Annotation of the results obtained in 2023
An expedition lasting 14 days (July 15 – July 28, 2023) was conducted to collect samples. The expedition route, the total length of which was about 4000 km (250 km per day, not counting the sampling work), was adjusted along the way taking into account the situation on the ground, the plan for the number of established sites and time restrictions. In total, during the third year of the project, 247 samples were selected: 152 samples from living trees (4 collections), 95 samples from wooden buildings (7 collections). Chronologies were constructed based on various parameters of tree rings for five sample plots and for four architectural objects. During the third year of the project, sample preparation and optical density measurements were carried out for 266 cores from living trees and architectural wood from the Novgorod, Kostroma, Saratov, Voronezh, Rostov regions and Perm Territory. The optical density of the southernmost pine chronologies (T45S, V08S) was measured to subsequently assess changes in the climatic response of optical density along the north-south gradient. Analysis of materials from excavations in the city of Tula made it possible to construct a tree-ring chronology (TRC) lasting 187 years. Its comparison with the DKH, built for the upper part of the Western Dvina River basin, made it possible to obtain absolute dating for the Tula chronology. It turned out that this DKH covers the period 1582-1768. The analysis of new samples from excavations at the Trinity-Sergius Lavra and their addition to materials analyzed in previous years of work on the project made it possible to construct four farmhouses with a duration of 309, 141, 102 and 121 years. These chronologies are named SP1, SP2, SP3 and SP4 respectively. For the first of these DKH (SP1), absolute dates (1090-1398) were obtained by comparison with the Western Dvina DKH in 2022 (correlation coefficient 0.30, t-test 5.54). In the same way, dates were obtained (1394-1514) for DKH SP4 (correlation coefficient 0.36, t-criterion 4.15). It has not yet been possible to obtain reliable dating for materials from the cemetery in Gorokhovets (Vladimir region). However, it was possible to show that these materials fall into two practically synchronous, but well-separated groups (lasting 237 and 202 years). For materials from Arkhangelsk, excavation site on the street. Terekhina, 6. Of the four samples, two were dated; the resulting dates were: 1783 for the pile and 1775 for the crown of the log house. For excavation materials on the street. Chumbarova-Luchinsky d. 20, 7 dates were obtained for the pile foundation of the foundation (1900-1906), most of which 5 fall in 1906. Another 6 datings of individual pine piles fell in the 18-19 centuries. Also in 2023, several collections of archaeological material from Kaluga and Perm were measured (see Table 2). To date, six DKHs have been built for subfossil materials of the Sura River (including samples measured and dated in 2023). The age of the youngest of them, GR2 (duration 253 years), was determined in 2022 based on three radiocarbon dates - its last tree ring formed in the interval 1836-1863. For DKH GR6 (duration 435 years) in 2023, two radiocarbon dates were obtained: 750±70 (IGAN-10470) and 450±70 (IGAN-10471). This made it possible to establish that the last annual ring of this DKH was formed in the interval 1336-1673. Thus, the floating farmhouses built to date cover a time period of almost two thousand years (Fig. 6). To investigate the possibility of dating archaeological material based on optical density, this parameter was measured for 22 archaeological samples from the Kostroma region. It turned out that due to the poor preservation of wood, it is not always possible to obtain the same long series of measurements for optical density as for the width of the rings. The main conclusion from the experiment is that optical density is poorly suited for dating archaeological samples, although if they are well preserved it can show better results, especially in the northern part of the region The results of the dendroclimatic analysis of optical density showed that both types of chronologies (based on 15% and 30% of the minimum pixel brightness) have a similar response to climate, which is observed in the same months. At two sites – A22S and T03S – the correlation value is slightly higher when using a 15% chronology. In the case of chronologies from the V08S site, the correlation coefficients are almost the same for the two methods of constructing chronologies, and for the chronologies at the T45S site, the signal for 15% of the variant is practically absent (with the exception of one month). Thus, the results of our study indicate the heterogeneity of the response in two types of chronologies: the 15% version of the chronology turned out to be more climatically sensitive only for pine trees growing in the center of the EPR (Yaroslavl region); for pine trees from sites T03S and V08S the signal is similar for both chronology options , and for site T45S a significant climatic response was found only for the 30% chronology variant. Our results support the importance of testing different cutoff parameters during the process of obtaining absorbance chronologies. An interesting result was the confirmed change in the climate signal in optical density when moving from north to south - a positive response to temperature changes to a negative one. Moreover, the response to temperature is stronger than to the Palmer drought severity index, which indirectly indicates a response to temperature, and not to aridity. A physiological explanation of this relationship through the parameters of the tree ring structure requires additional research. The results of a correlation analysis of the chronologies of maximum and minimum density, calculated on the basis of quantitative indicators of wood anatomy (cellular measurements) showed that the closeness of the relationship largely depends on which percentile is chosen for calculating the chronology. The 0% and 25% chronologies show a similar response: positive for June temperatures for LWAD (r = 0.38 – 0.44), negative for the variability of the dryness index in June, July and September for LWAD (r = -0.26 – -0.34). The 50% and 75% based chronologies have the highest correlation values for precipitation, reaching -0.42, -0.39, -0.19 (March, July and August, respectively). Previously, we found that there was no relationship between the growth of pines from site A22S and summer temperature, and the relationship between optical density (dBI) and June temperature turned out to be low (r=0.198, p<0.05). However, when using anatomical latewood density, the relationship with temperature, precipitation and the Palmer Drought Severity Index is significantly enhanced. An update of three reconstructions based on the optical density of coniferous tree rings for the north (Solovki), center (Kaluga) and south (Caucasus) of the EPR and analysis of their spatial correlation with the temperature field showed that these reconstructions can serve to describe the variability of summer temperatures in the past almost always the entire western part of European Russia. The quality indicators of the reconstruction of the updated drought atlas suggest its superiority over the previous version. Comparing the “self-calibrating” Palmer index (scPDSI), calculated on the basis of model data on temperature and precipitation, with the results of paleo-reconstructions from the Drought Atlas of European Russia, a study was carried out in four regions: region 1 (46-50 N, 40-50 E), region 2 (50-54 N, 32-42 E), region 3 (50-54 N, 42-52 E), region 4 (50- 60 N, 32-52 E). Over a large area (region 4), the reconstructions show no change in aridity during the industrial period compared to the pre-industrial period, while three of the four models show a decrease in drought. The industrial period also stands out for its more consistent model and reconstruction data, especially for mean and extreme values. Analysis of temperature reconstructions indicates similarities to the INM model in the Caucasus, but with a pronounced temperature response to volcanic eruptions. The MRI model shows less warming in the second half of the 20th century. Future climate estimates for all models have been completed, and although the INM model shows good results for some regions, the overall accuracy of the reconstructions remains questionable. Changes in future humidity and the likelihood of droughts are also analyzed, and the ssp585 scenario assumes an increase in aridity towards the end of the 21st century, supported by changes in the relationship of precipitation to temperature. The models also predict different trends in total soil moisture content. In total, in 2023, members of the project’s creative team presented 8 reports at four conferences. Two articles were published: in the journal Brief Communications of the Institute of Archeology (indexed by WoS) and Geosphere Research (indexed by Scopus).

 

Publications

---

Annotation of the results obtained in 2021
During the first year of the project, 1,140 samples were сollected: 466 samples from living trees, 247 samples from wooden buildings and 427 archaeological samples. The blue intensity (blue intensity of late wood BI and the difference in the blue intensity of late and early wood δBI) of 341 cores was measured for the Arkhangelsk, Vologda, Leningrad, Kostroma and Yaroslavl regions. It is shown that the constructed chronologies of blue intensity have high correlation coefficients with chronologies of maximum X-ray density from neighboring locations. The tree-ring width was measured for 10 sites of living trees, 10 local tree-ring chronologies up to 380 years long were constructed. The tree-ring width from 10 old buildings was measured, 10 local tree-ring chronologies up to 462 years long were constructed, 7 of them were dated. Dating, in particular, received such objects of cultural heritage as the chapel of Mary Magdalene, St. Nicholas Church from village Visokiy Ostrov and St. Nicholas Church from village Tukholi (Vitoslavlitsy Museum, Veliky Novgorod), Elijah Church (Kostroma Museum of Wooden Architecture), Bazhenov House from Vavchuga village, Arkhangelsk region. 88 samples of subfossile oak wood from the alluvial deposits of the Sura River were collected and processed, 6 floating chronologies up to 581 years long were constructed. Materials from 11 archaeological collections were measured. A total of 883 samples were measured during the first year of the project: 405 samples from living trees, 160 samples from wooden buildings, 230 archaeological samples and 88 samples of buried wood. Samples were prepared for cellular measurements. The first cellular measurements were obtained for pine trees from the Yaroslavl region. 183 cellulose samples were prepared to measure the ratio of stable carbon isotopes in annual rings. At the moment, 93 samples have been measured. A preliminary analysis of the response function for the period 2010-2020 was carried out, which showed a statistically significant response to aridity conditions.

 

Publications

---

Annotation of the results obtained in 2022
In 2022, organization of saw cut collections was completed (previously, the core collection was organized). The storage room was equipped with racks, the collections of saw cuts were placed on them. Over 1,000 saw cuts have been downsized, greatly reducing the storage space required. The created repository has enough space to replenish with new samples for at least the next 10 years. A booklet-memo on the selection of dendrochronological samples for distribution among the project volunteers was created and passed to the coordinator of the project for restoration of wooden churches. A 14-day expedition (May 28-June 10, 2022) was conducted to collect samples. The total length of the route was about 3500 km. Stands up to 300 years old were found and eight living tree plots were established. Most importantly, the samples were taken in the relatively forestless part of the Russian Plain, which expands the geographic coverage of the dendrochronological network to the south and will contribute both to the study of the climatic response of tree-ring parameters in these areas and to the developent of new composite chronologies. A total of 790 samples were collected during the second year of the project: 426 samples from living trees (10 collections), 289 samples from wooden structures (18 collections) and 75 archaeological samples (10 collections). 20 tree-ring chronologies were built for different tree species (pine, spruce, oak) with a duration of 100 years or more. During the second year of the project, samples were prepared and optical density was measured (optical density of late wood BI and the difference between optical density of late and early wood δBI) for 181 architectural wood cores from the Arkhangelsk, Novgorod, Vologda, Kostroma and Yaroslavl regions. Optical density was also measured for the first 10 archaeological samples from the Kostroma and Vologda regions. To build chronologies of cellular parameters, 23 microsections were measured for sample A22A21A. For each annual ring, we received data on 36 parameters. The length of the chronologies is132 years - 1889-2020. For the same sample plot, a chronology of stable carbon isotopes was obtained based on the measurement of 183 samples from individual tree rings and covering the period 1900–2020. Correlation analysis showed that most of the cellular parameters are characterized by significant positive correlations with the scPDSI index (r=0.36-0.47, p<0.05) during the growing season of the current year, from May to September. Significant positive correlations were obtained mainly for precipitation in May of the current year (r=0.34-0.43) and temperatures in January and February of the current year (r= from -0.25 to -0.45). We also compared the new C13 carbon isotope chronology with average monthly precipitation and air temperature. The analysis revealed a negative relationship with June and July temperatures, as well as a positive relationship with May, June and July precipitation. The relationship with the PDSI dryness index also turned out to be statistically significant for July, August and September. In general, the results obtained indicate the presence of a relationship with moisture availability. Further work to measure more series is likely to increase the tightness of the relationship. An analysis of the climatic response in the chronologies of the optical density of wood revealed the presence of a strong temperature signal in the chronologies located in the north. Thus, the optical density obtained for the territory of Solovki, the north of the Arkhangelsk and the east of the Leningrad regions depends on the variability of summer temperature (r = 0.2-0.55). At the same time, the highest correlations are observed in two pine chronologies with July, I24S and O02S from the Arkhangelsk and Leningrad regions, and reach the values r = 0.54 and 0.52. The only chronology of optical density in which no response to summer temperatures was found was A22SD, located to the south of all the other studied chronologies, in the Yaroslavl region. Three chronologies the width of annual rings BelBog, C04S, N13 from Solovki, Vologda, Novgorod regions have a positive response to winter air temperature.The value of the correlation coefficients with precipitation variability is much lower than with temperature and varies from 0.18 to 0.34. In 2022, dendrochronological analysis of samples of subfossil oak wood from alluvial deposits of the Sura River was continued. Fifty-seven new accessions were studied, some of which were identified as seven new trees, and some were included in the trees identified last year. This, in particular, forced us to reconsider the composition of one of the floating tree-ring chronologies built last year. At the present stage, the total number of studied subfossil individual trees from the Sura River is 56. Historical records of weather and climate were searched for two continuous historical periods: the entire 17th century and the first third of the 19th century (1801-1829). The main stage of the work consisted in collecting and presenting source information on the history of climate in Russia in the 17th century. This work included identifying relevant historical sources; systematic review of historical sources in search of news about climatic, weather, geological and astronomical phenomena; identification of relevant news; localization and dating of discovered news; a critical assessment of the information contained in the news; entering data into the table and (if necessary) drawing up an expert commentary. In the course of the work, tens of thousands of pages of historical sources of various types were systematically reviewed: narrative (chronicles, chronicles), office work (business correspondence, diplomatic, administrative and military documentation) and sources of personal origin (diaries and notes of contemporaries), as well as hundreds of pages of reference and scientific literature. The key result of the work done on the study of written sources of the 17th century was the identification and presentation in a structured tabular form of more than 300 evidence. The results of working with newspapers of the first third of the 19th century were a table with printed mentions of weather events in the news and an archive of photographs of daily meteorological observations. In total, 476 mentions of weather events were recorded for this period. Work was also carried out on the development of existing methods for the reconstruction of climatic indicators based on tree-ring data. The self calibrated Palmer drought severity index (scPDSI) and summer temperature were chosen as the considered climatic indicators. Four methods of spatial reconstruction were considered: Ensemble Point-by-point Regression (EPPR), Linear Principal Component Regression (lm), Artificial Neural Networks (ANNs), and Modified Pointwise Regression using Partial Least Squares Regression (PPR+PLSR). The last two methods were developed as part of this study. To test the proposed methods, the territory of North America was chosen, since it has the largest number of tree-ring chronologies, as well as the results of reconstructions by other authors. The results of scPDSI and temperature reconstruction based on pseudo-proxy simulation showed the superiority of the method (PPR+PLSR), while ANN showed the worst results. The linear model, which is actually a one-layer, one-node version of ANN, is worse than EPPR for scPDSI, but better for temperature, which is most likely due to the more continuous spatial distribution of temperature values, which means it is easier to describe in terms of EOF. In 2022, 6 papaers on the subject of the project were published, including 4 articles in Q1, one article in Scopus and one article in the RSCI. The declared indicators for publications were overfulfilled by 1.4 times a year before the end of the project. Three reports were presented at one international and one Russian conference.

 

Publications

---