ORIGINAL ARTICLEThe dendroclimatic value of oak stable isotopes
Introduction
Changes in tree-ring stable oxygen (δ18O; 18O/16O) and carbon (δ13C; 13C/12C) isotopes are modulated mainly by plant physiological processes associated with photosynthetic CO2 uptake (Farquhar et al., 1989; Gessler et al., 2014), as well as by the transpiration of water (Barbour and Farquhar, 2004). These factors are often influenced by environmental perturbations and/or changes in weather and climate. It has been shown that stable isotopes in tree rings reflect vegetation responses to changes in temperature, relative humidity, precipitation, and/or light (Roden et al., 2000; Leavitt, 2010). However, changes in the isotopic composition of atmospheric CO2 and source water must also be taken into account. Species-specific sensitivity to abiotic factors, as well as the overall constitution of the tree species including the root system, may lead to differences in the δ18O and δ13C ratios (Ponton et al., 2001; Song et al., 2014; Barbeta and Peñuelas, 2017).
Valuable climate information can be extracted from δ18O and δ13C isotopic ratios in tree rings (McCarroll and Loader, 2004; De Boer et al., 2019). Tree-ring stable isotopes can improve high-resolution reconstructions of hydroclimatic variability as compared to more traditional tree-ring parameters including tree-ring width and latewood density (Büntgen et al., 2020; Shestakova and Martínez-Sancho, 2021). Tree-ring stable isotopes have been used to reconstruct various hydroclimatic parameters, such as drought and sunlight duration in the European Alps (Kress et al., 2010); relative humidity, cloud cover, and drought in the eastern Carpathians (Nagavciuc et al., 2019); the Palmer drought severity index (PDSI) in China (Liu et al., 2019); the standardized precipitation evaporation index (SPEI) in southern Central Siberia (Shestakova et al., 2017); and precipitation amounts in Pakistan (Treydte et al., 2009). While there is evidence that δ18O can reflect changes in relative humidity and precipitation in semi-arid regions (Szejner et al., 2018), the links between δ18O and δ13C and climate signals are often stronger under wetter rather than drier conditions (Esper et al., 2018). In humid environments, variations in δ13C reflect the activity of Rubisco, the primary carbon-fixing enzyme in C3 plants (Hafner et al., 2014). In dry environments, δ13C is mainly controlled by stomatal conductance and decreases with increasing drought severity or vapor pressure deficit (McCaroll and Loader et al., 2004; Young et al., 2015). Fractionation of oxygen isotopes prior to and during wood formation is more complex than carbon fractionation (Gessler et al., 2014). Values of δ18O received increasing attention for long-term climate reconstruction in the last decade (Nakatsuka et al., 2020). In addition to providing reliable climate archives, δ18O chronologies have been shown to improve cross-dating compared to conventional tree-ring widths (Loader et al., 2020).
A multi-millennial standard oak ring-width chronology of ∼4000 core and disc samples from living trees, historical timbers, archaeological remains, and subfossil wood has been developed for the Czech Republic (Dobrovolný et al., 2015; Prokop et al., 2017; Dobrovolný et al., 2018). However, the suitability of stable isotopes from such a heterogeneous compilation has to be demonstrated before it can be used in paleoclimatic studies. Putative uncertainties related to species identification, geographical origin, cambial age, and the presence of polyvinyl may possibly limit the exploitation of all relict samples for climate reconstructions. Here, we present annually-resolved and absolutely-dated δ18O and δ13C ratios in 21 living oaks from different locations along elevational, longitudinal and ecological gradients in the Czech Republic. We use this dataset to elucidate the natural variability of oak stable isotopes. Moreover, we test whether polyvinyl acetate, an emulsion often used for the preservation of relict samples, has any influence on isotopic signals. More specifically, we aim to address the following issues: (1) The δ18O and δ13C ratios in the extracted alpha cellulose should be tested to ensure they are not influenced by the polyvinyl acetate emulsion. (2) Since the dataset includes two widespread oak species—English oak (Quercus robur L.) and sessile oak (Quercus petraea [Matt.] Liebl.)—which are anatomically indistinguishable from one another (Schoch et al., 2004), the effect of any species differences on δ18O and δ13C ratios must be investigated. Previous studies have revealed contradictory results (Ponton et al., 2001; Rybníček et al., 2016). (3) Uncertainties regarding the extent to which the geographical origin of the wood influences δ18O and δ13C signals must be clarified (Roden et al., 2005; Shestakova et al., 2017; Zalloni et al., 2019). (4) Age-related changes in physiological activity and tree-ring morphology, including tree‐ring width and maximum latewood density (Genet et al., 2010), might have an effect on tree-ring stable isotopes (Etien et al., 2008; Labuhn et al., 2013). If this is the case, it makes using historical wood samples for isotopic-based climate reconstruction particularly problematic.
Section snippets
Tree-ring network
Our sampling sites represent the natural, spatial and altitudinal distribution of oak (Q. robur and Q. petraea) in the Czech Republic (Fig. 1). Moreover, young (<50-year-old) and old (>100-year-old) trees were sampled to investigate the impact of tree age on stable isotopes. In total, 21 trees from seven localities in western Bohemia (sites #1, 2, and 6) and eastern Moravia (sites #3–5, and 7) were sampled (Table 1). This network includes trees from low (170–250 m a.s.l.; sites #3, and 5–7) and
Site effect
The TRW, δ18O, and δ13C time series (1920–2018 CE) of our sampled trees (Table 1) are shown in Fig. 3. Both the uncorrected δ13C time series and the time series corrected for the increase in atmospheric CO2 due to the burning of fossil fuels are shown (Figs. 3C, D). Hereafter, all presented δ13C time series refer to atmospherically corrected data.
Inter-site correlations of δ13C series, expressed as a Pearson’s correlation coefficient (r), range between 0.48 and 0.77 (p < 0.01) at most
Conclusions
Our annually-resolved and absolutely-dated δ18O and δ13C measurements of 21 living oak trees from the Czech Republic exhibit significant coherency across species, locations, and tree ages. Moreover, the applied extraction method removes any biases resulting from the polyvinyl acetate emulsions that are commonly used to preserve wood samples. The pooling of samples is often motivated by lower costs and shorter analysis times (Leavitt, 2010); however, non-pooled data enable the important
Declaration of Competing Interest
The authors report no declarations of interest.
Acknowledgements
This research was supported by the Czech Science Foundation (grant 17-22102S). The work by O.U. and A.A. on the development of the methods was supported by SustES – Adaptation strategies for sustainable ecosystem services and food security under adverse environmental conditions (CZ.02.1.01/0.0/0.0/16_019/0000797). The authors would also like to thank Inna Roshka and Josef Čáslavský for their work in the laboratory, and two anonymous referees who kindly comment on previous versions of this
References (54)
- et al.
Reconstruction of past precipitation δ18O using tree-ring cellulose δ18O and δ13C: a calibration study near Lac d’Annecy, France
Earth Planet. Sci. Lett.
(2006) - et al.
The core-microtome: a new tool for surface preparation on cores and time series analysis of varying cell parameters
Dendrochronologia
(2010) - et al.
No systematic effects of sampling direction on climate-growth relationships in a large-scale, multi-species tree-ring data set
Dendrochronologia
(2019) - et al.
Notes towards an optimal sampling strategy in dendroclimatology
Dendrochronologia
(2018) The use of altitude in ecological research
Trends Ecol. Evol.
(2007)Tree-ring C–H–O isotope variability and sampling
Sci. Total Environ.
(2010)- et al.
Tree-ring δ18O based PDSI reconstruction in the Mt. Tianmu region since 1618 AD and its connection to the East Asian summer monsoon
Ecol. Indic.
(2019) - et al.
Oxygen isotope dating of oak and elm timbers from the portcullis windlass, Byward Tower, Tower of London
J. Archaeol. Sci.
(2020) - et al.
Stable isotopes in tree rings
Quat. Sci. Rev.
(2004) - et al.
A mechanistic model for interpretation of hydrogen and oxygen isotope ratios in tree-ring cellulose
Geochim. Cosmochim. Acta
(2000)