Nitrogen, Parasites and Plants - Key Interactions in Boreal Forest Ecosystems

Joachim Strengbom

Umeå 2002

S\må vo • ±

Department of Ecology and Environmental Science Umeå University SE-901 87 Umeå

Sweden

AKADEMISK AVHANDLING

Som med vederbörligt tillstånd av rektorsämbetet vid Umeå universitet för erhållande av filosofie doktorsexamen i ekologi kommer att offentligen

försvaras fredagen den 18 januari 2002, kl. 10.00 i Stora hörsalen, KBC.

Examinator: Professor Lars Ericson

Opponent: Professor Gaius R Shaver, The Ecosystem Center, Marine Biological Laboratory, Woods Hole, Massachusetts, USA.

ISBN 91-7305-153-5 © Joachim Strengbom 2002

Printed by Solbädern Offset AB Cover: Bilberries (Vaccinium myrtillus)

by Joachim Strengbom

Organisation Document nameUMEÅ UNIVERSITY DOCTORAL DISSERTATIONDept, of Ecology and Environmental Science Date of issueSE-901 87 Umeå, Sweden January 2002

AuthorJoachim Strengbom

TitleNitrogen, Parasites and Plants - Key Interactions in Boreal Forest Ecosystem

AbstractIn the work described in this thesis I studied how increases in nitrogen (N) inputs may affect plant community structure in boreal forest understorey vegetation. These phenomena were investigated in N fertilization experiments and along a national N deposition gradient. After five years of N additions, large changes in understorey vegetation composition were observed in the fertilization study. In plots that received 50 kg N ha'1 year"1 (N2), the abundance of the dominant species, Vaccinium myrtillus, decreased on average by 32 %. No decrease was observed in control plots during the same period. In contrast, the grass Deschampsia flexuosa responded positively to increased N input, being on average more than five times as abundant in the N2 treatments as in controls.

Also an increase was seen in the incidence of disease caused by the parasitic fungus Valdensia heterodoxa on leaves of V. myrtillus following N additions. The parasite was on average nearly twice as abundant in N2 plots than in control plots. This could be explained by increased N concentrations in host plant tissue. Disease incidence also increased following experimental additions of glutamine to leaf surfaces of V. myrtillus, suggesting a causal connection between plant N concentration and performance of the fungus. The parasite also played a key role in the observed changes in understorey species composition. D. flexuosa was more abundant in patches in which V. myrtillus was severely affected by V heterodoxa. This suggests that V heterodoxa mediates the increased abundance of D. flexuosa following increased N additions. The fungus mediates changes in the composition of understorey vegetation mainly by increasing light availability via premature leaf loss of V. myrtillus.

The incidence of disease due to the parasite was on average higher in large than in smaller N-treated plots, indicating that the response to N fertilization is spatially scale dependent. This shows that using small plot sizes in experiments that simulate changed environmental conditions may be problematic, as important interactions may be underestimated.

Comparison of the occurrence of understorey species between regions with different rates of N deposition revealed that the occurrence of the two dwarf shrubs V. myrtillus and V. vitis- idaea was lower in regions with high N deposition compared to regions with low deposition. The opposite pattern was found for V heterodoxa. This is consistent with expectations from N fertilization experiments. For D. flexuosa no differences in occurrence were found between the different regions investigated.

The effects on vegetation and mycorrhizal fungi observed following N additions were also found to be long lasting. Nine years after termination of the fertilization, no signs of recovery were detected, and nearly 50 years after termination characteristic signs of N fertilization were found among bryophytes and mycorrhizal fungi. This suggests that the time needed for re-establishment of the original biota following N-induced changes may be substantial.

Key words: Critical loads, Deschampsia flexuosa, amino acids, mycorrhizal fungi, natural enemies, N deposition, N limitation, vegetational composition, Valdensia heterodoxa, Vaccinium myrtillus

Language: English ISBN: 91-7305-153-5 Number of pages: 41+5 app.

Signature: Date: 30 November 2001

Contents

Sammanfattning 2Introduction 4

N limitation in boreal forests 4Human alteration of the N cycle 5Responses of individual plants to increased N input 6

Plant community responses to increased N input. 6

Increased N input and forest production 7Increased N input and mycorrhizal fungi 8

Critical loads for N deposition 8

Natural enemies and increased N input 9Potential effects of spatial scale in simulation experiments 10Ecosystem responses to reductions in N input 10

Objectives 1 2

Methods used 13Studied species 13N fertilization experiments in areas with low background deposition 14 Natural gradients in N deposition 15Old fertilization experiments 16

Maj or results and discussion 18A conceptual model for N-induced changes in vegetation 22Importance of spatial scale 25Natural gradient in N deposition 25Ecosystem responses to decreased N input 27

Concluding remarks and future perspectives 29Acknowledgements 30References 31Tack 40Appendices: paper I-V

List of papers

Papers I-V

This thesis is based on the following papers, which will be referred to by the corresponding Roman numerals.

I. Strengbom, J., Nordin, A., Näsholm, T. & Ericson, L. 2002. Parasitic fungus mediates change in nitrogen exposed boreal forest vegetation. Journal o f Ecology 90, in press.

II. Strengbom, J., Näsholm, T. & Ericson, L. Light, not nitrogen, triggers expansion of the grass Deschampsia flexuosa in boreal forests. Manuscript.

III. Strengbom, J., Walheim, M., Näsholm, T. & Ericson, L. Regional differences in the occurrence of understorey species reflect nitrogen deposition in Swedish forests. Submitted manuscript.

IV. Strengbom, J. & Ericson, L. Importance of spatial scale for field simulation experiments: example from N fertilization of a boreal forest. Manuscript.

V. Strengbom, J., Nordin, A., Näsholm, T. & Ericson, L. 2001. Slow recovery of boreal forest ecosystem following decreased nitrogen input. Functional Ecology 15, 451-457.

Papers I and V are reproduced with the kind permission of the publishers.

Sammanfattning

Kväve är ett viktigt näringsämne för både växter och djur som behövs för uppbyggnaden av proteiner. Trots att jordens atmosfär består till 78 % av kväve är paradoxalt nog ämnet en bristvara i de flesta ekosystem. Detta beror på att ytterst fa av jordens organismer direkt kan utnyttja kvävgasen i atmosfären. Följaktligen har exempelvis växter anpassat sig till låg kvävetillgång. En sådan anpassning kan vara symbios med kvävefixerande bakterier som kan utnyttja det kväve som finns i atmosfären. Ett annat vanligt sätt för växter att tillgodose sitt behov av kväve är att leva i symbios med svampar, så kallade mykorrhizasvampar. Mykyorrhizasvampar levererar kväve till växten i utbyte mot kolföreningar. Växter har också utvecklat olika strategier för att hushålla med det kväve som en gång tagits upp. Långsam tillväxt eller att ha fleråriga blad är exempel på sådana strategier. Om kvävetillgången ökar är det inte säkert att växter med dessa anpassningar kan förändra sitt livssätt till de nya förhållandena.

Till följd av diverse mänskligt relaterade aktiviteter som ökad användning av konstgödsel eller förbränning av fossila bränslen har mängden kväve i omlopp mer än fördubblats jämfört med det som anses vara naturligt. Resultatet blir att haltema av olika kväveföreningar ökar i atmosfären. Kväveföreningarna kan då spridas långa sträckor, för att sedan blandas med regnvatten och falla ner (deponeras) som kväveberikad nederbörd. En ökad kvävetillgång kan leda till stora förändringar av växtsamhällen då arter som är anpassade till låg kvävetillgång lätt blir utkonkurrerade av arter som är anpassade till en högre kvävetillgång. Sådana vegetationsförändringar har skett på tidigare ljungdominerad hedmark i Nederländerna där ljungen ersatts av olika gräsarter då kvävedepositionen ökat.

Barrskogar i Skandinavien, boreala skogar, är typiska kvävebegränsade ekosystem. Många växtarter som blåbär och lingon har under lång tid anpassat sig till låg tillgång av kväve (se ovan). En ökad kvävedeposition kan tänkas leda till att dessa arter minskar i förekomst medan andra arter i stället ökar. I den här avhandlingen har jag studerat hur en ökad kvävetillgång påverkar faltskiktsvegetationen i boreala skogar samt vilka faktorer som är viktiga för sådana vegetationsförändringar. Jag har utnyttjat mig av gödslingsexperiment i områden med låg kvävedeposition och studerat skillnader mellan områden med hög respektive låg kvävedeposition. I Sverige följer kvävedepositionen en gradient från sydväst mot nordost. Depositionen är störst i sydvästra delen av landet för att sedan successivt minska i nordostlig riktning.

Efter fem år av kvävegödsling kan jag konstatera att långsamtväxande arter som blåbär minskat medan en gräsart, kruståtel, har ökat i förekomst. Även en parasitsvamp, som angriper blåbär och gör att bladen falls redan i mitten av sommaren, ökade efter kvävegödsling. Parasitsvampen visade sig vara mycket viktig för ökningen av kruståtel, då bladavfallet gör att ljustillgången för

2

kruståtel ökar. Eftersom kruståtel är en bättre konkurrent jämfört med blåbär vid förhöjd ljus- och kvävetillgång, leder detta till att gräset kan breda ut sig. Detta visar att samspelet mellan växter och deras naturliga fiender, som t. ex. parasitsvampar, kan vara viktiga för kväveorsakade vegetationsförändringar. Jag har också visat att förekomsten av parasitsvampen är högre i stora gödslade ytor än i små. Det betyder att man inte kan förvänta sig samma resultat om man gödslar en liten yta som om man gödslar en stor. Risken för att underskatta effekterna av en ökad kvävetillgång blir större om man använder sig av små ytor i ett gödslingsförsök.

Förekomsten av blåbärs- och lingonris visade sig också vara lägre i områden med hög kvävedeposition än i områden med låg deposition. Dessutom var en större andel av blåbärsriset angripet av parasitsvampen i områden med hög deposition av kväve. För kruståtel fanns det inga skillnader mellan dessa områden. Eftersom gradienten i kvävedeposition sammanfaller med många andra gradienter, t. ex. i nederbörd och temperatur, kan man inte säkert säga vad skillnaderna mellan områden beror på. En del av de skillnader som observerades härrör säkerligen från naturliga skillnader i förekomst av dessa arter. Mönstret i förekomst av blåbärs- och lingonris stämmer dock väl överens med resultat från både mina egna och andras gödslingsförsök, vilket tyder på att kväve kan ha bidragit till den låga förekomsten i områden med höga depositionsnivåer. Moderna skogsbruks metoder resulterar ofta i täta och mörka skogar, som kan påverka undervegetationen negativt. Detta är mest påtagligt i sydvästra delen av Sverige och har förmodligen också bidragit till den låga förekomsten av blåbärs- och lingon ris i den delen av landet.

För att kunna avgöra hur stor kvävedeposition man kan tillåta är det viktigt att veta om vegetationen kan återfå sin naturliga artsammansättning om kvävetillförseln minskar och i så fall hur lång tid denna återhämtningsprocess tar? Därför genomfördes en undersökning av två gamla gödslingsförsök. Undersökningen visade att effekterna på vegetationen ännu var påtagliga nio år efter det att kvävegödslingen hade upphört. Femtio år efter att gödslingen upphört gick det fortfarande att hitta skillnader mellan tidigare gödslade ytor och ogödslade kontrollytor. Det var framför allt en skillnad i förekomsten av vissa arter av mossor och i antalet fruktkroppar av mykorrhizabildande svampar. Detta visar att effekterna av en ökad kvävetillgång kan finnas kvar lång tid efter det att tillförseln upphört och att boreala ekosystem återhämtar sig långsamt.

3

Introduction

Nitrogen (N) is essential for life as we know it, but even though the atmosphere of our planet contains 78 % N, limited supplies of the element are available to the vast majority of living organisms (White 1994, Vitousek & Howarth 1991). This natural paradox is due to the fact that most organisms are incapable of converting N2, which is the dominant form of N in the atmosphere, into more biologically reactive forms. However, during the long history of earth many organisms have developed adaptations to meet this shortage of N. For example, some plant species live in symbiosis with bacteria that can fix N2 and consequently use it as an N source and others have symbiotic relationships with mycorrhizal fungi. Both of these kinds of symbiosis may increase the availability of N for plants. Other organisms have evolved strategies for N conservation and use the available N in an efficient way. However, if the supply of N increases, for instance due to increased N deposition, such strategies may be sub-optimal and species with these strategies may then be competitively excluded by other species that are better adapted to the new environmental conditions. Thus, deeper knowledge about the magnitude of these effects and the processes involved is needed.

N limitation in boreal forestsNitrogen is the growth-limiting nutrient in boreal forest ecosystems, as it is in many terrestrial ecosystems (Tamm 1991). The generally low N availability in boreal forest is due to a range of different factors. The relatively low annual temperature and short growing season in boreal areas makes decomposition of organic material slow as it inhibits the activity of decomposing micro organisms and, thus, causes low rates of N mineralization (Swift et al. 1979, van Cleve & Alexander 1981). In addition, the N-conserving features of plants from boreal forests result in low concentrations of N in dead and senescing parts of plants, and thus low N concentrations in the organic litter. Organic litter with low N concentrations decomposes more slowly than litter with higher N concentrations, which inhibits mineralization and further reduces N availability (van Cleve et al. 1983, Wedin & Tilman 1990, van der Krift & Berendse 2001). The slow decomposition also leads to the accumulation of organic matter, which may slow down the decomposition rate still further, as a thick layer of organic matter tends to reduce soil temperature and, thus, the activity of decomposers (Gosz 1981).

Due to the low mineralization rate, N tends to accumulate during the forest succession and mature boreal forests often have large N pools. However, the majority of the N in these pools is not available for primary production as it is immobilized by plants in standing biomass, by microbes or as non-decomposed organic matter in the soil (Tamm 1991). Some of the N may also be chemically immobilized in complex, degradation-resistant organic compounds (Johnson 1992). The relative importance of this type of immobilization is not fully

4

understood, but mineralization of N from these compounds is believed to be very slow (Johnson 1992, Mengel 1996).

Human alteration of the N cycleVarious anthropogenic processes, such as industrial fixation of N2 for use as fertilizer, increased cultivation of N-fixing crops and combustion of fossil fuels, which both mobilize N and fix N2, have increased the input of biologically reactive forms of N into the global N cycle (Vitousek et al. 1997). Compared with natural N fixation, these human-related activities have more than doubled the input of N into the global N cycle (Galloway et al. 1995, Vitousek et al. 1997).

Increased emissions of sulphur (S) and N compounds from the burning of fossil fuels have changed both atmospheric and precipitation chemistry, and subsequently also increased the input of these compounds by deposition to large areas of the western world (Galloway et al. 1995). Compared to the substantial reductions in emissions of S that have occurred in western parts of Europe in the last decade, reductions in N emissions have been marginal in this period (Tarrasón et al. 2000). N deposits may consist of either NOx or NHy compounds. There are large differences in the residence time in the atmosphere for these compounds: 1-2 days for NOx and 1-2 hours for NHy (Homung & Langan 1999). Problems associated with trans-boundary N pollution are thus almost always connected with NOx compounds. However, in the future NHy deposition may become more important further away from its sources as its residence time in the atmosphere is dependent on the atmospheric S0 2

concentration (Homung & Langan 1999), which at present is decreasing over large areas of Europe (Tarrasón et al. 2000). Rates of deposition of NHy may also be difficult to measure exactly, as it is deposited in gaseous form and data on NHy deposition is often estimated from models. Although uncertainties in the exact values may be large, deposition of NHy may be substantial close to the emission source (Lövblad et al. 1995, Lee 1998). Its short residence time in the atmosphere and local origins may also explain the large geographical variation in NHy deposition.

The total rates of N deposition in Europe show large geographical variations, ranging from an average of 1-2 kg N ha" 1 year" 1 in the northern parts of the Nordic countries to 47 kg N ha" 1 year" 1 in the Netherlands (Tarrasón et al. 2000). In Sweden, rates of N deposition range from 1-2 kg N ha" 1 year" 1 in the NE to more than 20 kg N ha" 1 year' 1 in the SW of the country (Lövblad et al. 1995). The composition of N deposits also varies between different geographical regions. For example deposition of NHy dominates in the Netherlands (Aerts & Bobbink 1999), while N deposition in Sweden consists of almost equal parts of NOx and NHy (Lövblad et al. 1995).

Responses of individual plants to increased N inputDirect damage to plants caused by N deposition is rarely observed with N concentrations found in current deposits (Homung & Langan 1999), but it has been noticed under experimental conditions with higher concentrations (Dueck et al. 1988). Although the concentrations near NHy point sources may be high enough to cause direct damage, most effects of N deposition are mediated by soil-plant interactions.

Increased N availability in the soil is almost always followed by increased N uptake by plants. If increased uptake of N is balanced by increased growth, no change in N concentration of plant tissue is to be expected. But as many plant species from N-limited environments have adaptations that promote N conservation, e.g. slow intrinsic growth rates and low N turnover rates, such plants may be constricted to a slow growth rate even if N availability increases. This may lead to increased N concentrations in their tissues. Increased N concentrations, partly due to increased amino acid levels in plant tissues, is also commonly reported following increases in N deposition or N fertilizer applications (Chapin et al. 1986, Lähdesmäki et al. 1990, Näsholm & Ericsson 1990, Näsholm e/<z/. 1994, Nordin et al. 1998).

Plant community responses to increased N input.Many plant species from N-limited ecosystems have slow potential growth rates, low N turnover rates and low concentrations of N in their senescing parts, i.e. high N use efficiency (e.g. Vitousek 1982, Berendse & Aerts 1987). Differences in N use efficiency may explain why some plant species are competitively superior under high N availability while others are superior competitors under low N availability (e.g. Chapin & Shaver 1989, Aerts & van der Peijl 1993, Aerts & Chapin 2000). Species adapted to low N conditions will only be successful competitors on soils with low N levels and are likely to be competitively excluded if the N availability increases, as their N conservation strategy may be sub-optimal under the new conditions (Chapin 1980, Aerts & van der Peijl 1993). Since many species of high conservation interest tend to be weak competitors and associated with nutrient-poor sites, an increase in anthropogenic N deposition is likely to result in changes in the plant community structure and reduced biodiversity in formerly N-limited ecosystems {cf. Bobbink et al. 1998).

The most well known example of N-induced vegatational changes comes from heathlands in the Netherlands where grasses like Molinia caerulea and Deschamspsia flexuosa have replaced dwarf shrubs such as Erica tetralix and Calluna vulgaris (Heil & Diemont 1983, Aerts & Berendse 1988, Aerts et al. 1990). Changes in understorey plant community structure have also been reported from boreal forests following N additions both in North America (e.g. Nams et al. 1993, Turkington et al. 1998) and Europe (Malmström 1949, Tamm 1991, Mäkipää 1994, Kellner & Redbo-Torestensson 1995). In all these

studies rapid increases were observed in the abundance or frequency of fast- growing grasses and herbs like D. flexuosa, Festuca altaica, Epïlobium angustifolium and Achillea millefolium. Reductions in the abundance of dwarf shrubs, such as Arcostaphylos uva-ursi or Vaccinuum myrtillus, have also been reported in several studies (e.g. Nams et al. 1993, Mäkipää 1994), but the reduction of V. myrtillus seems to be less pronounced on drier sites (Mäkipää 1994). Reductions in the abundance of lichens and some bryophytes are also commonly reported following N additions in boreal forests (Malmström 1949, Persson 1981, Dirkse & Matakis 1992).

Changes in forest understorey vegetation due to N deposition in northern Europe are less clear than those observed in N fertilization experiments. Conflicting results, ranging from no effects (e.g. Nygaard & 0degaard 1999) to changes towards a more nitrophilous flora in deciduous forests (Falkengren- Grerup & Eriksson 1990, Falkengren-Grerup et al 1998) and changes in coniferous forests consistent with results from fertilization experiments (0kland 1995) have been reported. National surveys on changes in understorey vegetation have also mentioned increased N deposition as a possible explanation for changes observed since 1950 (e.g. Reinikainen et al 2000). The conflicting results show that estimates of vegetational changes along gradients or over longer periods may be difficult to evaluate, as it may be difficult to isolate the effect of N deposition from other factors, for instance changes in silvicultural practice or natural geographical variation.

Increased N input and forest productionAs tree growth and vitality is of great economic interest for both governments and companies in countries like Sweden, the possibility that increased deposition of N, and other pollutants that tend to increase acidity in soil may decrease forest productivity has caused considerable concern. Some authors have suggested that the negative effects of N saturation, including nutrient imbalances in trees and potentially massive losses of nutrients due to increased acidity and leakage could cause forest decline (e.g. Sverdrup et al 1994, Wright et al 1995)

Gundersen (1991) presented a conceptual model describing the development of a forest ecosystem from N limitation towards N excess, with subsequent forest decline, following N deposition. The proposed process can be divided into three stages. In the first (production) phase, the increased N input results in increased N uptake by plants, causing primary production to increase. N availability in this phase does not exceed the N demands of the plants, and no leakage of N occurs, but ground vegetation may change towards domination by more nitrophilic species. In the second (destabilisation) phase, several ecosystem processes may be destabilised. Plant growth will no longer be N- limited and N will accumulate in plant tissue. This may increase frost- or parasite- induced damage and nutrient imbalances may occur. As the N

7

availability exceeds the demand of the biota, N leakage will start to increase. The third (decline) phase is associated with falling forest productivity due to soil acidification caused by the N 03" excess. At this stage N leakage will be substantial with loss of aluminium and base cations to run-off water.

Although all of these different phases may occur, coniferous forests can sustain substantial levels of N deposition without reaching the decline phase. Binkely & Högberg (1997) concluded that there was no evidence for reduced health or productivity in Swedish forests. Even in some areas with high N deposition, increased stand growth has been found following N fertilizer application, indicating that the systems concerned had still not reached N saturation (Binkely & Högberg 1997).

Increased N input and mycorrhizal fungiThe symbiosis between mycorrhizal fungi and plants in N-limited ecosystems is of great importance for nutrient uptake by plants since fungi provide plants with nutrients such as N in exchange for carbon (C) (e.g. Read et al. 1995). When N availability increases following fertilization or deposition, major changes in both abundance and species composition may occur in the mycorrhizal fungal community (Brandrud 1995, Boxman et al. 1998a, Lilleskov et al. 2001). The negative effects of increased N input are most pronounced in sporocarp formation (Brandrud 1995, Wiklund et al. 1995, Lilleskov et al. 2001). It is not fully understood why sporocarp production decreases following increased N input, but it has been hypothesised that relative shortage of carbon could be important (Wallander 1994). Following increases in N input the C/N ratio in soil organic matter will decrease and as the fungi are incapable of modulating their N uptake, they will receive a relative shortage of C. As production of sporocarps requires large amounts of C, a reduction in sporocarp production may be an adaptation for C conservation. Although this may describe a possible mechanism for the observed reductions in sporocarps, it does not explain the large differences in response seen between different species or genera. Moreover, it does not explain why species diversity and the number of mycelia-infected root tips also decreases following increases in N input (Kårén 1997, Boxman et al. 1998a, Wöllecke et al. 1999, Peter et al. 2001). How changes in mycorrhizal community composition or diversity may affect plant communities is not well understood, but it has been suggested that mycorrhizal fungal diversity is important for the maintenance of plant biodiversity and productivity (van der Heijden et al. 1998, Baxter & Dighton 2001).

Critical loads for N depositionThe term critical load is widely used in Europe when acceptable levels of pollutants such as S or N are discussed. The definition of critical load I use here follows that given by Nilsson & Grennfelt (1988): A quantitative estimate o f

the exposure to one or more pollutants below which significant harmful effects on specified elements o f the environment do not occur, according to present knowledge.

Natural enemies and increased N inputNatural enemies, including various herbivores (Huntley 1991) or pests and parasites (Harper 1990) can have profound effects on plant community structure. N concentration in plant tissues, either food or host tissues, is important for the establishment, survival and growth rates of natural enemies (Crawely 1993, White 1994, Marschner 1995). Increased concentrations of N compounds, e.g. free amino acids, in host plant tissues following increases in N input may thus increase the susceptibility of plants to different natural enemies (Schoeneweiss 1975).

Although the effect of increased N availability to the host plant on natural enemies varies from positive to negative (Waring & Cobb 1992) the positive effects tend to dominate, at least in natural ecosystems (Paul 1990). Both parasitic fungi and insect herbivores have been observed to increase following N additions to N limited ecosystems (Nordin et al 1998, Flückiger & Braun 1999). For instance, heather beetle larvae (Lochmaea suturalis) feeding on C. vulgaris with high N concentrations have a faster life cycle and higher final pupal weight than those feeding on plants with lower N concentrations (Heil & Diemont 1983, Power et al. 1995, 1998). This may lead to increases in population density of the beetle (Brunstig & Heil 1985), and thus to both more intensive and more frequent outbreaks of the beetle (Brunstig & Heil 1985, Berdowski 1987, Berdowski & Zeilinga 1987). Increased rates of N deposition are believed to be responsible for the decreased time intervals between outbreaks of this herbivore in the Netherlands (Berdowski 1993).

Experiments investigating competition between C. vulgaris and M. caerulea show that as long as C. vulgaris can form a closed canopy it is a superior competitor even under increased N conditions (Aerts & Heil 1993). However, after opening of the C. vulgaris canopy following herbivory by the heather beetle, grasses such as M. caerulea or D. flexousa can take advantage of their higher growth rates under the increased N conditions and competitively exclude C. vulgaris (Aerts et al. 1990, Aerts & Heil 1993). This suggests that the large-scale vegetational transitions of heathlands in the Netherlands cannot be explained by inter-specific plant competition alone and that natural enemies may play a key role in N-induced transitions of the vegetation. For other ecosystems, the importance of natural enemies in N-induced vegetation changes is not well known. Other factors such as frost or drought damages may have similar effects on the canopy cover as herbivory. Increased frost sensitivity has also been reported after N fertilizer applications to C. vulgaris (Berdowski 1993).

Potential effects of spatial scale in simulation experimentsAlmost all ecological experiments are conducted on smaller spatial and temporal scales than those of the system of interest, which may lead to misinterpretation of the results and misleading generalisations (Levin 1992). For example half of nearly 100 community ecology field studies analysed by Kareiva & Andersen (1988) were conducted using plots smaller than 1 m in diameter. When logistical and/or financial factors place a low limit on the plot size used in ecological experiments, it is important to know how such simplifications may affect their results and conclusions. This issue has major implications for experiments that simulate large-scale environmental processes such as the deposition of N or the effects of global warming. Data from such experiments are often used as inputs for parameterised models and are thus implicitly assumed to hold for larger areas, like landscape or even greater scales.

The spatial scale at which experiments are conducted has been shown to be important when assessing various ecological phenomena (e.g. Kareiva 1985, Bach 1988a, Petersen et al. 1997), for instance the significance of predator- prey interactions in aquatic ecosystems (e.g. Luckinbill 1974, Englund & Olsson 1996, Englund 1997). Although no studies of terrestrial ecosystems (to my knowledge) have found spatial scale effects of this kind they most certainly exist. For example, it seems likely that the interaction between C. vulgaris and the heather beetle could be spatially scale-dependent. Prior to an outbreak of this herbivore, there is a build-up phase in which the population density of the herbivore slowly increases. It seems unlikely that such increased density could occur if the increased food quality, due to increased N input, was restricted to the spatial scale used in most ecological experiments. If the critical load for N- induced vegetational changes was estimated from such small-scale experiments, N induced effects are likely to be underestimated.

Ecosystem responses to reductions in N inputA central question to consider when discussing the effects of increased N inputs to an ecosystem is whether the effects observed are reversible or not. If they are reversible, how quickly will the original processes and species composition of the ecosystems recover? Such questions are of crucial importance for the assessment of critical loads of N.

Studies of ecosystem recovery following reductions in N input have either involved building roofs to reduce the wet deposition of N (e.g. Boxman et al. 1998b), or old fertilization experiments, in which the fertilization has been terminated, have been exploited (e.g. Bowman & Steltzer 1998, Quist et al. 1999). Such studies have reported sharp decreases in exchangeable N in soil (Quist et al. 1998) and N 03‘ in run-off water following reductions in N input (Wright et al. 1993, Bredemeier et al. 1998, Xu et al. 1998) indicating that some ecosystem processes may recover rapidly. Although some chemical soil

10

parameters may recover rapidly, such as those mentioned above, the effects on plant community structure might persist for considerably longer (Vinton & Burke 1995). Decomposition of litter from fast-growing species generally exceeds that from slow-growing species and changes in species composition following increases in N input may thus increase nutrient cycling rates (Berendse 1990, Wedin & Tilman 1990, Hobbie 1992, Hobbie 1996). This might create feed-forward processes on N circulation that prevent or slow down the recolonization of the original vegetation (Wedin & Tilman 1990, Bowman & Steltzer 1998). Although it is uncertain if such processes could lead to a new stable equilibrium, it seems likely that they could increase the time needed for the original vegetation to re-establish. Knowledge of the time needed for this recovery may be important when evaluating the effects of increased N input on our ecosystems, and the maximal levels of emissions that will not harm them.

11

Objectives

The main objective of the work described in this thesis was to gain deeperknowledge about the effects of N inputs on understorey vegetation inconiferous forest ecosystems. The initial responses, following both increasesand decreases in N input have been given special attention.

The main questions addressed were:• How are natural enemies affected by biochemical changes, especially

increased N concentrations in plants, following increases in N input? (I)• Do natural enemies such as parasitic fungi play an important in the

observed N-induced changes in vegetation? (I, II)• What is the mechanism whereby natural enemies affect the competitive

relationship between plants in boreal forests? (II)• Does the plot size used in simulation experiments affect the response to

N additions? If so, what are the implications of using small plots in such simulation experiments? (Ill)

• Are there correlations between the response patterns of plants and natural enemies in N fertilization experiments, and in their occurrences along a national gradient of N deposition? (IV)

• How will the understorey vegetation and mycorrhizal fungi respond to long-term N additions? (V)

• Are the effects of increased N input on boreal forest ecosystems reversible? If so, is the time needed for re-establishment of the original biota likely to be short or prolonged? (V)

• How will the answers to the above questions influence our assessment of the critical load of N, below which no harmful effects on ecosystem processes or function can be detected? (Thesis)

12

Methods used

To study ecosystem effects of increased N inputs, two different methods were used. In Papers I-III and V, I used N fertilization experiments and in Paper IV I compared differences between regions along a natural N deposition gradient. Although each of these methods has its limitations, together they may provide answers to questions like those addressed in this thesis.

Studied speciesThis thesis is focused on just a few species, which all share two important features: they are all very common in understorey vegetation of boreal forest and changes in their abundance (due, for instance, to increased N deposition), are also likely to have large effects on other organisms associated with them. The two ericaceous dwarf shrubs V. myrtillus and V. vitis-idaea are among the most common plant species in Sweden, and large numbers of other organisms are associated with them (Niemelä et al 1982). They also have wide geographical distributions and may serve as model organisms for many other slow-growing species adapted to low N availability. The grass D. flexuosa is also common in boreal forests, although its abundance may be rather low. Because this species is known to respond positively to increased N availability this species may serve as a model for species favoured by high N levels.

The parasitic fungus Valdensia heterodoxa is found both in North America and Europe where it attacks V. myrtillus (its primary host), as well as species such as V vitis-idaea, Linnaea borealis, and Gymnocarpium dryopteris (Kujala 1946, Gjaerum 1970, Norvell & Redhead 1994). Its sexual stage is named Valdensinia heterodoxa (Eriksson 1974) and is a fungus belonging to the Ascomycetes. This parasite causes a brown spot disease on the leaves of the host, which may result in premature leaf loss (Norvell & Redhead 1994). Although it normally occurs only at low densities in the understorey, increased incidence leads to premature leaf loss and visible patches appear in the V. myrtillus cover. Infection occurs in early summer when fruit bodies develop from sclerotia, which overwinter in the veins of V. myrtillus leaves (Norvell & Redhead 1994). This fungus may also disperse asexually via star shaped conidia. These conidia are larger than the spores and are probably involved, almost entirely, in short distance dispersal (Kujala 1946, Redhead and Perrin 1972).

13

N fertilization experiments in areas with low background depositionIn order to determine key features of N-induced changes in vegetation, especially the initial responses and the mechanisms responsible for the changes, it could be advantageous to use fertilization experiments in areas with low background deposition. In areas with higher rates of N deposition changes may already have occurred and important factors may be more difficult to detect.

In 1996 a fertilization experiment was established within the Svartberget Experimental Forest in northern Sweden (64°14'N, 19°46'E), 70 km NW of Umeå. The area is located in the middle boreal zone (Ahti et a l 1968). The experimental site is a late successional Norway spruce {Picea abies) forest of Vaccinium-Myrtillus type sensu Kalliola (1973). The ericaceous shrub V. myrtillus dominates the understorey vegetation and co-dominant species are V. vitis-idaea, L. borealis and D. flexuosa. Some herbs like Maianthemum bifolium, Melampyrum pratense, M. sylvaticum, Solidago virgaurea, and Trientalis europaea are also found, but all at low frequencies.

The total background deposition of N in the study area is estimated to be low: 2-3 kg N ha" 1 year" 1 (Lövblad et al 1992). The experimental area includes plots of different sizes (1, 10, 100, 1000 and 5000 m2) plus controls. These plots were assigned randomly to three different N treatments: 0, 12.5, and 50 kg N ha" 1 year" 1 (n= 6 for each treatment and plot size permutation). Experimental N additions started in 1996 and were repeated in 1997, 1998, 1999 and 2000. Solid NH4NO3 was added in the form of granules by hand at the onset of the growth period, at the end of May or the beginning of June, each year.

In each plot, five sub-plots were permanently established for scoring the vegetation by a point intercept method, using a frame with 30 random points within an area of 20 x 60 cm (Jonasson 1988). For each species, the number of contacts was used as a measure of abundance. The vegetation analysis was repeated at the end of July for each of the five years during the period 1996- 2000.

Leaves of V. myrtillus were collected to estimate the incidence of disease caused by V. heterodoxa. From each plot 500 leaves were randomly sampled at the end of August or the beginning of September in each of the five years. A correction for patches with premature leaf loss was applied, so that leaves that had already fallen were assigned as diseased. Leaves were placed in paper bags, brought to the laboratory and either dried in an oven (40° C for 24 h) or pressed. Eighty of the original 500 leaves from each plot were randomly chosen, and leaves were checked for disease symptoms caused by the parasitic fungus Valdensia heterodoxa (a stable mean value was obtained after scoring roughly 70 leaves). In Paper III, the incidence of disease was compared between fertilized plots of different sizes (scoring the incidence of disease on V. myrtillus leaves from the central square metre of all fertilized plots). A total

14

of 1 0 0 leaves were sampled from each plot and disease incidence was scored as described above.

Amino acids were analysed in 30 current annual V myrtillus shoots that were sampled randomly from each experimental plot sized 1 0 0 m2 as well as from the controls (n=6 ), in June during the three years 1996 to 1998. The samples were immediately frozen on dry ice. The frozen leaves were separated from the shoots and milled in liquid N2. About 30 mg fresh weight was transferred into Eppendorf vials and amino acids were extracted by mixing the frozen plant material with 1 ml 10 mM HC1 for 60 min at 4°C. After centrifugation at 14 000 rpm for 30 min, the supernatant was pipetted into a second vial. Amino acids were analysed by HPLC with a fluorescence detector as their 9- fluorenylmethylformate (FMOC) derivatives (Näsholm et al 1987).

Paper II focused on an unfertilized area close to the larger fertilization experiment at the Svartberget Experimental Forest. In early June 1999, 40 plots sized 0.05 m2 were laid out. The plots were randomly used as controls or assigned to one of the following three treatments (n=10): application of 1 5N- labelled NH4NO3 equivalent to 50 kg N ha-1 (15N atom excess, 1%), increasing the light availability for the grass by reducing above-ground competition from V. myrtillus, and a combination of these two treatments. All non-fertilized plots also received a dose of 1 5N-labelled NH4NO3 (15N atom excess, 98%) equivalent to 0.5 kg N ha' 1 to trace the amount of N taken up under normal N conditions. In the treatment designed to reduce above-ground competition, V. myrtillus tillers were gently bent away from the 0.05 m2 circular plot, and were prevented from bending back by thin wires tied to small plastic pegs. This was done in early June 1999. At the same time, all leaves of D. flexuosa were counted in each plot. N was applied 10 days after leaf scoring by injection with a syringe beneath the bryophyte layer. In each plot, the N solution was divided and injected in five aliquots. All plots received the same volume of solution (100 ml). At the end of the growth period, i.e. mid-September, all leaves of D. flexuosa were scored and all above-ground biomass of D. flexuosa and V. myrtillus was harvested.

Total N and 15N atom percentages were analysed in oven-dried (80 °C for 24 h) milled plant material using a Europa Scientific Mod 20-20 stable isotope analyser interfaced with a Europa Scientific ANCA-NT preparation module (Ohlson & Wallmark 1999).

Natural gradients in N depositionIn Paper IV, the occurrence of three common forest understorey species, V. myrtillus, V. vitis-idaea, D. flexuosa and the parasite V. heterodoxa was compared between regions with different levels of N deposition along the natural N deposition gradient in Sweden. In total, 557 conifer-dominated forest stands were inventoried during the summer of 2000. The presence or absence

15

of the three species was recorded in twenty 0 . 0 1 m2 circular subplots per sample plot. The subplots were placed along the four cardinal points 2, 4, 6 , 8

and 10 m from the centre of the plot. When V. myrtillus was present in a subplot, the presence or absence of the fungus V. heterodoxa on the leaves was recorded. In addition, the stand size of V. myrtillus was recorded as being either small (< 0.05 m2) or large (> 0.05 m2).

Old fertilization experimentsThe two experimental sites considered in Paper V, the Norrliden site (64°23'N 19°45'E) and a site at 64°10'N 19°35'E designated Hesselman’s site are both old forest fertilization experiments. The understorey vegetation is fairly similar at the two sites and is classified as Vaccinium-Myrtillus-type according to Kalliola (1973), but the Norrliden site is slightly drier.

The Norrliden site was established in 1971. Three N treatments (N1-N3) were applied to plots sized 30 x 30 m. The treatments were replicated three times (n=3) with three unfertilized control plots. N was added once every year as granules of NH4NO3. The N1 and N2 treatments are still being fertilized, but the N addition to the N3 plots was terminated in 1990. The total amounts of N that the different treatments received up to 1999 were as follows: N l, 990; N2, 1980; and N3,2160 kg N ha"1. The average yearly doses of N the plots received were: 34, 6 8 and 108 kg N ha" 1 yr"1, respectively.

At Hessleman’s site, N additions started in 1937. A solution of ammonium nitrate was added by irrigation to an experimental plot once every second week throughout each growing season between 1937 and 1951. The total amount of N added to this plot corresponds to 1447 kg N ha"1. A control plot was also established, which received the same amount of water, without the ammonium nitrate, as the N-treated plot. The size of both of these plots is 10 x 16 m. Apart form the control, one treatment consisted of repeated additions of NH4 NO3

(applied on 32 occasions between 1944-1946) and the other two consisted of the NH4 NO3 additions plus one application of either 57kg phosphorus ha“1, or 1000 kg wood ashes ha"1. The N was added by irrigation with water during the growing season. The plot sizes of these treatments were 26 x 30 m, except for the plot that received phosphorus as well as N (15 x 30 m). The total amount of N added to these plots was 540 kg N ha"1.

In 1999, the understorey vegetation was analysed at the two sites, using the point intercept method (see above). The vegetation was analysed at 200 random points along each diagonal of the plots at Norrliden (a total of 400 points per plot). At Hesselman’s site the method was adjusted to fit smaller plot sizes. Each plot was examined with the same number of points per unit area as at the Norrliden site. In addition to the formerly fertilized plots and the control plot, vegetation data were obtained from one extra control close to the original experimental area.

16

Sporocarps of mycorrhizal fungi were scored in all fertilized plots and control plots at both sites. All species belonging to genera known to be mycorrhizal fungi according to Hansen & Knudsen (1992) were collected. Fruit bodies were sampled on one occasion, at the beginning of September 2000. All fruit bodies in each plot were collected and identified to species or genus level. At Hesselman’s site three extra control plots with a radius of 5 m were examined in addition to the treatment plots and the original control plot.

17

Major results and discussion

After five years of N additions to the experimental site at Svartberget, large changes were observed in the understorey vegetation. Although I have not observed any major changes in the number of species present, considerable shifts in the plant community structure have occurred. The dominant species, V. myrtillus, has decreased significantly in abundance as a result of the increased N input during the period 1996-2000 (Table 1). In the 1000 m2 plots that received an annual dose of 50 kg N ha' 1 (N2), the abundance of V. myrtillus decreased on average by 32 % compared to 17 % in plots receiving 12.5 kg N ha' 1 year' 1 (Nl), while a marginal increase was observed in the control plots during the same period (Fig. 1). This pattern is consistent with the response ofV. myrtillus observed following long-term N additions (28 years) with N doses of 34-108 kg N ha' 1 (Paper V). In contrast to V myrtillus, the grass D. flexuosa showed a positive response to increased N input (Table 1). After five years of N addition, D. flexuosa was on average four times as abundant in the Nl treatments as in controls and more than five times as abundant in the N2 treatments (Fig. 1).

500(a)

o(ÜcoÜ

0 )nE3

a>oc<0

73C3JQ<

300 -

300

250

200

150

100

50

0

-C-N1-N2

(b)

1996 1997 1998

Year1999 2000

Figure 1. Vegetation responses to the three nitrogen treatments (C = control, N l = 12.5 kg N ha'1 year'1, N2 = 50 kg N ha -1 year'1) during the five years 1996-2000 in terms o f the abundance of (a) V. myrtillus and (b) D. flexuosa. Vertical bars denote ± one S.E. (n=6).

18

Table 1. Repeated measures ANOVAs on the abundance of D. flexuosa, V. myrtillus and V. heterodoxa for the study period 1996-2000. In cases of heterosphericity (identified by Mauchly’s test of sphericity) degrees of freedom were adjusted by Huynh-Feldt epsilon within the SPSS 10.05 routine.

S ource df MS F P

V. mvrtillusBetween subjectsN 2 45.175 2.435 0.121Error 15 18.555Within subjectsYear 4 20.434 15.441 0 . 0 0 0

Year x N 8 4.511 3.409 0.003Error 60 1.323

D. flexuosaBetween subjectsN 2 233.752 3.863 0.044Error 15 60.504Within subjectsYear 4 42.383 29.495 0 . 0 0 0

Year x N 8 12.915 8.988 0 . 0 0 0

Error 60 1.437

V. heterodoxa*

Between subjectsN 2 0.872 11.122 0.001Error 15 0.078Within subjectsYear 2.755 0.944 59.936 0 . 0 0 0

Year x N 5.510 0.040 2.508 0.041Error 41.32 0.016

The general pattern of increasing abundance of a relatively fast-growing species such as D. flexuosa and decreasing abundance of more slow-growing dwarf shrubs following increased N input is in accordance with general models of plant competition (e.g. Tillman 1988, Aerts & van der Peijl 1993). The increased abundance of D. flexuosa is also consistent with other fertilization experiments from boreal forests (e.g. Kellner & Mårshagen 1991, Mäkipää 1994, Kellner & Redbo-Thorstensson 1995). However, the reduced abundance of V. myrtillus is not in accordance with the results of all fertilization studies from boreal areas. Some studies have reported insignificant responses, or even increases following N additions (Kellner & Mårshagen 1991, Kellner & Redbo-Thorstensson 1995) and Mäkipää (1994) found no negative responses in the abundance of dwarf shrubs such as V. myrtillus or V. vitis-idaea in dry sites, while on more mesic sites there was a negative response to N additions. This suggests that the response of V. myrtillus depends on site characteristics. In Paper IV, I found that in regions with high N deposition, V. myrtillus was more

19

common in forest stands dominated by Scots pine than in stands dominated by Norway spruce, while there were no differences between the two forest stand types in regions with lower N deposition. Because pine is generally found on drier and less productive soils than spruce this suggests that negative effects of N deposition on the abundance of dwarf shrubs should be less pronounced on dry sites with low productivity, and greater on more mesic and fertile sites. Some studies conducted in pine-dominated stands (e.g. Kellner & Mårshagen 1991, Kellner 1993) have found V myrtillus to be unaffected, or even to increase following N additions. Thus, these studies support the idea that sites with lower productivity may be resistant to N-induced changes of the understorey plant community structure. A possible explanation for the difference in responses is that on less fertile sites like pine heaths the organic layer may be too thin to retain large amounts of added N. As such soils are also often well drained, a large part of the added N may rapidly leave the system through leaching. Another possible explanation is that at low productivity sites fewer nitrophilous species or important natural enemies may be present. No substantial changes in the vegetation are likely to occur until such species have colonized the fertilized area. If the last explanation holds true, the observed effects may be short-term responses and the time needed for more profound changes following the establishment of new species may be substantial. If the sensitivity to increased N input is low in sites with lower productivity, then estimates of critical loads of N deposition for understorey vegetation of the Vaccinuum-Myrtillus type should not be based solely on data from such sites, as more mesic and fertile sites are likely to be more sensitive to N deposition.

oc0 )

T3Oc

0,8

0,6o 0 4 CO U ’ ^ CO% 0,2 Q

1996 1997 1998 1999 2000

Year

Figure 2. Response of Valdensia heterodoxa to the three nitrogen treatments (C = control, N1 = 12.5 kg N ha'1 year'1, N2 = 50 kg N ha _1 year'1) during the five years 1996-2000.

20

Besides the changed abundance of V. myrtillus and D. flexuosa observed in the fertilization experiment at Svartberget there was an increase in disease due to the parasitic fungus V heterodoxa on leaves of V myrtillus following N additions (Table 1). After five years of N treatment the incidence of this parasite was on average 1.5 times higher in N1 plots and nearly two times higher in N2 plots than in control plots (Fig. 2). This increase could be due to increased N concentrations in host plant tissues. Increased incidence of the fungus was paralleled by an increased concentration of free amino acids, notably glutamine and arginine, in V. myrtillus leaves (Paper I). Although the increased N concentration per se may explain the enhanced performance of the fungus, decreased levels of toxic carbon-based compounds, like various phenolics or tannins, following the N additions cannot be excluded as possible factors promoting the fungus, as such compounds may inhibit the growth of herbivores and parasites {cf. Bennett & Wallsgrove 1994). However, disease incidence was also found to increase following glutamine additions to the leaf surface of V. myrtillus (Paper I), which suggests there may be a causal connection between glutamine concentrations on the leaf surface and performance of the parasitic fungus. Although not entirely equivalent to raised glutamine levels in leaves following soil additions of N, this suggests that the parasite V. heterodoxa responded to the altered biochemistry of its host.

V. heterodoxa was also found to play a key role in observed changes in understorey species composition. If disease is severe, the parasite causes premature leaf shedding in its host, resulting in readily visible patches with premature leaf losses in the cover of V. myrtillus. In the study reported in Paper I, I found that the abundance of D. flexuosa was higher in patches diseased by the parasite than outside such patches, regardless of the level of N addition. This suggests that V. heterodoxa mediates the increased abundance of D. flexuosa following increased N additions. The fungus probably mediates changes in the composition of the understorey vegetation by increasing light availability due to the premature leaf losses in V. myrtillus. This hypothesis was tested in a separate experiment (Paper II) in which the effects of increased N availability and reduced above-ground competition from V myrtillus (increased light availability) on the growth-rate of D. flexuosa were tested. This study revealed that addition of N increased neither growth rates nor biomass of D. flexuosa, while increased light availability (reduced above-ground competition) did. This indicates that light availability, at least in the short term, is more important than N availability for both growth rate and above-ground biomass accumulation of D. flexuosa in boreal forests. It also suggests that natural enemies such as parasitic fungi can mediate changes in abundance of understorey vegetation by altering light availability. These suggestions are consistent with vegetational transitions observed in dry C. vulgaris heathlands in the Netherlands (Brunsting & Heil 1985, Aerts 1989). In this system, herbivory by the heather beetle seems to have similar effects to those of V. heterodoxa in boreal forests.

21

A conceptual model for N-induced changes in vegetationThe way that natural enemies may mediate N-induced changes in vegetation in an N-limited ecosystem can be summarized in the general conceptual model presented below (Fig. 3).

Changed plant community structure

3a.

If possible, increased growth

2a. tIncreased

1.Increased N

N input -► uptake byplants

2b. i

Changed plant community structure

7a.

Plants with high RGR

Increased light availability for other understorey plants

6a.

t -tReduced competitive ability for the host plant

\ t 6b.

Leaf losses in host plant

Plants with low RGR

t 5.

Increased concentration of free aa in leaf tissue

3b.

Increased susceptibility to attacks from natural enemies (NA)

IncreaseddiseaseincidencebyNA

Figure 3. Conceptual model for N-induced vegetational changes mediated by a natural enemy.

1. Increased uptake of N by plants following increased N input is a well known phenomenon (e.g. Chapin 1980, Aerts & Chapin 2000, Paper I).

2. Following N uptake, plants may respond in different ways depending on factors such as light availability and life history strategies.

2a. Generally plants with potentially high relative growth rates (RGRs) respond to increased N uptake by increasing their growth rate following increased N uptake. If species like V. myrtillus or C. vulgaris form a dense cover in the understorey vegetation, other understorey plants may then be light-limited (Berendse et al 1987, Aerts et a l 1990, Paper II). Inside a forest, dense tree cover may reduce light availability even further. If so, increased growth rates may not be possible even if the nutrient status of the plants improves.

22

2b. Plants adapted to N limitation are often characterized by slower RGRs than those adapted to more fertile conditions (Chapin 1980, Aerts & van der Peijl 1993). Although such plants, e.g. V. myrtillus, may respond to increases in N input by increasing their growth rates, they may not be able to increase them sufficiently (due to life history constraints) to balance the increased N uptake. Such “luxury uptake” will thus result in increased N concentrations in their tissues (Lipson et al. 1996, Nordin et al 1998, Paper I). This is often expressed in increased concentrations of free amino acids such as arginine and glutamine (Ohlson et a l 1995, Nordin et al 1998, Paper I).

3a. Increased N inputs may result in increased abundance of plants capable of rapid conversion of assimilated N into new biomass. This is likely to occur as long as the light availability is not limiting, e.g. in forest gaps, following clear cutting or if there are openings in the cover of other understorey species such as V myrtillus.

3b. Increased N concentration in plant tissue may increase susceptibility to attacks from different natural enemies (Schoeneweiss 1975).

4. Increased N concentration may result in increased attacks from natural enemies like the parasitic fungus V. heterodoxa. This was shown in Paper I, both in the fertilization experiment and following artificial addition of glutamine solution to V. myrtillus leaves. It is also consistent with other studies that have found increased performance of different parasites or herbivores following N additions (e.g. Brunsting & Heil 1985, Power et al. 1995, 1998, Nordin et al. 1998, Flückiger & Braun 1999).

5. If V. myrtillus is heavily diseased by the fungus this will result in premature leaf losses (Paper I).

6 a. The premature leaf loss may also increase light availability for other understorey species, as there will be openings in the otherwise dense cover of V. myrtillus.

6 b. Such damage is likely to affect host plants negatively and reduce their competitive ability compared to other species such as D. flexuosa. Although I have not quantified such effects, it seems likely that V. myrtillus is negatively affected by the large losses in leaf area (and thus photosynthetic capacity) caused by the increased disease incidence.

7a. Leaf losses by the dominant species will increase light availability and may release other species from light limitations. Such species will thus be able to increase their growth rate and subsequently increase in abundance. This was demonstrated in Paper I, where the abundance of D. flexuosa was much higher in patches with premature leaf losses than outside such patches. The crucial role of light availability indicated by the increases in both growth rate and above-ground biomass production following reductions in above-ground competition from V. myrtillus, is suggested by the results in Paper II. This is in accordance with the suggested mechanism for changes in vegetation following increased N

23

input to C. vulgaris dominated heathlands (Aerts 1989, Aerts et al. 1990).

7b.Reduced competitive ability for the dominant species, V. myrtillus, following intensive attacks from the parasite may result in increased abundance of other species.

However, the conceptual model presented above has several restrictions. The initial cover of the dominant species must be sufficient to create light limitations for other species. It is difficult to judge how dense such cover has to be, as it will also depend on whether there is a tree canopy above and how dense it is. Although V. myrtillus is a deciduous dwarf shrub and is thus not as effective in reducing light availability for other species as (for instance) C. vulgaris, the overall light availability is lower inside a boreal forest than in open heathland. This may explain why even small changes in light availability may be important for understorey species in boreal forests (Paper II). For example, Erica tetralix in open heathlands does not reduce the light availability as much as C. vulgaris, because of its low stature, so the grass M. caerulea can become dominant following increased N input without the E. tetralix cover being opened by a natural enemy (Aerts et a l 1990). If the N-favoured species has a taller stature than the dominant species, increased abundance may also be possible without additional interaction from a natural enemy. This shows that the response will be dependent upon the growth form or plant architecture of both dominant and subdominant species.

Following N and P additions to tundra vegetation, Betula nana responded by producing more new meristems and branches in a study by Chapin & Shaver (1996) and Bret-Harte et a l (2001). These adjustments to plant architecture enabled B. nana to gain dominance and overtop other species, and thus reduce their light availability. This may explain the simultaneous reductions in biomass of other species that were also observed (Bret-Harte et al 2001). Although this example is quite different from the one I present, the importance of plant architecture and light availability following fertilization are common to both studies, suggesting that such phenomena may also frequently occur in other ecosystems. Moreover, pair-wise competition experiments between two evenly sized plants may reveal little information about the effects of increased N input on plant community structure in nature, as the size differences and the advantage of being dominant may, at least in the short term, overwhelm other differences such as higher growth rate under increased N conditions.

24

Importance of spatial scaleIf natural enemies are important mediators of changes in plant community structure induced by increases in N deposition, the effects of their interactions with other species should be considered in any experiments designed to assess the impact of changes in environmental conditions. The results reported in Paper III show that the spatial scale of an experiment can have a significant effect on the observed response of the parasitic fungus V. heterodoxa. The incidence of disease caused by this fungus was on average higher in large than in small fertilized experimental plots, suggesting that perimeter to area ratios may be important for the response of this natural enemy to increased N inputs. Thus, using small plots in experiments simulating large-scale environmental changes like increases in N deposition may underestimate the effects of the treatment, as the plots may be too small to allow for population increases of important natural enemies. In the absence of such natural enemies, the rate of vegetational transition may be greatly decreased or may not even occur at all (Brunsting & Heil 1985, Berdowski 1987, Berdowski & Zelinga 1987, Aerts et al. 1990, Papers I and II). It has already been shown that results from small- scale experiments can be misleading if used to draw general conclusions (Englund 1997, Petersen et al. 1997, Englund et al. 2001). However, this problem may often be underestimated, and researchers need to be more aware of when and where such effects are likely to occur when planning experiments to evaluate processes affecting larger areas than their test plots. This may be especially important if results from experiments are used for deriving factors such as the critical load of N that a system can sustain without detectable harmful effects on ecosystem functions or processes.

Natural gradient in N depositionAre the results from the previously mentioned fertilization experiments at Svartberget generally applicable to coniferous forests at larger scales and can the same patterns be observed in nature following N deposition? To answer these questions the survey presented in Paper IV was conducted. Although it may be impossible to draw definitive conclusions about N-induced effects from descriptive data such as those gathered in the survey, they may be useful in combination with experimental data. The results in Paper IV show that the occurrences of the two dwarf shrubs, V. myrtillus and V. vitis-idaea, are lower in regions with high N deposition rates compared to regions with lower rates. The abundance of V. myrtillus was also lower, as shown by higher frequencies of patchy occurrence of the species in regions with high N deposition. In addition, V. myrtillus was also more frequently diseased by V. heterodoxa in these regions. Because these results are consistent with data from the previously described fertilization experiments (Paper I) and also with results from Paper V, it seems likely that N deposition has contributed to the low occurrence of the two dwarf shrubs found in regions with high N deposition. Although I have no data on how large the negative effects of this fungus are for V. myrtillus, the possibility cannot be excluded that the low and patchy

25

occurrence of the shrub observed in regions with high N deposition rates and its decreased abundance following N additions (Papers I and V) may be due to the fungus.

However, the climatic and edaphic differences between the regions covered are certainly important too. Modem silvicultural practice may also have contributed to the observed patterns. Large increases in standing volume, especially in south-western Sweden, have occurred since 1950, which most certainly has affected the light availability for understorey species. Such changes may be important and may partly explain the low occurrences of the dwarf shrubs in regions with high N deposition. Increased N availability for trees, following increased N deposition, has positive effects for tree growth, at least in the short term, and may have contributed to the observed increase in standing volume (Elfving & Tegnhammar 1996, Binkley & Högberg 1997). N deposition may thus also have indirectly affected the understorey vegetation by enhancing tree growth, with subsequent reductions in light availability for understorey species.

If the low occurrences of the target species are due to N deposition, then relatively low rates of deposition may have negative effects on ground vegetation. Large areas, therefore, may be affected since these relatively low rates of N deposition occur over extensive areas of northern Europe (Tarrasón et al 2000). Furthermore, since a large number of other organisms are associated with the two dwarf shrubs included in this study (Eriksson 1974, Niemelä et a l 1982) they can be regarded as keystone species in boreal forests, and reductions in their occurrence are likely to affect a wide range of other organisms in coniferous forest ecosystems.

One would expect a higher occurrence of D. flexuosa in regions with high N deposition rates, as numerous fertilizer application experiments have shown that this species is amongst the most favoured by increased N input (e.g. Tamm 1991, Kellner & Redbo-Tostensson 1995, paper I and V). Surprisingly, there were no differences in the occurrence of D. flexuosa between regions with high and low N deposition rates. One possible explanation for this is that increased abundance may not necessarily be reflected in increased frequency, as reported following N fertilization. Since only the presence or absence (frequency) was recorded in Paper IV, it is impossible to identify any differences in abundance. Moreover, N-induced changes in the vegetation may already have surpassed the optimal conditions for D. flexuosa, which is not an extremely nitrophilous species (Falkengren-Grerup 1995, Diekmann & Falkengren-Grerup 1998) and other more nitrophilous species may have begun to competitively exclude D. flexuosa from regions with higher rates of N deposition.

26

Ecosystem responses to decreased N inputIn Paper V, I addressed questions related to the ways understorey vegetation and sporocarp production of mycorrhizal fungi respond to long-term N additions and how the biota will react to reductions in N input. At Norrliden, the change in understorey vegetation following long-term addition of N in large annual doses was substantial. Plots that had received 6 8 and 108 kg N ha" 1 year" 1 also showed the largest shifts in species composition. In these treatments, the grass D. flexuosa had increased in abundance while the formerly dominant species V. myrtillus had decreased compared to control plots. The greatest changes in species composition were observed amongst the bryophytes: their species assemblage being almost completely changed after 28 years of N addition. The dominant species in control plots, Pleurozium schreberi, was almost absent from the N2 and N3 treatments. Instead the most abundant bryophytes were species such as Brachythecium starkei, B. reflexum and Plagiothecium denticulatum, which all showed marked increases.

No signs of rapid recovery in either species composition or abundance of individual species were detected nine years after the N addition was terminated in the N3 plots at Norrliden. The large effects on the vegetation composition caused by the N treatment were still present. At the second site investigated in this study, Hesselman’s site, there were no differences in the abundance of vascular plants between the controls and formerly fertilized plots. Some results, however, showed that the biota still differed 47 years after the termination of the treatments. The parasitic fungus V heterodoxa was more frequent in formerly fertilized plots than in control plots. The species composition of bryophytes also showed differences between control plots and formerly N fertilized plots. For instance, the N-favoured bryophytes B. reflexum and P. denticulatum were found only in the formerly fertilized plots. These two species increased dramatically after the N additions at the Norrliden site, and they are known to increase in abundance following increases in N availability (Kellner & Mårshagen 1991, Mäkipää 1995). Moreover, Malmström (1949) reported that growth of the dominant bryophyte of the site, Hylocomium splendens, was negatively affected by the fertilization. This effect of N fertilization still persists. In the present study, H. splendens was less abundant in formerly fertilized plots than in control plots. In other studies, this bryophyte has been reported to be sensitive to increases in N input (e.g. Dirkse & Martakis 1992). The study suggests that no rapid recovery in understorey vegetation can be expected in boreal forests after sharp decreases in N input. This is in accordance with other studies that have found ecosystem effects from nutrient treatments present several years after the treatments were terminated (Milchunas & Lauenroth 1995, Vinton & Burke 1995).

In addition sporocarp formation by mycorrhizal fungi showed clear differences between control and fertilized plots at both sites investigated. More sporocarps of species belonging to N-sensitive genera such as Cortinarious and Russula (e.g. Brandrud 1995) were present in the control plots than in the still fertilized

27

plots at the Norrliden site and in formerly fertilized plots at Hesselman’s site. N-correlated reductions in sporocarp abundance of species belonging to these genera have been found both in fertilization experiments (Brandrud 1995, Wiklund et al. 1997) and along N deposition gradients (Lilleskov et al. 2001). Below-ground structures of the mycorrhizal fungi community seem to be less sensitive to changes in N input, at least in short-term studies (Kårén & Nylund 1997, Jonsson et al 2000). However, some studies have also detected effects on below-ground parts (Kårén 1997, Peter et al 2001). Although the negative effects of increased N input may be small on below-ground parts, a decreased production of sporocarps may have important effects on the persistence of the fungi in the ecosystem. There are indications that colonisation by spores from sporocarps are important even in undisturbed forests, at least for species belonging to the Russula genus (Redecker et al. 2001). If N deposition reduces sporocarp production, such changes will reduce spore production and may in the long-term also change species composition and reduce biodiversity among mycorrhizal fungi.

The relatively prolonged effect of fertilization noted in Paper V indicates that N circulation in the formerly N fertilized plots may still be altered. Although the amount of exchangeable N in the soil at the Norrliden site had returned to control levels within a few years of the fertilization being terminated (Quist et al. 1998), there may still be an increased rate of N cycling within the system due to feed-forward processes from the altered vegetation. Each time a leaf is shed, it carries with it approximately half of its maximum N capital (Chapin & Kedrowski 1983, Chapin & Shaver 1989). Thus, fast-growing species such as grasses, which have a larger biomass turnover rate than dwarf shrubs, may produce litter that is both more abundant and more easily decomposed (Berendse 1990, Wedin & Tilman 1990, Hobbie 1992, 1996). This will also increase N mineralization rates and thus enable rates of N circulation to increase. Such feed-forward processes have been suggested to inhibit the recovery of ecosystems following reductions in N input (Bowman & Steltzer 1998).

The view of slow ecosystem recovery presented here conflicts with that of several authors who have interpreted observed decreases in N leakage and nutrient levels in trees following reductions in input as signs of rapid ecosystem recovery (Wright et al. 1993, Bredemeier et al. 1998, Xu et al. 1998). The results presented in Paper V suggest that the time needed for re-establishment of the original biota in formerly fertilized plots may be substantial. It could be argued that the prolonged effects of N fertilization on bryophytes and mycorrhizal fungi at Hesselman’s site may be due to slow re-colonization rates. The plot sizes at this site were relatively small and because the surrounding vegetation remained unaffected by the N treatment, it seems unlikely that the slow re-establishment of species was caused by dispersal limitations. The rate at which vegetation may re-establish in these plots is also likely to be much faster in such small plots than in larger areas or whole landscapes that have

28

been affected by increased N inputs, like those caused by atmospheric N deposition.

Concluding remarks and future perspectives

In this thesis I have considered how both increases and reductions in N input may affect plant community structure in boreal understorey vegetation. I have found that the parasitic fungus V hétérodoxe plays a critical role in N-induced changes in the vegetation. The positive response of this fungus to increases in N concentration in host plant tissue resulted in severe leaf losses by the host, V. myrtillus. This generated openings in the V. myrtillus canopy cover, allowing increased growth of the grass D. flexuosa, mainly due to increased light availability. This species can thus take advantage of its higher growth rate under increased N conditions and gain dominance. A similar mechanism is known to be important in N-induced vegetational transitions in C. vulgaris dominated heathlands. Because natural enemies in combination with changed environmental conditions have been shown to affect both the direction and rate of transition in boreal forests and open heathlands (two markedly differing ecosystems), it seems likely that such effects could be important in other ecosystems as well.

Both parasites and herbivores may influence the shape and structure of plant communities in many ecosystems (e.g. Crawely 1983, Huntely 1991 Castello et al 1995). The importance of such interactions is likely to be affected by changes in a range of environmental factors, including temperature and N availability (e.g. Petchy et al. 1999). If such natural enemies are important for plant community structure, it may be difficult or even impossible to predict changes following alterations in environmental conditions, like increases in N deposition, without considering interactions involving natural enemies.

Since the population density and frequency of outbreaks by different parasites are often both stochastic and dependent on a wide range of different environmental factors, long-term studies may be needed to predict their effects on the plant community. Moreover, not only the temporal scale, but also the spatial scale of experiments may be important for assessing effects on plant communities. In this thesis I have shown that the plot size used in N fertilization experiments may affect the incidence of disease by V. heterodoxa. Such effects may have implications for the interpretation of results from experiments simulating changes in different environmental factors, as small plot sizes may not allow population increases of potentially important natural enemies. I believe that this issue needs more attention to increase our knowledge about when and why such spatial scale dependence may occur.

In this thesis I also addressed questions about what happens to plant communities following reductions in N input. I found differences in plant

29

species and mycorrhizal fungi composition still to be present between controls and formerly fertilized plots nearly 50 years after the N additions were terminated. This suggests that the time needed for re-establishment of the original biota is substantial. Although I do not know the mechanisms behind this, I speculate that the changes in species composition may contribute to the slow recovery rate, as they may affect N circulation rates. Several authors have suggested that changes in species composition or changes in biodiversity may alter ecosystem processes like N circulation (e.g. Hobbie 1992, 1996, Chapin et al. 1997, Bowman & Steltzer 1998). I believe that more experimental data on these mechanisms are needed. Another issue that is related to this one, is how changes in composition or diversity of mycorrihizal fungi will affect the plant community. Some studies suggest that the diversity of mycorrhizal community may be an important factor affecting the productivity and biodiversity of vascular plant communities (van der Heijden 1998, Baxter & Dighton 2001). New PCR techniques for determining the below ground species composition of these fungi may provide valuable new information on such effects in the near future (Leake 2001).

Finally, I want to mention the need for more information about the connections between the C and N cycles. Because increased N deposition may increase CO2

sequestration, due to increased growth by plants, it has been hypothesised that high N deposition may reduce the rate of C 0 2 increase in the atmosphere. However, the connection between C and N cycles do not seem to be that simple. In a study by Nadelhoffer et al. (1999), increased N inputs to a temperate forest in North America made only a minor contribution to C0 2

sequestration. More experimental data concerning this issue are needed in order to understand if and how N deposition may affect C0 2 enrichment at regional and global scales.

Acknowledgements

Thanks to Mattias Edman, Lars Ericson, Nicholas Kruys, Torgny Näsholm and Annika Nordin for providing comments on earlier versions of this thesis. I would also like to thank John Blackwell for correcting my English. The work in this thesis has been funded by grants from the MISTRA research programme International and National Abatement Strategies for Transboundary Air Pollution (ASTA), the Swedish Environmental Protection Agency (both to L. Ericson) and the Gunnar and Ruth Björkman foundation.

30

References

Aerts, R. 1989. Aboveground biomass and nutrient dynamics of Calluna vulgaris and Molinia caerulea in dry heathland. Oikos 56, 31-38.

Aerts, R. & Berendse, F. 1988. The effect of increased nutrient availability on vegetation dynamics in wet heathlands. Vegetatio 76, 63-69.

Aerts, R., Berendse, F., de Caluwe, H. & Schmitz, M. 1990. Competition in heathland along an experimental gradient of nutrient availability. Oikos 57,310-318.

Aerts, R. & Heil, G. W. (eds) 1993. Heathlands: Patterns and Processes in a Changing Environment. Kluwer Academic Publishers, the Netherlands.

Aerts, R. & van der Peijl, M. J. 1993. A simple model to explain the dominance of low-productive perennials in nutrient-poor habitats. Oikos 6 6 , 144-147.

Aerts, R. & Bobbink, R. 1993. The impact of atmospheric nitrogen deposition on vegetation processes in terrestrial, non-forest ecosystems In The impact o f nitrogen deposition on natural and semi-natural ecosystems (ed Langan, S. J.) Kluwer Academic Publishers, Dordrecht, The Netherlands.pp 85-122.

Aerts, R. & Chapin III, S.F. 2000. The mineral nutrition of wild plants revisited: A re-evaluation of processes and patterns. Advances in Ecological Research 30, 1-67.

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

Baxter, J. W. & Dighton, J. 2001. Ectomycorrhizal diversity alters growth and nutrient acquisition of grey birch {Betula populifolia) seedlings in host- symbiont culture conditions. New Phytologist 152, 139-149.

Bach, C. E. 1988a. Effects of host plant patch size on herbivore density: patterns. Ecology 69, 1090-1102.

Bach, C. E. 1988b. Effects of host plant patch size on herbivore density: underlying mechanisms. Ecology 69, 1103-1117.

Bennett, R. N. & Wallsgrove, R. M. 1994. Tansley Review No. 72. Secondary metabolites in plant defence mechanisms. New Phytologist 127, 617-633.

Berdowski, J. J. M. 1987. Transition from heathland to grassland initiated by the heather beetle. Vegetatio 72, 167-173.

Berdowski, J. J. M. 1993. The effect of external stress and disturbance factors on C<z//wH<z-dominated heathland vegetation. In Heathlands: Patterns and Processes in a Changing Environment, (eds Aerts, R. & Heil, G. W.) Kluwer Academic Publishers, the Netherlands, pp 85-124.

Berdowski, J. J .M. & Zeilinga, R. 1987. Transition from heathland tograssland: damaging effects of the heather beetle. Journal o f Ecology 75, 159-175.

Berendse, F. 1990. Organic matter accumulation and nitrogen mineralization during secondary succession in heathland ecosystems. Journal o f Ecology 78,413-427.

Berendse, F. & Aerts, R. 1987. Nitrogen-use-efficiency: a biologically meaningful definition? Functional Ecology 1, 293-296.

31

Berendse, F., Oudhof, H. & Bol, J. 1987. A comparative study on nutrient cycling in wet heathland ecosystems. Oecologia 74, 174-184.

Binkley, D. & Högberg, P. 1997. Does atmospheric deposition of nitrogen threaten Swedish forests? Forest Ecology and Management 92, 119-152.

Bobbink, R., Hornung, M, & Roelofs, J. G. H. 1998. The effects of air-borne pollutants on species diversity in natural and semi natural European vegetation. Journal o f Ecology 8 6 , 717-738.

Bowman, W. D. & Steltzer, H. 1998. Positive feedbacks to anthropogenicnitrogen deposition in Rocky mountain alpine tundra. Ambio 27, 514-517.

Boxman, A. W., Blanck, K., Brandrud, T. E., Emmett, B. A., Gundersen, P., Hogervorst, R. F., Kjonaas, O. J., Persson, H. & Timmermann, V. 1998a. Vegetation and soil biota response to experimentally-changed nitrogen inputs in coniferous forest ecosystems of the NITREX project. Forest Ecology and Management 101, 65-79.

Boxman, A. W., van der Ven, P. J. M. & Roelofs, J. G. M. 1998b. Ecosystem recovery after a decrease in nitrogen input to a Scots pine stand at Ysselsteyen, the Netherlands. Forest Ecology and Management 101, 155- 163.

Brandrud, T. E. 1995. The effects of experimental nitrogen addition on the ectomycorrhizal fungus flora in an oligothrophic spruce forest at Gårdsjön, Sweden. Forest Ecology and Management 71, 111-122.

Bredemeier, M., Blanck, K., Dohrenbusch, A., Lamersdorf, N., Meyer, A. C., Murach, D., Parth, A. & Xu, Y.-J. 1998. The Solling roof project-site characteristics, experiments and results. Forest Ecology and Management 101,281-293.

Bret-Harte, M. S., Shaver, G. R., Zoemer, J. P., Johnstone, J. F., Wagner, J. L., Chavez, A.S., Gunkelman IV, R. F., Lippert, S. C. & Laundre, J. A. 2001. Developmental plasticity allows Betula nana to dominate tundra subjected to an altered environment. Ecology 82, 18-32.

Brunsting, A. M. H. & Heil, G. W. 1985. The role of nutrients in theinteractions between a herbivorous beetle and some competing plant species in heathlands. Oikos 44, 23-26.

Castello, J. D., Leopold, D. J. & Smallidge, P. J. 1995. Pathogens, patterns, and processes in forest ecosystems. BioScience 45, 16-24.

Chapin, III, S. F. 1980. The mineral nutrition of wild plants. Annual Review o f Ecology & Systematics 11, 233-260.

Chapin, III, S. F. & Kedrowski, R. A. 1983. Seasonal changes in nitrogen and phosphorous fractions and autumn retranslocation in evergreen and deciduous taiga trees. Ecology 64, 376-391.

Chapin, III, S. F., Shaver, G. R., & Kedrowski, R. A. 1986. Environmental control over carbon, nitrogen and phosphorous fractions in Eriophorum vaginatum in Alaskan tussock tundra. Journal o f Ecology 74, 167-195.

Chapin, III, S. F. & Shaver, G. R. 1989. Differences in growth and nutrient use among arctic plant growth forms. Functional Ecology 3, 73-80.

32

Chapin, III, S. F. & Shaver, G. R. 1996. Physiological and growth responses of artic plants to a field experiment simulating climatic change. Ecology 77, 822-840.

Chapin, III, S. F., Walker, B. H., Hobbs, R. J., Hooper, D. U., Lawton, J. H., Sala, O. E. & Timan, D. 1997. Biotic control over functioning of ecosystems. Science 277, 500-504.

Crawely, M. J. 1993. Herbivory. The Dynamics o f Animal-Plant Interactions. Studies in ecology volume 10. Blackwell Scientific Publications, London.

Diekmann, M. & Falkengren-Grerup, U. 1998. A new species index for forest vascular plants: development of functional indices based on mineralization rates of various forms of soil nitrogen. Journal o f Ecology 8 6 , 269-283.

Dirkse, G. M. & Martakis, G .F. P. 1992. Effects of fertilizer on bryophytes in Swedish experiments on forest fertilization. Biological Conservation 59, 155-161.

Dueck, T. A., Dii, E. W. & Pasman, F. J. M. 1988. Adaptation of grasses in the Netherlands to air pollution. New Phytologist 108, 167-174.

ElfVing, B. & Tegnhammar, L. 1996. Trends of tree growth in Swedish forests 1953-1992: An analysis based on sample trees from the National Forest Inventory. Scandinavian Journal o f Forest Research 11, 26-37.

Englund, G. 1997. Importance of spatial scale and prey movements in predator caging experiments. Ecology 78,2316-2325.

Englund, G. & Olsson T. 1996. Treatment effects in a predator cagingexperiment: influence of predation rate and prey movements. Oikos 77, 519-528.

Englund, G., Cooper, S. D. & Samelle, O. 2001. Application of a model of scale dependence to quantify scale domains in open predation experiments. Oikos 92, 501-514.

Eriksson, B. 1974. On Deuteromycetes on Diapensales and Ericales in Fennoscandia. Svensk Botanisk Tidskrift 6 8 , 235-253.

Falkengren-Grerup, U. 1995. Long-term changes in flora and vegetation in deciduous forests of southern Sweden. Ecological Bulletins 44, 215-226

Falkengren-Grerup, U. & Eriksson, H. 1990. Changes in soil, vegetation and forest yield between 1947 and 1988 in beech and oak sites of southern Sweden. Forest Ecology and Management 38, 37-53.

Falkengren-Grerup, U., Brunet, J. & Diekmann M. 1998. Nitrogenmineralisation in deciduous forest soils in south Sweden in gradients of soil acidity and deposition. Environmental Pollution 102, suppl. 1,415- 420.

Flückiger, W. & Braun, S. 1999. Nitrogen and its effect on growth, nutrient status and parasite attacks in beech and Norway spruce. Water, Air and Soil Pollution 116, 99-110.

Gjaerum, H. B. 1970. A curious fungus on Vaccinium myrtillus. Blyttia 28, 159-163.

33

Galloway, J. N., Schlesinger, W. H., Levy II, H., Michaels, A. & Schnoor, J. L. 1995. Nitrogen fixation: atmospheric enhancement—environmental response. Global Biogeochemical Cycles 9, 235-252.

Gosz, J. R. 1981. Nitrogen cycling in coniferous ecosystems. In Terrestrial Nitrogen Cycles (eds Clark, F. E. & Rosswall, T.). Ecological Bulletins 33,405-426.

Gunderssen, P. 1991. Nitrogen deposition and the forest nitrogen cycle: role of denitrification. Forest Ecology and Management 44, 15-28.

Hansen, L. & Knudsen, H. 1992. Nordic macromycetes. Vol. 2. Nordsvamp, Helsinki.

Harper, J. H. 1990. Pests, pathogens and plant communities: an introduction. In Pests, Pathogens and Plant Communities, (eds Burdon, J. J. & Leather, S. R.) Blackwell Scientific Publications, Oxford.pp 3-14.

Heil, G. W. & Diemont, W. H. 1983. Raised nutrient levels change heathland into grassland. Vegetatio 53, 113-120.

Hobbie, S. E. 1992. Effects of plant species on nutrient cycling. Trends in Ecology & Evolution 7, 336-339.

Hobbie, S. E. 1996. Temperature and plant species control over litterdecomposition in Alaskan tundra. Ecological Monographs 6 6 , 503-522.

Hornung, M. & Langan, S. J. 1999. Nitrogen deposition: sources, impacts and responses in natural and semi-natural ecosystems. In The impact o f nitrogen deposition on natural and semi-natural ecosystems (ed Langan,S. J.) Kluwer Academic Publishers, Dordrecht, The Netherlands, pp 1-14.

Huntly, N. 1991. Herbivores and the dynamics of communities and ecosystems. Annual Review in Ecology & Systematics 22, 477-503.

Jonasson, S. 1988. Evaluation of the point intercept method for the estimation of plant biomass. Oikos 52, 101-106.

Johnson, D. W. 1992. Nitrogen retention in forest soils. Journal o f Environmental Quality 21, 1-12.

Jonsson, L. Dahlberg, A. & Brandrud, T. E. 2000. Spatiotemporal distribution of an ectomycorrhizal community in an oligotrophic Swedish Picea abies forest subjected to experimental nitrogen addition: Above- and below- ground views. Forest Ecology and Management 132, 143-156.

Kalliola, R. (1973) Suomen Kasvimaantiede. Werner Söderström Porvoo, Helsinki.

Kareiva, P. 1985. Finding and losing host plant by Phyllotreta: patch size and surrounding habitat. Ecology 6 6 , 1809-1816.

Kareiva, P. & Andersen, M. 1988. Spatial aspects of species interactions: the wedding of models and experiments. In Community Ecology (ed. A. Hastings). Springer-Verlag, New York, pp 38-54.

Kårén, O. 1997. Effects o f air pollution andforest regeneration methods on community structure o f ectomycorrhizal fungi. PhD thesis, Department of Forest Mycology and Pathology, Swedish University of Agricultural Sciences, Uppsala, Sweden.

Kårén, O. & Nylund, J-E. 1997. Effects of ammonium sulphate on thecommunity structure and biomass of ectomycorrhizal fungi in a Norway

34

spruce stand in southwestern Sweden. Canadian Journal o f Botany 75, 1628-1642.

Kellner, O. 1993. Effects on associated flora of sylvicultural nitrogenfertilization repeated at long intervals. Journal o f Ecology 30, 563-574.

Kellner, O. & Mårshagen, M. 1991. Effects of irrigation and fertilization on the ground vegetation in a 130-year-old stand of Scots pine. Canadian Journal o f Forest Research 21, 733-738.

Kellner, O. & Redbo-Torstensson, P. 1995. Effects of elevated nitrogendeposition on the field-layer vegetation in coniferous forests. In Effects o f acid deposition and troposheric ozon on forest ecosystems in Sweden (eds Staaf, H. & Tyler, G.). Ecological Æw//efr‘/w(Copenhagen) 44, 227-237.

Kujala, V. 1946. Sterförmige Diasporen bei Saliastrum (Gloeosporium) myrtilli (Allesch.) Kujala. Memoranda societatis pro fauna et flora Fennica 22, 137-141.

Lähdesmäki, P., Pakonen, T., Saari, K. & Havas, P. 1990. Environmentalfactors affecting basic nitrogen metabolism and seasonal levels of various nitrogen fractions in tissue of bilberry, Vaccinium myrtillus. Holarctic Ecology 12, 19-30.

Leake, J. R. 2001. Is diversity of ectomycorrhizal fungi important for ecosystem function? New Phytologist 152, 1-8.

Lee, J. A. 1998. Unintentional experiments with terrestrial ecosystems:ecological effects of sulphur and nitrogen pollutants. Journal o f Ecology 86, 1-12.

Levin, S. A. 1992. The problem of pattern and scale in ecology. Ecology 73, 1943-1967.

Lilleskov, E, A., Fahey, T. J. & Lovett, G. M. 2001. Ectomycorrhizal fungal aboveground community change over an atmospheric nitrogen deposition gradient. Ecological Applications 11, 397-410.

Lipson, D. A., Bowman, W. D., & Monson, R. K. 1996. Luxury uptake and storage in the rhizomatous alpine herb Bistorta bistortioides. Ecology 77, 1277-1285

Lövblad, G., Andersen, B., Hovmand, M., Jofffe, S., Pedersen, U. & Reisell, A. 1992. Mapping deposition o f sulphur, nitrogen and base cations in the Nordic countries. IVL Report B 1055. Swedish Environmental Research Institute, Göteborg, Sweden

Lövblad, G.,Kindbom, K., Grennfelt, P., Hultberg, H. & Westling, O. 1995. Deposition of acidifying substances in Sweden. In Effects o f acid deposition and troposheric ozon on forest ecosystems in Sweden (eds Staaf, H. & Tyler, G.). Eclogical Bulletins (Copenhagen) 44, 17-34.

Luckinbill, S. 1974. The effect of space and enrichment on a predator-prey system. Ecology 55,1142-1147.

Mäkipää, R. 1994. Effects of nitrogen on the humus layer and groundvegetation under closed canopy in boreal coniferous stands. Silva Fennica 28,81-94.

Mäkipää, R. 1995. Sensitivity of forest-floor mosses in boreal forest to nitrogen and sulphur deposition. Water, Air and Soil Pollutions 85, 1239-1244.

35

Malmström, C. 1949. Studien über Waldtypen und Baumartenverteilung im län Västerbotten (in Swedish with German summary) Meddelande från statens skogsforkningsinstitut, Stockholm 37(11), 1-231.

Marschner, H. 1995. Mineral Nutrition o f Higher Plants. Academic Press, London.

Mengel, K. 1996. Turnover of organic nitrogen in soil and its availability to crops. Plant and Soil 181, 83-93.

Milchunas, D.G. & Lauenroth, W.K. 1995. Inertia in plant community structure: State changes after cessation of nutrient-enrichment stress. Ecological Applications 5, 452-458.

Nadelhoffer, K. J., Emmet, B. A., Gundersen, P., Kjonaas, O. J., Koopmans, C. J., Schleppi, P., Tietema, A. & Wtight, R. F. 1999. Nitrogen deposition makes a minor contribution to carbon sequestration in temperate forests. Nature 398, 145-148.

Nams, V. O., Folkard, F. G. & Smith, J. N. M. 1993. Effects of nitrogen fertilization on several woody and nonwoody boreal species. Canadian Journal o f Botany 71, 93-97.

Näsholm, T., Sandberg, G. & Ericsson, A. 1987. Quantitative analysis of amino acids in conifer tissues by high-performance liquid chromatography and fluorescence detection of their 9-fluorenylmethyl chloroformate derivatives. Journal o f Chromatography 396, 225-236.

Näsholm, T., & Ericsson, A. 1990. Seasonal changes in amino acids, protein and total nitrogen in needles of fertilized Scots pine trees. Tree Physiology 6 , 267-281.

Näsholm, T., Edfast, A-B., Ericsson, A. & Nordén, L-G. 1994. Accumulation of amino acids in boreal coniferous forest plants in response to increased nitrogen availability. New Phytologist 126, 137-143.

Nilsson, J. & Grennfelt, P. 1988. Critical Loads for Nitrogen and Sulfur.Report from a workshop held at Skokloster, Sweden 19-24 March 1988. Miljorapport 1988, 15, 1-418. Copenhagen.

Niemelä, P., Tahvanainen, J., Soijonen, J., Hokkanen, T. & Neuvonen, S. 1982. The influence of host plant growth form and phenology on the life strategies of Finnish macrolepidopterous larvae. Oikos 39, 164-170.

Nordin, A., Näsholm, T. & Ericson, L. 1998. Effects of simulated N deposition on understorey vegetation of boreal coniferous forest. Functional Ecology 12, 691-699.

Norvell, L. L & Redhead, S. A. 1994. Valdensia heterodoxa (Sclerotiniaceae) in the United States. Canadian Journal o f Forest Research 24, 1981-1983.

Nygaard, P. H. & 0degaard, T. 1999. Sixty years of vegetation dynamict in a south boreal coniferous forest in southern Norway. Journal o f Vegetation Science 10, 5-16.

Ohlson, M., Nordin, A. & Näsholm, T. 1995. Accumulation of amino acids in forest plants in relation to ecological amplitude and nitrogen supply. Functional Ecology 9, 596-605.

Ohlson, K. E. A. & Wallmark, P.H. 1999. Novel calibration with correction for drift and non-linear response for continuous flow isotope ratio mass

36

spectrometry applied to the determination of ô1 5N, total nitrogen, du C and total carbon in biological material. Analyst 124, 571-577.

0kland, R. H. 1995. Changes in occurrence and abundance of plant species in a Norwegian coniferous forest, 1988-1993. Nordic Journal o f Botany 15, 415-438.

Paul, N. D. 1990. Modification of the effects of plant pathogens by other components of natural ecosystems. In Pests, Pathogens and Plant Communities, (eds Burdon, J. J. & Leather, S. R.) Blackwell Scientific Publications, Oxford, pp 81-96.

Persson, H. 1981. The effect of fertilization and irrigation on the vegetation dynamics of a pine-heath ecosystem. Vegetatio 46, 181-192.

Petchey, O. L., McPherson, P. T., Casey, T. M. & Morin, P. J. 1999. Environmental warming alters food-web structure and ecosystem function. Nature 402, 69-72.

Peter, M., Ayer, F. & Egli, S. 2001. Nitrogen addition in a Norway spruce stand altered macromycete sporocarp production and belowground ectomycorrhizal species composition. New Phytologist 149, 311-325.

Petersen, J. E., Chen, C. C. & Kemp, M. W. 1997. Scaling aquatic productivity: experiments under nutrient- and light-limited conditions. Ecology 78, 2326-2338.

Power, S. A., Ashmore, M. R., Cousins, D. A. & Ainsworth, N. 1995. Long term effects of enhanced nitrogen deposition on a lowland dry heath in southern Britain. Water, Air and Soil Pollution 85, 1701-1706.

Power, S. A., Ashmore, M. R. & Cousins, D. A. 1998. Impacts and fate of experimentally enhanced nitrogen deposition on British lowland heath. Environmental Pollution 102, suppl. 1, 27-34.

Quist, M.E., Näsholm, T., Lindeberg, J., Johannisson, C., Högbom, L. & Högberg, P. 1999. Responses of a nitrogen-saturated forest to a sharp decrease in nitrogen input. Journal o f Environmental Quality 28, 1970- 1977.

Read, D.J., Lewis, D. H., Fitter, A. H. & Alexander, I. J. (eds) 1992. Mycorrhizas in Ecosystems. University Press, Cambridge.

Redecker, D., Szaro, T. M., Bowman, R. J. & Bums, T. D. 2001. Small genets of Lactarius xanthgalactus, Russula cremoricolor and Amanita francheti in late-stage ectomycorrhizal successions. Molecular Ecology 10, 1025- 1034.

Redhead, S. A. and Perrin, P. W. 1972. Asterbolus: a new parasitichyphomycete with a novel dispersal mechanism. Canadian Journal o f Botany 50, 409-412.

Reinikainen, A., Mäkipää, R., Vanha-Majamaa, I. & Hotanen, J-P. 2000.Kasvit muuttuvassa metsäluonnossa. Kustannusosakeyhtiö Tammi, Jyväskylä. (Finnish with summary in English).

Schoeneweiss, D. F. 1975. Predisposition, stress, and plant disease. Annual Review o f Phytopathology 13, 193-211.

37

Sverdrup, H., Warfvinge, P. & Nihlgård, B. 1994. Assessment of soilacidification effects on forest growth in Sweden. Water, Air, and Soil Pollutions 78, 1-36.

Swift, M. J., Heal, O. W. & Anderson, J. M. 1979. Decomposition in Terrestrial Ecosystems. Blackwell Scientific publications, Oxford.

Tamm, C. O. 1991. Nitrogen in terrestrial ecosystems. Ecological Studies no 81. Springer-Verlag, Berlin.

Tarrasón, L., Olendrzynski, K., Posch, M., Stören, E., Bartnicki, J. & Vestreng, V. 2000. Transboundary acidification and eutrophication in the EMEP area in 1998. Transboundary acidification and eutrophication in Europe. (eds Tarrasón, L. & Schaug, J.), Norwegian Meteorological Institute. Research Report no. 101. EMEP summary report. Norway, pp. 9-40.

Tilman, D. 1988. Plant Strategies and the Structure and Dynamics o f Plant Communities. Princeton Univ. Press, Princeton.

Turkington, R., John, E., Krebs, C. J., Dale, M.R.T., Nams, V. O., Boonstra,R., Boutin, S., Martin, K., Sinclair, A. R. E. & Smith, J. N. M. 1998. The effects of NPK fertilization for nine years on boreal forest vegetation in northwestern Canada. Journal o f Vegetation Science 9, 333-346.

van Cleve, K. & Alexander, V. 1981. Nitrogen cycling in tundra and boreal ecosystems. In Terrestrial Nitrogen Cycles (eds. Clark, F. E. & Roswall, T.). Ecolocical Bulletins 33, 375-404.

van Cleve, K., Oliver, L. & Schlentner, R. 1983. Productivity and nutrient cycling in taiga forest ecosystems. Canadian Journal o f Forest Research 13, 747-766.

van der Krift, T. A. J. & Berendse, F. 2001. The effect of plant species on soil nitrogen mineralization. Journal o f Ecology 89, 555-561.

van der Heijden, M. G. A., Klironomos, J. N., Ursic, M., Moutoglis, P., Streitwolf-Engel, R., Boiler, T., Wiemken, A. & Sanders, I. R. 1998. Mycorrhizal fungal diversity determines plant biodiversity, ecosystem variability and productivity. Nature 396, 69-72.

Vinton, M. A. & Burke, I. C. 1995. Interactions between individual plant species and soil nutrient status in short grass steppe. Ecology 76, 1116- 1133.

Vitousek, P. M. 1982. Nutrient cycling and nutrient use efficiency. American Naturalist 119, 553-572.

Vitousek, P.M. and Howarth, R.W. 1991. Nitrogen limitation on land and in the sea: how can it occur? Biogeochemistry 13, 87-115.

Vitousek, P. M., Aber, J. D., Howarth, R. W., Likens, G. E., Matson, P. M., Schindler, S. W., Schlesinger, W. H. & Tilman, D. G. 1997. Human alteration of the global nitrogen cycle: sources and consequences. Ecological Applications 7, T il-750.

Wallander, H. 1994. A new hypothesis to explain allocation of dry matter between mycorrhizal fungi and pine seedlings in relation to nutrient supply. Plant and Soil 168-169,243-248.

38

Warring, G. L. & Cobb, N. J. 1992. The impact of plant stress on herbivore population dynamics. In Insect-plant interactions. voi 4.(ed. Bemays, L.). Academic Press, New York.

Wedin, D. A. & Tilman, D. 1990. Species effects on nitrogen cycling: a test with perennial grasses. Oecologia 84, 433-441.

White, T. C. S. 1993. The Inadequate Environment. Nitrogen and the Abundance o f Animals. Springer, Berlin, Heidelberg, New York.

Wiklund, K., Nilsson, L. O. & Jacobsson, S. 1995. Effect of irrigation,fertilization, and artificial drought on basidioma production in a Norway spruce stand. Canadian Journal o f Botany 73,200-208.

Wöllecke, J., Münzenberger, B. & Hüttl, R. F. 1999. Some effects of N on ectomycorrhizal diversity of Scots pine (Pinus sylvestris L.) in northeastern Germany. Water, Air, and Soil Pollution 116, 135-140.

Wright, R. F., Lotse, E. & Semb, A. 1993. RAIN project: Results after 8 years of experimentally reduced acid deposition to a whole catchment. Canadian Journal o f Fisheries and Aquatic Science 50, 258-268.

Wright, R. F., Brandrud, T-E., Clemensson-Lindell, A., Hultberg, H., Kjonaas, O. J., Moldan, F., Persson, H. & Stuanes, A. O. 1995. NITREX project: ecosystem response to chronic additions of nitrogen to a spruce-forested catchment at Gårdsjön, Sweden. In Effects o f acid deposition and troposheric ozon on forest ecosystems in Sweden (eds Staaf, H. & Tyler, G.). Ecological Bulletins (Copenhagen) 44, 322-334.

Xu, Y.-J., Blanck, K., Bredemeier, M. & Lamersdorf, N.P. 1998.Hydrochemical input-output budgets for a clean rain and drought experiment at Soiling. Forest Ecology and Management 101,295-306.

39

Tack

Efter fem års doktorandarbete är det naturligtvis många jag vill tacka. Till att böija med Lars Ericson som har varit min handledare under denna tid. Jag hoppas att din stora kunskap och ständiga nyfikenhet på allt som har med ekologi att göra har smittat av sig på mig. Om inte annat så har det varit mycket inspirerande att ha dig som handledare. Jag vill också tacka för att jag har fått arbeta självständigt och fått komma med egna idéer. Jag kommer att ha stor nytta av dessa erfarenheter i framtiden.

Jag vill också tacka mina medförfattare Torgny Näsholm, Annika Nordin och Mats Walheim för att ni bidragit med många nya tankar och infallsvinklar till avhandlingen. Torgny och Annika har dessutom varit mycket mer än bara medförfattare. Jag har verkligen uppskattat alla diskussioner vi haft under årens lopp, även de som inte handlat om själva arbetet. Ett speciellt tack vill jag rikta till Torgny som fatt stå ut med otaliga frågor från min sida och i praktiken fungerat som biträdande handledare. I detta sammanhang förtjänar också Naturvårdsverket ett tack för initiativet till samarbetet med Torgny och Annika.

Mitt doktorandarbete har till stor del varit finansierat av Mistras ASTA-projekt. Tack vare detta projekt har jag fatt chansen att bredda mina kunskaper även utanför mitt forskningsområde. Genom projektet har jag också fatt lära känna många nya och trevliga människor. Tack till er alla!

Utan hjälp från alla fältarbetare hade aldrig den här avhandlingen blivit färdig. Jag är skyldig er ett stort tack. Inte bara för att ni har utfört ett utmärkt arbete, utan också för er positiva inställning trots enformiga arbetsuppgifter och ibland uselt väder. Speciellt tack till Carina Bengtsson, Andreas Garpebring, Anders Hedefalk, Jens Ingerlund, Niklas Johansson och Hanna Oscarsson som stått ut med både mig och arbetet under flera säsonger. Jag har verkligen uppskattat att jobba med er. Tack också till Margareta Zetherström för trevligt bemötande och hjälp med alla kemiska analyser. Hans-Göran Nilsson och övrig personal vid Svartbergets försökspark har underlättat mitt fältarbete både praktiskt och socialt. Tack!

Goda kollegor kan man inte få för många av. Jag har verkligen haft många bra arbetskamrater under min tid som doktorand. Alla medarbetare vid Institutionen för ekologi och geovetenskap är värda ett stort tack. Många av er har hjälpt mig med olika saker som rört mitt arbete, inte minst T/A personalen. Kanske är det ändå viktigast att det varit trivsamt att umgås med er alla i samband med lunch, fika, innebandy, fester m m. Några av mina arbetskamrater har varit och är mycket mer än bara arbetskamrater. Henrik, min nygamla rumskompis, har jag haft mycket gemensamt med under de här åren. Vi påböijade doktorandutbildning samtidigt och kommer att disputera nästan samtidigt. Dessutom med samma handledare. Jag undrar vad vi skall hitta på tillsammans härnäst? Nie, tack för att du är en person som alltid far mig

40

på gott humör. Dina kunskaper i det engelska språket har också varit nyttiga. Mina många och långa diskussioner med Johan har nog påverkat en hel del av innehållet i den här avhandlingen, även om fotboll kanske fatt oproportionellt litet utrymme. Jag hoppas att det kan bli fler pratstunder i framtiden trots att du nu far ett eget rum. Mattias och jag har spenderat mycket tid tillsammans, vilket resulterat i en åtskilliga trevliga minnen. Inte minst under våra något udda semesterresor. Jag lär aldrig glömma en viss sommarnatt med träsksångaren vid Ängesbyn. Blåbärspajen kommer snart! Jag vill också tacka alla er som varit med mig på diverse doktorandresor, parsemesterar och inte minst oförglömliga veckor på Öland. Sådana resor har varit välbehövliga andningshål i vardagen.

Alla som känner mig vet att fagelskådning är bland det bästa jag vet. Det är konstigt att så fort man far en kikare runt halsen försvinner alla tankar på jobbet. Som tur är finns det många likasinnade vid och utanför institutionen, vilket inneburit att det alltid är någon som vill ut på en liten runda. Tack till alla er som frestat och låtit er frestas. Jag vill också passa på att tacka alla i båtgänget. Köpet av vår båt har ökat mitt välbefinnande högst påtagligt. Det är alltid trevligt att komma ut på sjön eller bara få åka ut och vända på båten tillsammans med er.

Jag vill också tacka mina föräldrar och min syster för att ni alltid ställt upp och stöttat mig i allt jag gjort. Under de här åren har ni har säkert ibland undrat vad det skulle bli av det hela. Det nu i alla fall blivit en liten ”bok”. Ett extra tack till min pappa för att han hjälpt mig med att konstruera ramar till punktfrekvensanalyserna. Sist men kanske mest vill jag tacka Hanna. Du känner redan till det mesta i den här avhandlingen utan att behöva läsa den. Tack för allt du hjälpt mig med och för att du orkat lyssna. Viktigast har ändå varit ditt kärleksfulla stöd och den glädje som du ständigt ger mig.

41