Forest Ecology and Management 214 (2005) 142–157
www.elsevier.com/locate/foreco
Broadleaved tree species in conifer-dominated forestry:
Regeneration and limitation of saplings in southern Sweden
Frank Götmark a,*, Jonas Fridman b, Göran Kempe b, Björn Norden c
b
a
Department of Zoology, Göteborg University, Box 463, SE-40530 Göteborg, Sweden
Department of Forest Resource Management and Geomatics, Swedish University of Agricultural Sciences, S-901 83 Umeå, Sweden
c
Botanical Institute, Göteborg University, Box 461, SE40530 Göteborg, Sweden
Received 14 October 2004; received in revised form 1 April 2005; accepted 4 April 2005
Abstract
Forests and forestry in Sweden are dominated by conifers. Silviculture using mixed or broadleaved stands is often
recommended, but the degree to which broadleaves regenerate naturally needs to be clarified. The Swedish National Forest
Inventory is here used for a region-wide study of broadleaved saplings (1.3 m tall to 4.9 cm dbh) regenerated naturally. For 12
species (taxa) in young forests (<7 m tall) and high forests (7 m), sapling densities were related to seven forest types and three
productivity classes. Birch had highest densities in all but two broadleaved forest types. Birch, oak, rowan and sallow had 70–
85% of their total sapling populations in conifer-dominated forest types, indicating good potential for mixed stands. Beech, lime,
hornbeam, ash and elm were mostly restricted to ‘noble’ (hardwood) forests. The regeneration success (saplings per mature tree)
for birch, rowan and oak was highest in conifer-dominated forest; beech was about equally successful in conifer-dominated and
broadleaved forests, and ash was very successful in broadleaved forest. Oak regeneration may be problematical in broadleaved
forests, but we suggest this is not the case in conifer-dominated forests (where oaks have rarely been studied). Sapling densities
of the species in the forest types were not consistently correlated with productivity, but birch and aspen generally regenerated
strongest at intermediate and at high productivity, respectively. In noble forests, oak, ash and elm regenerated strongest at low
productivity. The role of asexual regeneration (sprouting) remains to clarify. Our results suggest that lime, elm, ash and some
other trees currently are limited mainly by poor dispersal, rather than habitat availability. The results are promising for various
forms of mixed-species forestry that does not require planting (or little planting) and that would be beneficial for nature
conservation.
# 2005 Elsevier B.V. All rights reserved.
Keywords: Sustainable forestry; Betula; Populus; Quercus; Fagus; Sorbus; Salix; Fraxinus; Ulmus; Tilia; Acer
1. Introduction
* Corresponding author. Tel.: +46 31 7733650/883436;
fax: +46 31 416729; mobile: +46 31 2309315.
E-mail address: [email protected] (F. Götmark).
In many parts of Europe and elsewhere, forestry is
based mostly on even-aged conifer stands harvested
under short rotations by clearcutting (or removal of
0378-1127/$ – see front matter # 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.foreco.2005.04.001
F. Götmark et al. / Forest Ecology and Management 214 (2005) 142–157
almost all large trees). In repeated thinnings, most of
the broadleaved trees are cut or harvested. There are
disadvantages with such forestry, if it comes to
dominate as in Sweden. Even-aged conifer stands
generally have low value for wildlife (SEPA, 1994;
Gustafsson and Ahlén, 1996; Hunter, 1999), may be
susceptible to pests and global warming (Young and
Giese, 2003; Sykes and Prentice, 1996; Bradshaw
et al., 2000) and may create more acid soils (Nordborg
and Olsson, 1999). Norway spruce (Picea abies),
common in Europe, is also susceptible to wind-throw.
Forest owners have been encouraged to plant or to
favour native broadleaved trees for forestry and
biodiversity purposes (Persson, 1990; Gustafsson,
2000; Zerbe, 2002). In these contexts, the benefits of
mixed coniferous-broadleaved forests are often mentioned (Mosandl and Kleinert, 1998; Olsthoorn et al.,
1999; Thelin et al., 2002; Johansson, 2003).
In northern Europe, the most common broadleaved
trees are silver birch (Betula pendula) and hairy birch
(B. pubescens), forming 10% of the total wood volume
and 53% of the volume of broadleaves in southern
Sweden (see below). High seed production and rapid
growth in open clear-cut areas favour birches; however,
they are mostly eliminated in repeated thinnings
(Simard et al., 2004, and references therein). In
Sweden, usually only a few mature trees per hectare
(mainly birch, aspen Populus tremula, and/or Scots pine
Pinus sylvestris) are retained for biodiversity at
clearcutting. Given the goal of changing the forest
composition to resemble more natural conditions
(SOU, 1992), the broadleaved trees need more
attention. For both forestry and conservation work,
the factors limiting regeneration of broadleaved tree
species are of interest. Beside historical human use of
forests (Peterken, 1996; Lindbladh et al., 2000), tree
species may be limited by, e.g., seed production, seed
dispersal, and availability of habitat for growth (Clarke
et al., 1999). Most of the present south Swedish forest
consists of planted coniferous trees, whereas natural
forests in this region were dominated by broadleaves, or
mixed coniferous-broadleaved forests (Lindbladh et al.,
2000 and references therein). Conversion of especially
spruce forest to mixed or broadleaved stands is
desirable (SOU, 1992; Spiecker et al., 2004). Here
we analyse regeneration of 12 broadleaved tree species
that occur in or have colonized the forests of southern
Sweden. Moreover, to examine habitat limitation in
143
these species, we analyse regeneration in relation to site
productivity (site index).
Below, we first briefly present regeneration patterns
(sapling densities) of broadleaves in the forests. The
data are then used to answer the following three
questions: (1) do the 12 broadleaved tree species differ
in regeneration (i.e. production of small trees, saplings)
in the forest types; in particular, does regeneration differ
in coniferous versus broadleaved forests, indicating
conditions that limit some species? (2) Are some
species more successful in producing saplings (per
capita mature trees) in coniferous than in broadleaved
forest? (3) To what extent is regeneration in the species
related to, or limited by site productivity?
Broadleaved trees in conifer-dominated forestry
occur naturally as seedlings and saplings, while their
densities as larger trees are determined mainly by the
cutting regime, where thinning is important. In
Sweden, regeneration of broadleaves has not been
analysed at regional level, which is done here for
saplings using data from the Swedish National Forest
Inventory (NFI). We selected the southern part of
Sweden for study, where most of the broadleaved tree
species occur. Although seedlings of trees (<50 cm
tall) may be ‘persistent juveniles’ for long periods
(Grime et al., 1988; Tapper, 1992), saplings more
reliably indicate regeneration potential of the species,
as mortality is lower for saplings than for seedlings.
We studied saplings defined as stems from breast
height (>1.3 m) up to stem diameter of 4.9 cm, dbh
(referred to as 0.1–4.9 cm dbh). The majority of these
saplings had not been affected by thinning (the cohort
is dominated by small saplings about 0.1–2.5 cm dbh;
see also below), and so regeneration was mainly
determined by the natural processes of seed production, seed dispersal, germination, growth, or sprouting
from cut stems or roots. For larger saplings (diameter
interval 5.0–9.9 cm dbh in our data sets), densities
were much lower, mainly due to thinning. We
compared regeneration in all major forest types in
the landscape of southern Sweden; three coniferous
ones (strongly predominating in area), one mixed, and
three broadleaved forest types. We analysed regeneration in young forest and in higher forest where
saplings form advance regeneration that may be
important for the stand after cutting (Greene et al.,
1999; Nyland, 2002). We end by discussing factors
limiting the broadleaved tree species in relation to
144
F. Götmark et al. / Forest Ecology and Management 214 (2005) 142–157
their ecological traits, and by considering implications
for forestry and conservation.
2. Materials and methods
2.1. Study area and tree species selected
The study area of southern Sweden (Fig. 1) is a
transition zone between the boreal forest in central and
northern Scandinavia and the temperate (nemoral)
forest in continental Europe; nemoral forest also
occurs in southernmost Sweden (Fig. 1; Diekmann,
1994; SEPA, 1994; Nilsson, 1997). The northern limit
of our study area coincides with the approximate limit
of the distribution of oak (Fig. 1; the northern counties
Uppsala, Västmanland, Örebro, and Värmland except
the municipalities Torsby, Hagfors and Filipstad, are
included). North of this limit, species richness and
abundance of broadleaves is much lower than to the
south (Gustafsson and Ahlén, 1996; Nilsson, 1990).
The study area is lowland (less than 300 m above
sea level) where glaciation ended 15,000–10,000 years
ago, containing mostly moraine soils on granite or
gneiss (bedrock sometimes exposed, in a mosaic
landscape of forest, small fields, wetlands and lakes).
There are several larger lowland agricultural areas with
fertile soil. In July, the mean precipitation decreases
from about 100 mm in the west to about 55 mm in the
east, and the temperature varies from about 14–17 8C.
About 80% of the land area would be forested in the
absence of human influence (Gustafsson and Ahlén,
1996) and the present forest proportion is about 56%
(Forestry Statistics, 2000). Several thousand years ago,
oaks (Quercus spp.), small-leaved lime (Tilia cordata),
hazel (Corylus avellana) and other species probably
dominated the forests, but during the last thousand
years human influence is thought to have strongly
favoured conifers, especially so in the last 150 years
(Lindbladh et al., 2000). Presently, most stands are less
than 60 years old; about 10% of the stands are 100 years
or older (Forestry Statistics, 2000). Table 1 (see ‘Total’)
shows the current tree species composition by wood
volume in the study area.
In Swedish forest legislation, 13 broadleaved trees
are ‘noble’ (historical term) and important for wood
production and biodiversity (Berg et al., 1994;
Gustafsson and Ahlén, 1996; Löf, 2001). Four of
these trees have very limited distribution and were
excluded. We included the following nine noble trees:
pedunculate oak (Q. robur), sessile oak (Q. petraea),
beech (Fagus sylvatica), common ash (Fraxinus
Fig. 1. Map of the study area in southern Sweden, coinciding approximately with the distribution of noble broadleaved forest or tree species,
especially oaks. The northern counties Uppsala, Västmanland, Örebro, and Värmland except municipalities Torsby, Hagfors and Filipstad, were
included, forming the northern limit of the study area. The area includes two (temperate) vegetation zones in Sweden, the nemoral and
boreonemoral zone. Beech and hornbeam are largely restricted to the southern (nemoral) zone, to which we added the county of Kronoberg in
south central Sweden where these trees also occur. The largest easternmost island, Gotland, is not included in the study.
F. Götmark et al. / Forest Ecology and Management 214 (2005) 142–157
145
Table 1
Tree species by wood volume for forest land, as defined in text, in the study area (‘Total’) and for seven forest types used in the analyses (see
Table 3, and text), given as percentage based on volume for each forest type (total volume in last row)
Species
Norway spruce (Picea abies)
Scots pine (P. sylvestris)
Birches (B. pendula/pubescens)
Aspen (P. tremula)
Oaks (Quercus robur/petraea)a
Common alder (Alnus glutinosa)
Beech (Fagus sylvestris)a
Sallow (Salix caprea)
Common ash (Fraxinus excelsior)a
Rowan (Sorbus aucuparia)
Grey alder (Alnus incana)
Wych elm (Ulmus glabra)a
Lime (Tilia cordata)a
Norway maple (Acer platanoides)a
Hornbeam (C. betulus)a
Gean (Cherry) (P. avium)a
Other broadleaves
Sum
Total volume (million cubic m)
Forest type
Spruce
Pine
Pine/spruce
Birch
Noble
Other broadl.
Mixed
Total
81.12
10.37
5.51
1.15
0.58
0.52
0.17
0.18
0.03
0.16
0.12
0.00
0.02
0.01
0.00
0.01
0.03
100.00
609.6
15.72
78.05
4.84
0.34
0.49
0.21
0.06
0.06
0.01
0.12
0.01
0.00
0.00
0.03
0.00
0.01
0.03
100.00
349.5
42.41
37.78
14.86
1.95
1.03
1.13
0.08
0.39
0.01
0.24
0.08
0.00
0.00
0.02
0.00
0.00
0.02
100.00
121.8
14.86
7.56
67.24
2.02
2.00
2.74
0.08
1.65
0.54
0.73
0.08
0.02
0.02
0.06
0.02
0.02
0.42
100.00
50.4
3.10
2.49
6.17
2.63
33.06
2.77
36.02
0.63
5.68
0.84
0.02
1.91
1.47
0.77
1.19
0.28
0.96
100.00
42.9
9.83
4.17
21.26
20.65
4.67
28.06
0.42
3.27
1.29
1.82
1.69
0.77
0.15
0.42
0.04
0.13
1.33
100.00
52.1
32.19
16.44
29.34
8.82
4.40
4.00
1.07
1.15
0.42
0.81
0.79
0.03
0.03
0.10
0.10
0.03
0.26
100.00
38.2
49.62
31.26
10.08
2.12
2.04
1.90
1.38
0.40
0.30
0.29
0.17
0.10
0.07
0.07
0.05
0.02
0.14
100.00
1264.5
Data for trees above breast height, NFI 1998–2002.
a
‘Noble’ tree species as defined by Swedish forestry Act (see text). Table based on the whole study area in Fig. 1, also for beech and hornbeam.
excelsior), small-leaved lime (Tilia cordata), Norway
maple (Acer platanoides), hornbeam (Carpinus
betulus), wych elm (Ulmus glabra), and gean (wild
cherry) (Prunus avium). Beech and hornbeam are
mainly restricted to the southern part of the study area
(see Fig. 1, and Lindgren, 1970; Björkman, 1998) and
sapling densities given below are based on this part. Of
other broadleaved trees, we included silver and hairy
birch, aspen, sallow (Salix caprea) and rowan (Sorbus
aucuparia). Although Q. petraea is more common
than Q. robur in xeric-mesic forest, the two oaks
overlap considerably in range, stand types and traits,
and were pooled. B. pubescens is more common in
mesic-moist forest than B. pendula, otherwise the two
birches overlap and are similar, and were pooled. In
total, 12 species (taxa) were included, and studied in
forest producing at least 1 m3 per hectare and year
(Swedish definition of ‘forest land’). Xeric and wet
forests of lower productivity were excluded. We also
excluded broadleaves mostly occurring in forested
wetland (alder Alnus glutinosa and Salix spp.).
Ecological traits relevant to regeneration of the
species are summarized in Table 2. Asexual regeneration by sprouting may be important for seedlings/
saplings (Del Tredici, 2001), but has generally been
neglected (Bond and Midgley, 2001) and little is
known about less common species (Table 2). Classification of shade-tolerance tends to vary among
studies; we focused on saplings (young trees) and
Scandinavian conditions (Table 2).
2.2. The national forest inventory (NFI) and
analyses
The Swedish NFI was established in 1923 (Thorell
and Östlin, 1931). In 1983, a new design was
introduced (Ranneby et al., 1987) and has been used
since. The information recorded in the NFI is detailed
and comprises tree, stand, and site data (Lindroth,
1995; SLU, 1997). The inventory covers the whole of
Sweden with a stratified systematic cluster sample.
Each cluster consists of permanent or temporary
circular sample plots with radius 10 and 7 m,
respectively, located along the sides of a square or
rectangle. Certain measurements, such as forest type,
height and closure are made with plot radius of 20 m,
others such as sapling counts are made with radius 5
or 3.5 m (SLU, 1997). The permanent plots are
146
F. Götmark et al. / Forest Ecology and Management 214 (2005) 142–157
Table 2
Ecological traits relevant to regeneration of the broadleaved tree species (taxa) included in the present studya (for latin names, see Table 1)
Species
Shade-tolerance b
Seed
weight (mg)c
Seed production/
mature treed
Dispersal
agent
Sprouting
capacitye
Distribution
in Sweden
Birches
Aspen
Sallowg
Oaks
Rowang
Gean
Beech
Hornbeamg
Maple
Ash
Elm
Lime
Intolerant
Intolerant
Intolerant
Intolerant/mid tol.
Mid tolerant
Mid tol./tolerant
Very tolerant
Tolerant
Tolerant
Tolerant
Very tolerant
Tolerant
0.14
Low
0.09
3500
2.6
150
250
40
140
80
3.5
35
Very high
Very high
Very high
Low/moderate
High/moderate
Moderate
Low/moderate
Moderate
Moderate
Moderate
High
Moderate
Wind
Wind
Wind
Animals
Animals
Animals
Animals
Wind
Wind
Wind
Wind
Wind
Moderate
Very highf
Very high
High
Moderate/high?
High?
Moderate?
High
Moderate/high
High/very high
High
Very high
Country-wide
Country-wide
Country-wide
South-central
Country-wide
South-central
South
South
South-central
South-central
South-central
South-central
a
Based on Rinne et al. (1987), Grime et al. (1988), Kauppi et al., 1988, Mossberg (1992), Johansson (1992), Suszka et al. (1996), Karlsson
(2001; pers. comm.), Deiller et al. (2003), Fällman et al. (2003), and own studies (Götmark, pers. obs.) of closed canopy broadleaved forest, and
cutting in such forest (shade-tolerance, sprouting). The two species of birches and oaks were pooled (Table 1).
b
Refers mainly to shade tolerance of seedlings/saplings; this trait may change with age/height of trees (e.g. Diekmann, 1996).
c
For sallow, rowan and elm, weight of germinule (Grime et al., 1988).
d
Number of seeds per tree.
e
Refers to small individuals (seedlings/saplings/poles), sprouting capacity or importance generally declines with stem diameter (Del Tredici,
2001; Johnson et al., 2002).
f
Sprouting by root suckers.
g
Relatively small trees (occasionally high, 15 m or more).
inventoried in 5–10 year intervals. In general, a 5-year
period is needed to attain acceptable precision in the
estimates. We present data based on all available plots
in the study area during 1998–2002 (in total 15,082
plots). For calculation of volumes, forest areas, and
standard error of means from NFI, see formulas in
Fridman (2000) and Fridman and Walheim (2000).
We excluded forest land (plots) containing no or
very few saplings/trees (stand closure in NFI equal to
0). Such forest land, for instance recent clear-cuts, is not
included in estimates of forest areas. Data from the NFI
plots were divided into (1) young forest, on average less
than 7 m tall, and (2) high forest, on average 7 m or
more (usually not very ‘old’, thus referred to as ‘high’).
In plots (area within 20 m radius from center), if the
average tree height weighed by basal area is higher than
7 m, weighing of height by basal area is used, but if it is
lower, height is the arithmetic mean of the main trees
assumed to be used in a crop. The forest types used in
our analyses are defined by tree basal area (Table 3).
Three types were dominated by conifers (Norway
spruce, Scots pine, and mixed spruce/pine) with limit
for conifer stand dominance set relatively low (>60%)
to study regeneration of broadleaved trees in these
forests. For broadleaved forest types, forest areas were
smaller (Table 3). They were defined as conventional in
the NFI (at least 70% dominance, e.g. Lindroth, 1995).
‘Other broadleaved forest’ was dominated by alder,
birch and aspen (see Table 1, with resulting wood
volumes for defined forest types). For plot sample sizes
in the forest types, see Table 3.
We report mean sapling densities (stems, 0.1–
4.9 cm dbh, per hectare), but areas of the forest types
differed considerably and we also estimated total
number of saplings per forest type, using these data for
further analyses. Number of saplings (population size)
was calculated as sapling densities per ha times the
estimated area (ha) of each forest type. For
comparison and for analyses, mean densities of large
trees of the 12 species in high forest (7 m) were
estimated. Densities of saplings may be affected by
thinning, but in young forest only 23% of the stand
area had been thinned since stand establishment. In
high forest, 57% of the area had evidence of thinning
from after 1975. We cannot estimate the number or
proportion of cut stems, but thinned stands had about
equal or higher sapling densities than unthinned stands
(Götmark et al., unpublished data).
F. Götmark et al. / Forest Ecology and Management 214 (2005) 142–157
147
Table 3
Definition of forest types, their area (in parenthesis, %), and number of plots studied (included) in the NFI, 1998–2002
Type of forest
Coniferous forest
Norway spruce
Young forest (<7 m)
High forest (>7 m)
Pine forest
Young
High
Mixed spruce/pine forest
Young
High
Broadleaved forest
Birch forest
Young
High
‘Noble’ broadleaved foresta
Young
High
Other broadleaved forest
Young
High
Mixed forest
Coniferous/broadleaved
Young
High
Total forest, study area
Composition, basal area in %
Area in 1000 ha (%)
At least 60% spruce
3168 (43)
796
2372
2244 (31)
417
1826
690 (9.5)
123
567
6649
1682
4967
4417
813
3604
1351
238
1113
438 (6.0)
89
349
208 (2.9)
17
191
271 (3.7)
25
246
952
189
763
538
39
499
633
55
578
265 (3.6)
36
229
7284 (100)
542
79
463
15082
At least 60% pine
At least 60% spruce and pine
At least 70% birch
At least 70% broadleaved trees and at least 50% ‘noble’ trees
At least 70% of other broadleaves
Mixed; remaining forest, after other forest types selected
No. of plots
a
For ‘noble’ trees: at least 50% of the total volume must consist of oaks, beech, ash, lime, maple, hornbeam, elm or gean/wild cherry
(generally accepted definition).
For each tree species, we calculated per capita
regeneration success, which is the ratio between
density of saplings and density of large (mature) trees.
For high forest, we compared two groups: the four
coniferous/mixed forest types (pooled), and the three
broadleaved types (pooled), calculating means that
were weighed by the areal extent of forest types
included within each group. Our success measure
indicates the relative (approximate) production of
saplings per mature tree in the species. In the first
analysis, mature (reproductive) trees were defined as
stems 15 cm dbh; for sallow and rowan, smaller
trees starting to reproduce earlier, we set the limit
lower (10 cm dbh). In the second analysis, we
included stems 20 cm for all species except sallow
and rowan [we could not find useful data for size at
maturity, but Loehle (2000) reported that the typical
age at sexual maturity in small trees (mean height
11 m) was 14 years, whereas medium-sized trees
(height 21 m) and tall trees (30 m) did not differ much
in this respect (typical reproductive age 33 and 35
years, respectively)]. We did not calculate per capita
regeneration success for trees in young forest.
Although there are scattered larger trees in young
forest (retained for conservation or shelterwood), most
seed trees would be at the edge of higher forest nearby
(Karlsson, 2001) and cannot be estimated from the
NFI.
Site productivity was based on the site index for
each study plot, determined from curves relating tree
age to tree height (based on conifers), and/or from the
type of lower forest vegetation (SLU, 1997). In the
NFI, the site index is translated into a productivity
measure (m3 tree volume per ha and year). We studied
regeneration of broadleaved trees in three classes
(intervals) of productivity, as follows: (1) low: 1–
5.9 m3 per ha and year, (2) intermediate: 6.0–8.9 m3,
and (3) high productivity: >9.0 m3.
Since the NFI is based on sampling with partial
replacement, the estimators are complex (see Fridman
148
F. Götmark et al. / Forest Ecology and Management 214 (2005) 142–157
P < 0.05 for both spruce and in pine forest), but birch
predominated among the saplings also in high forest.
Some species, such as aspen and rowan, regenerated in
most forest types, while saplings of other species
mostly were restricted to one or a few forest types
(Table 4). In young forest, saplings of oak were more
common than those of beech (P < 0.05); in high forest,
densities of these two species were more similar, but in
noble forest beech had higher sapling densities than
oak, ash and rowan (P < 0.05 in all cases, data from
southern part of study area used, Fig. 1).
and Walheim, 2000 for explanations and formulas).
Although systematic sampling is used in the NFI, for
this study standard error (S.E.) was estimated
assuming random sampling of tracts, which is
conservative as it might overestimate the true S.E.
Confidence intervals (95%) for means were based on
S.E., and significant differences reported below
(P < 0.05) based on these confidence intervals.
3. Results
3.2. Sapling populations in coniferous and
broadleaved forests
3.1. Sapling densities
Table 4 shows estimates of sapling densities for the
12 species (taxa) in each forest type. Birch generally
had highest sapling densities, but in noble and in other
broadleaved forests other species dominated, and had
relatively high sapling densities. The density of birch
saplings was higher in young than in high forest (e.g.
Total sapling populations were estimated from
Tables 3 and 4 (area densities). Birch formed 81%
(about 3.065 109 saplings) of all saplings in young
forest and 66% (about 3.107 109) of all saplings in
high forest. Birch was the major broadleaf element in
Table 4
Mean sapling densities (stems, >1.3 tall to 4.9 cm dbh, per hectare)a for 12 taxa in seven forest types, defined in Table 3 (data from NFI 1998–
2002, for number of plots see Table 3)
Species/taxab
Forest height (m)
Spruce
Pine
Spruce-pine
Mixed
Birch
Noble
Other broadleaved
Birches
<7
7
<7
7
<7
7
<7
7
<7
7
<7
7
<7
7
<7
7
<7
7
<7
7
<7
7
<7
7
1896
358
197
31
32
32
15
23
0
0
15
4.0
0
0.9
1.3
5.1
8.2
3.9
0.8
1.8
28
7.8
107
54
1426
618
23
30
43
60
0
8.8
0
0.2
0
4.0
0
0.5
0
0.8
30
1.3
0
1.6
21
11
27
57
2582
640
45
44
36
40
0
71
0
0
0
0.7
0
0
0
5.4
0
7.7
0
1.2
23
28
71
60
4077
776
306
117
462
85
0
56
0
2.3
0
5.5
0
0
0
2.0
0
16.2
3.9
0
34
34
235
208
5123
1097
381
103
29
56
0
16
0
1.2
163
91
0
0.4
0
8.3
0
44
0
17
54
31
128
157
363
81
199
87
1600
146
137
364
294
37
784
234
24
38
80
33
201
95
0
23
21
3.3
71
162
1426
305
3698
452
352
136
0
66
0
8.5
214
137
0
0
193
42
4.0
0.9
6.2
14
167
50
893
274
Aspen
Oaks
Beech c
Elm
Ash
Lime
Maple
Hornbeamc
Gean/Cherry
Sallow
Rowan
a
b
c
Zero does not necessarily mean that there were no trees (rounded numbers).
For full name, see Table 1.
Based on the southern part of the study area, where these two species occurred (see Fig. 1).
F. Götmark et al. / Forest Ecology and Management 214 (2005) 142–157
149
Fig. 2. The proportion (%) of saplings (total populations, 0.1–4.9 cm dbh) of broadleaved trees occurring in coniferous and mixed forest types.
The other saplings occurred in three forest types dominated by broadleaved trees (Table 3). Population sizes were estimated on the basis of
densities and forest areas (Tables 3 and 4).
five of the seven forest types (Table 4). High forest
contained more than twice as many saplings as young
forest (67.8% of all saplings), due to its large area. The
broadleaved tree species differed considerably in the
extent to which their sapling population occurred in or
had colonized coniferous and mixed forest (Fig. 2). In
high forest, birch, oak, sallow and rowan had highest
population proportion in coniferous and mixed forest,
whereas lime, hornbeam, ash and elm had lowest
population proportion there. In young forest, the result
was generally similar but gean, beech and hornbeam
were more common in conifer and mixed forest
(Fig. 2). In young forest, several species had small
populations and the estimates may partly be influenced by chance, but for lime, ash and elm 70–100%
of the saplings grew in broadleaved forest (Fig. 2).
3.3. Regeneration success of the tree species in
coniferous and broadleaved forests
Table 5 shows estimates of densities of large trees,
used to calculate per capita regeneration success in high
forest. Fig. 3a (mature trees >15 cm dbh, for rowan and
sallow >10 cm) shows that in coniferous forests, rowan
had the highest per capita regeneration success,
followed by oak and birch that had similar value, and
then aspen, beech and sallow in this order. In
broadleaved forests, ash was outstanding with a value
of 66, followed by maple and rowan; birch and sallow
had weakest per capita regeneration success there.
Birch was especially successful in coniferous
forests, compared to in broadleaved forests
(Fig. 3a); oak and sallow had about twice as high
per capita regeneration success in coniferous than in
broadleaved forests, whereas aspen and beech had
relatively similar regeneration success in coniferous
and broadleaved forests.
Fig. 3b (mature trees >20 cm dbh) shows largely
the same picture, but regeneration success of birch,
aspen and oak became higher in coniferous forests,
and the values for beech were similar in coniferous and
broadleaved forests. Again, regeneration of ash was
outstanding in broadleaved forests (Fig. 3b). Sallow
and rowan were not included in Fig. 3b as they mature
as small trees, and the data for gean was too uncertain
to include in Fig. 3b.
150
F. Götmark et al. / Forest Ecology and Management 214 (2005) 142–157
Table 5
Mean density of large mature trees (stems, >15 cm dbh, per hectare)a for 12 species (taxa) in high forest (7 m), of seven types defined in
Table 3 (NFI, 1998–2002)
Species/taxab
Spruce
Pine
Spruce-pine
Mixed
Birch
Noble
Other broadleaved
Birches
Aspen
Oaks
Beech c
Elm
Ash
Lime
Maple
Hornbeamc
Gean/Cherry
Sallowd
Rowand
26.5
4.3
1.9
2.0
0
0.1
0
0
0
0
0.8 (2.0)
0.6 (1.7)
17.7
1.0
1.9
1.6
0
0
0
0.1
0
0
0.2 (0.8)
0.1 (0.9)
67.9
7.6
4.2
2.1
0
0
0
0
0
0
1.7 (3.7)
0.9 (2.6)
102
30.9
12.7
19.1
0.2
1.1
0.2
0.3
0.4
0.2
2.8 (7.7)
2.7 (6.1)
207
5.0
3.4
1.0
0.1
0.8
0
0.2
0
0
3.7 (9.0)
1.2 (3.9)
24.8
7.5
107
123
5.7
20.6
3.2
3.6
6.3
1.0
2.9 (3.7)
3.7 (7.9)
93.7
79.0
15.3
5.9
1.0
3.3
0.5
1.5
0.2
0.9
17.1 (34.2)
5.4 (18.0)
a
b
c
d
Zero does not necessarily means there were no trees (rounded numbers).
For full name, see Table 1.
Based on the area where these species mainly occurred (Fig. 1).
In parenthesis, estimates based on stems >10 cm dbh, used in analyses (see text).
3.4. Regeneration of the tree species in relation to
productivity
For the forest types studied, we found complex
relationships between sapling densities of the species
and forest productivity (see Figs. 4–6, with significance of differences). For young forest, only data for
coniferous forest types were meaningful to analyse,
whereas for high forest six forest types were analysed.
For birch, sapling densities peaked at intermediate
productivity in six of nine comparisons, and in eight of
nine comparisons densities decreased from intermediate to high productivity conditions (Figs. 4-6). For
aspen, densities increased with increasing productivity
or peaked at high productivity in six of nine forest type
comparisons, and similarly for rowan in five of nine
comparisons. Sallow and especially oak sapling
densities showed highly variable relationships to
productivity. Surprisingly, in noble forest, densities
of oak, elm and ash saplings decreased with increasing
productivity (Fig. 6a). For beech, sapling density was
highest at high productivity in noble forest (Fig. 6a),
though this was not the case in the two other
broadleaved forest types.
At high productivity, seedlings and saplings might
grow at higher density, but thinning of the forest may
be more common, explaining, for instance, the pattern
observed for birch. We re-analysed the data using only
unthinned forest (plots) and found almost identical
patterns as those presented in Figs. 4–6 (data not
shown), so thinning did not influence these results.
Thinning generally had weak influence on sapling
densities in 1998–2002 (Götmark et al., unpublished
data).
4. Discussion
4.1. Regeneration of broadleaves: the role of
dispersal and persistence
Four species (birch, oak, sallow and rowan) had a
large proportion of their total sapling population in
coniferous and mixed forests, for oak almost 75%.
These four species represent two different modes of
seed dispersal; production of many small seeds easily
dispersed by wind (birch, sallow) or production of
fewer seeds dispersed by animals (acorns or in rowan
seeds in fleshy fruits). If seed dispersal is important for
the present occurrence of these trees in southern
Sweden, both strategies seem to be successful (cf.
Willson, 1993). Birch, sallow, oak, and rowan are also
favoured by the initially open conditions on clear-cuts
(cf. Table 2). The major dispersal agent of acorns in
Sweden and other parts of Europe, the jay Garrulus
glandarius (Bossema, 1979), usually nests in high
densities in conifer-dominated forests, which offer
good protection for their nests. Jays hoard acorns
F. Götmark et al. / Forest Ecology and Management 214 (2005) 142–157
Fig. 3. Per capita regeneration success in high forest (7 m) for ten
species (taxa) of broadleaves, measured as the ratio between density
of saplings and density of mature trees, either with mature trees (a)
larger than 15 cm dbh, or (b) larger than 20 cm dbh. For sallow and
rowan, mature trees larger than 10 cm dbh were used. For each
species, the measure was based on mean value for coniferous/mixed
forest (four types, means weighed by forest area of type) and mean
value for broadleaved forest (three types, means weighed by forest
area of type). Data from NFI 1998–2002.
hundred of metres or kilometres from oak trees,
preferring open conditions or forest edges where some
of the hoarded acorns germinate (Bossema, 1979).
Clear-cutting combined with jays may be a major
reason for oak colonization of coniferous forest. In
151
Fig. 4. Mean stem densities (saplings/ha, 0.1–4.9 cm dbh) for five
species (taxa) of broadleaves in young forest (<7 m) in relation to
site productivity, in (a) spruce forest, (b) pine forest, and (c)
coniferous (spruce/pine) forest. In comparisons, different letters
above bars indicate significant differences (P < 0.05), same letters
indicate non-significant differences (P > 0.05). The productivity
measure is based on the site index, and expressed as wood volume
increment per year in three classes; low productivity (1–5.9 m3 per
ha and year), intermediate productivity (6.0–8.9 m3), and high
productivity (>9.0 m3). For birch, mean sapling densities are given
above bars in two cases. Data from NFI, 1998–2002.
addition, Frost (1997) (see also Nilsson, 1985) found
higher survival of oak seedlings in coniferous than in
broadleaved forest, which is consistent with our
finding of higher per capita regeneration success in
152
F. Götmark et al. / Forest Ecology and Management 214 (2005) 142–157
Fig. 5. Mean stem densities (saplings/ha, 0.1–4.9 cm dbh) of six
species (taxa) of broadleaves in high forest (>7 m) in relation to site
productivity, in (a) spruce forest, (b) pine forest, and (c) coniferous
(spruce/pine) forest. For birch, mean sapling densities are given
above bars in five cases. For further explanation, see Fig. 4.
coniferous forest. Mosandl and Kleinert (1998) stated
that jays were the major reason for successful
regeneration of oak in pine forests in Germany, where
hoarding by jays apparently takes place also under
pine canopy cover. A comparison of our three coniferdominated forest types shows that average oak
densities were highest in pine forests, because jays
often hoard there, because the availability of light in
these stands is relatively good, and/or because oaks
compete well in the soil of pine stands.
The fruits of rowan are often eaten by migrating
birds in the autumn, mainly thrushes and waxwings.
The seeds survive passage through the birds’ digestive
system, leading to effective long-distance seed
dispersal. Many migrating birds roost or seek cover
Fig. 6. Mean stem densities (saplings/ha, 0.1–4.9 cm dbh) of 10
species (taxa) of broadleaves in high forest (>7 m) in relation to site
productivity, in (a) noble forest, (b) other broadleaved forest, and (c)
birch forest. For birch, sapling densities are given above bars for low
and intermediate productivity. For further explanation, see Fig. 4.
in coniferous forest that covers a much larger area than
broadleaved forest.
Beech colonized southern Sweden late (about 3000
years ago) and seems slow in its range expansion
(Björkman, 1998). Beech apparently lacks efficient
dispersal agents, though Nilsson (1985) reported that
in years with few acorns and many beech nuts, jays
dispersed nuts, which may explain northward migration. In high noble forest in the south, sapling density
of beech was high, and shade-tolerance probably
contributes to its success there. However, to some
extent beech is colonizing spruce and coniferous
forest, in agreement with Nilsson’s (1985) indication
of high sapling survival in such forest, and occasional
dispersal by jays.
Most of the trees with very low proportion of their
saplings in coniferous forest are shade-tolerant and
wind-dispersed (lime, hornbeam, ash, elm, maple).
F. Götmark et al. / Forest Ecology and Management 214 (2005) 142–157
Compared to birch and sallow, seed production per
tree is lower and the seed and its appendage is heavier/
larger in these species (Table 2), probably making
long-distance wind dispersal more difficult. Therefore, beside historical factors (Lindbladh et al., 2000),
dispersal limitation is one plausible or contributing
reason why these species currently are uncommon as
saplings in coniferous forest. The rates of seed fall in
still air for aspen, maple, hornbeam and ash (terminal
velocity) were 11, 107, 120 and 200 cm/s, respectively
(Levin and Kerster, 1974), which is consistent with
dispersal limitation and our findings. In southern
Sweden, birch seeds were more readily dispersed into
a clear-cut than seeds of elm and maple, which
dispersed better than hornbeam seeds (but elm and
birch did not differ much in dispersal distance;
Karlsson, 2001). For long-distance dispersal (rare)
updrafts are critical (Greene and Johnsson, 1995).
Southern Sweden is dominated by forest, lakes and
mires that may impede wind dispersal of heavy seeds.
An alternative or contributing factor explaining
sapling regeneration is sprouting, the main form of
regeneration in aspen, where sexual reproduction
currently is rare. Limited knowledge about this
‘persistence niche’ (Bond and Midgley, 2001) for
several species (Table 2) makes it difficult to evaluate
its role. Sprouting in (cut) saplings is more common
and important than sprouting in (cut) larger trees
(review in Del Tredici, 2001). Precommercial and
commercial thinning imply cutting of sapling and pole
trees of broadleaves, but the degree to which plants die
or become seedling sprouts has mainly been studied in
birch and aspen in Scandinavia. In hardwood forest in
US (three studies), 58–78% of ‘‘seedlings’’ were
sprouts that may survive for long (Del Tredici, 2001).
To judge from closely related American broadleaves,
saplings of oak in Europe should have higher
sprouting capacity than saplings of birch and maple
in Europe (Del Tredici, 2001, pp. 131–134; Johnson
et al., 2002; Table 2). If true, oak regeneration is
produced, or maintained also in this way.
The trees with low proportion of saplings in
coniferous forest, such as lime, ash and elm, sprout
readily when pruned or cut (Table 2). Historically,
sprouting probably facilitated their survival under
pressure from human exploitation; pollarding for
forage was common, so not all mature trees of these
species were eliminated (SEPA, 1994; Nilsson, 1997).
153
4.2. Regeneration success of broadleaved trees
For high forest, the per capita regeneration success
was higher in coniferous than in broadleaved forests
for birch, oak, rowan, sallow and to some extent aspen.
Between 1985 and 2000 in coniferous forests, birch
sapling densities remained approximately constant
while sapling densities of both oak and beech declined
about 50%, probably mainly due to deer browsing
(Götmark et al., unpublished data). If data from 1985
had been used, per capita regeneration success for oak
would have been much higher than for birch in
coniferous forest, and beech would also have had
higher success in coniferous forests.
For birch, oak, rowan, sallow and aspen there are
several alternative explanations for higher regeneration success in coniferous than in broadleaved forests.
First, light availability may be lower in high broadleaved forests than in high coniferous forests. As the
four species are mainly shade-intolerant, their
regeneration may be poor in darker broadleaved
forests. Most high conifer stands have been thinned in
forest management, which may keep stands relatively
open, and allow regeneration of these species. But the
broadleaved forests contained much birch and aspen,
and were probably relatively open (and also partially
cut, for e.g. fuel-wood). Noble forest is likely to be
darker, because there is presently little management
activities in these stands (Hamilton and Mirton, 1998).
Light availability may play a role, but is probably not a
major factor.
Second, in contrast to the broadleaves, Norway
spruce and Scots pine die after cutting, and lower
competition in conifer stands may facilitate regeneration of birch, oak, rowan and aspen there after
thinning. In addition, clear-cutting is less common in
broadleaved stands, which means less ground disturbance in the stands over time, and probably reduced
regeneration of birch and rowan. For oak, it would be
interesting to study whether jays prefer hoarding
acorns on ground disturbed during clear-cutting.
Aspen had somewhat higher per capita regeneration
success in coniferous forests, possibly due to lower
competition there.
The shade-tolerant maple and especially ash had
high per capita regeneration success in broadleaved
forest, but unfortunately these trees were too rare in
coniferous forest to estimate regeneration success
154
F. Götmark et al. / Forest Ecology and Management 214 (2005) 142–157
there. Ash has recently been reported to be successful
in Europe (Tapper, 1996; Harmer et al., 1997; Marigo
et al., 2000; Kerr and Cahalan, 2004). In lime,
regeneration through seeds is weak compared with for
instance ash and maple (pers. obs.). In addition to poor
dispersal, lime may be limited by factors controlling
sexual reproduction, as in north-western England
(Pigott and Huntley, 1981). In elm, with good seed
dispersal in Karlsson’s (2001) study, mortality through
Dutch elm disease might influence sapling densities.
4.3. The influence of habitat productivity on
regeneration
In studies of regeneration of trees in relation to
productivity, one potential problem is non-random
selection of study stands; in other words, the full range
of productivity conditions for a species may not be
represented. The NFI offers an advantage in its
systematic sampling that covers whole regions. For the
forest types studied, we found that birch sapling
densities were reduced at high productivity levels,
which may be due to increased competition from
herbs, shrubs and other trees (Harvey and Bergeron,
1989; review in Greene et al., 1999). A similar
relationship was found for essentially all forest types,
suggesting that under the current conditions birch
regenerate best in soils of intermediate productivity.
At high productivity, herbs, grasses and dwarf-shrubs
may suppress germination and growth of birch seeds
that have small nutrient reserves. Thinning did not
influence, or bias our results (see above).
Interestingly, the tree species that most strongly
rely on regeneration by sprouting (aspen) showed the
strongest positive association between productivity
and regeneration. The root suckers of aspen are tied
physiologically to the root system of the parent tree, at
least initially. Possibly, during such linked conditions,
regeneration and productivity will usually be positively correlated, under the assumption that the shoots
(sprouts) are less influenced by other vegetation
surrounding them than are shoots growing from seeds.
In high noble forest, sapling densities were
generally highest at low and intermediate productivity,
except for beech. For ash, we found a negative
correlation between sapling density and productivity.
Undergrowth of competing shrubs such as hazel is
common in noble forests of high productivity, and may
reduce tree sapling densities there. Alternatively,
productive noble forest have more closed canopies and
subcanopies (due to, e.g., limited cutting) leading to
lower sapling recruitment. However, ash seedlings and
saplings in southern Sweden are shade-tolerant (pers.
obs.). Our results indicate that regeneration of ash,
elm, maple and oak can be relatively high on forest
land of low and intermediate productivity, but further
studies of stem distributions and growth rates in noble
forests are of interest. Ash was considered ‘‘very
demanding with respect to soil moisture, pH and
nitrogene’’ in southern Sweden by Diekmann (1996).
The variables in Diekmann’s study should reflect
productivity, though his conclusion for ash might be
influenced by the particular stands selected for that
study.
In the study area, most trees regenerate under a
wide range of conditions. Scots pines are common on
xeric forest land, but also on moist-wet mires. Alders
are common in swamp forest, but also on mesic soils
(Table 1). For beech, there are four stand types,
differing markedly in soil pH and nutrient level
(Lindgren, 1970). We found regeneration of the tree
species under a wide range of nutrient (productivity)
levels. There are probably several reasons for these
broad habitat distributions. We suggest that the
extinction of the great majority of the European tree
species during the last ice age (Latham and Ricklefs,
1993) led to broader habitat distributions of remaining
species. Generalist species might have survived the ice
age better than specialists, contributing to the present
wide distributions.
4.4. Implications for forestry and conservation
Naturally grown saplings of broadleaves in coniferous forests should be re-considered, as resources
for future forests. Under certain conditions, forestry
based on mixed species stands is productive (Mosandl
and Kleinert, 1998; Olsthoorn et al., 1999; Thelin
et al., 2002; Johansson, 2003; Simard et al., 2004 and
references therein). Since conifer and hardwood trees
partly have different species communities, mixed
forests are favourable also for biodiversity (e.g.
Gustafsson and Ahlén, 1996).
Our results suggest that birch, aspen, oak and rowan
would be easy to favour in forestry by natural
regeneration from present conditions. In high forest,
F. Götmark et al. / Forest Ecology and Management 214 (2005) 142–157
careful thinning or partial cutting is required, as
advance regeneration may be damaged or die from
exposure (discussion and references in Laiho et al.,
1994; Greene et al., 1999; Nyland, 2002, pp. 283–
285). One problem in southern Sweden is browsing of
oak, beech and some other trees by roe deer and moose
(Götmark et al., 2005; unpublished data). Thinning of
birch, oak, and beech saplings in young coniferous
forest may be reduced and relatively dense broadleaved-dominated stands created, potentially swamping foraging ungulates and reducing damages per tree
(Vivås and Saether, 1987). Dense stands potentially
give good timber quality, but even if trees are
damaged, they may be used, for e.g. biofuel.
For long, oak regeneration has been weak and
considered a problem in temperate forest (e.g. Watt,
1919; Kelly, 2002; Abrams, 2003). This seems to be
the case in broadleaved forests in Sweden, when oak is
compared to other trees. However, our data indicate,
and we suggest that oak regeneration is stronger and
generally not problematical in conifer-dominated
forests, a hypothesis that should be tested in future
studies.
Among the tree species largely limited to broadleaved forest, the high per capita regeneration success
of ash under natural conditions is promising for
various forms of forestry (see also Kerr, 2004).
Between 1985 and 2000 in the study area, for a group
of trees where ash formed more than 50% of the
saplings, stem densities (0–10 cm dbh) more than
doubled (Götmark et al., unpublished data). Thus, deer
browsing had little effect on recruitment of ash
saplings (see also Götmark et al., 2005).
Forestry based on broadleaves requires integration
with conservation work (e.g. Nilsson et al., 2001,
2002). For instance, old hollow oaks contain
endangered species of lichens and beetles (Ranius
and Jansson, 2000), so planning for oak regeneration is
especially important in areas where such oaks occur.
Pure or mixed stands arising from regeneration
described here may be used for harvesting of
energy-rich hardwood for biofuel (Johansson,
2003), which has environmental advantages compared
to fossil fuel. To strengthen biodiversity values,
natural succession and self-thinning in some proportion of the stands is motivated (Nordén et al., 2004a,b).
Our study suggests that natural regeneration of
broadleaves offers a good potential for creating mixed
155
stands in conifer-dominated forests in the future, at
least in Sweden. This would reduce the costs of tree
planting. Initially, changes in pre-commercial thinning
regimes should be implemented, to favour broadleaved trees.
Acknowledgements
The Swedish Research Council (VR) and The
Swedish Foundation for International Cooperation in
Research and Higher Education (STINT) provided
financial support for F.G. This article was written
while F.G. was Visiting Scientist at University of
Wisconsin, Department of Forest Ecology and
Management at Madison, USA. F.G. thanks especially
Craig Lorimer, and Raymond Gurie and others there,
for hospitality; thanks also to Julie and Sophie. We
thank Matts Karlsson, anonymous reviewers, and
especially David Greene for valuable comments that
improved the manuscript. Also, thanks to all people
involved in collecting valuable data for the Swedish
National Forest Inventory. We dedicate our study to
oaks, beeches, ashes and other fellows in the forest.
References
Abrams, M.D., 2003. Where has all the white oak gone? Bioscience
53, 927–939.
Berg, Å., Ehnström, B., Gustafsson, L., Hallingbäck, T., Jonsell, M.,
Weslien, J., 1994. Threatened plant, animal, and fungus species
in Swedish forests: distribution and habitat associations. Conserv. Biol. 8, 718–731.
Björkman, L., 1998. Bokens historia i södra Sverige – en litteraturöversikt. Svensk Bot. Tidskr. 91, 573–583 (English summary).
Bond, W.J., Midgley, J.J., 2001. Ecology of sprouting in woody
plants: the persistence niche. Trends Ecol. Evol. 16, 45–51.
Bossema, I., 1979. Jays and oaks: an eco-ethological study of a
symbiosis. Behaviour 70, 1–117.
Bradshaw, R.H.W., Holmqvist, B.H., Cowling, S.A., Sykes, M.T.,
2000. The effects of climate change on the distribution and
management of Picea abies in southern Sweden. Can. J. For. 30,
1992–1998.
Clarke, J.S., et al., 1999. Interpreting recruitment limitation in
forests. Am. J. Bot. 86, 1–16.
Deiller, A.-F., Walter, J.-M.N., Trémolières, M., 2003. Regeneration
strategies in a temperate hardwood floodplain forest of the Upper
Rhine: sexual versus vegetative reproduction of woody species.
For. Ecol. Manage. 180, 215–225.
Diekmann, M., 1994. Deciduous forest vegetation in Boreo-nemoral
Scandinavia. Acta Phytogeographica suecica 80, 112.
156
F. Götmark et al. / Forest Ecology and Management 214 (2005) 142–157
Diekmann, M., 1996. Ecological behaviour of deciduous hardwood
trees in Boreo-nemoral Sweden in relation to light and soil
conditions. For. Ecol. Manage. 86, 1–14.
Del Tredici, P., 2001. Sprouting in temperate trees: a morphological
and ecological review. Bot. Rev. 67, 121–140.
Fällman, K., Ligné, D., Karlsson, A., Albrektson, A., 2003. Stem
quality and height development in a Betula-dominated stand
seven years after precommercial thinning at different stump
heights. Scand. J. For. Res. 18, 145–154.
Forestry Statistics, 2000. Official Statistics of Sweden. Swedish
University of Agricultural Science, Umeå.
Fridman, J., 2000. Conservation of forest in Sweden: a strategic
ecological analysis. Biol. Conserv. 96, 95–103.
Fridman, J., Walheim, M., 2000. Amount, structure and dynamics of
dead wood on managed forestland in Sweden. For. Ecol. Manage. 131, 23–36.
Frost, I., 1997. Dispersal and establishment of Quercus robur. Ph.D.
thesis, Dissertation from the Faculty of Science and Technology
305, Uppsala University, Sweden.
Götmark, F., Berglund, Å., Wiklander, K., 2005. Browsing damage on
broadleaved trees in semi-natural temperate forest in Sweden,
with a focus on oak regeneration. Scand. J. For. Res. 20, 228–234.
Greene, D.F., Johnsson, E.A., 1995. Long-distance wind dispersal of
tree seeds. Can. J. Bot. 73, 1036–1045.
Greene, D.F., Zasada, J.C., Sirois, L., Kneeshaw, D., Morin, H.,
Charron, I., Simard, M.J., 1999. A review of the regeneration
dynamics of North American boreal forest tree species. Can. J.
For. Res. 29, 824–839.
Grime, J.P., Hodgson, J.G., Hunt, R., 1988. Comparative Plant
Ecology. Unwin Hyman, London.
Gustafsson, L., 2000. Red-listed species and indicators: vascular
plants in woodland key habitats and surrounding production
forest in Sweden. Biol. Conserv. 92, 35–43.
Gustafsson, L., Ahlén, I. (Eds.), 1996. National Atlas of Sweden:
Geography of Plants and Animals. Almqvist & Wiksell International, Stockholm.
Hamilton, G., Mirton, A., 1998. Sköter vi ädellövskogen? (Is the
noble forest being managed?). Meddelande 12. Skogsstyrelsen,
Jönköping (in Swedish).
Harmer, R., Kerr, G., Boswell, R., 1997. Characteristics of lowland
broadleaved woodland being restocked by natural regeneration.
Forestry 70, 199–210.
Harvey, B.D., Bergeron, Y., 1989. Site patterns of natural regeneration following clear-cutting in northwestern Quebec. Can. J. For.
Res. 19, 1458–1469.
Hunter, M.L. (Ed.), 1999. Maintaining Biodiversity in Forest
Ecosystems. Cambridge University Press, Cambridge, UK,
pp. 335–372.
Johansson, T., 1992. Sprouting of 10- to 50-year-old Betula pubescens
in relation to felling time. For. Ecol. Manage. 53, 283–296.
Johansson, T., 2003. Mixed stands in Nordic countries – a challenge
for the future. Biomass Bioenergy 24, 365–372.
Johnson, P.S., Shifley, S.R., Rogers, R., 2002. The Ecology and
Silviculture of Oaks. CABI Publishing, Wallingford.
Karlsson, M., 2001. Natural regeneration of broadleaved tree species
in southern Sweden. Doctoral thesis, Silvestria 196, Swedish
University of Agricultural Sciences, Alnarp.
Kauppi, A., Kiviniitty, M., Ferm, A., 1988. Growth habits and crown
architecture of Betula pubescens Ehrh. of seed and sprout origin.
Can. J. For. Res. 18, 1603–1613.
Kelly, D.L., 2002. The regeneration of Quercus petraea (sessile oak)
in southwest Ireland: a 25-year experimental study. For. Ecol.
Manage. 166, 207–226.
Kerr, G., 2004. The growth and form of ash (Fraxinus excelsior) in
mixture with cherry (Prunus avium), oak (Quercus petraea and
Quercus robur), and beech (Fagus sylvatica). Can. J. For. Res.
34, 2340–2350.
Kerr, G., Cahalan, C., 2004. A review of site factors affecting the
early growth of ash (Fraxinus excelsior L.) For. Ecol. Manage.
188, 225–234.
Latham, R.E., Ricklefs, R.E., 1993. Continental comparisons of
temperate-zone tree species diversity. In: Ricklefs, R.E., Schulter, D. (Eds.), Species Diversity in Ecological Communities.
University of Chicago Press, Chicago, pp. 294–314.
Laiho, O., Lähde, E., Norokorpi, Y., Saksa, T., 1994. Undergrowth as
a regeneration potential in Finnish forests. In: Bamsey, C.R.
(Ed.), Proceedings: Innovative Silviculture Systems in Boreal
Forests. Clear Lake Ltd., Edmonton, Alta., pp. 90–94.
Levin, D.A., Kerster, H.W., 1974. Gene flow in seed plants. Evol.
Biol. 7, 139–185.
Lindbladh, M., Bradshaw, R.H.W., Holmqvist, B.H., 2000. Pattern
and process in south Swedish forests during the last 3000 years
sensed at stand and regional scales. J. Ecol. 88, 113–128.
Lindgren, L., 1970. Beech forest vegetation in Sweden – a survey.
Bot. Notiser 123, 401–424.
Lindroth, S. (Ed.), 1995. Skog och mark i Sverige. Forest and Land
in Sweden.Kommentus Förlag, Stockholm (In Swedish).
Loehle, C., 2000. Strategy space and the disturbance spectrum: a lifehistory model for tree species coexistence. Am. Nat. 156, 14–33.
Löf, M., 2001. Uthålligt skogsbruk i ädellövskog. Rapport 19,
Institutionen för Sydsvensk skogsvetenskap. Alnarp, SLU, Sweden (In Swedish, English summary.)
Marigo, G., Peltier, J.-P., Girel, J., 2000. Success in the demographic
expansion of Fraxinus excelsior L. Tree 15, 1–13.
Mosandl, R., Kleinert, A., 1998. Development of oaks (Quercus
petraea (Matt.) Liebl.) emerged from bird-dispersed seeds under
old-growth pine (Pinus silvestris L.) stands. For. Ecol. Manage.
106, 35–44.
Mossberg, B., 1992. Den nordiska floran. Wahlström and Widstrand,
Brepols, Belgium.
Nilsson, N.-E. (Ed.), 1990. The National Atlas of Sweden: The
Forests. SNA Publishing, Stockholm.
Nilsson, S.G., 1985. Ecological and evolutionary interactions
between reproduction of beech Fagus silvatica and seed eating
animals. Oikos 44, 157–164.
Nilsson, S.G., 1997. Forests in the temperate-boreal transition:
natural and man-made features. Ecol. Bull. 46, 61–70.
Nilsson, S.G., Hedin, J., Niklasson, M., 2001. Biodiversity and its
assessment in boreal and nemoral forests. Scand. J. For. Res. 3,
10–26.
Nilsson, S.G., Niklasson, M., Hedin, J., Aronsson, G., Gutowski, J.,
Linder, P., Ljungberg, H., Mikusiski, G., Ranius, T., 2002.
Densities of large living and dead trees in old-growth temperate
and boreal forest. For. Ecol. Manage. 161, 189–204.
F. Götmark et al. / Forest Ecology and Management 214 (2005) 142–157
Nordborg, F., Olsson, S., 1999. Changes in soil mineralogy and
exchangeable cation pools in stands of Norway spruce planted
on former pasture land. Plant Soil 207, 219–229.
Nordén, B., Ryberg, M., Götmark, F., Olausson, B., 2004a. Relative
importance of coarse and fine woody debris for the diversity of
wood-inhabiting fungi in temperate broadleaf forests. Biol.
Conserv. 117, 1–10.
Nordén, B., Götmark, F., Tönnberg, M., Ryberg, M., 2004b. Dead
wood in semi-natural temperate broadleaf woodland: contribution of coarse and fine dead wood, attached dead wood and
stumps. For. Ecol. Manage. 194, 235–248.
Nyland, R.D., 2002. Silviculture, Concepts and Applications.
McGraw-Hill, Boston.
Olsthoorn, A.F.M., Bartelink, H.H., Gardiner, J.J., Pretzsch, H.,
Hekhuis, H.J., Frank, A. (Eds.), 1999. Management of
Mixed-Species Forest: Silviculture and Economics. IBN Scientific Contribution 15. IBN Scientific, Wageningen.
Persson, J., 1990. Rikare skog (Richer forest). Skogsstyrelsen
(National Board of Forestry), Jönköping (in Swedish).
Peterken, G.F., 1996. Natural Woodland. Ecology and Conservation
in Northern Temperate Regions. Cambridge University Press,
Cambridge.
Pigott, C.D., Huntley, J., 1981. Factors controlling the distribution of
Tilia cordata at the northern limits of its geographical range.
New Phytol. 87, 817–839.
Ranius, T., Jansson, N., 2000. The influence of forest regrowth,
original canopy cover and tree size on saproxylic beetles associated with old oaks. Biol. Conserv. 95, 85–94.
Ranneby, B., Cruse, T., Hägglund, B., Jonasson, H., Swärd, J., 1987.
Designing a new national forest survey for Sweden. Studia
Forestalia Suecica 177, 1–29.
Rinne, P., Kauppi, A., Ferm, A., 1987. Induction of adventitious
buds and sprouts on birch seedlings (Betula pendula Roth and B.
pubescens Ehrh.) Can J. For. Res. 17, 545–555.
SEPA (Swedish Environmental Protection Agency), 1994. Biological Diversity in Sweden. SEPA, Solna, Sweden.
Simard, S., Sachs, D.L., Vyse, A., Blevins, L.L., 2004. Paper birch
competitive effects vary with conifer tree species and stand age
in interior British Columbia forests: implications for reforestation policy and practice. For. Ecol. Manage. 198, 55–74.
157
SLU, 1997. Instruktion för fältarbetet vid Riksskogtaxeringen år
1997 (Field manual for the NFI, 1997). Sveriges Lantbruksuniversitet. Institutionen för skoglig resurshushållning och geomatik, Umeå, Sweden.
SOU (Swedish Public Reports), 1992 Skogspolitiken inför 2000talet (Report 76). Miljödepartementet, Stockholm (In Swedish).
Spiecker, H., Hansen, J., Klimo, E., Skovsgaard, J.P., Sterba,
H., von Teuffel, K. (Eds.), 2004. Norway Spruce Conversion
– Options and Consequences. Brill, Leiden.
Suszka, B., Muller, C., Bonner-Masimbert, M.B., 1996. Seeds of
Forest Broadleaves: From Harvest to Sowing. INRA Editions,
Paris.
Sykes, M.T., Prentice, I.C., 1996. Climate change, tree species
distribution and forest dynamics: a case study in the mixed
conifer/northern hardwood zone of Europe. Clim. Change 34,
161–177.
Tapper, P.-G., 1992. Demography of persistent juveniles in Fraxinus
excelsior. Ecography 15, 385–392.
Tapper, P.-G., 1996. Tree dynamics in a successional Alnus-Fraxinus
woodland. Ecography 19, 237–244.
Thelin, G., Sverdrup, H., Holmqvist, J., Rosengren, U., Lindén, M.,
2002. Sustainability in spruce and mixed species stands. In:
Sverdrup, H., Stjernquist, I. (Eds.), Developing Principles and
Models for Sustainable Forestry in Sweden. Kluwer, Dordrecht,
pp. 337–354.
Thorell, K.E., Östlin, L.E., 1931. The national forest survey of
Sweden. J. For. 4, 585–591.
Vivås, J., Saether, B.-E., 1987. Interactions between a generalist
herbivore, the moose Alces alces and its food source: an
experimental study of winter foraging behaviour in relation to
browse availability. J. Anim. Ecol. 56, 509–520.
Watt, A.S., 1919. On the causes of failure of natural regeneration in
British oak woods. J. Ecol. 17, 173–203.
Willson, M.F., 1993. Dispersal mode, seed shadows, and colonization patterns. Vegetation 108, 261–280.
Young, R.A., Giese, R.L. (Eds.), 2003. Introduction to forest
ecosystem science and management. Wiley, Hoboken, NJ.
Zerbe, S., 2002. Restoration of natural broad-leaved woodland in
Central Europe on sites with coniferous forest plantations. For.
Ecol. Manage. 167, 27–42.