Hydrol. Earth Syst. Sci., 24, 673–696, 2020 https://doi.org/10.5194/hess-24-673-2020 © Author(s) 2020. This work is distributed under the Creative Commons Attribution 4.0 License. Freshwater pearl mussels from northern Sweden serve as long-term, high-resolution stream water isotope recorders Bernd R. Schöne1, Aliona E. Meret2, Sven M. Baier3, Jens Fiebig4, Jan Esper5, Jeffrey McDonnell6, and Laurent Pfister7 1Institute of Geosciences, University of Mainz, Mainz, 55128, Germany 2Naturhistoriska riksmuseet, Stockholm, 114 18, Sweden 3Agilent Technologies Sales & Services GmbH & Co. KG, Frankfurt am Main, 60528, Germany 4Institute of Geosciences, J. W. Goethe University, Frankfurt am Main, 60438, Germany 5Department of Geography, University of Mainz, Mainz, 55128, Germany 6Global Institute for Water Security, University of Saskatchewan, Saskatoon, SK S7N 3H5, Canada 7University of Luxembourg, Faculty of Science, Technology and Medicine, 2 Avenue de l’Université, 4365, Esch-sur-Alzette, Luxembourg Correspondence: Bernd R. Schöne ([email protected]) Received: 29 June 2019 – Discussion started: 29 July 2019 Revised: 27 November 2019 – Accepted: 16 January 2020 – Published: 17 February 2020 Abstract. The stable isotope composition of lacustrine sed- consistent ontogenetic trends, but rather oscillated around an iments is routinely used to infer Late Holocene changes in average that ranged from ca. −12:00 to −13:00 ‰ among precipitation over Scandinavia and, ultimately, atmospheric the streams studied. Results of this study contribute to an circulation dynamics in the North Atlantic realm. However, improved understanding of climate dynamics in Scandinavia such archives only provide a low temporal resolution (ca. and the North Atlantic sector and can help to constrain eco- 15 years), precluding the ability to identify changes on inter- hydrological changes in riverine ecosystems. Moreover, long annual and quasi-decadal timescales. Here, we present a isotope records of precipitation and streamflow are pivotal new, high-resolution reconstruction using shells of freshwa- to improve our understanding and modeling of hydrological, ter pearl mussels, Margaritifera margaritifera, from three ecological, biogeochemical and atmospheric processes. Our streams in northern Sweden. We present seasonally to annu- new approach offers a much higher temporal resolution and ally resolved, calendar-aligned stable oxygen and carbon iso- superior dating control than data from existing archives. tope data from 10 specimens, covering the time interval from 1819 to 1998. The bivalves studied formed their shells near equilibrium with the oxygen isotope signature of ambient water and, thus, reflect hydrological processes in the catch- 1 Introduction ment as well as changes, albeit damped, in the isotope sig- 18 nature of local atmospheric precipitation. The shell oxygen Multi-decadal records of δ O signals in precipitation and isotopes were significantly correlated with the North Atlantic stream water are important for documenting climate change Oscillation index (up to 56 % explained variability), suggest- impacts on river systems (Rank et al., 2017), improving the ing that the moisture that winter precipitation formed from mechanistic understanding of water flow and quality control- originated predominantly in the North Atlantic during NAOC ling processes (Darling and Bowes, 2016), and testing Earth years but in the Arctic during NAO− years. The isotope sig- hydrological and land surface models (Reckerth et al., 2017; nature of winter precipitation was attenuated in the stream Risi et al., 2016; see also Tetzlaff et al., 2014). However, the water, and this damping effect was eventually recorded by common sedimentary archives used for such purposes typi- the shells. Shell stable carbon isotope values did not show cally do not provide the required, i.e., at least annual, tem- poral resolution (e.g., Rosqvist et al., 2007). In such stud- Published by Copernicus Publications on behalf of the European Geosciences Union. 674 B. R. Schöne et al.: Freshwater pearl mussels as long-term, high-resolution stream water isotope recorders ies, the temporal changes in the oxygen isotope signatures 80 years (von Hessling, 1859) to over 200 years (Ziuganov of meteoric water are encoded in biogenic tissues and abio- et al., 2000; Mutvei and Westermark, 2001), offering an in- genic minerals formed in rivers and lakes (e.g., Teranes and sight into long-term changes in freshwater ecosystems at an McKenzie, 2001; Leng and Marshall, 2004). In fact, many unprecedented temporal resolution. studies have determined the oxygen isotope composition of Here, we present the first absolutely dated, annually re- diatoms, ostracods, authigenic carbonate and aquatic cellu- solved stable oxygen and carbon isotope record of freshwater lose preserved in lacustrine sediments to reconstruct Late pearl mussels from three different streams in northern Swe- Holocene changes in precipitation over Scandinavia and, ul- den covering nearly 2 centuries (1819–1998). We test the timately, atmospheric circulation dynamics in the North At- ability of these freshwater pearl mussels to reconstruct the lantic realm (e.g., Hammarlund et al., 2002; Andersson et al., NAO index and associated changes in precipitation prove- 18 2010; Rosqvist et al., 2004, 2013). With their short residence nance using shell oxygen isotope data (δ Os). We evalu- times of a few months (Rosqvist et al., 2013), the hydro- ate how shell oxygen isotope data compare to δ18O values 18 18 logically connected, through-flow lakes of northern Scandi- in stream water (δ Ow) and precipitation (δ Op) as well navia are an ideal region for this type of study. Their isotope as limited existing environmental data (such as stream wa- signatures – while damped in comparison with isotope sig- ter temperature); we also evaluate how these variables relate 18 18 nals in precipitation – directly respond to changes in the pre- to each other. We hypothesize that δ Ow and δ Op values cipitation isotope composition (Leng and Marshall, 2004). are positively correlated with one another as well as with However, even the highest available temporal resolution of the North Atlantic Oscillation index. In other words, during such records (15 years per sample in sediments from Lake positive NAO years, oxygen isotope values in stream wa- Tibetanus, Swedish Lapland; Rosqvist et al., 2007) is still ter and shells are higher than during negative NAO years, insufficient to resolve inter-annual- to decadal-scale variabil- when shell and stream water oxygen isotope values tend to ity, i.e., the timescales at which the North Atlantic Oscilla- be more depleted in 18O. In addition, we explore the phys- tion (NAO) operates (Hurrell, 1995; Trouet et al., 2009). The ical controls on shell stable carbon isotope signatures. Pre- NAO steers weather and climate dynamics in northern Scan- sumably, these data reflect changes in the stable carbon iso- dinavia and determines the origin of air masses from which tope value of dissolved inorganic carbon, which, in turn, is meteoric waters form. While sediment records are still of vi- sensitive to changes in primary production. We leverage past tal importance for century-scale and millennial-scale varia- work in Sweden that has shown that the main growing season tions, new approaches are needed for finer-scale resolution of M. margaritifera occurs from mid-May to mid-October, on the 1–100-year timescale. with the fastest growth rates occurring between June and Au- One underutilized approach for hydroclimate reconstruc- gust (Dunca and Mutvei, 2001; Dunca et al., 2005; Schöne et tion is the use of freshwater mussels as natural stream water al., 2004a, b, 2005a). We use changes in the annual incre- stable isotope recorders (Dettman et al., 1999; Kelemen et al., ment width of M. margaritifera to infer water temperature, 2017; Pfister et al., 2018, 2019). Their shells can provide sea- because growth rates are faster during warm summers and sonally to annually resolved, chronologically precisely con- result in broader increment widths (Schöne et al., 2004a, b, strained records of environmental changes in the form of 2005a). Because specimens of a given population react sim- variable increment widths (which refers to the distance be- ilarly to changes in temperature, their average shell growth tween subsequent growth lines) and geochemical properties patterns can be used to estimate climate and hydrological (e.g., Nyström et al., 1996; Schöne et al., 2005a; Geist et al., changes. Consequently, increment series of specimens with 2005; Black et al., 2010; Schöne and Krause, 2016; Geeza overlapping life spans can be crossdated and combined to et al., 2019, 2020; Kelemen et al., 2019). In particular, sim- form longer chronologies covering several centuries (Schöne ilar to marine (Epstein et al., 1953; Mook and Vogel, 1968; et al., 2004a, b, 2005a). Killingley and Berger, 1979) and other freshwater bivalves (Dettman et al., 1999; Kaandorp et al., 2003; Versteegh et al., 2009; Kelemen et al., 2017; Pfister et al., 2019), Margar- 2 Material and methods itifera margaritifera forms its shell near equilibrium with the 18 oxygen isotope composition of the ambient water (δ Ow) We collected 10 specimens of the freshwater pearl mussel (Pfister et al., 2018; Schöne et al., 2005a). If the fractiona- M. margaritifera from one river and two creeks in Nor- tion of oxygen isotopes between the water and shell carbon- rland (Norrbotten County), northern Sweden (Figs. 1, 2; ate