Geochemical Insights Into an Active Calcareous Tufa Depositing System in Southern Germany

Available online at www.sciencedirect.com ScienceDirect

Procedia Earth and Planetary Science 17 ( 2017 ) 328 – 331

15th Water-Rock Interaction International Symposium, WRI-15 Geochemical insights into an active calcareous tufa depositing system in southern Germany

Simon M. Ritter1,a, Margot Isenbeck- Schrötera, Andrea Schröder-Ritzraub, Christian Scholza and Norbert Frankb

aInstitute of Earth Sciences, Heidelberg University bInstitute of Environmental Physics, Heidelberg University

Abstract

Calcareous tufa deposition in bicarbonate-rich karstic spring waters is essentially linked to complex microbiological communities that alter the geochemistry of the deposited tufa and hydrochemistry of the creek water. A monthly monitoring of the creek water chemistry of a natural tufa depositing site in the Franconian Alb in southern Germany reflects the tufa formation well, which is expressed in a decrease of calcium and bicarbonate concentrations and SIcalcite oversaturation. The accompanying decrease of barium ions is most likely due to chemoselective chelation by Extracellular Polymeric Substances (EPS) of the tufa biofilms that favour divalent ions with low charge densities in the order of Ba>Sr>Mg. Tufa geochemistry is altered with respect to increasing Mg/Ca ratios downstream reflecting incresing Mg/Ca ratios downstream of the creek water due to continouing low-Mg-Calcite precipitation. Especially the decrease of barium holds the potential to monitor seasonal variabilities of the relative portion of biofilm influence on tufa formation, which is important regarding the use of calcareous tufa as an archive of paleoclimate information. © 20172017 Published The Authors. by Elsevier Published B.V. Thisby Elsevier is an open B.V. access article under the CC BY-NC-ND license Peer(http://creativecommons.org/licenses/by-nc-nd/4.0/-review under responsibility of the organizing). committee of WRI-15. Peer-review under responsibility of the organizing committee of WRI-15 Keywords: Calcareous Tufa; Extra Polymeric Substances (EPS); biofilms; fresh water carbonates

1. Introduction

Calcareous tufa is formed by precipitation of calcium carbonate under subaerial cool-ambient temperature conditions in various depositional systems, where calcium and bicarbonate-rich waters are involved, such as creeks, springs or lakes in limestone karstic settings1. While the calcite precipitation in such settings is partially solely

* Corresponding author. Tel.: +49 6221 546004; fax: +49 6221 545503 E-mail address: [email protected]

1878-5220 © 2017 Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the organizing committee of WRI-15 doi: 10.1016/j.proeps.2016.12.083 Simon M. Ritter et al. / Procedia Earth and Planetary Science 17 ( 2017 ) 328 – 331 329

physical, due to CO2-degasing and subsequent oversaturation of calcium carbonate, tufa formation is yet essentially linked to complex microbiological communities2,3. Laboratory flume experiments revealed large inventories of Ca2+ and other ions (Ba, Sr, Mg) in tufa biofilms chelated by extra cellular substances (EPS) and that this chelation process is chemoselective in favour of divalent ions with low charge densities in the order of Ba>Sr>Mg4. This may conflict with the use of fast growing tufa deposits as paleoclimate archive for temperature reconstructions with Mg/Ca-Thermometry3,5 in regard to the alteration of Mg/Ca ratios of calcite in the presence of EPS4,6. This study examines a natural calcareous tufa depositing site with the focus on quantifying the role of biofilms in tufa formation and its effects on the water and tufa geochemistry. The investigated Kaisinger creek is located 2 km south-east of the municipal Greding in the southern Franconian Alb, about 50 km south of the city Nürnberg in Germany. In the typical cuesta landscape, rapid draining of the limestone plateaus takes place in karstic cavern systems and so, many CO2-supersaturated karstic springs discharge along the valleys. The Kaisinger creek is subdivided into an upstream section, where no tufa deposition occurs, a tufa depositing section with three distinguishing tufa morphologies and a downstream section with only minor tufa deposition (Fig. 1a, b). Tufa in the Kaisinger Brunnenbach is deposited in three main sedimentary facies7: (i) Tufa-barrages composed of irregularly voided phytoherm boundstones with a high content of organic remains that dam the water forming the typical terraces. (ii) Laminated Tufa-Stromatolites predominantly occurring in areas of fast water flow and (iii) Pool-Sediments composed of fragments of (i) and (ii), as well as calcite-encrusted organic remains that are accumulated in the numerous pools in between the damming Tufa-barrages.

2. Methods

Water samples for determination of dissolved main cation and anion contents were taken at 15 sampling locations along the creek stream course (Fig. 1). Samples for determination of main-cations where filtered (0.45 µm) and acidified to pH < 2, stored cool in PP centrifuge tubes for subsequent ICP-OES analysis. Samples for determination of anion contents were filtered and stored cool in PP centrifuge tubes for immediate (less than 1 day after sampling) subsequent Ion-chromatography and Total Carbon Analyzer analysis. On-site parameters temperature, pH and electrical conductivity were monitored. For determination of Mg/Ca, Sr/Ca and Ba/Ca ratios in tufa samples, 200 mg of powdered tufa were digested in 22 % nitric acid and subsequently analyzed by ICP-OES. Relative uncertainties of measurement (2V) are 5 % for calcium, 1.3 % for magnesium, 2.5 % for strontium and 2.7 % for 8 barium. Saturation indices and CO2 partial pressures of the water samples were calculated with PhreeqC .

3. Results

The water of the Kaisinger creek is a Ca-HCO3-type. Minor constituents are sulfate, nitrate, chloride, magnesium, sodium and potassium, which concentrations remain constant along the stream course, thus they behave chemically conservative referring to tufa formation in the Kaisinger creek. The pH values rise along the stream course from around 7.1 at the spring to around 8.1 at the beginning of the tufa section and remain at values of around 8.1 throughout the tufa section (Fig. 1c). At the resurgence point, pH- values are lower, but increase again further downstream. The tufa depositing section in the Kaisinger creek is well reflected by SIcalcite > 0 (Fig. 1d). The PCO2 decreases downstream due to CO2-degassing, leading to increasing SIcalcite and thus an oversaturation and subsequent deposition of calcareous tufa (Fig. 1d). A slightly higher PCO2 and calcite undersaturation is observed at the resurgence point of the creek. Further downstream, PCO2 declines again and calcite saturation increases leading to another, yet smaller, tufa deposition. Calcium concentrations strongly decrease in the tufa forming section of the creek while the main decrease occurs in section Tufa-2 and Tufa-3 (Fig. 1e). Barium concentrations show a similar, though weaker, decline in the tufa forming sections. The decreases of calcium and barium can be expressed as the relative difference of the maximum value (mean of samples of the upstream section + Tufa 1) and the minimum value (mean of the two lowest values of samples in the tufa section Tufa-1 – Tufa-3). The relative decreases are 'Camax-min = 14.6 ± 1.5 % and 'Bamax-min = 6.4 ± 1.3 %, whereas the uncertainties are calculated from the mean deviations and the relative uncertainty of measurement of each element. Magnesium and strontium concentrations remain constant along the stream course,

330 Simon M. Ritter et al. / Procedia Earth and Planetary Science 17 ( 2017 ) 328 – 331

thus an increase of Mg/Ca and Sr/Ca is observed in the tufa forming section (Fig. 1f). Ba/Ca ratios show a weaker increase though. The Mg/Ca, Sr/Ca and Ba/Ca ratios of the analyzed tufa samples are without exception much lower than that of the creek water (Tab. 1). Regarding the phytoherm framestone tufa samples, there is a trend of increasing Mg/Ca downstream. Stromatolite tufa samples seem to contradict this trend showing the lowest Mg/Ca ratio of all samples. The analyzed pool sediments show element ratios that match those of tufa samples measured upstream of the pool sediment tufa.

Tab. 1: Mg/Ca, Sr/Ca and Ba/Ca weight ratios of HNO3-leachates of different tufa samples compared to the mean creek water of the upstream section and the main tufa forming sections of Tufa-1and T-2 + Tufa-3 (see Fig. 1). Values are the mean of measured samples and given uncertainties are calculated mean deviations. Mean water element weight ratios are calculated from six individual sampling campaigns.

Type of sample distance to amount of Mg/Ca Sr/Ca Ba/Ca spring [m] samples *10-3 *10-4 *10-4 Phytoherm framestone tufa 550 3 1.55 ± 0.07 1.08 ± 0.02 0.79 ± 0.04 Phytoherm framestone tufa 900 5 1.78 ± 0.11 1.04 ± 0.04 0.61 ± 0.04 Phytoherm framestone tufa 950 2 1.92 ± 0.15 1.07 ± 0.04 0.67 ± 0.03 Pool sediment tufa 950 2 1.73 ± 0.06 1.14 ± 0.02 0.73 ± 0.00 Stromatolite tufa 1020 4 1.38 ± 0.02 1.09 ± 0.05 0.65 ± 0.04 Mean tufa samples 16 1.64 ± 0.17 1.07 ± 0.05 0.67 ± 0.07 Mean water of upstream section + Tufa-1 0–650 30 58.6 ± 3.2 7.20 ± 0.2 2.22 ± 0.14 Mean water of tufa section T-2+Tufa-3 750–950 42 64.5 ± 4.8 7.90 ± 0.3 2.31 ± 0.07

4. Discussion

The tufa formation in the Kaisinger creek is well reflected in the hydro-chemistry, especially by the decrease of dissolved calcium concentrations and over-saturation with respect to calcite (Fig. 1 d, e).

Figure 1: Composite plot showing the morphological features Thalweg (a) and the schematic tufa morphology with three distinct tufa types Tufa 2+ 1–3 (b) of the Kaisinger creek. The tufa formation is reflected in the creek water chemical parameters pH (c), SIcalcite and log PCO2 (d), Ca and Ba2+ concentrations (e) and the weight ratios of dissolved Mg/Ca, Sr/Ca and Ba/Ca (f). Results are exemplary for six individual sampling campaigns, shown samples were taken in August 2014.

Simon M. Ritter et al. / Procedia Earth and Planetary Science 17 ( 2017 ) 328 – 331 331

The accompanying, slight, yet determin-able decrease of dissolved barium cannot be explained by inorganic calcite precipitation as distribution coefficients of barium into calcite are usually very low. Although barite (BaSO4) is highly insoluble, precipitation is unlikely because of an undersaturation with respect to Barite (SIBarite = -0.2). Under the reasonable assumption that the total decrease of Ca2+ and Ba2+ is due to tufa formation (calcite precipitation) a distribution coefficient of barium for partitioning from creek water into tufa DBa-water can be calculated by the relative decrease of barium and calcium concentrations ('Camax-min and 'Bamax-min) in the tufa forming creek section yielding DBa-calcite (water) = 0.44 ± 0.07. Considering the tufa sample Ba/Ca ratios a distribution coefficient DBa-calcite (tufa) can be calculated using the mean of the tufa samples Ba/Ca and the mean of the water Ba/Ca of the upstream section (Tab. 1). This yields a DBa-Tufa = 0.30 ± 0.04. Both calculated dis-tribution coefficients DBa-water and DBa-tufa are much higher than the reported distribution coefficient DBa-calcite = 0.012 ± 0.005 of barium for partitioning from aqueous solution into calcite9. This can be attributed to the extensive incorporation of Ba2+ ions into biofilms associated with the tufa formation as suggested by results of laboratory experiments with natural tufa samples4. Regarding the geochemistry of the tufa samples, the trend of increasing Mg/Ca ratios downstream, which is expected by the observed increasing Mg/Ca ratios in the creek water, is only valid for the phytoherm framestone tufa samples. The lower Mg/Ca ratios of the stromatolite tufa samples from even further downstream (Tab. 1) seem to contradict this trend. Stromatolite tufa are mainly composed of encrusted cyanobacteria and green algae biofilms10 and may give another hint for the important role of biofilms in tufa formation. Low Mg/Ca in those tufa samples is linked to the preferentially chelation of Ca2+ over Mg2+ in the microenvironment around the EPS molecules resulting in lower Mg/Ca ratios of calcite as expected from the bulk water Mg/Ca at a given temperature4,6. Therefore, the lower Mg/Ca of the stromatolite tufa compared to that of the phytoherm framestone tufa is not only influenced by the bulk water Mg/Ca, but reflect an even more extensive portion of biofilms in the formation of this tufa type.

5. Conclusion

All measured chemical parameters of the creek water are in good agreement with the tufa forming section in the Kaisinger creek. The role of biofilms with respect to chemoselective chelation processes of EPS in tufa formation, that was already demonstrated on laboratory scale4, is now confirmed in a natural tufa depositing site by altering the geochemistry of tufa and even the hydrogeochemistry of the creek water. The fact that the influence of biofilms is detectable in terms of the decrease of Ba2+ in the tufa forming creek section holds the potential to monitor seasonal variabilities of the relative portion of biofilm influence on tufa formation. The introduced tufa depositing site Kaisinger creek is capable to act as a field laboratory to study geochemical processes that are relevant with regard to the use of tufa as a valuable archive of paleoclimate information.

References

1. Capezzuoli, E., Gandin, A. & Pedley, M., 2014. Decoding tufa and travertine (fresh water carbonates) in the sedimentary record: The state of the art. Sedimentology, 61(1), pp.1–21. 2. Andrews, J.E. & Brasier, A.T., 2005. Seasonal records of climatic change in annually laminated tufas: Short review and future prospects. Journal of Quaternary Science, 20(5), pp.411–421. 3. Pedley, M., 2014. The morphology and function of thrombolitic calcite precipitating biofilms: A universal model derived from freshwater mesocosm experiments. Sedimentology, 61, pp.22–40 4. Rogerson, M. et al., 2008. New insights into biological influence on the geochemistry of freshwater carbonate deposits. Geochimica et Cosmochimica Acta, 72(20), pp.4976–4987. 5. Ihlenfeld, C. et al., 2003. Climatic significance of seasonal trace element and stable isotope variations in a modern freshwater tufa. Geochimica et Cosmochimica Acta, 67, pp.2341–2357 6. Saunders, P. et al., 2014. Mg/Ca ratios in freshwater microbial carbonates: Thermodynamic, kinetic and vital effects. Geochimica et Cosmochimica Acta, 147, pp.107–118. 7. Pedley, H.M., 1990. Classification and environmental models of cool freshwater tufas. Sedimentary Geology, 68(1-2), pp.143–154. 8. Parkhurst, B.D.L. & Appelo, C. a J., 1999. User’s Guide To PHREEQC (version 2) — a Computer Program for Speciation, and Inverse Geochemical Calculations. Exchange Organizational Behavior Teaching Journal, D(Version 2), p.326. 9. Tesoriero, A. & Pankow, J., 1996. Solid solution partitioning of Sr 2+, Ba 2+, and Cd 2+ to calcite. Geochimica et Cosmochimica Acta, 60(6), pp.1053–1063. 10. Shiraishi, F., 2008. Microbial metabolisms and calcification in freshwater biofilms. Dissertation at Georg-August-Universität Göttingen.