Species composition and vegetation structure of ... - Wiley Online Library

41 downloads 0 Views 3MB Size Report
montane forest at the summit of Mt. Doi Inthanon, Thailand .... the Himalayan mountain range (MacDonald et al. ...... ests of Mount Kinabalu, Sabah, Malaysia.
Nordic Journal of Botany

Species composition and vegetation structure of an upper montane forest at the summit of Mt. Doi Inthanon, Thailand Soontorn Khamyong, Anne Mette Lykke, Dusit Seramethakun and Anders S. Barfod

Khamyong, S., Lykke, A. M., Seramethakun, D. & Barfod, A. S. 2004. Species composition and vegetation structure of an upper montane forest at the summit of Mt. Doi Inthanon, Thailand. - Nord. J. Bot. 23: 83-97. Copenhagen. ISSN 0107055X. Upper montane forest (UMF) within Doi Inthanon National Park, Northern Thailand, was investigated by means of fifty, 40 x 40 m stratified random plots situated between 2080 and 2565 m altitude. The aim was to address a number of community ecological questions concerning woody species composition and structural heterogeneity of the forest. A total of 7474 individuals of trees and woody climbers 2 15 cm gbh (girth at breast height) were included in the study and these were identified to 47 species, 39 genera and 26 families. The average density was 934 individualsha and the average stem basal area was 71.8 m2ha. The most important species were: Quercus eumorpha, Sjzygium angkae, Litsea martabanica, Helicia nilagirica, Lindera caudata, Schima wallichii, Osmanthus fiagrans, Eurya acuminata, Myrsine semiserrata and Ilex umbellulata and the most important families were Fagaceae, Lauraceae, Theaceae and Myrtaceae. Altitude was the most important environmental variable explaining species composition and vegetation structure. Most of the calculated vegetation variables showed significant correlation with altitude: species richness, family richness, diversity, density and crown cover declined with altitude, average tree height was uncorrelated with altitude and basal area increased with altitude. An analysis of size class distributions indicated good forest conditions and reverse-J-shaped age class distribution of most species.

S.Khamyong, Department of Soil Science & Conservation, Chiang Mai Universiw, Thailand. - A. M. Lykke, Department of Systematic Botany, Universitetsparken, Building 137, University of Aarhus, 8000 Aarhus, Denmark. E-mail: [email protected] D. Seramethahun. Central Laboratory, Naresuan University. Thailand. - A. S. Bafled, Department of Systematic Botany, Universitetsparken, Building 137, University of Aarhus, 8000 Aarhus, Denmark.

Introduction Evergreen forests above 1000 m altitude are called “hill evergreen forest” by most Thai foresters and botanists. In the literature, the same vegetation type is referred to in various ways, e.g. temperate evergreen forest (Robbins & Smitinand 1966), montane rain forest m t m o r e 1975, Santisuk 1988), lower

montane forest (Ohsawa 1991), montane evergreen forest (Pooma & Barfod 2001), primary evergreen seasonal forest (Maxwell 81 Elliott 2001) and highland forest (Gardner et al. 2000). In Thailand, the montane forests have been further divided into lower montane forest (LMF) and upper montane forests (UMF) (Santisuk 1988). Depending on local climatoedaphic conditions, the altitudinal range of LMF is

Accepted 1-9-2004 Nord. J. Bot. 23(1)

83

Fig. 1. Upper montane forest at the summit of Mt. Doi Inthanon.

from 1000 m to 1700 or 1800 m’s altitude above which upper montane forest takes over. Santisuk (1988) made the distinction between upper montane rain forest and upper montane scrub forest. On Doi Inthanon, most of the primary forest belongs to the first subtype (Fig. 1). Where the mist belt prevails, UMFs are also called cloud forest (Hamilton et al. 1995, Werner 1995). The UMF of Doi Inthanon is defined as cloud forest by Hamilton et al. (1995), whereas Werner (2001) considers the cloud cover on Doi Inthanon too infrequent to be a true cloud forest. Upper montane forests are characterized by simple stratification, lack of emergent trees, abundance of epiphytes, few lianas and presence of a number of characteristic plant families known from montane vegetation types throughout tropical South-East Asia, e.g. Ericaceae, Fagaceae, Lauraceae, Myrtaceae, Symplocaceae and Theaceae. The trees are often of small stature and characterized by umbrellashaped crowns, small leaves, gnarled stems and branches that are covered by epiphytes such as or-

84

chids, ferns, lichens and mosses (Whitmore 1975, Singh 1987, Santisuk 1988, Ohsawa 1991, 1995, Werner 2001, Ashton 2003). UMFs play an important role in the hydrologic cycle due to their ability to prevent run-off after heavy downpours and thereby ensure continued streamflow into lower-lying areas and protection against soil erosion (Hamilton et al. 1995, Ateroff & Rada 2000). In addition, forests with frequent cloud cover augment the amount of precipitation received by the ecosystem through condensation of mist in the canopy (Vogelmann 1973, Bruijnzeel & Proctor 1995). The enhanced knowledge of these important ecological functions of upper montane forests combined with limited distribution, widespread destruction and extremely slow recovery rate of these forests, has increased the conservation interest in upper montane forests (Ewe1 1980, Hamilton et al. 1995, Werner 2001). The importance of UMF for protection of water-sheds has also become an important local issue (Isager 2001).

Nord. J. Bot. 23( I )

The upper montane forest at the summit of Doi Inthanon is unique to Thailand, but only about 5 km2 of primary montane forest remains between 2000 m and 2565 m altitude. Kiichler and Sawer (1967), Chuchip (1989) and Santisuk (1988) have conducted botanical surveys on the mountain including both LMF and UMF, which provided basic botanical information. Sri-Ngemyuang et al. (2003) made quantitative studies in LMF, whereas little is known about the quantitative floristic and structural characteristics of the UMF and how these are related to dynamic processes and underlying topographic, climatic and edaphic variables. The sampling design of the present study has been conceived to provide a thorough description of species richness, floristic composition and structural characteristics along an altitudinal gradient in UMF.

Study site The Doi Inthanon National Park is located in the Chiang Mai Province about 50 km southwest of Chiang Mai City. The national park was established in 1972 and expanded in 1978; it now covers 482.4 km2 with ranges across three districts: Chom Thong, Mae Wang and Mae Chaem (Fig. 2). Most of the park is mountainous and belongs to a southern extension of the Himalayan mountain range (MacDonald et al. 1993). The altitude range between 400 m and 2565 m

at the summit of Doi Inthanon, which is the highest point in Thailand. Five major forest types are found within the Doi Inthanon National Park: dry dipterocarp forest (DDF), mixed deciduous forest (MDF), dry evergreen forest (DEF), pine forest (PF) and montane forest (MF). The different forest types overlap in their altitudinal ranges: DDF: 400-1300 m; MDF: 400800 m; DEF: 600-1000 m; PF: 700-1700 m; and MF: 1000-2565 m. The area has a monsoonal climate with seven months of rainy season (May through November) and five months of dry season (December through April). At the summit of Mt. Doi Inthanon the average annual rainfall (1982-1999) is 2228 mm. Here the average annual air temperature is 13" C, with absolute minimum of 1" C and absolute maximum of 23" C (Doi Inthanon Military Base 2001). At Chiang Mai City situated at 330 m altitude, the average annual rainfall is 1164 mm (Chiang Mai University 1995) and at 1200 m altitude the corresponding figure is 1686 mm (Royal Project 1995). The bedrock underlying Doi Inthanon National Park forms part of a geological formation mainly composed of gneissic granite that extends discontinuously along the western mountain range of Northern Thailand (MacDonald et al. 1993). Various soil types have been recorded from Doi Inthanon such as Entisols, Inceptisols, Alfisols and Ultisols (Khamyong et al. 1996). Ultisols are especially com-

Fig. 2. Doi Inthanon National Park with location of study area and 50 vegetation plots at 2080-2565 m altitude.

Nod. J. Bot. 23( I )

85

mon within the park. In the UMF, the soils are char7 acterized by high accumulation of organic matter and extreme acidity; these soils are classified as Ultisols, suborder Humult. About 4000 people live inside the Doi Inthanon National Park in 31 villages and 624 households (Anon. 1989). These mainly belong to the Karen Hilltribe ethnic group but a few villages are populated with Mnong Hilltribe and local Thais. The most widespread agricultural practices are permanent agriculture, shifting cultivation and extractivism. The UMF zone on Doi Inthanon has escaped cultivation due to difficult access, steep slopes and cold climate.

Methods Fifty plots were established in UMF of Doi Inthanon during the period from October 1999 to August 2000, all woody plants with a girth at breast height (gbh) of 15 cm or more were identified to species. The following years (2001-2003) the plots were revisited in order to collect specimens in flower or h i t for species verification. The dimensions of the plots were 40 x 40 m, i.e. a total of 8 ha were investigated. Each plot was further divided into 16, 10 x 10 m subplots. The actual length of the plots and subplots was calculated based on a horizontal projection so that down-slope length of the plot increased with inclination. Accordingly the length of a subplot was 10 m for a inclination of 0-30%, 11 m for 30-50%, 12 m for 50-66% and 13 m for >66%. The plots were established according to a stratified random sampling design. Some deviation from a purely random sampling proved necessary from a cost-benefit point of view. To select the positions of the plots we used a topographic map (150 000 scale) on which boundaries between the major forest types were roughly indicated and Landsat TM images from 1999 where classes from an unsupervised classification was used as a simple proxy for vegetation types. The 50 plots form part of a larger sampling programme that included 300 plots from all the major forest types represented in the national park. Sampling of woody plants included measurement of gbh by plastic tape and estimation of height and crown diameter. Voucher specimens were collected for all species and kept for reference at the Department of Agriculture, Chiang Mai University. On each plot, one tree was tagged with a metal label and GPS coordinates and overall orientation of the plot were noted in order to enable relocation in future studies. Notes were taken on important environmental parameters such as altitude and soil. Five soil

86

samples were taken from the A horizon (0-5 cm) in each plot and pooled for analysis of texture, pH and organic matter. Species names follow IPNI (2004) Selected plant ecological indices were calculated to give an insight into diversity and structure of the forest. For all species, density, frequency, dominance and importance value index (IVI) were calculated according to Krebs (1994). For all families, density, diversity, dominance and family importance value (FIV) were calculated according to Mori & Boom (1983). A Principal Component Analysis (PCA) was carried out in order to identify important species associations and to identify the most important environmental variables for explaining species composition. PCA is particularly useful for vegetation data covering short gradients like the present dataset. The present dataset had a gradient length of 1.8 The PCA was carried out on a logtransformed plot-species matrix with variance/covariance as cross-product matrix (MacCune & Mefford 1999). A plot-environmental variable matrix was used as secondary matrix, with altitude and soil characteristics as environmental variables. A Spearman correlation was used to test for significant relationships between altitude and vegetation variables: species richness, family richness, diversity (Shannon Index), density, average tree height, basal area and crown cover, no. of stems 5100 cm gbh, no. of stems 100-200 cm gbh and no. of stems >200 cm gbh. Correlations were tested at anlevel 0.01. Finally, size class distributions were graphed for all species with at least eight individuals in the sample, excluding lianas. Species population trends were predicted on the basis of size class distributions of individual tree species.

Results and discussion Species composition The woody plants sampled within 50 plots each of 0.16 ha were identified to a total of 47 species, 39 genera and 26 families (Table 1). These included 26 big trees, 5 medium-sized trees, 12 small trees, 1 shrubby tree and 3 climbers (Appendix 1). Average density of all species over 15 cm gbh was 934 treesha (1018 stemsiha). The species with highest densityiha were Helicia nilagirica (108), Litsea martabanica (107), Quercus eumorpha (98), Syzygium angkae (73), Lindera caudata (66), Rapanea yunnanensis (64), Myrsine semiserrata (62), Eurya acuminata (60), Osmanthus fragrans (5 l), Ilex

Nard. J. Bal. 23( I )

Table 1. Density, frequency, dominance and importance value index (IVI) of 47 woody species in upper montane forest at Mt. Doi Inthanon. species name

Family

Density

Fre-

Domi- Relative Relative Relative denfiedomi(cm2/ha) sity quency nance

M

(no. indiv. quency nance bereus eumorpha $aygium angkae ,itsea martabanica felicia nilagirica hdera caudata ;chima wallichii kmanthus fragrans k y a acuminata 4yrsine semiserrata l a umbellulata tapanea yunnanensis 4acropanax dispermus lcer laurinum itsea beusekomii itsea sp. htanopsis acuminatissima +mplocos sumuntia ithocarpus echinops Yaeocarpus sphaericus ithocarpus aggregatus 4yrica esculenta torsjeldia glabra asminum attenuatum Jyssa javanica 'hoebepaniculata ldinandra intergerrima :ornus oblonga 'hoebe cathia asminum dispermum 'yrenaria diospyricarpa 'acciniumsprengelii 4elodinus cochinchinensis %ereus glabricupula Yamellia sp.- 1 Lmingtonia populnea hrennoidea wallichii 'runus phaeosticta kgelhardtia spicata (ithocalpusgarrettianus Lmplocos macrophylla brbus granulosa lhododendron arboreum 'odocalpus neriifolius h s sp.-3 "richillaconnaroides 'olygala arillata lphanamixispolystachya

lord. J. Bot. 23(1)

Fagaceae Myrtaceae Lauraceae Proteaceae Lauraceae Theaceae Oleaceae Theaceae Myrsinaceae Aquifoliaceae Myrsinaceae Araliaceae Aceraceae Lauraceae Lauraceae Fagaceae Symplocaceae Fagaceae Elaeocarpaceae Fagaceae Myricaceae Myristicaceae Oleaceae Nyssaceae Lauraceae Theaceae Comaceae Lauraceae Oleaceae Theaceae Ericaceae Apocynaceae Fagaceae Theaceae Hamamelidaceae Rubiaceae Rosaceae Juglandaceae Fagaceae Symplocaceae Rosaceae Ericaceae Podocarpaceae Moraceae Meliaceae Polygalaceae Meliaceae

/ha)

(%)

98.25 72.75 107.13 108.38 66.00 19.63 51.00 59.50 61.75 34.00 63.63 22.75 30.50 17.63 14.50 9.63 16.00 5.38 9.25 5.38 6.00 5.75 3.63 4.88 3.00 4.63 3.25 1.75 4.75 2.75 7.00 3.00 1.75 1s o 1.oo 1.oo 1.88 1.oo 0.13 0.75 0.50 0.75 0.13 0.13 0.13 0.13 0.13

92 86 94 100 100 70 92 90 96 86 50 78 54 66 72 14 36 26 26 22 28 24 34 26 26 20 26 14 22 24 10 18 14 18 8 14 10 8 2 10 6 2 2 2 2 2 2

175098 100734 50386 42444 40185 84949 35379 21413 521 1 21616 8553 11598 15372 10009 6360 19165 1552 12884 6306 9848 3869 4472 165 221 1 2725 3540 1372 6644 240 532 1126 128 1888 40 1 4023 20 1 190 966 3898 71 202 575 123 101

13 4 2

10.52 7.79 11.47 11.60 7.06 2.10 5.46 6.37 6.61 3.64 6.81 2.44 3.26 1.89 1.55 1.03 1.71 0.58 0.99 0.58 0.64 0.62 0.39 0.52 0.32 0.50 0.35 0.19 0.51 0.29 0.75 0.32 0.19 0.16 0.11 0.11 0.20 0.1 1 0.01 0.08 0.05 0.08 0.01 0.01 0.01 0.01 0.01

5.34 4.99 5.45 5.80 5.80 4.06 5.34 5.22 5.57 4.99 2.90 4.52 3.13 3.83 4.18 0.81 2.09 1.51 1.51 1.28 1.62 1.39 1.97 1.51 1.51 1.16 1.51 0.81 1.28 1.39 0.58 1.04 0.8 1 1.04 0.46 0.8 1 0.58 0.46 0.12 0.58 0.35 0.12 0.12 0.12 0.12 0.12 0.12

24.36 14.02 7.01 5.91 5.59 11.82 4.92 2.98 0.73 3.01 1.19 1.61 2.14 1.39 0.88 2.67 0.22 1.79 0.88 1.37 0.54 0.62 0.02 0.3 1 0.38 0.49 0.19 0.92 0.03 0.07 0.16 0.02 0.26 0.06 0.56 0.03 0.03 0.13 0.54 0.01 0.03 0.08 0.02 0.01 0.00 0.00 0.00

40.21 26.79 23.93 23.31 18.46 17.98 15.72 14.57 12.90 11.64 10.90 8.57 8.54 7.11 6.61 4.5 1 4.02 3.88 3.38 3.22 2.80 2.63 2.38 2.34 2.21 2.15 2.05 1.92 1.82 1.76 1.49 1.38 1.26 1.26 1.13 0.95 0.81 0.71 0.67 0.67 0.43 0.28 0.15 0.14 0.13 0.13 0.13

87

lowed by Myrsine semiserrata, Litsea martabarnica, Quercus eumorpha, Osmanthus fiagrans, Eurya acuminata, Syzygium angkae, and Ilex umbellulata with 96-86% frequency. Species that colonize with high abundance in restricted areas were identified by calculating the abundance of each species as the total number of individuals divided by the number of plots with its presence. A relatively high abundance, if present at all, was observed for Rapanea yunnanensis (20 individualslplot), Castanopsis acuminatissima (12) and Vaccinium sprengelii (1 l), which means that these species colonize limited area but grow abundantly when present. The dominance was calculated on the basis of the basal area; the total basal area was 71.9 m2/ha. Quercus eumorpha had the highest dominance (17.5 m2/ ha), followed by Syzygium angkue (10.1 m%a) and

umbellulata (34) and Acer laurinum (31). These 11 species accounted for 415 of the individuals, they were all big trees, except for Myrsine semiserrata and Rapanea yunnanensis. Nineteen species had a density of 3 individuals per ha or less, of these seven were considered rare species in the area (Camellia sp., Melodinus cochinchinensis, Phoebe cathia, Prunus phaeosticta, Pyrenaria diospyricarpa, Quercus grabicupula and Sorbus granulosa) whereas the remainder were more common at lower altitudes or pioneers in open areas. The rare species represented all life forms (Appendix 1). The 11 most abundant species had a frequency above 86%, except for Rapanea yunnanensis (50%) and Acer laurinum (54%), which had a relatively clumped distribution. Two species, Helicia nilagirica and Lindera caudata, had 100% frequency, fol-

Table 2. Density, diversity, dominance and family importance value (FIV) of 26 plant families in upper montane forest at Mt. Doi Inthanon.

Family

Fagaceae Lauraceae Theaceae Myrtaceae Proteaceae Myrsinaceae Oleaceae Aquifoliaceae Aceraceae Symplocaceae Araliaceae Ericaceae Rosaceae Meliaceae Elaeocarpaceae Myristicaceae Myricaceae Nyssaceae Hamamelidaceae Cornaceae Apocynaceae Juglandaceae Rubiaceae Podocarpaceae Moraceae Polygalaceae

88

Density Diversity Dominance (no. indivha) (no. of species) (cm2ha) 120.50 210.00 88.00 72.75 108.38 125.38 59.37 34.00 30.50 16.75 22.75 7.75 2.38 0.25 9.25 5.75 6.00 4.88 1.oo 3.25 3.00 1.oo 1.oo 0.13 0.13 0.13

6 6 5 1 1 2 3 1 1 2 1 2 2 2 1 1 1 1 1 1 1 1 1 1 1 1

222 116 110 100 42 13 35 21 15 1 11 1

781 309 834 734

444

765 784 616 372 623 598 701 392 15 6 306 4 472 3 869 2 211 4 023 1 372 128 966 20 1 123 101 4

Relarive density 12.90 22.48 9.42 7.79 11.60 13.42 6.36 3.64 3.26 1.79 2.44 0.83 0.25 0.03 0.99 0.62 0.64 0.52 0.1 1 0.35 0.32 0.1 1 0.11 0.01 0.01 0.01

Relative Relative diversity dominance 12.77 12.77 10.64 2.13 2.13 4.26 6.38 2.13 2.13 4.26 2.13 4.26 4.26 4.26 2.13 2.13 2.13 2.13 2.13 2.13 2.13 2.13 2.13 2.13 2.13 2.13

31.00 16.18 15.42 14.02 5.91 1.92 4.98 3.01 2.14 0.23 1.61 0.24 0.05 0.00 0.88 0.62 0.54 0.3 1 0.56 0.19 0.02 0.13 0.03 0.02 0.01 0.00

FVI

56.66 5 1.43 35.48 23.93 19.63 19.59 17.72 8.77 7.53 6.27 6.18 5.32 4.56 4.28 4.00 3.37 3.31 2.96 2.79 2.67 2.47 2.37 2.26 2.16 2.16 2.14

Nord. I. 601.23(1)

Table 3. Plot averages and ranges of compositional and structural variables.

Species richness Family richness Diversity Density Height (m) No. of stems Basal area (m’) Crown cover (m’)

mean

s. d.

min

max

17.2 11 2.32 149.5 13.9 162.9 11.5 5477

4.6 3 0.24 60.1 1.7 74.9 2.0 1403

10 7 1.74 61 .O 9.7 65.0 7.8 2698

29 18 2.81 418.0 17.8 496.0 16.6 8417

Schima wallichii (8.5 m2/ha); these three species accounted for 50% of the basal area. The remainder 44 species each had a basal area of 5 m2/ha or less, and 27 species had a basal area of 0.5 m’ha or less (Table 1). The biggest individual in the forest was Schima wallichii with a basal area of 4.2 m’. Average tree height was 13.4 m with a range from 2 to 50 m, particularly Fagaceae species were of tall stature (Appendix 1). The highest total crown cover was found for Quercus eumorpha (8 901 m2/ha), Syzygium angkae (3398), Litsea martabanica (2753), and Schima wallichii (2491), and these four species accounted for more than half of the total crown cover. A highly significant correlation was found between basal area and crown cover (rho = 0.97, p = 0.000). Myrsine semiserrata and Rapanea yunnanensis had a particularly high crown cover compared to dominance, whereas Schima wallichii and Syzygium angkae had a particularly low crown cover compared to dominance. Importance value index (IVI) combines relative density, relative frequency and relative dominance into a measure that can be use to indicate the ecological influence of each species in the forest. The species with highest IVI were: Quercus eumorpha (40.2), Syzygium angkae (26.8), Litsea martabanica (23.9), Helicia nilagirica (23.3), Lindera caudata (18.5), Schima wallichii (18.0), Osmanthus fragrans (15.7), Eurya acuminata (14.6), Myrsine semiserrata (12.9) and Zlex umbellulata (11.6). These 10 species accounted for 2/3 of the IVI value. The families of Fagaceae, Lauraceae and Theaceae had highest species diversity (6, 6 and 5 species, respectively). The family with highest density was Lauraceae (210 individualslha), and the family with highest dominance was Fagaceae (22.3 m2/ha). The most important families according to FIV were

Nod.J. BoI. 23(I )

Fagaceae, Lauraceae, Theaceae and Myrtaceae (Table 2). A number of plot averages were calculated for the 50, 0.16 ha plots: average species richness was 17 species/plot, average density was 149.5 individuals/ plot (162.9 stems/plot), average height was 13.9 m, average diversity (Shannon Index) was 2.3, average basal area was 11.5 mz/plot and average crown cover was 5477 mzor 3.4 times the plot area (Table 3).

Relations between vegetation and altitude Altitude is a complex gradient that is highly correlated with temperature, light intensity, rainfall and condensation, and furthermore it has been found to be correlated with various soil parameters (Grubb 1977, Vetaas 1997, Givnish 1999, Hansen 2001, Ashton 2003). The variation of soil variables, however, is complex and gradients at many spatial levels are often involved (Chang-Fu et al. 1998). Within the altitudinal range of the present study, the measured soil characteristics were not significanty correlated with altitude. The Principal Component Analysis (PCA) showed that altitude was positively correlated with the major species gradients, which is revealed by the first PCA axis (Fig. 3). Cornus oblonga, Helicia nilagirica, Osmanthus fiagrans, Schima wallichii and Syzygium angkae were characteristic for high altitudes. Helicia nilagirica, Osmanthus j-agrans, Schima wallichii and Syzygium angkae were found throughout the range, but had higher abundance at high altitudes. Cornus oblonga was practically confined to high altitude. The remainder species had higher abundance at lower elevations with about half of the species growing throughout the altitudinal range, e.g. Eurya acuminata, Ilex umbellulata, Lindera caudata, Litsea

89

+PO71

w72

+pm2

pm

+

+

Po88

+ +Po61 pm4

+PO97

PO1

+

+P101

sma fra

PO80

I

Po87

p u e r euni

A-

Cast acu

Svmi D

PO81

Phne cat

+

PO69

+

+Po78

PO74

PO91

i--

Iors sp

+

PO82

+

PO83

Fig. 3. Principal Component Analysis (PCA) of 50 vegetation plots from UMF. The first and the second axes explains 31 % and 14 % of the variance, respectively. Altitude is correlated with the major vegetation gradient.

sp., Litsea beusekomii, Litsea martabanica, Macropanax dispermus, Myrsine semiserrata and Quercus eumorpha, and about half limited to altitudes below 2400 m, e.g. Horsfeldia glabra, Lithocarpus aggregatus, Lithocarpus echinops, Myrica esculenta, Nyssa javanica and Pyrenaria diospyricarpa in some cases with a few outlying occur-

90

rences at higher altitude, e.g. Acer laurinum, Adinandra intergerrima, Elaeocarpus sphaericus, Rapanea yunnanensis and Symplocos sumuntia. Altitude generally plays a major role on species composition and vegetation structure (Chang-Fu et al. 1998, Singh et al. 1994, Vetaas 8c Chaudhary 1998). In the present study, most of the vegetation

Nord. J . Bot. 23( I )

30

Average tree height

25

0 0

20

.

0

15 0.

0

0

0

0

0 "

880

0

0. 0.

10

.

8.00 86

0

I

18 15-

12 -

Family richness

0 00 0

0

Basal area 0

.

0

0

0 0.0

0 .

0 0

0

0 . 0 0 0 00 0

9-

0 0 0 0.0

00 00 0 00 0 0

67

3

1

Stems >200 cm gbh

23

2

0.

-

0

0 0 .

400-

3 3

U .-

.z

.+

100-

O 0

0 00 O

.

0

8

i8'80 0

0 . 0

0

%

30-

u-

20-

E

0

C

"

Stems 100-200 cmgbh

40-

0

0

300200-

-

Density

0

c)

0

$&

0 0

0

0

'Oi

.

;;fi:

$ 0 880 O0 080°00 0 .

0

0 0. 0

1

I

1

I

Crown cover

-1 0000

-

Stems 15-100 cm gbh

0

2 E

.

8000

Y

5 6000 8

5 4000

$00

?! __-.

2000 2100 2200 2300 2400 2500 2600

2ooo 2100 2200 2300 2400 2500 2600

Altitude (m)

Altitude (m)

Fig. 4. Relationship between altitude and compositional and structural vegetation variables.

variables showed significant (a = 0.01) correlation with altitude (Fig. 4); species richness (rho = -0.81), family richness (rho = -0.76),diversity (rho = -0.75),

Nord. I. Bol. 23( I )

density (rho = -0.47) and crown cover (rho = -0.48) were all negatively correlated with altitude, average tree height was uncorrelated with altitude, and only

91

Carlanoqsiq acumihalfssrma

,

Schima wallichii

lam

286

-"

I

Lilhocarpus echinops

0 160

1Rn

M

Syminglmia populnea

rn

Enge/hamYii, spicala 5

1 1 . 1 1 1 92

Nord. J. Bot. 23( I )

basal area was positively correlated with altitude (rho = 0.52). The positive correlation was caused by a relatively large number of big individuals at higher altitude; the number of stems with gbh over 200 cm showed a significant positive correlation with altitude (rho = 0.41). No particularly large-growing species explained this patern, in contrast many species were represented by larger individuals at higher altitudes, supposingly because higher altitude areas have been less disturbed than lower lying ones (Anderson 1993, Pooana & Barfod 2001, Janzaji et al. 2004). Individuals in the size class from 100 to 200 cm gbh were evenly distributed throughout the altitudinal range, whereas the number of stems with gbh less than 100 cm showed a significant negative correlation with altitude (rho = -0.49). We found an increase in basal area and a decrease in density and species richness with increasing altitude, but the relationship is by no means universal. Both density and basal area were found in Himalaya to decrease with altitude (Singh et al. 1994, Kanai et al. 1975) whereas in Mexico they were found to increase with altitude (Vhzquez & Givnish 1998). The results presented by Kitayama (1992, 1995) and Givnish (1999) confirm that the number of woody species generally decrease with altitude. But in contrast to findings by Singh (1987), Nakashisuka et al. (1991), and Kappelle et al. (1995) tree height was not found to decrease with altitude, supposingly because higher altitude areas have been less disturbed than lower lying ones.

Size class distribution Size class distributions can be used as indicators of changes in population structure and species composition (Newbery & Gartlan 1996, Poorter et al. 1996). Distribution curves that drop exponentially with increasing dbh (reverse-J-shaped) are characteristic for species with continuous rejuvenation. Curves showing little or no drop in the lower dbh classes indicate that recruitment is unsustainable and that long term changes in species composition of the plant community studied is to be expected (Hall & Bawa 1993). In undisturbed environments the shapes of the size class distribution curves may be explained to some degree by phenological patterns or population dynamic parameter. Species with fast growth or with high survival rate in early stages will thus have flatter curves than on average (Condit et al. 1988). Al-

though it is not possible to predict the fate of a species from the shape of the size class distribution curve alone, a constant lack of rejuvenation in many species gives a good indication that the forest ecosystem is disturbed (Lykke 1998). Most species in the UMF followed the reverseJ shaped distribution with plenty of individuals in the small size .classes that indicates a good rejuvenation (Fig. 5). A group of species, including four relatively common species, Lithocarpus echinops, Schima wallichii, Castanopsis acuminatissima and Myrica esculenta, had a flat distribution with relatively few individuals in the small size classes, which indicate unstable recruitment and probably declining populations. The pioneer species, Engelhardtia spicata, Rhododendron arboreum and Symingtonia populnea, were not represented by individuals smaller than 30 cm gbh in any of the investigated plots indicating that the forest has reached a climax where pioneer species are shaded out. In general the forest seems in a good condition with reverse-J-shape distributions of most species indicating a stable species composition and a good rejuvenation potential.

Conclusion During the last century the forested area in Northern Thailand has decreased dramatically (Fox et al. 1995). This has raised both local, national and international interest in the remaining well-protected forests for conservation of biodiversity and preservation of important ecological functions. The UMF at the summit of Mt. Doi Inthanon has until recently been uncultivated and relatively undisturbed . in contrast to lower laying areas, which explains the dense vegetation and many large trees. During the last century, however, a site has been cleared for a military base and a road has been cut through the forest to the summit. This has led to some wind destruction of trees along open areas and following soil erosion. Inside the remaining forest patches, however, the forest condition seems relatively good with stable population structures of most species. As the UMF is of limited size (less than 5 km2)and vulnerable due to fragmentation, it is important to survey the area to ensure a continued healthy forest structure. The present study has for the first time provided a quantitative description of species richness, floristic composition and structural characteristics in the UMF of Mt. Doi Inthanon, which has created a

Fig. 5 . Size class distributions for all species with at least eight individuals in the sample, excluding lianas; x-axes give the size class and y-axes the number of individuals. Species are arranged after size class distribution slope.

Nord. 1. Bol. 23(1)

93

basis for fbture surveys as well as a large number of permanent plots that can be revisited in the future for monitoring vegetation dynamics.

Acknowledgements - This paper is one output of the research component on biodiversity and ecology under the project, “Forest and People in Thailand”, supported by DANCED (Denmark). The authors would like to thank four students of Chiang Mai University; Tanongsak Parathai, Danai Sancanthong, Tanun Hongsak and Chuttima Pradittivej for help during field work. Thanks also to M. Kanzaki, Kyoto University, who helped to identify some plant species and to T. Santisuk and R. Pooma for constructive comments on the manuscript.

References Anderson, E. 1993. Plants and people in the golden triangle. Ethnobotany of hill tribes of northern Thailand. Dioscorides, Portland. Anonymous. 1989. The master plan for management of the Doi Inthanon National Park, Chiang Mai Province. Thailand Faculty of Forestry, Kasetsart University, Bangkok. (In Thai). Ashton, P. S. 2003. Floristic zonation of tree communities on wet tropical mountains revised. - Persp. P1. Ecol. Evol. Syst. 6: 87-104. Ateroff, M. & Rada, F. 2000. Deforestation impact on water dynamics in Venezuelan Andean cloud forest. - Ambio 29: 440-444.

Bruijinzeel, J. A. & Proctor, J. 1995. Hydrology and biogeochemistry of tropical montane cloud forests: What do we really know ? - In: Hamilton, L. S., Juvik, J. 0. & Scatena, F. N. (eds), Tropical. montane cloud forest. SpringerVerlag, New York. Ecol. Studies 110: 38-78. Chang-Fu, H., Zueng-Sang, C., Yueh-Mei, H.,Kuoh-Chieng Y. & Tsung-Hsin, H. 1998. Altitudinal zonation of evergreen broad-leaved forest on Mount Lopei, Taiwan. - J. Veg. Sci. 9: 201-212. Chiang Mai University. 1995. Precipitation recorded at Faculty of Agriculture. - Chiang Mai University (19861995).

Chuchip, K. 1989. Land use studies and plant community analysis for situation analysis of Doi Inthanon National Park, Changwat, Chiang Mai. - Master Thesis, Kasetsart University, Bangkok. (In Thai). Condit, R., Sukumar, R., Hubbell, S. P. & Foster, R. B. 1998. Predicting population trends from size class distributions: a direct test in a tropical tree community. - Amer. Nat. 152: 495-509.

Doi Inthanon Military Base. 2001. Precipitation data recorded at the military base at the summit of Doi Inthanon (1982-1999).

Ewel, J. 1980. Tropical succcession: manifold routes to maturity. - Biotropica 12, suppl.: 2-7. Fox, J., Krummel, J., Yarnasarn, S., Ekasingh, M. & Podger, N. 1995. Land use and landscape dynamics in Northern

94

Thailand assessing change in three upland watersheds. Ambio 24: 328-334. Gardner, S., Sidisunthornand, P. & Anusamsunthorn, V. 2000. A field guide to forest trees of northern Thailand. Kobfai, Bangkok. Givnish, T. J. 1999. On the causes of gradients in tropical tree diversity. - J. Ecol. 87: 193-210. Grubb, P. J. 1977. Control of forest growth and distribution on wet tropical mountains: with special reference to mineral nutrition. - Annu. Rev. Ecol. Syst. 8: 83-107. Hall, P. & Bawa, K. 1993. Methods to assess the impact of extraction of non-timber tropical forest products on plant populations. - Econ. Bot. 47: 234-247 Hamilton, L. S., Juvik, J. 0. & Scatena, F. N. 1995. Tropical montane cloud forests. - Springer, New York. Hansen, P. K. 200 1. Environmental variability and agro-ecological stratification. - In: Poulsen, E., Skov, F., Lakanavichian, S., Thanisawanyangkura, S., Borgtoft, H. & H~jris,0. (eds), Forest in culture - culture in forest: perspective from Northern Thailand. Research Centre on Forest and people in Thailand, Tjele, pp. 163-176. Hara, M., Janzaki, M., Mizuno, T., Noguchi, H., Sringernyuang, K., Teejuntul, S., Sungpalee, C., Ohkubo, T., Yamakura, T., Sahunalu, P., Dhanmanonda, P. & Bunyavejchewin, S . 2002. The floristic composition of tropical montane forest in Doi Inthanon National Park. Northem Thailand, with special reference to its phytogeographical relations with montane forests in tropical Asia. - Nat. Hist. Res. 7: 1-17. IPNI. 2004. The international Plant Names Index. www.ipni.org. Isager. L. 2001. History and people of northern Thailand. In: Poulsen, E., Skov, F., Lakanavichian, S. Thanisawanyangkura, S., Borgtoft, H. & H~jris.0. (eds), Forest in culture - culture in forest: perspectives from northern Thailand. - Research centre on Forest and People in Thailand. Tjele, pp. 85-115. Kanai, H., Shakya, P. R. & Shrestha, T. B. 1975. Vegetation survey of central Nepal. - In: Ohashi, H. (ed.), The flora of eastern Himalaya. University of Tokyo, Tokyo, pp. 416-423.

Kanzaki, M., Haw, M., Yamakura, T., Ohkubo, T., Tamura, M. N., Sri-ngernyuang, K., Sahunalu, P., Teejuntuj, S. & Bunyavejchewin, S. 2004. Doi Intahnon Forest Dynamics Plot, Thailand. - In: Losos, E. C. & Leigh Jr., E. G. (eds), Tropical Forest Diversity and Dynamics. Findings from a large-scale plot network. Chicago Press, Chicago, pp. 474-48 1.

Kappelle, M., van Uffelen, J.-G. & Cleef, A. M. 1995. Altitudinal zonation of montane Quercus forests along two transects in Chimp6 National Park, Costa Rica. Vegetatio 119: 119-153. Khamyong, S., Pramprasit, S. & Pinthong, J. 1996. The relationship between plant associations and soil characteristics in a dry dipterocarp forests in Thailand. - In: Appanah, S . & Khoo, K. C. (eds), Proceedings fifth round-table conference on Dipterocarps. Forest Research Institute, Malaysia, pp. 117-137. Kitayama, K. 1992. An altitudinal transect study of the vegetation on Mount Kinabalu, Borneo. - Vegetatio 102: 149- 17 1. - 1995. Biophysical conditions of the montane cloud forests of Mount Kinabalu, Sabah, Malaysia. - In: Hamilton,

Nord. J. Bot. 23(1)

L. S., Juvik, J. 0. & Scatena, F. N. (eds), Tropical montane cloud forest. Springer-Verlag, New York. Ecol. Studies 110: 183-197. Krebs, C. J. 1994. Ecology: the experimental analysis of distribution and abundance. - Harper Collins, New York. Kiichler, A. W. & Sawyer, 0. 1967. A study of the vegetation near Chiangmai, Thailand. - Trans. Kansas Acad. Sci. 70: 281-347. Lykke, A. M. 1998. Assessment of species composition change in savanna vegetation by means of woody plants’ size class distributions and local information. - Biodiv. Conserv. 7: 1261-1275. MacDonald, A. S., Ban; S. M., Dunning, G. R. & Yaowanoiyothin, W. 1993. The Doi Inthanon metamorphic core complex in NW Thailand: age and tectonic significance. J. S.-E. Asian Earth Sci. 8: 117-125. Maxwell, J. F. & Elliott, S. 2001. Vegetation and Vascular Flora of Doi Suthep - Pui National Park, Northern Thailand. - The Biodiversity Research and Training Program (BRT), pp. 14-15. McCune, B. & Mefford, M. J. 1999. PC-ORD. Multivariate analysis of ecological data, vers. 4. - MjM Software Design, Gleneden Beach. Mori, S. A. & Boom, B. M. 1983. Ecological importance of Myrtaceae in an Eastern Brazilian wet forest. - Biotropica 15: 68-70.

Nakashisuka, T., Yusop, Z. & Nik, A. R. 1991. Altitudinal zonation of forest communities in Selangor, Peninsular Malaysia. - Trop. Forest Sci. 4: 233-244. Newbery, D. McC. & Gartlan, J. S. 1996. A structural analysis of rain forest at Korup and Douala-Edea, Cameroon. Proc. Royal SOC.Edinburgh 104B: 107-224. Ohsawa, M. 1991. Structural comparison of tropical montane rain forest along latitudinal and altitudinal gradients in south and east Asia. - Vegetatio 97: 1-10. - 1995. The montane cloud forest and its gradational changes in Southeast Asia. - In: Hamilton, L. S., Juvik, J. 0. & Scatena, F. N. (eds), Tropical montane cloud forests. Springer, New York. Ecol. Studies 110: 254-265. Pooma, R. & Barfod, A. 2001. Vegetation types in Northern Thailand. - In: Poulsen, E., Skov, F., Lakanavichian, S., Thanisawanyangkura, S., Borgtoft, H. & Hejris, 0. (eds), Forest in culture - culture in forest: perspective from

Nod. J. Bot. .?XI)

Northern Thailand. Research Centre on Forest and People in Thailand, Tjele, pp. 11-15. Poorter, L., Bongers, F., van Rompaey, R. S. A. R. & de Klerk, M. 1996. Regeneration of canopy tree species at five sites in West African moist forest. - Forest Ecol. Managem. 84: 61-69. Robbins, R. G. & Smitinand, T. 1966. A botanical ascent of Doi Inthanon. - Nat. Hist. Bull. Siam SOC.21: 205-227. Royal Project. 1995. Precipitation recorded at Royal Project Station, Doi Inthanon National Park. - Chiang Mai (1 993- 1995).

Santisuk, T. 1988. An account of the vegetation of northern Thailand. - Franz Steiner, Stuttgart. Singh, S. P. 1987. Forest vegetation of the Himalaya. - Bot. Rev. 53: 80-192. - , Adhikari, B. S. & Zobel, D. B. 1994. Biomass, productivity, leaf longivity, and forest structure in the central Himalaya. - Ecol. Monogr. 64: 401-421. Sri-Ngernyuang, K., Kanzaki, M., Mizuno, T., Noguchi, H., Teejuntuk, S., Sungpalee, C., Hara, M., Yamakura, T., Sahunalu, P., Dhanmanonda, P. & Bunyavejchenwin, S. 2003. Habitat differentiation of Lauraceae species in a tropical lower mintane forest in northern Thailand. Ecol. Res. 18: 1-14. Vizquez, G. J. A. & Givnish, T. J. 1998. Altitudinal gradients in tropical forest composition, structure, and diversity in the Sierra de Manantalh. - J. Ecol. 86: 999-1020. Vetaas, 0. R. 1997. The effect of canopy disturbance on species richness in a central Himalayan oak forest. - PI. Ecol. 132: 29-38.

Vetaas, 0. R. & Chaudhary, R. P. 1998. Scale and speciesenvironment relationships in a central Himalayan oak forest, Nepal. - PI. Ecol. 134: 67-76. Vogelmann, H. W. 1973. Fog precipitation in the cloud forests of Eastern Mexico. - BioScience 23: 96-100. Werner, W. L. 1995. Biogeography and ecology of the upper montane rain forest of Sri Lanka (Ceylon). - In: Hamilton, L. S., Juvik, J. 0. & Scatena, F. N. (eds), Tropical montane cloud forest. Springer-Verlag, New York. Ecol. Studies 110: 343-352. Werner, W. 2001. Sri Lanka’s magnificent cloud forests. WTH, Colombo. Whitmore, T. C. 1975. Mountain rain forest. - In: Tropical rain forest of the far East. Clarendon, Oxford, pp. 243-259.

95

Appendix 1 Characteristics of 47 species sampled in uper montane forest at Mt. Doi Inthanon. Species name

Family

Acer laurinum Hassk. Adinandra intergerrima T. Anderson ex Dyer Aphanamixis polystachya (Wall) R. Parker Camellia sp. Castanopsis acuminatissima (Blume) A. DC. var. acuminatissima Cornus oblonga Wall. var. siamica Geddes Elaeocarpus sphaericus (Gaertn.) K. Schum. Engelhardtia spicata Blume var. colebrookana (Lindl. ex. Wall.) Kuntze Eurya acuminata DC. var. wallichiana Dyer Ficus sp. Helicia nilagirica Bedd. Horsfeldia glabra (Blume) Warb Ilex umbellulata Loes. Jasminum attenuatum Roxb. ex G. Don Jasminum dispennum Wall. ssp. forrestianum (Kobuski) P.S.Green Lindera caudata (Nees) Hook. f. Lithocarpus aggregatus Bamett Lithocarpus echinops Hjelmqvist. Lithocarpus garrettianus (Craib) A. Camus Litsea beusekomii Kostermans Litsea martabanica (Kurz) L. f. Litsea sp. Macropanax dispennus (Blume) Kuntze Melodinus cochinchinensis (Lour.) Merr. Myrica esculenta Buch.-Ham. Myrsine semiserrata Wall. Nyssa javanica (Blume) Wangerin. Osmanthus fiagrans Wall. Phoebe cathia (D. Don) Kosterm. Phoebe paniculata Nees Podocarpus neriifolius D. Don Polygala arillata Buch.-Ham. ex D. Don Prunus phaeosticta (Hance) Maxim. Pyrenaria diospyricarpa Kurz Quercus eumorpha Kun Quercus glabricupula Bamett Rapanea yunanensis Mez Rhododendron arboreum Sm. ssp. delavayi (Franch.) Chamb. Schima wallichii (DC.) Korth Sorbus granulosa (Bertol.) Rehder Symingtonia populnea (R. Br. ex Grift.) Steenis Symplocos macrophylla Wall ex DC. ssp. sulcafa (Kurz) Noot. var. sulcata Symplocos sumuntia Buch.-Ham. ex D.Don Syzygium angkae (Craib) Chantar. & J. Pam. ssp. angkae Tarennoidea wallichii (Hook. f.) Tirveng. & Sastre Trichilla connaroides (Wight & Am.) Bentv. Vaccinium sprengelii (G. Don) Sleumer

Aceraceae Theaceae Meliaceae Theaceae Fagaceae Cornaceae Elaeocarpaceae Juglandaceae Theaceae Moraceae Proteaceae Myristicaceae Aquifoliaceae Oleaceae Oleaceae Lauraceae Fagaceae Fagaceae Fagaceae Lauraceae Lauraceae Lauraceae Araliaceae Apocynaceae Myricaceae Myrsinaceae Nyssaceae Oleaceae Lauraceae Lauraceae Podocarpaceae Polygalaceae Rosaceae Theaceae Fagaceae Fagaceae Myrsinaceae Ericaceae Theaceae Rosaceae Hamamelidaceae Symplocaceae Symplocaceae Myrtaceae Rubiaceae Meliaceae Ericaceae

96

Nord. 1. Bot. 23(1)

Life form big tree big tree big tree small tree big tree medium tree big tree big tree big tree medium tree big tree medium tree big tree climber climber big tree big tree big tree big tree small tree big tree big tree medium tree climber small tree small tree big tree big tree big tree big tree big tree shrubby tree small tree medium tree big tree big tree small tree small tree big tree small tree big tree small tree small tree big tree small tree big tree small tree

Nod. J. Bot. 23(1)

Max. height (m) 45 34 5 14 44 35 43 26 38 20 33 37 50 24 25 40 48 45 47 27 40 42 25 20 20 15

45 38 45 34 20 6 19 18 50 34 20 13 45 19 35 13 14 43 16 13 15

Average height (m) 14.7 13.6 5.0 10.8 20.3 13.1 13.3 16.8 11.8 20.0 12.3 12.4 13.6 14.4 14.0 15.4 19.5 22.9 47.0 13.1 13.2 12.4 12.6 14.1 13.6 7.1 13.6 11.9 15.3 15.7 17.0 6.0 9.3 8.7 18.4 14.4 9.2 13.3 23.7 13.1 24.1 11.9 8.1 15.1 9.7 13.0 7.5

Av. no. of stems

Crown cover (m2/ha)

1.o

1066 233 1 33 796 71 358 53 1278 6 1935 187 1015 19 22 2351 50 1 63 5 88 523 2753 342 533 13 249 859 143 1530 83 121 9 1 20 28 8901 92 925 39 249 1 12 141 6 190 3398 20 2 160

1.2 1.o

1.2 1.4 1.2 1.1 1.3 1.2 1.o 1.o 1.1 1.1 1.o 1.o 1.1 1.1 1.o 1.o 1.1 1.1 1.o 1.1 1.o 1.3 1.1 1.o 1.2 1.5 1.1 2.0 1.o 1.2 1 .o 1.1 1.1 1.1

2.0 1.1 1.5 1 .o

1.2 1.o 1.o 1.0 1.o 1.8

97