Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / The vegetation to be employed in techniques proposed by the natural engineering or soil bioengineering needs to fill out some requisites related to ecological, phytossociological and of reproduction aspects. The techniques in natural engineering use the plants in different angles in relation to the land. For this reason, the present work aims to evaluate the effects of different angles of stem cuttings planting of Phyllanthus sellowianus Müll. Arg., Salix humboldtiana Willd. and Sebastiania schottiana (Müll. Arg.) Müll. Arg., in order to determine the most suitable form to the planting of these species. The experiment was conducted in the Laboratory of Silviculture, in the Federal University of Santa Maria (UFSM), Santa Maria, Rio Grande do Sul, in the period of August to November of 2007. The treatments consisted in the three species of stem cuttings planting, in three different angles of planting in relation to the land (10°, 30° and 90°). The stem cuttings with 30 cm of length, with variable diameter from 1.0 to 4.5 cm, were collected in the central region of the state of the Rio Grande do Sul, in the final of August month. Two thirds of the stem cutting s bases were inserted in inert substrate (sand). The experimental design was completely randomized, with 30 repetitions for treatment. After 30 and 60 days, the following parameters were observed: the number of sprouted stem cuttings, number and length of the sprouts, and after 90 days, besides these parameters, were determined the survival, the length, the diameter, and the dry mass of the shoots, and also the length, the diameter, the dry mass and the volume of the roots. The data were statistically analyzed through the statistical programs and Excel. Sebastiania schottiana did not present sprouting, not even rooting, which perhaps could be explained by the stem cuttings time of collection, where that should be tested in other periods of year. The planting angle showed influence on some variables for Phyllanthus sellowianus and Salix humboldtiana. Both the species presented high values of survival, with a tendency for major results for stem cuttings planted in angle of 10° and 90° for Phyllanthus sellowianus (100%) and in angle of 30° for Salix humboldtiana (87%). Significant correlations were observed between the variables of the air part and of the radicial system and the area of transverse section of the stem cuttings, for the different planting angles. Both the two species presented the majority of the roots located on the basis of the stem cuttings, except for the ones planted in angle of 30°, which had the roots distributed along the whole buried portion. The realization of new studies is suggested with other species and also with another angles of planting, as well as the use of bigger diametric breadth. / A vegetação a ser empregada em técnicas propostas pela engenharia natural ou bioengenharia de solos necessita preencher alguns requisitos relacionados a aspectos ecológicos, fitossociológicos e de reprodução. As técnicas de engenharia natural utilizam as plantas em diferentes ângulos em relação ao terreno. Por esta razão, o presente trabalho tem por objetivo avaliar os efeitos de diferentes ângulos de plantio de estacas de Phyllanthus sellowianus Müll. Arg., Salix humboldtiana Willd. e Sebastiania schottiana (Müll. Arg.) Müll. Arg., a fim de determinar a forma mais adequada para o plantio destas espécies. O experimento foi conduzido no Laboratório de Silvicultura, na Universidade Federal de Santa Maria (UFSM), Santa Maria, Rio Grande do Sul, no período de agosto a novembro de 2007. Os tratamentos consistiram no plantio de estacas das três espécies, em três ângulos diferentes de plantio em relação ao terreno (10°, 30° e 90°). As estacas com 30 cm de comprimento, com diâmetro variável de 1,0 a 4,5 cm, foram coletadas no final de agosto, na região central do estado do Rio Grande do Sul. Dois terços da base das estacas foram inseridos em substrato inerte (areia). O delineamento experimental foi o inteiramente casualizado, com 30 repetições por tratamento. Após 30 e 60 dias, foram observados os seguintes parâmetros: número de estacas brotadas, número e comprimento dos brotos, e, aos 90 dias, além destes parâmetros, foram determinados a sobrevivência, o comprimento, o diâmetro, a massa seca dos brotos e o comprimento, a massa seca e o volume das raízes. Os dados foram analisados estatisticamente através de programas estatísticos e Excel. Sebastiania schottiana não apresentou brotação, nem enraizamento, o que talvez possa ser explicado pela época de coleta das estacas, devendo ser testada em outras épocas do ano. O ângulo de plantio mostrou influência sobre algumas variáveis para Phyllanthus sellowianus e Salix humboldtiana. Ambas as espécies apresentaram altos valores de sobrevivência, com tendência de maiores resultados para estacas plantadas em ângulo de 10° e 90° para Phyllanthus sellowianus (100%) e em ângulo de 30° para Salix humboldtiana (87%). Observaram-se correlações significativas entre as variáveis da parte aérea e do sistema radicial e a área da seção transversal das estacas, para os diferentes ângulos de plantio. As duas espécies apresentaram a maioria das raízes localizadas na base das estacas, exceto para as estacas plantadas em ângulo de 30°, as quais tiveram as raízes distribuídas ao longo de toda a porção enterrada. Sugere-se a realização de novos estudos com outras espécies e também com outros ângulos de plantio, assim como a utilização de maior amplitude diamétrica.

Ziel des Forschungsprojektes war der Gewinn umfassender verjüngungsökologischer Kenntnisse zu den Pionierbaumarten Sandbirke (Betula pendula Roth), Salweide (Salix caprea L.) und Eberesche (Sorbus aucuparia L.) im Hinblick auf eine natürliche, eingriffsfreie Wiederbewaldung von Schadflächen (z.B. nach Sturmwurf). Die Abschätzung des Besiedlungserfolges von Schadflächen durch Pionierbaumarten ist aufgrund unzureichender verjüngungsökologischer Kenntnisse gegenwärtig noch mit großen Unsicherheiten verbunden. Die Untersuchungen fanden auf fünf 4-12 ha großen Kyrill-Sturmwurfflächen in den Hoch- und Kammlagen (750-900 m ü. NN) des Thüringer Waldes statt. Alle potenziellen Samenbäume der Pionierbaumarten wurden in den angrenzenden, geschlossenen Fichtenbeständen lokalisiert. Als Versuchsdesign wurde in Abhängigkeit der vorgefundenen Samenbaumdichten und -verteilungen ein Kreuz- bzw. Sterntransekt auf den Sturmwurfflächen etabliert. Entlang der Transektlinien wurden alle 20 m Samenfallen installiert. Als Samenfallen kamen für die Sandbirke Netztrichterfallen (0,2 m²), für die Salweide Klebfallen (0,043 m²) und für die endozoochore Ausbreitung durch frugivore Vogelar-ten Kotfallen (0,25 m²) zum Einsatz. Für die Modellierung der Samenausbreitung von Sandbirke und Salweide wurden inverse Modelle bzw. geostatistische Modelle erstellt. Zudem wurden auf einer der Sturmwurfflächen genetische Nachkommenschaftsanalysen bei Sal-weide mittels Kern-DNA-Primer durchgeführt. Die Bodensamenbankuntersuchungen fanden in jeweils drei geschlossenen Birkenbeständen, Fichten-Birken-Beständen, Fichtenbeständen mit einer einzeln eingemischten Birke und reinen Fichtenbeständen im Tharandter Wald und Thüringer Wald statt. Mittels eines 10,2 cm breiten Stechzylinders wurden 10 cm tiefe Bodenproben gewonnen. Die Lagerung der Proben fand im Kaltgewächshaus statt, wo alle 14 d die gekeimten Samen erfasst wurden. Weiterhin wurde ein Eingrabungsexperiment installiert. Dafür wurden Sandbirkensamen, Ebereschensamen und Ebereschenfrüchte in 2 cm, 5 cm und 10 cm tiefen Mineralboden vergraben und in sechsmonatigen Intervallen jeweils eine Keimprobe zum Test der verbliebenen Keimfähigkeit entnommen. Die Auswertung der Bodensamenbankuntersuchungen erfolgte mittels GLM und GLMM. Während der zweijährigen Untersuchung zur zeitlichen und räumlichen Samenausbreitung von Salix caprea auf fünf Sturmwurfflächen konnten ein schwächeres und ein stärkeres Samenjahr nachgewiesen werden. Des Weiteren erstreckte sich der Samenflugzeitraum im Frühjahr in Abhängigkeit von den klimatischen Bedingungen über 12 Wochen in 2015 und 6 Wochen in 2016. Die höchsten Samenmengen von 23-156 Samen je Falle wurden jeweils unter den Kronen von Salweiden-Samenbäumen nachgewiesen. Ab einer Entfernung von 350 m zum Samenbaum bis zur untersuchten Distanz von 870 m wurden, unabhängig von der Distanz, der Hangneigung, der Anzahl der Samen-bäume und der Windrichtung (Anisotropie), im Durchschnitt 0,6-2,1 Samen je Falle er-fasst. Die genetischen Analysen zur Nachkommenschaft ergaben, dass 29 % der untersuch-ten Verjüngungspflanzen von einem der 20 lokalisierten Elternbäume in der bewaldeten, 500 m breiten Suchzone abstammten. Die Ausbreitungsdistanzen der nachweislich am erfolgreichsten verjüngten Samenbäume betrugen dabei 550-800 m. Insgesamt zeigte die Salweidenverjüngung eine höhere Allel-Variation, als die 20 Elternbäume, was auf einen externen Genfluss und lange Samen- und Pollenausbreitungsdistanzen hinweist. Im Zuge des zweijährigen Untersuchungszeitraums zur Samenausbreitung von Betula pendula auf zwei Kyrill-Sturmwurfflächen konnten ein Mastjahr und ein Zwischenjahr nachgewiesen werden. Die Ergebnisse der inversen Modellierung mittels isotroper Mo-delle ergaben dabei flächenunabhängig Produktionsmengen für einen Samenbaum von 20 cm im Bhd von 300.000-366.000 Samen je Baum im Zwi-schenjahr 2015 und 1.430.000-1.530.000 Samen je Baum im Mastjahr 2016. Mittels räumlicher Modellierung der Samenausbreitung konnte keine Anisotropie belegt werden. Unabhängig von den beprobten Flächen und Un-tersuchungsjahren, belegen die Modellschätzungen allesamt eine isotrope Ausbreitung der Samen. Die mittleren Ausbreitungsdistanzen (MDD) beliefen sich dabei hangaufwärts auf 86-97 m und hangabwärts auf 367-380 m. Maximal ab-gelagerte Samendichten von 2.015 n m-² im Zwischenjahr und 9.557 n m-² im Mastjahr fanden sich bis 40-50 m Entfernung zum Samenbaum. Die Untersuchungen der endozoochoren Samenausbreitung auf fünf Sturmwurfflächen weisen auf eine bevorzuge Nutzung der Vogelarten von Rast- und Sitzgelegenheiten (Strukturelemente) auf Freiflächen zum Absetzen von Kot hin (2,7 Kothaufen je m²). Unter Freiflächenbedingungen - ohne Strukturelemente - ergaben sich im Mittel 0,4 Kothaufen je m². Die höchsten mittleren Kotdichten wurden unter aufra-genden Totästen (20 n m-²), umgeklappten Wurzeltellern (4,6 n m-²) und Hochstubben (3,9 n m-²) nachgewiesen. Schwach dimensionierte Verjüngungspflanzen der Sandbirke, Eberesche und Fichte, und Strukturelemente unter einem Meter Höhe wurden dagegen weitgehend für das Absetzen von Kot gemieden. Das Vermögen zum Aufbau einer Bodensamenbank durch Betula pendula und Sorbus aucuparia unterschied sich deutlich. 56-100 % der eingegrabenen Sandbirkensamen wa-ren auch nach 2,5 Jahren keimfähig, wohingegen lediglich 3-16 % der eingegrabenen Ebereschensamen ohne Fruchthülle und 0-19 % der eingegrabenen Ebereschensamen mit Fruchthülle vital waren. Die Auswertung mittels GLM prognostizierte einen kom-pletten Verlust der Keimfähigkeit nach 12 Jahren, 4,5 Jahren und 3 Jahren der Sandbirkensamen, sowie der Ebereschensamen mit und ohne Fruchthülle. Ein Einfluss der Lagerungstiefe war nur für Sandbirke nachweisbar. Die Untersuchungen der Bodensamenbank von Birke in Fichtenbeständen mit unter-schiedlichen Birkensamenbaumanteilen ergab einen straffen Zusammenhang zwischen der Anzahl von Samenquellen und den nachgewiesenen Samendichten im Boden. In den Birkenbeständen fanden sich stets die höchsten Dichten von 489-1.142 Birkensamen je m². Die Analyse unterschiedlicher Bodenschichten zeigte zudem signifikant abneh-mende Birkensamendichten mit zunehmender Bodentiefe. Die Ergebnisse der Untersuchungen zeigen, dass die Fruktifikation von Betula pendula, Salix caprea und Sorbus aucuparia durch klimatische Verhältnisse beeinflusst wird, weshalb die drei Pionierbaumarten nicht alljährlich hohe Samenmengen produzieren (Mastjahre und Zwischenjahre). Zum Ausgleich von Produktionsdefiziten in den Zwischenjahren unter-scheiden sich die Pionierbaumarten in ihrer Strategie. Dies gilt es bei der Umsetzung des Konzeptes einer natürlichen, eingriffsfreien Wiederbewaldung von Schadflächen nach Sturmwurf durch die Naturverjüngung von Pionierbaumarten zu beachten. Die einzige der drei Pionierbaumarten, die dem allgemeinen Bild einer Pionierbaumart ent-spricht, ist die Salweide. Ihr Besiedlungserfolg ist allein von den aktuellen, alljährlich variierenden, aber dennoch stets hohen Samenproduktionsmengen und den enorm weiten Aus-breitungsdistanzen (>800 m) abhängig. Hinsichtlich der Samenausbreitung haben die Him-melsrichtung, die Position der Samenbäume und die Anzahl vorhandener Samenquellen ab einer Distanz von 50 m zur Schadfläche keinen bedeutenden Einfluss auf die abgelagerten Samenmengen mehr. Die auf 86-380 m limitierte Samenausbreitung von Sandbirke wurde dagegen stark vom Geländerelief (Hangneigung), der Position der Samenbäume (Tal, Kuppe, Hanglage) und der Distanz der Samenbäume zur Sturmwurffläche beeinflusst. Zum Ausgleich der limitierten Samenausbreitung und deutlich reduzierten Samenmengen im Zwischenjahr ist Sandbirke jedoch zum Aufbau einer short-term persistenten Bodensamenbank befähigt. Den gesamten Verjüngungszyklus betrachtend entspricht die Eberesche eher einer Schluss-waldbaumart. Unter ungünstigen klimatischen Bedingungen kommt es häufig zum kompletten Ausfall der Samenproduktion. Ihr enormes Wiederbewaldungspotential von Sturmwurfflächen speist sich hauptsächlich aus dem Aufbau einer Sämlingsbank und weni-ger durch den aktuellen Samenregen oder der short-term persistenten Bodensamenbank. Die limitierte Samenausbreitung von Sandbirke und Eberesche macht eine „räumliche Optimierung“ der Samenbaumpositionen durch die Forstwirtschaft erforderlich. Aufgrund der allgegenwärtigen Omnipräsenz von Weidensamen ist dies für Salweide nicht zwingend not-wendig. Das detailreiche Wissen zur Verjüngungsökologie der untersuchten Pionierbaumar-ten ermöglicht eine gezielte waldbauliche Steuerung im Sinne des Vorhalts und der Pflege von Pionierbaumarten im Wirtschaftswald. Dies ist gegenwärtig und zukünftig vor allem von besonderer Bedeutung, da aufgrund der zu erwartenden Zunahme von Schadereignissen und deren Unvorhersehbarkeit die Fähigkeit der Wälder zur natürlichen Wiederbewaldung von Schadflächen durch Pionierbaumarten zunehmend an Interesse gewinnen wird.:Table of abbreviations III Summary IV Zusammenfassung VIII Chapter 1 – General introduction 1 1.1 Introduction 2 1.1.1 Importance and relevance of the study 2 1.1.2 Research interest - regeneration ecology 5 1.3 Aims, scope and hypotheses 9 1.4 Study outline 12 1.5 References 13 Chapter 2 – Seed dispersal capacity of Salix caprea L. assessed by seed trapping and parentage analysis 25 2.1 Abstract 26 2.2 Introduction 26 2.3 Materials and methods 29 2.3.1 Study area 29 2.3.2 Experimental design 31 2.3.3 Genetic analysis 32 2.3.4 Seed trap data analysis 33 2.3.5 Geostatistical models 34 2.4 Results 36 2.4.1 Temporal patterns of seed dispersal 37 2.4.2 Dispersal distance and spatial patterns of seed dispersal 37 2.4.3 Genetic parentage analysis 40 2.5 Discussion 42 2.5.1 Seed production and temporal patterns of seed dispersal 42 2.5.2 Dispersal distance and spatial patterns of seed dispersal 43 2.5.3 Genetic parentage analysis 45 2.6 Conclusions for silvicultural practice 46 2.7 References 48 Chapter 3 – Restrictions on natural regeneration of storm-felled spruce sites by silver birch (Betula pendula Roth) through limita-tions in fructification and seed dispersal 57 3.1 Abstract 58 3.2 Introduction 58 3.3 Materials and methods 60 3.3.1 Study area 60 3.3.2 Experimental design 63 3.3.3 Data analysis 64 3.3.4 Seed dispersal model 64 3.3.5 Simulations for practical management decisions 66 3.4 Results 66 3.4.1 Seed production 66 3.4.2 Seed dispersal and spatial patterns 68 3.5 Discussion 71 3.5.1 Seed production 71 3.5.2 Directionality 72 3.5.3 Spatial patterns and seed dispersal distances 72 3.6 Seed dispersal scenarios for silvicultural management decisions 74 3.6.1 Seed dispersal scenarios 74 3.6.2 Conclusions for silvicultural management decisions 76 3.7 References 78 Chapter 4 – The impact of structural elements on storm-felled sites on endozoochorous seed dispersal by birds – a case study 85 4.1 Abstract 86 4.2 Introduction 86 4.3 Materials and methods 88 4.3.1 Study area 88 4.3.2 Experimental design 90 4.3.3 Data analysis 91 4.4 Results 92 4.5 Discussion 95 4.6 Conclusions for silvicultural practice 97 4.7 References 98 Chapter 5 – Soil seed banks of pioneer tree species in European tempe-rate forests: a review 104 5.1 Abstract 105 5.2 Introduction 105 5.3 Methods of literature search 107 5.4 Species-specific reproductive ecology determining the potential of soil seed banks 110 5.5 Characterization and classification of soil seed banks 112 5.5.1 Soil seed bank of Betula spp. 114 5.5.2 Soil seed bank of Alnus glutinosa (L.) GAERTN. 116 5.5.3 Soil seed banks of Salix spp. and Populus tremula L. 117 5.5.4 Soil seed bank of Sorbus aucuparia L. 118 5.6 Conclusions 119 5.7 References 120 Chapter 6 – Do birch and rowan establish soil seed banks sufficient to compensate for a lack of seed rain after forest disturbance? 134 6.1 Abstract 135 6.2 Introduction 135 6.3 Materials and methods 137 6.3.1 Study areas 137 6.3.2 Data collection 139 6.3.3. Statistical analysis 140 6.4 Results 141 6.4.1 Study A - Artificial seed burial experiment 141 6.4.2 Study B - Soil core sampling in the forest 145 6.5 Discussion 147 6.5.1 Study A - Artificial seed burial experiment 147 6.5.2 Study B - Soil core sampling in the forest 149 6.6 Conclusions 151 6.7 References 152 Chapter 7 – General discussion 160 7.1 Discussion of important aspects of regeneration ability 161 7.1.1 Fructification and seed production in Salix caprea, Betula pendula and Sorbus aucuparia 165 7.1.2 Ecological processes within the regeneration cycle of Salix caprea 167 7.1.3 Ecological processes within the regeneration cycle of Betula pendula 168 7.1.4 Ecological processes within the regeneration cycle of Sorbus aucuparia 169 7.2. Conclusions for silviculture and management recommendations 172 7.3 References 175 Table of appendix i / The aim of the study was to obtain comprehensive knowledge of the regeneration ecology of the pioneer tree species silver birch (Betula pendula Roth), goat willow (Salix caprea L.) and rowan (Sorbus aucuparia L.). The findings should contribute to better management of the natural regeneration of disturbed sites (e.g., windthrown sites) by pioneer tree species. Insufficient knowledge of the regeneration ecology of pioneer tree species renders forest managers’ abilities to assess the success of regeneration of windthrown sites uncertain. The study took place in the years 2015 and 2016. The study sites were located on the slopes and mountain tops (plateaus) of the Thuringian Forest (715-900 m a.s.l.), on five windthrown open areas (4-13 ha) created by the storm ‘Kyrill’ in January 2007. All seed trees of pioneer tree species were mapped within the forested search zone around each site. This zone extended 200 m for silver birch and rowan and 500 m for goat willow. Following the mapping of these seed trees and an analysis of their spatial distribution, seed traps were placed along two or four crossing line transects, with intervals of 20 m between traps. The traps were funnel shaped net seed traps for silver birch (0.2 m²), seed traps with a sticky, non-drying glue for goat willow (0.043 m²) and dropping traps for seeds dispersed endozoochorously by frugivorous birds (0.25 m²). A phenomenological model and model-based geostatistics were used to investigate silver birch and goat willow seed dispersal. For goat willow a parentage analysis was performed at one of the study sites using nuclear-DNA-primers. The soil seed bank study was carried out in three birch stands, spruce stands with admixed birch, spruce stands with one isolated birch tree and pure spruce stands in the Tharandter Forest and in the Thuringian Forest. Soil core samples with a diameter of 10.2 cm were taken from the litter layer and the mineral soil to a depth of 10 cm. The soil samples were placed in a greenhouse and seed germination was checked every 14 days. An artificial seed burial experiment was also carried out. Silver birch seeds, rowan seeds and rowan fruits were buried in mineral soil at depths of 2 cm, 5 cm and 10 cm. At intervals of 6 months sample sets were removed from the soil and the germination capacity checked. The analysis of the soil seed banks was based on generalized linear mixed models (GLMM) and generalized linear models (GLM). The 2-year study of the temporal and spatial dispersal of seeds of Salix caprea on five Kyrill-felled areas involved one year with lower seed production and one with more bountiful seed production. The duration of the spring seed rain was about 12 weeks in 2015, and only 6 weeks in 2016 because of contrasting weather conditions. The highest seed numbers of 23-156 n per trap occurred close to the base of the seed trees. Beyond 350 m from the seed trees, up to the maximum distance in the study of 870 m, the average numbers of seeds per trap (0.6-2.1 seeds) were independent of the dispersal distance, inclination, the number of seed sources and the dispersal direction (anisotropy). Parentage analyses showed that 29% of the saplings stemmed from one of the 20 parent trees within the 500 m search zone extending from the edge of the open area. The seed dispersal distances of the most successful seed parents were between 550-800 m. The saplings revealed a higher allelic variation than the 20 parent trees, indicating external gene flow and long seed and pollen dispersal distances. During the 2-year study of Betula pendula seed dispersal on two Kyrill-felled areas there was a mast year and a non-mast year. Independent of the site, the seed production rate of a silver birch seed tree with a mean diameter at breast height (dbh) of 20 cm predicted by isotropic inverse models was approximately 300,000-366,000 seeds in 2015 and 1,430,000-1,530,000 seeds per tree in the mast year 2016. Directionality (anisotropic inverse modelling) of seed dispersal around an individual seed tree could not be confirmed. The model results revealed the isotropic model (no directionality) to be an appropriate approach for all sites and years. The mean dispersal distances (MDD) were 86 m and 97 m (uphill) and 367 m and 380 m (downhill). The maximum seed numbers occurred within 40-50 m of a seed tree, amounting to 2,015 n m-² in the non-mast year and 9,557 n m-² in the mast year. The study of endozoochorous seed dispersal on the five sites felled by the storm Kyrill showed a preference of frugivorous birds for perches and resting places (structural elements) from which to defecate onto open areas (2.7 droppings per m²). On completely open areas – with no structural elements – an average of 0.4 droppings per m² was recorded. The highest mean bird dropping density was observed under towering dead branches (20 n m-²), upturned root plates (4.6 n m-²) and high stumps (3.9 n m-²). Young, small diameter silver birch, rowan and spruce trees, and structural elements less than 1 m in height generally, were avoided by frugivorous birds as a place from which to defecate. The abilities of Betula pendula and Sorbus aucuparia to form a soil seed bank differed. Between 56-100 % of the buried silver birch seeds were still viable after 2.5 years, whereas only 3-16 % of the rowan seeds buried without pulp and 0-19 % of the rowan seeds within pulp were viable. The maximum durations of storage in the soil predicted for silver birch seeds and rowan seeds with and without pulp by GLM were 12 years, 4.5 years and 3 years. An influence of the storage depth was found for silver birch seeds only. The investigation of the soil seed banks of birch in three birch stands and nine spruce forests with different numbers of admixed birch seed trees showed a strong correlation between the number of seed sources and the seed density in the soil. The birch stands contained the highest mean densities of viable birch seeds in soil, between 489-1,142 n m-². The analysis of the different soil layers showed significantly declining birch seed densities with increasing soil depth across all sites. The results of the study showed that the fructification of Betula pendula, Salix caprea and Sorbus aucuparia is influenced by weather conditions, with the three pioneer tree species failing to produce high numbers of seeds every year (mast and non-mast years). The three species differed in their strategies to compensate for low seed production in non-mast years. This must be considered when implementing a concept for the reforestation of disturbed sites based on natural regeneration by pioneer tree species. Goat willow was the only one of the three specie studied with characteristics corresponding to the general assumptions made about pioneer tree species. The regeneration success of goat willow is dependent upon the variable but generally high annual seed production and long seed dispersal distances (> 800 m). The azimuth direction, position and number of seed trees have no meaningful influence on seed numbers at a distance of more than 50 m from the seed source. The limited mean seed dispersal distances of 86-380 m determined for silver birch were influenced by site inclination, the seed tree location (valley, slope or plateau) and the distance between the seed tree and the windthrown site. Silver birch seed shadow is also influenced by the number of seed sources. To compensate for the limited dispersal distances and the significantly lower seed production in non-mast years, silver birch is able to build up a short-term persistent soil seed bank. The regeneration cycle of rowan is more reminiscent of that of a shade-tolerant tree species. Unfavorable weather conditions often result in a complete failure to produce seeds. The enormous regeneration potential of rowan on disturbed sites stems primarily from a seedling bank, which is built up over years. The seed rain in any given year and its short-term persistent soil seed bank are of secondary importance. Forest management targeting a ‘spatial optimization’ of silver birch and rowan seed trees is necessary to ensure successful natural regeneration given the limited seed dispersal. The omnipresence of goat willow seeds renders specific spatial management measures for its establishment unnecessary. Detailed knowledge of the regeneration ecology of the studied pioneer tree species makes possible an approach to silviculture that is targeted to the conservation and revitalization of pioneer tree species in managed forests. The expected increase in the frequency of disturbances, and their unpredictability, means that the ability of forests to naturally regenerate using pioneer tree species is likely to grow in importance.:Table of abbreviations III Summary IV Zusammenfassung VIII Chapter 1 – General introduction 1 1.1 Introduction 2 1.1.1 Importance and relevance of the study 2 1.1.2 Research interest - regeneration ecology 5 1.3 Aims, scope and hypotheses 9 1.4 Study outline 12 1.5 References 13 Chapter 2 – Seed dispersal capacity of Salix caprea L. assessed by seed trapping and parentage analysis 25 2.1 Abstract 26 2.2 Introduction 26 2.3 Materials and methods 29 2.3.1 Study area 29 2.3.2 Experimental design 31 2.3.3 Genetic analysis 32 2.3.4 Seed trap data analysis 33 2.3.5 Geostatistical models 34 2.4 Results 36 2.4.1 Temporal patterns of seed dispersal 37 2.4.2 Dispersal distance and spatial patterns of seed dispersal 37 2.4.3 Genetic parentage analysis 40 2.5 Discussion 42 2.5.1 Seed production and temporal patterns of seed dispersal 42 2.5.2 Dispersal distance and spatial patterns of seed dispersal 43 2.5.3 Genetic parentage analysis 45 2.6 Conclusions for silvicultural practice 46 2.7 References 48 Chapter 3 – Restrictions on natural regeneration of storm-felled spruce sites by silver birch (Betula pendula Roth) through limita-tions in fructification and seed dispersal 57 3.1 Abstract 58 3.2 Introduction 58 3.3 Materials and methods 60 3.3.1 Study area 60 3.3.2 Experimental design 63 3.3.3 Data analysis 64 3.3.4 Seed dispersal model 64 3.3.5 Simulations for practical management decisions 66 3.4 Results 66 3.4.1 Seed production 66 3.4.2 Seed dispersal and spatial patterns 68 3.5 Discussion 71 3.5.1 Seed production 71 3.5.2 Directionality 72 3.5.3 Spatial patterns and seed dispersal distances 72 3.6 Seed dispersal scenarios for silvicultural management decisions 74 3.6.1 Seed dispersal scenarios 74 3.6.2 Conclusions for silvicultural management decisions 76 3.7 References 78 Chapter 4 – The impact of structural elements on storm-felled sites on endozoochorous seed dispersal by birds – a case study 85 4.1 Abstract 86 4.2 Introduction 86 4.3 Materials and methods 88 4.3.1 Study area 88 4.3.2 Experimental design 90 4.3.3 Data analysis 91 4.4 Results 92 4.5 Discussion 95 4.6 Conclusions for silvicultural practice 97 4.7 References 98 Chapter 5 – Soil seed banks of pioneer tree species in European tempe-rate forests: a review 104 5.1 Abstract 105 5.2 Introduction 105 5.3 Methods of literature search 107 5.4 Species-specific reproductive ecology determining the potential of soil seed banks 110 5.5 Characterization and classification of soil seed banks 112 5.5.1 Soil seed bank of Betula spp. 114 5.5.2 Soil seed bank of Alnus glutinosa (L.) GAERTN. 116 5.5.3 Soil seed banks of Salix spp. and Populus tremula L. 117 5.5.4 Soil seed bank of Sorbus aucuparia L. 118 5.6 Conclusions 119 5.7 References 120 Chapter 6 – Do birch and rowan establish soil seed banks sufficient to compensate for a lack of seed rain after forest disturbance? 134 6.1 Abstract 135 6.2 Introduction 135 6.3 Materials and methods 137 6.3.1 Study areas 137 6.3.2 Data collection 139 6.3.3. Statistical analysis 140 6.4 Results 141 6.4.1 Study A - Artificial seed burial experiment 141 6.4.2 Study B - Soil core sampling in the forest 145 6.5 Discussion 147 6.5.1 Study A - Artificial seed burial experiment 147 6.5.2 Study B - Soil core sampling in the forest 149 6.6 Conclusions 151 6.7 References 152 Chapter 7 – General discussion 160 7.1 Discussion of important aspects of regeneration ability 161 7.1.1 Fructification and seed production in Salix caprea, Betula pendula and Sorbus aucuparia 165 7.1.2 Ecological processes within the regeneration cycle of Salix caprea 167 7.1.3 Ecological processes within the regeneration cycle of Betula pendula 168 7.1.4 Ecological processes within the regeneration cycle of Sorbus aucuparia 169 7.2. Conclusions for silviculture and management recommendations 172 7.3 References 175 Table of appendix i

La culture de saules (Salix sp.) est une pratique courante en Europe et en Amérique du Nord pour produire de la biomasse végétale. Cependant, le développement d’outils moléculaires est très récent. De plus, la phylogénie des saules est incomplète. Il y a un manque d’information pour les programmes de sélection d'espèces indigènes et pour la compréhension de l’évolution du genre. Le genre Salix inclut 500 espèces réparties principalement dans les régions tempérées et boréo-arctique de l’hémisphère nord. Nous avons obtenu l’ensemble des espèces retrouvées naturellement en Amérique (121 indigènes et introduites). Dans un premier temps, nous avons développé de nouveaux outils moléculaires et méthodes : extraction d’ADN, marqueurs microsatellites et gènes nucléaires. Puis, nous avons séquencé deux gènes chloroplastiques (matK et rbcL) et la région ITS. Les analyses phylogénétiques ont été réalisées selon trois approches : parcimonie, maximum de vraisemblance et Bayésienne. L’arbre d’espèces obtenu a un fort support et divise le genre Salix en deux sous-genres, Salix et Vetrix. Seize espèces ont une position ambiguë. La diversité génétique du sous-genre Vetrix est plus faible. Une phylogénie moléculaire complète a été établie pour les espèces américaines. D’autres analyses et marqueurs sont nécessaires pour déterminer les relations phylogénétiques entre certaines espèces. Nous affirmons que le genre Salix est divisé en deux clades. / Fast growing willows (Salix sp.) are increasingly used in Europe and North America for biomass production and other environmental applications. However, the development of molecular tools is recent. The phylogeny of willows is incomplete, which slows down the selection of suitable native species and the development of improvement programs. The genus Salix includes approximately 500 species worldwide, and these are mainly located in temperate and cold regions of the Northern Hemisphere. We gathered leaf material from all 121 willows of North America (species native and introduced). We developed three molecular tools-methods: DNA extraction, SSR markers, and nuclear genes. We sequenced two chloroplast genes matK and rbcL and the ITS region. Phylogenetic analyses were carried out using parsimony, maximum likelihood and Bayesian approaches. The species tree provides strong support for a division of the genus into two subgenera, Salix and Vetrix. Sixteen species have ambiguous positions. A complete molecular phylogeny of American willows has been established. It needs to be confirmed and further resolved using other molecular data. Nonetheless, the genus clearly has two clades.

In this study, extracts from bark, leaves and catkins of Salix aegyptiaca L. were investigated for their antioxidant content by 2,2-diphenyl-2-picrylhydrazyl hydrate (DPPH) free radical quenching assay, total phenolic and total flavonoid assays. The highest antioxidant activity (19 ug/ml IC50 for inhibition of DPPH radical activity), total phenolic content (212 mg gallic acid equivalents/g of dried extract) and total flavonoid (479 mg catechin equivalents/g of dried extract) was observed in the ethanolic extract of bark. High performance liquid chromatography (HPLC) analyses revealed the presence of gallic acid, caffeic acid, vanillin and p-coumaric acid, myricetin, catechin, epigallocatechin gallate, rutin, quercetin as well as salicin. In addition, the anti-proliferative effects of the ethanolic extracts on colorectal cancer cell lines (HCT-116 and HT-29) were examined by an MTT cell viability assay while their apoptotic effects were assayed by acridine orange staining and caspase 3 activity. The results indicate that the ethanolic extract of bark of S. aegyptiaca can strongly inhibit cell proliferation and induces apoptosis in a dose dependent manner on both cell lines. We propose that extracts from this plant may be utilized as a source of health promoting antioxidants. Our data provide a perspective for more detailed study of biochemical pathways associated with the cancer preventive effects of active components of the extracts from S. aegyptiaca.

La culture de saules (Salix sp.) est une pratique courante en Europe et en Amérique du Nord pour produire de la biomasse végétale. Cependant, le développement d’outils moléculaires est très récent. De plus, la phylogénie des saules est incomplète. Il y a un manque d’information pour les programmes de sélection d'espèces indigènes et pour la compréhension de l’évolution du genre. Le genre Salix inclut 500 espèces réparties principalement dans les régions tempérées et boréo-arctique de l’hémisphère nord. Nous avons obtenu l’ensemble des espèces retrouvées naturellement en Amérique (121 indigènes et introduites). Dans un premier temps, nous avons développé de nouveaux outils moléculaires et méthodes : extraction d’ADN, marqueurs microsatellites et gènes nucléaires. Puis, nous avons séquencé deux gènes chloroplastiques (matK et rbcL) et la région ITS. Les analyses phylogénétiques ont été réalisées selon trois approches : parcimonie, maximum de vraisemblance et Bayésienne. L’arbre d’espèces obtenu a un fort support et divise le genre Salix en deux sous-genres, Salix et Vetrix. Seize espèces ont une position ambiguë. La diversité génétique du sous-genre Vetrix est plus faible. Une phylogénie moléculaire complète a été établie pour les espèces américaines. D’autres analyses et marqueurs sont nécessaires pour déterminer les relations phylogénétiques entre certaines espèces. Nous affirmons que le genre Salix est divisé en deux clades. / Fast growing willows (Salix sp.) are increasingly used in Europe and North America for biomass production and other environmental applications. However, the development of molecular tools is recent. The phylogeny of willows is incomplete, which slows down the selection of suitable native species and the development of improvement programs. The genus Salix includes approximately 500 species worldwide, and these are mainly located in temperate and cold regions of the Northern Hemisphere. We gathered leaf material from all 121 willows of North America (species native and introduced). We developed three molecular tools-methods: DNA extraction, SSR markers, and nuclear genes. We sequenced two chloroplast genes matK and rbcL and the ITS region. Phylogenetic analyses were carried out using parsimony, maximum likelihood and Bayesian approaches. The species tree provides strong support for a division of the genus into two subgenera, Salix and Vetrix. Sixteen species have ambiguous positions. A complete molecular phylogeny of American willows has been established. It needs to be confirmed and further resolved using other molecular data. Nonetheless, the genus clearly has two clades.

Diss. (sammanfattning) Umeå : Sveriges lantbruksuniv., 2002. / Härtill 6 uppsatser.

Diss. (sammanfattning) Uppsala : Sveriges lantbruksuniversitet, 2005. / Härtill 5 uppsatser.

Diss. (sammanfattning) Uppsala : Sveriges lantbruksuniversitet, 2005. / Härtill 4 uppsatser.

Diss. (sammanfattning) Uppsala : Sveriges lantbruksuniversitet, 2006. / Härtill 5 uppsatser.

Älgförvaltningsområden har sedan 2012 varit en av de rumsliga skalor där älgförvaltningen i Sverige sker. I denna studie har Riksskogstaxeringens data över förekomst, täckningsgrad och betningsgrad av viltfoder per älgförvaltningsområde (2011–2015) analyserats tillsammans med Skogsstyrelsens prognos över foderproducerande ungskog (fodpro) och avskjutningsstatistik (avskjutningstäthet) från den nationella jaktdatabasen (Viltdata.se). Resultaten visar på både likheter och skillnader mellan undersökta älgförvaltningsområden (ÄFO). Björk var den viltfoderart med högst täckningsgrad i alla ÄFO samtidigt som den hade lägst betestryck. Rönn var den enda enskilda viltfoderart där täckningsgraden skiljde sig signifikant åt mellan några ÄFO, vilket tyder på att älgförvaltningsområdena är lika varandra i fråga om kvalitet av viltfoder. Trots signifikanta skillnader i den totala täckningsgraden av viltfoder mellan några av älgförvaltningsområdena tyder resultatet på att områdena även är lika varandra i fråga om kvantitet av viltfoder. Både avskjutningstätheten och några av betningsgraderna för tall skiljde sig signifikant åt mellan alla ÄFO. Det fanns svaga, positiva signifikanta samband mellan avskjutningstätheten och betningsgraden ”hård” för både tall och salix. Dessa samband skulle kunna bero på att en ökad älgtäthet genererat ett högre betestryck på tall och salix vilket i sin tur sannolikt medfört en högre avskjutning av älg. Det fanns inga signifikanta skillnader mellan Riksskogstaxeringens och Skogsstyrelsens skattningar av viltfoder. Metoderna för skattningarna skiljer sig dock åt i flera avseenden vilket gör att de får olika tillämpningsområden.