parts (Czuppon et al., 2022). Speleothems collected from the non-ventilated cave rooms were further studied by U-Th series dating, carbonate and fluid inclusion H-C-O isotope compositions, and Sr concentrations. A cca. 15 cm long and 7 cm thick stalagmite was selected for detailed investigations that yielded U-Th ages from 8.2 to 9.7 ka BP. The dataset includes 11 U-Th ages (with an average 2σ age uncertainty of 160 years), 20 AMS 14C analyses, 175 carbonate C-O isotope analyses, 19 fluid inclusion H isotope compositions, and 76 Sr concentration data. The calcite O and inclusion water H isotope records show about 2 and 0.5 ‰ fluctuations, respectively, between 8.4 and 8.1 ka, with a slight negative δ18O peak at about 8.3 ka, which may be attributed to the 8.2 ka event. Opposite to the slight isotope changes at about 8.3 ka, the H isotope record shows a ~10 ‰ negative δ2H peak at 8.5 ka. No similar isotope peak is detected in European speleothem records, except one stalagmite from the Limnon Cave (Greece, Peckover et al., 2019). Further to the east, the exceptionally well dated Jeita cave record (Chen et al., 2015), and the isotope records from Oman (Hoti Cave, Neff et al., 2001; Qunf Cave, Fleitmann et al., 2007) contain characteristic isotope peaks around 8.5 ka. We compute rolling-window correlation between the Vacska and Jeita cave records in every 250-yr windows between 8.0-9.0ka to find that two records show significant correlations only between 8.4 and 8.6 ka (r~0.6). However, the inferred „8.5 ka event” does not appear unequivocally in other speleothem records from the North Atlantic region to China, hence we hypothesise that this climate anomaly may have been confined to the region from Oman through the Eastern Mediterranean to the Carpathian Basin. To further test this idea, we analyzed TraCE-21ka simulations. Very similar correlations between global precipitation, surface temperature and 500-hPa geopotential heights were found with both the Vacska and Jeita records, reinforcing the synchronization event between the two records. Furthermore, both records significantly correlate with tropical Indian Ocean surface temperatures around 8.5 ka, which hints at possible atmospheric teleconnections underlying the 8.5 ka event.


Chen et al. (2015): doi:10.1002/2015GL065397.

Czuppon et al. (2021): doi: 10.1556/24.2021.00109

Czuppon et al. (2022): Springer Verlag ISSN: 2364-4591.

Fleitmann et al. (2007): doi:10.1016/j.quascirev.2006.04.012

Neff et al. (2001): doi:10.1038/35077048

Peckover et al. (2019): doi:10.1016/j.palaeo.2019.109252


Dana Felicitas Christine Riechelmann1, Bernd Reinhard Schöne1, Klaus Peter Jochum2, Denis Scholz1

1Institute for Geosciences, Johannes Gutenberg University Mainz, Mainz, 55128, Germany; 2Climate Geochemistry Department, Max Planck Institute for Chemistry, Mainz, 55128, Germany

Five stalagmites were sampled from different excavations in Dechen Cave (northwest Germany) and dated to Marine Isotope Stages (MIS) 9 and 11. MIS 9 is a very interesting interglacial due to its comparatively high concentration of greenhouse gasses (Lang & Wolff, 2011) and high temperatures (Petit et al., 1999), which are on the same level as during early industrial times (Robertson et al., 2001). Speleothem records from this time period are rare, in particular from central Europe.

The stalagmites were precisely dated by the 230Th/U-method using MC-ICP-MS (Mainz University). Trace element concentrations were determined using LA-ICP-MS (Max Planck Institute for Chemistry, Mainz) and stable carbon and oxygen isotope composition were measured using CF-IRMS (Mainz University).

The stalagmites show growth during MIS 9a and 9e and one stalagmite also shows a growth phase during MIS 11. This is also an interesting interglacial because even less stalagmite records exist from this time period, in particular from central Europe. Two stalagmites covering Termination IV and show a strong increase in the 18O values indicating the temperature increase during the termination. The timing of this increase corresponds well with the LR04 benthic foraminifera stack (Lisiecki & Raymo, 2005). First results from the trace element analyses indicate an increase of precipitation and vegetation density/activity during Termination IV.

Lang, N., Wolff, E.W., 2011. Interglacial and glacial variability from the last 800 ka in marine, ice and terrestrial archives. Climate of the Past 7, 361-380.

Lisiecki, L.E., Raymo, M.E., 2005. A Pliocene-Pleistocene stack of 57 globally distributed benthic δ18O records. Paleoceanography 20.

Petit, J.R., Jouzel, J., Raynaud, D., Barkov, N.I., Barnola, J.M., Basile, I., Bender, M., Chappellaz, J., Davis, M., Delaygue, G., Delmotte, M., Kotlyakov, V.M., Legrand, M., Lipenkov, V.Y., Lorius, C., Pépin, L., Ritz, C., Saltzman, E., Stievenard, M., 1999. Climate and atmospheric history of the past 420,000 years from the Vostok ice core, Antarctica. Nature 399, 429.

Robertson, A., Overpeck, J., Rind, D., Mosley-Thompson, E., Zielinski, G., Lean, J., Koch, D., Penner, J., Tegen, I., Healy, R., 2001. Hypothesized climate forcing time series for the last 500 years. Journal of Geophysical Research: Atmospheres 106, 14783-14803.

A high-resolution stalagmite record of Asian monsoon variations during Marine Isotope Stage 11 from Central China

Hongbin Zhang1, Griffiths Michael2, Shuyu Xue1, Junhua Huang1, Hai Cheng3, Youfeng Ning3, Shucheng Xie1

1State Key Laboratory of Biogeology and Environmental Geology, China University of Geosciences, Wuhan, 430074, China.; 2Department of Environmental Science, William Paterson University, Wayne, NJ 07470, USA; 3Institute of Global Environmental Change, Xi’an Jiaotong University, Xi’an, 710049, China.

Marine Isotope Stage 11 (MIS 11) has been widely considered as the most appropriate analogue to the current Holocene interglacial given similarity in orbital configuration and characteristic length in its interglacial state compared with any of the other interglacials during the last 500 ka. Deciphering the climatic features during MIS11 may thus help in understanding how the climate of the current interglacial evolved, and importantly help to assess the long-term stability of interglacial climates.

Here, we present a new multiproxy speleothem records from Haozhu Cave (HZZ) in central China, which document changes in East Asian summer monsoon (EASM) and local hydrology during 410 to 350 ka BP, covering the MIS11c, MIS11b and MIS11a substages. The HZZ δ18O record is consistent with published speleothem records from nearby Sanbao and Yongxing caves, and closely follows the northern high latitude summer insolation changes. Interestingly, our δ13C and trace elements ratios (Sr-Mg-Ba/Ca), which are significantly correlated suggesting that prior calcite precipitation (PCP) was the dominant driver of these proxies, are antiphased with the δ18O. Specifically, under drier conditions reduced infiltration within the karst fracture network likely favored CO2 degassing and PCP in the epikarst, leading to enriched trace elements relative to calcium in the stalagmite. Together, the composited δ13C and trace element records imply generally wetter (drier) conditions during periods of higher (lower) δ18O.

On orbital timescales, higher northern summer insolation would have enhanced the land-sea contrast and southerly winds, intensifying the EASM and thus leading to lower δ18O. At the same time, increased higher northern latitude warming would have decreased the equator-to-pole temperature gradient, leading to a more northerly displaced Westerly Jet and thus earlier transition to the Meiyu and Summer rainfall regimes. Consequently, a shorter Meiyu and longer Midsummer stage would have resulted in overall drier conditions in central China, as observed in our δ13C and trace element records. On millennial timescales, the δ18O and trace element changes during MIS11a show similar patterns, with higher δ18O (weak monsoon interval) corresponding to lower trace element ratios (wet), similar to the previous finding in Haozhu Cave during last deglaciation, implying a close correlation between northern high-latitudes temperature and East Asian hydroclimate.

Climate variability in SE Spain revealed by several flowstone deposits covering the last 400 ka

Michael Weber1, Alexander Budsky1,2, Hubert Vonhof3, Denis Scholz1

1Johannes Gutenberg-Universität Mainz, Germany; 2Landesmuseum Kärnten, Klagenfurt am Wörthersee, Austria; 3Max-Planck-Institute for Chemistry, Mainz, Germany

Modern climate in southeastern Spain is characterised by a strong seasonality of rainfall, with the annual precipitation sum usually not exceeding 300 mm. In addition, high temperatures lead to increased evapotranspiration, which further limits water availability. This causes a limited number of (high-resolution) speleothem deposits in the area. Cueva Victoria is one of the few speleothem deposits in SE Spain, and previous studies have shown that speleothem deposition occurred during Marine Isotope Stages (MIS) 5, 3 and the early Holocene, revealing hydrological changes on millennial and orbital time scale.

Here we present several flowstone records from Cueva Victoria covering MIS 11 – 6. U-series dating reveals at least episodical growth during both glacial and interglacial periods. Interestingly, speleothem growth rates were significantly reduced during the interglacial maxima, i.e., MIS 11c, 9e and 7e, compared to other interglacial phases. This change in growth rate as an indicator of water availability in Cueva Victoria is supported by changes in the (234U/238U)initial activity ratio, indicating wetter conditions during MIS 11a and 9c. Preliminary stable Ca isotope data support this trend in hydrology and demonstrate the potential of these proxies on the orbital time scale.

In addition, further flowstone deposits covering MIS 8 – 6 show distinctly different growth characteristics during these phases. One flowstone shows almost continuous growth during MIS 8 and large parts of MIS 7. Stable isotope data (δ13C and δ18O) show a drying trend with less vegetation cover as MIS 8 progresses, followed by a significant shift in δ13C and δ18O at Termination III / the onset of MIS 7e around 245 ka, indicating an increase in precipitation and vegetation density. Thereafter, δ13C values do not vary significantly until growth stops around 210 ka. In contrast, another flowstone shows only very short and episodic growth during MIS 7. However, the δ13C values show a similar value to the continuously growing sample, and the growth episodes correlate with the maxima of 7e, 7c and 7a, as well as with MIS 6.5.

To better understand climate variability in this currently semi-arid terrestrial climate environment of SE Spain, we will use further proxy data to establish a multi proxy approach that includes stable isotopes of C, O, Ca, as well as fluid inclusions to better understand precipitation and temperature evolution during the last 400 ka.

Evaluating the precession forcing on South Korean climate

Daniel M. Cleary1,2, Jasper A. Wassenburg1,2, Nitesh Sinha1,2, Sayak Basu1,2, Kyoung-nam Jo3, Timmermann Axel1,2

1Center for Climate Physics, Institute for Basic Science, Busan, Republic of Korea, 46241; 2Pusan National University, Busan, Republic of Korea, 46241; 3Department of Geology, Kangwon National University, Chuncheon, Republic of Korea

Chinese speleothem carbonate d18O records are dominated by precession cycles. This is in contrast with a rainfall reconstruction from the Chinese loess plateau and d18O seawater reconstructions from the East China Sea, which both show dominant eccentricity cycles. The interpretation of East Asian monsoon speleothem carbonate d18O is highly debated. One possible interpretation involves a variable balance between 18O-depleted rainfall derived from the Indian ocean and 18O-enriched rainfall derived from the western Pacific Ocean.

South Korean speleothem d18O records are expected to be more affected by the Pacific Ocean compared to the Chinese speleothems and thus may contribute significantly to the ongoing discussion. Former publications have shown that South Korean speleothem d13C records correspond well with Chinese d18O records, however South Korean d18O records do not. Here, we present new speleothem carbonate d18O and d13C, and fluid inclusion d18O and d-excess records that cover entire precession cycles from marine isotope stage (MIS) 7, MIS9 and MIS11. With the fluid inclusion d-excess we are able to address the evaporation conditions at the moisture source and its potential influence on rainfall d18O.

Ocean-land interactions determined the specific timing and structure of Termination II

David Domínguez-Villar1, Juan A. Vázquez-Navarro2, Kristina Krklec3, Sonja Lojen4, José A. López-Sáez5, Miriam Dorado-Valiño5, Ian J. Farchild6

1Department of Geology, Universidad de Salamanca, Spain; 2Department of Geography, Universidad Autónoma de Madrid, Spain; 3Department of Soil Science, Univeristy of Zagreb, Croatia; 4Jožef Stefan Institute, Slovenia; 5Institute of History-Centre for Human and Social Sciences, Spanish National Research Council (CSIC), Spain; 6School of Earth and Environmental Sciences, University of Birmingham, UK

Glacial terminations commence when ice-sheets from the Northern Hemisphere reach a supercritical size and their ablation is enhanced due to the increase of insolation during the Northern Hemisphere summer. Terminations reach a no-return point when the thermohaline circulation shuts down abruptly and triggers the bipolar seesaw mechanism that causes non-linear modifications on the global climate. Because of progressive variations of the solar forcing, suitable temporal windows for the occurrence of the bipolar seesaw mechanism exceed 10 ka, which represent a limitation on understanding the precise inception time of the bipolar seesaw mechanism and its causes. Therefore, controls that affect the structure and timing of terminations are still poorly understood.

We studied a tufa deposit from the Iberian Peninsula that covers Termination II (T-II) and the previous interglacial period. The studied tufa sequence provides a continuous record due to its specific hydrological and geomorphological setting and enables the possibility to study multiple proxies difficult to study in other karst records. The chronology of the tufa sequence was previously dated by U-Th and geomagnetism techniques and the oxygen isotope signal of the carbonates was synchronized to speleothem records from the Mediterranean to provide the age model of the deposit. We used the same chronology to synchronize ocean sediments from the North Atlantic to correlate major climate events in a common timescale.

We identify two stages within T-II. The first stage started with the increase of boreal summer integrated solar insolation, and during this stage three millennial climate oscillations were recorded. These oscillations resulted from complex ocean-atmosphere interactions in the Nordic seas, caused by the progressive decay of Northern Hemisphere ice-sheets. The second stage commenced after a glacial outburst that caused the collapse of the Thermohaline Circulation, a massive Heinrich event (HS11), and the onset of the bipolar seesaw mechanism that eventually permitted the completion of T-II. The pace of the millennial oscillations during the first stage of T-II controlled the onset of the second stage, when the termination became a non-reversible and global phenomenon that accelerated the deglaciation. During the last two terminations, the bipolar seesaw mechanism was triggered by different detailed climate interactions, which suggests the occurrence of different modes of terminations.

Orbital controls on East Asian hydroclimate during Marine Isotope Stage 6

Michael Griffiths1, Hongbin Zhang2, Clay Tabor3, Shuyu Xue2, Junhua Huang4, Hai Cheng5, Shucheng Xie2

1Department of Environmental Science, William Paterson University of New Jersey, Wayne, USA; 2State Key Laboratory of Biogeology and Environmental Geology, China University of Geosciences, Wuhan, China; 3Department of Geosciences, University of Connecticut, Storrs, USA; 4State Key Laboratory of Geological Processes and Mineral Resources, School of Earth Sciences, China University of Geosciences, Wuhan, China; 5Institute of Global Environmental Change, Xi’an Jiaotong University, Xi’an, China

Speleothem oxygen isotope records (δ18Oc) from China have provided detailed insights into past Asian summer monsoon variability on both orbital and millennial time scales. Recent multiproxy reconstructions and isotope-enabled general circulation model simulations suggest that δ18Oc represent the strength of the southerly Asian monsoon flow and fractionation of water vapor en route to the cave sites (Zhang et al., 2018; He et al., 2021). These findings reveal that central-eastern China was wetter during millennial-scale North Atlantic cooling episodes of the last deglaciation (Heinrich Stadial 1 and the Younger Dryas) coincident with an overall weaker Asian monsoon flow inferred from δ18Oc. While climate model simulations suggest similar hydroclimate patterns across China on orbital time scales—i.e., drier in the south and wetter in the north with higher summer insolation (e.g., Kong et al., 2017; Tabor et al., 2018)—we still lack robust proxy records to test model skill.

To this end, we present a new multiproxy speleothem record from Haozhu Cave in central-eastern China during MIS6 based on growth frequency, δ18Oc and δ13Cc from a total of 15 stalagmites. Results from ~90 U-Th dates show orbital-scale periods of active speleothem growth during MIS6a, MIS6c, and MIS6e when summer insolation was low, and periods of scarce speleothem growth during MIS6b and MIS6d when summer insolation was high. During these active growth periods, the δ13Cc records show decreasing trends as insolation was declining, while at the same time δ18Oc was increasing. We interpret these opposing isotopic trends to reflect enhanced karst dewatering and/or reduced C3 vegetation (higher δ13Cc), and stronger monsoonal flow (lower δ18Oc) during periods of higher summer insolation. These findings are consistent with recent isotope tracer-enabled Community Earth System Model (iCESM1.2) sensitivity experiments conducted under various combinations of precession-eccentricity (Tabor et al., 2018): higher summer insolation leads to both reduced rainfall and increased evaporation in central-eastern China, while at the same time increased monsoonal flow over South Asia brings more isotopically depleted precipitation across interior East Asia. In line with results from other previous modeling efforts under different orbital configurations (e.g., Kong et al., 2017), we propose that orbital changes in equator-pole temperature gradient modulated north-south migration of the westerly jet during MIS 6. Lower northern summer insolation during MIS6a, MIS6c, and MIS6e, enhanced the equator-to-pole temperature gradient and delayed the northward shift of the westerly jet, leading to a prolonged Meiyu season in central China and increased summer rainfall.


He, C., et al., 2021. Hydroclimate footprint of pan-Asian monsoon water isotope during the last deglaciation. Science Advances, 7, p.eabe2611.

Kong, W., Swenson, L.M. and Chiang, J.C., 2017. Seasonal transitions and the westerly jet in the Holocene East Asian summer monsoon. Journal of Climate, 30, 3343-3365.

Tabor, C.R. et al., 2018. Interpreting precession‐driven δ18O variability in the South Asian monsoon region. Journal of Geophysical Research: Atmospheres, 123, 5927-5946.

Zhang, H., et al., 2018. East Asian hydroclimate modulated by the position of the westerlies during Termination I. Science, 362, 580-583.

Orbital-to-millennial scale changes of the Central American Monsoon over the last glacial-interglacial cycle: Insights from Guatemalan Speleothem Records

Giuseppe Lucia1, Matthew Lachniet1, Amos Winter2, Juan Pablo Bernal3, Hai Cheng4

1University of Nevada Las Vegas, United States of America; 2Department of Earth and Environmental Systems, Indiana State University, USA; 3Centre for Geoscience, Universidad Nacional Autónoma de México, México; 4School of Human Settlement and Civil Engineering, China

Monsoon circulation changes may have dramatic impacts on the human agricultural and water supply systems in the tropics and subtropics. A comprehensive understanding of the forcing mechanisms and the different regional dynamics operating on paleoclimate timescales can help inform future projections, but such data is still limited for Central America. Various speleothem-based monsoon records, particularly in the East Asian and South American monsoon regions, document long-term variability linked to orbital precession cycles on ~19-23-kyr timescales. However, other monsoonal systems do not follow the ‘canonical’ orbital-insolation forcing. In Central America, for example, integrated terrestrial and marine paleo-hydroclimate proxies exhibit heterogeneous responses with a large sub-regional differentiation, evincing large uncertainties about the major drivers of millennial to orbital-scale Central American monsoon (CAM) changes. Here we present preliminary oxygen-isotope time series of three U/Th-dated stalagmites from the eastern Guatemalan highlands spanning the interval from 130 to 4 thousand years. The oxygen-isotope data exhibits synchronous co-variation with some global and regional records, notably, we notice significant correlation between speleothem δ18O and Caribbean Sea and Western Tropical Atlantic surface temperature (SST) records. Contrarily, local orbital summer insolation at 15°N does not exhibit an obvious control on the convention strength within the CAM. We also observe abrupt monsoon weakenings (high δ18O values) coincident with some North Atlantic cold periods, i.e., Heinrich stadials, despite tropical Atlantic and Caribbean SSTs remaining apparently high. Our preliminary results suggest that, in Central America, the rainfall response is more strongly related to SST than variations in orbital parameters, but is also related to ocean circulation and impacts on regional ocean-atmosphere interactions.

Paleoprecipitation data from northeast Greenland during MIS 11 and the mid-Pleistocene transition: preliminary results of a speleothem fluid inclusion study

Gabriella Koltai1, Anika Donner1, Xianglei Li2, Lawrence R. Edwards2, Christoph Spötl1, Yuri Dublyansky1, Gina E. Moseley1

1University of Innsbruck, Austria; 2University of Minnesota

The High Arctic is known to be a highly sensitive region for climate change. In response to global climate change, near-surface air temperatures in the Arctic have risen at a faster rate than the global average and this amplification is predicted to continue in the future. Climate models agree that a warmer and sea ice-free Arctic will lead to higher evaporation rates and a concomitant precipitation rise in the region. The Greenland ice cores provide a continuous record since the last interglacial, however, other well-dated proxy records from the High Arctic are scarce, hampering our understanding on past temperature variability and associated changes of the hydrological cycle under warmer than present-day conditions.

In the last few years, caves hosting inactive speleothems (mostly flowstone) were discovered in Silurian limestones at 80°N in northeast Greenland, approximately 35 km east of the coast and 60 km from the margin of the Greenland Ice Sheet. These caves are largely horizontal with a maximum length of 100 m and passages between ca. 350 m and 670 m a.s.l. The caves terminate at ice plugs, breakdown or sediment fill. Today, the area is in continuous permafrost and receives only ca. 200 mm of precipitation annually. The presence of inactive speleothems of Quaternary age indicates that significantly different climatic and environmental conditions prevailed in the past.

During speleothem growth, tiny amounts of dripwater are trapped as fluid inclusions. Their stable isotope composition provides insight into paleo-precipitation. Here we present preliminary results obtained from a flowstone that was deposited during Marine Isotope Stage (MIS) 11 and compare these to a stalagmite that grew during the mid-Pleistocene transition (MPT). Both specimens are dominated by translucent columnar calcite fabric that is rich in primary fluid inclusions. The latter vary in shape and size with elongated, thorn-shaped and down-arrow fluid inclusions being most common. Individual inclusions vary between 10–90 µm in length in the MIS 11 speleothem and 5–50 µm in the MPT speleothem. Large elongated inclusions up to 200 µm are also present in both specimens; these are often confined to distinct areas comprising macroscopically translucent calcite. Our preliminary stable isotope data indicates that the δD values of the paleo-dripwater were somewhat more depleted during the MPT (–182‰ to –135‰) compared to MIS 11 (–173‰ to –125‰). These initial results demonstrate the potential of speleothems to hold key information on NE Greenland´s paleo-precipitation.

Petrographic and geochemical properties of a subaqueous speleothem from a Macedonian cave – preliminary results and insight into MIS 7 paleoclimate record of the Balkan Peninsula

Marjan Temovski, Kata Molnár, Danny Vargas, Gabriella Ilona Kiss, László Palcsu

Isotope Climatology and Environmental Research Centre, Institute for Nuclear Research, ELKH, Bem tér 18/c, 4026, Debrecen, Hungary

The petrographic and geochemical properties of a subaqueous calcite coating were studied for its paleoclimate record. The speleothem was deposited in a fracture-controlled cave that formed in travertine deposits of a river terrace. The fossil Podot 2 Cave is located in Crna Reka valley, in the southern parts of N. Macedonia, in an output zone where intermediate to regional karst groundwater systems discharge cold to lukewarm waters. The cave is located ~60 m above the river bed and the current position of the Gugjakovo karst springs, that have relatively constant, but somewhat increased temperature (17.6°C), due to some contribution (≤16%) of old (~15 ka) thermal water.

The P2 speleothem is yellowish to brownish in color, ~25 mm thick of which the bottom ~20 mm is composed of densely laminated, mammillary calcite, that is covered by ~5 mm spar layer, with larger transparent crystals. Thin-section microscopy shows that the speleothem is composed dominantly of columnar calcite, with mosaic calcite fabrics found in sections of the mammillary part with high colloid density. The spar part is composed mostly of elongated columnar calcite, with no colloids present. The mammillary part has several pronounced boundaries, at some of which there is interruption in crystal growth with presence of micritic calcite, while others show continuous crystal growth across boundaries under cross polarized light. U-Th dating (n=10) indicates that the speleothem grew between 241±3 ka and 165±2 ka BP, with a major growth hiatus between the mammillary and the spar part. Linear age-depth model was constructed for the spar part, while the age of the mammillary part was modeled with StalAge. The age-depth model, together with the petrographic analysis suggest continuous growth in the mammillary part that covers the first half of MIS7, and grew mostly during the MIS7e interstadial. Growth rate is slow with highest values (2.5-3.9 mm/ka) for sections that grew during MIS7e, and significantly lower (0.5 mm/ka) for the younger mammillary section, similar to the spar part (0.1 mm/ka).

Stable isotope composition along a hand-drilled profile at ~1 mm increment, shows variation of up to 1.7 ‰ in δ18O and up to 4 ‰ in δ13C values. The spar part has similar δ18O but much higher δ13C values than the mammillary part, which combined with the petrographic observations suggests that it grew in an isolated pool environment, likely after lowering of the water table. The stable isotope composition of the mamillary part shows decreasing trend in δ18O values throughout MIS7e, and much less pronounced coeval increase in δ13C values. These trends are comparable to, but opposite of what was found for Lake Ohrid calcite, where increase in δ18O is interpreted as a transition to drier conditions in a generally warm and wet climate. With overall MIS7e warming indicated by a number of regional paleoclimate records, the lowering of δ18O values likely reflects an increase in winter precipitation, as local cave monitoring shows winter bias in dripwater stable isotope composition.

Pleistocene hydroclimatic variability recorded in high-altitude speleothems from the Făgăraș Mountains (Romania)

Maria-Laura Tîrlă1,2,3, Gelu Costin4, Virgil Drăgușin2,3, Vasile Ersek5, Constantin Marin3, Ionuț C. Mirea3, Relu D. Roban6

1Faculty of Geography, University of Bucharest, Romania; 2Research Institute of the University of Bucharest (ICUB), Romania; 3Emil Racoviță Institute of Speleology, Romanian Academy, Bucharest, Romania; 4Department of Earth, Environmental and Planetary Sciences, Rice University, Houston (TX), USA; 5Department of Geography and Environmental Sciences, Northumbria University, Newcastle upon Tyne, UK; 6Faculty of Geology and Geophysics, University of Bucharest, Romania

High-altitude speleothems record information about past environments – primarily climate and hydrology – of particular terrestrial settings like glaciokarst islands. In this study we present petrographic and geochemical data from three flowstones and bedrock samples from the Făgăraș Mountains (South Carpathians, Romania) in order to investigate the controlling factors on karstification and speleothem geochemical proxy records. The Făgăraș Mountains karst developed in dolomitic and calcitic marbles with interbanded silicate rocks. Prior to this study, speleothem U-Th ages from M6, M5, and M3-R2 caves (~2400 m a.s.l.) were reported at >560 to 526 ka, 440 to 398 ka, and 330 to 95 ka (here excluding uncertainties) and it was shown that they formed mostly during interglacials.

Petrographic determinations were based on detailed thin section study using optical microscopy and electronic imaging (backscattered electrons), assisted by energy dispersive spectrometry (EDS) and wavelength dispersive spectroscopy (WDS) analysis. Elemental analysis was performed using a field emission electron probe microanalyzer at 250-µm intervals, with a beam diameter of 5 µm for major elements and 20 µm for trace elements. A complete geochemical profile was done only for the youngest flowstone (M3-R2-3), and partial sections (2.5 cm) for each of the older flowstones (M5-1 and M6-1). For stable isotope analysis, sampling was conducted at 5-mm intervals.

Mg/Ca ratio apparently decreases with age, recording large shifts in flowstone M6-1 and a sudden drop at the base of sample M3-R2-3. A petrographic transition from micron-thick laminae to columnar calcite layers, accompanied by another drop in Mg/Ca ratio occurred at ~3 cm from the base of M3-R2-3 (age ~200 ka), probably under more humid climatic conditions and changes in water inflow. Ba and Sr contents are very low in all samples, barely detectable in both speleothems and marble and do not show any trend. δ13C values as low as -10‰ indicate a higher soil CO2 contribution to speleothem calcite, detrital-rich and moonmilk layers have values of -4.5‰ to -2.7‰, and bedrock marble δ13C is ~1‰.

The variability of trace element concentrations in speleothem calcite could reflect shifts in local hydrology under control of groundwater recharge patterns. We also explored the possibility that higher Mg content in older speleothem layers could indicate the presence of Mg-rich dolomitic marbles on the groundwater pathway, which were later removed by erosion. Speleothem δ13C and δ18O data provided information on the succession of distinctive climatic episodes, ranging from a higher variability in MIS 11 to a relatively stable, warm and dry climate during MIS 5. This study laid more light on how speleothems recorded site-specific changes that occurred at high altitudes in the South Carpathians during the Middle to Late Pleistocene, and thus contributing to a better understanding of regional climatic and landscape evolution in Central-Eastern Europe.

Račiška pečina speleothem dome section: Plio-Holocene multi-proxy records

Nadja Zupan Hajna1, Andrej Mihevc1, Pavel Bosák2, Petr Pruner2, Helena Hercman3, Ivan Horáček4, Jan Wagner5, Stanislav Čermák2, Jacek Pawlak3, Paula Sierpień3, Šimon Kdýr2, Lucie Juřičková5, Astrid Švara1

1Karst Research Institute Research Centre of the Slovenian Academy of Sciences and Arts, Slovenia; 2Institute of Geology of the Czech Academy of Sciences; 3Institute of Geological Sciences of the Polish Academy of Sciences; 4Department of Zoology, Faculty of Sciences, Charles University; 5Department of Paleontology, National Museum, Czech Republic

The sedimentary sequence of the flowstone dome from the Račiška pečina represents an exceptional chronostratigraphic record from the late Pliocene to the present. It is one of the best preserved cave records of the paleoenvironmental change for the last 3.4 million years in general. The occasional deposition of speleothem layers was interrupted by sedimentation of clayey to silty material infiltrated from the surface above the cave. A detailed chronology of the Račiška pečina section was established based on magnetostratigraphy and isotope-oxygen stratigraphy and correlated with paleontological, U-series, and radiocarbon results. The Račiška pečina section was deposited from about 3.4 Ma to about 80 ka (MIS Km3 to MIS 5). In the sequence, the Pliocene to Pleistocene transition at 2.59 Ma, the presence of the Olduvai subchron between 1.78-1.925 Ma, and the Matuyama/Brunhes geomagnetic field reversal are well recorded by magnetostratigraphy. The section was divided into two segments separated by a major unconformity (hiatus) for stable isotope studies. The paleoenvironmental changes associated with these disconformities were well expressed by the changing stable isotope values. The lower segment correlates better with the regional Medi curve, while the upper segment correlates with the global LR04 curve, indicating changes in the main factors controlling environmental conditions in the area. The sequence contains an important and representative Early and Late Pleistocene fauna (e.g., Ursus ex gr. Spelaeus and Apodemus cf. atavus, Borsodia sp., Pliomys sp., and Clethrionomys cf. glareolus) and probably the first known fossil cave gastropod (i.e. Zospeum sp.). In the upper part of the section, above the remains of cave bears in the yellow clay layer older than ~72 ka, a soot material from three layers was radiocarbon dated to ~11 ka, ~9 ka, and ~3 ka. Lengthy and detailed analyzes of speleothem dome have shown that they contain a huge variety of data that cannot be collected in a single borehole or stalagmite.

Speleothems record timing of rainfall in the northwest Sahara during the middle to late Pleistocene

Hamish Owain Couper1, Christopher Day1, Julia Barrott1, Said Maouche2, Aboubakr Deramchi2, Stacy Carolin3, Mohamed El Messaoud Derder2, Abdelkarim Yelles Chaouche2, Gideon Henderson1

1Department of Earth Sciences, University of Oxford, United Kingdom; 2Centre de Recherche en Astronomie Astrophysique et Géophysique, CRAAG. Route de l'Observatoire, B.P 63 Bouzareah, Algérie; 3Department of Earth Sciences, University of Cambridge, Cambridge, UK

The hyper-arid Saharan desert belt stretching across North Africa is an important part of the global climate system, with dust export shown to influence climate systems such as ENSO and distant monsoon systems. Understanding climate dynamics and potential future changes in this region is however difficult due to a paucity in both instrumental and high-resolution paleoclimate data. There is strong evidence for periods of increased rainfall across large parts of North Africa during the late Quaternary, termed ‘Green Sahara’ periods, which contribute to regional aquifer recharge and improved human population connectivity across the Sahara. There is, however, currently limited evidence regarding: i) precisely where and when rainfall occurred and; ii) the sources of moisture contributing to increased rainfall at the northern-most reaches of the Sahara.

In this study, we present new proxy reconstructions from the northern limits of the presently hyper-arid Sahara Desert, to identify moisture sources, timing and latitudinal extent of rainfall change at precise locations. We do this using several ancient fossil stalagmites collected from cave sites in the desert foothills of the central Saharan Atlas Mountains, Algeria. High-precision U-Th chronology and stable-isotope measurements on calcite samples from multiple cave sites contribute towards an east-west transect of records. Due to the locations of the caves, stalagmite growth periods and stable isotope records provide direct evidence of where and when there was significantly increased rainfall in this region, and help us to identify potential sources of moisture through time. We present these results, and their implications for a more detailed reconstruction of the occurrence of increased rainfall in northwest Africa.

Termination I in the central Mediterranean

Andrea Columbu1, Christoph Spötl2, Jens Fohlmeister3, Hsun-Ming Hu4, Veronica Chiarini5, John Hellstrom6, Hai Cheng7, Chuan-Chou Shen4, Jo De Waele5

1University of Parma, Italy; 2University of Innsbruck, Austria; 3German Federal Office for Radiation Protection, Berlin, Germany; 4National Taiwan University, Taipei, Taiwan; 5University of Bologna, Italy; 6University of Melbourne, Australia; 7Xi’an Jiaotong University, China

The last deglaciation (Termination I, T-I) was the most recent global-scale climate transition. It involved a drastic temperature increase guiding massive melting of ice sheets, with a concurrent reorganization of inter- and intrahemispherical atmospheric and ocean circulation patterns.

T-I lasted ~3000 years in Greenland (1), although it was not a linear process. A rapid temperature increase at 4.6 ka (Bølling-Allerød, BA) was followed by a return towards glacial-like conditions (12.9 ka, Younger Dryas, YD), before the last warming that led to the Holocene (11.7 ka). Other secondary climate oscillations characterized T-I too (2). Some of these intra-deglaciation global warmings were particularly rapid, at times occurring at centennial or even decadal timescale. Comprehensively, T-I provides an interesting analogue to the current climate change.

Yet, it is not clear how T-I-related dynamics occurring at the polar regions and/or in the oceans impacted the terrestrial environment at mid latitudes, in terms of rainfall and temperature variation and related ecological changes. This is especially true for the Mediterranean area, considering that its climate is connected – and controlled – by processes occurring in the Atlantic and Arctic. In Italy for example, T-I records of adequate chronological resolution are virtually absent.

Here, we present a high-resolution reconstruction of rainfall changes in the Central Mediterranean across T-I. It is based on a novel δ18O-δ13C time series from a southern Italian stalagmite, deposited in Sant’Angelo Cave (SA1, Ostuni, Apulia) between 47.8+0.41/-0.39 and 6.63+0.62/-0.57 ka. In the period from 20 to 9 ka, multiple U-Th ages (n=22) result in a final age model with an average uncertainty of <0.3 ka and a resolution of ~25 years. Climate proxies (δ18O-δ13C, 0.2 mm spacing) were anchored to this chronology. The resulting δ18O-δ13C timeseries demonstrates high similarities with Greenland’s reconstructions, along with interesting divergences. The reliability of SA1-δ18O in recording palaeoclimate information was ascertained by a statistical inter-cave replication test with a recently published speleothem record from a nearby site (3). The interpretation of SA1 allows to: 1) accurately constrain the timing of the hydroclimate across T-I; 2) evaluate the impact of BA, YD and early Holocene, as well as other associated events especially in terms of rainfall amount and provenance variability; and 3) understand the spatio-temporal relation between the Atlantic/Greenland domain, the western-central Mediterranean realm and possible influences on the Middle East and Asian monsoonal sectors throughout T-I. We discuss this new record within the framework of regional studies based on glacial, marine and continental archives, with the aim of providing a better comprehension of the timing and structure of T-I in Southern Italy and, by extension, of the central Mediterranean area.


1 NGRIP 2004. High-resolution record of Northern Hemisphere climate extending into the last interglacial period. Nature, 431, 147-151.

2 Cheng H et al 2020. Timing and structure of the Younger Dryas event and its underlying climate dynamics. PNAS, 117, 23408-23417.

3 Columbu A et al 2020. Speleothem record attests to stable environmental conditions during Neanderthal-Modern Human turnover in Southern Italy. Nature Ecology & Evolution, 4, 1188-1195.

Uncoupling of subtropical moisture availability and temperature over multiple glacial cycles: insights from speleothems from the Naracoorte Caves in southern Australia

Rieneke Weij1,2,3, J.M. Kale Sniderman1, Jon D. Woodhead1, John C. Hellstrom1, Russell N. Drysdale1,4, Liz Reed5,6, Steven Bourne7, Jay Gordon1,8

1School of Geography, Earth and Atmospheric Sciences, University of Melbourne, Melbourne, Australia; 2Human Evolution Research Institute, University of Cape town, Rondebosch, South Africa; 3Department of Geological Sciences, University of Cape town, Rondebosch, South Africa; 4Environnements, Dynamiques et Territoires de la Montagne, UMR CNRS, Université de Savoie-Mont Blanc, Le Bourget du Lac, France; 5Ecology and Evolutionary Biology, School of Biological Sciences, University of Adelaide, Adelaide, Australia; 6Earth and Biological Sciences (Palaeontology), South Australian Museum, Adelaide, Australia; 7Limestone Coast Landscape Board, Mount Gambier, Australia; 8Climate Works Australia, Melbourne, Australia

Studies of past climate change are pivotal for understanding the future terrestrial response to further global warming, especially in the Southern Hemisphere (SH) semi-arid subtropics, where the annual moisture balance is negative. However, from these regions, well-dated records that cover multiple glacial-interglacial cycles are scarce. Here, we present a radiometrically-dated record of subtropical moisture availability using U-Th age frequency distributions and pollen analysis of speleothems from the World Heritage Naracoorte Cave Complex in South Australia. Additionally, we provide a framework for the utility of speleothem age frequency distributions for palaeoclimate assessment.

We identify five periods of increased speleothem growth (with maxima at 303 ka, 192 ka, 96 ka, 24 ka and 6 ka) and confirm, using quantitative climate reconstructions based on fossil pollen extracted from the speleothems, that periods of increased speleothem growth represent times of enhanced moisture availability in the past. We show that speleothem growth peaks do not coincide with maximum interglacial temperatures, but in fact lag them by ~25 ka over the past three glacial-interglacial cycles, and that times of highest moisture availability were confined to parts of the Mid to Late Pleistocene glacial periods, rather than interglacials. By comparing with a range of SH subtropical records, we demonstrate a hemisphere-wide pattern during the last glacial-interglacial cycle in which intervals of high moisture are driven by lower evaporation during cooler conditions. This implies that temperature and moisture availability can be strongly uncoupled in the SH subtropics.

Assessing the potential use of ‘stable’ uranium isotopes as a paleoredox proxy in speleothems

Ashley N. Martin1, Monika Markowska2, Stefan Weyer1

1Leibniz University Hannover, Germany; 2Max Planck Institute for Chemistry, Mainz, Germany

The typical application of uranium (U) isotopes in speleothems is determining accurate and reliable age chronologies, but ‘stable’ U isotopes could additionally provide important information about paleoenvironmental conditions. Some studies have investigated the disequilibrium between 234U-238U as a paleohydrology proxy, but the stable isotope fractionation between the two long-lived, primordial isotopes (238U and 235U) remains largely unexplored. Variations in 238U/235U arise during redox processes due to the nuclear field shift effect, whereby 238U is preferentially reduced and 235U is favourably adsorbed onto to Fe-Mn (oxy)hydroxides. Moreover, U isotopes are not strongly fractionated during the precipitation of calcite and aragonite. Here we measure the U isotope composition of a range of speleothems from several caves in different climates (arid Arabia, temperature Australia and tropical Indonesia) to firstly assess if there are significant variations and, if so, examine which processes may have generated these variations.

Speleothem carbonate δ238U ranged from -0.52 to -0.11‰ (n=13), providing further evidence that there is indeed significant fractionation between 238U and 235U in cave environments. This range of values is unlikely to simply reflect local variations in the bedrock as the upper continental crust exhibits a narrow range of δ238U from -0.2 to -0.4‰ and, thus, measured δ238Uspeleothem both above and below this range suggest that redox processes in the unsaturated zone affect U isotopes prior to speleothem precipitation. No correlations between δ238U and other parameters were observed when comparing samples from all sites together, but systematics trends were found within single speleothems. For instance, δ238Uspeleothem from temperate Australia ranged from ca. -0.1 to -0.4‰ (n=3), and correlated with molybdenum concentrations, δ234U and Th/U. These parameters have important implications for palaeoenvironmental reconstructions, for instance, molybdenum is suggested to be a useful proxy for recording volcanic eruptions in speleothems, whereas Th contents and δ234U are key parameters that affect U-Th age determinations. This pilot dataset offers a tantalising hint at the potential information that could be obtained from variations of 238U/235U in speleothem archives.

Help! My fluid inclusion isotope values are not on the meteoric water line!

Hubert Vonhof1, Elan Levy1, Monika Markowska1, Denis Scholz2, Stefan de Graaf1

1Max Planck Institute for Chemistry, Mainz, Germany; 2Johannes Gutenberg-Universität, Mainz, Germany

The analysis of fluid inclusion oxygen and hydrogen isotope ratios is increasingly applied in speleothem paleoclimate reconstructions. The technique has great potential, because the oxygen and hydrogen isotope values are a direct proxy of the isotope values of past rainfall above the cave. Speleothem fluid inclusion isotope time series can thus be used to reconstruct changes in past precipitation and/or atmospheric circulation, in a way that was previously only possible for ice cores.

Pioneering work in the late 1970’s led to the widely supported view that, while hydrogen isotope values of fluid inclusion water in speleothems can give useful results, the oxygen isotope values are likely altered due to post depositional exchange of oxygen isotopes with the CaCO3 host mineral. With new continuous-flow techniques that were developed over the past ca. 15 years, it has become clear that, while diagenetic alteration of fluid inclusion isotope values can take place indeed, it is not a process that occurs very commonly in speleothem calcite. Multiple recent publications have presented what appear to be unaltered hydrogen and oxygen isotope data from speleothem fluid inclusions, even in specimens up to several million years old.

In several studies that used combined oxygen and hydrogen isotope data from speleothems, the quality assessment of the isotope data has been based on demonstrating their proximity to the global or local meteoric water lines (GMWL or LMWL), which represent the accepted empirical relationship between the hydrogen and oxygen isotope values of rainfall. The assumption is that, since diagenetic alteration of the originally rainwater-like fluid primarily influences the oxygen isotope ratios, the isotope data can only remain on the MWL if no post-depositional alteration takes place. Another line of reasoning in quality control of fluid inclusion isotope data is to check if fluid inclusion oxygen isotope data yield expected temperatures when applying commonly-used oxygen isotope equilibrium equations for the speleothem calcite – fluid inclusion water system.

We will present some recent case studies that show distinct off-MWL patterns, or anomalous oxygen isotope equilibrium temperatures. For some of these cases, we believe diagenetic alteration or analytical artefacts play a role. In other cases, we believe the fluid inclusion isotope data to be unaltered. We will discuss underlying mechanisms and possible next steps in quality assessment of fluid inclusion isotope data for paleoclimate reconstruction.

Systematic assessment of reproducibility and accuracy of the analysis of hydrogen and oxygen isotopic composition of fluid inclusions in speleothems using the Heidelberg fluid inclusion line

Yao Wu1, Tobias Kluge1,2, Ting-Yong Li3, Sophie F. Warken1,4, Martina Schmidt1, Norbert Frank1, Werner Aeschbach1

1Institute of Environmental Physics, Heidelberg University, Germany; 2Institute of Applied Geosciences, Karlsruhe Institute of Technology, Germany; 3Yunnan Key Laboratory of Plateau Geographical Processes & Environmental Changes, Faculty of Geography, Yunnan Normal University; 4Institute of Earth Sciences, Heidelberg University, Germany

Here we present recent improvements of precision and accuracy of water isotope fluid inclusion measurements on speleothems as well as first data from speleothems of Yangzi cave. Sub-gram carbonate samples are crushed in a heated and continuous water vapour flushed extraction line. A cavity-ring-down laser spectrometer (CRDS) is used to continuously analyze the gas mixture and provides the water volume and δ18O, δ2H values of the fluid inclusions after peak integration and background subtraction (Warken et al. 2021, Weißbach et al. (in press)). Our replicated injections of water standards (N = 300) over a time period of 4 months show a precision (1σ) of ± 0.5 ‰ for δ18O and ± 1.5 ‰ for δ2H in case of an extracted water amount <1 µl and of ± 0.15 ‰ and ± 0.4 ‰, respectively, above 1 µl. The absolute accuracy is better than ± 0.08‰ for δ18O and ± 0.3 ‰ for δ2H. The external reproducibility (1σ) based on replicate fluid inclusion analysis of speleothem samples from the same growth layers with water amounts > 0.2 µl corresponds to the standard deviation of the reference materials (± 0.5 ‰ for δ18O and ± 1.5 ‰ for δ2H, N = 16).

The technique is used to explore the water isotope fluid inclusion of speleothems from Yangzi Chinese caves to trace the provenance of precipitation during the past climate cycles. Numerous speleothems have been dated and stable isotopes measured. Moreover, for speleothem YZ1 we have collected a first in-situ laser ablation Mg/Ca, Sr/Ca time-series. Here, we will discuss the first preliminary fluid inclusion measurements in light of observed Monsson and climate variability since termination II.

Systematic analysis of correction factors and an open-source software for MC-ICPMS 230Th/U-dating

Inga Kristina Kerber1,2, Fabian Kontor1, René Eichstädter1, Athulya Babu1, Andrea Schröder-Ritzrau1, Sophie Warken1,2, Norbert Frank1

1Institute for Environmental Physics, Heidelberg University, Heidelberg, Germany; 2Institute for Earth Sciences, Heidelberg University, Heidelberg, Germany

230Th/U dating is a crucial tool for accessing the timing of events of the last few hundred thousand years in secondary carbonates, typically measured using multi-collector inductively coupled plasma mass spectrometers (MC-ICP-MS). Here, we present our measurement protocol at Heidelberg University, designed for combined Faraday cup (FC) and secondary electron multiplier (SEM) measurements. Furthermore, we show our Python-based algorithm equipped with a graphical user interface (GUI) that comprises all raw data treatment, corrections and age calculation. Lastly, we demonstrate the influence of variations in correction factors on final age accuracy.

For the measurement protocol, we describe our method for the transfer of tailing reference solutions measured on a SEM with a retarding potential quadrupole (RPQ) to FCs without RPQ and prove the method’s stability. Furthermore, we re-determined fade-out times of FCs, i. e. for the 1013 Ω amplifier, at typical signal intensities. On masses 229 amu and 230 amu, we observe “ghost signals”, for which we introduce our correction method. Our data evaluation software comes with features that ensure reproducibility and enable user-friendly reanalysis of measurements: customized calculation constants with templates, and automated reanalysis of old data with new constants. Result files contain all relevant used constants and are saved automatically.

We systematically analyse the effect of undetected variation in different MC-ICP-MS raw data corrections on the accuracy of the activity ratios (234U/238U) and (230Th/238U) and the ages. To do so, three speleothem and coral samples of different isotopic concentrations and ages were employed. Here, we find that hydride correction does not contribute any relevant offset in accuracy. However, frequent measurements of process blank and tailing correction are crucial, otherwise tailing can cause up to permille-level deviations in both young and old high intensity samples. Process blank in turn mainly affects young samples and could cause up to permille level offsets there. The largest offset however would be introduced by not applying the “ghost signal” corrections. This leads to deviations from the reference value of more than 10% for young samples, and still up to 1% for older samples. In everyday lab routine, our protocol allows for (234U/238U) measurements on the ε-level and (230Th/238U) determinations on the permille scale.

Precise and accurate 230Th/U dating of small-scale and short-term speleothem growth phases – comparison of different sampling strategies

Michael Weber, Jennifer Klose, Denis Scholz

Johannes Gutenberg-Universität Mainz, Germany

Climate reconstructions based on speleothems largely dependent on a precise and accurate age model. The most commonly used dating technique for speleothems is 230Th/U dating, which allows precise age determination of speleothem samples up to 600 ka. State-of-the-art MC-ICP-MS systems in combination with calibrated spike solutions allow highly precise and accurate age determinations, even for small sample sizes (<<50 mg). However, there are still limitations in the sampling strategy, in particular for sampling of very thin growth layers and short growth phases. The variable growth behavior of different speleothems (e.g., stalagmites and flowstones) potentially requires adapted sampling strategies. Potential approaches, such as micromilling or the direct coupling of a laser ablation (LA) system with the MC-ICP-MS system can enable the sampling of small-scale growth features.

Here we present different sampling methods for 230Th/U dating of speleothems, including both solution-based and LA-MC-ICP-MS analyses. While LA-MC-ICP-MS dating allows fast and direct analysis without any chemical treatment, solution analyses rely on the careful sampling and separation of U and Th from other matrix elements. For the solution-based analyses, we used three different sampling strategies: Cutting, hand-held drilling and micromilling. Cutting speleothem samples allows for fast and easy collection of relatively large subsamples, with the possibility of leaching prior to chemical separation to minimize potential contamination. In contrast, hand-held drilling and micromilling are more prone to potential contamination during sampling. However, thin growth layers can be targeted using a micromill with fully automated sampling procedures. Several flowstones have been sampled using these three sampling strategies, and the resulting ages are evaluated in terms of detrital contamination and the resulting age uncertainties. Furthermore, speleothem samples were dated using LA-MC-ICP-MS. This technique is limited to speleothem samples with relatively high U concentrations (ideally >0.5 µg/g), but allows for fast data acquisition and even better sampling control as the micromill. However, laser ablation dating allows not only the age determination of small-scale growth layers, but also a better understanding of the detrital distribution and open-system behaviour around the hiatus. By application of line scans across a hiatus, we identified highly variable (230Th/238U) and (234U/238U) activity ratios directly below and above a hiatus. These results are supported by trace element mapping across several growth stops. Around each growth stop, the spatial distribution of detrital elements (e.g., Th, Pb, Al) appears to be irregular and not necessarily parallel to the growth layers.

A new U-Th laboratory for dating speleothems

László Palcsu1, Marjan Temovski1, Danny Vargas1, Gabriella Ilona Kiss1, Gergely Surányi2

1Institute for Nuclear Research, Debrecen, Hungary; 2MTA-ELTE Geological, Geophysical and Space Sciences Research Group (Hungarian Academy of Sciences at Eötvös University), Budapest, Hungary

In 2019 a Neptune Plus multi-collector ICP-MS (Thermo Scientific) was installed at the Isotope Climatology and Environmental Research Centre (ICER), Institute for Nuclear Research, Debrecen (Hungary). The detector configuration allows us to use six and three Faraday collectors with 1011 and 1013 Ohm resistors, respectively, and a SEM (secondary electron multiplier) for sensitive analyses of tiny ion intensities. The instrument is equipped with two Aridus3 desolvating nebulizer systems. To avoid cross contamination, one nebulizer is used only for 230Th-234U dating of carbonates, while the other for different isotope ratios, like 87Sr/86Sr or δ234U. The sample preparation is performed in a Class 1000 clean laboratory using ion exchange chromatography. The sensitivity of the measurement system is usually 2000 V/ppm, which makes it possible to provide accurate ages even from 5‒10 ng of uranium. Our triple spike has been prepared from an IRMM-3636a double spike solution mixed with an in-house 229Th standard. This latter thorium spike was prepared from a NIST SRM 4328C. The 233U/229Th ratio of the spike was first determined gravimetrically and then improved with a series of analyses of an infinite age (in secular equilibrium) calcite sample (Wildermann-1). The internal calibration of the mass spectrometric measurement of uranium and thorium is performed with an in-house mixture of the

CRM-112A uranium standard and the IRMM-3636a double spike solution. Each uranium isotope (233U, 234U, 235U, 236U, 238U) is determined with multi-collection, when 234U is detected in a central position with either a Faraday-collector connected to a 1013 Ohm resistor amplification or a SEM. The mass discrimination is determined by the ratio of 238U/235U, which is assumed to be natural in speleothem samples. The intensity of the 233U spike is also determined with 1013 Ohm, providing accurate 235U/233U ratios even at lower intensities. Thorium isotopes are measured using the magnet peak jump method, when 229Th, 230Th and 232Th are determined in multi-collection (230Th with SEM, others with Faraday), and then after peak jump only 229Th with SEM. During a measurement run, first the uranium sample solution is analysed, and then, the magnet is set to 230Th (at central position), while the thorium ion beams are aligned only with the dispersion quadrupole lens without changing the cup configuration set for uranium. This approach allows us to determine precise isotope ratios of uranium and thorium. The poster will present the status report on establishing U-Th dating of speleothems at ICER, with preliminary results from our measurements of already dated speleothems.

Acknowledgement: We thank Prof. Christoph Spötl for providing a piece of Wildermann-1 calcite. Prof. Hai Cheng is acknowledged for measuring a few stalagmite samples of us.

Ion Microprobe Method Development for High-Resolution Oxygen Isotope Analysis in Speleothems

Ian J. Orland1,2, Kouki Kitajima2, Cameron J. Batchelor2, Shaun A. Marcott2, R. Lawrence Edwards3,4, Yongjin Wang4, Hai Cheng5, John W. Valley2

1Wisconsin Geological and Natural History Survey, University of Wisconsin-Madison, United States of America; 2Department of Geoscience, University of Wisconsin-Madison, United States of America; 3Department of Earth and Env. Sciences, University of Minnesota-Twin Cities, United States of America; 4Department of Geography, Nanjing Normal University, China; 5Institute of Global Environmental Change, Xi’an Jiaotong University, China

Ion microprobe analysis of δ18O in speleothems began two decades ago, and initial studies targeted rapid climate change events. Despite early challenges with analytical accuracy, the smaller spot-size of ion microprobe analyses vs. conventional drill- and microdrill-sampling techniques proved effective at interrogating abrupt δ18O changes. Since then, methodology and sample preparation developments have increased the spatial resolution (to ≤10 µm), as well as the accuracy and precision (to as low as 0.2‰, 2 s.d.) of δ18O measurements by ion microprobe. When combined with micro-imaging of annually-laminated samples, these improvements can allow for analyses at sub-seasonal resolution and comparisons of intra-annual δ18O variability.

More recently, a methodological advance at the WiscSIMS lab in the University of Wisconsin-Madison pairs the open-source geospatial software QGIS to an automated sampling protocol on the ion microprobe. Instrumental settings were also optimized to shorten analysis time, including reduced durations of beam centering and ion counting, while maintaining high precision (0.3‰, 2 s.d.). The new approach not only results in a nearly 3x increase in sample analysis rate (from 10 spots/hr to 28 spots/hr), but also allows users to pre-select points for unattended analysis. As a result, it is possible to analyze long, high-resolution δ18O records that are precisely targeted using micro-imaging. Two examples include a 16 cm traverse of δ18O measurements placed once per annual band (>6,500 spots) along the growth axis of a Hulu Cave (China) stalagmite, and a 4.5 cm traverse of δ18O measurements (>850 spots) along the growth axis of an annually-banded Cave of the Mounds (USA) stalagmite.

For both of these stalagmite samples, QGIS also provided the digital framework for critical post-analysis steps. Reflected-light imaging was used as a datum for spatial registration – both samples had been cut into multiple chips for ion microprobe analysis – and to verify or edit final δ18O spot locations. Point data from other analyses (i.e. drilled U-Th and δ18O data, but potentially also including linear-traverse and raster proxy data from laser or imaging techniques) were imported and spatially registered, then projected onto a single linear traverse that was used to construct the age model. In each case, these steps were crucial for reconstructing reliable sub-annual-resolution δ18O records. Documentation and tutorials for these methods are publicly available via the WiscSIMS lab and QGIS website.

These examples demonstrate QGIS as an integrated platform for the analysis, post-processing, interpretation, and visualization of high-resolution and/or multi-proxy speleothem data for paleo-environmental studies. WiscSIMS has further leveraged this software to facilitate high-speed, precisely-targeted, unattended ion microprobe analysis of δ18O at sub-annual resolution in speleothems, thereby improving the accessibility, cost effectiveness, and potential scientific impact of this tool for the speleothem science community. Partially supported by U.S. NSF Grants 1805629, 1702816, 1702407, 1603065, 1658823 and the 111 project of China (D19002).

Validation of glacial-interglacial rainfall variations in southwest Sulawesi using Mg/Ca and δ18O in speleothems

Alena Kay Kimbrough1,2, Michael K Gagan1,2,3, Gavin B Dunbar4, Wahyoe S Hantoro5,1, Chuan-Chou Shen6,7, Hsun-Ming Hu6,7, Hai Cheng8,9, R. Lawrence Edwards9, Hamdi Rifai10, Bambang Suwargadi5

1School of Earth, Atmospheric and Life Sciences, University of Wollongong, Australia; 2Research School of Earth Sciences, The Australian National University, Australia; 3School of Earth and Environmental Sciences, The University of Queensland, Australia; 4Antarctic Research Centre, Victoria University of Wellington, New Zealand; 5Research Center for Geotechnology, Indonesian Institute of Sciences, Indonesia; 6High-Precision Mass Spectrometry and Environment Change Laboratory (HISPEC), National Taiwan University, Taiwan; 7Research Center for Future Earth, National Taiwan University, Taiwan; 8Institute of Global Environmental Change, Xi’an Jiatong University, China; 9Department of Earth Sciences, University of Minnesota, USA; 10Department of Physics, Universitas Negeri Padang, Indonesia

Speleothem δ18O is widely used as a proxy for rainfall amount in the tropics on glacial-interglacial to interannual scales. However, uncertainties in the interpretation of this renowned proxy makes producing robust paleoclimate and environmental records difficult. For this reason, a multiple proxy approach in speleothem research is becoming increasingly valuable. Here, we present paired measurements of Mg/Ca and δ18O for multiple stalagmites from Sulawesi, Indonesia. Collectively, the stalagmites span two glacial-interglacial transitions from 380 to 330 ky BP and 230 to 170 ky BP. The stalagmite Mg/Ca confirms that coupled measurements of δ18O record major shifts in tropical rainfall in southwest Sulawesi over glacial-interglacial transitions.

We find that minor element ratios (Mg/Ca, Sr/Ca) deposited in the Sulawesi stalagmites are primarily sourced from bedrock and subsequently altered by prior calcite precipitation. In Sulawesi, Mg/Ca in slower-growing stalagmites is more sensitive to prior calcite precipitation, a reliable proxy for changes in karst infiltration and local rainfall. Mg/Ca systematically lags the corresponding δ18O by 100s of years during glacial-interglacial transitions, indicating a delay in karst recharge. When paired, the Mg/Ca and δ18O corroborate the shifts from drier glacials to wetter interglacials in the core of the Australasian monsoon domain. Deviations between the two proxies may reveal concurrent changes in rainfall amount and moisture-transport pathways indicative of regional monsoon circulation.

WlCount: Software to detect and count laminae in stalagmites

Fabio Oriani1, Pauline Treble2,3, Andy Baker3,2, Gregoire Mariethoz1

1University of Lausanne, Switzerland; 2ANSTO, Australia; 3UNSW Sydney, Australia

Speleothem science is increasingly adopting annual laminae counting to produce chronologies for speleothem-based records of environmental change. Annual laminae counting offers a precise method for chronology construction or supplementation and can be based on seasonal variations in physical (e.g., speleothem fabric and particulate matter) or chemical (e.g., trace element concentrations and UV fluorescence) composition (Baker et al., 2021). The manual identification and count of laminae from images has become common practice although this can be time consuming, open to human error and carries larger uncertainty for blurry, irregular, or discontinuous laminar structures. We present a novel image-analysis approach to detect and count laminae in geoscientific imagery, called WlCount (Oriani et al., 2022). Based on Dynamic Time Warping and Wavelet analysis, WlCount first aligns persistent vertical features to increase the continuity of the laminar structure. Then, using a graphic interface, the user extracts the signal component which reflects the mean position and allows the automatic count of the laminae. WlCount is a potent and convenient tool to help geoscientists construct precise chronologies, reducing both human bias and the amount of time required to process these datasets. WlCount is open-source and freely available to the scientific community ( It can also be applied to any geoscientific studies requiring the quantitative analysis of laminated structures from an image (e.g., tree-rings, corals, and lake sediments) in addition to speleothems.


Baker, A., et. al. (2021), The Properties of Annually Laminated Stalagmites-A Global Synthesis. 59(2), e2020RG000722.

Oriani, F., et al. (2022), WlCount: Geological lamination detection and counting using an image analysis approach. Computers & Geosciences, 160, 105037.

Autogenic clastic sedimentation in iron formation caves

Ira D. Sasowsky1, Augusto S Auler2, John W. Senko1, Hazel A. Barton1

1University of Akron, United States of America; 2Instituto do Carste, Brazil

Autogenic clastic sediments are found in caves throughout the world, and primarily occur as breakdown (incasion) which forms from the collapse of cave ceiling and walls. Ranging in size from millimeters to tens of meters, these materials are easily differentiated from allogenic fluvial sediments by their high angularity and lithology. If the pieces are large, it may be apparent as to where they have detached from the cave wall, allowing their virtual reassembly as in a jigsaw puzzle. Breakdown is an ongoing process which occurs throughout the life of a cave. It has been generally accepted, though, that significant collapse occurs in many limestone caves during the interval in their genesis when they make the transition from phreatic to vadose conditions. This is related to the loss of buoyant support which occurs from the transition. As such, the emplacement of breakdown may be a marker for significant paleohydrologic events. Minimum event ages may be assigned by dating drip stone that has accumulated on top of the materials. Breakdown is also known to be provoked by arenization and cryogenic processes. In the present paper, an unusual mode of breakdown formation is described.

In several regions of Brazil extensive banded iron formations (BIF) constitute a durable and unique landscape. The rocks, which are composed primarily of silica and hematite, are of >2.5 Ga age and have formed a resistant crust called canga which drapes the upper parts of the landscape. Over 3,000 of such caves, mostly <100 m length, occur in this rock, frequently at the interface between BIF and canga. The morphology of the caves suggests beads-on-a-string, where semi-spherical rooms are connected by smaller passages. While initially suggestive of hypogene origin, we now know that these caves form by a unique mechanism whereby microbially facilitated weathering behind the cave walls leads to weakening and failure of the rock. This is an ongoing process that is still under investigation.

This behind-the-wall (sub muros) weathering presents a previously undescribed mode for generation of breakdown material. The floors of the cave are littered with un-transported irregular blocks of various sizes, as well as fine grained sediments. The finer material may represent disaggregated breakdown, or the sub muros. Transport processes are fairly limited because most of the caves do not have significant streams. This mechanism of breakdown generation is unlikely to be found in limestone landscapes due to differences in rock chemistry. The process is driven by water, and in the current climate ~110 cm of rain is received annually. Skylight entrances, in many cases the sole human-sized access point, attest that the breakdown process can progress to the land surface. What remains most poorly understood is how the material is then removed.

Black sediments in karst caves of the Eastern Alps, Austria

Barbara Gruber1,2, Lukas Plan2, Susanne Gier1, Jörn Peckmann3, Daniel Birgel3

1University of Vienna, Austria; 2Natural History Museum Vienna, Austria; 3University of Hamburg, Germany

In the Eastern Alps, black sediments in caves were already described in 1913 by Bock et al. and were first studied by Schauberger in 1957. He interpreted the thin black layers, which cover clays and limestones in Hirlatzhöhle (Dachstein), as soot and reported a similar occurrence in the nearby Dachstein-Mammuthöhle. Schauberger assumed that the black layers derive from wildfires. In the 1996, black crusts on cave walls were described by Menne from caves with active stream and were identified as microbial iron and manganese minerals.

To date, no comprehensive study has been carried out for the black coatings on cave sediments in the Eastern Alps. The aim of this study is to characterise 42 samples taken from 12 karst-caves in the Eastern Alps of Austria. Host rocks were (a) non-metamorphic Triassic limestones of the Northern Calcareous Alps, (b) weakly metamorphic Palaeozoic limestones north of Graz (Styria), and (c) Miocene limestones of the Vienna Basin. For comparison, one sample was taken in a limestone cave in the Swabian Jura (Germany).

Four types of black deposits can be distinguished:

• submillimetre thin, soot-like layers on top of clay and limestone deposits;

• millimetre to few centimetre-thick crusts on walls of caves with active stream, as well as palaeo-phreatic caves;

• submillimetre-thick crusts on boulders and flowstones;

• layers and lenses within clastic cave sediments or coatings of pebbles.

Analyses included macroscopic outcrop and thin section descriptions. In addition, mineral composition was determined using X-ray diffraction (XRD; Panalytical X’Pert PRO), and organic (TOC) and inorganic carbon (TIC) contents were analysed using a multiphase carbon and moisture detector (LECO RC 612). To constrain the potential presence of microorganisms, lipid biomarker analysis was performed on selected samples. The lipids were analysed with a gas chromatograph coupled to a mass spectrometer.

The soot-like layers show a TOC content of up to 18%. XRD allowed determination of minerals only from the clay layers below the black coatings. Detected biomarker did not include polyaromatic hydrocarbons like coronene, which are often used as evidence for wildfires, but included (bacterial) fatty acids and other bacterial compounds like hopanoids. Total Carbon content for the majority of black crusts is approximately 1 % or less with two exceptions reaching up to 9 %. Some crusts consist of iron oxyhydroxides like goethite and lepidocrocite as well as manganese oxides (todorokite and others).

Surprisingly, also hydroxyapatite was found in four samples, even though a significant contribution of bat excrement or mammal bones can be ruled out because of the remoteness of the locations. Most studied caves are several kilometres away from the entrance and several 100 m below the surface. Thin sections partly show a 3D dendritic appearance, but no clear visual evidence for a microbial formation.

Chemical analysis using ICP-OES of the bulk samples as well as extraction of oxides in oxalate and dithionite are ongoing and will hopefully give further insights into the diverse origins of black sediments in caves.

Cave evolution and valley incision in the Balkan Peninsula – new insight from U-Th and 10Be/26Al burial dating of cave deposits in the Crna Reka river basin (N. Macedonia)

Marjan Temovski1,2, Alexander Wieser3, Oscar Marchhart3, Mihály Braun1, Balázs Madarász4, Gabriella Ilona Kiss1, László Palcsu1, Zsófia Ruszkiczay-Rüdiger2

1Isotope Climatology and Environmental Research Centre, Institute for Nuclear Research, ELKH, Bem tér 18/c, 4026, Debrecen, Hungary; 2Institute for Geological and Geochemical Research, Research Centre for Astronomy and Earth Sciences, ELKH, Budaörsi út 45, 1112 Budapest, Hungary; 3University of Vienna, Faculty of Physics - Isotope Physics, Austria, Waehringer Strasse 17, 1090 Wien, Austria; 4Geographical Institute, Research Centre for Astronomy and Earth Sciences, ELKH; Budaörsi út 45, 1112 Budapest, Hungary

Information on Quaternary incision rate of rivers and related tectonic uplift is generally lacking in the Balkan Peninsula, especially its central parts, even though the region has experienced dynamic tectonic evolution since the Miocene, related to the Aegean extensional system. Caves offer a possibility to indirectly determine valley incision rates by dating cave deposits related to cave levels that correspond to past base levels. Recent research on cave morphology and sediments in the lower part of the Crna Reka river basin (southern part of N. Macedonia) identified cave development related to base level rise due to Pliocene-Early Pleistocene fill-up of basins and paleovalleys and establishment of lacustrine environments, as well as cave development related to base level lowering due to valley incision as a result of Pleistocene uplift and draining of the lacustrine systems.

As radiometric ages are available for a few caves only, deposits from selected caves, located in different sections of the Crna Reka river basin were sampled for U-Th dating of speleothems and burial age determination of clastic sediments using the cosmogenic radionuclide (CRN) pair of 26Al and 10Be. The latter method has been applied for the first time in this region. For CRN burial age determination, samples were collected from three caves, at two of which (Ristea Peštera, Temna Peštera – Dragožel) they represent clastic sediments infilling cave passages. At one location (Melnička Peštera), the cave-hosting bedrock, composed of quartz-containing carbonate breccia, was sampled. It represents an alluvial fan supposedly of Pliocene to Early Pleistocene age. Speleothem samples for U-Th dating were collected from four caves, at two of which (Temna Peštera – Dragožel, Liljarnikot) they are flowstones toping clastic sediment sections, while at the other two (Čulejca, Podot 2) they are mammillary calcite coatings, indicators of past karst water table. Preliminary U-Th results indicate Middle Pleistocene age (≤430ka) of the studied speleothems. The cave morphology and sediments together with new U-Th and CRN ages will be presented.

This research was supported by the NKFIH FK124807, Radiate Transnational Access 22002703-ST and GINOP-2.3.2-15-2016-00009 projects. We are indebted to Silke Merchel and Peter Steier for their contribution to CRN measurements.

Clastic sediments of the fossil-bearing upper member of Malapa cave in the Cradle of Humankind, South Africa

Dakalo Portia Maphanda1, Tebogo Vincent Makhubela2, Paloma de la Peña Alonso1, Zubair Jinnah1

1University of the Witwatersrand, South Africa; 2University of Johannesburg, South Africa

Clastic sediments deposited in caves are important paleoenvironmental and climatic records and
proxies. In the Cradle of Humankind (Cradle), a UNESCO World Heritage Site, clastic deposits
also play an important role in preserving hominin fossils, which necessitates the need to
understand those clastic deposits. In the Malapa cave, a site in the Cradle where Australopithecus sediba was discovered, the clastic sediments encasing the fossils have been previously studied to understand the depositional context. Recent excavations have exposed fossiliferous and barren clastic deposits, which we assign to the upper member because they seem to sit stratigraphically higher than the deposits in the two previously studied pits. In this study, we investigate the upper member deposits to understand their connection with the six stratigraphic facies (A-F) already established for the Malapa cave. We collected fourteen samples from the new deposits that either belonged to facies A-F or new facies not yet described. In addition, we collected one facies E and two facies F samples from the new deposits. The petrography of these samples was analyzed using optical microscopy and scanning electron microscope with energy dispersive spectroscopy (SEM-EDS) to evaluate the upper member stratigraphic facies and units. Preliminary findings show that the collected samples are different in terms of composition, grains, clasts, textures, and porosity and have led to the description of a new facies G. Facies E is defined by a dark-colored sample with an abundance of Fe-Mn coated chert grains, quartz, and peloids. Facies F in the upper member is defined by peloidal grainstone and sandy peloidal grainstone [2]. The facies F samples studied here are moderate to well-sorted and composed of peloid grains with interstitial drusy and granular sparry calcite cement and do not have calcite veins. The peloids are rounded brown to dark brown with grain size ranging from <500µm to 4mm, and Fe-Mn oxides are few (<1%). Facies G is defined by micrite groundmass, calcite veins, quartz and chert clasts, opaques minerals (Fe-Mn oxides), dolomite clasts, and the absence of peloids. Micrite occurs abundantly as the matrix or groundmass with intraclasts of chert, dolomite, quartz, Mn-oxides coated grains and sometimes calcite grains are present as framework grains. These samples are matrix-supported, poorly sorted, and have abundant coated grains. Some samples contain fractured clasts or grains while other samples have veins cutting through grains. The grain size varies significantly from a micrometer scale to >6mm. No fossil bones identified in the facies G. Facies G differs from facies A-D and F because it is poorly sorted, and lacks peloids and horizontal layering, but it is similar to facies E even though it lacks the abundant fossil content seen in facies E. According to [1] and [2], facies F sediments were waterlogged with evidence seen from peloids forming a dark brown muddy matrix and water damage on bone fossils [2]. Facies G depositional environment may be fluvial.

[1] Dirks et al., 2010, Science, 328(5975), 205-208
[2] Pickering et al., 2011, Science, 333(6048), 1421-1423.


Pavel Bella1,2, Michał Gradziński3, Helena Hercman4, Stanisław Leszczyński3, Wojciech Nemec5

1State Nature Conservancy, Slovak Caves Administration, 031 01 Liptovský Mikuláš, Slovak Republic; 2Department of Geography, Pedagogical Faculty, Catholic University in Ružomberok, 034 01 Ružomberok, Slovak Republic; 3Institute of Geological Sciences, Jagiellonian University, 30-387 Kraków, Poland; 4Institute of Geological Sciences, Polish Academy of Sciences, 00-818 Warszawa, Poland; 5Department of Earth Science, University of Bergen, 5007 Bergen, Norway

Cave sediments have recently attracted considerable research interest as an archive of the past climate changes, with a focus on carbonate speleothems. Relatively little sedimentological research has thus far been conducted on cave clastic deposits to reconstruct the hydrology of karstic tunnels, where the subterranean rivers are highly constrained by bedrock morphology and no longer free to operate in their ‘normal’ land-surface alluvial manner. This research issue is addressed by the present case study from the Demänová Cave System (DCS) of Slovakia.

The DCS is located in the north-trending Demänová Valley in the Low Tatra Mts of Slovakia and has a length in excess of 40 km. The system includes ten interconnected caves and is developed in the Anisian carbonate rocks of the Krížna Nappe that makes the northern sedimentary cover of the granitoid crystalline core of the Low Tatra Mts. Glacial tills indicate that the upper part of the Demänová Valley was glaciated at least twice during the Pleistocene. The DCS formed in a narrow canyon downstream of the glaciated part of the valley, due to the corrosion and erosion by water entering the karst system at the contact of the granitoids and carbonates. The DCS is a world-known case of a multi-level karst system. This study is from the lowest level, active as an underground branch of the modern Demänovka River.

The deposits studied are a relic compound sidebar of the underground Demänovka River, with a preserved thickness of ca. 4 m thick and a lateral extent of a few hundred metres. The deposits vary stratigraphically from pebble gravel to sand and mud, their many intervening flowstone layers and stalagmites have been dated by U-series isotopic method, and their stacking pattern has been recognized by the outcrop-scale mapping of depositional unconformities. Ten allostratigraphic units are distinguished, with a time span from >349 ka to the Holocene, and the tunnel morphodynamics is reconstructed in considerable detail.

Four units are gravelly and interpreted as a record of high-power flow of the underground river, genetically related to deglaciation phases in the upper part of the catchment. Two muddy units represent distinct episodes of flow ponding, attributed to (independently documented) downstream cave-roof collapses – probably caused by local post-orogenic earthquakes.

The study as a whole demonstrates that a detailed sedimentological analysis of the relics of cave-fill deposits can be an important contribution to an understanding of the hydrological history of karstic-cave fluvial tunnels.

The study was partly financed by the National Science Centre grants 2017/25/B/ST10/01430 and 2016/21/B/ST10/01483.

Glimpse on environmental conditions of the past from cave sediments from Dachstein (Austria)

Lukas Plan1, Franziska Holzer1,2, Knud Bethke2, Susanne Gier2, Denis Scholz3, Philipp Häuselmann4

1Karst and Cave Group, Natural History Museum, Vienna, 1010, Austria; 2Department of Geoology, Univ. of Vienna, Vienna, 1090, Austria; 3Johannes Gutenberg Univ., Mainz, 55128, Germany; 4Swiss Inst. for Speleology and Karst Studies, La Chaux-de-Fonds, 2301, Switzerland

The Dachstein is one of the large (580 km²) karst massifs in Austria that ranges from Lake Hallstatt at 507 m asl. up to the summit at 2995 m asl. It is dominated by Upper Triassic limestone. 750 caves with a cumulative length of 280 km have been documented so far. For this study, five major caves in the North were investigated: Hirlatzhöhle (113 km length); Dachstein-Mammuthöhle (67 km), Dachstein-Rieseneishöhle (3 km), Schönberghöhle (9 km), and Günter-Stummer-Höhle (1 km).

About 40 sediment profiles or outcrops between 855 and 1960 m asl. were analyzed with respect to stratigraphy, grain size, mineral composition (XRD), heavy minerals, and carbonate content. 26Al/10Be burial age dating was applied on quartz pebbles, and flowstones were dated with the 230Th/U disequilibrium method. Furthermore, cave morphology was investigated to reveal paleo-hydrologic conditions. Even though there are often mixtures of reworked sediments, our interpretation differentiates between four main types of clastic sediment:

(1) The oldest sediments are poorly sorted, often greenish, and can reach at least 2.5 m in thickness. They range from silt to very coarse gravel and are dominated by quartz and mica; the greenish color derives from chlorite. The origin is the Eocene to early Miocene Augenstein Formation. As the flow velocity for transporting the gravel corresponds to the one indicated by scallops, we assume that it was deposited during an initial phase of speleogenesis. Layers of moderately to well sorted quartz gravel or greenish sands are interpreted as a fractionation of this sediment. Dating of quartz rich samples gave burial ages older than 3.5 Ma. So far, this sediment was only found in the Giant Cave level (between 1200 to 1500 m a.s.l.).

(2) Subsequently, a dark brown fine-grained sediment without layering was deposited. It can be classified as slack-water facies and is attributed to a paleosoil that derived from the Augenstein Formation. It was found at all levels, in the most elevated one at 1960 m a.s.l. it contains boehmite and anatase. These minerals suggest a formation under (sub)tropical conditions and could hint to a formation during the Mid-Miocene Climatic Optimum (around 15 Ma) or even before.

(3) Only at some places, a coarse sediment or conglomerate with rounded limestone components occurs. Grain size ranges from sand to pebbles. In two outcrops, layering indicates fluvial transport, but at one site, which is close to the surface, it seems that gravels and boulders (up to 0.8 m in diameter) were pushed in by glacier ice.

(4) The youngest sediment is a bright, finely layered silty clay. In general, the thicknesses of at least 6 m in the low-lying caves decreases with altitude. It is almost omnipresent up to around 1700 m a.s.l. It is very likely that the layering represents varves. The carbonate content is generally high (up to 88 %). It derived from glacial erosion, and the slack-water facies is due to glacial backflooding. Deposition during the Pleistocene is supported by 230Th/U-dating of flowstone layers that grew between 5 ka and 0.5 Ma.

La Galiana multilevel cave system (Soria, N Iberia): a multi-chronological approach and implications for landscape evolution since the Pliocene

Miguel Bartolomé1, Belén Oliva-Urcia2, Toshiyuki Fujioka3, Alicia Medialdea3, Alfonso Dodero4, Gerardo Benito5, Mikel Calle6, Juan José Villalaín7, Hai Cheng8, R.L Edwards9, Carlos Sancho4, Ana Moreno1

1Instituto Pirenaico de Ecología, Spain; 2Universidad Autónoma de Madrid, Spain; 3National Research Centre on Human Evolution, Spain; 4University of Zaragoza, Spain; 5Museo Nacional de Ciencias Naturales, Spain; 6University of Turku, Finland; 7University of Burgos, Spain; 8Xi'an Jiaotong University, China; 9University of Minnesota, USA

Karstic systems with different topographic levels are of high interest to reconstruct landscape evolution. Each level marks the position of the phreatic surface that matches the water discharge level of the karstic system. La Galiana system (Lobos river canyon, Iberian Range, Duero basin, Spain) is composed of three staircase horizontal galleries (La Galiana Alta, La Galiana Baja and the Cueva del Lago) and Ucero spring located at 1099, 966, 961 and 953 m a.s.l. respectively. These three cave levels represent distinct stages of stabilized water-table linked to local fluvial base levels. Caves develop in a monoclinal series of Cretaceous limestones, which are cut by Astaracian erosion surface, whose exhumation took place at the end of Turolian period (9-5.3 Ma). An OSL (Optically Stimulated Luminescence) age from Cueva del Lago indicates the cave formation occurred before 180 ka. U-Th ages derived from La Galiana Baja speleothems indicate that the formation of that level was older than 224±15 ka (MIS 7). On the other hand, in La Galiana Alta, a large flowstone (240 cm in thickness) dated beyond the U-Th limit (>600 ka). This flowstone, overlies a detrital sequence formed by 120 cm of gravels and sands with cross stratification. Both deposits form a basal sequence that fill a vadose cave canyon, indicative of incision after cave formation. The burial age of detrital sequence based on TCND (Terrestrial Cosmogenic Nuclide Dating) gave an age ranging from 3.5-4.5 Ma. Paleomagnetic measurements carried out in the basal sequence in La Galiana Alta reveal a dominant reverse polarity, thus indicating that those deposits were older than 780 ka, although some samples show normal polarity. Cave formation requires high incision rates, and the basal sequence formed after vadose canyon development in the cave post-dating the Astaracian surface erosion formation (~11.6 Ma) and exhumation (9-5.3 Ma). The geomorphological context, radiometric and palaeomagnetic data allow placing the flowstone formation in La Galiana Alta during the Gilbert Chron (6-3.5 Ma). The flowstone in La Galiana Alta, could be related to warm and probably humid periods during the Pliocene (e.g mid-Piacenzian Warm Period, Pliocene Climate Optimum). Altitude differences between galleries (133 and 13 m, respectively) and the geochronology results provide an estimation of the maximum fluvial incision rates between 39 and 58 mm/ka, similar to incision rates observed in other studies based on fluvial terraces in the Duero basin. This work can shed light on the endorheic-exoreic transition timing in the Duero basin.

Late Holocene Vegetation dynamics and anthropogenic impact inferred from cave sediments and Bat guano in Nigeria

Kenechukwu Daniel1, Emuobosa Akpo Orijemie2

1University of Nigeria, Nsukka, Nigeria; 2Department of Archaeology and Anthropology, University of Ibadan, Nigeria

Palynological analyses of surface sediments and stratified bat guano recovered from Onualor-Ogadike cave, southeastern Nigeria were conducted. The results were used to infer the vegetation history and decipher human impact on the regional late Holocene environment in the study area. The surface samples revealed the occurrence of the pollen of rainforest and secondary forest taxa which reflected the present-day vegetation of the Onualor-Ogadike cave. The pollen assemblages from the five bat guano layers in an archaeological context within the cave yielded evidence of alternating palaeoenvironmental conditions. Lowland rainforest and humid conditions (layers 5 and 3) and open vegetation and dry conditions (layers 4, 2-1) were recognised. Layer 4, which was dominated by the pollen of secondary (open) forests and Guinea savanna elements yielded at date of 4650 ±30 BP. Layer 4 was succeeded by layer 3, the constituent vegetation of which included several economic plants such as Elaeis guineensis (oil palm), Dacryodes edulis (African Plum), and Blighia sapida (Ackee apple tree) and Parinari cf. curatellifolia (Mobola Plum). In contrast, the periods of inferred dry conditions (layers 2-1) had sparse evidence of human presence, an indication that the periods might have been too harsh for continuous human occupation. It has been demonstrated that bat guano, as has been found for other tropical areas, contain properties for the preservation of pollen and spores, the evidence of which could be harnessed for ecological anthropology studies in the tropical areas of Africa.

Reconstructing extreme rainfall events from speleothem flood records

Marc Luetscher, Pierre-Yves Jeannin, Manon Trottet, Eric Weber

SISKA, Switzerland

Speleothems have emerged as a promising archive for the reconstruction of paleofloods on short and long time-scales (Denniston and Luetscher, 2017). Here we investigate to which extent singular flood layers from Milandre stalagmites may provide quantitative interpretations of past hydrological recharge. The karst system, NW Switzerland drains a c. 13 km2 hydrological catchment comprising forests, grassland and cultivated areas. Meteoric precipitation averages 1070 mm·a-1 for an effective recharge estimated at 520 mm·a-1. The total discharge at the karst springs ranges between 30 and 3000 ls-1 with peak flow being measured during the winter half-year. The speleological network comprises c. 10.5 km of cave passages including a 4 km long perennial cave stream fed by two main tributaries which contribute to 25 and 35% of the total discharge, respectively. During heavy rainfall events, backflooding raises the karst water table by c. 30 m, remobilizing large amounts of silt which redeposits on the top of speleothem formations during flow recession.

Based on a detailed cave survey supported by LIDAR measurements and photogrametric imaging, we model the hydraulic response of the stream using SWMM 5 which has proved to be well suited to simulate turbulent flows in variably saturated conduits. Model results are then calibrated with cave monitoring data including hydraulic head and discharge measured at 10 min intervals. Preliminary results support fast rising water levels when the total discharge exceeds 2.2 m3s-1 and provide a head-discharge relation that can be used to interpret stalagmite flood layers at different elevations. Two stalagmites sampled at 5 m interval in elevation reveal distinct flood layers which can be attributed to millennial-scale floods with simulated discharges exceeding instrumental records by up to 10%.


Denniston R., Luetscher M., 2017. Speleothems as high-resolution paleoflood archives. Invited by Quaternary Science Reviews, 170, 1-13.

Sedimentological Analysis of varve-like “Bright” Cave Loam from Caves in the Northern Rim of the Dachstein Massif (Austria)

Franziska Holzer1,2, Susanne Gier1, Lukas Plan2

1Department of Geology, University of Vienna, Vienna, 1090, Austria; 2Karst and Cave Group, Natural History Museum, Vienna, 1010, Austria

The Dachstein Massif, with an extent of 580 km2, is one of the largest karst massifs in Austria. In the caves developed there are often depositions of clastic sediments. A beige and brown, laminated, fine-grained sediment is present in caves of the Dachstein Massif below an altitude about 1,700 m a.s.l. This “bright” cave loam represents the topmost layer in a number of caves of the area and has a maximum thickness of 6 m. The thickness of the layer decreases with an increasing sea level.

There is a theory that the mm to cm laminations of the bright cave loam are glacio-lacustrine varves caused by seasonal or longer climatic rhythms. To further investigate this theory seven samples were taken from two different caves: Hirlatzhöhle and Dachstein-Mammuthöhle.

Sedimentological methods used were grain size analysis and X-ray diffraction (XRD) for the bulk- and clay mineralogy, which was the main focus of this work. Wherever possible, the laminae of the bright cave loam were analysed separately. However, this could not be done with the samples from Mammuthöhle, as the lamination is not as clearly developed as in Hirlatzhöhle.

The results of the bulk mineralogy and the semiquantitative analysis show that darker laminae mainly consist of calcite (32 % – 75 %) with secondary amounts of quartz (2 % – 5 %), dolomite, ankerite, muscovite, and chlorite. K-feldspar and goethite occur in traces only. Overall, the bright laminae have the same mineralogical constituents, but show slightly higher amounts of calcite and quartz.

Bulk clay mineralogy shows a layer pattern for clay minerals except smectite. Illite is the most abundant clay mineral with 50 – 76 % and is together with kaolinite and chlorite slightly enriched in the darker laminae.

The dark and bright laminae can also be distinguished by using grain size analysis. The dark laminae have a clearly higher percentage of clay-sized particles. The maximum grain size of both lamina types is coarse silt. Components up to medium silt are very well sorted, which can be seen by the steep ascent of the cumulative grain size curve that flattens rapidly afterwards.

In conclusion, the bright cave loam was probably formed under conditions of slack-water facies and can be seen as a result of glacial backflooding in Pleistocene. This also explains the altitude control of the bright cave loam. The studied sediments indicate that the depositions are very likely to be varves: The lightly-coloured, carbonate-rich, and coarser-grained laminae originate from more energetic meltwater during spring and summer; in winter there is lower discharge and the remaining clay minerals are sedimented and result in the thin, darker laminae. However, the cause for the rhythmicity of the varves can not be established.

Tracing the source of the clastic sediments using rockmagnetism in the Urșilor Cave (Apuseni Mountains, Romania)

Cristian George Panaiotu1, Cristian Necula1, Relu Dumitru Roban2, Alexandru Petculescu3, Ionut Cornel Mirea3, Luchiana Faur3, Silviu Constantin3

1Faculty of Physics, University of Bucharest, Magurele, 077125, Romania; 2Faculty of Geology and Geophysics, University of Bucharest, Bucharest, 010041, Romania; 3“Emil Racovița” Institute of Speleology, Bucharest, 010986, Romania

Several studies have shown that the magnetic susceptibility data of cave sediments can be used to reconstruct paleoenvironments outside the cave at the time of deposition. These magnetic susceptibility variations are attributed to soils with climate-dependent magnetic properties that are washed, blown, or tracked into the cave where they accumulate, creating the changes observed in rock magnetic data. To test this hypothesis, we did a rockmagnetic study of the sediments from the Urșilor cave (Apuseni Mountains, Romania) and the soils above the cave. Our main objective was the detailed characterization of the ferromagnetic mineralogy preserved in the cave sediments and its links with potential soil sources. In the cave, we selected for sampling four sections (2-3 m high) belonging to interior cave facies, consisting of fine sediments, with some sand layers. According to the optically stimulated luminescence ages, the deposition of sediments started during marine isotope stage 4. Around the cave, we sampled both soils developed on Permian rocks and Jurassic limestones. We measured for all samples: variation of magnetic susceptibility with frequency (976 and 15616 Hz), isothermal remanent magnetization (IRM), anhysteretic remanent magnetization, and the anisotropy of magnetic susceptibility. On selected soils and cave sediments samples, we also measured the frequency dependence of magnetic susceptibility at 13 frequencies (between 128 and 512000 Hz), low and high temperature magnetic properties, stepwise acquisition of isothermal remanent magnetization (up to 8.5 T) with AF demagnetization at 200 mT after each step to highlight the contribution of high coercivity minerals. IRM acquisition before and after AF demagnetization shows the presence of low coercivity minerals (magnetite and/or maghemite) and high coercivity minerals (mainly hematite and goethite). The ferromagnetic mineralogy is similar in soils and sediments. Present-day soils have various degrees of frequency dependence, but most soils have high magnetic susceptibility with a significant frequency dependence. Cave sediments show a frequency dependence similar to that of present-day soils. Our rockmagnetic measurements indicate that the clastic cave sediments have magnetic properties that are mainly controlled by the magnetic properties of soils transported into the cave by erosion. Large variations in the magnetic susceptibility of cave sediments probably reflect major climatic oscillations outside the cave. However, due to a lack of dating precision, the correlation with specific isotope stages must be regarded with caution. Acknowledgment: The research leading to these results has received funding from the EEA Grants 2014-2021, under Project contract no. EEA-RO-NO-2018-0126.

Tropical Rainfall Response to Millennial and Orbital Scale Forcing: A 70ka Multi-Proxy Speleothem Record from the Eastern Amazon Basin

Dylan Parmenter1, Emma Brown1, Francisco W. da Cruz Jr2, Hai Cheng3, Xianfeng Wang4, Augusto S. Auler5, Xinyuan Zheng1, R. Lawrence Edwards1

1University of Minnesota, United States of America; 2University of São Paulo, São Paulo, Brazil; 3Xi’an Jiaotong University, Xi’an, China; 4Nanyang Technological University, Singapore; 5Brazilian Karst Research Institute, Belo Horizonte, Brazil

The Amazon Basin is an ideal location for testing the Held and Soden hypothesis1, which holds that wet regions should get wetter in response to global warming. In part to test this idea, we have extended the existing cave record from Paraiso Cave2 (4.07oS, 55.45oW) beyond 45 ka. The full record now covers 70 ka, allowing us to investigate 9 additional millennial scale events, as well as events that correlate with the MIS 4/3 transition. This record fills in a critical gap in the interpretation of South American hydroclimate records, as the eastern Amazon Basin is widely regarded as the gateway for moisture into the continent. The extended oxygen isotope record has δ18O values ranging between -3 and -6‰ and for overlapping sections replicates the old record. The most prominent events in the extended record are a pluvial anomaly between 47.0 and 48.9 ka, correlative with Heinrich Stadial 5, and a negative shift accompanying the MIS 4/3 transition between 65 to 60 ka. Our stable isotope findings record a previously observed2 millennial scale anti-phased relationship between Amazon rainfall on one hand, and Asian Monsoon strength and Greenland temperature on the other. On the orbital scale, we see correlative shifts between Amazon rainfall and atmospheric CO2, consistent with the Held and Soden idea. New trace metal data gives us a secondary proxy for local hydroclimate conditions to pair with the full 70ka δ18O record. The strong co-variation between Mg/Ca, Sr/Ca, and Ba/Ca ratios in PAR samples makes them ideal3 for indicating prior calcite precipitation resulting from aridity. The trace metal data strongly support our interpretation of the δ18O record, indicating that local moisture availability varies in concert with overall rainfall patterns.

1Held, I. M., & Soden, B. J. (2006). Journal of Climate, 19, 5686–5699.

2Wang, X. F. et al. (2017). Nature, 541, 204–207.

3Fairchild, I. J. et al. (2000). Chemical Geology, 166(3), 255–269.