See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/328205347

A 781-year oak tree-ring chronology for the Middle Ages archaeological dating
in Maramureș (Eastern Europe)

Article  in  Dendrochronologia · October 2018

DOI: 10.1016/j.dendro.2018.10.006

CITATIONS READS

0 112

4 authors:

Nechita Constantin Olafur Eggertsson

Forest Research and Management Institute / Stefan cel Mare University of Suceava Iceland Forest Service
14 PUBLICATIONS   97 CITATIONS    60 PUBLICATIONS   531 CITATIONS   

SEE PROFILE SEE PROFILE

Ovidiu Badea Popa Ionel

National Research and Development Institute in Forestry ”Marin Dracea” National Research and Development Institute in Forestry ”Marin Dracea”
60 PUBLICATIONS   477 CITATIONS    63 PUBLICATIONS   615 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Flakaliden View project

Evaluation and analysis of forest ecosystems from Romania in the pan-European (Level I) and intensive (Level II) grids of forest monitoring View project

All content following this page was uploaded by Nechita Constantin on 15 December 2018.

The user has requested enhancement of the downloaded file.

 Dendrochronologia 52 (2018) 105–112

Contents lists available at ScienceDirect

Dendrochronologia
journal homepage: www.elsevier.com/locate/dendro

A 781-year oak tree-ring chronology for the Middle Ages archaeological T

dating in Maramureș (Eastern Europe)
⁎
Constantin Nechitaa,c, Olafur Eggertssonb, Ovidiu Nicolae Badeaa, Ionel Popaa,
a
National Institute for Research and Development in Forestry "Marin Drăcea", Calea Bucovinei, 73 bis, 725100, Câmpulung Moldovenesc, Romania
b
Icelandic Forest Research, Mógilsá, IS-162 Reykjavík, Iceland
c
Department of Geography, Universităţii 13, 720229, Ștefan cel Mare University of Suceava, Romania

A R T I C LE I N FO A B S T R A C T

Keywords: From 1997 to the present, a sustained project in the Maramureș region in Romania was completed with the
Tree-ring chronology construction of a 781-year oak tree-ring chronology. A total of 395 samples from living trees and 429 from
Quercus archaeological wood were analysed with dendrochronological methods. The study aimed to provide the scien-
Sapwood tific community with a new oak chronology that could be applied in dendroarchaeology, dendroclimatology and
Mean segment length
dendroecology studies and also for interpreting past socio-economic events. Furthermore, we studied the number
Average growth rate
of sapwood rings and the growth pattern for different time periods. The chronology covered the continuous
period of A.D. 1236‒2016. From the 824 samples collected, we separated 271 tree-ring series into a new
chronology (A.D. 1406–2016), which fulfilled all the criteria necessary to reconstruct past climate and en-
vironmental changes. The resulting new tree-ring chronology indicated robust signal series intercorrelation
(r = 0.55) and an average mean sensitivity of 0.21. Based on an analysis of the sapwood, we recommend es-
timating a number between 8 and 32 rings to the date of final ring for 95% confidence range of mean on the for
felling in the Maramureș region. Additionally, we observed that the number of sapwood rings was not correlated
with altitude or oak tree species.

1. Introduction palaeoclimate reconstruction (Mann, 2007; Büntgen et al., 2010). Sig-

nificant information from the chronologies may also be retained for the
Two multi-millennium tree-ring chronologies exist in Europe cov- indirect assessment of anthropogenic activity and human susceptibility
ering most of the Holocene or even longer: the 10,429 (8480 BCE) year (Büntgen et al., 2011; Allué et al., 2017). The millennial chronologies
oak and pine chronology from Germany (Friedrich et al., 2004) and the have also been successfully applied to reconstruct geomorphological
7272-year oak chronology built in Ireland (Pilcher et al., 1984). During processes, soil evolution (Compostella et al., 2013) and historical events
the last decades, several multi-centennial oak chronologies have been (Trouet et al., 2017).
published in Europe, including Slovenia (Čufar et al., 2008a, b, c), Northern Romania represents a specific case in the region of
Hungary (Kern et al., 2009), France (Tegel et al., 2010), Portugal (Leal southeast Europe in respect to the response of oak radial growth to
et al., 2015), Spain (Souto-Herrero et al., 2017), Ukraine (Netsvetov climate parameters (Nechita et al., 2017). Oak growth in Maramureș is
et al., 2017), Poland (Pritzkow et al., 2016) and the Czech Republic controlled by precipitation from late spring to early summer and, on the
(Prokop et al., 2017). The potential of prolonging the multi-millennial opposite side of the Carpathian Chain, by temperature from winter to
oak chronologies in Europe has been highlighted in several studies late spring (Nechita et al., 2017). This feature was one of the primary
(Krąpiec, 2001; Pearson et al., 2014; Rădoane et al., 2015). reasons for the construction of a long oak chronology for the Mar-
The importance of building such large databases with tree-ring amureș region, although the area is relatively restricted, it can have
measurements for historical periods is primarily initiated by den- applicability in dendroclimatology and dendroarchaeology studies. The
drochronological dating studies of wooden objects (Ważny et al., 2014). region offers excellent conditions for developing a regional chronology
The evaluation of the quality of the historical chronologies for climatic for NW Romania, using wood from archaeological, historical, ethno-
reconstruction is commonly discussed and debated (Büntgen et al., graphical wooden monuments and subfossil wood that exists in large
2008; Tegel et al., 2010), although they are often applied for deposits along the Iza River. The abundant of old wooden monuments

⁎
Corresponding author.
E-mail address: [email protected] (I. Popa).

https://doi.org/10.1016/j.dendro.2018.10.006
Received 28 May 2018; Received in revised form 5 October 2018; Accepted 8 October 2018
Available online 10 October 2018
1125-7865/ © 2018 Published by Elsevier GmbH.
C. Nechita et al. Dendrochronologia 52 (2018) 105–112

in this region can be highlighted using dendrochronological dating, countries and N-America, especially the sessile oak from the Rona re-
because information is lacking about the time of construction of many gion, which has high economic value (Horj, 2006).
of them. The material provided can also improve the existing chron-
ology by increasing the replication and therefore the quality of the 3. Material and methods
chronology.
In Maramureș, the constructing of wooden churches is an old tra- In this study, we collected samples from living trees and historical
dition. UNESCO recognizes eight of these churches (in total 43) as constructions. The dendrochronological material from living trees was
World Heritage Sites. The wooden churches in Bârsana, Budești Josani collected from old-growth oak (Quercus robur L.) and sessile oak
and Ieud were dated using an oak chronology constructed at the end of (Quercus petraea (Matt.) Liebl.) from eight sites (Table 1).
the 20th century (Eggertsson and Babos, 2003). The “Church on the From each site, 20 to 50 trees were cored randomly with a Pressler
Hill” in Ieud, is the oldest church, which is dated to 1364. The precise borer following a standard dendrochronological methodology (Cook
dating information established regarding the felling date of timber used and Kairiukstis, 1990; Speer, 2010). Ultimately, a database summar-
for the different parts of the buildings is possible only by the use of the izing 395 samples was obtained representing one core from each tree
dendrochronological dating method, which indicates the importance of (Fig. 1a).
this type of research. Two sites with both oak and sessile oak were investigated (Borlești
Part of this historical wood is sold to various regions of the world in and Coștui), to obtain additional information regarding interspecies
which the law requires that the restoration of monuments must be compatibility for the newly formed chronology, because in terms of
performed with wood from periods close to those of the building con- wood anatomy, the species could not be separated. Most of the sam-
struction. In Maramureș, much of the historical monuments require pling sites were chosen from the central depression of Maramureș, be-
restoration. By highlighting the importance of old wood through den- cause the building timber was most likely provided from nearby forests
drochronological techniques, we demonstrated to the regional autho- (Fig. 1b). Additionally, four sampling sites located on the eastern cliff of
rities and national decision makers the importance of providing funds the volcanic Mountains of Maramureș were included in the chronology
for the conservation and restoration of the wooden monuments. to increase the area of representativity (Nechita et al., 2017).
To extend the tree-ring chronology back in time, we also collected
2. Study area samples from historical buildings located in the region. The timber for
the extension of the Maramureș regional oak tree-ring chronology ori-
The study area is located in the northern part of Romania, i.e in the ginated from old historical buildings (houses, barns and mills), included
Romanian part of the historical Maramureș region, (Fig. 1b). The cli- actually in the patrimony of the Maramureș museum in Sighetul
mate is characterized by Atlantic influence with a mean annual pre- Marmaţiei (Fig. 1c). The sampled buildings represented the oldest and
cipitation of 770 mm/year and mean annual temperature of 5.5 °C well-preserved monuments in the region. The samples from wooden
based on average values extracted from CRU TS3.23 (0.5° x 0.5°) for the constructions were collected in 1997 and 2016. In most cases, samples
region (Harris et al., 2014). Predominant tree species are pedunculate were obtained by coring, but we also collected larger samples using a
oak (Quercus robur L.), sessile oak (Quercus petraea (Matt.) Liebl.), chain saw. At present, a market exists for old wood. Accordingly, we
European beech (Fagus sylvatica L), European hornbeam (Carpinus be- also collected wood from various private old wood stacks that were
tulus L) and sweet chestnut (Castanea sativa Mill.). From the first land open for purchase.
survey and mapping of the region it is known that the natural oak sites The wooden material was processed in the dendrochronological
have been heavily exploited for economical interest (1769–1773 laboratory at the National Research and Development Institute for
Franzisco-Josephinische Landesaufnahme) (Sabau, 2016). Oaks and Silviculture, Marin Drăcea, using standard methods (Baillie, 1982;
beech were clear‐cut in the mid -1800s to provide wood for the Austro- Speer, 2010). After preparation, the wooden materials were scanned at
Hungarian Empire, and later wood was sold to other European 3200 dpi (interpolated from 2400 true dpi) using an Epson Expression

Fig. 1. (a) The temporal samples distribution

from living trees and historical wood. (b)
Geographical positions of the study sites. (c)
Examples of living, historical wood and
method of obtaining samples from historical
buildings (d) Full dataset Maramureș oak TRW
chronology.

106
C. Nechita et al. Dendrochronologia 52 (2018) 105–112

Table 1
Descriptive statistics of the living raw tree-ring width chronologies.
Site names Species Long. E Lat. N Series intercorrelation Average mean sensitivity No. trees Length (yrs.) RW (mm/yr.)

Baia Mare Q. robur 23°39′ 47°39′ 0.62 0.30 45 1831-2011 1.30

Ocna Șugatag Q. robur 23°56′ 47°47′ 0.40 0.26 40 1809-2016 1.86
Rona de Sus Q. petraea 24°1′ 47°52′ 0.52 0.22 23 1773-2016 1.68
Săpânţa Q. robur 23°42′ 47°59′ 0.68 0.26 53 1846-2015 1.74
Satu Mare Q. robur 22°47′ 47°49′ 0.62 0.24 43 1882-2011 2.39
Valea Stejarului Q. robur 23°57′ 47°53′ 0.42 0.22 31 1835-2016 1.34
Borlești Q.robur 23°16′ 47°41′ 0.65 0.22 44 1870-2011 2.03
Borlești Q. petraea 23°16′ 47°41′ 0.60 0.24 54 1785-2011 1.57
Coștui Q. robur 24°4′ 47°52′ 0.49 0.21 22 1827-2016 1.49
Coștui Q. petraea 24°4′ 47°52′ 0.50 0.20 40 1853-2016 2.13

12000XL flatbed scanner and measured to the nearest 0.001 mm using over 20-year windows lagged by 10 years, using ARSTAN software
image analysis program Cybis™ CooRecorder and CDendro software (Cook and Krusic, 2005). Individual raw measurements were detrended
(Larsson and Larsson, 2017). CooRecorder 9.0 allowed us to import an using ARSTAN software by applying a 67% cubic smoothing spline with
image of the wood sample, identify the ring boundaries and measure a 50% cut-off frequency to remove effects of disturbance events and
the tree-ring parameters. stand competition. The autocorrelation was removed by applying a
Visual and statistical crossdating were both performed and checked filter to the detrended measurements. For removing biases caused by
using TSAP-Win (Rinn, 2010) and COFECHA (Holmes, 1983) software. extreme values, we used bi-weight robust mean to obtain a mean
The regional chronology was compiled from living trees and historical chronology.
wood. The chronology was constructed from oaks samples (combined
both sessile oak and pedunculate oak) with the best visual pattern-
matching and statistical verification results. First, we crossdated his- 4. Results
torical wood internally where multiple samples were available from the
same site/building, since for some cored buildings the date of con- 4.1. The Maramureș oak tree-ring chronology
struction/reconstruction is known. The chronology construction pro-
gresses step by step with caution, using dating checks such as: iterative In the Maramureș (MM) a set of 824 samples were studied, and the
replacement and removal of series to test the stability of the statistical absolutely dated samples resulted in a 781-year chronology covering
results, comparison with others regional tree-ring chronologies. For the time span 1236–2016 suitable for dendrochronological dating
dating quality control and teleconnection assessment, the graphical- (Fig. 1a, d). Summary chronology descriptive statistics were presented
chronology match was statistically verified using the common statistical in Table 2. The length of individual tree-ring series cored both dated
parameters: coefficient of coincidence Gleichläufigkeit (Glk), correla- and undated varied between 23 and 293 years with the mean length of
tion coefficient (ccoeff), t-value, t-value according to Hollstein, Baillie the individual tree-ring series of 125 years. The statistical parameters in
and Pilcher, and cross-date Index (CDI) (Knibbe, 2004; Rinn, 2010; several cases were unsatisfactory for the chronology to be utilized later
Baillie and Pilcher, 1973). as a proxy for climate reconstruction. Altogether, 553 samples re-
The Glk represents the percentage slope equivalence of a given presenting over 67% of the total data set could not be successfully used
sample and reference with the overlapping of both series (Eckstein and in the new tree-ring chronology, because of low correlation. These in-
Bauch, 1969). The correlation coefficient is a percentage of overall si- dividual tree-ring series had to be excluded from the final MM chron-
milarity between two series. When considering the length (N) of over- ology. Therefore, we eliminated individual samples with weak corre-
lapping series, the correlation coefficient becomes more valuable and is lation with the mean chronology (below 0.4).
incorporated in a t-test. We also used the t-values calculated after de- Based on these criteria, we extracted a subset of tree ring width
trending, as proposed by Hollstein (1980), to exclude any age trend that (TRW) data and we created a chronology from 271 wood samples, in-
might significantly affect the resulting statistics. Tree-ring series were cluding 134 living trees and 137 historical wood samples, covering the
graphically matched to one another using statistical parameters; at the 611-year period between 1406 and 2016, including only the samples
end, they were averaged to produce a mean sample chronology. CDI is a that could be strongly correlated (Fig. 2a, b). Several descriptive sta-
coefficient calculated in TSAPWin© that includes Glk and t-values cal- tistics commonly used in dendrochronology were calculated for subset
culated with different detrending options (Rinn, 1996). of tree-ring width data to evaluate the chronology strength. These in-
Haneca et al. (2005) proposed t-test scores ranging between 5.0 and cluded the intercorrelation of the mean series (SI = 0.55) which is a
7.5 resulting in an acceptable t-test threshold of 6.0 for matching tree- measure of the strength of the common signal to sampled trees at the
ring series to a reference chronology. However, Sass-Klassen et al. site.; mean sensitivity (MS), an indicator of the mean relative change
(2008) in their work for establishing new reference chronologies for the between consecutive ring widths showed values of 0.21.
dendroprovenancing of timber indicated the Glk must be at least 50% to The newly established regional chronology fulfils all the constraints
have any statistical significance, in the tree-ring series matching. In required for use in climate reconstruction for the Maramureș region
addition, an empirically tested 99.9% confidence threshold (tBP > 3.7) (Buras, 2017): EPS > 0.85, different age classes occurred, and the
can be also considered as a useful guidance in evaluation of the cross-
dating results and teleconnection pattern (Fowler and Bridge, 2017). Table 2
For our reference chronologies, several overlapping records were Full dataset chronology summary.
relatively short but the correlation was high, which explained why we Sub-sets No of series Start End Period MSL AGR Rbar EPS
carefully analysed the t-test and the other statistical parameters men-
tioned to justify good matches. At the final stage, absolutely dated in- Living 395 1773 2016 244 73 1.60 0.13 0.94
Historical 429 1236 1999 764 52 1.36 0.11 0.81
dividual tree-ring series were included into a regional chronology. To All 824 1236 2016 781 62 1.54 0.29 0.94
assess the quality of the chronology, expressed population signal (EPS)
(Wigley et al., 1984) and inter-series correlation (Rbar) were computed *mean segment length (MSL), average growth rate (AGR), inter-series correla-
tion (Rbar) and the Expressed Population Signal (EPS).

107
C. Nechita et al. Dendrochronologia 52 (2018) 105–112

Fig. 2. (a) Sub-dataset Maramureș TRW

chronology indices (blue) and sample depth
(black); (b) blue line – TRW for living trees and
sample depth; red line – TRW for historical
wood and sample depth; black rectangle in-
dicates periods with low replication (minimum
of 14 series); (c) highlighting the common
coverage period, respectively the 18th century
with cross-dating statistics; blue line – living
trees; red line – historical wood; line – TRW;
dotted line – sample depth; (d) red line - the
expressed population signal (EPS); blue line -
the inter-series correlation; both EPS and Rbar
were calculated over 20-year windows lagged
by 10 years (For interpretation of the refer-
ences to colour in this figure legend, the reader
is referred to the web version of this article.).

correlation of living trees with climate was high and stable over time, 1987) and 95% confidence interval of mean falls between 8 and 32
according to Nechita et al. (2017). The Rbar value of the new chron- rings. The average value for all sites in the region was 14 sapwood
ology ranged from 0.16 (1566 y) to 0.36 (1816 y) (Fig. 2d). The signal rings, both for sessile and pedunculate oak. A slightly lower number of
to noise ratio (SNR) varied between 5.91 and 58.5. The chronology was sapwood rings (12 rings) was identified for sessile oak located at lower
replicated with at least 14 samples, with that lowest replication be- elevations when individual sites were analyzed. The total number of
tween 1780 and 1820 and the maximum replication in the 20th century rings counted for living trees from bark to pith varied between 86 and
of 135 samples. t-values calculated using the Baillie-Pilcher (tBP) and 244 rings, and the average site value varied between 118 and 220 rings
Hollstein (tH) algorithm highlight the correlation strength between (Fig. 3b).
chronologies represented by living and historical material in their
overlapping period (Fig. 2c).
4.3. Age related growth trends based on the MM dataset

4.2. Sapwood rings of oaks in NW Romania For the full dataset of living trees, age-related ring width trends
indicated a brief increase in ring width over time (Fig. 4). For the first
We analysed the average number of sapwood rings in the full da- 50 years, the trend of juvenile wood was almost the same for both the
taset of living tree samples. The results indicated that the number of living and the historical wood. Between 50 and 150 years from first tree
sapwood rings ranged from 6 to 40, with an average for each sampling ring was observed a linear trend of growing both in living and historical
site between 10 and 16 rings (Fig. 3a). The distribution of the number wood. The only difference in growth trend observed was an increasing
of sapwood rings conforms to a log-normal distribution (Hillam et al., trend after 150 years in living trees for a small period, followed by a

Fig. 3. (a) Distribution of sapwood rings of

living trees for Maramureș region (Romania);
(b) Age distribution of living trees samples.
Graphical representation includes diamond
box-style with percentile ranges of 10, 25, 75,
and 90; whisker range-confidence interval with
a probability of 95%.

108
C. Nechita et al. Dendrochronologia 52 (2018) 105–112

Fig. 4. Square symbol: age versus average growth rate (green, historical wood; blue, living trees). Continuous line: mean average growth in report with year starting
from the first tree ring (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.).

downward trend. A similar upward trend was not observed in the his- were included in the new Maramureș (MM) oak TRW chronology sui-
torical material (Fig. 4). table as proxy. A total of 553 samples collected from both living and
historical wood samples were excluded from the proxy chronology.
4.4. Teleconnections of MM chronology These samples were excluded for multiple reasons (low number of tree
rings, low correlation with mean chronology). Some samples of his-
Comparison of MM chronology with regional chronology of SE torical tree-ring data, were excluded because of an insufficient number
Slovenia (Čufar et al., 2008a, b) (site #12) indicate a moderate corre- of tree rings. The suggestion of Cook and Kairiukstis (1990) is not to
lation value (tv = 5.5 / tBP = 4.6 / tH = 4.8). However, using an up- include samples with less than 40 rings per sample. Other studies
dated chronology of Slovenia (Čufar et al., 2014 unpublished), we can propose the minimum of even 50 tree rings per sample (Baillie, 1995;
observe higher t values (tv = 11.2 / tBP = 5.7 / tH = 6.4) (site #13) Haneca et al., 2009). Experience with precision dating of art objects
(Table 3.). attests to the fact that even time series with more than 100 tree rings
Teleconnection statistics with living tree ring chronologies from SE may not be sufficient for successful dating (Horovitz, 2012). In our case,
Europe highlight wide range of variability (Table 3). The area covering only 18 individual series contained less than 40 tree rings, and 35
the sites with tBP / tH > 6 is indicative of the region where newly MM contained less than 50 rings, which was less than 4% of the entire data
updated chronology has the highest potential to be used as a reference set. However, the abundant historical timber and subfossil wood of
chronology. Outside this core region (including sites #1, #4, #8, #9), different ages in the region offer plausible opportunities to extend the
the dating potential of MM is lower. While the areas with tBP < 3.7 chronology further back in time and to improve the replication.
(including sites #5, #6, #10, #11) are already outside the zone were Several historical samples with more than 200 tree rings had low
MM can be used as a reference chronology (Fig. 5). crossdating statistics. The explanation might be the provenance of
timber outside the Maramureș territory or more likely that the samples
were ancient. Some beams might be from older structures and then re-
5. Discussion
used in a younger building. Several holes, carving or joining stem was
observed in different position of the cored beams compared to the
5.1. The Maramureș oak tree-ring chronology
current one which provide structural evidence to support the hypoth-
esis. Historical documents indicate a long presence of civilization in the
In this study, we collected and processed 824 samples resulting in
study area (Drimba, 1987; Filipascu, 1940; Ivanciuc, 1998). The tu-
103,371 rings measured and dated, resulting a 781-year long chron-
multuous historical wars, burns and destructions were the main factor
ology suitable for dendrochronological dating. A total of 271 samples

Table 3
Teleconnection between Maramureș oak chronology and living oak chronologies from SE Europe / regional oak chronologies.
No Location Length (year) Date end Glk tv tBP tH Source

1 Serbia – Sremska Mitrovica 163 2003 64***

0.7 4.5 4.7 Čufar et al. (2014)
2 Hungary - Baktalórántháza 278 2008 64*** 4.0 6.0 6.2 Kern et al. (2013)
3 Hungary - Debrecen 221 2009 65*** 6.4 5.7 6.4 Árvai et al. (2018)
4 Romania -Târgu Neamţ 216 2009 59*** 11.1 4.1 4.8 Nechita (2013)
5 Romania - Caraorman 282 2006 55 0.3 2.6 2.4 Nechita (2013)
6 Romania – Roșiorii de Vede 166 2009 54 4.7 2.9 3.4 Nechita (2013)
7 Romania - Cașolţ 225 2008 66*** 6.4 6.7 6.4 Nechita (2013)
8 Ukraine - Czerniatyn 197 2009 59*** 2.0 5.4 4.8 Ważny et al. (2014)
9 Ukraine - Severynivka 217 2009 63*** 3.6 5.0 4.5 Ważny et al. (2014)
10 Bulgaria - Zvezdec 223 2009 55 13.0 3.1 3.5 Ważny et al. (2014)
11 Bulgaria – Shumensko Plateau 115 2009 64** 0.8 2.6 3.3 Ważny et al. (2014)
12 SE Slovenia 548 2003 61*** 5.5 4.6 4.8 Čufar et al. (2008a, b, c)
13 SE Slovenia 853 2003 59*** 11.2 5.7 6.7 Čufar et al. (2008a, b, c),(2014) unpublished
14 East Austria 777 2016 58*** 3.7 7.3 6.6 Grabner et al. (2014)
15 North Central Hungary 626 1995 58*** 7.3 7.4 7.3 Grynaeus, unpublished

*Glk – signature Gleichläufigkeit; tv – t-value computed without any standardization; t-values calculated using the Baillie-Pilcher algorithm (tBP); t-values calculated
using the Hollstein algorithm (tH). *#12 - #15 represent Regional European Oak Chronology.

109
C. Nechita et al. Dendrochronologia 52 (2018) 105–112

Fig. 5. Map of SE Europe with teleconnection

between Maramureș oak TRW chronology and
local oak chronologies from both sides of
Carpathians. Sites chronologies corresponding
to the site numbers of the map, listed in the
Table 3. The numbers assigned to lines re-
present tH-values; tH > 3.7 was marked with
red (For interpretation of the references to
colour in this figure legend, the reader is re-
ferred to the web version of this article.).

of periodic reconstruction of the buildings. Growth anomalies and Chain, induces a local pattern on the response of trees to climate
narrow rings were also reasons for excluding samples from the proxy (Nechita et al., 2017). At a large scale, the former MM chronology was
chronology. Several other studies highlight similar problems (Haneca strongly correlated (tH = 9.6) with East Austrian Oak Chronology re-
et al., 2009; Kolář et al., 2013; Prokop et al., 2017). ference (Grabner et al., 2014) at a distance of almost 600 km, but
The resulting chronology will have a potential to be used in pa- showed a lower correlation (tBP = 8.61) in the comparison with the
leoclimate reconstruction owing the robust common signal indicated by much closer Czech oak chronology (Prokop et al., 2017). The compar-
the EPS (Buras, 2017) and the well-expressed and time stable correla- ison between updated MM oak chronology and East Austrian Oak
tion between oak radial growth and precipitation found in this region Chronology (site #14) indicate tBP = 7.3 (tH = 6.6). Furthermore, the
(Nechita et al., 2017). The quality of the chronology could be improved North Central Hungary oak chronology (site #15) located between
at the end of 18th and early the 19th century when the EPS reached the Austria and Maramureș indicate almost the same indices tBP = 7.4
lowest thresholds of 0.87 and 0.86, respectively, by adding more his- (tH = 7.3), justifying the spatial stability of teleconnection.
torical wood series. The parameters of the chronology indicated robust According with tBP -value (3.1) was observed insignificant correla-
signal strength as analysed by statistical parameter values: inter-series tion with oak chronology from Bulgaria - Zvezdec (#10), located in
correlation (Rbar) and the expressed population signal (EPS) (Fig. 2b) protected nature reserve natural forest. Contrary to our expectations,
(Wigley et al., 1984). Additionally, we noted the existence of subfossil correlation of oak chronology from Sremska Mitrovica, Serbia, (Čufar
oaks imbedded in the riverbed of the Iza River, which indicate the et al., 2014) with MM chronology show significant values for tBP (4.5)
potential to extend the chronology further back in time. and tH (4.7). We noticed that comparing with South Austria and Hun-
Two widespread species, pedunculate oak and sessile oak, pre- gary the correlation is lower, but even for this area MM chronology has
dominate within the oak forests of Maramureș. However, both species realistic potential for being used as a reference chronology.
could not be distinguished from each other by anatomy Ważny et al. (2014) attest to the necessity for constructing a long
(Schweingruber, 1993; Schoch et al., 2004) in archaeological and his- oak chronology in south-eastern Europe for "bridging gaps" to link north
torical wood, and we can only suppose they existed in mixed popula- central European and east Mediterranean tree-ring networks. The new
tions. Feuillat et al. (1997) and Mosedale et al. (1998) attempted to MM chronology come to fill this gap.
distinguish Quercus robur L. and Quercus petraea (Matt.) Liebl. based on Ważny et al. (2014) and Nechita et al. (2017) noticed the role of
the ratio between the number of earlywood vessels and the latewood Carpathians Chain to induce low correlation between eastern and
ratio; however, this approach was not used in this study. Nechita et al. western side of this mountains, also in the south. MM chronology seems
(2017) studied the common climatic signal for both species. The results to cross-date well with SE Poland oak chronology (Krąpiec, 1998) (tBP /
indicated irrelevant differences between the species responses to pre- tH > 3.7), explain by extended influence of North Atlantic air mases in
cipitation in April and June. This information highlights the importance our study area. Even so, tree ring signal can be modified by local
of the MM oak tree-ring chronology, which can be used as a proxy for condition, genetic factors, provenance of the regeneration material,
the reconstruction of past climate and environmental changes for the disturbances or anthropogenic influences, common in lower altitude.
last 611 years.
Several studies, analysing the teleconnection between oak chron-
ologies in Europe, have referred on the former Maramureș oak tree-ring 5.2. Sapwood rings of oaks in NW Romania
chronology in an unpublished form. Information outlined in Kolář et al.
(2012) and Ważny et al. (2014) for the past 600 years corroborates with Commonly, when dendrochronological samples are collected for
that of a previous study on the Maramureș living tree-ring data (Nechita dating from constructions, they do not contain the complete sapwood,
et al., 2017), indicating the necessity of building a long oak chronology because most builders remove a part of the outermost sapwood rings.
in eastern Europe. The primary reason is because of the Carpathians The estimation of missing sapwood tree rings is of great importance in
dendroarchaeological dating, which differ in number from west to east

110
C. Nechita et al. Dendrochronologia 52 (2018) 105–112

Europe (Baillie et al., 1985; Prokop et al., 2017). The number of sap- earliest/latest felling date. For timbers with no sapwood the pos-
wood tree rings for the Maramureș region varied between 6 and 40. sible earliest felling date could be estimated by the addition of 8
Mean sapwood ring number ranged from 10.6 to 16.8. According to our years to the date of the final ring.
results, sapwood tree-ring variability did not depend on altitude or
interspecific characteristics. Special thanks also go to anonymous reviewers which help us to
The number of mean sapwood rings found in our study (n = 14.5) is improve quality of this manuscript, as well as Katarina Čufar, Zoltán
consistent with the results of Prokop et al. (2017) in the Czech Republic Kern and Tomasz Ważny for share with us their chronologies.
(n = 14.6). Compared with other studies in different parts of the Eur-
opean continent such as: Finland and western Estonia (4–21), eastern Acknowledgements
Estonia, Latvia, and Lithuania (6–18) (Sohar et al., 2012), south Finland
(7–24) (Baillie et al., 1985), Ireland (14–62) (Baillie, 1982), the British This work was supported by a grant from the Romanian National
Isles (14–66) (Hillam et al., 1987), north west England and north Wales Authority for Scientific Research and Innovation, CNCS/
(10–55) (Hughes et al., 1981), north England (10–60), south England CCCDI–UEFISCDI, project number PN-III-P2-2.1-PED-2016-1058,
(4–57) (Miles, 1997), north France (12–49) (Pilcher, 1987), Belgium within PNCDI III, “A new technique regarding dendrochronological
(13–39) (Haneca et al., 2009), Germany (7–66) (Hollstein, 1965,1980), dating. Statistical, biological and chemical approach” (DendroTECH).
Hungary (7–36) (Grynaeus, 1998), and Poland (9–36) (Ważny, 1990), EEA Financial Mechanism 2009–2014, contract no. 18SEE “CLIMFOR”
our results confirmed a decreasing trend of the number of sapwood also supported this work. Thanks to Roibu Catalin Constantin and
rings, expressed in absolute range, from west to east. Mursa Andrei for their support and for assistance in fieldwork. Special
thanks to Alexandru Dumitru Babos who initiated the dating work of
5.3. Age related growth trends based on the MM dataset the Maramureș churches.

Differences in growth pattern between living trees and archae- References

ological/historical wood are also reported in other studies. The average
tree-ring width for living trees was 1.60 mm/y for north Romania and Allué, E., Picornell-Gelabert, L., Daura, J., Sanz, M., 2017. Reconstructing of the pa-
only 1.01 mm/y for historical wood. Whereas in Slovakia, the mean leoenvironment and anthropogenic activity from the Upper Pleistocene/Holocene
anthracological records of the NE Iberian Peninsula (Barcelona, Spain). Quat. Int.
growth rate for recent wood is 1.79 mm/y and that for historical wood 457, 172–189.
1.33 mm/y (Prokop et al., 2016). In central Germany (Büntgen et al., Árvai, M., Morgós, A., Kern, Z., 2018. Growth-climate relations and the enhancement of
2010), the difference between growth rate levels and age trends in drought signals in pedunculate oak (Quercus robur L.) tree-ring chronology in Eastern
Hungary. iForest-Biogeosci. For. 11, 267–274.
living trees and archaeological wood is smaller than that in Slovakia Baillie, M.G.L., 1982. Tree-Ring Dating and Archaeology. Croom Helm, London and
and Romania (living = 1.47 mm/y, archaeological = 1.28 mm/y). One Canberra 274 pp.
of the reasons might be the high proportion of juvenile wood in the Baillie, M.G.L., 1995. A Slice Through Time: Dendrochronology and Precision Dating. B.T.
Batsford, Ltd., London 176 pp.
living tree-data set.
Baillie, M.G.L., Pilcher, J.R., 1973. A simple cross-dating program for tree-ring research.
The physiological tree growth models indicate a decreasing trend in Tree-Ring Bull. 33, 7–14.
old aged trees (age trend). We have observed an increase in tree-ring Baillie, M.G.L., Hillam, J., Briffa, K.R., D.M, Brown, 1985. Re-dating the English art-
historical tree-ring chronologies. Nature 315, 317–319.
growth in the old aged trees. This finding is in accordance to previous
Büntgen, U., Frank, D.C., Wilson, R., Career, M., Urbinati, C., Esper, J., 2008. Testing for
studies by D’Arrigo et al. (2008); Esper et al. (2008) and Oberhuber tree-ring divergence in the European Alps. Global Change Biol. 14, 2443–2453.
et al. (2008). This phenomenon might be explained by fertilization Büntgen, U., Trouet, V., Frank, D.C., Leuschner, H.H., Friedrichs, D., Luterbacher, J.,
produced by atmospheric nitrogen and CO2 (Jonson and Abrams, 2009; Esper, J., 2010. Tree-ring indicators of German summer drought over the last mil-
lennium. Quat. Sci. Rev. 29, 1005–1016.
Lu et al., 2010; McMahon et al., 2010). Another explanation and a more Büntgen, U., Tegel, W., Nicolussi, K., McCormick, M., Frank, D.C., Trouet, V., Kaplan,
reasonable may be that the forest environment is different; in earlier J.O., Herzig, F., Heussner, K.U., Wanner, H., Luterbacher, J., Esper, J., 2011. 2500
days, the forest was denser (a pristine forest), whereas the management years of European climate variability and human susceptibility. Science 331,
578–582.
was selective cuttings. Buras, A., 2017. A comment on the expressed populational signal. Dendrochronologia 44,
130–132. https://doi.org/10.1016/j.dendro.2017.03.05.
6. Conclusions Compostella, C., Trombino, L., Caccianiga, M., 2013. Late Holocene soil evolution and
tree line fluctuation in the Northern Apennines. Quat. Int. 289, 46–59.
Cook, E.R., Krusic, P.J., 2005. ARSTAN v. 41d: A Tree-Ring Standardization Program
i) In total, eight sites with living oaks and 28 historical monuments Based on Detrending and Autoregressive Time Series Modeling, With Interactive
were sampled. From the full dataset a 781-year long oak TRW Graphics. Tree Ring Laboratory. Lamont-Doherty Earth Observatory of Columbia
University, Palisades, New York, NY.
chronology was built suitable for dendroarchaeological dating. A Cook, E.R., Kairiukstis, L.A. (Eds.), 1990. Methods of Dendrochronology - Applications in
subset of 271 samples were included in the new Maramureș (MM) the Environmental Sciences. Kluwer Academic Publishers, Dordrecht. https://doi.
oak TRW chronology suitable as climate proxy studies. The time org/10.1007/978-94-015-7879-0. p 394.
Čufar, K., Sekelj Ivančan, T., Grabner, M., 2008a. The dendrochronological dating of
span of the proxy chronology is 611 years.
wood from the site of Torčec–Gradić in northern Croatia. Podravina 7 (13), 30–40.
ii) The newly formed MM oak chronology offers an opportunity to Čufar, K., de Luis, M., Eckstein, D., Kajfež-Bogataj, L., 2008b. Reconstructing dry and wet
unveil past climate and environmental changes back to 1406 CE. summers in SE Slovenia from oak tree-ring series. Int. J. Biometeorol. 52, 607–615.
The chronology can also be used to date cultural objects and in- Čufar, K., De Luis, M., Zupančič, M., Eckstein, D., 2008c. A 548-year tree-ring chronology
of oak (Quercus spp.) for southeast Slovenia and its significance as a dating tool and
terpret ethnographic and social events of the past. climate archive. Tree-Ring Res. 64 (1), 3–15.
iii) The teleconnection with other chronology from nearby indicates Čufar, K., Šefc, B., de Luis, M., Morgós, A., Grabner, M., Merela, M., Trajković, J., 2014.
strong pattern with East Austrian, North-Central Hungary, South- Tree-ring chronology of pedunculate oak (Quercus robur) and its potential for de-
velopment of dendrochronological research in Croatia. Drvna industrija 65 (2),
East Slovenia and Ukraine oaks chronologies. Low correlation 129–137.
coefficients were obtained with local chronologies of living trees D’Arrigo, R., Wilson, R., Liepert, B., Cherubini, P., 2008. On the ‘Divergence Problem’ in
from Bulgaria and South and SE Romania. Northern Forests: a review of the tree-ring evidence and possible causes. Global
Planet. Change 60, 289–305.
iv) Finally, as a highlight, the number of sapwood tree rings in Drimba, O., 1987. Istoria culturii și civilizaţiei românești. Editura Știinţifică și
northern Romania (Maramureș) coincided with that of studies Pedagogică, Bucuresti 2 404 pp (in Romanian).
conducted in different regions of Europe. At present the best Eckstein, D., Bauch, J., 1969. Beitrag zur Rationalisierung eines dendrochronologischen
Verfahrens und zur Analyse seiner Aussagesicherrheit. Forstwiss. Centralb 88,
available estimate is 14 sapwood rings (95% confidence limit: 230–250.
8–32) for the Maramures oaks. Consequently, for timbers with in- Eggertsson, O., Baboș, A.D., 2003. Dendrochronological Dating in Maramures with
complete sapwood 14 rings can be assumed to estimate the likely Special Emphases on Objects from the Maramures Museum in Sighetu Marmatiei.

111
C. Nechita et al. Dendrochronologia 52 (2018) 105–112

Muzeul Maramuresului, Sighetu Marmatiei, Romania, pp. 40–49. (10), 2688–2700.

Esper, J., Niederer, R., Bebi, P., Frank, D., 2008. Climate signal age effects-evidence from Mann, M., 2007. Climate over the past two millennia. Annu. Rev. Earth Planet. Sci. 35,
young and old trees in the Swiss Engadin. For. Ecol. Manag. 255, 3783–3789. 111–136.
Feuillat, F., Dupouey, J.L., Sciama, D., Keller, R., 1997. A new attempt of discrimination McMahon, S.M., Parker, G.G., Miller, D.R., 2010. Evidence for a recent increase in forest
between Quercus petraea and Quercus robur based on wood anatomy. Can. J. For. Res. growth. Proc. Natl. Acad. Sci. 107 (8), 3611–3615.
27, 343–351. Miles, D., 1997. The interpretation, presentation and use of tree-ring dates. Vernac. Arch.
Filipascu, A., 1940. Istoria Maramuresului. Editura Universul, Bucuresti 270 pp (In 28, 40–56.
Romanian). Mosedale, J.R., Feuillat, F., Baumes, R., Dupouey, J.L., Puech, J.L., 1998. Variability of
Fowler, A.M., Bridge, M.C., 2017. Empirically-determined statistical significance of the wood extractives among Quercus robur and Quercus petraea trees from mixed stands
Baillie and Pilcher (1973) t statistic for British Isles oak. Dendrochronologia 42, and their relation to wood anatomy and leaf morphology. Can. J. For. Res. 28 (7),
51–55. 994–1006.
Friedrich, M., Remmele, S., Kromer, B., Hofmann, J., Spurk, M., Felix, K.K., Orcel, C., Nechita, C., 2013. Rețeaua națională de serii dendrocronologice pentru stejar și gorun.
Küppers, M., 2004. The 12.460 year Hohenheim oak and pine tree-ring chronology Editura Silvică, București, pp. 279 (in Romanian).
from Central Europe – a unique annual record for radiocarbon calibration and pa- Nechita, C., Popa, I., Eggertsson, O., 2017. Climate response of oak (Quercus spp), an
leoenvironment reconstruction. Radiocarbon 46 (3), 1111–1122. evidence of a bioclimatic boundary induced by Carpathians. Sci. Total Environ.
Grabner, M., Wächter, E., Karanitsch-Ackerl, S., Bolka, M., 2014. New regional chron- 599–600, 1598–1607. https://doi.org/10.1016/j.scitotenv.2017.05.118.
ologies for eastern Austria–a basis for dendroprovenancing and dendroclimatology. Netsvetov, M., Sergeyev, M., Nikulina, V., Kornyenko, V., Prokopuk, Y., 2017. The climate
In: García-González, I., Souto-Herrero, M. (Eds.), EuroDendro 2014. Book of ab- growth relationship of pedunculate oak in steppe. Dendrochronologia 44, 31–38.
stracts, Lugo, Spain, pp. 64 8–12 September. Oberhuber, W., Kofler, W., Pfeifer, K., Seeber, A., Gruber, A., Wieser, G., 2008. Long-term
Grynaeus, A., 1998. Sapwood analysis on archaeological remains in Hungary. changes in tree-ring-climate relationships at Mt. Patscherkofel (Tyrol, Austria) since
Stravinskiene, V., Juknys, R. (Eds.), Dendrochronology and Environmental Trends. the mid 1980s. Trees 22, 31–40.
Proceedings of the International Conference 17–21 June 1998 18–21. Pearson, C.L., Ważny, T., Kuniholm, P.I., Botić, K., Durman, A., Seufer, K., 2014. Potential
Haneca, K., Ważny, T., Van Acker, J., Beeckman, H., 2005. Provenancing Baltic timber for a new multimillennial tree-ring chronology from subfossil Balkan river oaks.
from art historical objects: success and limitations. J. Archaeolog. Sci. 32, 261–271. Radiocarbon 51, S51–S59.
Haneca, K., Čufar, K., Beeckman, H., 2009. Oaks, tree-rings and wooden cultural heritage: Pilcher, J.R., 1987. A 700 year dating chronology for northern France. In: Ward, R.G.W.
a review of the main characteristics and applications of oak dendrochronology in (Ed.), Applications of Tree-Ring Studies: Current Research in Dendrochronology and
Europe. J. Archaeolog. Sci. 36, 1–11. Related Subjects, pp. 127–139 BAR International Series 333.
Harris, I., Jones, P.D., Osborn, T.J., Lister, D.H., 2014. Updated high-resolution grids of Pilcher, J.R., Baillie, M.G.L., Schmidt, B., Becker, B., 1984. A 7.272-year tree-ring
monthly climatic observation - the CRU TS3.10 dataset. Int. J. Climatol. 34, 623–642. chronology for western Europe. Nature 312 (8), 150–152.
https://doi.org/10.1002/joc.3711. Pritzkow, C., Ważny, T., Heußner, K.U., Słowiński, M., Bieber, A., Dorado Liňán, I., Helle,
Hillam, J., Morgan, R.A., Tyers, I., 1987. Sapwood estimates and the dating of short ring G., Heinrich, I., 2016. Minimum winter temperature reconstruction from average
sequences. In: Ward, R.G.W. (Ed.), Applications of Tree-Ring Studies: Current earlywood vessel area of European oak (Quercus robur) in N-Poland. Palaeogeogr.
Research in Dendrochronology and Related Subjects, pp. 165–185 BAR International Palaeoclimatol. Palaeoecol. 449, 520–530.
Series 333. Prokop, O., Kolář, T., Büntgen, U., Kyncl, J., Kyncl, T., Bošeľa, M., Choma, M., Barta, P.,
Hollstein, E., 1965. Jahrringchronologische Datierung von Eichenhölzern ohne Rybníček, M., 2016. On the paleoclimatic potential of millennium-long oak ring
Waldkante. Bonner Jahrbücher 165, 12–27. width chronology from Slovakia. Dendrochronologia 40, 93–101.
Hollstein, E., 1980. Mitteleuropäische Eichenchronologie. Mainz, 273 pp. . Prokop, O., Kolář, T., Kyncl, T., Rybníček, 2017. Updating the Czech millennia-long oak
Holmes, R.L., 1983. Computer-assisted quality control in tree-ring dating and measure- tree-ring width chronology. Tree-Ring Res. 73 (1), 47–52. https://doi.org/10.3959/
ment. Tree-Ring Bull. 43, 69–78. 1536-1098-73.1.47.
Horj, P., 2006. Istoria gospodăririi padurilor de pe Valea Vaserului. Ed Saga, Cluj, Rădoane, M., Nechita, C., Chiriloaei, F., Rădoane, N., Popa, I., Roibu, C., Robu, D., 2015.
Romania. Late Holocene fluvial activity and correlations with dendrochronology of subfossil
Horovitz, I., 2012. Copper as a support for easel paintings. In: Stoner, J.H., Rushfield, R. trunks: case studies of northeastern Romania. Geomorphology 239, 142–159.
(Eds.), Conservation of Easel Paintings. Routledge, Abingdon, USA and Canada, pp. Rinn, F., 1996. Computer program for time series analysis and presentation. TSAP.
99–106 ISBN: 978-0-7506-8199-5. Version 3.0. Reference Manual. Heidelberg, 262 pp.
Hughes, M.K., Milsom, S.J., Leggett, P.A., 1981. Sapwood estimates in the interpretation Rinn, F., 2010. TSAP – Reference Manual. Frank Rinn, Heidelberg, Germany.
of tree-ring dates. J. Archaeolog. Sci. 8, 381–390. Sabau, V., 2016. Statistica pădurilor din România pe anul 1929. Bucovina Forestieră 16
Ivanciuc, T., 1998. Noi dovezi privind locuirea Maramuresului din protoistorie. In: In: (1), 117–140. https://doi.org/10.4316/bf.2016.010.
Voda, Dragos (Ed.), Maramures. Vatra de istorie milenară 4. pp. 15–22 Cluj Napoca, Sass-Klassen, U., Vernimmen, T., Baitinger, C., 2008. Dendrochronological dating and
Romania. ISSN 973-98025-8-3. provenancing of timber used as foundation piles under historic buildings in the
Jonson, S.E., Abrams, M.D., 2009. Age class, longevity and growth rate relationships: Netherlands. Int. Biodeterior. Biodegrad. 61, 96–105.
protected growth increases in old trees in the eastern United States. Tree Physiol. 29 Schoch, W., Heller, I., Schweingruber, F.H., Kienast, F., 2004. Wood Anatomy of Central
(11), 1317–1328. https://doi.org/10.1093/treephys/tpp068. European Species. Online version:. www.woodanatomy.ch.
Kern, Z., Grynaeus, A., Morgos, A., 2009. Reconstructed precipitation for southern Schweingruber, F.H., 1993. Trees and Wood in Dendrochronology. Springer-Verlag,
Bakony Mountains (Transdanubia, Hungary) back to 1746 AD based on ring widths of Berlin 402 pp.
oak trees. Idöjárás 113, 299–314. Sohar, K., A, Vitas, A, L.äänelaid, 2012. Sapwood estimates of pedunculate oak (Quercus
Knibbe, B., 2004. Personal Analysis System for Tree-Ring Research 4 - Instruction Manual. robur L.) in eastern Baltic. Dendrochronologia 30 (1), 49–56.
SCIEM, Vienna. Souto-Herrero, M., Rozas, V., García-González, I., 2017. A 481-year chronology of oak
Kolář, T., Kyncl, T., Rybníček, M., 2012. Oak chronology development in the Czech earlywood vessels as an age-independent climatic proxy in NW Iberia. Global Planet.
Republic and its teleconnection on a European scale. Dendrochronologia 30, Change 155, 20–28.
243–248. Speer, J.H., 2010. Fundamentals of Tree-Ring Research. The University of Arizona Press,
Kolář, T., Rybníček, M., Tegel, W., 2013. Dendrochronological evidence of cockchafer Tucson 333 pp.
(Melolontha sp.) outbreaks in subfossil tree-trunks from Tovačov (CZ Moravia). Tegel, W., Vanmoerkerke, Büntgen, U., 2010. Updating historical tree-ring records for
Dendrochronologia 31, 29–33. climate reconstruction. Quat. Sci. Rev. 29 (17–18), 1957–1959.
Krąpiec, M., 1998. Oak dendrochronology of Neoholocene in Poland. Folia Quaternaria Trouet, V., Dominguez-Delmas, M., Pearson, C., Pederson, N., Rubino, D., 2017. Dendro-
69, 5–133. archeo-ecology in Northern America and Europe: re-purposing historical materials to
Krąpiec, M., 2001. Holocene dendrochronological standards for subfossil oaks from the study ancient human-environment interactions. Dendroecology 365–394.
area of southern Poland. Stud. Quat. 18, 47–63. Ważny, T., 1990. Aufbau und Anwendung der Dendrochronologie für Eichenholz in
Larsson, L.A., Larsson, P.O., 2017. CDendro and CooRecorder (v.9.1) [Software]. Polen. Dissertation. Hamburg University.
Available from. Cybis Elektronik and Data AB, Saltsjöbaden, Sweden. http://www. Ważny, T., Lorentzen, B.E., Köse, N., Akkemik, Ü., Boltryk, T., Güner, T., Kyncl, J., Kyncl,
cybis.se. T., Nechita, C., Sagaydak, S., Vasileva, J.K., 2014. Bridging the gaps in the tree-ring
Leal, S., Campelo, F., Luisa Luz, A., Fatima Carneiro, M., Andrade Santos, J., 2015. records: creating a high-resolution dendrochronological network for Southeastern
Potential of oak tree-ring chronologies from Southern Portugal for climate re- Europe. Radiocarbon 56 (4), 39–50.
constructions. Dendrochronologia 35, 4–13. Wigley, T.M.L., Briffa, K.R., Jones, P.D., 1984. On the average value of correlated time
Lu, X., Mo, J., Gilliam, F.S., Zhou, G., Fang, Y., 2010. Effects of experimental nitrogen series, with applications in dendroclimatology and hydrometeorology. J. Appl.
additions on plant diversity in an old‐growth tropical forest. Global Change Biol. 16 Meteorol. Climatol. 23, 201–213.

112

View publication stats