aspects of heterogeneity - diva portal145107/fulltext01.pdfaspects of heterogeneity: ... avläggande...

32
Aspects of heterogeneity: Department of Ecology and Environmental Science Umeå University 2006 Marcus Åström effects of clear-cutting and post-harvest extraction of bioenergy on plants in boreal forests

Upload: vannhu

Post on 20-Mar-2019

223 views

Category:

Documents


0 download

TRANSCRIPT

Aspects of heterogeneity:

Department of Ecology and Environmental ScienceUmeå University

2006

Marcus Åström

effects of clear-cutting and post-harvest extraction of bioenergy on plants in boreal forests

Department of Ecology and Environmental ScienceUmeå UniversitySE-901 87 Umeå

Sweden

AKADEMISK AVHANDLING

som med vederbörligt tillstånd av rektorsämbetet vid Umeå universitet för avläggande av fi losofi e doktorsexamen i Ekologi, offentligen försvaras fredagen

den 8 december kl. 13.00 i sal Björken, Sveriges lantbruksuniversitet,

Umeå

Examinator: Professor Christer Nilsson, Umeå universitet

Fakultetsopponent: Dr. Katherine Frego, Biology Department, University of New Brunswick, Saint John, Canada

Aspects of heterogeneity:

Marcus Åström

effects of clear-cutting and post-harvest extraction of bioenergy on plants in boreal forests

2006

OrganisationUMEÅ UNIVERSITYDept. of Ecology and Environmental ScienceSE-901 87 Umeå, Sweden

AuthorMarcus Åström

Title

Document typeDOCTORAL DISSERTATION

Date of publicationNovember 2006November 2006November

Aspects of heterogeneity: effects of clear-cutting and post-harvest extraction of bioenergy on plants in boreal forests

Language: English

Signature:

ISBN: 91-7264-192-4 Number of pages: 26 + 4 papers

Abstract. The objectives of this thesis are to evaluate (1) the infl uence of slope aspect on boreal plant respon-ses to clear-cutting and (2) the effects of post-harvest extraction of bioenergy (logging residues or slash) on plant composition, richness and performance in clear-cuts. Such insight is essential for understanding chan-ges in species composition and richness in response to clear-cutting and application of intensifi ed harvesting systems. The focus is on productive and managed spruce dominated forests and focal organisms are mosses, liverworts (i.e. bryophytes) and vascular plants. Space-for-time substitution studies were performed in south- and north-facing slopes located in 10 forests and 10 adjacent clear-cut stands in central Sweden. Differences between forests and clear-cuts were interpre-ted as effects of clear-cutting. The results show that the response of all three focal groups differed between aspects. More species were lost in south-facing slopes and clear-cutting reduced species richness of liverworts as well as of bryophytes and vascular plants associated with sheltered habitats. By contrast, clear-cutting caused no reduction in any group and more species were added in north-facing slopes. As a result north-fa-cing clear-cuts generally had higher species richness than their forest counterparts. The disparate patterns in species’ response between aspects were most likely caused by initial microclimatic differences and a greater microclimatic change in south-facing slopes, in response to clear-cutting. A paired comparative study of conventionally harvested (i.e. slash left) and slash-harvested clear-cut stands was performed 5-10 years after clear-cutting in south-central Sweden. Both the species composition and the richness of mosses and liverworts were affected by slash harvest, whereas the composition of vascular plants was not. Slash harvest also reduced richness of mosses and liverworts associated with forests and organic substrates (e.g. dead wood and litter). Species richness of vascular plants and bryophytes associated with inorganic substrates (i.e. mineral soil) was unchanged. Differences between conventionally harvested stands and slash-harvested stands were most likely a result of reduced cover of organic material reducing substrate availability and shelter in the latter. Increased mechanical disturbance in slash-harvested stands that destroys remnant forest vegetation and favours pioneers may also play a role. A bryophyte transplant experiment was performed in seven clear-cuts in central Sweden and monitored over one vegetation period. The results show that logging residues (or slash) and forest edges may shelter ground-dwelling bryophytes by buffering the clear-cut microclimate. In conclusion, both slope aspect and extraction of forest bioenergy affect plant survival in clear-cut boreal forests. As surviving plant populations facilitate re-colonisation, north-facing slopes and conventionally har-vested clear-cuts (i.e. slash left) may potentially recover faster than south-facing slopes and slash-harvested clear-cuts.

Key words: aspect; clear-cutting; disturbance; habitat quality; liverworts; logging residues; mosses; resilience; slash; slope; topography

SammanfattningSyftet med denna avhandling är att utvärdera (1) betydelsen av skogsmarkens lutning samt (2) betydelsen av GROT-uttag (GRenar Och Toppar) för växters (bladmossor, levermossor och kärlväxter) överlevnad och etablering efter avverkning av grandominerade skogar. Betydelsen av skogsmarkens lutning utvärderades genom att jämföra syd- och nordslutt-ningar i äldre skog med motsvarande sluttningar på hyggen eller i ungskog. Resultaten visar att lutningen i stor grad styr hur både mossor och kärlväxter reagerar på skogsavverkning. Ett större antal arter försvann i sydsluttningar som blev fattigare på levermossor och skogslevande mossor och kärlväxter. I nordsluttningar försvann överlag inte lika många arter och ett större antal arter tillkom vilket gjorde att nordvända hyggen generellt blev artrikare jämfört med nordvända skogar. Resultaten tyder på att skillnaden i växters respons mellan olika sluttningar beror på initiala skillnader i artsammansättning och artrikedom mellan nord- och sydslutt-ningar och en större mikroklimatisk förändring efter avverkning i sydsluttningar. Effekter av risanpassad avverkning och GROT-uttag utvärderades genom att jämföra 14 plant- och ungskogar som skördats på GROT med 14 plant- och ungskogar där inget GROT-uttag skett. Resultaten visar att både artsammansättning och artrikedom bland bladmossor och levermossor påverkas av GROT-uttag. Framförallt missgynnades levermossor och bladmoss-arter som är beroende av skyddade miljöer och organiska substrat. Däremot påverkades inte artsammansättning eller artrikedom bland kärlväxter. Orsaken till de påvisade skillnaderna bedöms vara (1) minskad substratmängd för ved- och förnalevande arter, (2) minskat skydd mot uttorkning och (3) ökad grad av störning på GROT-skördade hyggen. Betydelsen av grenar och toppar som skydd under hyggesfasen utvärderades i ett experi-ment där skogslevande markmossor transplanterades till skyddade miljöer (nordvända skogs-kanter) och relativt exponerade hyggesmiljöer (hyggescentrum). I varje miljö täcktes hälften av mosstransplantaten med färska grangrenar och deras tillväxt och vitalitet följdes under ca en tillväxtsäsong. Resultaten visar att GROT skyddar skogslevande växter genom att dämpa variationerna i hyggesklimatet men också att effekterna var störst i den mer extrema hygges-miljön. De huvudsakliga slutsatserna är att både skogsmarkens lutning och GROT-uttag påverkar skogslevande arters överlevnad under hyggesfasen. Lokal överlevnad påverkar återkolonisa-tion. Det är därför troligt att nordvända skogsbestånd och bestånd där inget GROT-uttag skett kommer att återhämta sig fortare än sydvända bestånd och GROT-skördade bestånd efter av-verkning.

LIST OF PAPERS

INTRODUCTIONClear-cuttingExtraction of forest bioenergy

FOCAL ORGANISMS

OBJECTIVES

STUDY AREAS

METHODSSpace-for-time substitution studies Space-for-time substitution studies SThe comparative studyThe bryophyte transplant experiment

MAJOR RESULTS AND DISCUSSIONAJOR RESULTS AND DISCUSSIONAJOR RESULAspect modifi es boreal plant respboreal plant respboreal onses to clear-cuttingSouth- and north-facing forestsPost-harvest extraction of logging residuesThe sheltering effect of forest edges and logging residues

SYNTHESIS

IMPLICATIONS FOR MANAGEMENTSelection of standsRetention of treesRetention of treesRetention of trSubstrateMechanical disturbance

PERSONAL REFLECTIONS

ACKNOWLEDGEMENTS

REFERENCES

TACK!

1

123

4

5

6

6899

1010111213

14

1516161617

17

18

18

25

TABLE OF CONTENTS

LIST OF PAPERS

This thesis is a summary of the following papers, which will be referred to by their Roman numerals.

INTRODUCTION

Historically, external and internal disturban-ces such as fi re, insect attacks, tree-falls and uprootings were the main driving forces sha-ping successional development in boreal fo-rest ecosystems in Sweden (Zackrisson 1977; Engelmark 1999). However, since the mid- and late- 1800s, large shifts have taken pla-ce in the Swedish boreal forest ecosystems, mainly due to large scale tree-harvesting for saw and pulp timber. After the introduction of clear-felling in the 1950s forest management has mainly focused on silviculture that pro-motes even-aged and even-spaced monocul-tures dominated by conifers (Ebeling 1959; Axelsson 2001). Since then, practices such as selective cleaning, thinning, clear-cutting and short rotation periods have caused major chan-ges in forest structure and habitat heterogen-eity. Consequently, single-storied stands with a high proportion of clear-cuts and younger

stands have replaced the multi-storied forest mosaics, which contained large volumes of dead wood, that dominated the boreal forest landscape in the past (Östlund 1993). Exo-tic species, particularly the North American lodgepole pine (Pinus contortalodgepole pine (Pinus contortalodgepole pine ( )Pinus contorta)Pinus contorta , have also been introduced (Engelmark et al. 2001). At present, protected forest in Sweden ac-counts for roughly 1% of the area of pro-ductive forest land below the limit for alpine forests (Anonymous 2005), whereas mana-ged forests older than 100 years and younger than 50 years account for roughly 22% and 52%, respectively, of the cover of productive land (Anonymous 2006). This transforma-tion has heavily affected the spatial distri-bution of plants and animals (Bernes 1994).

Knowledge of the effects of forest mana-gement and of the spatial and temporal sta-tus of managed forests is of great impor-tance for developing sustainable mana-gement strategies. Species assemblages

Åström, M., Dynesius, M., Hylander, K. and Nilsson, C. Slope aspect modifi es community responses to clear-cutting in boreal forests. Ecology, in press.

Åström, M., Dynesius, M. and Nilsson, C. Responses of understorey vascular plants to clear-cutting of boreal forests in north- and south-facing slopes. Ma-nuscript.

Åström, M., Dynesius, M., Hylander, K. and Nilsson, C. 2005. Effects of slash harvest on bryophytes and vascular plants in southern boreal forest clear-cuts. Journal of Applied Ecology 42, 1194-1202.

Åström, M., Dynesius, M. and Nilsson, C. Logging residues and forest edges affect bryophyte vitality and growth in boreal forest clear-cuts. Manuscript.

Papers I and I and I III are reproduced with permission from the publishers. III are reproduced with permission from the publishers. III

I.

II.

III.

IV.

-1-

affect ecosystem function and may, for ex-ample, infl uence net primary production, retention of moisture and nutrients, nitro-gen-fi xing, and tree regeneration (Fahey et al. 1991; Shelton and Cain 2000; Hanseen 2003). Furthermore, the quality of the mana-ged forest matrix is likely to infl uence lands-cape-scale resilience and conservation efforts in adjacent reserves (Bengtsson et al. 2003). Maintaining or promoting species diversity in managed forests has been suggested as a me-ans of reducing the need for reserves and/or improve their function (Saunders et al. 1991; Bengtsson et al. 2003); compensatory effects of (remnant) habitats within managed forests where species survive may be considered step-ping stones for genes and/or species and they may also infl uence population vitality or (re-) colonisation. In contrast, increased fragmen-tation eventually drives species to extinction (Tilman et al. 1994; Brotons et al. 2003).

Clear-cuttingFew forest management practices have arou-sed as much public attention and debate as cle-ar-cutting. The forest microclimate and hence transformations induced by clear-cutting ge-nerally show a strong relationship with tree structure (Chen et al. 1999; Drever and Lertz-man 2003), although other vegetation strata and physical factors are also important (Chen et al. 1999; Prevost and Pothier 2003). Cano-py removal causes drastic changes in micro-climate by increasing the input of short-wave radiation which, in turn, elevates daytime soil, near-ground and ground-surface temperatu-res (Keenan and Kimmins 1991; Chen et al. 1995). Clear-cuts generally experience higher near-ground wind velocity as well (Chen et al. 1993). Clear-cutting transforms substrate availabi-lity by removing trees and by disturbing the ground or patchy substrates such as woody debris or boulders, but it also creates new mi-crohabitats, such as patches of exposed mine-ral soil and fresh woody debris. In addition,

tree harvest and subsequent transformations in microclimate and substrate availability af-fect edaphic conditions. Clear-cutting of bo-real forests tends to increase soil moisture by reducing evapotranspiration and because a higher proportion of the precipitation reach-es the ground (McColl 1977; Zobel 1993). Furthermore, greater inputs of short-wave ra-diation and woody debris tend to stimulate de-composition, which increases soil pH and the amount of nutrients available to plants (Swift et al. 1979; Nykvist and Rosén 1985; Fisher and Binkley 2000). Changes in microclimate, substrate avai-lability and edaphic conditions are of critical importance for the survival and colonisation of forest organisms after clear-cutting. In ge-neral, organisms associated with late-succes-sional stages are negatively affected whereas pioneers are favoured. However, different groups of organisms may respond differently to clear-cutting. For example, although large changes in the abundance of forest vascular plants may occur, few species disappear com-pletely (Kardell 1992; Esseen et al. 1997). Si-multaneously, pioneer species are favoured, which sometimes enhances total richness of vascular plants (Dyrness 1973; Kardell 1992; Nykvist 1997; Bergquist et al. 1999). Other organisms, such as forest bryophytes, are more negatively affected; they cannot effec-tively control water uptake or loss (Proctor 1990) and they inhabit exposed and elevated substrates, like woody debris, that desiccates easily (Hylander et al. 2005). This is fatal for bryophyte species, especially liverworts, that have not evolved tolerance to desiccation. Although negatively affected, many fo-rest organisms may survive in microhabitats where the post-harvest conditions are more similar to those of the pre-harvest situation. Therefore, the effects of tree harvest are linked to the intensity and spatial extent of disturbance (Halpern 1988; Pimm 1991) but also to environmental heterogeneity (Frank and McNaugthon 1991; Auerbach and Schmi-

-2-

da 1987). For example, remnant canopies may moderate the microclimate and its negative ef-moderate the microclimate and its negative ef-moderate the microclimate and its negative effects on forest bryophytes (Fenton and Frego 2005; Hylander et al. 2005; Heithecker and Halpern 2006), and shade from shrubs, logs, slash and stumps may improve survival and growth of shade-adapted plants in otherwise exposed habitats (Busby et al. 1978; Minore 1986; Gray and Spies 1997; Bråkenhielm and Liu 1998). Furthermore, disturbance regime and effects of disturbance may be related to physical landscape features that may play an important role in structuring vegetations pat-terns (Wimberly and Spies 2001). Given the strong infl uence that topography has on mi-croclimate and species distribution (Chen et al. 1999) it is reasonable to assume that bio-tic responses to clear-cutting vary with slope aspect. For example, Olivero and Hix (1998) and Økland et al. (2003) suggested that plant persistence and re-colonisation after disturban-ce may be affected by slope aspect, although they did not specifi cally address this question. Understanding the interaction between slope aspect and clear-cutting is essential for under-standing the effect that clear-cutting has on community dynamics, species composition and species richness in managed forest lands-capes with heterogeneous topography.

Extraction of forest bioenergyIn recent decades (from the 1970’s oil-crisis and onward), large scale commercial systems for post-harvest extraction of slash (i.e. log-ging residues composed of treetops, branches and twigs) have been introduced for bioenergy purposes. Converting fossil-based energy sys-tems into CO2-neutral systems, such as those based on bioenergy, is an essential strategy for reaching targets set by the Kyoto Proto-col (Gustavsson et al. 1995; Gururaja 2003). In regions with forestry industry, post-harvest extraction of slash is likely to increase in re-sponse to increased public demand for bioen-ergy but also because slash removal facilitates site-preparation and regeneration of commer-

cial tree species (Egnell et al. 1998; Lundborg cial tree species (Egnell et al. 1998; Lundborg cial tree species (Egnell et al. 1998; Lund1998; Hakkila 2003). In fact, in southern Swe-den, extraction of slash following clear-cutting of spruce dominated stands is already a rule rather than an exception and the demand for forest bioenergy is increasing rapidly (ErikLing, Sveaskog, personal communication). To date, commercial slash extraction has most commonly been practiced following cle-ar-felling of well-stocked coniferous forests on sites that have few technical constraints and moderate or high site productivity. Typi-cally, the logging residues are piled within the clear-cut area during tree harvest. Forwarding (extraction) of logging residues may occur shortly after timber hauling or after a period of needle-dehiscing to minimise loss of nu-trients. The latter method is recommended by the Swedish Forestry Agency except in areas with high nitrogen deposition (Anonymous 2001). Although extraction of slash could help combat global warming and may offer bet-ter regeneration of commercial tree-species, it may also have ecological drawbacks. For example, due to loss of shading structures and reduction in surface roughness, slash extraction augments the effects of clear-fel-ling on microclimate by increasing near-ground wind velocities and daytime tempe-ratures (Mahendrappa and Kingston 1994; Proe et al. 1994). In addition, it contributes to further changes in substrate availability as branches, tree-tops and needles are lost (Eg-nell et al. 1998). This reduces habitat hete-rogeneity and substrate availability that may be important components for many species of plants, animals and fungi (Samuelsson et al. 1994; Gunnarsson et al. 2004; Nordén et al. 2004; Nittérus 2006). Therefore, slash ex-traction may represent an intensifi cation of forest management that counteracts efforts to enhance conditions for wood-inhabiting or-ganisms in managed boreal forests (e.g. crea-tion of coarse woody debris (CWD; >10 cm diameter). Also, increased mechanical distur-

-3-

bance by forwarders extracting the logging residues is likely to increase damage to stru-tural components such as the ground surface and CWD (Egnell et al. 1998; Hakkila 2003; Rudolphi and Gustafsson 2005). In previous studies, slash extraction has been reported to affect post-harvest vegetation dy-namics (Olsson and Staaf 1995; Bråkenhielm and Liu, 1998; Bergquist et al. 1999), and to reduce tree growth (Egnell and Leijon 1999; Egnell and Valinger 2002) and the abundan-ce of soil arthropods due to reduced input of organic material and nutrients (Bengtsson et al. 1997). It has also been proposed that log-ging residues may provide shelter to drought- sensitive species (McInnis and Roberts 1994; Bråkenhielm and Liu 1998). With respect to composition, richness and abundance of bo-real forest plants, previous studies examining the effects of slash harvesting have been res-tricted to few sites and small plots. As a re-sult, the effects were evaluated only for a few abundant species, most of which were vascu-lar plants (e.g. Olsson and Staaf 1995; Nykvist 1997; Bergquist et al. 1999). Few studies have focused on less common vascular plants or bryophytes. In addition, most previous studies were designed as manipulative experiments in which the logging residues were evenly spread (sometimes manually) or excess slash added. These procedures have little resemblance to commercial forest practices in Sweden, which are highly mechanised and result in clear-cuts where slash is either abundant or scarce. This is even more apparent in clear-cuts that will be subjected to extraction of logging residues as branches and tree tops are piled. Under-standing the effects of this intensifi ed harves-ting system in practice, as well as evaluating the mechanisms behind possible effects, are essential for sustainable and responsible ma-nagement of boreal forest landscapes.

FOCAL ORGANISMS

This thesis focuses on mosses, liverworts

(bryophytes) and vascular plants. They are all important constituents of boreal forest ecosys-tems, both in terms of abundance and species richness. They also infl uence primary produc-tion, cycling of nutrients, retention of water and nutrients and provide nesting material, substrate and food for a wide array of orga-nisms (e.g. Shelton & Cain 2000; Gentili et al. 2005; Solga and Frahm 2006). The distribution and abundance of these th-ree groups are highly affected by disturbances such as clear-cutting. In general, late-succes-sional species are negatively affected while pioneers are favoured (e.g. Bråkenhielm and Liu 1998; Uotila and Kuoki 2005). Howe-ver, vascular plants and bryophytes represent a gradient in sensitivity to exposure caused primarily by ecophysiological differences. Vascular plants may regulate water uptake more effectively because their roots enables them to retrieve water and nutrients from the subsurface soil and because their ecophysio-logical regulation of water loss is more ad-vanced. Consequently, few species disappear completely following clear-cutting although large changes in their relative abundance may occur (Esseen et al. 1997). In contrast, bryop-hytes are poikilohydric (i.e. most species lack effective mechanisms for controlling water uptake or loss) and are therefore particularly dependent on the ambient microclimate for survival, growth, reproduction and dispersal (Schofi eld 1985; Proctor 1990). Although there is large variation in tolerance within groups, the liverworts are generally more sen-sitive to an altered microclimate than mosses and so a greater proportion decline when ex-posed (Marshall and Proctor 2004; Hylander et al. 2005). Furthermore, due to their large ecophysiological amplitude, both mosses and liverworts in boreal forests may grow on most substrates. However, a larger proportion of the liverworts are restricted to elevated substrates such as stumps, logs and tree-bases. These habitats desiccate more easily when exposed, thus contributing to the relatively higher sen-

-4-

sitivity of liverworts. Bryophyte species with different substrate/habitat associations have been shown to respond differently to changes in microclimate (Hylander et al. 2005). In pa-pers [I] and [I] and [I III], I used habitat and substrate III], I used habitat and substrate IIIspecifi city as indicators of the ecological ef-fects of clear-cutting and how this effect is modifi ed by slope aspect and extraction of logging residues.

All vascular plants dealt with in this thesis are exclusively ground-living. Vascular plants may also be used as ecological indi-cators of environmental conditions as their distribution and abundance are highly af-fected by light availability, as well as the

moisture and nutrient content of the soil.In paper [II]II]II I employed Ellenberg indica-

tor values (Ellenberg 1992), which are com-monly used for describing environmental change based on observed changes in species composition (Bennie et al. 2006; Fanelli et al. 2006). They may be used to monitor ef-fects of disturbance and associated changes in resource availability (Fanelli et al. 2006) and are particularly effective in comparative studies at larger spatial scales (Pignatti et al. 2001). In paper [II]II]II they were used to explore which environmental factors mediate compo-sitional change in response to clear-cutting in north- and south-facing slopes.

OBJECTIVES

The broad objective of this thesis is to improve our understanding of the effects of clear-cut-ting and post-harvest extraction of logging residues on the species richness and the abundance of boreal forests organisms. The results and ideas presented in this thesis will hopefully pro-vide some insight on how to adapt management and mitigate negative effects of these practi-ces. The focus is on productive and managed spruce-dominated boreal forests and the study organisms are mosses, liverworts and vascular plants.

Specifi c objectives are:

To study the effect of slope aspect on the biotic responses to clear-cutting [papers I and I and I II].II].II

To investigate the infl uence of slope aspect on species composition and rich-ness [papers I and I and I II].II].II

To study the effects of commercial post-harvest extraction of logging resi-dues on species composition and richness [paper III]. III]. III

To evaluate the sheltering effect of logging residues and stand edges on the vitality and growth of ground-dwelling forest bryophytes [paper IV].IV].IV

-5-

STUDY AREAS

The work presented in this thesis originates from three different regions (Fig. 1). The ma-ture forests in these areas are dominated by Norway spruce (Picea abiesNorway spruce (Picea abiesNorway spruce ( L. Karst) and Scots Pine (Pinus sylvestrisScots Pine (Pinus sylvestrisScots Pine ( L.) with scatte-red broadleaves (Betula pubescensred broadleaves (Betula pubescensred broadleaves ( Ehrh., B. pendula Roth, and Populus tremula L.). The forests are intensively managed for wood pro-duction, and clear-cutting has been the main form of timber harvest for the last 60 years (Wastenson and Nilsson 1995). The fi eld work for papers [I] and [I] and [I II] was II] was IIaccomplished in the middle boreal zone (Ahti et al. 1968) in the province of Ångermanland, between latitudes 63°35’ and 64°05’ N (Fig. 1a). The length of the growing season is ~150 days. Mean annual precipitation is ~750 mm and mean temperatures for January and July are -9ºC and +15ºC, respectively (Raab and Vedin 1995). The study sites were distributed along the valleys of the Öre and Lögde ri-vers, with altitudes ranging 120 to 190 m a.s.l. which is below the former highest coast line. This area is characterised by thick delta se-diments that have formed terraces intersected by ancient gully systems with scarps, ridges and ravines up to 50 m high. The fi eld work for paper [III] was conducted III] was conducted IIIin Örebro County, in the southern boreal zone (Ahti et al. 1968) in south central Sweden bet-ween latitudes 58°40’ and 59°35’ N (Fig. 1c). In this region post-harvest extraction of log-ging residues is a norm rather than an excep-tion. The length of the growing season ranges from 180 to 190 days and mean annual preci-pitation is 700–800 mm (Sjörs 1999). Mean temperatures in January and July are -4ºC and +16°C, respectively (Sjörs 1999). The study sites were located 95 to 260 m a.s.l. The fi eld work for paper [IV] was car-IV] was car-IVried out in the middle boreal zone of Ahti et al. (1968) in the province of Ångermanland between latitudes 62°50’ and 63°05’ N (Fig. 1b). The growing season is ~150 days. Mean

annual precipitation is ~750 mm and mean temperatures for January and July are –7°C and +16°C, respectively (Raab and Vedin 1995). Sites were located 151 to 239 m a.s.l.

METHODS

Most of the study sites and stands in all four studies presented in this thesis were located to bilberry-spruce forests (Piceeto-Vaccinietum bilberry-spruce forests (Piceeto-Vaccinietum bilberry-spruce forests (myrtilli) or low-herb spruce forests (Piceeto ) or low-herb spruce forests (Piceeto ) or low-herb spruce forests (oxalidetum acetosellae; Sjörs 1965). Site pro-ductivities range from medium to high (domi-nant height at age 100 years ranges from 22 to 33 m) and the soils are typically mesic. The parent soil material consists of sandy-loamy till or sandy-loamy sediments with underly-ing bedrock of gneiss or granite and soils are mainly podsolics (Fredén 1994). The term “stand” is defi ned as a cluster of trees and associated vegetation of similar struc-ture growing under similar conditions (Oliver and Larsson 1990). The stand is suffi ciently similar to be considered a fairly homogenous ecological unit and is managed uniformly. For convenience, the term “clear-cut” or “clear-cut stand” refers to former forest stands where all commercially valuable trees have been re-cently harvested (≤ 22 years), although scat-tered remnant trees or regenerated small trees may be present. From a successional point of view, the tem-poral scale of my studies on effects of clear-cutting and slash extraction can be considered short-term because they mainly focus on ini-tial effects of canopy disturbance and altered habitat conditions (Halpern and Spies 1995). Papers [I and I and I II] include clear-cuts that were II] include clear-cuts that were IIharvested 3–22 years before our studies. Pa-per [III] includes 5–10 year old clear-cuts, III] includes 5–10 year old clear-cuts, IIIwhereas paper [IV] only deals with the fi rst IV] only deals with the fi rst IVgrowing season following clear-cutting. Ho-wever, it is often the extremes that govern the distribution patterns of species (Crawford 1989), and clear-cutting is the most extreme (positive or negative) event for the majority of

-6-

plants encountered in managed boreal forests.Throughout this thesis I principally use the

term “richness” or “species richness” for descri-bing the number of species within a defi ned spa-tial unit (also called α-diversity). I use the word “composition” as species richness and iden- tity combined with an abundance parameter (in this case, frequency). The word “turnover” is used for describing the change or difference in species richness or composition between stands, hence “species turnover” or “compo-sitional turnover” (also called β-diversity).

The main focus of all my studies (except pa-per [IV]) was to evaluate the effects on boreal IV]) was to evaluate the effects on boreal IVplants on large spatial scales. Therefore, I used large plots, 200–1000 m2 in size. Larger plots allow studies of a larger fraction of the local species pool and tend to emphasise differen-ces between stands, whereas small plots tend to emphasise differences within stands (Dyne-sius 2001; Mills and McDonald 2004). More-over, results from small study plots may sug-gest that species are locally absent when their local abundance is low (Dynesius 2001).

-7-

Figure 1. Distribution of the fi eld sites in the four different studies; (a) sites for papers [I] and [I] and [I II] II] IIwhere each marker represents four stand types (south- and north-facing forests and clear-cuts), (b) experimental sites for paper [IV] and (c) sites for paper [IV] and (c) sites for paper [IV III] where circles and triangles represent III] where circles and triangles represent IIIconventionally and slash-harvested stands, respectively.

a)

b)

c)

Space-for-time substitution studiesSpace-for-time substitution studiesSWe conducted space-for-time substitution stu-dies to analyse the effects of slope aspect on understorey boreal forest bryophytes [I] and I] and Ivascular plants [II] and the effects of slope II] and the effects of slope IIaspect on their response to clear-cutting. We compared species composition and richness in a south- and a north-facing slope located in a forest stand adjacent to south- and north-facing slopes located in a clear-cuts in ten different sites. All stands/clear-cuts were, or had been dominated by spruce. All forty slo-pes were sedimentary without surface rocks or boulders and the four slopes at each site were paired with respect to slope inclination and slope length. In each of the four slopes, we randomly located a plot midpoint that was vertically matched (among the four slopes at each site) in relation to the valley bottoms and the ridge tops. Around the midpoints, a 200 m2 (10 m x 20 m) sample plot was positioned with the long side perpendicular to the slope contours. Each 200 m2 plot was divided into fi ve 40 m2 (4 m x 10 m) subplots in which we re-corded presence of all mosses, liverworts and vascular plants. In paper [I], mosses and li-I], mosses and li-Iverworts were classifi ed into fi ve ecological groups according to their substrate and habitat preferences based on Hallingbäck (1996) and personal experience from the region. In paper [II], we similarly classifi ed vascular plants into II], we similarly classifi ed vascular plants into IIeight ecological groups based on Ellenberg in-dicator values (Ellenberg et al. 1992). In paper [II], we calculated mean Ellenberg indicator II], we calculated mean Ellenberg indicator IIvalue for light, moisture, pH and nitrogen for each subplot. We also estimated the total co-vers (m2) of the following understorey (< 2 m height) growth forms: bryophytes, forbs, ferns, graminoids, dwarf shrubs and shrubs and sap-lings. The frequency of individual species was calculated as the number of subplots where a species was present (up to fi ve subplots per 200 m2 plot), whereas occupancy was the number of 200 m2 plots of each kind where a species was found (up to ten plots for each

of the four stand types). By using occupancy, we calculated plot richness and the number of species ”lost“ and ”gained“ in response to clear-cutting in each aspect. Lost species were those found only in the forest plot and gained species those found only in the clear-cut plot.

Using frequencies, we analysed composi-tional differences using ordination analysis. In paper [I], we used non-metric multi-dimensi-I], we used non-metric multi-dimensi-Ional scaling (NMS or NMDS), blocked multi- response permutation procedures (MRBP) and Monte Carlo permutation tests in PC-ORD (McCune and Mefford 1999) to analyse dif-ferences between the four kinds of slopes. We also used MRBP to analyse (i) the difference in compositional change (compositional tur-nover) between aspects in response to clear-cutting, and (ii) the difference in compositio-nal change between forests and clear-cuts in response to aspect. In these analyses we used the differences in frequencies between forests and clear-cuts or between aspects. In paper [II], we used principal compo-II], we used principal compo-IInents analysis (PCA), partial redundancy analysis (pRDA) and Monte Carlo permuta-tion tests available in CANOCO (ter Braak and Šmilauer 1997–2003). However, in paper [II], we also analysed the role of other vari-II], we also analysed the role of other vari-IIables apart from slope and aspect by applying pRDA and the forward selection option in CA-NOCO. Furthermore, in paper [II], we used II], we used IIpRDA site scores to calculate the Euclidian (Pythagorean) distance between the north-fa-cing and south-facing plot and between the forest and the clear-cut plot in aspect and site. These are measures of dissimilarity between plots and hence express compositional change in response to aspect and clear-cutting. In both papers [I] and [I] and [I II] correlations were II] correlations were IIused to identify associations between NMS or PCA site scores and environmental variables important for separating the four different slo-pes. Differences in richness, lost and gained species, Euclidian dissimilarity (paper [II] II] IIonly), growth forms and frequency of indivi-dual species between the four kinds of slopes

-8-

were compared using the Wilcoxon’s signed rank test. For analysis of differences in plot occupancy of individual species we employed the sign test.

The comparative study We conducted a comparative study to analyse the effects of commercial post-harvest extrac-tion of forest bioenergy (logging residues or slash) on composition and richness of vascu-lar plants and bryophytes. By using a database provided by Sveaskog (Sweden’s publicly ow-ned forest company) containing roughly 8000 clear-cut stands, we selected fourteen pairs of clear-cut stands, of which one in each pair had been subjected to post-harvest extraction of slash several months after clear-felling. Each pair was matched according to stand variab-les (such as time since logging, area, surface structure, and site index [dominant height (m) at age 100 years] and vegetation type). Within each stand we established a 20 m x 50 m (1000 m2) sample plot that was placed to obtain matching plots with high degree of pair-wise similarity with respect to substrate hete-rogeneity and covers of CWD (coarse woody debris), stones, boulders and moist ground, and small trees, respectively. We did not con-trol for covers of slash and mesic ground. Each sample plot was divided into fi ve 10 m x 20 m (200 m2) subplots. In each subplot we recorded presence of all mosses, liverworts and vascular plants and estimated covers (m2) of the following growth forms: dwarf shrubs, graminoids, herbs and ferns, bryophytes and shrubs and saplings. The frequency of each individual species was calculated as in papers [I] and [I] and [I II]. All mosses and liverworts were II]. All mosses and liverworts were IIsubdivided into ecological groups based on substrate (organic and non-organic) and habi-tat (forest and open) associations.

We use partial redundancy analysisWe use partial redundancy analysisW(pRDA) and Monte Carlo permutation test (CANOCO; ter Braak and Šmilauer 1997–2003) to analyse differences in species com-position between slash harvest plots (SHP)

and conventionally harvested plots (CHP). The pair-wise matching of plots was evalua-ted by analysing heterogeneity (Shannon in-dex of diversity) and composition (pRDA) of all substrates beside mesic ground and slash. By using an ANOVA and pRDA site scores we tested if there was any difference in response between species associated with different ha-bitats or substrates. Differences in richness were analysed at two scales (200 m2 and 1000 m2). Differences between treatments with re-spect to richness, substrate heterogeneity of matched substrates, covers of substrates and different growth forms and species frequen-cies were compared using the Wilcoxon’s sig-ned rank test.

The bryophyte transplant experimentIn paper [IV] we used a phytometer approach IV] we used a phytometer approach IVto evaluate the sheltering effect of logging re-sidues (spruce branches) and edge effects on vitality and growth of eight different forest bryophyte (six pleurocarpous mosses and two leafy liverworts) species in a controlled ma-nipulative experiment. The bryophytes were transplanted to four 0.7 m x 2.1 m experimen-tal plots in seven different clear-cuts. Two of these plots were positioned close to north-fa-cing forest edges (< 10 m from the edge) and two plots were located in the centre of seven clear-cuts (50 to 300 m from the edge). In each of the four plots, we planted three circu-lar transplants of each species. In each posi-tion, half of the transplants were covered by spruce branches and the other half was left ex-posed. Edges were at least 15 m high and 30 m wide. The experiment was initiated in May 29 and June 5, 2003 and terminated Septem-ber 15–20, after about 15 weeks or 105–114 days after transplantation. We measured light intensity (W/m2) and the internal maximum daytime temperature (°C) inside the transplants and soil water con-tent (%) in all four plots. The bryophytes were evaluated by measuring the vertical projection of each transplant at the start and at the end

-9-

of the experiment. By using peripheral projec-tions and circle-approximation, we calculated radial growth (mm) of each transplant. We also classifi ed the vitality of each transplant 23–29 and 105–114 days after transplanta-tion. As a measure of the net change in cover (growth minus dead tissue) of each species we measured the cover (cm2) of living green tis-sue at the start and the end of the experiment and calculated the difference. Radial growth of each species was analy-sed using two-way ANOVAs where cover (sheltered vs. exposed) and position (edge vs. centre) were considered fi xed factors. One-way ANOVAs and Tukey or Dunnet’s C tests were applied to locate differences among tre-atments in plot temperature, soil water con-tent, radial growth and net change in cover of each species. Reductions in radiation input and temperature caused by branch cover were compared between plots at the edge and in the clear-cut centre using Mann-Whitney U-test due to non-normality. Vitality was analysed using non-parametric methods as data con-stituted discrete classes. The Kruskal-Wal-lis H-test analysed differences in fi nal vita-lity of each species among treatments and the Wilcoxon’s signed rank test was used to ana-lyse reductions in vitality 4 and 15 weeks af-ter transplantation. The relationships between radiation, temperature and pooled growth of all species (mean of all transplants within a plot) and between radiation and pooled vita-lity were explored using Pearson and Spear-man rank correlations, respectively. Pearson correlations were also applied to investigate the effect of soil water content on the pooled growth in each of the four treatments. In ana-lyses of individual species we used plot avera-ges (from three transplants) as input data.

MAJOR RESULTS AND DISCUSSION

Aspect modifi es boreal plant responses to clear-cutting [I and I and I II]II]II

Papers [I] and [I] and [I II] provide evidence that slo-II] provide evidence that slo-II

pe aspect modifi es biotic responses to clear-cutting. In response to clear-cutting, covers of graminoids, shrubs/saplings and forbs showed a similar response between aspects and were higher in clear-cuts than forests. However, covers of dwarf shrubs, vascular cryptograms (ferns, horsetails, club mosses) and bryophytes differed in response. The former decreased in south-facing slopes while increasing in north-facing slopes.Vascular cryptogams and bryop-hytes showed a greater decline in south-facing slopes, although their covers declined in both aspects (paper [I]; paper [I]; paper [I II], Table 3).II], Table 3).II With respect to bryophytes, species compo-sition changed more in south-facing slopes as a consequence of both greater decline in abun-dance (frequency) of individual species (paper [I], Table 2) and greater loss of species (pa-I], Table 2) and greater loss of species (pa-Iper [I], Fig. 3). In south-facing slopes, clear-I], Fig. 3). In south-facing slopes, clear-Icutting reduced richness of liverworts, forest species and species associated with wood and bark whereas no ecological group declined in north-facing slopes. Although typical pioneer species increased in both aspects in response to clear-cutting, more species were generally gained in north-facing slopes, somewhat re-gained in north-facing slopes, somewhat re-gainedducing the difference in compositional change (paper [I], Fig. 3).I], Fig. 3).I

With respect to vascular plants the magni-tude of compositional change in response to clear-cutting was similar in both north- and south-facing slopes. However, the causes be-hind the turnover differed between aspects. Like bryophytes, more vascular plants were lost in south-facing slopes and richness, as well as occupancy and frequency of species associated with shady habitats declined. In contrast to south-facing slopes, neither rich-ness of any ecological group or occupancies and frequencies of individual species declined in north-facing slopes (paper [II], Fig 3; Table II], Fig 3; Table II4). Furthermore, the number of gained spe-cies of vascular plants tended to be greater in north-facing slopes and clear-cutting increased total richness and richness of all ecological groups in this slope aspect. In south-facing

-10-

slopes, clear-cutting only favoured vascular plants associated with sunny habitats and total richness did not differ in relation to south-fa-cing forests (paper [II], Fig 3).II], Fig 3).II The differences in the response of boreal plants to clear-cutting between south- and north-facing slopes are most likely directly or indirectly related to microclimatic differences between aspects. Trees effectively buffer radi-ation inputs (Chen et al. 1999) and this is ex-emplifi ed by the fact that south-facing slopes in clear-cuts but not in forests showed higher mean Ellenberg indicator values for light (pa-per [II], Table 3). The greater compositional II], Table 3). The greater compositional IIchange in bryophyte composition, the diffe-rences in growth form responses and the grea-ter loss of both vascular plants and bryophytes in south-facing slopes are most likely caused by clear-cutting accentuating microclimatic differences between aspects. Evidently, ex-cess radiation loads (and hence temperatures) in south-facing clear-cuts increased the mor-tality and decreased the abundance of drought -sensitive forest species (cf. Hannerz and Hå-nell 1997; Jalonen and Vanha-Majamaa 2001). In contrast, due to higher levels of shade and moisture, north-facing slopes retained more drought-sensitive species and their abundance declined less. However, differences in radiation input bet-ween aspects are also likely to have indirect effects on species responses. Clear-cutting of boreal forests tends to increase soil moisture by reducing evapotranspiration and because greater amounts of precipitation reaches the ground (McColl 1977; Zobel 1993). Ho-wever, if radiation input is too high the top-soil may desiccate (Edwards and Ross-Todd 1983). In our case, the Ellenberg indicator va-lue for moisture suggests that only the north-facing clear-cuts shifted towards more moist soil conditions (paper [II], Table 3). This may II], Table 3). This may IIbe because too much radiation, and the subse-quent desiccation of the topsoil prevented es-tablishment of species associated with moist habitats. Furthermore, clear-cutting of boreal

forests tends to increase nitrogen availabi-lity, as well as pH (Keenan and Kimmins 1993). While clear-cutting caused differences in mean indicator values for light between aspects, it removed differences in indicator values for nitrogen (paper [II], Table 3)II], Table 3)II . In addition, only north-facing clear-cuts showed higher indicator values for pH compared to their forest counterparts (paper [II], Table 3). II], Table 3). IIConsequently, the fl ora on north-facing slopes experienced a greater shift towards more eu-trophic conditions, suggesting that the edaphic changes in response to clear-cutting were lar-ger in this aspect.

South- and north-facing forests [I and I and I II]II]IISouth- and north-facing mature forests dif-fered in bryophyte and vascular plant species composition. This difference between aspects was probably a consequence of microclima-tic differences affecting habitat availability, as well as growth and soil conditions. South-facing forests had higher covers of broad-leaved trees, forbs, graminoids and vascu-lar cryptogams and were generally richer in vascular plants (paper [II], Fig. 2a, Tables 1 II], Fig. 2a, Tables 1 IIand 2). These effects of aspect are expected as small openings in the tree canopy in south-facing slopes allow more radiation to reach the forest fl oor and will also result in greater microclimatic variation than in north-facing slopes (Cantlon 1953). Increased radiation in-put vouches for greater population sizes and increases species richness provided that wa-ter is not limiting (Pausas and Austin 2001; Hawkins et al. 2003), and as the number of habitats increases more species can exist (Ma-cArthur 1972). South-facing forests also had higher mean Ellenberg indicator values for pH and nitrogen, suggesting that these sites were more productive (paper [II], Table 3). This is II], Table 3). This is IIlikely because pH and productivity are linked to decomposition rates that generally increase with high radiation input and high soil tempe-ratures (Nykvist and Rosén 1985; Kennan and Kimmins 1993; Fisher and Binkley 2000).

-11-

South-facing forests were also richer in mos-ses (paper [I], Fig. 2a), probably as a result of I], Fig. 2a), probably as a result of Igreater cover of broadleaved trees, woody de-bris, bare ground, and vascular plants in these plots. Woody debris (e.g. Kruys and Jonsson 1999), disturbed mineral soil (Jonsson and Esseen 1990) and through-fall from base-rich broadleaved trees (Weibull and Rydin 2005) have been shown to raise moss diversity in spruce forests. Moreover, rich litter produced by e.g. herbs and broadleaved trees and the generally more eutrophic conditions in south-facing forests probably stimulated mosses as-sociated with more rich conditions such as the moss R. roseum (paper [I], Table 1; Koponen I], Table 1; Koponen I1966). The north-facing forests generally had hig-her cover of bryophytes and some bryophytes associated with moist and shady forest habi-tats (e.g. the liverwort Lophozia obtusa; paper [I], Table 1) were more frequent in the more I], Table 1) were more frequent in the more Imoist north-facing slopes. Because liverworts and species associated with wood, bark and sheltered habitats are particularly dependent on moister microclimates (Marschall and Proctor 2004; Hylander 2005), we expected to fi nd higher richness of these groups in north-facing forests. However, we did not fi nd such a pattern (paper [I], Fig. 2a), probably be-I], Fig. 2a), probably be-Icause our studied forests were poor in convex substrates such as boulders and coarse woody debris (CWD). Such structures are vital for boreal forest liverworts and wood-inhabiting species (e.g. Hylander et al. 2005). Nonethe-less, only north-facing forests displayed a po-sitive relationship between CWD cover and richness of species associated with wood and bark. It is therefore likely that CWD-rich fo-rests would show larger differences between south- and north-facing slopes in the richness of these species groups.

Post-harvest extraction of logging residues[III]III]III

Paper [III] provides evidence that post-harIII] provides evidence that post-harIII -vest extraction of logging residues affects spe-

cies composition and richness in clear-cuts. Extraction of logging residues reduced bryop-hyte cover by half and raised graminoid cover by 10 % but had no signifi cant effect on other growth forms (paper [III], Table 3). III], Table 3). III Species composition of both mosses and liverworts differed signifi cantly between CHPs (con-ventionally harvested plots) and SHPs (slash harvested plots). Relative to CHPs, species richness of mosses and liverworts was redu-ced in SHPs (paper [III], Fig. 1). The liver-III], Fig. 1). The liver-IIIworts experienced the greatest reduction with approximately one-third of the species disap-pearing. Also, species richness of both forest and non-forest bryophytes associated with or-ganic substrates was reduced in SHPs (paper [III], Fig. 3). In contrast, slash harvesting did III], Fig. 3). In contrast, slash harvesting did IIInot affect the number of species associated with inorganic substrates (neither forest nor non-forest species) or species composition and richness of vascular plants (paper [III], III], IIIFigs. 1 and 3). There are at least three possible causes for the difference in bryophyte composition and richness. Firstly, loss of woody debris is li-kely to infl uence species able to colonise this substrate in clear-cuts. Fine woody debris (FWD) (<10 cm diameter) is important for wood-living organisms (Kruys and Jonsson 1999; Nordén et al. 2004) and there was con-siderable post-harvest colonisation on logging residues (M. Åström, personal observation). Consequently, it is not surprising that SHPs had fewer bryophytes associated with organic substrates in open habitats. Secondly, logging residues may also infl uence the microclimate (Mahendrappa and Kingston 1994; Proe et al. 1994, 2001). The signifi cant infl uence of the habitat association (forest vs. non-forest) on the liverwort response to slash harvest (paper [III], Fig. 2, Table 3), III], Fig. 2, Table 3), III and the decline of liver-worts and forest bryophyte richness in slash harvested plots (paper [III], Fig. 1, 3), III], Fig. 1, 3), III sug-gest that some of the changes in composition may be linked to a shelter effect (see paper [IV]). Thirdly, post-harvest IV]). Thirdly, post-harvest IV extraction of log-

-12-

ging residues requires additional machinery and enables more effective site preparation which may result in greater mechanical dis-turbance of the ground surface (Kardell 1992; Hakkila 2003). This may be detrimental to remnant populations (Jalonen andVanha-Ma-jamaa 2001) but may also stimulate establish-ment of new populations (Granström 1986). It is likely that that mechanical disturbance to disturbance to disturplants, soil and other substrates (cf. Rudolphi and Gustafsson 2005) contributed to the dif-ferences in bryophyte cover, composition and richness. With respect to vascular plants, this study suggests that they are not seriously affected, even though minor changes may occur. It has been suggested that slash harvesting affects vascular plants due to reduced nitrogen inputs (e.g. Olsson and Staaf 1995; Bråkenhielm and Liu 1998), fewer physical barriers that prevent regeneration (Fahey et al. 1991; Olsson and Staaf 1995) and reduced survival of remnant forest vascular plants (e.g. Bergquist et al. 1999). The previously reported effect of slash harvest on nitrofi lous species (e.g. Olsson and Staaf 1995; Bråkenhielm and Liu 1998) was not evident in our study, maybe because the branches and tree-tops were harvested after a period of needle dehiscing. This practice re-tains some of the needle-mass, reducing ne-gative impacts on the soil nitrogen budget (Egnell et al. 1998), and hence on nitrofi lous vascular plants. However, in our study area, N-deposition is relatively high (Lövblad et al. 1995), and may counteract nitrogen los-ses and subsequent effects on vascular plants in SHPs. Based on this study we cannot rule out the possibility of short- or long-term nu-trient effects on plants in regions with lower rates of N-deposition. The higher graminoid cover in SHPs is most probably caused by the lower cover of slash in these plots allowing graminoid expansion (Fahey et al. 1991; Ols-son and Staaf 1995), but it may also be due to mechanical disturbance, which facilitates establishment of pioneer graminoid species.

Our relatively large plots make it possible to evaluate many species. However, we cannot dismiss the possibility that commercial slash harvesting might have negative effects on vascular plants that were not included, or oc-curred at low frequencies, in this study.

The sheltering effect of forest edges and logging residues [IV]IV]IV

Paper [IV] shows that logging residuesIV] shows that logging residuesIV may shelter boreal bryophytes in clear-cuts by buf-fering the clear-cut microclimate. Cover by spruce branches only signifi cantly infl uenced growth of the liverwort P. asplenioides, which tended to grow better in covered plots irre-spective of the position in the clear-cut. Ho-wever, for all other species branch cover ten-ded to reduce growth at the edge but increase it in the clear-cut centre. Plot position in the clear-cut signifi cantly infl uenced growth of all species and growth was generally higher in edge plots than in plots in the clear-cut centre (paper [IV], Fig. 2, Table 1).IV], Fig. 2, Table 1).IV There were signifi cant reductions in vitality just four weeks after transplantation and by the end of the experiment, all individual species differed in vitality among the four treatments. Species located in plots covered by spruce branches at the edge appeared unaffected and all of them increased in cover. Although spe-cies in exposed plots at the edge and in plots covered by spruce branches in the clear-cut centre showed comparably lower vitalities, all species except P. asplenioides increased in cover. In contrast, all species located in ex-posed plots in the clear-cut centre were de-teriorated with low vitality. Here, cover was reduced for all species except P. schreberi and the responses typically differed from the oth-er treatments, particularly in relation to edge plots. The change in cover in the other three treatments was generally similar although the increase in cover tended to be lower in the co-vered plots in the centres of the clear-cuts (pa-per [IV], Figs. 3 and 4).IV], Figs. 3 and 4).IV Our results show that both position and co-

-13-

ver modifi ed the microclimate and this was confi rmed by measurements of radiation and temperature (paper [IV], Fig. 1). Radia-IV], Fig. 1). Radia-IVtion and temperature affect ambient moisture conditions (Chen et al. 1993), which in turn helps control bryophyte growth. Bryophytes are only metabolically active when wet and, therefore, dehydration negatively infl uences growth rates (Busby et al. 1978; Vitt 1989; Proctor 1990). Although litter input may affect bryophyte growth, due to nutrient additions (Frego and Carleton 1995), the main effect of spruce branch cover, in our case, was most li-kely related to light and moisture conditions. At the edge, the slightly lower growth in cove-red plots suggests that the cumulative effect of shading by adjacent trees and by spruce bran-ches produced suboptimal growth conditions (Sveinbjörnsson and Oechel 1992). At the clear-cut centre, however, shading by spruce branches reduced evaporative stress and pro-moted growth. Bryophyte vigour may be rapidly affected by decreased shelter (Busby et al. 1978; Hy-lander et al. 2002) and many forest species are sensitive to high radiation input, high tem-peratures, extensive droughts and fl uctuating microclimates (Clausen 1964; Proctor 1990; Marschall and Proctor 2004). For example, temperatures over 30°C may be lethal to bry-ophyte tissue (Furness and Grime 1982) and the fact that such temperatures were frequently recorded in the exposed plots helps explain the generally higher vitality in covered plots. The two liverworts were among the three species showing lowest vitality in the exposed plots in the clear-cut centre. This is not surprising sin-ce liverworts are generally less tolerant than mosses (Marshall and Proctor 2004). The change in cover of living shoots (inclu-ding both growth and shoot mortality) illustra-tes that the effect of cover by spruce branches is greater in the more extreme microclimate of the clear-cut centre rather than at the edge. The over-all negative change in cover of spe-cies located in plots not covered by spruce

branches suggests that species in a fully ex-posed habitat are unlikely to persist when the tree-canopy is completely removed. In cont-rast, the positive net response of species situa-ted in habitats sheltered by spruce branches or adjacent trees may survive and/or grow even though they may be initially damaged by an abrupt change in microclimate. It is likely that the potential for ground-dwelling bryophytes (particularly liverworts) to grow and survive through the clear-cut phase diminishes when logging residues are harvested. This is consistent with paper [III] III] IIIin which we found lower bryophyte cover and lower richness of liverworts and bryophytes associated with forests in clear-cuts where logging residues had been harvested. Even though factors such as the loss of woody sub-strate or mechanical damage contributed to this change, paper [IV]V]V suggests that the buf-fering capacity of logging residue on micro-climate may play a crucial role, particularly in fully exposed habitats.

SYNTHESIS

Topographical variation and logging residues infl uence the spatial distribution of species by modifying the degree of microclimatic change following clear-cutting. Although slo-pe aspect and logging residues most certainly differ in importance, they are both signifi cant aspects of heterogeneity affecting the survival of forest species after clear-cutting. Given that north-facing clear-cuts and conventionally harvested clear-cuts retain more forest spe-cies and because surviving populations can facilitate re-colonisation (Bengtsson 2002), these clear-cuts may approach pre-disturban-ce community composition more rapidly than south-facing clear-cuts or clear-cuts subjected to post-harvest slash extraction. It is likely that the interaction between slo-pe aspect and canopy disturbance has impor-tant ecological consequences at the landscape scale. Hilly landscapes are more species rich

-14-

than fl at landscapes, presumably because they have higher habitat heterogeneity. Assuming that fl at landscapes show a response that is intermediate to that of the south- and north-facing slopes, our results suggest that hilly landscapes may also be rich in species becau-se north-facing slopes serve as refugia after canopy disturbance.

South-facing forests may be more hetero-geneous because small openings in the cano-py creates larger microclimatic variation than in north-facing forests (Cantlon 1953, Fig. 2a). However, clear-cutting in south-facing slopes probably reduces habitat heterogen-eity by making most microhabitats equally sunny and dry. In contrast, clear-cutting may increase habitat heterogeneity in the shadier north-facing slopes by increasing radiation in some exposed habitats while ground micro-topography and structures, such as stumps, logs and shrubs maintain shady microhabitats by intercepting more radiation input than in south-facing slopes (Fig. 2b).

The effect of post-harvest extraction of log-ging residues at the landscape scale is likely to depend on the spatial extent of slash har-vest at stand- and landscape-scales, other fo-

restry practices, and the state of the forest in general. For example, the speed of re-co-lonisation in areas subjected to slash extrac-tion will partly depend on the status of the surrounding matrix of clear-cuts and forest, as well as the distance to the nearest viable source population. Furthermore, in contrast to CWD, very few species are specifi cally tied to FWD such as branches and tree-tops. Howe-ver, most species may grow on both FWD and CWD and logging residues are therefore like-ly to be comparatively more important in sites and landscapes poor in CWD (cf. Kruys and Jonsson 1999). In addition, the role of logging residues as shelter only appears to be impor-tant in the most exposed habitats (paper [IV]). IV]). IVIn other words, the management practices and the status of the managed forests are likely to dictate the importance of logging residues at the landscape scale.

IMPLICATIONS FOR MANAGEMENT

Previous studies and the results of this the-sis suggest that liverworts and species asso-ciated with woody debris in forests with high humidity are among the most negatively af-humidity are among the most negatively af-humidity are among the most negatively af

-15-

Figure 2. Schematic diagram of the infl uence of (a) small openings in the tree canopy in south- and north-facing forests (after Cantlon 1953) and (b) stumps and shrubs on the radiation input in south- and north-facing clear-cuts.

a)

b)Increasedheterogeneity?

Decreasedheterogeneity?

North-facing South-facing

fected by clear-cutting and residue extraction in boreal coniferous forests. Many of these species are already adversely affected by fo-rest management.

Selection of standsThere is an increasing body of evidence that suggests that some boreal forest habitats re-tain more species throughout the clear-cut phase of succession. Dynesius et al. (2001) found that bryophyte species richness was proportionally less reduced by clear-cutting in streamside forests compared to upland ones. They also suggested that streamside forests may be more effective as refugia and/or si-tes for establishment after clear-cutting. The-se results coincide with the results of paper [I]; although streamside forests are different I]; although streamside forests are different Ifrom north-facing slopes, they are both relati-vely moist despite disturbance of the canopy. When considering the buffering capacity of branches and tree tops in clear-cuts, this ef-fect is most likely larger in the more exposed south-facing slopes (cf. [IV]). This may sug-IV]). This may sug-IVgest that it is better to extract logging residues in north-facing slopes. However, naturally sheltered sites that provide moist microha-bitats and enhance survival and viability of forest species are probably essential for the maintenance of species richness at the lands-cape scale and for re-colonisation of sites where species do not persist after clear-cut-ting. This is likely to be more important than the initial shelter-effect of logging residues in exposed habitats. Furthermore, higher humi-dity and microclimatic continuity increases the ability to inhabit dead wood and the sub-sequent colonisation of bryophytes associated with dead wood (cf. [I]). Therefore, extrac-I]). Therefore, extrac-Ition of logging residues that causes loss of substrate and increased disturbance to rem-nant populations should be avoided in natu-rally moist sites (e.g. north-facing slopes) as this activity may inhibit their function.

Retention of trees Given that the sheltering effect of tree bran-ches is not so pronounced in less extreme mi-croclimates (paper [IV]), it is likely that the negative effects of logging residue extraction will be mitigated by creation of shadier and less exposed clear-felled areas. This could be achieved by retaining trees during clear-cut-ting operations or by avoiding cleaning of un-dergrowth before or after clear-cutting.

In response to clear-cutting the understoreywas more negatively affected in south-facing slopes (paper [I])I])I . A logical management im-plication of these fi ndings is that cutting prac-tices on south-facing slopes should be more restricted and should never remove the tree cover, entirely. However, some intolerant spe-cies (liverworts) also declined in north-facing slopes. If aiming to conserve drought-sensi-tive species negatively affected by forestry, I suggest that trees should be retained on north-facing slopes. Such species are less likely to survive on south-facing slopes irrespective of tree retention. To my knowledge, no study has analyzed edge-effects in relation to aspect but I believe that edge-effects induced by light (but not wind) are smaller in shadier north-fa-cing slopes. This would make groups of trees on north-facing slopes more effective in pre-serving the forest fl ora than on south-facing slopes.

SubstrateContinuity of dead wood in different stages of decay is crucial for many wood-inhabiting organisms (e.g. Samuelsson et al. 1994) and FWD may be an important substrate affecting plant richness in both forests (Kruys and Jons-son 1999) and clear-cuts (paper [III]III]III ). Howe-ver, in contrast to CWD, few species are spe-cifi cally dependent on FWD although most species will grow on both FWD and CWD, given suitable conditions. In addition, coar-ser substrates are not as easily overgrown by

-16-

ground vegetation as fi ner substrates and thus offer a wider temporal window for establish-ment (de Jong et al. 2004). It is therefore li-kely that protection and creation of CWD may mitigate the loss of fi ner substrates. In landscapes subjected to extraction of logging residues this would be particularly important. Furthermore, for wood-inhabiting bryophy-tes, this thesis suggests that CWD should not be created or placed arbitrarily, but rather be located in shadier (and hence moister) habi-tats, such as north-facing forests or clear-cuts (paper [I]). Although only the former exhibi-I]). Although only the former exhibi-Ited a positive relationship between richness of wood-inhabiting bryophytes and cover of CWD, logs located in north-facing clear-cuts are most likely easier to colonize in later stages of succession. Given that north-facing slopes also retain more species following clear-cut-ting, the dispersal distances will be shorter.

Mechanical disturbance The method used and the timing of post-har-vest residue extraction are potentially very important factors infl uencing the ecological effects of forest bioenergy extraction. As an additional mechanised step is introduced, the risk of greater disturbance to the ground surfa-ce and vegetation may increase. Such change will favour pioneer species but will negatively affect remnant plant populations as indicated in paper [III]. In paper [III]. In paper [III III], the logging resi-III], the logging resi-IIIdues were generally extracted several months after clear-cutting (in July-August). This prac-tice allows some dehiscing of needles which may mitigate negative effects on soil nutrient status (Anonymous 2001). However, because ground-water levels tend to rise after clear-cut-ting (Zobel 1993), delayed residue extraction may increase risks of poor bearing capacity of the ground when the residues are forwarded. In practice, this may cause increased distur-bance in areas with poor bearing capacity but it may also increase the spatial extent of the disturbance as machines are re-routed to av-oid e.g. moist ground.

-17-

To my knowledge no study has compared the impact of delayed extraction with the impact of extraction immediately after clear-cutting with respect to damages to plants and soil. Ho-wever, because of above mentioned reasons it is likely that extraction of un-dehisced logging residues immediately after clear-felling reduce mechanical disturbance. This is defi nitely the case when the ground is frozen and covered by a protective layer of snow when the trees are harvested, but not when the residues are extracted in late summer or autumn. Healthy vegetation in clear-cuts has been shown to be very important to reduce leaching of nutrients (e.g. Boring et al. 1981; Staaf and Olsson 1994; Palviainen 2005). Therefore, delayed residue extraction may be of little use if it re-sults in increased damage to the soil surface and the ground vegetation. In addition, it has been suggested that leaving the residues for a few months to dehisce is a relatively ineffec-tive method for retaining needles and, conse-quently nutrients.

PERSONAL REFLECTIONS

Drought-sensitive and wood-inhabiting plants will most likely become less abundant, due to post-harvest extraction of logging residues following clear-cutting of coniferous forests. However, if appropriate environmental consi-deration is taken in all forestry measures (e.g. creation/protection of CWD and protection of important habitats), and if extraction is li-mited to areas with few technical constraints (e.g. areas with a smooth surface structure), within “production stands” (which is most-ly the case today), I do not believe that this practice, poses a substantial threat to plant diversity in the boreal forest landscape. Fine woody debris is constantly produced in all successional stages through natural proces-ses, and forest management, for example in cleaning and thinning. This ensures a steady supply of substrate (cf. de Jong et al. 2004). Furthermore, no rare or red-listed species

-18-

(Gärdenfors 2005) were encountered in any of the clear-cuts studied. Such species are typi-cally confi ned to old forest stands characteri-zed by high heterogeneity, a humid micro-cli-mate, and a large volume of broadleaved trees and/or CWD (Gustafsson et al. 2004). It is un-likely that their future is jeopardized by post-harvest slash extraction in production stands. However, if precautionary measures are not applied and the boreal forest landscape is ”va-cuumed” for logging residues (and CWD) for bioenergy, the landscape will most likely be-come less rich in wood-living and drought-in-tolerant species. To conclude, extraction of forest bioenergy is a tool that can be used to combat anthropo-genic global warming by replacing fossil fu-els. Although this practice presents ecological drawbacks it may foster long-term ecological benefi ts by retarding global warming and as-sociated effects, such as habitat alteration and species loss.

ACKNOWLEDGEMENTS

I would like to thank Christer Nilsson, Mats Dynesius, Birgitta Malm Renöfält and Roland Jansson for comments on earlier versions of this thesis. Thanks also to Jane Catford for linguistic improvements. I would also like to thank Sveaskog, Holmen Skog, SCA Skog, The Church of Sweden, The Swedish Forest Agency and several private landowners for providing sites and data. The research presen-ted in this thesis was supported by grants from the Swedish Energy Agency (to C. Nilsson)and the Björkman Fund (and the Björkman Fund (and the B to M. Åström).

REFERENCES

Ahti, T., Hämet-Ahti, L. & Jalas, J. 1968.Vegetation zones and their sections in north western Europe. Annales Botanici Fennici 5: 169-211.

Anonymous 2001. Rekommendationer vid uttag av skogsbränsle och kompensations-

gödsling. Meddelande 2-2001. Skogssty-relsen, Skogsstyrelsens förlag, Jönköping, Sweden.

Anonymous 2005. Hur stor andel är skydd-ad produktiv skog? Naturvårdsverket, http://www.naturvardsverket.se, 20 Okto-ber 2006.

Anonymous 2006. Statistical Yearbook of Forestry. Offi cial Statistics of Sweden. Swe-dish Forest Agency, Jönköping, Sweden.

Auerbach, M. & Schmida, A. 1987. Spatial scale and the determinants of plant species richness. Trends in Ecology and Evolution 2: 238-242.

Axelsson, A. L. 2001. Forest Landscape Change in Boreal Sweden 1950-2000. Ph.D. thesis, Department of Forest Vegeta-tion Ecology, Swedish University of Agri-cultural Sciences, Umeå, Sweden.

Bengtsson, J., Persson, T. & Lundkvist, H. 1997. Long-term effects of logging residue addition and removal on macroarthropods and enchytraeids. Journal of Applied Eco-logy 34: 1014-1022.

Bengtsson, J., Angelstam, P., Elmqvist, T., Emanuelsson, U., Folke, C., Ihse, M. & Ny-ström, M. 2003. Reserves, resilience and dynamic landscapes. Ambio 32: 389-396.

Bengtsson, J. 2002. Disturbance and resi-lience in soil animal communities. Europe-an Journal of Soil Biology 38:119-125.

Bennie, J., Hill, M.O., Baxter, R. & Huntley, B. 2006. Infl uence of slope and aspect on long-term vegetation change in British chalk grasslands. Journal of Ecology 94: 355-368.

Bergquist, J., Örlander, G. & Nilsson, U. 1999. Deer browsing and slash removal affect fi eld vegetation on south Swedish clear-cuts. Forest Ecology and Management 115: 171-182.

Bernes, C. (Ed.) 1994. Biological diversity inSweden: a country study. Monitor 13. Swe-dish Environmental Protection Agency, Solna, Sweden.

Boring, L.R., Monk, C.D. & Swank, W.T.

-19-

1981. Early regeneration of a clear-cut southern Appalachian forest. Ecology 62: 1244-1253.

Bråkenhielm, S. & Liu, Q. 1998. Long-term effects of clear-felling on vegetation dyna-mics and species diversity in a boreal pine forest. Biodiversity and Conservation 7: 207-220.

Brotons, L., Monkkonen, M. & Martin, J.L. 2003. Are fragments islands? Landscape context and density-area relationships in boreal forest birds. American Naturalist 162: 343-357.

Busby, J.R., Bliss, L.C. & Hamilton, C.D. 1978. Microclimate control of growth ra-tes and habitats of boreal forest mosses, Tomenthypnum nitens and Hylocomium splendens. Ecological Monographs 48:95-110.

Canton, J.E. 1953. Vegetation and microcli-mates on north and south slopes of Cus-hetunk mountain, New Jersey. Ecological Monographs 23: 241-270.

Chen, J.Q., Franklin, J.F. & Spies, T.A. 1993. Contrasting microclimates among clearcut edge, and interior of old-growth Douglas-fi r forest. Agricultural and Forest Meteoro-logy 63: 219-237

Chen, J.Q., Franklin, J.F. & Spies, T.A. 1995. Growing-season microclimatic gradients from clear-cut edges into old-growth Dou-glas-Fir forests. Ecological Applications 5: 74-86.

Chen, J.Q., Saunders, S.C., Crow, T.R., Nai-man, R.J., Brosofske, K.D., Mroz, G.D., Brookshire, B.L. & Franklin J.F. 1999. Mi-croclimate in forest ecosystem and lands-cape ecology. BioScience 49:288-297.

Clausen, E. 1964. The tolerance of hepatics to desiccation and temperature. Bryologist 67: 411-417.

Crawford, R.M.M. 1989. Studies in plantsurvival: ecological case histories of plant adaptation to adversity. Studies in ecology. Volume 11. Blackwell Scientifi c, Oxford, UK.

de Jong, J., Dahlberg, A. & Stokland, J.N.2004. Vilken kvalitet och kvantitet av död ved krävs för att bevara biologisk mång-fald i skogen? I: de Jong, J. & Almstedt, M. (red.) 2004. Död ved i levande skogar - Vilket mål bör vi sträva efter och hur når vi det? Centrum för biologisk mångfald, Upp-sala universitet och Sveriges lantbruksuni-versitet, Uppsala.

Drever, R.C. & Lertzman, K.P. 2003. Effects of a wide gradient of retained tree structure on understorey light in coastal Douglas-fi r forests. Canadian Journal of Forest Re-search 33: 137-146.

Dynesius, M. 2001. Spatial and evolutionary aspects of species diversity, species traits, and human impacts with examples from bo-real riparian and forest plant communities. Ph.D. thesis, Department of Ecology and Environmental Sciences, Umeå University, Umeå, Sweden.

Dyrness, C.T. 1973. Early stages of plant succession following logging and burning in the western cascades of Oregon. Ecology 54: 57-69.

Ebeling, F. 1959. Skogarna och deras vård i övre Norrland från och med 1930-talet. pp. 413-443 in Arpi, G. (Ed.). Sveriges skogar under 100 år II. Kungliga domänstyrelsen, Stockholm, Sweden.

Edwards, N.T. & Ross-Todd, B.M. 1983. Soil carbon dynamics in a mixed deciduous forest following clear-cutting with and wit-hout residue removal. Soil Science Society of America Journal 47: 1014-1021.

Egnell, G. & Leijon, B. 1999. Survival and growth of planted seedlings of Pinus syl-vestris and Picea abies after different levels of biomass removal in clear-felling. Scan-dinavian Journal of Forest Research 14: 303-311.

Egnell, G., Nohrstedt, H.-Ö., Weslien, J., Westling, O. & Örlander, G. (1998) Mil-jökonsekvensbeskrivning av skogsbräns-leuttag, asktillförsel och övrig närings-kompensation. Skogsstyrelsens Förlag,

-20-

Jönköping, Sweden. Egnell, G. & Valinger, E. 2002. Survival,

growth, and growth allocation of planted Scots pine trees after different levels of bio-mass removal in clear-felling. Forest Eco-logy and Management 177: 65-74.

Ellenberg, H., Weber, H.E., Düll, R., Wirth, V., Werner, W. & Paulisson, D. 1992. Zei-gerwerte von Pfl anzen in Mitteleuropa. Scripta Geobotanica. 18.

Engelmark, O., Sjöberg, K., Andersson, B., Rosvall, O., Ågren, G.I., Baker, W.L., Barklund, P., Björkman, C., Despain, D.G., Elfving, B., Ennos, R.A., Karlman, M., Knecht, M.F., Knight, D.H., Ledgard, N.J., Lindelöw, Å., Nilsson, C., Peterken, G.F., Sörlin, S. & Sykes. M. T. 2001. Ecological effects and management aspects of an exo-tic tree species: the case of lodgepole pine in Sweden. Forest Ecology and Manage-ment 141: 3-13.

Engelmark, O. 1999. Boreal forest disturban-ces. In Walker, L.R. (Ed.). Ecosystems on disturbed ground. Elsevier. Amsterdam. pp. 161-186.

Esseen, P.-A., Ehnström, B., Ericson, L. & Sjöberg, K. 1997. Boreal forests. Ecologi-cal Bulletins 46: 16-47.

Fahey, T.J., Hill, M.O., Stevens, P.A., Hor-nung, M. & Rowland, P. 1991. Nutrient accumulation in vegetation following con-ventional and whole-tree harvest of Sitka spruce plantations in north Wales. Forestry 64: 271-288.

Fanelli, G., Tescarolla, P. & Testi, A. 2006. Ecological indicators applied to urban and suburban fl oras. Ecological Indicators 6: 444-457.

Fenton, N.J. & Frego, K.A. 2005. Bryophyte (moss and liverwort) conservation under remnant canopy in managed forests. Biolo-gical Conservation 122: 417-430.

Fisher, R.F. & Binkley, D. 2000. Ecology and management of forest soils. Wiley. New York.

Frank, D.A., McNaugthon, S.J. 1991. Stabi-

lity increases with diversity in plant com-munities: empirical evidence from 1998 Yellowstone drought. Oikos 62: 360-362.

Fredén, C. (ed.). 1994. The National Atlas ofSweden. Geology – the Bedrock, Soils, Landforms, Groundwater. Geological Sur-vey of Sweden, Uppsala, Sweden.

Frego, K.A., & Carleton, T.J. 1995. Micro-site tolerance of four bryophytes in a ma-ture Black Spruce stand, reciprocal trans-plants. Bryologist 98: 452-458.

Furness, S.B. & Grime, J.P. 1982. Growth ra-tes and temperature responses in bryophy-tes, II. A comparative study of species of contrasted ecology. Journal of Ecology 70: 525-536.

Gärdenfors, U. 2005. (Ed.) Rödlistade arter i Sverige 2005. [The 2005 redlist of Swe-dish species]. Artdatabanken, Uppsala, Sweden.

Gentili, F., Nilsson, M.C. Zackrisson, O. DeLuca, T.H., & Sellstedt, A. 2005. Phy-siological and molecular diversity of fea-thermoss associative N2-fi xing cyanobac-teria. Journal of Experimental Botany 56: 3121-3127.

Granström, A. 1986. Seed banks in forest soils and their role in vegetation succes-sion after disturbance. PhD thesis, Swedish University of Agricultural Sciences, Umeå, Sweden.

Gray, A.N. & Spies, T.A. 1997. Microsite controls on tree seedling establishment in conifer forest canopy gaps. Ecology 78: 2458-2473.

Gunnarsson, B., Nittérus, K. & Wirdenäs, P. 2004. Effects of logging residue remo-val on ground-active beetles in temperate forests. Forest Ecology and Management 201: 229-239.

Gururaja, J. 2003. Energy for sustainable development: review of national and inter-national energy policies. Natural Resources Forum 27: 53-67.

Gustafsson, L., Hylander, K. & Jacobsson, C. 2004. Uncommon bryophytes in Swe-

-21-

dish forests: key habitats and production forests compared. Forest Ecology and Ma-nagement 194: 11-22.

Gustavsson, L., Börjesson, P., Johansson, B.& Svenningsson, P. 1995. Reducing CO2 emissions by substituting biomass for fossil fuels. Energy 20: 1097-1113.

Hakkila, P. 2003. Developing technology for large-scale production of forest chips. Tech-nology Programme Report 5/2003, Natio-nal Technology Agency, Helsinki, Finland.

Hallingbäck, T. 1996. Ekologisk katalog över mossor. [The bryophytes of Sweden and their ecology]. Swedish Species Infor-mation Centre, Uppsala, Sweden.

Halpern, C.B. 1988. Early successional path-ways and the resistance and resilience of fo-rest communities. Ecology 69: 1703-1715.

Halpern, C.B. & Spies, T.A. 1995. Plant spe-cies diversity in natural and managed fo-rests of the Pacifi c Northwest. Ecological Applications 5: 913-934.

Hannerz, M. & Hånell, B. 1997. Effects on the fl ora in Norway spruce forests following clearcutting and shelterwood cutting. Forest Ecology and Management 90: 29-49.

Hanssen, K.H. 2003. Natural regeneration of Picea abies on small clear-cuts in SE Nor-way. Forest Ecology and Management 180: 199-213.

Hawkins, B.A., Field, R., Cornell, H.V., Cur-rie, D.J., Guegan, J.-F., Kaufman, D.M., Kerr, J.T., Mittelbach, G.G., Oberdorff, T., O’Brien, E.M., Porter, E.E. & Turner, J.R.G. 2003. Energy, water, and broad-sca-le geographic patterns of species richness. Ecology 84: 3105–3117.

Heithecker, T.D. & Halpern, C.B. 2006. Va-riations in microclimate associated with dispersed-retention harvests in coniferous forests of western Washington. Forest Eco-logy and Management 226: 60-71.

Hylander, K. 2005. Aspect modifi es the mag-nitude of edge effects on bryophyte growth in boreal forests. Journal of Applied Eco-logy 42: 518-525.

Hylander, K., Dynesius, M., Jonsson, B.G. &C. Nilsson. 2005. Substrate form determi-nes the fate of bryophytes in riparian buffer strips. Ecological Applications 15: 674-688.

Hylander, K., Jonsson, B.G. & Nilsson, C. 2002. Evaluating buffer strips along bore-al streams using bryophytes as indicators. Ecological Applications 12: 796-806.

Jalonen, J. & Vanha-Majamaa, I. 2001. Im-mediate effects of four different felling met-hods on mature boreal spruce forest under-storey vegetation in southern Finland. Fo-rest Ecology and Management 146:25-34.

Jonsson, B.G. & Esseen, P.-A. 1990. Treefall disturbance maintains high bryophyte di-versity in boreal spruce forests. Journal of Ecology 78: 924-936.

Kardell, L. 1992. Vegetationsförändring, plantetablering samt bärproduktion efter stubb- och ristäkt. Report No. 50, Swedish University of Agricultural Sciences, Upp-sala, Sweden.

Keenan, R.J. & Kimmins, J.P. 1993. The ecological effects of clear-cutting. Environ-mental Reviews 1: 121-144.

Kruys, N. & Jonsson, B.G. 1999. Fine woo-dy debris is important for species richness on logs in managed boreal spruce forests in northern Sweden. Canadian Journal of Fo-rest Research 29: 1295-1299.

Koponen, T. 1966. Distribution of Rhodo-bryum roseum in Finland in the light of line survey and frequency maps. Annales Bota-nici Fennici 3: 399-405.

Lövblad, G., Kindblom, K., Grennfelt, P., Hultberg, H. & Westling, O. 1995. Depo-sition of acidifying substances in Sweden.Ecological Bulletins 44: 17-34.

Lundborg, A. 1998. A sustainable forest fuel system in Sweden. Biomass and Bioenergy 15: 399-406.

MacArthur, R.H. 1972. Geographical Ecology.Harper and Row, New York.

MacHattie, L. B. & McCormack, R. J. 1961. Forest microclimate: a topographic study in Ontario. Journal of Ecology 49: 301-323.

-22-

Mahendrappa, M.K. & Kingston, D.G.O. 1994. Intensive harvesting impacts on soil temperature and solution chemistry in the maritimes region of Canada. New Zealand Journal of Forestry Science 24: 402-414.

Marschall, M. &. Proctor, M.C.F. 2004. Are bryophytes shade plants? Photosyntheticlight responses and proportions of chlorop-hyll a, chlorophyll b and total carotenoids. Annals of Botany 94: 593-603.

McColl, J.G. 1977. Retention of soil water following forest cutting. Soil Science So-ciety of America Journal. 41: 984-988.

McCune, B. & Mefford, M. J. 1999. PC-ORD. Multivariate analysis of ecological data. MjM Software Design, Gleneden Be-ach, Oregon, USA.

McInnis, B.G. & Roberts, M.R. 1994. The effects of full-tree and tree-length harvest on natural regeneration. Northern Journal of Applied Forestry 11: 131-137.

Mills, S.E. & McDonald S.E. 2004. Predic-tors of moss and liverwort species diversity of microsites in conifer-dominated boreal forests. Journal of Vegetation Science 15: 189-198.

Minore, D. 1986. Germination, survival and early growth of conifer seedlings in two habitat types. Pacifi c Northwest Research Station, Portland, Oregon, USDA Forest Service Research Paper PNW-348.

Nittérus, K. 2006. Biodiversity & Biofuel.Harvest of logging residue affects forest Arthropods. Ph.D. thesis, Department of Plant and Environmental Sciences. Göte-borg University.

Nordén, B., Rydberg, M., Götmark, F. &Olausson, B. 2004. Relative importance of coarse and fi ne woody debris for the diver-sity of wood-inhabiting fungi in temperate broadleaf forests. Biological Conservation 117:1-10.

Nykvist, N. 1997. Changes in species occur-rence and phytomass after clear-felling, prescribed burning and slash removal in two Swedish spruce forests. Studia Forest-

alia Suecica 201: 1-33.Nykvist, N. & Rosén, K. 1985. Effects of

clearfelling and slash removal on the acidity of northern coniferous forests soils. Forest Ecology and Management 11: 157-169.

Økland, T., Rydgren, K., Halvorsen Økland, R., Storaunet, K.O. & Rolstad, J. 2003. Variation in environmental conditions, un-derstorey species number, abundance and composition among natural and managed Picea abies forest stands. Forest Ecology and Management 177: 17-37.

Oliver, C.D. & Larsson, B.C. 1990. Forest stand dynamics. McGraw-Hill, New York, USA.

Olivero, A.M. & Hix, D.M. 1998. Infl uence of aspect and stand age on ground fl ora of southeastern Ohio forest ecosystems. Plant Ecology 139: 177-187.

Olsson, B.A. & Staaf, H. 1995. Infl uence of harvesting intensity of logging residues on ground vegetation in coniferous forests. Journal of Applied Ecology 32: 640-654.

Östlund, L. 1993. Exploitation and structural changes in the north Swedish boreal forest 1800-1992. Ph.D. thesis, Department of Forest Vegetation Ecology, Swedish Uni-versity of Agricultural Sciences, Umeå, Sweden.

Palviainen, M., Finér, L., Mannerkoski, H., Piirainen, S. & Starr, M. 2005. Responses of ground vegetation species to clear-cut-ting in a boreal forest: above ground bio-mass and nutrient contents during the fi rst 7 years. Ecological Research. 20: 652-660.

Pausas, J.G. & Austin, M.P. 2001. Patterns of plant species richness in relation to diffe-rent environments: an appraisal. Journal of Vegetation Science 12: 153-166.

Pignatti, S., Bianco, P.M., Fanelli, G., Gua-rino, R. Petersen, J. & Tescarollo, P. 2001. Reliability and effectiveness of Ellenberg’s indices in checking fl ora and vegetation changes induced by climatic variations. In: Walter, J.R., Burga, C.A. & Edwards, P.J. (Eds.). Fingerprints of Climate Changes:

-23-

Adapted Behaviour and Shifting Species Ranges. Kluwer Accademy/Plenum Publis-hers. Pp. 281-304.

Pimm, S. L. 1991. The balance of nature? Ecological issues in the conservation of species and communities. Chicago Univer-sity Press, Illinois, USA.

Prevost, M. & Pothier, D. 2003. Partial cuts in a trembling aspen-conifer stand: effects on microenvironmental conditions and re-generation dynamics. Canadian Journal of Forest Research 33: 1-15.

Proctor, M.C.F. 1990. The physiological ba-sis of bryophyte production. Botanical Jour-nal of the Linnean Society 104: 61- 77.

Proe, M.F., Dutch, J. & Griffi ths, J. 1994. Harvest residue effect on microclimate, nu-trition, and early growth of sitka spruce (Pi-cea sitchensis) seedlings on a restock site. New Zealand Journal of Forestry Science 24: 390-401.

Proe, M.F., Griffi ths, J. & McKay, H.M. 2001. Effect of whole-tree harvesting on microclimate during establishment of se-cond rotation forestry. Agricultural and Fo-rest Meteorology 110: 141-154.

Raab, B. & H. Vedin. (Eds.). 1995. National Atlas of Sweden: climate, lakes and rivers. SNA Publisher, Stockholm, Sweden.

Rudolphi, J. & Gustafsson, L. 2005. Effects of forest-fuel harvesting on the amount of deadwood on clear-cuts. Scandinavian Journal of Forest Research 20: 235-242.

Rydgren, K. &Hestmark, G. 1997. The soil propagule bank in a boreal old-growth spru-ce forest: changes with depth and relation-ship to aboveground vegetation. Canadian Journal of Botany 75: 121-128.

Samuelsson, J., Gustafsson, L. & Ingelög, T. 1994. Dying and Dead Trees: A Review of Their Importance for Biodiversity. Swedish Threatened Species Unit, Uppsala, Swe-den.

Schofi eld, W.B. 1985. Introduction to bryolo-gy. (1st ed.). New Jersey: Blackburn Press.

Shelton, M.G., & Cain, M.D. 2000. Regene-

rating uneven-aged stands of loblolly and shortleaf pines: the current state of know-ledge. Forest Ecology and Management 129: 177-193.

Sjörs, H. 1965. Forest regions. In: The plant cover of Sweden. Acta Phytogeographica Suecica 50. pp. 48-63.

Sjörs, H. 1999. Swedish Plant Geography (Eds. H. Rydin, P. Snoeijs & M. Diekmann). The background: geology, climate and zona-tion. Acta Phytogeographica Suecica 84. pp. 5-14.

Solga, A. & Frahm, J.P. 2006. Nitrogen accu- mulation by six pleurocarpous moss species and their suitability for monitoring nitrogen deposition. Journal of Bryology 28: 46-52.

Staaf, H. & Olsson, B.A. 1994. Effects of slash removal and stump harvesting on soil water chemistry in a clear-cutting in SW Sweden. Scandinavian Journal of Forest Research 9: 305-310.

Sveinbjörnsson, B. & Oechel, W.C. 1992. Controls on growth and productivity of bry-ophytes, environmental limitations under current and anticipated conditions. In: Ba-tes, J.W. & Farmer, A.M. (Eds.) Bryophy-tes and lichens in a changing environment. (pp. 77-102). Oxford University press.

Swift, M.J., Heal, O.W. & Anderson, J.M. 1979. Decomposition in terrestrial ecosys-tems. 1st ed., University of Chicago Press, Berkeley, US.

ter Braak, C.J.F. & Šmilauer, P. 1997-2003. Canoco for Windows version 4.51. Biome-tris, Plant Research International, Wage-ningen, The Netherlands.

Tilman, D. & Downing, J.A. 1994. Biodiver-sity and stability in grasslands. Nature 367: 363-365.

Uotila, A. & Kouki, J. 2005. Understorey vegetation in spruce-dominated forests in eastern Finland and Russian Karelia: Suc-cessional patterns after anthropogenic and natural disturbances. Forest Ecology and Management 215: 113-137.

Vitt, D.H. 1989. Patterns of growth of the

-24-

drought tolerant moss, Racomitrium micro-carpon, over a three year period. Lindbergia 15: 181-187.

Wastenson, L., & Nilsson, N.-E. (Eds.)1995.National Atlas of Sweden: the forests. Na-tional Board of Forestry, Jönköping, Swe-den.

Weibull,H. & Rydin, H. 2005. Bryophyte sp- ecies richness on boulders: relationship to area, habitat diversity and canopy tree spe-cies. Biological Conservation 122: 71-79.

Wimberly, M.C. & Spies, T.A. 2001. Infl uen-ces of environment and disturbance on fo-rest patterns in coastal Oregon watersheds. Ecology 82: 1443-1459.

Zackrisson, O. 1977. Infl uence of forest fi reson the north Swedish boreal forest. Oikos 29: 22-32.

Zobel, M. 1993. Changes in pine forest com-munities after clear cutting: a comparison of two edaphic gradients. Annales Botanici Fennici 30: 131-137.

-25-

TACK

Efter fem år som doktorand vid EMG är det många som jag vill tacka, både inom och utanför universitetets väggar. Särskilt tack till:

Christer Nilsson, min ”strategiska handledare” som har varit en fantastisk hjälp med försöks-planering, författande och en massa annat. Du har alltid varit tillgänglig för frågor och diskus-sioner och du höjer det mesta till en högre nivå. Du har varit en stor inspirationskälla men du har också gett mig fria händer att komma med egna idéer och att både lyckas och misslyckas.

Mats Dynesius, min ”skogshandledare” som har hjälpt mig med planering, fältarbete och så mycket annat. Jag vet ingen som är så skogligt allmänbildad som du och som med ständig iver tar tag i och diskuterar nya projekt och idéer. Du har också varit en klippa när det gäller förfat-tandet av artiklarna. Du har fi ngranskat de vetenskapliga resonemangen och hittat svagheterna i mina påståenden hur jag än försökt dölja dem.

Kristoffer Hylander, min ”mosshandledare” som lärt mig i stort sett allt jag kan om mossor(men inte allt han kan). Tack för att du hjälp mig med alla pyttesmå och skumma mosskollekter och tack för alla dina bidrag till artiklarna

Alla andra i forskargruppen som har bidragit med goda idéer, skämt, skratt, hjälp, trams, allvar, sällskap, samtal, terapi, fi ka, kaffe, uppmuntran, kunskap, information, diskussion och strunt. Utan er hade min tid varit otroligt mycket tråkigare. Ett doktorandprojekt är ett lagarbete och ni är ett fantastiskt lag. Ett stort tack till dig Roland för att du tålmodigt har svarat på alla frågor Roland för att du tålmodigt har svarat på alla frågor Rolandkring statistik och annat. Tack också Erik och Erik och Erik Niclas för att allt inte växer och gror.

Sandra, Lotta, Patrik, Kicki och Martina som med glada miner har grävt, inventerat och plan-terat oavsett väder och vind. Tack Patrik för den oförglömliga första fältsäsongen som vi till-Patrik för den oförglömliga första fältsäsongen som vi till-Patrikbringade i Kilsbergen och på Dylta Bruks herrgård. Tack allihop för tiden i ”huset som Gud glömde” mitt emot pappersbruket i Veja.

Carl-Gustaf och Carl-Gustaf och Carl-Gustaf Vivika Åkerhielm för att Patrik och jag fi ck ”vakta” herrgården sommaren Patrik och jag fi ck ”vakta” herrgården sommaren Patrik2002.

Karolina Nittérus, Göteborgs universitet för din positiva anda, och för att du lärt mig en del om kryp, blåa tänder, badrumsfaror och annat.

Botaniklärarna på Institutionen för skoglig vegetationsekologi, SLU samt personalen på bota-niska trädgården i Córdoba som bidragit till mitt stora intresse för växter samt mina exjobbs-handledare John Jeglum och Johan Svensson som gjorde mig intresserad av forskning.

TA-personalen: Lenita, Marie, Kaij och Kaij och Kaij Lars-Inge samt alla ni andra som får undervisning och annat jobb att fl yta på. Utan er skulle bygget och min dator stå still.

Andreas Renöfält vid SGS som gjorde det möjligt för mig att arbeta som miljörevisor inom skogsbruket vid sidan av mina doktorandstudier. Teorin är en sak men praktiken är en annan.

Sveaskog, SCA Skog, Holmen Skog och Skogsstyrelsen som har bidragit med data och för- sökslokaler. STEM och Björkmans som fi nansierat min forskning.

Lotta för den schyssta illustrationen.

Matt Groening med familj samt Matt Groening med familj samt Matt Groening Myra för att ni får mig att glömma arbetet.

Gamla skogiskamrater, polare samt nya kollegor och gamla kollegor: Fredrik B., Ola L., Snark-Jörgen, Åsa G, Marcos, Lotta S, Patrik, Olle F, Johnny, Katarina, Micke, Emma, Johanna, Zlatko, Anna, Erik, Niclas, Marie, Birgitta, Tuss, Elisabet, Cathy, Jim, Carl-Johan, Mats, inne-bandygänget och många fl er. Mycket kul har det blivit. Tack för: innebandy, Ecuador, Kana-rieöarna, undervisning, Bodö, Umeå, fi ske, mat, öl, vin, snacks, skidturer, bivacker, camping och vandring men också för vardagligt trams.

Mamma, Jan, Petra, Lars, Karl, Stina, Lisa, Hanna, Jesper, Mormor och Örbytomtarna som inspirerat och uppmuntrat men ibland också undrat vad fasiken man håller på med.

Sandra! Vad kan jag säga. Du har en stor del in denna avhandling. Du har stöttat och hjälpt mig helhjärtat i vått och i torrt. Viktigast är dock allt roligt vi gör tillsammans och den glädje som du ständigt ger mig.

-26-

Copyright © Marcus Åström ISBN: 91-7264-192-4Cover by Lotta Ström

Printed by Solfjädern Offset AB, Umeå, Sweden