Sten Svantesson
Uppsats för avläggande av naturvetenskaplig magisterexamen i
Växtekologi 60 hp
Institutionen för biologi och miljövetenskap Göteborgs universitet
Ecological requirements of corticioid fungi
– a study on species richness and community
composition in south-eastern Norway
0
Contents
Abstract ................................................................................................. 1
Sammanfattning ................................................................................... 1
Introduction .......................................................................................... 2
Materials and methods ........................................................................ 6
Study sites .............................................................................................................................. 6
Field methods ......................................................................................................................... 7
Explanatory variables ............................................................................................................. 8
Studied taxa and species identifications ................................................................................. 9
Species categories ................................................................................................................ 10
Generalised linear mixed-effects models (GLMM) analysis ............................................... 11
Non-metric multidimensional scaling (NMDS) analysis ..................................................... 12
Results ................................................................................................. 14
General ................................................................................................................................. 14
Determinants of species richness ......................................................................................... 14
Ordination of species composition and correlated variables ................................................ 18
Discussion ........................................................................................... 21
Conclusions and implications for conservation .............................. 27
Acknowledgements ............................................................................ 27
References ........................................................................................... 28
Appendix 1 .......................................................................................... 33
Appendix 2 .......................................................................................... 37
1
Abstract
Although the corticioid fungi are a species rich and ecologically important group, with many
Red Listed species, knowledge about their ecological requirements is limited. Previous
ecological studies have often excluded them, due to their inconspicuous fruit bodies and time-
consuming identification. In the cases where corticioids have been included they have very
seldom been analysed separately from other fungal groups living in dead wood.
At six sites of old-growth spruce forest in south-eastern Norway, I made an inventory of 48
logs in total, equally divided between the decay classes 3, intermediately decomposed, and 4,
well decomposed. All fruit bodies of corticioid species were collected. In addition to decay
class 16 explanatory variables were analysed for correlations with species composition and
the richness of all, Red Listed, common, occasional and rare species. I also aimed to find the
best models explaining species richness and study whether the richness of Red Listed species
varied with the richness of non-Red Listed species.
I found on average 17.2 ± 4.5 (S.D.) species per log. The respective species richness of all,
common and occasional species were all positively correlated with log size. The total species
richness was also greater on logs created due to stem breakage than on uprooted logs, and
there were more occasional species on logs with more area of polypore hymenophores than on
logs with less polypore hymenophore area. The number of Red Listed species per log
increased with ground contact and the total number of non-Red Listed species, decreased with
bark coverage and was higher on broken logs and logs in decay class 4 than on uprooted logs
and logs in decay class 3. The best models explaining the species richness of the different
categories were quite consistent with the correlations the categories formed with the single
variables. The species composition on the studied logs was correlated with ground contact,
plant coverage, bark coverage, log type (uprooted or broken), polypore hymenophore area,
decay class and richness of Red Listed species.
I conclude that even within a single, narrowly defined kind of dead wood, a great variation
in microenvironments exists, and that many of these environments host differently
composited and species rich communities of corticioid fungi. The number of species found on
logs in this study was higher than what has previously been reported in any studies of similar
substrates, further underscoring the importance dead wood has for biodiversity. The rare
species needed many different kinds of logs, and hence a diversity of dead wood is required to
host them. Both the rare and the Red Listed species needed well decayed logs, a substrate that
today can almost only be found in the most natural forests.
Sammanfattning
Fastän skinnsvamparna är en artrik och ekologiskt viktig grupp med många rödlistade arter är
kunskapen om deras ekologiska krav mycket begränsad. Med anledning av sina oansenliga
fruktkroppar och tidskrävande artbestämning har de ofta blivit exkluderade ur tidigare
ekologiska studier. I de fall då de inkluderats har de mycket sällan separerats från andra
vedlevande svampar.
På sex lokaler med gammelskog av gran i sydöstra Norge genomförde jag en inventering
av sammanlagt 48 lågor, jämnt fördelade mellan nedbrytningsklasserna 3, intermediärt
nedbruten, och 4, väl nedbruten. Alla fruktkroppar samlades in. Utöver nedbrytningsklass
analyserade jag 16 olika variabler för att finna korrelationer mellan dessa och artrikedom och
artsammansättning av alla, rödlistade, vanliga, tillfälliga och sällsynta arter. Jag hade också
för avsikt att finna de modeller som bäst kunde förklara artrikedom samt ta reda på huruvida
rikedomen av rödlistade arter var korrelerad med rikedomen av icke rödlistade arter.
2
Jag fann i snitt 17.2 ± 4.5 (S.D) arter per låga. De respektive rikedomarna av alla, vanliga
och tillfälliga arter var samtliga positivt korrelerade med lågastorlek. Den totala artrikedomen
var också större på lågor som skapats till följd av stambrott än på rotvältor, och det fanns fler
tillfälliga arter på lågor med större fruktkroppsareal av tickor än på lågor med mindre
fruktkroppsareal. Antalet rödlistade arter per låga ökade med markkontakt och rikedom av
icke rödlistade arter, minskade med barktäckning och var högre på avbrutna lågor och lågor i
nedbrytningsklass 4 än på rotvältor och lågor i nedbrytningsklass 3. De modeller som bäst
kunde förklara artrikedomen i de undersökta kategorierna var tämligen lika de korrelationer
som förekom mellan kategorierna och de enskilda variablerna. Artsammansättningen på de
studerade lågorna korrelerade med variablerna: markkontakt, växttäckning, barktäckning,
lågatyp (avbruten eller rotvälta), fruktkroppsareal av tickor, nedbrytningsklass och rikhet av
rödlistade arter.
Jag drar slutsatsen att det även inom en enda, snävt definierad typ av död ved förekommer
en stor variation i mikromiljöer och att många av dessa miljöer hyser både olikt sammansatta
och artrika samfund av skinnsvampar. Antalet arter funna på stockarna i denna studie var
större än vad som tidigare rapporterats i någon studie av liknande substrat – ett resultat som
ytterligare understryker betydelsen död ved har för den biologiska mångfalden. De sällsynta
arterna behövde många olika sorters lågor och således krävs en mångfald av död ved för att
kunna härbärgera dem. Både de sällsynta och de rödlistade arterna behövde väl nedbrutna
lågor, ett substrat man idag nästan enbart finner i de mest naturliga skogarna.
Introduction
Wood-decaying fungi are in several senses very important organisms. Their action of decay
plays a crucial role in the nutrient recycling in the forests of the world (Harmon et al. 1986;
Lonsdale et al. 2008). Through constituting accessible nutrition and creating microhabitats
they make existence possible for, or start the food chains of, the many organisms living
directly in dead wood or who in turn are nutritionally dependent upon the ones that do
(Lonsdale et al. 2008; Gjerde et al. 2009). Siitonen (2001) found that in Finland there are
4000-5000 species of dead wood-dependent organisms, representing 20-25 % of the total
number of forest-inhabiting species.
Distinguished by their simple, often small and inconspicuous, crust-like fruit bodies that in
most cases need the aid of a microscope to be identified (Hjortstam et al. 1988), the corticioid
fungi constitute a major group of wood-inhabiting fungi; only in the Nordic countries there
are 555 species, accounting for approximately 27 % of all the wood-inhabiting fungal species
in the area (Stokland and Meyke 2008). The corticioids make up a highly polyphyletic group,
representing almost all major clades of the homobasidiomycetes (Larsson et al. 2004), and
though often referred to as wood decayers a smaller portion of them are in fact
ectomycorrhizal or litter and humus decayers (Stokland and Meyke 2008).
3
For 144 species, or 29 % of all the corticioid species present in Norway, the risk to go
nationally extinct is either 5 % or more, or is very hard to evaluate, and consequently they are
Red Listed (Hofton unpublished; Kålås et al. 2010). The reasons for this are that the
populations of some corticioid fungi seem to be very small and many are considered to be
declining (Hofton unpublished; Kålås et al. 2010). These conditions are in turn generally
regarded to be caused by the loss of natural forest (Kålås et al. 2010), but, even though such
information is badly needed in conservation, the relative importance of the factors involved
remains deficiently known, as knowledge about the ecological requirements of corticioid
fungi at the community level is very limited in available literature. The entire group has
received less attention than many other macrofungal groups, due to the often inconspicuous
fruit bodies of its members and the difficulties involved with their identification (Hjortstam et
al. 1988).
Knowledge about the ecological requirements of corticioid fungi at the community level is
mainly limited to what can be interpreted from studies addressing wood-inhabiting fungi in
general; there has been very little separate research on the ecology of corticioids. In studies on
species richness of wood-inhabiting fungi on Norway spruce Picea abies (L.) H. Karst in the
taiga region, providing a value for the mean number of species fruiting per log (Renvall 1995;
Lindblad 1998; Edman et al. 2004; Berglund et al. 2005; Juutilainen et al. 2011; Olsson et al.
2011; Stokland and Larsson 2011) this measure varies from 3.2 (Renvall 1995) to almost 11
(Lindblad 1998).
A positive relationship has been found to exist between the size of a dead wood substrate
and the number of wood-inhabiting fungi fruiting on it (Lonsdale et al. 2008; Bader et al.
1995; Heilmann-Clausen and Christensen 2004; Lindhe et al. 2004), consistent also in studies
with a large proportion of corticioid species included (Renvall 1995; Høiland and Bendiksen
1997; Lindblad 1998; Stokland and Larsson 2011). On the other hand corticioid fungi seem to
need less space than other basidiomycetic fungi; in a study "investigating the relative
importance of coarse (diameter >10 cm) and fine woody debris (1–10 cm) for fungi in
broadleaf forests in southern Sweden" Nordén et al. (2004) made nearly 80 % and more than
70 % of all their records of corticioid and stereoid fungi, respectively, on fine woody debris,
4
while the corresponding numbers for polypores and agarics were about 50 % and 40 %,
respectively. Renvall (1995) found that the species composition of wood-inhabiting fungi,
among them many corticioids, on spruce logs in Northern Finland, differs with substrate size -
a result confirmed by later studies; Stokland and Larsson (2011) found many corticioid
species to be specialised on logs larger or smaller than 30 cm in diameter, and Küffer et al.
(2008) and Juutilainen et al. (2011) found the communities of corticioid fungi occurring on
dead wood pieces with a diameter smaller than about one cm to be different from those on
larger pieces. Tikkanen et al. (2006) concluded that of all Red Listed (following Rassi et al.
2001), boreal, aphyllophoraceous fungi (also including species that do not live in dead wood)
in Finland, less than 5 % were specialised in dead wood with a diameter smaller than 10 cm,
more than 50 % in dead wood larger than 10 but smaller than 30 cm in diameter, and nearly
30 % in logs larger than 30 cm in diameter. About 10 % of the species were indifferent to the
diameter of dead wood.
Decay class has in several studies (Renvall 1995; Høiland and Bendiksen 1997; Lindblad
1998; Stokland and Larsson 2011) been found to be the quality-related variable with the
highest impact on species richness of wood-inhabiting fungi on coniferous trees in the Nordic
countries. Concerning spruce, all these studies found medium decayed logs to be the most rich
in species. Several studies also indicated the presence of species-specific differences in
relation to decay class (Renvall 1995; Høiland and Bendiksen 1997; Lindblad 1998; Küffer
and Senn-Irlet 2005; Tikkanen et al. 2006; Stokland and Larsson 2011).
Renvall (1995) found seven different successional pathways of wood-inhabiting fungi on
spruce in northern Finland; six pathways occurred on broken logs, while the species
community on uprooted logs constituted the seventh. He showed that the uprooted log
pathway had a mean species richness which was lower than any of the pathways of the broken
logs. He further argued that the type of death a tree suffers is linked to the identity of its
primary decayers, which start successional pathways with different species composition and
richness. Heilmann-Clausen and Christensen (2003) found a corresponding difference in
species richness of Red Listed wood-inhabiting fungi between uprooted and broken logs of
beech Fagus sylvatica L. in Denmark, for which they ascribed the same explanation. Lindblad
(1998) also included log type in her study but found no significant relation.
5
In a study by Mahmood et al. (2001) certain species of the corticioid, ectomycorrhizal
genus Piloderma Jülich turned out to colonise base rich ash granules experimentally placed in
soil, while other corticioid species did not, indicating that the calcium content of the ground
may affect the corticioid funga. Stokland and Kauserud (2004) also showed that soil
properties may indirectly have an impact on the community of wood-inhabiting fungi: the
strictly wood-decaying polypore, Phellinus nigrolimitatus (Romell) Bourdot & Galzin was
found to be more common in high productive than in medium and low productive forests.
Stokland and Larsson (2011) concluded that rare wood-inhabiting species on spruce were
reduced by forestry to a considerably higher degree than common species, thus implicating
that species of different occurrence frequency may have different ecological requirements.
Halme and Kotiaho (2012) showed that the outcome of any study on wood-inhabiting fungi
with annual fruit bodies is heavily dependent on the time of season, at which the data
collection is performed. I have taken both of these issues into account in this master's thesis, a
study focused on Norway spruce within the project Habitat fragmentation and Pathways to
Extinction in dead-wood dependent fungi (PATHEXT), led by Jenni Nordén and Karl-Henrik
Larsson at the University of Oslo. I have aimed to analyse the ecological requirements of
corticioid fungi at a community level through the following questions:
1. How is the composition and richness of corticioid fungi regarded as all species and
Red Listed species as well as rare, occasional and common species correlated with:
– the properties of the logs they inhabit?
– the properties of the surroundings of the logs they inhabit?
2. What model – i.e. what set of variables – best explains the richness of all, Red Listed,
common, occasional and rare corticioid species, respectively?
3. Does the richness of Red Listed corticioid species vary with the richness of:
– common, occasional and rare non-Red Listed species, respectively?
– all non-Red Listed species?
6
Tab
le 1
: B
ackg
rou
nd
in
form
ati
on
of
the s
tud
y s
ites. ”D
ista
nce t
o s
mall s
ite»
den
ote
s t
he d
ista
nce b
etw
een
th
e c
en
ter
co
ord
inate
s o
f th
e s
am
ple
plo
ts in
a s
ite p
air
(F
ig.
1).
Alt
itu
de is g
iven
as t
he r
an
ge b
etw
een
th
e lo
west
an
d t
he h
igh
est
po
sit
ion
ed
su
rvey
ed
lo
g. C
od
es u
sed
to
den
ote
veg
eta
tio
n s
ecti
on
s: O
C: in
dif
fere
nt;
O1: slig
htl
y
ocean
ic;
O2: m
ark
ed
ly o
cean
ic.
Materials and methods
Study sites
The fieldwork was performed at six 200 x
200 m sample plots of the PATHEXT
project, which held 60 preselected logs each.
The plots were positioned in sites (Table 1),
and the sites were situated in pairs; each pair
constituting one small and one large site.
The large sites were nature reserves and the
small sites comprised a woodland key
habitat (Sweden) and two nature type
localities (Norway; in forest, areas similar to
woodland key habitats; see further
Direktoratet for naturforvaltning 2007)
located to south-eastern Norway and south-
western Sweden (Fig. 1). The sites were
situated in the boreonemoral and middle
boreal vegetation zones (Dahl et al. 1986;
Gusfafsson and Ahlén 1996) and in the
indifferent (OC) to the markedly oceanic
(O2) vegetation sections (Moen 1999). The
altitude of the sample plots varied from 23
(Skee) to 569 m.a.s.l. (Rudskampen),
measured from the lowest to the highest
positioned log, respectively. All the sample
plots and great parts of the sites constituted
old, spruce-dominated forests with a lot of
dead wood (Norwegian Directorate for
Nature Management 2012; Swedish Forest
Agency 2012; personal observations). The
history of the sample plots and the sites was
varying, but all the sites have been left
OC
, O1
O2
O1
O1
O1
O2
Bo
reo
nem
ora
l
Mid
dle
bo
real
Mid
dle
bo
real
Bo
reo
nem
ora
l
Bo
reo
nem
ora
l
33
7-3
98
29
3-3
47
52
9-5
51
54
0-5
69
15
3-1
75
23
-53
17
.62
2
- 18
.16
2
- 44
.94
6
-
24
48
.1
13
.1
18
49
.9
8.2
39
2.4
5.1
Lar
ge
Sm
all
Lar
ge
Sm
all
Lar
ge
Sm
all
Ty
pe
Nat
ure
res
erv
e
Nat
ure
ty
pe
loca
lity
Nat
ure
res
erv
e
Nat
ure
ty
pe
loca
lity
Nat
ure
res
erv
e
Wo
od
lan
d k
ey h
abit
at
Sit
e (
na
me;
mu
nic
ipa
lity
; co
un
ty)
San
dal
slia
; D
ran
ged
al;
Tel
emar
k
Ru
dsk
amp
en;
Nan
nes
tad
; A
ker
shu
s
Tjø
stø
l; A
rem
ark
; Ø
stfo
ld
Sk
ee;
Str
öm
stad
; V
ästr
a G
öta
lan
d (
SE
)
Veg
eta
tion
secti
on
Veg
eta
tion
zon
e
Bo
reo
nem
ora
l, m
idd
le b
ore
al
Alt
itu
de
(m.a
.s.l
.)
Dis
tan
ce
to s
mall
sit
e (
km
)
Siz
e
(ha)
Siz
e
cla
ss
Mø
rkv
assj
uv
et;
No
me,
Dra
ng
edal
;
T
elem
ark
Sp
ålen
-Kat
no
sa;
Jev
nak
er, L
un
ner
,
R
ing
erik
e; B
usk
eru
d, O
pp
lan
d
7
Figure 1: The geographic position of the large (red) and small (blue) study sites.
untouched by the extensive cuttings of the 20th
century and in the sample plots there were few
or no visible signs of human impact (Norwegian Directorate for Nature Management 2012;
Swedish Forest Agency 2012; personal observations).
Field methods
At every sample plot I selected four logs each in the decay classes 3, intermediately
decomposed, or 4, well decomposed (for the decay classification, see Hottola and Siitonen
2008), with a diameter at breast height (DBH) in the range of 20 to 40 cm and a length of at
least eleven meters. The field work team measured the DBH and the total length of the logs.
We visually estimated the degree of bark and plant coverage and ground contact for each log
and noted whether they had fallen down due to breakage or uprooting. We also checked
8
Table 2: the
explanatory variables.
whether the tops had been broken or remained unbroken. Tops broken above the point where
the logs were five cm in diameter were still counted as unbroken. I identified the vegetation
types (see below) present within a distance of approximately two metres to each side of the
logs according to the older of the two standard systems used in Norway (Fremstad 1997). The
identification was made to the lowest level possible – in nearly all cases to subtype. We
carefully surveyed the logs for fruit bodies of corticioid fungi, collecting specimens of the
great majority that we could not identify in the field. We visually estimated the hymenophore
area of the polypores present, counting both living and dead fruit bodies. The field work was
conducted during October and early November 2011. We noted the survey date of each log to
account for seasonal effects (see further Halme and Kotiaho 2012).
Explanatory variables
In addition to the variables obtained directly in the field and site
size (Table 1), I chose to regard the effect of vegetation types in
four different ways and to calculate the volume and surface area
of the included logs, thence making the number of explanatory
variables 17 (Table 2).
By vegetation types I refer in this thesis to different types of
spruce-containing forests, swamps and fens separated by the
presence, absence and dominance of certain species of vascular
plants, bryophytes and lichens, in all vegetation layers (see further
Fremstad 1997). Halvorsen et al. (2009) showed that vegetation
types in forests are clearly defined by the combination of three
different ecoclines. I therefore choose to analyse the effect of
these rather than of the vegetation types per se. The calcium
content of the ground turned out to be the only ecocline with
enough variation to be analysed. I did so classifying it into three
different scales: a fine ordered scale (range 1-5), a coarse ordered
scale (range 1-3) and as presence/absence of richer vegetation
types. The rich vegetation types include: low herb woodland, tall
fern woodland, tall herb Norway spruce forest and rich swamp
woodland; the vegetation types belonging to the three highest
categories of Halvorsen’s et al. (2009) five-point scale of calcium
content. I also included the number of vegetation types as a variable.
Explanatory variables
DBH
Length
Volume
Surface area
Log type
Decay class
Bark coverage
Plant coverage
Ground contact
Calcium, fine scale
Calcium, coarse scale
Presence of richer vegetation types
Number of vegetation types
Top intactness
Polypore hymenophore area
Site size
Time of survey
9
To calculate the volume of the unbroken logs (the logs whose tops remained unbroken) I used
the formula of Laasasenaho (1982):
V = 0.022927·d1.91505
·0.99146d·h
2.82541·(h-1.3)
-1.53547/1000
where d is the DBH in cm, and h is the length in metres.
To calculate the volume of the broken logs I first used a linear model to see if the DBH of
the unbroken logs explained their length. I included all unbroken logs among the 60 logs
present at each sample plot in the PATHEXT data set and found unbroken log length to be
significantly explained by DBH at all plots but Mørkvassjuvet. Through linear regression I
could thus predict the length the broken logs had while they were still unbroken and
subsequently also calculate their unbroken volume, using the same formula as before. I
deduced the volume of the missing tops of the broken logs, assuming that they and the entire
logs, while unbroken, could be seen as parts of the same cone. I then subtracted the volume of
the top from the volume of the log. All these steps can be summarised as the formula:
V = (0.022927·d1.91505
·0.99146d·h
2.82541·(h-1.3)
-1.53547/1000)-(1/3·(d/200·(h-hbroken)/(h-
1.3))2·(h-hbroken)·π)
where V is the volume of the broken log, d is the DBH in cm, h is the predicted length of the
unbroken log and hbroken is the length of the broken log. The few broken logs that were longer
than their model-predicted full lengths were treated as unbroken logs, using their actual
lengths as full lengths.
To calculate the surface area of the logs I used the equation of the surface area of a cone,
subtracting the area of the top from the ones that were broken.
Studied taxa and species identifications
The study was mainly limited to corticioid fungi in a wide sense Appendix 1), i.e. apart from
species with effused, smooth to spiny fruit bodies, also including such whose fruit bodies
consist of separately growing spines (e.g. Mucronella Fr. spp. and Henningsomyces candidus
(Pers.) Kuntze). All specimens with spiny fruit bodies were thus included and identified to
species level. Specimens belonging to the genera Calocera (Fr.) Fr., Dacrymyces Nees,
Pseudotomentella Svrcek and Tomentella Pers. ex Pat. were included but only identified to
genus level (with the exception of the spiny T. fibrosa (Berk. & M.A. Curtis) Kõljalg).
10
Corticioid Heterobasidiomycetes were otherwise excluded, as were all Ascomycetes, with the
exception of Sebacina calcea (Pers.) Bres. and Camarops tubulina (Alb. & Schwein.) Shear –
two easily identified species. Due to doubts about the correctness of the stricter species
delimitations of Athelia epiphylla Pers., Litschauerella clematitis (Bourdot & Galzin) J.
Erikss. & Ryvarden, Thanatephorus fusisporus (J. Schröt.) Hauerslev & P. Roberts and
Tubulicrinis borealis J. Erikss., I chose to treat these species in broader concepts than what is
commonly recognised. I included specimens belonging to undescribed but clearly distinct
species in the analyses and assigned them names consisting of the genus name – if known –
and a number (e.g. Hyphodontia J. Erikss. sp. 1). The many specimens identified as
Botryobasidium candicans J. Erikss. most certainly belong to a separate, undescribed species,
as B. candicans is a species limited to deciduous wood (Eriksson and Ryvarden 1973), but
since they otherwise fit the species description well, and for want of a better name, I chose to
include them under B.candicans. I encountered specimens of Botryobasidium aureum
Parmasto or B. conspersum J. Erikss. or B. ellispsosporum Hol.-Jech. lacking their respective
anamorphs, and since these three species then cannot be separated I treated them as B. vagum
group indet.
I identified the specimens using a 300-1600x phase contrast microscope, a 50x stereo loupe
and by sight. The nomenclature follows the Species Nomenclature Database (Artsdatabanken
2012).
Species categories
To assign species to the frequency categories common, occasional and rare I directly used the
categories and classification provided in Stokland and Larsson (2011) for the major part of the
species. Their study covered roughly the same geographical area as mine, and their material is
the largest in Norway so far, with collections of polypores and corticioids from 1138 spruce
logs and 992 pine Pinus sylvestris L. logs from 90 managed and 34 natural or near natural
spruce or pine forests. Stokland and Larsson (2011) classified a species as rare if it was found
on 10 or fewer of their logs, occasional if it was found on 11-30 logs, and common if it was
found on more than 30 logs. With the exception of a few species that are deficiently known
(K.-H. Larsson, pers. comm.), I treated all officially described corticioid species absent from
their material as rare. I assigned the following taxa and groups of species to no category:
deficiently known species, Heterobasidiomycetes, Calocera spp., Mucronella spp.,
Henningsomyces candidus, Sphaerobolus stellatus Tode, Camarops tubulina and undescribed
species with the exception of Hyphoderma velatum K.-H. Larsson ad int (undescribed species
11
with an interimistic name) and Trechispora minuta K.-H. Larsson ad int. The last two were
instead considered rare (K.-H. Larsson, pers. comm.).
The Red List status follows The 2010 Norwegian Red List for Species (Kålås et al. 2010).
Generalised linear mixed-effects models (GLMM) analysis
To test the explanatory variables against each of the five species richness measures chosen as
response variables (see Introduction), and find the best models explaining these richnesses, I
used a generalised linear mixed-effects models (GLMM) analysis. It was also used to test the
richness of non-Red Listed common, occasional and rare species as well as the richness of all
non-Red Listed species against the richness of Red Listed species. As the response variables
were discrete, I applied a Poisson distribution. Site (n = 6) and site pair (n = 3) were included
as a nested random effect – i.e. site nested in site pair – to control for repeated measurements
within each site and site pair, and also to control for possible variation associated with each
site and site pair, respectively. Polypore hymenophore area was log10-transformed due to great
variation in the variable values.
To find the best models explaining the species richness groups studied, I tested every
possible combination of all explanatory variables against each response variable and ordered
the models according to their goodness of fit, as measured by the Akaike information criterion
(AIC). The models with the lowest AIC were considered to have the highest goodness of fit.
Applying a parsimonious approach (Burnham and Anderson 1998; 2002), I then added 2.0
AIC units to all models including more explanatory variables than one, and regarded the
model with the lowest AIC thus yielded as the best. In the cases where the best model
constituted a set of several explanatory variables, I also tested for all possible interactions of
those.
When performing the model selection tests I could not include correlated explanatory
variables. From a selection of such I therefore selected the one which resulted in the lowest
AIC-value together with the response variable. All log size measures – length, DBH, volume
and surface area – were correlated except for length and DBH, why either those two or
volume or surface area were included in the same test. All three calcium content variables
were also correlated.
Due to the moderate size of the data set and the big number of explanatory variables, some
variable combinations yielded singular convergence, and were considered unreliable. All taxa
12
that were clearly separate, at the level of individual logs, were included in the analysis
(Appendix 1).
For the GLMM, I utilised the function lmer(), in the R package lme4 (Bates et al. 2011).
Non-metric multidimensional scaling (NMDS) analysis
To analyse the relationship between the species composition – regarded as all and Red Listed
species as well as rare, occasional and common species – and the properties of the logs and
their surroundings, I performed a two-dimensional non-metric multidimensional scaling
(NMDS) analysis. NMDS is considered a good method to apply when analysing the effects of
many and possibly intercorrelated variables (Oksanen et al. 2012) on species communities. In
contrast to GLMM it takes the individual species identities into account.
I localised the logs in the ordination space according to their pairwise dissimilarities in
species composition. For this I used the Jaccard dissimilarity index (Oksanen et al. 2012), as it
was the one that performed the best when different indices were compared. I then plotted the
species on the average distance of their host trees' locations. I square-root transformed
polypore hymenophore area, due to great variation in the values of this variable. Subsequently
I fitted it and the other 16 explanatory variables (Table 2), site and site pair – in order to
control for possible variation they could be associated with – and the richness of Red Listed
and non-Red Listed species in the ordination space in the way that provided the maximum
correlation between each of them and the log scores (Oksanen et al. 2012). I then used a
permutation test with 1000 permutations to find out whether the fitted variables were non-
randomly distributed in the ordination space.
In the NMDS plot I decided to use isoclines to show the probability of a species being Red
Listed in different parts of the ordination space. As the real threat situation for species
belonging to the category DD is very unclear (they would with better knowledge be assigned
to any of the categories LC, NT, VU, EN or CR; Artsdatabanken 2009) I drew the isoclines
both including and excluding them from the Red Listed species.
All taxa that were clearly separate, at the level of the data set, were included (Appendix 1).
Hence the NMDS analysis spanned five species less than the GLMM.
For the NMDS, I utilised the R package vegan (Oksanen et al. 2012).
All statistical analyses and the volume calculations were performed in the statistical
programming environment R, version 2.15.0 (R Development Core Team 2012).
13
Estimate AIC
All species
Intercept 2.344 0.147 15.918 < 2e-16 48.54
Surface area 0.061 0.017 3.604 0.0003
Intercept 2.068 0.208 9.937 <2e-16 48.89
DBH 0.028 0.007 3.838 0.0001
Intercept 2.524 0.105 23.960 <2e-16 49.85
Volume 0.645 0.190 3.405 0.0007
Intercept 2.363 0.174 13.571 <2e-16 52.30
Length 0.031 0.011 2.945 0.0032
Intercept 2.890 0.054 53.760 <2e-16 55.74
Type (uprooted) -0.187 0.083 -2.240 0.0250
Intercept 2.772 0.066 41.820 <2e-16 57.85
Site size (small) 0.138 0.070 1.980 0.0477
Redlisted species
Intercept -1.302 0.621 -2.098 0.0359 59.49
Non-Red Listed species 0.081 0.035 2.341 0.0192
Intercept 0.288 0.200 1.441 0.1495 60.08
Decay class (3) -0.614 0.299 -2.056 0.0398
Common species
Intercept 2.128 0.170 12.522 <2e-16 49.17
Surface area 0.046 0.020 2.286 0.0223
Intercept 2.251 0.119 18.866 <2e-16 49.25
Volume 0.507 0.223 2.273 0.0230
Intercept 1.886 0.247 7.640 2.17e-14 57.87
DBH 0.022 0.009 2.549 0.0108
Occasional species
Intercept -1.421 0.513 -2.771 0.0056 61.65
Surface area 0.209 0.055 3.779 0.0002
Intercept -0.741 0.362 -2.046 0.0408 63.97
Volume 2.117 0.609 3.475 0.0005
Intercept -1.984 0.747 -2.654 0.0079 64.69
DBH 0.081 0.025 3.287 0.0010
Intercept -1.373 0.565 -2.430 0.0151 65.97
Length 0.109 0.033 3.316 0.0009
Intercept -0.250 0.341 -0.731 0.4647 71.13
Polypore hymenophore area 0.115 0.057 2.032 0.0422
Standard
errorZ value P value
Table 3: The variables significantly explaining the richness of all, Red Listed, common and occasional species.
The richness of rare species was not explained by any variable and is hence not included.
14
Results
General
The 48 logs of the survey yielded a total of 827 species occurrences (Appendix 1). These were
identified to146 different taxa; 136 officially described and ten undescribed species. The
species (or rather taxon) richness varied between 7 and 31 per log with a mean of 17.2 ± 4.5
(S.D). Twenty-two of the species were Red Listed, with a species richness varying between 0
and 4 per log and a mean of 1.1 ± 1.1 (S.D.). Red Listed species occurred only in the
frequency groups rare and occasional.
Determinants of species richness
The richness of all, common and occasional species were positively correlated with log size
(Table 3): they all increased with the size measures DBH, surface area and volume, and the
richness of all and occasional species also with length. The richness of all species was also
correlated with log type and site size, broken logs and logs at small sites having significantly
more species than uprooted logs and logs at large sites, respectively. The number of
occasional species increased with the area covered by polypore hymenophores.
There were more Red Listed species on logs in decay class 4 than on logs in decay class 3
(Table 3). The richness of Red Listed species also increased with the richness of non-Red
Listed species (Fig. 2).
6 8 10 12 14 16 18 20 22 24 26
0
1
2
3
4
Richness of non-Red Listed species
(species/log)
Ric
hn
ess
of
Re
d L
iste
d s
pe
cie
s
(sp
ecie
s/lo
g)
Figure 2: The richness of Red Listed species in relation to the richness of non-Red Listed
species. GLMM prediction (Table 3) shown as a line.
15
Estimate AIC
All species
Intercept 1.891 0.243 7.789 6.77e-15 43.07
Length 0.025 0.011 2.352 0.0187
DBH 0.022 0.007 2.989 0.0028
Type (uprooted) -0.192 0.084 -2.292 0.0219
Redlisted species
Intercept 0.562 0.192 2.930 0.0034 55.04
Bark -2.592 1.119 -2.316 0.0206
Decay class (3) -0.512 0.298 -1.721 0.0852
Standard
errorZ value P value
The most parsimonious models explaining the species richness in the different species
categories were quite consistent with the significance of the single variables (Tables 3 and 4);
the richness of all species was best predicted by length+DBH+log type (Fig. 3). It increased
with DBH and length, and broken logs had more species than uprooted logs. The most
parsimonious models for common (Fig. 4) and occasional (Fig. 5) species included only DBH
and surface area respectively. Albeit the number of Red Listed species present on a log could
not be significantly predicted by bark coverage alone –due to singular convergence (see
Materials and methods) – the combination bark coverage+decay class turned out to provide
the best model for it (Fig. 6). The model showed that the richness of Red Listed species
increased with decreasing bark coverage and that there were more species on logs in decay
class four than in decay class three.
The richness of rare species was not explained by any explanatory variable and the most
parsimonious model was non-significant. The ten most parsimonious models of each analysed
species group are listed in Appendix 2.
Table 4: The most parsimonious models predicting the richness of all and Red Listed species (Figs. 3 and 6
respectively). The richness of common and occasional species were most parsimoniously predicted by single
variables, and are thence presented in Table 3.
16
Figure 3: The most parsimonious model predicting the richness of all species
on broken (a) and uprooted (b) logs (Table 4).
17
20 22 24 26 28 30 32 34 36 38
0
5
10
15
20
25
DBH (cm)
Ric
hn
ess
of
co
mm
on
sp
ecie
s
(sp
ecie
s/lo
g)
4.5 5.5 6.5 7.5 8.5 9.5 10.5 11.5 12.5
0
1
2
3
4
Ric
hn
ess
of
occa
son
al
spe
cie
s
(sp
ecie
s/lo
g)
Figure 4: The most parsimonious model describing the richness of common species. GLMM
prediction (Table 3) shown as a line.
Figure 5: The most parsimonious model describing the richness of occasional species. GLMM
prediction (Table 3) shown as a line.
Surface area (m2)
18
Ordination of species composition and correlated variables
Six of the 17 explanatory variables tested (Table 2) and the richness of Red Listed species
correlated with the ordination configuration of the NMDS (Table 5; Fig. 7); ground contact
and plant coverage showed the strongest correlations, followed by richness of Red Listed
species, bark coverage, log type, polypore hymenophore area and decay class.
Logs in decay class 4 had more ground contact and plant coverage but less area covered by
polypore hymenophores and bark than logs in decay class 3 (Fig. 7). Uprooted logs had more
bark coverage than broken logs.
The number of Red Listed species per log was positively correlated with ground contact
and negatively correlated with bark coverage (Fig. 7). There were more Red Listed species
per log on broken than on uprooted logs.
The Red Listed species were concentrated to logs in decay class 4, with much ground
contact and plant coverage but little bark coverage and polypore hymenophore area (Fig. 7).
More precisely, the direction of increasing probability of a species being Red Listed as NT,
VU and EN was very similar to the direction of increasing plant coverage (Fig 7a). This
probability increased also with ground contact, decreasing polypore hymenophore area and
was higher for decay class 4 than 3. The species Red Listed as Data Deficient turned out to
respond differently than the other Red Listed species (Fig. 7b).
0 10 20 30 40 50 60 70 80 90
0
1
2
3
4
5
Bark coverage (%)
Ric
hn
ess
of
Re
d L
iste
d s
pe
cie
s
(sp
ecie
s/lo
g)
Figure 6: The most parsimonious model predicting the richness of Red Listed species on
logs in decay class 3 (blue) and 4 (green). GLMM predictions (Table 4) shown as lines.
19
Though distributed to a great variation of log characteristics, as measured by the significant
variables of the ordination, the rare species displayed the same pattern as the Red Listed
species (Fig. 7); they were concentrated to logs in decay class 4, with much ground contact
and plant coverage but little bark coverage and polypore hymenophore area. A number of rare
species also seemed to prefer logs with a high richness of Red Listed species.
Common species were overrepresented on logs with little ground contact and plant
coverage (Fig. 7) but much bark coverage and polypore hymenophore area, and on logs with
intermediate values on these variables.
Variable P(>r)
Ground contact 0.3957 <0.001
Plant coverage 0.3718 <0.001
Log type 0.1241 0.003
0.2153 0.006
Decay class 0.0899 0.011
Bark coverage 0.1563 0.017
0.1205 0.043
0.1060 0.071
Site size 0.0523 0.087
DBH 0.0663 0.229
Location 0.0535 0.291
0.0523 0.313
Site 0.1242 0.316
Calcium, coarse scale 0.0492 0.332
Volume 0.0288 0.518
Top intactness 0.0137 0.548
Surface area 0.0249 0.559
0.0113 0.587
Calcium, fine scale 0.0673 0.645
Time 0.0099 0.813
Length 0.0048 0.894
r2
Richness of Red Listed species
Polypore hymenophore area
Richness of non-Red Listed species
Number of vegetation types
Presence of richer vegetation types
Table 5: The coefficients of determination and
empirical p-values assessing the goodness of fit and
significance of the explanatory variables (Table 2), the variables
site and site pair and the richness of Red Listed and non-Red
Listed species fitted onto the NMDS ordination space (Fig. 7).
20
a) b)
Figure 7: Two-dimensional NMDS plots of the studied taxa, variables and logs significantly
correlating with the ordination configuration (Table 5). The symbols and their colors
indicate frequency and Red List status, respectively, of the taxa. Isoclines show how the
probability of a species being Red Listed in all categories, excluding (a) or including (b)
DD, changes in the ordination space. Explanatory variables and the richnesses of Red Listed
(RL) species are shown in black; continuous variables as vectors and categorical variables
by their centroids.
21
Discussion
Basing an inventory of corticioid fungi on fruit bodies collected during just one season may
be problematic. Halme and Kotiaho (2012) concluded that surveys of corticioid fungi should
preferably be performed at least at three consecutive occasions, in order to be give a good
representation of the funga of the studied logs. However, even with fruit body data from only
one season I think that some of the results are quite clear and worthy of discussion.
The mean number of species per log (17.2) was higher in my study than in previous studies
of wood-inhabiting fungi on Norway spruce in the taiga region (Renvall 1995; Lindblad 1998;
Edman et al. 2004; Berglund et al. 2005; Juutilainen et al. 2011; Olsson et al. 2011; Stokland
and Larsson 2011). The highest number occurring is that of Lindblad's study (1998), which
states a mean richness of less than 11species per log, on logs in decay class 2 in a natural
forest. In order to make a correct comparison of results two things may have to be taken into
account though: the delimitation of the surveyed decay classes and the size of the studied
logs; my study only included logs in decay class 3 and 4, 20-40 cm in DBH and at least 11 m
long. None of the studies initially mentioned in this paragraph have included logs with both
their lower limits for log size set as high as mine and few of them have presented separate
results for decay class 3 and 4. The most similar studies, however, seems to be Renvall (1995)
and Stokland and Larsson (2011). Renvall (1995) surveyed logs longer than 1.5 m, with a
base diameter exceeding 10 cm, and displayed separate results for decay class 3 and 4. He
found the logs of both classes to have a mean richness of slightly more than 4 species per log.
Stokland and Larsson (2011) surveyed logs longer than 60 cm, with a maximum diameter
larger than 10 cm and displayed separate results based on both decay class and diameter. In
decay class 3 their logs had a mean species richness of 7.9 and 9.1 species per log, for logs
with a maximum diameter between 20 and 30 cm, and more than 30 cm, respectively.
Corresponding values for logs in decay class 4 and 5 – which were combined – were 6.3 and
7.9 species per log. Whether the cause of the high mean species richness I acquired is the high
quality of the sample plots, the large size of the logs or the thorough survey remains unclear,
but it is evidently clear that coarse woody debris (CWD) can harbour more corticioid species
per log than what has previously been shown - a matter perhaps not very surprising given the
small amount of substrate many corticioid species seem to cope with (Nordén et al. 2004;
Juutilainen et al. 2011).
As expected, larger logs had a greater total species richness and held more common and
occasional species than smaller logs. Also Renvall (1995), Høiland and Bendiksen (1997),
22
Lindblad (1998) and Stokland and Larsson (2011) found a positive correlation between the
size of investigated resource units and the number of wood-inhabiting fungi fruiting on them.
More interestingly, broken logs had a greater total species richness, held more Red Listed
species and had a different species composition than uprooted logs. Renvall (1995), found
broken logs of spruce in northern Finland to hold more and different wood-inhabiting fungi
than uprooted logs, and Heilmann-Clausen and Christensen (2003), made the same result for
Red Listed wood-inhabiting fungi on beech in Denmark. Both studies ascribed their results to
Renvall's (1995) empirically supported theory of successional pathways (see Introduction),
i.e. believed the observed differences in species richness and composition between uprooted
and broken logs, to be caused by the outcome of separate and distinct chains of species
interactions, initially determined by a difference in the funga of primary decayers that
colonises the two log types. Mine and Heilmann-Clausen and Christensen's (2003) results
concerning log type could perhaps be of interest from a conservational point of view: if a
greater percentage of the trees in small areas, compared to large areas, become uprooted as a
result of heavier impact by strong wind on small areas, then these areas would suffer a greater
loss of Red Listed species – at least when measured as species per log.
Uprooted logs had more bark coverage than broken logs, and logs with more bark coverage
held less Red Listed species and had a different species composition than logs with less bark
coverage. The only earlier study which has found a relationship between any of these
variables made a result contradictory to mine: Lindblad (1998) discovered a positive
correlation between total species richness of wood-inhabiting fungi and bark coverage. A
theory to why the uprooted logs had more bark coverage than the broken logs is that the
former probably died and fell healthy, with all their bark still on, whereas the latter possibly
got broken because they were dead or dying and had parts of their bark removed already
while they were still standing.
Logs in decay class 4 held more Red Listed species and had a different species
composition than logs in decay class 3. A number of studies have shown that Red Listed
corticioid fungi exhibit species-specific differences in their use of decay classes (Renvall
1995; Høiland and Bendiksen 1997; Lindblad 1998; Stokland and Larsson 2011) but none
have shown this entire group of fungi to vary in species richness between decay classes.
Tikkanen et al. (2006), however, found boreal, Red Listed (following Rassi et al. 2001),
aphyllophoraceous fungi in Finland to most often utilise logs in decay class 3 as their primary
substrate, on a scale from 1 to 5, with addition of the class kelo (wood dry and hard, bark
lost). The second most employed logs used as primary substrate belonged to decay class 4,
23
followed by classes 2, kelo and 1. No species was found to primarily live on logs in decay
class 5. The difference between mine and Tikkanen's et al (2006) results could be caused by
many factors, but dissimilar inclusiveness of data (aphyllophoraceous fungi inhabiting logs of
boreal tree species vs. corticioid fungi inhabiting spruce logs), geographical areas or methods
(different decay classification) are probably the most likely candidates.
Logs at small sites held more species than logs at large sites (marginally significant). The
direction of this correlation is surprising but my data include only three site pairs and the
effect of chance or variables not included might therefore be considerable. According to
previous studies dead wood amount (Lonsdale et al. 2008; Junninen and Komonen 2011;
Stokland and Larsson 2011), continuity (Lonsdale et al. 2008; Junninen and Komonen 2011)
and connectivity (Jönsson et al. 2008; Lonsdale et al. 2008; Junninen and Komonen 2011;
Nordén et al. 2012; Norros et al. 2012) have all proved to affect species richness of wood-
inhabiting fungi positively. However, viewed more in general, small sites with more edges
(ecotones) can also be rich in species (Leopold 1933; Odum 1983; Walker et al. 2003; Senft
2009).
Logs with more polypore hymenophore area had a greater richness of occasional species
and a different species composition than logs with less polypore hymenophore area. A
possible explanation may be that these corticioid species are specialised in degrading wood
partially decomposed by certain polypores, or live of the living or dead polypore mycelia. For
example, some species of Sistotrema Fr., seem to favour growing on dead polypores (K.-H.
Larsson 2012, pers.comm.).
Logs with more ground contact held more Red Listed species and had a different species
composition than logs with less ground contact. Lindblad (1998) reported ground contact to
be the variable the strongest correlated with species richness of wood-inhabiting fungi on the
spruce logs of her study. Heilmann-Clausen and Christensen (2003) found also the species
richness of wood-inhabiting fungi on beech logs to increase with ground contact. Lindblad
(1998) argued that it was the increased moisture content of a log when it comes in contact
with ground, as found by Harmon et al. (1986), that caused the increase in species richness.
The moisture content of logs have indeed been found to be directly correlated with the wood-
inhabiting funga, but rather in the form of species composition (Boddy et al. 1989), not
richness. An alternative hypothesis was suggested by Heilmann-Clausen and Christensen
(2003): "Logs with a high degree of soil contact are likely to be buffered against fluctuations
in temperature and especially water content compared to logs with little soil contact". Yet
24
another idea might be to search for the cause in the characteristics of the species, not the log; a
number of corticioid species, such as several of those belonging to the genera Piloderma
(Eriksson et al. 1981), Tomentella (Kõljalg et al. 2000), Amphinema P. Karst. (Larsson et al.
2004) and Tylospora Donk (Larsson et al. 2004), have been proved to be ectomycorrhizal. For
their presence to impact on the species richness and composition of a log it must needs have
contact with ground, and for the same reason their numbers will most probably also increase
with the amount of contact – at least up to a certain point.
Logs with more plant coverage had a different species composition than logs with less
plant coverage. The only study which has discovered any aspects of the wood-inhabiting
funga to be correlated with this variable before seems to be that of Lindblad (1998), who
found plant coverage to be positively correlated with species richness on her spruce logs.
Heilmann-Clausen and Christensen (2003) did not find a correlation between moss coverage
and species richness of wood-inhabiting fungi on beech. These authors did not, however,
analyse species composition in this respect. Plant coverage was positively correlated to
ground contact, and perhaps it may be that it is correlated with the moisture content and
stability of the logs.
Logs with more Red Listed species held more non-Red Listed species and had a different
species composition than logs with less Red Listed species. Berglund and Jonsson (2003)
showed that corticioid fungi on spruce exhibit nestedness, i.e. species in species-poor sites
comprise a non-random subset of the species pool in richer ones. Rare species consequently
have a tendency to occur only at the most species rich sites. The species communities of
single logs do not seem to have been analysed for patterns of nestedness in any study, but as
18 of the 22 Red Listed species in my study were classified as rare (Appendix 1), the
existence of such a relationship among corticioids would probably explain my result. It would
then also be possible to identify indicator species of species rich logs or sites.
The most parsimonious models predicting the richness of common and occasional species
included DBH and surface area, respectively. No previous studies seem to have made separate
models for species of different frequency categories and it is difficult to compare also the
models of total species richness and richness of Red Listed species to those of others, since
the set of explanatory variables included in the analyses selecting a model varies considerably
between studies. I have found two comparable studies, both of which presented models that
were more similar to each other than to mine. The best model of Lindblad (1998) – which
explained the total richness of wood-inhabiting fungi on spruce logs – included ground
contact, decay class, log length and penetration value (as obtained from a penetrometer
25
measuring log hardness). The best model of Heilmann-Clausen and Christensen (2003) –
which explained the total richness of wood-inhabiting fungi on beech – comprised ground
contact, log age, log complexity (a measure of the number of bole forks and branches thicker
than 50 cm) and ground coverage of Anemone nemorosa L. The study of Heilmann-Clausen
and Christensen (2003) included a total of 19 explanatory variables and Lindblad's study
(1998) 17. Both studies had seven variables in common with mine, and in both cases these
comprised all but one of the variables included in my best models. Heilmann-Clausen and
Christensen's (2003) study also featured a model of the richness of Red Listed species but as it
included the variables log age, log complexity and log type, it did not have much in common
with mine. Worth noticing is that my measure of decay class is not entirely comparable to
those of the other studies mentioned, since it only includes two classes. With so big
differences in both our sets of explanatory and response variables – corticioid vs. wood
inhabiting fungi – it is hard to draw any further conclusions than that there is little consistency
in the current models of species richness of wood-inhabiting and corticioid fungi.
Logs in decay class 4 had more ground contact and plant coverage but less area covered by
polypore hymenophores and bark than logs in decay class 3. These correlations can be
explained by the physical changes taking place in a log when it passes from decay class 3 to
4; it loses more bark, the amount of area covered by polypore fruit bodies on it decreases and
a greater proportion of it settles to the ground, where it gets more moist (Harmon et al.1986)
and covered by plants.
Similarly, rare and Red Listed corticioid species were concentrated to logs in decay class 4
– with much ground contact and plant coverage but little bark coverage and polypore
hymenophore area. A possible explanation to why rare and Red Listed species were found to
prefer this substrate might be derived from studying the occurrence frequency of it. According
to Kruys's et al. (1999) study from northern Sweden, the amount of dead wood of spruce in
managed forests drastically decrease with decay, between each of the medium and late decay
classes. Including all pieces exceeding 5 cm in diameter and 30 cm in length, they found only
1.6 pieces of dead wood in late decay classes per ha, accounting for 3.4 % of all the pieces
found. In Norway managed forests are reported to cover 92-95 % of the total forest area
(Stokland et al. 2003) and hold on average 10 m3 dead wood per ha (Larsson and Hylen
2007), while mesic, spruce-dominated, old-growth forests in the southern and middle boreal
zones of Fennoscandia have been found to hold 90-120 m3 dead spruce wood per ha (Siitonen
2001). Perhaps a greater decline of wood in decay class 4 than in decay class 3 has caused the
26
species which prefer decay class 4 to become rare and Red Listed to a higher extent than the
species which prefer decay class 3.
The richness of common species only increased with different measures of log size and the
richness of occasional species increased with both log size and polypore hymenophore area,
while the richness Red Listed species did not change with any measure of log size but instead
correlated with four different explanatory variables related to the quality of the logs. Red
Listed corticioid species seem to be limited by the quality rather than the size of dead wood
substrates. This is interesting and has not been shown in any previous study.
The fact that the Data Deficient species behaved differently than the other Red-Listed
species in my NMDS-configuration is something I think is worth attention and which I hope
will make others be more cautionary about the juxtapositioned nature of these species (see
Materials and methods); if one aims to draw any conclusions about species threatened by
extinction, due to rareness, population decline or a combination thereof, I think one should
make very sure that that indeed is what one has been studying. I thoroughly regret the
inclusion of Data Deficient species in the Red-listed species in the GLMM analysis of my
study, as it surely added an unnecessary amount of error to it.
Further results could have been extracted from my analyses by comparing the occurrence
patterns of individual species or groups of species. This was not done merely due to lack of
time.
Lastly I would like to put words to a question, and the subsequent personal view derived
from it, awoken by the finding of species like Repetobasidium vestitum J. Erikss. & Hjortstam
(Appendix 1) – a corticioid whose known distribution in the world is limited to a few, widely
separated logs in northern Europe (Eriksson et al. 1981) – and developed through
conversations with my supervisor Karl-Henrik Larsson: how do the very rare corticioid
species, that occur only at one or a few logs in the small number of forest where they are
found, manage to survive? It strikes me as unreasonable that their occurrence could be traced
only to very specific demands on general environmental variables, such as the ones analysed
in this study; no niche can be that immensely rare, and even if it were: how would a sufficient
amount of spores be able to travel between across so widely separated substrates?
27
Conclusions and implications for conservation
This study has that even within a single, narrowly defined kind of dead wood, a great
variation in microenvironments exists, and that many of these environments host differently
composited and species rich communities of corticioid fungi. Numerous questions also remain
to be answered concerning species diversity and coexistence in dead wood. The number of
species found on logs in this study was clearly higher than what has previously been reported
in any studies of similar substrates, further underscoring the importance dead wood has for
biodiversity.
The rare species found in this study needed many different kinds of logs, and hence a
diversity of dead wood is required to host them. Both the rare and the Red Listed species
needed well decayed logs, a substrate that today can almost only be found in the most natural
forests. Large areas of natural spruce forest have to be protected to harbour also the rarest
corticioid species.
Acknowledgements
First and foremost I want to thank my three supervisors! Jenni Nordén, I am very grateful for
your enthusiasm, for our happy moments in the forest and by the microscope and for the great
amount of work you have laid in me and my project all along. Karl-Henrik Larsson, I am
most thankful for the support you have lent me to structure my work, for our conversations of
interesting things great and small and for being my mentor in the part I liked the best: the
understanding and identification of corticioid fungi. Björn Nordén, I thank you for your
optimism and your solutions, for our interesting discussions and generally for seeing me
through this. I also want to very much thank Ronny Steen: without your help I would have
spent considerably more hours with the statistics, a discipline you slowly made start to like
and begin to understand. I want to thank Johan Rydlöv, Vidar Brodin, Anna Norberg and
Anders Aas for help and support during my field work. I am grateful to Anders Grönberg for
his good grammatical aid and to my mother for the fine map. Last, but not least, I want to
thank my friend Olli Manninen, without whose initial inspiration this would not have come to
pass.
The subsistence allowance during the field work of my thesis was partly funded by the P.
A. Larsson foundation. For this I am very thankful.
28
References
Artsdatabanken. 2009. Norsk rødliste 2010 – Veileder til rødlistevurdering, versjon 2.1. Artsdatabanken,
Trondheim.
Artsdatabanken. 2012. Norwegian Species Nomenclature Database, version 180311 250412, accessed
30 April 2012.
Bader P., Jansson S. and Jonsson B. G. 1995.Wood-inhabiting fungi and substratum decline in selectively
logged boreal spruce forests. Biological Conservation. 72: 355-362.
Bates D., Maechler M. and Bolker B. 2011. lme4: Linear mixed-effects models using S4 classes, R package
version 0.999375-42.
Berglund H. and Jonsson B. G. 2003. Nested plant and fungal communities; the importance of area and habitat
quality in maximizing species capture in boreal old-growth forests. Biological Conservation. 112: 319-328.
Berglund H., Edman M., and Ericson L. 2005. Temporal variation of wood-fungi diversity in boreal old-growth
forests: implications for monitoring. Ecological Applications. 15: 970-982.
Boddy L., Owens E. M. and Chapela I.H. 1989. Small scale variation in decay rate within logs one year after
felling: Effect of fungal community structure and moisture content. FEMS Microbiology Letters. 62: 173-
183.
Burnham K. P. and Anderson D. R. 1998. Model selection and inference: a practical
information-theoretic approach. - Springer, New York. 353 pp.
Burnham K. P. and Anderson D. R. 2002. Model selection and multimodel inference: a practical information-
theoretic approach, 2nd edition. - Springer, New York. 488 pp.
Dahl E., Elven R., Moen A. and Skogen A. 1986. Vegetasjonskart over Norge 1 : l 500 000, Nasjonalatlas for
Norge. - Statens kartverk, Hønefoss.
Direktoratet for naturforvaltning. 2007. Kartlegging av naturtyper - Verdisetting av biologisk mangfold.
DN-håndbok 13 2. utgave. - Direktoratet for naturforvaltning, Trondheim. 254 pp.
Edman M., Kruys N. and Jonsson B. G. 2004. Local dispersal sources strongly affect colonization patterns of
wood-decaying fungi on spruce logs. Ecological Applications. 14: 893-901.
Eriksson J. and Ryvarden L. 1973.The Corticiaceae of North Europe, Vol. 2, Pp: 59-286. - Fungiflora, Oslo.
29
Eriksson J., Hjortstam K. and Ryvarden L. 1981. The Corticiaceae of North Europe, Vol. 6, Pp: 1048-1246. -
Fungiflora, Oslo.
Fremstad E. 1997. Vegetasjonstyper i Norge. NINA temahefte. 12: 1-279.
Gjerde I., Brandrud T. E., Ohlson M. and Ødegaard F. 2009. Skoglandskapet – Miljøforhold og påvirkninger på
rødlistearter. Artsdatabanken, Trondheim.
Gustafsson L., and Ahlén I. 1996. Sveriges Nationalatlas, växter och djur, vegetationszoner i Sverige. - Sveriges
Nationalatlas, Stockholm. 160 pp.
Halme P. and Kotiaho J. S. 2012. The importance of timing and number of surveys in fungal biodiversity
research. Biodiversity and Conservation. 21: 205-219.
Halvorsen R., Andersen T., Blom H. H., Elvebakk A., Elven R., Erikstad L., Gaarder G., Moen A., Mortensen
P. B., Norderhaug A., Nygaard K., Thorsnes T. and Ødegaard F. 2009. Naturtyper i Norge (NiN), version
1.0.0. - Artsdatabanken, Trondheim. , accessed 27 April 2012.
Harmon M. E., Franklin J. F., Swanson F. J., Sollins P., Gregory S. V., Lattin J. D., Anderson N. H., Cline S. P.,
Aumen N. G., Sedell J. R., Lienkaemper G. W., Jr. Cromack K. and Cummins K. W. 1986. Ecology of
coarse woody debris in temperate ecosystems. Advances in Ecological Research. 15:133-302.
Heilmann-Clausen J. and Christensen M. 2003. Fungal diversity on decaying beech logs – implications for
sustainable forestry. Biodiversity and Conservation. 12: 953-973.
Heilmann-Clausen J. and Christensen M. 2004. Does size matter? On the importance of various dead wood
fractions for fungal diversity in Danish beech forests.Forest Ecology and Management. 201: 105-117.
Hjortstam K., Larsson K.-H. and Ryvarden L. 1988. The Corticiaceae of North Europe, Vol. 1, pp. 1-58. -
Fungiflora,Oslo.
Høiland K. and Bendiksen E. 1997. Biodiversity of wood-inhabiting fungi in a boreal coniferous forest in Sør-
Trøndelag County, Central Norway. Nordic Journal of Botany. 16: 643-659.
Hofton T. H. 2010. Corticiaceae assessments for the 2010 Norwegian Red List, xl-sheet. BioFokus, Oslo.
Unpublished.
Hottola J., Siitonen J. 2008. Significance of woodland key habitats for polypore diversity and red-listed species
in boreal forests. Biodiversity and Conservation. 17: 2559-2577.
30
Jönsson M. T., Edman M. and Jonsson B. G. 2008.Colonization and extinction patterns of wood-decaying fungi
in a boreal old-growth Piceaabies forest. Journal of Ecology. 96: 1065-1075.
Junninen K. and Komonen A. 2011. Conservation ecology of boreal polypores: A review. Biological
Conservation. 144: 11-20.
Juutilainen K., Halme P., Kotiranta H. and Mönkkönen M. 2011. Size matters in studies of dead wood and wood-
inhabiting fungi. Fungal Ecology. 4: 342-349.
Kålås J.A., Viken Å., Henriksen S. and Skjelseth, S. (eds.). 2010. The 2010 Norwegian Red List for Species. -
Norwegian Biodiversity Information Centre, Norway. 480 pp.
Kõljalg U., Dahlberg A., Taylor A. F. S., Larsson E., Hallenberg N., Stenlid J., Larsson K. H., Fransson P. M.,
Kårén O. and Jonsson L. 2000. Diversity and abundance of resupinate thelephoroid fungi as ectomycorrhizal
symbionts in Swedish boreal forests. Molecular Ecology. 9: 1985-1996.
Küffer N., Gillet F., Senn-Irlet B., Aragno M. and Job D. 2008. Ecological determinants of fungal diversity on
dead wood in European forests.Fungal Diversity. 30: 83-95.
Küffer N. and Senn-Irlet B. 2005. Diversity and ecology of wood-inhabiting aphyllophoroid basidiomycetes on
fallen woody debris in various forest types in Switzerland. Mycological Progress. 4: 77-86.
Kruys N., Fries C., Jonsson B. G., Lämås T. and Ståhl G. 1999. Wood-inhabiting cryptogams on dead Norway
spruce (Picea abies) trees in managed Swedish boreal forests. Canadian Journal of Forest Research. 29: 178-
186.
Laasasenaho J. 1982. Taper curve and volume function for pine, spruce and birch. Communicationes Instituti
Forestalis Fenniae. 108: 1–74.
Larsson J. Y. and Hylen G. 2007. Skogen i Norge. Statistikk over skogforhold og skogressurser i Norge
registrert i perioden 2000 – 2004. Viten fra Skog og landskap. 1/07. - Norsk institutt for skog og landskap,
Ås. 91 pp.
Larsson K.-H., Larsson E. and Kõljalg U. 2004. High phylogenetic diversity among corticioid
homobasidiomycetes. Mycological Research. 108: 983-1002.
Leopold, A. 1933. Game management. - Charles Scribner’s Sons, New York. 481 pp.
31
Lindblad I. 1998. Wood-inhabiting fungi on fallen logs of Norway spruce: relations to forest management and
substrate quality. Nordic Journal of Botany. 18: 243-255.
Lindhe A., Asenblad N., Toresson H. G. 2004. Cut logs and high stumps of spruce, birch, aspen and oak – nine
years of saproxylic fungi succession. Biological Conservation. 119: 443-454.
Lonsdale D., Pautasso M. and Holdenrieder O. 2008. Wood-decaying fungi in the forest: conservation needs and
management options. European Journal of Forest Research. 127: 1-22.
Mahmood S., Finlay R. D., Erland S and Wallander H. 2001. Solubilisation and colonisation of wood ash by
ectomycorrhizal fungi isolated from a wood ash fertilised spruce forest. FEMS Microbiology Ecology. 35:
151-161.
Moen, A. 1999.National Atlas of Norway: Vegetation. - Norwegian Mapping Authority, Hønefoss. 200 pp.
Nordén B., Ryberg M., Götmark F., Olausson B. 2004. Relative importance of coarse and fine woody debris for
the diversity of wood-inhabiting fungi in temperate broadleaf forests. Biological Conservation.117: 1-10.
Nordén J., Penttilä R., Siitonen J., Tomppo E. and Ovaskainen O. 2012. Specialist species of wood-inhabiting
fungi struggle while generalists thrive in fragmented boreal forests. Submitted paper.
Norros V., Penttilä R., Suominen M. and Ovaskainen O. 2012. Dispersal may limit the occurrence of specialist
wood decay fungi already at small spatial scales. Oikos. 121: 961-974.
Norwegian Directorate for Nature Management. N.d. Naturbase. , accessed 14 March 2012.
Odum, E. P. 1983. Basic ecology. - Saunders College Publishing, Philadelphia. 613 pp.
Oksanen J., Blanchet F. G., Kindt R., Legendre P., Minchin P. R., O'Hara R. B., Simpson G. L., Solymos P.,
Stevens M. H. H. and Wagner H. 2012. vegan: Community Ecology Package, R package version 2.0-3.
Olsson J., Jonsson B. G., Hjältén J. and Ericson L. 2011. Addition of coarse woody debris – The early fungal
succession on Picea abies logs in managed forests and reserves. Biological Conservation. 144: 1100-1110.
R Development Core Team. 2012. R: A language and environment for statistical computing, version 2.15.0. R
Foundation for Statistical Computing, Vienna.ISBN 3-900051-07-0, http://www.R-project.org/.
Rassi P., Alanen A., Kanerva T. and Mannerkoski I. (eds.). 2001. The 2000 Red List of Finnish species. -
Ympäristöministeriö & Suomen ympäristökäskus, Helsinki. 432 pp. In Finnish with English summary
32
Renvall P. 1995. Community structure and dynamics of wood-rotting basidiomycetes on decomposing conifer
trunks in northern Finland. Karstenia. 35: 1-51.
Senft A. R. 2009. Species diversity patterns at ecotones. Master's thesis. University of North Carolina.
Siitonen J. 2001. Forest management, coarse woody debris and saproxylic organisms: Fennoscandian boreal
forests as an example. Ecological Bulletins. 49: 11-41.
Stokland J. N., Eriksen R., Tomter S. M., Korhonen K., Tomppo E., Rajaniemi S., Söderström U., Toet H. and
Riis-Nielsen T. 2003. Forest biodiversity indicators in the Nordic countries – status based on national forest
inventories. TemaNord 514: 1-106.
Stokland J. N. and Kauserud H. 2004. Phellinus nigrolimitatus – a wood-decomposing fungus highly influenced
by forestry. Forest Ecology and Management. 187: 333-343.
Stokland J. N. and Larsson K.-H. 2011. Legacies from natural forest dynamics: Different effects of forest
management on wood-inhabiting fungi in pine and spruce forests. Forest Ecology and Management. 261:
1707–1721.
Stokland J. N. and Meyke E. 2008. The saproxylic database: an emerging overview of the biological diversity in
dead wood. Revue d’Ecologie (Terre Vie). 63: 29-40.
Swedish Forest Agency. N.d. Skogens Pärlor. , accessed 14 March 2012.
Tikkanen O.-P., Martikainen P., Hyvärinen E., Junninen K. and Kouki J. 2006. Redlisted boreal forest species of
Finland: associations with forest structure, tree species, and decaying wood. Annales Zoologici Fennici.
43: 373-383.
33
Taxon Frequency Red List status
Amphinema byssoides Common LC 32
Amyloathelia crassiuscula Rare LC 1
Amylocorticium cebennense Common LC 7
Amylocorticium pedunculatum Rare DD 1
Amylocorticium subincarnatum Rare EN 1
Amyloxenasma allantospora Rare NE 1
Asterodon ferruginosus Occasional LC 2
Asterostroma laxum Rare NE 3
Athelia bombacina Occasional LC 2
Athelia decipiens Common LC 12
Athelia epiphylla s.l. Common LC 17
Athelia neuhoffii Rare LC 1
Athelia singularis Rare Not Red Listed 3
Athelia subovata Rare NE 1
Athelicium stridii Rare LC 1
Athelopsis glaucina Occasional LC 2
Athelopsis subinconspicua Common LC 9
Boidinia furfuracea Common LC 9
Botryobasidium botryosum Common LC 7
Botryobasidium candicans Rare LC 9
Botryobasidium intertextum Common LC 13
Botryobasidium medium Rare DD 1
Botryobasidium obtusisporum Common LC 11
Botryobasidium indet. No category Not Red Listed 7 x
No category LC/NE 2
Botryobasidium subcoronatum Common LC 9
Botryohypochnus isabellinus Common LC 1
Byssocorticium pulchrum Rare LC 5
Calocera spp. No category Not Red Listed 3
Camarops tubulina No category VU 2
Ceraceomyces eludens Common LC 22
Ceraceomyces sp. 1 No category Not Red Listed 1
Ceraceomyces tessulatus Occasional LC 3
Numbers of
occurences
(logs)
Absence
from NMDS
analyses (x)
Botryobasidium vagum group
indet.
Appendix 1
A list of all taxa present in the GLMM analyses, i.e. in the study. To the right those absent
from the NMDS analyses are indicated. Taxa assigned with the Red List status Not Red Listed
are either species that had not been found in Norway when the Red List was made,
undescribed species or taxa including several species, of which at least one is non-Red Listed.
34
Cerinomyces crustulinus Occasional LC 4
Coniophora olivacea Common LC 6
Corticiaceae sp. 1 No category Not Red Listed 2
Corticiaceae sp. 2 No category Not Red Listed 1
Corticiaceae sp. 3 No category Not Red Listed 1
Dacrymyces spp. Common Not Red Listed 12
Fibricium lapponicum Rare VU 2
Globulicium hiemale Common LC 16
Henningsomyces candidus No category LC 1
Hymenochaete fuliginosa Common LC 4
Hyphoderma argillaceum Common LC 11
Hyphoderma capitatum Rare NT 2
Hyphoderma definitum Common LC 12
Hyphoderma occidentale Rare LC 1
Hyphoderma sibiricum Occasional LC 3
Hyphoderma velatum in herb. Rare Not Red Listed 4
Hyphodontia abieticola Occasional LC 3
Hyphodontia alutacea Common LC 13
Hyphodontia alutaria Common LC 1
Hyphodontia curvispora Rare VU 1
Hyphodontia hastata Common LC 4
Hyphodontia pallidula Common LC 12
Hyphodontia subalutacea Common LC 4
Hypochnicium c.f. subrigescens Rare NE 1
Hypochnicium lundellii Rare LC 1
Hypochnicium punctulatum Rare LC 1
Jaapia ochroleuca Occasional LC 14
Laxitextum bicolor Rare LC 1
Leptosporomyces galzinii Common LC 2
Leucogyrophana romellii Common LC 1
Litschauerella clematidis s.l. Rare NE 1
Lobulicium occultum Rare LC 4
Luellia furcata Rare NE 1
Membranomyces delectabilis Rare LC 2
Mucronella bresadolae No category DD 4
Mucronella calva No category LC 17
Mucronella indet. No category Not Red Listed 2 x
Paullicorticium allantosporum Rare NT 1
Paullicorticium ansatum Occasional NT 13
Paullicorticium pearsonii Rare LC 2
Peniophorella pallida Rare LC 1
Peniophorella praetermissa Common LC 17
Phanerochaete laevis Common LC 1
Phanerochaete sanguinea Common LC 1
35
Phlebia centrifuga Rare NT 1
Phlebia lilascens Rare NE 2
Phlebia livida Occasional LC 3
Phlebia segregata Common LC 11
Phlebia subcretacea Occasional LC 9
Phlebia subulata Rare VU 8
Phlebiella borealis Rare LC 1
Phlebiella christiansenii Rare DD 4
Phlebiella subflavidogrisea Rare NT 1
Phlebiella tulasnelloidea Occasional LC 1
Phlebiella vaga Common LC 19
Physodontia lundellii Rare VU 1
Piloderma byssinum Common LC 28
Piloderma fallax Common LC 8
Piloderma olivaceum No category NE 1
Piloderma sp. 1 No category Not Red Listed 1
Piloderma sp. 2 No category Not Red Listed 1
Piloderma sp. 3 No category Not Red Listed 1
No category Not Red Listed 1 x
Piloderma sphaerosporum Common LC 19
Protodontia piceicola No category VU 2
Pseudohydnum gelatinosum No category LC 3
Pseudoxenasma verrucisporum Rare LC 6
Repetobasidium vestitum Rare DD 1
Resinicium bicolor Common LC 13
Resinicium furfuraceum Common LC 4
Scytinostroma odoratum Rare LC 1
Sebacina calcea No category LC 2
Serpula himantioides Rare LC 2
Sidera lunata Occasional LC 1
Sistotrema brinkmannii Common LC 1
Sistotrema c.f. alboluteum Rare NT 1
Sistotrema citriforme Rare VU 1
Sistotrema diademiferum Rare LC 1
Sistotrema muscicola Occasional LC 2
Sistotrema pistilliferum Rare DD 1
Sistotremastrum suecicum Common LC 1
Sistotremella perpusilla Rare LC 5
Sphaerobasidium minutum Common LC 11
Sphaerobolus stellatus No category LC 1
Stypella vermiformis No category NE 12
Thanatephorus fusisporus s.l. Occasional LC/NE 1
Thelephoraceae spp. Common Not Red Listed 21
Piloderma indet.; P. fallax or P.
olivaceum
36
Tomentella fibrosa Common NE 1
Tomentellopsis echinospora Common LC 13
Trechispora farinacea Common LC 13
Trechispora laevis Common LC 3
Trechispora minima Occasional LC 4
Trechispora minuta ad int Rare Not Red Listed 1
Trechispora indet. No category Not Red Listed 3 x
Trechispora subsphaerospora Common LC 10
Trechispora verruculosa Rare LC 1
Tretomyces lutescens Rare DD 1
Tubulicrinis accedens Common LC 12
Tubulicrinis borealis s.l. Common LC/NE 27
Tubulicrinis calothrix Common LC 1
Tubulicrinis chaetophorus Rare VU 1
Tubulicrinis globisporus Occasional LC 1
Tubulicrinis medius Common LC 1
Tubulicrinis propinquus Occasional LC 1
Tubulicrinis subulatus Common LC 10
Tylospora asterophora Common LC 7
Tylospora fibrillosa Common LC 38
Vesiculomyces citrinus Common LC 6
Xylodon asperus Common LC 16
Xylodon borealis Rare NE 2
Xylodon brevisetus Common LC 25
Xylodon indet. No category Not Red Listed 3 x
Xylodon sp. 1 No category Not Red Listed 2
37
Tim e Length DBH
All species
43.07 0 0 0 1 0 0 0 0 0 0 0 - - 1 1 -
44.00 0 0 0 1 0 0 0 1 0 0 0 - - 1 1 -
44.26 0 0 1 1 0 0 0 0 0 0 0 - - 1 1 -
44.32 0 0 0 1 0 0 1 0 0 0 0 - - 1 1 -
44.37 1 0 0 1 0 0 0 0 0 0 0 - - 1 1 -
45.39 1 0 0 0 0 0 0 0 0 0 0 - - 1 1 -
44.72 0 0 0 1 0 0 0 0 0 0 1 - - 1 1 -
44.87 0 1 0 1 0 0 0 0 0 0 0 - - 1 1 -
44.99 0 0 0 1 1 0 0 0 0 0 0 - - 1 1 -
45.01 0 0 0 1 0 0 0 0 1 0 0 - - 1 1 -
Red Listed species
55.04 1 0 0 0 1 0 0 0 0 0 - 0 - 0 0 -
55.01 1 0 0 0 1 0 0 0 0 0 - 0 - 1 0 -
56.15 1 0 0 0 0 0 0 0 0 0 - 0 - 0 0 - x
55.30 1 0 0 0 1 0 0 1 0 0 - 0 - 0 0 -
54.57 1 0 0 0 1 0 0 1 0 0 - 0 - 1 0 -
56.32 1 0 0 0 0 0 0 0 0 0 - 0 - 1 0 -
55.96 1 0 0 0 1 0 0 0 0 0 - 0 - 0 1 -
55.98 1 1 0 0 0 0 0 1 0 0 - 0 - 0 0 -
56.66 1 0 1 0 0 0 0 0 0 0 - 0 - 0 0 - x
56.12 1 0 0 1 1 0 0 0 0 0 - 0 - 0 0 -
AIC
value
Bark
coverage
Plant
coverage
Ground
contact
Log
type
Decay
class
Top
intactness
Log10
polypore
hym enophore
area
Area
size
Num ber of
vegetation
types
Calcium ,
fine
scale
Calcium ,
coarse
scale
Presence
of richer
vegetation
types
Surface
area
Singular
convergence
Appendix 2
A list of the ten most parsimonious models of each analysed species group. Presence/absence of variables in the models is indicated by 1/0, and
unreliable models due to singular convergence by an x.
38
Tim e Length DBH
Common species
46.81 0 0 0 0 0 0 0 0 0 0 1 - - 0 1 -
47.28 0 1 0 0 0 0 0 0 0 0 0 - - 0 1 -
47.87 0 0 0 0 0 0 0 0 0 0 0 - - 0 1 -
46.98 0 1 0 1 0 0 0 0 0 0 0 - - 0 1 -
47.76 0 0 0 1 0 0 0 0 0 0 0 - - 0 1 -
46.64 0 1 0 0 0 0 1 0 0 0 0 - - 1 1 -
47.49 0 1 0 0 0 0 1 0 0 0 0 - - 0 1 - x
47.65 0 1 0 0 0 0 0 0 0 0 1 - - 0 1 -
48.37 0 0 0 0 1 0 0 0 0 0 0 - - 0 1 -
47.80 0 0 0 0 1 0 0 0 0 0 1 - - 0 1 -
Occasional species
61.65 0 0 0 0 0 0 0 0 0 0 - - 0 - - 1
58.98 0 1 0 0 0 0 1 1 0 1 - - 0 - - 1
60.01 0 1 0 0 0 0 1 0 0 1 - - 0 - - 1
61.42 0 1 0 0 0 0 0 0 0 0 - - 0 - - 1
60.97 0 1 0 0 0 0 0 0 0 1 - - 0 - - 1
61.26 0 1 0 0 0 0 0 1 0 0 - - 0 - - 1
60.54 0 1 0 0 0 0 1 1 0 0 - - 0 - - 1
60.68 0 1 0 0 0 0 1 1 0 1 - - 0 - - 0
61.74 0 1 0 0 0 0 0 0 1 0 - - 0 - - 1
62.39 0 0 0 0 0 0 0 0 0 1 - - 0 - - 1
Rare species
55.26 0 0 1 0 0 0 0 0 0 0 - - 0 0 0 -
55.27 0 1 0 0 0 0 0 0 0 0 - - 0 0 0 -
55.35 0 0 0 0 1 0 0 0 0 0 - - 0 0 0 -
55.18 0 0 0 0 1 0 0 0 0 0 - - 0 1 0 -
56.12 0 0 0 0 0 0 0 0 0 0 - - 0 0 0 -
55.44 0 0 1 0 0 0 0 0 0 0 - - 0 1 0 -
AIC
value
Bark
coverage
Plant
coverage
Ground
contact
Log
type
Decay
class
Top
intactness
Log10
polypore
hym enophore
area
Area
size
Num ber of
vegetation
types
Calcium ,
fine
scale
Calcium ,
coarse
scale
Presence
of richer
vegetation
types
Surface
area
Singular
convergence
39
Time Length DBH
Rare species cont.
55.55 0 0 0 1 1 0 0 0 0 0 - - 0 0 0 -
55.69 0 0 0 0 1 1 0 0 0 0 - - 0 0 0 -
55.21 0 0 0 1 1 0 0 0 0 0 - - 1 0 0 -
55.99 0 1 0 0 1 0 0 0 0 0 - - 0 0 0 -
AIC
value
Bark
coverage
Plant
coverage
Ground
contact
Log
type
Decay
class
Top
intactness
Log10
polypore
hymenophore
area
Area
size
Number of
vegetation
types
Calcium,
fine
scale
Calcium,
coarse
scale
Presence
of richer
vegetation
types
Surface
area
Singular
convergence