Dispersal and persistence of an epiphytic lichen in ...

Dispersal and persistence of an epiphytic lichen in a dynamic pasture-woodland landscape

Inauguraldissertation der Philosophisch-naturwissenschaftlichen Fakultät der Universität Bern Vorgelegt von Silke Werth von Deutschland Leiter der Arbeit: PD Dr. C. Scheidegger Eidgenössische Forschungsanstalt WSL, Birmensdorf ZH

Dispersal and persistence of an epiphytic lichen in a dynamic pasture-woodland landscape

Inauguraldissertation der Philosophisch-naturwissenschaftlichen Fakultät der Universität Bern

Vorgelegt von

Silke Werth von Deutschland

Leiter der Arbeit: PD Dr. C. Scheidegger Eidgenössische Forschungsanstalt WSL, Birmensdorf ZH

Von der Philosophisch-naturwissenschaftlichen Fakultät angenommen.

Bern, 20. Juni 2005

Der Dekan Prof. Dr. P. Messerli

Contents Synthesis

1

Introduction

1

Conclusions

7

Hierarchy theory and a conceptual model of epiphyte dynamics

7

Biology of Lobaria pulmonaria

9

Dispersal versus establishment limitation

11

Implications for conservation of L. pulmonaria

17

Chapter 1

23

Chapter 2

49

Chapter 3

71

Chapter 4

105

Chapter 5

137

Summary

149

Appendix 1 (Population genetic sampling)

153

Appendix 2 (Transplants)

165

Appendix 3 (Snow samples)

167

Acknowledgements

169

Curriculum vitae of Silke Werth

171

Synthesis

Introduction Traditional sylvopastoral landscapes exhibit dynamics in the form of shifting mosaics of forested and open areas if disturbances are at intermediate levels (Olff et al. 1999). If grazing pressure is too high, wooded pastures develop into grasslands; in the absence of grazing, successional change leads to climax forests (Gillet et al. 2002). Wooded pastures would not be predicted as a favoured type of habitat for epiphytes requiring long ecological continuity of the forest canopy or for dispersal-limited epiphytes with long generation times, because shifting mosaic dynamics imply limited time for fulfilling life cycles, including dispersal to new patches of habitat. However, the epiphytic lichen Lobaria pulmonaria, which is considered to be dispersal-limited (Walser et al. 2001; Walser 2004), and has a long generation time compared with other foliose lichens (Scheidegger & Goward 2002), has a hotspot of its occurrence in Switzerland in a sylvopastoral landscape in and around the Parc Jurassien Vaudois (Zoller et al. 1999; Rychen 2002). L. pulmonaria is widely distributed in the nemoral and boreal zones of the northern hemisphere, with some additional occurrences in the southern hemisphere (Yoshimura 1971). This tripartite lichenised fungus is associated with the green algae Dictyochloropsis reticulata and cyanobacteria of the genus Nostoc (Geitler 1966; Jordan 1970). The generation time in lichens is defined as the average age at which a population of a lichen produces propagules, be it soredia, isidia, thallus fragments, or ascospores. Generation time in populations of L. pulmonaria in central Europe has been estimated to be 30 years or longer (Scheidegger & Goward 2002). In Europe and North America, L. pulmonaria is mainly associated with old-growth forests (Lesica et al. 1991; Rose 1992; Radies & Coxson 2004). Owing to air-pollution and intensive forest management, the species has faced a severe decline in central and northern Europe (Hallingbäck & Martinsson 1987), where it is red-listed in several countries (Wirth et al. 1996; Aptroot et al. 1999; Scheidegger et al. 2002; Søchting & Alstrup 2002). The population of the L. pulmonaria in the pasture-woodland landscape in the Parc Jurassien Vaudois is not only one of the largest in Switzerland, but also seems to exhibit genetic variation comparable in magnitude to that of populations in the Swiss

1

Synthesis

Alps and the Swiss Plateau (Zoller et al. 1999), despite the fact that the forests of the Parc Jurassien Vaudois are and were subjected to disturbances, mainly in the form of grazing pressure on juvenile trees, mainly uneven-aged forestry, some stand-level timber logging, and also stand-replacing disturbances (Vittoz 1998; Kalwij et al. in press). Genetic diversity of populations of L. pulmonaria would be expected to be low in this system for several reasons. Firstly, disturbances affecting carrier trees such as Acer pseudoplatanus resulting in reductions of L. pulmonaria population sizes may lead to instant loss of rare alleles (Nei et al. 1975). Secondly, new populations in disturbed areas which were founded by only few individuals may be subject to genetic drift, i.e., random changes in allele frequencies, leading to fixation or further loss of alleles even if overall mean allele frequencies remain constant among populations (Hartl & Clark 1997). Whether the spatial genetic signature differed among disturbance types in L. pulmonaria interested me greatly. Large population sizes of a putatively dispersal-limited organism with high genetic variation in a dynamic landscape represent an apparent contradiction, which is why I was particularly interested in studying patterns of genetic diversity in sites differing in disturbance history, as well as dispersal and establishment in L. pulmonaria. Dispersal and establishment are key processes in the life history of organisms (Clobert et al. 2001). The there has been a debate about the role of dispersal in structuring in the composition of local communities, namely if local species assemblages are a function of dispersal ability or of site conditions (Leibold et al. 2004; Ozinga et al. 2005a; Ozinga et al. 2005b). The relative importance of different environmental factors, such as regionalscale versus local factors in structuring communities can be determined by an investigation of communities along regional gradients, while simultaneously measuring site conditions (Ohmann & Spies 1998; Peterson & McCune 2001), and subjecting the resulting dataset to gradient analysis techniques such as constrained ordination (Økland 1990; Borcard et al. 1992). This study aimed (1) to determine if and how genetic structure and diversity of L. pulmonaria are influenced by disturbance type, (2) to partition spatial genetic structure to a clonal and a recombinant component, (3) to reveal to which degree the distribution of L. pulmonaria in the sylvopastoral landscape is limited by propagule availability versus establishment, (4) to assess the significance of a putative barrier to gene flow in the

2

Synthesis

sylvopastoral landscape and to determine the spatial extent of populations in the studied landscape, (5) to determine the relative importance of local site conditions vs. regional gradients in shaping epiphytic lichen communities. These issues can be summarised in five main questions: (i)

What is the effect of two types of stand-level disturbances on the genetic diversity and spatial genetic structure in L. pulmonaria?

(ii)

Are there spatially separate populations of L. pulmonaria in the study area, and do large open pastures represent effective barriers to gene flow in this species?

(iii) Is the distribution of L. pulmonaria in a sylvopastoral landscape limited by dispersal or establishment, or by both? (iv) What is the spatial extent of clonal versus recombinant genetic structure? (v)

How important are local environmental conditions versus large-scale, regional gradients in structuring epiphytic macrolichen communities? The relationship between disturbance type and spatial genetic structure and

genetic diversity is analysed in the first chapter. The second chapter determines how many populations are present in the study area, and assesses the effectiveness of a putative barrier, a large pasture, to restrict past gene flow in L. pulmonaria. In the third chapter, the role of dispersal versus establishment in limiting the distribution of L. pulmonaria in the study area is investigated. The spatial extent of overall genetic structure, as well as of its clonal and recombinant component are assessed in the fourth chapter. The importance of local environmental conditions versus large-scale environmental gradients for the species composition of epiphytic macrolichen communities is investigated in the fifth chapter. We furthermore investigated the implications of stand-replacing and stand-level disturbance under different levels of local versus global dispersal for persistence of L. pulmonaria using modelling approaches. The resulting manuscript is not included in this thesis. We discuss the results gained in the light of hierarchy theory. Novel and exciting aspects of the biology of L. pulmonaria are presented. The role of dispersal limitation vs. establishment limitation in structuring the

3

Synthesis

local distribution of L. pulmonaria is reviewed, and implications for conservation of L. pulmonaria are discussed.

Influence of disturbance type on genetic structure and diversity Old-forest associated lichens are commonly assumed to be negatively affected by tree logging or natural forest disturbances. However, we found that genetic diversity of Lobaria pulmonaria depends on the type of disturbance. Using 895 thalli collected from 41 plots of 1 ha (demes) corresponding to the disturbance categories stand-replacing disturbance, intensive logging and uneven-aged forestry, we determined fragment length at six mycobiont-specific microsatellite loci. There was evidence for multiple independent colonisations of demes located in areas affected by disturbance at forest stand level. Using spatial autocorrelation methods, we determined the spatial scale of similar genetic structure discriminating among the clonal and recombinant component of genetic variation and among disturbance type. Spatial autocorrelation of gene diversity was strong at distances up to 150 m in all three disturbance categories, with the strongest autocorrelation for demes affected by stand-replacing disturbance. The spatial autocorrelation was predominantly attributed to clonal dispersal of vegetative propagules. After accounting for the clonal component, we did not find significant spatial autocorrelation. This pattern may indicate low dispersal ranges of clonal propagules, but random dispersal of sexual ascospores, indicating that there was no dispersal limitation of ascospores. Genetic diversity was highest in demes affected by intensive logging at forest-stand level, and lowest in demes affected by stand-replacing disturbance. Our study exemplified that genetic diversity of some epiphytic lichens may not necessarily be impacted by stand-level disturbances for many centuries.

Spatial extent of populations and barriers to gene flow As epiphytes as are strongly affected by the population dynamics of their carrier trees, large dispersal rates and corresponding high levels of gene flow may be needed for populations to persist at a landscape scale. Here we analysed landscape-level

4

Synthesis

population genetic data for L. pulmonaria and found indeed high amounts of past gene flow at the landscape level in microsatellite data from 895 thalli of L. pulmonaria genotyped at six fungal loci. Bayesian analysis revealed the presence of at least three spatially intermingled gene pools in the studied landscape. We also assessed the presence and effectiveness of potential barriers to gene flow and show that an a priori assumed barrier in the landscape, a large unforested area, did not significantly restrict past gene flow. Our results indicate that population genetic data at the landscape level may be relevant for our understanding of the organisation of genetic diversity and for population persistence of L. pulmonaria.

Dispersal versus establishment limitation Dispersal is a critical population process for the dynamics and persistence of metapopulations, but is difficult to quantify. We analysed 240 DNA extracts derived from snow samples by a L. pulmonaria specific RealTime-PCR assay allowing for the discrimination among propagules originating from a single, isolated source tree, and originating from other locations. Samples which were detected as positives by RealTimePCR were additionally genotyped for six L. pulmonaria microsatellite loci. Both molecular approaches demonstrated substantial dispersal from other than local sources. In a landscape approach, we analysed 240 snow samples with RealTime-PCR and detected propagules not only in forests where L. pulmonaria was present, but also in large unforested pasture areas and in forest patches where L. pulmonaria was not found. Monitoring of soredia of L. pulmonaria transplanted to maple bark after two vegetation periods showed high variance in growth among forest stands, but no significant differences among transplantation treatments. Hence, in all likelihood not the amount of propagules is what is limiting the old-forest lichen L. pulmonaria from occurring in younger forests in localities where large populations are present, but ecological constraints at the stand level may be. Further research is needed to identify the factors responsible for this.

5

Synthesis

Spatial extent of clonal vs. recombinant genetic structure A geostatistical perspective on spatial genetic structure may explain methodological issues of quantifying spatial genetic structure and suggest new approaches to addressing them. We use a variogram approach to (i) derive a spatial partitioning of molecular variance, gene diversity, and genotypic diversity for microsatellite data under the infinite allele model (IAM) and the stepwise mutation model (SMM), (ii) develop a weighting of sampling units to reflect ploidy levels or multiple sampling of genets, and (iii) show how variograms summarize the spatial genetic structure within a population under isolation-by-distance. The methods are illustrated with data from a population of the epiphytic lichen Lobaria pulmonaria, using six microsatellite markers, and illustrated the spatial extent of the clonal and recombinant component of genetic variation. The clonal component was large in the studied population. Variogram-based analysis not only avoids bias due to the underestimation of population variance in the presence of spatial autocorrelation, but also provides estimates of population genetic diversity and the degree and extent of spatial genetic structure accounting for autocorrelation.

Regional versus local factors shaping epiphytic macrolichen communities To investigate the relative influences of human impact, macroclimate, geographic location and habitat related environmental differences on species composition of boreal epiphytic macrolichen communities, lists of epiphytic macrolichen species from deciduous forests recorded in Troms county in northern Norway were subjected to constrained ordination. By Canonical Correspondence Analysis with variance partitioning, the relative amounts of variance in macrolichen species composition attributable to site factors (human impact, environmental differences among sites), regional factors (macroclimate), and the spatial component were quantified. Regional factors were most important in determining epiphytic macrolichen communities, but also site conditions such as forest stand properties contributed significantly to macrolichen community composition. Epiphytic macrolichen communities were assembled along a macroclimatic gradient from the coastline to the interior of central northern Norway.

6

Synthesis

Conclusions Hierarchy theory and a conceptual model of epiphyte dynamics Landscape ecological research has pinpointed the importance of temporal and spatial scales in ecology (Wiens 1989; Keitt et al. 1997; Bunnell & Huggard 1999; Li & Wu 2004; Wu 2004). Hierarchy theory assumes that processes operating at large spatial scales, the so-called “forcing functions”, shape processes at smaller scales. In the same way, processes at fine spatial scales influence processes at the next lower scale (Allen & Starr 1982; O'Neill et al. 1986; Allen & Hoekstra 1990; Muller 1992). From a hierarchy theory point of view, one may view topography, climatic conditions, the natural disturbance regime, landscape management and the level of acidic precipitation in the study area as the large-scale forcing factors shaping the dynamics of tree populations (Fig. 0-1). The spatial configuration and dynamics of the carrier tree population crucially influences epiphyte populations (Snäll et al. 2005a; Snäll et al. 2005b; Wagner et al. in prep.). Ultimately, processes that influence the carrier tree population, or habitat properties of carrier trees, determine the population dynamics of epiphyte communities, and thus of single epiphyte species. Species are assembled in communities along ecological gradients (Legendre & Legendre 1998). Macroclimatic gradients drive tree species composition on a regional scale, whereas topographical gradients are decisive at the forest stand level (Ohmann & Spies 1998; Bugmann & Solomon 2000). However, our results from epiphytic macrolichen communities in north Norway show that this is not only true for tree species composition. Macroclimatic gradients were found to be most important in structuring community composition of epiphytic macrolichens, relative to environmental factors at the level of forest stands, or spatial structure. Under different climatic conditions, other tree species than Picea abies, which is not an important carrier tree of L. pulmonaria in the study area, might have dominated the landscape. Had the forest in the study area been composed predominantly by a late successional tree species which provided more suitable habitat for L. pulmonaria, the effect of stand-replacing disturbances on the spatial occurrence and genetic diversity of L. pulmonaria might have been markedly different. In the latter scenario, the largest population sizes and highest genetic diversity of L. pulmonaria would be expected to

7

Synthesis

occur in old-growth forest stands. Further studies are needed to determine if this scenario is true.

Figure 0-1. Conceptual model of epiphyte population dynamics in central European temperate forests.

Microsite parameters like bark suitability or bark chemical and textural properties may influence the lower level processes in the hierarchy, e.g., the dynamics of epiphyte communities. The chemical properties of bark substrates may be influenced by wet acidic deposition, lowering the bark pH (Bates 1992; Gauslaa & Holien 1998). However, pollution influence is most likely constant across the study area. Macroclimatic conditions may affect the breadth of the ecological amplitude of epiphytic lichens. At their northern distribution limit in the northern Norway, some oceanic species like Lobaria pulmonaria, Lobaria hallii and Degelia plumbea occur only in the most continental sites in the interior, characterised by high summer temperature sums and comparatively low annual precipitation (see chapter five), and in the interior sites, they were predominantly occurring on calcareous schist rock substrates. At the northern outposts of their distribution, these species’ ecological amplitude was narrowed

8

Synthesis

to the climatically most favourable sites and to a favourable substrate which was rare in the study area in northern Norway. Some lichen species which are habitat specialists and confined to a narrow range of host trees in central Europe may show a wider ecological amplitude in a coastal oceanic climate, allowing them to colonise tree species which are not suitable habitat in most of central Europe, such as Picea abies. In boreal spruce forests of central Norway, an oceanic lichen species assemblage including L. pulmonaria frequently grows on P. abies branches (Haugan et al. 1995; Tønsberg et al. 1996; Holien 1998). Low bark pH may lead to a narrowing of the ecological amplitude of epiphytes, confining some epiphytes species to the most alkaline substrates (Gauslaa 1995; Wirth 1995; Wirth et al. 1996). This effect has been observed for L. pulmonaria, which used to grow on acidic substrates such as P. abies twigs in interior areas of Norway in the past, but has not been found on this substrate type for the last 60 years (Tønsberg et al. 1996).

Biology of Lobaria pulmonaria The clonal component of genetic variation is very important in the sylvopastoral landscape in Switzerland, particularly on a local scale (chapter 1, chapter 4). However, the absence of significant linkage among the majority of loci showed that also recombination may be important in the genetic structure of the L. pulmonaria population from the Parc Jurassien Vaudois, a result which supports that L. pulmonaria is an outcrossing species (Walser et al. 2004). A further evidence of recombination is that fertile thalli of L. pulmonaria occur in the Parc Jurassien Vaudois, even though they are not frequent; during our field work, we found apothecia on a total of 15 out of 298 trees (5 %). However, the presence of four significant pairwise associations of loci points towards some non-random (i.e. assortative) mating in the total population. Also the analysis of population structure using assignment tests indicated assortative mating, as it revealed the presence of three gene pools in the study area, all of which occurred spatially intermingled. These results point towards heterothallism in L. pulmonaria, i.e. mating occurring only among compatible mating types. However, further studies, e.g. of spore multilocus genotypes, or of mating type genes, are needed for proving whether or not L. pulmonaria is a heterothallic fungus. In general, the conditions under which ascospores

9

Synthesis

are produced in L. pulmonaria are still very little understood and further research is needed, which could focus on a wide range of subjects including paternity analysis of ascospores, dispersal ranges of ascospores, and ecological conditions favouring the development of apothecia. In this study, in addition to the transplanted soredia, eight thallus fragments of thalli containing soredia, but not apothecia were transplanted onto each of the 55 trees which we transplanted soredia of L. pulmonaria to. When we revisited some of the sites three years after experiment establishment, some fragments had developed apothecia. What would be needed in the future is a pairwise transplantation study involving overlapping transplantation of fragments of the same microsatellite multilocus genotype onto unoccupied potential host trees at different distances, versus transplantation of very different multilocus genotypes. Long-term monitoring of the fragments could then determine under which conditions apothecia are developed, and if presence of another multilocus genotype is necessary for ascomata development in L. pulmonaria. Fifteen of the 895 analysed thalli had two alleles at one or more microsatellite loci. Most of these occurred on a single tree and were of the same multilocus genotype. Also a further five thalli collected from a single tree in the surroundings of the plot where snow samples had been collected (point source approach) showed two alleles at a microsatellite locus. There are three likely reasons for this pattern. (1) The occurrence of several alleles in the same thallus may simply represent a genome duplication affecting a microsatellite locus. A contradiction to this view is that in a few thalli, two alleles were found at two loci. This makes genome duplication less likely, unless the two microsatellite loci would be located next to each other on a chromosome. As the microsatellite loci were not significantly linked, they seemed to lie on different linkage groups, so that genome duplication is not a very likely explanation (Walser et al. 2004; Werth et al. in prep.). (2) Dikaryotic and diploid hyphae may have been extracted, which contained the allele of the “mother” thallus plus that of a “father”. What is strange, however, is that all thalli from two trees contained exactly the same alleles. (3) Another, more likely reason might be that the thalli with more than a single allele per locus do not contain a single mycobiont, but several mycobiont individuals, which may have spread clonally within the tree. A more detailed study is needed to be conclusive on this subject,

10

Synthesis

genotyping multiple extracts from the thalli where several alleles were found, perhaps in combination with genotyping of sterile cultured thallus hyphae. Together with information on heterothallism, this kind of study may provide very interesting insights into the biology of L. pulmonaria.

Dispersal versus establishment limitation Recruitment in plant or lichen populations is the number of new individuals entering a given population per generation. Recruitment is a key process for population dynamics, to which both dispersal and establishment contribute (Begon et al. 1999; Nathan & Müller-Landau 2000). Dispersal has to be sufficient to provide diaspores at a given site within a population. In the following, limited availability of propagules at a particular habitat will be termed dispersal limitation (Nathan & Müller-Landau 2000). Establishment may be influenced not only by biotic factors like competition, but also by abiotic factors such as site conditions (Begon et al. 1999). Inability of diaspores to establish at a given habitat is in the following referred to as establishment limitation (Nathan & Müller-Landau 2000). Dispersal limitation and establishment limitation are not mutually exclusive, but may simultaneously reduce recruitment in a population. There has been an ongoing debate on whether local species pools are a function of ecological niches or of the dispersal ability of species (Leibold et al. 2004; Ozinga et al. 2005b). Also for epiphytic lichen communities, these concepts have been implicitly adopted in the form of the dispersal limitation hypothesis, stating that limited availability of propagules in young forests is the major reason why many epiphytic lichens are confined to old-growth forests (Sillett et al. 2000b; Hilmo 2002). According to an alternative hypothesis, the ecological gradient hypothesis, which corresponds to the niche assembly view of local communities, old-growth dependence in lichens is a function of favourable ecological conditions (Dettki et al. 2000). Both hypotheses are not necessarily mutually exclusive. It is however important to distinguish between the two hypotheses, because each one leads to a different conservation strategy. If the ecological gradient hypothesis applies, habitat quality of primeval forests would be equal to that of secondary forests once comparable microhabitats had developed. The ultimate conservation strategy

11

Synthesis

would be to prolong the rotation cycle. Forestry practices would not matter directly, as long as forests had sufficient time for regrowing to mature or even old-growth, and for old-growth dependent lichens it would not make a difference whether a forest was clearcut or whether remnant trees were retained at final harvest. If the dispersal limitation hypothesis holds, on the other hand, it would be of primary importance to retain as many old-growth forests as diaspore source for old-growth associated species as possible, and instead of clearcutting whole forest stands, old trees should be retained at harvest to facilitate recolonisation (Sillett et al. 2000b). An important question is why L. pulmonaria is not found in many parts of the study area in Switzerland where seemingly suitable habitat is present. The distribution of L. pulmonaria could be either limited by propagule availability, or by establishment. One may argue that dispersal of every species on the planet is limited because the number of its propagules is finite, hence limiting distribution. Decreasing propagule density with distance from source is a common phenomenon (Ouborg et al. 1999; Cain et al. 2000; Nathan & Müller-Landau 2000; Clobert et al. 2001) and this alone does not allow the conclusion that dispersal limits the distribution of a given species. Furthermore, the term dispersal limitation is meaningless unless an explicit spatio-temporal context is given. Most transplantation experiments of propagules concluding that lichen species are dispersal limited have not defined what they mean with dispersal limitation. Studies involving transplantation of propagules do actually test establishment limitation, i.e. the ability of propagules to establish successfully at a given site (Nathan & Müller-Landau 2000). Hence, such studies allow limited conclusions on recruitment limitation in terms of dispersal limitation, i.e. the availability of propagules at a particular site (Nathan & Müller-Landau 2000). Molecular, population-genetic approaches as the variogram method employed in this study (chapters one and four) only allow inference of “realised” dispersal, i.e. dispersal events which were followed by successful establishment of propagules, and thus they integrate propagule limitation with establishment limitation (Nathan & Müller-Landau 2000). I propose the following definition for dispersal limitation: a given epiphytic lichen species is dispersal limited if it does not manage to colonise a forest stand with suitable habitat within three times its generation time. Three times the generation time is the time scale at which population decline of endangered

12

Synthesis

species is assessed for conservation purposes such as in the IUCN criteria for Red Lists of threatened species (IUCN 2001), and may therefore also be a relevant time scale for a definition of dispersal limitation. The generation time in L. pulmonaria is 30 years or more for vegetative propagules, and even probably longer for ascospores (Scheidegger & Goward 2002). Our model organism has been able to colonise an area affected by standreplacing disturbance within 132 years after disturbance, and has already reached a rather large population size; the disturbed area has been colonised multiple times and has by now reached a comparatively high genetic diversity, given the strength of the disturbance event. This alone makes it unlikely that L. pulmonaria is dispersal limited in the study area in the Parc Jurassien Vaudois in terms of our rather wide definition. Assuming that the main carrier tree species of L. pulmonaria in the study area, Sycamore maple (Acer pseudoplatanus), needs to be about 50 years old to be suitable for colonisation by L. pulmonaria, this would leave 82 years or about three generations for L. pulmonaria to colonise the stand. L. pulmonaria has managed to successfully colonise the disturbed area within these three generations. Our data provide therefore evidence that L. pulmonaria is not dispersal limited. This is not very surprising. Most lichens have small, lightweight propagules, and thus one would expect many lichens to be efficient dispersers (Bailey 1976; Galloway & Aptroot 1995). Lichen colonisation of newly emerged volcanic islands three years after the last phase of eruption (Kristinsson 1972) and rapid colonisation of anthropogenic substrates by the lichen Lecanora dispersa (Bailey 1976) are only two examples of efficient dispersal and establishment in lichens. As compared to e.g. seeds of most forest trees, propagules of L. pulmonaria are lightweight and hence well suited for wind dispersal. Also our results concerning the number of colonisation events to found demes disagree with dispersal limitation. There was evidence of multiple colonisations of demes affected by stand-replacing disturbances, and the minimum number of colonisation events was not related to disturbance level. Furthermore, ANOVA showed that there was no relationship between disturbance and genetic distance, suggesting that demes of L. pulmonaria were independently colonised several times from different sources after a disturbance event.

13

Synthesis

Throughout our study area, however, we found 1-ha plots which were not colonised by L. pulmonaria even though they contained maple trees which seemed to be suitable for colonisation (Kalwij et al. in press). An alternative definition of dispersal limitation might thus be that a species is dispersal limited if it does not manage to colonise all suitable habitat. However, the problem with this definition is that experimental evidence is needed to show that habitat is actually suitable, and not only unoccupied due to an unknown, unfavourable environmental factor. This is what has often been attempted to be tested with experimental transplantation of lichen propagules (Scheidegger 1995; Sillett et al. 2000a; Sillett et al. 2000b; Hilmo & Såstad 2001). While giving valuable information on juvenile development of lichens, the problem with these studies is that transplanted lichen propagules or fragments are often monitored for a few years only, and that effects of e.g. competition may become visible much later. In a longterm study where transplanted thallus fragments of Lobaria amplissima had been observed for two decades, the number of transplanted fragments declined drastically in the second decade to less than half of the initially transplanted number of thallus fragments (Gilbert 2002). Even though the latter example suffers from a low sample size (14), it may still illustrate the main problem of short term transplantation studies, namely that it is not predictable whether transplanted propagules or thallus fragments which have been observed for a few years will be able to survive in the long run. For transplanted propagules, long-term chances of survival are probably even lower than for transplanted thallus fragments owing to their small size, which may make them susceptible to being outcompeted by fast-growing bryophytes or lichens, or being grazed by herbivores. First signs of competition by fast-growing bryophytes and lichens were also evident in our experimental transplantation of L. pulmonaria symbiotic propagules (chaper 3), but were not pronounced enough to be quantified. In the years to come, however, it is likely that a large number of juvenile thalli of L. pulmonaria developing from transplanted symbiotic propagules will be overgrown and outcompeted by competitors. Thus it would be very interesting to re-visit the sites in five or ten years time to find out where adult thalli of L. pulmonaria developed from the transplanted propagules. That dispersal limitation was found to be less important in the L. pulmonaria population in the study area is probably mainly a result of its large population size.

14

Synthesis

Dripzone effects are known to occur in lichens, with establishment occurring on otherwise less suitable habitat if huge amounts of propagules are available nearby (Goward & Arsenault 2000). Also in the study area, I found two trees of Picea abies and a single Abies alba to be occupied by L. pulmonaria in localities where L. pulmonaria population density was very high and adult thalli of L. pulmonaria were situated directly above or at a few meters distance from the respective “unsuitable” tree. This example illustrates that if excessive amounts of propagules are available, eventually some may be able to establish even on seemingly unsuitable habitat. On the other hand, in the study area, propagules of L. pulmonaria seem to be able to reach most currently unoccupied sycamore and beech trees, if a dispersal distance of 300 m is used and currently occupied trees are connected with currently unoccupied trees at distances up to 300 m (results not shown). So why is there a large proportion of currently unoccupied potential carrier trees which look as if they were suitable habitat? It might be a question of competition by fastgrowing epiphytes, or of allelopathic effects by secondary metabolites of other lichen species, hindering establishment of L. pulmonaria, but this is not likely to explain the large differences between stands which we found. Alternatively, the reason may be differential microclimate among stands, favouring propagule establishment in some sites, while being unfavourable in others. In addition to local unavailability of cyanobacteria, this seems to be another very likely explanation for the large among-stand differences in the number of successfully established symbiotic propagules, but more studies are needed to be conclusive on this subject. We observed a large clonal component. Recurrent multilocus genotypes result from three processes: soredia are produced via mitotic cell divisions in clonal propagation, ascospores may be produced via self-fertilisation producing ascospores identical with the thallus multilocus genotype in homothallic fungi, or via outcrossing, by chance resulting in an ascospore multilocus genotype identical to a parent multilocus genotype. Variogram analysis of gene diversity H and of genotype diversity D showed that spatial autocorrelation due to clonal dispersal or selfing was confined to distances up to 150 m. However, it remains unclear why the spatial extent of autocorrelation is shorter under unevenaged forestry than that under stand-level logging and stand-replacing disturbance. Within the time since the stand-replacing disturbance (132 years), the

15

Synthesis

majority of propagules resulting from vegetative propagation or selfing established within a few hundred meters’ distance from their origin. However, this does not necessarily mean that vegetative propagules or ascospores produced by selfing are dispersal limited in large populations, as the one investigated in the Parc Jurassien Vaudois. Decreasing propagule densities with distance from a source is a common feature also of species that are not dispersal-limited (Nathan & Müller-Landau 2000; Clobert et al. 2001), such as the generalist lichen Hypogymnia physodes (Armstrong 1987). This pattern of propagule distribution common to most species may induce spatial autocorrelation at short distances. However, a truly dispersal-limited organism would not have been able to colonise a disturbed area within about a century, and even establish deme sizes larger than those of undisturbed demes. Therefore, we conclude that at a time-scale of about a century and the spatial extent of the disturbance, vegetative propagules of L. pulmonaria were not dispersal limited in our study area. Dispersal limitation thus depends on the time scale investigated. When the effect of recurrent multilocus genotypes was accounted for, a small, but statistically non-significant amount of spatial autocorrelation remained in demes affected by stand-replacing disturbance in the first two distance classes. This means that pairs of multilocus genotypes within trees and within demes tended to be more similar to each other than pairs of multilocus genotypes at larger distances, and is a sign of local, i.e., within tree or within deme, dispersal of ascospores. However, ascospores may also serve as dispersal agents over longer distances, e.g. for the colonisation of disturbed forest stands. The absence of spatial autocorrelation at larger distances suggests that there were ascospores produced by outcrossing, which had dispersed over long distances, which might have originated from occurrences of L. pulmonaria bordering our study area in the SW and NE, or from sources further apart. Hence, we consider that ascospores of L. pulmonaria were not dispersal-limited at the temporal and spatial scale represented in our investigation. The variogram analysis showed clearly that the dispersal range of clonal propagules was lower than that of ascospores. As clonal propagules were abundant in the study area – almost every thallus we found exceeding 5 cm in diameter contained soredia – they may be important for the rapid short-distance colonisation of trees within forest stands (Walser et al. 2001; Walser 2004). In contrast, the probably less abundant

16

Synthesis

ascospores may serve as dispersal agents over longer distances, e.g. for the colonisation of distant forest patches, or of areas affected by disturbances (Bailey 1976). Implications for conservation of L. pulmonaria Future conservation strategies for L. pulmonaria and species of similar ecological demands need to conserve large populations allowing for production of ascospores, and to establish conditions under which large populations will be obtained in the future. In areas where the species is very rare, harvest of single trees containing the species may already be harmful for the remaining specimen. In these areas, like in the small but genetically diverse populations of the Swiss Plateau, no forest management activities should be performed under which remaining carrier trees are lost, while for long-term persistence of the population, it should be attempted to establish and promote future carrier trees in the vicinity. Localities in the Swiss Jura and in the Alps are of major importance not only to protect extant populations of the species in central Europe, but may also be relevant for potential future range expansion of the species to areas where L. pulmonaria was formerly frequent, but has declined owing to air pollution or intensive forest management (Schöller 1997). Recolonisation of former lichen deserts by various lichen species has been observed and is thought to be the result of decreased levels of acidic deposition (van Dobben 1996; Heibel et al. 1999; Kricke & Feige 2001). Hopefully, also species like L. pulmonaria will be able to expand into areas where they formerly were frequent, but do not occur currently. Also in this respect, it is essential to protect the large and viable populations in the Jura Mountains and Alps of Switzerland.

References Allen THF, Hoekstra TW (1990) The confusion between scale-defined levels and conventional levels of organization in ecology Journal of Vegetation Science 1, 512. Allen THF, Starr TB (1982) Hierarchy: perspectives for ecological complexity. University of Chicago Press, Chicago. Aptroot A, van Herk CM, Sparrius LB, van den Boom PPG (1999) Checklist van de Nederlandse Korstmossen en Lichenicole Fungi Buxbaumiella 50, 4-64.

17

Synthesis

Armstrong RA (1987) Dispersal in a population of the lichen Hypogymnia physodes Environmental and Experimental Botany 27, 357-363. Bailey RH (1976) Ecological aspects of dispersal and establishment in lichens. In: Lichenology: Progress and Problems (eds. Brown DH, Hawksworth DL, Bailey RH) pp. 215-247. Academic Press, London. Bates JW (1992) Influence of chemical and physical factors on Quercus and Fraxinus epiphytes at Loch Sunart, western Scotland: a multivariate analysis Journal of Ecology 80, 163-179. Begon M, Harper JL, Townsend CR (1999) Ecology: individuals, populations and communities, Third ed., repr. edn. Blackwell, Oxford. Borcard D, Legendre P, Drapeau P (1992) Partialling out the spatial component of ecological variation Ecology 73, 1045-1055. Bugmann HKM, Solomon AM (2000) Explaining forest composition and biomass across multiple biogeographical regions Ecological Applications 10, 95-114. Bunnell F, Huggard DJ (1999) Biodiversity across spatial and temporal scales: problems and opportunities Forest Ecology and Management 115, 113-126. Cain ML, Milligan BG, Strand AE (2000) Long-distance seed dispersal in plant populations American Journal of Botany 87, 1217-1227. Clobert J, Danchin E, Dhondt AA, Nichols JD (2001) Dispersal. Oxford University Press, Oxford. Dettki H, Klintberg P, Esseen P-A (2000) Are epiphytic lichens in young forests limited by local dispersal? Ecoscience 7, 317-325. Galloway DJ, Aptroot A (1995) Bipolar lichens: a review Cryptogamic Botany 5, 184191. Gauslaa Y (1995) The Lobarion, an epiphytic community of ancient forests threatened by acid rain Lichenologist 27, 59-76. Gauslaa Y, Holien H (1998) Acidity of boreal Picea abies-canopy lichens and their substratum, modified by local soils and airborne acidic depositions Flora 193, 249-257. Geitler L (1966) Die Chlorococcalen Dictyochloris und Dictyochloropsis, nov. Gen. Österreichische Botanische Zeitschrift 113, 155–164. Gilbert O (2002) A transplant operation involving Lobaria amplissima; the first twenty years Lichenologist 34, 267-269. Gillet F, Besson O, Gobat J-M (2002) PATUMOD: a compartment model of vegetation dynamics in wooded pastures Ecological Modelling 147, 267-290. Goward T, Arsenault A (2000) Cyanolichen distribution in young unmanaged forests: A dripzone effect? The Bryologist 103, 28-37. Hallingbäck T, Martinsson PO (1987) The retreat of two lichens, Lobaria pulmonaria and L. scrobiculata in the district of Gäsene (SW Sweden) Windahlia 17, 27-32. Hartl DL, Clark AG (1997) Principles of population genetics. Sinauer, Sunderland. Haugan R, Holien H, Rydgren K (1995) Liaheia, Brønnøy kommune, Nordland, en oseanisk granskog med verdens nordligste forekomst av rund porelav, Sticta fuliginosa (Dicks.) Ach. Blyttia 1, 15-24. Heibel E, Lumbsch HT, Schmitt I (1999) Genetic variation of Usnea filipendula (Parmeliaceae) populations in western Germany investigated by RAPDs suggests reinvasion from various sources. American Journal of Botany 86, 753-757.

18

Synthesis

Hilmo O (2002) Growth and morphological response of old-forest lichens transplanted into a young and an old Picea abies forest Ecography 25, 329-335. Hilmo O, Såstad SM (2001) Colonization of old-forest lichens in a young and an old boreal Picea abies forest: an experimental approach Biological Conservation 102, 251-259. Holien H (1998) Lichens in spruce forest stands of different successional stages in central Norway with emphasis on diversity and old growth species Nova Hedwigia 66, 283-324. IUCN (2001) IUCN red list categories and criteria: Version 3.1. IUCN Species Survival Commission. IUCN, Gland, Switzerland and Cambridge, UK. Jordan W (1970) The internal cephalodia of the genus Lobaria The Bryologist 73, 669681. Kalwij JM, Wagner HH, Scheidegger C (in press) Effects of stand-replacing disturbance events on the spatial distribution of a lichen indicator of forest conservation value Ecological Applications. Keitt TH, Urban DL, Milne BT (1997) Detecting critical scales in fragmented landscapes Conservation Ecology (online) 1, 4. Available from the Internet. URL: http://www.consecol.org/vol1/iss1/art4. Kricke R, Feige GB (2001) Biomonitoring with lichens in the Ruhr area for air quality assessment - 1966 to 2000 Gefahrstoffe - Reinhaltung der Luft 61, 163-166. Kristinsson H (1972) Studies on lichen colonization in Surtsey 1970 Surtsey Research Programm Report 5, 77. Legendre P, Legendre L (1998) Numeric ecology. Elsevier, Amsterdam. Leibold MA, Holyoak M, Mouquet N, Amarasekare P, Chase JM, Hoopes MF, Holt RD, Shurin JB, Law R, Tilman D, Loreau M, Gonzalez A (2004) The metacommunity concept: a framework for multi-scale community ecology Ecology Letters 7, 601613. Lesica P, McCune B, Cooper SV, Hong WS (1991) Differences in lichen and bryophyte communities between old-growth and managed second-growth forests in the Swan Valley, Montana Canadian Journal of Botany 69, 1745-1755. Li HB, Wu JG (2004) Use and misuse of landscape indices Landscape Ecology 19, 389399. Muller F (1992) Hierarchical approaches to ecosystem theory Ecological Modelling 63, 215-242. Nathan R, Müller-Landau HC (2000) Spatial patterns of seed dispersal, their determinants and consequences for recruitment Trends in Ecology and Evolution 15, 278-285. Nei M, Maruyama T, Charkaborty R (1975) The bottleneck effect and genetic variability in populations Evolution 29, 1-10. Ohmann JL, Spies TA (1998) Regional gradient analysis and spatial pattern of woody plant communities of Oregon forests Ecological Monographs 68, 151-182. Økland RH (1990) Vegetation ecology: theory, methods and applications with reference to Fennoscandia Sommerfeltia Supplement 1, 1-233. Olff H, Vera FWM, Bokdam J, Bakker ES, Gleichman JM, De Maeyer K, Smit R (1999) Shifting mosaics in grazed woodlands driven by the alternation of plant facilitation and competition Plant Biology 1, 127-137.

19

Synthesis

O'Neill RV, De Angelis DI, Waide JB, Allen TFH (1986) A hierarchical concept of ecosystems. Princeton University Press, Princeton, NJ. Ouborg NJ, Piquot Y, Van Groenendael JM (1999) Population genetics, molecular markers and the study of dispersal in plants Journal of Ecology 87, 551-568. Ozinga WA, Hennekens SM, Schaminee JHJ, Bekker RM, Prinzing A, Bonn S, Poschlod P, Tackenberg O, Thompson K, Bakker JP, van Groenendael JM (2005a) Assessing the relative importance of dispersal in plant communities using an ecoinformatics approach Folia Geobotanica 40, 53-67. Ozinga WA, Schaminee JHJ, Bekker RM, Bonn S, Poschlod P, Tackenberg O, Bakker J, van Groenendael JM (2005b) Predictability of plant species composition from environmental conditions is constrained by dispersal limitation Oikos 108, 555561. Peterson EB, McCune B (2001) Diversity and succession of epiphytic macrolichen communities in low-elevation managed conifer forests in Western Oregon Journal of Vegetation Science 12, 511-524. Radies DN, Coxson DS (2004) Macrolichen colonization on 120-140 year old Tsuga heterophylla in wet temperate rainforests of central-interior British Columbia: a comparison of lichen response to even-aged versus old-growth stand structures Lichenologist 36, 235-247. Rose F (1992) Temperate forest management: its effects on bryophyte and lichen floras and habitats. In: Bryophytes and lichens in a changing environment (eds. Bates JW, Farmer AM) pp. 211-233. Clarendon Press, Oxford. Rychen N (2002) Räumlich explizite Verbreitungsmodelle für die epiphytische Flechte Lobaria pulmonaria in der Schweiz. Diploma thesis, University of Berne. Scheidegger C (1995) Early development of transplanted isidioid soredia of Lobaria pulmonaria in an endangered population. Lichenologist 27, 361-374. Scheidegger C, Clerc P, Dietrich M, Frei M, Groner U, Keller C, Roth I, Stofer S, Vust M (2002) Rote Liste der gefährdeten Arten der Schweiz: Baum- und erdbewohnende Flechten. Bundesamt für Umwelt, Wald und Landschaft (BUWAL), Bern, Eidg. Forschungsanstalt WSL, Birmensdorf, und Conservatoire et Jardin botaniques de la Ville de Genève CJBG. Scheidegger C, Goward T (2002) Monitoring lichens for conservation: Red Lists and conservation action plans. In: Monitoring with lichens - monitoring lichens (eds. Nimis PL, Scheidegger C, Wolseley PA) pp. 163-181. Kluwer Academic Publishers, Dordrecht. Schöller H (1997) Anthropogene Lebensraumgestaltung und Biodiversität von Flechten. In: Flechten - Geschichte, Biologie, Systematik, Ökologie, Naturschutz und kulturelle Bedeutung., Vol. 27 pp. 247. Senkenbergische Naturforschende Gesellschaft, Frankfurt a. M. Sillett SC, McCune B, Peck JE, Rambo TR (2000a) Four years of epiphyte colonization in Douglas-fir forest canopies The Bryologist 103, 661-669. Sillett SC, McCune B, Peck JE, Rambo TR, Ruchty A (2000b) Dispersal limitations of epiphytic lichens result in species dependent on old-growth forests Ecological Applications 10, 789 - 799. Snäll T, Ehrlen J, Rydin H (2005a) Colonization-extinction dynamics of an epiphyte metapopulation in a dynamic landscape Ecology 86, 106-115.

20

Synthesis

Snäll T, Pennanen J, Kivistö L, Hanski I (2005b) Modelling epiphyte metapopulation dynamics in a dynamic forest landscape Oikos 109, 209-222. Søchting U, Alstrup V (2002) Danish Lichen Checklist. Version 1. Available at http://www.bi.ku.dk/lichens/dkchecklist/. Tønsberg T, Gauslaa Y, Haugan R, Holien H, Timdal E (1996) The threatened macrolichens of Norway Sommerfeltia 23, 1-258. van Dobben HF (1996) Decline and recovery of epiphytic lichens in an agricultural area in The Netherlands (1900-1988) Nova Hedwigia 62, 477-485. Vittoz P (1998) Flore et végétation du Parc jurassien vaudois: typologie, écologie et dynamique des milieux. PhD thesis, University of Lausanne. Wagner HH, Werth S, Kalwij JM, Scheidegger C (in prep.) Modelling the persistence and genetic structure of an epiphytic lichen in a dynamic landscape Landscape Ecology. Walser JC (2004) Molecular evidence for limited dispersal of vegetative propagules in the epiphytic lichen Lobaria pulmonaria American Journal of Botany 91, 12731276. Walser JC, Gugerli F, Holderegger R, Kuonen D, Scheidegger C (2004) Recombination and clonal propagation in different populations of the lichen Lobaria pulmonaria Heredity 93, 322-329. Walser JC, Zoller S, Büchler U, Scheidegger C (2001) Species-specific detection of Lobaria pulmonaria (lichenized ascomycete) diaspores in litter samples trapped in snow cover Molecular Ecology 10, 2129-2138. Werth S, Wagner HH, Holderegger R, Kalwij JM, Scheidegger C (in prep.) Genetic diversity of an old-forest associated lichen is affected by stand-replacing disturbances Molecular Ecology. Wiens JA (1989) Spatial scaling in ecology Functional Ecology 3, 385-397. Wirth V (1995) Die Flechten Baden-Württembergs. Eugen Ulmer, Stuttgart. Wirth V, Schöller H, Scholz P, Ernst G, Feuerer T, Gnüchtel A, Hauck M, Jacobsen P, John V, Litterski B (1996) Rote Liste der Flechten (Lichenes) der Bundesrepublik Deutschland Schriftenreihe für Vegetationskunde 28, 307-368. Wu JG (2004) Effects of changing scale on landscape pattern analysis: scaling relations Landscape Ecology 19, 125-138. Yoshimura I (1971) The genus Lobaria of Eastern Asia Journal of the Hattori Botanical Laboratory 34, 231-364. Zoller S, Lutzoni F, Scheidegger C (1999) Genetic variation within and among populations of the threatened lichen Lobaria pulmonaria in Switzerland and implications for its conservation Molecular Ecology 8, 2049-2059.

21

Synthesis

22

Chapter 1 - Genetic diversity is affected by stand-replacing disturbances

Chapter 1

Title of the paper: Genetic diversity of an old-forest associated lichen is affected by stand-replacing disturbances

For submission to: Molecular Ecology

Authors: S. Werth, H. H. Wagner, R. Holderegger, J. M. Kalwij and C. Scheidegger

Running head: Genetic diversity is affected by stand-replacing disturbances

Keywords: Lichenised ascomycetes, stand-replacing disturbance, population history, landscape genetics, spatial autocorrelation, microsatellites, Lobaria pulmonaria

Author’s address: Section of Ecological Genetics, Swiss Federal Research Institute WSL, Zürcherstrasse 111, CH-8903 Birmensdorf, Switzerland

Word count: 7140

Abstract word count: 272

23


Abstract Old-forest associated lichens are commonly assumed to be negatively affected by tree logging or natural forest disturbances. However, in this study performed in a spruce-dominated sylvopastoral landscape in the Swiss Jura Mountains, we found that genetic diversity of the epiphytic old-forest lichen Lobaria pulmonaria depends on the type of disturbance. We collected 923 thalli from 41 sampling plots of 1 ha (demes) corresponding to the disturbance categories stand-replacing disturbance, intensive logging and uneven-aged forestry, and analysed the thalli at six mycobiont-specific microsatellite loci. We found evidence for multiple independent colonisations of demes located in areas affected by disturbance at forest stand level. Using spatial autocorrelation methods, the spatial scale of similar genetic structure can be determined, discriminating among the clonal and recombinant component of genetic variation. Spatial autocorrelation of gene diversity was strong at short distances up to 150 m in all three disturbance categories, with the strongest autocorrelation for demes affected by standreplacing disturbance. The spatial autocorrelation was predominantly attributed to clonal dispersal of vegetative propagules. After accounting for the clonal component, we did not find significant spatial autocorrelation. This pattern may indicate low dispersal ranges of clonal propagules, but random dispersal of sexual ascospores, indicating that there was no dispersal limitation of ascospores. Genetic diversity was highest in demes affected by intensive logging at forest-stand level, and lowest in demes affected by stand-replacing disturbance. Our study exemplified that genetic diversity of epiphytic lichen demes may not necessarily be impacted by stand-level disturbances for many centuries. Given that forest management retains large populations that serve as a diaspore source, genetic diversity of epiphytic lichens may be high even under intensive logging.

24


Introduction Genetic diversity of natural populations, an important component of biodiversity, may be affected by disturbance (Bradshaw 2004; Baucom et al. 2005; Volis et al. 2005). In forested systems, wind-throws and fires belong to the natural disturbance regime. In many parts of Central Europe, human influence has added various types of forest disturbances (Bradshaw 2004). The traditional sylvopastoral landscapes found in many mountainous areas of Europe are regularly disturbed by cattle grazing, timber logging and other forest management practices (Kirby et al. 1995; Gillet et al. 2002). At the same time, however, these systems often have a high species richness and may harbour e.g. populations of endangered old-forest species such as the epiphytic lichen Lobaria pulmonaria. If a disturbance event leads to a significant reduction in the habitat of a particular population, the local population size generally declines. This may lead to a population bottleneck, involving an instant loss of rare alleles (Hartl & Clark 1997). If population size remains small over an extended time period, rare alleles may further be lost due to genetic drift, leading to a continued decrease of genetic diversity over time (Nei et al. 1975), particularly in terms of allelic richness (Widmer & Lexer 2001). Subsequently, other population genetic consequences of small population sizes may occur, as the population enters an extinction vortex (Tanaka 2000; Frankham et al. 2002). In self-sterile organisms such as some lichen mycobionts, the likelihood that compatible mating types are present is lower in small than in large populations, and their absence may prevent sexual reproduction (Zoller et al. 1999; Taylor et al. 2000; Walser et al. 2004). On the other hand, if disturbance has caused the local extinction of a species, the habitat is open for recolonisation during succession. In many cases, the habitat is recolonised by few founders, leading to low genetic diversity in the newly founded population (founder event, Hartl & Clark 1997). However, as new individuals (or gametes in plants and fungi) enter the population over time, genetic diversity increases and the footprints of a severe bottleneck or of founder events are successively erased. Lichens with a long generation time may exhibit low population growth rates, which is one reason why lichen species are considered sensitive to disturbances such as logging (Gilbert 1977; Seaward 1982). Consequently, random sampling effects such as 25


bottlenecks or founder events in the population history should be imprinted in the genetic diversity of lichens for a comparatively long period. Recent studies in population genetics of lichens have mainly focused on large spatial scales, comparing populations among regions and continents (Högberg et al. 2002; Myllys et al. 2003; Printzen & Ekman 2003; Printzen et al. 2003; Walser et al. 2005). Very little is known about the genetic diversity, gene flow, and population dynamics of lichens within landscapes, and about the effect of disturbances on epiphytic lichen populations. To overcome this gap in knowledge, we performed a population genetic study at the landscape scale focussing on the comparison of demes of an epiphytic lichen within a sylvopastoral landscape where stand-replacing disturbances could be reconstructed (Kalwij et al. in press). Our model organism, the epiphytic lichen Lobaria pulmonaria, is described as an indicator lichen of forest canopy continuity in Europe (Rose 1992). The species is also thought to be dispersal-limited (Scheidegger et al. 1995; Walser et al. 2001). Since the haploid fungus L. pulmonaria produces both vegetative and sexual propagules (Yoshimura 1971), its genetic structure consists of a clonal as well as a recombinant component. The species is possibly selfsterile (heterothallic, Zoller et al. 1999), and has a generation time of 30 years or longer in central European populations (Scheidegger & Goward 2002). Owing to forest management with short rotation cycles and to air pollution, L. pulmonaria has suffered a severe decline in Central Europe during the last decades. The species is considered vulnerable in the Red List of Switzerland (Scheidegger et al. 2002) and is endangered (Hallingbäck & Martinsson 1987; Wirth 1995; Wirth et al. 1996; Türk & Hafellner 1999; Søchting & Alstrup 2002) or extinct in other European countries (Aptroot et al. 1999). In our study area, a sylvopastoral landscape in the Swiss Jura Mountains with a spatial extent of 46°28’-34’N and 06°06’-16’E at an elevation of 1300-1450 m, L. pulmonaria exhibits large population sizes. The carrier trees of L. pulmonaria, sycamore maple (Acer pseudoplatanus L.), and beech (Fagus sylvatica L.), are scarce and irregularly distributed within forest patches (Fig. 1-1, Kalwij et al. in press).

26


Figure 1-1. Distribution of Lobaria pulmonaria carrier trees in the study area in the in the Swiss Jura Mountains, following Kalwij et al. (in press). The mapping of L. pulmonaria carrier trees was based on circular sampling plots of 1 ha. (Digital data from Landeskarte der Schweiz: (c) Bundesamt für Landestopographie)

Forests in the entire study area are subject to tree-level forest management, i.e., unevenaged forestry, where individual trees of all size classes are harvested in order to maintain a constant size class distribution. Two forest stands in the north and west of the study area (Fig. 1-2) were additionally affected by 19th century stand-level disturbance (Kalwij et al. in press; Bolli et al. submitted). The northern stand was subjected to a 19th century stand-replacing disturbance, whereas the western stand underwent intensive, stand-level timber logging for charcoal production (Vittoz 1998). Large population sizes of L. pulmonaria were found both in the areas affected by stand-level disturbance and in some areas under tree-level forest management (Fig. 1-1, Kalwij et al. in press). This 27


raises the questions of how the recolonisation process of such disturbed habitats takes place, and whether gene flow is so rare that low genetic diversity is observed on formerly disturbed forest patches even after more than 100 years.

Figure 1-2. Demes of Lobaria pulmonaria in the Swiss Jura Mountains subjected to moleculargenetic analyses, and reconstructed stand-level disturbances (Kalwij et al. in press).

This paper addresses the following questions: (1) Were demes of L. pulmonaria, i.e. assemblages of thalli of the species within 1-ha sampling plots, in the area affected by a stand-replacing disturbance founded by only few independent colonisation events? (2) Is there spatial genetic structure within the population, and if so, does the structure differ between the three disturbance types? (3) Does the spatial genetic structure differ between clonal and recombinant components, and what is the spatial patch size of vegetative propagules and ascospores? (4) What is the effect of stand-level disturbances on different aspects of genetic diversity in L. pulmonaria demes and on genetic distance among 28


demes, and does the former effect depend on deme size in terms of the number of colonised trees? We discuss the results in the context of conservation strategies for L. pulmonaria.

Methods Sampling design Our sampling was based on Kalwij et al. (in press), where a random sample of 251 sampling plots of 1 ha in the wooded areas of the northern part of the study area was searched and the occurrence of L. pulmonaria and its carrier trees maple and beech were mapped. Using aerial photographs from 1933, Kalwij et al. (in press) delineated two severe stand-level disturbances reported in historical documents (c.f. Vittoz 1998): (i) Intensive, stand-level logging between 1850 and 1900, possibly selective for P. abies, for charcoal production, in the following referred to as “logging”, (ii) large-scale logging in 1870, followed by wind throw and, in the next year, a two-week fire, in the following named “stand-replacing disturbance”. We refer to plots under uneven-aged forestry as “undisturbed”, as they did not undergo stand-level disturbance. Our study plots (demes) were a subsample of the sampling plots of Kalwij et al. (in press), assuring a similar distribution of deme sizes in each area. In the two areas affected by stand-level disturbances, both disturbed and undisturbed demes, i.e. demes outside the reconstructed perimeter of disturbance (Fig. 1-2) were sampled. A total of 20 undisturbed 1-ha plots were investigated. From within the reconstructed perimeter of intensive logging, nine plots were chosen, whereas in the area affected by the stand-replacing disturbance, 12 plots were investigated. A hierarchical random sample of 923 thalli was collected from the 41 plots. Within each plot, all carrier trees of L. pulmonaria found by Kalwij et al. (in press) were searched for L. pulmonaria. A maximum of 24 thalli per deme were randomly selected from different trees. If there were fewer than 24 trees colonised by L. pulmonaria, multiple thalli were sampled from the same tree. If there were fewer than 24 thalli in a plot, every thallus found was included. From each thallus, we collected a single lobe tip of a minimum area of 3 cm2, minimising the impact of collecting on the sampled populations, but still providing enough material for DNA extractions. 29


Molecular analysis Sample preparation for DNA extractions followed Walser et al. (2003). Total DNA was isolated using the DNeasy 96 plant kit (Qiagen, Hilden, Germany) according to the manufacturer’s protocol. Six fungal microsatellite loci (Walser et al. 2003) were analysed. A modified set of PCR primers (table 3 in Walser et al. 2004) was used in two multiplex PCR reactions of 10 µL each. The first multiplex PCR contained 1 µL genomic DNA, 25 nM of the primers for LPu03 and LPu09, 125 nM of the primers for LPu15, 1× PCR buffer (Sigma, Saint Louis, MO, USA), 1.87 mM MgCl2, 50 µM of dNTPs (Promega, Madison, WI, USA), and 0.75 U DNA polymerase (Sigma, Saint Louis, MO, USA). The second multiplex differed in that primer concentration was 125 nM for all of the six primers used (loci LPu16, LPu20, LPu27). We used the fluorescent labelling of primers proposed by Walser et al (2004). The amplification protocol of both multiplex amplification reactions began with an initial denaturation at 94°C for 120s, 29 cycles of denaturation at 94°C for 60s, annealing at 57°C for 60s and extension at 72°C for 60s, followed by a final extension at 72°C for 45 minutes. All amplification reactions were performed with a PTC-100 thermal cycler (MJ Research, Waltham, USA). Fragment sizes of PCR products were determined on an ABI3100-avant automatic sequencer (Applied Biosystems, Foster City, CA, USA). Alleles were sized with an internal size standard (ROX 500, Applied Biosystems, Foster City, CA, USA). A reference sample of L. pulmonaria with known allele sizes was run on each plate to check repeatability of the allele sizing. Subsequently, genotyping was done using GENOTYPER 2.1 (Applied Biosystems, Foster City, CA, USA). Data analysis There were missing values for 13 thalli, and in 15 thalli, multiple alleles were found in at least one locus, most of them from the same tree. These latter 15 samples most likely consisted of more than one genetic individual of the haploid mycobiont. Alternatively, by chance, fertile dikaryotic and diploid hyphae within thalli might have 30


been used for extraction. We thus excluded 28 thalli from statistical analyses (895 thalli remaining). On every tree from which we collected L. pulmonaria, we estimated the total thallus area of L. pulmonaria in cm2. If the cumulative thallus area was larger than 1 dm2, we estimated it as the number of DINA4-size sheets covered by L. pulmonaria on the tree. Furthermore, we roughly estimated the number of thalli on each tree from which we collected L. pulmonaria, counting up to 24 thalli or, for larger numbers, using the following thallus number classes: 25-50, 51-75, 76-100, 101-150, 151-200, 200-300, 300400. We calculated the sum of the number of thalli or of the median class value over all trees on which we had sampled L. pulmonaria within each plot (deme census size), as well as the sum of the thallus area per plot (deme thallus area). Note that by this procedure, thallus number and thallus area were underestimated in plots where more than 24 trees were present, which was the case in three undisturbed plots (34, 30 and 26 trees colonised by L. pulmonaria) and in one logged plot (104 trees colonised by L. pulmonaria). To test for linkage disequilibrium in terms of a significant association between pairs of loci, we calculated Ȥ2-tests with P-values computed by Monte Carlo permutations with 2000 replicates using a reduced data set without recurrent genotypes. To get a measure of the strength of the association between two loci x and y, we calculated Cramér’s V2 (Agresti 1984): V 2

F 2 >n minr 1, c 1 @1 , where Ȥ2 is the chi-square

statistic, r corresponds to the number of alleles at locus x, c corresponds to the number of alleles at locus y, and n is the sample size. V2 is similar to a correlation coefficient, with values between 0 and 1 with larger values representing stronger association. We estimated the minimum number of colonisation events (C), as the number of alleles at the most variable locus for demes (Walser et al. 2003). To test for an association of C with disturbance, we computed a Ȥ2-test using R (Anonymous 2004). To test if there was spatial genetic structure in L. pulmonaria within the study area, and if the structure differed between disturbance levels, we calculated three different types of variograms for each level of disturbance using the methods described in Wagner et al. (2005): (1) a variogram of gene diversity H without accounting for recurrent genotypes, representing the overall spatial genetic structure, (2) a variogram of 31


genotype diversity D that reflects the clonal component, and (3) a variogram of H weighted for recurrent genotypes, reflecting the sexual component. A lag distance of 50 m was chosen, with the first distance class containing only pairs of samples from the same tree and the last distance class containing all pairs of samples separated by more than 450 m. Statistical significance of spatial autocorrelation was assessed for each variogram and each distance class using a one-sided permutation test with 100 permutations (Wagner et al. 2005). Exponential variogram models were fitted with the library gstat in R (Pebesma 2004). To quantify genetic diversity of demes for a comparison of different disturbance levels, we calculated Nei’s unbiased gene diversity (H, Nei 1978) for each deme as H

1§ r · ¨ ¦ hk ¸ , where hk r©k 1 ¹

mk n 2 (1 ¦ xik ) , xik is the frequency of the ith allele at the n 1 i 1

kth locus in a subpopulation, mk is the number of alleles at the kth locus, and r the number of loci analysed. Furthermore, we calculated the number of multilocus genotypes (G) and the percentage of multilocus genotypes per deme (M), i.e. the number of multilocus genotypes divided by the number of thalli. For the calculation of H, G and M, we wrote our own code in the R statistical package (Anonymous 2004), since we found no R library which had formulae implemented for haploid data. Our sampling design was unbalanced at the tree level, i.e., the number of thalli sampled per tree differed among plots. To overcome this problem, we based our analyses on averaged values from 1000 reduced data sets, created by randomly selecting one thallus per tree. We performed analysis of covariance (ANCOVA) of G, H and M as a function of the log-transformed number of L. pulmonaria carrier trees per deme as covariate and disturbance as factor (three levels). Furthermore, to see whether there were significant differences in G, M and H between pairs of disturbance types, we used analysis of variance (ANOVA) and calculated Tukey’s Honest Significant Difference as implemented in the R statistical package (formula TukeyHSD, R Development Core Team 2004). To assess if there were major differences in genetic distance among disturbance levels, we calculated Nei’s minimum genetic distance between demes (Dm, Takezaki & Nei 1996) as Dm = (Ja + Jb)/2 – Jab, where Ja and Jb are the average identity over loci within 32

subpopulation

a

and

b,

respectively,

calculated

as


Ja

1 r mj 2 ¦ ¦ xaij and J b r j i

1 r mj 2 ¦ ¦ xbij , where Jab is the average identity of r j i

subpopulations a and b, calculated as J ab

1 r mj ¦ ¦ xaij xbij . Here, xaij and xbij are the r j i

frequency of the ith allele at the jth locus in subpopulation a and b, respectively. Dm was also calculated with a self-written code in R (Anonymous 2004). To account for the unbalanced sampling design, we used an averaged value of Dm based on 1000 reduced data sets (see above). The pairwise genetic distances Dm were grouped as comparisons among disturbed plots within each disturbed area (d-d), comparisons of disturbed plots in the North and West subpopulations with plots in the East subpopulation (d-u), and comparisons among plots in the East subpopulation (u-u). Based on these data, an ANOVA of the effect of disturbance on genetic distance was calculated, which is similar to analyses performed by Hilfiker et al. (2004). We expected larger genetic distances in the d-d group as compared to the u-u group in the case of independent colonisations of the disturbed plots from different sources. One problem of this analysis is the nonindependence of pairwise genetic distances, which may potentially lead to false significances (type I error). In the case of significant results, this analysis therefore should not be over-interpreted. To assess if there were major differences in genetic distance among disturbance levels, principal coordinate analysis (PCoA) of genetic distance Dm was calculated for a two-dimensional ordination space with the library stats (function cmdscale, R Development Core Team 2004) of the R statistical package. For interpretation of the ordination, we used the deme-level disturbance classification.

Results

Non-spatial analysis Deme census sizes of Lobaria pulmonaria in terms of the mean number of thalli were highest in the area affected by stand-replacing disturbances, followed by the logged area (Table 1-1). The mean cumulative thallus area per plot was largest in the area affected by the stand-replacing disturbance (Table 1-1). We found apothecia of L. pulmonaria in few plots, most of which were not affected by stand-replacing or standlevel disturbances. 33


Table 1-1. The number of sampling plots investigated (Number of demes), average deme census size (Number of thalli), and thallus area (Area) of Lobaria pulmonaria in one-hectare plots from the Swiss Jura Mountains, in relationship to disturbance. Four plots have missing data, and are thus not shown. The table also lists the number of plots in which apothecia of L. pulmonaria were observed (Apothecia), with the total number of trees on which apothecia were observed in parentheses. Disturbance level of Area [dm2] Number of Apothecia Number of deme thalli demes Stand-replacing 572.9 930 1 (2) 11 disturbance Logging 405.6 810 2 (6) 9 Undisturbed (all areas) 413.8 717 5 (7) 17

We found some evidence of linkage disequilibrium, as in the Ȥ2-tests, four out of 15 pairs of microsatellite loci differed significantly from random association, and two further pairs were marginally significant (P < 0.1; Table 1-2). However, the associations were generally very weak, with a maximum V2 of 0.32. Also, the majority of pairwise comparisons of loci did not differ from random association, suggesting that loci were located at separate linkage groups. Table 1-2. Results of tests for linkage disequilibrium among pairwise combinations of six fungusspecific microsatellite loci in Lobaria pulmonaria from the Swiss Jura Mountains. In the upper diagonal, Cramér’s V2 is listed, which may be interpreted as a correlation coefficient (values close to one indicating a strong association of loci), and in the lower diagonal the P-value is listed, and underlined if the respective association is statistically significant. LPu27

LPu20

LPu27

LPu16 0.169

LPu15

LPu09

LPu03

0.221

0.104

0.099

0.052

0.191

0.201

0.110

0.137

0.184

0.180

0.079

0.201

0.321

LPu20

0.300

LPu16

0.022

0.217

LPu15

0.211

0.153

0.007

LPu09

0.535

0.957

0.057

0.016

LPu03

0.086

0.520

0.274

0.021

0.084 0.618

The minimum number of colonisation events C per deme ranged from one to eight. There was no statistically significant association between C and disturbance (Ȥ214 = 16.6, P > 0.05). Among the 919 L. pulmonaria samples analysed, we found 176 multilocus genotypes (M = 19.2 %). Gene diversity H increased significantly with the number of colonised trees per plot, whereas the proportion of multilocus genotypes M decreased 34


(Fig. 1-3). There was a significant effect of disturbance on H (P = 0.002; Table 1-3) and on M (P = 0.005; Table 1-3). H and M were highest in logged demes, intermediate in undisturbed demes, and lowest in demes affected by the stand-replacing disturbance. In the ANCOVA model testing the effects of disturbance and number of trees colonised by L. pulmonaria on the number of multilocus genotypes G, there was a significant difference in slopes between demes affected by the stand-replacing disturbance and the other two disturbance levels (P = 0.025; Table 1-3).

Table 1-3. Analysis of covariance testing the effect of disturbance (factor) and of the number of Lobaria pulmonaria trees per deme (number of trees = covariate) on the number of multilocus genotypes (G) per deme, the percentage of multilocus genotypes (M) in demes and Nei’s unbiased gene diversity (H). Diversity parameter G

M

H

Source of variation Number of trees Disturbance Number of trees × Disturbance Error Total Number of trees Disturbance Error Total Number of trees Disturbance Error Total

SS

df

MS

F

P

220.0

1

220.0

49.72

24.6 36.1

2 2

12.3 18.1

2.78 4.08

30 % of the study plot area was open water. Presence-absence data of corticolous, epiphytic macrolichen species were recorded on lower stems (< 2 m) of deciduous trees in a total of 69 study plots. The sampling of trunk habitats did not include the lowest parts of the trunks with a terricolous bryophyte cover, often with terricolous/muscicolous lichens such as Peltigera spp. and Cladonia spp. extending continuously upwards, often to stem heights of 3 cm. Macrolichens were defined as the species treated by Krog et al. (1994). Due to the great efforts needed to obtain reliable species determinations, the genus Cladonia was excluded from the analyses. Explanatory variables A total of 71 explanatory variables was used to characterise the plots. They were grouped into four sets of explanatory variables: {C}, a set of 15 macroclimatic,

Fig, 1. Map of the study area in central and southern Troms county, northern Norway. Dots indicate sampling locations.

- Epiphytic macrolichen communities along regional gradients in northern Norway microclimatic and topographic variables; {E}, a set consisting of 34 forest structure and growth substrate related environmental variables – referred to as environmental variables in the following; {H}, consisting of 10 human impact variables and {S}, composed of 12 spatial variables. Details about all explanatory variables are given in App. 1. Monthly and annual normal values (reference period 1961-1990; Aune 1993; Førland 1993) for precipitation and annual temperature sums were included in set {C}. Extrapolated data for the study plots were provided by The Norwegian Meteorological Institute, Klimaavdelingen. Climatic oceanicity was determined from a map in Moen (1999), distances to open sea and closest seashore were determined from maps of scale 1:500 000. Moisture and light conditions of the plots were estimated using a five-point scale, ranging from dry to wet and from low light to high light situations, respectively. Further variables were the mean slope angle of the study plot, plot elevation, plot aspect, a plot heat index (Parker 1988), a 15 point irradiation index with highest values for S and SSW exposure and aspect unfavourability, i.e. the deviation of aspect from SSW exposure (T. Økland 1996). Most of these latter variables characterize the microclimate of the plots. In the set of environmental variables {E}, a rock suitability index, giving an approximate indication of the quality of a given study plot’s rock habitats for lichen growth was included. This was done since some lichens which are predominantly epiphytic in central or southern Europe (Wirth 1995) frequently grow on both rocks and bark substrates in northern Norway (Krog et al. 1994), e.g. Degelia plumbea, Lobaria scrobiculata, L. pulmonaria and Pannaria conoplea. Suitable rock substrates can be important as a refuge of such lichen species in the case of forest disturbance. Based on the number and diameter of tree stems (DBH > 5 cm) in a 100-m2 subplot, tree density and basal area were calculated. Tree species composition was identified as presence-absence of species. The type of deciduous forest was determined and coded as binary variables. Decay class (Linder et al. 1997), diameter, number, density, position and decay class of dead wood (basal diameter > 10 cm) was recorded in 100-m2 subplots. The number and density of all young trees (DBH < 5 cm) present in 100-m2 subplots was recorded. Distances to the closest river and/or lake were determined from topographic maps, because they were assumed to be indicative of the moisture level of the study plots. A landscape patchiness index was defined by means of the sum of the length of boundaries between forest and non-forested mires and lakes, as taken from topographical maps in circular areas of 500-m radius around the study plots. In the set of human impact related variables {H}, the

201

presence or absence of Picea spp. or Larix spp. plantations in visible distance from the study plots was recorded. In the 500-m radius area mentioned above, the distance to and length of outlines of structures related to human settlements and of clearly human induced deforestation were estimated from topographic maps. Recently, parameters similar to the latter were used successfully to define an index of human impact by Gombert et al. (2004). The overall human impact on a plot was estimated using criteria suggested by Trass et al. (1999). For our purpose, the 13 point human impact scale of Trass et al. (1999) was reversed, with the starting point of 1. Note that the three highest human impact categories were missing in our data set. To enable detection of complex, large-scale spatial patterns, latitude and longitude, and their quadratic and cubic combinations defining a polynomial three-dimensional trend surface were included, as recommended by Borcard et al. (1992). In addition, geographic location of plots was included as binary variables, defined by plot affiliations to the municipalities of Tromsø, Storfjord and Målselv. A bark sample was collected from a specimen of the dominant tree species in each study plot. Bark pH was measured with a pH-meter after soaking bark samples in double distilled water for three hours. The age of the largest deciduous tree in a study plot was determined from a core taken 10 cm above ground level. Statistical analysis We performed DCA (Detrended Correspondence Analysis) ordination (Hill & Gauch 1980) to reveal the main gradient structure of the data set. Global nonmetric multidimensional scaling (Kruskal 1964) gave results comparable to DCA and thus corroborated DCA results (R. Økland 1996). Kendall’s Y correlation coefficients between environmental variables and DCA axes and their significance was calculated (Sokal & Rohlf 1995). Since we were unable to interpret DCA axes three and four in a biologically meaningful way, results are only shown for the first two axes. All explanatory variables recorded on a continuous measurement scale were transformed to zero skewness, followed by ranging (Økland et al. 2001). As a prelude to variation partitioning, variables were selected separately for each of the four sets by the forward selection procedure in CANOCO, followed by Monte Carlo permutation tests. The threshold statistical significance value for inclusion of a variable in a particular set of explanatory variables was Bonferroni corrected to F = 0.05. Statistically significant variables were checked for collinearity, and the rejection criterion was a variance inflation factor >5.

202

Werth, S. et al.

Using partial CCA (ter Braak 1988), we partitioned variance in macrolichen community composition onto the four sets of explanatory variables following Økland (2003). For all statistical analyses, the programs CANOCO for Windows 4.0 (ter Braak & Šmilauer 1998) and R software Version 2.0.0 (Anon. 2004) were used.

Results A total of 71 epiphytic macrolichen species was found in the 69 study plots (Table 1). Gradient lengths of the first two DCA axes were 1.950 and 1.899, implying that two short gradients were present in the data. The first DCA axis was significantly related to Betula pubescens forest, poor in nutrients, and to the study plots’ location in Tromsø municipality in the coastal part of the study area (Table 2, Fig. 2). DCA axis two was significantly correlated with latitude, combinations of latitude and longitude, oceanicity and the amount of precipitation (see also App. 2). Most of the variables significantly related to DCA axis two were strongly correlated with each other (Fig. 2). Species optima for lichens associated with green algal photobionts were separated from optima of lichens associated with cyanobacteria along DCA axes 1 and 2 (Fig. 3).

Fig. 2. DCA ordination diagram of macrolichen communities in deciduous forests of northern Norway. The biplot shows study plots () and significantly correlated environmental variables (arrows and X) along the first two axes.

Fig. 3. DCA ordination diagram of macrolichen communities in deciduous forests of northern Norway showing the species optima along the first two axes. Cyanolichens are underlined, tripartite lichens underlined and in italics; the remaining macrolichen species are chlorolichens.

Fig. 4. Fraction of the total variation explained (FTVE) and shared variance in epiphytic macrolichen communities of northern Norway, partitioned to four sets of explanatory variables: forest and environmental variables {E}, macroclimate, microclimate and topography {C}, human impact {H} and spatial variables {S}. Arrows among variable sets indicate their shared FTVE. Arrows pointing to macrolichen species composition show the fraction of the total variance which was explained by a respective set of explanatory variables, when covariance with all remaining sets had been removed. Dashed lines indicate that a specific FTVE was not statistically significant.

- Epiphytic macrolichen communities along regional gradients in northern Norway -

203

Table 1. Species list. The table includes abbreviations (Abbr.), thermophily (T), growth form (GF), photobiont (PB) and frequencies in coastal (35) and inland (34) study plots. The symbol ‘+’ indicates that a species is thermophilic, i.e. restricted to the middle boreal zone, with the symbol ‘c’ added in cases where only occurrences in corticolous habitats can be considered thermophilic. For functional groups, A = alectorioid; F = foliose; P = shrub-formed pendulous and S = squamulose lichen. For photobionts, C = cyanobiont and G = chlorobiont. Taxon

Abbr.

Bryoria capillaris Bryoria fuscescens Bryoria glabra Bryoria implexa Bryoria simplicior Collema furfuraceum Collema fasciculare Collema nigrescens Evernia prunastri Hypogymnia austerodes Hypogymnia bitteri Hypogymnia physodes Hypogymnia tubulosa Imshaugia aleurites Leptogium saturninum Lobaria amplissima Lobaria hallii Lobaria pulmonaria Lobaria scrobiculata Melanelia exasperata Melanelia exasperatula Melanelia fuliginosa Melanelia olivacea Melanelia subaurifera Nephroma bellum Nephroma expallidum Nephroma parile Nephroma resupinatum Pannaria pezizoides Parmelia omphalodes Parmelia saxatilis Parmelia sulcata Parmeliopsis ambigua Parmeliopsis hyperopta Peltigera aphthosa Peltigera canina Peltigera collina Peltigera didactyla Peltigera degenii Peltigera leucophlebia Peltigera malacea Peltigera membranacea Peltigera neopolydactyla Peltigera ponojensis Peltigera praetextata Peltigera retifoveata Peltigera rufescens Phaeophyscia ciliata Phaeophyscia nigricans Phaeophyscia orbicularis Phaeophyscia sciastra Physcia adscendens Physcia aipolia Physcia caesia Physcia dubia Physcia stellaris Physcia tenella Physconia distorta Physconia perisidiosa Platismatia glauca Protopannaria pezizoides Psoroma hypnorum Ramalina farinacea Sphaerophorus globosus Tuckermanopsis chlorophylla Tuckermannopsis sepincola Usnea subfloridana Vulpicida pinastri Xanthoria candelaria Xanthoria elegans Xanthoria parietina

Bry cap Bry fus Bry gla Bry imp Bry sim Col fas Col fur Col nig Eve pru Hyp aus Hyp bit Hyp phy Hyp tub Ims ale Lep sat Lob amp Lob hal Lob pul Lob scr Mel eta Mel ela Mel ful Mel oli Mel sub Nep bel Nep exp Nep par Nep res Par omp Par sax Par sul Par tri Par amb Par hyp Pel aph Pel can Pel col Pel deg Pel did Pel leu Pel mal Pel mem Pel neo Pel pon Pel pra Pel ret Pel ruf Pha cil Pha nig Pha orb Pha sci Phy ads Phy aip Phy cae Phy dub Phy ste Phy ten Pho dis Pho per Pla gla Pro pez Pso hyp Ram far Sph glo Cet chl Cet sep Usn sub Vul pin Xan can Xan ele Xan par

T

+ + + + + + + + + + + + + + +

+

+

+ + +

+c + +

+

+

+c

GF

PB

A A A A A F F F P F F F F F F F F F F F F F F F F F F F F F F S F F F F F F F F F F F F F F F F F F F F F F F F F F F F S S P P F F P F F F F

G G G G G C C C G G G G G G C CG C CG C G G G G G C CG C C G G G C G G CG C C C C CG C C C C C C C G G G G G G G G G G G G G C CG G G G G G G G G G

Frequency Coast Inland 1.4 27.5 2.9 2.9 15.9 0.0 0.0 1.4 2.9 0.0 0.0 46.4 10.1 1.4 13.0 0.0 1.4 1.4 13.0 11.6 1.4 2.9 47.8 27.5 30.4 1.4 26.1 10.1 2.9 21.7 50.7 5.8 49.3 39.1 0.0 2.9 1.4 1.4 1.4 1.4 0.0 21.7 2.9 1.4 8.7 0.0 0.0 2.9 1.4 1.4 1.4 1.4 20.3 1.4 1.4 8.7 2.9 8.7 1.4 4.3 10.1 7.2 2.9 1.4 18.8 2.9 8.7 33.3 2.9 0.0 0.0

0.0 23.2 1.4 0.0 29.0 1.4 11.6 1.4 1.4 5.8 4.3 49.3 4.3 4.3 15.9 1.4 7.2 2.9 29.0 14.5 0.0 0.0 49.3 37.7 36.2 1.4 30.4 15.9 1.4 15.9 49.3 10.1 47.8 40.6 5.8 7.2 7.2 0.0 2.9 1.4 2.9 11.6 8.7 0.0 10.1 1.4 1.4 0.0 0.0 1.4 0.0 0.0 27.5 0.0 0.0 14.5 0.0 11.6 0.0 5.8 7.2 1.4 1.4 2.9 7.2 2.9 7.2 47.8 0.0 1.4 1.4

204

Werth, S. et al.

Table 2. The ten variables most strongly correlated (Kendall’s Y) with two DCA axes. Asterisks indicate that a respective correlation is significant at a Bonferroni-corrected overall significance level F = 0.05. Variable

Explanation

Betula_herb Tromso Aln_inc Dist_town TM_SUM4 Bet_pub Sal_cap Sor_auc Tree_age Open_forest X X2 X3 X2Y XY XY2 Pr_year Pr_S_T4 Pr_S_T0 Oceanicity

Study plot situated in birch forest poor in nutrients Study plot located in Tromsø municipality Presence of Alnus incana Distance to closest town Temperature sum of all months exceeding 4°C mean air temperature Presence of Betula pubescens Presence of Salix caprea Presence of Sorbus aucuparia Tree age Proportion of open area to forested area Longitude Longitude ^2 Longitude ^3 Longitude ^2*Latitude Longitude *Latitude Longitude *Latitude^2 Annual normal precipitation Sum of precipitation in all months exceeding 4 °C in mean air temperature Sum of precipitation in all months exceeding 0 °C in mean air temperature Oceanicity of climate

Of 71 explanatory variables, 11 were statistically significant and were therefore used for variation partitioning. Significant variables included in {C} were oceanicity, distance to open sea, precipitation and temperature sum; in {E} the suitability of rock habitats, the presence of Alnus incana and Betula pubescens forest with a field layer dominated by herbs; in {H} the overall human impact index only and in {S} longitude and location of study plots in Tromsø and Storfjord municipalities. The full model (with all 11 significant variables) explained 26 % of the total inertia (3.44 inertia units, IU). The largest amount of variance was explained by the set of macroclimatic variables (48.0%), followed by the set of spatial variables (33.7%) and environmental variables (30.3%). It is noteworthy that human impact only explained 9.1 % of the variance (Table 3). None of the variable sets alone explained statistically significant amounts of variance, when variance due to the remaining sets had been partialled out (Table 3; Fig. 4). The variance attributable to second, third and fourth order partial intersections of variable sets was marginal (< 1.7%). The bark pH ranges of most deciduous tree species were higher than expected (i.e. ca. 2 pH units; Table 4). There was no significant difference in the bark pH of Alnus incana and Populus tremula (t-test, p = 0.061). The maximum age found for tree cores of Salix pentandra, S. caprea and Prunus padus was ca. 70 years, whereas that of Betula pubescens was 180 years.

Kendall’s Tau

Axis

*–0.39 *–0.29 0.27 –0.23 0.23 –0.21 0.19 0.19 –0.18 –0.16 *–0.38 *0.38 *–0.38 *–0.37 *–0.36 *–0.34 *0.34 *0.33 *0.31 *0.30

DCA1 DCA1 DCA1 DCA1 DCA1 DCA1 DCA1 DCA1 DCA1 DCA1 DCA2 DCA2 DCA2 DCA2 DCA2 DCA2 DCA2 DCA2 DCA2 DCA2

Table 3. Partitioning of the variation in epiphytic macrolichen communities on the four sets of explanatory variables: forest and environmental variables {E}, macroclimate, microclimate and topography {C}, human impact {H} and spatial variables {S}. The denotation of the respective term calculated (Denotation), the sum of all canonical eigenvalues (EV), the fraction of the total variance explained (FTVE), and the p-value of the respective term (p) are shown in the table. Variation explained (EV) is given in inertia units, IU (total inertia was 3.44 IU, total variance explained was 0.909 IU) as well as fraction of the total variation explained (FTVE). The symbols ‘’ and ‘’ indicate unions and intersections of variable sets, while ‘ | ’ stands for the Boolean operator NOT; n.p. no test performed. Asterisks indicate that a respective fraction of total explained variance is statistically significant at a Bonferronicorrected F of 0.05. Denotation E C S H E | (C H S) C | (E H S) S | (E H C) H | (E S C) (E S) | (C H) (E H) | (C S) (E C) | (H S) (S H) | (E C) (S C) | (E H) (H C) | (E S) (C H S) | E (E H S) | C (E C H) | S (E S C) | H ECHS

EV

FTVE

p

0.276 0.436 0.306 0.083 0.214 0.304 0.184 0.058 0.015 0.000 0.023 0.002 0.080 0.000 0.007 0.001 0.007 0.007 0.010

*30.3 *48.0 *33.7 9.1 23.5 33.4 20.2 6.4 1.6 0.0 0.3 0.0 0.8 0.0 0.8 0.1 0.8 0.8 1.1

< 0.001 < 0.001 < 0.001 0.011 0.005 0.005 0.015 0.302 n.p. n.p. n.p. n.p. n.p. n.p. n.p. n.p. n.p. n.p. n.p.

- Epiphytic macrolichen communities along regional gradients in northern Norway Discussion Our DCA and partial CCA results show that human impact is not a key factor structuring epiphytic macrolichen communities in deciduous forests of central Troms county. This finding is in accordance with that of a study of forest vascular plant vegetation in the northwestern United States in which clearfelling accounted for only 2 % of the total variation explained (Ohmann & Spies 1998). The effect of logging on lichen species depends on population size and habitat preferences. Rare species are at higher risk of becoming locally extinct because of logging than common species. Logging will probably have a long-term negative effect only on species restricted to certain habitat types; species that are rare and irregularly distributed in a landscape. Dispersal ability may become a restricting factor if logging not only reduces the total area of available habitat, but also number of favourable landscape patches as has been hypothesised for deciduous forests of central Europe (Wirth 1999). Proposals for lichen species that can be used as indicators of human impact on forests (Pfefferkorn & Türk 1996) or ecological continuity of forests (Rose 1992) are rarely built on statistical analysis of empirical data. This perpetuates the problem of confounding habitat qualities with human impacts such as logging regimes. Forests that have remained undisturbed for a longer time usually contain many specific habitats suitable for the growth of different epiphytic lichens, such as large old trees, logs and snags (Peterson & McCune 2001; Jüriado et al. 2003). This is mainly due to a shift in forest structure with time since disturbance, and may thus be independent of logging per se. The results of a study performed in central Norwegian coastal spruce forests in which the covariance between environmental and human effects was corrected for led to the conclusion that moderate selective logging did not have significant effects on several epiphytic macrolichen species (Rolstad et al. 2001). The latter study included several old-growth associated lichens belonging to the Lobarion community, which were classified as indicators of forest continuity (Rose 1992; Kuusinen 1996) or hemerophoby (Trass et al. 1999). Likewise, forest thinning did not have the anticipated negative effect on old-growth dependent cyanolichens, but rather led to a slight increase of alectorioid and drought-resistant lichens in a study performed in the northwestern United States (Peterson & McCune 2001). However, the latter form of forest management affected neither the density nor the frequency of large old trees, which may act as source of lichen propagules for enhanced colonisation of trees remaining after management. In contrast, in a study focusing on branch macrolichen communities,

205

Table 4. pH value of deciduous tree bark in the study area in Troms county, showing species (Tree species), minimum pH (Min), mean pH (Mean), maximum pH value measured (Max), number of bark samples analysed to determine the latter values (n), maximum age among all tree cores of a species (Maxage) and the number of tree cores taken (Cores). Tree species

Min

Mean

Max

n

Alnus incana Betula pubescens Populus tremula Prunus padus Salix caprea Salix myrsinifolia Salix pentandra Sorbus aucuparia

4.5 4.1 5.0 5.1 4.8 4.8 5.5 4.6

5.3 4.8 5.7 5.5 5.3 5.9 5.7 5.8

6.7 6.1 6.9 5.8 6.9 6.6 6.1 6.4

26 33 8 3 18 25 15 21

Maxage Cores 103 180 92 64 76 106 63 114

17 48 7 3 12 9 3 12

cyanolichens were common in old-growth stands, but virtually absent in secondary, even-aged stands (Radies & Coxson 2004). Logging is likely to lead to an increase of solar radiation on remnant trees to levels destructive for several old forest lichens (Gauslaa & Solhaug 2000). In valley bottom sites in northern Norway that receive a low amount of solar irradiation due to a low sun angle, this effect is probably not of importance; explaining the low importance of human impact found in our study. The climate of Troms county can be characterised as oceanic to suboceanic, with the climatically most favourable sites for lichens belonging to the Lobarionpulmonariae alliance situated in inland valley bottoms. The habitat of the Lobarion is closely related to macroclimate, in particular humidity (Barkman 1958). In a favourable oceanic climate, Lobarion species more quickly colonise young stands than in drier climates (Peterson & McCune 2001) and may, therefore, be more resilient to logging than under growth conditions further from their climatic optimum (Anonby 1994). In contrast, for thermophilic lichen species reaching their northern distribution limits in northern Norway, logging and other human impacts were assumed to have strong effects. However, no evidence of such an effect is found in the present data set. Both the DCA and the CCA results show strong relationships between epiphytic macrolichen species composition and macroclimatic factors. DCA axis two reflected the gradient from moist coastal to drier interior sites. Typical coastal species with low optima along this axis are Xanthoria parietina, X. candelaria and Physcia tenella, while Lobaria amplissima, L. pulmonaria, L. hallii and Collema fasciculare are characteristic interior species. Fundamental importance of macroclimate for lichen communities accords with the findings of McCune et al. (1997) and Peterson & McCune (2001) from the USA. In a Swedish investigation of epiphyllic algae on spruce needles and epiphytic lichens on trunks of Pinus

206

Werth, S. et al.

sylvestris, however, macroclimatic variation only accounted for 14.1 % of the total variation in species composition (Liu & Bråkenhielm 1995). Oceanicity, precipitation and local temperature sums are the main determinants of macrolichen community composition in our study. There may be several reasons for this. First, there is a strong climatic gradient in the study area from oceanic coastal sites with low temperature sums and high annual precipitation to inland sites characterised by a more continental climate with high temperature sums and lower annual precipitation (Aune 1993; Førland 1993). Thermophilic lichens in northern Norway may experience a trade-off between demands for precipitation and temperature – demands for precipitation favour a coastal distribution, temperature requirements are satisfied in climatically favourable interior sites. Species with an oceanic distribution in central Europe such as Collema fasciculare, C. nigrescens, Degelia plumbea, Lobaria amplissima, L. pulmonaria, L. scrobiculata, Pannaria conoplea and Sphaerophorus globosus (Schauer 1965; Wirth 1995) may be limited by low temperature sums in coastal sites of northern Fennoscandia. In Norway, at latitudes exceeding 68° N these oceanic species are restricted to locally moist habitats at the most continental sites, where demands for temperature and humidity are both met (Elvebakk & Sandvik 1980). Some lichen species are good indicators of favourable local climatic conditions in northern Norway (Ingebrigtsen unpubl.). As many as 26 of the recorded species can be considered thermophilic in northern Fennoscandia, in the sense that their distributions are limited to the thermically defined middle boreal zone (Moen 1999). These species are confined to interior sites characterised by high summer temperatures. The presence of a thermophilic species group is also reflected in the DCA ordination, in which cyanolichens and tripartite lichens (that contain both cyanobacteria and green algae as photobionts), many of which are considered thermophilic in northern Norway (Ingebrigtsen unpubl.), obtain low optima on DCA axis one. Contrary to the suggestions of Barkman (1958), species typical of the nitrophilous Xanthorion parietinae alliance do not seem to be thermally limited in Troms. Many of these species may, however, be favoured by nutrient richness, e.g. brought about by shoreline dust: thus Xanthoria parietina, X. candelaria, Physcia tenella and Melanelia exasperatula were only found at coastal sites in Troms, where they occurred sparsely as epiphytes close to rock habitats by the shoreline where the species were abundantly present. Another major gradient in epiphytic macrolichen communities found in this study by DCA ordination was due to environmental variation, and in CCA, environ-

mental variables accounted for 30.3 % of the total variation explained. Substrate-related variation in epiphyte communities has been emphasized in many local scale studies (Oksanen 1988; Burgaz et al. 1994). Variation due to forest stand properties, carrier tree species and bark properties is often considered significant (Bates 1992; Hyvärinen et al. 1992; Gustafsson & Eriksson 1995). In our study, the presence of Alnus incana was significantly influencing epiphytic macrolichen communities. The bark pH conditions of A. incana are commonly considered as acidic. However, in calcareous areas or in habitats influenced by dust, tree species normally with acidic bark may have higher pH and epiphyte communities characteristic of rich bark habitats may establish (Du Rietz 1944). This effect was also prevalent in the study area, where the bark pH values of A. incana did not differ significantly from those of the rich bark tree species Populus tremula. These particular bark conditions, as well as the ability of A. incana to live longer than other floodplain tree species (e.g. Salix pentandra, Prunus padus), may account for its particular importance for epiphytic macrolichen communities. The significance of a rock suitability index in CCA indicates that suitable rock habitats are important for macrolichen communities in the study area, where they might operate as lichen refuges from forest disturbance. Spatial variation was also proven important by DCA ordination and variation partitioning, accounting for 33.7% of the total variation. This is comparable to the 27.2% of the total variation accounted for by spatial variables in the study of lichen communities on Pinus sylvestris in Sweden by Liu & Bråkenhielm (1995). In the present study, geographic location was more important for lichen communities than geographic extent of the study area, which is in accordance with the results of Ohmann & Spies (1998). Organisms are distributed neither regularly nor randomly within communities, but respond to gradients, or aggregate in patches (Legendre & Fortin 1989). Future research should be directed towards the spatial distribution of species and to their dispersal biology, since the latter is crucial for our understanding of immigration and species turnover in communities.

Acknowledgements. We thank the Norwegian Meteorological Institute for providing climate data and Tove Midtun, University of Tromsø, for help with the figures. Furthermore, we thank Rolf Holderegger for his very helpful comments on an earlier draft of the manuscript. Funding for this project was provided by the Norwegian Research Council, the German Academic Exchange Service, the Kometen Fund (University of Tromsø) and the Norwegian Institute of Nature Research. This study was supported by the Swiss National Science Foundation (SNF) under the NCCR Plant Survival.

- Epiphytic macrolichen communities along regional gradients in northern Norway References Anon. 2004. R Version 2.0.0 for Windows. URL http://cran.rproject.org. The R foundation for statistical computing. Anonby, J.E. 1994. Epifyttisk lavflora i vestnorsk furuskog – sammenligning av en naturskog og en kulturskog. Blyttia 2: 81-87. Aune, B. 1993. Air temperature normals, normal period 19611990. Det Norske Meteorologiske Institutt, Oslo, NO. Barkman, J.J. 1958. Phytosociology and ecology of cryptogamic epiphytes: including a taxonomic survey and description of their vegetation units in Europe. Gorcum, Assen, NL. Bates, J.W. 1992. Influence of chemical and physical factors on Quercus and Fraxinus epiphytes at Loch Sunart, western Scotland: a multivariate analysis. J. Ecol. 80: 163-179. Birks, H.J.B. 1993. Is the hypothesis of survival on glacial nunataks necessary to explain the present-day distributions of Norwegian mountain plants? Phytocoenologia 23: 399-426. Borcard, D., Legendre, P. & Drapeau, P. 1992. Partialling out the spatial component of ecological variation. Ecology 73: 1045-1055. Burgaz, A.R., Fuertes, E. & Escudero, A. 1994. Ecology of cryptogamic epiphytes and their communities in deciduous forests in mediterranean Spain. Vegetatio 112: 73-86. Du Rietz, G.E. 1944. Om fattigbark- og rikbarksamhällen. Sven. Bot. Tidskr. 39: 147-150. Elvebakk, A. & Sandvik, S.M. 1980. Ny funn av vanleg blåfiltlav (Parmeliella plumbea) i Troms. Polarflokken 4: 87-94. Førland, E.J. 1993. Precipitation normals, normal period 1961-1990. Det Norske Meteorologiske Institutt, Oslo, NO. Fremstad, E. 1997. Vegetasjonstyper i Norge. Norsk Institutt for Naturforskning, Trondheim, NO. Gauslaa, Y. & Solhaug, K.A. 2000. High-light-intensity damage to the foliose lichen Lobaria pulmonaria within a natural forest: the applicability of chlorophyll fluorescence methods. Lichenologist 32: 271-289. Gombert, S., Asta, J. & Seaward, M.R.D. 2004. Assessment of lichen diversity by index of atmospheric purity (IAP), index of human impact (IHI) and other environmental factors in an urban area (Grenoble, southeast France). Sci. Total Environ. 324: 183-199. Graae, B.J., Økland, R.H., Petersen, P.M., Jensen, K. & Fritzbøger, B. 2004. Influence of historical, geographical and environmental variables on understorey composition and richness in Danish forests. J. Veg. Sci. 15: 465-474. Gustafsson, L. & Eriksson, I. 1995. Factors of importance for the epiphytic vegetation of aspen Populus tremula with special emphasis on bark chemistry and soil chemistry. J. Appl. Ecol. 32: 412-424. Hedenås, H. & Ericson, L. 2003. Response of epiphytic lichens on Populus tremula in a selective cutting experiment. Ecol. Appl. 13: 1124-1134. Hill, M.O. & Gauch, H.G.J. 1980. Detrended correspondence analysis: an improved ordination technique. Vegetatio 42: 47-58.

207

Hyvärinen, M., Halonen, P. & Kauppi, M. 1992. Influence of stand age and structure on the epiphytic lichen vegetation in the middle-boreal forests of Finland. Lichenologist 24: 165-180. Jüriado, I., Paal, J. & Liira, J. 2003. Epiphytic and epixylic lichen species diversity in Estonian natural forests. Biodivers. Conserv. 12: 1587-1607. Krog, H., Østhagen, H. & Tønsberg, T. 1994. Lavflora: norske busk- og bladlav. Universitetetsforlag, Oslo, NO. Kruskal, J.B. 1964. Nonmetric multidimensional scaling – a numerical method. Psychometrika 29: 115-129. Kuusinen, M. 1996. Cyanobacterial macrolichens on Populus tremula as indicators of forest continuity in Finland. Biol. Conserv. 75: 43-49. Legendre, P. & Fortin, M.J. 1989. Spatial pattern and ecological analysis. Vegetatio 80: 107-138. Lid, J. & Lid, D.T. 1994. Norsk flora. Norske Samlaget, Oslo, NO. Linder, P., Elfving, B. & Zackrisson, O. 1997. Stand structure and successional trends in virgin boreal forest reserves in Sweden. For. Ecol. Manage. 98: 17-33. Liu, Q.H. & Bråkenhielm, S. 1995. A statistical approach to decompose ecological variation. Water Air Soil Pollut. 85: 1587-1592. McCune, B. & Lesica, P. 1992. The trade-off between species capture and quantitative accuracy in ecological inventory of lichens and bryophytes in forests in Montana. Bryologist 95: 296-304. McCune, B., Dey, J., Peck, J., Heiman, K. & Will-Wolf, S. 1997. Regional gradients in lichen communities of the southeast United States. Bryologist 100: 145-158. Meffe, G.K. & Carroll, C.R. 1997. Principles of conservation biology. Sinauer, Sunderland, MS, US. Moen, A. 1999. National atlas of Norway: vegetation. Norwegian Mapping Authority, Hønefoss, NO. Ohmann, J.L. & Spies, T.A. 1998. Regional gradient analysis and spatial pattern of woody plant communities of Oregon forests. Ecol. Monogr. 68: 151-182. Økland, R.H. 1990. Vegetation ecology: theory, methods and applications with reference to Fennoscandia. Sommerfeltia Suppl. 1: 1-233. Økland, R.H. 1996. Are ordination and constrained ordination alternative or complementary strategies in general ecological studies? J. Veg. Sci. 7: 289-292. Økland, R.H. 2003. Partitioning the variation in a plot-byspecies data matrix that is related to n sets of explanatory variables. J. Veg. Sci. 14: 693-700. Økland, R.H. & Eilertsen, O. 1994. Canonical correspondence analysis with variation partitioning: some comments and an application. J. Veg. Sci. 5: 117-126. Økland, R.H., Økland, T. & Rydgren, K. 2001. Vegetationenvironment relationships of boreal spruce swamp forests in Østmarka Nature Reserve, SE Norway. Sommerfeltia 29: 129. Økland, T. 1996. Vegetation-environment relationships of boreal spruce forests in ten monitoring reference areas in Norway. Sommerfeltia 22: 1-349. Oksanen, J. 1988. Impact of habitat, substrate and microsite classes on the epiphyte vegetation: interpretation using

Werth, S. et al.

exploratory and canonical correspondence analysis. Ann. Bot. Fenn. 25: 59-71. Parker, K.C. 1988. Environmental relationships and vegetation associates of columnar cacti in the northern Sonoran Desert. Vegetatio 78: 125-140. Peterson, E.B. & McCune, B. 2001. Diversity and succession of epiphytic macrolichen communities in low-elevation managed conifer forests in western Oregon. J. Veg. Sci. 12: 511-524. Pfefferkorn, V. & Türk, R. 1996. Flechten als Zeiger des Hemerobiegrades in terrestrischen Biotopen. Sauteria 8: 181-192. Qian, H., Klinka, K., Økland, R.H., Krestov, P. & Kayahara, G.J. 2003. Understorey vegetation in boreal Picea mariana and Populus tremuloides stands in British Columbia. J. Veg. Sci. 14: 173-184. Radies, D.N. & Coxson, D.S. 2004. Macrolichen colonization on 120-140 year old Tsuga heterophylla in wet temperate rainforests of central-interior British Columbia: a comparison of lichen response to even-aged versus old-growth stand structures. Lichenologist 36: 235-247. Richardson, D.H.S. 1992. Pollution monitoring with lichens. Richmond, Slough, UK. Rolstad, J., Gjerde, I., Storaunet, K.O. & Rolstad, E. 2001. Epiphytic lichens in Norwegian coastal spruce forest: historic logging and present forest structure. Ecol. Appl. 11: 421-436. Rose, F. 1992. Temperate forest management: its effects on bryophyte and lichen floras and habitats. In: Bates, J.W. & Farmer, A.M. (eds.) Bryophytes and lichens in a changing environment, pp. 211-233. Clarendon Press, Oxford, UK.

Santesson, R., Moberg, R., Nordin, A., Tønsberg, T. & Vitikainen, O. 2004. Lichen-forming and lichenicolous fungi of Fennoscandia. Museum of Evolution, Uppsala University, Göteborg, SE. Schauer, T. 1965. Ozeanische Flechten im Nordalpenraum. Portugaliae Acta Biologica B 8: 17-229. Sokal, R.R. & Rohlf, F.J. Biometry. ed. 3. Freeman, San Francisco, CA, US. ter Braak, C.J.F. 1988. Partial canonical correspondence analysis. In: Bock, H.H. (ed.) Classification and related methods of data analysis: proceedings of the first conference of the International Federation of Classification Societies (IFCS), pp. 551-558. North-Holland, Amsterdam, NL. ter Braak, C.J.F. & Smilauer, P. 1998. CANOCO reference manual and users guide to Canoco for Windows: software for canonical community ordination (version 4). Microcomputer Power, Ithaca, NY, US. Tømmervik, H., Høgda, K.A. & Solheim, I. 2003. Monitoring vegetation changes in Pasvik (Norway) and Pechenga in Kola Peninsula (Russia) using multitemporal Landsat MSS/ TM data. Remote Sens. Environ. 85: 370-388. Trass, H., Vellak, K. & Ingerpuu, N. 1999. Floristical and ecological properties for identifying of primeval forests in Estonia. Ann. Bot. Fenn. 36: 67-80. van Herk, C.M., Mathijssen-Spiekman, E.A.M. & de Zwart, D. 2003. Long distance nitrogen air pollution effects on lichens in Europe. Lichenologist 35: 347-359. Wirth, V. 1995. Die Flechten Baden-Württembergs. Eugen Ulmer, Stuttgart, DE. Wirth, V. 1999. Gefährdete Flechtenbiotope in Mitteleuropa. Natur Museum 129: 12-21. ‹

208

Received 23 September 2004; Accepted 9 February 2005. Co-ordinating Editor: R.H. Økland.

For App. 1, see JVS/AVS Electronic Archives; www.opuluspress.se

App. 1. Explanation of environmental variables used to characterise the study plots, variable name (Variable), method used to determine variable and explanation of variable (method and explanation), units of measurement (units), explanatory variable set the variable was included into, E= environmental and forest variables, S=Spatial variables, C= Micro-and macroclimatic variables, H= human impact variables (VS), transformation formula (Trans), c value used in transformation (c), minimum value (Min), mean value (Mean), maximum value (Max), n.a. not applicable. Variables marked with (*) were ranged prior to transformation to zero skewness. Variable Aln_inc

Method and explanation Presence/absence of Alnus incana in study plot.

Aln_Sal _for

VS E

Trans None

n.a.

0.0

0.5

Max 1.0

Study plot situated in alder-willow forest, composed of Alnus incana, Salix pentandra and/or S. myrsinifolia, (type E3a in Fremstad 1997).

E

None

n.a.

0.0

0.1

1.0

Alnus_for

Study plot situated in Alnus incana forest (type C3 in Fremstad 1997).

E

None

n.a.

0.0

0.3

1.0

Altit

Average plot altitude above sea level, approximated from GPS position and topographical maps. Basal area of logs, calculated from measurements of log diameters. In tree stumps, the top diameter was measured, and in logs the collar diameter.

m

C

ln(c+x)

69.80378

10.0

122.8

440.0

m2/ha

E

ln(c+x)

2.12570

0.0

8.9

38.6

Basal area of trees, calculated from circumference measurements of deciduous trees in 1.3 m stem height. Presence/absence of Betula pubescens in study plot.

m2/ha

E

ln(c+x)

10.69264

1.9

26.3

71.5

E

None

n.a.

0.0

0.9

1.0

Study plot situated in nutrient-rich birch forest with a field layer dominated by forbs (type C1 and C2 in Fremstad 1997). Study plot situated in nutrient-poor birch forest, dominated by dwarf shrubs (types A3, A4, A5 in Fremstad 1997).

E

None

n.a.

0.0

0.2

1.0

E

None

n.a.

0.0

0.2

1.0

D_Salix _for

Presence/absence of dry Salix caprea forest in study plot.

E

None

n.a.

0.0

0.1

1.0

Dens _young

Density of young trees. The number of young trees (collar diameter lesser than or equal to 5 cm) was counted, and the number of young trees per ha was calculated. The distance to the closest house, determined from topographical maps. The distance to the closest area cleared from forest, determined from topographical maps. The distance to the closest river was measured from topographical maps.

#/ha

E

ln(c+x)

0.00853

0.0

0.3

12.0

km

H

ln(c+x)

0.10958

0.1

1.4

7.6

km

H

ln(c+x)

0.01996

0.0

0.7

9.4

km

E

ln(c+x)

0.00890

0.0

0.2

1.1

The distance to the closest road, determined from topographical maps. The distance to the closest sea shore, determined from maps of scale 1:500000. The distance to the closest town, determined from topographical maps. The proportion of forest edge length to total length of edge was determined in circular macroplots of radius 500m surrounding the study plots. The length of forest edge and the total edge length were determined from topographical maps. Study plot heat index (Parker 1988), calculated as tanĮ1*cosĮ2 , where Į1 was plot inclination and Į2 was unfavourability.

km

H

ln(c+x)

0.03922

0.0

0.4

2.6

km

C

ln(c+x)

11.15712

0.1

17.6

54.5

km

H

ln(c+x)

2.84107

0.2

12.3

38.5

%

E

e^cx

0.00651

16.1

53.6

87.6

C

ln(c+x))

6.73740

-6.1

2.6

185.7

H

None

n.a.

1.0

4.6

10.0

Bas_logs

Bas _trees

Bet_pub Betula _forb Betula _herb

Dist_house Dist _open

Dist_river Dist_road Dist_sea Dist _town Forest _total

Heat_index

Hum_imp

The overall human impact index was recorded following the classification of Trass et al. (1999). Trass’ 13-point human impact scale was reversed, the

Units

c

Min

Mean

App. 1. Internet supplement to: Werth, S.; Tømmervik, H. & Elvebakk, A. 2005. Epiphytic macrolichen communities along regional gradients in northern Norway. J. Veg. Sci. 16: 199-208.

Human_nat

Human _total

Insolation

Landsc

Light Log_dec

Log_dens

beginning point was set to 1 (lowest value of human impact). The proportion of human derived edge length to natural edge length was determined in macroplots (see above) from topographical maps; defined as "human derived" were settlements, gravel pits, paths, powerlines, football pitches and farming area. Defined as "natural edge" were outlines of mire, lakes, rivers, and alike. The proportion of human derived edge length to total length of edge. Measured from topographical maps; defined as "human derived" were settlements, gravel pits, and farming area. 15-point insolation index with highest values in S and SSW. Plots with a slope of 0 were given the value zero. Landscape unaffectedness, following the scale in Trass et al. (1999). 1= strongly affected, 3= unaffected. Study plot light conditions, 1= shady, 5= sunny. Log decay, following the classification in Linder et al. (1997). 1= weakly decayed, 3= strongly decayed. Density of logs, determined by counting all logs exceeding 10 cm collar diameter, and calculating the density per ha.

%

H

ln(c+x)

0.07630

0.0

0.3

1.3

%

H

ln(c+x)

22.14012

0.0

21.2

56.0

C

None

n.a.

0.0

5.7

15.0

H

None

n.a.

1.0

2.1

3.0

C

None

n.a.

2.0

3.3

5.0

E

None

n.a.

1.0

2.2

3.0

E

ln(c+x)

0.01384

0.0

0.1

0.4

#/ha

M_Salix _for

Study plot situated in moist Salix forest, i.e. swamp-forests comprising Salix pentandra and/or S. myrsinifolia.

E

None

n.a.

0.0

0.0

1.0

Maalselv

Location of study plot in Målselv municipality, as determined from topographical maps. Maximum log diameter of study plot, determined from measurements of log diameters. Maximum tree diameter of study plot, calculated by transforming tree circumference values at 1.3 m stem height to diameters, and selecting the maximum tree diameter of each plot

S

None

n.a.

0.0

0.3

1.0

cm

E

ln(c+x)

17.30933

0.0

19.9

51.0

cm

E

ln(c+x)

7.06790

8.2

27.9

64.6

cm

E

ln(c+ln(c+x))

0.39422

6.3

13.5

38.0

C

None

n.a.

1.0

2.9

5.0

Max_log

Max_tree

Med_tree

Moist

Median tree diameter of study plot, calculated from tree circumference at 1.3 m stem height, transformed to diameter. Study plot moisture conditions. 1=Wet to moist. Area either within the floodplains of a river, showing a strong seasonal pattern in moisture, or in a swamp forest, permanently moist with surface water in brooks or small pools, water table at or near the forest ground. Rich in moistureindicating plants such as Caltha palustris, Climacium dendroides, Potentilla palustris, Sphagnum sp. or Matteuccia struthiopteris. 2=Between moist and mesic. Little permanent water on the ground, indicators of both moist and mesic conditions. 3=Mesic. Little or no permanent water on the ground, and plants such as Vaccinium myrtillus, Gymnocarpium dryopteris, Phegopteris connectilis. 4=Between mesic and dry. No permanent water on the ground, a mixture of species of mesic and dry conditions. 5=Dry. No permanent water on the ground. Often with plant species showing adaptations to dry conditions. Higher plant vegetation with Empetrum nigrum, Vaccinium vitis-idaea,and/or Polystichum lonchitis.


Nat_nr

pH-lab

Number of naturally derived lines, line segments and patches, as determined for macroplots from topographical maps. Mires, lakes and river segments were taken into consideration. Proportion of naturally derived edge length to total length of edge. Measured from topographical maps; defined as "natural edge" were outlines of bogs, fens, forest, lakes and rivers. Total number of lines, line segments and patches, corresponding to the number of landscape elements in a macroplot. Determined for using topographical maps. Oceanity of climate, determined from maps of oceanity of the study area shown in Moen et al. (1999). Proportion of edge length of non-forested area to edge length of forested area, determined for macroplots using topographical maps. Presence of tree plantation (spruce, larch) in visibility of the study plot. Soil pH value.

Pop_tre

Presence of Populus tremula in study plot.

E

None

n.a.

0.0

0.1

1.0

Populus _for Pr_S_T0

Study plot situated in Populus tremula forest. Sum of monthly precipitation for all months exceeding 0°C in mean air temperature (reference period 19611990), interpolated for the locations of the study plots. Sum of monthly precipitation in all months exceeding 4°C in mean air temperature (reference period 19611990), interpolated for the locations of the study plots. Annual normal precipitation (reference period 1961-1990), interpolated for the locations of the study plots.

E

None

n.a.

0.0

0.1

1.0

mm

C

ln(c+x)

229.56638

208.0

440.6

837.0

mm

C

ln(c+x)

225.36732

171.0

310.4

477.0

mm

C

e^cx

0.00019

319.0

808.4

1275.0

Nat_total

Nr_landsc_ elem

Oceanity

Open _forest

Plantation

Pr_S_T4

Pr_year

%

%

E

ln(c+x)

1032.955

0.0

0.5

1.0

E

ln(c+x)

24.85430

0.0

25.1

71.2

E

ln(c+x)

C

None

E

ln(c+x)

H E

32890

0.0

0.5

1.0

1.0

2.3

4.0

0.07370

0.0

0.7

4.0

None

n.a.

0.0

0.6

1.0

ln(c+ln (c+x))

2813663.2346

4.1

5.5

7.7

n.a.

Road_cat

Category of road closest to the study plot, with the categories following the key of topographical maps.

H

None

n.a.

1.0

1.8

3.0

Rock_suit

Suitability of rock substrates, calculated as A+B+C. (A) Bedrock type, (1) gneiss or granite, (2) acidic to neutral schist or arkose, or (5) calcareous schist. (B) Surface conditions, (1) hard, (2) intermediate, or (-1) eroded. (C) Height of exposed rock surface, (1) 0.00 - 0.49m, (2) 0.5 - 0.9 m, (3) > 0.9 m. Presence of Salix caprea in study plot.

E

None

n.a.

1.0

3.3

10.0

Sal_cap

E

None

n.a.

0.0

0.2

1.0

Sal_myr

Presence of Salix myrsinifolia in study plot.

E

None

n.a.

0.0

0.3

1.0

Sal_pen

Presence of Salix pentandra in study plot.

E

None

n.a.

0.0

0.1

1.0

Sea _west

Distance to open sea, as determined from maps of scale 1:500000.

km

C

e^cx

11.15712

0.1

17.6

54.5

Sib

Percentage of trees with more than one stem. Slope angle of study plot, as determined with clinometer. Presence of Sorbus aucuparia in study plot. Study plot located in Storfjord municipality. Temperature sum of all months exceeding 0°C (reference period 19611990).

%

E

ln(c+x)

0.25687

0.0

0.3

1.0

°azimuth

C

ln(c+x)

6.55405

0.0

11.2

39.0

E

None

n.a.

0.0

0.2

1.0

S

None

n.a.

0.0

0.1

1.0

°C

C

e^cx

0.11612

38.4

50.3

54.7

°C

C

e^cx

0.13445

35.6

48.0

52.4

Slope Sor_auc Storfjord TM _SUM0 TM _SUM4

Temperature sum of all months exceeding 4°C (reference period 19611990).


Tree _age Tree _dens

Tromso

Unfav

Water _total

Approximate age of tree layer, 1= young forest, 3= old forest. Density of trees. The number of trees (collar diameter larger than 5 cm) was counted, and the number of trees per ha was calculated. Study plot located in Tromsø municipality, determined from topographical maps. Unfavourability of plot aspect, i.e. the absolute deviation of plot aspect from SSW (202.5°). Proportion of water edge length to total length of edge. Determined using topographical maps.

E

None

n.a.

1.0

2.2

3.0

E

ln(c+ln(c+x))

0.54047

0.1

0.6

30.0

S

None

n.a.

0.0

0.3

1.0

°

C

ln(c+x)

1.80244

0.0

45.5

162.5

%

E

ln(c+x)

11.03673

0.0

15.5

79.6

#/ha

X (*)

(*) Longitude

S

ln(c+x)

0.29023

384268.0

422761.6

487446.0

X2 (*)

(*) Longitude^2

S

e^cx

-1.80436

X2Y (*)

(*) Longitude^2*Latitude

S

ln(c+x)

0.23539

X3 (*)

(*) Longitude^3

S

ln(c+x)

0.15735

XY (*)

(*) Longitude*Latitude

S

ln(c+x)

0.36348

XY2 (*)

(*) Longitude*Latitude^2

S

ln(c+x)

0.42230

Y (*)

(*) Latitude

S

ln(c+x)

0.89363

1.476 *10^11 1.140 *10^18 5.674 *10^16 2.967 *10^12 2.267 *10^19 7618977.0

1.793 *10^11 1.377 *10^18 7.642 *10^16 3.245 *10^12 2.491 *10^19 7676801.5

2.376 *10^11 1.824 *10^18 1.158 *10^17 3.743 *10^12 2.874 *10^19 7737570.0

Y2 (*)

(*) Latitude^2

S

e^cx

-0.77010

Y3 (*)

(*) Latitude^3

S

ln(c+x)

0.84303

5.804 *10^13 4.422 *10^20

5.893 *10^13 4.524 *10^20

5.986* 10^13 4.632 *10^20


App. 2. Kendall’s Tau correlation of the first four DCA axes with environmental variables investigated. Asterisks indicate that the significance of a correlation at a Bonferroni-corrected level of Į = 0.05. Variable DCA1 DCA2 DCA3 DCA4 Aln_inc -0.05 -0.11 -0.04 0.18 Aln_Sal_for -0.05 -0.11 -0.04 0.18 Alnus_for -0.10 0.15 0.19 -0.16 Altit 0.08 -0.27 0.28 -0.15 Bas_logs -0.09 -0.16 -0.16 0.15 Bas_trees 0.15 -0.14 -0.22 -0.14 Bet_pub 0.12 -0.17 -0.14 0.14 Betula_forb 0.00 *0.10 0.10 -0.14 Betula_herb *0.02 0.10 0.04 0.14 D_Salix_for -0.03 -0.18 -0.12 0.13 Dens_young 0.19 0.13 0.08 -0.13 Dist_house -0.16 0.06 -0.20 0.12 Dist_open 0.23 -0.20 -0.06 0.12 Dist_river 0.06 0.05 0.02 -0.12 Dist_road 0.13 -0.12 -0.20 -0.11 Dist_sea 0.10 -0.06 0.17 -0.11 Dist_town -0.04 0.05 0.16 -0.11 Forest_total -0.04 -0.18 -0.12 0.11 Heat_index 0.07 0.02 0.00 0.09 Hum_imp 0.19 0.27 -0.08 0.08 Human_nat 0.07 0.18 -0.02 -0.08 Human_total 0.15 0.28 0.09 0.07 Insolation 0.07 0.11 -0.08 0.07 Landsc 0.11 -0.15 0.22 -0.06 Light -0.03 0.01 0.08 -0.06 Log_dec 0.04 -0.11 0.04 0.06 Log_dens 0.09 0.07 -0.03 0.06 M_Salix_for 0.01 -0.13 -0.02 0.06 Maalselv 0.05 -0.15 -0.21 -0.05 Max_log 0.07 -0.04 -0.18 -0.05 Max_tree -0.14 0.07 -0.17 -0.05 Med_tree 0.14 -0.07 0.17 0.05 Moist -0.14 0.07 -0.17 -0.05 Nat_nr 0.08 -0.09 0.13 -0.05 Nat_total 0.15 -0.18 -0.08 0.05 Nr_landsc_elem 0.06 -0.12 0.08 0.05 Oceanity -0.13 *-0.38 0.06 -0.05 Open_forest 0.13 0.38 -0.06 0.05 pH_lab -0.13 -0.38 0.06 -0.05 Plantation -0.01 -0.11 -0.02 0.05 Pop_tre -0.29 0.15 -0.24 -0.04 Populus_for 0.09 -0.21 0.19 -0.04 Pr_S_T0 -0.04 *0.20 -0.19 -0.04 Pr_S_T4 0.05 *-0.05 -0.13 -0.04

Pr_year Road_cat Rock_suit Sal_cap Sal_myr Sal_pen Sea_west Sib Slope Sor_auc Storfjord TM_SUM0 TM_SUM4 Tree_age Tree_dens Tromso Unfav Water_total X X2 X2Y X3 XY XY2 Y Y2 Y3

0.06 *0.12 -0.21 0.04 -0.13 -0.37 -0.14 -0.36 0.06 0.01 -0.16 -0.34 0.09 -0.09 -0.39 0.03 0.04 0.27 0.12 0.05 -0.07 0.34 0.15 -0.05 -0.04 0.11 0.15 0.00 0.05 0.04 *-0.05 0.33 -0.05 0.31 -0.18 0.11 -0.04 *0.11 0.09 *0.30 0.05 *0.12 0.01 *0.07 -0.01 *-0.15 0.15 *0.11 -0.23 -0.17 0.12 0.24 0.27 -0.15

0.11 0.08 0.06 0.06 -0.05 0.04 0.02 0.26 0.17 -0.17 -0.14 -0.11 0.10 -0.09 -0.27 -0.17 -0.16 0.14 0.10 -0.08 0.06 -0.04 -0.03 -0.01 0.10 0.06 0.01

-0.04 0.04 -0.04 -0.04 -0.04 -0.04 0.04 0.03 -0.03 -0.03 -0.03 0.03 -0.03 0.03 0.02 -0.02 -0.02 0.02 -0.02 0.02 -0.02 -0.02 0.02 -0.01 0.00 0.00 0.00

App. 2. Internet supplement to: Werth, S.; Tømmervik, H. & Elvebakk, A. 2005. Regional gradients in epiphytic macrolichen communities of northern Norway. J. Veg. Sci. 16: xxx-yyy.

Summary

Summary This study investigated dispersal and genetic structure of the putatively dispersal-limited epiphytic lichen Lobaria pulmonaria by means of direct dispersal estimates derived from snow samples, and by population-genetic approaches. From 41 sampling plots of 1 ha (demes) corresponding to the disturbance categories stand-replacing disturbance, intensive logging and unevenaged forestry, 923 thalli of L. pulmonaria were collected, and analysed at six mycobiont-specific microsatellite loci in order to investigate genetic diversity and spatial genetic structure. Old-forest associated lichens are commonly assumed to be negatively affected by tree logging or large-scale, natural forest disturbances. Contrary to this view, in this study, highest genetic diversity was found in demes of L. pulmonaria affected by 19th century stand-level logging. However, genetic diversity was lowest in demes affected by 19th-century stand-replacing disturbance. This result exemplifies that genetic diversity strongly depends on the type of disturbance, and may not necessarily be impacted by stand-level disturbance for many centuries. Bayesian analysis of population structure revealed the presence of three spatially intermingled gene pools in the studied landscape, a result which indicated that assortative mating might occur in L. pulmonaria. High levels of gene flow among areas and the uneffectiveness of a putative barrier to past gene flow, a 1 km-wide unforested area, implied that the study area (13 km2) is occupied by a single, continuous population of L. pulmonaria. The number of colonisation events was not significantly lower in demes affected by stand-replacing disturbance than in demes affected by other disturbance types, showing that demes in disturbed areas were founded multiple times from independent sources, followed by rapid clonal spread at forest-stand level. Multiple alleles at one or several microsatellite loci found in a few of the 895 thalli of L. pulmonaria which were investigated pointed towards the presence of multiple conspecific fungal strains in a single thallus. Using spatial autocorrelation methods, the spatial scale of similar genetic structure was determined, discriminating among the clonal and recombinant component 149

Summary

of genetic variation. Spatial autocorrelation of gene diversity was strong at distances up to 150 m in three disturbance categories (unevenaged forestry, stand-replacing disturbance, stand-level logging), with the strongest autocorrelation for demes affected by stand-replacing disturbance. The spatial autocorrelation was predominantly attributed to clonal dispersal of vegetative propagules. After accounting for the clonal component, no significant spatial autocorrelation remained. This pattern may indicate low dispersal ranges of clonal propagules, but random dispersal of sexual ascospores, i.e., that there was no dispersal limitation of ascospores. Using a L. pulmonaria-specific RealTime-PCR assay, 240 DNA extracts of snow sample filtrates were genotyped, enabling a discrimination among propagules originating from a single, isolated source tree and from elsewhere. Samples which were detected as positives by RealTime-PCR were additionally genotyped at six L. pulmonaria microsatellite loci. Both molecular approaches demonstrated substantial dispersal from other than local sources. In a landscape approach, we analysed 240 snow samples with RealTime-PCR and detected propagules not only in forests where L. pulmonaria was present, but also in large unforested pasture areas and in forest patches where L. pulmonaria was not found. Monitoring of soredia of L. pulmonaria transplanted to maple bark after two vegetation periods showed a significant effect of transplantation substrate (gauze vs. bark) and high variance in growth among groups of forest stands, but no significant differences among four transplantation treatments (north-exposed, high tree density; north-exposed, low tree density; south-exposed, high tree density; south-exposed, low tree density). There was no difference in numbers of established soredia transplanted to freestanding trees or to trees within forests. This field experiment indicated that ecological constraints at the level of adjacent forest stands may hinder colonisation of seemingly suitable trees, but did not identify the environmental factors responsible for this. The combined result of the spatial analysis, the rapid recolonisation of disturbed areas from multiple sources within three generations of L. pulmonaria, and the large amount of dispersal over long distances have shown that the amount of propagules 150

Summary

reaching a particular site is probably not a major factor limiting the old-forest lichen L. pulmonaria from colonising potential carrier trees in the study area. In the large population investigated, dispersal limitation therefore seems to play a minor role, as compared to establishment limitation.

151

Appendix 1

Appendix 1: Details on the trees from which lichens were collected for molecular analyses. Plot W009 W009 W009 W009 W009 W009 W009 W009 W009 W009 W009 W009 W009 W009 W009 W009 W013 W013 W013 W013 W013 W013 W013 W013 W013 W013 W013 W013 W013 W013 W013 W013 W013 W013 W013 W013 W014 W014 W014 W014 W014 W014 W014 W014 W015 W015 W021 W021 W021 W021 W032 W032

Area North North North North North North North North North North North North North North North North East East East East East East East East East East East East East East East East East East East East North North North North North North North North North North North North North North East East

Disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Unevenaged forestry Unevenaged forestry

TreeID W09_46 W09_46 W09_47 W09_47 W09_50 W09_50 W09_51 W09_51 W09_52 W09_52 W09_53 W09_53 W09_55 W09_55 W09_56 W09_56 W13_100 W13_100 W13_74 W13_74 W13_87 W13_87 W13_88 W13_88 W13_89 W13_89 W13_90 W13_90 W13_93 W13_93 W13_94 W13_94 W13_97 W13_97 W13_99 W13_99 W14_1 W14_1 W14_22 W14_22 W14_23 W14_23 W14_34 W14_34 W15_3 W15_3 W21_105 W21_105 W21_83 W21_83 W32_203 W32_203

X_Tree 505243 505243 505239 505239 505192 505192 505193 505193 505202 505202 505199 505199 505205 505205 505210 505210 507744 507744 507706 507706 507724 507724 507734 507734 507739 507739 507741 507741 507732 507732 507725 507725 507750 507750 507747 507747 505060 505060 505053 505053 505060 505060 505033 505033 504890 504890 505621 505621 505637 505637 507409 507409

Y_Tree 156188 156188 156179 156179 156142 156142 156142 156142 156148 156148 156150 156150 156155 156155 156160 156160 155153 155153 155116 155116 155072 155072 155079 155079 155077 155077 155083 155083 155096 155096 155115 155115 155125 155125 155150 155150 156043 156043 156048 156048 156043 156043 156104 156104 155960 155960 156351 156351 156401 156401 154770 154770

153

Appendix 1

Plot W032 W032 W035 W035 W035 W035 W035 W035 W036 W036 W036 W036 W036 W036 W036 W036 W037 W037 W037 W037 W040 W040 W040 W040 W040 W040 W040 W040 W043 W043 W043 W043 W043 W043 W043 W043 W046 W046 W046 W046 W046 W046 W046 W046 W046 W046 W046 W046 W046 W046 W050 W050 W051 W051 W051 W051

154

Area East East East East East East East East North North North North North North North North North North North North North North North North North North North North East East East East East East East East North North North North North North North North North North North North North North East East North North North North

Disturbance Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry

TreeID W32_215 W32_215 W35_41 W35_41 W35_50 W35_50 W35_54 W35_54 W36_1 W36_1 W36_171 W36_171 W36_198 W36_198 W36_201 W36_201 W37_88 W37_88 W37_96 W37_96 W40_1 W40_1 W40_45 W40_45 W40_50 W40_50 W40_69 W40_69 W43_100 W43_100 W43_107 W43_107 W43_108 W43_108 W43_93 W43_93 W46_136 W46_136 W46_137 W46_137 W46_142 W46_142 W46_143 W46_143 W46_144 W46_144 W46_145 W46_145 W46_148 W46_148 W50_35 W50_35 W51_1 W51_1 W51_23 W51_23

X_Tree 507380 507380 507226 507226 507212 507212 507221 507221 505455 505455 505454 505454 505451 505451 505457 505457 505359 505359 505415 505415 506057 506057 506077 506077 506057 506057 506129 506129 507206 507206 507244 507244 507245 507245 507253 507253 505024 505024 505025 505025 505000 505000 504993 504993 505004 505004 505007 505007 505022 505022 507306 507306 506022 506022 506075 506075

Y_Tree 154786 154786 154721 154721 154735 154735 154736 154736 156113 156113 156113 156113 156116 156116 156150 156150 156381 156381 156385 156385 156427 156427 156432 156432 156417 156417 156405 156405 154387 154387 154300 154300 154300 154300 154391 154391 155863 155863 155856 155856 155912 155912 155910 155910 155914 155914 155918 155918 155946 155946 154751 154751 156428 156428 156467 156467

Appendix 1


Area North North North North North North North North North North North North North North North North North North North North North North North North North North North North North North North North North North North North North North North North North North North North North North North North North North North North North North East East

Disturbance Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Stand-replacing disturbance Stand-replacing disturbance Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Unevenaged forestry Unevenaged forestry


X_Tree 506076 506076 506065 506065 506027 506027 505983 505983 505600 505600 504620 504620 504612 504612 505305 505305 504974 504974 504972 504972 504958 504958 504947 504947 504965 504965 505423 505423 505415 505415 505463 505463 505479 505479 505491 505491 505635 505635 505630 505630 505658 505658 505658 505658 505645 505645 505648 505648 505657 505657 505668 505668 505650 505650 507479 507479

Y_Tree 156461 156461 156455 156455 156440 156440 156474 156474 155579 155579 156066 156066 156076 156076 156315 156315 155863 155863 155870 155870 155857 155857 155811 155811 155856 155856 156055 156055 156099 156099 156106 156106 156063 156063 156058 156058 155469 155469 155463 155463 156271 156271 156271 156271 156263 156263 156258 156258 156242 156242 156228 156228 156232 156232 154600 154600

155

Appendix 1


156

Area East East East East East East East East East East East East East East North North North North East East East East East East East East East East East East North North North North North North North North North North East East East East East East East East East East North North North North North North

Disturbance Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance


X_Tree 507496 507496 507507 507507 507512 507512 507511 507511 507472 507472 507480 507480 507447 507447 505968 505968 505969 505969 507510 507510 507451 507451 507476 507476 507477 507477 507451 507451 507567 507567 505608 505608 505595 505595 505605 505605 505607 505607 505609 505609 507760 507760 507653 507653 507653 507653 507662 507662 507710 507710 505560 505560 505524 505524 505604 505604

Y_Tree 154628 154628 154636 154636 154639 154639 154644 154644 154651 154651 154674 154674 154625 154625 156269 156269 156268 156268 154791 154791 154676 154676 154738 154738 154735 154735 154676 154676 154750 154750 156132 156132 156105 156105 156119 156119 156125 156125 156133 156133 155185 155185 155273 155273 155273 155273 155242 155242 155291 155291 156222 156222 156214 156214 156262 156262

Appendix 1


Area North North North North North North North North North North North North North North North North North North North North North North North North North North North North North North North North North North North North West West West West North North North North North North North North North North North North North North North North

Disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-replacing disturbance Stand-level logging Stand-level logging Stand-level logging Stand-level logging Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry


X_Tree 505622 505622 505628 505628 505630 505630 505630 505630 505620 505620 505560 505560 505560 505560 505607 505607 505569 505569 505606 505606 505578 505578 505551 505551 505580 505580 505538 505538 505527 505527 505552 505552 505545 505545 505589 505589 505068 505068 505059 505059 506181 506181 506241 506241 506186 506186 506187 506187 506193 506193 506195 506195 506200 506200 506221 506221

Y_Tree 156246 156246 156246 156246 156245 156245 156247 156247 156219 156219 156224 156224 156224 156224 156222 156222 156225 156225 156220 156220 156209 156209 156165 156165 156199 156199 156171 156171 156179 156179 156171 156171 156170 156170 156185 156185 154363 154363 154327 154327 156605 156605 156577 156577 156596 156596 156577 156577 156587 156587 156586 156586 156580 156580 156599 156599

157

Appendix 1


158

Area North North North North North North North North North North North North North North North North North North North North North North North North North North North North North North West West West West West West West West West West West West West West West West West West West West West West West West West West

Disturbance Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging


X_Tree 506218 506218 506230 506230 506230 506230 506208 506208 506201 506201 506180 506180 506195 506195 506219 506219 506219 506219 506227 506227 506220 506220 506215 506215 506247 506247 506237 506237 506238 506238 504719 504719 504712 504712 504708 504708 504696 504696 504698 504698 504641 504641 504651 504651 504661 504661 504623 504623 504624 504624 504627 504627 504651 504651 504670 504670

Y_Tree 156598 156598 156608 156608 156611 156611 156623 156623 156612 156612 156553 156553 156558 156558 156583 156583 156583 156583 156561 156561 156548 156548 156545 156545 156563 156563 156565 156565 156603 156603 154214 154214 154208 154208 154209 154209 154205 154205 154194 154194 154205 154205 154210 154210 154208 154208 154237 154237 154244 154244 154247 154247 154234 154234 154259 154259

Appendix 1


Area West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat

Disturbance Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown


X_Tree 504688 504688 504599 504599 504563 504563 504562 504562 504570 504570 504568 504568 504584 504584 504597 504597 504597 504597 504604 504604 504608 504608 504609 504609 504612 504612 504621 504621 504607 504607 504600 504600 504589 504589 504604 504604 504597 504597 504579 504579 504573 504573 504560 504560 504556 504556 157263 157263 157296 157296 157301 157301 157266 157266 157266 157266

Y_Tree 154184 154184 154115 154115 153995 153995 153999 153999 154025 154025 154032 154032 154054 154054 154048 154048 154047 154047 154037 154037 154028 154028 154120 154120 154120 154120 154134 154134 154001 154001 153995 153995 154068 154068 154095 154095 153989 153989 153995 153995 153994 153994 154002 154002 154003 154003 507454 507454 507432 507432 507406 507406 507429 507429 507398 507398

159

Appendix 1


160

Area G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat G. Rolat West West West West West West West West West West West West West West West West West West

Disturbance Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry


X_Tree 157249 157249 157246 157246 157243 157243 157235 157235 157228 157228 157230 157230 157252 157252 157255 157255 157260 157260 157244 157244 157249 157249 157260 157260 157272 157272 157272 157272 157281 157281 157272 157272 157230 157230 157237 157237 157223 157223 504781 504781 504739 504739 504772 504772 504762 504762 504762 504762 504762 504762 504781 504781 504742 504742 504765 504765

Y_Tree 507398 507398 507397 507397 507397 507397 507396 507396 507389 507389 507381 507381 507389 507389 507385 507385 507381 507381 507372 507372 507368 507368 507366 507366 507367 507367 507362 507362 507390 507390 507409 507409 507411 507411 507426 507426 507410 507410 153964 153964 153939 153939 153961 153961 153956 153956 153971 153971 153976 153976 153954 153954 153956 153956 153950 153950

Appendix 1


Area West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West East East East East East East East East East East East East East East East East

Disturbance Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry


X_Tree 504749 504749 504819 504819 504826 504826 504808 504808 504805 504805 504807 504807 504792 504792 504783 504783 504770 504770 504768 504768 504771 504771 504791 504791 504750 504750 504759 504759 504775 504775 504787 504787 504913 504913 504930 504930 504934 504934 504931 504931 507045 507045 507033 507033 507045 507045 507050 507050 507061 507061 507097 507097 507110 507110 507107 507107

Y_Tree 153921 153921 153953 153953 153952 153952 153946 153946 153944 153944 153942 153942 153939 153939 153921 153921 153910 153910 153907 153907 153922 153922 153944 153944 153974 153974 153985 153985 153996 153996 153975 153975 154503 154503 154523 154523 154467 154467 154485 154485 154194 154194 154184 154184 154188 154188 154192 154192 154191 154191 154211 154211 154211 154211 154227 154227

161

Appendix 1


162

Area East East East East East East East East East East East East East East East East East East East East East East East East East East East East West West West West West West West West West West West West West West West West West West West West West West West West West West West West

Disturbance Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry


X_Tree 507147 507147 507129 507129 507057 507057 507066 507066 507071 507071 507072 507072 507069 507069 507093 507093 507083 507083 507154 507154 507116 507116 507061 507061 507075 507075 507142 507142 505349 505349 505355 505355 505382 505382 505393 505393 505319 505319 505358 505358 505360 505360 505370 505370 505351 505351 505395 505395 505091 505091 505101 505101 505101 505101 505084 505084

Y_Tree 154184 154184 154170 154170 154204 154204 154216 154216 154215 154215 154219 154219 154217 154217 154263 154263 154253 154253 154191 154191 154158 154158 154199 154199 154195 154195 154183 154183 154693 154693 154693 154693 154729 154729 154728 154728 154779 154779 154707 154707 154712 154712 154705 154705 154696 154696 154715 154715 154959 154959 154964 154964 154966 154966 154952 154952

Appendix 1


Area West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West

Disturbance Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging


X_Tree 505095 505095 505117 505117 505074 505074 505058 505058 505060 505060 504723 504723 504728 504728 504698 504698 504695 504695 504681 504681 504682 504682 504669 504669 504675 504675 504673 504673 505131 505131 505129 505129 505135 505135 505137 505137 505173 505173 505154 505154 504433 504433 504427 504427 504426 504426 504431 504431 504420 504420 504420 504420 504418 504418 504417 504417

Y_Tree 154942 154942 154944 154944 154970 154970 154989 154989 154990 154990 154695 154695 154705 154705 154636 154636 154687 154687 154681 154681 154667 154667 154690 154690 154696 154696 154699 154699 154421 154421 154411 154411 154407 154407 154412 154412 154366 154366 154348 154348 154248 154248 154255 154255 154257 154257 154265 154265 154264 154264 154264 154264 154262 154262 154260 154260

163

Appendix 1

Plot W229 W229 W229 W229 W229 W229 W229 W229 W230 W230 W230 W230 W230 W230 W230 W230 W230 W230 W230 W230 W238 W238 W238 W238 W238 W238 W238 W238 W238 W238 W238 W238 W55a W55a W55b W55b W55b W55b W55b W55b W55b W55b

164

Area West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West West North North North North North North North North North North

Disturbance Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Stand-level logging Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry Unevenaged forestry

TreeID W229_7064 W229_7064 W229_7077 W229_7077 W229_7099 W229_7099 W229_7109 W229_7109 W230_7190 W230_7190 W230_7200 W230_7200 W230_7206 W230_7206 W230_7209 W230_7209 W230_7246 W230_7246 W230_7252 W230_7252 W238_7499 W238_7499 W238_7501 W238_7501 W238_7502 W238_7502 W238_7526 W238_7526 W238_7527 W238_7527 W238_7533 W238_7533 W55-1 W55-1 W55-2 W55-2 W55-3 W55-3 W55-4 W55-4 W55-5 W55-5

X_Tree 504417 504417 504395 504395 504399 504399 504422 504422 504755 504755 504756 504756 504747 504747 504733 504733 504802 504802 504828 504828 504546 504546 504548 504548 504554 504554 504613 504613 504601 504601 504601 504601 504617 504617 504661 504661 504641 504641 504623 504623 504631 504631

Y_Tree 154291 154291 154271 154271 154212 154212 154230 154230 154209 154209 154254 154254 154244 154244 154230 154230 154222 154222 154243 154243 154580 154580 154586 154586 154587 154587 154572 154572 154548 154548 154545 154545 155838 155838 156264 156264 156270 156270 156275 156275 156270 156270

Appendix 2

Appendix 2: Details on the trees which received transplants. Tree, tree number; Easting, East coordinate; Northing, North coordinate (both Swiss Grid, map date CH1903); Block, number of the forest stand; Exposition, exposition of forest stand; Tree density, tree density; Mean_B, mean number of soredia counted on bark substrate; Mean_G, mean number of soredia counted on gauze substrate; Mean_B and Mean_B refer to the control of the transplants in 2003, 15 months after transplant establishment. Tree Easting Northing Block Exposition SW01 503330 150707 1 North SW02 503483 150636 1 South SW03 503708 151049 1 South SW04 503686 151044 1 North SW05 503460 151146 2 South SW06 503441 151202 2 North SW07 503581 151262 2 South SW08 503548 151288 2 North SW09 504048 151769 3 South SW10 504021 151748 3 North SW11 504000 151824 3 North SW12 504066 151815 3 South SW13 506500 157098 4 North SW14 506484 157143 4 South SW15 506517 157172 4 North SW16 506498 157197 4 South SW17 506500 157200 5 South SW18 506579 157230 5 South SW19 506602 157215 5 North SW20 506627 157229 5 North SW21 505058 156294 6 North SW22 505019 156287 6 North SW23 505010 156294 6 South SW24 504965 156243 6 South SW25 503636 154112 7 South SW26 503609 154156 7 South SW27 503610 154099 7 North SW28 503652 154117 7 North SW29 505043 152917 8 South SW30 504956 152908 8 South SW31 505140 153012 8 North SW32 505234 153131 8 North SW33 505741 153711 9 South SW34 505738 153720 9 South SW35 505702 153726 9 North SW36 505724 153754 9 North SW37 506529 155333 10 South SW38 506540 155467 10 South SW39 507297 154927 10 North SW40 507326 154927 10 North

Tree density Mean_B Mean_G High 1.0 0.1 High 0.9 9.1 Low 1.2 3.6 Low 13.9 100.3 High 0.2 0.9 High 3.0 17.4 Low 7.1 6.4 Low 1.9 10.9 Low 4.3 13.2 Low 15.1 22.0 High 9.9 21.3 High 3.4 19.9 High 0.8 2.7 High 0.1 0.6 Low 0.3 21.7 Low 0.1 10.8 High 0.2 1.2 Low 0.8 18.6 High 0.7 6.7 Low 0.0 8.6 Low 0.3 7.1 High 0.2 7.3 High 1.9 13.1 Low 0.0 0.6 Low 7.6 7.1 High 9.9 2.0 High 0.9 2.6 Low 6.2 1.3 High 1.1 3.0 Low 21.8 89.1 Low 0.6 1.1 High 0.4 17.2 Low 9.6 111.3 High 46.7 75.6 High 10.7 91.0 Low 1.1 58.4 Low 19.7 63.3 High 43.1 147.3 High 18.4 82.2 Low 31.0 65.7

165

Appendix 2

Tree Easting Northing Block SW41 508054 156245 11 SW42 507850 156392 11 SW43 507830 156375 11 SW44 508017 156526 11 SW45 505841 158810 NA SW46 506030 158774 NA SW47 509180 157302 NA SW48 505179 152857 NA SW49 504669 152611 NA SW50 504653 152673 NA SW51 503517 150980 NA SW52 503642 150465 NA SW53 503953 150064 NA SW54 503844 150097 NA SW55 503417 150595 NA

166

Exposition North South South North NA NA NA NA NA NA NA NA NA NA NA

Tree density Mean_B Mean_G High 5.1 48.2 High 3.9 115.6 Low 2.0 3.3 Low 7.9 40.2 Freestanding 14.1 33.0 Freestanding 1.6 3.1 Freestanding 0.2 2.3 Freestanding 2.9 1.7 Freestanding 11.3 26.8 Freestanding 7.8 24.3 Freestanding 0.1 19.1 Freestanding 23.8 17.0 Freestanding 0.2 6.0 Freestanding 24.2 11.8 Freestanding 98.3 112.7

Appendix 3

Appendix 3: Central coordinates of 1-ha plots in the Parc Jurassien Vaudois in Switzerland where snow was collected. Per plot, 10 snow samples were collected each. Plot R02 R08 R14 R20 R21 R22 R26 R29 Wer001 Wer006 Wer009 Wer013 Wer020 Wer027 Wer043 Wer056 Wer075 Wer079 Wer082 Wer091 Wer092 Wer096 Wer098 Wer100

X_Coord 507143 507153 507137 506518 506738 506944 506284 506193 505768 506449 505196 507701 507255 505843 507246 507471 505633 505677 507464 506679 507474 507703 505556 507575

Y_Coord 154874 155319 155761 155238 155998 155269 155894 155406 155598 155386 156167 155130 154576 155722 154353 154871 155495 156278 154629 155500 155602 155245 156221 155104

Stratum Open area, situated in pasture Open area, situated in pasture Open area, situated in pasture Open area, situated in pasture Open area, situated in pasture Open area, situated in pasture Open area, situated in pasture Open area, situated in pasture Without Lobaria pulmonaria Without Lobaria pulmonaria More than three trees colonised by Lobaria pulmonaria More than three trees colonised by Lobaria pulmonaria Without Lobaria pulmonaria Without Lobaria pulmonaria More than three trees colonised by Lobaria pulmonaria Without Lobaria pulmonaria More than three trees colonised by Lobaria pulmonaria More than three trees colonised by Lobaria pulmonaria More than three trees colonised by Lobaria pulmonaria Without Lobaria pulmonaria Without Lobaria pulmonaria More than three trees colonised by Lobaria pulmonaria More than three trees colonised by Lobaria pulmonaria Without Lobaria pulmonaria

167

Acknowledgements

Acknowledgments Christoph Scheidegger has been the best supervisor that I could have dreamt of. Christoph, you have always been positive, encouraging and optimistic. I am very thankful for enumerous exciting scientific discussions and your never-ending enthusiasm for lichens, which made that I fortunately never forgot that some of the transparent liquids I was juggling with in the lab actually originally were most-beautiful-and-ever-sofascinating lichen Lobaria pulmonaria (Yes, Rolf and Felix – lichens ARE beautiful. And only a living Lobaria is a good Lobaria, to state it clearly this one time). During the course of my thesis, I have had a great number of fascinating scientific discussions with Helene Wagner. It has often surprised me how Helene could solve problems that would cost me a day in five minutes, and here I’m not only talking about the R programming. Helene, your advice and guidance has really been a great inspiration for and input into my thesis. Even though I assume that you are more fascinated by numbers, models and ecological theory than by soredia, isidia and apothecia, I hope that the work on spatial genetic structure, clonality and recombination of Lobaria pulmonaria has been equally fantastic for you as it was for me. Brigitta Ammann accepted me as an external student at the University of Berne. You were always really supportive and enthusiastic about my work, Brigitta, and I really admire the breadth of your knowledge. Sharing an office with Jesse Kalwij for over three years was a very good experience – it never got boring, and now that I think of it, I cannot come up with very many instances where we shared opinions on any particular subject. I very much appreciate your cheerfulness and openness, Dutch lifestyle, and your true pioneer spirit, e.g. when it came to building showers in a cottage in the Marchairuz carstic area where not even water was accessible (except for the plenty rainfalls, which both of us got a good share of). Rolf Holderegger, Felix Gugerli, and Susan Hoebee gave lots of input on the analysis of population genetic data, and they revised manuscripts, posters and talks. I am very glad that you took the time – thank you for everything! My honest thanks are due to Daniela Csencsics, Jesse Kalwij, Johanna Scheidegger, Svetlana Tschabanenko and my dad Klaus Werth for field assistance, to poor Fabienne Föry who filtered 1000 snow samples during weeks, and to Marcus Hofer, who prepared the snow samples and assisted in the snow-sample DNA extractions. Geography students Stefan Schmid and Lukas Bischof have

169

Acknowledgements

constructed extraordinarily beautiful maps of the study area displaying the multilocus genotype data. I’m sorry that the maps were not included in the thesis, but they will be published in the modelling paper by Helene Wagner which is about to be submitted. I’m very grateful to Henrik Hedenås for numerous discussions on lichen ecology and dispersal, and to Gwenaël Jacob for the exchange of ideas on population genetics. I would like to express my gratitude to the friendly Genetic Ecology team, and particularly to the members of the lab team who have been supportive. It was nice to be a Ph.D. student at WSL. The WSL is a great place for performing a Ph.D., given the many friendly people who are working here, and all the facilities of the WSL. The WSL library team is doing a job that goes bejond daily business in obtaining literature that is hard to get access to – thanks to Gret Nebel, Claudia Berger, Alois Kempf, Christine Matter and others. Thanks to Rosemarie Honegger (University of Zürich), I learned how to obtain sterile cultures of lichen apobionts. I’m grateful to Sebastien Sachot (Centre de conservation de la faune et de la nature, Vaud) for permits for performing winter fieldwork in Marchairuz and to the Vogelwarte Sempach and the Kanton Vaud for permits for catching birds in Marchairuz. Fränzi Korner-Nievergelt (University of Zürich) helped greatly with discussions on bird catching and bird dispersal of lichens, and so did Beatrice Miranda from the WSL. This research is part of a project funded by the Swiss National Science Foundation (SNF) under the NCCR Plant Survival. Thanks to the NCCR Plant Survival, I could improve my research skills considerably by taking courses at the NCCR graduate school of the University of Neuchatel. I’m very grateful to my friend Shyam Nyati for his incredible cheerfulness and kindness, and for sharing a lot of ideas about Indian culture, lifestyle, and also recipes with me over the last three years. Thanks to my flatmates Simone Ammann and Sandy Nussbaumer, my stay in Zürich was a pleasure and I even felt a bit at home. My parents Erika and Klaus Werth, my brother Henning Werth and my sister in law Birgitt Windschuh have always been very supportive during these years, and without my parents’ generous financial aid, I would never have been able to study biology in the first place. Thanks to all of you!!

170

Curriculum vitae Silke Werth

Curriculum vitae of Silke Werth Education 1/2002 – 06/2005

Ph.D. studies at the University of Berne Institute/University: Landscape at Swiss Federal Institute of Forest, Snow and Landscape Research WSL (Switzerland) and Institute of Plant Sciences, University of Bern (Switzerland) Dissertation: “Dispersal and persistence of the epiphytic lichen Lobaria pulmonaria in a dynamic landscape”. Topics: Landscape genetics, genetic diversity in relation to stand-level disturbance, dispersal parameters Advisors: Prof. Dr. Brigitta Ammann (Institute of Plant Sciences, University of Berne), PD Dr. Christoph Scheidegger (Swiss Federal Institute of Forest, Snow and Landscape Research WSL), PD Dr. Helene Wagner (WSL)

1/2000 – 12/2001

Master’s studies at the University of Tromsø, Norway (Cand. Scient.) Studies in: Biology (community ecology and lichenology) University: Institute of Biology, University of Tromsø Thesis: “Key factors for epiphytic macrolichen vegetation in deciduous forests of Troms county, northern Norway: human impact, substrate, climate or spatial variation?” Topics: Vegetation surveys, analysis of epiphytic macrolichen communities, environmental gradients, response of lichens to human impact, forest stand properties, geographic location and macroclimate Advisor: Dr. Arve Elvebakk (University of Tromsø) Co-advisor: Dr. Hans Tømmervik (Norwegian Institute for Nature Research, Tromsø)

11/1997 – 12/1999 Diploma studies at the Technical University of Munich Studies in: Ecological biology (applied ecology, animal ecology, limnology, vegetation ecology) University: Technical University of Munich 10/1995 – 10/1997 Bachelor’s studies at the University of Ulm (Bachelor’s degree) Studies in: Biology (undergraduate level) University: University of Ulm

171

Curriculum vitae Silke Werth

Grants and awards April 2004

NASA-MSU award, to aid the participation in the US-IALE symposium; additionally a travel grant by the NCCR Plant Survival

September 2003

Travel grant by NCCR Plant Survival to participate in the German Ecological Society meeting in Halle, Germany

September 2001

Award for best poster presentation. International Association of Vegetation Science IAVS Symposium, Freising, Germany

August 2001

Grant for field expenses, Kometenfund, University of Tromsø, Norway

August 2000 – April 2001

Annual grant by the German Academical Exchange Service DAAD

August 1999 – July 2000

Annual grant, Norwegian Research Council

March 1996

Award (regional environmental award) for a vegetation mapping study in a peat bog and its implications for species conservation

Computer literacy (PC) Statistical software

R, S+, PC-Ord, Canoco, SPSS

Population genetics software Arlequin, Structure, Barrier, GDA, SGS, Genepop, Microsat, Genescan, Genemapper, Sequencing Analysis, Autoassembler, Sequence Detection Analysis Programming

R, C++ (basic knowledge)

GIS

ArcGIS, ArcView, Idrisi

Web design

Dreamweaver, Express Thumbnail Creator

Office applications

Microsoft

Languages German

mother tongue

English

fluent

Norwegian

excellent, lived and studied in Norway

French & Spanish

basic knowledge

Other interests Mental training, hiking, bicycle tours, reading, flamenco, nature observation, bird watching

172