Functional attributes including physiological, life history and ecological traits are directly related to variation in elevation that change with climatic factors as well as land area surface. Temperature, air pressure, and solar radiation are the climatic factors that vary with elevation influencing on the distribution of species [9] . Elevation, slope inclination and aspect exposition are the topographic factors playing an important role in determining the abundance of PFTs. Elevation is a complex combination of related climatic variables closely correlated with numerous other environmental properties like soil texture, nutrients, substrate stability and many more [10] .

Land-use change includes modified and fragmented habitats, which are more vulnerable to invasive species and plays an important role in shaping and maintaining ecosystem structures and functions. According to [11] , land-use changes involve two main impacts on the biosphere: first, the conversion (i.e. natural habitats altered for human use) and second, intensification (e.g. greater intensity and frequency of disturbance, increased use of external inputs). Habitat loss and fragmentation are the major factors which cause biodiversity loss by decreasing suitable habitats of species and thereby forcing the species to extinction [12] . Land conversion for agricultural purposes, population growth, socio-economic conditions drive land cover changes.

The present study deals with the variation in plant functional traits along the different land-use types and altitudinal gradient in Sagarmatha National Park and Buffer Zone. Further, the study aims to know the major traits the dominant plant species present in the study area; to assess variation in the traits of dominant species among major land use types; and to identify the plant species best suited for specific altitude and land-use type.

Sagarmatha National Park (SNP) is located in the north-eastern part of Nepal in Solukhumbu district. SNP covers an area of 1148 sq∙km of Namche and Khumjung VDCs. The park is located between 27˚46'19'' to 27˚06'45''N latitudes and 86˚30'53'' to 86˚99'08''E longitudes. The buffer zone was declared in January 1, 2002, with an area of 275 sq∙km [13] . This study has been carried out starting from Surkey (2200 m) upto Khumjung (3800 m). Annual precipitation occurs in the monsoon season from June to September, maximum in the July and the remainder of the year is fairly dry. Annual precipitation is 984 mm in Namche Bazar, 733 mm in Khumjung and 1043 mm in Tengboche. The mean temperature of the coldest month, February is −0.11˚C. Mainly from July to August, the average annual rainfall is approximately 100 mm. Figure 1 reveals the map of the study area.

Two field investigations at April and September in 2011 were made for collecting samples. Sampling was done from 2200 - 3800 m asl varying approx. 400 m. Two different aspects (east and west facing) of each valley were chosen. In each aspect four land-use types were selected. A transect of 25 m long and 2.5 m wide was laid [14] . Eight plant functional traits were examined for 60 plant species belonging to 31 families, collected from 40 sampled plots. Eight functional traits i.e. Growth form, plant height, clonality, spinescence, leaf dry matter content, stem specific density, twig dry matter content, twig drying time were conducted and analysed.

Land-use types category included four different types namely, Natural forest, Exploited forest, Meadows and Cultivated land [14] . Natural forests are area

with either pristine character or with a low anthropogenic influence. Exploited forests are areas with a relatively closed canopy but with an altered stand structure and tree species composition compared to natural forests, either because of intensive exploitation for agriculture, livestock, and collection of fodder and fuel woods or plantation after intensive domestic or commercial forest management. Meadows are open grasslands with less than 20% tree coverage. Cultivated lands are intensively managed, fertilized, sometimes irrigated and yearly ploughed areas.

For Laboratory techniques, the protocol for standardized and easy measurement of plant functional traits worldwide [15] was followed. Functional traits were either directly measured in the forest, at the research station in the evening after collecting, in the laboratory or deduced later from measured traits.

For analyses of quantitative traits, the overall data was divided into three growth forms (i.e. herb, shrub and tree). Each trait was then tested to find significant difference with land-use types and altitude for three growth forms. After a normality test (Shapiro Wilk), traits such as plant height, LDMC and SSD of herbs and shrubs were square root transformed to meet the normality assumption while the remaining data was considered normal and left untransformed.

One-way Analysis of Variance (ANOVA) was used to find the variability across means of quantitative traits with land-use types and altitude. Post hoc analysis (Tukey’s HSD test) at 5% level of significance was carried out within the results of ANOVA to find more pairwise significant difference of traits with environmental variables. ANOVA was also used to find the altitude wise variation in quantitative traits along different land-use types. Spearman’s correlation was done to test the relationship among functional traits in overall altitude and to explore the relationship among functional traits for each five altitude. ANOVA was performed using statistical computer program R version 2.12.1. Correlation was done in SPSS Statistics 17.0.

Nine different types of growth forms were recorded from the study site. The plant species collected (both dicots and monocots) included 12 trees, 19 shrubs, 28 herbs and 1 climber. The herbaceous plants included 5 short basal, 1 long basal, 1 semi basal, 13 erect leafy, 3 tussocks and 5 palmoid. The most abundant form was the herbaceous erect leafy followed by trees and shrubs. Trees were dominant in natural forest. Short basal and erect leafy herbs were found to be dominant on meadow. The number of shrubs was higher in the exploited forest in comparison to other land-use types (Figure 2(a)). Along the altitudinal gradient, trees and erect leafy herbs decreased in number in comparison to other growth forms. Dwarf shrub and tussocks were dominant at 3400 m (Figure 2(b)).

Plant height of the investigated species ranged from 0.03 - 15 m. Among the

herb and shrub species, height showed highly significant difference with altitude. Height of the tree species showed significant difference with land-use type and altitude (Table 1). From post hoc analysis, significant difference was observed between meadow and natural forest (p < 0.01). Outlier was observed at 3000 m showing the highest height of herb. Herb height showed the lowest value at 3800 m with outlier (Figure 3(a)). Shrub height showed the lowest value at 2600 m and the highest value at 3000 m and 3800 m (Figure 3(b)). Taller tree was present at natural forest and tree with less height were observed in meadows (Figure 3(c)).

The stolon consisting species such as Anaphalis busua, Androsacae sarmentosa, Fragaria nubicola, and F. indica were dominant in exploited forest. Diversity of clonal species was more in meadow (Figure 4(a)). Non-clonal species were dominant in all the altitudes. At 3400 m, stolon, bulb and adventitious bud forming species were dominant. At 2200 m, non-clonal species were most dominant followed by stolon and vegetative bud forming species (Figure 4(b)). Species with vegetative buds were Arundinaria maling, Impatiens urticifolia and Rosa sericea; and others clonal belowground were Commelina maculata, Iris clarkei, Malaxis cylindristachya, Senecio diversifolius and Rumex nepalensis.

Among the investigated species, only eight consisted of spines. The species with longer spines were dominant mostly at the meadows (Figure 5(a)). In comparison to lower altitude, spine consisting species decreased at highest altitude (Figure 5(b)). Spinescence was observed in Berberis aristata, Cotoneaster microphyllus, Dipsacus inermis, Rosa sericea, and Zanthoxylum armatum, occurring at the meadow whereas Rubus ellipticus and Quercus semecarpifolia were occurred at the natural forests.

The highest LDMC was recorded in Rhododendron arboreum (736.360 mg/g) followed by Pinus wallichiana, Quercus semecarpifolia and Abies spectabilis

Note: ***Significant at p < 0.001; **Significant at p < 0.01; *Significant at p < 0.05; p value without asterisk being insignificant.

while the lowest LDMC was of Pilea umbrosa (35 mg/g). LDMC of herb was significant with land-use type. From post hoc analysis, significant difference (p < 0.01) was found in natural and exploited forest. LDMC of shrub was found significantly different with altitude. Similarly, LDMC of trees showed significant difference with land-use type and altitude (Table 2). Lowest LDMC of tree was

Note: ***Significant at p < 0.001; **Significant at p < 0.01; *Significant at p < 0.05; p value without asterisk being insignificant.

observed in natural forest (Figure 6(a)). Highest value of shrub LDMC was observed at 2600 m with outliers and the lowest value at 3400 m (Figure 6(b)). Highest value of herb LDMC was observed at 3400 m and lowest at 2200 m with outliers (Figure 6(c)).

Altogether 27 trees and shrub species were collected and among them stem specific density of 15 shrubs were measured. The highest value of SSD was recorded in Rhododendron lepidotum (9.35 mg/mm³) while the lowest value was observed in Anaphalis busua (0.19 mg/mm³). Stem specific density for herbs showed significant difference with land-use type (Table 3). More difference was seen in between exploited forest and cultivated land (p < 0.001) as well as natural forest and exploited forest (p < 0.001) from post hoc analysis. Similar result was observed from boxplot showing lowest value of SSD at exploited and natural forest with presence of outliers and highest value at natural forest (Figure 7).

Twig dry matter content was calculated for tree and shrubs. The highest value of TDMC was 875 mg/gm in Gaultheria fragrantissima, while the lowest was 251.57 mg/gm in Abies spectabilis. TDMC of shrub species was found significant with

Note: ***Significant at p < 0.001; **Significant at p < 0.01; *Significant at p < 0.05; p value without asterisk being insignificant.

land-use and altitude (Table 4). More difference was found between meadow and exploited forest. Similarly, twig dry matter content for trees showed significant difference with land-use type (Table 4). Outlier was present showing the highest value of tree TDMC at the natural forest and lowest mean value was observed at meadow (Figure 8).

There was not much difference in the twig drying time of the plant species. TDT ranged from 2 - 4 days. The twigs of Juniperus indica, Pieris formosa, Pinus wallichiana and Quercus semecarpifolia dried in 4 days, while Osbekia took the lowest drying time i.e. 2 days. All the other remaining species took 3 days to dry out completely.

Spearman’s Correlation Analysis revealed significant relationships among a number of traits (Table 5). Significant correlation (p < 0.01) were observed between Plant height and LDMC (r = 0.553), Plant height and SSD (r = 0.462), Plant height and TDMC (r = 0.802), Plant height and TDT (r = 0.801), LDMC and TDMC (r = 0.610), LDMC and TDT (r = 0598.), SSD and TDMC (r = 0.371), SSD and TDT (r = 0.426), and TDMC and TDT (r = 0.976).

Note: ***Significant at p < 0.001; **Significant at p < 0.01; *Significant at p < 0.05; p value without asterisk being insignificant.

Note: **Correlation significant at p < 0.01; *significant at p < 0.05; other correlations being insignificant.

The correlation coefficient value showed significantly different result within same altitudes as well as other altitudes. Along each of the five altitudinal gradients, plant height showed positive correlation with TDMC and TDT. Higher significant correlation between traits was found in the lower altitude (2200 m). LDMC also showed positive correlation with TDMC and TDT at 2200 m and 2600 m.

The result achieved shows variation in different traits along altitude as well as land-use types. Among the various growth forms, tree was dominant in natural forest while different categories of herbs were less dominant. This might be because herbs do not grow well under canopy due to difficulty in light intensity much reaching to the ground [16] [17] . Shrubs and erect leafy were found to be dominant at meadow and exploited forest which may be an adaptation to grazing. The height and positioning of the foliage may be both adaptations and responses to grazing by different herbivores [15] . Along the altitudinal gradient, tree and erect leafy herb species decreased in number whereas other herb and dwarf shrub species were found to be dominant along higher altitude. Similar result of increment in herbs at higher altitudes was found by Wana and Beierkuhnlein (2009) [18] in southwest Ethiopian highlands. The distinction of growth forms informs us about the specific adaptations to environmental conditions such as climate and wind [19] [20] , solar energy partitioning [21] , water use efficiency [22] and resource partitioning [23] in the ecosystem.

Herb and shrub height showed a significant difference with the altitude. A study by Bhattarai and Vetaas (2003) [24] in the Himalayan range indicates that woody species display a uni-modal pattern of diversity while herbaceous species do not show any relationship to altitude which is different from the result of present study. The reason behind it could possibly be the effect of an incomplete gradient length of the study ranging of 2200 - 3800 m asl. Plant height in case of shrubs showed significant difference between natural forest and exploited forest as well as between meadow and exploited forest. Significant difference was observed with land-use types mainly between meadow and natural forest in case of tree height. It may be due to the presence of very few trees of short stature in meadow, and low disturbance such as grazing, mowing, firing, etc in natural forest in comparison to the meadow. Short plants were found most dominant in the meadows because they have higher quality and growth rate, and are more tolerant to herbivory than taller plants [25] .

Rhizomes and other clonal species were dominant in high altitudes for adaptation to chilling and grazing, similar to the finding of Wana and Beierkuhnlein (2009) [18] . Increase of clonal species in the colder sites was also found by [26] in East Ladakh, Western Himalayas. Diversity of clonal species was high in the meadows as they grow well in harsher environment [10] and nutrient-poor conditions [27] . In response to severe environments, plants either develop adaptations that enable protection of renewing buds or spread the risk of bud mortality by multiplying buds through clonal growth [28] . Clonal behavior may also be an effective means of short distance spread under circumstances of poor seed dispersal or seedling recruitment [15] .

Plant species with spines were mostly shrubs and were dominant at the meadows as disturbance level is high due to grazing, trampling, etc. Another reason may be defense mechanism to protect them from herbivores. They can play an additional role in reducing heat or drought stress [15] by dissipating heat loading on the surface of the leaves and stems or absorbing solar radiation, in addition to their role as a mechanical deterrence against vertebrate herbivory [29] . Spines were found to be more abundant in lower altitudinal ranges in response to drought and grazing which is similar to the findings Wana and Beierkuhnlein (2009) [18] and Dalacho (2009) [20] .

Leaf dry matter content of tree varied significantly with land-use type and altitude. Significant differences were found between meadow and cultivated land as well as natural forest and cultivated land. In the natural forest, trees with high LDMC were present while in the cultivated and meadow more herbs with low LDMC were present. Species with low LDMC are associated with highly disturbed environments, intensive land-use and colder sites [30] . At higher altitude, fast-growing and shorter species also have low LDMC and lower toughness [31] . Plants found in highlight, less intensive land-use and warmer sites usually have thick leaves with high LDMC [30] [32] . LDMC was significant with altitude in the herbaceous species. Spinescence, growth form, clonality, Twig drying time and plant height were positively correlated with LDMC.

Stem specific density for herbs showed significant difference with land-use type. More difference was observed between exploited forest and cultivated (p < 0.001) as well as natural forest and exploited (p < 0.001). Trees present in the natural forest had more stem specific density than the ones present in the exploited forest. Best competitors for light are those plants that allocate large fractions of their biomass to above ground organs i.e. stems and leaves [1] . SSD is correlated with plant height and TDMC. The higher the plants are the higher is their stem density as well as twig dry matter content. A dense stem provides the structural strength that a plant needs to stand upright and the durability it needs to live sufficiently long. The lower the stem density is the higher is their plant growth rate.

Twig dry matter content of tree differed significantly with altitude and land-use type. High significance was between meadow and exploited forest. Low TDMC may be positively correlated with high potential relative growth rate. Twig drying time was positively correlated to twig dry matter content in case of trees. Twigs with high dry matter are expected to dry out relatively quickly during the dry season in fire-prone regions [15] . Highest TDMC value in this study was recorded in Gaultheria fragrantissima which indicates that it has stronger avoidance characteristics than the others as reported by Olivero (2011) [33] .