Understanding the deglacial relationship between carbon isotopes and temperature in stalagmites from Western Europe
- 1Department of Earth Sciences, University of Oxford, South Parks Road, Oxford OX1 3AN, UK (franziska.lechleitner@earth.ox.ac.uk)
- 2Swiss Federal Institute for Forest, Snow and Landscape Research, Zürcherstrasse 111 8903 Birmensdorf, Switzerland
- 3Laboratory for Ion Beam Physics, ETH Zürich, Otto-Stern-Weg 5, 8093 Zürich, Switzerland
- 4Department of Earth Sciences, ETH Zürich, Sonneggstrasse 5, 8092 Zürich, Switzerland
The last deglaciation was a period of rapid and profound climatic change in Western Europe. Speleothem carbon isotope (δ13C) records from mid-latitude Western Europe have consistently shown large and reproducible excursions over this time period, strikingly similar to available temperature reconstructions from other archives. The mechanism behind the temperature sensitivity of speleothem δ13C, however, remains poorly constrained, due to the complex interplay of multiple processes affecting this proxy.
Here we use a multi-proxy approach and forward modelling of karst processes to investigate what drives the response of speleothem δ13C to the last deglaciation in Western Europe. We present new proxy data (14C and δ44Ca) from speleothem Candela from El Pindal Cave, northern Spain, which covers the period from the Last Glacial Maximum (25 ka BP) to the Early Holocene (8 ka BP). Previously published stable isotope data (Moreno et al., 2010) revealed a pronounced decrease in δ13C over the deglaciation (~8‰ VPDB) which closely tracks regional temperature records from the Iberian Margin. We make use of the different sensitivities of ancillary proxies (14C, Mg/Ca, and δ44Ca) to processes in soil and karst to quantify their relative importance on the δ13C shift. For this, we use the forward modelling software CaveCalc (Owen et al., 2018) to generate a large ensemble of possible solutions, from which the ones closest matching the data are chosen and evaluated.
Our preliminary results suggest that in-cave and karst processes (carbonate host rock dissolution and reprecipitation) cannot explain the full amplitude of the δ13C shift over the deglaciation, and that changes in soil δ13C are to some extent translated to the speleothem carbonate δ13C. The possibility of quantitatively disentangling processes in the soil from other karst processes could allow the reconstruction of past soil activity from speleothems.
References:
Moreno, A., Stoll, H., Jiménez-Sánchez, M., Cacho, I., Valero-Garcés, B., Ito, E., Edwards, R.L., 2010. A speleothem record of glacial (25-11.6 kyr BP) rapid climatic changes from northern Iberian Peninsula. Glob. Planet. Change 71, 218–231. doi:10.1016/j.gloplacha.2009.10.002
Owen, R.A., Day, C.C., Henderson, G.M., 2018. CaveCalc: A new model for speleothem chemistry & isotopes. Comput. Geosci. doi:10.1016/J.CAGEO.2018.06.011
How to cite: Lechleitner, F. A., Day, C. C., Wilhelm, M., Haghipour, N., Kost, O., Henderson, G. M., and Stoll, H. M.: Understanding the deglacial relationship between carbon isotopes and temperature in stalagmites from Western Europe, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-7466, https://doi.org/10.5194/egusphere-egu2020-7466, 2020.
Comments on the display
AC: Author Comment | CC: Community Comment | Report abuse
I really appreciated reading your display. Wonderful work.
One very small comment / question that occurred to me. Would an alternative /additional final interpretation include a change between just soil CO2 and both soil + vadose zone CO2. Its the same vegetation change betwee grass and forest.
Dear Andy, thanks for your supportive comment.
A change in the isotopic composition of the initial gas through a change in the relative importance of different CO2 sources (atmosphere, soil, vadose zone) with different d13C values could lead to the observed shift in d13C. We are going to test this in our dataset using a mixing line approach and we also have some monitoring data that we hope will help us.
Does this answer your question?
It does :)
I like your contribution a lot. Thank you for sharing.
There are already a lot of information given, with respect to the methods. But I am not sure if I understand everything correctly. So I list here a few points/questions, which answers will help me to understand the methods even better.
Page 4
How did you prescribe soil turnover rate?
What do you mean with not changing "Dissolution/reprecipitation"? In contrast to that you say, that you changed the dissolution regime. And below you showed some examples where you accounted for PCP, which I would interpret as 'reprecipitation'.
What exactly do you mean by 'atmospheric exchange'?
Page 5:
d44Ca-PCP plot: I don't understand how you obtain the f_Ca values for the d44Ca and Mg/Ca values. Please explain.
d44Ca-PCP plot: Would this be a good method to determine the host rock d44Ca? Or expressed in other words: How does the d44Ca of the initial solution (f_Ca = 1) compare with the d44Ca of the host rock above your cave?
Page 6:
Why are the DCF errors so large? They are between 5 and 10 %. How good is the age-depth model?
Dear Jens,
thank you for your supportive comment and your questions.
Soil turnover rate: We prescribed this by changing the F14C value of the soil gas (between 100-80 pMC) to simulate whether a significant old carbon pool in the soil would have an effect in the DCF. The range is relatively large and will need to be adjusted to more realistic parameters though.
Not changing "Dissolution/reprecipitation": Apologies, I think this might be formulated in a misleading way in the figure. What we mean is that the degassing/precipitation mode and the calcite supersaturation limit were not allowed to vary for the individual solutions. Degassing/precipitation was set to "kinetic" (as defined in PHREEQC) and the simulations were run in a multiple-step degassing mode, where a loop of degassing-precipitation steps is followed until the solution is not supersaturated anymore.
Atmospheric exchange: This was defined as a fraction of the initial gas that is composed of atmospheric air, so that the initial gas is a mixture between soil gas and atmosphere.
d44Ca-PCP: The f_ca values were obtained by comparing the measured stalagmite value to the d44Ca value of the overlying bedrock and using the published fractionation factor by Owen et al. (2016, EPSL).
d44Ca-PCP plot: We have measured the d44Ca of the bedrock, therefore there is no need to calculate it. I'm not sure it would be possible to estimate the bedrock d44Ca from the stalagmite d44Ca values alone, as it is not possible without the former to estimate the degree of PCP the stalagmite carbonate has experienced.
DCF error: The age model is not very well constrained, especially for the older part. We will double check the error however, as it might be overestimated at the moment. Thank you for pointing this out.