Radial growth-based assessment of sites effects on pedunculate and [629558]

Radial growth-based assessment of sites effects on pedunculate and
greyish oak in southern Romania
Constantin Nechitaa,b, Irina Macoveic, Ionel Popaa,e,⁎, Ovidiu Nicolae Badeaa,
Ecaterina Nicoleta Apostola, Ólafur Eggertssond
aNational Institute for Research and Development in Forestry "Marin Dr ăcea", Calea Bucovinei, 73 bis, 725100, Câmpulung Moldovenesc, Romania
bDepartment of Geography, Universit ății 13, 720229, Ștefan cel Mare University of Suceava, Romania
cDepartment of Pharmacognosy, Faculty of Pharmacy, Grigore T. Popa University of Medicine and Pharmacy Iasi, 16 Universitatii Str., 700115, Iasi, Ro mania
dIcelandic Forest Research, Mógilsá, IS-162 Reykjavík, Iceland
eINCE- Mountain Economy Center CE-MONT, Vatra Dornei, Romania
HIGHLIGHTS
•The dendroclimatology of Quercus robur
andQ. pedunculi florawas studied.
•Different growth patterns among sites
and within species were observed.
•There is a relationship between offset
earlywood and climatic drivers between
varieties.
•Winter and spring represent key sea-
sons in separating tardive from praecox
varieties.GRAPHICAL ABSTRACT
New species?
Q.pedunculiflora var. praecox andtardiveSubspecies?
Incomplete speciation process?Climate
dataSpring
PhenologyQuercus
robur L.
Quercus
peduncu
liflora
K. Koc h.Tree
ring
width
abstract article info
Article history:
Received 28 February 2019Received in revised form 31 July 2019Accepted 31 July 2019Available online 31 July 2019
Editor: Cristina AponteThis study focuses on the climate growth drivers of Quercus robur L. (pedunculate oak) and Q. robur subsp.
pedunculi floraK. Koch. (greyish oak), occurring in the biodiversity of three sites in southern Romania. We deter-
mined the degree of tolerance of the greyish oak, between the tardive and praecox varieties, to environmental
stress, between 1951 and 2016. Total tree ring-width (RW), and earlywood (EW) and latewood (LW) measure-
ments were subject of periodical and monthly climate-growth analysis. Our results revealed a moderate relation-ship between climate and tree-growth. A signi ficant and positive relationship was observed between RW and
previous growing season precipitation. Mean and minimum temperatures affected both positive and negative
tree-rings during the growing season. We also observed that winter and spring represent key seasons for differ-entiating tardive from praecox varieties, affecting the intra-annual variability of ring-width, and EW and LW pa-
rameters. The correlation between the tree-ring measurements and daily climate data shows a clear offset of the
starting growth between greyish oak varieties. A weak in fluence of stressors on tree-growth at the sites was ob-
served through pointer year and resilience components analysis.
© 2019 Elsevier B.V. All rights reserved.Keywords:
Pedunculate oakGreyish oakGrowth/daily climate dependencePointer yearsResilience components
1. Introduction
Global temperatures increased at different rates, depending on the
region, starting in the early 1970s ( Huijun, 2001 ;D'Arrigo et al.,
2008 ). One of the forest ecosystems that may severely suffer from aScience of the Total Environment 694 (2019) 133709
⁎Corresponding author at: National Institute for Research and Development in Forestry
"Marin Dr ăcea", Calea Bucovinei, 73 bis, 725100, Câmpulung Moldovenesc, Romania.
E-mail address: [anonimizat] (I. Popa).
https://doi.org/10.1016/j.scitotenv.2019.133709
0048-9697/© 2019 Elsevier B.V. All rights reserved.
Contents lists available at ScienceDirect
Science of the Total Environment
journal homepage: www.elsevier.com/locate/scitotenv

changing water balance and rising temperatures is the Quercus –
dominated forest. As a consequence, tree vitality and species resilience
to environmental stress conditions will decrease ( Kunz et al., 2018 ;
Gazol et al., 2018 ). Thus, a need to obtain information on the trees spe-
cies' potential for adaptation has emerged in the last four decades
(Montwé et al., 2018 ).
Nowadays the dendrochronological interest in oak-growth, espe-
cially in floodplain forest ecosystems, in relation to environmental
drivers has grown. Most studies have been performed in Central and
Western Europe ( Baillie, 1992 ;Filippo et al., 2010 ;Dole žal et al.,
2016 ), but there is also increasing interest in Eastern European and
the Balkans area ( Mikac et al., 2018 ;Árvai et al., 2018 ;Stojanovi ć
et al., 2018 ;Dobrovolný et al., 2018 ;Vanhellemont et al., 2019 ;Mrak
et al., 2019 ). Even so, in many studies, sessile ( Q. petraea )a n dp e d u n c u –
late oaks have been independently analysed ( Árvai et al., 2018 ;Perkins
et al., 2018 ). The species-rich oak genus Quercus is challenging ( Muir
et al., 2000 ) due to interspecies hybridisation, which can potentially ad-
dress the mechanism of speciation; morphological and genetic investi-
gations have focused more on deciphering this aspect, but in most
cases, scienti fic opinion is divided, as in the case of the greyish oak
(Gömöry et al., 2001 ;Curtu et al., 2011 ;Apostol et al., 2017 ).
Recent studies have leaned more towards an ecological speciation
process, with the greyish oak being treated as a distinct taxonomic
unit rather than a subspecies of Q. robur (Curtu et al., 2011 ). This aspect
may be the reason why Q. pedunculi flora, distinct from Q. robur ,a p p e a r s
to have been poorly investigated, with only a few studies on forest ge-
netics clearly distinguishing between these species ( Postolache et al.,
2019 ;Popescu et al., 2019 ). However, certain authors consider the
greyish oak to be a hybrid of the pedunculate oak, thus there is no
clear consensus on the status of Q. pedunculi flora(Cuza, 2007 ;Tîcu,
2012 ). The grevish oak is considered to be thermophilous and xeric,
compared to the pedunculate oak ( Beldie, 1952 ), features that confer in-
creased plasticity and drought resistance. Both species hold important
status in future forests, in the context of climate change and deserti fica-
tion processes, in southern Romania.
To the best of our knowledge, this study represents the first dendro-
chronological approach in investigating and reinforcing the ability of
the greyish oak to easily adapt to extreme conditions through develop-
ing tardive and praecox varieties ( Chesnoiu et al., 2009 ;Curtu et al.,
2011 ;Apostol et al., 2017 ). A dendroecological approach was used on
two close species of oak ( Q. robur and Q. pedunculi flora) with different
ecological preferences in order to observe and compare their adaptive
potential. The study improves the level of current knowledge regarding
specimen and species adaptation under current environmental condi-
tions, with reference to future scenarios.
In this study, we compared the relationship between the tree-ring,
earlywood (EW) and latewood (LW) width indices of Q. robur and
Q. pedunculi flora, using two varieties (tardive and praecox), and climatic
factors in floodplain forests in three regions of southern Romania. The
aims of the study were to: (i) determine if climate was the main growth
driver in the region; (ii) assess growth changes in Q. robur and
Q. pedunculi floratrees under the effects of local climate; and (iii) com-
pare the degree of tolerance to environmental stress of the greyish
oak between the tardive and praecox varieties. We expected that:
(1) there would be differences in oak tree growth between sites and
within species; and (2) that the environmental driver in fluences
would be detectable in the time-series of oak ring growth, which
would show dissimilarities between tardive and praecox trees.
2. Materials and methods
2.1. Study sites
The study area is located in southern Romania, close to the Danube
River ( Fig. 1 ). It is characterised by strong climatic gradients, with
high annual minimum temperatures (6 –7° C , Tmin) and low annualmean precipitation ( Pmean, 500 mm/year). Droughts are frequent due
to the in fluence of Mediterranean climate and late frosts occurs due to
continental climate. Additionally, the limiting effects of precipitation
are increased by the daily thermal amplitudes in early spring, while
the maximum temperatures ( Tmax) may exceed 35 and 43 °C in sum-
mer. The annual Pmean ranges between 251 mm/year (in 2000) and
992 mm/year (in 2014). The first frost occurs in late October, the last
at the end of April.
The three study sites included plots containing two broadleaved spe-
cies of pedunculate oak (at Ciupercenii Vechi [CIV]) and greyish oak (at
Brani ștea Bistre țforest [BRB] and Brani ștea Catârilor [BRC]). The dis-
tance between the sites was almost 50 km. The first site, CIV, is a Natura
2000 protected area (Ciuperceni-Desa, ROSCI0039) that includes flora
and fauna speci fic to wetlands in the floodplain of the Danube River
(43°53 ′38”N, 24°14 ′43″E). The protected area has different habitat
types, including natural eutrophic lakes, oligotrophic ponds, Pannonian
salty swamps, sandy Pannonian meadows, low-altitude meadows, clus-
ters of riparian vegetation ( Salix,Populus ), and cultivated arable land.
The second site, BRB, is located close to C ălugăreni Lake on two relief
forms –floodplain and a terrace of the Danube (43°54 ′24”N, 23°31 ′7″E).
The meadow is directly controlled by the river discharge, which creates
a wide variety of soils, mainly young, less developed alluvial ones, with
unstable structures. The terrace is characterised mainly by gley
processes.
The third site, BRC, is a protected area of national forestry interest for
Q. pedunculi flora,Q. pubescens andQ. virgiliana (43°56 ′9”N, 22°54 ′14″E).
Two habitat types are present –Euro-Siberian sylvosteppe and areas
with Ponto-Sarmatian forest vegetation speci fic to the area, with sand
dunes. The oak trees grow in association with Acer pseudoplatanus L.,
Tilia cordata Mill., Fraxinus excelsior L.,Crataegus monogyna Jacq., Rubus
fruticosus L.,Prunus spinosa L. and Rosa canina L. The invasive species
Robinia pseudoacacia L. is also present. A monumental oak tree was
cored at this site (200 years old, ~5 m in circumference) –a remnant
of an ancient floodplain forest.
2.2. Tree-ring data sampling and processing
Thefieldwork took place in the first week of April 2017. Samples for
dendrochronological analysis were collected using a 5-mm-increment
borer. As the stands are located in protected areas, a minimum number
of trees was sampled. For each sampled tree, GPS coordinates, diameter
at breast height and spring phenological observations were registered.
At the BRB and BRC sites, we sampled both types of greyish oak, thelate (tardive) and early (praecox) flush of leaves, hereinafter tandp.A
number of 9 t and 21 p trees were sampled from BRB, respectively
23 t and 11 p from BRC sites ( Table 1 ). This phenological variability
was not observed in Q. robur (CIV), and we collected only 20 samples
from trees with leaves already formed ( Fig. 1 ).
The cores were air-dried and prepared according to standard den-
drochronological methods ( Speer, 2010 ). The samples were scanned
at high resolution (2400 pixels), using an Epson Expression A3
11000XL Pro scanner and SilverFast Ai v.8.7 software. The images
were used to identify tree-ring width (RW) boundaries and to measure
EW and LW RWs, using CooRecorder v.9.0 image analysis software
(Larsson and Larsson, 2017 ), up to a precision of 0.01 mm. All RW sam-
ples measured were cross-dated visually and statistically, using
CDendro and COFECHA ( Holmes, 1983 ), respectively.
Each tree-ring component series was detrended using a 20-year
spline function with a 50% frequency-response cutoff, using ARSTAN
software ( Cook and Kairiukstis, 1990 ), in order to preserve the high-
frequency variability. An autoregressive model was applied to remove
the autocorrelation with previous RW. The Akaike (1974) information
criteria were used to obtain the autoregressive moving-average process.
Additionally, the Keith-Briffa rbar-weighted method of variance
stabilisation was used. The mean site chronology was built by averaging
the annual index values by bi-weight robust estimation ( Cook, 1985 ).2 C. Nechita et al. / Science of the Total Environment 694 (2019) 133709

The tree-ring index-series were further pre-whitened using an
autoregressive model and moving-average time-series modelling
(Cook, 1985 ), in order to obtain the mean chronologies.
The resulting residual chronologies for each species and phenologi-
cal group (tardive and praecox) were used in all analyses of the
growth/climate relationship. The statistical quality of the chronologies
was assessed by means of standard basic statistics ( Briffa and Jones,
1990 ). Two widely-used parameters –expressed populational signal
(EPS) and average correlation between series (Rbar) –were applied to
measure the chronology reliability (common variability among individ-
ual tree-ring series through time), considering EPS N0.85 as a reliable
threshold ( Wigley et al., 1984 ). Rbar is independent of sample size,
and provides information on the chronology signal strength (common
variance).
2.3. Statistical analysis
We compared the growth to the monthly and periodic mean Pmean,
andTmin and mean temperature ( Tmean) from the E-OBS v.17 climate
database ( http://climexp.knmi.nl ;Haylock et al., 2008 ) for 1950 –2017.
The climate/growth correlations were calculated using Pearson's corre-
lation coef ficient and DendroClim 2002 ( Biondi and Waikul, 2004 ), with
monthly temperatures and precipitation values starting from the Sep-
tember of the previous year up to the September of the current year.
The time stability of the correlation coef ficients was calculated using a
26-year moving interval. The statistical signi ficance of the growth/cli-
mate correlation was tested using a bootstrap procedure and 1000 rep-
lications. The signi ficant statistical difference between correlation
coefficients obtained for both phenoforms was tested using Fisher r-
to-ztransformation ( Steiger, 1980 ;Howell, 2011 ).
The periodic climate-growth correlation, derived from the daily cli-
mate data, was analysed using the dendroTools package ( R
Development Core Team, 2008 ;Jevšenak and Levani č,2 0 1 8 ). The corre-
lation started with a 20-day window, shifting by one day at each itera-
tion, and after the routine was completed, the calculation started
again with a 21-day window, up to 120 days. The moving correlations
with the tree-ring variables were calculated for all the time windows
from the previous year's 1st July to 30th October of the current year,
for the common period 1951 –2016.
The pointer years as an ecological indicator recording the reaction of
the trees to environmental stress factors, were calculated usingnormalized Cropper values |C|, which re flect the number of standard de-
viation (SD) from the local mean ( Cropper, 1979 ). Thresholds were ap-
plied for the interpretation of extreme climatic in fluence from the
occurrence of extreme positive and negative deviations (|C| N1.0).
‘Weak ’growth deviation was de fined when at least 50% of the tree-
ring time-series showed an event year with a growth increase (positive
pointer year) or decrease (negative pointer year) of at least 40% of the
average growth over the previous four years.
To quantify the tree-ring responses to environmental stressors (e.g.
drought, insect outbreaks, frost events) of pedunculated and greyish
oak (tardive and praecox), we calculated resilience, recovery and resil-
ience growth indices, in accordance with Lloret et al. (2011) .T h er e s i s –
tance index ( Rt) expressed the ratio of growth between the negative
pointer year and the previous period, whereas the recovery index ( Rc)
accounted for the growth reaction following the negative pointer year.The resilience index ( Rs) was a measure of the ratio of growth between
the previous and following stress periods.
3. Results
3.1. Radial growth pattern across species
The mean segment length of the tree-ring growth ranged from
76 years (at BRB) to 112 years (CIV). The tardive trees were fast-
growing, having a larger diameter than the praecox trees at the same
age. The mean values of EW varied from 0.99 mm year
−1(at BRB tardive
trees) to 0.96 mm year−1(at BRB praecox trees). In contrast, at BRC, the
EW width varied between 0.82 mm year−1for the tardive and
0.90 mm year−1for the praecox chronology. The mean LW width was
constantly higher in the case of the tardive than the praecox variety at
both sites ( Fig. 2 ).
At the pedunculate oak site, the average growth rate was generally
greater than that of the greyish oak, being 1.18 mm year−1for the EW
and 2.66 mm year−1for the LW ( Fig. 2 ). After 1980, a changing growing
ratio of the EW and LW at BRB was noted. Moderate and poor correla-
tions were found between the EW and LW for all sites (0.33 Nrb
0.80). A poorer correlation was found between varieties of EW (BRB, r
= 0.69; BRC, r= 0.61), while a good correlation was observed for the
LW (BRB, r= 0.73; BRC, r= 0.79). A clear disagreement between the
tardive and praecox trees was observed after 1980, shown in both the
EW and LW growth chronologies ( Fig. 3 ).
II
IV
IIILegend
Major rivers
Q. pedunculiflora BRC site
Q. pedunculiflora BRB site
Q. robur CIV site
Borders
0 50 kmCIV
BRB BRCVII
V
0 50 kmCIVBRB BRC
BULGARIASERBIAHUNGARYUKRAINE
MOLDOVA
I
VIV
Fig. 1. Study area in southern Romania, showing sampling sites: BRC –Brani ștea Catarilor; BRB –Brani ștea Bistre ț;C I V –Ciupercenii Vechi. Main climatic zones for Romania (according to
the Atlas of Romania, 1978): I –North Atlantic; II –Baltic; III –Eurasian, strongly continental, with accentuated aridity; IV –Pontic; V –Temperate continental transitional; VI –
Mediterranean.3 C. Nechita et al. / Science of the Total Environment 694 (2019) 133709

The EPS values were above the threshold of 0.85 only in the LW and
RW chronologies at all sites after 1950. The series intercorrelation indi-
cated that EW tardive trees were less homogenous (0.29) than EW
praecox trees, which re flected greater homogeneity (0.53) at BRB. How-
ever, at BRC, for EW, the chronologies of both the tardive and praecox
trees had similar intercorrelations. Mean sensitivity emphases a high re-
sponsive capacity of the tree-ring series to environmental factors, espe-
cially in LW praecox chronologies ( Table 1 ). The Rbar statistics were
relatively low in the EW tardive chronology, suggesting a small amount
of common signal between our dendrochronological records.
3.2. Growth in relation to monthly climate
The precipitation and temperature effects were summarised ac-
cording to the three seasonal groups: (1) the autumn of theprevious growing season; (2) the winter and spring; and (3) the
current growing season. All correlations were tested for signi ficance
(pb0.05).
The climatic conditions of the previous autumn were of low impor-
tance in tree-ring formation at BRB, with the exception of a moderate ef-
fect of precipitation on LW praecox and RW both varieties ( Fig. 4 ). At
BRC, in addition to a signi ficant positive in fluence of precipitation,
reflected in both the LW and RW phenoform chronologies, a signi ficant
correlation was obtained for the EW with Tmean. For the autumn, a pos-
itive effect of all climatic drivers was found for EW at CIV, whereas the
Tmin stimulated LW growth ( Fig. 4 ).
Winter and spring represented key seasons that differentiated
the tardive from the praecox at the greyish oak sites, affecting the
intra-annual variability of all pa rameters. Thus, at BRB, the EW
tardive trees re flected no signi ficant response to climatic drivers,Table 1
Descriptive statistics of Q. robur andQ. pedunculi florasites and tree-ring chronologies parameters for earlywood (EW), latewood (LW) and total ring width (RW). For Q. pedunculi florawere
analysed both tardive (t) and praecox (p) varieties.
Parameter Site EWt EWp LWt lWp rWt rWp
N BRB 9 21 9 21 9 21
Age ± SD (yrs.) 52 ± 12 56 ± 9 52 ± 12 56 ± 9 52 ± 12 56 ± 9
DHB ± SD (cm) 38.5 ± 10 39.5 ± 9.2 38.5 ± 10 39.5 ± 9.2 38.5 ± 10 39.5 ± 9.2
SI 0.29 0.53 0.51 0.62 0.57 0.67
MS 0.23 0.23 0.40 0.47 0.24 0.29
Rbar 0.09 0.24 0.39 0.51 0.31 0.50
N BRC 23 11 23 11 23 11
Age ± SD (yrs.) 86 ± 15 80 ± 19 86 ± 15 80 ± 19 86 ± 15 80 ± 19
DHB ± SD (cm) 51.8 ± 11 50 ± 12 51.8 ± 11 50 ± 12 51.8 ± 11 50 ± 12
SI 0.33 0.37 0.54 0.60 0.57 0.62
MS 0.29 0.28 0.44 0.45 0.28 0.29
Rbar 0.09 0.15 0.29 0.33 0.25 0.31
N CIV 20 20 20
Age ± SD (yrs.) 84 ± 24 84 ± 24 84 ± 24
DHB ± SD (cm) 97.8 ± 24 97.8 ± 24 97.8 ± 24
SI 0.35 0.59 0.61
MS 0.22 0.40 0.27
Rbar 0.16 0.36 0.35
*N–number of trees included in chronologies; DHB –diameter at breast height; SI –series intercorrelation; MS –mean sensitivity; Rbar –average correlation between series.
Fig. 2. a) Box-whisker plots of mean growth over common growth period (1951 –2016), with variability for mean chronology. b) Relationship between mean chronology of
Q. pedunculi floravarieties (tardive and praecox) for earlywood (EW), in Brani ștea Bistre țforest (BRB) –black, Brani ștea Catârilor (BRC) –blue and latewood (LW), BRB (green), BRC
(red). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)4 C. Nechita et al. / Science of the Total Environment 694 (2019) 133709

in contrast with the EW praecox ones, which was sensitive, espe-
cially to the winter and spring Tmins. The LW praecox trees was
positively dependent on the current growing season's temperatureand precipitation. The LW tardive trees responded only to the cur-
rent growing season, with a positive effect for precipitation and
Tmin, and a negative effect for Tmean ( Fig. 4 ).
Fig. 3. Mean raw ring width chronologies for Q. pedunculi floravarieties. Blue line –tardive trees mean ring width; red line –praecox trees mean ring width; grey lines represent each
individual ring width. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Fig. 4. Pearson's product moment correlation of chronologies of Q. pedunculi floravarieties, tardive (t) and praecox (p) with precipitation (left) and temperature (right) data from EOBS v.17
for 1950 –2017 for the period previous September to current August in the case of latewood (EW) and total ring width (RW). For earlywood the period analysed was be tween previous
September to current May.5 C. Nechita et al. / Science of the Total Environment 694 (2019) 133709

The relationship between the EW praecox variety (BRC) and April
Tmean of the current growing season was signi ficant. An additional neg-
ative in fluence of Tmin in May was also calculated. The EW tardive trees
responded signi ficantly only in January, negatively to temperature and
positively to precipitation. The LW tardive trees at BRC experienced a
June and July negative temperature in fluence ( Fig. 4 ).
The EW RW of Q. robur (CIV) was positively correlated with the
Tmean of the previous December and the Tmin of the previous Novem-
ber and December and the current March. The EW responded signi fi-
cantly to the Pmean of the previous September and November. The
LW responded signi ficantly only to the Tmin from the previous October
and November. The RW was controlled by rainfall and Tmin (in the au-
tumn of the previous growing season –September –November) ( Fig. 4 ).
The difference between the correlation coef ficients ( pb0.1, two-
tailed tests of signi ficance), indicated a signi ficant disagreement be-
tween the EW, both phenoforms for January and February precipitation
at BRB. RW emphasised the same signi ficant difference, but for Tmin. At
BRC, no signi ficant difference was calculated between the tree
phenoforms.
The temporal stability of the correlation coef ficients re flects changes
in time between the BRB chronologies and precipitation. The EW
praecox trees showed a negative signi ficant dependence with Pmean
in May. RW and LW tardive varieties were positively correlated with
the April Tmin, with decreasing correlation coef ficient values from the
beginning of the 1970s ( r= 0.46) up to the present ( r=0 . 3 1 ) .F o r
BRC, positive correlation was calculated between LW both phenoform,
RW praecox trees and the precipitation of the previous November. The
CIV chronologies emphasised a time-signi ficant correlation only be-
tween the EW and the previous December's Tmin ( Fig. 5 ).
3.3. Growth in relation to periodic climate
Several differences occurred among the growth-climate relation-
ships at the stand, species, and RW type- and variety levels. Major differ-
ences were observed only in the EW, while for the LW and RW, theagreement was relatively comparable. Thus, at BRB, the relationship be-
tween the Tmean and EW praecox trees was negatively signi ficantly
correlated in the first week of March and mid-April. In contrast, for
EW tardive trees, a positively insigni ficant dependence was obtained,
starting at the beginning of February ( Fig. 6 ). At BRC, the EW tardive
trees were negatively correlated starting at the beginning of January
until mid-February, and the EW praecox ones had no signi ficant re-
sponse to the Tmean ( Fig. 7 ). In the RW, this aspect was less evident
and, in the LW, no signi ficant differences were observed.
The results showed a moderate relationship between RW and cli-
mate, sustained in multiple cases for over a month. Generally, there
was a persistent difference between the tardive and praecox trees, in
both the sites and species. Thus, at BRB, the EW both phenoformsshowed a relatively divergent and offset correlation with the Tmean,
with both cases being below signi ficance level. The praecox correlated
mainly negatively, the tardive positively. At BRC, the tardive correlated
negatively signi ficantly in the interval 18th December –27th February.
In the RW, the differences were mitigated by mixing signals from the
EW and RW.
3.4. Site species-speci fic extreme growth pattern
One positive pointer year (1966) and three negatives (2007, 1993,
1955) were recorded at the site level (BRB). The tardive variety showed
high sensitivity, recording two positive pointer years (1999, 1995) and
four negatives (2012, 2007, 1993 and 1988), while the praecox
responded only to the negative conditions in 2007 and 1997. A diverse
behaviour was observed at BRC, as the site pointer year showed growth
reduction in 1983, 1965 and 1955. Between the varieties, only the
tardive responded negatively in 1983. For the CIV RW chronology,
1979 and 1965 were positive years, and 1984, 1978 and 1957 were neg-
ative ones.
Growth reductions were seen in the greyish oak in 1955, while for
the pedunculate oak, this occurred in 1957. At the site level, the tree
growth at BRC was less resistant to negative factors ( Rt= 0.48, Rs=
1.37) than at BRB ( Rt=0 . 6 9 , Rs= 1.50). No signi ficant differences be-
tween tardive and praecox tree-growth were obtained; for example, at
BRB in 2007, the tardive Rtwas 0.67, while for the praecox, it was 0.69.
In all cases, the Rcexceeded the mean unit value for the 4 years prior to
growth depression.
4. Discussion4.1. Chronology characteristics
The mean growth value reported for pedunculate oak in this study
(4.54 mm year
−1) showed that they are fast-growing trees, experienc-
ing optimal conditions. The values for this parameter may be region-speci fic, with lower values having been reported in the literature:
north-eastern Hungary –0.81 mm (EW), 1.09 mm (LW) ( Kern et al.,
2013 ); north-eastern Germany –1.71 –1.81 mm (RW) ( Scharnweber
et al., 2011 ); Estonia –1.2–2.6 mm (RW) ( Sohar et al., 2014 ); central
Poland –1.59 –2.43 mm (RW) ( Ważny and Eckstein, 1991 ;Bronisz
et al., 2012 ); and south-eastern Slovenia –1.71 mm (RW) ( Čufar et al.,
2008 ). A broad study of the natural geographical range of pedunculate
and sessile oak in south-eastern Central Europe (Alpine South, Conti-
nental and Pannonian ranges) indicated mean growths of 0.71 mm to
3.69 mm (RW) ( Čufar et al., 2014a ). The oak's mean radial growth in
Romania, based on measurements from N60 sites, varied between 1.24
and 3.6 mm ( Nechita, 2013 ), with cross-regional peculiarities, such as
2.04 mm in southern Romania ( Nechita and Chiriloaei, 2018 ),
1980 1990 2000 2010-0.4-0.30.30.40.5
BRB_LWt vs. April Tmin
BRB_LWp vs. May Pmean
BRC_LWt vs. pNov Pmean
BRC_LWp vs. pNov Pmean
CIV_EW vs. pDec TminCorrelation coefficients
YearLegend
Fig. 5. Temporal changes in signi ficant bootstrapped correlation coef ficients ( Pb0.05) using intervals of 26 years lagged 5 years, between earlywood (EW) and latewood (LW) ring indices
and average minimum temperature and precipitation amount calculated for the September of the previous year up to the September of the current year. Fo rQ. pedunculi floraboth
varieties, tardive (t) and praecox (p) were compared.6 C. Nechita et al. / Science of the Total Environment 694 (2019) 133709

2.53 mm in the Western Plain ( Nechita and Popa, 2012 ), and 2.93 mm in
the NE and 1.82 mm in the north-west ( Nechita et al., 2017 ).
The mean radial growth of greyish oak in this study ranged between
2.11 mm and 2.84 mm, ( Fig. 2 a), signi ficantly lower than the mean tree-
ring growth measured for Q. pedunculi flora(3.79 mm) in south-easternRomania (Ianca region, Popa and Nechita unpublished data, 2019). The
mean RW indicated that the sites were located in the natural greyish
oak distribution, which is relatively restricted to south-easten Europe
(Schwarz, 1993 ;Menitsky, 2005 ). To the best of our knowledge, this
study is the first to separately dendrochronologically investigate25-Jul.
13-Sep.
02-Nov.22-Dec.10-feb.01-apr.21-may.
10-jul.20406080100120Time span (days)EWt vs.P m e a n
25-Jul.13-Sep.02-Nov.22-Dec.10-feb.01-apr.21-may.10-jul.20406080100120LWtvs.P m e a n
25-Jul.13-Sep.
02-Nov.22-Dec.10-feb.01-apr.
21-may.
10-jul.20406080100120EWt vs.T m e a n
25-Jul.13-Sep.
02-Nov.22-Dec.10-feb.01-apr.21-may.10-jul.20406080100120LWtvs.T m e a n
-0.3-0.10.10.30.5Legend
25-Jul.
13-Sep.02-Nov.
22-Dec.10-feb.
01-apr.
21-may.10-jul.20406080100120EWp vs.P m e a n
25-Jul.13-Sep.02-Nov.22-Dec.10-feb.
01-apr.
21-may.10-jul.20406080100120LWp vs.P m e a n
25-Jul.13-Sep.
02-Nov.
22-Dec.10-feb.
01-apr.
21-may.10-jul.20406080100120EWp vs.T m e a n
25-Jul.13-Sep.
02-Nov.22-Dec.10-feb.
01-apr.21-may.10-jul.20406080100120LWp vs.T m e a n
Fig. 6. Climate-growth correlation between precipitation and temperature for Q. pedunculi flora(BRB). Iteration starts on 1st July of the previous year and ends on 30nd October of current
year, with a 20-day window shift, and each iteration with one day.
25-Jul.13-Sep.02-Nov.22-Dec.10-feb.01-apr.
21-may.10-jul.20406080100120EWt vs. PmeanTime span (days)25-Jul.13-Sep.02-Nov.22-Dec.
10-feb.01-apr.21-may.10-jul.20406080100120LWtvs.P m e a n
25-Jul.13-Sep.02-Nov.22-Dec.10-feb.01-apr.
21-may.10-jul.20406080100120EWt vs.T m e a n
25-Jul.13-Sep.02-Nov.
22-Dec.10-feb.01-apr.21-may.10-jul.20406080100120LWtvs.T m e a n
-0.4-0.200.20.4Legend
25-Jul.
13-Sep.02-Nov.22-Dec.10-feb.01-apr.
21-may.10-jul.20406080100120
EWp vs. Pmean
25-Jul.13-Sep.02-Nov.
22-Dec.
10-feb.01-apr.21-may.10-jul.20406080100120
LWp vs. Pmean
25-Jul.13-Sep.02-Nov.22-Dec.10-feb.01-apr.21-may.10-jul.20406080100120
EWp vs.T m e a n
25-Jul.
13-Sep.02-Nov.22-Dec.
10-feb.01-apr.21-may.
10-jul.20406080100120
LWp vs.T m e a n
Fig. 7. Climate-growth correlation between precipitation and temperature for Q. pedunculi flora(BRC). Iteration starts on 1st July of the previous year and ends on 30nd October of current
year, with a 20-day window shift, and each iteration with one day.7 C. Nechita et al. / Science of the Total Environment 694 (2019) 133709

Q. robur and Q. pedunculi florausing both praecox and tardive varieties,
which indicate a high plasticity and adaptability to environmental
conditions.
Until now, the few studies dealing with pedunculate and greyish oak
distributions have more often focused on south-eastern Europe, de-
scribing the presence of Q. pedunculi floraas an independent species
that relies most on morphological characteristics and habitat prefer-ences ( Georgescu and Morariu, 1948 ;Vuckovic, 1984 ;Stănescu et al.,
1997 ). Even though the genetic structure shows a relative similarity be-
tween Q. pedunculi floraand Q. robur ,(Gömöry et al., 2001 ;Muir and
Schloetterer, 2005 ), further investigations are needed to establish
their systematic positions. Nuclear divergence, using isozyme markers,
has indicated a greater genetic diversity of pedunculate oak in Europe
(Streiff et al., 1998 ;Gömöry and Schmidtova, 2007 ;Gömöry et al.,
2010 ;Pettenkofer et al., 2019 ).
A recent combined morphological and genetic study on pedun-
culate and greyish oak populations from Romania indicated a low
level of nuclear divergence between the species, but a large vari-
ance in the discriminant function values of the leaf morphological
characteristics ( Curtu et al., 2011 ). Moreover, bud burst and
flowering phenology ( Chesnoiu et al., 2009 ) showed a differentia-
tion between these two taxa ( Apostol et al., 2017 ). A closer look
at the greyish oak supports the hypothesis that Q. pedunculi florais
an incipient species that has recently evolved to withstand drought
stress through several adaptations, such as the hairiness on the ab-
axial leaf surface ( Curtu et al., 2011 ).Ganatsas and Tsakaldimi
(2013) confirmed that the seeds of Q. pedunculi floraare less sensi-
tive to desiccation than those of Q. pubescens and Q. coccifera ,t h i s
aspect being important in forestry practice and in the management
of forest genetic resources.
Studies on the genetic diversity of oak species in southern Romania
(Popescu and Postolache, 2009 ) have indicated the presence of one or
two haplotypes in most of the pedunculate oak populations. Autochtho-
nous populations with three haplotypes were rarely observed, mostly in
regions where different post-glacial recolonization routes met. A larger
number of chloroplast haplotypes have been observed in the CIV popu-
lation ( Popescu and Postolache, 2009 ), which may indicate that this for-
est stand was arti
ficially planted with alien seeds. This aspect was
confirmed by accessing the management plans for the plot; these men-
tioned the year of plantation and an unknown provenance of the seeds.
The dendrochronological analysis of the tree-rings indicated a clear
disagreement, over approximately 20 years after the mid-1980s, be-
tween the praecox and tardive varieties. Whilst at BRB, the tardive
growth was greater than the praecox growth, at BRC, the trend was
the opposite. In our opinion, the change in the behaviour of the greyish
oak population in southern Romania needs to be studied using complex
wood anatomy, phenology and genetical studies, using larger trees
populations.
Thus, mean sensitivity is a measure of the relative changes between
adjacent ring widths, with values N0.20 indicating high variability
(Fritts, 1976 ). In our study, the comparative statistics indicated that all
the mean sites chronologies exhibited the threshold for EW and RW
(0.22 –0.29) ( Table 1 ). Moreover, LW (0.40 –0.47) showed a higher var-
iability in greyish oak than pedunculate oak, and in praecox vs. tardive
trees. This might be a consequence of passive stressors from the winter
and spring causing an asymmetric alteration of the year-by-year tree-
ring formation process. However, the series intercorrelation was higher
than the critical value ( N0.32), which is comparable to those calculated
from other temperate regions (0.19 –0.84) ( Grissino-Mayer, 2001 ;
Rozas, 2005 ;Kern et al., 2013 ;Čufar et al., 2014b ;Cater and Levanic,
2015 ).
Generally, the mean correlations between the series, through se-
ries and mean site chronology, respectively, increase with negative
climatic stress action ( Fritts, 1976 ;Grissino-Mayer, 2001 ). In our
case, under the same climatic-site conditions, the series intercorre-
lation difference between tard ive (0.29) and praecox (0.53) isunusually high, especially at BRB for EW ( Table 1 ). This may be in-
fluenced by decreasing LW width and increasing EW width
(Fig. 2 b).
According to Sass-Klaassen et al. (2011) , LW is responsible for the
majority of annual tree RW variability. Our study shows that EW may
also induce variability and signals in tree RW, to almost the same degree
as LW. This aspect has not been apparent in other studies, and the mech-anism was evident only after 1980, and is speci fic to the praecox (at
BRB) more than tardive trees. A reduction in LW has often been associ-
ated with oak tree decline ( Levani čet al., 2011 ;Tulik, 2014 ). This was
not evident in our case, although the effect of the previous year's condi-
tions on seasonally-decreasing total ring width was observed ( González
and Eckstein, 2003 ).
4.2. Inter- and intraspeci fic oak climate-growth sensitivity
In this study, both pedunculate and greyish oak RW (EW and LW)
showed a clear, but not very strong, response to climate. The Tmin and
Tmean during the autumn of the previous growing season and the
early spring appeared to be the main climatic drivers controlling EW
growth. The LW and total RW were mainly affected by the current
growing season's temperature conditions. These results are comparableto those obtained from other ring-porous tree species in similar flood-
plain geoclimatic contexts ( Stojanovi ćet al., 2015 ;Tumajer and Treml,
2016 ). Also, a week positive in fluence of precipitation in winter and
spring was observed.
We demonstrated that, on a site-speci ficl e v e l ,t h e Tmean in
spring (beginning of April) is responsible for fast-growing EW in
tardive (at BRB) and praecox (at BRC) trees. Accordingly, the LW
praecox trees ring widths were wide, even when controlled by the
divergent action of the same climate driver; that is, temperature
positive (BRB) and negative (BRC), ( Figs. 6 ;7). Being warmer in
winter and spring, the Mediterranean climate provides suitable
conditions for the greyish oak to grow larger at BRB that at BRC,
where the temperates of the continental transitional climate arecolder, and late frosts are more frequent.
The monthly climate-growth models performed better at the site
level than at the intervariety level, as seen in the case of greyish oak.
Their performances were good enough for the implementation of differ-
ent climatic drivers, although the models underestimated the positive
and negative intervals with signi ficant correlation, which may have
advantaged or held back tardive and praecox tree-growth. Similar re-
sponse was reported by Rybnicek et al. (2016) which demonstrated
that the tree growth response to climate parameters depends on local
site characteristics, rather than on the oak species.
The response at all three sites shows a positive in fluence of precipi-
tation, while, with regard to temperature, BRB and CIV were mainly pos-
itively correlated (LW and RW), a phenomenon observed on a large
scale in north-eastern Carpathians oak chronologies ( Nechita et al.,
2017 ). For the EW, the opposite behaviour was noted –negative in
the case of BRB and positive for CIV. Other studies, on diverse geograph-
ical conditions, have also reported late autumn and early spring temper-
ature and precipitation signals in EW ( Fonti et al., 2007 ;Matisons et al.,
2015 ;Nechita et al., 2017 ;Mikac et al., 2018 ). Positive temperatures in
spring may have also caused an increase in respiration and, thus, trees
may have consumed the nutrients earlier, causing an earlier start in veg-
etation cycle ( Fritts, 1976 ).
A closer look at the intraspeci fic resolution of EW versus temper-
ature at BR reveals a greater sensitivity to positive temperature
from the previous growing se ason than at BRB. According to
Pérez-de-Lis et al. (2016) , EW formation and budburst onset is cor-
related with thermal condition, which is species-speci fic. This as-
pect has been observed in the greyish oak trees of southern
Romania ( Chesnoiu et al., 2009 ;Apostol et al., 2017 ). Since this phe-
nomenon is more obvious in the praecox trees, it may also have
been in fluenced by an earlier cambial activity, starting with the8 C. Nechita et al. / Science of the Total Environment 694 (2019) 133709

bud burst and leaf expansion processes ( Rozas, 2005 ). Moreover, for
oak species, the onset of vessel enlargement takes place prior to bud
swelling, based on the previous year's resources ( Bréda et al., 2006 ;
Sass-Klaassen et al., 2011 ;Pérez-de-Lis et al., 2016 ).
The most interesting fact to explain is the capacity of the greyish
oak to start growing earlier, with the praecox starting approxi-
mately 2 –3 weeks earlier than the tardive variety under the same
environmental conditions. The spring leaf phenology between
20th March and 10th May in 406 greyish oak trees (2008 –2016)
has even indicated three categories of trees –praecox (24th –31st
March), intermediate (1st –14th April) and tardive (15th –18th
April) ( Chesnoiu et al., 2009 ;Apostol et al., 2017 ). Another signi fi-
cant fact is the historical budburst onset reported in Romania,
which was one week earlier for t he praecox than the pedunculate
oak ( Tomescu et al., 1967 ) and, in 2008, N20 days for the tardive
(Chesnoiu et al., 2009 ) variety.
Both the Tmin and Tmean, the latter being speci fict oC I V ,i n d i c a t ea
positive in fluence over an earlier start in vegetation cycle, with the tem-
perature oscillation in spring partially explaining the start of lea fing
(Askeyev et al., 2005 ). After the 1980s, and even more evident in the
last decade ( Hopkins and Kirby, 2007 ), the complex mechanism of
firstflowering and first lea fing occurs offset along with the first appear-
ance of moths and butter flies. As, previously observed ( Tomescu et al.,
1967 ), the same gap of approximately 15 –20 days has been identi fied
in England ( Sparkes et al., 1997 ) and is even more evident in north-
eastern Europe ( Askeyev et al., 2005 ).
4.3. Site species-speci fic stress reaction pattern
The increase in spring temperature disrupt the synchronicity among
a wide variety of vegetal and animal organisms ( Roy and Sparks, 2000 ;
Visser and Holleman, 2001 ). We have con firmed the presence of several
negative pointer years that occurred simultaneously with insect out-
breaks, such as in 1997 (at BRC), with 60% green oak tortrix moth
(Tortrix viridana ). We observed this year only in the praecox variety.
Also, it was noted that the green oak tortrix moth produced defoliation,
together with the gypsy moth ( Lymantria dispar ). The statistics regard-
ing the degree of defoliation produced by the gypsy moth in 1975 –1985
affected a nearby region of 9029 ha (in 1976) and 30,031 ha (in 1980) in
Dolj County (Romania).
Lower oak stand defoliation has been reported for the green oak tor-
trix, with a large number of hectares (10,066 ha) having been infested in
1983, according to Simionescu et al. (1992) . The national forest health
inventory for 1986 –2000 showed various degrees of defoliation in
nearby areas ( Simionescu et al., 2001 ). According to this study, we
know that the 1980s experienced the highest defoliation by the gypsy
moth, with N80% (50,272 ha) of all the oak forest, having been affected
by a high degree of defoliation in 1987, and similar phenomenon occur-
ring in 1986, 1988 and 1995. The green oak tortrix has affected the oak
forests of the region, yearly, since 1986 (4759 ha), with two maximum
inflexion years in 1992 (30,859 ha) and 2000 (34,717 ha) ( Simionescu
et al., 2001 ).
According to Tomescu and Ne țoiu (2006) , in 1948, a regional defoli-
ation was recorded, with severe intensity. In 1951, it was widespread,
occurring from the south (Banat, Cri șana) all the way up to the north-
west (Maramure ș). The most signi ficant defoliation caused by the
green oak tortrix were reported in 1960 –1964 and 1980 –1984. The pe-
riodicity of the gypsy moth is 7 –10 years, 10 years for the green oak tor-
trix. Between these two species, the tortrix starts earlier in the spring,
while the gypsy moth arrives after the leaves are formed ( Netoiu,
2000 ). Increasing numbers of oak stands were affected after the
1980s, a phenomenon that has been previously reported ( Rossiter
et al., 1988 ;Ivashov et al., 2002 ).
Our results show that pointer years relate not only to the biotic
events such as insect's infestation but they are also caused by extreme
climate, both positive and negative. The correlative relationshipbetween radial growth and local climate varies among sites, species
and varieties, which leads to uneven appearance of pointer years. This
could be explained through species range of tolerance, to which it is
added climatic in fluences and different vegetative periods, in the case
of greyish oak. The interactions between all factors that are related to
occurrence of pointer years are complex and dif ficult to detect.
This knowledge, in addition to our findings regarding oak tree sensi-
tivity to climate and their response to stress events (resilience compo-
nents), reinforces the importance of evaluating biological and
ecological processes when large response variability is present
(Walther et al., 2002 ). Even though the praecox trees are affected by
moth larvae, a recent study has indicated that tardive trees are less ex-
posed ( Kulfan et al., 2018 ). This may explain the divergent growth rates
in tardive and praecox trees.
We noticed that Q. pedunculi floratrees have several signi ficant ways
to mitigate, and adapt to, climate change ( Edenhofer et al., 2014 ). Con-
sidering its radial growth is under the mean, compared to the peduncu-
late oak, and based on its resilient components and the pointer year
analysis of the current study, no critical stress factors affect its tree-
ring formation.
5. Conclusions
In the changing climate, the oak forest ecosystems of south-
eastern Europe are expected to be exposed to increasingly extreme
temperatures and drought, inducing rising tree mortality. Our
dendroecological study offers insights into the decryption of tree-
growth responses to environmental biotic and abiotic stressors in
Q. pedunculi floratrees. We observed differences in tree growth
and response to climate drivers between the two greyish oak sites
analysed. Moreover, a dissociated growth reaction in the greyish
oak was observed between the praecox and tardive varieties over
approximately 20 years after the mid-1980s.
A complex of factors, other than genetic peculiarities, are affecting
the performance of Q. pedunculi floratrees, such as local site conditions,
regional climatic in fluences, and climatic drivers, especially tempera-
ture. The latter seems to be responsible for the earlier start of bud
burst and flowering phenology, disrupting synchronicity with the leaf
moths. We observed that the winter and spring climatic conditions rep-
resent a key season differentiating the tardive from praecox trees ingreyish oak sites.
Both Q. robur andQ. pedunculi floraring width indices showed a clear,
but not very signi ficant, response to climate. The earlywood growth ap-
pears to be controlled by minimum and mean temperature during the
autumn of the previous growing season and by those of the early spring.
in comparison with latewood and total ring width that were mainly af-
fected by the current growing season's temperature conditions.
The results of our study demonstrate that Q. pedunculi florais more
adaptive than the pedunculate oak, which is itself highly tolerant to a
variety of stressors. In terms of overall growth rates, Q. robur is superior
to the greyish oak, showing no negative in fluence of climate on tree-
growth. The high ecological value of greyish oak was presented in our
study, based on the last 67 years of climate evaluations.
Acknowledgments
This work was supported by a grant from the Romanian National Au-
thority for Scienti fic Research and Innovation, CNCS/CCCDI –UEFISCDI,
project number PN-III-P2-2.1-PED-2016-1058, in PNCDI III, A new tech-
nique regarding dendrochronological dating. Statistical, biological and
chemical approach (DendroTECH), by a mobility grant from the
Romanian Ministry of Research and Innovation, CNCS-UEFISCDI, project
number 1533/2018, in PNCDI III, and by a PN19070502 grant. Special
thanks to Flaviu Popescu, Dragos Postolache and Constantin Netoiu for
their support, advice, assistance in the field and for their critical prere-
views. We address our warmest thanks to Eliza-Maria Cosma.9 C. Nechita et al. / Science of the Total Environment 694 (2019) 133709

References
Akaike, H., 1974. A new look at the statistical model identi fication. Selected Papers of
Hirotugu Akaike. Springer, New York, NY, pp. 215 –222.
Apostol, E.N., Curtu, A.L., Daia, L.M., Apostol, B., Dinu, C.G., Șofletea, N., 2017. Leaf morpho-
logical variability and intraspeci fic taxonomic units for pedunculate oak and greyish
oak (genus Quercus L ., series Pedunculate Schwz.) in Southern Carpathian Region
(Romania). Sci. Total Environ. 609, 497 –505. https://doi https://doi.org/10.1016/j.
scitotenv.2017.05.274 .
Árvai, M., Morgós, A., Kern, Z., 2018. Growth-climate relations and the enhancement of
drought signals in pedunculate oak ( Quercus robur L.) tree-ring chronology in eastern
Hungary. iForest-Biogeosciences and Forestry 11 (2), 267. https://doi https://doi.org/
10.3832/ifor2348-011 .
Askeyev, O.V., Tischin, D., Sparks, T.H., Askeyev, I.V., 2005. The effect of climate on the
phenology, acorn crop and radial increment of pedunculate oak ( Quercus robur )i n
the middle Volga region, Tatarstan, Russia. Int. J. Biometeorol. 49 (4), 262 –266.
https://doi https://doi.org/10.1007/s00484-004-0233-3 .
Baillie, M.G.L., 1992. Dendrochronology and past environmental change. Proceedings of
the British Academy. vol. 77, pp. 5 –23.
Beldie, Al, 1952. In: Nyarady, E. (Ed.), Quercus genus. Flora R.P.R vol. Volume I. Academia
Publishing House, pp. 224 –261 ( in Romanian ).
Biondi, F., Waikul, K., 2004. DENDROCLIM2002: a C++ program for statistical calibration
of climate signals in tree-ring chronologies. Comput. Geosci. 30, 303 –311.
Bréda, N., Huc, R., Granier, A., Dreyer, E., 2006. Temperate forest trees and stands under
severe drought: a review of ecophysiological responses, adaptation processes andlong-term consequences. Ann. For. Sci. 63 (6), 625 –644. https://doi.org/10.1051/for-
est:2006042 .
Briffa, K., Jones, P.D., 1990. Basic chronology statistics and assessment. In: Cook, E.,
Kairiukstis, L. (Eds.), Methods of Dendrochronology: Applications in the Environmen-
tal Sciences. Kluwer Academic Publishers, Dordrecht, pp. 137 –152.
Bronisz, A., Bijak, S., Bronisz, K., Zasada, M., 2012. Climate in fluence on radial increment of
oak ( Quercus sp.) in central Poland. Geochronometria 39 (4), 276 –284 (https://doi:
0.2478/s13386-012-0011-7).
Cater, M., Levanic, T., 2015. Physiological and growth response of Quercus robur in
Slovenia. Dendrobiology 74, 3 –12.https://doi 10.12657/denbio.074.001 .
Chesnoiu, E.N., Șofletea, N., Curtu, A.L., Toader, A., Radu, R., Enescu, M., 2009. Bud burst
andfl
owering phenology in a mixed oak forest from Eastern Romania. Ann. For.
Res. 52 (1), 199 –206.
Cook, E.R., 1985. A Time Series Analysis Approach to Tree-ring Standardization. Ph.D. the-
sis. The University of Arizona (171 p).
Cook, E.R., Kairiukstis, L.A., 1990. Methods of Dendrochronology: Applications in the En-
vironmental Sciences. Kluwer Academic Publishers, Dordrecht (394 p).
Cropper, J.P., 1979. Tree-ring skeleton plotting by computer. Tree-Ring Bulletin. vol. 39,
pp. 47 –59.
Čufar, K., De Luis, M., Eckstein, D., Kajfe ž-Bogataj, L., 2008. Reconstructing dry and wet
summers in SE Slovenia from oak tree-ring series. Int. J. Biometeorol. 52 (7), 607.
https://doi.org/10.1007/s00484-008-0153-8 .
Čufar, K., Grabner, M., Morgós, A., Del Castillo, E.M., Merela, M., De Luis, M., 2014a. Com-
mon climatic signals affecting oak tree-ring growth in SE Central Europe. Trees 28(5), 1267 –1277. https://doi.org/10.1007/s00468-013-0972-z .
Čufar, K., Šefc, B., De Luis, M., Morgós, A., Grabner, M., Merela, M., Trajkovi ć, J., 2014b.
Tree-ring chronology of pedunculate oak ( Quercus robur ) and its potential for devel-
opment of dendrochronological research in Croatia. Drvna industrija 65 (2), 129 –137.
Curtu, A.L., So fletea, N., Toader, A.V., Enescu, M.C., 2011. Leaf morphological and genetic
differentiation between Quercus robur L. and its closest relative, the drought-
tolerant Quercus pedunculi floraK. Koch. Ann. For. Sci. 68 (7), 1163 –1172. https://doi
https://doi.org/10.1007/s13595-011-0105-z .
Cuza, P., 2007. The dynamics of sapling of pedunculate oak ( Quercus robur L.) in the poly-
morphic population and consanguineous descendants. Ambiental environment 2
(32), 10 –15 (in Romanian) .
D'Arrigo, R., Wilson, R., Liepert, B., Cherubini, P., 2008. On the ‘divergence problem ’in
northern forests: a review of the tree-ring evidence and possible causes. Glob. Planet.
Chang. 60 (3 –4), 289 –305. https://doi.org/10.1016/j.gloplacha.2007.03.004 .
Dobrovolný, P., Rybní ček, M., Kolá ř, T., Brázdil, R., Trnka, M., Büntgen, U., 2018. May –July
precipitation reconstruction from oak tree-rings for Bohemia (Czech Republic) since
AD 1040. Int. J. Climatol. 38 (4), 1910 –1924. https://doi.org/10.1002/joc.5305 .
Dole žal, J., Lehe čková, E., Sohar, K., Altman, J., 2016. Oak decline induced by mistletoe,
competition and climate change: a case study from central Europe. Preslia. 88,
323
–346.
Edenhofer, O., et al. (Eds.), 2014. Climate Change 2014: Mitigation of Climate Change.
Contribution of Working Group III to the Fifth Assessement Report of the Intergov-
ernmental Panel on Climate Change. Cambridge University Press, United Kingdom
and New York, NY, USA.
Di Filippo, A., Alessandrini, A., Biondi, F., Blasi, S., Portoghesi, L., Piovesan, G., 2010. Climate
change and oak growth decline: Dendroecology and stand productivity of a Turkey
oak ( Quercus cerris L.) old stored coppice in Central Italy. Ann. For. Sci.. 67(7),
706 –706.
Fonti, P., Solomonoff, N., García-González, I., 2007. Earlywood vessels of Castanea sativa
record temperature before their formation. New Phytol. 173 (3), 562 –570.
Fritts, H., 1976. Tree Rings and Climate. vol. 567. Academic Press, London.
Ganatsas, P., Tsakaldimi, M., 2013. A comparative study of desiccation responses of seeds
of three drought-resistant Mediterranean oaks. For. Ecol. Manag. 305, 189 –194.
Gazol, A., Camarero, J.J., Vicente-Serrano, S.M., Sánchez-Salguero, R., Gutiérrez, E., de Luis,
M., Sangüesa-Barreda, G., Novak, K., Rozas, V., Tíscar, P.A., et al., 2018. Forest resilience
to drought varies across biomes. Glob. Chang. Biol. 24, 2143 –2158. https://doi.org/
10.1111/gcb.14082 .Georgescu, C.C., Morariu, I., 1948. Monograph of Oaks in Romania. The Universe,
Bucharest (26 p. ( in Romanian ).
Gömöry, D., Schmidtova, J., 2007. Extent of nuclear genome sharing among white oak spe-
cies ( Quercus L. subgen. Lepidobalanus (Endl.) Oerst.) in Slovakia estimated by
allozymes. Plant Syst. Evol. 266 (3 –4), 253 –264. https://doi.org/10.1007/s00606-
007-0535-0 .
Gömöry, D., Yakovlev, I., Zhelev, P., Jedináková, J., Paule, L., 2001. Genetic differentiation of
oak populations within the Quercus robur /Quercus petraea complex in Central and
Eastern Europe. Heredity 86 (5), 557 –563.
Gömöry, D., Longauer, R., Paule, L., Krajmerová, D., Schmidtová, J., 2010. Across-species
patterns of genetic variation in forest trees of Central Europe. Biodivers. Conserv. 19
(7), 2025 –2038. https://doi.org/10.1007/s10531-010-9823-z .
González, I.G., Eckstein, D., 2003. Climatic signal of earlywood vessels of oak on a mari-
time site. Tree Physiol. 23 (7), 497 –504.
Grissino-Mayer, H.D., 2001. Evaluating crossdating accuracy: a manual and tutorial for the
computer program COFECHA. Tree-Ring Research 57 (2), 205 –221.
Haylock, M.R., Hofstra, N., Klein Tank, A.M.G., Klok, E.J., Jones, P.D., New, M., 2008. A
European daily high-resolution gridded data set of surface temperature and precipi-tation for 1950 –2006. J. Geophys. Res. Atmos. 113 (D20).
Holmes, R.L., 1983. Computer-Assisted Quality Control in Tree-ring Dating and Measure-
ment. Tree-Ring Bulletin. vol. 43 pp. 59 –82.
Hopkins, J.J., Kirby, K.J., 2007. Ecological change in British broadleaved woodland since
1947. Ibis 149 (s2), 29 –40.https://doi.org/10.1111/j.1474-919X.2007.00703.x .
Howell, D.C., 2011. Statistical Methods for Psychology. 5
thed. Duxbury, Paci ficG r o v e ,C A .
Huijun, W., 2001. The weakening of the Asian monsoon circulation after the end of
1 9 7 0 ' s .A d v .A t m o s .S c i .1 8( 3 ) ,3 7 6 –386.
Ivashov, A.V., Boyko, G.E., Simchuk, A.P., 2002. The role of host plant phenology in the de-
velopment of the oak leafroller moth, Tortrix viridana L. (Lepidoptera: Tortricidae).
For. Ecol. Manag. 157 (1 –3), 7 –14.
Jevšenak, J., Levani č, T., 2018. dendroTools: R package for studying linear and nonlinear
responses between tree-rings and daily environmental data. Dendrochronologia 48,
32–39.
Kern, Z., Patkó, M., Kázmér, M., Fekete, J., Kele, S., Pályi, Z., 2013. Multiple tree-ring proxies
(earlywood width, latewood width and δ13C) from pedunculate oak ( Quercus robur
L.), Hungary. Quat. Int. 293, 257 –267.
Kulfan, J., Sarva šová, L., Parák, M., Dzurenko, M., Zach, P., 2018. Can late flushing trees
avoid attack by moth larvae in temperate forests? Plant Prot. Sci. 54 (4), 272 –283.
https://doi.org/10.17221/11/2018-PPS .
Kunz, J., Löf fler, G., Bauhus, J., 2018. Minor European broadleaved tree species are more
drought-tolerant than Fagus sylvatica but not more tolerant than Quercus petraea .
For. Ecol. Manag. 414, 15 –27.https://doi.org/10.1016/j.foreco.2018.02.016 .
Larsson, L.A., Larsson, P.O., 2017. CDendro and CooRecorder (v.9.1) [Software]. Cybis
Elektronik and Data AB, Saltsjöbaden, Sweden Available from. http://www.cybis.se .
Levani č,T . ,Čater, M., McDowell, N.G., 2011. Associations between growth, wood anatomy,
carbon isotope discrimination and mortality in a Quercus robur forest. Tree Physiol.
31, 298 –308.
Lloret, F., Keeling, E.G., Sala, A., 2011. Components of tree resilience: effects of successive
low-growth episodes in old ponderosa pine forests. Oikos 120 (12), 1909 –1920.
Matisons, R., Jansons, J., Katrevi čs, J., Jansons, Ā., 2015. Relation of tree-ring width and ear-
lywood vessel size of alien Quercus rubra L. with climatic factors in Latvia. Silva
Fennica 49 (4), 1 –14.
Menitsky, Y.L., 2005. Oaks of Asia. Science Publishers (549 p).
Mikac, S., Žmega č,A . ,T r l i n ,D . ,P a u l i ć,V . ,O r šanić, M., Ani ć, I., 2018. Drought-induced shift
in tree response to climate in floodplain forests of South-eastern Europe. Sci. Rep. 8
(1), 16495. https://doi.org/10.1038/s41598-018-34875-w .
Montwé, D., Isaac-Renton, M., Hamann, A., Spiecker, H., 2018. Cold adaptation recorded in
tree rings highlights risks associated with climate change and assisted migration. Nat.
Commun. 9 (1), 1574. https://doi.org/10.1038/s41467-018-04039-5 .
Mrak, T., Štraus, I., Grebenc, T., Gri čar, J., Hoshika, Y., Carriero, G., Paoletti, E., Kraigher, H.,
2019. Different belowground responses to elevated ozone and soil water de ficit in
three European oak species ( Quercus ilex ,Q. pubescens and Q. robur ). Sci. Total Envi-
ron. 651, 1310 –1320. https://doi.org/10.1016/j.scitotenv.2018.09.246 .
Muir, G., Schloetterer, C., 2005. Evidence for shared ancestral polymorphism rather than
recurrent gene flow at microsatellite loci differentiating two hybridizing oaks
(Quercus spp.). Mol. Ecol. 14 (2), 549 –561.
Muir, G., Fleming, C.C., Schltterer, C., 2000. Taxonomy: species status of hybridizing oaks.
Nature 405 (6790), 1016. https://doi.org/10.1038/35016640 .
Nechita, C., 2013. The National Dendrochronological Network of Pedunculate and Sessile
Oak. vol. 279. Forestry Publishing House, Bucharest (in Romanian).
Nechita, C., Chiriloaei, F., 2018. Interpreting the effect of regional climate fluctuations on
Quercus robur L. trees under a temperate continental climate (southern Romania).
Dendrobiology 79, 77 –89.https://doi.org/10.12657/denbio.079.007 .
Nechita, C., Popa, I., 2012. The relationship between climate and radial growth for the oak
(Quercus robur L.) in the western plain of Romania. Carpathian Journal of Earth and
Environmental Sciences 7 (3), 137 –144.
Nechita, C., Popa, I., Eggertsson, Ó., 2017. Climate response of oak ( Quercus spp .), an evi-
dence of a bioclimatic boundary induced by the Carpathians. Sci. Total Environ.
599, 1598 –1607. https://doi.org/10.1016/j.scitotenv.2017.05.118 .
Netoiu, C., 2000. The effect of some defoliator insects on physiological processes in
Quercineae. Acta Horti Botanici Bucurestiensis 28, 337 –349.
Pérez-de-Lis, G., Rossi, S., Vázquez-Ruiz, R.A., Rozas, V., García-González, I., 2016. Do
changes in spring phenology affect earlywood vessels? Perspective from the
xylogenesis monitoring of two sympatric ring-porous oaks. New Phytol. 209 (2),521 –530.
Perkins, D., Uhl, E., Biber, P., du Toit, B., Carraro, V., Rötzer, T., Pretzsch, H., 2018. Impact of
climate trends and drought events on the growth of oaks ( Quercus robur L. and10 C. Nechita et al. / Science of the Total Environment 694 (2019) 133709

Quercus petraea (Matt.) Liebl.) within and beyond their natural range. Forests 9 (3),
108. https://doi.org/10.3390/f9030108 .
Pettenkofer, T., Burkardt, K., Ammer, C., Vor, T., Finkeldey, R., Müller, M., Krutovsky, K.,
Vornam, B., Leinemann, L., Gailing, O., 2019. Genetic diversity and differentiation of
introduced red oak ( Quercus rubra ) in Germany in comparison with reference native
North American populations. Eur. J. For. Res., 1 –11https://doi.org/10.1007/s10342-
019-01167-5 .
Popescu, F., Postolache, D., 2009. The variability of the oaks populations in the Romania,
the result of the interaction between the postglacial evolution of vegetation and
the anthropogenic in fluences. Forest Journal 124, 49 –54 (in Romanian) .
Popescu, F., Postolache, D., Curtu, A.L., 2019. The current state of forest genetic resources
in Romania. In: Šijačić-Nikoli ć, M., Milovanovi ć,J . ,N o n i ć,M .( E d s . ) ,F o r e s t so fS o u t h –
east Europe under a Changing Climate. Advances in Global Change Research. vol 65.Springer, Cham, pp. 61 –72.https://doi.org/10.1007/978-3-319-95267-3_5 .
Postolache, D., Curtu, A.L., Șofletea, N., Popescu, F., 2019. Conservation and management
of Romanian forest genetic resources in the context of climate change. In: Šijačić-
Nikoli ć, M., Milovanovi ć, J., Noni ć, M. (Eds.), Forests of Southeast Europe under a
Changing Climate. Advances in Global Change Research. vol. 65. Springer, Cham,pp. 389 –399. https://doi.org/10.1007/978-3-319-95267-3_32 .
R Development Core Team, 2008. R: A Language and Environment for Statistical Comput-
ing. R Foundation for Statistical Computing, Vienna, Austria ISBN 3-900051-07-0.
http://www.R-project.org .
Rossiter, M., Schultz, J.C., Baldwin, I.T., 1988. Relationships among defoliation, red oak
phenolics, and gypsy moth growth and reproduction. Ecology 69 (1), 267 –277.
Roy, D.B., Sparks, T.H., 2000. Phenology of British butter flies and climate change. Glob.
Chang. Biol. 6 (4), 407 –416.
Rozas, V., 2005. Dendrochronology of pedunculate oak ( Quercus robur L.) in an old-
growth pollarded woodland in northern Spain: tree-ring growth responses to cli-
mate. Ann. For. Sci. 62 (3), 209 –218. https://doi.org/10.1051/forest:2005012 .
Rybnicek, M., Cermak, P., Prokop, O., Zid, T., Trnka, M., Kolar, T., 2016. Oak ( Quercus spp .)
response to climate differs more among sites than among species in central Czech
Republic. Dendrobiology 75, 55 –
65.
Sass-Klaassen, U., Sabajo, C.R., den Ouden, J., 2011. Vessel formation in relation to leaf
phenology in pedunculate oak and European ash. Dendrochronologia 29 (3),
171 –175. https://doi.org/10.1016/j.dendro.2011.01.002 .
Scharnweber, T., Manthey, M., Criegee, C., Bauwe, A., Schröder, C., Wilmking, M., 2011.
Drought matters –declining precipitation in fluences growth of Fagus sylvatica L. and
Quercus robur L. in north-eastern Germany. For. Ecol. Manag. 262 (6), 947 –961.
https://doi.org/10.1016/j.foreco.2011.05.026 .
Schwarz, O., 1993. Quercus L. In: Tutin, T.G., Burges, N.A., Chater, A.O. (Eds.), Flora
Europaea. Cambridge University Press, Cambridge, pp. 72 –76.
Simionescu, A., Ni țescu, C., Vl ădescu, D., Vl ăduleasa, A., 1992. Phytosanitary Condition of
Forests in Romania During the Period 1976 –1985. vol. 309. Inter-Media Publishing
House, Bucharest ( in Romanian) .
Simionescu, A., Mihalciuc, V., Lupu, D., Vl ăduleasa, A., Badea, O., Fulicea, T., 2001. The
Health Status of the Forests in Romania During 1986 –2000. Mu șatinii Publishing
House, Suceava (934 p. ( in Romanian ).
Sohar, K., Läänelaid, A., Eckstein, D., Helama, S., Jaagus, J., 2014. Dendroclimatic signals of
pedunculate oak ( Quercus robur L.) in Estonia. Eur. J. For. Res. 133 (3), 535 –549.
Sparkes, T.H., Carey, P.D., Coombes, J., 1997. First lea fing dates of trees in Surrey between
1947 and 1996. London Naturalist 76, 15 –20.Speer, J.H., 2010. Fundamentals of Tree-ring Research. University of Arizona Press (333 p).
Stănescu, V., Șofletea, N., Popescu, O., 1997. Forest Woody Flora of Romania. Ceres Pub-
lishing House. Bucharest (451 p ( in Romanian ).
Steiger, J.H., 1980. Tests for comparing elements of a correlation matrix. Psychol. Bull. 87
(2), 245.
Stojanovi ć,D . B . ,L e v a n i č, T., Matovi ć, B., Orlovi ć, S., 2015. Growth decrease and mortality
of oak floodplain forests as a response to change of water regime and climate. Eur.
J. For. Res. 134 (3), 555 –567. https://doi.org/10.1007/s10342-015-0871-5 .
Stojanovi ć,B . D . ,L e v a n i č, T., Matovi ć, B., Stjepanovi ć, S., Orlovi ć, S., 2018. Growth response
of different tree species (oaks, beech and pine) from SE Europe to precipitation over
time. Dendrobiology 79, 97 –110.
Streiff, R., Labbé, T., Bacilieri, R., Steinkellner, H., Glössl, J., Kremer, A., 1998. Within-
population genetic structure in Quercus robur L. and Quercus petraea (Matt.) Liebl.
assessed with isozymes and microsatellites. Mol. Ecol. 7 (3), 317 –328.
Tîcu, L., 2012. Species of seedlings and descendants of pedunculate oak ( Quercus robur L.)
in the scienti fic reserve “Plaiul Fagului ”. Ambiental environment 6 (66), 22 –28 (in
Romanian) .
Tomescu, R., Ne țoiu, C., 2006. Control of the broad leave's mains defoliators in Romania in
2005. Proceedings of IUFRO Working Party 7.03. 10 Workshop on “Methodology of
Forest Insect and Disease Survey in Central Europe ”, (CD). Federal Research & Train-
ing Centre for Forests, Natural Hayards & Landscape, Gmunden –Austria, pp. 263 –270.
Tomescu, A., Florescu, I.I., Mihalache, A., Ștrimbei, M., Avramescu, C., 1967. Phenological
Research on the Main Autochthonous Forest Species of RSR –Synthesis for the Period
1956 –1965. vol. 221. Technical Documentation Centre for the Forest Economy,
Bucharest ( in Romanian) .
Tulik, M., 2014. The anatomical traits of trunk wood and their relevance to oak ( Quercus
robur L.) vitality. Eur. J. For. Res. 133 (5), 845 –855. https://doi.org/10.1007/s10342-
014-0801-y .
Tumajer, J., Treml, V., 2016. Response of floodplain pedunculate oak ( Quercus robur L.)
tree-ring width and vessel anatomy to climatic trends and extreme hydroclimaticevents. For. Ecol. Manag. 379, 185 –194. https://doi.org/10.1016/j.foreco.2016.08.013 .
Vanhellemont, M., Sousa-Silva, R., Maes, S.L., Van den Bulcke, J., Hertzog, L., De Groote,
S.R., Acker, J.V., Bonte, D., Martel, A., Lens, L., Verheyen, K., 2019. Distinct growth re-
sponses to drought for oak and beech in temperate mixed forests. Sci. Total Environ.
650, 3017 –3026. https://doi.org/10.1016/j.scitotenv.2018.10.054 .
Visser, M.E., Holleman, L.J., 2001. Warmer springs disrupt the synchrony of oak and win-
ter moth phenology. Proc. R. Soc. Lond. Ser. B Biol. Sci. 268 (1464), 289 –294.
Vuckovic, B., 1984. A new habitat of Quercus pedunculi floraC. Koch in Serbia (Yugoslavia).
Zbornik radova Instituta za sumarstvo i drvnu industriju 22 –23, 111 –113.
Walther, G.R., Post, E., Convey, P., Menzel, A., Parmesan, C., Beebee, T.J., Fromentin, J.M.,
Hoegh-Guldberg, O., Bairlein, F., 2002. Ecological responses to recent climate change.
Nature 416 (6879), 389.
Ważny, T., Eckstein, D., 1991. The dendrochronological signal of oak ( Quercus spp.) in
Poland. Dendrochronologia 9, 35 –49.
Wigley, T.M.L., Briffa, K.R., Jones, P.D., 1984. On the average of correlated time series, with
applications in dendroclimatology and hydrometeorology. J. Appl. Meteorol. Climatol.
23, 201 –213.11 C. Nechita et al. / Science of the Total Environment 694 (2019) 133709