Journal of Plant Research

, Volume 125, Issue 4, pp 539–546

The advancing timberline on Mt. Fuji: natural recovery or climate change?


    • Sado Station, Field Center for Sustainable Agriculture and Forestry, Faculty of AgricultureNiigata University
  • Takehiro Masuzawa
    • Department of Biology, Faculty of ScienceShizuoka University
Regular Paper

DOI: 10.1007/s10265-011-0465-3

Cite this article as:
Sakio, H. & Masuzawa, T. J Plant Res (2012) 125: 539. doi:10.1007/s10265-011-0465-3


The alpine timberline on Mt. Fuji (central Japan) is at 2,400–2,500 m above sea level. Over a 21-year period (1978–1999), we tracked changes in this vegetation boundary on a transect at a site impacted by the 1707 volcanic eruption. The timberline advanced rapidly upwards during this time period. Dominant tree species at the timberline (Alnus maximowiczii, Salix reinii, and Larix kaempferi) colonized sites that were initially largely free of vegetation at higher altitudes. Seedlings of L. kaempferi were particularly abundant at the border of advancing vegetation. According to tree age, we found that this was the first canopy species in the colonized areas. L. kaempferi is drought resistant, and this probably contributes to its establishment capability in the high-altitude climate. Most seedlings of Abies veitchii invaded patches of herbs and shrubs. These vegetation patches in the upper kampfzone provide important shelter for seedlings of invading tree species. We predict that the upward advance of the alpine timberline is a recovery process following the volcanic eruption, and that climate change may accelerate this advance.


Abies veitchiiAge structureAlpineClimate changeLarix kaempferiSeedling establishment


The alpine timberline is a vegetation boundary marking the forest limit on high mountains. The zone around the alpine timberline is termed the “kampfzone,” and trees struggle to survive there (Tranquillini 1979). In this zone, the timberline migrates upwards and downwards over time. The main plant-limiting factors at the timberline are low air and soil temperatures, carbon limitation, frost damage, winter desiccation, wind, and snow (Holtmeier 2009). Ecophysiological investigations in this harsh climate contribute greatly to our knowledge of plant adaptations to the environment, especially morphological adaptations.

It is thought that global warming effects will first become evident in polar and high altitude ecosystems (Grabherr et al. 1994; Kullman 2001; Sanz-Elorza et al. 2003; Sturm et al. 2001; Wardle and Coleman 1992). Grabherr et al. (1994) demonstrated significant ecological impacts of global warming in the upwards advance of alpine-nival flora. Hence, long-term ecological research on alpine timberline dynamics is likely to provide integrated warning signals of climate change.

Mt. Fuji (central Japan) has an alpine timberline ecotone (kampfzone) that has been studied intensively in recent decades (Maruta 1996; Maruta and Masuyama 2009; Masuzawa 1985; Masuzawa and Suzuki 1991; Sakio and Masuzawa 1987, 1992; Tanaka et al. 2008; Yura 1988). The dynamics of the system have been investigated by aerial photography and vegetation research (Maruta and Masuyama 2009; Tanaka et al. 2008); however, there are very few long-term ecological data on the changes in vegetation, and there is especially limited information on the establishment and growth of plants around the timberline. We investigated changes in forest dynamics at the Mt. Fuji timberline near Hoei crater by analyzing forest structure and the establishment of tree seedlings over a period of 21 years. Specifically, we addressed the following questions:
  1. 1.

    Has the alpine timberline of Mt. Fuji advanced over time?

  2. 2.

    What is the mechanism by which the timberline advances?


Materials and methods

Study site

Mt. Fuji (3,776 m) is the highest mountain in Japan. Most of the slopes are covered with vegetation up to ca. 2,500 m. Our study site (35°21′N, 138°45′E) is located at the timberline (ca. 2,400 m) on the southeastern slope, where the forest is recovering from damage caused in the year 1707 by the most recent volcanic eruption of Hoei-Zan, a parasitic crater (Fig. 1). The timberline vegetation here comprises the following deciduous dwarf trees: Alnus maximowiczii Call., Salix reinii Franch. et Savat., Larix kaempferi Carriere, and Betula ermanii Cham. Above the timberline, there are stands of perennial herbs. Down-slope from the dwarf tree vegetation, the forest composition changes to coniferous evergreen trees, among which Abies veitchii Lindley and Picea jezoensis var. hondoensis Rehder are dominant (Fig. 2; Masuzawa 1985).
Fig. 1

Study site. Location on the southeastern slope of Mt. Fuji near the Hoei Crater (central Japan)
Fig. 2

Mt. Fuji timberline. Image shows many patches of perennial herbs at the upper krummholz limit. Dominant species are Polygonum weyrichii var. alpinum, Carex doenitzii, Hedysararum vicioides, and Arabis serrata

The timberline climate on Mt. Fuji is cold, very windy, and with little snow cover (ca. 30 cm in depth from November to February). Annual mean air temperature is 1.1°C, with the highest and lowest monthly means of 11.8°C in August and −9.5°C in February (Masuzawa 1985). Annual precipitation is about 4,500 mm (Ito, 1964). Precipitation levels are high year-round, especially during the summer growing season. Relative humidity is high from May to October, and particularly high from June to September (mean >80%) when afternoons are frequently foggy (Masuzawa 1985).

The surface substratum consists of basalt scoria from the volcanic eruptions of Hoei in 1707. This scoria is easily moved by the freeze–thaw cycle and by strong wind; the ground surface is very unstable (Oka 1980). The nitrogen and carbon content of the soil is very low, 0.02 and 0.3%, respectively, at the upper timberline (Masuzawa 1985).


In August 1978, we established a 220-m-long permanent belt-transect (10 m wide) from the upper timberline zone down-slope into the coniferous evergreen forest (Fig. 3). The transect comprised 22 contiguous plots (10 × 10 m). All living trees (≥130 cm tall) were numbered and identified to species; diameter at breast height (DBH; diameter at 130 cm above ground level) and total height measurements were recorded in uppermost plots at 0–130 m along the transect, and in lower plots at 180 and 220 m (Masuzawa 1985). For dwarf bushes such as A. maximowiczii and S. reinii, the longest stems of individual plants were selected for these measurements. Basal area was calculated from the DBHs of all plants. Twenty-one years later (in 1999), we repeated the measurements.
Fig. 3

Forest profile for the Mt. Fuji timberline. Gray canopy indicates evergreen coniferous trees (Abies veitchii and Picea jezoensis var. hondoensis)

Increment cores were taken with a borer from three canopy tree species, i.e., L. kaempferi, A. veitchii, and P. jezoensis var. hondoensis. The largest tree (measured in DBH) of each species in each plot was selected to estimate its age. The increment borer was screwed into trunks about 40 cm above ground level, as close as possible to the substratum. Tree age was estimated from the sum of the number of annual rings in each core and the mean age of 40-cm-tall saplings. When the borer missed a tree center, the number of annual rings in the missing part was extrapolated from mean radial growth data.

We recorded seedling establishment in the eight uppermost plots (1–8). Data were used to investigate the upwards advance of vegetation. The number, diameter, and height of new seedlings established between 1978 and 1999 was measured in each plot in 1999. We recorded the shortest distance between each seedling and the edge of vegetation patches containing dwarf trees and herbs.

The advance rate of the alpine timberline before 1978 was estimated from the age of the largest L. kampferi tree in each plot because this is the dominant species of the timberline. The colonized age of this tree on each plot was estimated from the relation between largest age and distance. The advance rate was calculated from the difference in age between plot 5 and plot 22. The rate from 1978 to 1999 was estimated from the newly established seedlings over 130 cm in height.


Change in forest structure over time

The timberline vegetation comprised three dwarf tree species, viz., A. maximowiczii, S. reinii and L. kaempferi. All of these species were found to have colonized upwards by more than 20 m during the 21 years between observations (Fig. 4). The increase in the number of L. kaempferi trees was especially marked in upslope plots 3–6. S. reinii numbers declined in mid-transect plots (8–12), and this species was absent from plots 11 and 12. Numbers of A. maximowicziii and L. kaempferi did not change in the down-slope plots, except that the former disappeared from plots 18 and 22.
Fig. 4

Distribution of tree numbers across the slope gradient between 1978 and 1999. White circles and black circles show tree numbers in 1978 and 1999, respectively

The average heights of L. kaempferi and P. jezoensis var. hondoensis increased over time in all plots (Fig. 5). The largest height increases for S. reinii and A. maximowiczii were recorded in plots 8 and 9, respectively. In the case of L. kaempferi, the large height increases were found below plot 8. The sum of total basal area (BA) for all species increased over the 21-year study period, except in plot 22 (Fig. 6). The total basal area of S. reinii and A. maximowiczii increased at elevations above plot 7 at the timberline and decreased below plot 8. Moreover, increases in these two species in plot 7 were 6.7- and 2.1-fold, respectively. The BA of L. kaempferi increased 3.6-, 1.8-, and 1.4-fold in plot 8, 9, and 10, respectively.
Fig. 5

Distribution of tree heights across the slope gradient between 1978 and 1999. White circles and black circles show tree heights in 1978 and 1999, respectively
Fig. 6

Distribution of tree basal areas across the slope gradient between 1978 and 1999. White circles and black circles show tree basal areas in 1978 and 1999, respectively

Age structure of dominant tree species

The ages of L. kaempferi, A. veitchii, and P. jezoensis var. hondoensis individuals increased over time towards the down-slope end of the transect (Fig. 7). There were significant positive relationships between tree age and distance to barren land for all three species (P < 0.01). A maximum age of 208 years was recorded for L. kaempferi in plot 16, and this was the oldest species in many plots. A. veitchii trees were the youngest among the three species (t test, P < 0.01).
Fig. 7

Tree ages of three main timberline species across the slope gradient. For significant linear regressions (P < 0.05), slope and R2 are provided. Broken line, two dot chain line and solid line show Larix, Picea and Abies, respectively

Establishment of seedlings at the upper timberline

L. kaempferi colonized the upper area of the timberline, with 196 newly established seedlings between 1978 and 1999 (Table 1). These seedlings were distributed widely throughout the timberline, but most had colonized plots 4–6 (Fig. 8). Maximum seedling height was 325 cm (in plot 5). This tallest individual grew 15.5 cm year−1 on average. There were fewer and smaller newly established seedlings of A. veitchii and P. jezoensis var. hondoensis than L. kaempferi. Most A. veitchii seedlings invaded vegetation patches at the timberline (plot 7), whereas few P. jezoensis seedlings became established within this part of the transect.
Table 1

Number and height of Larix seedlings established from 1978 to 1999


No. of seedlings

Seedling height (cm)







12 ± 8




32 ± 38




52 ± 60




74 ± 79




77 ± 74




44 ± 23




11 ± 9

Fig. 8

Number of seedlings of three tree species that newly invaded and established between 1978 and 1999

Establishment sites (Table 2) differed significantly among the three woody species (Kruskal–Wallis test, P < 0.01). A. veitchii invaded pre-existing vegetation patches, and seedlings were on average 43.1 cm distant from the vegetation patch margin. In contrast, L. kaempferi seedlings established very close to vegetation patch edges.
Table 2

Distance from the edge of vegetation patch


No. of seedlings

Distance from patch (cm)

Larix kaempferi



Abies veitchii



Picea jezoensis var. hondoensis



− shows that seedlings is in the plant patch

Advance rate of the alpine timberline

The colonized ages of L. kaempferi in each plot were estimated from the relationship between the highest age and distance (Fig. 7). Based on these data, the advance rate of the alpine timberline before 1978 was 7.6 m per 10 years in plots 5–22. The rate from 1978 to 1999 was estimated from newly established seedlings over 130 cm in height, and was about 10 m in 10 years (Figs. 4, 8). There was no clear difference in the advance rate of the alpine timberline before or after 1978. On the other hand, there were significant differences in tree shape of L. kaempferi between newly established seedlings after 1978 and individuals already established in 1978. Figure 9 shows the positive relationship between diameter and height in new seedlings (P < 0.0001). In addition, tree heights also had a positive relation to seedling age (P < 0.0001). Trees established before 1978 were table-shaped, where new seedlings were not. However, there was no significant difference between diameter and height of individuals from before 1978.
Fig. 9

Relationship between tree diameter and height. Black circles and solid line are established seedlings in the upper kampfzone (plots 2–7) from 1978 to 1999. White triangles are trees present in 1978 in the upper kampfzone (plots 5–7)


The vegetation on the mountain was displaced downwards by a volcanic eruption by Hoei-Zan in 1707, and our study site was bare ground 300 years ago. On Mt. Fuji, the timberline on the western slope that escaped the eruption in 1707 is at 2,800 m asl, with a limit at 2,900 m (Oka 1992). Hence, it is expected that vegetation near the study site will progress toward the same altitude as the timberline on the western slope. It might be argued that the recent advance upwards in our study site is part of a recovery process following the eruption. Indeed, the alpine timberline on the southeastern slope has advanced upwards for about 200 years (Fig. 7). Maruta and Masuyama (2009) also showed the advance of the timberline using aerial photography, and reached the conclusion that the cause of the advance of the timberline ecotone on the southern slope affected by the eruption of Mt. Fuji is natural recovery through succession. In this study site, the advance rate of L. kaempferi trees was 7.6 m per 10 years before 1978, while a rough rate estimate from seedling establishment after 1979 gave a figure of about 10 m in 10 years (Figs. 4, 8). Hence, the advance rate has not changed significantly.

However, the shape of L. kaempferi trees at our study site have changed. Maruta and Masuyama (2009) classified the form of this species on the south slope of Mt. Fuji into five categories. Table-shaped trees (formed by the continuous death of main shoots) occurred only in the upper kampfzone, whereas erect trees with symmetrical branches occurred in down-slope sections of the timberline. These changes in tree shape may represent the natural course of primary succession. Akasaka and Tsuyuzaki (2005) showed that stunted and branched stems with higher root allocation in L. kaempferi is an adaptation to bare ground for the effective acquisition of light, water, nutrients, and high tolerance to wind. At our study site 21 years ago, most L. kaempferi trees were table-shaped in the upper kampfzone. Most trees present in 1978 had reached a ceiling as table-shaped trees (Fig. 9). On the other hand, trees that colonized after 1978 were erect from the start, having retained their main stems, and table shapes were absent (Fig. 9). Slatyer (1976) demonstrated that low temperature kills Eucalyptus pauciflora at the timberline in the Snowy Mountains of Australia. Tranquillini (1979) argued, however, that tree death in winter at the timberline is attributable to desiccation stress. On Mt. Fuji, the primary factor causing winter desiccation damage and krummholz formation in timberline larch (L. kaempferi) is abrasion by fine, wind-blown volcanic gravel (Maruta 1996). It is expected that bark abrasion will be reduced and high plant water content will be maintained in mild winters, as found by Maruta (1996) for shoots of krummholz larches in the winter of 1985–1986. Measurements from the meteorological station at the summit of Mt. Fuji show that there has been a reduction in wind velocity during recent winters (Japan Meteorological Agency 2011).

Increasing air temperature and CO2 concentration will increase photosynthetic rate and growth period, and may also decrease desiccation stress. Such climate change is correlated with upward vegetation advances in many mountain ranges, e.g., the European Alps (Grabherr et al. 1994). From 1933 to 2010, the mean air temperature at the summit observatory on Mt. Fuji has gradually increased (Fig. 10), and the rate of increase has significantly accelerated since the 1980s. There is not a significant positive relationship between year and temperature before 1978 (P > 0.43), but there is one after 1979 (P < 0.01). The mean CO2 concentration at the summit of Mt. Fuji was about 335 ppm in August 1981 (Nakazawa et al. 1984), and had increased to about 388 ppm by August 2010 (Sunaga et al. 2011). Curtis and Wang (1998) suggested that elevated CO2 might enhance net assimilation rate and tree growth. Yazaki et al. (2004) found that assimilation in L. kampferi became saturated at an intercellular CO2 concentration of 600 ppm, regardless of mineral nutrient supply. Climate changes could affect photosynthesis of L. kaempferi trees, leading to increased growth rates and changes in shape. Such increases may accelerate upwards movement of the alpine timberline on Mt. Fuji during the natural recovery process after volcanic eruptions.
Fig. 10

Mean air temperature from 1933 to 2010 at the summit observatory on Mt. Fuji. White squares and the solid line show temperatures after 1979. Black circles and the dotted line show temperatures before 1978 (Japan Meteorological Agency 2011)

The dominant high-altitude trees on Mt. Fuji (A. maximowiczii, S. reinii, and L. kaempferi) colonized upper sections of the timberline over our 21-year study period (Figs. 4, 8). A. maximowiczii and S. reinii are pioneer shrubs with many stems, and they tend to disappear from forests with tall trees due to shading. A. maximowiczii dwarf forest is highly productive (Sakio and Masuzawa 1987) and has a high nitrogen content in leaves (Sakio and Masuzawa 1992); therefore, this forest type could play an important role in nitrogen supply to soil below this forest. L. kaempferi was a particularly active dominant tall tree in this zone; tree age analysis (Fig. 7) and patterns of seedling establishment (Fig. 8) demonstrated that L. kaempferi was the first colonizer among canopy tree species to the high altitude zones, while A. veitchii was the last colonizer. S. reinii is particularly dominant in vegetation patches below 50 cm in height in the herb layer of upper sections of the timberline, and may provide important shelter for A. veitchii seedlings. L. kaempferi is drought resistant and is able to colonize very dry landscape areas. A. veitchii seedlings invaded existing herb and shrub patches, where climatic conditions are moderate. These vegetation patches in the upper kampfzone play important roles in tree seedling invasion, indicating that facilitative interactions among alpine plants increase with stress (Callaway et al. 2002). Yura (1988, 1989) suggested that L. kaempferi seedlings can avoid desiccation by extending roots deep into the soil faster than A. veitchii seedlings. This may help explain why L. kaempferi is able to become established in dry barren ground outside herb/shrub patches.

In conclusion, the timberline of Mt. Fuji at our study site will continue to advance upward in the future as a natural recovery process from the volcanic eruption of 1707. Although the influence of climate change on the advance of the timberline was not clear in this research apart from a change in tree shape, climate-related changes such as increases in air temperature and CO2 concentration may accelerate the advance of the timberline in the future. Therefore, long-term ecological monitoring is needed to elucidate the dynamics of the timberline ecotone on Mt. Fuji in relation to climate change.


The authors are indebted to Dr. F. Konta for his advice, and to members of the Laboratory of Plant Ecology, Shizuoka University, for their kind assistance during field work. A part of this investigation was financed by a Grant-in-Aid for Scientific Research (B) (No. 19310008) from the Ministry of Education, Culture, Sports, Science and Technology. We would also like to thank anonymous reviewers and an editor for very constructive comments and suggestions.

Copyright information

© The Botanical Society of Japan and Springer 2011