B.Sc. – thesis May 2018

The influence of plant size and fertilization on catkin production in

birch (Betula pubescens Ehrh.)

Julia C. Bos

Faculty of Natural resources and Environmental

Sciences

ii

B.Sc. – thesis May 2018

The influence of plant size and fertilization on catkin production in birch (Betula pubescens

Ehrh.)

Julia C. Bos

Academic advisor: Ása L. Aradóttir

Agricultural University of Iceland Faculty of Nature and Environmental Sciences

iii

Statement by author I hereby declare that the writing of the following thesis is based on my own

observations, it is my own work and has never, in whole or part, been submitted for a

higher degree.

________________________________

Julia C. Bos

iv

Abstract

Downy birch Betula pubescens Ehrh. is the only tall growing native birch species in

Iceland. Around the time of the settlement, birch woodlands used to cover large areas

of Iceland, but today only little is left. Regeneration of birch is a key process in

woodland expansion and is important for land reclamation and restoration. Success of

natural regeneration depends largely on seed production and seed establishment.

Therefore, information about factors influencing these processes is of value.

Plant seed production is affected by a number of factors, including nutrient availability,

moisture supply, weather conditions, plant density and predation. The main objective

of this project is to assess the effects of plant size and fertilization on seed production,

using data from experiments conducted at two research sites in North and South Iceland

in 1992-1994: Hálsmelar and Gunnlaugsskógur.

A positive correlation was found between plant size and catkin production. Youngest

catkin producing plants were on average 32 to 97 centimetres long and their estimated

age was 6 to 16 years. Data further suggested that small size of plants might be a

limiting factor in reproductive maturity as percentage of catkin bearing plants increased

with plant size.

High and low fertilizer treatments increased catkin production in Gunnlaugsskógur, but

not in Hálsmelar. However, fertilization did increase plant growth at both sites and

could therefore possibly increase catkin production over a longer period of time.

Key words: Birch (Betula pubescens Ehrh.), catkin production, plant size, fertilization.

v

Acknowledgements

I would like to thank my advisor Ása L. Aradóttir for her guidance during this study

and for giving me the opportunity to work with the data used in this study.

Further, I would like to thank Ása L. Aradóttir, Járngerður Grétarsdóttir and others

involved in setting up the experiments and collecting the data in 1992-1994.

Last but not least, I would like to thank my father Tjalling Bos and my sister Naomi D.

Bos for all their support.

vi

Contents

1. Introduction ............................................................................................................ 1

2. Materials and methods ........................................................................................... 3

2.1. Species description .......................................................................................... 3

2.2. Site description ................................................................................................ 4

2.2. Experimental design and measurements ......................................................... 5

2.3. Data analysis ................................................................................................... 5

3. Results .................................................................................................................... 7

3.1. Relationship between age and size .................................................................. 7

3.2. Relationship between age/length and plants bearing female catkins .............. 8

3.4. Relationship between age/length and female catkin production ..................... 9

3.5 Effect of fertilizer treatment on female catkin production and growth ......... 10

4. Discussion ............................................................................................................ 13

4.1. Relationship between birch plant size/age and catkin production ................ 13

4.2. Effect of fertilization on catkin production ................................................... 14

4.3. Influence of weather factors and climate change on birch seed production . 15

5. Conclusions .......................................................................................................... 17

6. References ............................................................................................................ 18

1

1. Introduction

Downy birch Betula pubescens Ehrh. is the only tall growing native birch species in

Iceland. Before the first settlement of Iceland, around 874 AD, according to estimates

approximately 27% of the total land area of Iceland was covered by birch forests (Snorri

Sigurðsson, 1977). However, much of this has been lost, largely due to cutting for

firewood and overgrazing by sheep. Today, birch forests only cover approximately

1.5% (1506 km2) of the land (Arnor Snorrason et al., 2016). The distribution of these

forests is quite unequal. Most are found in West Iceland and only a few in East Iceland.

Since 1989, the birch forest cover has increased by 130 km2 or 9%. Approximately

41500 km2 of Iceland lies below the tree line, but only little is covered by trees. This

shows a potential for further natural expansion of birch forests.

Regeneration of birch in Iceland is important as it allows restoration of woodlands and

reclamation and rehabilitation of degraded land (Aradóttir, 2007). This can help reduce

soil erosion and restore ecosystem function and biodiversity and increase carbon

sequestration. Restoration can be promoted by planting seedlings or by direct seeding.

These plants can then serve as seed sources for further colonization and increase natural

regeneration (Aradóttir & Halldórsson, 2018). Both seed production and plant

establishment are important for the natural regeneration success of birch and the

restoration of birch woodland (Fenner & Thompson, 2005). An increased

understanding of influential factors of birch seed production could therefore be very

useful. Plant seed production is affected by abiotic factors such as nutrient availability,

moisture supply and weather conditions, and biotic factors such as plant density and

predation (Fenner & Thompson, 2005). These factors are also likely to influence birch

seed production.

Large annual fluctuations in pollen catch and seed set have been described for birch

(Sarvas, 1952; Ranta et al., 2008). The seed supply is irregular due to the varying seed

production each year and the relatively short-lived nature of the seeds (Atkinson, 1992),

but annual amounts of male and female catkins are positively correlated (Sarvas, 1952;

Masaka & Maguchi, 2001). Flowering depends mostly on winter chilling which is

required to break dormancy and is largely forced by spring temperatures (Linkosalo,

2000; Emberlin et al., 2002). However, catkin production is also influenced by weather

2

variables the year before flowering since that is when the buds are formed (Ranta et al.,

2008).

Other factors that may influence plant seed production are plant age and size (Thomas,

2011), although not many recent studies have examined the relationship of age, size

and birch seed production. Nutrient availability can also influence seed production,

which means that fertilization might enhance seed production (Wang et al., 2011).

Masting, the large fluctuations in annual seed production, is a characteristic of trees in

the Betulaceae family (Sarvas, 1952; 1955). Different hypotheses for this phenomenon

have been suggested and tested. The predator satiation hypothesis explains the variation

in seed production as an adaptation against seed predators; there is an inverse

correlation between the number of produced seeds and the percentage of seeds

consumed by seed predators (Crone & Rapp, 2014). The wind pollination hypothesis

assumes an increase in the efficiency of pollination in years with large seed production

(Smith, Hamrick & Kramer, 1990). The weather-tracking hypothesis suggests that

weather conditions, such as temperature and sunlight hours, cause this variation in seed

production, however weather conditions alone cannot account for masting behaviour

(Masaka & Maguchi, 2001). Another hypothesis is the resource budget model which

suggests that masting is regulated by weather factors together with the system of

resource allocation among years (Ranta, Oksanen, Hokkanen, Bondestam & Heino,

2005).

The aim of this study is to assess the influence of different factors on birch seed

production. The research questions are:

- At what age and size do birches start to flower and produce seeds?

- What is the relationship between catkin production and birch plant size and age?

- Can catkin production in birch be stimulated with fertilization?

3

2. Materials and methods

2.1. Species description

Betula pubescens Ehrh. is a species of the Betula genus in the Betulaceae family. As

described by Atkinson (1992), this genus consists of 52 species, 24 varieties, and 8

natural interspecific hybrids. However, the taxonomy of this genus is quite complex,

partly due to the hybridization between species. For example, hybrids exist between the

diploid Betula pendula, the diploid Betula nana, and the tetraploid B. pubescens, which

are all European species (Truong, Palmé & Felber, 2007). Hybridization between B.

nana and B. pubescens occurs in Iceland (Anamthawat-Jónsson & Thórsson, 2003).

Species of the Betula genus are distributed throughout the northern temperate region

(Atkinson, 1992). The wide distribution of birch is largely the result of their ability to

colonize bare areas and to grow on nutrient-poor soils. However, they do not establish

well when there is vegetation, and they are quite intolerant to shade, extreme low

temperatures, and drying winds (Anderson, Cooke, Elkington & Read, 1966). There are

differences in distribution between species. For example, B. pubescens extends further

to the north and the east than B. pendula, while B. pendula extends further south in

Europe and Asia (Atkinson, 1992).

Birch trees are monoecious. Individual plants can produce both male and female

flowers. The flowers are wind-pollinated in catkin inflorescences (Atkinson, 1992).

Male catkins are located at long shoots, and female catkins at short shoots. The buds of

both female and male catkins are formed during the summer of the year before

flowering. However, male catkins already become visible before winter, while female

catkins overwinter as primordia and emerge at bud burst in spring. Flowering takes

place during May-June in Iceland (Hörður Kristinsson, 2012). Female flowers usually

become receptive one day before male flowers on the same tree shed pollen (Atkinson,

1992). Both self-fertilization as well as out-crossing can take place. But, as Perala &

Alm (1990) reviewed, seed set is much better for flowers on the same tree if fertilized

by out-crossing rather than by self-fertilization. After fertilization has taken place, male

catkins fall off the trees and female catkins develop over the summer and small winged

seeds are formed. Seed fall occurs in autumn and winter. The seeds are wind dispersed,

but usually do not travel further than twice the tree height, although there have been

occasional exceptions of dispersal over many kilometres (Holm, 1994). Seed

germination takes place during the following spring or summer.

4

2.2. Site description

This study used data collected in 1992-1994 at two sites: Gunnlaugsskógur in

Rangárvellir in South-Iceland (63°52’N 20°12’W, 100-110 m a.s.l.) and Hálsmelar in

Fnjóskadalur in North-Iceland (65°44’N 17°54’W, 80-100 m a.s.l.) as a part of a larger

project ‘Birch Ecology: Seed Production and Seed Dispersal’ (Ása L. Aradóttir,

personal communication).

Gunnlaugsskógur is a small woodland established by seeding of birch in a protected,

fenced off reclamation area in 1939 and 1945 and later by transplantation of the seeded

plants into a nearby lavafield (Aradóttir & Arnalds, 2001). Dispersal from the initial

birch stands subsequently led to much establishment of birch plants, resulting in a

patchy dispersal pattern, with cores of old birch plants being surrounded by a band of

younger plants (Aradóttir, Robertson & Moore, 1997; Aradóttir & Arnalds, 2001). The

plots used at this site were situated on rather poor, shallow soils and included many

birch plants of different age and size (Aradóttir, 1991). Mean temperature at the nearest

weather station, Hella, for the study years is presented in Table 1.

Hálsmelar in Fnjóskadalur valley is a site with remnants of an old birch forest that

expanded through natural regeneration in the previous decades. Several centuries ago,

almost the entire valley used to be forested, however, part of these woodlands has been

destroyed due to unsustainable land use, mostly agricultural practises, which has led to

soil erosion and bare hill sides (Bjarnason, 1980). The plots in this research area

included many birch plants of different age and size and were situated on poor eroded

soils on a slope to the north and north-east (Ása L. Aradóttir, personal communication).

Gravel and only little vegetation was found on top and to sides of this slope. Mean

temperatures of the nearest weather station, Lerkihlíð, for the study years is presented

in Table 1.

Both research areas were protected from grazing by sheep.

Table 1. Mean temperatures (°C) in Hella, South Iceland (63°50’ N 20°22' W), and Lerkihlíð, North

Iceland (65°43' N 17°53’ W), in 1991-1994 (Source of data: Icelandic Meteorological Office, 2007).

1991 1992 1993 1994 1991 1992 1993 1994

Mar. - May. 3 3 3 2 2 1 2 1

Jun. - Aug. 12 9 10 10 7 9 8 10

Sep. - Nov. 3 3 5 4 2 2 4 1

Dec. - Feb. 0 0 -3 -2 -1 -1 -4 -3

Hella, South Iceland Lerkhlíð, North Iceland

5

2.2. Experimental design and measurements

At each site, three blocks with each three plots were used, a total of 9 plots. Plots were

100 m2 (33.3 m x 3 m) in Gunnlaugsskógur and 50 m2 (16.7 m x 3 m) in Hálsmelar.

This was because of a higher plant density in Hálsmelar than in Gunnlaugsskógur.

About 600 plants were marked at either site.

For each plant, its length and basal diameter were measured and number of female

catkins was counted in the spring of 1992. This was repeated in 1993 and 1994. Height

was also measured in 1993 and 1994.

The three plots in each block received a different fertilizer treatment in June 1993;

control (no fertilizer), low or high fertilization level (2.5 g/m2 and 7.5 g/m2 of 20% N -

4,3 % P - 8,2 % K - 2 % S - 4% Ca respectively).

To estimate plant age, 50 plants of various sizes were selected at each site for age

determination and establishment of the relationship between size and age. In 1992, the

length, height, basal diameter, and diameter perpendicular to the basal diameter were

measured for every plant, after which the plants were cut at soil surface level. The

lowermost 1-2 cm section of each stem was cut out, placed in a plastic bag, and stored

in a cooled storage until further handling. The sections or slices were put in a 50:50

mixture of isopropanol and ethanol for a few days and then dried at room temperature.

Subsequently, the slices were polished with rough and fine sandpaper until the surface

was smooth and the age of the trees was determined by counting the annual rings under

a microscope (Ása L. Aradóttir, personal communication).

2.3. Data analysis

The relationship between the age and size of birch plants was determined by a linear

regression of annual ring count to length, height, basal diameter, and diameter

perpendicular to basal diameter separately for each site. For each site, the parameter

giving the highest R2 was used to estimate age of all the birch plants at that site in 1992

and subsequently the plants were divided into the following age categories: 0-5, 6-10,

7-15, 15-20, 21-25, and >25 years. Age of birch plants in 1993 and 1994 was estimated

by adding respectively one and two years to the estimated age in 1992. Plants in a

certain age category in 1992 could therefore be in a different category in 1993 or 1994.

Plots were used as a base unit. The relationship between plant size or age and the

number of plants bearing female catkins was examined by counting the total number as

6

well as the percentage of birch plants bearing female catkins in spring in all three years

at each site.

To assess onset of reproductive stage, the mean, median and minimum age of the

youngest plant with female catkins in each plot was summarized for each year.

The relationship between size or estimated age and number of female catkins per plant

was analysed by linear regressions for each site in 1992 and 1994, after log-

transforming the catkin number data (natural logarithm). Only plants with catkins were

included in these regressions. The catkin data from 1993 were not analysed in the same

way because of the small number of birch plants bearing female catkins that year.

The effect of fertilizer treatment on the change in female catkin production from 1992

to 1994 was studied by conducting an ANOVA for each site. First, catkin number data

were transformed by taking the natural logarithm. Then, the mean increase in catkin

production was calculated for each plot and used as the dependent variable in the

ANOVA with block and fertilizer treatment as classification variables. The means of

fertilizer treatments within a site were compared with Tukey’s HSD.

Effect of fertilizer treatment on growth of plants from 1993 to 1994 was studied by

conducting an ANOVA for each site. The mean increase in length was calculated for

each plot and used as the dependent variable in the ANOVA with fertilizer treatment as

classification variable. The means of fertilizer treatments within a site were compared

with Tukey’s HSD.

Linear regression and ANOVA were done with JMP®, Version 13.1.0. SAS Institute

Inc., Cary, NC, 1989-2007.

7

3. Results

3.1. Relationship between age and size

Linear regressions showed a significant positive relationship between age and all size

parameters for the birch plant samples taken in 1992 (Figure 1), where length showed

the strongest correlation with age (Table 2). Therefore, length was subsequently used

to estimate the age of all the plants in the study plots in 1992.

The regression lines were steeper for Hálsmelar than Gunnlaugsskógur reflecting that

similar sized plants were generally older in Hálsmelar than in Gunnlaugsskógur (Figure

1).

Figure 1. Relationship between various size parameters and plant age determined by counting of annual

rings of about 50 birch plants from Gunnlaugsskógur (G) and Hálsmelar (H) in 1992. Regression lines

are shown for each site. Regression results are presented in Table 2.

8

Table 2. Results of regressions between various plant size parameters and age, showing R2, P-values,

and formulas of regression lines, and number of birch plants (N) for Gunnlaugsskógur and Hálsmelar.

The age of all the plants on each site was estimated by using the prediction formula for

length for that site (Table 2). Table 3 shows the distribution of plants among age

categories.

Table 3. Number of plants and their length range of each estimated age category for Gunnlaugsskógur

and Hálsmelar in 1992. NA is number of plants without estimated length and age.

3.2. Relationship between age/length and plants bearing female catkins

The percentage of plants bearing female catkins increased with age, but it also

fluctuated among years (Table 4). Percentage of plants with catkins was higher in

Gunnlaugsskógur than in Hálsmelar for all age categories.

Table 4. Total number and percentage of plants in each age category with flowering female catkins for

Gunnlaugsskógur and Hálsmelar in 1992, 1993 and 1994.

Site Est. age N % N % N %

Gunnlaugsskógur 0-5 2 1 0 0 - -

6-10 95 26 6 1 77 19

11-15 49 74 16 14 122 71

16-20 4 100 3 50 13 93

21-25 - - - - - -

>25 - - - - - -

Hálsmelar 0-5 - - - - - -

6-10 1 0.4 1 0.6 0 0

11-15 6 4 5 3 7 4

16-20 14 13 11 10 25 20

21-25 30 56 20 30 40 49

>25 4 80 7 70 13 81

Number of plants with catkins

1992 1993 1994

Site 0-5 6-10 11-15 16-20 21-25 >25 NA Total

Gunnlaugsskógur Nr. of plants 169 360 66 4 0 0 7 606

Length (cm) 10-26 27-85 86-138 >138 - - - -

Hálsmelar Nr. of plants 0 263 154 106 54 5 11 593

Length (cm) - 10-47 48-89 90-132 133-174 >174 - -

Estimated age categories

R2

P-value Formula N R2

P-value Formula N

Length 0.78 <.0001 Age = 3.21+0.09*Length 47 0.67 <.0001 Age = 4.96+0.12*Length 50

Height 0.66 <.0001 Age = 3.60+0.09*Heigth 49 0.54 <.0001 Age = 6.46+0.12*Height 50

Basal diameter (D1) 0.74 <.0001 Age = 2.86+0.44*D1 50 0.58 <.0001 Age = 6.10+0.58*D1 50

Diameter perpendicular

to D1 (D2)0.73 <.0001 Age = 2.62+0.48*D2 50 0.63 <.0001 Age = 5.29+0.68*D2 50

Gunnlaugsskógur Hálsmelar

9

Age of the youngest birch plants bearing female catkins was on average 6 to 10 years

in Gunnlaugsskógur and on average 16 years in Hálsmelar (Table 5). Length of the

youngest birch plants bearing female catkins was on average 86 to 97 cm in

Gunnlaugsskógur and on average 32 to 74 cm in Hálsmelar (Table 5).

Table 5. Mean, median, minimum age and length of youngest birch plants bearing flowering female

catkins for Gunnlaugsskógur and Hálsmelar in 1992, 1993 and 1994.

3.4. Relationship between age/length and female catkin production

Linear regressions showed a significant positive relationship between length and catkin

production in 1992, and between length and catkin production and age and catkin

production in 1994 (Figure 2 and 3 respectively). However, the correlations were not

very strong, as was indicated by a low R2 in all cases (Table 6).

The regression lines were steeper for Gunnlaugsskógur than Hálsmelar reflecting that

plants of similar size and age generally produced more catkins in Gunnlaugsskógur than

in Hálsmelar (Figure 2 and 3).

Figure 2. Relationship between length of birch plants bearing female catkins and log number of

flowering female catkins in 1992. Regression lines are shown for each site (Gunnlaugsskógur (G) and

Hálsmelar (H)). Regression results are presented in Table 6.

Site Mean Median Min. Mean Median Min. Mean Median Min. Mean Median Min. Mean Median Min. Mean Median Min.

Gunnlaugsskógur 6 6 4 94 103 10 10 10 8 86 93 50 8 8 6 97 89 37

Hálsmelar 16 17 9 32 34 33 16 16 9 74 63 31 16 15 14 50 45 67

Length (cm) Length (cm)

1994

Est. age Est. age

1992 1993

Est. ageLength (cm)

10

Figure 3. Relationship between length (left) and estimated age (right) of birch plants bearing female

catkins and log number of flowering female catkins. Regression lines are shown for each site

(Gunnlaugsskógur (G) and Hálsmelar (H)) in 1994. Regression results are presented in Table 6.

Table 6. Results of regressions between length of birch plants bearing female catkins and log number of

flowering female catkins in 1992 and length or age of birch plants bearing female catkins and log number

of flowering catkins in 1994, showing R2, P-values, formulas for regression lines, and number of birch

plants included (N) for Gunnlaugsskógur and Hálsmelar.

3.5 Effect of fertilizer treatment on female catkin production and growth

The effect of fertilizer treatments on the increase in female catkin production between

1992 and 1994 was significant in Gunnlaugsskógur but not in Hálsmelar (Table 7,

Figure 4). In Gunnlaugsskógur, the average increase in number of catkins was

significantly higher for the low and high fertilizer treatments than the control, but there

was no significant difference in average increase in number of catkins between the two

fertilizer treatments.

The effect of fertilizer treatments on birch growth (increase in length) between 1993

(year of fertilization) and 1994 was significant for both sites (Table 7, Figure 5). In

Gunnlaugsskógur, average growth was significantly higher for the high fertilizer

treatment than the control and the low fertilizer treatment, while the low fertilizer

treatment showed no significant difference in average growth from the control. In

Year R2

P-value Formula N R2

P-value Formula N

1992 Length 0.21 <.0001 Log nr. of catk.= 0.18 + 0.03 * Length 150 0.25 <.0001 Log nr. of catk.= -1.25 + 0.02 * Length 55

Length 0.20 <.0001 Log nr. of catk.= 0.29 + 0.23* Length 215 0.19 <.0001 Log nr. of catk.= -0.65 + 0.01 * Length 84

Est. age 0.14 <.0001 Log nr. of catk.= -0.19 + 0.22 * Est. age 215 0.24 <.0001 Log nr. of catk.= -1.83 + 0.14 * Est. age 85

Gunnlaugsskógur Hálsmelar

1994

11

Hálsmelar, average growth of both fertilizer treatments was significantly higher than

the control, but there was no significant difference in average growth between the low

and high fertilizer treatments.

Table 7. Results of ANOVA of the effect of fertilizer treatment on mean increase in catkin production

between 1992 and 1994 and growth of plants between 1993 and 1994 for Gunnlaugsskógur and

Hálsmelar, showing degrees of freedom (df), F-values and P-values.

Figure 4. Results of Tukey test of the effect of different fertilizer treatments (Control = 0 g/m2, Low =

2.5 g/m2, High = 7.5 g/m2) on increase in catkin production between 1992 and 1994 in Gunnlaugsskógar

(G) and Hálsmelar (H), showing mean increase in number of catkins per plant for each fertilizer

treatment, and significance of difference between treatments. Averages for Gunnlaugsskógur marked

with the same letter are not significantly different at =0.05; there was not significant difference among

fertilizer treatments at Hálsmelar.

Classification variable df F-value P-value F-value P-value

Fertilizer treatment 2 102 0.0004 0.71 0.547

Block 2 36.5 0.0027 0.13 0.882

Growth Fertilizer treatment 2 9.6 0.014 20 0.002

Gunnlaugsskógur Hálsmelar

Increase in catkin

production

12

Figure 5. Results of Tukey test of the effect of different fertilizer treatments (Control = 0 g/m2, Low =

2.5 g/m2, High = 7.5 g/m2) on growth of plants between 1993 and 1994 in Gunnlaugsskógar (G) and

Hálsmelar (H), showing mean increase of plant length per fertilizer treatment for each site and

significance of difference between treatments. For each site, averages marked with the same letter are

not significantly different at =0.05.

13

4. Discussion

4.1. Relationship between birch plant size/age and catkin production

This study showed that there was a positive correlation, although not very strong,

between plant size and number of female catkins. Since a positive relationship was

found between plant size and age, number of catkins also increased for estimated age.

This supports the idea that reproductive allocation increases with size and age in

iteroparous organisms as described by Thomas (2011). Plants in Gunnlaugsskógur with

similar size and age as in Hálsmelar generally produced more female catkins. The

difference in number of catkins for plants of similar age could be explained by the

difference in plant size; Plants in Hálsmelar were generally smaller than in

Gunnlaugsskógur. The difference in number of catkins of plants of similar size might

be explained by environmental factors (Perala & Alm, 1990), such as soil conditions,

or mean temperature or precipitation the year before flowering and the year of

flowering. Reduced reproductive success is possibly also connected to increased

elevation (Sveinbjörnsson, Kauhanen & Nordell, 1994). However, in this case the two

sites were located at similar elevations and therefore this was likely not a contributing

factor.

Previous studies on different plant species show that plant size can influence

reproductive success (Aarssen & Taylor, 1992; Ollerton & Lack, 1998; Weiner,

Campbell, Pino & Echarte, 2009) and some even suspect that it is a more important

responsible factor of variation in plant fecundity than age (Herrera, 1991). According

to Wareing (1959) and Brinkman (1974), birch plants can flower for the first time at an

age of 5 to 15 years, but recent studies on this subject are lacking. This study showed

that onset of reproductive stage took place in plants of on average 32-97 centimetres in

length and 6-16 years old. Onset of reproductive stage generally took place at higher

age in Hálsmelar than in Gunnlaugsskógur, likely due to slower growth of the plants.

This could mean that the smaller size of the plants limits the number of plants bearing

female catkins in Hálsmelar. This would support the idea by Longman and Wareing

(1959) and Wareing (1959) that plant size is important for the transition from vegetative

to reproductive stage in birch. Robinson and Wareing (1969) studied this change from

juvenile to adult phase in woody species, including Betula verrucose, and suggest that

although the change is correlated with a certain size (phase change takes place after the

14

meristems have undergone a certain number of cell divisions), it is not the primary

factor determining the stage at which it occurs.

For both sites, the percentage of plants bearing female catkins was considerably higher

for older age categories. The same pattern was described by Aradóttir (1991). The

higher percentage of catkin-bearing plants in older categories and the positive

correlation between size of plants and number of female catkins showed that larger and

older plants are more likely to produce catkins and to produce more catkins than

younger ones.

4.2. Effect of fertilization on catkin production

Effect of fertilizer treatments on birch seed production showed that both the low and

the high fertilizer treatment increased mean female catkin production significantly

compared to the control in Gunnlaugsskógur. This is in accordance with several other

studies that show increased seed production as a result of fertilization in other plant

species (Nams, Folkard & Smith, 1993; Sperens, 1997; Drenovsky & Richards, 2005;

Karlsson, 2006; Kristín Svavarsdóttir & Ása L. Aradóttir, 2006). Fertilizer treatment

did not show a significant effect on catkin production in Hálsmelar. This might be due

to for example different soil conditions such as soil moisture, nutrient uptake, or

weather conditions limiting catkin production and possibly hiding potential differences

in fertilizer treatments.

Even though catkin production was not significantly increased by fertilization in

Hálsmelar, both low and high fertilizer treatments in Hálsmelar increased growth

significantly compared to control. In Gunnlaugsskógur, the high fertilizer treatment

significantly increased growth compared to control. Since a positive correlation was

found between size and catkin production, this increase in growth might accelerate time

to reproductive maturity and increase seed production, as was also suggested by

Aradóttir (1991) and described by a study on the effect of NPK fertilization on

flowering in Betula platyphylla Suk. (Wang et al., 2011).

A possible explanation for the increase in growth but not in catkin production in

Hálsmelar might be that although growth was increased by fertilization, it might take

some time before plants in Hálsmelar reach a certain size, which plants in

Gunnlaugsskógur already had reached, at which they are reproductive mature or start

to produce significantly more catkins.

15

Since fertilization did directly or indirectly increase catkin production, this method

could be used to promote regeneration of birch.

4.3. Influence of weather factors and climate change on birch seed production

Observed differences in catkin production between Gunnlaugsskógur and Hálsmelar

could (partly) be explained by weather factors such as precipitation and temperature;

the mean temperatures in summer and winter were lower in Hálsmelar than in

Gunnlaugsskógur in 1991-1994 (Icelandic Meteorological Office, 2007). Ranta et al.

(2008) also found in their study that there is a connection between the number of catkins

and sum of sunlight hours in June, which explains part of the annual variation of catkin

numbers. It might be possible that sum of sunlight hours in spring not only explain

annual variation but also perhaps variation between sites. However, other factors such

as soil factors or genetic variation might also contribute to the observed differences

between the two sites.

Beside the differences between the two sites, differences were also found between

years. In 1993, catkin production was generally lower and there were fewer plants

bearing female catkins, especially for Gunnlaugsskógur, when compared to 1992 and

1994. This pattern was found for both sites and showed synchronization of female

catkin production of different populations over a large geographic area. Such

synchronization was also found by other studies for male catkins (Ranta et al., 2008;

Yasaka et al., 2009). The lower catkin production in 1993 could be due to weather

affecting the amount of flowering; the summers of 1992 and 1993 were relatively cold,

especially in Gunnlaugsskógur. However, weather variables generally show less

variation than seed production and may therefore not explain year-to-year variation

alone (Koenig & Knops, 2000). Several studies suggest that masting might be regulated

by weather factors in combination with a system of resource allocation between years

(Masaka, & Maguchi, 2001; Ranta et al., 2005; Crone & Rapp, 2014). However, other

influential or explanatory factors such as seed predation also are suggested in other

studies. This study did not focus on the influential factors of the fluctuation of seed

production between years and therefore could not completely support or reject any of

these hypotheses.

The results of this study were based on data from 1992 to 1994, however, weather

factors and their effects on birch catkin production could change as a result of climate

change. It has been predicted that annual mean temperatures in the lowlands in Iceland

16

are likely to increase 1.5-2.5 °C in the 21st century (Halldór Björnsson et al., 2008). The

positive relationships between size and catkin production are not likely to change, but

the age at which plants start to produce catkins might. A study on climatic effects on

birch growth in North Iceland, in the same valley as one of the sites in this study, shows

that above-average temperatures in summer, especially in June, can increase size of

plants (Levanič & Eggertsson, 2008). Such increase in plant growth could mean that

plants reach a certain size at a younger age and therefore might start to produce catkins

and larger number of catkins earlier in the future than they did in this study. Nakamura

et al. (2016) also found that especially tall Betula ermanii Cham. trees (18-20 m)

allocate extra resources to bud and flower production, rather than to plant growth, as a

response to increased aboveground temperatures. This shows that response might be

size-dependent.

17

5. Conclusions

Birch catkin production increased with size and age. However, environmental

conditions could cause differences in catkin production for plants of the same size and

age between sites.

Onset of reproductive stage was on average at a size of 32-97 centimetres and an age

of 6-16 years. Age at onset of reproductive stage varied between sites, which could be

explained by the differences in size of plants.

High and low fertilizer treatments increased catkin production in Gunnlaugsskógur, but

not in Hálsmelar. However, fertilization did increase plant growth at both sites and

could therefore possibly increase catkin production over a longer period of time.

Fertilization could therefore be used to promote birch regeneration success in birch

forest restoration and land reclamation.

Further, data suggested that increase in growth, which could be caused by higher

temperatures due to global warming, might accelerate reproductive maturity and

increase seed production. Therefore, it would be good to study this under current

climatic conditions or in relation to difference in temperature. Such possible increase

in seed production due to climatic warming could promote birch regeneration and

restoration of birch forests in Iceland. However, other factors also influence birch

regeneration. For example, it should be noted that there was no grazing by sheep in the

research areas. Therefore, connecting this study to other factors affecting seed

production, viability, dispersal and establishment could be valuable for better

understanding of the factors controlling birch regeneration.

18

6. References

Aarssen, L.W. & Taylor, D.R. (1992). Fecundity allocation in herbaceous plants. Oikos,

65, 225-232.

Anamthawat-Jónsson, K. & Thórsson, A.T. (2003). Natural hybridisation in birch:

triploid hybrids between Betula nana and B. pubescens. Plant Cell, Tissue and

Organ Culture, 75(2), 99-107.

Anderson, D.J., Cooke, R.C., Elkington, T.T. & Read, D.J. (1966). Studies on structure

in plant communities. II. The structure of some dwarf-heath and birch-copse

communities in Skjaldfannardalur, North-West Iceland. Journal of Ecology, 54,

781-793.

Aradóttir, Á.L. (1991). Population biology and stand development of birch (Betula

pubescens Ehrh.) on disturbed sites in Iceland. Ph.D. Dissertation, Texas A&M

University, Texas.

Aradóttir, Á.L. (2007). Restoration of birch and willow woodland on eroded areas. In

Halldórsson, G., Oddsdóttir, E.S. & Eggertsson, Ó. (Eds.), Effects of

afforestation on ecosystems, landscape and rural development (pp. 67-74).

Copenhagen: Nordic Council of Ministers.

Aradottir, Á.L. & Arnalds, Ó. (2001). Ecosystem degradation and restoration of

birch woodlands in Iceland. In Wielgolaski, F.E. (Ed.), Nordic Mountain Birch

Ecosystems (pp. 295-308). Carnforth: UNESCO, Paris, and Parthenon

Publishing.

Aradóttir, Á.L. & Halldórsson, G. (2018). Colonization of woodland species

during restoration: seed or safe site limitation? Restoration Ecology, early view,

https://doi.org/10.1111/rec.12645.

Aradottir, Á.L., Robertson, A. & Moore, E. (1997). Circular statistical analysis of

birch colonization and the directional growth response of birch and black

cottonwood in south Iceland. Agricultural and Forest Meteorology, 84, 179-

186.

Arnor Snorrason, et al. (2016). Náttúrulegt birki á Íslandi - Ný úttekt á útbreiðslu

þess og ástandi [The natural birch woodland in Icelands — a new assessment of

distribution and state]. Náttúrufræðingurinn 86, 97-111.

Atkinson, M.D. (1992). Betula Pendula Roth. (B. Verrucosa Ehrh.) and B. Pubescens

Ehrh. Journal of Ecology, 80(4), 837-870.

19

Bjarnason, Á.H. (1980). The history of woodland in Fnjóskadalur. Acta

Phytogeographica Suecica, 68, 31-42.

Brinkman, K.A. (1974). Betula L. Birch. In Schopmeier, C.S. (Ed.), Seeds of woody

plants of the United States. USDA Agricultural Handbook 450 (pp. 252-257).

Washington, DC.: U.S. Department of Agriculture, Forest Service.

Crone, E.E. & Rapp, J.M. (2014). Resource depletion, pollen coupling, and the ecology

of mast seeding. Annals of the New York Academy of Sciences, 1322, 21-34.

Drenovsky, R.E. & Richards, J.H. (2005). Nitrogen addition increases fecundity in the

desert shrub Sarcobatus vermiculatus. Oecologia, 143(3), 349-356.

Emberlin, J., Detandt, M., Gehrig, R., Jaeger, S., Nolard, N. & Rantio-Lehtimäki, A.

(2002). Responses in the start of Betula (birch) pollen seasons to recent changes

in spring temperatures across Europe. International Journal of Biometeorology,

46, 159-170.

Fenner, M. & Thompson, K. (2005). The Ecology of Seeds. Cambridge: Cambridge

University Press.

Halldór Björnsson et al. (2008). Hnattrænar loftslagsbreytingar og áhrif þeirra á

Íslandi. Skýrsla vísindanefndar um loftslagsbreytingar. Reykjavík: Ministry for

the Environment.

Herrera, C.M. (1991). Dissecting factors responsible for individual variation in plant

fecundity. Ecology, 72(4), 1436-1448.

Holm, S.O. (1994). Reproductive patterns of Betula pendula and B. pubescens coll.

along a regional altitudinal gradient in northern Sweden. Ecography, 17, 60-72.

Hörður Kristinsson (2012). Íslenska plöntuhandbókin. Reykjavík: Mál og menning.

Icelandic Meteorological Office (2007). Tímaraðir fyrir valdar veðurstöðvar. Visited

28. February 2018 at http://www.vedur.is/vedur/vedurfar/medaltalstoflur.

Karlsson, C. (2006). Fertilization and release cutting increase seed production and stem

diameter growth in Pinus sylvestris seed trees. Scandinavian Journal of Forest

Research, 21, 317-326.

Koenig, W.D. & Knops, J.M.H. (2000). Patterns of annual seed production by

northern hemisphere trees: A global perspective. The American Naturalist,

155(1), 59-69.

Kristín Svavarsdóttir & Ása L. Aradóttir (2006). Áhrif áburðar á stærð, blómgun of

fræframleiðdlu gulvíðis og loðvíðis. In Kristín Svavarsdóttir (Ed.), Innlendar

20

víðitegundir: Líffræði og notkunarmöguleikar í landgræðslu (pp. 33-41).

Gunnarsholt: Landgræðsla ríkisins.

Levanič, T. & Eggertsson, O. (2008). Climatic effects on birch (Betula pubescens

Ehrh.) growth in Fnjoskadalur valley, northern Iceland. Dendrochronologia, 25,

135-143.

Linkosalo, T. (2000). Analyses of the spring phenology of boreal trees and its response

to climate change. Ph.D. dissertation. University of Helsinki, Finland.

Longman, K.A. & Wareing, P.F. (1959). Early induction of flowering in birch

seedlings. Nature, 184, 2037-2038.

Masaka, K. & Maguchi, S. (2001). Modelling the masting behaviour of Betula

platyphylla var. japonica using the resource budget model. Annals of Botany,

88, 1049-1055.

Nakamura, M., Makoto, K., Tanaka, M., Inoue, T., Son, Y. & Hiura, T. (2016). Leaf

flushing and shedding, bud and flower production, and stem elongation in tall

birch trees subjected to increases in aboveground temperature. Trees, 30, 1535-

1541.

Nams, V.O., Folkard, N.F.G. & Smith, J.N.M. (1993). Effects of nitrogen fertilization

on several woody and nonwoody boreal forest species. Canadian Journal of

Botany, 71, 93-97.

Ollerton, J. & Lack, A. (1998). Relationships between flowering phenology, plant size

and reproductive success in Lotus corniculatus (Fabaceae). Plant Ecology, 139,

35-47.

Perala, D.A. & Alm, A.A. (1990). Reproductive ecology of birch: A review. Forest

Ecology and Management, 32, 1-38.

Ranta, H., Hokkanen, T., Linkosalo, T., Laukkanen, L., Bondestam, K. & Oksanen, A.

(2008). Male flowering of birch: Spatial synchronization, year-to-year variation

and relation of catkin numbers and airborne pollen counts. Forest Ecology and

Management, 255, 643-650.

Ranta, H., Oksanen, A., Hokkanen, T., Bondestam, K. & Heino, S. (2005). Masting by

Betula-species; applying the resource budget model to north European data sets.

International Journal of Biometeorology, 49, 146-151.

Robinson, L.W. & Wareing, P.F. (1969). Experiments on the juvenile-adult phase

change in some woody species. New Phytologist, 68, 67-78.

21

Sarvas, R. (1952). On the flowering of birch and the quantity of seed crop.

Communicationes Instituti Forestalis Fenniae, 40, 1-38.

Sarvas, R. (1955). Investigations into the flowering and seed quality of forest trees.

Communicationes Instituti Forestalis Fenniae, 45, 1-37.

Smith, C.C., Hamrick, J.L. & Kramer, C.L. (1990). The advantage of mast years for

wind-pollination. American Naturalist, 136, 154-166.

Snorri Sigurðsson (1977). Birki á Íslandi (útbreiðsla og ástand). In Hákon

Guðmundsson, et al. (Eds.), Skógarmál (pp. 146-172). Reykjavík: Edda.

Sperens, U. (1997). Long-term variation in, and effects of fertiliser addition on, flower,

fruit and seed production in the tree Sorbus aucuparia (Rosaceae). Ecography,

20(6), 521-534.

Sveinbjörnsson, B., Kauhanen, H. & Nordell, O. (1994). Treeline ecology of mountain

birch in the Torneträsk area. Ecological Bulletins, 45, 65-70.

Thomas, S.C. (2011). Age-related changes in tree growth and functional biology: The

role of reproduction. In Meinzer, F., Lachenbruch, B. & Dawson, T. (Eds.),

Size- and age-related changes in tree structure and function (pp. 33-64).

Dordrecht: Springer.

Truong, C., Palmé, A.E. & Felber, F. (2007). Recent invasion of the mountain birch

Betula pubescens ssp. tortuosa above the treeline due to climate change: genetic

and ecological study in northern Sweden. Journal of Evolutionary Biology,

20(1), 369-380.

Wang, S., Jiang, J., Li, T., Li, H., Wang, C., Wang, Y. & Liu, G. (2011). Influence of

nitrogen, phosphorus, and potassium fertilization on flowering and expression

of flowering-associated genes in white birch (Betula platyphylla Suk.).

Plant Molecular Biology Reporter, 29(4), 794-801.

Wareing, P.F. (1959). Problems of juvenility and flowering in trees. Journal of the

Linnean Society (Botany), 56, 282-289.

Weiner, J., Campbell, L.G., Pino, J. & Echarte, L. (2009). The allometry of

reproduction within plant populations. Journal of Ecology, 97(6), 1220-1233.

Yasaka, M., Kobayashi, S., Takeuchi, S., Tokuda, S., Takiya, M. & Ohno, Y. (2009).

Prediction of birch airborne pollen counts by examining male catkin numbers

in Hokkaido, northern Japan. Aerobiologia, 25, 111-117.