ArticlePDF Available

Genetic variation in phenology and growth among and within Norway spruce populations from two altitudinal transects in Mid-Norway

Authors:
  • Norwegian Institute of Bioecoomy Research, Ås, Norwy

Abstract and Figures

Progenies from open pollinated cones collected in natural populations of Norway spruce ( (L. Karst.)) distributed along two altitudinal transects in Mid-Norway were tested in the nursery, in short term tests and in long-term field trials. The populations showed clinal variation related to the mean annual temperatures of the populations, with the earliest bud flush and cessation of shoot elongation and lowest height at age nine years for the high altitude populations. Within population variation was considerable as the narrow sense heritability for these traits was 0.67, 0.31 and 0.09 in one transect and 0.55, 0.18 and 0.14 in the other transect, respectively. Lammas shoots occurred in the short term trials with large variation in frequency between years. There was significant family variation for this trait, but also interactions between populations and year. The variance within populations was considerably larger in the populations from low altitude compared to the high-altitude populations. Significant genetic correlations between height and phenology traits and damage scores indicate that families flushing early and ceasing growth late were taller. Taller families also had higher frequencies of damages. Selection of the top 20% families for height growth in short term tests at age nine years gave a simulated gain of 11% increased height growth at age 18 years in long term trials at altitudes similar to those of origin of the populations. The gain was negative when high altitude populations were selected based on testing in the lowland. Picea abies
Content may be subject to copyright.
1
SILVA FENNICA
Silva Fennica vol. 53 no. 1 article id 10076
Category: research article
https://doi.org/10.14214/sf.10076
http://www.silvafennica.
Licenced CC BY-SA 4.0
ISSN-L 0037-5330 | ISSN 2242-4075 (Online)
The Finnish Society of Forest Science
Tore Skrøppa1 and Arne Steenrem
Genetic variation in phenology and growth among and
within Norway spruce populations from two altitudinal
transects in Mid-Norway
Skrøppa T., Steenrem A. (2019). Genetic variation in phenology and growth among and within
Norway spruce populations from two altitudinal transects in Mid-Norway. Silva Fennica vol. 53
no. 1 article id 10076. 19 p. https://doi.org/10.14214/sf.10076
Highlights
Norway spruce populations distributed along each of two altitudinal transects showed strong
clinal relationships between the annual mean temperatures of the sites of the populations and
height and phenology traits in short term tests and height in eld trials.
Large variation was present among families within populations for height and phenology
traits and with a wider range within than among populations.
Correlation patterns among traits were dierent for provenances and families.
Abstract
Progenies from open pollinated cones collected in natural populations of Norway spruce (Picea
abies (L.) Karst.) distributed along two altitudinal transects in Mid-Norway were tested in the
nursery, in short term tests and in long-term eld trials. The populations showed clinal variation
related to the mean annual temperatures of the populations, with the earliest bud ush and cessation
of shoot elongation and lowest height at age nine years for the high altitude populations. Within
population variation was considerable as the narrow sense heritability for these traits was 0.67,
0.31 and 0.09 in one transect and 0.55, 0.18 and 0.14 in the other transect, respectively. Lammas
shoots occurred in the short term trials with large variation in frequency between years. There was
signicant family variation for this trait, but also interactions between populations and year. The
variance within populations was considerably larger in the populations from low altitude compared
to the high-altitude populations. Signicant genetic correlations between height and phenology
traits and damage scores indicate that families ushing early and ceasing growth late were taller.
Taller families also had higher frequencies of damages. Selection of the top 20% families for height
growth in short term tests at age nine years gave a simulated gain of 11% increased height growth
at age 18 years in long term trials at altitudes similar to those of origin of the populations. The
gain was negative when high altitude populations were selected based on testing in the lowland.
Keywords Picea abies; adaptation; clinal variation; populations; families; bud ush; height; tree
breeding
Address 1 Norwegian Institute of Bioeconomy Research, P.O. Box 115, 1431 Ås, Norway
E-mail tore.skroppa@nibio.no
Received 20 November 2018 Revised 21 February 2019 Accepted 8 March 2019
2
Silva Fennica vol. 53 no. 1 article id 10076 · Skrøppa et al. · Genetic variation in phenology and growth among…
1 Introduction
Norway spruce (Picea abies (L.) Karst.) has its north-western distribution in the central and northern
part of Norway, a region where the glacial and postglacial history of the species has been intensively
debated based on evidences from palaeodata and from modern and ancient DNA (Giesecke and
Bennett 2004; Latalowa and van der Knaap 2006; Kullman 2008; Parducci et al. 2012a, 2012b;
Birks et al. 2012). Based on pollen data, Norway spruce established approximately 3000 years ago
in the region (Giesecke and Bennet 2004; Latalowa and van der Knaap 2006), and based on pollen
data and genetic data combined, Norway spruce in the region originate from a refugium located
on the Russian plain (Tollefsrud et al. 2008). Recently, however, analyses of ancient DNA in lake
sediments suggest that Norway spruce also was present in the Mid-Norway (Trøndelag County)
already 10 300 year ago (Parducci et al. 2012 a). The genetic variation among and within the present
Norway spruce populations in this region will be inuenced by their evolutionary history and the
adaptive processes that have taken place, in particular related to the northern climatic constraints.
No studies have so far been made presenting the patterns of variation of phenotypic traits based
on measurements in trials in the region.
During the last 30 years a number of studies have shown that Norway spruce in the boreal
region can adjust its adaptive phenology by a rapid and most likely epigenetic mechanism, through
a kind of a long-term memory of the climatic conditions during the seed maturation period (sum-
marized in: Johnsen et al. 2009). Phenology and hardiness of progenies are inuenced in a way
such that seed production in a cold environment advances bud set and cold acclimation in the
autumn as well as dehardening and ushing in the spring, whereas a warm reproductive envi-
ronment delays the timing of these events. In a recent study, Solvin and Steenrem (2019) have
performed experiments with Norway spruce provenances from seed lots produced in years with
dierent temperatures. They show that trees from warm seed years had later bud ush, bud set
and growth cessation. It has been suggested that these eects contribute, together with directional
selection, to the steep clinal variation observed in adaptive traits in Norway spruce (Johnsen and
Skrøppa 2000; Skrøppa et al. 2007).
A few provenance studies have provided information about variation in quantitative traits
in Norway spruce provenances from Mid-Norway in comparison with provenances from more
southern Norwegian origins or from the wide natural range of the species. Norway spruce prov-
enances from Mid-Norway were included in two international IUFRO trials and showed high
mortality and inferior growth when they were planted at more southern latitudes (Langlet 1960;
Fottland and Skrøppa 1989; Persson and Persson 1992). Two other studies were performed with
a representative set of provenances from the north-western region, tested at sites in Mid- and
North-Norway. These showed clinal variation related to provenance latitude and altitude for
phenology and growth traits (Bergan 1994; Skrøppa and Steenrem unpublished). Only one
study (Dietrichson 1969) has characterized variability both among and within populations from
Mid-Norway, based on four open-pollinated families from each of four populations, tested in
southern Norway. It showed signicant variation both within and among populations for phenol-
ogy traits and height growth at the age of four years. In that limited material the variation within
and among populations was lower than within and among 11 populations from southern Norway.
Clinal variation in adaptive traits with latitude and altitude has been demonstrated in several
studies with Norway spruce provenances from a wide latitudinal range in the Nordic countries
(Dietrichson 1969; Krutzsch 1975; Dæhlen et al. 1995), and from altitudinal transects in Central
Europe (Holzer 1993; Skrøppa and Magnussen 1993; Modrzynski 1995; Oleksyn et al. 1998). In
Sweden, the clinal variation among provenances is stronger in the northern than in the southern
part of the country (Danusevicius and Person 1998).
3
Silva Fennica vol. 53 no. 1 article id 10076 · Skrøppa et al. · Genetic variation in phenology and growth among…
The objective of this study was to characterize the patterns of variation of phenology and
growth traits among and within populations of Norway spruce along two altitudinal transects in
Mid-Norway. Results are presented from three types of trials with the same materials: early tests in
nurseries, short term trials on former agricultural land, and long term eld trials. The study provides
information about the relationships between traits measured at dierent ages for populations and
families. Further objectives were to demonstrate the response in long term eld trials when selection
for height growth is made in the short term trials and to provide information about the potential
gain from selection in these for the breeding programme for Norway spruce in Mid-Norway. It is
the rst study presenting variation among and within spruce populations from Mid-Norway from
trials located in the region.
2 Materials and methods
2.1 Seed collection
In 1992, cones were collected in four natural Norway spruce populations in the southern part
of Trøndelag County (Mid-Norway), close to latitude 63°N and between altitudes 300–630 m,
and in eight populations in the northern part of Trøndelag, close to latitude 64°N and between
altitudes 30–600 m (Fig. 1). The distance between the most remote populations was 78 km in the
rst and 146 km in the second collection. These two collections will be referred to as the southern
and northern transect. The intention was to obtain seeds from 15 randomly chosen trees in each
Fig. 1. Maps of the Nordic range of Norway spruce
(left) and magnied Mid-Norway region with sampled
populations (dots), nursery trial (triangles), short term
trials (plus signs) and long term trials (turned triangles).
4
Silva Fennica vol. 53 no. 1 article id 10076 · Skrøppa et al. · Genetic variation in phenology and growth among…
population. Some trees, however, in particular in populations close to the altitudinal tree limit, did
not produce enough lled seeds for their ospring to be included in further trials. Only one family
remained from the population at 600 m in the northern transect and this population was excluded
from the study. The seed lots from each tree were kept separate to allow for characterization of
variability among half-sib families within populations.
Daily mean temperatures for 1 × 1 km grid cells in Mid-Norway were obtained from the
Norwegian Meteorological Institute (sharki.oslo.dnmi.no/) for long-term annual mean temperature
for the reference period 1961–1990 and for the seed year 1992. Average temperature values for
nearby grid cells with approximately same altitude as the actual populations where cones were
collected were used to model temperature proles for these sites, as described by Solvin and Stef-
fenrem (2019). For the long-term temperatures, the annual means for the period 1961–1990 varied
between 0.7 and 5.1 °C, and the mean temperatures for 1992 of the three months July, August and
September, covering the seed maturation period, varied from 9.4 to 12.2 °C. The annual mean
temperatures were closely related to the altitude in each transects, with Pearson correlations coef-
cients r = –0.77 in the southern and r=–0.96 in the northern transect.
2.2 Nursery trial
Filled seeds from 136 families from 11 populations were sown and germinated in multipot con-
tainers at Stiklestad Nursery (63°48´N, 11°33´E, altitude 45 m) in the last week of May 1995. The
experimental design was four replicates (blocks), each with 41–42 sown seed per family plot. The
cultivation followed standard nursery routines as regards watering and fertilization. Starting from
August 17 the same year, each seedling in three replicates was visually examined for visible apical
bud. This was done once a week until September 6. The accumulated percentage of seedlings with
bud was calculated per plot for each of the four assessments.
2.3 Short term trials
Three short terms trials were planted on former cultivated elds in 1997 with the two-year-old
seedlings (Fig. 1). One trial “Selbu” (63°15´N, 10°48´E, altitude 240 m) was established in the
southern part of Trøndelag with the 55 families from the four populations in the southern transect.
The 81 families from seven populations from the northern transect were planted at two sites,
Stiklestad, the same site as the nursery, and Kvatninga (64°28´N, 11°39´E, altitude 15 m), both
located in northern Trøndelag. The experimental design was randomized single tree plots with 30
replicates at 1m spacing at all sites.
Measurements were made of tree heights in the trials at ages 6 and 9 years. At the same
time, each tree was scored for “stem damage”: occurrence of double stems, spike knots or other
visible stem damage. The stage of terminal bud ush was scored according to the Krutzsch scale
(Krutzsch 1973) at one occasion at the beginning of each of growing seasons six and eight. At
Selbu and Stiklestad, the leader shoot length was measured at one date close to the termination
of shoot elongation (June 30), and a second time after shoot growth cessation. The percentage of
elongation attained at the rst date will be used a measure of the timing of leader shoot growth
cessation. Lammas shoots occurred at Stiklestad and Kvatninga, and assessments were made of
such occurrence with more than 2 cm shoot extension on leader, at ages six, eight and nine years
at Stiklestad and ages six and nine years at Kvatninga.
5
Silva Fennica vol. 53 no. 1 article id 10076 · Skrøppa et al. · Genetic variation in phenology and growth among…
2.4 Field trials
From both transects, northern and southern, eld trials were planted at three forest sites in Trøndelag
County, between altitudes 240 and 400 m, the same year as the short term trials were established
(Fig. 1). From families with sucient number of vital seedlings, a sample of 2–4 seedlings was
pooled to represent the population. Seedlings were planted in single tree plots with one tree from
each population per replicate in 40 blocks at 2 m spacing. Thus, the population is the experimen-
tal unit, although family identity was kept for individual trees. Measurements of tree height and
diameter were made 16 years after planting at age 18 years from seed, and assessments of double
stems and spike knots were made at the same time.
2.5 Statistical analyses
To obtain more normal distributions and homogeneity of variances, transformations of some traits
were made before analyses of variance were performed. The bud set percentage per plot and the
shoot growth percentage of individual trees were transformed by the arcsine square root trans-
formation. The scoring of bud ush (classes), lammas growth and damage (binary observations)
were transformed to normal scores, within blocks and sites, by the Blom method (PROC RANK,
SAS Institute 2003).
All statistical analyses were performed with procedures in SAS (SAS Institute 2003) using
PROC GLM, PROC VARCOMP and PROC MIXED for analyses of variance and estimation of
variance components. In the analyses of variance, populations were considered as xed eects as
they in each transect were sampled along altitudinal gradients at approximately the same latitude.
Families were considered as random and it was assumed that there were similar levels of variance
within each population.
Analyses of variance of traits measured in the nursery trial and the short term trial at Selbu
(southern transect) were made based on the model 1:
YPFBE
ijk iijkijk


1
() ,
here Yijk is the is the observed value for the family j from population i block k, µ is the grand mean,
P is the xed eect of population, F the random eect of family, B is the eect of block and E is
the residual error. The random eects were assumed to be normal and independently distributed
with expectation zero and variance components σ2F, σ2B and σ2E, respectively.
In the combined analyses of measurements from Stiklestad and Kvatninga (northern transect)
a xed eect of site (S), its xed eect of the interaction with populations (SP) and random eect
of interaction with families (SPF) were added. The block eect was then nested within site, and
the analyses were based on the model 2:
YSPSPF SPFB E
ihjk hi ih ij ih jkhihjk
  

2
() () () .
The model for the analyses of the eld trials (model 3) included sites, populations from
both transects, site-by-population interactions and blocks nested within sites, in addition to the
residual error:
YSPSPB E
igkgijgkgigk
  

3
() .
6
Silva Fennica vol. 53 no. 1 article id 10076 · Skrøppa et al. · Genetic variation in phenology and growth among…
Variance components were estimated for the random eects, and p-values of the appropriate
F-tests are presented. The narrow sense heritability estimates of bud set and the traits at Selbu were
estimated as h2 = 2F/(σ2F + σ2E), assuming that the family variance component approximately
equals one third of the additive genetic variance as open-pollinated progenies are more closely
related than half-sibs (Squillace 1974). In the combined analyses of Stiklestad and Kvatninga, the
site-by-family interaction was included in the denominator of the heritability estimate. Its standard
error was calculated by the Taylor expansion for variances of ratios (Lynch and Walsh 1998) in
SAS/IML. In the short term trials, genetic correlations between height, bud ush and shoot elon-
gation and their standard errors were estimated by calculations made in SAS/IML based on the
output from multivariate analyses in PROC MIXED (Holland 2006). Family breeding values were
estimated and used to study “genetic” Pearson-correlations between bud set in the nursery and
eld data, and between traits that caused convergence problems (e.g. damages) in the multivariate
analysis in PROC MIXED.
Pearson correlation coecients were calculated between population means of traits meas-
ured in the short term trials and the long-term mean and 1992 mean of months July to September
at each population locality, and also between traits and altitude and mean annual temperatures of
the populations.
2.6 Estimation of potential gain from selection for height based on short term testing
In the national breeding strategy of Norway spruce, two breeding zones are dened for Mid-Norway;
low altitude zone below 250 m altitude, and high altitude zone above 250 m altitude (Edvardsen
et al. 2017). The families from the populations in Trøndelag were assumed to belong to three
subsets of populations, below or above altitude 250 m, in the two breeding zones, containing the
families from four, four and three populations, respectively. For each of these subsets the 20%
tallest families were selected based on predicted BLUP family values for tree height at age nine
years measured in the short term trials. Least square (LS) means of families were calculated from
the measurements in the eld tests, and the mean values of the trees from selected families were
compared to the means of all families in the subset.
3 Results
3.1 Variation in bud set in both transects
The mean percentages of seedlings with visible terminal buds at the end of the rst growth season
were 7.4%, 67.8%, 99.0% and 100% at the four dates August 17th, 23rd, 31st and September 6th,
respectively. The variation among the populations was largest for those from the northern transect
which also have the widest range in altitude, from 20 to 500 m. The populations from the highest
altitudes in the northern transect were the rst to set bud, and those from low altitudes were the
latest. A strong relationship was present between the bud set percentages and the mean annual
temperature of the population sites with correlation coecients r = –0.80 and r = –0.92 the south-
ern and northern transect, respectively (Fig. 2). Signicant variation among families was present
within all populations, with a range on August 23rd from 27 to 81% in the population from 50 m
altitude with the lowest mean value, and from 51 and 100% in the population from 400 m altitude
with the highest mean, both populations from the northern transect. The estimate of the heritability
based on plot means was 0.54 and 0.53 (Table 1), respectively.
7
Silva Fennica vol. 53 no. 1 article id 10076 · Skrøppa et al. · Genetic variation in phenology and growth among…
Fig. 2. Norway spruce populations from Mid-Norway. Relationships between the long-term
annual mean temperatures at the population sites and population means of bud set, bud ush,
growth cessation and height at age nine years. Black circles denote the southern and white circles
the northern transect populations. 1) Bud set assessed in the common nursery trial at Stiklestad,
2) Growth cessation measured in the short term trials at Selbu and Stiklestad, 3) Bud ush as-
sessed at Selbu and at Stiklestad and Kvatningen, 4) Height measured at Selbu and at Stiklestad
and Kvatningen.
Table 1. Norway spruce populations and families from Mid-Norway. Analyses of variance
of transformed proportion of seedlings with apical bud measured in the nursery at August
23 in the nursery (Stiklestad). Shown are estimates of variance components, p-values and
heritabilities at plot level (standard error in parentheses).
Southern transect populations Northern transect populations
Variance p-value Variance p-value
Populations 0.013 < 0.0001
Families (populations) 0.016 0.0003 0.015 <0.001
Blocks 0.128 <0.0001 0.032 0.001
Populations × blocks 0.004 0.028 0.009 <0.0001
Error 0.037 0.031
Heritability 0.54 (0.17) 0.53 (0.18)
8
Silva Fennica vol. 53 no. 1 article id 10076 · Skrøppa et al. · Genetic variation in phenology and growth among…
3.2 Variation in the short term trials
3.2.1 Populations from the southern transect tested at Selbu
At age nine years, the mortality was 21% at Selbu, and the mean height was 127 cm. Signicant
dierences were present both among and within populations for height, the bud ushing scores and
shoot growth cessation (Table 2). The trees from the populations from the lowest altitudes 300 and
400 m were the tallest; they had a latest bud ush in the spring and the latest cessation of leader
growth (Fig. 3a). However, as the range of variation among families within populations was wide
and there was overlap in the genetic variation among all populations for these traits (Fig. 3a). The
estimate of heritability was as low as 0.09 for height at the age nine years, but was much higher for
bud ush (h2 = 0.67) and shoot growth cessation (h2 = 0.32). For damage scores, the dierences both
among and within populations were small, and the heritability was low (Table 2). The frequency
of lammas shoots was negligible in this trial.
Quite strong and positive genetic correlations were estimated between the timing of bud
ush and growth cessation (Table 3). The last trait was negatively correlated to height, showing that
the best growing families had the latest termination of shoot growth. A signicant positive genetic
correlation (Table 3) was present between damage and height, showing that the tallest families
within each population tended to have a higher frequency of damaged trees. It is remarkable that
no genetic relationships were present between bud set at the end of the rst growth season and any
of the traits measured in the short term test (Table 3).
At the population level, strong relationships were present between height and the timing of
bud ush and growth cessation with absolute values of correlation coecients higher than 0.90
(not shown). These correlations were closely related to the relationships present between the annual
mean temperature of the populations and height, growth cessation and bud ush, as shown in Fig. 3.
The general pattern was that lower altitude populations with the highest mean temperatures were
ushing and ceasing growth later, and grew taller. The correlation coecients between the traits
and the mean temperatures of the months July, August and September of the seed year 1992 were
lower than those based on the mean annual mean temperatures.
Table 2. Norway spruce populations and families from Mid-Norway. Analyses of variance of traits measured in the
short term trial at Selbu. Shown are estimates of variance components for random eects, p-values from the F-test and
heritabilities (standard error in parentheses).
Height
age 9
Bud ush 1)
age 8
Growth cessation 2)
age 9
Damage
age 9
Variance p-value Variance p-value Variance p-value Variance p-value
Populations 0.0004 0.0017 0.0009 0.92
Families (populations) 33.68 <0.0001 0.18 <0.0001 0.0028 <0.0001 0.008 0.04
Blocks 298.60 <0.0001 0.0013 <0.0001
Error 1086.105 0.64 0.0233 0.500
Heritability 0.09 (0.04) 0.67 (0.12) 0.32 (0.09) 0.05 (0.03)
1) Scored according to Krutzsch (1973). 2) Proportion of growth accomplished at June 30.
9
Silva Fennica vol. 53 no. 1 article id 10076 · Skrøppa et al. · Genetic variation in phenology and growth among…
Fig. 3. Norway spruce populations and families from Mid-Norway. Means and range
of variation shown for height, shoot growth cessation and bud ush score in the short
term trials for the southern (a) and northern transect populations (b), ranked according
to altitude.
Table 3. Norway spruce families from Mid-Norway. Estimates of genetic correlations
(with standard errors or p-values) for traits measured in the short term trial at Selbu.
P-values are given when inferences are based on Pearson-correlations between BLU-
predicted breeding values.
Bud ush Growth cessation 1) Damage Bud set 2)
Height 0.20 (0.23) –0.46 (0.21) 0.38 (p = 0.004) –0.08 (ns)
Bud ush 0.65 (0.12) 0.18 (ns) –0.05 (ns)
Growth cessation –0.10 (ns) 0.03 (ns)
Damage 0.07 (ns)
1) Proportion of shoot growth accomplished at June 30. 2) Measured in the nursery (Stiklestad).
10
Silva Fennica vol. 53 no. 1 article id 10076 · Skrøppa et al. · Genetic variation in phenology and growth among…
3.2.2 Northern transect populations tested at Stiklestad and Kvatninga
The mortality was 30% at Stiklestad and 1% at Kvatninga at age nine years, and the mean heights
at were 171 and 160 cm, respectively. Population means of height across the two sites ranged from
148 cm to 183 cm, and the trees from the lowland populations were on average taller than those
from higher altitudes (Fig. 3b). In the analysis of variance, signicant variation for height at age
nine years was found both among populations and among families within populations (Table 4).
Signicant interactions were also present between populations and sites and between families and
sites. The variance component of the last interaction was, however, one third of the size of the
component between families. The estimate of heritability for height was 0.14.
The trees in the populations from altitudes 400 to 500 m and with the lowest mean annual
temperatures had the earliest bud ush (Fig. 2) and signicant variation was found (Table 4) both
among populations and families, with no interactions with test sites present. The estimate of herit-
ability of this trait was 0.55.
On the last day of June, the trees from the populations at high altitudes had nearly com-
pleted 90% of their shoot growth, whereas those from low altitude populations had completed
nearly 75% of their annual growth (Fig. 3b). Highly signicant variation was present both among
populations and among families (Table 4), and the estimate of heritability was 0.18. Less, but still
signicant, variation was present among families for damage scores. The heritability for this trait
was, however, low.
The percentages of trees with lammas shoots were 6.8, 18.1 and 22.8 at Stiklestad in 2000,
2002 and 2003, respectively. At Kvatninga, these gures were 25.2 and 5.6 in 2000 and 2003. The
analysis of variance presented in Table 4 was based on the assessments in 2003 at Stiklestad and
2000 at Kvatninga. There was no signicant variation among populations due to a strong interac-
tion between sites and populations. In a separate analysis of the assessments at Stiklestad 2003,
the variation among populations was highly signicant (p < 0.0001) with a range of variation from
12 to 34%. Here the variation among families was in particular large, with a range of 11–52%
and 18–64% among families in the two lowland populations (30 and 50 m), as well as the 400 m
Table 4. Norway spruce populations and families from Mid-Norway. Analyses of variance of traits measured in the
short term trials at Stiklestad and Kvatninga. Height, bud ush, lammas shoots and damage are from measurements/
assessments at both sites, while growth cessation was measured only at Stiklestad. Shown are estimates of variance
components for random eects, p-values from the F-test and heritabilities (standard error in parentheses).
Height
age 9
Bud ush 1)
age 6
Growth cessation 2)
age 9
Lammas shoots
age 6 3) and 9 4)
Damage
age 9
Variance p-value Variance p-value Variance p-value Variance p-value Variance p-value
Sites 0.004
Populations <0.0001 <0.0001 0.0001 0.16 0.48
Sites × populations 0.02 0.48 <0.0001 0.78
Families (populations) 60.06 <0.0001 0.14 <0.0001 0.0016 0.0003 0.030 <0.0001 0.007 0.02
Sites × families
(populations) 19.98 <0.0001 0.0 0.001 0.32 0.001 0.37
Replicates (site) 117.86 <0.0001 0.0005 0.008
Error 1250.81 0.63 0.0238 0.464 0.499
Heritability 0.14 (0.04) 0.55 (0.08) 0.18 (0.07) 0.18 (0.04) 0.04 (0.02)
1) Scored according to Krutzsch (1973). 2) Proportion of shoot growth accomplished at June 30. 3) Assessed at Kvatninga.4) Assessed
at Stiklestad.
11
Silva Fennica vol. 53 no. 1 article id 10076 · Skrøppa et al. · Genetic variation in phenology and growth among…
population (0 to 39%). The interaction between sites, years and populations is clearly seen in Fig. 4
as the two high altitude populations from 400 and 500 m perform dierently, with high levels of
lammas shoots at Kvatninga and lower levels at Stiklestad.
The northern transect populations originate from altitudes between 30 and 500 m above
sea level and cover two zones in the breeding program of Norway spruce in this region (Edvard-
sen 2017). When analyses of variance were made for these two zones separately (0–250 m: four
populations; and 250–500 m: three populations), the estimates of heritability were reduced for the
high altitude zone. The largest changes were present for ushing, with estimates 0.57 and 0.36,
while smaller changes occurred for height, with estimated values 0.15 and 0.13 for the two zones,
respectably.
Positive genetic correlations of similar magnitude to that observed in the southern transect
were present between the timing of ushing and shoot growth cessation (Table 5). A positive cor-
relation was also found between ushing and height growth, and between lammas growth and
ushing and height, and also between damage and height and ushing. Similar to the southern
transect families, no signicant genetic relationships were present between terminal bud set and
the traits measured in the short term tests.
Fig. 4. Norway spruce populations from Mid-Norway. Re-
lationship between the percentages of trees with lammas
shoots at Kvatninga 2000 (white circles) and 2003 (trian-
gles) and Stiklestad 2003 (black circles) and the altitude of
the populations.
Table 5. Norway spruce families from Mid-Norway. Estimates of genetic correlations (with standard errors or p-values)
for traits measured in the short term trial at Stiklestad and Kvatninga and for bud set assessed in the nursery (Stiklestad).
P-values are given when inference are based on Pearson-correlations between BLU-predicted breeding values.
Bud ush Growth cessation 1) Lammas shoots Damage Bud set 2)
Height 0.58 (0.12) –0.09 (0.22) 0.32 (0.16) 0.42 (p < 0.0001) –0.24 (p = 0.04)
Bud ush 0.62 (0.16) 0.45 (0.12) 0.43 (p < 0.0001) –0.17 (ns)
Growth cessation 0.32 (0.26) 0.29 (p = 0.01) 0.17 (ns)
Lammas shoots 0.27 (p = 0.01) –0.08 (ns)
Damage –0.05 (ns)
1) Proportion of shoot growth accomplished at June 30. 2) Measured in the nursery (Stiklestad).
12
Silva Fennica vol. 53 no. 1 article id 10076 · Skrøppa et al. · Genetic variation in phenology and growth among…
The Pearson correlation coecient at the population level between the timing of ushing
and growth cessation was 0.92, and between height growth and ushing and growth cessation,
–0.91 and –0.98, respectively.
The populations showed strong trends related to their mean annual temperatures for height,
bud ush and growth cessation (Fig. 2) with absolute values of Pearson correlation coecients
between the population means in the range 0.92 to 0.99.
3.3 Variationintheeldtrialsforbothtransects
The mean heights in the three eld trials 16 years after planting were 281, 384 and 384 cm, and the
mortality was 8, 12 and 23%. There were small dierences among populations for mortality and
the frequency of trees with spike knots and double leaders. In the analyses of variance of height
and diameter across the three sites signicant variation was present among the 11 populations for
height (p = 0.01), but not for diameter (p = 0.13). No interactions were present between sites and
populations. Strong clinal relationships were present between the population mean height and
diameter and the annual mean temperatures of the populations, as shown in Fig. 5. The populations
from the lower altitudes and with the highest mean temperatures were the tallest, except for one
population from low altitude (30 m) and quite high mean annual temperature that had a low mean
height and deviated from the trend. The populations from latitude 63°N had higher annual mean
temperatures and were taller than those from the same altitudes at latitude 64°N.
3.4 Relationshipsbetweennursery,shorttermandeldtrials
At the population level, strong relationships were present between mean tree heights in the eld
trials and heights and the timing of the growth season in the short term trials, as well as for bud
set, all with absolute values of the correlation coecients higher than 0.80 (data not shown). This
corresponds to the clinal variation observed in these traits relative to the mean temperatures of the
populations. No such relationships were found between the traits measured at an early age and
mortality or the frequency of double leaders or spike knots in the eld trials.
Fig. 5. Norway spruce populations from Mid-Norway. Mean heights and diameters of the pop-
ulations at age 18 years from seed in three eld trials plotted against the mean annual tem-
peratures of the populations. Black circles denote the southern and white circles the northern
transect populations.
13
Silva Fennica vol. 53 no. 1 article id 10076 · Skrøppa et al. · Genetic variation in phenology and growth among…
3.5 Response to selection for height in the short term trials
For each of the three subsets of families, the mean performance in the eld trials of the best 20%
of families selected for height in the short term trials are presented in Table 6. For the four popula-
tions in the southern transect the mean of the selected families was 11% above the total mean of the
trees from all families in these populations. The selected families from low altitude in the northern
transect were on average 4% taller than all families, while there was a small negative eect of
the selection in the high altitude populations. There were only small dierences in mortality and
damage between the two groups of families.
4 Discussion
4.1 Patterns of variation
The Norway spruce populations tested were sampled along two altitudinal transects approxi-
mately at latitudes 63°N and 64°N in Trøndelag. A striking result is the clinal relationships pre-
sent between the mean annual temperatures of the populations in each of the two transects and
height and phenology traits in the short term tests and height in the eld test. This resembles the
clinal variation patterns related to latitude and altitude found in earlier provenance studies with
Norway spruce (Dietrichson 1969; Krutzsch 1975; Holzer 1993; Skrøppa and Magnussen 1993;
Dæhlen et al. 1995; Modrzynski 1995; Oleksyn et al. 1998). The study also demonstrates that
populations in the southern transect have better height growth than populations from the same
altitude in the northern transect, and that this dierence can correspond to a change in altitude
from 100 to 300 m.
It has been assumed that long-term adaptation to the climatic conditions has played a major
role for the observed clinal variation patterns found for phenology traits such as bud burst in the
spring and growth cessation and bud set in the autumn (Eriksson et al. 2013). It may be advanta-
geous for northern and high altitude populations to respond rapidly to high temperatures in the
spring and likewise react to short night lengths and lower temperatures before the end of the summer
for building up hardiness. However, there are other factors that may contribute to the patterns of
variability found among populations. We have observed that trees in the low altitude populations
Table 6. Norway spruce populations and families from Mid-Norway. Mean values across three
eld trials at age 18 year of selected and all families for each of three subset of populations.
Subset of populations
Height
cm
Diameter
mm
Mortality
%
Damage
%
Southern transect populations > 250 m altitude
Selected families 390 49 11.5 29.4
All families 350 46 18.1 21.7
Northern transect populations < 250 m altitude
Selected families 368 44 18.8 17.8
All families 353 43 16.4 23.8
Northern transect populations > 250 m altitude
Selected families 313 37 13.1 28.3
All families 321 40 19.0 21.5
14
Silva Fennica vol. 53 no. 1 article id 10076 · Skrøppa et al. · Genetic variation in phenology and growth among…
ower earlier than those in the higher altitude populations due to higher spring temperatures in
the lowland. Pollen exchange occurs between populations in the hilly landscape, and it is likely
that the late owering trees at the lower altitude will be pollinated by pollen from early ower-
ing individuals from higher altitudes. At the same time there will be a positive assortative mating
by owering time within populations. Since both owering time and the timing of bud ush are
inuences by temperature sums in spring, this fact can contribute to the variation observed in the
last trait. It should generally counteract the clinal pattern of variation observed; see discussion in
Soularue and Kremer (2012, 2014).
Another contributing factor to the observed clinal variation may be the epigenetic memory
eect of temperature conditions during seed maturation that in particular inuences phenology
traits of Norway spruce seedlings and young trees (Johnsen et al. 2009). High temperatures during
seed maturation imply delayed phenology and lower temperatures the opposite. Recent results from
trials with Norway spruce provenances from seed lots collected in Mid-Norway in years with dif-
ferent climatic conditions conrm that such an epigenetic memory eect is an important component
of the clinal patterns found in phenology traits (Solvin and Steenrem 2019). In our study, high
values were found for the correlation coecients between such traits and both for the long term
annual mean temperature and for the mean temperature during the seed maturation period for each
population. It is therefore not possible with our data to separate between inuences of long-term
genetic adaptation to temperature conditions and epigenetic eects of temperatures the specic
seed year. Due to the close relationship between phenology traits and early growth similar clinal
variation patterns may be present for height and diameter growth. However, there may still be
deviations from the general pattern among populations within the same altitudinal transect, most
likely due to dierences in local climatic conditions. Examples are the population at altitude 630
m in the southern transect in Trøndelag and that at altitude 30 m in the northern transect. Genetic
variation among stands may also occur due to dierences in long-term population history, such
as inbreeding due to small population size. Quite large dierences among stands within the same
provenance were found by Dietrichson (1973) who tested Norway spruce families, stands and
provenances from altitude 600–750 m in southern Norway.
Another important result is the large variation found among families within populations. It
was signicant for most traits and had a wider range than the variation among populations (Fig. 2).
This agrees with earlier reports on within-population variability in Norway spruce in southern
Norway (Dietrichson 1971, 1973; Skrøppa 1982, 1991). The variation among families within
populations was in particular large for the timing of bud ush, which had its largest range of vari-
ation with two units on the Krutzsch scale, corresponding to eight to ten days, in the two lowland
populations from the northern transect. In the two populations from 400 and 500 m altitude, the
range of variation was half that size. This trait is evidently of high climatic adaptive value and
with the strongest selection pressure at high altitudes. One reason for the reduced variability within
populations at high altitudes may also be due to a reduced eective population size. This was
demonstrated in the population from 600 m in the northern transect from which only one out of
ten trees produced seeds with a sucient number of seedlings to be included in the trials. Similar
patterns of within-population variability in adaptive traits, with less variation among families in
a northern than in three more southern Norway spruce populations in Sweden, was demonstrated
by Ekberg et al. (1985). The range of variation among families in height growth was also reduced
in the high altitude populations, but to a smaller extent than for the adaptive traits. These dier-
ences in genetic variability among populations for the timing of bud ush are also reected in the
dierences in heritability estimates for the two altitude zones. Our estimates are comparable to
those of Ekberg et al. (1985) who presented estimates at the family level. The families within low
altitude populations from Nord-Trøndelag have similar or even larger range of variation than is
15
Silva Fennica vol. 53 no. 1 article id 10076 · Skrøppa et al. · Genetic variation in phenology and growth among…
present for the same traits within populations in southern Norway as demonstrated by Dietrichson
(1969) and Skrøppa (1982).
The assessments of lammas shoots that were made in three years at Stiklestad and in two years
at Kvatninga, showed large dierences both between years, between sites and a strong interaction
between populations and sites. The interaction could be an eect due to dierences in age when
the analysed assessments were made; six years at Kvatninga and nine years at Stiklestad. It could
also be caused by dierent responses of populations to variation in soil and nutrient conditions or
to temperature and precipitation, which clearly exists between the two sites. At the family level,
no interactions were present.
4.2 Trait relationships
Quite strong correlations, positive or negative, were present at the population level between bud
set, the timing of ushing and growth cessation and height. A number of studies have presented
similar relationships for provenances, e. g. Skrøppa and Magnussen (1993), who discussed the
interrelationships between traits that are triggered by the same environmental factors such as tem-
perature and photoperiod and show the same clinal patterns of variation.
The correlations were quite dierent for the same traits for families within populations. Such
dierences in correlations at the provenance and family level have earlier been demonstrated and
discussed by several authors, e.g. Ekberg et al. (1994), Skrøppa et al. (1999), Johnsen and Skrøppa
(2000). A striking result is the lack of genetic relationship between terminal bud set at the end of
the rst growth season and other traits, similar to results found and discussed by Skrøppa (1991)
and Johnsen and Skrøppa (2000). This result clearly shows that selection for higher frost hardiness
based on the timing of terminal bud set of one year-old seedlings does not have the intentional
eect for improvements of Norway spruce.
A quite high positive genetic correlation was found between the timing of ushing and
height growth for the families from the northern transect. This is contrary to what has been found
in progeny trials with this species in the southern part of the Nordic countries (Hannerz et al. 1999;
Skrøppa and Steenrem 2015), where negative correlations have been estimated. In the southern
areas late spring frosts occur frequently and cause damage on trees from early ushing families
and are detrimental for height growth. Such frost events did not occur on the test sites Stiklestad
and Kvatninga, and this may explain these dierences. Another reason might be that the limited
growing seasons at these latitudes benet earlier ushing families, as these might utilize the
favourable growing conditions in June and July more eectively provided that they do not cease
the leader elongation too early.
The relationships found between the frequency of lammas shoots and the phenology traits
indicate that families that have an early growth start and early growth cessation are more prone to
develop lammas shoots than those that have a later development. This corresponds to the results
from progeny tests in southern Norway (Skrøppa and Steenrem 2017). During the last years an
increasing occurrence of lammas shoots has been reported in Norway spruce plantations on the
most productive sites in the lowland area in the south-eastern part of Norway (Kvaalen et al. 2010;
Granhus et al. 2018). If their occurrence increases, with development of multiple tops (forking)
the following years, this may be a negative factor for the production of high quality timber. In
these trials, the genetic correlation between the frequencies of lammas shoots and damage was
signicant, but moderate (r = 0.27).
16
Silva Fennica vol. 53 no. 1 article id 10076 · Skrøppa et al. · Genetic variation in phenology and growth among…
4.3 Implications for tree breeding
The results from these trials support the division into two altitudinal breeding zones for Mid-
Norway, below and above 250 m altitude (Edvardsen 2017). However, more results from on-going
trials should give further information to the question of whether there is need for a breeding zone
for the most northern area north of latitude 65°N.
The selection of families in the short term tests was ecient for the southern transect popula-
tions and worked partly for the low altitude populations in the northern transect, based on the results
from the long term eld trials. However, it did not produce any gain for the high altitude populations
in the northern transect. This shows that the low altitude test sites at Stiklestad and Kvatninga are
not appropriate for testing for high altitude localities. The higher gain for the southern compared
with the more northern families may be due to the more appropriate altitude of the short term test
(240 m) relative to the eld trials all located at altitudes above 240 m and the transfer from low to
higher altitudes. It stresses again the importance of choosing appropriate sites for progeny testing.
There are genetic correlations between phenology and growth that must be considered in
a breeding program. As individuals with later growth cessation tend to be taller, there is a risk
of unintentional changes in hardiness as selection for growth goes on for generations. Likewise,
genetic correlations between height and damage show that selection for growth must be weighed
against the frequency of damage.
Acknowledgements
The seed collections in 1992 were nancially supported by the Norwegian Ministry of Agriculture
and Food. The nursery and short term trials were established at Stiklestad and Kvatninga nurser-
ies with support from sta from Skogplanter Midt-Norge and from Skogselskapet i Trøndelag.
Measurements in the eld trials, statistical analyses and the writing of the manuscript were done as
part of the project “Klimatilpasset gran i Midt-Norge” that was nancially supported by Allskog,
Skogtiltaksfondet, Utviklingsfondet for skogbruket, Det norske Skogfrøverk and the European
Union’s Horizon 2020 research and innovation programme under grant agreement No 773383
(B4EST). We would like to thank all the institutions for their support and several colleagues that
have contributed, in particular Ragnar Sand, Øyvind Meland Edvardsen, Torstein Myhre and Hans
Christian Brede.
References
Bergan J. (1994). Planter etter frø fra utvalgte granbestand (Picea abies (L.) Karst.) på forskjel-
lige breddegrader plantet under ulik jord- og klimaforhold i Troms. [Plants of seed collected
from a selection of spruce stands (Picea abies (L.) Karst.) at dierent latitudes planted under
various soil and climatic conditions in Troms]. Reports of the Norwegian Forest Research
Institute 5/94: 1–51.
Birks H.H., Giesecke T., Hewitt G.M., Tzesakis P.C., Bakke J., Birks H.J.B. (2012). Comment on
“Glacial survival of boreal trees in northern Scandinavia”. Science 338: 742-a.
Dæhlen A.G., Johnsen Ø., Kohmann K. (1995). Autumn frost hardiness of young seedlings of
Norway spruce from Norwegian provenances and seed orchards. Skogforsk Research Paper
1/95: 1–24. ISBN 82-7169-701-3. [In Norwegian with English summary].
Danusevicius D., Person B. (1998). Phenology of natural Swedish populations of Picea abies as
17
Silva Fennica vol. 53 no. 1 article id 10076 · Skrøppa et al. · Genetic variation in phenology and growth among…
compared with introduced seed sources. Forest Genetics 5: 211–220.
Dietrichson J. (1969). Growth rhythm and yield a related to provenance, progeny and environment.
FAO IUFRO FO-FTB-69-2/3, 2nd World Consultation on Forest Tree Breeding, Washington
7–16 August 1969.
Dietrichson J. (1971). A summary of studies on genetic variation in forest trees grown in Scandi-
navia with special reference to the adaptation problem. Meddelelser fra Det norske skogfor-
søksvesen 29: 21–59.
Dietrichson J. (1973). Genetic variation among trees, stands, and provenances of Norway spruce
in alpine southern Norway. IUFRO Norway spruce Working Party s2-2-11. Biri Norway. 11 p.
Edvardsen Ø.E., Steenrem A., Johnskås O.R., Johnsen Ø., Myking T., Kvaalen H. (2017). Skog-
frøverkets strategi for skogplanteforedling 2010–2040 (revidert 2017). Stiftelsen det norske
Skogfrøverk. 22 p. [In Norwegian with English abstract].
Ekberg I., Eriksson G., Weng Y. (1985). Between and within-population variation in growth rhythm
and plant height in four Picea abies populations. Studia Forestalia Suecica167: 1–14.
Ekberg I, Eriksson G, Namkoong G, Nilsson C, Norell L. (1994). Genetic correlations for growth
rhythm and growth capacity at ages 3–8 years in provenance hybrids in Picea abies. Scandina-
vian Journal of Forest Research 9(1–4): 25–33. https://doi.org/10.1080/02827589409382809.
Eriksson G., Ekberg I., Clapham D. (2013). Genetics applied to forestry. An introduction. SLU,
Uppsala, Sweden. 206 p. ISBN 978-91-576-9187-3.
Fottland H., Skrøppa T. (1989). The IUFRO 1964/68 provenance experiment with Norway spruce
(Picea abies) in Norway. Variation in mortality and height growth. Reports of the Norwegian
Forest Research Institute 43(1): 1–30.
Giesecke T., Bennett K.D. (2004). The Holocene spread of Picea abies (L.) Karst. in Fennoscandia
and adjacent areas. Journal of Biogeography 31(9): 1523–1528. https://doi.org/10.1111/j.1365-
2699.2004.01095.x.
Granhus A., Metslaid M., Kvaalen H., Søgaard G. (2018). Tree, stand and site characteristics
aecting the occurrence of lammas growth and multiple tops in eld-grown Norway spruce.
New Forests. https://doi.org/10.1007/s11056-018-9664-2.
Hannerz M., Sonesson J., Ekberg I. (1999). Genetic correlations between growth and growth
rhythm observed in a short-term test and performance in long-term eld trials of Norway
spruce. Canadian Journal of Forest Research 29(6): 768–778. https://doi.org/10.1139/x99-056.
Holland J.B. (2006). Estimating genotypic correlations and their standard errors using multivari-
ate restricted maximum likelihood estimation with SAS Proc MIXED. Crop Science 46(2):
642–654. https://doi.org/10.2135/cropsci2005.0191.
Holzer K. (1993). The evolution of alpine Norway spruce during immigration into high altitudes
and its consequences. Norway spruce provenances and breeding. Proceedings of IUFRO
(S2.2-11) Symposium, Latvia, Riga. p. 68–78.
Johnsen, Ø; Skrøppa T. (2000). Provenances and families show dierent patterns of relationship
between bud set and frost hardiness in Picea abies. Canadian Journal of Forest Research
30(12): 1858–1866. https://doi.org/10.1139/x00-113.
Johnsen Ø., Kvaalen H., Yakovlev I., Dæhlen O.G., Fossdal C.G., Skrøppa T. (2009). An embryonic
memory from time of embryo development aects climatic adaptation in Norway spruce. In:
Gupta L., Wiesniewski M., Tanio K. (eds.). Plant cold hardiness: from the laboratory to the
eld. CAB International. p. 99–107. https://doi.org/10.1079/9781845935139.0099.
Krutzsch P. (1973). Norway spruce development of buds. International Union of Forest Research
Organizations, Vienna. Report IUFRO S2.02.11.
Krutzsch P. (1975). Die Pantzschulergebnisse eines Inventirerende Fichetenherkunftsversuches
(Picea abies Karst. and Picea obovata Ledeb.). Royal College of Forestry, Department of
18
Silva Fennica vol. 53 no. 1 article id 10076 · Skrøppa et al. · Genetic variation in phenology and growth among…
Forest Genetics. Research Notes 14/1975.
Kvaalen H., Søgaard G., Granhus A., Fløistad I.S., Hanssen K.H., Steenrem A., Skrøppa T. (2010).
Høstskudd og toppskader – et omfattende problem på god mark I lavlandet. Skogeieren 2010
(10): 18–19. [In Norwegian].
Kullmann L. (2008). Early postglacial appearance of tree species in northern Scandinavia: review
and perspective. Quaternary Science Reviews 27(27–28): 2467–2472. https://doi.org/10.1016/j.
quascirev.2008.09.004.
Langlet O. (1960). Mellaneuropeiska granprovenienser i svenskt skogsbruk. Kungl. Skogs- och
Lantbruksakademiens Tidskrift 1960: 259–329. [In Swedish].
Latalowa M, van der Knaap W.O. (2006). Late quaternary expansion of Norway spruce Picea
abies (L.) Karst. in Europe according to pollen data. Quaternary Science Reviews 25(21–22):
2780–2805. https://doi.org/10.1016/j.quascirev.2006.06.007.
Lynch M., Walsh B. (1998). Genetics and analysis of quantitative traits. Sinauer Associates, Inc.
980 p.
Oleksyn J., Modrzynski J., Tjoelker M.G., Zytkowiak R., Reich P.B., Karolewski P. (1998). Growth
and physiology of Picea abies populations from elevational transects: common garden evidence
of altitudinal ecotypes and cold adaptation. Functional Ecology 12(4): 573–590. https://doi.
org/10.1046/j.1365-2435.1998.00236.x.
Modrzynski J. (1995). Altitudinal adaptation of Norway spruce (Picea abies (L.) Karst.) progenies
indicates small role of introduced populations in the Karkonosze Mountains. Silvae Genetica
44(2–3): 70–75.
Parducci L., Jørgensen T., Tollefsrud M.M., Elverland E., Alm T., Fontana S.L., Bennett K.D.,
Haile J., Matetovici I., Suyama Y., Edwards M.E., Andersen K., Rasmussen M., Boessenkool
S., Coissac E., Brochmann C., Taberlet P., Houmark-Nielsen M., Krog Larsen N., Orlando
L., Gilbert M.T.P., Kjær K.H., Alsos I.G., Willerslev E. (2012a). Glacial survival of boreal
trees in northern Scandinavia. Science 335(6072): 1083–1085. https://doi.org/10.1126/sci-
ence.1216043.
Parducci L., Edwards M.E., Bennett K.D., Alm T., Elverland E., Tollefsrud M.M., Jørgensen T.,
Houmark-Nielsen M., Krog Larsen N., Kjær K.H., Fontana S.L., Alsos I.G., Willerslev E.
(2012b). Response to Comment on “Glacial survival of boreal trees in northern Scandinavia”.
Science 338: 742-b.
Persson B, Persson A. (1992). Survival, growth and quality of Norway spruce (Picea abies (L.)
Karst.) provenances at three Swedish sites of the IUFRO 1964/68 Picea abies provenance
experiment. Report 29. SLU, Sweden. 67 p.
SAS Institute (2003). SAS/STAT user’s guide, version 9. SAS Institute, Cary, N.C., USA.
Skrøppa T. (1982). Genetic variation in growth rhythm characteristics within and between natural
populations of Norway spruce. A preliminary report. Silva Fennica 16(2): 160–167. https://
doi.org/10.14214/sf.a15075.
Skrøppa T. (1991). Within population variation in autumn frost hardiness and its relationship to
bud-set and height growth in Picea abies. Scandinavian Journal of Forest Research 6(1–4):
353–363. https://doi.org/10.1080/02827589109382673.
Skrøppa T., Magnussen S. (1993). Provenance variation in shoot growth components of Norway
spruce. Silvae Genetica 42: 111–120.
Skrøppa T., Steenrem A. (2015). Selection in a provenance trial of Norway spruce (Picea abies L.
Karst.) produced a landrace with desirable properties. Scandinavian Journal of Forest Research
31(5): 439–449. https://doi.org/10.1080/02827581.2015.1081983.
Skrøppa T., Steenrem A. (2017). Høstskudd og toppskader i genetiske forsøk med gran; variasjon
og sammenhenger med vekst og vekstrytme. NIBIO Rapport 3/23/2017. 30 p. [In Norwegian
19
Silva Fennica vol. 53 no. 1 article id 10076 · Skrøppa et al. · Genetic variation in phenology and growth among…
with English abstract].
Skrøppa T., Hylen G., Dietrichson J. (1999). Relationships between wood density components and
juvenile height growth and growth rhythm traits for Norway spruce provenances and families.
Silvae Genetica 48(5): 235–239.
Skrøppa T., Kohmann J., Johnsen Ø., Steenrem A., Edvardsen Ø.M. (2007). Field performance
and early test results of ospring from two Norway spruce seed orchards containing clones
transferred to warmer climates. Canadian Journal of Forest Research 37(3): 515–522. https://
doi.org/10.1139/X06-253.
Solvin T.M., Steenrem A. (2019). Modelling the epigenetic response of increased temperature
during reproduction on Norway spruce phenology. Scandinavian Journal of Forest Research
34(2): 83–93. https://doi.org/10.1080/02827581.2018.1555278.
Soularue J.-P., Kremer A. (2012). Assortative mating and gene ow generate clinal phenological
variation in trees. BMC Evolutionary Biology 12: 179. https://doi.org/10.1186/1471-2148-
12-179.
Soularue J.-P., Kremer A. (2014). Evolutionary responses of tree phenology to the combined eects
of assortative mating, gene ow and divergent selection. Heredity 113: 485–494. https://doi.
org/10.1038/hdy.2014.51.
Squillace A.E. (1974). Average genetic correlations among ospring from open-pollinated forest
trees. Silvae Genetica 23: 149–156.
Tollefsrud M.M., Kissling R., Gugerli F., Johnsen Ø., Skrøppa T., Cheddadi R., Van der Knapp
O., Latalowa M., Terhürne-Berson R., Litt T., Geburek T., Brochmann C., Sperisen C. (2008).
Genetic consequences of glacial survival and postglacial colonization in Norway spruce: com-
bined analysis of mitochondrial DNA and fossil pollen. Molecular Ecology 17(18): 4134–4150.
https://doi.org/10.1111/j.1365-294X.2008.03893.x.
Total of 43 references.
... The differences in mean values that were observed for most traits of families from Diallel 1 and Diallels 2 and 3 confirm results from earlier common garden trials of Norway spruce provenances from different altitudes both in southern Norway (Dietrichson 1969, Skrøppa et al. 2007) and Mid-Norway (Skrøppa and Steffenrem 2019). Seedlings from natural populations from low altitudes are taller both in the nursery and in short-term tests in the lowland than those from populations originating from higher altitudes. ...
... They also have a later terminal bud set and cessation of shoot growth. Similar results, with clinal variation related to either the mean annual temperature or the altitude of the populations, were recently presented by Skrøppa and Steffenrem (2019) for Norway spruce populations in Mid-Norway. They found for most traits differences between populations separated by 200 m in altitude. ...
... The results from the diallel trials show similar patterns of variation for height at ages 7 and 10 and diameter at age 10 years as have been found in earlier studies (Dietrichson 1969;Ekberg et al. 1985;Skrøppa and Steffenrem 2019). The GCA variance is the dominating component in all three diallels, indicating that the largest part of the genetic variation for height Table 6 Pearson correlation coefficients between BLUP-values of the same trait for each parent in the diallel and in the OP families. ...
Article
Full-text available
The complete diallel cross is the only mating design that provides estimates of variance components of general combining (GCA), specific combining ability (SCA), maternal and reciprocal effects, in addition to heritabilities and genetic correlations. To obtain such estimates, complete diallels were made among 10 trees in each of three natural Norway spruce populations from altitude 300 and 500 m in southern Norway. Seedlings from families from the diallels and open pollinations were tested in short-term tests on agricultural soil at one site at altitude 85 m until age 10 years from seed. Tree height at ages 7 and 10 years and diameter at age 10 had strongly significant GCA variance components within each population. The components for SCA and maternal effects were small and not significant, indicating low levels of non-additive genetic variation. For the days of initiation and cessation of the shoot elongation period the GCA components were dominating and had the highest heritability estimates in two of the diallels. Estimates of genetic correlations between traits measured in earlier nursery trials and height and diameter in the short-term trials had low and not consistent values in the three diallels. The duration of the shoot growth period and rate of growth showed positive relationships with height and diameter. Strong relationships were present between half-sib family means from the diallels and open-pollinated families for height, diameter and phenology traits. Progeny trials testing open-pollinated half-sib families from natural populations can be used for selection of candidates for the initial breeding populations.
... They also have a later terminal bud set. Similar results, with clinal variation related to either the mean annual temperature or the altitude of the populations, were recently presented by Skrøppa and Steffenrem (2019) for Norway spruce populations in Mid-Norway. Differences between populations separated by 200 m in altitude, as we have here, were present for most traits. ...
... Dormling 1973, Krutzsch 1986, Hannerz and Westin 2000. The results from our study, from Skrøppa (1991) with a subset of families from Diallels 2 and 3, and from Skrøppa and Steffenrem (2019) show that there is considerable genetic variation for this trait also within natural populations and that the GCA variance component has the highest value. Significant GCA variance components were also found for this trait by Eriksson et al. (1978) for seedlings from different provenances grown in growth chambers. ...
Article
Full-text available
The complete diallel cross is the only mating design that provides estimates of variance components of general combining (GCA), specific combining ability (SCA), maternal and reciprocal effects, in addition to heritabilities and genetic correlations. To obtain such estimates, complete diallels were made among 10 trees in each of three natural Norway spruce populations from altitude 300 and 500 m in Norway. Nursery trials were performed with the families from these crosses and families from seeds collected from open pollination. Traits measured and analysed are seed weight, germination rate, germination percentage, terminal bud set, and seedling heights the first and second years. The seedlings from the population at origin 500 m had lower seed weight, lower heights and earlier bud set than those from the two populations from lower altitude. A considerable variation was present among families within each diallel, and the GCA variance components had the highest values and were significant for most traits. Variance components for SCA and maternal effects were also significant for some traits, but with inconsistent values in the three diallels. A strong relationship was present between the weight of the seed lots from the maternal parent and mean family height after one and two growing seasons. The highest estimate of heritability was observed for bud set, with similar values in all three diallels.
... The aboveground phenology and morphological traits of young Norway spruces were also investigated as they could provide indications of the initiation of an active growing period and features that may explain differences in growth rate (Skrøppa & Steffenrem, 2019). In Norway spruce, bud burst is known to show both phenotypic plasticity and genetic diversity (Salmela et al., 2020(Salmela et al., , 2024. ...
Article
Full-text available
Our study delved into the relationship between root‐associated fungi, gene expression and plant morphology in Norway spruce cuttings derived from both slow‐and fast‐growing trees. We found no clear link between the gene expression patterns of adventitious roots and the growth phenotype, suggesting no fundamental differences in the receptiveness to fungal symbionts between the phenotypes. Interestingly, saplings from slow‐growing parental trees exhibited a higher richness of ectomycorrhizal species and larger roots. Some ectomycorrhizal species, typically found on mature spruces, were more prevalent on saplings from slow‐growing spruces. The ericoid mycorrhizal fungus, Hyaloscypha hepaticola , showed a stronger association with saplings from fast‐growing spruces. Moreover, saplings from slow‐growing spruces had a greater number of Ascomycete taxa and free‐living saprotrophic fungi. Aboveground sapling stems displayed some phenotypic variation; saplings from fast‐growing phenotypes had longer branches but fewer whorls in their stems compared to those from the slow‐growing group. In conclusion, the observed root‐associated fungi and phenotypic characteristics in young Norway spruces may play a role in their long‐term growth rate. This suggests that the early interactions between spruces and fungi could potentially influence their growth trajectory.
... For shoot elongation in the field trials, this relationship was weaker. Such negative associations were reported from experiments in Mid-Norway by Skrøppa and Steffenrem (2019) and a similar the tendency was observed at the site in Northern Sweden reported by Helmersson and Westin (2019). However, it is contradictory to Skrøppa and Steffenrem (2016) who found a positive relationship between late bud flush in the short-term trial in the lowland of Southern Norway (also at Hogsmark) and height growth in a 17 year-old field trial in the same lowland region. ...
Article
Full-text available
Genetic variation and performance of up to 40-year-old Norway spruce ( Picea abies (L.) Karst.) families from half-diallel crosses made in natural populations where studied, and the results were compared with results from nursery test of seedlings (1-2 years from seed) and short-term farm-field tests at 6-10 years from seed. The diallel analyses revealed significant levels of additive genetic variance for growth and phenology traits. The non-additive genetic variance was generally small to moderate. Strong genetic correlations for growth performance were found between the short-term and long-term trials but not between the nursery and long-term trials. Similarly, genetic correlations for phenology traits were strong between short-term trials and long-term trials, whereas the nursery tests generally did not predict phenology at older age well. The findings indicate that early selection for growth and adaptive traits based on short-term trials is effective for improvement of long-term performance in field if the test site is not outside the normal range of deployment of the reproductive material.
... 0.3 ) when difference in latitude was used as the transfer variable. So, the predicted optimal transfers for northern harsh sites were almost as long as for the southern mild sites, which conflicts with practical experience and several previous studies of material in Sweden and Norway (Kroon and Rosvall, 2004;Persson and Persson, 1992;Skrøppa and Steffenrem 2019;Skrøppa and Steffenrem 2020). For example, Kroon and Rosvall (2004) found that local material provided outstanding performance at the harsh sites in northern Sweden, while for northern coastal sites with a slightly milder climate, the optimal latitudinal transfer for height was 4.3 . ...
Article
Full-text available
Norway spruce is a major industrial tree species in Fennoscandia and future productivity of the species must be secured by matching the variation in adaptation of the species with suitable sites for optimized performance. An appropriate transfer model for forest reproductive material (FRM) is crucial for regeneration of productive forests in the changing climatic conditions that are predicted to occur in Fennoscandia. We have developed a transfer model for prediction of height of Norway spruce in Norway, Sweden, and Finland, using data acquired from 438 progeny and provenance trials with 1919 genetic entries of local and transferred origins. Transfer of genetic material at a given site was expressed in terms of the difference in daylength (photoperiod) between the site and its origin. This variable best reflected the nonlinear response to transfer that has been commonly reported in previous studies. Apart from the transfer variable, the height prediction model included the age of material when height measurements were acquired, annual temperature sum over 5 °C, precipitation during the vegetation period, and interaction terms between test site and transfer variables. The results show that long northward transfers (4-5° latitude) seem to be optimal for relatively mild sites in southern parts of the countries where growing season is longer, and shorter northward transfers (2-4° latitude) for harsher northern sites with shorter growing seasons. The transfer model also predicts that southward transfers of Norway spruce would result in height growth reductions. The developed model provides foundations for development of common or national recommendations for genetically improving Norway spruce material in Fennoscandia.
... Traits of seasonal phenology are often both variable and heritable within populations (Li & Adams, 1993;Matziris, 1994;Skrøppa & Steffenrem, 2019), giving scope for evolutionary change. ...
Article
Full-text available
Plant phenology is not only manifested in the seasonal timing of vegetative and reproductive processes but also has ontogenetic aspects. The adaptive basis of seasonal phenology has been considered mainly in terms of climatic drivers. However, some biotic factors as likely evolutionary influences on plants’ phenology appear to have been under‐researched. Several specific cases of putative biotic factors driving plant phenology are outlined, involving both herbivores and pathogens. These illustrate the diversity of likely interactions rather than any systematic coverage or review. Emphasis is on woody perennials, in which phenology is often most multifaceted and complicated by the ontogenetic aspect. The complete seasonal leaf fall that characterizes deciduous plants may be a very important defense against some pathogens. Whether biotic influences drive acquisition or long‐term persistence of deciduousness is considered. In one case, of leaf rusts in poplars, countervailing influences of the rusts and climate suggest persistence. Often, however, biotic and environmental influences likely reinforce each other. The timing and duration of shoot flushing may in at least some cases contribute to defenses against herbivores, largely through brief periods of “predator satiation” when plant tissues have highest food value. Wide re‐examination of plant phenology, accommodating the roles of biotic factors and their interplays with environments as additional adaptive drivers, is advocated toward developing and applying hypotheses that are observationally or experimentally testable. It is postulated that not only biotic factors, especially pathogens, but also herbivores have played an under‐rated role in the evolution of plant seasonal ecology. A selection of case histories is reviewed, a classic case being deciduousness in the poplar rust pathosystem.
... In Lithuania, at an age of 35 years, a negative correlation between Dbh and knot diameter was obtained (Šilinskas et al. 2020), with a favorable impact on wood quality, which recommends the use of Dbh in selection. In Norway, because of the high correlation between Th and the number of damaged trees, it was also recommended to avoid selection by Th (Skrøppa & Steffenrem 2019). Also, our research leads to the recommendation of the Dbh trait for selection, after the stability traits (Ts, Cd, Cs). ...
Article
Full-text available
The aim of our study was to analyze the stability traits between 33 Norway spruce provenances tested in five field trials across different environmental conditions, in two major variants of the Romanian Carpathians: outside of the natural distribution range (ONR) and in the natural habitat (INR). To justify the early selection, we selected 40-year-old trees and measured tree height (Th), breast height diameter, pruning height, crown diameter, tree volume, tree slenderness (Ts), pruning height ratio, and crown slenderness, which were then compared on a time series with measurements from trees at 30 and 10 years old, respectively. All provenances reacted to the changes in the environmental conditions, presenting higher Th in the warmer ONR environments, compared with the results of the mountain INR trials, with negative consequences on the stand's stability. In all trials, highly significant differences resulted between, and especially within provenances, suggesting a high potential for adaptation in the future climate change scenario. An analysis of the stability traits suggests that we must avoid ONR afforestation with Norway spruce. All the elite provenances (Marginea, Gurghiu, Comandău, and Sudrigiu), together with Câmpeni and Turda, were highlighted, both for the stability and growth traits, whereas the local provenances and the standard IUFRO provenance were ranked below the average of the trials. The age-age significant correlations and the ranking of the provenances show that no major changes occurred in the last ten years, confirming the backward selection performed at the age of 30 years. The juvenile-mature correlations were also strong but the different evolutions in time of the elite provenances eliminate the possibility of a juvenile selection. The forward selection strategy, for the best trees belonging to the six mentioned provenances, according to Ts, can be applied in the INR trials.
Article
Full-text available
Establishing sustainable plantations with genetic diversity equivalent to that of natural populations is vital for successful reforestation efforts. In this study, we present an innovative approach for selecting populations suitable for reforestation, taking into account their genetic uniqueness using Inter Simple Sequence Repeats (ISSR) markers. Our investigation focused on six populations of Picea obovata Ledeb and six populations of Larix sibirica Ledeb, collected from the Northern and Middle Urals. We found that the indicators of genetic diversity were significantly higher in L. sibirica compared to P. obovata, while the number of rare alleles was greater in Siberian spruce (R = 19). Among the P. obovata populations, the Cherdyn's forestry exhibited notably high genetic diversity, and for L. sibirica, the Gainy's forestry stood out in this regard. Moreover, the genetic subdivision of the six P. obovata populations (G ST = 0.331) was higher than that of the six L. sibirica populations (G ST = 0.177). To ensure optimal seed selection considering the genetic originality coefficient (GOC) and population differentiation, we recommend utilizing the P. obovata population from Gainy's forestry with a GOC of 0.554 and the L. sibirica population from Cherdyn's forestry with a GOC of 0.372. These populations harbor typical alleles characteristic of the research region, making them ideal candidates for seed selection. Furthermore, the specific alleles identified can serve as valuable markers for determining the geographic origin of P. obovata and L. sibirica wood, aiding in efforts to trace the sources of these species in forestry and trade practices.
Book
Full-text available
The research objective was to improve the productive and adaptation potential of Norway spruce, the second largest species in Romania’s forests (depending on the proportions), with the purpose of obtaining stands with higher bioaccumulation potential, better wood quality and, especially, with a higher resistance to combined action of abiotic disturbing factors, wind and snow. In order to accomplish the general objective it was necessary to achieve the following specific objectives: 1. Reassessment of natural populations with narrow-crowned Norway spruce in composition and the inclusion of the selected ones in forest genetic resources category (FGR), to conserve them in situ. 2. Comparative analysis of narrow and normal crown form spruce growing and adaptation performances (even of hybrids between the two forms), wood resistance tests, as well as analyses of the environment conditions in which narrow crown transmission rate is high. 3. Comparative analysis of vegetation and flowering phenophases development, as well as of monthly radial increment, at the two forms of spruce, in a natural population. 4. Establishment of a working protocol for in vitro multiplication of narrow crowned Norway spruce. 5. Setting up of a clonal seed orchard for ex situ conservation of valuable narrow-crowned spruce populations and to produce seeds from this ideotype. Research material and methods: The research required both field work and laboratory analysis. The following investigations have been carried out: a) Reevaluation of all 25 populations chosen by Pârnuță (2008), selection, field measurement and materialization of narrow-crowned spruce trees, and collection of the necessary data to include the populations who meet the legal requirements in the forest genetic resources (FGR) category. b) Measurements carried out in the 5 full-sib comparative trials on the main phenotypic traits: tree’s height, last year hight increment, diameter at breast height, height from the ground to the first green branch, crown diameter, number of branches per whorl, growth and finesse of the branches. c) Measurements carried out in the 2 half-sib comparative trials, targeting the same traits described above. d) Performing wood resistance tests. e) Monthly radial increment monitoring, in Predeal population. f) Phonological observations in Predeal spruce natural population. g) Analysing of the existed narrow-crowned spruce seed orchard (Soveja). h) Vegetative multiplication (grafting) and establishment of a new pendula spruce seed orchard. i) In vitro micropropagation of narrow-crowned Norway spruce. Narrow-crowned spruce is composed by two forms/varieties, first being in biggest proportion in Romania (above 80%): - Picea abies (L.) Karst. f. pendula (Lawson) Sylven, with thin and slender first-order branches; - Picea abies var. columnaris (Jacq.) Carr. with short and horizontal branches. To constitute FGR of narrow-crowned spruce, sampling stage consisted in verifying in the field of 25 population selected by Pârnuță, in 1991, and those which have not been subjected to catting treatment methods (followed by natural or artificial regeneration), for which it was has been received the agreement of the administrator or owner, and which have been fulfilled the condition of existence of at least 15 narrow-crowned spruce trees, have been selected. Next, we proceeded to their description, measurement and field materialization of narrow-crown spruce trees. Finally, a record sheet of the FGR was drawn up and minutes were signed with the administrators of the respective forests. Comparative analysis of growing and adapting performances of normal and narrow-crown Norway spruce, and of diallel combinations (complete and factorial) between the two forms, it was carried out in 5 trials of biparental descendants (Comandău, Ilva Mică, Lepșa 1 & 2, and Voineasa), installed in 1996. These trials were constituted from seedlings obtained by seeds collected from three narrow-crowned spruce trees and three normal crown spruce tree, chosen in Stâna de Vale population (in 1990), in which it was realize controlled pollination between them, in a complete (P x P and N x N) and factorial (P x N and N x P) diallel system. In all 5 full-sib trials, the experimental design was a randomized blocks type, with 4 replications and 3-9 trees per unitary plot, and the planting scheme was 2.5 m x 2.5 m. in every trial were introduced 38-41 variants, including both open pollinated families and crossbreeding diallel between normal and pendula spruce trees, together with consanguineous (Budeanu et al. 2019b, 2020). Comparative analysis of the two forms of spruce was realized also in the two comparative trials of maternal descendance, Măneciu and Soveja, installed in the spring of 1994, in which are tested 24 pendula families and the same number of pyramidalis, originated from 8 provenances (3 families/ provenance, from each form). The experimental device, in both tests, is the a randomized blocks type with 4 replications and 4-12 trees/ unitary plot, planted at 2 m x 2 m. From 4 natural populations and 4 comparative trials increment cores were collected, and from a population and a trial logs were harvested for wood strength study, shoots for grafting, and also shoots and seeds for in vitro micropropagation. Comparative analysis for mechanic resistance of normal and narrow-crowned spruce wood it was realized both for standing trees (in Coşna population), and also using wood specimen collected from a mature population (Stâna de Vale), as well as from their descendants (juvenile stage) tested in Comandău trial. Testing the strength of standing trees (torsional strength) was realized using the Resistograph. In the laboratory were determined some mechanic properties of wood: elasticity module, as well as the resistance at bending, compression and shear). In the Predeal population, at the beginning of 2019, on the 10 from 14 pendula trees and also 10 pyramidalis neighbours, similar dimensional, distributed in diameter categories, in 60-86 cm interval, were installed permanent increasing belts, at 1.30 m height, which were read at an interval of 30 days in the vegetative rest, and also from 15 to 15 days in vegetation season. Using the methodology previously used by Pârnuță (2008), but also the ICP Forests manual (Beuker et al. 2016), phonological observations on the growing and flowering in the 2021 and 2022 vegetation seasons, focused on ten pendula spruce trees and ten pyramidalis neighbours, similar dimensional, chosen in Predeal population, were performed. Regularly were carried out drone flights in order to optimize as much as possible the data collected using the binoculars. To install a seed orchard, besides obtaining the grafted seedlings, another activities are necessary: site choosing, field and soil preparing, experimental device carried out and seed orchard installing. Vegetative multiplication by grafting (cloning) involve the development of following activities (Budeanu et al. 2021b): rootstock seedlings preparation, plus-tree selection and scions collection, development of the grafting action, and grafted seedlings care. Seedlings production by in vitro somatic micropropagation involve such specific activities: obtaining of the biologic material, identification of the optimum explant types and also the most efficient sterilization methods, determination of optimum hormonal balances to obtain high reactivity percentages to in vitro crop conditions, determination of adequate crop environment in multiplication and rooting and, in the end, establishing of an in vitro micropropagation protocol (multiplication, rooting, in vitro transfer). The results of the research can be summarized as follows: 1. About narrow-crowned spruce adaptability in natural stands and their selection for in situ conservation: a) Seven forest genetic resources have been selected, distributed in all divisions of Romanian Carpathians, in which narrow-crowned spruce trees (pendula form and columnaris variety) will be conserved in situ. b) The crown diameter of selected narrow-crowned spruce trees represent only a half of the value achieved by the pyramidal crown form trees (normal form), in the same stands, which ensure the highest resistance, especially due to less amount of snow retained in canopy. In the same time, pendula trees present, in generally, higher values for trees’ volume and lower for the number of branches in whorl ant branches thickness, an economically favourable results. 2. About the testing of the two spruce forms (and the hybrids between them), in half-sib and full-sib trials: a) The trials results were strongly influenced by the testing site, which requires an extreme caution in the movement of forest reproduction materials. b) Pendula spruce trees obtained superior results than normal spruce for the most important traits involved in selection (trees’ slenderness, survival rate, number of branches in whorl, branches thickness and crown diameter), at the same time for trees’ slenderness was obtained also the highest inheritance rate transmission (expressed by heritability and genetic gain), being the more appropriate trait for the next stage of selection. c) Consanguineous trees registered a reduced survival tare (42%) and a majority are dominated and they will disappear in following years. 3. About the wood resistance tests: a) Laboratory tests on the elasticity module and the resistance to bending, compression and shear, at the two spruce forms, highlighted the narrowcrowned spruce superiority (statistic very significantly), for samples collected in 140 years population (Stâna de Vale), while among its descendants (Comandău trial, 23 years) differences were statistically insignificant, also in favour of narrow-crowned spruce. b) Analyses performed on the standing tree indicated also the superiority (statistic significant) of narrow-crowned spruce trees, the mean amplitude indicated by Resistograph being 5,2% higher than the normal spruce crown form. c) Wood density highlighted also the superiority of pendula form and, more importantly, its superior adaptability (testing site influence was insignificant). A remarkable result was registered by pendula spruce from Stâna de Vale, in Măneciu test. d) About latewood, results from the two half-sib experiments were contradictory, the only constant is related to the superiority of Stâna de Vale pendula spruce provenance. 4. About monthly radial increment monitoring and phonological observations, performed in Predeal population: a) At pendula spruce trees from Predeal population, radial increments started with about 1 week delay and were bigger in the second half of the vegetation season, while the normal spruce registered bigger radial growths in the first half of the bioactive season. b) Radial increments were negatively and very significant influenced by the air temperature (especially in June and August) and positively, also very significant, by precipitations (June, July March, current year, and September-October, previous year), without existing a distinction between the two forms. c) Phenological analyses carried out in two consecutive years (2021 and 2022) on 10 narrow-crowned spruce trees and 10 neighbours that relate to the normal crown shape, respectively, showed significant differences from one year to another, regarding the period from the complete opening of the buds until the end of the growths, the start of vegetation season being registered at approximately the same date, in both years and for both spruce forms. d) Especially in 2021, from the moment of simultaneously reaching of stage 2 (buds starts to open), the growth phenophases were developed much faster at normal spruce, the slight tardiness of pendula spruce provide additional protection in front of late frosts. e) Maximum receptivity period of female strobilus is overlap at the two spruce forms existing possibility of inter-fecundity between them. 5. About the establishment of a narrow-crowned Norway spruce seed orchard: a) Promotion of narrow-crowned Norway spruce (Picea abies pendula form) in Romanian forests, and also the necessity of cloning and ex situ conservation of the last trees belongs to this ideotype, founded the objective of establishing a pendula spruce seed orchard. b) In B1 (Predeal), E2 and E3 (Apuseni mountains) provenance regions, also in Soveja seed orchard, 29 pendula spruce trees/ ramets were selected, from which grafts were collected, at the moment when rootstock seedlings began to start the vegetation season. c) Vegetative multiplication by grafting, using the method in side-veneer grafting, had a medium success percentage of 37% respectively, at the end have been obtained 661 grafted seedlings which are related at a total of 25 clones. In the seed orchards were introduced 506 grafted seedlings, while the rest is kept in reserve. The adopted plantation scheme was 4 m x 4 m. 6. Establishing of a work protocol for in vitro micropropagation of the narrow-crowned Norway spruce trees. a) For the initiation of in vitro cultures for narrow-crowned Norway spruce, the most suitable type of plant material was in vitro germinated seeds (on culture medium), followed by buds and nodal segments. The most effective sterilization method used 0.2% mercury chloride for 35 minutes and 70% ethanol for 1 minute, which significantly influenced the survival of the explants. The most suitable culture medium was the MS type, supplemented with BAP, in a concentration of 1 mg/l, and for a good survival of the explants it is necessary to add kinetin, 1 mg/l, while for obtaining high reactivity percentages, the addition of 2- IP 0.5 mg/l is required. High rooting percentages were obtained on MS medium supplemented with BAP (1 mg/l), to which 0.1 mg/l IBA was added. b) The research created the premises for the use of in vitro cultures for the ex situ conservation of forest genetic resources, the conservation of germinal plasma through tissue cultures being an alternative method of preserving genetic diversity in the forest tree species. Keywords: Carpathian mountains, climatic changes, forest genetics, forestry, ideotype trees, Norway spruce, spruce resilience, trees breeding strategy.
Preprint
Full-text available
Plant phenology is manifested in the seasonal timing of vegetative and reproductive processes, but also has ontogenetic aspects. The adaptive basis of seasonal phenology has been considered mainly in terms of climatic drivers. However, some biotic factors as likely evolutionary influences on plants’ phenology appear to have been under-researched. Several specific cases of putative biotic factors driving plant phenology are outlined, involving both herbivores and pathogens. These illustrate the diversity of likely interactions rather than any systematic coverage or review. Emphasis is on woody perennials, in which phenology is often most multi-faceted and complicated by the ontogenetic aspect. The complete seasonal leaf fall that characterises deciduous plants may be a very important defence against some pathogens. Whether biotic influences drive acquisition or long-term persistence of deciduousness is considered. In one case; of leaf rusts in poplars, countervailing influences of the rusts and climate suggest persistence. Often, however, biotic and environmental influences likely reinforce each other. The timing and duration of shoot flushing may in at least some cases contribute to defences against herbivores, largely through brief periods of ‘predator satiation’ when plant tissues have highest food value. Wide re-examination of plant phenology, accommodating the roles of biotic factors and their interplays with environments as additional adaptive drivers, is advocated, towards developing and applying hypotheses that are observationally or experimentally testable.
Article
Full-text available
Temperature during seed maturation can induce an epigenetic memory effect in growth phenology of Norway spruce (Picea abies (L.) Karst.) that lasts for several years. To quantify the epigenetic modifications induced by natural climatic variation, common garden experiments with plants originating from different provenances and seed years were performed. Plants from warmer seed years showed delayed phenology with later bud flush, bud set and growth cessation. This effect was quantified by linear models of phenology traits as a function of climate indices for the origin and seed year of the plants. Significant effects of the temperature during seed production (seed year) was found for the bud set in seedlings in their first growing season and for bud flush and growth cessation in the 7th-8th growing season from seed. The models suggest that growth start and growth cessation are delayed 0.7–1.8 days per 100 additional degree days experienced by the seed during embryo development and seed maturation. Models that include factors that are known to induce epigenetic effects could be used to better predict future performance of forest reproductive material.
Article
Full-text available
In the Nordic-Baltic region, there has been a growing concern about an increasing occurrence of multiple tops in young stands of Norway spruce. There is however a lack of documentation on the amount of such damages, and the causal agents involved. In two separate studies in SE Norway, we assessed the frequency of multiple tops in young sapling-sized stands, and studied the relationship between the occurrence of multiple tops and lammas growth the previous growing season on the sample trees. Study 1 included 44 planted and 10 naturally regenerated stands, while Study 2 included 68 planted stands with information on seed source. Among sample trees with multiple tops, 57% (Study 1) and 32% (Study 2) had signs of lammas growth the previous autumn, either in the form of an extended leading shoot or swollen bud. Site index as well as sample tree height were positively correlated to the occurrence of both lammas growth and multiple tops in both studies. In Study 1 we show that the probability of lammas growth was significantly higher in planted than in naturally regenerated stands. In Study 2 we show that it was higher in stands planted with seedlings grown from stand-origin seeds compared with improved seed material. Furthermore, the results of both studies show that lammas growth occurs most frequently among the dominant trees in the stand.
Article
Elongation of leading shoots were measured in Picea abies during the growth seasons of 1978 and 1980. Considerable genetic variation was found in shoot elongation parameters, both between different provenances and between full-sib families within the same stand. The central and eastern European provenances generally had a longer growth season than the Nordic ones, the main reason for their better growth. -from Author
Article
Measurements of heights at ages 2, 4 and 7 years and of flushing, early shoot elongation and lammas growth at age 7 years were made in nurseries and in a short-term trial with seedlings from provenances and open pollinated families of Picea abies. Fifteen of the provenances and 45 of the families were planted in a long-term field trial. At age 29 years from seed, basic wood density and its components: earlywood and latewood density and latewood percentage from increment cores were measured by X-ray analysis. The trees of the northern provenances, which were the shortest ones, had the earliest growth start and the lowest proportion of trees with lammas growth, and had also the highest wood density and proportion of latewood. Moderate to strong negative relationships were found between the wood density traits and early height growth and growth rhythm traits, suggesting that the latter ones could be good predictors of provenance differences in wood density. For the families, much weaker and in most cases non-significant phenotypic correlations were found between the two groups of traits. The genetic correlation coefficients were in most cases negative and had low absolute values. The possible causes of the differences in relationship patterns between provenances and families are discussed.
Article
Selections were made within East European Norway spruce provenances in a provenance trial in northern Sweden and the selected clones were grafted in a Norwegian seed orchard mixed with plus trees of local origin. Full-sib families were generated by controlled crosses, producing sets of both inter- and intraprovenance families, which were planted in two short-term and in long-term trials. Measurements were made of frost damage, tree height, annual shoot growth rhythm and lammas shoots, and genetic parameters were estimated. The intraprovenance families from selected trees of East European origin showed the least variation in all traits and had the lowest heritability values. The variation among the intraprovenance families of local origin was large and with the highest estimates of heritability. The means of the provenance hybrid families were intermediate between those of the intraprovenance families. Strong genetic correlations were present between the growth rhythm traits measured, and in the family group of local origin also between these traits and frost damage, as well as for lammas shoots and height. The progenies of the selected parents of East European origin seem to be a desirable land race with less damage, lower percentage of lammas shoots and less variability in growth capacity.
Article
The timing of bud burst (TBB) in temperate trees is a key adaptive trait, the expression of which is triggered by temperature gradients across the landscape. TBB is strongly correlated with flowering time and is therefore probably mediated by assortative mating. We derived theoretical predictions and realized numerical simulations of evolutionary changes in TBB in response to divergent selection and gene flow in a metapopulation. We showed that the combination of the environmental gradient of TBB and assortative mating creates contrasting genetic clines, depending on the direction of divergent selection. If divergent selection acts in the same direction as the environmental gradient (cogradient settings), genetic clines are established and inflated by assortative mating. Conversely, under divergent selection of the same strength but acting in the opposite direction (countergradient selection), genetic clines are slightly constrained. We explored the consequences of these dynamics for population maladaptation, by monitoring pollen swamping. Depending on the direction of divergent selection with respect to the environmental gradient, pollen filtering owing to assortative mating either facilitates or impedes adaptation in peripheral populations.Heredity advance online publication, 11 June 2014; doi:10.1038/hdy.2014.51.
Article
Norway spruce (Picea abies [L.] Karst.) is a broadly distributed European conifer tree whose history has been intensively studied by means of fossil records to infer the location of full-glacial refugia and the main routes of postglacial colonization. Here we use recently compiled fossil pollen data as a template to examine how past demographic events have influenced the species' modern genetic diversity. Variation was assessed in the mitochondrial nad1 gene containing two minisatellite regions. Among the 369 populations (4876 trees) assayed, 28 mitochondrial variants were identified. The patterns of population subdivision superimposed on interpolated fossil pollen distributions indicate that survival in separate refugia and postglacial colonization has led to significant structuring of genetic variation in the southern range of the species. The populations in the northern range, on the other hand, showed a shallow genetic structure consistent with the fossil pollen data, suggesting that the vast northern range was colonized from a single refugium. Although the genetic diversity decreased away from the putative refugia, there were large differences between different colonization routes. In the Alps, the diversity decreased over short distances, probably as a result of population bottlenecks caused by the presence of competing tree species. In northern Europe, the diversity was maintained across large areas, corroborating fossil pollen data in suggesting that colonization took place at high population densities. The genetic diversity increased north of the Carpathians, probably as a result of admixture of expanding populations from two separate refugia.