Natural Small-Scale Disturbances and Below-Canopy Solar

Natural Small-Scale Disturbances and

Below-Canopy Solar Radiation Effects on the Regeneration

Patterns in a Nothofagus betuloides Forest

– A Case Study from Tierra del Fuego, Chile

Thesis submitted in partial fulfilment of the requirements of

the degree Doctor rer. nat. of the

Faculty of Forest and Environmental Sciences,

Albert-Ludwigs-Universität,

Freiburg im Breisgau, Germany

by

Alvaro Andrés Promis Baeza

Freiburg im Breisgau, Germany

2009

Dean: Prof. Dr. Heinz Rennenberg

Supervisor: Prof. Dr. Dr. h.c. Albert Reif

Second reviewer: Prof. Dr. Helmut Mayer

Thesis defence: 23th January 2009

für meine geliebte Frau Carolina,

und meinen Sohn Martín.

Ohne ihre Liebe, Begleitung und Unterstützung

wäre dieser Traum für mich sehrr schwierig gewesen.

“Tierra del Fuego, Fireland,

so often described with wistful tones by explorers and writers,

is a land of legends

where the fires of the Indians have long died

but which still attracts the interest of researchers”

(Tuhkanen et al. 1989-1990)

“… for in these still solitudes,

Death, instead of Life,

seemed the predominant spirit”

(Darwin 1839)

“whoever eats the calafate returns for more”

(Moore 1983)

Acknowledgments

I

ACKNOWLEDGMENTS

Many people have contributed to the completion of my thesis in various ways. I

would like to thank the following people and institutions for helping me to achieve my

dream.

First at all, I would like to thank Prof. Dr. Dr. h.c. Albert Reif not only for his

help, support and confidence, but also for his friendship. I cannot imagine having had a

better supervisor for my Ph.D. I see in him an excellent person, one who encouraged me

to overcome the problems (“immer gibt es Probleme”), and who is so very generous

towards his students.

I would also like to thank Prof. Dr. Helmut Mayer for all his help and support as

my second supervisor.

I sincerely thank Dr. Stefanie Gärtner, whose constructive feedback and

comments at various stages have contributed significantly to this thesis. Similarly, I

must also thank Dr. Dirk Schindler for his help in developing the below-canopy solar

radiation chapter.

My sincere thanks to Prof. Dr. Jürgen Bauhus and Prof. Dr. Dr. h.c. Jürgen Huss,

whose comments and small contributions helped to improve my knowledge.

Special thanks also to Dr. Gustavo Cruz of the Department of Silviculture at the

University of Chile. His help throughout all of the study has been invaluable. His

research project FONDEF D02I1080 ‘Incorporation of coihue de Magallanes forests

into forest management for purposes of diversification and to increase production in the

XII Region (Chile)’ and the ‘Programa de Bosques Patagónicos’ of the University of

Chile supported a huge part of the field work carried out as part of this study. I thank

Gustavo Cabello and Marcelo Díaz, who provided help in the field, and Eric Campos

for helping me with the tree-ring measurements.

There were a number of people at the Waldbau-Institut in Freiburg who enriched

my everyday life in various ways. I am grateful to all of you.

I have to thank the German Academic Exchange Service (DAAD) for granting

me the scholarship; I am especially grateful to Ms. Maria Hartmann (Ref. 415) of the

DAAD office, who was always ready to answer questions and provide help.

I would like to thank those involved in the International Ph.D. Programme

‘Forestry in Transition’ of the Faculty of Forest and Environmental Sciences of the

Acknowledgments

II

University of Freiburg, especially Ms. Esther Muschelknautz. The international

programme was a source of intensive academic guidance and financial assistance, and

allowed me to participate in various seminars.

I am grateful to the Wildlife Conservation Society (WCS) in Chile, especially to

Dr. Bárbara Saavedra and Ricardo Muza, for providing logistical support within the

Karukinka Natural Park in Tierra del Fuego, at both Puerto Arturo and Río Bueno

research stations.

I would also like to thank the owner of the forests studied, Mr. Joaquín Soto and

his family, who provided both the research area and logistical support.

I am grateful to Dr. David Butler-Manning for proofreading the thesis.

A special thank you goes for Bernd Künemund for help with the translation of

the thesis review to German. I am also grateful to Rodrigo Vargas for printing and

binding the thesis.

I must thank Dr. Carlos Magni, head of the Department of Silviculture at the

University of Chile for his help, especially during the last phase of the preparation of

the thesis.

I would like to thank all of my family and the family of my wife for their

support. They were always present during our time in Germany. Whether by email,

letter, phone or parcel the distance was always made seem smaller.

One of the most important people this whole time has been my wife Carolina.

She has been with me through every moment of my Ph.D. work. I am very grateful to

her for the many sacrifices she has made to support me during this time. I would also

like to thank my son Martín for expanding our small family and accompanying me

through the last part of my work.

Contents

III

TABLE OF CONTENTS

1 THESIS REVIEW, ZUSAMMENFASSUNG AND RESUMEN 1

1.1 Thesis Review 1

1.1.1 Introduction: Disturbance and Ecology of Tree Species

Regeneration in Forests 1

1.1.2 The Nothofagus betuloides Forests 3

1.1.3 Hypothesis and Objectives 5

1.1.4 Study Area 6

1.1.4.1 Forest vegetation 6

1.1.4.2 Site conditions 7

1.1.5 Data Collection 7

1.1.6 Thesis structure 8

1.1.7 Short summary of results 11

1.2 Zusammenfassung 13

1.2.1 Einführung: Einfluss von Störungen auf die Verjüngung

in Waldbeständen 13

1.2.2 Die Nothofagus betuloides Wälder 15

1.2.3 Hypothesen und Ziele 17

1.2.4 Untersuchungsgebiete 18

1.2.4.1 Waldvegetation 18

1.2.4.2 Standortbedingungen 19

1.2.5 Datenaufnahme 20

1.2.6 Struktur der Doktorarbeit 21

1.2.7 Ergebnisse 23

1.3 Resumen 25

1.3.1 Introducción: Perturbación y Ecología de la Regeneración

De Especies Arbóreas en los Bosques 25

1.3.2 Los Bosques de Nothofagus betuloides 28

1.3.3 Hipótesis y Objetivos 30

1.3.4 Área de Estudio 31

1.3.4.1 Vegetación 31

1.3.4.2 Condiciones del sitio 32

Contents

IV

1.3.5 Toma de Datos 33

1.3.6 Estructura de la Tesis 34

1.3.7 Resumen de los Resultados 36

2 Nothofagus betuloides (Mirb.) Oerst 1871 (FAGALES:

NOTHOFAGACEAE) FORESTS IN SOUTHERN PATAGONIA

AND TIERRA DEL FUEGO 39

2.1 The Genus Nothofagus 39

2.2 Nothofagus betuloides – the Southernmost Evergreen Tree

Species and its Forests 40

2.2.1 Biology 41

2.2.2 Ecology 43

2.2.2.1 Geographical distribution 43

2.2.2.2 Autecology: Germination and juvenile growth 44

2.2.2.3 Synecology: Vegetation types 45

2.2.2.4 Forest texture 49

2.2.2.5 Forest dynamics and structure 50

2.3 Forest Use in the Past, the Present and the Future 52

2.3.1 The Past 52

2.3.2 The Present 54

2.3.3 The Future 55

3 SMALL-SCALE NATURAL DISTURBANCES AND TREE

DEVELOPMENT PROCESSES IN TWO FORESTS

DOMINATED BY Nothofagus betuloides – A CASE STUDY

FROM TIERRA DEL FUEGO 57

3.1 Abstract 57

3.2 Introduction 58

3.3 Material and Methods 61

3.3.1 Study Area 61

3.3.2 Sampling Design for the Assessment of the Forest

Structure 63

3.3.3 Measuring Gap Characteristics 63

3.3.4 Assessing Disturbance Dynamics and Radial Growth

Contents

V

Responses of Juvenile Trees 64

3.3.5 Statistical Analyses 65

3.4 Results 65

3.4.1 Forest Structures 65

3.4.2 Canopy Gap Characteristics 68

3.4.3 Disturbance Dynamics and Radial Growth Responses of

Juvenile Trees 70

3.5 Discussion 75

3.5.1 Causes and Characteristics of Canopy Gaps 75

3.5.2 Disturbance Dynamics and Radial Growth Responses of

Juvenile Trees 82

3.6 Conclusions 84

4 COMPARISON OF CANOPY STRUCTURES AND SOLAR

RADIATION TRANSMITTANCES ESTIMATED USING FOUR

DIFFERENT PROGRAMMES FOR THE ANALYSIS OF

HEMISPHERICAL PHOTOGRAPHS 85

4.1 Abstract 85

4.2 Introduction 86


4.3.1 Study Areas 88

4.3.2 Photographic Source Material 90

4.3.3 Image Processing 90

4.3.4 Image Analyses 92

4.3.4.1 The software used and their settings 92

4.3.4.2 Calculated canopy structure and light

environments 93

4.3.5 Statistical Analyses 96

4.4 Results 97

4.4.1 Comparison of Canopy Structure Estimates 97

4.4.2 Comparison of Solar Radiation Transmittance Estimates 101

4.4.2.1 Cosine-corrected solar radiation transmittance

estimates 101

Contents

VI

4.4.2.2 Non-cosine-corrected solar radiation transmittance

estimates 101

4.5 Discussion 109

4.5.1 Canopy Structure and Solar Radiation Transmittance 109

4.5.2 Other aspects requiring consideration when selecting a

programme 112

4.6 Conclusions 114

5 EFFECTS OF CANOPY STRUCTURE AND STAND

PARAMETERS ON THE VARIABILITY OF SOLAR

RADIATION TRANSMITTANCE IN AN UNEVEN-AGED

EVERGREEN Nothofagus betuloides FOREST 116

5.1 Abstract 116

5.2 Introduction 117


5.3.1 Study Area 119

5.3.2 Measuring Below-Canopy Solar Radiation 121

5.3.3 Canopy Structure and Stand Parameters Measurements 123

5.3.4 Statistical and Regression Analysis to Explain the Solar

Radiation Transmittance 125

5.4 Results 125

5.4.1 Transmission of Solar Radiation into the Forest 125

5.4.2 Relationships between Solar Radiation Transmittances,

Canopy Structure and Stand Parameters 128

5.5 Discussion 131

5.5.1 Solar Radiation Transmittances 131

5.5.2 Influences of Canopy Structures and Stand Parameters on

Solar Radiation Transmittances 133

5.6 Conclusions 135

6 EFFECTS OF NATURAL SMALL-SCALE DISTURBANCES ON

BELOW-CANOPY SOLAR RADIATION AND REGENERATION

PATTERNS IN AN OLD-GROWTH Nothofagus betuloides FOREST

IN TIERRA DEL FUEGO, CHILE 137

Contents

VII

6.1 Abstract 137

6.2 Introduction 138

6.3 Materials and Methods 141

6.3.1 Study Area 141

6.3.2 Selection of Canopy Gaps 142

6.3.3 Young Tree Measurements 142

6.3.4 Measuring Below-Canopy Solar Radiation 143

6.3.5 Statistical Analysis 144

6.4 Results 145

6.4.1 Solar Radiation Transmittance 145

6.4.2 Young Tree density and the Influence of Solar Radiation 146

6.4.3 Relations between Height, Diameter at Root Collar and

Young Tree Age 148

6.4.4 Relative Height Increment and Radial Growth with

Respect to Age and Solar Radiation 149

6.5 Discussion 152

6.5.1 Below-Canopy Solar Radiation Conditions 152

6.5.2 Regeneration Pattern and the Relationship between Young

Growth and Below-Canopy Solar Radiation 154

6.5.3 Browsing Effects 154

6.5.4 Silvicultural Implications for Old-Growth N. betuloides

Forests 155

7 CONCLUSIONS AND SILVICULTURAL IMPLICATIONS 157

7.1 Disturbance and Stand Dynamics 157

7.2 Light and Tress Species Regeneration: Testing the

Hypothesis 158

7.2.1 Below-canopy solar radiation transmittances 159

7.2.2 Regeneration patterns of Nothofagus betuloides 160

7.3 Shade Tolerance of Nothofagus betuloides 161

7.4 Silvicultural Implications for Old-Growth Nothofagus

betuloides Forests 162

8 REFERENCES 165

Chapter 1 Thesis review, Zusammenfassung and resumen

1

CHAPTER 1

THESIS REVIEW, ZUSAMMENFASSUNG AND RESUMEN

1.1 Thesis Review

1.1.1 Introduction: Disturbance and Ecology of Tree Species Regeneration in

Forests

Natural plant communities are spatially heterogeneous and dynamic systems

(Sousa 1984). Disturbances are major impacts that affect their structures and species

compositions (White 1979, Sousa 1984), by bringing about changes to the state of the

structural and physical variables in the ecosystem (White and Jentsch 2001). One

important step towards understanding the natural dynamics of natural communities,

therefore, is ascertaining the prevailing disturbance regimes (White et al. 1999).

Not all disturbances are equal. They differ with respect to spatial patterns (size

and shape), temporal parameters (frequency, return period, turnover time and

seasonality), specificity (probability of disturbance as a factor of species, age and size)

and magnitude (intensity and severity) (Sousa 1984, White et al. 1999).

Forest communities are frequently subjected to disturbances (Oliver and Larson

1996). The effects of natural disturbances on the dynamics and structure of forests have

been studied in tropical, temperate and boreal forests (e.g. Platt and Strong 1989,

Denslow and Spies 1990, McCarthy 2001, Kuuluvainen 2002).

Large-scale disturbances are more severe than small-scale ones, but they are less

frequent (Sousa 1984, Turner et al. 1998). Small-scale canopy disturbances (gaps)

occurring as a result of single or multiple tree-falls may be more frequent in old-growth

forest ecosystems (Oliver and Larson 1996), and affect a larger area over time (Spies et

al. 1990).

Therefore in forest ecosystems, the severity of disturbances influences stand

structures, species composition, the growth rates of the surviving trees, the regeneration

dynamics, and the species diversity (Connell 1978, Canham et al. 1994, Oliver and

Larson 1996).


2

The development phase of a forest affected by small-scale canopy disturbances

has been termed the gap phase (Watt 1947), the period during which tree regeneration is

stimulated, the regenerating trees typically fill the gap and, finally, one or a group of

them replace the previous canopy trees (Busing and Brokaw 2002). Large canopy gaps

are more likely to be filled by shade-intolerant tree species, whereas small gaps are

almost always filled by shade-tolerant tree species (Grubb 1977).

In forests, the availability of the resources (light, water, soil nutrients) required

for plant growth is modified by disturbances and the occurrence of small canopy gaps

(Canham and Marks 1985). A gradient of differing site conditions exists from the area

on the ground directly below the disturbed canopy to the undisturbed surrounding forest

(Riklefs 1977, Denslow 1980). The substrate conditions are frequently also altered by a

disturbance. When trees are uprooted or snapped contrasting substrate conditions

(microsites) are created on the forest floor (Collins et al. 1985).

The most obvious changes brought about by gaps can be observed in the light

environment below canopy level (Canham et al. 1990, Gray et al. 2002). Understanding

the effects of solar radiation in the forest understorey is important for understanding

forest dynamics, as solar radiation affects plant regeneration patterns such as

germination, establishment, growth and survival (Grant 1997). The differences in the

below-canopy solar radiation between gaps and closed stands have been described for

different forest types (Denslow 1980, Canham et al. 1990, de Freitas and Enright 1995,

Gray et al. 2002).

The importance of the role solar radiation plays in forest development means

that there has been much interest in measuring below-canopy solar radiation. Several

instruments have been developed to measure either directly or indirectly the below-

canopy solar radiation environment. Many ecologists and foresters prefer indirect

approaches due to the difficulties inherent in measuring solar radiation directly

(Jennings et al. 1999). Hemispherical photography has been a widely applied as a means

of calculating indirectly the light environment in forests (Rich et al. 1993, Comeau et al.

1998, Gendron et al. 1998, Clearwater et al. 1999, Engelbrecht and Herz 2001).

The spatial heterogeneity of microenvironments within the forest, and the

varying regeneration niches, facilitate the establishment of seedlings of different tree

species and promote the coexistence of various species (Grubb 1977). Even under

undisturbed canopies heterogeneity can be high, due to variations in canopy

composition, height, thickness and foliage density (Lieberman et al. 1989, Veblen


3

1992). At high latitudes the proportion of the solar radiation that reaches the forest floor

and extends into the stand beyond the edges of canopy gaps is higher than at lower

latitudes, resulting in a more homogeneous distribution of below-canopy solar radiation

within a stand (Canham et al. 1990). Therefore in high latitudes, including Nothofagus

forests in the southern hemisphere, the regeneration dynamic, especially of some shade-

tolerant species, may be less affected by canopy gaps themselves (Canham 1990,

Veblen 1992).

1.1.2 The Nothofagus betuloides Forests

Nothofagus betuloides is an evergreen endemic tree species of the Chilean and

Argentinean subantarctic forests. At present the forest area in Chile with N. betuloides

as either the dominant or as a minor species amounts to approximately 2,739,906

hectares, corresponding to 20.4 % of the total national forest area (CONAF-CONAMA

1999a). Of this surface 1,793,098 hectares comprises the evergreen N. betuloides forest

type, and 946,808 hectares the evergreen-deciduous N. betuloides – N. pumilio forest

sub-type. In southern Patagonia and Tierra del Fuego, N. betuloides occurs in

approximately 1,396,947 hectares (53 % of the total forest area). Today, these forests

are no longer endangered, as more than 69 % of the these forests in southern Patagonia

and Tierra del Fuego are currently designated either ‘State Protected Wild Area’

(SNASPE) or is private nature reserve (Karukinka) administrated by the Wildlife

Conservation Society (WCS) (Arroyo et al. 1996, CONAF-CONAMA 1999b).

Natural large and small-scale disturbances in Nothofagus forests in Tierra del

Fuego are shaped by the strong winds (Rebertus et al. 1997, Puigdefábregas et al. 1999).

Storms can cause the blow-down of entire stands (Rebertus et al. 1997, Puigdefábregas

et al. 1999), and wave-like patterns of canopy gaps on sites predisposed to wind

disturbances have been reported for pure N. betuloides and N. pumilio forests, and

mixed N. betuloides - N. pumilio forests (Rebertus and Veblen 1993b, Rebertus et al.

1993, Puigdefábregas et al. 1999). Small-scale canopy gaps < 200 m2 in size are created

by the windthrow of individual trees (Rebertus and Veblen 1993a, Gutiérrez 1994).

However, in pure primary old-growth N. betuloides forests in Tierra del Fuego discrete

gaps are not always apparent, as openings always occur as interwoven gap complexes in

the canopy layer (Rebertus and Veblen 1993a). In places, the presence of patches of

younger trees adjacent to larger patches of old-growth forest lend the forest a multi-


4

cohort stand structure, which results in the occurrence of a patch mosaic pattern to the

structure of these Nothofagus forests (Gutiérrez et al. 1991).

The regeneration dynamics of southern South American Nothofagus forests, and

the importance of disturbances in this context, have been well summarised by Veblen et

al. (1996) and Pollmann and Veblen (2004). They showed that at low elevations and

under a milder climate, where Nothofagus species compete with other shade-tolerant

forest species, large-scale disturbances appear to be important for the successful

regeneration of Nothofagus. At higher elevations and latitudes, where site conditions are

generally suboptimal and the forest species diversity is low as a consequence, the

Nothofagus species regenerated after both large-scale (e.g., landslides, fires, blow-

downs) and small-scale, tree-fall disturbances.

In Tierra del Fuego, N. betuloides is capable of establishing in small canopy

gaps (Rebertus and Veblen 1993a, Gutiérrez 1994, Arroyo et al. 1996). Advance

regeneration present on the ground at the time of gap creation is released by the

formation of these small gaps in the canopy, that which is not killed or damaged by

falling canopy trees (Veblen et al. 1996). However, the establishment and growth of N.

betuloides can be impeded where there is a dense coverage of understorey trees and

shrubs such as Drimys winteri and Maytenus magellanica (Rebertus et al. 1993a,

Gutiérrez et al. 1991, Gutiérrez 1994, Veblen et al. 1996).

Since the late 19th century the forests of the coast and much of the interior of

southern Patagonia and Tierra del Fuego have been logged according to a selective

system of logging, known in the region as floreo. This means, the best, largest and

healthiest trees were selectively cut for timber, and the poor quality, badly formed and

unhealthy trees were left standing, or the remaining forests were simply burned

(Martínez Pastur et al. 2000, Cruz et al. 2007a).

At present, Chilean law dictates that N. betuloides forests suitable for timber

production must be managed according to either a selection or a shelterwood system

(Donoso 1981), both of which are designed to promote natural regeneration after

logging. However, many stands are still managed as floreo, or where the shelterwood

system is being applied only the regeneration felling is currently being applied (Cruz et

al. 2007a). Even if carried out correctly, the application of the shelterwood system will

result in a homogenisation and simplification of the typically uneven-aged structure of

this forest type (Cruz et al. 2008).


5

Very recently, there has been a trend towards an intensification of the utilisation

of N. betuloides forests. New studies of the ecology of the N. betuloides forests, their

distribution, silviculture, wood properties, industrial yields, and also the methods of

drying N. betuloides timber, have been carried out (see Cruz and Caldentey 2007).

An understanding of the natural stand dynamics of these forests, including the

role of natural disturbances, will provide the basis upon which they may be managed as

a renewable resource; that continues to provide the necessary timber while at the same

time preserving their species diversity and structural richness (Lindenmayer and

Franklin 2002).

1.1.3 Hypothesis and Objectives

The hypothesis central to this study is that the establishment and growth of

juvenile N. betuloides is affected by the occurrence of natural, small-scale disturbances

and the existence of a gradient in the below-canopy solar radiation, with solar radiation

penetrating into the stand through these gaps and extending outwards to areas below the

undisturbed forest canopy.

The principal objectives of this study are to:

1. characterise the small, naturally occurring canopy gaps in forests dominated by

N. betuloides,

2. analyse the canopy gap dynamics in forest dominated by N. betuloides,

3. compare canopy structure variables and below-canopy solar radiation using four

different programmes for the analysis of hemispherical photographs,

4. analyse the relationship between canopy structures and stand parameters and the

variation in the below-canopy solar radiation in the N. betuloides forest, and

5. determine the effects of the natural, small-scale disturbances on:

a. the below-canopy solar radiation environment,

b. the regeneration patterns (density and growth rates) of N. betuloides,

c. the radial growth responses of juvenile trees growing in the vicinity of

the canopy gaps.


6

1.1.4 Study Area

The study area is located at the ‘Estancia Olguita’ on the southeastern side of the

Río Cóndor (53 ° 59 ’ S, 69 ° 58 ’ W) and on the southwestern Chilean side of Tierra

del Fuego. Cattle grazing continues in the Río Cóndor valley to this day, where the low

riversides and the entire reverie are especially heavily impacted (Arroyo et al. 1996).

However, no evidence of cattle grazing was found in the study forest.

1.1.4.1 Forest vegetation

The study was conducted in two primary, uneven-aged old-growth forests with

no evidence of human impact. The natural state of these forests allowed for a study of

the effects of canopy disturbances on the below-canopy solar radiation environment, the

regeneration patterns, and the radial growth responses of juvenile N. betuloides trees.

Two forests of approximately 20 ha each one were studied, a pure N. betuloides forest

and a mixed evergreen-deciduous N. betuloides - N. pumilio forest. In the pure forest,

the complete work was carried out. In the mixed forest, only the natural small-scale

disturbances and the tree development processes were characterised.

The plant species diversity of the pure N. betuloides forest was low. There were

only few species in the shrub layer, which was dominated mainly by the tall shrub

Berberis ilicifolia. Less frequent and lower in cover were B. buxifolia, and the smaller

shrubs Pernettya mucronata, and P. pumila. The diversity of vascular plant species in

the ground flora was also low, the most frequent species being Adenocaulon chilense,

Luzuriaga marginata, Senecio acanthifolius, Rubus geoides, Uncinia lechleriana, the

ferns Blechnum penna-marina, Asplenium dareoides, Grammitis magellanica, and filmy

ferns, mostly Hymenophyllum secundum and H. tortuosum. A very pronounced layer of

abundant mosses and liverworts were found on the ground and on decaying tree trunks,

dominated by the mosses Dicranoloma chilense, D. robustum, and Pyrrhobryum

mnioides, and the liverworts Gakstroemia magellanica, Plagiochila obovata, and

Chiloscyphus magellanica. Worth mentioning are lichens like Cladonia sp. on the

ground and Pseudocyphellaria sp. on decaying tree trunks.

The mixed evergreen – deciduous N. betuloides – N. pumilio forest is regarded

as a transition between the evergreen and the deciduous Nothofagus forests. N.

betuloides dominates the more humid and poorly drained sites. Shrub coverage was


7

low, dominated by B. ilicifolia and P. mucronata. There were fewer filmy ferns

(Hymenophyllaceae family) and bryophyte species than in the pure N. betuloides forest

(Young 1972, McQueen 1976, Pisano 1977, Gajardo 1994, Luebert and Pliscoff 2006).

1.1.4.2 Site conditions

The climate of the area belongs to the Northern Antiboreal sub-zone, with a

mean air temperature between 9-9.5 °C during the warmest month of the year and

remaining above zero in the coldest month. The mean annual rainfall is around 500-600

mm, but can reach up to 900 mm. The wind direction is commonly west to

southwesterly, with average speeds between 14-22 km h-1 and maximum speed of > 100

km h-1 (Tuhkanen 1992).

The study area is part of the subalpine zone defined by Frederiksen (1988). The

relief is characterised by valleys running parallel to the Andes, which in most places

have been glacially deepened forming U-valleys. Two soil formation processes have

been described for N. betuloides forests, both of wich occurred in the study area. These

are podsolisation on well-drained sites, and hydromorphism in areas where there is

waterlogging (Pisano 1977, Puigdefábregas et al. 1999, Gerding and Thiers 2002,

Romanyà et al. 2005). The soils are normally shallow (< 50 cm), loamy in texture,

acidic (pH 3.4-5.5) and not very fertile (Gerding and Thiers 2002, Romanyà et al. 2005,

Thiers and Gerding 2007). The accumulation of deep layers of organic matter on the

forest floor and large amounts of decaying wood have been described for this forest type

in Tierra del Fuego (Gutiérrez et al. 1991). Most of the plant roots and nutrients are

located towards the bottom part of a thick raw humus layer, i.e., at soil depth of only 4-

10 cm (Pisano 1977, Gutiérrez et al. 1991, Gerding and Thiers 2002, Romanyà et al.

2005).

1.1.5 Data Collection

The data collection occurred over the course of two periods of field work, the

first of which lasted from January to March 2006. In this period the forest stands were

selected, and 36 parallel transects (300 m) were located at 30 m intervals. The transects

were used in the assessment of the canopy species compositions and the forest

structures employing the point-centred quarter method, and in the recording of the


8

canopy gaps. All characteristics of the canopy gaps were measured. Increment cores

were taken from tree trunks to assess the disturbance dynamics and radial growth

responses of juvenile trees. Thirteen canopy gaps in the pure N. betuloides forest were

selected for the study of the effects of the canopy gaps on the below-canopy solar

radiation and the regeneration. The transects were established running through the

canopy gap centres to the adjacent undisturbed canopies. The hemispherical

photographs were taken and plots for the seedling and sapling surveys were carried out.

The second period of field work lasted from January to February 2007. The data

collection focused on studying the effects of canopy structure and stand parameters on

the variability of solar radiation transmittance in the pure N. betuloides forest.

Hemispherical photographs were taken and stand measurements were made in fixed

plots of 225 m2.

In April 2007 another two sets of 26 hemispherical photographs were made in

two additional broadleaf forests in order to compare the estimates of the canopy

structures and below-canopy light environments provided by four different programmes

for the analysis of hemispherical photographs. One set of hemispherical photographs

came from the Weberstedter Holz in the Hainich National Park in Germany (51 ° 01 ’

N, 10 ° 04 ’ E), a temperate mixed deciduous beech forest on limestone, dominated by

Fagus sylvatica. The other set of hemispherical photographs was taken in a very humid

submontane tropical cloud forest, in the Sierra de Lema Forest (Canaima National Park)

in Venezuela (05 ° 53 ’ N, 61 ° 26 ’ W).

1.1.6 Thesis structure

Each of the following chapters of the thesis has been published on or submitted

to a scientific journal for publication. The paper title, the authorship, the relevant journal

and a brief summary is given for each.

Chapter 2 Nothofagus betuloides (Mirb.) Oerst 1871 (FAGALES:

NOTHOFAGACEAE) forests in southern Patagonia and Tierra del

Fuego

published on Anales del Instituto de la Patagonia 36: 53-68

Promis, A., Cruz, G., Reif, A., Gärtner, S.


9

The aim of this study was to review biological and ecological aspects of N.

betuloides, as well as the characteristics of the vegetation types where it occurs, the

forest structures, forest dynamics and the use of the forests in the past, present and

future. It is a review of the present state of knowledge in the relation to the

southernmost part of the species’ distribution.

Chapter 3 Small-scale natural disturbances and tree development processes in

two forests dominated by Nothofagus betuloides – A case study from

Tierra del Fuego

submitted

Promis, A., Gärtner, S., Reif, A., Cruz, G.

The objectives of the study were to compare the characteristics of natural canopy

gaps found in the pure N. betuloides forest and the mixed N. betuloides – N. pumilio

forest, and to analyse the gap dynamics and the role of natural disturbances in the radial

growth responses of juvenile trees. Two questions in particular were addressed: Do

juvenile N. betuloides trees growing in pure and mixed natural forests respond similarly

following the creation of small canopy gaps? In a mixed forest, do the responses of N.

betuloides and N. pumilio differ? The objective was to provide a deeper understanding

of the natural dynamics of these forest types in Tierra del Fuego.

Chapter 4 Comparison of canopy structures and solar radiation transmittances

estimated using four different programmes for the analysis of

hemispherical photographs

submitted

Promis, A., Gärtner, S., Butler-Manning, D., Durán-Rangel, C., Reif, A.,

Cruz, G., Hernández, L.

There have been many studies involving the use of hemispherical photographs to

estimate indirectly forest structures and below-canopy solar radiation environments.

There are both commercial and free software packages for the analysis of hemispherical

photographs. The costs of investment might represent an advantage of the free

programmes over the commercial, but are there differences in the usability and the

results provided by the different softwares? As yet, little has been documented about the


10

differences in their outputs and technical applications from a user (ecologists and

foresters) perspective. The objective was to show the most commonly used canopy

structures (canopy openness and effective plant area index) and below-canopy solar

radiation variables (direct, diffuse and global transmittances) estimated from digital

hemispherical photographs taken under two forest canopy conditions (gap and closed

canopy) and in forest located at different latitudes (northern, equatorial and southern

hemisphere).

Chapter 5 Effects of canopy structure and stand parameters on the variability

of solar radiation transmittance in an uneven-aged evergreen

Nothofagus betuloides forest

submitted

Promis, A., Schindler, D., Reif, A., Cruz, G.

There are knowledge gaps regarding the effects of canopy structure and stand

parameters on the spatial variation of the below-canopy solar radiation environment in

N. betuloides forests. The objectives of the study, therefore, were to analyse the effects

of the forest canopy on the transmittance of solar radiation to the forest floor, and to

evaluate whether canopy structures and stand parameters can explain the variation in the

below-canopy solar radiation in an uneven-aged N. betuloides forest.

Chapter 6 Effects of natural small-scale disturbances on understorey light and

regeneration patterns in an old-growth Nothofagus betuloides forest

in Tierra del Fuego, Chile

submitted


What are the effects of small canopy gaps on the regeneration dynamics of N.

betuloides forest? It is not yet known whether these small canopy gaps result in a

differentiation in the availability of the below-canopy solar radiation between areas

located beneath natural canopy gaps and areas beneath undisturbed canopy. If there are

differences in the solar radiation levels, do these influence the density and growth rates

of the young trees in the forest? Another question concerned whether the presence of

canopy gaps influences the browsing habits of Lama guanicoe and the age distribution


11

of the young trees in N. betuloides forest. All of these questions were addressed in the

study.

1.1.7 Short summary of results

The canopy gaps observed were small, with an average size of 51 m2 in the pure

N. betuloides forest and 107 m2 in the mixed N. betuloides – N. pumilio forest. Only 2.0

% of the total canopy area of the pure forest was covered by canopy gaps and 4.6 % by

expanded gaps. In the mixed forest the gap areas were 2.6 and 2.5 times greater,

respectively. An increase in the frequency of disturbances affecting the forests since

1920-30 was observed.

The mean number of gap-makers per gap, i.e., the trees that have created the

canopy gaps, was 2.2 trees in the pure forest and 2.5 trees in the mixed forest. The most

common type of damage in the pure forest was snapping, compared to uprooting in the

mixed forest.

The greatest changes in radial increment after release from restricted growth

were found for N. pumilio (18.4 %) and N. betuloides (11.2 %) growing in the mixed

forest and the lowest for N. betuloides (2.4 %) in the pure forest.

An analysis of hemispherical photographs using one commercial (HemiView)

and three free programmes (Gap Light Analyzer, hemIMAGE and Winphot) revealed

that all four programmes calculated similar canopy structures and below-canopy solar

radiation estimates. Only the effective plant area index with an ellipsoidal leaf angle

distribution calculated for canopy gaps demonstrated a weak correlation. Consequently,

the costs associated with the analysis of hemispherical photographs can be reduced by

using free software, which can be downloaded from the internet.

The below-canopy solar radiation in the N. betuloides forest was affected by a

high level of horizontal variability, the vertical heterogeneity of the forest canopy, and

also by the low angles of the sun’s path over the course of the growing season (October

to March). The below-canopy direct radiation appeared to be variable in space and time,

while the variability of the diffuse radiation was lower.

The direct solar radiation during the growing season correlated poorly with the

canopy structure and stand parameters, the variable that best fit the data was the plant

area index (R2 = 0.263). Although poorly correlated with simple stand parameters,

diffuse and global solar radiation were very sensitive to canopy openness (R2 = 0.963


12

and 0.833). However, using only variables of canopy structure and stands parameters

measured in the forest a high heterogeneity in the spatial structures of the uneven-aged

forest was found affecting the below-canopy diffuse and global solar radiation. This

explained 75 % of the variation in diffuse solar radiation and 73 % of global solar

radiation, when basal area, crown area projected to the ground, crown volume, stem

volume and the averaged equivalent crown radius were combined in a common model.

The availability of the non-cosine-corrected direct, diffuse and global solar

radiation transmitted into the pure forest ranged from 3.2 to 19.4 %, 3.1 to16.7 %, and

3.2 to 17.6 %, respectively.

The seedlings and saplings of N. betuloides showed the greatest shade tolerance,

compared to other South American Nothofagus species, regenerating even under very

shady conditions, and apparently not requiring the presence of large gaps to establish.

This resulted in a more continuous process of forest regeneration.

The solar radiation transmittances did not correlate with either the relative radial

growth or the relative height increment of the young trees. The growth of the young

trees correlated only with the age of the plants. Inverse polynomial curves explained 70

% of the variance in relative radial growth and 50 % of the variance in relative height

increment.

The proportions of seedlings browsed by L. guanicoe were low (0.7-2.8 % of the

total). Browsing damage to young trees was observed below canopy gaps as well as

beneath undisturbed canopies, demonstrating no particular habitat preference.

The heterogeneous canopy of the primary old-growth uneven-aged N. betuloides

forest with only very small canopy gaps produced a variety of mosaics in the

understorey with seedlings and saplings present in a large range of ages and heights.

These results suggest that N. betuloides exhibits greater shade-tolerance. It can

successfully grow beneath closed canopies and in the vicinity of very small canopy gaps

with low radial growth, until larger disturbances to the canopy bring abrupt increases.


13

1.2 Zusammenfassung

1.2.1 Einführung: Einfluss von Störungen auf die Verjüngung in Waldbeständen

Natürliche Pflanzengesellschaften sind räumlich heterogene, dynamische

Systeme (Sousa 1984). Störungen verändern die Struktur sowie die physikalische

Umwelt von Ökosystemen (White und Jentsch 2001). Aus diesem Grund wurden sie als

der Hauptfaktor, welcher die Struktur und Artenzusammensetzung beeinflusst,

angesehen (White 1979, Sousa 1984). Die Beschreibung von Störungsregimen ist also

ein wichtiger Schritt auf dem Weg zum Verständnis der natürlichen Dynamik in

Pflanzengesellschaften (White et al. 1999).

Störungen können abhängig von ihrer Art völlig unterschiedliche Auswirkungen

haben. Störungen unterscheiden sich in ihrem räumlichen (Größe und Form) und

zeitlichen Auftreten (Frequenz, Dauer, Zeitpunkt). Störungen können das gesamte

Ökosystem betreffen oder nur Teile (Arten, Altersklassen, Durchmesserklassen). Zudem

kann das Ausmaß der Störung sehr unterschiedlich sein (Sousa 1984, White et al. 1999).

Waldgesellschaften sind häufig Störungen ausgesetzt (Oliver und Larsern 1996). Die

Auswirkungen natürlicher Störungen auf die Dynamik und die Struktur von

Waldbeständen wurden in tropischen, temperaten und borealen Wäldern erforscht (Platt

und Strong 1989, Denslow und Spies 1990, McCarthy 2001, Kuuluvainen 2002).

Großflächige Störungen haben stärkere Auswirkungen als kleinflächige

Störungen, sie treten aber seltener auf (Sousa 1984, Turner et al. 1998). Kleine

Störungen durch das Absterben eines oder mehrer Bäume sind in alten

Waldökosystemen häufiger (Oliver und Larson 1996) und betreffen über längere Zeit

gesehen große Flächen (Spies et al. 1990).

Folglich beeinflusst in Waldökosystemen die Intensität von Störungen die

Bestandesstruktur, die Artenzusammensetzung, das Zuwachsverhalten der überlebenden

Bäume, die Verjüngungsdynamik und die Diversität (Connell 1978, Canham et al. 1994,

Oliver und Larson 1996).

Die Entwicklung von Verjüngung in Wäldern, hervorgerufen durch kleinflächige

Kronendachöffnungen, wurde als Lückendynamik beschreiben (Watt 1947).

Entstehende Lücken werden durch sich verjüngende Bäume gefüllt, von denen einer

oder mehrere die vorherigen Bäume im Kronendach ersetzt. Größere Lücken im


14

Kronendach können auch durch schattenintolerante Baumarten geschlossen werden,

kleine Unterbrechungen im Kronendach dagegen werden in der Regel durch

schattentolerante Arten gefüllt (Grubb 1977).

In Wäldern wird die Verfügbarkeit von Ressourcen, die für das

Pflanzenwachstum notwendig sind (Licht, Wasser, Nährelemente), durch Störungen und

das Auftreten von kleinen Öffnungen im Kronendach modifiziert (Canham und Marks

1985). Es entsteht ein Gradient unterschiedlicher Standortbedingungen vom Zentrum

einer Lücke bis zum ungestörten, die Lücke umgebenden Bestand (Riklefs 1977,

Denslow 1980). Die Standortbedingungen verändern sich, insbesondere wenn Bäume

entwurzelt oder abgebrochen sind, wodurch Mikrostandorte auf dem Waldboden

entstehen (Collins et al. 1985).

Die offensichtlichsten Veränderungen, verursacht durch Lücken im Kronendach,

können im Strahlungshaushalt des Bestandesinneren beobachtet werden (Canham et

al.1990, Gray et al. 2002). Um Walddynamik zu verstehen, ist es wichtig den Einfluss

des Strahlungshaushalts auf den Unterstand zu kennen. Der Strahlungshaushalt

beeinflusst den gesamten Verjüngungsvorgang wie die Keimung, die Etablierung von

Jungwüchsen, deren Wachstum und Überleben (Grant 1997). Unterschiede im

Strahlungshaushalt des Bestandesinneren innerhalb von Lücken sowie unter einem

geschlossenen Kronendach wurden für unterschiedliche Waldtypen beschrieben

(Denslow 1980, Canham et al. 1990, de Freitas und Enright 1995, Gray et al. 2002).

In vielen Forschungsprojekten wurden die Strahlungsintensitäten im

Bestandesinneren gemessen. Es wurde eine Anzahl von Messinstrumenten entwickelt,

die entweder die direkte oder die indirekte Strahlung im Bestandesinneren messen

können. Viele Ökologen und Forstwissenschaftler bevorzugen den indirekten Ansatz

um die Strahlungsintensität abzuschätzen, da bei der Messung von direkter Strahlung

eine Reihe von Problemen auftritt (Jennings et al. 1999). Hemisphärische Photographie

beispielsweise ist ein weit verbreitetes Mittel zur indirekten Berechnung der Strahlung

in Wäldern (Rich et al. 1993, Comeau et al. 1998, Gendron et al. 1998, Clearwater et al.

1999, Engelbrecht und Herz 2001).

Die räumliche Heterogenität und die damit verbundene Vielfalt an

unterschiedlichen Mikrostandorten führen zu ganz unterschiedlichen

Verjüngungsnischen, in denen sich unterschiedliche Arten etablieren und nebeneinander

koexistieren können (Grubb 1977). Selbst unter einem ungestörten Kronendach findet

man eine hohe kleinstandörtliche Heterogenität, da Kronendächer im Hinblick auf die


15

Artenzusammensetzung, die Höhe, die Dichte und den Blattflächenindex ganz

unterschiedlich gestaltet sein können (Lieberman et al. 1989, Veblen 1992).

Insbesondere in höheren Breitengraden fällt ein größerer Anteil der Sonnenstrahlung am

Rande der Lücke oder sogar im angrenzenden Bestand auf den Unterstand bzw.

Waldboden. Dies führt insgesamt gesehen zu homogeneren Strahlungsverhältnissen im

Bestandesinneren (Canham et al. 1990). Aus diesem Grund ist es möglich, dass die

Verjüngung von schattentoleranten Arten in größeren Breiten, also auch in den

Nothofagus-Wäldern der südlichen Hemisphäre, durch die Lückendynamik in Wäldern

nur relativ wenig beeinflusst wird (Canham 1990, Veblen 1992).

1.2.2 Die Nothofagus betuloides Wälder

Nothofagus betuloides ist eine immergrüne endemische Baumart der

chilenischen und argentinischen subantarktischen Wälder. In Chile bedecken die Wälder

mit N. betuloides als Haupt- oder Nebenbaumart eine Fläche von ungefähr 2,739,906

ha. Dies entspricht 20,4 % der gesamten nationalen Waldfläche Chiles (CONAF-

CONAMA 1999a). Von dieser Fläche gehören 1,793,098 ha zu der immergrünen N.

betuloides-Assoziation, die restlichen 947,808 ha gehören dem N. betuloides – N.

pumilio-Mischwäldern an. In Südpatagonien und der Tierra del Fuego wächst N.

betuloides auf ungefähr 1,396,947 ha. Dies entspricht 53% der gesamten Waldfläche.

Heute sind diese Wälder nicht mehr gefährdet, da mehr als 69% der Landfläche in

Südpatagonien und der Tierra del Fuego in „State Protected Wild Areas“ (SNASPE)

oder in privaten Naturreservaten (Karukinka), verwaltet durch die „Wild Conservation

Society“ (WCS), liegen (Arroyo et al. 1996, CONAF-CONAMA 1999b).

Natürliche groß- und kleinflächige Störungen in Nothofagus-Wäldern in Tierra

del Fuego werden vor allem durch starke Stürme verursacht (Rebertus et al. 1997,

Puigdefábregas et al. 1999). Stürme können großflächige Windwürfe ganzer

Waldbestände verursachen (Rebertus et al. 1997, Puigdefábregas et al. 1999).

Wellenartige Unterbrechungen des Kronendachs wurden in topografisch

windexponierten N. betuloides-, N. pumilio- Beständen sowie in Mischwäldern mit N.

betuloides und N. pumilio beobachtet (Rebertus und Veblen 1993b, Rebertus et al.

1993, Puigdefábregas et al. 1999). Kleinere Unterbrechungen des Kronendachs unter

200 m² entstehen durch Entwurzelung oder Stammbruch einzelner Bäume (Rebertus

und Veblen 1993a, Gutiérrez 1994). In reinen, alten N. betuloides- Primärwäldern von


16

Tierra del Fuego sind einzelne Lücken im Kronendach oftmals nicht deutlich sichtbar,

da die Lücken immer als mit dem vorhandenen Kronendach verflochtene

Lückenkomplexe auftreten (Rebertus und Veblen 1993a). An anderen Orten hat das

Vorhandensein von Verjüngungskegeln, angrenzend an Althölzer zum Entstehen von

„multi-kohorten“ mosaikartigen Bestandesstrukturen in Nothofagus-Wäldern geführt

(Gutiérrez et al. 1991).

Die Regenerationsdynamik der südlichen südamerikanischen Nothofagus-

Wälder und der Einfluss von Störungen auf die Verjüngung wurde von Veblen et al.

(1996) und Pollman und Veblen (2004) sehr gut zusammengefasst. Sie zeigten, dass in

niedrigen Meereshöhen und bei milderem Klima, wo Nothofagus-Arten mit anderen

schattentoleranten Baumarten vergesellschaftet sind und konkurrieren, großflächige

Störungen wichtig für die Verjüngung sind. In größeren Meereshöhen und höheren

Breitengraden, also unter suboptimalen Standortsbedingungen und geringerer

Artenvielfalt, scheinen die Nothofagus-Arten in der Lage zu sein, sich nach

großflächigen Störungen (z.B Erdrutsche, Waldbrände, Windwürfe) als auch nach

kleinflächigen Störungen zu verjüngen.

N. betuloides ist in Tierra del Fuego in der Lage, sich auch nach kleinflächigen

Störungen zu etablieren (Rebertus und Veblen 1993a, Gutiérrez 1994, Arroyo et al.

1996). Keimlinge, Jungwüchse und fortgeschrittenere Verjüngung, die zum Zeitpunkt

der Störung schon vorhanden sind, können entstandene Lücken schnell schließen, wenn

sie durch die Störung nicht beschädigt wurden (Veblen et al. 1996). Die Etablierung

und das Wachstum von N. betuloides kann aber durch dichte Bodenvegetation oder das

Vorhandensein von Sträuchern und Bäumen wie z.B. Drimys winteri oder Maytenus

magellanica verhindert werden (Rebertus et al. 1993a, Gutiérrez et al. 1991, Gutiérrez

1994, Veblen et al. 1996).

Seit dem späten 19. Jahrhundert wurden die Wälder an der Küste und ein großer

Teil der Wälder im Landesinneren im südlichen Patagonien und der Tierra del Fuego

einzelstammweise genutzt. Das waldbauliche Vorgehen ist unter dem Namen floreo in

der Region bekannt. Dies bedeutet, dass die qualitativ besten, größten und gesündesten

Bäume selektiv genutzt wurden, und die qualitativ schlechten, kranken Bäume stehen

gelassen wurden oder der verbleibende Wald völlig abgebrannt wurde (Martínez Pastur

et al. 2000, Cruz et al. 2007a).

Im Moment müssen die für die Holzproduktion geeigneten N. betuloides-

Wälder in Chile aufgrund gesetzlicher Bestimmungen entweder Plenterung oder im


17

Schirmschlagverfahren bewirtschaftet werden. Dadurch soll die natürliche Verjüngung

der Bestände nach der Holzernte gefördert werden. Trotzdem werden in der Praxis

immer noch viele Bestände in der herkömmlichen Weise (floreo) bewirtschaftet, oder

nur die Regenerationshiebe des Schirmschlagverfahrens werden durchgeführt (Cruz et

al. 2007a). Durch die Anwendung des Schirmschlagverfahrens wird es zu einer

Homogenisierung von Artenzusammensetzung und Strukturen dieser Wälder kommen

(Cruz et al. 2008).

Seit kurzem ist in Chile ein Trend hin zur Intensivierung der Nutzung von

Nothofagus-Wäldern zu beobachten. Neue Studien über die Ökologie der N. betuloides-

Wälder, ihre Verbreitung, Waldbau, Holzeigenschaften, Methoden zur Holztrocknung

und Erträge wurden durchgeführt (Cruz und Caldentey 2007).

Das Verständnis der natürlichen Bestandesdynamik inklusive der Rolle

natürlicher Störungen wird die Basis darstellen, aufgrund derer die Wälder als

erneuerbare Ressource unter Beibehaltung des Artenreichtums und der strukturellen

Vielfalt bewirtschaftet werden können (Lindenmayer und Franklin 2002).

1.2.3 Hypothesen und Ziele

Die zentrale Hypothese dieser Arbeit geht davon aus, dass die Etablierung und

das Wachstum von N. betuloides- Jungwüchsen durch das Vorhandensein von

kleinflächigen Störungen beeinflusst wird. Diese Störungen führen dazu, dass sich ein

Gradient verändernder Strahlungsbedingungen vom ungestörten Bestandesinneren hin

zum Zentrum kleinerer Lücken entwickelt.

Die Hauptziele dieser Studie sind:

1. Die natürlichen kleinflächigen Störungen im Kronendach in N. betuloides -

Wäldern werden charakterisiert,

2. die Dynamik von Lücken im Kronendach von N. betuloides- Wäldern werden

analysiert,

3. die Kronendachstruktur wird mit der Strahlungsintensität im Bestandesinneren

in Beziehung gesetzt und mit Hilfe von vier unterschiedlichen Programmen

analysiert,

4. die Zusammenhänge zwischen Struktur des Kronendachs, Bestandesparametern

und Unterschieden im Strahlungshaushalt des Bestandesinneren von N.

betuloides-Wäldern werden erklärt,


18

5. der Effekt natürlicher, kleinflächiger Störungen wird untersucht, hierbei auf

a. die Strahlungsverhältnisse im Bestandesinneren,

b. die Verjüngungsmuster (Dichte und Wachstumsraten) von N. betuloides

c. das Dickenwachstum junger, um die Störungsfläche herum stehender

Bäume.

1.2.4 Untersuchungsgebiete

Die Untersuchungsgebiete liegen am „Estancia Olguita“ an der südöstlichen

Seite des Río Cóndor (53 ° 59 ’ S, 69 ° 58 ’ W) sowie an der südwestlichen,

chilenischen Seite von Tierra del Fuego. Bis heute wird Waldweide im „Río Cóndor

Tal“ praktiziert, wo die niedrigen Flussauen stark durch den Menschen beeinflusst sind

(Arroyo et al. 1996). In den Untersuchungsgebieten wurden jedoch keine Hinweise auf

Verbiss durch Weidetiere gefunden.

1.2.4.1 Waldvegetation

Die Feldaufnahmen wurden in zwei primären, ungleichaltrigen Althölzern

durchgeführt. In beiden Beständen fanden sich keine Hinweise auf menschliche

Einflüsse. Der annähernd natürliche Zustand dieser Wälder erlaubte es, die Einflüsse

von Störungen im Kronendach auf den Strahlungshaushalt, die Verjüngungsmuster und

das Dickenwachstum junger N. betuloides-Bäume zu untersuchen. Zwei Waldbestände

mit einer Fläche von jeweils 20 ha wurden ausgewählt und im Detail untersucht. Es

handelte sich um einen reinen N. betuloides- Wald und einen gemischten N. betuloides

– N. pumilio- Bestand. In dem Reinbestand wurden alle Untersuchungsansätze verfolgt

und alle Methoden angewandt. Im Mischwald wurden die kleinflächigen Störungen

sowie Baumentwicklungsprozesse charakterisiert.

Der reine N. betuloides- Bestand erwies sich als relativ artenarm. In der

Strauchschicht fanden sich nur wenige Arten. Dominiert wurde die Strauchschicht

durch den hochwüchsigen Strauch Berberis ilicifolia. Weniger häufig und mit einer

geringeren Deckung kamen B. buxifolia und die kleineren Sträucher Pernettya

mucronata und P. pumila vor. In der Krautschicht fanden sich ebenfalls nur wenige

Arten. Die häufigsten Arten hier waren Adenocaulon chilense, Luzuriaga marginata,

Senecio acanthifolius, Rubus geoides, Uncinia lechleriana, die Farne Blechnum penna-


19

marina, Asplenium dareoides, Grammitis magellanica, und Hautfarne, insbesondere

Hymenophyllum secundum und H. tortuosum. Eine mächtige Moosschicht überzog den

Boden und vermodenrdes Todholz. Dominierende Arten der Moosschicht waren

Dicranoloma chilense, D. robustum und Pyrrhobryum mnioides sowie die Lebermoose

Gakstroemia magellanica, Plagiochila obovata und Chiloscyphus magellanica.

Ewähnenswerte Flechten waren Cladonia sp. auf dem Boden sowie Pseudocyphellaria

sp. auf Moderholz.

Der N. betuloides–N. pumilio-Mischwald kann als ein Übergang zwischen den

immergrünen Nothofagus-Wäldern und den Nothofagus-Laubwäldern angesehen

werden. N. betuloides dominierte vor allem auf den nässeren Standorten. Die Deckung

der Strauchschicht war niedrig. Dominierende Arten in der Strauchschicht waren B.

ilicifolia und P. mucronata. Es fanden sich weniger Hautfarne und Moose als in den

reinen N. betuloides-Wäldern (Young 1972, McQueen 1976, Pisano 1977, Gajardo

1994, Luebert und Pliscoff 2006).

1.2.4.2 Standortbedingungen

Das Untersuchungsgebiet befindet sich in der nördlichen antiborealen Sub-Zone.

Die Durchschnittstemperatur im Untersuchungsgebiet liegt zwischen 9,0 und 9,5 °C im

wärmsten Monat des Jahres. Im kältesten Monat bleibt die Temperatur über 0 °C. Es

fallen zwischen 500 und 600 mm Niederschläge pro Jahr, in Ausnahmefällen bis zu 900

mm. Der Wind weht gewöhnlich aus west bis südwest, mit durchschnittlichen

Windgeschwindigkeiten zwischen 14 und 22 km h-1. Die maximale

Windgeschwindigkeit liegt weit über 100 km h-1 (Tuhkanen 1992).

Von Frederiksen (1988) wurde das Untersuchungsgebiet der subalpinen Zone

zugeordnet. Das Relief ist charakterisiert durch während der Eiszeiten entstandene U-

Täler, die parallel zu den Anden verlaufen. Für die N. betuloides-Wälder wurden zwei

bodenbildende Prozesse beschrieben, welche auch im Untersuchungsgebiet zu erkennen

waren. Es handelt sich um Podsolierungsvorgänge auf gut entwässernden Standorten

sowie um hydromorphologische Prozesse auf stauwasserbeeinflussten Böden (Pisano

1977, Puigdefábregas et al. 1999, Gerding und Thiers 2002, Romanyà et al. 2005). Die

Böden sind in der Regel flachgründig (< 50 cm), lehmig, sauer (pH 3,4 – 5,5) sowie

nährstoffarm (Gerding und Thiers 2002, Romanyà et al. 2005, Thiers und Gerding

2007). Eine Akkumulation großer Humusauflagen auf dem Boden sowie die


20

Ansammlung großer Mengen an liegendem Todholz wurden für die Waldtypen in der

Tierra del Fuego bereits früher beschrieben (Gutiérrez et al. 1991). Die meisten Wurzeln

als auch die Nährelemente befinden sich im oberen Teil der dicken Rohhumusauflage in

einer Bodentiefe von nur 4 bis 10 cm (Pisano 1977, Gutiérrez et al. 1991, Gerding und

Thiers 2002, Romanyà et al. 2005).

1.2.5 Datenaufnahme

Die Geländeaufnahmen wurden in zwei Phasen durchgeführt. Die erste Phase

dauerte von Januar bis März 2006. In der ersten Phase wurden die

Untersuchungsbestände ausgewählt sowie 36 parallele Transekte (300 m) im Abstand

von 30m zueinander markiert. In den Transekten wurden die Baumarten, die das

Kronendach bilden, ermittelt, die Waldstruktur mit der „Point Centred Quarter Method“

erhoben und die Lücken im Kronendach aufgenommen. Alle Merkmale der Lücken im

Kronendach wurden gemessen. Es wurden Zuwachsbohrungen an einigen Bäumen

durchgeführt, um den Einfluss der Störungsdynamik auf das Zuwachsverhalten junger

Bäume abzuschätzen. 13 Lücken im Kronendach wurden ausgewählt, um den Effekt der

Lücken auf die Strahlungsverhältnisse im Bestandesinneren sowie auf die Verjüngung

der reinen N. betuloides-Wälder zu untersuchen. Transekte wurden vom Zentrum der

Lücken bis in den angrenzenden Bestand mit ungestörtem Kronendach angelegt.

Hemisphärische Fotografien wurden aufgenommen und Probepunkte für die

Sämlingsuntersuchungen ausgewählt.

Die zweite Phase der Geländeaufnahmen dauerte von Januar bis Februar 2007.

Der Schwerpunkt lag auf der Erhebung der Struktur des Kronendachs und den

Bestandesparametern sowie deren Auswirkungen auf die Strahlungsverhältnisse in

reinen N. betuloides-Wäldern. Es wurden erneut hemisphärische Fotografien

aufgenommen und Bestandesmessungen in 225 m2 großen Probepunkten durchgeführt.

Mit dem Ziel, die Struktur des Kronendachs mit der Strahlung im

Bestandesinneren zu vergleichen, und diese dann mit vier unterschiedlichen

Programmen zu berechnen, wurden im April 2007 zwei Sets hemisphärischer

Fotografien mit jeweils 26 Bildern in Laubwäldern aufgenommen und durch Vergleiche

mit entsprechenden Fotografien aus anderen Waldtypen der Erde analysiert. Eine

Bildreihe, aufgenommen von BUTLER-MANNING, stammt vom Weberstedter Holz im

Hainich Nationalpark in Deutschland (51 ° 01 ’ N, 10 ° 04 ’ E). Es handelt sich um


21

einen temperaten Buchenmischwald auf Kalkgestein. Die zweite Bildreihe,

aufgenommen von DURÁN RANGEL, stammt aus einem sehr feuchten submontanem

tropischen Nebelwald im „Sierra de Lema Forest“ (Canaima National Park) in

Venezuela (05 ° 53 ’ N, 61 ° 26 ’ W).

1.2.6 Struktur der Doktorarbeit

Die folgenden Kapitel der Doktorarbeit wurden entweder zur Publikation in

wissenschaftlichen Zeitschriften eingereicht oder publiziert:

Kapitel 2 Nothofagus betuloides (Mirb.) Oerst 1871 (FAGALES:


Fuego

Publiziert bei Anales del Instituto de la Patagonia 36: 53-68

Promis A, Cruz G, Reif A, Gärtner S

Das Ziel dieser Arbeit war die Erstellung eines Reviews über die Biologie und

die Ökologie von N. betuloides. Des Weiteren sollten die Pflanzengesellschaften, in

denen N. betuloides auftritt, die Walddynamik, die Waldnutzung in der Vergangenheit,

der Gegenwart und der Zukunft charakterisiert werden. Hierfür wurde der aktuelle

Stand des Wissens zusammen getragen.

Kapitel 3 Small-scale natural disturbances and tree development processes in


Tierra del Fuego

Eingereicht

Promis A, Gärtner S, Reif A, Cruz G

Das Ziel der Untersuchung war der Vergleich natürlicher Kronendachlücken in

reinen N. betuloides Wäldern mit denen in gemischten N. betuloides – N. pumilio-

Wäldern. Die Lückendynamik sowie der Einfluss von Störungen auf das

Dickenwachstum junger Bäume wurden analysiert. Insbesondere zwei Fragen sollte

nachgegangen werden: Ist die durch die Entstehung von kleinen Lücken induzierte

Wachstumsreaktion junger Bäume in Rein- und Mischbeständen gleich? Unterscheiden


22

sich in Mischbeständen N. pumilio und N. betuloides in ihre Reaktion? Allgemein sollte

ein tieferes Verständnis für die natürliche Dynamik dieser Waldbestände in Tierra del

Fuego geschaffen werden.

Kapitel 4 Comparison of canopy structures and solar radiation transmittances



Eingereicht

Promis A, Gärtner S, Butler-Manning D, Durán-Rangel C, Reif A, Cruz

G, Hernández L

Es wurde bereits eine Vielzahl unterschiedlicher Studien durchgeführt, in denen

indirekt mit Hilfe hemisphärischer Photografie die Bestandesstruktur sowie die

Strahlungsverhältnisse im Bestandesinneren beschrieben worden sind. Zur Auswertung

hemisphärischer Photografien stehen kommerziell vertriebene und im Netz frei

verfügbare Programme zur Verfügung. Die kostenlose Verfügbarkeit ist ein Vorteil der

Open Source Software. Es stellt sich jedoch die Frage, ob darüber hinaus auch

Unterschiede in der Nutzerfreundlichkeit sowie den berechneten Ergebnissen zwischen

den verschiedenen Programmen bestehen. Bis heute wurde wenig über die Unterschiede

der verschiedenen Programme geschrieben, insbesondere aus der Sicht eines Ökologen

oder Forstwissenschaftlers. Das Ziel dieser Arbeit war, die am häufigsten verwendeten

Kronendach- und Strahlungsparameter (Kronenschluss, Blattflächenindex, direkte und

indirekte Strahlung, Globalstrahlung) mit Hilfe hemisphärischer Fotografien zu

ermitteln. Es wurden Bestände mit unterschiedlichem Kronenschluss (durch Lücken

unterbrochener Kronenschluss, geschlossenes Kronendach) und Bestände

unterschiedlicher Breitengrade (nördliche und südliche Hemisphäre sowie am Äquator)

untersucht.

Kapitel 5 Effects of canopy structure and stand parameters on the variability



Eingereicht

Promis A, Schindler D, Reif A, Cruz G


23

Bisher ist über die Auswirkung der Struktur des Kronendachs und der

Bestandesparameter auf die räumliche Variabilität der Strahlungsverhältnisse in N.

betuloides- Beständen wenig bekannt. Aus diesem Grund war es Ziel dieser Studie, den

Effekt des Kronenschlusses auf die zum Waldboden durchdringende Strahlung zu

beschreiben. Zusätzlich sollte untersucht werden, ob Kronendachstruktur und

Bestandesparameter die Variabilität der Strahlungsverhältnisse in einem

ungleichaltrigen N. betuloides- Bestand erklären.

Kapitel 6 Effects of natural small-scale disturbances on understorey light and



Eingereicht

Promis A, Gärtner S, Reif A, Cruz G

Untersucht wurden die Auswirkungen kleinflächiger Unterbrechungen im

Kronendach auf die Verjüngungsdynamik in N. betuloides- Wäldern. Kleine Lücken im

Kronendach führen zu Unterschieden der Lichtversorgung im Umfeld der Lücken. Es

stellt sich die Frage, wie stark diese unterschiedlichen Strahlungsverhältnisse die Dichte

und das Wachstum junger Bäume im Unterstand beeinflussen. Indirekt hängen davon

die Ausprägung der Bodenvegetation, ja sogar das Äsungsverhalten des Guanaco (Lama

guanicoe) ab. Weitere Forschungsfragen sind die Verteilung junger Bäume in N.

betuloides-Wäldern im Zusammenhang mit Bestandeslücken.

1.2.7 Ergebnisse

Die untersuchten Lücken im Kronendach waren relativ klein. So hatten die

Lücken im reinen N. betuloides-Bestand eine durchschnittliche Größe von 51 m². Die

Lücken in dem untersuchten N. betuloides – N. pumilio- Bestand hatten eine

durchschnittliche Größe von 107 m². Nur 2 % der gesamten Kronendachfläche waren

Lücken, 4,6 % konnten als erweiterte Lücken („expanded gaps“) eingestuft werden. Im

untersuchten Mischwald waren die Lücken 2,6- und 2,7-mal größer. Ein Ansteigen der

Häufigkeit von Störungen konnte für die Periode von 1920 bis 1930 beobachtet werden.

Die durchschnittliche Zahl an geworfenen Bäumen, durch welche die Lücken

entstanden sind, war im Reinbestand 2,2 und im Mischwald 2,5. Im Reinbestand


24

entstanden die Lücken meistens durch das Abbrechen einzelner Bäume, im Mischwald

dagegen durch die Entwurzelung von Bäumen.

Die größte Veränderung im radialen Zuwachsverhalten nach Freistellung konnte

für N. pumilio (18,4 %) und N. betuloides (11,2 %) im Mischwald festgestellt werden.

Der geringste Zuwachs wurde (bei kleineren Lücken!) bei N. betuloides (2,4 %) im

Reinbestand ermittelt werden.

Die Analyse der hemisphärischen Fotografien mit einem kommerziellen

vertriebenen (HemiView) und drei frei verfügbaren Software-Programmen (Gap Light

Analyzer, hemIMAGE and Winphot) zeigte, dass alle verwendeten Programme ähnliche

Kronendachstrukturen sowie Strahlungsverhältnisse für das Bestandesinnere

berechneten. Nur die Ergebnisse des Blattflächenindex, welche mit einem elliptischen

Blatt – Winkel – Verteilung („ellipsoidal leaf angle distribution“) für Lücken im

Kronendach berechnet wurde, zeigte eine schwache Korrelation. Die Kosten zur

Analyse hemisphärischer Photografien lassen sich also durch die Verwendung frei

verfügbarer Software, welche aus dem Internet herunter geladen werden kann,

reduzieren.

Die Höhe der Einstrahlung im Bestandesinneren des N. betuloides-Waldes

wurde durch eine hohe horizontale und vertikale Heterogenität im Bestandesaufbau

beeinflusst. Hinzu kommt der niedrige Winkel der Sonne während der

Vegetationsperiode (Oktober bis März). Die direkte Einstrahlung schien räumlich und

zeitlich stark zu variieren, während die indirekte Einstrahlung weitgehend konstant

blieb.

Die direkte Einstrahlung während der Vegetationsperiode zeigte eine geringe

Korrelation mit der Kronendachstruktur und den Bestandesparametern. Hierbei zeigte

der Blattflächenindex die beste Korrelation mit der Summe der direkten Einstrahlung

(R² = 0,263). Obwohl die diffuse und globale Strahlung kaum mit einfachen

Bestandesparametern korreliert werden konnte, besteht zwischen der Strahlungssumme

und dem Kronenschlussgrad ein starker Zusammenhang (R² = 0,963 und 0,833). Die

Kronendach- und Bestandesparameter weisen eine große räumliche Heterogenität auf,

welche die diffuse und globale Strahlung im Bestandesinneren beeinflusst. Durch die

Kombination der Grundfläche, der Kronenschirmfläche, des Kronenvolumens, des

Stammvolumens und dem durchschnittlichen Kronenradius in einem Modell konnten 75

% und 73 % der Variabilität der diffusen und globalen Strahlung erklärt werden. Die

Verfügbarkeit der nicht-Kosinus-korrigierten direkten, diffusen und globalen


25

Sonnenstrahlung, welche in den Reinbestand eindringen konnte, reichte von 3,2 bis 19,4

%, von 3,1 bis 16,7 % und von 3,2 bis 17,6 %.

Die Sämlinge und Jungwüchse von N. betuloides zeigten die größte

Schattentoleranz. N. betuloides verjüngt sich sogar unter sehr schattigen Bedingungen

und ist bei der Verjüngung offensichtlich nicht auf das Vorhandensein von Lücken im

Kronendach angewiesen. Dies führt zu einem kontinuierlich ablaufenden

Verjüngungsprozess in N. betuloides-Wäldern. Die Strahlungsmenge im

Bestandesinneren korrelierte weder mit dem relativen Dickenwachstum noch mit dem

relativen Zuwachs junger Bäume. Das Wachstum der jungen Bäume korrelierte nur mit

deren Alter. Inverse polynomische Kurve erklären 70% der Variabilität des radialen

Wachstums und 50 % der Variabilität des relativen Höhenzuwachses.

Der Anteil von durch Lama guanicoe verbissenen Sämlingen war sehr niedrig

(0,7-2,8 %). Der Verbiss junger Bäume war unter Lücken im Kronendach ebenso häufig

wie unter dem geschlossenen Kronendach.

Das heterogene Kronendach in dem primären, ungleichaltrigen N. betuloides-

Altholz mit nur sehr kleinen Öffnungen im Kronendach führte zu einem Mosaik

unterschiedlich alter und großer Jungwüchse im Unterstand.

Fazit: Die Ergebnisse deuten darauf hin, dass N. betuloides eine relativ hohe

Schattentoleranz aufweist. Die Baumart kann um sehr kleine Lücken herum mit

einem geringen Radialzuwachs ausharren und auf plötzliche, größere Störungen

im Kronendach mit erhöhtem Zuwachs reagieren.

1.3 Resumen

1.3.1 Introducción: Perturbación y Ecología de la Regeneración de Especies

Arbóreas en los Bosques

Las comunidades naturales de plantas son sistemas dinámicos que se encuentra

heterogéneamente distribuidas en el espacio (Sousa 1984). Las perturbaciones han sido

consideradas fuentes de gran impacto, las que afectan las estructuras de estas

comunidades así como también la composición de especies (White 1979, Sousa 1984),

pues modifican variables físicas y estructurales en los ecosistemas (White y Jentsch


26

2001). Así, la descripción de los regimenes de perturbaciones es un importante paso

hacia el entendimiento de la dinámica natural de las comunidades (White et al. 1999).

No todas las perturbaciones son parecidas, pues ellas difieren en patrones

espaciales (tamaño y forma), parámetros temporales (frecuencia, período de rotación,

estacionalidad), especificidad (probabilidad de perturbaciones por especies, edad, y

clases de tamaño), y magnitud (intensidad y severidad) (Sousa 1984, White et al. 1999).

Comunidades boscosas están frecuentemente sujetas a perturbaciones (Oliver y

Larson 1996). Los efectos de las perturbaciones naturales en la dinámica y estructura de

los bosques han sido estudiados en bosques tropicales, templados y boreales (Platt y

Strong 1989, Denslow y Spies 1990, McCarthy 2001, Kuuluvainen 2002).

Perturbaciones de gran escala son más severas que las de pequeña escala, sin

embargo son menos frecuentes (Sousa 1984, Turner et al. 1998). Perturbaciones de

pequeña escala, tales como los claros de dosel producidos por la muerte de uno o varios

árboles, pueden ser más frecuentes en ecosistemas forestales en fases maduras (Oliver y

Larson 1996) y afectan un área más grande en el tiempo (Spies et al. 1990).

Por lo tanto en ecosistemas forestales, la severidad de las perturbaciones influye

en las estructuras de rodal, en la composición de especies, en las tasas de crecimiento de

los árboles sobrevivientes, en la dinámica de la regeneración, y en la diversidad de

especies (Connell 1978, Canham et al. 1994, Oliver y Larson 1996).

La fase de desarrollo de un bosque afectado por perturbaciones de claros de

dosel de pequeña escala ha sido descrita como fase de claros (Watt 1947), donde la

regeneración de los árboles es estimulada por la muerte de uno o más árboles del dosel

arbóreo, llegando a ocupar el espacio y finalmente uno o varios de los árboles juveniles

remplazarán el o los árboles del dosel previamente muertos (Busing y Brokaw 2002).

Espacios dejados por claros de dosel grandes presentan una mayor probabilidad de ser

ocupados por especies arbóreas intolerantes a la sombra, mientras que los claros de

dosel pequeños son frecuentemente ocupados por especies arbóreas tolerantes a la

sombra (Grubb 1977).

En los bosques, las perturbaciones y la ocurrencia de claros de dosel modifican

la disponibilidad de recursos (luz, agua, nutrientes del suelo) requeridos para el

crecimiento de las plantas (Canham y Marks 1985). Así, se puede encontrar un

gradiente de condiciones de sitio diferentes, desde el centro de un área perturbada hasta

el bosque de las cercanías que no presenta signos de perturbación (Riklefs 1977,

Denslow 1980). Además, se alteran las condiciones del substrato, y nuevas condiciones


27

son creadas en el piso del bosque, particularmente cuando los árboles muertos son

descales o se produjeron porque el tronco se quebró (Collins et al. 1985).

El cambio más obvio debido a los claros de dosel puede ser observado en el

ambiente de radiación solar bajo el dosel arbóreo (Canham et al.1990, Gray et al. 2002).

La comprensión de los efectos de la radiación solar en el interior del bosque es

importante para entender las dinámicas forestales, pues la radiación solar afecta los

patrones de regeneración de las plantas, tales como la germinación, el establecimiento,

el crecimiento y la sobrevivencia (Grant 1997). Diferencias en la radiación solar dentro

de claros de dosel y bajo doseles cerrados han sido descritas para diferentes tipos

forestales (Denslow 1980, Canham et al. 1990, de Freitas y Enright 1995, Gray et al.

2002).

Consiguientemente ha habido mucho interés en medir la radiación solar bajo el

dosel arbóreo. Varios instrumentos han sido desarrollados para medir directa o

indirectamente el ambiente de la radiación solar en el interior de un bosque. Ecologistas

y silvicultores prefieren muchas veces aproximaciones indirectas para estimar la

radiación solar, debido a las dificultades inherentes en la medición de la radiación solar

directamente (Jennings et al. 1999). Fotografías hemisféricas han sido una herramienta

ampliamente usada para calcular indirectamente el ambiente de la radiación solar en el

interior de los bosques. Varios estudios, que han utilizado esta herramienta, han

mostrado patrones similares de la radiación solar en el interior del bosque (Rich et al.

1993, Comeau et al. 1998, Gendron et al. 1998, Clearwater et al. 1999, Engelbrecht y

Herz 2001).

Por lo tanto, la heterogeneidad especial de microambientes en el bosque, referido

al nicho de regeneración, facilita el establecimiento de las plantas de regeneración de

diferentes especies arbóreas y promueve la coexistencia de especies (Grubb 1977).

Sin embargo, existe también una alta heterogeneidad bajo los doseles no

perturbados debido a las variaciones en el dosel, tales como la composición, la altura, la

espesura y la densidad del follaje (Lieberman et al. 1989, Veblen 1992). Particularmente

a altas latitudes, una gran y significante proporción de la radiación solar alcanza el piso

del bosque más allá de los límites de los claros del dosel, produciendo una más

homogénea distribución de la radiación solar dentro del rodal (Canham et al. 1990). Por

lo tanto, a altas latitudes, donde se presentan bosques como los de Nothofagus en el

hemisferio sur, la dinámica de la regeneración, especialmente de especies tolerantes a la


28

sombra, puede ser menos afectada por los mismos claros de dosel (Canham 1990,

Veblen 1992).

1.3.2 Los Bosques de Nothofagus betuloides

Nothofagus betuloides es una especie arbórea siempreverde endémica de los

bosques subantárticos de Chile y Argentina. En Chile, la superficie de bosques en los

que participa N. betuloides alcanza aproximadamente 2.739.906 ha, correspondiendo al

20,4 % de la superficie nacional de bosques (CONAF-CONAMA 1999a). De esta

superficie, 1.793.098 ha corresponden al tipo forestal siempreverde de N. betuloides y

946.808 ha conforman el subtipo forestal siempreverde - deciduo de N. betuloides – N.

pumilio. N. betuloides se presenta aproximadamente en 1.396.947 ha en la Patagonia

Austral y en Tierra del Fuego (53 % de la superficie total de bosques). Actualmente

estos bosques no se encuentran en peligro, pues al menos más del 69 % de la superficie

de bosques en la Patagonia Austral y en Tierra del Fuego se encuentran dentro del

Sistema Nacional de Áreas Silvestres Protegidas del Estado (SNASPE) o forman parte

de un parque natural privado en el lado chileno de Tierra del Fuego (Karukinka)

administrado por Wildlife Conservation Society (WCS) (Arroyo et al. 1996, CONAF-

CONAMA 1999b).

Perturbaciones naturales de gran y pequeña escala afectan los bosques de

Nothofagus en Tierra del Fuego. Estos bosques son modelados por la ocurrencia de

fuertes vientos (Rebertus et al. 1997, Puigdefábregas et al. 1999). Tormentas de viento

pueden causar la caída de la mayoría de los árboles de un rodal (Rebertus et al. 1997,

Puigdefábregas et al. 1999), y patrones de bandas de mortalidad en sitios predispuestos

a fuertes vientos han sido reportados para bosques puros de N. betuloides, de N.

pumilio, y también para bosques mixtos de N. betuloides - N. pumilio (Rebertus y

Veblen 1993b, Rebertus et al. 1993, Puigdefábregas et al. 1999). Claros de dosel de

tamaños menores a 200 m2 en superficie son creados por la caída de árboles producida

por la acción del viento (Rebertus y Veblen 1993a, Gutiérrez 1994). Sin embargo, en

bosques adultos, vírgenes y puros de N. betuloides en Tierra del Fuego, los discretos

claros de dosel pueden no ser aparentes, y estas aberturas ocurren generalmente como

un complejo de claros entretejidos en el dosel (Rebertus y Veblen 1993a). También, la

presencia de manchas de árboles juveniles colindantes a grandes parches de bosques


29

adultos ha formado una estructura de rodal multietánea, lo que resulta en un patrón de

mosaico de parches en estos bosques de Nothofagus (Gutiérrez et al. 1991).

La dinámica de regeneración de los bosques australes sudamericanos de

Nothofagus y la importancia de las perturbaciones en ella ha sido bien resumida por

Veblen et al. (1996) y Pollmann y Veblen (2004). Ellos demostraron que a bajas

elevaciones, en condiciones de clima templado, donde especies de Nothofagus compiten

con otras especies forestales tolerantes a la sombra, perturbaciones de gran escala

parecen ser importantes para la regeneración de Nothofagus. A elevaciones más altas, en

sitios de condiciones subóptimas, donde la riqueza de especies forestales es menor, las

especies de Nothofagus parecen ser capaces de regenerar tanto después de

perturbaciones de gran escala (deslizamientos de tierra, fuego, caída masiva de árboles

por efecto del viento), así como también después de la ocurrencia de perturbaciones de

pequeña escala o claros de dosel, donde pocos árboles mueren.

N. betuloides en Tierra del Fuego es capaz de establecerse en claros de dosel de

pequeña escala (Rebertus y Veblen 1993a, Gutiérrez 1994, Arroyo et al. 1996). Plantas

de regeneración, regeneración avanzada y árboles jóvenes presentes al momento de la

formación del claro, y que no han sido dañados ni muertos, son liberados por la creación

de esta perturbación en el dosel de pequeña escala (Veblen et al. 1996). Sin embargo, el

establecimiento y crecimiento de N. betuloides pueden ser impedidos si existe un alto

nivel de cobertura de árboles y arbustos en el sotobosque tales como Drimys winteri y

Maytenus magellanica (Rebertus et al. 1993a, Gutiérrez et al. 1991, Gutiérrez 1994,

Veblen et al. 1996).

Desde finales del siglo 19 bosques costeros y del interior de la Patagonia Austral

y de Tierra del Fuego han sido intervenidos a través de un sistema selectivo de corta de

árboles conocido como floreo. Con este sistema se cortaban selectivamente los mejores

árboles desde un punto de vista maderero, y eran dejados en pie aquellos árboles que

presentaban mala forma y con bajas calidades madereras, o simplemente el bosque era

quemado después de la corta (Martínez Pastur et al. 2000, Cruz et al. 2007a).

Actualmente la legislación forestal chilena especifica que los bosques

productivos de N. betuloides, desde un punto de vista maderero, deben ser manejados

bajo los sistemas silviculturales de cortas de protección o de selección (Donoso 1981),

ambos diseñados para promover regeneración natural después de la cosecha. Sin

embargo, muchos rodales todavía son intervenidos a través del floreo, o solamente ha

sido aplicada la corta de regeneración del sistema silvicultural de cortas de protección


30

(Cruz et al. 2007a). Sin embargo, la aplicación del sistema silvicultural de cortas de

protección promueve una homogenización y simplificación de la típica estructura

multietánea de estos bosques (Cruz et al. 2008).

Recientemente ha existido una tendencia a la intensificación de la utilización de

los bosques de N. betuloides. Nuevos estudios han sido llevados respecto a la ecología

de estos bosques, su distribución, silvicultura, propiedades de la madera y de su secado,

y de producción maderera industrial (ver Cruz y Caldentey 2007).

La comprensión de la dinámica natural de rodales incluyendo el rol de las

perturbaciones naturales debería proveer la base en la cual estos bosques deberían ser

manejados como un recurso renovable, preservando la diversidad de especies y la

riqueza estructural (Lindenmayer y Franklin 2002).

1.3.3 Hipótesis y Objetivos

La hipótesis central de este estudio es que el establecimiento y el crecimiento de

árboles juveniles de Nothofagus betuloides son afectados por la ocurrencia de

perturbaciones de pequeña escala, debido a la existencia de un gradiente de condiciones

de radiación solar bajo el dosel del bosque, el que se extendería desde los claros de

dosel hasta bajo el dosel del bosque no perturbado.

Los principales objetivos del estudio son:

1. caracterizar las perturbaciones de pequeña escala de claros de dosel en bosques

dominados por N. betuloides,

2. analizar la dinámica de claros de dosel en bosques dominados por N. betuloides,

3. comparar variables de estructura de rodal y de transmisión de radiación solar

bajo el dosel del bosque usando cuatro programas diferentes para el análisis de

fotografías hemisféricas,

4. analizar como las estructuras del dosel y los parámetros de rodal explican la

variación en la radiación solar bajo el dosel de un bosque de N. betuloides

5. determinar los efectos de las perturbaciones de pequeña escala en:

a. las condiciones de radiación solar bajo el dosel del bosque,

b. los patrones de la regeneración (densidad y tasas de crecimiento) de N.

betuloides,

c. las respuestas de crecimiento radial de árboles juveniles que se

encuentran creciendo al rededor de los claros de dosel.


31

1.3.4 Área de Estudio

El área estudio se localiza al suroeste del lado chileno de Tierra del Fuego, en la

“Estancia Olguita”, específicamente en la vertiente sureste del Río Cóndor (53 ° 59 ’ S,

69 ° 58 ’ W). Hasta la fecha todavía existe pastoreo en el valle del Río Cóndor, el que se

debe principalmente a ganado vacuno, donde han sido fuertemente impactadas las

terrazas ribereñas bajas del río (Arroyo et al. 1996). Sin embargo en el área de estudio

no fueron encontradas evidencias de pastoreo por ganado.

1.3.4.1 Vegetación

El estudio fue llevado a cabo en dos bosques primarios y maduros que

presentaron una estructura multietánea, y no evidenciaban alteración antrópica. Así, el

estado natural de estos bosques nos permiten estudiar el efecto de las perturbaciones en

el dosel arbóreo sobre la radiación solar en el interior del bosque, los patrones de la

regeneración y las respuestas del crecimiento radial de los árboles juveniles de N.

betuloides. Dos bosques, de aproximadamente 20 ha cada uno, fueron estudiados, un

bosque puro de N. betuloides y un bosque mixto siempreverde-deciduo de N. betuloides

- N. pumilio. En el bosque puro se llevo a cabo completamente el trabajo. Mientras que

en el bosque mixto fueron solamente caracterizadas las perturbaciones de pequeña

escala y los procesos de desarrollo de los árboles.

El bosque puro de N. betuloides tenía una baja riqueza de especies de plantas.

Pocas especies arbustivas fueron encontradas, siendo la más frecuente y abundante en

cobertura el arbusto Berberis ilicifolia. Menos frecuentes y con menores coberturas

fueron encontradas B. buxifolia, Pernettya mucronata, y P. pumila. La riqueza de

especies de plantas vasculares que crecen en el piso del bosque también fue baja, siendo

las más frecuentes Adenocaulon chilense, Luzuriaga marginata, Senecio acanthifolius,

Rubus geoides, Uncinia lechleriana, y los helechos Blechnum penna-marina, Asplenium

dareoides, Grammitis magellanica, y los helechos película, mayoritariamente

Hymenophyllum secundum e H. tortuosum. Sin embargo, fue encontrada sobre el suelo

y madera en descomposición una abundante y profunda capa de musgos y hepáticas, la

que estaba dominada por las especies de musgos Dicranoloma chilense, D. robustum, y

Pyrrhobryum mnioides, y las hepáticas Gakstroemia magellanica, Plagiochila obovata,

y Chiloscyphus magellanica. Digno de mencionar son los líquenes como los del género


32

Cladonia sp. los que fueron encontrados creciendo sobre el suelo y los del género

Pseudocyphellaria sp. sobre madera en descomposición.

El bosque mixto siempreverde-deciduo de N. betuloides – N. pumilio puede ser

considerado como una transición entre los bosques de Nothofagus siempreverdes y

deciduos. N. betuloides domina los sitios más húmedos y pobremente drenados. La

cobertura arbustiva en el interior del bosque fue baja, dominada por B. ilicifolia y P.

mucronata. El sotobosque presentó menor cantidad de helechos películas (familia

Hymenophyllaceae) y de especies de briófitas que el bosque puro de N. betuloides

mencionado anteriormente (Young 1972, McQueen 1976, Pisano 1977, Gajardo 1994,

Luebert y Pliscoff 2006).

1.3.4.2 Condiciones del sitio

Climáticamente el área de estudio se inserta en la subzona Antiboreal del norte,

que presenta una temperatura media del aire en el mes más cálido de entre 9.0 - 9.5 °C y

para el mes más frío mayor a 0 ºC. La precipitación anual promedio es

aproximadamente entre 500 y 600 mm, pero puede alcanzar valores de hasta 900 mm.

Los vientos generalmente provienen del oeste y suroeste, registrándose velocidades

promedio de entre 14 y 22 km h-1 y velocidades máximas sobre 100 km h-1 (Tuhkanen

1992).

El área de estudio es parte de la zona subalpina definida por Frederiksen (1988).

El relieve es caracterizado por valles que corren paralelos a los Andes, los que fueron

profundamente modificados por la glaciación. Dos tipos de suelos han sido descritos

para los bosques de N. betuloides, los que también fueron encontrados en el área de

estudio. Suelos podzólicos en los sitios bien drenados, y suelos muy húmedos en áreas

mal drenadas y con condiciones de anegamiento moderadas a altas (Pisano 1977,

Puigdefábregas et al. 1999, Gerding y Thiers 2002, Romanyà et al. 2005). Los suelos

son normalmente delgados (< 50 cm), de textura arcillosa, ácidos (pH 3.4-5.5) y no muy

fértiles (Gerding y Thiers 2002, Romanyà et al. 2005, Thiers y Gerding 2007). Una

enorme acumulación de materia orgánica en el suelo y grandes cantidades de madera en

descomposición han sido descritos para este tipo forestal en Tierra del Fuego (Gutiérrez

et al. 1991). La mayoría de las raíces de las plantas y de los nutrientes están localizadas

en la parte baja de una gruesa capa de humus, a una profundidad aproximada en el suelo


33

de entre 4 y 10 cm (Pisano 1977, Gutiérrez et al. 1991, Gerding y Thiers 2002,

Romanyà et al. 2005).

1.3.5 Toma de Datos

Se llevaron a cabo en total dos períodos de toma de datos en terreno.

El primer período de trabajo de campo se prolongó desde Enero a Marzo de

2006. En este período se seleccionaron los rodales y se trazaron 36 transectos paralelos

de 300 m a intervalos de 30 m cada uno. Los transectos fueron usados tanto para hacer

la descripción de la estructura de los bosques y de la composición de especies a través

del método de los cuartos, así como también para registrar los claros de dosel. Todas las

características de los claros de dosel encontrados fueron medidas. Tarugos de

incremento fueron tomados a árboles juveniles para posteriormente evaluar la dinámica

de las perturbaciones y las respuestas de crecimiento radial de los árboles. Un total de

13 claros de dosel fueron seleccionados para estudiar sus efectos sobre la radiación solar

en el interior del bosque, y sobre la regeneración en el bosque puro de N. betuloides. Un

transecto fue trazado en cada uno de los claros de dosel. Estos transectos iban desde el

centro del claro de dosel hasta el interior del bosque adyacente no perturbado. Las

fotografías hemisféricas fueron tomadas, y también fueron instaladas las parcelas para el

muestreo de las plantas de regeneración.

El segundo trabajo de campo se prolongó desde Enero a Febrero de 2007. La

toma de datos se enfocó en el estudio de los efectos de las estructuras del dosel y los

parámetros de rodal en la variabilidad de la transmisión de radiación solar en el bosque

puro de N. betuloides. Las fotografías hemisféricas fueron tomadas y las mediciones de

rodal fueron hechas en parcelas fijas de 225 m2.

Además, durante Abril de 2007 dos juegos de 26 fotografías hemisféricas,

tomadas en otros bosques de latifoliadas, fueron pedidas prestadas. Esto con el ánimo de

comparar las variables de estructuras de rodal y transmisiones de radiación solar en el

interior del bosque estimadas usando cuatro programas diferentes para el análisis de las

fotografías hemisféricas. Un juego de fotografías hemisféricas fue tomado en un bosque

mixto deciduo dominado por Fagus sylvatica, creciendo en suelo de piedra caliza,

localizado en el Parque Nacional de Weberstedter Holz en Alemania (51 ° 01 ’ N, 10 °

04 ’ E). El otro juego de fotografías hemisféricas fue tomado en un bosque tropical


34

submontano nublado, muy húmedo, localizado en el Bosque de Sierra de Lema (Parque

Nacional Canaima) en Venezuela (05 ° 53 ’ N, 61 ° 26 ’ W).

1.3.6 Estructura de la Tesis

Los siguientes capítulos de la tesis fueron publicados o sometidos a publicación

a revistas científicas.

Capítulo 2 Nothofagus betuloides (Mirb.) Oerst 1871 (FAGALES:


Fuego

publicado en Anales del Instituto de la Patagonia 36: 53-68

Promis, A., Cruz, G., Reif, A., Gärtner, S.

El ánimo de este trabajo fue revisar los aspectos biológicos y ecológicos de N.

betuloides, así cómo también las características de los tipos vegetacionales donde

ocurre, estructuras de los bosques, dinámica de los bosques, y uso de los bosques en el

pasado, el presente y el futuro. Por lo tanto, fue revisado el presente estado del arte de la

especie, con especial énfasis en su distribución austral.

Capítulo 3 Small-scale natural disturbances and tree development processes in


Tierra del Fuego

sometido


Los objetivos de este estudio fueron comparar las características de los claros de

dosel formados naturalmente en un bosque puro de N. betuloides y en un bosque mixto

de N. betuloides – N. pumilio, analizar la dinámica de los claros y el rol de las

perturbaciones naturales en el crecimiento radial de los árboles juveniles. Dos preguntas

fueron establecidas. ¿Son similares las respuestas de los árboles juveniles de N.

betuloides que crecen en bosques puros y mixtos después de la creación de pequeños

claros de dosel? ¿En bosques mixtos, las respuestas de N. pumilio y N. betuloides son


35

diferentes? La idea fue la de entregar una profunda comprensión de la dinámica natural

de estos tipos de bosques den Tierra del Fuego.

Capítulo 4 Comparison of canopy structures and solar radiation transmittances



sometido

Promis, A., Gärtner, S., Butler-Manning, D., Durán-Rangel, C., Reif, A.,

Cruz, G., Hernández, L.

Existen muchas investigaciones en las que se han usado las fotografías

hemisféricas para estimar indirectamente las estructuras del bosque y el ambiente de

radiación solar bajo el dosel arbóreo. Para el análisis de las fotografías hemisféricas

existen programas computacionales gratis y otros que deben ser comprados. El costo de

inversión podría representar una ventaja de los programas computacionales que son

gratis sobre los que deben ser comprados. Sin embargo, ¿existen diferencias en la

utilización y los resultados obtenidos por los diferentes tipos de programas

computacionales? Además, poco ha sido documentado respecto a las diferencias en los

resultados y en las aplicaciones técnicas de los programas computacionales desde un

punto de vista del usuario (ecologistas y silvicultores). El objetivo fue comparar los

resultados de las variables usadas comúnmente, de estructuras de dosel (abertura de

dosel e índice de área de planta) y transmisiones de radiación solar (directa, difusa y

global) en el interior del bosque, estimadas a partir de fotografías hemisféricas usando

cuatro programas computacionales, en dos condiciones de dosel de copa (claros de

dosel y doseles cerrados), y en bosques localizados a diferentes latitudes (hemisferios

norte y sur y zona ecuatorial).

Capítulo 5 Effects of canopy structure and stand parameters on the variability



sometido

Promis, A., Schindler, D., Reif, A., Cruz, G.


36

Debido a la falta de conocimientos respecto a los efectos de las estructuras del

dosel y los parámetros de rodal en la variación espacial de la radiación solar bajo el

dosel de copas en bosques de N. betuloides, el objetivo fue analizar los efectos del dosel

en la transmisión de la radiación solar al piso del bosque, y evaluar si las estructuras del

dosel y los parámetros del rodal explican la variación en la radiación solar bajo el dosel

de copas en un bosque multietáneo de N. betuloides.

Capítulo 6 Effects of natural small-scale disturbances on understorey light and



sometido


¿Cuáles son los efectos de los claros de dosel de pequeña escala en las dinámicas

de la regeneración en el bosque de N. betuloides? No se sabe si estos pequeños claros de

dosel resultan en una disponibilidad de radiación solar diferente bajo el dosel de copas

entre áreas bajo claros de dosel y aquellas partes del dosel del bosque no perturbadas. Si

hay diferencias en las transmisiones de radiación solar, ¿éstas influyen la densidad y las

tasas de crecimiento de los árboles jóvenes en este bosque? ¿Los claros de dosel

influyen los hábitos de ramoneo de Lama guanicoe y la distribución de edades de los

árboles jóvenes en el bosque de N. betuloides? Todas estas preguntas fueron explicadas

con el estudio.

1.3.7 Resumen de los resultados

Los claros de dosel encontrados fueron pequeños, con un tamaño promedio de

51 m2 en el bosque puro de N. betuloides y 107 m2 en el bosque mixto de N. betuloides

– N. pumilio. Solo 2,0 % del dosel de copas del bosque puro correspondió a claros de

dosel, y 4,6 % a claros de dosel expandidos. En el bosque mixto las áreas del dosel en

claros fueron 2,6 y 2,5 veces más grandes, respectivamente. Un incremento en la

frecuencia de las perturbaciones fue estimado desde los años 1920-30.

El número promedio de árboles que han creado los claros de dosel fue de 2,2

árboles por claro en el bosque puro y 2,5 árboles por claro en el bosque mixto. El tipo


37

de daño más frecuente fue el de troncos quebrados en el bosque puro y de descalces en

el bosque mixto.

Los mayores cambios en incremento radial después de la liberación de períodos

de crecimiento restringido fueron registrados para N. pumilio (18,4 %) y N. betuloides

(11,2 %) creciendo en el bosque mixto, en cambio los menores se encontraron para N.

betuloides (2,4 %) en el bosque puro.

A partir del análisis de fotografías hemisféricas usando un programa

computacional comercial (HemiView) y tres programas computacionales gratis (Gap

Light Analyzer, hemIMAGE y Winphot) se demostró que todos los programas

calcularon estimaciones similares respecto a estructuras del dosel de copas y el medio

ambiente de radiación solar en el interior del bosque. Solo los resultados de índice de

área de planta efectiva medida con una distribución elíptica del ángulo de hoja,

calculado en claros de dosel, demostró una correlación baja. Por lo tanto, se podrían

reducir los costos asociados con los análisis de fotografías hemisféricas usando

programas computacionales gratis, los cuales pueden ser obtenidos a través de Internet.

La radiación solar transmitida bajo el dosel de copas en e bosque de N.

betuloides fue afectada por un alto nivel de variabilidad horizontal, una heterogeneidad

vertical del dosel de copas, y también por menores ángulos de la trayectoria del sol

durante la estación de crecimiento vegetacional (entre Octubre y Marzo). La radiación

solar directa bajo el dosel de copas mostró ser variable tanto en el espacio como

también en el tiempo, mientras la transmisión de la radiación solar difusa mostró menor

variabilidad.

El monto de radiación solar directa en la estación de crecimiento vegetativo

mostró una baja correlación con las estructuras del dosel y los parámetros de rodal,

siendo el índice de área de planta, la variable que mejor se ajusta a los datos (R2 =

0.263). Aunque pobremente correlacionadas con los parámetros de rodal, las

radiaciones solares difusa y global estimadas durante la estación de crecimiento

vegetativo fueron sensitivas a la variable de abertura del dosel (R2 = 0.963 y 0.833,

respectivamente). Una alta heterogeneidad en las estructuras espaciales del bosque

multietáneo fue encontrada, la que afectaría las radiaciones solares difusa y global en el

interior del bosque. 75 % y 73 % de las variaciones en las radiaciones solares difusa y

global, respectivamente, estimadas durante el período de crecimiento vegetativo, fueron

explicadas cuando fueron combinadas en un modelo las variables de área basal,

proyección de área de copa, volumen de copa y el radio equivalente de copa medio.


38

Las transmisiones de las radiaciones solares directa, difusa y global (estimadas

sin aplicación de ley del coseno) en el interior del bosque puro de N. betuloides fue de

entre 3,2 a 19,4 %, 3,1 a 16,7 %, y 3,2 a 17,6 %, respectivamente.

Las plantas de regeneración de N. betuloides mostraron una alta tolerancia a la

sombra, regenerando incluso bajo condiciones muy sombrías, y aparentemente no

requerirían de la presencia de grandes claros de dosel para establecerse. Esto ha

resultado en un proceso de regeneración del bosque más continuo

Las transmisiones de radiaciones solares no se correlacionaron con el

crecimiento radial relativo ni con el incremento en altura relativo de los árboles

juveniles. El crecimiento de los árboles jóvenes se correlacionó solamente con la edad

de las plantas, y una curva polinómica inversa explicó el 70 % de la varianza en el

crecimiento radial relativo y el 50 % de la varianza en el incremento en altura relativo.

Las proporciones de plantas de regeneración ramoneadas por L. guanicoe fueron

bajas (entre 0,7-2,8 % del total). El daño por ramoneo a las plantas de regeneración fue

observado tanto en claros de dosel como también bajo el dosel del bosque sin

perturbación, demostrándose que el animal no presenta un hábitat de preferencia.

El heterogéneo dosel del bosque maduro, multietáneo y primario de N.

betuloides, que solamente presentó claros de dosel muy pequeños, ha influido en que

exista una variedad de mosaicos en el interior del bosque, con plantas de regeneración

presentes en un amplio rango de edades y alturas.

Estos resultados sugieren que N. betuloides presenta una alta tolerancia a la

sombra. Esta especie puede crecer alrededor de claros de dosel muy pequeños, con

bajos crecimientos radiales, y sujeta a incrementos abruptos producto de la ocurrencia

de perturbaciones del dosel de copas más grandes.

Chapter 2 Nothofagus betuloides forests

39

CHAPTER 2

Nothofagus betuloides (Mirb.) Oerst 1871 (FAGALES:

NOTHOFAGACEAE) FORESTS IN SOUTHERN PATAGONIA

AND TIERRA DEL FUEGO

2.1 The Genus Nothofagus

Southern Patagonia and Tierra del Fuego have always been of great interest for

botanists because of their evergreen plant formations and biogeographical perspectives.

Antarctic families like the Winteraceae and genera like Gunnera and Gaultheria

represent the ancient Gondwana flora. The biogeography of Nothofagus, the southern

beech, has been the classical study that supports the sequence of Gondwana break-up

and the linkage of the Austral biota between South America and Australasia (Swenson

et al. 2001).

In the past the genus Nothofagus was included in the Fagaceae family

(Rodríguez and Quezada 2003), and seemed to be closely related to Fagus (Hill and

Dettmann 1996). More recent genetic studies have shown that the genus of Nothofagus

belongs to a new monogeneric family, the Nothofagaceae (Hill and Jordan 1993, Hill

and Dettmann 1996, Manos 1997, Jordan and Hill 1999).

There are currently 36 recognised Nothofagus species (Table 2.1), 26 of which

occur in Australasia and the remaining 10 in South America. Of the total number, 78 %

are evergreen and 22 % deciduous. The evergreen Nothofagus species can grow equally

well at tropical latitudes in the mountains of New Guinea (13 species, van Royen 1983),

and in subantarctic areas. Nothofagus betuloides (Mirb.) Oerst is the species with the

southernmost distribution. However, natural hybridisation between evergreen

Nothofagus species has been demonstrated in South America (Donoso and Atienza

1983, Donoso et al. 2004) as well as in New Zealand (Wardle 1984).


40

Table 2.1: Distribution and leaf characteristics of the Nothofagus species.

Species Distribution Leaves Species Distribution Leaves

Brassospora Fuscospora

N. aequilateralis New Caledonia Evergreen N. alessandri Chile Deciduous

N. balansae New Caledonia Evergreen N. fusca New Zealand Evergreen

N. baumanniae New Caledonia Evergreen N. gunnii Australia Deciduous

N. brassii New Guinea Evergreen N. solandri New Zealand Evergreen

N. carrii New Guinea Evergreen N. truncata New Zealand Evergreen

N. codonandra New Caledonia Evergreen Lophozonia

N. crenata New Guinea Evergreen N. alpina Chile - Argentina Deciduous

N. discoidea New Caledonia Evergreen N. cunninghamii Australia Evergreen

N. flaviramea New Guinea Evergreen N. glauca Chile Deciduous

N. grandis New Guinea Evergreen N. macrocarpa Chile Deciduous

N. nuda New Guinea Evergreen N. menziesii New Zealand Evergreen

N. perryi New Guinea Evergreen N. moorei Australia Evergreen

N. pseudoresinosa New Guinea Evergreen N. obliqua Chile - Argentina Deciduous

N. pullei New Guinea Evergreen Nothofagus

N. resinosa New Guinea Evergreen N. antarctica Chile - Argentina Deciduous

N. rubra New Guinea Evergreen N. betuloides Chile - Argentina Evergreen

N. starkenborghii New Guinea Evergreen N. dombeyi Chile - Argentina Evergreen

N. stylosa New Guinea Evergreen N. nitida Chile Evergreen

N. womersleyi New Guinea Evergreen N. pumilio Chile - Argentina Deciduous

source: Hill and Dettmann 1996, Ogden et al. 1996, Read and Brown 1996, Read and

Hope 1996, Rodríguez and Quezada 2003.

2.2 Nothofagus betuloides – the Southernmost Evergreen Tree

Species and its Forests

N. betuloides is known commonly as ‘coihue de Magallanes’ in Chile and

‘guindo’ in Argentina. It was also well known to the Indian tribes of South America

who referred to it as coigüe (Mapuche name), yerkianop (Alacaluf name), ouchpaya

(Ona or Shelknam name), and shushchi (Yahgan or Yamana name).


41

2.2.1 Biology

N. betuloides is one of the longest-living South American Nothofagus species,

with specimens reaching 500-600 years of age (Veblen et al. 1996) or more (628 years,

Gutiérrez et al. 1991). The trees (Figure 1.1) can grow to heights of 20 m, and in a few

cases up to 35 m, but can also remain as only meter low shrubs in subantarctic

shrublands. The trunk can reach 2 m in diameter. The bark is grey or reddish in colour

and relatively smooth. The twigs are puberulent, rarely glabrous. The leaves (12-28 x 8-

18 mm) are ovate-elliptical to elliptical–suborbicular, acute to obtuse, cuneate at the

base, serrate, subcoriaceous and glabrous (Figure 2.2) (Moore 1983).

Figure 2.1: Nothofagus betuloides growing in Río Bueno, Tierra del Fuego (Photo: A.

Promis).

N. betuloides is a hermaphrodite. The male flowers are solitary, with a perianth

4-4.5 cm long, 5-7 lobes, and has 10-16 stamens. Female flowers occur in groups of

three (Moore 1983, Rodríguez and Quezada 2003). Depending on the location, N.

betuloides flowers between September and December (Rodríguez and Quezada 2003).

However, the South American Nothofagus species do not flower regularly and in some


42

years widespread non-flowering occurs, and seed production is therefore intermittent

(Báez et al. 2002).

Figure 2.2: Leaves and seeds of Nothofagus betuloides (Photo: A. Promis).

The fruits of N. betuloides (Figures 2.2 and 2.3) are nuts occurring in threes, 5-6

x 4-4.5 mm, triquetrous, glabrous; cupule 4-partite (Moore 1983). They mature during

summer and are dispersed mainly by gravity and wind between March and May

(Donoso and Donoso 2007, Ibarra et al. 2007). Although nut development in

Nothofagus is often close to 100 %, the majority of the seeds produced are often not

viable (Báez et al. 2002).

Figure 2.3: Seeds of Nothofagus betuloides (Photo: G. Cruz).

The wood of N. betuloides has a slight lustre, and a fine and homogeneous

texture. The grain is generally straight. The growth rings are annual but not clearly


43

visible. The sapwood has a yellowish-white colour, and the heartwood is light pink to

reddish-brown. The wood density has been classified as low, with a basic density of 615

kg m-3, and an air dry density of 620 kg m-3. With respect to the strength properties of

the species, the wood has a low modulus of rupture (708 kg cm-2), a medium modulus

of elasticity (103 t cm-2), and also a medium maximum crushing strength parallel to the

grain when dry (445 kg cm-2) (Pérez 1983, Manso et al. 2007). The wood is long

lasting, even without treatment. The durability ranges from durable to very durable

when it is not exposed to conditions promoting rapid decay, e.g., ground contact

(Donoso and Donoso 2007).

The timber can be used as poles, flooring, indoor panelling, roofing (joist and

rafter), pillars, furniture, decorative veneers and panels (Pérez 1983, Manso et al. 2007).

2.2.2 Ecology

2.2.2.1 Geographical distribution

N. betuloides is an endemic tree species of the

Chilean and Argentinean subantarctic forests. In Chile

it occurs from the Valdivia district (40 ° 31 ’ S) to the

archipelago of Cape Horn (55 ° 31 ’ S) (Rodríguez and

Quezada 2003). In Argentina it occurs mainly between

48° S to the southernmost tip of Tierra del Fuego

(Veblen et al. 1996) (Figure 2.4).

In the northern part of its distribution, N.

betuloides grows at high elevations in both the coastal

(above 800 m a.s.l.) and the Andean Cordillera (above

900 m a.s.l.), approaching the treeline. In the south it

forms forests in elevations from sea level to upper

treeline on the southern and western sides of Tierra del

Fuego (McQueen 1976, Tuhkanen et al. 1989-1990,

Donoso and Donoso 2007).

Figure 2.4: Map of the distribution of Nothofagus

betuloides in Chile and Argentina (source: Dimitri 1972, Ibarra et al. 2007).


44

2.2.2.2 Autecology: Germination and juvenile growth

Germination is epigeous and occurs during the spring of the year in which the

seed is produced (Donoso and Donoso 2007). Most Nothofagus species do not develop

a persistent seed bank (Veblen et al. 1996). Germination can occur in sunny, semi-shady

or shady places (Roig et al. 1985). The seed dormancy of N. betuloides is readily broken

(Martínez Pastur et al. 1994). Pre-germinative treatments have demonstrated that with a

cold-moist stratification of the seeds for a period of up to 120 days the germination rate

reaches up to 45 % (Donoso and Donoso 2007). The immersion of seeds in 100 ppm

giberelic acid for 12 hours results in a germination rate of 33 %. The application of

potassium nitrate has no effect (Martínez Pastur et al. 1994).

The natural regeneration of N. betuloides is through seed. Vegetative

reproduction by means of sprouting at the base of the trunks of living trees has also

been observed on the upper slopes of Tierra del Fuego, where the trees are stunted and

crooked (Krummholz) (Gutiérrez et al. 1991).

For all Nothofagus species in South America, seedling establishment occurs best

under moderately high light levels and where bare mineral soil has been exposed

(Veblen et al. 1996). For example, N. betuloides is able to grow as a pioneer species on

open sites with mineral soils, including moraines (Armesto et al. 1992), recently

deglaciated areas (Pisano 1978), and forest road banks. In old-growth forests, seedlings

often establish on fallen logs (Roig et al. 1985, Veblen et al. 1996).

Positive rates of net photosynthesis of up to three hours (around noon) have been

reported for N. betuloides seedlings growing under a closed canopy in Tierra del Fuego,

and up to ten hours in canopy gaps (Squeo and Cabrera 1995, Arroyo et al. 1996). As N.

betuloides has a relatively low light compensation point for net positive photosynthesis,

seedlings can establish under closed canopy conditions. N. betuloides can form an

abundant stock of persistent seedlings or saplings, surviving in the understorey for many

decades and even exceeding one hundred years (Rebertus and Veblen 1993a, Veblen et

al. 1996) at high elevations or latitudes, where there is little competition from ground

flora.

The relationship between foliar weight and leaf area is on average lower for

seedlings growing in shade (9.3 mg cm-2) than for those situated in canopy gaps (11 mg

cm-2). The foliar nitrogen concentration of the seedlings is similar under both light

conditions (ca. 1.2 %, Squeo and Cabrera 1995).


45

2.2.2.3 Synecology: Vegetation types

N. betuloides forests have been well described phytogeographically, and given

the following designations: subantarctic evergreen forest (Skottsberg 1910, Godley

1960), subantarctic evergreen rain forest (Schmithüsen 1956), Nothofagetum betuloidis

(Oberdorfer 1960, Pisano 1977), evergreen forest (Pisano 1977, Moore 1983, Pisano

1997), N. betuloides forest type (Donoso 1981, Veblen et al. 1983), sub-region of the

microphyllus evergreen forest (Gajardo 1994), and Andean and coastal temperate

evergreen forest (Luebert and Pliscoff 2006).

N. betuloides forests are predominantly found where an oceanic cold temperate

climate prevails, characterised by a mean temperature of around 8.9 ºC in the warmest

month (between 8.5 and 10 °C) and 2.7 ºC in the coldest (between 1.0 and 3.5 °C). The

rainfall ranges between 800-2,000 mm year-1; i.e., there are no arid months (Pisano

1977, Tuhkanen 1992). The climate of the mixed evergreen – deciduous forest is less

oceanic, characterised by a mean temperature between 9.0 to 9.5 ° C, and even 11 °C, in

the warmest month, and between 0.5 to 2.5 °C in the coldest month (Tuhkanen 1992).

N. betuloides can also grow as creeping shrub in transition to the tundra near the Pacific

Ocean. The climate there has a mean temperature of 8.8 ºC in the warmest and 4.4 ºC in

the coldest month. The rainfall exceeds 2,000 mm year-1, and reaches up to 4,846 mm

year-1 (McQueen 1976, Pisano 1977, Pisano 1981).

The climate conditions, where the pure evergreen N. betuloides forest and the

mixed evergreen forests are found, until now are only poorly known. Nonetheless, we

hypothesize that the temperature would be affecting the distribution of the tree species

in these forests, with mild winters adjacent to the sea and at low elevations, and

decreasing minimum temperatures further inland.

In southern Patagonia and Tierra del Fuego, four characteristic forest types

dominated by N. betuloides, and additionally occurrences in subantarctic shrublands,

can be distinguished: (1) the pure evergreen N. betuloides forest, (2) the mixed

evergreen N. betuloides – Drimys winteri forest, (3) the mixed evergreen N.

betuloides – D. winteri – Pseudopanax laetevirens forest, and (4) the mixed evergreen

– deciduous N. betuloides – Nothofagus pumilio forest (Pisano 1977). (5) Outside

closed forests, N. betuloides is able to grow as creeping shrub in subantarctic

Krummholz formations and moorland.


46

(1) Pure evergreen N. betuloides forest is mainly located in inland southern

Patagonia and Tierra del Fuego, and also at the treeline in some places protected from

strong winds (Figure 2.5).

Figure 2.5: Pure evergreen Nothofagus betuloides forest in Río Condor, Tierra del

Fuego (Photo: A. Promis).

Two soil formation processes have been described for pure N. betuloides forests,

podsolisation on well-drained sites and hydromorphism in areas with moderate to high

waterlogging (Pisano 1977, Puigdefábregas et al. 1999, Gerding and Thiers 2002,

Romanyà et al. 2005). These soils are normally shallow (< 50 cm), loamy in texture,

acidic (pH 3.4-5.5) and not very fertile (Gerding and Thiers 2002, Romanyà et al. 2005,

Thiers and Gerding 2007). Accumulation of huge layers of organic matter on the forest

floor and large amounts of decaying wood have been described for this forest type in

Tierra del Fuego (Gutiérrez et al. 1991). Most of the plant roots and nutrients are

located in the bottom part of a thick raw humus layer, i.e., in a soil depth of only 4 to10

cm (Pisano 1977, Gutiérrez et al. 1991, Gerding and Thiers 2002, Romanyà et al. 2005).

The N. betuloides forests form a very dense forest with few vascular plant

species. Few individuals of the tree species D. winteri and Maytenus magellanica occur

locally. There are few species in the shrub layer, which is dominated mainly by

Berberis ilicifolia. Less frequent and lower in cover are Empetrum rubrum, Fuchsia

magellanica, Pernettya mucronata, and Ribes magellanicum. Vascular plant species of

the ground flora are Adenocaulon chilense, Luzuriaga marginata, Senecio acanthifolius,

Gunnera magellanica, the ferns Blechnum magellanicum, Blechnum penna-marina,


47

Asplenium dareoides, and filmy ferns, mostly Hymenophyllum pectinatum,

Hymenophyllum secundum and Hymenophyllum tortuosum (Pisano 1977, Moore 1983,

Gajardo 1994). Worth mentioning is the dense cover of lower plants, dominating the

forest floor, the decaying tree trunks, and as epiphytes on the stems and branches of the

trees. Characteristic mosses are Acrocladium auriculatum, Dendroligotrichum

dendroides, Polytrichadelphus magellanicus, Dicranoloma robustum, and Ptychomnion

cygnisetum. The most common liverworts are Marchantia berteroana, Gakstroemia

magellanica, Schistochila lamellata, and Lepidozia filamentosa (McQueen 1976, Pisano

1977, Moore 1983). Worth mentioning are lichens like Cladonia sp. on the ground, and

a large number of epiphytes.

A study of forest hydrology in a pure evergreen mature N. betuloides forest

located in Tierra del Fuego showed that the canopy intercepts around 41 % of the total

rainfall (779 mm year-1). The direct throughfall was 59 %, and the stemflow averaged

less than 0.1 % of the total rainfall. The amount of water percolated was around 31 % of

the gross precipitation, and the water yield 33 % (Frangi and Ritcher 1994). The average

mass of fallen coarse woody debris in stands on the Argentinean side of Tierra del

Fuego has been measured at 52 Mg ha-1 (dry weight), with low decay rates for small

branches (k=0.17). This might be related to the lower summer temperatures and

saturated water conditions, which can reduce aeration of the laid logs on the ground

(Frangi et al. 1997).

(2) Near the coast below ca. 200 m a.s.l., D. winteri is becoming a more regular

component of the canopy, forming a mixed evergreen N. betuloides – D. winteri forest

(Figure 2.6). The precipitation ranges between 900-2,000 mm year-1, and there are low

thermal amplitudes, indicating hyperoceanic climate (Pisano 1977). The soils are deep,

but often poorly drained and waterlogged (Thiers and Gerding 2007).

Between 8-12 % of the trees are D. winteri, with isolated individuals of M.

magellanica and Embothrium coccineum. Floristically this association is very similar to

the pure evergreen N. betuloides forest. However, due to edaphic and climatic factors,

the decomposition of litter is slow. The forest floors are covered by decomposed and

decomposing tree trunks. These conditions facilitate the ferns A. dareoides, Cystopteris

fragilis, Grammitis magellanica, the lower plants and the filmy ferns mentioned above

with the addition of Hymenophyllum ferrugineum, Hymenophyllum peltatum, and

Serpyllopsis caespitose (Pisano 1977).


48

Figure 2.6: Coastal mixed evergreen Nothofagus betuloides – Drimys winteri forest

(Photo: G. Cruz).

(3) At elevations below ca. 100 m a.s.l., the tree P. laetevirens joins the canopy,

forming the coastal mixed evergreen N. betuloides – D. winteri – P. laetevirens

forest. The precipitation ranges from around 1,500-4,000 mm year-1. The soils are often

organic, with a deep accumulation of peat in the first horizon. The soil is covered by a

thick litter layer of coarse woody debris in different stages of decay (Pisano 1977). Tree

species such as D. winteri and P. laetevirens are very characteristic. The latter is a small

tree up to 5 m tall. The ground vegetation is species poor, most dominant are mosses

and liverworts on decaying wood.

(4) The mixed evergreen – deciduous N. betuloides – N. pumilio forest can be

regarded as a transition between the evergreen and the deciduous Nothofagus forests

(Figure 2.7). N. betuloides dominates the more humid and poorly drained sites. With

decreasing rainfall the deciduous N. pumilio gains dominance and the transition to the

deciduous forests is gradual. The soils are more fertile than those of the pure evergreen

N. betuloides forest or the mixed evergreen N. betuloides – D. winteri forest (Thiers and

Gerding 2007). They are predominantly brown podsolic, with increased podsolisation in

stands dominated by the deciduous N. pumilio (Pisano 1977). The tree species D.

winteri and M. magellanica are present in low numbers. The coverage of the shrub

understorey is low, dominated by B. ilicifolia and P. mucronata. There are fewer filmy

ferns (Hymenophyllaceae sp.) and bryophyte species than in the forest types mentioned


49

previously (Young 1972, McQueen 1976, Pisano 1977, Gajardo 1994, Luebert and

Pliscoff 2006).

Figure 2.7: Mixed evergreen – deciduous Nothofagus betuloides – Nothofagus pumilio

forest (Photo: A. Reif).

(5) N. betuloides also can grow in the Magellan Mooreland, there being a minor

species in shrubland dominated by Pilgerodendron uviferum, or as a low creeping shrub

associated with Schoenus antarcticus and Carpha alpina var. schoenoides (Pisano

1977) and mosses including Sphagnum magellanicum.

2.2.2.4 Forest texture

The structure of the forests of southern Patagonia and Tierra del Fuego is shaped

principally by wind. Wind acts as an agent of both coarse and fine-scale disturbance.

Storms can cause the blow-down of entire stands (Rebertus et al. 1997, Puigdefábregas

et al. 1999). Wavelike patterns of gap formation have been documented for both pure N.

betuloides and pure N. pumilio forests in Tierra del Fuego, and for mixed N. betuloides -

N. pumilio forests, mostly on sites predisposed to wind disturbance (Rebertus and

Veblen 1993b, Rebertus et al. 1993, Puigdefábregas et al. 1999).

In most cases, windthrow of individual trees creates canopy gaps smaller than

200 m2 (Rebertus and Veblen 1993a, Gutiérrez 1994). In pure N. betuloides stands,

canopy openings can occur as interwoven gap complexes where discrete gaps are not

apparent (Rebertus and Veblen 1993a). It was observed small canopy gaps (51 m2) in a


50

pure, uneven-aged N. betuloides forest in Tierra del Fuego (see Chapter 3). In a mixed,

uneven-aged N. betuloides – N. pumilio forest larger canopy gaps were found, with an

average size of 107 m2 (see Chapter 3). In Tierra del Fuego a patch mosaic pattern has

been observed in Nothofagus forests, with patches of younger trees neighbouring larger

patches of old forest, forming multicohort stands (Gutiérrez et al. 1991).

2.2.2.5 Forest dynamics and structure

Disturbances of different frequencies and intensities on different sizes of areas

are a major factor shaping the forest structures. The importance of coarse and fine-scale

disturbances to the forest dynamics and structures of southern South American

Nothofagus forests has been well summarised by Veblen et al. (1996) and Pollmann and

Veblen (2004).

Coarse-scale disturbance

Coarse-scale disturbances are generally necessary for the regeneration of

Nothofagus species at lower elevations and under a milder climate, where a number of

shade-tolerant rain forest species are dominant, and a dense understorey competes with

juveniles of Nothofagus.

Large-scale disturbance at higher elevations or latitudes can initiate regeneration

patterns forming even-aged N. betuloides stands (Donoso and Donoso 2007). Examples

are the pure, even-aged secondary N. betuloides forest which established after a fire on

the Argentinean side of Tierra del Fuego at the end of the 1950s (Martínez Pastur et al.

2002), or the pure, even-aged secondary forest of the coastal Chilean side of southern

Patagonia and Tierra del Fuego, which recovered through natural succession after

periods of colonisation during which the forests were selectively logged, burned and

grazed.

Fine-scale disturbance

At higher elevations and at higher latitudes, where species richness is low,

regeneration is less dependent on coarse-scale disturbance (Pollmann and Veblen 2004).

Here canopy gaps are more important for the regeneration of the Nothofagus species,


51

e.g., in southern Patagonia and Tierra del Fuego. The wind-induced snapping and

uprooting of trees were the most common types of mortality observed in a pure N.

betuloides forest and in N. betuloides forest mixed with N. pumilio (see Chapter 3). In

the pure N. betuloides forest 52 % of the tree-falls were snap and 30 % uprooting. This

appears to be characteristic of other Nothofagus dominated forests in South America. In

the mixed N. betuloides – N. pumilio forest, however, the uprooting of trees was by far

the most frequent cause of tree mortality, accounting for 70 % of the tree-falls,

compared to only 24 % snapped trees. This largely corresponded with the findings from

Tierra del Fuego (Rebertus and Veblen 1993a).

The lower resistance to stem breakage and the prevalence of snapped trees in the

evergreen N. betuloides forests might be related to crown dieback (Rebertus and Veblen

1993b, Rebertus et al. 1993). An additional cause might be due to the abundance of the

magellanic woodpecker (Campephilus magellanicus). Its occurrence in forests in Tierra

del Fuego has been related to the density of N. betuloides and the occurrence of snags

(Vergara and Schlatter 2004). Magellanic woodpeckers primarily consume the larvae of

wood-boring coleopteras in large and decaying trees, and also drill holes in large and

healthy trees to access the phloem sap (Schlatter and Vergara 2005). Although not fatal

in itself, the activity of the woodpecker may lead to secondary damage by diseases and

insects. Insects and extreme climatic events may also reduce the vigour of trees,

inducing partial crown mortality. This, too, facilitates fungal attacks and heart rot of the

bole. The decaying standing deadwood enlarges the potential of trees suited for cavity

creation by magellanic woodpeckers (Ojeda et al. 2007).

Both N. betuloides and N. pumilio can establish in small tree-fall canopy gaps

(Veblen 1989, Rebertus and Veblen 1993a, Gutiérrez 1994, Arroyo et al. 1996, Cuevas

2003, Fajardo and de Graaf 2004, Cavieres and Fajardo 2005). Juvenile trees are also

released by the creation of these small-scale disturbances in the canopy (Veblen et

al.1996), although the growth strategies of the species differ. The observation by

Rebertus and Veblen (1993a) that N. betuloides is more shade tolerant than N. pumilio

has been proved in Tierra del Fuego (see Chapter 3).

In the mixed evergreen N. betuloides – D. winteri forest at low elevations near

the Tierra del Fuego coast, the D. winteri seedlings grow well under dense canopies

and, after formation of gaps in the canopy, can respond faster than N. betuloides,

sometimes even impeding the establishment of the latter (Rebertus et al. 1993a,

Gutiérrez 1994, Veblen et al. 1996). Neither saplings nor pole stage N. betuloides are


52

likely to be found where there are high numbers of D. winteri and M. magellanica in the

understorey (Gutiérrez et al. 1991).

2.3 Forest Use in the Past, the Present and the Future

2.3.1 The Past

The historical forest use has strongly influenced the present forest cover. In the

Magellan Region of Chile (southern Patagonia), the history of forest use can been

divided into four periods (Cruz et al. 2007b); (1) an indigenous period (10,000 B.C.-

1843); (2) a colonisation period (1843-1953); (3) an oil and industrial period (1953-

1980); and (4) a forest management and industrial expansion period (1980-2004).

(1) Five different indigenous tribes inhabited Patagonia thousands of years

before the arrival of the Spaniards (16th century). The Strait of Magellan was inhabited

by two nomadic tribes, the canoe aborigines (the Alacalufes or Kawéskar and the

Yámanas or Yaganes) and the land aborigines (the Onas or Selk’man and the Haush or

Manek’enk on the large island of Tierra del Fuego or Karukinká, and the Tehuelches or

Patagones on the continent). The indigenous people mainly used the forests to collect

fungi and berries, and wood for fire, bows, arrows, domestic utensils and for

constructing huts. Canoes were built with the bark of N. betuloides. There was also a

spiritual attachment, with the Onas people believing that spirits inhabited the forests

(Gusinde 1944, de Agostini 1945, 1956, Martinic 1982, 1992, Vairo 1997).

(2) Harvesting of the N. betuloides forests began in 1843, when the Bulnes’ Fort

was built on the eastern side of the Brunswick Peninsula (Figure 2.8), a consequence of

the colonisation policies in southern Chile. In 1848 the city of Punta Arenas (Sandy

Point) was founded. At this time wood became an important resource for construction

and as fuel. The first sawmill with a hydropower system was established in Punta

Arenas in 1861, and in 1875 steam power was introduced to the region in the form of

locomotives. In the late 19th century there were several sawmills located on the

Brunswick Peninsula (Martinic 1992, Cruz et al. 2007b).


53

Figure 2.8: Map of southern Patagonia and Tierra del Fuego showing the past locations

of sawmills.

From the 1880s, many forests were converted to farmland and pasture through a

process of logging and burning. In southern Patagonia and Tierra del Fuego,

approximately 200,000-300,000 ha were transformed to pasture (Cruz and Lara 1987,

Cruz et al. 2007b).

In the late 19th and early 20th centuries the timber industry was centred on the

Chilean side of southern Patagonia and Tierra del Fuego, but with the majority of the

wood harvested destined for the Argentinean side of Patagonia and Tierra del Fuego,

and the Falklands. New sawmills were established near forest dominated by N.

betuloides, such as those situated on or near Dawson Island, the Whiteside Channel, the

Almirantazgo Fjord, Navarino Island (Beagle Channel), the Otway Sound and Skyring

Sound (Figure 1.8). During this period N. betuloides provided around 80 % of the

timber traded. However, restrictions on the importation of wood put in place by the

Argentinean government impeded the timber industry in Chile, with the lowest

exportation rates recorded between 1951 and 1952 (Martinic 1992, Cruz et al. 2007b).

(3) In the middle of the 20th century an intensive harvesting of the forests began

on both the Chilean and the Argentinean (Gea-Izquierdo et al. 2004) sides of southern

Patagonia. This coincided with the beginning of the oil and industrial period. At this

time the forest industry concentrated on forests dominated by the deciduous N. pumilio,

because oil exploration, and the related infrastructure, was frequently situated at

locations near to these forests. The demand for wood from N. betuloides forests, which


54

were located near urban areas and the coast, decreased. The harvested trees were logged

selectively, with the best, largest and healthiest timber trees removed from the stands

(known as ‘floreo’ in the region). Only the poor quality, badly shaped and unhealthy

trees remained (Martínez Pastur et al. 2000, Klepeis and Laris 2006, Cruz et al. 2007a).

Many other stands were burned and converted to farmland.

The introduction of fossil fuels after the 1950s as a new energy resource

replaced the use of firewood. At that time N. betuloides accounted for around 15 % of

the total wood volume traded in the Chilean Patagonia and Tierra del Fuego (Cruz et al.

2007b).

(4) A new period of forest use in Chile began in the 1980s. Specific silvicultural

methods for the management of the native forests were developed. It was demonstrated

that N. betuloides forests in Chile could be managed employing either a selection or a

shelterwood system (Donoso 1981). In reality, however, the stands were only rarely

managed, with a small number of regeneration cuts made as part of a shelterwood

system (Cruz et al. 2007a).

In the early 1990s a new, productive development of wood and forest

management began. Wood chips were exported for a short time (1991-1997), and a

furniture factory using solid wood was founded (Cruz et al. 2007b).

2.3.2 The Present

At present the forest surface where N. betuloides participates is approximately

1,396,947 ha in the Magellan Region in Chile, accounting for 53 % of the total forest

area. In terms of forest structure, 51 % are old-growth, 6 % are secondary growth, 9 %

represent a transition between mature and secondary growth, and 34 % are 2 to 8 m tall

shrubland (CONAF-CONAMA 1999b).

A new period of forest conservation on private land began in 2004. In the 1990s

the American Trillium Corporation purchased 625,000 ha on the Chilean and 185,000

ha on the Argentinean side of Tierra del Fuego. However, the firm failed in its attempts

to establish a sustainable logging operation. Goldman Sachs, a global investment

banking firm, acquired the loans and land, and finally donated the property to the

Wildlife Conservation Society (WCS) in 2004 (Duncan 2006). At present, the property

on the Chilean side of Tierra del Fuego is managed by WCS as a private reserve. The

principal aim of the Karukinka Reserve is to preserve the wildlife existing there, and to


55

restore its ecosystems (Karukinka 2007, WCS 2007). This private reserve protects more

than 20 % of the N. betuloides forests of southern Patagonia and Tierra del Fuego in

Chile (Arroyo et al. 1996).

Additionally, more than 49 % of the N. betuloides forests of Chilean Patagonia

and Tierra del Fuego are protected as State Protected Wild Areas (SNASPE) (CONAF-

CONAMA 1999b).

It can be estimated, that ca. 280,000 ha of pure N. betuloides forests and forests

mixed with N. pumilio are suitable for timber production. Because of logistical reasons,

lack of ecological and silvicultural knowledge, and problems in the wood drying

process, the use of these forests at present is marginal (< 1 % of wood production in

2004) (Cruz et al. 2007b). Sawn wood of N. betuloides is only produced in one big

sawmill for the furniture production industry, consuming < 5 % of the harvested timber

(Cruz et al. 2007b).

2.3.3 The Future

The N. betuloides forest area suitable for timber production in the Magellan

Region of Chile has been estimated ca. 280,000 ha. Additionally, the forested area

covered by the deciduous N. pumilio, the most important forest resource in southern

Patagonia and Tierra del Fuego, has been estimated ca. 200,000 ha (Cruz et al. 2007b).

For the future development of a commercial forest industry, a system of

sustainable forest management is required. This includes (1) increasing the managed

forest area of stands dominated by N. betuloides, (2) improving current timber

management (carrying out intermediate practices such as thinning), (3) incorporating

the forests selectively logged in the past in silvicultural management schemes, (4)

applying new silvicultural treatments in order to maintain the uneven-aged and multi-

layer structure of the forest, based on an ecological understanding of the natural stand

development, and including the role of natural disturbances, providing the basis on

which the forest may be managed as a renewable resource and trying to leave biological

legacies in order to maintain a higher diversity and richness of species.

New studies of the ecology of the N. betuloides forests, their distribution,

silviculture, wood properties, industrial yields, and also methods of drying have been

carried out in the meantime (see Cruz and Caldentey 2007). The application of this new

knowledge in combination with a diversification of the forest industry within the


56

framework of sustainable forest management would allow the derivation of goods and

services from forest landscapes, while certain levels of biodiversity and ecosystem

processes could also be maintained.

Chapter 3 Disturbances in Nothofagus betuloides forests

57

CHAPTER 3

SMALL-SCALE NATURAL DISTURBANCES AND TREE

DEVELOPMENT PROCESSES IN TWO FORESTS DOMINATED

BY Nothofagus betuloides – A CASE STUDY FROM TIERRA DEL

FUEGO

3.1 Abstract

In order to better understand natural disturbance regimes and tree development

processes, a characterisation of canopy gaps and a study of disturbance dynamics were

undertaken in a pure uneven-aged evergreen Nothofagus betuloides forest and a mixed

uneven-aged evergreen-deciduous N. betuloides – N. pumilio forest. Canopy gaps were

measured along transects, and the gap-makers characterised. Disturbance dynamics

were studied by means of abrupt releases in radial growth revealed in increment cores.

These cores were taken from juvenile trees growing at the edges of canopy gaps. The

canopy gaps observed were small, with an average size of 51 m2 in the pure forest and

107 m2 in the mixed. Only 2.0 % of the total canopy area of the pure forest was canopy

gaps, and 4.6 % expanded gaps. The gap areas in the mixed forest were 2.6 and 2.5

times greater, respectively. The mean number of gap-makers was 2.2 trees in the pure

forest and 2.5 trees in the mixed forest. The most common type of damage in the pure

forest was snapping, compared to uprooting in the mixed forest. An increase in the

frequency of disturbances affecting the forests since 1920-30 was observed. The

greatest changes in radial increment after release from restricted growth were found for

N. pumilio (18.4 %) and N. betuloides (11.2 %) growing in the mixed forest, and the

lowest for N. betuloides (2.4 %) in the pure forest. These results suggest that N.

betuloides exhibits greater shade-tolerance. It can grow surrounding very small canopy

gaps, with low radial growth, subject to abrupt increases upon larger disturbances to the

canopy.

Keywords: canopy gaps, disturbance dynamics, Nothofagus, dendroecology, Tierra del

Fuego


58

3.2 Introduction

The effects of natural disturbances on the dynamics and structure of forests have

been studied in tropical, temperate and boreal forests (e.g. Platt and Strong 1989,

Denslow and Spies 1990, McCarthy 2001, Kuuluvainen 2002). Gap dynamics are

characterised by small or micro-scale disturbances to the mature forest canopy

(McCarthy 2001). The severity of disturbances can influence stand structure, species

composition and the growth rates of surviving trees (Oliver and Larson 1996), and may

increase the availability of resources for plant growth (Veblen 1992) through the

creation of special microhabitats (Runkle 1985). Small-scale canopy disturbances,

although less dramatic than large-scale disturbances, may be more frequent and affect a

larger area over time (Spies et al. 1990). The focus of this study was on gap

characteristics and disturbance dynamics, and their role in the radial growth responses

of juvenile trees, in a pure evergreen Nothofagus betuloides (Mirb.) Oerst. forest and a

mixed evergreen-deciduous N. betuloides - Nothofagus pumilio (Poepp. et Endl.)

Krasser forest.

In South America, the temperate evergreen forest is largely confined to southern

Chile (from 37 ° 45 ’ S to the southernmost tip of the continent), and neighbouring

southern Argentina. These forests tend to have a multi-layered structure. They have a

reduction in species diversity in a southerly direction, and with increasing altitude

(Ovington 1983, Veblen et al. 1983), likely caused by the declining temperatures and

shorter growing seasons (Hildebrand-Vogel and Vogel 1995). Nothofagus species such

as N. betuloides, N. dombeyi (Mirb.) Oerst. and N. nitida (Phil.) Krasser largely

dominate these forests (Ovington 1983). To the north and to the east of its distribution,

the temperate evergreen forest is replaced mainly by the deciduous N. pumilio forest

(Hildebrand-Vogel et al. 1990, Hildebrand-Vogel and Vogel 1995).

Evergreen N. betuloides forests have been described in a variety of different

associations or forest types depending on the author and phytogeographic region. These

include the subantarctic evergreen forest (Skottsberg 1910, Godley 1960), subantarctic

evergreen rain forest (Schmithüsen 1956), Nothofagetum betuloidis association

(Oberdorfer 1960, Pisano 1977), evergreen forest (Moore 1983), N. betuloides forest

type (Donoso 1981, Veblen et al. 1983), sub-region of the microphyllus evergreen forest

(Gajardo 1994), and the Andean and coastal temperate evergreen forests (Luebert and


59

Pliscoff 2006). In general, they have been characterised as species-poor in the canopy,

and species-rich in the cryptogam layer. These forest types occur on sites influenced by

high precipitation levels (> 800 mm year-1) and low temperatures, and where soils are

shallow, with partially decomposed organic matter overlying a bleached podzolic

horizon (Skottsberg 1910, Schmithüsen 1956, Pisano 1977, Moore, 1983). The mixed

evergreen and deciduous N. betuloides - N. pumilio forest has been characterised as a

transitional forest between pure Nothofagus forests. N. betuloides dominates the more

humid and poorly drained sites. With decreasing rainfall the deciduous N. pumilio gains

dominance. The floristic composition is similar to that described above, but with fewer

Hymenophyllaceae and bryophyte species (Young 1972, Pisano 1977, Donoso 1981,

Moore 1983, Veblen et al. 1983, Gajardo 1994, Luebert and Pliscoff 2006).

It has been demonstrated that more than 50 % of the surveyed forests belonging

to these forest types, specifically those forests in southern Patagonia and Tierra del

Fuego suitable for timber production, have been impacted by human influences in the

past (Cruz et al. 2007a). These influences include grazing, logging and burning (Cruz et

al. 2007a). In fact, many of these anthropogenic disturbances have affected the process

of forest stand dynamics, resulting in fragmented and isolated patches of regeneration or

secondary forest (Martínez Pastur et al. 2000).

The world’s southernmost temperate evergreen forests are located on Tierra del

Fuego and Cape Horn archipelagos, and are dominated by N. betuloides. The structure

of these forests was shaped principally by wind. Wind acts as both a coarse and a fine-

scale disturbance agent. In the case of the former, it can cause the blow-down of entire

stands (Rebertus et al. 1997, Puigdefábregas et al. 1999), and in the latter it results in

the windthrow of individual trees, creating canopy gaps smaller than 200 m2, such as

occurs in pure N. pumilio stands and N. betuloides forests mixed with either Drimys

winteri J.R.Forst. & G. Forst or N. pumilio (Rebertus and Veblen 1993a, Gutiérrez

1994). In pure N. betuloides stands, canopy openings can occur as interwoven gap

complexes. In this case discrete gaps might not be apparent (Rebertus and Veblen

1993a). Uprooting and the snapping of boles are the most predominant modes of tree-

fall (Rebertus and Veblen 1993a). On Tierra del Fuego, wavelike patterns of gap

formation have also been documented for both pure N. betuloides and pure N. pumilio

forests, and for mixed N. betuloides - N. pumilio forests. Those patterns appear to be

limited to sites predisposed to wind disturbance (Rebertus and Veblen 1993b, Rebertus

et al. 1993, Puigdefábregas et al. 1999).


60

The importance of coarse and fine-scale disturbances to the forest dynamics of

southern South American Nothofagus forests has been summarised by Veblen et al.

(1996) and Pollmann and Veblen (2004). They demonstrated that canopy gaps appear to

be important in the regeneration dynamics of these Nothofagus forests. In this context,

both N. betuloides and N. pumilio can establish beneath small tree-fall canopy gaps

(Veblen 1989, Rebertus and Veblen 1993a, Arroyo et al. 1996, Cuevas 2003, Fajardo

and de Graaf 2004, Cavieres and Fajardo 2005). Juvenile trees are also released by the

creation of these small-scale disturbances in the canopy (Veblen et al. 1996), although

the growth strategies of the species differ. For example, N. betuloides is more shade

tolerant than N. pumilio. Individuals of this species are able to survive in the

understorey for a long time, even more than 100 years (Veblen et al. 1996). Features of

shade tolerant trees are that they slowly grow into the canopy in the absence of any

disturbance, and persist suppressed beneath a closed canopy until periodic openings in

the canopy are created (Canham 1985, 1989). Shade-intolerant species require relative

large gaps because of their need for relatively high illumination, however (Canham

1990). The growth of N. pumilio alternatively appears to be facilitated by lower

minimum temperatures in larger canopy gaps (Veblen et al. 1996).

However, questions still remain regarding the effects of canopy gaps on stand

development in Nothofagus forests. With this study we want to address the following

questions: are the responses of juvenile N. betuloides trees growing in pure and mixed

natural forests following the creation of small canopy gaps similar? In mixed forests, do

the responses of N. pumilio and N. betuloides trees differ?

To address these questions, the specific objectives of the study were (1) to

compare the characteristics of natural canopy gaps found in two forest types dominated

by evergreen N. betuloides with no human impact. One is a pure N. betuloides forest

and the other is a mixed forest with the deciduous N. pumilio, and (2) to analyse the gap

dynamics and the role of natural disturbances in the radial growth responses of juvenile

trees, thereby providing a deeper understanding of the natural dynamics of these forest

types.


61

3.3 Material and Methods

3.3.1 Study Area

Two forest areas dominated by the evergreen N. betuloides with no evidence of

human impact were studied in the summer 2006 on the south western end of the Isla

Grande of Tierra del Fuego. The two forest areas studied were a pure N. betuloides

forest (forest type 14 sensu Veblen et al. 1983) and a mixed evergreen-deciduous N.

betuloides - N. pumilio forest (forest type 16 sensu Veblen et al. 1983). The larger was a

pure evergreen N. betuloides forest and covered an area of 20 ha, whereas the smaller

was a mixed evergreen-deciduous N. betuloides - N. pumilio forest covering 18.6 ha.

The study was located at the ‘Estancia Olguita’, on the south-eastern side of the Río

Cóndor (53 °59 ’ S, 69 °58 ’ W) (Figure 3.1). Access to this remote area is difficult.

Because of the remoteness, human impact is very slight. Therefore, the study area is

excellent for the study of the natural dynamics of these forest types.

The climate of the area corresponds to that of the northern antiboreal sub-zone,

which has a mean temperature of between 9.0-9.5 °C in the warmest month of the year

and slightly above 0 °C in the coldest month. The annual rainfall is 500-600 mm, but

can reach up to 900 mm. The prevailing wind direction is west to south-west, with an

average speed of 14-22 km hr-1 and a maximum speed of in excess of 100 km hr-1

(Tuhkanen 1992).

The study area is part of the subalpine zone (Frederiksen 1988). The relief is

characterised by valleys running parallel to the Andes, which in most places have been

glacially deepened forming U-valleys. The most frequent soil formation processes in N.

betuloides forests have been described as: (1) podzolisation on well-drained sites and

(2) hydromorphism in areas with moderate to high waterlogging (Pisano 1977,

Puigdefábregas et al. 1999, Gerding and Thiers 2002, Romanyà et al. 2005). These soils

are normally shallow (< 50 cm). They are loamy in texture, acid (pH 3.4-5.5) and not

very fertile (Gerding and Thiers, 2002, Romanyà et al. 2005, Thiers and Gerding 2007).

Most of the plant roots and nutrients are located in the deep layer of poorly decomposed

organic material (4-10 cm) (Pisano 1977, Gutiérrez et al. 1991, Gerding and Thiers

2002, Romanyà et al. 2005).


62

Figure 3.1: Map of South America, Tierra del Fuego and the forests studied on the Río

Cóndor.

Chile


63

3.3.2 Sampling Design for the Assessment of the Forest Structure

The sampling design used to assess the canopy species composition and the

forest structure was a grid system combined with the point-centred quarter method

(PCQ, Cottam and Curtis 1956). A total of 36 parallel transects (300 m) were located at

30 m intervals, following a north - south direction. Sample points were placed every 30

m along each transect. At every point a PCQ was done by measuring the distance to the

nearest live tree with a dbh ≥ 5.0 cm. For each individual, its dbh (cm) was measured

and the species and the crown class were determined (dominant, co-dominant,

intermediate and suppressed). At every second sample point the dominant tree height

was estimated by averaging the heights of the two tallest trees.

3.3.3 Measuring Gap Characteristics

In order to determine the canopy gap and the expanded gap sizes, all gaps (≥ 20

m2) found along the transects were measured. Canopy gaps were defined as the

horizontal projection of a canopy opening to the ground surface (Runkle 1982). In this

study, a gap was considered closed if the vegetation growing below the opening in the

canopy was ≥ 2-3 m tall. The expanded gap area was defined as the area formed by the

gap in the canopy plus the adjacent area delimited by the bases of the edge trees (Runkle

1982). Both gap types were measured on the basis of between 5-8 radii, following a

method similar to that applied in N. pumilio forests in southern Chile by Fajardo and de

Graaf (2004). The gap centre was determined visually. The distances from the gap

centre and the azimuths to the bases of almost every edge tree were recorded, as were

the distances to the corresponding horizontal crown projections. All gaps were digitised

using ArcView GIS Version 3.2 (ESRI, Redlands, CA, USA), and their areas and

perimeters calculated. The topography at the location of each gap was also registered, in

terms of slope, aspect and elevation.

The diameter at breast height (dbh), height or length, type of damage (uprooted,

standing dead, snap) and fall direction was recorded for each gap-maker, i.e., those trees

that resulted in the creation of the gap.

The proportion of the total forest area made up of both canopy gaps and

expanded gaps was calculated from the length of the transect sections they intercepted

and the proportion of the total transect length per forest (sensu Runkle 1992).


64

3.3.4 Assessing Disturbance Dynamics and Radial Growth Responses of Juvenile

Trees

The frequency and intensity of disturbances over time, was studied on the basis

of radial growth patterns of trees, with release periods estimated by means of variations

in the diameter increment. The death of neighbouring canopy trees may lead to a sudden

increase in resource availability, which can in turn produce an abrupt increase in tree

growth. A tree-ring growth release represents a post-disturbance environment, whereby

the surviving trees respond to an increase in resource availability (Copenheaver and

Abrams 2003). Increment cores were examined to determine any abrupt changes in

radial growth, potentially highlighting the dynamics of disturbance in this particular

forest ecosystem (Lorimer 1985, Lorimer and Frelich 1989).

Between three and four juvenile trees were selected in each canopy gap and

cored. Increment cores were taken at a height of about 1 m from the base of the stem.

All suitable cores (n = 203; 85 and 118 for the pure and mixed forest, respectively) were

dried, mounted, sanded and processed at the Department of Silviculture of the

University of Chile, following the standard procedures outlined by Stokes and Smiley

(1968). Ring widths were recorded to the nearest 0.01 mm using a microscope.

Schulman’s (1956) convention for processing increment cores extracted from trees in

the southern hemisphere was followed for the dating of annual rings. The year in which

radial growth began was ascertained for each tree-ring, and the cores were visually

crossdated.

The nature of the understorey release was classed according to the major and

moderate release criteria (Lorimer and Frelich 1989, Copenheaver and Abrams 2003).

Comparing the radial growth rate for the 10 years prior to release with that of the

subsequent 10 years, a major understorey release was defined as an average growth

increase ≥ 100% lasting for at least 10 years, whereas a moderate release was defined as

an average growth increase ≥ 50% sustained for at least 10 years (Copenheaver and

Abrams 2003). The periods not classified as release were defined as restricted growth

periods.

The disturbance intensity was defined as the percentage of all of the trees

sampled showing signs of understorey release in each decade (Lorimer and Frelich

1989, Frelich and Lorimer 1991).


65

The analysis revealed only a small portion of the heterogeneity of the forests, as

only the radial growth responses of juvenile trees to disturbance frequencies and

intensities were examined.

3.3.5 Statistical Analyses

The non-parametric Mann-Whitney U test was used to compare forest structures

between the forests (Sokal and Rohlf 2000). The variables compared in each case were

stocking density, canopy gap size, the canopy gap area as a proportion of the total forest

area, referred to as the gap fraction, the characteristics of gap-makers and radial growth

rates. All of the statistical analyses were performed using SPSS 15.0 for Windows

(SPSS Inc.).

3.4 Results

3.4.1 Forest Structures

Two uneven-aged forests were surveyed. When graphed, the diameter

distributions of the trees fit a power function (Figure 3.2).

The stocking density and the basal area differed between the two forests.

Whereas the pure N. betuloides forest had 1,362 trees ha-1 and a basal area of 105 m2 ha-

1 there were 604 trees ha-1 in the mixed forest, and a basal area of 73 m2 ha-1 (Table 3.1).

The pure forest revealed a higher number of trees in the smallest diameter classes than

the mixed forest, with more than 500 trees of dbh ≤ 10 cm. The distributions of the

relative stocking density and the relative basal area according to the tree crown class

classification were very similar in both forests. Suppressed and intermediate trees were

most frequent but contributed < 15 % to the total basal area. The dominant trees by

contrast accounted for 63 % and 56% of the basal area in the pure and the mixed forest,

respectively (Figure 3.3). The tree species representation within each of the social

classes in the mixed forest demonstrated the importance of N. betuloides. It accounted

for > 80% of the total tree number in the dominant and suppressed classes, indicating

continuity of the species composition in the future (Figure 3.3).


66

Figure 3.2: Tree frequency per 5 cm diameter class for trees ≥ 5 cm dbh in both (a) the

pure evergreen N. betuloides forest and (b) the mixed evergreen-deciduous N.

betuloides – N. pumilio forest. Lines in the graphs correspond to power function models

describing the tree diameter distribution.

Table 3.1: Structural variables of the pure evergreen N. betuloides forest and the mixed

evergreen-deciduous N. betuloides – N. pumilio forest. Numbers in brackets are ranges.

Forest Stocking

(trees ha-1)

Basal area

(m2 ha-1)

dbh

(cm)

Dominant tree

height (m)

N. betuloides 1,362 105 6-118 25 (16-31)

N. betuloides – N. pumilio 604 73 5-120 27 (21-36)

y= 24946.90391*x –1.78603 n=20; R2=0.82300; RMSE=0.60720; p<0.001

(a)

0

100

200

300

400

500

600

≤ 10

11-

15

16-

20 2

1-25

26-

30 3

1-35

36-

40 4

1-45

46-

50 5

1-55

56-

60 6

1-65

66-

70 7

1-75

76-

80 8

1-85

86-9

091

-95

96-

100

>100

dbh (cm)

Stoc

king

(tre

es h

a-1) Nothofagus betuloides

y= 4399.25074*x –1.36603 n=20; R2=0.81241; RMSE=0.48121; p<0.001

0

100

200

300

400

500

600

≤ 10

11-

15

16-

20 2

1-25

26-

30

31-

35

36-

40

41-

45 4

6-50

51-

55

56-

60

61-

65

66-

70 7

1-75

76-

80

81-

85

86-9

0

91-9

5 9

6-10

0

>100

dbh (cm)

Stoc

king

(tre

es h

a-1) Nothofagus betuloides Nothofagus pumilio


67

Figure 3.3: Relative stocking density and basal area presented according to crown class

(Sup, suppressed; int, intermediate; cod, co-dominant; dom, dominant) for both (a) the

pure evergreen N. betuloides forest and (b) the mixed evergreen-deciduous N.

betuloides – N. pumilio forest.

Relative stocking density of N. betuloides

Relative stocking density of N. pumilio

Relative basal area of N. betuloides

Relative basal area of N. pumilio

The two forests differed in terms of the heights of the dominant trees. The

canopy height in the mixed forest was 27 m, compared to only 25 m in the pure forest

(Table 3.1).

(a)

0

10

20

30

40

50

Dom Cod Int Sup

Classification of the tree crown

Rel

ativ

e st

ocki

ng (%

tree

s ha

-1)

0

10

20

30

40

50

60

70

Relative basal area (%

m2 ha

-1)

N. betuloides

(b)

0

10

20

30

40

50

Dom Cod Int Sup

Classification of the tree crown

Rel

ativ

e st

ocki

ng (%

tree

s ha

-1)

0

10

20

30

40

50

60

70

Relative basal area (%

m2 ha

-1)

N. betuloides N. pumilio


68

3.4.2 Canopy Gap Characteristics

There were 25 gaps > 20 m2 in the pure N. betuloides forest and 40 in the mixed

forest (Table 3.2). The mean sizes of both the canopy gaps and the expanded gaps were

significantly smaller in the pure N. betuloides forest (51 m2 and 164 m2, respectively)

than in the mixed N. betuloides - N. pumilio forest (107 m2 and 299 m2, respectively)

(Mann-Whitney U test, p < 0.001). All of the canopy gaps in the pure forest were < 100

m2, compared to 68 % of the gaps in the mixed forest. Roughly 75 % of the expanded

gaps were < 200 m2 in the pure forest compared to only 43 % in the mixed forest

(Figure 3.4).

Table 3.2: Number of gaps, characteristics of canopy and expanded gaps and the gap

fraction determined for each forest. Identical letters mean that there was no significant

difference (Mann-Whitney U test, p > 0.05). Numbers in brackets are standard errors.

N. betuloides N. betuloides – N. pumilio

Area (ha) 20.0 18.6

Slope characteristics

range of slope inclinations (º) 0-10 0-12

most common relative elevation mid and upper slope mid slope

Number of gaps 25 40

Canopy gap

mean gap area (m2) 51 a (± 4.7) 107 b (± 21.5)

size range (m2) 21-98 21-605

mean gap perimeter (m) 34 (± 2.0) 48 (± 5.6)

Expanded gap

mean gap area (m2) 164 a (±15.4) 299 b (± 35.9)

size range (m2) 82-441 69-1,126

mean gap perimeter (m) 56 (± 2.8) 75 (± 6.3)

Gap fraction (% of forest)

canopy gaps 2.0 a 5.2 b

expanded gaps 4.6 a 11.6 b


69

Figure 3.4: Frequency of (a) canopy gaps (size class 20 m2) and (b) expanded gaps (size

class 40 m2) for both the pure evergreen N. betuloides forest and the mixed evergreen-

deciduous N. betuloides – N. pumilio forest.

The percentage gap area, determined for both gap types as a function of the

combined length of transect situated within gaps and the total transect length, was

significantly lower for the pure N. betuloides forest (Mann-Whitney U test, p < 0.001).

The percentage canopy gap area and expanded gap area in the mixed forest was 2.6 and

2.5 times greater than in the pure forest, respectively (Table 3.2).

The characteristics of the gap-makers in both forests did not differ significantly

in terms of the average number of dead trees per gap (Mann-Whitney U test, p = 0.527).

There were an average of 2.2 dead trees in the gaps located in the pure N. betuloides

forest and 2.5 in the mixed N. betuloides - N. pumilio forest (Table 3.3). About 30 % of

the gaps present in both forests were created by the death of a single tree.

The most common type of damage in the pure N. betuloides forest was snap,

with uprooting most prevalent in the mixed forest. This difference between the type of

0

10

20

30

40

50

30 50 70 90 110 130 150 170 190 210 230 > 250

Gap size classes (m2)

Freq

uenc

y (%

)N. betuloides N. betuloides - N. pumilio

(a)

0

10

20

30

40

50

40 80 120 160 200 240 280 320 360 400 440 480 > 500

Expanded gap size classes (m2)

Freq

uenc

y (%

)

N. betuloides N. betuloides - N. pumilio(b)


70

damage observed in the two forests was statistically significant (Mann-Whitney U test,

p < 0.001). The number of standing dead trees did not differ significantly between the

two forests (Mann-Whitney U test, p = 0.388), although these were more frequently

observed in the pure N. betuloides forest (Table 3.3).

Table 3.3: Characteristics of gap-makers and the relative frequency of tree mortality

types in both forests. Identical letters indicate no significant difference (Mann-Whitney

U test, p > 0.05). Numbers in brackets are standard errors.

N. betuloides N. betuloides – N. pumilio

Gap-makers per gap

mean number 2.2 a (± 0.2) 2.5 a (± 0.3)

range 1-5 1-8

mean dbh (cm) 55 a (± 2.7) 66 b (± 5.6)

dbh range (cm) 19-142 35-103

most frequent fall orientation south-east south-east

Tree mortality types (%)

snapped 52 a 24 b

uprooted 30 a 70 b

standing dead 18 a 6 a

3.4.3 Disturbance Dynamics and Radial Growth Responses of Juvenile Trees

The N. betuloides juvenile trees selected in the two forests studied exhibited a

smaller average dbh and height than the N. pumilio juvenile trees. Both the mean and

the maximum radial growth recorded for N. betuloides was lower in individuals

sampled from the pure forest than those located in the mixed forest. Whereas N. pumilio

demonstrated the greatest mean radial growth, N. betuloides revealed a greater

maximum radial growth, of 7.96 mm yr-1 (Table 3.4).

The recruitment of N. betuloides juvenile trees occurred continually over the

time period in question; from 1810 to 1970 in the pure forest and 1820 to 1990 in the

mixed forest (Figure 3.5). N. betuloides revealed a fairly constant recruitment rate

between 1920 and 1950 in the pure forest and up until 1980 in the mixed forest. The

recruitment of N. pumilio in the mixed forest was more recent than that of N. betuloides,


71

only occurring after 1850 (Figure 3.5). The recruitment rate of N. pumilio was higher

than that of N. betuloides in the mixed forest, but appeared to decline after 1950, with

N. betuloides tending to dominate the forest composition with time (Figure 3.5).

Table 3.4: Characteristics of all recorded juvenile trees, the number and mean duration

of major and moderate periods of release, radial growth patterns of juvenile trees,

percentage of the mean annual radial increment due to release, and the percentage of the

lifespan in release by species and forest. Identical letters indicate no significant

difference between the annual radial growth under restricted growth and in release

(Mann-Whitney U test, p > 0.05). Numbers in brackets are standard errors.

Pure forest Mixed forest

N. betuloides N. betuloides N. pumilio

juvenile trees 85 70 48

mean age (yr) 81 (3.4) 91 (4.2) 97 (4.3)

mean height (m) 10 (0.5) 11 (1.2) 14 (1.2)

mean dbh (cm) 13 (± 0.6) 18 (± 0.9) 23 (± 1.1)

mean radial growth (mm yr-1) 0.62 (± 0.01) 0.75 (± 0.01) 0.87 (± 0.01)

highest radial growth (mm yr-1) 3.21 7.96 7.16

juvenile trees with at least one release

period (%) 41 60 74

number of major releases 10 20 19

mean duration of major releases (yr) 11 (± 0.54) 12 (± 0.53) 12 (± 0.48)

number of moderate releases 39 51 41

mean duration of moderate releases

(yr)15 (± 0.54) 17 (± 0.94) 17 (± 0.99)

lifespan in release (%) 22 (±1.66) 23 (±1.55) 21 (±1.28)

mean annual radial growth under

restricted growth (mm yr-1)

0.61 a

(± 0.03)

0.80 a

(± 0.09)

0.84 a

(± 0.08)

mean annual radial growth in release

(mm yr-1)

0.62 a

(± 0.04)

0.89 a

(± 0.14)

1.00 b

(± 0.17)

mean annual radial growth increment

due to release (%) 2.4 11.2 18.4


72

Figure 3.5: Recruitment in (a) the pure evergreen N. betuloides forest and (b) the mixed

evergreen-deciduous N. betuloides – N. pumilio forest. Bars show the percentage of the

trees cored recruited in each decade.

Changes to the radial growth of juvenile trees were also analysed to determine

disturbance frequencies and intensities. A total of 112 increment cores were analysed to

study the occurrence of release events.

In general, the responses of N. betuloides and N. pumilio to disturbances were

very irregular, revealing many longer periods of restricted growth followed by both

moderate and major releases (Figure 3.6).

The juvenile trees sampled in this study revealed between 0 and 3 periods of

release during their lives. A large percentage (74 %) of the cored N. pumilio juvenile

trees showed at least one period of release, compared to only 60 % of N. betuloides trees

in the mixed forest and 41 % in the pure forest (Table 3.4).

(a)

02468

1012141618

1800 1820 1840 1860 1880 1900 1920 1940 1960 1980 2000

Decade

Per

cent

age

of tr

ees

N. betuloides

(b)

02468

1012141618

1800 1820 1840 1860 1880 1900 1920 1940 1960 1980 2000

Decade

Per

cent

age

of tr

ees

N. betuloides

N. pumilio


73

Figure 3.6: Examples of radial growth patterns of juvenile trees growing surrounding

canopy gaps in both forests. Radial growth in solid line and growth increases (%) in

dashed line. The arrowhead and the whole arrow indicate moderate and major periods of

release, respectively.

The duration of the periods of release was very similar for the two species. On

average, both N. betuloides and N. pumilo in the mixed forest enjoyed 17 years

moderate release and 12 years major release. In the pure forest, N. betuloides juvenile

trees enjoyed an average of only 15 years moderate and 11 years major release (Table

3.4). This indicated, therefore, that 22 % of the lifespan of N. betuloides juvenile trees

growing in the pure forest was characterised by a period of release and 23 % in the

mixed forest. Similarly, 21 % of the lifespan of N. pumilio was spent in release (Table

3.4).

0

2

4

6

8

1850 1875 1900 1925 1950 1975 2000

Year

Rad

ial g

row

th (m

m y

r-1)

-200

-150

-100

-50

0

50

100

150

200

Gro

wth

incr

ease

(%)

▼ ▼

N. betuloides in the pure stand

0

2

4

6

8

1850 1875 1900 1925 1950 1975 2000

Year

Rad

ial g

row

th (m

m y

r-1)

-200

-150

-100

-50

0

50

100

150

200

Gro

wth

incr

ease

(%)

▼ ▼

N. betuloides in the mixed stand

0

2

4

6

8

1850 1875 1900 1925 1950 1975 2000

Year

Rad

ial g

row

th (m

m y

r-1)

-200

-150

-100

-50

0

50

100

150

200

Gro

wth

incr

ese

(%)

N. betuloides in the pure stand

0

2

4

6

8

1850 1875 1900 1925 1950 1975 2000

Year

Rad

ial g

row

th (m

m y

r-1)

-250-200-150-100-50050100150200250

Gro

wth

incr

ease

(%)

N. betuloides in the mixed stand

0

2

4

6

8

1850 1875 1900 1925 1950 1975 2000

Year

Rad

ial g

row

th (m

m y

r-1)

-200-150

-100-50

050

100150

200250

Per

cent

age

grow

th c

hang

e

▼

N. pumilio in the mixed stand

0

2

4

6

8

1850 1875 1900 1925 1950 1975 2000

Year

Rad

ial g

row

th (m

m y

r-1)

-100-50050100150200250300350400450

Gro

wth

incr

ease

(%)

N. pumilio in the mixed stand


74

Frequent periods of moderate release after 1870 were detected for the pure N.

betuloides forest and after 1860 for the mixed forest (Figure 3.7). However, the periods

of release appeared to increase after 1930-1940. The distribution of the major periods of

release revealed an increase in the frequency of disturbances affecting the mixed forest

after 1930. The frequency of major disturbances affecting the pure forest was lower, and

only in the years since 1940 (Figure 3.7). These results suggest a possible change to

both the frequency and the intensity of the disturbances impacting upon these forests, as

caused possibly by both autogenic and allogeneic factors.

Figure 3.7: Periods of (a) major and (b) moderate release for juvenile trees growing

surrounding canopy gaps in both forests in the decades between 1850 and 1980.

A comparison of the average annual radial growth rates for both species during

periods of release and while restricted growth (Table 3.4) revealed that N. betuloides

growing in the pure forest had the lowest annual radial increments (2.4 %), and the

mean annual radial growth under restricted growth and in release did not show

statistical differences (Mann-Whitney U test, p= 0.874). In the mixed forest the highest

(a)

0

5

10

15

20

25

1850 1860 1870 1880 1890 1900 1910 1920 1930 1940 1950 1960 1970 1980

Decade

Num

ber o

f maj

or re

leas

es N. betuloides N. betuloides - N. pumilio

(b)

0

10

20

30

40

50

1850 1860 1870 1880 1890 1900 1910 1920 1930 1940 1950 1960 1970 1980

Decade

Num

ber o

f mod

erat

e re

leas

es

N. betuloides N. betuloides - N. pumilio


75

radial growth increments observed were those of N. pumilio (18.4 %), whit annual

radial growths in release statistically higher than during restricted growth (Mann-

Whitney U test, p = 0.028). In the case of N. betuloides growing in the mixed forest,

there was also an increment in the mean annual growth (11.2 %), albeit it was not

significant (Mann-Whitney U test, p = 0.260) when the mean annual growth in

restricted growth and in release were compared.

3.5 Discussion

3.5.1 Causes and Characteristics of Canopy Gaps

In the pure evergreen N. betuloides forest the average sizes of canopy and

expanded gaps were 51 and 164 m2 (Table 3.2), respectively. These were smaller than

canopy and expanded gaps observed in other studies of Nothofagus dominated forests in

South America and in New Zealand (Tables 3.5, 3.6 and 3.7). In the mixed evergreen-

deciduous N. betuloides – N. pumilio forest the canopy and expanded gaps were 107

and 299 m2 (Table 3.2), respectively. These larger gap sizes were similar to the findings

from elsewhere (Tables 3.5, 3.6 and 3.7).

The gap fractions measured in the pure N. betuloides forest were 2.0 % for the

canopy gaps and 5.2 % for the expanded gaps, and higher in the mixed N. betuloides –

N. pumilio forest with 4.6 and 11.6 %, respectively (Table 3.2). In both cases, the

proportion of gaps in the forests was generally smaller than recorded in other

Nothofagus dominated forests (Tables 3.5, 3.6 and 3.7). Indeed, the results obtained for

the pure forest reveal the smallest documented gap fraction.

The average number of gap-makers was similar in both forests, at 2.2 and 2.5

dead trees per gap in the pure and in the mixed forest, respectively (Table 3.3). This

feature, namely multiple gap-makers per gap, is similar to observations from Tierra del

Fuego, where many gaps underwent secondary expansion as a result of the death of

trees bordering the gaps (Table 3.8). This has also been observed in New Zealand’s

Nothofagus forests (Table 3.9).


76

Table 3.5: Canopy and expanded gap characteristics in some Nothofagus dominated

forests of Tierra del Fuego. Numbers in brackets are ranges. A ‘-’ means that no figure

was given in the reference.

Ref

eren

ces

Reb

ertu

s and

Veb

len

(199

3a)

Cue

vas (

2003

)

Cav

iere

s and

Faja

rdo

(200

5)

Expa

nded

- - - - - - - - - -

Gap

frac

tion

(%)

Can

opy

9.1

34.8

15.6

12.1

11.6

- - - - -

Expa

nded

208

(55-

1665

)

173

(66-

1249

)

268

(76-

>147

6)

- - - - -

Mea

n ga

p ar

ea (m

2 ) ± S

E

Can

opy

106

(19-

372)

61 (2

6-89

8)

104

(18-

1476

)

90.6

± 1

7.5

62.7

± 1

2.4

36.0

± 9

.3

11.1

± 0

.8

-

Nº

gaps

disc

rete

gap

s wer

e no

t app

aren

t

25

30

15

33

20

11

3 35

Alti

tude

(m. a

.s.l.)

400

300

150

350

350

450

530

630

690 -

Loca

tion

54º5

4’S,

66º

57’W

54º5

5’S,

66º

54’W

54º5

5’S,

66º

57’W

54º5

5’S,

66º

54’W

54º4

4’S,

67º

53’W

54º4

4’S,

68º

00’W

54º1

3’S,

68º

41’W

54º1

2’S,

68º

45’W

Fore

st c

ompo

sitio

n

N. b

etul

oide

s

N. b

etul

oide

s –

D w

inte

ri

N. p

umili

o –

N. b

etul

oide

s

N. p

umili

o

N. p

umili

o

N. p

umili

o


77


forests of South America (Continent). Numbers in brackets are ranges. A ‘-’ means that

no figure was given in the reference.

Ref

eren

ces

Hei

nem

ann

et

al. (

2000

)

Faja

rdo

and

de

Gra

af (2

004)

Veb

len

(198

5)

Veb

len

(198

9)

Expa

nded

-

23.8

24.4

13.3

26

29

Gap

frac

tion

(%)

Can

opy

-

13.6

12.1

8.6 - -

Expa

nded

413

± 34

.6

289

(126

-744

)

189

(60-

394)

120-

1532

294-

1144

334-

3462

Mea

n ga

p ar

ea (m

2 ) ± S

E

Can

opy

95 ±

13.

2

113

(21-

312)

80 (6

-229

)

- - -

Nº

gaps

15

15

20

35

12

7

Alti

tude

(m. a

.s.l.)

1200

1500

900

800

850

850

Loca

tion

41ºS

, 71º

W

36º6

0’S,

71º

30’W

45º5

2’S,

72º

00’W

39º3

3’S

40º3

8’S

40º3

8’S

Fore

st c

ompo

sitio

n

N. p

umili

o

N. p

umili

o

N. d

ombe

yi –

Laur

elio

psis

phili

ppia

na –

Saxe

goth

aea

cons

picu

a

N. d

ombe

yi –

Aust

roce

drus

chile

nsis

N. d

ombe

yi –

A. c

hile

nsis

–

N. a

ntar

ctic

a


78


forests of New Zealand. Numbers in brackets are ranges. A ‘-’ means that no figure was

given in the reference.

Ref

eren

ces

Stew

art (

1986

)

Stew

art e

t al.

(199

1)

Expa

nded

-

11.4

-12.

9

25.2

-30.

7

24.8

-28.

8

Gap

frac

tion

(%)

Can

opy

8.6-

37.3

3.7-

4.1

14.4

-14.

9

9.8-

10.7

Expa

nded

-

260

(126

-443

)

295

(91-

939)

353

(112

-834

)

Mea

n ga

p ar

ea (m

2 ) ± S

E

Can

opy

100-

1000

68 (2

4-20

0)

140

(33-

528)

107

(33-

490)

Nº

gaps

- 50

50

50

Alti

tude

(m. a

.s.l.)

-

400-

500

400-

500

400-

500

Loca

tion

45º0

2’S,

167

º34’

E

42º1

3’S,

172

º15’

E

42º2

0’S,

172

º12’

E

42º2

3’S,

172

º16’

E

Fore

st c

ompo

sitio

n

N. m

enzi

esii

–

Wei

nman

nia

race

mos

a

N. f

usca

–

N. m

enzi

essi


79

Table 3.8: Mean number of gap-makers per gap and causes of tree mortality in some

Nothofagus dominated forests of South America. A ‘-’ means that no figure was given

in the reference.

Ref

eren

ces

Reb

ertu

s and

Veb

len

(199

3a)

Faja

rdo

and

de

Gra

af (2

004)

Veb

len

(198

5)

Veb

len

(198

9)

Oth

er

17

0 11

17

0 0

Stan

ding

dea

d

- 13

3 - 0 0

Upr

ootin

g

45

40

37

29

82

74

Tree

mor

talit

y ty

pes

(%)

Snap

38

47

49

54

18

26

Mea

n no

. gap

-mak

ers

per g

ap

- - 4 4 4 2 1.8 1 - -

Alti

tude

(m. a

.s.l.)

400

300

150

350

350

1500

900

800

850

850

Loca

tion

54º5

4’S,

66º

57’W

54º5

5’S,

66º

54’W

54º5

5’S,

66º

57’W

54º5

5’S,

66º

54’W

54º4

4’S,

67º

53’W

54º4

4’S,

68º

00’W

36°6

0’S,

71°

30’W

45°5

2’S,

72°

00’W

39°3

3’S

40°3

8’S

40°3

8’S

Fore

st c

ompo

sitio

n

Sout

h A

mer

ica

– Ti

erra

del

Fue

go

N. b

etul

oide

s

N. b

etul

oide

s –

D w

inte

ri

N. p

umili

o –

N. b

etul

oide

s

N. p

umili

o

Sout

h A

mer

ica

– C

ontin

ent

N. p

umili

o

N. d

ombe

yi –

L. p

hilip

pian

a –

S. c

onsp

icua

N. d

ombe

yi –

A. c

hile

nsis

N. d

ombe

yi –

A. c

hile

nsis

–

N. a

ntar

ctic

a


80

Table 3.9: Mean number of gap-makers per gap and causes of tree mortality in some

Nothofagus dominated forests of New Zealand.

Ref

eren

ces

Stew

art e

t al.

(199

1)

Oth

er

2 1 5

Stan

ding

dea

d

37

40

17

Upr

ootin

g

25

22

25

Tree

mor

talit

y ty

pes

(%)

Snap

33

37

53

Mea

n no

. gap

-mak

ers

per g

ap

2.6

5.3

3.9

Alti

tude

(m. a

.s.l.)

400-

500

Loca

tion

42°1

3’S,

172

°15’

E

42°2

0’S,

172

°12’

E

42°2

3’S,

172

°16’

E

Fore

st c

ompo

sitio

n

N. f

usca

–

N. m

enzi

essi


81

The wind-induced snapping and uprooting of trees were the most common types

of mortality observed. In the pure N. betuloides forest 52 % of the gap-makers were

snapped and 30 % uprooted (Table 3.3). This also appears to be characteristic of other

Nothofagus dominated forests in South America and New Zealand (Tables 3.8 and 3.9).

On the other hand in the mixed N. betuloides – N. pumilio forest, the uprooting of trees

was by far the most frequent cause of tree mortality, responsible for 70 % of the gap-

makers, compared to 24 % snapped trees (Table 3.3). This largely corresponded with

the findings from Tierra del Fuego (Table 3.8).

The lower resistance to stem breakage and the prevalence of snapped trees as

gap-makers in the evergreen N. betuloides forest might be related to either a dieback

phenomenon or a disease-decline (Rebertus and Veblen 1993b, Rebertus et al. 1993),

because of the rotting limbs in the dominant and emergent canopy trees (personal

observation). Another explanation is that the intertwining roots of adjacent trees may

help to maintain tree stability until the roots and trunks of individual trees are weakened

by rot (Alfonoso 1940). An additional cause could be related to the abundance of the

magellanic woodpecker (Campephilus magellanicus (King.)). Its occurrence in forests

in Tierra del Fuego has been related to the density of N. betuloides and the occurrence

of snags (Vergara and Schlatter 2004). Magellanic woodpeckers primarily consume the

larvae of wood-boring coleopterans in large and decaying trees, and also drill holes in

large and healthy trees to access the phloem sap (Schlatter and Vergara 2005). Although

not fatal in itself, the activity of the woodpecker may lead to secondary damage by

diseases and insects. Insects and extreme climatic events may also reduce the vigour of

trees, inducing partial crown mortality. This, too, facilitates fungal attack and heart rot

of the bole. This increases the soft substrate and the area available for cavity creation by

magellanic woodpeckers (Ojeda et al. 2007).

Shallow rooting, topographic exposure and poor vitality can promote tree

uprooting (Schaetzl et al. 1989). In the Río Cóndor area shallow rooting is a feature of

mixed N. betuloides – N. pumilio forests, with roots often only able to penetrate soils to

depths of less than 35 cm (Gerding and Thiers 2002). This, and the lower stocking

density in the mixed forest (604 trees ha-1, Table 3.1), may increase the susceptibility of

trees to uprooting.


82

3.5.2 Disturbance Dynamics and Radial Growth Responses of Juvenile Trees

The low frequency of major growth releases in the pure evergreen N. betuloides

forest showed an absence of large-scale disturbance between 1880 and 1940 (Figure

3.7). The uneven-aged canopy structure indicated many decades of stability, with the

canopy mainly affected by fine-scale disturbances leading to the creation of small

canopy gaps. This corresponds with the findings from uneven-aged forests dominated

by Nothofagus fusca (Hook. F.) Oerst. at lower altitudes in New Zealand (June and

Ogden 1978).

In South American Nothofagus forests large-scale disturbances at higher

elevations or latitudes can induce regeneration patterns forming even-aged stands

(Veblen et al. 1996). Examples on Tierra del Fuego are the pure, even-aged secondary

N. betuloides forests established either after fire (Martínez Pastur et al. 2002), or after

periods of colonisation during which the forests were selectively logged, burned and

grazed (Cruz et al. 2007), or the common even-aged structures of N. pumilio forests

resulted of large blow-downs (Rebertus et al. 1997). However, the presence of small

canopy gaps during the last 150 years and the indications of senescence in the larger

canopy trees in the pure N. betuloides forest might represent a gradual break-up of an

even-aged structure (Rebertus et al. 1993).

The release patterns in the pure N. betuloides forest were characterised by longer

periods of restricted growth and low to moderate release. Here trees rely on the presence

of only very small canopy gaps in order to reach maturity. This affects the radial growth

responses of the trees, and reflects the capacity of N. betuloides to remain in shaded

conditions in the understorey for long periods (Rebertus and Veblen 1983a). According

to Veblen et al. (1996), N. betuloides can survive in a suppressed state as advance

regeneration for more than one hundred years. Typically the radial increment of N.

betuloides growing naturally is < 1.0 mm year-1, as observed by Young (1972). Similar

patterns have been documented for Nothofagus forests located in New Zealand, where

Nothofagus menziesii (Hook. F.) Oerst is able to persist and grow under more closed

canopies than other Nothofagus species (Wardle 1984).

N. betuloides can also behave as a pioneer species, often the first to establish on

moraines and deglaciated areas in Patagonia (Pisano 1978, Armesto et al. 1992, Moore

and Pisano 1997). This feature has also been described for N. menziesii after glacial

retreat in New Zealand (Wardle 1984). In a pure, even-aged secondary forest of N.


83

betuloides established after a fire at the end of 1950s, the average radial increment

measured over a the last four years of the study was approximately 0.87 mm year-1

(Martínez Pastur et al. 2002). This was higher than the increment recorded for the

juvenile trees of N. betuloides growing beneath the edges of canopy gaps in the uneven-

aged pure forest (0.60-0.65 mm year-1).

Individuals of both Nothofagus species growing in the mixed evergreen-

deciduous N. betuloides – N. pumilio forest are released by the creation of canopy gaps,

allowing them to grow into the main canopy (Veblen et al. 1996). Over the last few

decades and under the present climate conditions, the occurrence of major disturbances

appears to have increased in this forest type. A similar dynamic has also been recorded

in old-growth North Patagonian rain forests in Chile (Gutiérrez et al. 2004). The high

frequency of periods of moderate release indicates that small scale disturbances affect

this forest, resulting in a reduction in the recruitment of N. pumilio. As a consequence,

N. betuloides has tended to dominate the forest composition. This pattern has been also

observed in mixed Nothofagus forests in New Zealand. In these forests, in the absence

of disturbance the more shade tolerant species, N. menziessi, tends to replace both

Nothofagus solandri var. cliffortioides (Hook. F.) Poole and N. fusca in dense stands

(Wardle 1984, Ogden 1988).

The recruitment of N. pumilio in the mixed N. betuloides – N. pumilio forest

might also be impeded by the harmful effects of the large and very abundant native

herbivore Lama guanicoe Müller. L. guanicoe browses the deciduous Nothofagus

species present in Tierra del Fuego (Raedeke 1980), influencing the regeneration

process in disturbed N. pumilio forest (Arroyo et al. 1996, Rebertus et al. 1997). The

development of the regeneration of N. pumilio in small canopy gaps has been observed

to be more vulnerable than that of the evergreen N. betuloides due to the effects of L.

guanicoe (Cavieres and Fajardo 2005). This may serve to threaten the successful

recruitment of the deciduous species in the mixed N. betuloides – N. pumilio forest.

N. pumilio demonstrated greater radial growth increments upon gap creation

than N. betuloides. This behaviour is similar to findings by Runkle et al. (1997).

Studying sapling diameter growth in gaps in forest dominated almost completely by N.

fusca and the more shade tolerant N. menziessi, they observed that in general the faster

growth rates of N. fusca were sufficient to balance the greater abundance of N.

menziesii. Therefore, neither species attained dominance under gaps created in the

canopy.


84

3.6 Conclusions

The study of two uneven-aged forests dominated by N. betuloides on Tierra del

Fuego, one pure and the other mixed with N. pumilio, revealed that it is reasonable to

assume that (1) the response of N. betuloides and N. pumilio to disturbances is very

irregular and that many longer periods of restricted growth are followed by moderate or

major releases; (2) small canopy gaps are the prevalent disturbance type and that large

tree-fall gaps are very episodic; (3) the recruitment of both Nothofagus species occurs

continually; (4) the mean number of gap-makers is similar in both forests; (5) the most

common cause of tree mortality is snapping in the pure forest and uprooting in the

mixed forest; (6) there is a reduction in the recruitment of N. pumilio in the mixed

forest, and N. betuloides tends to dominate the forest composition; (7) the frequency and

intensity of disturbance in these forests appears to have changed since 1930-40; and (8)

the greatest changes to the radial increment occurred during periods of release. In the

mixed forest the radial growth of both species increased on average by 11.2 and 18.4 %

for both N. betuloides and N. pumilio, respectively. Whereas in the pure forest the

increment of N. betuloides growing beneath the small canopy gaps increased by only

2.4 %. The evergreen N. betuloides demonstrated greater shade-tolerance on Tierra del

Fuego.

Chapter 4 Programmes for analysing hemispherical photographs

85

CHAPTER 4

COMPARISON OF CANOPY STRUCTURES AND SOLAR

RADIATION TRANSMITTANCES ESTIMATED USING FOUR

DIFFERENT PROGRAMMES FOR THE ANALYSIS OF

HEMISPHERICAL PHOTOGRAPHS

4.1 Abstract

There have been many studies involving the use of hemispherical photographs to

estimate indirectly forest light environments. A variety of commercial and free software

packages are available for the analysis of hemispherical photographs. The costs of

investment might represent an advantage of the free programmes over the commercial,

but as yet little has been documented about the differences in their outputs and in the

technical applications from a user (ecologist and forester) perspective. The objective of

the study was to show the most commonly employed canopy structure (canopy

openness and effective plant area index) and solar radiation variables (direct, diffuse

and global solar radiation transmittances) estimated from digital hemispherical

photographs taken under two forest canopy conditions (gap and closed canopy), and in

forests located at different latitudes (northern, equatorial and southern hemisphere). The

hemispherical photographs were analysed using one commercial (HemiView) and three

free programmes (Gap Light Analyzer, hemIMAGE and Winphot). The results

demonstrated that all of the programmes computed similar estimates of both canopy

structures and below-canopy solar radiation. Only the results relating to the effective

plant area index with an ellipsoidal leaf angle distribution calculated for the canopy

gaps demonstrated a weak correlation. Other user aspects are also discussed, such as

costs, image formats, computer system requirements, etc.

Keywords: hemispherical photography; solar radiation transmittances; canopy

openness; plant area index; HemiView; Gap Light Analyzer; hemIMAGE; Winphot


86

4.2 Introduction

The greatest importance of solar radiation for plant life lies in the plants’

dependence upon photosynthesis for growth and development, and the dependence in

turn of photosynthesis on light (Barnes et al. 1998). There is, consequently, much

interest in measuring light. Several instruments have been developed to measure either

directly or indirectly the forest understorey light environment. Many comparisons of

direct and indirect methods for the estimation of below-canopy irradiation have been

conducted in order to determine the best way to measure the light environment

(Chazdon and Field 1987, Rich et al. 1993, Roxburgh and Kelly 1995, Comeau et al.

1998, Gendron et al. 1998, Clearwater et al. 1999, Machado and Reich 1999,

Engelbrecht and Herz 2001, Ferment et al. 2001, Bellow and Nair 2003). However,

many ecologists and foresters prefer indirect means of light estimation due to the

difficulties inherent in measuring light directly (Jennings et al. 1999).

Since its introduction (Evans and Coombe 1959), hemispherical photography

has become a widely applied means of calculating the light environment under forest

canopies. In many studies carried out worldwide, this method has been applied not only

to the measurement of light environments in the understorey, but also to the estimation

of canopy structure variables. A number of studies have demonstrated a high level of

agreement between both estimates (Rich et al. 1993, Comeau et al. 1998; Gendron et al.

1998, Clearwater et al. 1999, Engelbrecht and Herz 2001). However, in deeply shaded

environments, the applicability of hemispherical photographs for the calculation of

understorey light environments still needs to be verified conclusively (Roxburgh and

Kelly 1995, Machado and Reich 1999).

Hemispherical photography has been used to estimate canopy structure and

below-canopy solar radiation environments in a variety of forest types, including

temperate (Canham et al. 1990, Roxburgh and Kelly 1995, Collet and Chenost 2006),

tropical (Canham et al. 1990, Rich et al. 1993, Clearwater et al. 1999, Engelbrecht and

Herz 2001, Ferment et al. 2001) and boreal forests (Wright et al. 1998, Bartemucci et al.

2006).

With advances in technology, the use of conventional film photography has

largely given way to digital photography (Hale and Edwards 2002). To date, there have

been several comparisons made between the data provided by analogue and digital


87

photography, demonstrating comparable results (Englund et al. 2000, Frazer et al. 2000,

Hale and Edwards 2002). In the case of digital photography, the effects of image quality

and image size have also been evaluated (Englund et al. 2000, Frazer et al. 2001, Inoue

et al. 2004a), revealing some differences in the light and the forest canopy structure

estimates associated with digital camera type and image quality.

The theoretical basis for estimating the various components of solar radiation

using hemispherical photography were developed by Anderson (1964a, 1966). This

initial manual, and very tedious, means of assessment was problematic because of the

time required to analyse the complex photographs (Chan et al. 1986). In the intervening

period a variety of semi-automated and computerised techniques have been developed

for the calculation of canopy structure and the estimation from hemispherical

photographs of the diffuse and direct light penetrating through openings in the canopy

(Chan et al. 1986, Chazdon and Field 1987, Becker et al. 1989, Barrie et al. 1990, Smith

and Somers 1993, Walter and Torquebiau 2000). A range of software packages are

currently available for the analysis of hemispherical photographs (Comeau 2000). These

include both commercial and free versions, with the latter available for download from

the internet.

The question that arises now is whether there are differences in the usability and

the results provided by the different software solutions. Frazer et al. (1997) compared

two canopy characteristics (percent open sky and effective leaf area index) using two

programmes (Hemiphot and PAMAP GIS) in eight chronosequences of coastal

temperate rainforest in British Columbia, Canada. The authors documented high

correlations between the percent open sky (R2 = 0.98) and the effective leaf area index

(R2 = 0.96) results produced by the two programmes. Brunner (2002) wrote that the

hemIMAGE software calculates results for transmitted light very similar to those

computed by the GLI-C and Winphot software, but that the results differed significantly

from those produced by the Solarcalc software. There remains as yet a lack of studies

covering a wide range of environments, which this paper seeks to address; for example,

to determine whether different latitudes influence the below-canopy solar radiation

environments estimated from hemispherical photographs. Furthermore, there is as yet

no published study comparing the programmes commonly used for computing these

variables from hemispherical photographs, namely Gap Light Analyzer (Frazer et al.

2001, Bartemucci et al. 2006, Fahey and Puettmann 2007), hemIMAGE (Gärtner and

Reif 2005, Collet and Chenost 2006, Gáldhidy et al. 2006), HemiView (Hale and


88

Edwards 2002, Coops et al. 2004, Pardos et al. 2007) and Winphot (Machado and Reich

1999, Englund et al. 2000, Bellow and Nair 2003).

The objective of this study, therefore, was to compare the canopy structure

variables and solar radiation transmission estimates calculated using the four different

programmes. In order to draw general conclusions, the hemispherical photographs used

in the study 1) depicted two extreme forest canopy conditions (gap and closed canopy),

and 2) were drawn from forests at different latitudes – northern hemisphere, equatorial

and southern hemisphere – which have different canopy structure and solar radiation

characteristics. With this paper, we want to share our experience with the different

programmes with other users in order to help them to decide which programme provides

the information needed to answer their particular research questions.


4.3.1 Study Areas

The hemispherical photographs used in the evaluation were obtained from three

different forest ecosystems in which studies of gaps, regeneration and vegetation

dynamics were taking place. In order to have a variety of canopy structures and solar

radiation characteristics, contrasting latitudes and forest ecosystems were selected. The

details regarding the three sites are presented in Table 4.1. Besides the considerable

differences in the density of the forests studied, the different ecosystem characteristics

represented a complete range of broadleaf forest ecosystems. A gradient covering a

wide range of forest canopy conditions was sampled within each of the three study

areas.


89

Table 4.1 Study areas and their characterisation.

Top

heig

ht

(m)

34

40

31

Bas

al a

rea

(m2 h

a-1)

36

53

105

Stoc

king

(tree

s ha-1

)

519

700

1,36

2

Fore

st ty

pe

Tem

pera

te m

ixed

dec

iduo

us b

eech

fore

st o

n lim

esto

ne, d

omin

ated

by

Fagu

s syl

vatic

a

Trop

ical

clo

ud fo

rest

, ver

y hu

mid

subm

onta

ne fo

rest

Col

d te

mpe

rate

eve

rgre

en fo

rest

,

dom

inat

ed b

y N

otho

fagu

s bet

uloi

des

Slop

e

(°) 0 0

0-10

Alti

tude

(m a

.s.l.)

430

1,43

5

190

Loca

tion

Web

erst

edte

r Hol

z

Hai

nich

Nat

iona

l Par

k

(Ger

man

y) 1

(51°

01’

N, 1

0° 0

4’ E

)

Sier

ra d

e Le

ma

Fore

st,

Can

aim

a N

atio

nal P

ark

(Ven

ezue

la) 2

(05°

53’

N, 6

1° 2

6’ W

)

Río

Cón

dor,

Tier

ra d

el F

uego

(Chi

le) 3

(53°

59’

S, 6

9° 5

8’ W

)

1 But

ler-

Man

ning

(200

8), 2 H

erná

ndez

and

Cas

tella

nos (

2006

), 3 T

able

3.1


90

4.3.2 Photographic Source Material

A total of 78 hemispherical photographs were incorporated in the evaluation.

These comprised 13 photos made in gaps at each location and a further 13 under closed

canopies. All of the images were made using a digital Nikon Coolpix 990® camera

(Nikon Corporation, Tokyo, Japan) fitted with a Nikon FC-E8® fisheye converter

(Nikon Corporation, Tokyo, Japan), which has a field of view of 183°. The camera was

mounted on a tripod at a height of approximately 1.3 m above the ground. In the Sierra

de Lema cloud forest the photos were taken at a height of between 1.5-1.7 m above the

ground. The camera and the lens were arranged horizontally with the aid of a spirit

level, and pointed to the magnetic north. Automatic settings for aperture width and

shutter speed were selected (Englund et al. 2000, Inoue et al. 2004a). Details of the

photograph formats are presented in Table 4.2. Comparisons of the different image

qualities and image sizes obtained using a digital Nikon Coolpix 990 camera have

revealed no statistically discernable differences in the estimates of either gap fractions

or canopy openness (Inoue et al. 2004a). In so far as it was possible, the photographs

were only taken when the sky overhead was almost uniformly cloudy, or else shortly

after sunset. The reason for this was to avoid the occurrence of bright regions around the

sun and light reflection off foliage and woody structures, which can render thresholding

difficult.

4.3.3 Image Processing

All of the images were first cropped to squares to clearly define the image

boundaries and the image centre (refer to Brunner 2002). The reason for this is that the

fisheye photograph is a projection on a plane of a hemisphere, with the zenith at the

centre of the image and the horizons at the edges. This image renders it possible to

ascertain the distribution of canopy openings, and to estimate the solar radiation that

penetrates below the plant canopy (Rich 1990). The images were subsequently

converted to grey scales (256 levels) using the Adobe Photoshop® software (version 7.0

for Windows®, Adobe Systems Inc., San Jose, CA, USA).

A threshold value was then set for the separation of canopy and sky elements,

producing a binary black and white image (Anderson 1964a). This has in the past often

been performed manually (Roxburgh and Kelly 1995, Englund et al. 2000, Frazer et al.


91

2001, Brunner 2002, Hale and Edwards 2002), but this approach has been criticised for

being too subjective, with human interpretation a major source of error (Inoue et al.

2004a, A. Brunner, personal communication). Recently, automatic threshold setting

methods (Ishida 2004, Nobis and Hunziker 2005, Schwalbe et al. 2006) and automatic

gap fraction assessments for digital hemispherical photographs taken in forests

(Jonckheere et al. 2005) have been developed. In this study, all of the digital images

were converted to binary black and white pixels employing an automatic threshold

setting method based on edge detection (Nobis and Hunziker 2005), using the SideLook

software (http://www.appleco.ch/cms/index.php, Nobis 2005). The automatic threshold

was applied to all images by choosing the ‘edge value’ modus, and where necessary

‘local maxima’ were tested) (Nobis 2005).

Table 4.2: Details of the digital hemispherical photograph formats.

Location Colour Image

Quality* Pixels* Format

Weberstedter Holz

Hainich National Park

(Germany)

black and

white BASIC 2048 × 1536

1:16 compression

JPEG

Sierra de Lema Forest,

Canaima National Park

(Venezuela)

colour BASIC 2048 × 15361:16 compression

JPEG

Río Cóndor,

Tierra del Fuego (Chile)

black and

white HI 2048 × 1536

uncompressed

TIFF

* The Nikon Coolpix 990 allows for four image qualities, namely basic, normal, fine

and hi. Three image sizes are also possible; the largest is 2048 x 1536 pixels, the

medium 1024 x 768 pixels, and the lowest 640 x 480 pixels. Thus, an 8MB

CompactFlash memory card can store approximately 0 hi-quality images (each image is

larger than 9MB); between 5 and 48 fine-quality images, depending on the image size

used; 10 to 91 normal-quality images, and 19 to 161 basic-quality images.


92

4.3.4 Image Analyses

4.3.4.1 The software used and their settings

All images were analysed using one commercial software package called

HemiView (http://www.delta-t.co.uk, Delta-T Devices, Cambridge, UK, Rich et al.

1999), and three free programmes: (a) Gap Light Analyzer version 2.0

(http://www.ecostudies.org/gla/, Frazer et al. 1999), (b) hemIMAGE version 15-09-

2002 (http://www.umb.no/ina/ansatte/andrb_e.php, Brunner 2002) and (c) Winphot

version 5.0 (http://www.bio.uu.nl/~herba/Guyana/winphot/wp_index.htm, ter Steege

1996). All photographs were saved in the BMP format for analysis using HemiView,

Gap Light Analyzer and Winphot. The programme requirements of the hemIMAGE

software necessitated that the hemispherical photographs be converted to the GIF

format (Brunner 2002).

The lens used was the Nikon fisheye FC E8. The lens was originally designed to

produce a simple polar or equiangular projection (Herbert 1987), but new calibrations to

this lens type have since been made (Frazer et al. 2001, Inoue et al. 2004b). Three of the

programmes compared provided the option to set different calibrations to correct for the

lens distortion. The Coolpix 900 was selected for HemiView (Hale and Edwards 2002).

A third-order polynomial derived by Frazer et al. (2001) was set in the hemIMAGE

software (Nikon-Coolpix 950). This projection distortion is very close to the polar or

equiangular projection, which was also imported directly into Gap Light Analyzer as a

custom lens file (*.lns) (G. Frazer, personal communication). Lens options cannot be set

in Winphot, which assumes a polar or equiangular projection (H. ter Steege, personal

communication).

A uniform overcast sky (UOC) model was selected to describe the light intensity

of the diffuse sky (Monteith and Unsworth 1990). This model considers all regions of

the sky to be equally bright. As no actual measurements of the diffuse and direct

radiation above the study areas were available, a relative proportion of direct and diffuse

radiation equal to 0.5 was assumed for the three latitudes (Canham et al. 1990),

providing for a comparable and uniform data base. The results were calculated on the

basis of the specific vegetation periods corresponding to each site, namely April to

September in the Weberstedter Holz (Germany), all year round in the Sierra de Lema

(Venezuela) and from October to March in the Río Cóndor (Tierra del Fuego, Chile).


93

The option to set the entire year as the vegetation period for the Venezuelan forest did

not present a problem with any of the programmes, nor did the setting of a specific

vegetation period for the other two forests in either Gap Light Analyzer or hemIMAGE.

With HemiView specific periods can be calculated from the outputs obtained (refer to

Rich et al., 1999). Using Winphot a specific vegetation period comprising a maximum

of only 12 days can be set. In this case 7 days during the respective vegetation periods

were chosen for the Weberstedter Holz (1st April, 1st May, 1st June, 1st July, 1st August,

1st September and 30th September) and for the Río Cóndor (1st October, 1st November,

1st December, 1st January, 1st February, 1st March and 31st March). The option in

Winphot to include diffuse canopy light, which corresponds to the scattered radiation

transmitted or reflected from foliage, was not selected.

In HemiView and Gap Light Analyzer, the photographs were divided into 16

azimuth and 9 zenith regions (144 sky regions in total). Winphot and hemIMAGE,

alternatively, employ 89 fixed concentric rings, each one corresponding to a circular

sphere segment in the sky hemisphere (ter Steege 1996). These divisions are used by the

programmes to calculate canopy structures and solar radiation transmittances with

greater accuracy.

4.3.4.2 Calculated canopy structure and light environments

From hemispherical photographs it is possible to estimate canopy structure

variables (gap fraction, canopy openness, leaf area index, etc.), as well as potential solar

radiation (above canopy and the proportion transmitted into the forest). The basic and

most common outputs calculated using the programmes compared in this study are

listed in Table 4.3. The solar radiation calculations produced by all four programmes

were cosine-corrected. This is useful when comparing hemispherical radiation flux

estimates and measurements from cosine-corrected light sensors (Rich 1990).

HemiView and hemIMAGE also provide non-cosine-corrected transmitted solar

radiation outputs, which may be desirable for the purposes of measuring solar radiation

from all directions. This is an important feature of potential light interception by non-

flat surfaces (Rich et al. 1999); for example, a plant which has leaves oriented in many

directions (Rich 1990, Brunner 1998).


94

Table 4.3: List of the basic variables computed by the programmes HemiView (HV),

Gap Light Analyzer (GLA), hemIMAGE (hI) and Winphot (Wp). The variables

calculated by a particular programme are indicated by an ‘X.’ A ‘+’ means that the

variable can be derived from other variables computed by the programme.

Wp X

X + X X + + + + X + X

X

X

hI X

X

X

X

X

X

GLA

X

X X X X X X X X + X

X

HV

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X X

X

X

X

cosi

ne-c

orre

cted

non-

cosi

ne-c

orre

cted

cosi

ne-c

orre

cted

non-

cosi

ne-c

orre

cted

cosi

ne-c

orre

cted

non-

cosi

ne-c

orre

cted

cosi

ne-c

orre

cted

non-

cosi

ne-c

orre

cted

cosi

ne-c

orre

cted

non-

cosi

ne-c

orre

cted

cosi

ne-c

orre

cted

non-

cosi

ne-c

orre

cted

cosi

ne-c

orre

cted

non-

cosi

ne-c

orre

cted

cosi

ne-c

orre

cted

non-

cosi

ne-c

orre

cted

cosi

ne-c

orre

cted

non-

cosi

ne-c

orre

cted

Bas

ic O

utpu

ts

Abo

ve c

anop

y di

rect

sola

r rad

iatio

n

Abo

ve c

anop

y di

ffus

e so

lar r

adia

tion

Abo

ve c

anop

y gl

obal

sola

r rad

iatio

n

Bel

ow-c

anop

y di

rect

sola

r rad

iatio

n

Bel

ow-c

anop

y di

ffus

e so

lar r

adia

tion

Bel

ow-c

anop

y gl

obal

sola

r rad

iatio

n

Tran

smitt

ed d

irect

sola

r rad

iatio

n

Tran

smitt

ed d

ifuss

e so

lar r

adia

tion

Tran

smitt

ed g

loba

l sol

ar ra

diat

ion

Red

/ fa

r-re

d ra

tio

Gap

-fra

ctio

n (u

nwei

ghte

d op

enne

ss)

Can

opy

open

ness

(wei

ghte

d op

enne

ss)

Effe

ctiv

e le

af a

rea

inde

x

Leaf

ang

le d

istri

butio

n

Sola

r

Rad

iatio

n

Can

opy

The canopy structural characteristics measured were canopy openness (CO) and

effective plant area index (Le). Canopy openness was defined as the relative area of

visible sky weighted by the incident angle of cosine law from a point (Inoue et al.


95

2004a). Le was defined as the sum of all elements blocking canopy light (stems, twigs,

leaves), as hemispherical photographs do not distinguish between opaque objects

(stems) and photosynthetic tissue (Holst et al. 2004).

To obtain the Le, the programmes used methods based upon the determinations

of gap fractions in the canopy and inversion procedures. Thus, Le can be derived by

inverting Eq. (4.1) (Norman and Campbell 1989):

(4.1)

where Ti is the gap fraction at zenith angle θi, Kij is the extinction coefficient for a beam

at zenith angle θi and a leaf inclination angle (αj), and Li is the leaf area index at zenith

angle θi.

Using Eq. (4.1) Gap Light Analyzer and Winphot introduce other relationships

which finally produce Eq. (4.2) for the calculation of Le (Welles and Norman 1991):

(4.2)

where θi corresponds to fixed five viewing angles (7, 23, 38, 53 and 68), Ti is the gap

fraction around each viewing angle in bands of 15 degrees, and Wi is a fixed value

weighted to account for an area correction (0.034, 0.0104, 0.160, 0.218 and 0.484, for

the five angles referred to above). This is similar to the technique employed by the LAI

2000 plant canopy analyser. For the purposes of this evaluation, and as Le was

estimated over zenith angles of 0-75°, it is subsequently denoted Le-75 here.

Le was also estimated using an inversion algorithm for canopy transmission

employing an ellipsoidal leaf angle distribution (Le-E), which is incorporated into

HemiView and Winphot. The basis of this method is that the leaf angle distribution of a

canopy can be represented by the distribution of the area on the surface of an ellipsoid

of revolution (refer to Campbell 1986, Norman and Campbell 1989, Rich et al. 1999).

For an elliptical leaf angle distribution, the extinction coefficient – as defined by Eq.

(4.1) above (Campbell 1986, Norman and Campbell 1989) – is shown in Eq. (4.3):

(4.3)

iiji LxKT −=)ln(

( ) ( )∑=

−=5

1cosln2

iiiie WTL

θ

θ

( )( )[ ]

D

xxKK

i

ii

⎟⎟⎠

⎞⎜⎜⎝

⎛+

==

21

2tan,

2 θθ


96

where θi corresponds to the zenith angles, x corresponds to the ellipsoidal leaf angle

distribution parameter (ELADP), which is a ratio of vertical to horizontal foliage area

projections and describes the shape of the distribution. A spherical distribution occurs

when x = 1, whereas the canopy tends to be vertical and horizontal if x < 1 or x > 1,

respectively (Walter 1989-2006). D is an expression of a normalised ellipse area, which

is given by Eq. (4.4) in Winphot (ter Steege 1996) and by Eq. (4.5) in HemiView

(Wood 2001):

(4.4)

(4.5)

In Winphot x and L in Eqs. (4.1, 4.3 and 4.4) are solved using a Pascal

translation of the Basic programme (Norman and Campbell 1989), which also has been

adapted in HemiView (Rich et al. 1999).

Furthermore, a mean leaf angle or mean tilt angle (A) can be calculated using

Eq. (4.6) (Walter 1989-2006):

(4.6)

where A (the mean leaf angle) is in radians and x corresponds to ELADP.

4.3.5 Statistical Analyses

A linear regression analysis was performed to test the strength of the relationship

between the outputs (solar radiation transmittances and canopy structures) calculated by

the different programmes. The 78 hemispherical photographs used for this study were

stratified into two groups, according to the canopy condition. The analysis was carried

out separately for those images made under canopy gaps (n = 39) and those under

closed canopies (n = 39). For the regression analysis the goodness-of-fit was calculated

using the coefficient of determination (R2), the root mean square error (RMSE) and the

significance of the p-value (Sokal and Rohlf 2000). All of the statistical analyses were

performed using SPSS 15.0 for Windows (SPSS Inc.).

733.0)182.1(774.1 −++= xxD

708.0)12.1(702.1 −++= xxD

65.1)0.3(65.9)( −+= xxA


97

4.4 Results

4.4.1 Comparison of Canopy Structure Estimates

The canopy structures were estimated as percentage canopy openness (CO) and

using two definitions of effective plant area index (Le-E and Le-75) with three

programmes providing these outputs, namely HemiView, Gap Light Analyzer and

Winphot.

The canopies of the three forests were generally very dense, with the estimates

of canopy openness computed ranging between 11.1-22.6 % (Table 4.4).

The estimates of the effective plant area index integrated over the zenith angle 0-

75° (Le-75) were similar in each case, showing a range of values with an average of

between 3-4 m2 m-2 (Table 4.4). However, the maximum values of Le-75 in the

Weberstedter Holz (Germany) estimated by both of the programmes used were very

different (6.8 m2 m-2 with Gap Light Analyzer and 10.7 m2 m-2 with Winphot). The

tendency indicated by these values was also similar to the computed estimates of

effective plant area index using an ellipsoidal leaf angle distribution (Le-E) for all sites

(Table 4.4). In the case of the Weberstedter Holz there were again differences between

the maximum values computed (4.1 m2 m-2 with HemiView and 11.9 m2 m-2 with

Winphot).

The estimates of canopy openness revealed a strong, and high, correlation

between the different software outputs (Table 4.5 and Figure 4.1). This was the case for

both the hemispherical photographs taken beneath canopy gaps and those from under

closed canopies (n = 39; R2 > 0.993; p < 0.01 for all).

The computed estimates of effective plant area index (Le-75) were also highly

correlated between the two programmes that calculated this variable (Table 4.6), namely

Gap Light Analyzer and Winphot. The strong relationship was found under closed

canopy conditions (n = 39; R2 = 0.994; RMSE = 0.038; p < 0.01). However, below

canopy gaps one hemispherical photograph showed a strong deviation in the Le-75

estimate (Table 4.6 and Figure 4.1), which produced a decrease in the coefficient of

determination (n = 39; R2 = 0.859; RMSE = 0.492; p < 0.01).


98

Table 4.4: Characteristics of the canopy structures estimated for the three study areas

using the programmes HemiView (HV), Gap Light Analyzer (GLA) and Winphot (Wp).

The variables included are the canopy openness (CO), and the effective plant area

indexes integrated over the zenith angle 0-75° (Le-75) and calculated using an

ellipsoidal leaf angle distribution (Le-E). A ‘-’ means that the programme does not

compute outputs for the variable. Numbers in brackets are ranges.

CO (%) Le-75 Le-E

Weberstedter Holz, Hainich National Park, Germany (51° 01’ N, 10° 04’ E)

HV 18.5 (15.5-21.2) - 2.9 (2.3-4.1)

GLA 18.4 (15.4-21.1) 3.1 (2.5-6.8) -

Wp 19.2 (15.9-22.0) 3.2 (2.4-10.7) 3.2 (2.3-11.9)

Sierra de Lema Forest, Canaima National Park, Venezuela (05° 53’ N, 61° 26’ W)

HV 14.9 (11.2-21.4) - 3.8 (2.8-5.7)

GLA 14.8 (11.1-21.3) 3.8 (3.1-4.9) -

Wp 15.5 (11.4-22.6) 3.8 (3.1-4.9) 3.7 (3.0-4.8)

Río Cóndor, Tierra del Fuego, Chile (53° 59’ S, 69° 58’ W)

HV 14.9 (11.6-20.2) - 3.9 (2.6-6.4)

GLA 14.8 (11.6-20.1) 3.4 (2.4-4.3) -

Wp 15.4 (12.0-21.3) 3.4 (2.4-4.1) 3.3 (2.3-4.0)

The relationships between the estimates of Le-E computed using HemiView and

Winphot were very weak (Table 4.6). The relationship found under the closed canopy

conditions was characterised by large variation (n = 39; R2 = 0.053; RMSE = 1.111; p <

0.05). Below canopy gaps there was no correlation between the estimates provided by

both programmes (n = 39; R2 = 0.000; RMSE = 1.502; p = 0.939). In general, the

estimates of Le-E provided by HemiView deviated from those calculated using

Winphot, and vice versa, with great deviations between individual hemispherical

photographs (Figure 4.1).


99

Table 4.5: Linear regression coefficients for the comparison of the canopy openness

values associated with the two types of canopy condition estimated using the

programmes HemiView (HV), Gap Light Analyzer (GLA) and Winphot (Wp). The two

canopy conditions are canopy gaps and closed canopies. R2 is the coefficient of

determination; RMSE is the root mean square error. * indicates significance at the 5%

level, and ** at the 1% level.

Condition Variable Intercept Slope R2 RMSE p-value n

HV vs GLA 0.033 0.993** 0.998 0.109 0.000** 39

HV vs Wp 0.151 1.035** 0.995 0.184 0.000** 39Canopy

gap GLA vs Wp 0.149 1.041** 0.993 0.210 0.000** 39

HV vs GLA -0.023 0.998** 1.00 0.014 0.000** 39

HV vs Wp -0.044 1.033** 0.997 0.152 0.000** 39Closed

canopy GLA vs Wp -0.018 1.035** 0.997 0.155 0.000** 39

Table 4.6: Linear regression coefficients for the comparison of both effective plant area

indices associated with the two types of canopy condition estimated using the

programmes HemiView (HV), Gap Light Analyzer (GLA) and Winphot (Wp). The

variables are the effective plant area indices integrated over the zenith angle 0-75° (Le-

75) and calculated using an ellipsoidal leaf angle distribution (Le-E). The two canopy

conditions are canopy gaps and closed canopies. R2 is the coefficient of determination;

RMSE is the root mean square error. * indicates significance at the 5% level, and ** at

the 1% level.

Condition Variable Intercept Slope R2 RMSE p-value n

Le-75

Canopy gap GLA vs Wp -1.924** 1.529** 0.859 0.492 0.000** 39

Closed

canopy GLA vs Wp -0.044 0.996** 0.994 0.038 0.000** 39

Le-E

Canopy gap HV vs Wp 3.758** -0.018 0.000 1.502 0.939 39

Closed

canopy HV vs Wp 2.435** 0.273* 0.053 1.111 0.043* 39


100

(a) (b)

(c)

Figure 4.1: Scatter plots of the relationships between (a) the canopy openness estimated

under closed canopies using HemiView and Gap Light Analyzer, (b) the effective plant

area index integrated over the zenith angle 0-75° (Le-75) in canopy gaps estimated

using Gap Light Analyzer and Winphot, and (c) the effective plant area index calculated

from an ellipsoidal leaf angle distribution (Le-E) in canopy gaps estimated using

HemiView and Winphot. The solid lines correspond to the equations in Tables 4.5 (Fig.

a) and 4.6 (Figs. b and c). The broken lines represent a 1:1 reference.


101

4.4.2 Comparison of Solar RadiationTransmittance Estimates

All of the software packages evaluated in this study calculated cosine-corrected

solar radiation transmittance. However, only two estimated the non-cosine-corrected

solar radiation, namely HemiView and hemIMAGE.

4.4.2.1 Cosine-corrected solar radiation transmittance estimates

For all sites, the below-canopy cosine-corrected solar radiation values calculated

from the hemispherical photographs were low to moderate (Table 4.7). The cosine-

corrected direct, diffuse and global solar radiation transmittances ranged between 0.01-

0.41, 0.06-0.25, and 0.04-0.36, respectively. The highest direct and global solar

radiation transmittance values were always computed at the Sierra de Lema forest in

Venezuela.

All of the estimates of the cosine-corrected direct, diffuse and global solar

radiation transmittance under the two canopy variants incorporated in the study, namely

gap and closed canopy, were significant and strongly correlated (n = 39; R2 > 0.886; p <

0.01 for all). Therefore, all four programmes calculated very similar estimates of the

below-canopy solar radiation transmittance under both canopy conditions (Tables 4.8,

4.9 and 4.10 and Figure 4.2). However, in all cases some deviation was observed and

either overestimates or underestimates might be expected.

4.4.2.2 Non-cosine-corrected solar radiation transmittance estimates

In this case, only HemiView and hemIMAGE were compared, as the other two

programmes did not compute these variables. Again, the comparison was based on

hemispherical photos from gaps and closed stands.

As with the previous results, the estimates of the non-cosine-corrected direct,

diffuse and global solar radiation transmittances computed were between low to

medium solar radiation levels (Table 4.7). The estimation of non-cosine-corrected

direct, diffuse, and global solar radiation transmittances ranged between 0.01 and 0.34,

0.04 and 0.13, and 0.03 and 0.23, respectively. Again, the lighter below-canopy

environment was observed in Sierra de Lema Forest (Venezuela).


102

Table 4.7: Characteristics of the solar radiation transmittances of the three study areas

estimated using the programmes HemiView (HV), Gap Light Analyzer (GLA),

hemIMAGE (hI) and Winphot (Wp). The variables included are the direct (DIR),

diffuse (DIF) and global (GLO) solar radiation transmittances. A ‘-’ means that the

programme does not compute outputs for the variable. Numbers in brackets are ranges.

GLO

0.10

(0.0

7-0.

14)

-

0.10

(0.0

5-0.

15)

-

0.11

(0.0

3-0.

22)

-

0.11

(0.0

3-0.

23)

-

0.07

(0.0

4-0.

11)

-

0.07

(0.0

4-0.

10)

-

DIF

0.10

(0.0

7-0.

13)

-

0.10

(0.0

7-0.

13)

-

0.07

(0.0

4-0.

13)

-

0.07

(0.0

4-0.

13)

-

0.07

(0.0

4-0.

12)

-

0.07

(0.0

4-0.

12)

-

Non

-cos

ine-

corr

ecte

d so

lar r

adia

tion

trans

mitt

ance

s

DIR

0.11

(0.0

1-0.

17)

-

0.10

(0.0

1-0.

16)

-

0.16

(0.0

2-0.

34)

-

0.16

(0.0

2-0.

32)

-

0.07

(0.0

2-0.

13)

-

0.07

(0.0

2-0.

13)

-

GLO

0.15

(0.1

0-0.

20)

0.14

(0.0

9-0.

20)

0.14

(0.0

9-0.

20)

0.14

(0.0

7-0.

21)

0.15

(0.0

4-0.

31)

0.15

(0.0

4-0.

30)

0.15

(0.0

4-0.

30)

0.17

(0.0

4-0.

36)

0.10

(0.0

7-0.

17)

0.09

(0.0

6-0.

15)

0.09

(0.0

6-0.

15)

0.09

(0.0

5-0.

14)

DIF

0.16

(0.1

0-0.

23)

0.16

(0.1

0-0.

22)

0.16

(0.1

0-0.

23)

0.17

(0.1

1-0.

24)

0.12

(0.0

6-0.

23)

0.11

(0.0

6-0.

23)

0.11

(0.0

6-0.

23)

0.12

(0.0

6-0.

25)

0.11

(0.0

7-0.

22)

0.11

(0.0

7-0.

21)

0.11

(0.0

7-0.

21)

0.11

(0.0

7-0.

24)

Cos

ine-

corr

ecte

d so

lar r

adia

tion

trans

mitt

ance

s

DIR

0.12

(0.0

1-0.

19)

0.12

(0.0

1-0.

19)

0.12

(0.0

2-0.

19)

0.13

(0.0

2-0.

22)

0.19

(0.0

3-0.

39)

0.18

(0.0

3-0.

37)

0.18

(0.0

2-0.

39)

0.20

(0.0

3-0.

41)

0.08

(0.0

2-0.

15)

0.08

(0.0

2-0.

15)

0.08

(0.0

2-0.

15)

0.08

(0.0

2-0.

14)

Web

erst

edte

r Hol

z, H

aini

ch N

atio

nal P

ark,

Ger

man

y (5

1° 0

1’ N

, 10°

04’

E)

HV

GLA

hI

Wp

Sier

ra d

e Le

ma

Fore

st, C

anai

ma

Nat

iona

l Par

k, V

enez

uela

(05°

53’

N, 6

1° 2

6’ W

)

HV

GLA

hI

Wp

Río

Cón

dor,

Tier

ra d

el F

uego

, Chi

le (5

3° 5

9’ S

, 69°

58’

W)

HV

GLA

hI

Wp


103

Table 4.8: Linear regression coefficients for the comparison of the cosine-corrected

direct solar radiation transmittances associated with the two types of canopy condition


hemIMAGE (hI) and Winphot (Wp). The two canopy conditions are canopy gaps and

closed canopies. R2 is the coefficient of determination; RMSE is the root mean square

error. * indicates significance at the 5% level, and ** at the 1% level.

Condition Variable Intercept Slope R2 RMSE p-Value n

HV vs GLA 0.004** 0.930** 0.998 0.005 0.000** 39

HV vs hI 0.001 0.973** 0.997 0.005 0.000** 39

HV vs Wp -0.003 1.067** 0.994 0.010 0.000** 39

GLA vs hI -0.003 1.044** 0.997 0.006 0.000** 39

GLA vs Wp -0.007* 1.145** 0.992 0.011 0.000** 39

Canopy

Gap

hI vs Wp -0.003 1.096** 0.994 0.009 0.000** 39

HV vs GLA 0.003 0.959** 0.987 0.005 0.000** 39

HV vs hI 0.003 0.952** 0.938 0.011 0.000** 39

HV vs Wp -0.002 1.043 0.942 0.012 0.000** 39

GLA vs hI 0.002 0.981** 0.928 0.012 0.000** 39

GLA vs Wp -0.002 1.068** 0.920 0.013 0.000** 39

Closed

Canopy

hI vs Wp 0.002 1.029** 0.886 0.016 0.000** 39


104


diffuse solar radiation transmittances associated with the two types of canopy condition






HV vs GLA -0.001 0.993** 0.998 0.002 0.000** 39

HV vs hI 0.000 0.997** 1.000 0.001 0.000** 39

HV vs Wp 0.001 1.072** 0.994 0.003 0.000** 39

GLA vs hI 0.001 1.002** 0.998 0.002 0.000** 39

GLA vs Wp 0.002 1.077 0.992 0.004 0.000** 39

Canopy

Gap

hI vs Wp 0.001 1.075 0.994 0.003 0.000** 39

HV vs GLA 0.000 0.996** 0.999 0.001 0.000** 39

HV vs hI 0.000 0.999** 1.000 0.001 0.000** 39

HV vs Wp -0.001 1.047** 0.997 0.002 0.000** 39

GLA vs hI 0.000 1.003** 0.999 0.001 0.000** 39

GLA vs Wp 0.000 1.050** 0.996 0.002 0.000** 39

Closed

Canopy

hI vs Wp 0.000 1.047** 0.997 0.002 0.000** 39


105


global solar radiation transmittances associated with the two types of canopy condition






HV vs GLA -0.007* 0.987** 0.989 0.006 0.000** 39

HV vs hI -0.009** 1.010** 0.990 0.006 0.000** 39

HV vs Wp -0.052** 1.333** 0.952 0.019 0.000** 39

GLA vs hI -0.002 1.020** 0.995 0.005 0.000** 39

GLA vs Wp -0.044** 1.358** 0.973 0.014 0.000** 39

Canopy

Gap

hI vs Wp -0.041 1.330** 0.975 0.014 0.000** 39

HV vs GLA -0.003 1.013** 0.981 0.005 0.000** 39

HV vs hI -0.002 1.012** 0.955 0.008 0.000** 39

HV vs Wp -0.005 1.101** 0.921 0.011 0.000** 39

GLA vs hI 0.002 0.990** 0.955 0.008 0.000** 39

GLA vs Wp -0.002 1.090 0.944 0.010 0.000** 39

Closed

Canopyt

hI vs Wp 0.000 1.068** 0.929 0.011 0.000** 39


106

(a) (b)

(c)

Figure 4.2: Scatter plots of the relationships between (a) the cosine-corrected direct

solar radiation transmittances (DIR) beneath canopy gaps estimated using HemiView

and Gap Light Analyzer, (b) the cosine-corrected diffuse solar radiation transmittances

(DIF) beneath canopy gaps estimated using Gap Light Analyzer and Winphot and c) the

cosine-corrected global solar radiation transmittances (GLO) under closed canopies

estimated using hemIMAGE and Winphot. The solid lines correspond to the equations

in Tables 4.8 (Fig. a), 4.9 (Fig. b) and 4.10 (Fig. c). The broken lines represent a 1:1

reference.


107

Table 4.11: Linear regression coefficients for the comparison of the non-cosine-

corrected direct, diffuse and global solar radiation transmittances associated with the

two types of canopy condition estimated using the programmes HemiView (HV) and

hemIMAGE (hI). The two canopy conditions are canopy gaps and closed canopies. R2

is the coefficient of determination; RMSE is the root mean square error. * indicates

significance at the 5% level, and ** at the 1% level.


Non-cosine-corrected direct solar radiation transmittances

Canopy Gap HV vs hI -0.001 0.968** 0.996 0.006 0.000** 39

Closed Canopy HV vs hI 0.007 0.895** 0.922 0.010 0.000** 39

Non-cosine-corrected diffuse solar radiation transmittances

Canopy Gap HV vs hI 0.000 0.989** 1.000 0.000 0.000** 39


Non-cosine-corrected global solar radiation transmittances

Canopy Gap HV vs hI -0.015** 1.139** 0.967 0.009 0.000** 39


The linear regression analysis of all of the non-cosine-corrected solar radiation

transmittance values obtained for both the gaps and the closed canopy conditions also

demonstrated strong relationships between the estimates produced by the two

programmes (n = 39; R2 > 0.922; p < 0.01 for all). This meant that both computed very

similar results with respect to the transmittance of solar radiation irrespective of canopy

condition (Table 4.11 and Figure 4.3).


108

(a) (b)

(c)

Figure 4.3: Scatter plots of the relationships between (a) the non-cosine-corrected direct

solar radiation transmittances (DIR) beneath canopy gaps estimated using HemiView

and hemIMAGE, (b) the non-cosine-corrected diffuse solar radiation transmittances

(DIF) under closed canopies estimated using HemiView and hemIMAGE and c) the

non-cosine-corrected global solar radiation transmittances (GLO) under closed canopies

estimated using HemiView and hemIMAGE. The solid lines correspond to the

equations in Table 4.11. The broken lines represent a 1:1 reference.


109

4.5 Discussion

4.5.1 Canopy Structure and Solar Radiation Transmittance

All of the comparisons of both variable types, namely canopy structure and

below-canopy solar radiation environment, calculated using the different programmes

revealed a high level of agreement, with no statistically discernable differences between

the results produced. Irrespective of canopy condition, the results produced by all four

software packages proved to be very similar.

The outputs of the four programmes almost always revealed a strong positive

relationship (R2 > 0.859) with respect to the evaluation of the two contrasting canopy

conditions (gap and closed canopy). This coincided with the findings of Frazer et al.

(1997), who compared the canopy architecture (percent open sky and effective plant

area index) of different forest chronosequences in British Columbia using Hemiphot and

PAMAP GIS. However, a comparison of the effective plant area index derived for both

canopy gaps and under closed canopy conditions, calculated by HemiView and

Winphot on the basis of ellipsoidal leaf angle distributions (Le-E), revealed only weak

correlations. This may have been linked to the plant area index calculations, which are

quite sensitive to small changes in cover. Therefore, in environments with dense cover,

the computed plant area indexes might underestimate the actual situation (ter Steege

1996). However, the hemispherical photographs input into each programme had the

same format and resolution. The contrasting findings may have been due to different

gap-fraction inversion procedures or mathematical algorithms used by the programmes,

resulting in the calculation of different values. This can be observed in the different

parameters featured in the equations incorporated into Winphot and HemiView (Eqs.

4.4 and 4.5, respectively). These calculate the extinction coefficient given in Eq. 4.3 and

finally, by successive iterations, until the best effective plant area index (Le) is obtained

from Eq. 4.1. Furthermore, the mean leaf angle estimated in both conditions, under both

canopy gaps and closed canopy, showed variations. The weaker relationship was found

for the canopy gaps (R2 = 0.332), which indicates a high variability between the

programmes (Figure 4.4). The mean leaf angle variable can show the distribution of the

ellipsoidal leaf angle distribution parameter (ELADP) in Eqs. 4.3, 4.4, 4.5 and 4.6. Its

values are near to 0.0 when all canopy elements are vertical (90°). This was


110

(b)

n = 39; R2 = 0.612; RMSE = 5.735; p < 0.001

(a)

n = 39; R2 = 0.332; RMSE = 7.008; p < 0.001

predominantly the case in the canopy gaps, as estimated by HemiView, and contrasted

with the results obtained from the same hemispherical photographs with Winphot.

Values near 1.0 show that the canopy has a spherical distribution, and values of ELADP

→ ∞ when the canopy elements are more horizontal (0°).

Figure 4.4: Linear regression analysis of the mean leaf angle (°) (MLA) estimated by

HemiView and Winphot beneath a) canopy gaps and b) closed canopies. The solid lines

correspond to the equations a) MLA (Winphot) = 23.357 + 0.645 MLA (HemiView),

and b) MLA (Winphot) = 8.490 + 0.800 MLA (HemiView). The broken lines represent

a 1:1 reference, where the mean leaf angle (°) estimated by HemiView would be equal

to that estimated by Winphot.

The contrasting results may also have derived from the assumption of a random

distribution of canopy elements, leading to either an under- or an overestimation in the

calculation of the leaf area indexes. This possibility has been described for conifer


111

forests (Rich et al. 1999, Walter et al. 2003) and for forests with discontinuous canopies

(Walter et al. 2003). The hemispherical photographs made under the gaps fall into the

latter category. This can also be observed in Figure 4.4, where the estimated values of

mean leaf angle under closed canopy conditions were more closely related (R2 = 0.612).

However, both Gap Light Analyzer and Winphot also calculate the effective Le-75 on

the basis of the assumption that canopy elements are randomly distributed. Yet, in this

case (Table 4.6), the resulting values were strongly positively correlated (R2 = 0.859). It

would appear, therefore, that the contrasting results were not linked to the assumption of

a random distribution of canopy elements.

Although the comparison of the software packages resulted in similar

quantitative differences in the calculation of plant area indexes, these were all indirect

estimates. A bias in all of these measurements is, therefore, possible (Coops et al. 2004).

Additional data relating to, for example, foliage clumping, the shading effects of

branches and boles, slope corrections, etc., must be incorporated in order to improve

estimates. This is required because accurate estimates of plant area indices are necessary

in studies of forest ecology (Frazer et al. 2000, Walter and Torquebiau 2000, Walter et

al. 2003). Coops et al. (2004) stated that a more accurate estimate of the effective plant

area index can be achieved by comparing the estimates with actual leaf quantity

measurements, or other direct leaf area index measurements (refer to Jonckheere et al.

2004). The need for the validation of indirect methods remains. LAI measurements

may, therefore, prove important as a method of calibration (Jonckheere et al. 2004), and

calibration of the estimates of plant area index from hemispherical photographs might

be made with those derived from allometric models.

Through the regression analysis, strong relationships were found when all of the

outputs of solar radiation transmittances were compared (Tables 4.8, 4.9, 4.10 and 4.11,

and Figures 4.2 and 4.3), indicating that all of the programmes provided accurate

estimates of the below-canopy solar radiation. Ultimately, slight differences were

observed between the programmes.

Finally, and in spite of the fact that the length of the vegetation period in

Winphot was set to only 7 days for both the German and the Chilean forest, the results

computed did not differ from those calculated on the basis of the complete length of the

respective vegetation periods using the other programmes.


112

4.5.2 Other aspects requiring consideration when selecting a programme

All four of the software packages evaluated (HemiView, Gap Light Analyzer,

hemIMAGE and Winphot) are Windows-based programmes. The minimum system

requirements of each are Microsoft Windows 3.1 or later (Winphot), Microsoft

Windows 95 or later (HemiView and Gap Light Analyzer), and Windows NT 4.0 or

later (HemiView and Gap Light Analyzer). Other considerations in relation to the

software system requirements are listed in Table 4.12.

Table 4.12: Programme system requirements. A ‘-’ means that no indication of the

programme requirements has been given.

Programme Minimum operating

system

Minimum

RAM

Minimum

hard disk

Minimum video

display

HemiView Microsoft Windows

NT 4.0 and 95 16 Mbytes 10 Mbytes 16 colour VGA

Gap Light

Analyzer

Microsoft Windows

NT 4.0 and 95 64 Mbytes -

4 MB of 600 x

800 true-colour

hemIMAGE - - - -

Winphot Microsoft Windows

3.1 - - -

source: ter Steege 1996, Frazer et al. 1999, Rich et al. 1999, Brunner 2002.

One of the difficulties encountered using the programmes concerned the image

file formats supported. hemIMAGE requires GIF files saved in the grey scales format

(Brunner 2002), Winphot accommodates both BMP and PCX (ter Steege 1996),

whereas Gap Light Analyzer supports most common graphics formats, with the

exception of compressed TIFF, GIF and newer formats like FlashPix (Frazer et al.

1999). The commercial product, HemiView, supports the following image formats:

BMP, JPG, PCX, TIFF, TARGA and PCD (Rich et al. 1999). The latter two

programmes can also support colour photographs. However, not all of the

aforementioned file formats are produced by digital cameras. Graphics editing software

such as Adobe Photoshop or the free image manipulation programme ‘GIMP’ is,

therefore, necessary to convert the digital hemispherical photographs to the supported


113

file formats. This can potentially increase the time needed for the analysis of

hemispherical photographs.

The hemIMAGE software requires that all images are quadratic in shape

(Brunner 2002), so that the programme can calculate the boundaries of the circular

fisheye image and the zenith position. This, too, means that a graphics editing software

is required. Winphot also requires a graphics editor, for the conversion of the images to

be analysed to the required formats (refer to ter Steege 1996). Incorporated within

Winphot, Gap Light Analyzer and HemiView are registration or alienation processes to

determine correctly the horizon on each photograph (Frazer et al. 1999, Rich et al. 1999,

ter Steege 1996). Therefore, in the case of these three programmes, the cropping

procedure is not necessary. However, it was not tested whether these alienation

processes rendered the image preparation less subjective, and easier, or whether they

improved the accuracy of the subsequent analysis of the images.

Another factor often highlighted as a source of error associated with

hemispherical photography relates to the classification and distinction of visible sky

from obscured sky, commonly referred to as threshold setting. The programmes Gap

Light Analyzer, Winphot and HemiView have an inbuilt interactive threshold setting

tool, which can be applied to the complete photo, or to a segment thereof (Frazer et al.

1999, Rich et al. 1999). However, there are also problems associated with this

technique, caused by unevenness in the light conditions within hemispherical

photographs (Rich et al. 1999). Manual threshold setting procedures have also been

criticised for their subjectivity, with human interpretation considered a major source of

error (Inoue et al. 2004a, Ishida 2004, Nobis and Hunziker 2005, A. Brunner personal

communication). Automatic thresholding methods have been developed in recent times

(Ishida, 2004, Nobis and Hunziker 2005, Schwalbe et al. 2006), which might eliminate

this source of error. This may negate the need for the use of graphics editing software

and interactive threshold setting tools.

A correction for lens distortion is integrated into the hemIMAGE, Gap Light

Analyzer and HemiView software. The latter two programmes allow the user to input a

new, user-defined lens distortion (Frazer et al. 1999, Rich et al. 1999). In hemIMAGE

any new calibrations must be entered into the programme code by the programmer

(Brunner 2002). Winphot, alternatively, assumes a polar or equiangular projection and

lens options cannot be set (H. ter Steege personal communication). In this context, all


114

lens calibrations used are very close to the polar or equiangular projection, and

differences in the results produced by this setting procedure should not be expected.

The four programmes evaluated allow the user to store different file variables.

Different configurations for site (latitude, longitude, altitude, time zone, magnetic

correction for true north, slope, aspect), the time period (a specific day, growing season,

year, etc.) and for the radiation models (universal overcast model, standard overcast

model, percentage of both diffuse and direct radiation, etc.) can be saved. These stored

settings can also be used later for further analysis of the images (ter Steege 1996, Frazer

et al. 1999, Rich et al. 1999, Brunner 2002).

All of the results generated by the three free programmes are saved as text files

with different formats, all of which can be opened in a spreadsheet application like

Microsoft Excel for further analysis (ter Steege 1996, Frazer et al. 1999, Brunner 2002).

The spreadsheets generated by the commercial software HemiView can be saved

directly as MS Excel 5.0 files (Rich et al. 1999).

All four programmes were generally user friendly, but in each case it was

necessary to understand the functions, tools and programme-specific characteristics to

begin analysing the hemispherical photographs. Although each is accompanied by a

user manual (ter Steege 1996, Frazer et al. 1999, Rich et al. 1999, Brunner 2002), not all

aspects are covered in the manuals. Brunner (2002), for example, assumed that the

readers and users are familiar with the basic steps involved in the analysis of

hemispherical photographs. Each manual includes a list of references or a bibliography,

allowing readers and users to consult source literature directly.

4.6 Conclusions

It was possible to demonstrate that the characterisation of forest environment

using digital hemispherical photographs, evaluated using a commercial software

package (HemiView) and free software packages (Gap Light Analyzer, hemIMAGE

and Winphot), resulted in similar estimates of the most commonly used canopy

structure and solar radiation variables. Thus, irrespective of canopy condition and

latitude, the results produced by all four software packages proved to be very similar.

However, accurate estimates of forest canopy structure and below-canopy solar

radiation environments are needed, and in the case of some variables the outputs of the


115

programmes deviated, especially the outliers. This may lead to greater bias in the

analysis of vegetation patterns in the forest. The calculation of the effective plant area

index should be viewed with caution, as the comparison of the computed outputs

revealed varying estimates.

The outline of the contrasting methods employed for the analysis of

hemispherical photographs provided here might be used to develop a standard protocol

for the evaluation of hemispherical photographs made in broadleaf forests, because the

outputs of all four programmes were very similar.

Chapter 5 Below-canopy solar radiation transmittances

116

CHAPTER 5

EFFECTS OF CANOPY STRUCTURE AND STAND

PARAMETERS ON THE VARIABILITY OF SOLAR RADIATION

TRANSMITTANCE IN AN UNEVEN-AGED EVERGREEN

Nothofagus betuloides FOREST

5.1 Abstract

The transmission of the direct, diffuse and global solar radiation penetrating the

canopy of an uneven-aged, evergreen Nothofagus betuloides forest during the growing

season (October to March), and the effects of the canopy structure and stand parameters

on the below-canopy solar radiation regime, were analysed. Hemispherical photographs

were used to estimate the solar radiation transmittances and certain of the canopy

structure variables, namely gap fraction, canopy openness and plant area index. The

transmission of the solar radiation into the forest was not only affected by a high level

of horizontal and vertical heterogeneity of the forest canopy, but also by the low angles

of the sun’s path. The below-canopy direct solar radiation appeared to be variable in

space and time, whereas the transmission of the diffuse radiation exhibited lower

variability. The direct solar radiation correlated poorly with the canopy structure and

stand parameters. The variable that best fit the data was the plant area index (R2 =

0.263). Although the correlation with the simple stand parameters was poor, the diffuse

and global solar radiation were very sensitive to canopy openness (R2 = 0.963 and

0.833). However, the degree of heterogeneity of the forest – measured only on the basis

of canopy structure and stand parameters – was high, which had a corresponding effect

on the diffuse and global below-canopy solar radiation. As much as 75 % of the

variation in the diffuse and 73 % of the global solar radiation were explained by a

combination of the basal area, crown projection, crown volume, stem volume and the

average equivalent crown radius.

Keywords: Nothofagus betuloides, uneven-aged forest, hemispherical photographs,

below-canopy solar radiation, transmittance.


117

Nomenclature

BA basal area (m2 ha-1) GC centre of the gap

CA crown area projected to

the ground (m2)

GF gap fraction (%)

CA averaged crown area

projected to the ground

per plot (m2)

GLO growing season global solar radiation

transmittance (%)

CL crown length (m) Le-

60

plant area index calculated using the

meant tilt angle of the foliage integrated

over zenith angles from 0 to 60º (m2 m-2)

CO canopy openness (%) Le-

75

same as above but integrated over zenith

angles from 0 to 75º (m2 m-2)

Cr averaged equivalent

crown radius per plot (m)

Le-E plant area index employing an ellipsoidal

angle distribution (m2 m-2)

CSA crown surface area (m2) NGE northwestern gap edge

CV crown volume (m3) NUC northwestern undisturbed canopy

D stocking (nº trees ha-1) R equivalent crown radius (m)

DBH diameter at breast height

(cm)

SGE southeastern gap edge

DIF growing season diffuse

solar radiation

transmittance (%)

SUC southeastern undisturbed canopy

DIR growing season direct

solar radiation

transmittance (%)

SV stem volume (m3 ha-1)

5.2 Introduction

Biological and meteorological processes are influenced by solar radiation

(Anderson 1964c, Barnes et al. 1998). Both above and within forests, chemical, physical

and physiological processes are driven by the components of solar radiation (Holst and

Mayer 2005). A knowledge of the effects of solar radiation in the forest understorey is


118

important for an understanding forest dynamics, as solar radiation affects plant

regeneration patterns such as germination, establishment, growth and survival (Grant

1997).

The latitude, time of day, atmospheric clarity and altitude affect the total amount

of solar radiation reaching a forest canopy (Hutchison and Matt 1977, Barnes et al.

1998). The forest canopy is an exceedingly complex, three-dimensional structure that

changes with time (Anderson 1964b). Its irregular surface greatly influences the

reflection, transmission and the absorption of solar radiation (Grant 1997, Geiger et al.

2003). Even in tropical forests with no pronounced seasonal variation in stand structure,

the below-canopy changes in light throughout the year will differ from those occurring

in an open area (Anderson 1964b, Rich et al. 1993). Only a small percentage of the

incident radiation reaches the forest floor (Barnes et al. 1998). Therefore, measuring

solar radiation beneath a forest canopy is complicated by the highly irregular

distribution of the solar radiation in both space and time (Reifsnyder et al. 1971-1972,

Geiger et al. 2003).

The amount of direct solar radiation penetrating a forest canopy is highly

variable in space and time, as the sun’s position changes throughout the day, and from

day to day. This is exacerbated by the fact that the distribution of openings in the

canopy tends to be highly variable (Hutchison and Matt 1977, Barnes et al. 1998,

Geiger et al. 2003). The penetration of diffuse solar radiation by contrast is less

variable, as it depends upon the distribution and the level of sky brightness, the number,

size and spatial distribution of canopy openings, the canopy geometry, the spatial

distribution and optical characteristics of the forest biomass. Thus, within a forest,

diffuse solar radiation is more uniform in space and time than either direct or global

solar radiation (Hutchison and Matt 1977, Geiger et al. 2003).

Seasonal changes in the below-canopy solar radiation have been reported for

deciduous (Hutchison and Matt 1977, Baldocchi et al. 1984, Caldentey et al. 1999-2000,

Mayer et al. 2002, Holst and Mayer 2005, Holst et al. 2005) and coniferous forests

(Weiss 2000, Hardy et al. 2004).

Several attempts have been made to estimate the below-canopy solar radiation

from stand parameters (Comeau and Heineman 2003, Geiger et al. 2003). Individually

and in combination, the basal area, stocking density and height of trees, crown

dimensions, canopy cover, and the leaf area index have been correlated with the

availability of solar radiation in coniferous forests (Sampson and Smith 1993, Vales and


119

Bunnell 1988, Hale 2003, Sonohat et al. 2004) and in broadleaved forests (Comeau

2001, Pinno et al. 2001, Comeau and Heineman 2003, Piboule et al. 2005, Comeau et

al. 2006). However, the availability of solar radiation in the understorey of a

homogeneous, even-aged forest may be drastically different to that of a heterogeneous,

uneven-aged forest with the same leaf area (van Pelt and Franklin 2000, Piboule et al.

2005). The canopy of a heterogeneous, uneven-aged forest is characterised by a multi-

layered vertical distribution and an irregular horizontal distribution of structures, due to

the presence of canopy openings of different sizes (Franklin and van Pelt 2004), as has

been observed in old-growth Nothofagus betuloides (Mirb.) Oerst forest in Tierra del

Fuego (Rebertus and Veblen 1993, see Chapter 3).

The evergreen N. betuloides is one of the endemic forest species characteristic of

the Chilean and Argentinean sub-Antarctic forest. It grows from 40 ° 31’ S to 55 ° 31 ’

S (Rodríguez and Quezada 2003), and from sea level near the southern limit of its

distribution to altitudes of 1200 m a.s.l. near the northern limit of its distribution

(Veblen et al. 1996). N. betuloides exhibits a high ecological amplitude. It can

germinate in very shady environments (Veblen et al. 1996). There are indications that

for all South American Nothofagus species seedling establishment occurs best under

moderately high light levels (Table 5.1). However, N. betuloides appears to be more

shade-tolerant than other Nothofagus species (Veblen et al. 1996, see Chapter 3).

There are as yet numerous gaps in the knowledge of the effects of canopy

structure and stand parameters on the spatial variation of the below-canopy solar

radiation regime in an uneven-aged N. betuloides forest.

This study was carried out to (i) analyse the effects of the forest canopy on the

transmittance of solar radiation to the forest floor, and to (ii) evaluate whether canopy

structures and stand parameters explain the variation in the below-canopy solar

radiation in an uneven-aged N. betuloides forest.


5.3.1 Study Area

The study area is located in an old-growth, uneven-aged evergreen N. betuloides

forest (20 ha, with a stocking density of 1362 trees ha-1, and a basal area of 105 m2 ha-1)


120

with no direct evidence of human impact. The forest appears to be closely related to the

evergreen N. betuloides forest described by Veblen et al. (1983). Its soil is covered by a

thick layer of partially decomposed organic matter. The species diversity of the shrub

layer is poor, but there is a very pronounced layer of abundant epiphytic ferns, mosses

and liverworts, particularly on the slowly decaying tree trunks that lie on the forest

floor. The forest is located at the ‘Estancia Olguita,’ on the southeastern side of the Río

Cóndor (53 ° 59 ’S, 69 ° 58 ’W), which flows along the southwestern Chilean side of

Tierra del Fuego (Figure 3.1).

Table 5.1: Below-canopy solar radiation in some South American Nothofagus forests.

Numbers in parentheses are standard errors.

Site Altitude

(m a.s.l.)

Forest

composition

% diffuse and

direct radiation

% direct

radiation Reference

850 14.0 (1.6) 17.6 (1.1) Veblen et al. (1980)San Pablo

39°33 ′ S

72°03′ W 810

N. dombeyi 17.3 (1.5) 19.5 (1.4) Veblen et al. (1980)

1320 N. pumilio 21.7 (1.9) 26.7 (1.5) Veblen (1979)

1110 N. pumilio

N. betuloides39.9 (7.2) 38.9 (1.8) Veblen (1979)

Antillanca

Valley

40°47′ S

72°12′ W 1060 N. alpina 31.3 (2.2) 29.6 (1.0) Veblen et al. (1980)

1040 N. betuloides

N. pumilio 28.2 (1.8) 27.6 (1.2) Veblen (1979)

1020 N. pumilio

N. betuloides30.7 (2.5) 27.1 (1.1) Veblen et al. (1979)

Antillanca

Valley

40°47′S

72°12′ W 1000 N. pumilio 35.4 (3.5) 38.9 (1.8) Veblen (1979)

The climate of the area is that of the northern antiboreal sub-zone, with a mean

air temperature of between 9.0-9.5 °C in the warmest month of the year, and

temperatures remaining above zero in the coldest month. The mean annual rainfall is

around 500-600 mm, but can reach up to 900 mm. The wind direction is commonly

west to southwest, with average speeds of between 14-22 km h-1 and maximum speeds

of above 100 km h-1 (Tuhkanen 1992).


121

N

Centre of the Gap (GC)

Southeastern Gap Edge

(SGE)

Northwestern Gap Edge

(NGE)

Southeastern Undisturbed

Canopy (SUC)

Northwestern Undisturbed

Canopy (NUC)

Canopy Gap

5.3.2 Measuring Below-Canopy Solar Radiation

Hemispherical photographs were used to estimate indirectly the solar radiation

transmittance in the forest. The photographs were taken during the summer of 2006

along a gradient of solar radiation conditions, for which transects running from areas

beneath the undisturbed canopy to the centre of canopy gaps were established.

Thirteen canopy gaps (> 20 m2) were selected in a forest in Tierra del Fuego

dominated by N. betuloides. Their sizes ranged from 21-92 m2, with an average of 47 ±

6.3 m2 (± SE). A ‘gap’ was defined as the horizontal projection of a canopy opening to

the forest floor (Runkle 1982), and was considered closed if the vegetation growing

below the opening in the canopy was more than 2-3 m in height (see Chapter 3). Only

canopy gaps occurring on podsolic, well-drained and shallow soils (< 50 cm) were

selected.

Five hemispherical photographs were taken along a southeasterly to

northwesterly transect in each canopy gap (Figure 5.1). One hemispherical photograph

was taken at the centre of the gap (GC), two at the edges of the gap (southeast, SGE,

and northwest, NGE, of GC), and two more below undisturbed canopy (southeast, SUC,

and northwest, NUC, of GC) at a distance of half the height of the highest tree in the

vicinity of the gap (26 m ± 0.9). This distance was measured from the bases of the trees

at the edge of the gap.

Figure 5.1: Layout of transect in a canopy gap where the hemispherical photographs

were taken. The circle shape of the canopy gap does not illustrate the reality, because of

all gaps analyzed had different shapes.


122

0

1

2

3

4

5

6

7

8

J F M A M J J A S O N DMonth

Clo

udin

ess

(okt

as)

Punta Arenas (Chile)

Río Grande (Argentina)

Ushuaia (Argentina)

All hemispherical photographs were taken with a Nikon Coolpix 990® digital

camera (Nikon Corporation, Tokyo, Japan) fitted with a Nikon FC-E8® fisheye

converter (Nikon Corporation, Tokyo, Japan). The camera was mounted on a tripod at a

height of approximately 1.3 m above the ground. The camera and lens were levelled

with the aid of a spirit level and oriented to the magnetic north.

The digital images were processed according to Brunner (2002). This included

the manual setting of a threshold value to separate canopy and sky elements into a

binary black and white image and converted to greyscale. All images were analysed

using HemiView (Delta-T Devices, Cambridge, UK). Using the Coolpix 900 option

(Hale and Edwards 2002), the lens distortion was corrected. The universal overcast sky

(UOC) model was selected to describe the intensity of the diffuse radiation (Steven and

Unsworth 1980). The model assumes all regions of the sky to be equally bright. A

relative proportion of direct and diffuse radiation equal to 0.5 was assumed, because in

the vegetation period (October to March) the mean monthly cloudiness of southern

Patagonia and Tierra del Fuego ranges between 5.3 and 6.1 oktas, depending on the

location of the meteorological station (Figure 5.2).

Figure 5.2: Average monthly cloudiness for southern Patagonia (from the ‘Jorge C.

Scythe’ meteorological station in Punta Arenas (53°08’S; 70° 53’W; 6 m a.s.l.)) and

Tierra del Fuego (from the Río Grande (53° 80’S; 67° 78’W; 9 m a.s.l.) and Río Olivia

in Ushuaia (54°82’S; 68°30’W) meteorological stations (source: Butorovic (2003, 2004,

2005), Santana (2006, 2007), SMN (2007)).

The hemispherical photographs were divided into 16 azimuth and 9 zenith

regions (144 sky regions in total) in order to determine whether the diffuse solar


123

radiation transmittance originates from different sky regions in areas beneath disturbed

and undisturbed canopies. The photos were cosine corrected, so as to allow for

comparison with meteorological data (Anderson 1964c, Rich 1990).

All calculations of direct (DIR), diffuse (DIF) and global (GLO) solar radiation

transmittances below the canopy were estimated during the growing season (between

October and March).

5.3.3 Canopy Structure and Stand Parameters Measurements

The gap fraction (GF), canopy openness (CO) and plant area index (Le) were

estimated from the hemispherical photographs. GF was defined as the vertically

projected canopy area per unit ground area. CO was deemed to be the relative amount of

visible sky weighted by the incident angle of cosine law from a point (Inoue et al.

2004a). Le was described as the sum of all elements blocking canopy-light (stems,

twigs, leaves), as hemispherical photographs do not distinguish between opaque objects

(stems) and photosynthetic tissue (Holst et al. 2004).

Le was estimated using HemiView and Gap Light Analyzer (Frazer et al. 1999).

Both programmes use methods based upon the determination of gap fraction in the

canopy and inversion procedures (Norman and Campbell 1989). HemiView

incorporates an inversion algorithm of canopy transmission employing an ellipsoidal

leaf angle distribution to estimate the plant area index (Le-E). This means that the leaf

angle distribution of a canopy can be represented by the distribution of the area on the

surface of an ellipsoid of revolution (Norman and Campbell 1989, Rich et al. 1999).

Gap Light Analyzer uses the mean tilt angle of the foliage to estimate Le. The mean tilt

angle is calculated by a polynomial derived for a uniform leaf azimuth distribution and a

constant leaf normal angle (Welles and Norman 1991, Frazer et al. 1999). This

technique is similar to that used by the LAI-2000 Plant Canopy Analyzer (LI-COR, NE,

USA). For this study, Le was estimated over two zenith angle ranges from 0-60° (Le-

60) and from 0-75° (Le-75) with Gap Light Analyzer.

Stand measurements were made in 26 fixed plots of 225 m2 (15 x 15 m) during

the summer of 2007. In each plot, the diameter at breast height (DBH) for all trees with

a DBH ≥ 5 cm was recorded, and the stocking density (D), the basal area (BA) and the

stem volume (SV) per hectare were estimated. The centres of the plots were located

where the hemispherical photographs were taken (thirteen at GC, and thirteen at NUC).


124

The crown projection to the ground of each tree was also visually estimated, delimited

by four radii in four cardinal directions (to the north, south, east and west). The edge of

the crown was delimited with the aid of a clinometer. For each crown quarter, the crown

area (CA) projected to the ground was calculated using the formula of an ellipse, and

then all four quarters were added. SV was calculated from a species-specific allometric

regression based on DBH and the dominant height of the trees in the plot (Promis et al.

2007).

To reconstruct the crown shape of each tree, the equivalent crown radius (R) was

estimated, defined as the radius of a circle whose area is equal to the CA projected to

the ground (Piboule et al. 2005). The crown surface area (CSA) and the crown volume

(CV) were then computed, assuming that the crown had a parabolic geometric shape:

(5.1)

(5.2)

where CL (m) was the crown length. CL was defined as the length between the tree top

and the base of its crown, excluding epicormic shoots.

A regression analysis was performed to find a relationship between CL and

DBH. From a sub-sample of 70 trees, the total height (between 14-31 m), CL (between

2-10 m) and DBH (between 19-89 cm) were measured. Although significant, the best

relationship was a power function, which showed a large variation (n = 70; R2 = 0.147;

RMSE = 0.295; p < 0.01), indicating a high variability of CL among the trees in this old-

growth forest.

(5.3)

where DBH is the diameter at breast height (cm).

( ) 31.52L

22

LSA RC4R

C6RπC −×+×

××

=

2CR

πC L2

V×

×=

0.296L DBH1.707C ×=


125

5.3.4 Statistical and Regression Analysis to Explain the Solar Radiation

Transmittance

The non-parametric Kruskal-Wallis H-test and a post-hoc Mann-Whitney U-test

were used to analyse the variation in the transmission of cosine-corrected solar radiation

between areas beneath disturbed and undisturbed canopy.

A regression analysis was performed to test the strength of the relationships

between the solar radiation transmittance, the canopy structure and the stand

parameters. The CURVEFIT algorithm in SPSS 15.0 for Windows (SPSS Inc.) was

used. Linear, logarithmic, inverse, quadratic, compound, power, Schumacher’s

equation, growth and exponential functions were tested. The independent variables

tested were DIR, DIF, GLO, ln(DIR), ln(DIF), ln(GLO). The independent variables

used were BA, CA, CA, CO, Cr, CSA, CV, D, GF, Le-60, Le-75, Le-E, and SV. For the

regression analysis, the goodness-of-fit was calculated using the coefficient of

determination (R2), the root mean square error (RMSE) and the significance of the p-

value (Sokal and Rohlf 2000). A total of 26 and 65 observations were used for the

regression analysis, depending on whether the explanatory variables were calculated

from the fixed plots used for the stand measurements (BA, CA, CA, Cr, CSA, CV, D, and

SV) or hemispherical photographs (CO, GF, Le-60, Le-75, Le-E).

The non-linear regression procedure of SPSS 15.0 employing a Levenberg-

Marquardt algorithm was also used for evaluating the relationship between the solar

radiation transmittance with a combination of canopy structure and stand parameters

obtained from the stand measurements.

5.4 Results

5.4.1 Transmission of Solar Radiation Into the Forest

Between 3.5 and 22.2 % of the above canopy direct solar radiation was

transmitted to the forest floor during the growing season. The highest amount of DIR

(10.6 % ± 3.2, average ± SD) was recorded at the SUC. However, the values did not

differ significantly (n = 13, p > 0.05) from the mean transmittances calculated at the GC

(9.1 % ± 4.4) and SGE (8.7 % ± 3.9). Although the DIR was higher at the SGE, the


126

contrast with the DIR calculated at NUC (8.1 % ± 3.3) was not significant. The lowest

transmittance was computed at the NGE (5.9 % ± 3.1). In this case the contrast with the

values obtained for all other locations was statistically significant (Table 5.2).

The mean daily values of DIR differed between all of the plots (Figure 5.3). The

daily maximum DIR occurred at noon at two locations beneath the disturbed canopy (at

GC and SGE) (Figures 5.3a, b). At the NGE the maximum DIR transmission was more

variable; occurring at noon in October, at 9:00 in December and at 11:00 in February

(Figure 5.3c). Beneath the undisturbed canopy, the greatest DIR value was almost

always observed in the afternoon (between 13:00 and 14:00). The exception was in

October, when the maximum at the NUC occurred at 9:00 (Figures 5.3d, e).

Table 5.2: Descriptive statistics of the cosine-corrected direct, diffuse and global solar

radiation transmittances at the centre of the gaps (GC), the southeastern gap edge

(SGE), the northwestern gap edge (NGE), and below the southeastern undisturbed

canopy (SUC) and the northwestern undisturbed canopy (NUC). The standard deviation

is indicated in parentheses. Locations denoted by the same letter (a, b, c) did not differ

significantly (Kruskal-Wallis H-test and a post-hoc Mann-Whitney U-test, p > 0.05, n =

13).

Cosine-corrected solar radiation transmittances GC SGE NGE SUC NUC

Mean 9.1 ab 8.7 ab 5.9 c 10.6 a 8.1 b

(SD) (4.4) (3.9) (3.1) (3.2) (3.3) Direct

Range 4.9-22.2 4.2-18.5 3.5-13.3 6.0-15.7 4.1-17.3

Mean 13.9 a 12.5 ab 11.1 b 9.0 c 8.5 c

(SD) (4.8) (2.2) (2.7) (1.6) (2.1) Diffuse

Range 8.9-28.0 8.7-17.3 5.0-17.5 5.4-11.5 4.4-11.8

Mean 12.1 a 11.1 ab 9.2 bc 9.6 abc 8.3 c

(SD) (4.5) (2.7) (2.5) (1.7) (1.9) Global

Range 8.3-25.8 7.2-17.8 4.5-13.5 6.6-11.9 4.7-11.8

The DIF above the forest floor ranged from 4.4 to 28.0 % (Table 5.2). The

highest value (13.9 % ± 4.8) was calculated at the GC. This value did not differ

significantly (n = 13, p > 0.05) from the mean value calculated at the SGE (12.5 % ±


127

2.2). Although the DIF was higher at the SGE than at the NGE (11.1 % ± 2.7), the

difference was not significant. The lowest DIF values were calculated at the SUC and

NUC (9.0 % ± 1.6; 8.5 % ± 2.1).

Figure 5.3: Mean daily direct solar radiation transmittance for three months (October,

December and February) at (a) the centre of the gaps (GC), (b) the southeastern gap

edge (SGE), (c) the northwestern gap edge (NGE), and below the (d) southeastern

(SUC) and (e) northwestern (NUC) undisturbed canopy. LT means local time.

(a)

0

2

4

6

8

10

12

14

16

18

20

22

0 2 4 6 8 10 12 14 16 18 20 22 24time (LT)

Dire

ct s

olar

radi

atio

n tra

nsm

ittan

ce (%

)

OctoberDecemberFebruary

(b)

0

2

4

6

8

10

12

14

16

18

20

22

0 2 4 6 8 10 12 14 16 18 20 22 24time (LT)

Dire

ct s

olar

radi

atio

n tra

nsm

ittan

ce (%

)


(c)

0

2

4

6

8

10

12

14

16

18

20

22

0 2 4 6 8 10 12 14 16 18 20 22 24time (LT)

Dire

ct s

olar

radi

atio

n tra

nsm

ittan

ce (%

)


(d)

0

2

4

6

8

10

12

14

16

18

20

22

0 2 4 6 8 10 12 14 16 18 20 22 24time (LT)

Dire

ct s

olar

radi

atio

n tra

nsm

ittan

ce (%

)


(e)

0

2

4

6

8

10

12

14

16

18

20

22

0 2 4 6 8 10 12 14 16 18 20 22 24time (LT)

Dire

ct s

olar

radi

atio

n tra

nsm

ittan

ce (%

)



128

The high proportion of DIF beneath the gaps originated from 10-30° of the

zenith angles, with average values of 43 % at the GC, 46 % at the SGE and 44 % at the

NGE. Beneath the undisturbed canopy, the largest DIF values were observed within 30-

50° of the zenith; 47 % at the SUC and 43 % at the NUC (Figure 5.4)

Figure 5.4: Average distribution of diffuse solar radiation transmittance (DIF) at

different locations in the forest; at the centre of the gaps (GC), the southeastern gap

edge (SGE), the northwestern gap edge (NGE), and below the southeastern undisturbed

canopy (SUC), and the northwestern undisturbed canopy (NUC). Estimated from 10°

bands of zenith angle of the sky hemisphere.

The below-canopy GLO ranged from 4.5-25.8 % of the above-canopy GLO

(Table 4.2). There was no significant difference (n=13, p>0.05) between the mean GLO

at the GC (12.1 % ± 4.5) and at the SGE (11.1 % ± 2.7). By contrast, there was a

statistically significant difference between the GLO at the GC and the below-canopy

GLO at the NGE (9.2 % ± 2.5). However, the GLO values recorded at the two edges of

the gaps did not differ. The lowest estimates of GLO were found at the SUC and the

NUC (9.6 % ± 1.7, and 8.3 % ± 1.9).

5.4.2 Relationships between Solar Radiation Transmittances, Canopy Structure

and Stand Parameters

Significant correlations between below-canopy DIR, DIF and GLO and the

canopy structure and stand parameters were found (Tables 5.3, 5.4 and 5.5).

0

5

10

15

20

25

30

5 15 25 35 45 55 65 75 85Angle from zenith

Diff

use

sola

r rad

iatio

n tra

nsm

ittan

ce (%

)GC

SGE

NGE

SUC

NUC


129

Models fitted through regression analysis (Table 5.3, Figure 5.5a) suggest that

Le-E, although the coefficient of determination was low (R2 = 0.263), was the variable

that best explained the largest proportion of the variance of the below-canopy DIR.

Other variables explaining more than 20 % of the variation of the DIR were CO (R2 =

0.221) and Le-60 (R2 = 0.219).

Table 5.3: Regression models fitted for the estimation of the direct solar radiation

transmitted into the Nothofagus betuloides forest. b0, b1, b2 are the coefficients for the

models. R2 is the coefficient of determination; RMSE is the root mean square error. *

indicates significance at the 5 % level, and ** at the 1 % level.

Model b0 b1 b2 R2 RMSE p-value N

Ln (y) = exp (b0 + b1 Le-E) 1.059 -0.096 - 0.263 0.187 0.000** 65

Ln (y) = b0 + b1 Ln (Le-60) +

b2 Ln (Le-60)2 -0.864 6.661 -3.534 0.219 0.385 0.001** 65

Ln (y) = exp (b0 + b1 Le-75) 1.484 -0.256 - 0.159 0.200 0.001** 65

y = b0 + b1 CO + b2 CO2 9.289 -0.817 0.091 0.221 3.420 0.000** 65

y = b0 + b1 / Ln (Cr) 6.361 0.981 - 0.175 3.501 0.030* 26

The best models fitted for the estimation of DIF (Table 4.4 and Figure 4.5b)

used CO (R2 = 0.963). The significant variables explaining more than 20 % of the

variation of the DIF were GF (R2 = 0.839), Le-60 (R2 = 0.479), Le-75 (R2 = 0.396), CA

(R2 = 0.266), Cr (R2 = 0.265), BA (R2 = 0.256), CV (R2 = 0.245), and SV (R2 = 0.215).

For the prediction of below-canopy GLO, CO (R2 = 0.833) fitted the best model

(Table 5.5 and Figure 5.5). The other variables describing the canopy structure and

crown architecture that explained more than 20 % of the variation were GF (R2 =

0.638), Le-60 (R2 = 0.536), Le-75 (R2 = 0.434), and Cr (R2 = 0.296).


130

Table 5.4: Regression models fitted for the estimation of the diffuse solar radiation





y = b0 + b1 Ln (Le-60) + b2

Ln (Le-60)2 82.664 -106.063 37.870 0.479 2.646 0.000** 65

Ln (y) = exp (b0 + b1 Le-75) 1.660 -0.262 - 0.396 0.110 0.000** 65

y = b0 + b1 CO + b2 CO2 -0.247 1.499 0.012 0.963 0.707 0.000** 65

y = b0 + b1 GF + b2 GF2 4.788 0.355 0.009 0.839 1.469 0.000** 65

Ln (y) = b0 + b1 Ln (BA) 4.491 -0.482 - 0.256 0.340 0.007** 65

Ln (y) = b0 + b1 Ln (CA) 5.153 -0.484 - 0.266 0.338 0.006** 26

Ln (y) = exp (b0 + b1 / Ln

(CSA)) -0.864 10.900 - 0.183 0.157 0.026* 26

Ln (y) = b0 + b1 CV + b2 CV2 3.008 -0.002 0.000 0.245 0.350 0.035* 26

y = b0 + b1 / SV 6.790 3,581.576 - 0.215 4.242 0.015* 26

y = b0 + b1 / Ln (Cr) 7.835 1.496 - 0.265 4.106 0.006** 26

y = b0 + b1 / Ln (CA) 2.246 22.527 - 0.146 4.425 0.049* 26

Table 5.5: Regression models fitted for the estimation of the global solar radiation





y = b0 + b1 / Ln (Le-E) 4.784 6.579 - 0.137 2.889 0.002** 65

Ln (y) = exp (b0 + b1 Le-60) 1.555 -0.230 - 0.536 0.087 0.000** 65

Ln (y) = exp (b0 + b1 Le-75) 1.579 -0.247 - 0.434 0.096 0.000** 65

y = b0 + b1 CO + b2 CO2 3.362 0.623 0.042 0.833 1.281 0.000** 65

y = b0 + b1 GF + b2 GF2 8.024 -0.107 0.018 0.638 1.885 0.000** 65

Ln (y) = b0 + b1 / Ln (BA) 0.969 5.766 - 0.162 0.301 0.037* 26

Ln (y) = b0 + b1 / Ln (CA) 0.519 10.130 - 0.166 0.300 0.035* 26

y = b0 + b1 / SV 7.091 2,544.676 - 0.161 3.600 0.038* 26

y = b0 + b1 / Ln (Cr) 7.287 1.298 - 0.296 3.299 0.003** 26


131

5.5 Discussion

5.5.1 Solar Radiation Transmittances

A small but significant degree of spatial variation in the below-canopy solar

radiation was found in this study. The below-canopy solar radiation values were lower

than those estimated in forests of N. betuloides mixed with the deciduous Nothofagus

pumilio (Table 5.1). The differences may have been the result of the different forest

compositions, forest structures, times at which the solar radiation transmittances were

estimated (Gray et al. 2002) or the differing forest locations.

Canopy gaps alter the availability of many resources in the forest. The most

obvious changes are those to the below-canopy solar radiation (Canham et al.1990,

Gray et al. 2002). The differences in the below-canopy solar radiation occurring over

the course of the growing season observed beneath canopy gaps and beneath closed

canopies in this N. betuloides forest have also been described for other forest types

(Denslow 1980, Canham et al. 1990, de Freitas and Enright 1995, Gray et al. 2002).

Beneath the canopy gaps, the DIR showed the largest differences in the below-

canopy solar radiation. This corresponded with the findings of Gray et al. (2002). In

contrast with studies carried out in the northern hemisphere (Canham et al. 1990,

Bazzaz 1996, Gray et al. 2002), the southern edges of the canopy gaps evaluated in this

study received more solar radiation than the northern edges. Due to the changing angle

of solar elevation throughout the year (Geiger et al. 2003), the diurnal DIR patterns

within the canopy gaps were also reversed, as mentioned by Bazzaz (1996). On average,

the southeastern gap edge received the maximum DIR at noon over the growing season,

while it was between one and three hours later (in February and December,

respectively) at the northwestern gap edge. The amount of below-canopy DIR was also

affected by the distribution of canopy openings in the overstorey. The effects of the

canopy gaps on the below-canopy solar radiation extended to the areas below the

undisturbed canopy, as indicated by the fact that the highest average DIR was recorded

at the SUC (Canham et al. 1990, Gray et al. 2002).


132

Figure 5.5: Scatter plots of the best relationships between (a) the direct solar radiation

transmittances (DIR) during the growing season with the plant area index employing an

ellipsoidal leaf angle distribution (Le-E), (b) the diffuse solar radiation transmittances

(DIF) during the growing season with the canopy openness, and (c) the global solar

radiation transmittances (GLO) during the growing season with the canopy openness.

Lines are described by the equations in Table 5.3, 5.4. and 5.5.

(a)

1.0

1.5

2.0

2.5

3.0

3.5

2 3 4 5 6 7Le-E

Ln (D

IR)

(b)

0

5

10

15

20

25

30

2 4 6 8 10 12 14 16 18Canopy openness (%)

DIF

(%)

(c)

0

5

10

15

20

25

30

2 4 6 8 10 12 14 16 18Canopy openness (%)

GLO

(%)


133

The transmission of DIF exhibited less variability than DIR. The highest DIF

values were found in the canopy gaps rather than below undisturbed canopies. The

reason for this is that the penetration of diffuse light depends mainly on the sky

brightness and the canopy openings (Anderson 1964b, Reifsnyder 1971-1972,

Hutchison and Matt 1977). Thus, the contribution of DIF to the overall light climate

decreases with increasing distance from the centre of a canopy gap (Canham et al.

1990). Furthermore, the greater proportion of below-canopy DIF originated from two

bands of the sky, between 10-30º of the zenith angles in canopy gaps, and between 30-

50º of the zenith angles beneath undisturbed canopies. Similar patterns have been

observed in temperate and tropical forests (Canham et al. 1990).

The effects of canopy gaps on both the below-canopy GLO and on the forest

dynamics are not limited to the gaps themselves (van Pelt and Franklin 2000, Gray et al.

2002), as the largest differences in the below-canopy solar radiation transmittances

calculated were determined primarily by DIR (Hutchison and Matt 1977, Gray et al.

2002), which on average was higher at the SUC; that is, below the undisturbed canopy.

5.5.2 Influences of Canopy Structures and Stand Parameters on Solar Radiation

Transmittances

The results indicated that the directly transmitted component of the solar

radiation penetrating into the forest during the growing season was highly variable.

Although poorly correlated, five variables of canopy structure explained the variation,

with Le-E the independent variable exhibiting the best correlation. Stand parameters

were not correlated with DIR. This might be explained by the fact that the penetration of

DIR into the forest depends on the above-canopy DIR, the stand characteristics, and on

the number, size and spatial distribution of the canopy openings (Anderson 1964c,

Reifsnyder 1971-1972, Hutchison and Matt 1977). Moreover, at high latitudes in the

southern hemisphere, an occasional opening of the canopy low on the horizon increases

the DIR transmittance. The variability of the below-canopy DIR, therefore, is high.

The transmission of DIF was more uniform in space. Twelve variables

describing the canopy structure were correlated with DIF. The best equations fitted for

predicting and explaining the diffuse solar radiation transmitted into the forest during

the growing season used either CO or GF as independent variables. Battaglia et al.

(2002) found that this was also the case in a Pinus palustris forest. If one were to


134

assume that in this evergreen forest the CO and GF do not change during the growing

season, these variables would explain 96 % and 84 % of the variation in DIF,

respectively.

Independent stand parameter and canopy structure variables were poorly

correlated with DIF. Each factor alone (BA, CA, CSA, CV, SV, Cr, CA) explained

between 15 % and 27 % of the variation in DIF. Old, uneven-aged forests have a

complex structure and are spatially heterogeneous (Franklin and van Pelt 2004). In

single species stands, or within even-aged mixed stands with similar canopy

architecture, simple stand architecture parameters such as basal area, stocking density,

canopy cover, age, etc. have been used successfully in predicting below-canopy light

environments, especially the transmission of the diffuse component of the solar

radiation (Vales and Bunnell 1988, Comeau 2001, Hale 2001, Pinno et al. 2001,

Comeau and Heineman 2003, Hale 2003, Sonohat et al. 2004, Comeau et al. 2006).

When the aforementioned, readily measured canopy structure and stand

parameters are combined (BA, CA, CV, Cr and SV) they can explain 75 % of the

variation in DIF (Figure 5.6), without showing great differences to the observed values.

Figure 5.6: Comparion between the observed and the predicted diffuse solar radiation

transmittances (DIF) during the growing season. Predicted values were calculated using

the model DIF = 206.601 – 11.975 Ln (BA) – 13.980 Ln (CA) + 0.026 CV – 8.224E-06

CV2 + exp( -7.844 + 1.309 / Ln (Cr)) – 501.367 / Ln (SV) (n = 26; R2 = 0.749; RMSE =

2.751; p < 0.001). The solid line represents a 1:1 reference, where the predicted values

of the diffuse solar radiation transmittances during the growing season would be equal

to those predicted from the hemispherical photographs.

0

5

10

15

20

25

30

35

0 5 10 15 20 25 30 35Predicted DIF value (%)

Obs

erve

d D

IF v

alue

s (%

)


135

The variables that explained the variation in the below-canopy GLO best were

CO (83 %) and GF (64 %). Taken individually, other canopy architecture and stand

parameters (BA, CA, SV, Cr) explained between 16 % and 54 % of the variation in

GLO. However, the heterogeneity of the spatial distribution of structures, which is a

feature of old uneven-aged forests (Franklin and van Pelt 2004), also influences the

heterogeneity of the transmittance of global solar radiation in the understorey.

Therefore, when more than one of the canopy structure and stand parameter variables

measured in the forest were combined, a model explaining 73 % of the variation in the

global solar radiation could be developed (Figure 5.7).

Figure 5.7: Comparison between the observed and the predicted global solar radiation

transmittances (GLO) during the growing season. Predicted values were calculated

using the model GLO = -11.671 + 7.225 Ln (BA) + 122.952 Ln (CA) + exp( -10.788 +

1.656 / Ln (Cr)) – 1,310.796 / Ln (SV) (n = 26; R2 = 0.734; RMSE = 2.213; p < 0.001).

The solid line represents a 1:1 reference, where the predicted values of the global solar

radiation transmittances during the growing season would be equal to those predicted

from the hemispherical photographs.

5.6 Conclusions

The transmittance of solar radiation and its components onto the floor of a N.

betuloides forest is characterised by a high degree of spatial and temporal variability.

Although canopy gaps appear to increase the amount of solar radiation reaching the

0

5

10

15

20

25

30

35

0 5 10 15 20 25 30 35Predicted GLO value (%)

Obs

erve

d G

LO v

alue

s (%

)


136

forest floor, at this high southern latitude, a higher below-canopy DIR was found below

the undisturbed canopy to the southeast of the gap centre, at a distance from the gap

edge of half the height of the highest tree in the vicinity of the gap.

Canopy structure and stand parameters explain partly the variation in the below-

canopy solar radiation regime. The best predictors of the below-canopy DIF and GLO

are CO and GF, indicating a sensitivity to canopy openings in this forest. Readily

estimated canopy architecture and stand parameters are poorly correlated with the

transmittances of the solar radiation into the forest. However, by combining BA, CA,

CV, and SV a large amount of the variability of the below-canopy DIF and GLO can be

explained. This demonstrates a high degree of heterogeneity of the spatial structures of

this uneven-aged, evergreen forest, affecting the amount of solar radiation reaching the

forest floor.

Chapter 6 Regeneration patterns of a N. betuloides forest

137

CHAPTER 6

EFFECTS OF NATURAL SMALL-SCALE DISTURBANCES ON

BELOW-CANOPY SOLAR RADIATION AND REGENERATION

PATTERNS IN AN OLD-GROWTH Nothofagus betuloides FOREST

IN TIERRA DEL FUEGO, CHILE

6.1 Abstract

Over the last decade a shelterwood system promoting natural regeneration has

been applied in the southern Chilean old-growth Nothofagus forests. However, most of

these forests are naturally uneven-aged, and the effect of the application of this

silvicultural system has been the homogenisation and simplification of the stand

structure. An ecological understanding of natural disturbance processes is necessary to

improve the current silvicultural practices, and to develop new silvicultural systems in

southern Chile, in order to maintain the multi-layer structures and native biodiversity.

The research was aiming at analysing the effects of natural small-scale disturbances on

the below-canopy solar radiation conditions, the regeneration patterns (density and

growth rates), and the browsing damage to young trees caused by Lama guanicoe in an

old-growth Nothofagus betuloides forest. The study was carried out in 13 canopy gaps

(21-92 m2). The regeneration was sampled in 65 plots (4 m2) in and around the canopy

gaps along a solar radiation gradient. All N. betuloides seedlings and saplings were

counted in height classes, together with browsing effects caused by L. guanicoe. The

height growth and radial increment were measured for the tallest plant in each plot. The

results revealed that the availability of the non-cosine-corrected direct, diffuse and

global solar radiation transmitted into the forest ranged from 3.2 to 19.4 %, 3.1 to 16.7

% and 3.2 to17.6 %, respectively. The seedlings and saplings of N. betuloides exhibited

an high shade tolerance, apparently not requiring the presence of large gaps to establish.

This resulted in a more continuous process of forest regeneration. The solar radiation

transmittances did not correlate with either the relative radial growth or the relative

height increment of the young trees. The growth of the young trees correlated only with

the age of the plants, and inverse polynomial curves explained 70 % of the variance in

relative radial growth and 50 % of the variance in relative height increment. The


138

proportions of seedlings browsed by L. guanicoe were low (0.7 to 2.8 % of the total).

Browsing damage to young trees was observed below canopy gaps as well as beneath

closed canopies, demonstrating no particular habitat preference. The heterogeneous

canopy of the old-growth N. betuloides forest with only very small canopy gaps

produced a variety of mosaics in the understorey, with seedlings and saplings present in

a range of all ages and heights.

Keywords: Old-growth forest, Nothofagus betuloides, canopy gaps, solar radiation

transmittances, regeneration, Tierra del Fuego

6.2 Introduction

Although old-growth forests across the globe may share the same ecological

processes and have many similar attributes, such as spatial heterogeneity, they also

differ in many ways, and a single definition cannot be given for all forest types (Spies

2004, Lindenmayer 2008). However, old-growth forest can also be roughly termed

primary forest, which has been defined as forest comprising native species, and where

ecological processes have not been significantly disturbed by human activities, fulfilling

essential functions like the conservation of biological diversity, soil and water

conservation, carbon sequestration and the preservation of aesthetics, cultural and

religious values (FAO 2006). The structures in old-growth forests are heterogeneous,

with an apparent continuous or multi-layered vertical distribution of the canopy, and an

irregular horizontal distribution of vegetation structures due to the presence of multiple

canopy gaps or forest openings, created by mortality of canopy trees (Franklin and Van

Pelt 2004).

Small-scale canopy disturbances, although less dramatic than large disturbances,

may be more frequent and affect a larger area over time (Spies et al. 1990). The severity

of such disturbances can stimulate a change in the growth rates of the surviving trees

(Oliver and Larson 1996), through increasing the availability of the resources (light, soil

moisture, nutrients, microhabitats) promoting plant growth (Canham and Marks 1985,

Runkle 1985, Veblen 1992). The sizes of the canopy gaps affect the microclimate

conditions for seedling establishment, and a gradient in the resources available at the

ground level exists from below the undisturbed canopy to the gap centre (Denslow


139

1980). The regeneration dynamics are affected by the heterogeneity of resources in

forests (Canham et al. 1994). Nevertheless, young trees present and growing slowly

beneath dense canopies have a potential advantage in that they are in a position to

respond to the creation of canopy gaps following small-scale disturbances (Oliver and

Larson 1996, Messier et al. 1999). These features have also been observed in old-

growth Nothofagus betuloides (Mirb.) Oerst forest, where an abundant stock of advance

young trees, persistent seedlings or saplings, can survive in the understorey for a long

time, even in excess of one hundred years (Rebertus and Veblen 1993a, Veblen et al.

1996), until they can eventually ascend into the canopy following the creation of a gap

(Veblen et al. 1996).

The structure of the Nothofagus forests of Tierra del Fuego is shaped principally

by wind, acting as an agent of both large and small-scale disturbance (Rebertus et al.

1997, Puigdefábregas et al. 1999). The windthrow of individual trees creates canopy

gaps smaller than 200 m2 (Rebertus and Veblen 1993a, Gutiérrez 1994). In pure old-

growth N. betuloides forests in Tierra del Fuego discrete gaps might not be apparent, as

openings can occur as interwoven gap complexes in the canopy layer (Rebertus and

Veblen 1993a). Small canopy gaps (51 m2 on average) were observed in a pure uneven-

aged N. betuloides forest, with somewhat larger gaps (107 m2) recorded in a mixed

uneven-aged N. betuloides – Nothofagus pumilio (Poepp. et Endl.) Krasser forest (see

Chapter 3). Wavelike patterns of gap formation have also been documented for pure N.

betuloides, pure N. pumilio, and for mixed N. betuloides - N. pumilio forests in Tierra

del Fuego, seemingly limited to sites predisposed to wind disturbance (Rebertus and

Veblen 1993b, Rebertus et al. 1993, Puigdefábregas et al. 1999). Moreover, a patch

mosaic pattern has been found in these Nothofagus forests, forming multicohort stands,

with patches of younger trees alongside larger patches of old-growth forest (Gutiérrez et

al. 1991).

The small-scale disturbances mentioned above appear to be important in the

regeneration dynamics of southern South American Nothofagus forests situated at

higher elevations and at higher latitudes, where the diversity of tree species is poor

(Veblen et al. 1996, Pollmann and Veblen 2004), as is the case in the forests of Tierra

del Fuego. N. betuloides is capable of establishing beneath small tree-fall canopy gaps

(Rebertus and Veblen 1993a, Gutiérrez 1994, Arroyo et al. 1996). Seedlings and

saplings present on the ground at the time of gap formation are also released by the

creation of these small-scale disturbances in the canopy (Veblen et al. 1996, see Chapter


140

3). The successful establishment and growth of young N. betuloides trees can be

impeded where the ground coverage of understorey trees and shrubs such as Drymis

winteri J.R. et Forster and Maytenus magellanica (Lam.) Hooker is high (Rebertus et al.

1993a, Gutiérrez et al. 1991, Gutiérrez 1994, Veblen et al. 1996).

However, there are still gaps in the knowledge regarding the effects of small-

scale disturbances on regeneration dynamics in old-growth N. betuloides forests. It is

not known whether these small canopy gaps result in a differentiation in the availability

of below-canopy solar radiation between areas beneath natural gaps and areas where the

canopy conditions remain undisturbed. And if there are differences in solar radiation

transmittances, do these influence the density and growth rates of the young trees in old-

growth forests? The influence of canopy gaps on the browsing habits of Lama guanicoe

Müller (guanaco) and the age distribution of the young trees in an old-growth N.

betuloides forest are also unknown.

Since the late 19th century the forests of the coast and much of the interior of

southern Patagonia and Tierra del Fuego have been logged, with the best, largest and

healthiest timber trees (known as floreo in the region) selectively cut and the poor

quality, badly shaped and unhealthy trees left standing or the forest simply burned

(Martínez Pastur et al. 2000, Cruz et al. 2007a). Nevertheless, presently in Chile those

N. betuloides forests suitable for timber production are managed under either a selection

or a shelterwood system (Donoso 1981), designed to promote natural regeneration after

logging. However, many stands are hardly managed at all, with regeneration cuts the

only part of the shelterwood system implemented (Cruz et al. 2007a), resulting in a

homogenisation and simplification of the typically uneven-aged structure of this forest

type (Cruz et al. 2008).

It is argued that natural disturbances should be used as a guide to silviculture

(Mitchell et al. 2002), permitting the design of multiple new silvicultural systems to

help facilitate the acquisition of commodities from forest landscapes while at the same

time maintaining biodiversity and ecosystem processes (Lindenmayer and Franklin

2002). The management of primary or old-growth forest should be based on an

ecological understanding of natural stand development, including the role of natural

disturbances, like small-scale disturbances, providing the basis upon which the forest

may be managed as a renewable resource and at the same time leaving behind a

functioning biological system in which a high diversity of species is maintained

(Attiwill 1994 a,b, Coates and Burton 1997, Franklin et al. 2002).


141

As there are currently plans targeting an intensification of the utilisation of N.

betuloides forests, it is necessary to obtain a better idea of their natural dynamics. The

hypothesis central to this study was that the establishment and growth of young N.

betuloides is affected by the occurrence of small-scale disturbances due to the existence

of a gradient in the below-canopy solar radiation conditions from areas beneath canopy

gaps to those below an undisturbed forest canopy. The objectives of this study were to

determine the effects of natural small-scale disturbances (1) on the below-canopy solar

radiation conditions, (2) on the regeneration patterns (density and growth rates) of N.

betuloides, and (3) on the browsing damage by L. guanicoe. This knowledge is

necessary in to order to improve the silvicultural systems currently used in this forest

type, and may represent a basis for the promotion of new silvicultural practices in

southern Chile relying on a gap approach so as to ensure the maintenance of multi-

layered structures and the native biodiversity.

6.3 Materials and Methods

6.3.1 Study Area

The study was conducted in an old-growth, uneven-aged evergreen N. betuloides

forest (20 ha, 1,362 trees ha-1, 105 m2 ha-1, Table 2.1 and Figure 2.2) with no direct

evidence of human impact. The forest appeared to be closely related to the evergreen N.

betuloides forest type described by Veblen et al. (1983). The forest is located at the

‘Estancia Olguita’ on the southeastern side of the Río Cóndor (53 °59 ’ S, 69 ° 58 ’ W)

and on the southwestern Chilean side of Tierra del Fuego (Figure 3.1).

The climate of the area belongs to the Northern Antiboreal sub-zone, which has

a mean air temperature of 9.0-9.5 °C in the warmest month of the year and remains

above zero in the coldest month. The mean annual rainfall is around 500-600 mm, but

can reach up to 900 mm. The wind direction is commonly west to southwest, with

speeds on average of between 14-22 km h-1 and with a maximum windspeed in summer

of more than 100 km h-1 (Tuhkanen 1992).


142

6.3.2 Selection of Canopy Gaps

Thirteen canopy gaps (Table 6.1) recorded in a study of canopy gaps and

disturbance dynamics in a forest dominated by N. betuloides in Tierra del Fuego were

selected (see Chapter 3). A total of 25 canopy gaps (> 20 m2) were found along parallel

transects. A ‘gap’ was defined as the horizontal projection of a canopy opening to the

ground surface (Runkle 1982), and was considered closed if the vegetation growing

below the opening in the canopy was more than 2-3 m in height. The ‘expanded gap’

was defined as the area formed by the gap in the canopy plus the adjacent area delimited

by the bases of the edge trees (Runkle 1982) (see Chapter 3). Only canopy gaps

occurring on podsolic, well-drained and shallow soils (< 50 cm) were selected.

Table 6.1: Mean and range characteristics of the 13 selected canopy gaps. Numbers in

brackets are standard errors.

Canopy gap area (m2) Expanded gap area (m2) Gap-makers per gap

Mean 47 (± 6.3) 167 (± 11.7) 2 (± 0.3)

Range 21-92 110-278 1-5

6.3.3 Young Tree Measurements

All seedlings and saplings present in plots of 4 m2 (2 x 2 m) were counted during

the summer of 2006. The plots were located along a gradient of expected solar radiation

conditions, for which transects running from areas beneath undisturbed canopy to the

centre of canopy gaps were established (Figure 5.1).

Five plots were established in and around each canopy gap. One plot was located

in the centre of the gap, two at the edges of the gap (southeast and northwest of centre),

and two more below undisturbed canopy at a distance of half the height of the highest

tree in the vicinity of the gap (26 m ± 0.9). This distance was measured from the bases

of the trees at the edges of the gap.

The young tree (seedlings and saplings) growth was classified according to the

following classes: germinant and non-lignified, lignified and ≤ 20 cm tall, 21-50 cm tall,

51-100 cm tall, 100-200 cm tall and ≤ 5 cm in dbh. The number of seedlings and

saplings in each height class browsed by L. guanicoe was counted.


143

The height of the tallest seedling or sapling in each plot was measured, and its

annual height increments over the previous 5 years reconstructed by measuring the

lengths between the bud scars left on the bark of the axis by leaves and branches. This

plant was harvested and a stem disc cut at the root collar in order to measure the

diameter, age and the widths of the last 10 annual growth increments, where this was

possible. The plants harvested ranged in diameter at root collar from 1.5-25 mm and in

height from 12-248 cm. All discs (n = 65) were dried, sanded and processed at the

Department of Silviculture of the University of Chile. Ring widths were recorded to the

nearest 0.01 mm using a microscope. For the purposes of this study only the last annual

height and radial increment values were used because of the lack of any solar radiation

data for the other years.

The height increment and annual radial growth were expressed as ratios – to

total height and total radius, respectively - because the use of relative measures ruled

out biases associated with the comparison of absolute growth values for trees of

different sizes (Stancioiu and O’Hara 2006).

6.3.4 Measuring Below-Canopy Solar Radiation

The solar radiation transmittance was estimated indirectly using hemispherical

photographs. During the summer of 2006 a total of 65 hemispherical photographs were

taken, one at the location of each sampling plot.

All of the hemispherical photographs were taken with a Nikon Coolpix 990®

digital camera (Nikon Corporation, Tokyo, Japan) fitted with a Nikon FC-E8® fisheye

converter (Nikon Corporation, Tokyo, Japan). The camera was mounted on a tripod at a

height of approximately 1.3 m above the ground. The camera and lens were levelled

with the aid of a spirit level and oriented to the magnetic north.

The digital images were processed following the method developed by Brunner

(2002), including the manual setting of a threshold value to separate canopy and sky

elements into a binary black and white image (Anderson 1964a) and converted to

greyscale. All of the images were analysed using HemiView (Delta-T Devices,

Cambridge, UK) (Rich et al. 1999). Using the Coolpix 900 option (Hale and Edwards,

2002) the lens distortion was corrected. A universal overcast sky (UOC) model was

used to describe the intensity of the diffuse radiation (Steven and Unsworth 1980). The

model assumes all regions of the sky to be equally bright. As no actual measurements of


144

the diffuse and direct radiation were available for the study area, a relative proportion of

direct and diffuse radiation equal to 0.5 was assumed (Canham et al. 1990).

Direct, diffuse and global solar radiation transmissions were estimated during

the vegetation period (between October and March). The outputs were non-cosine-

corrected, which was desirable for the purposes of measuring solar radiation from all

directions (Rich 1990, Brunner 1998).

6.3.5 Statistical Analysis

In order to contrast the variation in the non-cosine-corrected solar radiation

(direct, diffuse and global), the density of the young growth (plants m-2), the numbers

browsed by L. guanicoe, the relative radial growth and the relative radial height

increment between areas beneath disturbed and undisturbed canopy, the values

calculated for the centre of the canopy gaps, the two gap edges and beneath the closed

canopy were compared using the non-parametric Kruskal-Wallis H-test, and the Mann-

Whitney U-test as a post-hoc test. The non-parametric approach was chosen for its

greater robustness in dealing with small samples, and as the variable values were not

normally distributed. The selected tests are less affected by the failure of assumptions

than parametric (Sokal and Rohlf 2000). The hypothesis tested was that the variables

mentioned above derived from the same population. Acceptance of the null hypothesis

would mean that there was no difference between the values obtained beneath each of

the canopy situations analysed.

Spearman’s rank correlation was used to test the strength of the relationships

between the non-cosine-corrected solar radiation (direct, diffuse and global), the density

of the young growth classed by height and the ratio of annual height increment to total

height, and the ratio of annual radial increment to the radius.

A regression analysis was performed to discern a relationship between the

height, the diameter at the root collar, the age, the relative radial growth and the relative

height increment. For the regression analysis the goodness-of-fit was evaluated using

the coefficient of determination (R2), the root mean square error (RMSE), and the

significance of the p-value. All statistical analyses were carried out using SPSS 15.0 for

Windows (SPSS, Inc.).


145

6.4 Results

6.4.1 Solar Radiation Transmittance

The transmission of the non-cosine-corrected direct solar radiation into the forest

during the growing season (from October to March) ranged between 3.2-19.4 % of the

above canopy solar radiation. The highest average value (9.5 %) was observed beneath

an undisturbed canopy, to the southeast of the gap centre. However, this did not differ

significantly (n = 13, p > 0.05) from the 8.2 % below-canopy direct solar radiation

recorded in the gap centre, the 8.0 % at the southeastern gap edge and the 8.2 %

recorded beneath the closed canopy to the northwest of the gap centre. The values

determined for the northwestern edges of the gaps by contrast were significantly lower

(average 5.5 %) (Table 6.2).

Table 6.2: Descriptive statistics for the non-cosine-corrected direct, diffuse and global

solar radiation transmittances in different locations in the forest – in gaps, at gap edges

and beneath undisturbed canopies. The standard error is in brackets. Identical letters

indicate no significant difference between the different locations in the forest (Kruskal-

Wallis H-test and Mann-Whitney U-test as a post-hoc test, p > 0.05, n = 13).

Gap edge Closed canopy Non-cosine-corrected solar

radiation transmittances Gap

SE NW SE NW

Mean 8.2 a 8.0 a 5.5 b 9.5 a 8.2 a

(SE) (1.0) (0.9) (0.8) (0.7) (1.1) Direct

Range 4.5-19.4 3.8-15.5 3.2-11.9 5.8-13.8 4.0-17.8

Mean 8.7 a 7.8 a 7.1 ab 6.1 b 6.3 b

(SE) (0.8) (0.4) (0.5) (1.0) (1.5) Diffuse

Range 5.8-16.7 5.4-10.2 3.2-11.0 4.0-7.7 3.1-8.8

Mean 8.6 a 7.9 a 6.6 a 7.2 a 6.9 a

(SE) (0.8) (0.5) (0.8) (0.3) (0.5) Global

Range 6.2-17.6 5.2-11.9 3.2-9.9 5.0-8.5 3.6-10.6


146

The non-cosine-corrected diffuse solar radiation in the understorey ranged

between 3.2-16.7 % of the solar radiation above the canopy (Table 5.2). The highest

values were found in the gap centres (average 8.7 %) and at the edges (7.8 % to the

southeast and 7.1 % to the northwest). These values did not differ significantly (n = 13,

p > 0.05). Lower, albeit not significantly different, diffuse solar radiation transmittances

were calculated beneath undisturbed canopies (6.1-6.3 %). These were also similar to

the values estimated at the northwestern edge of the gaps.

The range of global solar radiation transmittances lay between 3.2-17.6 % of the

total above canopy solar radiation (Table 6.2). Although the transmittances in the gap

centre were higher (average 8.6 %), there was no significant difference (n = 13, p >

0.05) between this value and the values for the gap edges (7.9 % to the southeast and

6.6 % to the northwest) or beneath the undisturbed canopy (7.2 % to the southeast and

6.9 % to the northwest).

6.4.2 Young Tree density and the Influence of Solar Radiation

Young trees were generally present in all of the plots, but with a high degree of

variation in density, which ranged between 2-93 plants m-2. The total density did not

differ significantly between any of the gap situations (n = 13, p > 0.05). The density

ranged from an average of 16.9 plants m-2 beneath the closed canopy to the southeast of

the gap centre and 24.4 plants m-2 beneath the closed canopy to the northwest of the gap

centre. Nor was there any significant difference in terms of the density of the young

trees under the different canopy gap situations according to the height classes analysed.

The highest numbers of seedlings found were in the height class ≤ 20 cm, with a range

of 1-86 plants m-2, and averages of between 12.2-19.9 plants m-2. The other height

classes were characterised by lower plant densities, with ranges of between 0-6

germinants and non-lignified plants m-2, 0-10 seedlings m-2 in the class 21-50 cm tall, 0-

8 plants m-2 between 51-100 cm tall, and 0-5 plants m-2 in the class 101-200 cm tall

(Table 6.3).

The average rates of browsing by L. guanicoe were low (between 0.7-2.8% of all

seedlings and saplings of N. betuloides sampled). L. guanicoe did not exhibit any

habitat preference, with young trees damaged in both gaps and beneath undisturbed

canopies (n = 13, p > 0.05). Seedlings between ≤ 20-100 cm showed the least a signs of

browsing (Table 6.3).


147

Table 6.3: Density of the young trees (plants m-2) and percentage of plants browsed by

Lama guanicoe at different locations in the forest – in gaps, at gap edges and beneath

undisturbed canopies. The standard error is in brackets. Identical letters indicate no

significant difference between the different locations in the forest (Kruskal-Wallis H-

test and Mann-Whitney U-test as a post-hoc test, p > 0.05, n = 13).

Gap edge Closed canopy Height classes (cm) Gap

SE NW SE NW

Mean 0.9 a

(0.2)

1.6 a

(0.5)

0.4 a

(0.1)

0.8 a

(0.2)

0.9 a

(0.3)

Range 0 – 2 0 – 6 0 – 2 0 – 3 0 – 4

Germinants

and non-

lignified Browsed (%) 0.0 a 0.0 a 0.0 a 0.0 a 0.0 a

Mean 12.5 a

(1.6)

15.9 a

(3.1)

13.5 a

(3.0)

12.2 a

(2.2)

19.0 a

(5.9)

Range 4 - 24 1 - 40 4 - 43 1 - 27 3 - 86

≤ 20 cm

tall

Browsed (%) 0.8 a 0.8 a 0.6 a 2.5 a 2.1 a

Mean 3.3 a

(0.7)

2.4 a

(0.6)

3.6 a

(0.8)

2.5 a

(0.5)

2.7 a

(0.6)

Range 0 – 7 0 – 8 0 – 10 0 – 7 0 – 6

21-50 cm

tall

Browsed (%) 1.2 a 3.1 a 1.6 a 2.3 a 8.5 a

Mean 1.8 a

(0.6)

1.0 a

(0.3)

1.8 a

(0.6)

0.5 a

(0.1)

1.1 a

(0.4)

Range 0 – 6 0 – 3 0 – 8 0 – 2 0 – 5

51-100 cm

tall

Browsed (%) 0.0 a 3.8 a 0.0 a 7.1 a 5.1 a

Mean 0.4 a

(0.2)

0.3 a

(0.1)

0.9 a

(0.3)

0.9 a

(0.4)

0.7 a

(0.3)

Range 0 – 2 0 – 2 0 – 3 0 – 5 0 – 3

101-200 cm

tall

Browsed (%) 0.0 a 0.0 a 0.0 a 0.0 a 0.0 a

Mean 18.9 a

(2.3)

21.3 a

(3.9)

20.3 a

(4.0)

16.9 a

(2.6)

24.4 a

(5.9)

Range 4 – 31 2 – 49 6 – 54 4 – 31 12 – 93 Total

Browsed (%) 0.7 a 1.2 a 0.7 a 2.4 a 2.8 a

There was a significant correlation between the non-cosine-corrected solar

radiation and the density of young trees in the height classes 21-50 cm and 51-100 cm

(Table 6.4). The density of young trees between 21-50 cm revealed a positive


148

correlation with the direct (Spearman rank correlation coefficient: 0.319) and the global

solar radiation (Spearman rank correlation coefficient: 0.405), and also with the diffuse

solar radiation (Spearman rank correlation coefficient: 0.293). The density of young

trees in the height class 51-100 cm was correlated with the diffuse and global solar

radiation (Spearman rank correlation coefficient: 0.275 and 0.335, respectively). The

density of young trees in the small classes (germinant and non-lignified, lignified and ≤

20 cm tall) and in the higher class (101-200 cm) showed no relation to the

transmittances of any type of below-canopy solar radiation. However, the global solar

radiation transmitted into the forest correlated with the overall young tree density

(Spearman rank correlation coefficient: 0.284).

Table 6.4: Spearman’s rank correlation matrix of the young tree density according to

height class in relation to the non-cosine-corrected direct, diffuse and global solar

radiation transmittances. * indicates significance at the 5% level, and ** at the 1% level.

Height class (cm) Direct Diffuse Global n

Germinants and non-lignified 0.061 0.211 0.154 65

≤ 20 cm tall 0.123 0.129 0.141 65

21-50 cm tall 0.319** 0.293* 0.405** 65

51-100 cm tall 0.209 0.275* 0.335* 65

101-200 cm tall 0.099 0.024 0.107 65

Total 0.222 0.235 0.284* 65

6.4.3 Relations between Height, Diameter at Root Collar and Young Tree Age

An allometric growth curve (upward concave and increasing) provided the best

fit for the height and the diameter at the root collar. Thus, the proportion of both

structural variables remained roughly constant, yielding a relatively high increase in

height with respect to the root collar diameter. The model explained 93% of the

variance in the height of the young trees. The model and regression coefficient values

and the goodness of fit are presented in Figure 6.1.


149

Figure 6.1: The relationship between the height and the root collar diameter of the

young trees. The line indicates the best model fitted for the estimation of the age based

on the diameter at root collar.

Furthermore, the allometric growth curves explained the ages of the plants with

79 % and 73 % confidence using the diameter at root collar and the height, respectively

(Figure 6.2). In the first case (age vs. diameter at root collar) the curve modelled was

downward concave and increasing, whereas in the second the curve was upward

concave and increasing.

6.4.4 Relative Height Increment and Radial Growth with Respect to Age and

Solar Radiation

The relative radial growth of the young trees, expressed as the ratio to the total

plant radius, did not differ significantly (n = 13, p > 0.05) between locations. The

average increment of plants growing in gaps was 0.047 mm mm-1, 0.044 mm mm-1 at

the southeastern edge and 0.040 mm mm-1 at the northwestern edge, and 0.038 mm mm-

1 beneath the canopy to the southeast of the gap centre and 0.054 mm mm-1 beneath the

canopy to the northwest (Figure 6.3). The relative height increment did not differ

between canopy situations either, with average values of 0.083 cm cm-1 in canopy gaps,

0.084 cm cm-1 and 0.098 cm cm-1 at the southeastern and northwestern gap edges

respectively, and 0.072 cm cm-1 and 0.084 cm cm-1 beneath the canopy to the

southeastern and the northwest of the centre respectively (Figure 6.3).

y= 8.52925*x 1.05115 n = 58; R2 = 0.93; RMSE = 0.197; p < 0.001


150

Figure 6.2: The relationship between the height and the root collar diameter of the

young trees. The line indicates the best model fitted for the estimation of the age based

on the diameter at root collar.

y= 3.86379*x 0.79822 n = 58; R 2 = 0.79; RMSE = 0.275; p < 0.001

(a)

y= 0.99899*x 0.69974 n = 58; R2 = 0.73; RMSE = 0.317; p < 0.001

(b)


151

Figure 6.3: Box-whisker plots indicating (a) the radial growth relative to total radius,

and (b) the total height increment to total height of the young trees assessed at different

locations in the forest – in gaps, at gap edges (southeast and northwest) and beneath

undisturbed canopies (southeast and northwest). The margins of the box are the upper

and lower quartiles, the difference between them the interquartile range (IQR). The

horizontal line within the box represents the median. The whiskers indicate the

minimum and maximum values. The small open circle indicates an outlier, values

between 1.5 and 3 times the IQR. The small asterisk indicates an extreme value, more

than 3 times the IQR. Identical letters indicate no significant difference between the

locations in the forest (Kruskal-Wallis H-test and Mann-Whitney U-test as a post-hoc

test, p > 0.05, n = 58).

Furthermore, neither the non-cosine-corrected direct, diffuse nor global solar

radiation transmitted into the forest correlated with either the relative radial growth or

(a)

a a a a a

(b)

a a a a a


152

the relative height increment (Table 6.5). They correlated negatively only with the age

of the plants (Spearman rank correlation coefficient: -0.733 and -0.488, respectively).

Table 6.5: Spearman’s rank correlation matrix of the relative height increment and

radial growth of young trees with respect to age and non-cosine-corrected direct, diffuse

and global solar radiation transmittances. * indicates significance at the 5% level, and

** at the 1% level.

Variables Relative height

increment

Relative radial

growth

Relative height increment 1.000

Relative radial growth 0.441** 1.000

Age -0.488** -0.733**

Direct solar radiation -0.248 -0.176

Diffuse solar radiation 0.157 -0.067

Global solar radiation -0.049 -0.139

Therefore, inverse polynomial curves explained the relationships between

relative radial growth and relative height increment and the age of the plants best

(Figure 6.4). In the models the age explained 70 % of the variance in the relative radial

growth and 50 % of the variance in the relative height increment.

6.5 Discussion

6.5.1 Below-Canopy Solar Radiation Conditions

The results indicated that there is a gradient in the below-canopy solar radiation

from the centre of the gap to the undisturbed canopy in old-growth N. betuloides forest

over the course of a growing season, as described by Ricklefs (1977), Denslow (1980),

Canham et al. (1990, 1994) and Lieffers et al. (1999). The characteristic features of old-

growth forest (Franklin and Van Pelt, 2004), such as the presence of small canopy gaps

and the heterogeneity of the canopy with a multi-layered vertical distribution and an

irregular horizontal distribution of canopy structures, result in a more heterogeneous


153

distribution of solar radiation in the understorey. In general, a major increase in solar

radiation has been found in the centre of gaps (Canham et al. 1990). As the studied

forest is located in the southern hemisphere, solar radiation was found to be greater at

the southeastern edge of gaps rather than the northwestern. This contrasts with the

observations made in forests of the northern hemisphere (Canham et al. 1990, Comeau

1998, Lieffers et al. 1999). Furthermore, higher transmittances of non-cosine-corrected

direct solar radiation were measured beneath the closed canopy to the southeast of the

gap centre as the effect of canopy gaps on the below-canopy solar radiation extends

beyond the edge of the gaps at higher latitudes, where the sun is low in the sky (Canham

et al. 1990, Gray et al. 2002).

Figure 6.4: The relationships between the age and (a) the radial growth relative to the

radius at the root collar, (b) the height increment relative to the total height. The lines

describe the best model fitted for the estimation of the relative radial growth and height

increment on the basis of the age of the young trees.

y= 0.03350 + 1.03822 / x n=58; R2=0.50; RMSE=0.044; p<0.001

(b)

y= 0.01759 + 0.54923 / x n=58; R2=0.70; RMSE=0.015; p<0.001

(a)


154

6.5.2 Regeneration Pattern and the Relationship between Young Growth and

Below-Canopy Solar Radiation

Only minor spatial differences in the natural regeneration processes were found

within the old-growth N. betuloides forest. The young tree density (Table 6.3) was

similar beneath both gaps and the closed canopy. These seedling and sapling densities

were between two and three times those recorded in Tierra del Fuego by Rebertus and

Veblen (1993a), with an average of 6.1 plants m-2 growing beneath the closed canopy

and 8.5 plants m-2 in gaps. A large number of seedlings were found in the lower height

classes, with fewer in the higher classes. This corresponded with findings from uneven-

aged forests dominated by red beech (Nothofagus fusca (Hook. F.) Oerst.) in New

Zealand, where the frequency distribution of the seedlings as a function of height

approached a negative exponential (June and Ogden 1975, Wardle 1984).

The density of plants belonging to the shortest class (≤ 20 cm tall) revealed no

influence of below-canopy solar radiation (Table 6.4). An explanation for this might be

that the young plants are more dependent upon below ground resources than the light

transmitted into the forest (Stancioiu and O’Hara 2006). For example, seedlings are

susceptible to drought, as has been described for N. pumilio (Heinemann et al. 2000)

and New Zealand Nothofagus species (Wardle 1984). The density of plants 21-100 cm

by contrast was correlated with the solar radiation (Table 6.4). However, the plant

density decreased with increasing height, probably as a result of competition for access

to light in this extremely shaded environment.

N. betuloides seedlings are able to establish in the shaded understorey of the old-

growth forest, and can survive and grow slowly (Rebertus and Veblen 1993a and see

Chapter 3). The death of canopy trees and the creation of small canopy gaps in a small-

scale forest texture promote an uneven pattern of regeneration in the old-growth N.

betuloides forest, with seedlings, saplings and pole stage trees growing into the canopy,

showing a great range of ages and heights (Peet and Christensen 1987, Oliver and

Larson 1996, Barnes et al. 1998).

6.5.3 Browsing Effects

The levels of browsing damage to young N. betuloides by L. guanicoe were very

low (Table 6.3), less than the effect on the deciduous N. pumilio. L. guanicoe feeds on


155

the leaves of the three Nothofagus species occurring in Tierra del Fuego, but prefers N.

pumilio (Raedeke 1980). L. guanicoe has become a serious problem for the successful

natural regeneration of N. pumilio, in both natural canopy gaps (Arroyo et al. 1996,

Rebertus et al. 1997, Cavieres and Fajardo 2005) and in harvested stands (Martínez

Pastur et al. 1999, Pulido et al. 2000). The results of this study failed to show any

particular preference of L. guanicoe for forest conditions, with trees browsed in gaps

and beneath the closed canopy.

6.5.4 Silvicultural Implications for Old-Growth N. betuloides Forest

As historically the canopy layer in old-growth N. betuloides forest has probably

been affected mainly by fine-scale disturbances (small gaps), the uneven-aged canopy

structure has been maintained over time (see Chapter 3). Therefore, the competitive

pressure exerted upon young trees by the trees in the canopy is great (Oliver and Larson

1996, Barnes et al. 1998). Just as with the regeneration of Nothofagus species in New

Zealand (Wardle 1984), seedlings and saplings of N. betuloides progress through the

understorey slowly but steadily, often remaining suppressed even after the death of a

large canopy tree. Thus, the age of the plants is a better predictor of both their size

(height and root collar diameter) and their growth rates than the transmittance of solar

radiation into the forest, as young N. betuloides may reduce height growth when

growing in shade, as is the case with the advance regeneration of boreal forests (Messier

et al. 1999).

From a silvicultural point of view, in order to maintain uneven-aged

characteristics in the stands the selection method can be employed to promote

regeneration, i.e., the trees are removed as individuals or in small groups (Smith et al.

1997, Lindenmayer and Franklin 2002). However, given the characteristics of N.

betuloides forest, with an abundant stock of advance young trees, the stand

establishment must typically occur rapidly after harvesting. This should be based on

silvicultural system oriented according to natural disturbances. Different harvesting

scenarios tested for temperate rain forest in southern Chile revealed that the structure

and composition of old-growth forest can be better maintained using the selection

method, although providing lower harvests (Rüger et al. 2007).

Knowing the distribution of gap sizes, their dispersion and dynamics can help to

explain much of the natural development of a stand (Coates and Burton 1997). Thus,


156

management strategies and silvicultural practices incorporating natural disturbances and

the natural structural development of forest ecosystems might be availed of to maintain

the uneven-aged structure of this old-growth forest, to preserve the native biodiversity,

to protect the ecosystem from exotic species invasions and to promote the conservation

of biotic and abiotic interactions necessary for natural regeneration after harvesting

(Franklin et al. 2002, Rüger et al. 2007).

In order to maintain the unique characteristics of individual old-growth stands,

tools integrating the features of stand structure into management guidelines are needed

(O’Hara et al. 2008). This study can serve as a framework for new silvicultural practices

to address multifunctional ecosystem management objectives in old-growth N.

betuloides forests of southern Patagonia and Tierra del Fuego.

Chapter 7 Conclusions and silvicultural implications

157

CHAPTER 7

CONCLUSIONS AND SILVICULTURAL IMPLICATIONS

7.1 Disturbance and Stand Dynamics

The regeneration patterns in South American Nothofagus forests located at

higher elevations or latitudes are influenced by large-scale disturbances, which lead to

the formation of even-aged stands (Veblen et al. 1996, Pollmann and Veblen 2004,

Donoso and Donoso 2007). In southern Patagonia and Tierra del Fuego N. betuloides

can grow as pioneer species on moraines and recently deglaciated areas (Pisano 1978,

Armesto et al. 1992, Moore and Pisano 1997), on land recently disturbed by fire

(Martínez Pastur et al. 2002, Cruz et al. 2007a), and along forest roads. The

development of N. betuloides stands can be described on the basis of the four general

physiognomic stages following major disturbances (sensu Oliver 1981), namely 1) stand

initiation, 2) stem exclusion, 3) understorey reinitiation, and 4) old-growth (Figure 7.1).

Figure 7.1: The four hypothetical stages of stand development following a large-scale

disturbance to a Nothofagus betuloides forest (source: Oliver 1981).

The old-growth stage has been defined by Oliver (1981) as the part of a forest’s

life cycle “when the overstorey gradually dies and the understorey slowly fills in to

replace it”. The younger trees growing upward usually include individuals of a range of

sizes and ages; hence the stand as a whole contains trees of a wide range of ages and

Stand initiation

stage

Stem exclusion

stage

Understorey reinitiation

stage

Old growth stage

Larg

e-sc

ale

dist

urba

nce


158

heights (Oliver and Larson 1996), forming an uneven-aged mosaic pattern of

successional stages (Oliver 1981).

Similar dynamics were observed in both the pure N. betuloides forest and in the

mixed N. betuloides – N. pumilio forest. Many decades of stability are indicated by the

uneven-aged canopy structures, facilitated by the occurrence of fine-scale disturbances

over a period of at least the last 150 years. These small disturbances lead to the creation

of small canopy gaps (Figure 7.2) with an average size of 51 m2 (± 4.7, SE) in the pure

N. betuloides forest and 107 m2 (± 21.5) in the mixed N. betuloides – N. pumilio forest.

These disturbances are contributed to by the senescence over time of the larger canopy

trees, the result of either a dieback phenomenon or a disease-decline. As a result, small

gaps are created in the canopy, with a gradual break-up of the even-aged structure. This

is especially evident in the pure N. betuloides forest. These development and the

presence of some older relic trees in the canopy support the idea that both the pure N.

betuloides and the mixed N. betuloides – N. pumilio forests are entering into an old-

growth stage or represent a transition to an old-growth stage (sensu Oliver and Larson

1996).

Silvicultural Implications

Figure 7.2: Schematic diagram of a small canopy gap in the ‘old-growth stage’ of the

Nothofagus betuloides forest.

7.2 Solar Radiation and Tree Species Regeneration: Testing the

Hypothesis

Hypothesis: The establishment and growth of young N. betuloides is affected by the

occurrence of natural, small-scale disturbances and the resultant existence of a gradient

in the below-canopy solar radiation conditions, with changing solar radiation intensities

from areas beneath canopy gaps to those below an undisturbed forest canopy.

Small canopy gap


159

7.2.1 Below-canopy solar radiation transmittances

Several instruments have been developed to measure either directly or indirectly

the below-canopy solar radiation environment. Many studies have demonstrated that

hemispherical photography (indirect technique) has a high level of accuracy as a means

to estimate canopy structures and below-canopy solar radiation environments (Canham

et al. 1990, Rich et al. 1993, Roxburgh and Kelly 1995, Wright et al. 1998, Clearwater

et al. 1999, Engelbrecht and Herz 2001, Bartemucci et al. 2006, Collet and Chenost

2006). The use of conventional film photography has largely given way to digital

photography (Hale and Edwards 2002), and comparisons between the data provided by

both techniques have been carried out, demonstrating comparable results (Englund et al.

2000, Frazer et al. 2000, Hale and Edwards 2002). A variety of commercial and free

software packages are available for the analysis of hemispherical photographs (Comeau

2000), with the latter available for download from the internet. The results of the study

presented in chapter 4 revealed no differences between the estimates of the most

commonly used canopy structure and solar radiation variables calculated by the

commercial (HemiView) and free software packages (Gap Light Analyzer, hemIMAGE

and Winphot).

The part of the hypothesis related to the existence of a gradient in the below-

canopy solar radiation conditions from areas beneath canopy gaps to those below an

undisturbed forest canopy appear to have been partially confirmed. There was a gradient

in the below-canopy solar radiation environment from the centre of the gap to the

undisturbed canopy in the pure old-growth N. betuloides forest during the growing

season, as was described in other studies (Ricklefs 1977, Denslow 1980, Canham et al.

1990, 1994, Lieffers et al. 1999). Nonetheless, although canopy gaps appear to increase

the amount of solar radiation reaching the forest floor, higher transmittances of direct

solar radiation during the growing season were found beneath the closed canopy, to the

southeast of the gap centre. At higher latitudes, the effect of canopy gaps on the below-

canopy solar radiation extends beyond the edge of the gaps, as was observed by

Canham et al. (1990) and Gray et al. (2002). The transmission of the solar radiation into

the forest, therefore, was affected not only by a high level of horizontal and vertical

heterogeneity of the forest canopy, but also by the low angles of the sun’s path.

The characteristic features of old-growth forest (Franklin and Van Pelt, 2004),

such as the presence of small canopy gaps and a heterogeneous canopy with a multi-


160

layered vertical distribution and an irregular horizontal distribution of canopy structures,

result in a more homogenous distribution of the solar radiation below the canopy.

Finally, it was observed that the canopy structure and stand parameters explain

partly the variation in the below-canopy solar radiation regime, and a combination of

the basal area, crown projection, crown volume, stem volume and the average

equivalent crown radius explained as much as 75 % of the variation in the diffuse and

73 % of the variation in the global solar radiation transmitted into the forest. This

demonstrates a high degree of heterogeneity of the spatial structures of the pure old-

growth forest, affecting the amount of solar radiation reaching the forest floor.

7.2.2 Regeneration patterns of Nothofagus betuloides

The regeneration of N. betuloides in the pure forest was not particularly linked to

the occurrence of canopy gaps. The regeneration density beneath gaps and under the

closed canopy was similar, and was higher than recorded in another study from Tierra

del Fuego (Rebertus and Veblen 1993a). A large number of the seedlings were in the

lower height classes, and fewer in the higher classes.

N. betuloides seedlings are able to establish, survive and grow slowly in the

shaded understorey of the pure old-growth forest.

No relationship was observed between the below-canopy solar radiation and the

density of plants belonging to the smallest class (≤ 20 cm tall), which is likely to have

been due to the fact that the young plants are more dependent upon below-ground

resources than the light transmitted into the forest (Stancioiu and O’Hara 2006). The

density of plants in the size class 21-100 cm was, by contrast, correlated with the solar

radiation.

The presence of small canopy gaps only, and the signs of senescence in the

larger canopy trees in the pure N. betuloides forest, have promoted an uneven pattern of

regeneration, with seedlings, saplings and pole stage trees growing into the canopy, and

the stand showing a great range of ages and heights, as has been observed in other old-

growth forests by Peet and Christensen (1987), Oliver and Larson (1996) and Barnes et

al. (1998).

The establishment of N. betuloides in the pure, old-growth forest is not affected

directly by the occurrence of small canopy gaps and the resultant changes to the below-


161

canopy solar radiation environment. However, the young N. betuloides trees need the

occurrence of small canopy gaps if they are to reach the maturity.

Therefore, longevity and high rates of survival are key traits of N. betuloides that

contribute to its persistence in old-growth forests.

7.3 Shade Tolerance of Nothofagus betuloides

Seedlings, saplings and juvenile trees of shade tolerant species are able to

establish, survive, and grow at low light levels, and persist suppressed in the shaded

understorey for long periods (Canham 1985, Tilman 1988, Barnes et al. 1998). As

overstorey trees die and a gap is created in the canopy, the trees gradually penetrate into

the canopy (Barnes et al. 1998). Saplings may eventually reach the canopy after two or

more periods of suppression (Canham 1985).

In this study it was shown that juvenile N. betuloides trees are able to survive

and grow under closed canopies for decades, indicating a high degree of shade

tolerance. The analysis of the growth responses of juvenile N. betuloides trees to

disturbances indicated irregular disturbance intervals, with many longer periods of

restricted growth followed by short to intermediate periods of release. Saplings revealed

between 0 and 3 periods of release during their lives, accounting for approximately 22-

23 % of their lifespan. Only 41 % of the cores taken in the pure forest and 60 % in the

mixed forest showed at least one period of release. For the juvenile trees of N.

betuloides, the greatest changes in radial increment after release from restricted growth

were found in the mixed forest (11.2 %), and the lowest in the pure forest (2.4 %).

Thus, the juvenile trees growing in the pure N. betuloides forest rely on the

presence of only very small gaps in order to reach maturity. This affects the radial

growth responses of the trees, and reflects the capacity of N. betuloides to remain in

shaded conditions in the understorey for long periods (Rebertus and Veblen 1983a,

Veblen et al. 1996). In the mixed evergreen-deciduous N. betuloides – N. pumilio forest,

by contrast, individual N. betuloides trees are released by the creation of larger canopy

gaps, allowing them to grow into the main canopy (Veblen et al. 1996).

Therefore, it could be shown, that N. betuloides is able to survive as juvenile tree

and grow for decades under closed canopies, indicating a high shade tolerance.


162

7.4 Silvicultural Implications for Old-Growth Nothofagus betuloides

Forests

The intention behind the new Chilean Native Forest Law, recently approved by

the Chilean parliament, is to protect, restore and improve the country's forests in a

sustainable way. The law does not consider timber production to be the forests’ only

function, but also caters for the promotion (and financing through annual funds) of

conservation, restoration and sustainable forest management.

At present there are plans targeting an intensification of the utilisation of N.

betuloides forests. However, the management of primary or old-growth forest should be

based on an ecological understanding of natural stand development, including the role

of natural disturbances such as small openings created in the canopy (Franklin et al.

2007). An understanding of forest ecology provides a basis upon which to manage the

forests as a renewable resource, by influencing the ecological processes, and leaving

behind a functioning biological system in which a high diversity of species is

maintained (Attiwill 1994 a,b, Coates and Burton 1997, Franklin et al. 2002).

The most important objective of silviculture to date has been timber production.

However, silvicultural systems have to integrate multiple and often conflicting

management objectives (Bauhus 1999). Thus, other forest functions, including water

regulation, wildlife management, animal rearing and wood pasture, recreation,

aesthetics, biodiversity conservation, carbon cycling and the production of other, non-

timber products are multi-purpose objectives of the forests that must also be catered for

in silvicultural methods (Smith et al. 1997, Bauhus 1999, Lindenmayer and Franklin

2002).

The silvicultural systems permitted for N. betuloides forests by the current

Chilean Forest Law are the shelterwood and the selection system. In reality, however,

the only part of the shelterwood system that is currently being applied in the barely

managed forests is the regeneration felling (Cruz et al. 2007a). As a consequence, the

once complex stand structures have been homogenised and simplified.

Different harvesting scenarios tested for temperate rain forest in southern Chile

revealed that the structure and composition of old-growth forests can be better

maintained using the selection method, although the yields provided are lower (Rüger et

al. 2007).


163

In order to maintain the uneven-aged characteristics of old-growth N. betuloides

stands, with an abundance of young trees in the understorey, the selection method

should be employed to promote regeneration; i.e., the trees are removed as individuals

or in small groups (Smith et al. 1997, Lindenmayer and Franklin 2002). This favours the

shade tolerant species, and trees typically establish rapidly after harvesting (Tappeiner

et al. 1997).

The same effect can be achieved using similar approaches such as Dauerwald or

continuous cover forest (Helliwell 1997, Schabel and Palmer 1999, Burschel and Huss

2003). This is a close to natural system of forest management based on uneven-aged

silviculture (Schabel and Palmer 1999). Efforts are focused on retaining the trees that

are growing vigorously while removing those trees that have ceased to grow as well

(Helliwell 1997, Smith et al. 1997, Burschel and Huss 2003). This system requires that

natural regeneration occurs in great abundance (Helliwell 1997, Burschel and Huss

2003), which was the found to be the case in the pure N. betuloides forest.

If the aim of forest management is to meet other objectives also, including the

maintenance of the uneven-aged structure of this old-growth forest, the preservation of

the native biodiversity, the protection of the ecosystem from the invasion of exotic

species and the conservation of biotic and abiotic interactions necessary for natural

regeneration after harvesting, management strategies and silvicultural practices

incorporating natural disturbances and the natural structural development of forest

ecosystems might be availed of (Franklin et al. 2002).

In this case, the presence of small canopy gaps in the pure uneven-aged old-

growth N. betuloides forest implies the application of a silvicultural system employing a

gap-based approach, which can be used to address the ecosystem management

objectives (Coates and Burton 1997). This will also allow for the achievement of certain

timber management goals (Franklin et al. 2007), which do not to have be ignored when

implementing a system of natural, disturbance-based silviculture (Seymour and Hunter

1999, Palik et al. 2002). However, focus attention on making a detailed and directed

assessment of disturbances and their effects on the tree regeneration so that natural

disturbance regimes can be used as a guide for regeneration efforts must be paid

(Puettmann and Ammer 2007).

All of the systems and approaches mentioned above (selection method, gap-

based approach, Dauerwald or continuous cover forest approach) may represent a basis

for the promotion of new silvicultural practices in the old-growth forests of southern


164

Chile in the context of the new Chilean Forestry Law as they will contribute to ensuring

the maintenance of multi-layered structures, the native biodiversity, and the ecosystem

processes, while at the same time helping to achieve the economic goals of commercial

timber management.

Chapter 8 References

165

CHAPTER 8

REFERENCES

Alfonso, J.L., 1940. Algunas consideraciones sobre los bosques de Tierra del Fuego.

Ing. Agronómica 1, 10-23.

Anderson, M.C., 1964a. Studies of the woodland light climate I. The photographic

computation of light conditions. J. Ecol. 52, 27-41.

Anderson, M.C., 1964b. Studies of the woodland light climate II. Seasonal variation in

the light climate. J. Ecol. 52, 643-663.

Anderson, M.C., 1964c. Light relations of terrestrial plant communities and their

measurement. Biol. Rev. 39, 425-486.

Anderson, M.C., 1966. Stand structure and light penetration. II. A theoretical analysis.

J. Appl. Ecol. 3, 41-54.

Armesto, J.J, Casassa, I., Dollenz, O., 1992. Age structure and dynamics of Patagonian

beech forests in Torres del Paine National Park, Chile. Vegetatio 98, 13-22.

Arroyo, M.T.K., Donoso, C., Murúa, R.E., Pisano, E.E., Schlatter, R.P., Serey, I.A.,

1996. Toward an ecologically sustainable forestry project. Concepts, analysis

and recommendations. Departamento de Investigación y Desarrollo, Universidad

de Chile, Santiago, Chile.

Attiwill, P.M., 1994a. The disturbance of forest ecosystems: the ecological basis for

conservative management. For. Ecol. Manage. 63, 247-300.

Attiwill, P.M., 1994b. Ecological disturbance and the conservative management of

eucalypt forests in Australia. For. Ecol. Manage. 63, 301-346.

Báez, P., Riveros, M., Lehnebach, C., 2002. Viability and longevity of pollen of

Nothofagus species in south Chile. New Zeal. J. Bot. 40, 671-678

Baldocchi, D., Hutchison, B., Matt, D., McMillen, R., 1984. Seasonal variations in the

radiation regime within an oak-hickory forest. Agric. For. Meteorol. 33, 177-

191.

Barnes, B.V., Zak, D.R., Denton, S.R., Spurr, S.H., 1998. Forest ecology. 4th Edition.

John Wiley & Sons, Inc, New York Chichester Weinheim Brisbane Singapore

Toronto.


166

Barrie, J., Greatorex-Davies, J.N., Parsell, R.J., Mars, R.H., 1990. A semi-automated

method for analyzing hemispherical photographs for the assessment of woodland

shade. Biol. Conserv. 54, 327-334.

Bartemucci, P., Messier, C., Canham, C.D., 2006. Overstory influences on light

attenuation patterns and understory plant community diversity and composition

in southern boreal forests of Quebec. Can. J. For. Res. 36, 2065-2079.

Battaglia, M.A., Mou, P., Pallik, B., Mitchell, R.J., 2002. The effect of spatially variable

overstory on the understory light environment of an open-canopied longleaf pine

forest. Can. J. For. Res. 32, 1984-1991.

Bauhus, J., 1999. Silvicultural practices in Australian native State forests - An

introduction. Australian Forestry 62, 217-222.

Bazzaz, F.A., 1996. Plants in changing environments. Linking physiological,

population, and community ecology. Cambridge University Press, Cambridge.

Becker, P., Erhart, D.W., Smith, A.P., 1989. Analysis of forest light environments. Part

I. Computerized estimation of solar radiation from hemispherical canopy

photographs. Agric. For. Meteorol. 44, 217-232.

Bellow, G., Nair, P.K.R., 2003. Comparing common methods for assessing understory

light availability in shaded-perennial agroforestry systems. Agric. For. Meteorol.

114, 197-211.

Brunner, A., 1998. A light model for spatially explicit forest stand models. For. Ecol.

Manage. 107, 19-46.

Brunner, A., 2002. Hemispherical photography and image analysis with hemIMAGE

and Adobe Photoshop, Horsholm, Denmark. Available from

http://www.umb.no/ina/ansatte/andrb_e.php.

Burschel, P., Huss, J., 2003. Grundriss des Waldbaus. 3., unveränderte Auflage. Eugen

Ulmer GmbH & Co., Stuttgart, Germany.

Busing, R.T., Brokaw, N., 2002. Tree species diversity in temperate and tropical forest

gaps: the role of lottery recruitment. Folia Geobot. 37, 33-43.

Butler-Manning, D., 2008. Stand structure, gap dynamics and regeneration of a semi-

natural mixed beech forest on limestone in central Europe – a case study.

Freiburger Forstliche Forschung Vol 38, Freiburg, Germany.

Butorovic, N., 2003. Resumen meteorológico año 2002, estación "Jorge C. Schythe"

(53°08' S; 70°53' O; 6 m s.n.m.). Ans. Inst. Pat. 31, 123-130.


167

Butorovic, N., 2004. Resumen meteorológico año 2003, estación "Jorge C. Schythe"


Butorovic, N., 2005. Resumen meteorológico año 2004. Estación "Jorge C. Schythe"

(53°08' S; 70°53' O; 6 msnm). Ans. Inst. Pat. 33, 65-71.

Caldentey, J., Promis, A., Schmidt, H., Ibarra, M., 1999-2000. Variación microclimática

causada por una corta de protección en un bosque de lenga (Nothofagus

pumilio). Ciencias Forestales 14, 51-59.

Campbell, G.S., 1986. Extinction coefficients for radiation in plant canopies calculated

using an ellipsoidal inclination angle distribution. Agric. For. Meteorol. 36, 317-

321.

Canham, C.D., 1985. Suppression and release during canopy recruitment in Acer

saccharum. B. Torrey Bot. Club 112, 134-145.

Canham, C.D., 1989. Different responses to gaps among shade-tolerant species.

Ecology 70, 548-550.

Canham, C.D., 1990. Suppression and release during canopy recruitment in Fagus

grandiflora. B. Torrey Bot. Club 117, 1-7.

Canham, C.D., Marks, P.L., 1985. The response of woody plants to disturbance:

patterns of establishment and growth. In: Pickett, S.T.A., White, P.S. (Eds.), The

ecology of natural disturbance and patch dynamics. Academic Press, London,

UK, pp.197-216.

Canham, C.D., Denslow, J.S., Platt, W.J., Runkle, J.R., Spies, T.A., White, P.S., 1990.

Light regimes beneath closed canopies and tree-fall gaps in temperate and

tropical forests. Can. J. For. Res. 20, 620-631.

Canham, C.D., Finzi, A.C., Pacala, S.W., Burbank, D.H., 1994. Causes and

consequences of resource heterogeneity in forests: interspecific variation in light

transmission by canopy trees. Can. J. For. Res. 24, 337-349.

Cavieres, L.A., Fajardo, A., 2005. Browsing by guanaco (Lama guanicoe) on

Nothofagus pumilio forest gaps in Tierra del Fuego, Chile. For. Ecol. Manage.

204, 237-248.

Chan, S.S., McCreight, R.W., Walstad, J.D., Spies, T.A., 1986. Evaluating forest

vegetative cover with computerized analysis of fisheye photographs. For. Sci.

32, 1085-1091.

Chazdon, R.L., Field, C.B., 1987. Photographic estimation of photosynthetically active

radiation: Evaluation of a computerized technique. Oecologia 73, 525-532.


168

Clearwater, M.J., Nifinluri, T., van Gardingen, P.R., 1999. Forest fire smoke and a test

of hemispherical photography for predicting understory light in Bornean tropical

rain forest. Agric. For. Meteorol. 97, 129-139.

Coates, K. D., Burton, P. J., 1997. A gap-based approach for development of

silvicultural systems to address ecosystem management objectives. For. Ecol.

Manage. 99, 337-354.

Collet, C., Chenost, C., 2006. Using competition and light estimates to predict diameter

and height growth of naturally regenerated beech seedlings growing under

changing canopy conditions. Forestry 79, 489-502.

Collins, B.S., Dunne, K.P., Pickett, S.T.A., 1985. Responses of forest herbs to canopy

gaps. In: Pickett, S.T.A., White, P.S. (Eds.), The ecology of natural disturbance

and patch dynamics. Academic Press, London, UK, pp. 217-234.

Comeau, P.G., 1998. LITE: A model for estimating light under broadleaf and conifer

tree canopies. Extension Note 23. B.C. Ministry of Forests Research Branch,

Stn. Prov. Govt. Victoria, BC, Canada. Available from

http://www.for.gov.bc.ca/hfd/pubs/Docs/En/En23.htm.

Comeau, P., 2000. Measuring light in the forest. Extension Note 42. Ministry of Forest

Research Program, British Columbia. Available from

http://www.for.gov.bc.ca/hfd/pubs/Docs/En/En42.htm.

Comeau, P.G., 2001. Relationships between stand parameters and understorey light in

boreal aspen stands. B.C. J. Ecosyst. Manag. 1, 8 (article 2).

Comeau, P.G., Heineman, J.L., 2003. Predicting understory light microclimate from

stand parameters in young paper birch (Betula papyrifera Marsh.) stands. For.

Ecol. Manage. 180, 303-315.

Comeau, P., Gendron, F., Letchford, T., 1998. A comparison of several methods for

estimating light under a paper birch mixed wood stand. Can. J. For. Res. 28,

1843-1850.

Comeau, P., Heineman, J., Newsome, T., 2006. Evaluation of relationships between

understory light and aspen basal area in the British Columbia central interior.

For. Ecol. Manage. 226, 80-87.

CONAF-CONAMA, 1999a. Catastro y evaluación de recursos vegetacionales nativos

de Chile. Informe nacional con variables ambientales. Santiago, Chile.

CONAF-CONAMA, 1999b. Catastro y evaluación de recursos vegetacionales nativos

de Chile. Informe Regional Duodécima Región. Santiago, Chile.


169

Connell, J.H., 1978. Diversity in tropical rain forests and coral reefs. Science 199, 1302-

1310.

Coops, N.C., Smith, M.L., Jacobsen, K.L., Martin, M., Ollinger, S., 2004. Estimation of

plant and area index using three techniques in a mature native eucalypt canopy.

Austral Ecol. 29, 332-341.

Copenheaver, C.A., Abrams, M.D., 2003. Dendroecology in young stands: case studies

from jack pine in northern lower Michigan. For. Ecol. Manage. 182, 247-257

Cottam, G., Curtis, J.T., 1956. The use of distance measures in phytosociological

sampling. Ecology 37, 451-460.

Cruz, G., Lara, A., 1987. Vegetación del área de uso agropecuario de la XII Región,

Magallanes y de la Antártica Chilena. INIA, Intendencia de la XII Región,

Santiago, Chile.

Cruz, G., Caldentey, J. (Eds.), 2007. Caracterización, silvicultura y uso de los bosques

de Coihue de Magallanes (Nothofagus betuloides) en la XII Región de Chile.

Facultad de Ciencias Forestales, Universidad de Chile, Santiago, Chile.

Cruz, G., Caprile, R., Promis, A., Cabello, G., 2007a. Structural and biometric

characterization of Nothofagus betuloides production forests in the Magellan

Region, Chile. J. Sustain. Forest. 24, 123-14.

Cruz, G., Rosenfeld, M., Cabello, G., Hidalgo, F., 2007b. Uso de los bosques de Coihue

de Magallanes. In: Cruz, G., Caldentey, J. (Eds.), Caracterización, silvicultura y

uso de los bosques de Coihue de Magallanes (Nothofagus betuloides) en la XII

Región de Chile. Facultad de Ciencias Forestales, Universidad de Chile,

Santiago, Chile, pp. 100-114.

Cruz, G., Promis, A., Schmidt, H., 2008. Silvicultural treatments for old-growth forests

dominated by Nothofagus betuloides in southern Patagonia, Chile. Abstract in

Proceedings of ‘Old Forest, New Management’ Sir Mark Oliphant Conference,

17-21 February 2008, Hobart, Australia, p. 138.

Cuevas, J.G., 2003. Gap characteristics in relation to forest structure and implications

for southern beech forest dynamics. Can. J. For. Res. 33, 1915-1922.

Darwin, C., 1839. Journal and remarks. In: Fitzroy, R. (Ed.), Narrative of the suveying

voyages of his Majesty’s ships adventure and Beagle between the years 1826

and 1836. Vol. III. Henry Colburn, London, pp. 1832-36.

de Agostini, A., 1945. Andes Patagónicos. Viajes de exploración a la Cordillera

Patagónica Austral. Buenos Aires, Argentina.


170

de Agostini, A., 1956. 30 Años en Tierra del Fuego. Ed. Peuser, Buenos Aires,

Argentina.

de Freitas, C.R., Enright, N.J., 1995. Microclimatic differences between and within

Canopy gaps in a temperate rainforest. Int. J. Biometeorol. 38, 188-193.

Denslow, J.S., 1980. Gap partitioning among tropical rainforest trees. Biotropica 12, 47-

55.

Denslow, J.S., Spies, T., 1990. Canopy gaps in forest ecosystems: an introduction. Can.

J. For. Res. 20, 619.

Dimitri, M.J., 1972. La región de los bosques Andino-Patagónicos. Sinopsis general.

Colección Científica del INTA, Buenos Aires, Argentina.

Donoso, C., 1981. Tipos forestales de los bosques nativos de Chile. Documento de

Trabajo Nº38. Investigación y Desarrollo Forestal (CONAF/PNUD/FAO),

Santiago, Chile.

Donoso, C., Atienza, J., 1983. Hibridación natural entre especies de Nothofagus

siempreverdes en Chile. Bosque 5, 21-34.

Donoso, C., Donoso, P., 2007. Nothofagus betuloides (Mirb.) Oerst. Coigüe de

Magallanes, Ouchpaya (onas), Guindo (zona Austral), Coigüe Colorado (Tierra

del Fuego). In: Donoso, C. (Ed.), Las especies arbóreas de los bosques

templados de Chile y Argentina. Autoecología. Marisa Cuneo Ediciones,

Valdivia, Chile, pp. 411-422.

Donoso, C., Premoli, A., Donoso, P., 2004 Variación en Nothofagus siempreverdes

sudamericanos. In: Donoso, C., Premoli, A., Gallo, L., Ipinza, R. (Eds.),

Variación intraespecífica en especies arbóreas de los bosques templados de

Chile y Argentina. Ed. Universitaria , Santiago, Chile, pp. 189-213.

Duncan, F., 2006. Temperate native forests in Chile. Management, conservation and

forest practices. Report to the J. W. Gottstein Memorial Trust.

Engelbrecht, B.M., Herz, H.M., 2001. Evaluation of different methods to estimate

understorey light conditions in tropical forests. J. Trop. Ecol. 17, 207-224.

Englund, S.R., O’Brien, J.J., Clark, D.B., 2000. Evaluation of digital and film

hemispherical photography and spherical densiometry for measuring forest light

environments. Can. J. For. Res. 30, 1999-2005.

Evans, G.C., Coombe, D.E., 1959. Hemispherical and woodland canopy photography

and the light climate. J. Ecol. 47, 103-113.


171

Fahey, R.F., Puettmann, K.J., 2007. Ground-layer disturbance and initial conditions

influence gap partitioning of understorey vegetation. J. Ecol. 95, 1098-1109.

Fajardo, A., de Graaf, R., 2004. Tree dynamics in canopy gaps in old-growth forests of

Nothofagus pumilio in southern Chile. Plant Ecol. 173, 95-105.

FAO, 2006. Global Forest Resources Assessment 2005. Progress towards sustainable

forest management. FAO Forestry paper 147. FAO, Rome, Italy.

Ferment, A., Picard, N., Gourlet-Fleury, S., Baralato, C., 2001 A comparison of five

indirect methods for characterizing the light environment in a tropical forest.

Ann. For. Sci. 58, 877-891.

Frangi, J.L, Richter, L.L., 1994. Balabces hídricos de bosques de Nothofagus de Tierra

del Fuego, Argentina. Rev. Fac. Agronomía, La Plata 70, 65-79.

Frangi, J.L., Richter L.L., Barrera, M.D., Aloggia, M., 1997. Decomposition of

Nothofagus fallen woody debris in forests of Tierra del Fuego, Argentina. Can.

J. For. Res. 27, 1095-1102.

Franklin, J.F., Van Pelt, R., 2004. Spatial aspects of structural complexity in old-growth

forests. J. Forest. 102, 22-28.

Franklin, J.F., Spies, T.A., Van Pelt., R., Carey, A.B., Thornburgh, D.A., Berg, D.R.,

Lindenmayer, D.B., Harmon, M.E., Keeton, W.S., Shaw, D.C., Bible., K., Chen,

J., 2002. Disturbances and structural development of natural forest ecosystems

with silvicultural implications, using Douglas-fir forests as an example. For.

Ecol. Manage. 155, 399-423.

Franklin, J.F., Mitchel, R.J., Palik, B.J., 2007. Natural disturbance and stand

development principles for ecological forestry. Gen. Tech. Rep. NRS-19.

Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern

Research Station, USA.

Frazer, G.W., Trofymow, J.A., Lertzman, K.P., 1997. A method for estimating canopy

openness, effective plant area index, and photosynthetically active photon flux

density using hemispherical photography and computerized image analysis

techniques. Can For Serv, Pac For Cent. Inf Rep No BC-X-373.

Frazer, G.W., Canham, C.D., Lertzman, K.P., 1999. Gap Light Analyzer (GLA):

Imaging software to extract canopy structure and gap light transmission indices

from true-colour fisheye photographs, users manual and program

documentation, Simon Fraser University, Burnaby, British Columbia, and the

Institute of Ecosystem Studies, Millbrook, New York.


172

Frazer, G.W., Trofymow, J.A., Lertzman, K.P., 2000. Canopy openness and leaf area in

chronosequences of coastal temperate rainforests. Can. J. For. Res. 30, 239-256.

Frazer, G.W., Fournier, R.A., Trofymow, J.A., Hall, R.J., 2001. A comparison of digital

and film fisheye photography for analysis of forest canopy structure and gap

light transmission. Agric. For. Meteorol. 109, 249-263.

Frederiksen, P., 1988. Soils of Tierra del Fuego. A satellite-based land survey approach.

Appendix: Methods and soil data - Annex with 5 plates. Folia Geographica

Danica, Tom. XVIII.

Frelich, L.E., Lorimer, C.G., 1991. Natural disturbance regimes in hemlock-hardwood

forests of the Upper Great Lakes Region. Ecol. Monogr. 61, 145-164.

Gajardo, R., 1994. La vegetación natural de Chile. Clasificación y distribución

geográfica. Ed. Universitaria, Santiago, Chile.

Gálhidy, L., Mihók, B., Hagyó, A., Rajkai, K., Standovár, T., 2006. Effects of gap size

and associated changes in light and soil moisture on the understorey vegetation

of a Hungarian beech forest. Plant. Ecol. 183, 133-145.

Gärtner, S., Reif, A., 2005. The response of ground vegetation to structural change

during forest conversion in the southern Black Forest. Eur. J. For. Res. 124, 221-

231.

Gea-Izquierdo, G., Martínez Pastur, G., Cellini, J.M., Lencinas, M.V., 2004. Forty years

of silvicultural management in southern Nothofagus pumilio primary forests.

For. Ecol. Manage. 201, 335-347.

Geiger, R., Aron, R.H., Todhunter, P., 2003. The climate near the ground. Sixth edition.

Rowman & Littlefield Publishers, Inc., Lanham.

Gendron, F., Messier, C., Comeau, P.G., 1998. Comparison of various methods for

estimating the mean growing season percent photosynthetic photon flux density

in forest. Agric. For. Meteorol. 92, 55-70.

Gerding, V., Thiers, O., 2002. Characterization of soils of Nothofagus betuloides

(Mirb.) Blume forests, in Tierra del Fuego. Chile. Rev. Chil. Hist. Nat. 75, 819-

833.

Godley, E.J., 1960. The botany of southern Chile in relation to New Zealand and the

Subantarctic. Proceedings of the Royal Society of London, Series B 152, 457–

475.

Grant, R.H., 1997. Partitioning of biologically active radiation in plant canopies. Int. J.

Biometeorol. 40, 26-40.


173

Gray, A.N., Spies, T.A., Easter, M.J., 2002. Microclimatic and soil moisture responses

to gap formation in coastal Douglas-fir forests. Can. J. For. Res. 32, 332-343.

Grubb, P.J., 1977. The maintenance of species-richness in plant communities: the

importance of the regeneration niche. Biol. Rev. 52, 107-145.

Gusinde, M., 1944. Urmenschen im Feuerland: vom Forscher zum Stammesmitglied.

Paul Zsolnay, Berlin, Germany.

Gutiérrez, A.G., Armesto, J.J., Aravena, J,C., 2004. Disturbance and regeneration

dynamics of an old-growth North Patagonian rain forest in Chiloé Island, Chile.

J. Ecol. 92, 598-608.

Gutiérrez, E., 1994. Els boscos de Nothofagus de la Terra del Foc com a paradigma de

dinàmica successional del no-equilibri. Treballs de la SCB 45, 93-121.

Gutiérrez, E., Vallejo V. R., Romaña, J., Fons, J., 1991. The subantarctic Nothofagus

forests of Tierra del Fuego: distribution, structure and production. Oecol. Aquat.

10, 351-366.

Hale, S.E., 2001. Light regime beneath Sitka spruce plantations in northern Britain:

preliminary results. For. Ecol. Manage. 151, 61-66.

Hale, S.E., 2003. The effect of thinning intensity on the below-canopy light

environment in a Sitka spruce plantation. For. Ecol. Manage. 179, 341-349.

Hale, S.E., Edwards, C., 2002. Comparison of film and digital hemispherical

photography across a wide range of canopy densities. Agric. For. Meteorol. 112,

51-56.

Helliwell, D.R., 1997. Dauerwald. Forestry 70, 375-379.

Hardy, J.P., Melloh, R., Koenig, G., Marks, D., Winstral, A., Pomeroy, J.W., Link, T.,

2004. Solar radiation transmission through conifer canopies. Agric. For.

Meteorol. 126, 257-270.

Heinemann, K., Kitzberger, T., Veblen, T.T., 2000. Influences of gap

microheterogeneity on the regeneration of Nothofagus pumilio in a xeric old-

growth forest of northwestern Patagonia. Argentina. Can. J. For. Res. 30, 25-31

Herbert, T.J., 1987. Area projection of fish-eye photographic lenses. Agric. For.

Meteorol. 39, 215-223.

Hernández, L., Castellanos, H., 2006. Crecimiento diamétrico arbóreo en bosques de

Sierra de Lema, Guyana Venezolana: primeras evaluaciones. Interciencia 31,

779-786.


174

Hildebrand-Vogel, R., Vogel, A., 1995. Evergreen broad-leaved forests of South

America. In: Box, E.O. (Ed.), Vegetation science in forestry. Kluwer Academic

Publishers, Dordrecht, pp. 125-140.

Hildebrand-Vogel, R., Godoy, R., Vogel, A., 1990. Subantarctic-Andean Nothofagus

pumilio forests. Vegetatio 89, 55-68.

Hill, R.S., Dettmann, M.E., 1996. Origin and diversification of the genus Nothofagus.

In: Veblen, T.T., Hill, R.S., Read, J. (Eds.), The ecology and biogeography of

Nothofagus forests. Yale University Press, New Haven, USA, pp. 11-24.

Hill, R.S., Jordan, G.J., 1993. The evolutionary history of Nothofagus (Nothofagaceae).

Aust. Syst. Bot. 6, 111–126.

Holst, T., Mayer, H., 2005. Radiation components of beech stands in Southwest

Germany. Meteorol. Z. 14, 107-115

Holst, T., Hauser, S., Kirchgäßner, A., Matzarakis, A., Mayer, H., Schindler, D., 2004.

Measuring and modelling plant area index in beech stands. Int. J. Biometeorol.

48, 192-201.

Holst, T., Rost, J., Mayer, H., 2005. Net radiation balance for two forested slopes on

opposite sides of a valley. Int. J. Biometeorol. 49, 275-284.

Hutchison, B.A., Matt, D.R., 1977. The distribution of solar radiation within a

deciduous forest. Ecol. Monogr. 47, 185-207.

Ibarra, M., Caldentey, J., Hidalgo, F., 2007. Antecedentes generales de la especie. In:

Cruz, G., Caldentey, J. (Eds.), Caracterización, silvicultura y uso de los bosques


Facultad de Ciencias Forestales, Universidad de Chile, Santiago, Chile, pp. 3-11.

Inoue, A., Yamamoto, K., Mizoue, N., Kawahara, Y., 2004a. Effects of image quality,

size and camera type on forest light environment estimates using digital

hemispherical photography. Agric. For. Meteorol. 126, 89-97.

Inoue, A., Yamamoto, K., Mizoue, N., Kawahara, Y., 2004b. Calibrating view angle

and lens distortion of the Nikon fish-eye converter FC-E8. J. For. Res. 9, 177-

181.

Ishida, M., 2004. Automatic thresholding for digital hemispherical photography. Can. J.

For. Res. 34, 2208-2216.

Jennings, S.B., Brown, N.D., Sheil, D., 1999. Assessing forest canopies and understorey

illumination: canopy closure, canopy cover and other measures. Forestry 72, 59-

73.


175

Jonckheere, I., Fleck, S., Nackaerts, K., Muys, B., Coppin, P., Weiss, M., Baret, F.,

2004. Review of methods for in situ leaf area index determination. Part I.

Theories, sensors and hemispherical photography. Agric. For. Meteorol. 121, 19-

35.

Jonckheere, I., Nackaerts, K., Muys, B., Coppin, P., 2005. Assessment of automatic gap

fraction estimation of forests from digital hemispherical photography. Agric.

For. Meteorol. 132, 96-114.

Jordan, G.J., Hill, R.S., 1999. The phylogenetic affinities of Nothofagus

(Nothofagaceae) leaf fossils based on combined molecular and morphological

data. Int. J. Plant Sci. 160, 1177-1188.

June, S.R., Ogden, J., 1975. Studies on the vegetation of Mount Colenso, New Zealand.

3. The population dynamics of Red Beech seedlings. Proc. New Zeal. Ecol. Soc.

22, 61-66.

June, S.R., Ogden, J., 1978. Studies on the vegetation of Mount Colenso, New Zealand.

4. An assessment of the processes of canopy maintenance and regeneration

strategy in a red beech (Nothofagus fusca) forest. New Zeal. J. Ecol. 1, 7-15.

Karukinka, 2007. Karukinka is a new model for the conservation of biodiversity.

Available from: http://www.karukinkanatural.cl/Karukinka/index-english.htm

(accessed 16.12.07).

Klepeis, P., Laris, P., 2006. Contesting sustainable development in Tierra del Fuego.

Geoforum 37, 505-518.

Kuuluvainen, T., 2002. Disturbance dynamics in boreal forests: defining the ecological

basis of restoration and management of biodiversity. Silva Fenn. 36, 5-11.

Lieberman, M., Lieberman, D., Peralta, R., 1989. Forests are not just swiss cheese:

canopy stereogeometry of non-gaps in tropical forests. Ecology 70, 550-552.

Lieffers, V.J., Messier, C., Stadt, K.J., Gendron, F., Comeau, P.G., 1999. Predicting and

managing light in the understory of boreal forests. Can. J. For. Res. 29, 796-811.

Lindenmayer, D., 2008. Perspectives on old-growth – insights from mainland eastern

Australia. Abstract in Proceedings of ‘Old Forest, New Management’ Sir Mark

Oliphant Conference, 17-21 February 2008, Hobart, Australia, p. 39.

Lindenmayer, D.B., Franklin, J.F.F., 2002. Conserving forest biodiversity. A

comprehensive multiscaled approach. Island Press, Washington DC, USA.

Lorimer, C.G., 1985. Methodological considerations in the analysis of forest

disturbance history. Can. J. For. Res. 15, 200-213.


176

Lorimer, C.G., Frelich, L.E., 1989. A methodology for estimating canopy disturbance

frequency and intensity in dense temperate forests. Can. J. For. Res. 19, 651-

663.

Luebert, F., Pliscoff, P., 2006. Sinópsis bioclimática y vegetacional de Chile. Ed

Universitaria, Santiago, Chile.

Machado, J., Reich, P.B., 1999. Evaluation of several measures of canopy openness as

predictors of photosynthetic photon flux density in deeply shaded conifer-

dominated forest understory. Can. J. For. Res. 29, 1438-1444.

Manos, P.S., 1997. Systematics of Nothofagus (Nothofagaceae) based on rDNA spacer

sequences (ITS): Taxonomic congruence with morphology and plastid

sequences. Am. J. Bot. 84, 1137–1155.

Manso, M., Ibarra, M, Nutto, L., Becker, G., 2007. Características tecnológicas y usos

de la madera. In: Cruz, G., Caldentey, J. (Eds.), Caracterización, silvicultura y

uso de los bosques de Coihue de Magallanes (Nothofagus betuloides) en la XII

Región de Chile. Facultad de Ciencias Forestales, Universidad de Chile,

Santiago, Chile, pp. 75-99.

Martínez Pastur, G., Arena, M.E., Fernández, C., 1994. Nota sobre la influencia del

ácido giberélico y del nitrato de potasio en la germinación de semillas de

Nothofagus betuloides (Mirb.) Oerst. Invest. Agrar. Sist. Recur. For. 3, 83-89.

Martínez Pastur, G., Peri, P.L., Fernández, C., Staffieri G., Rodríguez, D., 1999.

Desarrollo de la regeneración a lo largo del ciclo del manejo forestal de un

bosque de Nothofagus pumilio: 2. Incidencia del ramoneo de Lama guanicoe.

Bosque 20, 47-53.

Martínez Pastur, G., Cellini, J.M., Peri, P.L., Vukasovic, R.F., Fernández, M.C., 2000.

Timber production of Nothofagus pumilio forests by a shelterwood system in

Tierra del Fuego (Argentina). For. Ecol. Manage. 134, 153-162.

Martínez Pastur, G., Cellini, J.M., Lencinas, M.V., Vukasovic, R.F., Peri, P.L., Donoso,

S., 2002. Response of Nothofagus betuloides (Mirb.) Oerst. to different thinning

intensities in Tierra del Fuego, Argentina. Interciencia 27, 679-685.

Martinic, M., 1982. La Tierra de los Fuegos. Historia, geografía, sociedad, economía.

Artegraf Ltda., Punta Arenas, Chile.

Martinic, M., 1992. Historia de la Región Magallánica. Volúmenes I-II. Universidad de

Magallanes, Punta Arenas, Chile.


177

Mayer, H., Holst, T., Schindler, D., 2002. Mikroklima in Buchenbeständen - Teil I:

Photosynthetisch aktive Strahlung. Forstw. Cbl. 121, 301-321.

McCarthy, J., 2001. Gap dynamics of forest trees: a review with particular attention to

boreal forests. Environ. Rev. 9, 1-59.

McQueen, D.R., 1976. The ecology of Nothofagus and associated vegetation in South

America. Part I. Vegetation and soils in Southern Patagonia and Tierra del

Fuego. Tuatara 22, 38-68.

Messier, C., Doucet, R., Ruel, J.C., Claveau, Y., Kelly, C., Lechowicz, M.J., 1999.

Functional ecology of advance regeneration in relation to light in boreal forests.

Can. J. For. Res. 29, 812-823.

Mitchell, R.J., Palik, B.J., Hunter Jr., M.L., 2002. Natural disturbance as a guide to

silviculture. For. Ecol. Manage. 155, 315-317.

Monteith, J.L., Unsworth, M.H., 1990. Principles of environmental physics (2nd

edition). Edward Arnold, London.

Moore, D., 1983. Flora of Tierra del Fuego. Anthony Nelson, England, Missouri

Botanical Garden, USA.

Moore, D., Pisano, E., 1997. Biotic colonization of recently deglaciated areas in Fuego-

Patagonia – Phytogeographical considerations. Ans. Inst. Pat. 25, 21-46.

Nobis, M., 2005. SideLook 1.1. Imaging software for the analysis of vegetation

structure with true-colour photographs. Available from http://www.appleco.ch.

Nobis, M., Hunziker, U., 2005. Automatic thresholding for hemispherical canopy-

photographs based on edge detection. Agric. For. Meteorol. 128, 243-250.

Norman, J.M., Campbell, G.S., 1989. Canopy structure. In: Pearcy, R.W., Ehleringer,

J.R., Mooney, H.A., Rundel, P.W. (Eds.), Plant physiological ecology: field

methods and instrumentation. Chapman and Hall, London, pp 301-325.

Oberdorfer, E., 1960. Pflanzensoziologische Studien in Chile: Ein Vergleich mit

Europa. Flora et Vegetatio Mundi 2, 1-208.

O’Hara, K.L., Porter, D., Libby, W., 2008. The potential for uneven-aged silviculture in

restoration and management of old forests. Abstract in Proceedings of ‘Old

Forest, New Management’ Sir Mark Oliphant Conference, 17-21 February 2008,

Hobart, Australia, p. 94.

Ogden, J., 1988. Forest dynamics and stand-level dieback in New Zealand’s Nothofagus

forests. GeoJournal 17, 225-230.


178

Ogden, J., Stewart, G.H., Allen, R.B., 1996. Ecology of New Zealand Nothofagus

forests. In: Veblen, T:T., Hill, R:S., Read, J. (Eds.), The ecology and

biogeography of Nothofagus Forests. Yale University Press, New Haven, USA,

pp. 25-82.

Ojeda, V.S., Suarez, M.L., Kitzberger, T., 2007. Crown dieback events as key processes

creating cavity habitat for magellanic woodpeckers. Austral Ecol. 32, 436-445

Oliver, C.D., 1981. Forest development following major disturbances. For. Ecol.

Manage. 3, 153-168.

Oliver, C.D., Larson, B.C., 1996. Forest stand dynamics. Update edition. John Wiley &

Sons, Inc., New York, USA.

Ovington, J.D., 1983. Introduction. In: Ovington, J.D. (Ed.) Temperate broad-leaved

evergreen forests. Elsevier, Amsterdam, The Netherlands, pp. 1-4.

Palik, B.J., Mitchell, R.J., Hiers, J.K., 2002. Modeling silviculture after natural

disturbance to sustain biodiversity in the longleaf pine (Pinus palustris)

ecosystem: balancing complexity and implementation. For. Ecol. Manage. 155,

347-356.

Pardos, M., Montes, F., Aranda, I., Cañellas, I., 2007. Influence of environmental

conditions on germinant survival and diversity of Scots pine (Pinus sylvestris L.)

in central Spain. Eur. J. For. Res. 126, 37-47

Peet, R.K., Christensen, N.L., 1987. Competition and tree death. BioScience 37, 586-

595.

Pérez, V., 1983. Manual de propiedades físicas y mecánicas de maderas Chilenas.

Documento de Trabajo Nº37. Investigación y Desarrollo Forestal

(CONAF/PNUD/FAO), Santiago, Chile.

Piboule, A., Collet, C., Frochot, H., Dhôte, J.-F., 2005. Reconstructing crown shape

from stem diameter and tree position to supply light models. I. Algorithms and

comparison of light simulations. Ann. For. Sci. 62, 645-657.

Pinno, B.D., Lieffers, V.J., Stadt, K.J., 2001. Measuring and modelling the crown and

light transmission characteristics of juvenile aspen. Can. J. For. Res. 31, 1930-

1939.

Pisano, E., 1977. Fitogeografía de Tierra del Fuego - Patagonia Chilena. I Comunidades

vegetales entre las latitudes 52° y 56° S. Ans. Inst. Pat. 8, 121-250.

Pisano, E., 1978. Establecimiento de Nothofagus betuloides (Mirb.) Blume (Coigüe de

Magallanes) en un valle en proceso de desglaciación. Ans. Inst. Pat. 9, 107-128.


179

Pisano, E., 1981. Bosquejo fitogeográfico de Fuego-Patagonia. Ans. Inst. Pat. 12, 159-

171.

Pisano, E., 1997. Los bosques de Patagonia Austral y Tierra del Fuego Chilenas. Ans.

Inst. Pat., Ser. Cs. Nat. 25, 9-19.

Platt, W.J., Strong, D.R., 1989. Special feature. Gaps in forest ecology. Ecology 70, 535

Pollmann, W., Veblen, T.T., 2004. Nothofagus regeneration dynamics in South-Central

Chile: a test of a general model. Ecol. Monogr. 74, 615-634.

Promis, A., Ibarra, M., Schmidt, A., Hidalgo, F., 2007. Antecedentes dasométricos. In:

Cruz, G., Caldentey, J. (Eds.), Caracterización, silvicultura y uso de los bosques


Facultad de Ciencias Forestales, Universidad de Chile, Santiago, Chile, pp. 46-

50.

Puettmann, K.J., Ammer, C., 2007. Trends in North American and European

regeneration research under the ecosystem management paradigm. Eur. J. For.

Res. 126, 1-9.

Puigdefábregas, J., Gallart, F., Biaciotto, O., Allogia, M., del Barrio, G., 1999. Banded

vegetation patterning in a subantarctic forest of Tierra del Fuego, as an outcome

of the interaction between wind and tree growth. Acta Oecol. 20, 135-146.

Pulido, F., Díaz, B., Martínez Pastur, G., 2000. Incidencia de ramoneo del guanaco

(Lama guanicoe Müller) sobre la regeneración temprana en bosques de lenga

[Nothofagus pumilio (Poepp et Endl) Kraser]. Invest. Agr.: Sist. Recur. For. 9,

381-394.

Raedeke, K.J., 1980. Food habits of the guanaco (Lama guanicoe) of Tierra del Fuego,

Chile. Turrialba 30, 177-181.

Read, J., Brown, M.J., 1996. Ecology of Australian Nothofagus forests. In: Veblen,

T.T., Hill, R.S., Read, J. (Eds.), The ecology and biogeography of Nothofagus

Forests. Yale University Press, New Haven, USA, pp. 131-181.

Read, J., Hope, G.S., 1996. Ecology of Nothofagus forests of New Guinea and New

Caledonia. In: Veblen, T.T., Hill, R.S. Read, J. (Eds.), The ecology and

biogeography of Nothofagus forests. Yale University Press, New Haven, USA,

pp. 200-256.

Rebertus, A.J., Veblen, T.T., 1993a. Structure and tree-fall gap dynamics of old-growth

Nothofagus forests in Tierra del Fuego, Argentina. J. Veg. Sci. 4, 641-654.


180

Rebertus, A.J., Veblen, T.T., 1993b. Partial wave formation in old-growth Nothofagus

forests on Tierra del Fuego, Argentina. B. Torrey Bot. Club 120, 461-470.

Rebertus, A.J., Veblen, T.T., Kitzberger, T., 1993. Gap formation and dieback in

Fuego-Patagonian Nothofagus forests. Phytocoenologia 23, 581-599.

Rebertus, A.J., Kitzberger, T., Veblen, T.T., Roovers, L.M., 1997. Blowdown history

and landscape patterns in the Andes of Tierra del Fuego, Argentina. Ecology 78,

678-692.

Reifsnyder, W.E., Furnival, G.M., Horowitz, J.L., 1971-1972. Spatial and temporal

distribution of solar radiation beneath forest canopies. Agric. Meteorol. 9, 21-37.

Rich, P.M., 1990. Characterizing plant canopies with hemispherical photographs.

Instrumentation for studying vegetation canopies for remote sensing in optical

and thermal infrared regions. Remote Sens. Rev. 5, 13-29.

Rich, P.M., Clark, D.B., Clark, D.A., Oberbauer, S.F., 1993. Long-term study of solar

radiation regimes in a tropical wet forest using quantum sensors and

hemispherical photography. Agric. For. Meteorol. 65, 107-127.

Rich, P.M., Wood, J., Vieglais, D.A., Burek, K., Webb, N., 1999. Guide to HemiView:

Software for analysis of hemispherical photography. Delta-T Devices, Ltd.,

Cambridge, England.

Ricklefs, R.E., 1977. Environmental heterogeneity and plant species diversity: a

hypothesis. Am. Nat. 111, 376-381.

Rodríguez, R., Quezada, M., 2003. Fagaceae. In: Marticorena, C., Rodríguez, R. (Eds.),

Flora de Chile Vol. 2 (2) Berberidaceae – Betulaceae. Universidad de

Concepción, Chile, pp. 64-76.

Roig, F., Anchorena, J., Dollenz, O., Faggi, M., Méndez, E., 1985. Las comunidades

vegetales de la transecta botánica de la Patagonia Austral. In: Boelcke, O.,

Moore, M., Roig, F. (Eds.), Transecta botánica de la Patagonia Austral. Consejo

Nacional de Investigaciones Científicas y Técnicas (Argentina), Instituto de la

Patagonia (Chile), Royal Society (Gran Bretaña), Buenos Aires, Argentina, pp.

350-519.

Romanyà, J., Fons, J., Sauras-Yera, T., Gutiérrez, E., Vallejo, V.R., 2005. Soil-plant

relationships and tree distribution in old growth Nothofagus betuloides and

Nothofagus pumilio forests of Tierra del Fuego. Geoderma 124, 169-180.

Roxburgh, J.R., Kelly, D., 1995. Uses and limitations of hemispherical photography for

estimating forest light environments. New Zeal. J. Ecol. 19, 213-217.


181

Rüger, N., Gutiérrez, Á.G., Kissling, W.D., Armesto, J.J., Huth, A., 2007. Ecological

impacts of different harvesting scenarios for temperate evergreen rain forest in

southern Chile – A simulation experiment. For. Ecol. Manage. 252, 52-66.

Runkle, J.R., 1982. Patterns of disturbance in some old-growth mesic forest of eastern

North America. Ecology 63, 1533-1546.

Runkle, J.R., 1985. Disturbance regimes in temperate forests. In: Pickett S.T.A., White,

P.S. (Eds.), The ecology of natural disturbance and patch dynamics. Academic

Press, New York, USA, pp. 17-33.

Runkle, J.R., 1992. Guidelines and sample protocol for sampling forest gaps. Gen.

Tech. Rep. PNW-GTR-283. Portland, OR: U.S. Department of Agriculture,

Forest Service, Pacific Northwest Research Station, USA.

Runkle, J.R., Stewart, G.H., McClenahen, J.R., 1997. Temporal changes in height and

diameter growth for two Nothofagus species in New Zealand. J. Veg. Sci. 8,

437-446.

Sampson, D.A., Smith, F.W., 1993. Influence of canopy architecture on light

penetration in lodgepole pine (Pinus contorta var. latifolia) forests. Agric. For.

Meteorol. 64, 63-79.

Santana, A., 2006. Resumen meteorológico año 2005, estación "Jorge C. Schythe"


Santana. A., 2007. Resumen meteorológico año 2006, estación "Jorge C. Schythe"


Schabel, H.G., Palmer, S.L., 1999. The Dauerwald: its role in the restoration of natural

forests. J. Forest. 97, 20-25.

Schaetzl, R.J., Johmson, D.L., Burns, S.F., Small, T.W., 1989. Tree uprooting: review

of terminology, process, and environmental implications. Can. J. For. Res. 19,

1–11.

Schlatter, R.P., Vergara, P., 2005. Magellanic woodpecker (Campephilus magellanicus)

sap feeding and its role in the Tierra del Fuego forest bird assemblage. J.

Ornithol. 146, 188-190.

Schmithüsen, J., 1956. Die räumliche Ordnung der chilenischen Vegetation. Bonner

Geographische Abhandlungen 17, 1-86.

Schulman, E., 1956. Dendroclimatic change in semiarid America. University of Arizona

Press, Tucson. Arizona, USA.


182

Schwalbe, E., Maas, H.-G., Roscher, M., Wagner, S., 2006. Profile based sub-pixel-

classification of hemispherical images for solar radiation analysis in forest

ecosystems. Proceedings of the ISPRS Commission VII Symposium Remote

Sensing: From Pixels to Processes. International Society for Photogrammetry

and Remote Sensing (ISPRS), Enschede, the Netherlands, 8-11 May, pp 420-

425. Available from http://www.itc.nl/isprsc7/symposium2006/proceedings.aspx

Seymour, R., Hunter, M., 1999. Principles of ecological forestry. In: Hunter, M.L. (Ed.),

Maintaining biodiversity in forest ecosystems. Cambridge University Press,

Cambridge, UK, pp. 22-61.

Skottsberg, C., 1910. Botanische Ergebnisse der Schwedischen Expedition nach

Patagonien und dem Feuerlande 1907-1909. I. Übersicht über die wichtigsten

Pflanzenformationen Südamerikas S. von 41°, ihre geographische Verbreitung

und Beziehungen zum Klima. Kungl. Svenska Vetenskapsakademiens

Handlingar 46(3), 1-28.

SMN. Secretaría de Minería de la Nación de la República Argentina. 2007. <

http://www.mineria.gov.ar/ambiente/estudios/irn/tierradelfuego/tablametypluvio.

asp?pr=m12>, last accessed October 2007.

Smith, D.M., Larson, B.C., Kelty, M.J., Ashton, P.M., 1997. The practice of

silviculture. Applied forest ecology. Ninth Edition. John Wiley & Sons, Inc.,

New York, USA.

Smith, W.R., Somers, G.L., 1993. A system for estimating direct and diffuse

photosynthetically active radiation from hemispherical photographs. Comput.

Electron. Agric. 8, 181-193.

Sokal, R.R., Rohlf, F.J., 2000. Biometry. The principles and practice of statistics in

biological research. Third Edition. W.H. Freeman and Company, New York,

USA.

Sonohat, G., Balandier, P., Ruchaud, F., 2004. Predicting solar radiation transmittance

in the understory of even-aged coniferous stands in temperate forests. Ann. For.

Sci. 61, 629-641.

Sousa, W.P., 1984. The role of disturbance in natural communities. Ann. Rev. Ecol.

Syst. 15, 353-391.

Spies, T.A., 2004. Ecological concepts and diversity of old-growth forests. J. Forest.

102 , 14-20.


183

Spies, T.A., Franklin, J.F., Klopsch, M., 1990. Canopy gaps in Douglas-fir forests of the

Cascade Mountains. Can. J. For. Res. 20, 649-658.

Squeo, F., Cabrera, M., 1995. Ecofisiología vegetal. Comité Científico Proyecto Río

Cóndor. Estudio de Línea Base. Informe del Subproyecto 94-13.

Stancioiu, P.T., O’Hara, K.L., 2006. Regeneration growth in different light

environments of mixed species, multiaged, mountainous forests of Romania.

Eur. J. For. Res. 125, 151-162.

Steven, M.D., Unsworth, M.H., 1980. The angular distribution and interception of

diffuse solar radiation below overcast skies. Quart. J. Roy. Met. Soc. 106, 57-61.

Stewart, G.H., 1986. Forest dynamics and disturbance in a beech/hardwood forest.

Fiorland, New Zealand. Vegetatio 68, 115-126.

Stewart, G.H., Rose, A.B., Veblen, T.T., 1991. Forest development in canopy gaps in

old-growth beech (Nothofagus) forests, New Zealand. J. Veg. Sci. 2, 679-690.

Stokes, M., Smiley, T., 1968. An introduction to tree ring dating. University of Chicago

Press. Chicago, USA.

Swenson, U., Hill, R.S., McLoughlin, S., 2001. Biogeography of Nothofagus supports

the sequence of Gondwana break-up. Taxon 50, 1025-1041.

Tappeiner, J.C., Lavender, D., Walstad, J., Curtis, R.O., DeBell, D.S., 1997.

Silvicultural systems and regeneration methods: current practices and new

alternatives. In: Kohm, K.A., Franklin, J.F. (Eds.), Creating a forestry for the

21st century. The science of ecosystem management. Island Press, Washington,

D.C., USA.

ter Steege, H., 1996. Winphot 5.0: A programme to analyze vegetation indices, light and

light quality from hemispherical photographs. Tropenbos Guyana Reports 95-2,

Tropenbos Guyana Programme, Georgetown Guyana.

Thiers, O., Gerding, V., 2007. Variabilidad topográfica y edáfica en bosques de

Nothofagus betuloides (Mirb) Blume, en el suroeste de Tierra del Fuego. Chile.

Rev. Chil. Hist. Nat. 80, 201-211.

Tilman, D., 1988. Plant strategies and the dynamics and structure of plant communities.

Princeton University Press, Princeton, New Jersey, USA.

Tuhkanen, S.I., 1992. The climate of Tierra del Fuego from a vegetation geographical

point of view and its ecoclimatic counterparts elsewhere. Acta Botanici Fennici

145, 1-64.


184

Tuhkanen, S., Kuokka, I., Hyvönen, J., Stenroos, S., Niemelä, J., 1989-1990. Tierra del

Fuego as a target for biogeographical research in the past and present. Ans. Inst.

Pat., Ser. Cs. Nat. 19, 5-107.

Turner, M.G., Baker, W.L., Peterson, C.L., Peet, R.K., 1998. Factors influencing

succession: Lessons from large, infrequent natural disturbances. Ecosystems 1,

511-523.

Vairo, C., 1997. Los Yamana. Nuestra única tradición naval marítima autóctona. Zagier

& Urruty, Buenos Aires, Argentina.

Vales, D.J., Bunnell, F.L., 1988. Relationships between transmission of solar radiation

and coniferous forest stand characteristics. Agric. For. Meteorol. 43, 201-223

Van Pelt, R., Franklin, J.F., 2000. Influence of canopy structure on the understorey

environment in tall, old-growth, conifer forests. Can. J. For. Res. 30, 1231-1245.

Van Royen, P., 1983. The alpine flora of New Guinea 4. Cramer, Vaduz.

Veblen, T.T., 1979. Structure and dynamics of Nothofagus forest near timberline in

South-Central Chile. Ecology 60, 937-945.

Veblen, T.T., 1985. Forest development in tree-fall gaps in the temperate rainforests of

Chile. National Geographic Research 1, 162-183.

Veblen, T.T., 1989. Nothofagus regeneration in tree-fall gaps in northern Patagonia.

Can. J. For. Res. 19, 365-371.

Veblen, T.T., 1992. Regeneration dynamics. In: Glenn-Lewin, D.G., Peet, R.K., Veblen,

T.T. (Eds.), Plant succession. Chapman and Hall, London, UK, pp. 152-187.

Veblen, T.T., Veblen, A.T., Schlegel, F.M., 1979. Understorey patterns in mixed

evergreen-deciduous Nothofagus forests in Chile. J. Ecol. 67, 809-823.

Veblen, T.T., Schlegel, F.M., Escobar, B., 1980. Structure and dynamics of old-growth

Nothofagus forests in the Valdivian Andes, Chile. J. Ecol. 68, 1-31.

Veblen, T.T., Schlegel, F.M., Oltremari, J.V., 1983. Temperate broad-leaved evergreen

forests of South America. In: Ovington, J.D. (Ed.), Temperate broad-leaved

evergreen forests. Elsevier, Amsterdam, The Netherlands, pp. 5-31.

Veblen, T.T., Donoso, C., Kitzberger, T., Rebertus, A.J., 1996. Ecology of southern

Chilean and Argentinean Nothofagus forests. In: Veblen, T.T., Hill, R.S., Read,

J. (Eds.), The ecology and biogeography of Nothofagus forests. Yale University

Press, New Haven, Connecticut, USA, pp. 293-353.

Vergara, P., Schlatter, R.P., 2004. Magellanic woodpecker (Campephilus magellanicus)

abundance and foraging in Tierra del Fuego, Chile. J. Ornithol. 145, 343-351.


185

Walter, J.-M.N., 1989-2006. CIMES. A package of programs for the assessment of

canopy geometry through hemispherical photographs. Manual. Université Louis

Pasteur, Strasbourg, France.

Walter, J.-M.N., Torquebiau, E.F., 2000. The computation of forest leaf area index on

slope using fish-eye sensors. C.R. Acad. Sci. Paris, Life Sciences 323, 801-813.

Walter, J.-M.N., Fournier, R.A., Soudani, K., Meyer, E., 2003. Integrating clumping

effects in forest canopy structure: an assessment through hemispherical

photographs. Can. J. Remote Sens. 29, 388-410.

Wardle, J.A., 1984. The New Zealand beeches. Ecology, utilization and management.

New Zealand Forest Service, Christchurch, New Zealand.

Watt, A.S., 1947. Pattern and process in the plant community. J. Ecol. 35, 1-22

WCS, 2007. Wildlife Conservation Society. Available from: http://www.wcs.org/sw-

our_mission/institute (accessed 16.12.07).

Weiss, S.B., 2000. Vertical and temporal distribution of insolation in gaps in an old-

growth coniferous forest. Can. J. For. Res. 30, 1953-1964.

Welles, J.M., Norman, J.M., 1991. Instrument for indirect measurement of canopy

architecture. Agron. J. 83, 818-825.

White, P.S., 1979. Pattern, process, and natural disturbance in vegetation. Bot. Rev. 45,

229-299.

White, P.S., Jentsch, A., 2001. The search for generality in studies of disturbance and

ecosystem dynamics. In: Esser, K., Lüttge, U., Kadereit, J.W., Beyschlag, W.

(Eds.), Progress in botany Vol. 62, Berlin, Heidelberg, New York (Springer), pp.

399-449.

White, P.S., Harrod, J., Romme, H., Betancourt, J., 1999. Disturbance and temporal

dynamics. In: Szaro, R.C., Johnson, N.C., Sexton, W.T., Malk, A.J. (Eds.),

Ecological stewardship. A common reference for ecosystem management. Vol.

II. Elsevier, Amsterdam, The Netherlands, pp. 281-312.

Wood, J., 2001. Canopy LAI calculation. HemiView application note

Wright, E.F., Coates, K.D., Canham, C.D., Bartemucci, P., 1998. Species variability in

growth response to light across climatic regions in northwestern British

Columbia. Can. J. For. Res. 28, 871-886.

Young, S., 1972. Subantarctic rain forest of Magallanic Chile: distribution,

composition, and age and growth rate studies of common forest trees. In: Llano,


186

G.A. (Ed.), Antarctic terrestrial biology. Antarctic Research Series Vol. 20.

American Geophysical Union, Washington DC, USA, pp. 307-322.

Documents

Natural Small-Scale Disturbances and Below-Canopy Solar