The head of the project: Ekaterina A. Dolgova
The title of the project: Climate reconstruction based on millennia long tree-ring data from Solovki Island
Key words: dendrochronology, paleoclimate reconstruction, climate response function, last millennium, Solovki Island, tree-ring width, tree-ring maximum density, buried wood, Little Ice Age, Medieval Climate Anomaly
The study of trends in climatic variability and extreme climatic events in the Northern Hemisphere in the late Holocene is a modern task of paleoclimatology (Frank et al., 2010; Masson-Delmotte et al., 2013). Some of the reconstructions are based only on dendrochronological data (Esper et al., 2002; D'Arrigo et al., 2006; Frank et al., 2007; Wilson et al., 2007; Schneider et al., 2015; Stoffel et al. , 2015), but in most cases, multy-proxy data approach is the most often used (Jones et al., 1998; Mann et al., 1999; Mann et al., 2009; Moberg et al., 2005; Hegerl et al., 2007 ; Wahl and Ammann, 2007; Ljungqvist, 2010).
Most of the reconstructions in the Northern Hemisphere are presented as a single mean temperature curve, and only a few reconstructions have a spatial distribution (Briffa et al., 2002; Mann et al., 1998; Mann et al., 2009; Ljungqvist et al., 2012; Tingley and Huybers, 2013). The latter approach is key in understanding the spatiotemporal patterns of modern warming in context of other climatic epochs, and also allows comparing reconstructions with global climatic models and revealing external factors controlling low- to high-frequency climate variability. In recent published reconstructions for the Northern Hemisphere (IPCC, 2013, Section 5.3.5, Masson-Delmotte et al., 2013), we can clearly trace the coherence of climatic epochs: the warming of the 10th century with a subsequent cooling during next 500 years, a cooling around 1450-1850ss. and a warming trend that began in the mid-19th century. Although these reconstructions are generally similar, nevertheless, there are significant differences in the time and amplitude of some epochs (Esper et al., 2004). Most likely, the use of several proxies of paleoclimatic information leads to uncertainties in the final reconstruction. The discrepancies are attributed to different seasonal climatic responses, different resolution of proxies (not annual), and the lack of reliable dating (Wilson et al., 2016). The urgency of the development of a dendroclimatic network throughout the Northern Hemisphere remains unchanged even in the face of significant progress in this area.
In this work, we will try to fill this gap by creating a millennia long reconstruction of air temperatures based on dendrochronological data. Our previous dendrochronological research in the Solovki Island allowed us to develop the first tree-ring width chronology of conifers for the last 850 years (Matskovsky et al., 2011). Such a long duration was achieved due to including samples from living trees and architectural objects. Most likely, the extension of chronology due to architecture has almost exhausted itself. As new sources of old-aged wood, we propose the use of buried wood in lake sediments. This practice proved itself in neighboring Finland (Helama et al., 2017) with similar hydro-geomorphological conditions, when the wood obtained from the bottom of shallow lakes remains safe. If the cross-dating of the floating chronology does not give results, AMS-dating will be performed.
Reconstruction of air temperature will be obtained from the Blue Intensity measurements of conifers. At the same time, several topical problems will be solved: calibration of the Blue Intensity with respect to the X-ray maximum density, estimation of the possibility of a cross-dating of the floating chronology and an analysis of the stability of the climatic signal inferred from tree-ring data.
The problem of choosing the method of standardizing chronology is still relevant in dendroclimatology and a wrong choice led to biased results. We will develop the air temperature reconstruction based on two different techniques: one based on the standardization of the dendrochronological using RCS and linear regression; the second one - without standardizing the series by direct transition of absolute values of Blue Intensity values to instrumental data (Matskovsky and Helama, 2015).
Thus, the results obtained during the implementation of the project will cover the most actual problems of modern dendroclimatology.
The expected results
1. We will develop first a millenia-long composite Blue-Intensity chronology of the coniferous from Solovki Island. The chronology length increase will be possible due to the inclusion in the already existing chronology (850 years) of buried wood from bottom lake sediments. The success of this approach is shown in the example of the Finnish chronology, the duration of which has been significantly increased. The lack of millenia-long absolutely dated chronologies is still an important problem in the development of global paleoclimatic reconstructions (Wilson et al., 2016) and the creation of a thousand-year sensitive chronology can become part of spatial global reconstructions.
2. For the first time for this territory, the Blue Intensity parameter (BI) will be measured, and the difference between early and late wood (delta BI) will be measured to remove possible errors associated with a sharp color transition between sapwood and heartwood. This method, which is cheaper than traditional X-ray densitometry, has become widespread in recent years. All the studies we know show a strong summer temperature signal (for example, McCarroll et al., 2013; Wilson et al., 2014; Dolgova, 2016). Despite the advances in this field, the entire scientific community agrees that an adequate analysis of the strengths and weaknesses of this method requires obtaining as much data as possible from different geographic areas using different species of conifers (Rydval et al., 2014; Björklund et al 2014, 2015). The legality of using optical density will be shown by comparing it with the maximum density measured on identical samples. In this regard, the calibration of the optical density with respect to the maximum will be performed, and in the future, this calibration curve will be extended to the entire region of the North of the European part of Russia. This approach will be performed for the first time for Russian territory, and the positive results obtained during the project will become the basis for the use of optical density by other Russian dendrochronological laboratories.
3. For the first time for the study area, the results of the possibility of cross-dating of floating chronologies based on BI measurements will be presented. The problem of obtaining absolute datings of floating chronologies has always been acute.
Often, short length and weak climate sensitivity tree-ring width series made it difficult to accurately classify the series to a particular period. Recent research has shown that the use of BI (instead of the tree-ring width) in Scotland for the purpose of dating of architectural wood has significantly increased the statistical reliability of the result (Wilson et al., 2017). This gives us reason to assume that a similar situation will be observed in the Solovki Island. Measurement of BI of living and buried trees will allow us to assess how much this parameter is suitable for dating. We will review our previous dating of the floating series and give a comparative description of the statistical data for dating. The results of this approach will open up new opportunities in the field of archaeological research and will lead to more reliable dating results for the whole European territory of Russia.
4. An estimation of the climatic function of the BI-chronology will be given, with particular attention to the divergence problem of the series. The solution of this task in the modern dendroclimatology deserves special attention, since the inclusion of such series in large-scale reconstruction can lead to significant errors. Thus, in recent works, one of the criteria for using regional reconstruction is the lack of divergence (Wilson et al., 2016). If there is a divergence between the density and temperature series, the possible reasons for its occurrence will be analyzed.
5. The BI-chronology will be studied in terms of presence of long-term variability in it, as well as its ability to reflect the amplitude of extreme climatic events during the calibration period. We will estimate the spectra coherence of the BI and maximum density chronology with each other and with instrumental air temperature records.
6. For the first time for the Northern part of European territory of Russia, reconstruction of the air temperature for the last millennium will be achieved by two different methods. The first reconstruction will be obtained by traditional method in dendroclimatology - using linear regression. The second reconstruction will be obtained without standardization by direct transformation from absolute values of BI to temperature records. The results of comparison between two reconstructions and analyzing the reasons for the differences between the series will be presented.
7. The contribution of external factors affecting the climate will be assessed, by comparing the reconstruction with modeled radiation data (Schmidt et al, 2012), which in turn reflect the insolation variability and the consequences of volcanic activity in the Northern Hemisphere. Due to this analysis, it will be possible to determine the reasons for establishing cold or warm epochs. An assessment of the amplitude and duration of cooling due to large volcanic eruptions will be made by the method of superimposed epochs (SEA).
8. All the interim and final results will be posted on the web site dedicated to the implementation of the project. The results of the research will be published in high-ranking international journals.