Climatic variations during the Holocene inferred from radiocarbon and stable carbon isotopes in speleothems from a high-alpine cave

Abstract. Rapid and continuous analysis of radiocarbon (14C) concentration in
carbonate samples at spatial resolution down to 100 µm has been made
possible with the new LA-AMS (laser ablation accelerator mass spectrometry)
technique. This novel approach can provide radiocarbon data at a spatial
resolution similar to that of stable carbon (C) isotope measurements by
isotope ratio mass spectrometry of micromilled samples and, thus, can help
to interpret δ13C signatures, which otherwise are difficult to
understand due to numerous processes contributing to changes in the C-isotope
ratio. In this work, we analyzed δ13C and 14C on the
Holocene stalagmite SPA 127 from the high-alpine Spannagel Cave (Austria).
Both proxies respond in a complex manner to climate variability. Combined
stable carbon and radiocarbon profiles allow three growth
periods characterized by different δ13C signatures to be identified: (i) the
period 8.5 to 8.0 ka is characterized by relatively low δ13C
values with small variability combined with a comparably high radiocarbon
reservoir effect (expressed as dead carbon fraction, dcf) of around 60 %.
This points towards C contributions of host rock dissolution and/or from an
“old” organic matter (OM) reservoir in the karst potentially mobilized due
to the warm climatic conditions of the early Holocene. (ii) Between 8 and
3.8 ka there was a strong variability in δ13C with values ranging from
−8 ‰ to +1 ‰ and a generally lower dcf. The δ13C variability is most likely caused by changes in C exchange between
cave air CO2 and dissolved inorganic carbon in drip water in the cave,
which are induced by reduced drip rates as derived from reduced stalagmite
growth rates. Additionally, the lower dcf indicates that the OM reservoir
contributed less to stalagmite growth in this period possibly as a result of
reduced meteoric precipitation or because it was exhausted. (iii) In the
youngest section between 3.8 and 2.4 ka, comparably stable and low
δ13C values, combined with an increasing dcf reaching up to
50 % again, hint towards a contribution of an aged OM reservoir in the
karst. This study reveals the potential of combining high-resolution
14C profiles in speleothems with δ13C records in order to
disentangle climate-related C dynamics in karst systems.



S1. Anomalies in 12 C and F 14 C during LA-AMS
An unexpected signal was observed during radiocarbon analysis by LA-AMS in the bottom piece of stalagmite SPA 127 (Fig. S1). The signal intensity (top panel of Fig. S2) is expressed as the 12 Ccurrent recorded by the AMS. A typical signal evolution during the course of a measurement is a fast rise in the 12 C --current to about 5 µA after the laser is started, followed by a slower and steady increase to about 6 -8 µA (compare also Welte et al. (2016)). After 10 -15 minutes, the maximum lifetime of the sputter target in the ion source is reached as its CO2 outlet is blocked from deposits resulting in a decrease in 12 C current. A new sub-scan has to be started using a fresh sputter target. In the bottom section of SPA 127, five peaks were observed in the 12 C --current with corresponding F 14 C dips from approximately 0.2 down to 0.05. The repeated scan (B2, see main text), which was performed after removal of the top surface layer, showed a similar behavior (see Figure S3). We additionally investigated the regions showing these peaks using FTIR. These anomalies are less distinctly observed in the third LA-AMS scan ( Figures S2 and S3).

Interpretation of the anomlies
The five 12 C-current peaks correlating with strongly depleted F 14 C observed between 120 and 145 mm depth, i.e. from 8.4 to 8.0 ka BP, indicate that these layers are composed of a different material than the bulk. The higher 12 C-currents are associated with a matrix that converts more readily to CO/CO2 upon LA compared to CaCO3, a behavior that is known from organic substances with a higher oxygen/carbon stoichiometric ratio (Frick and Gunther, 2012). In order to ascertain whether the substance in these layers is inherent to the stalagmite or a contamination, its composition has been determined using FTIR. These measurements revealed that the anomalies were caused by epoxy resin ( Fig. S4 and S5).
Indeed, the stalagmite was glued in this section when it broke into two parts shortly after its removal from the cave in 2002 CE (Fig. S6). Although the speleothem was only glued on one location, we observed the glue at least at five positions coinciding with small cracks. This suggests that the glue was able to soak through these small cracks. Hence, attention should be paid when working with glued speleothems, especially, if the type of geochemical analyses is based on methods not easily able to differentiate the analyzed material such as e.g. laser ablation IRMS (Spötl and Mattey, 2006), while e.g. IRMS based on drilled material, which is acidified by H 3PO4, should be unaffected.
These findings underline that unlike conventional analytical methods applied to stalagmite samples for 14 C analysis, which provide exclusively the isotope composition of the CaCO3, LA-AMS additionally yields the 14 C content of OM captured in stalagmites. Despite the fact that this offers novel possibilities it also requires particular caution to distinguish between inherent OM and contaminants of organic origin. 14 C data used for the interpretation in this study stem from scan B3, which was largely unaffected by the epoxy because the scan was placed off the glued joint as confirmed by conventional 14 C analysis (see Fig. A4).

S2. Fourier Transform Infrared spectroscopy (FTIR) analysis
Fourier Transform Infrared spectroscopy (FTIR) was used to determine the specific composition of selected areas in our sample to clarify the causes of anomalies. FTIR is a standard non-invasive technique for material analysis (Derrick et al., 2000). The coupling of the IR spectrometer with a microscope enables micro-analysis, while the development of focal plane array (FPA) detectors allows to perform imaging instead of single point measurements. Attenuated total reflection (ATR) is a technique that requires no sample preparation other than a flat surface and is independent of the thickness of the sample as it is performed upon contact of the sample with the ATR crystal, which is a medium of high refractive index. The analysis of the speleothem was performed on a Spotlight 400 FT-IR Imaging System (Perkin Elmer, Massachusetts, USA), equipped with a MCT array (mercury cadmium telluride) detector. The stalagmite was placed under the microscope on a motorized stage and focused to measure the spectrum between 4000 and 520 cm -1 using a micro-ATR objective germanium crystal at 4 cm −1 spectral resolution and 32 scans. The detection area of a measurement was 300 x 200 μm 2 in diameter (spot size). An area of 1.9 x 3.4 mm in the vicinity of the abnormal 14 C/ 12 C behavior observed by LA-AMS was selected for rastering by FTIR.
In the regions where LA-AMS anomalies were observed, two different types of matrices were revealed by the FTIR spectra (Fig. S5), representing calcite and epoxy resin. The epoxy resin spectra were found for measurement points along a crack present in this area. The calcite matrix was found in all other measurement points outside of the crack.   Wavenumber (cm -1 ) %T %T Figure S5 Top: FTIR spectrum representative of measurements outside the crack (such as the pink circle in Figure 1 of SI) showing a calcite matrix. Bottom: FTIR spectrum representative of measurements within the crack (such as green circle in Figure 1 of SI) revealing epoxy in the matrix. S4 LA-AMS results (raw data) 14 C results are shown in Fig. S8A, where error bars are not displayed for reasons of clarity. The measured F 14 C ranges of both pieces of SPA 127 (T1, T2 and B3) range from about 0.55 at the top to approximately 0.15 towards the oldest part of the stalagmite. Between 0 and ca. 40 mm, the F 14 C varies around a comparably constant value of 0.45, followed by a pronounced dip between 40 mm and 70 mm. From 70 mm to 140 mm, F 14 C decreases from 0.4 to 0.2. The measurement precision is approximately 8% for the younger slab and 10% for the older piece with a spatial resolution of 280 µm per data point. In Fig. S8B, F 14 C data is depicted in grey with uncertainties corresponding to the analytical error. In order to reduce noise a SG filter has been applied with an interval length of 21 points and the polynomial order of 2. Stable C and O data are shown in Fig. S8 C.  Figure S9: Comparison of LA-AMS data with conventional AMS data using a gas ion source. In short, 1 -2 mg CaCO3 samples were analyzed following the procedure described in Wacker et al. (2013). The red lines indicate two locations that were identified as potentially contaminated by epoxy resin (see caption of Fig. 3 in SI). However, results of both methods are in good agreement, especially at a dft of 133 mm (first red line). At 148 mm, epoxy contamination cannot be ruled out as no conventional data point was sampled there. However, the generally high agreement, also in the oldest part of SPA 127, suggests that the LA-AMS data are robust. Figure S10: 11-point running correlation coefficients calculated for the dcf versus δ 13 C (top) and for δ 18 O versus δ 13 C (bottom).