Introduction

Nitrogen is generally considered the most limiting element for terrestrial vegetation. The biogeography of leaf nitrogen (N) is a challenging issue and has drawn great attentions [1]–[7]. Reich and Oleksyn found that at the global scale, leaf N concentration (N_mass) increased from the tropics to mid-latitudes and then remained stable or decreased at high latitudes [3]. Han et al. (2005) showed that in China, leaf N increased with increase in latitude (decreasing mean annual temperature [MAT]) in 753 plant species pooled together [8]. Recently, Han et al. analyzed leaf N in 1900 plant species across China and discovered that leaf N showed remarkable latitudinal and longitudinal trends driven by climate, soil, and plant functional type and that leaf N variation was explained more by precipitation than by temperature [6].

Climate and vegetation-type change markedly with altitude over a short distance, therefore, mountains are ideal sites for examining the biogeographical pattern of leaf N and the effects of climate, soil, and plant species on leaf chemistry. Altitudinal variation in leaf N has been studied intensely [9]–[15], but most of the studies focused on leaf N content per unit leaf area (N_area) rather than N_mass. Leaf N_area generally tends to increase with increase in altitude, irrespective of life form [9], [10]. However, no conclusive pattern of altitudinal variation in leaf N_mass has been detected because relatively few studies have been conducted and a limited number of plant species have been used. For example, Li et al. reported that leaf N_mass in the evergreen shrub Quercus aquifolioides decreased with increasing altitude [16], whereas Li et al. found that leaf N_mass in the deciduous shrub Hippophae rhamnoides first decreased and then increased with increasing elevation [15]. Luo et al. observed that leaf N_mass in Picea asperata, an evergreen conifer tree, decreased below 2950 m and then increased with increasing altitude [14]. Körner reported that leaf N_mass increased with altitude in herbaceous plants but was remarkably stable in evergreen woody plants [9]. All these findings were limited by the small number of species used in the studies; therefore, additional studies with a large number of plant species are required to obtain a more general representation of altitudinal variation in leaf N_mass.

Mount Gongga is an excellent site for examining altitudinal variation in leaf N because of its altitudinal span (from 1100 to 7600 m), abundant species and intact and varied vertical vegetation belts. In this study, we sampled plants from a broad altitudinal range and measured leaf N_mass. Our objective was to discover the altitudinal variation in leaf N_mass of plants in Mount Gongga and to explore the possible causes.

Materials and Methods

Study area

Mount Gongga, located on the southeastern side of the Qinghai–Tibet Plateau in Southwest China (29°20′–30°00′N, 101°30′–102°10′E), exhibits remarkable differences in terrain and climate between its eastern and western slopes. The elevation of the eastern slope of Mount Gongga varies from 1100 m (Dadu River valley) to 7600 m. This study was conducted in the Hailuogou region of the eastern slope. Two meteorological observatories are located in the Hailuogou region (at 1640 m and 3000 m). The mean annual precipitation (MAP) and MAT recorded by the two observatories are listed in Table 1. MATs at the sample sites were obtained through linear interpolation on the basis of temperature data recorded. In addition, Zhong et al. used precipitation records from the two observatories, combined with the regional hydrology data, and suggested that rainfall increases with increasing altitude on the eastern slope of Mount Gongga [17].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Mean annual precipitation (MAP) and mean annual precipitation from 2 meteorological observatories in Hailuogou on the eastern slope of Mount Gongga (Zhong et al. 1997).

https://doi.org/10.1371/journal.pone.0044628.t001

Mount Gongga has an intact and continuous vertical vegetation spectrum that can be observed over a vertical range of 4900 m, from the subtropical arid Dadu River valley to the snowline. A subtropical evergreen broad-leaved vegetation occurs between 1100 and 2200 m with semi-arid shrubs and grasses in the valley below 1500 m. The dominant shrubs and grasses are Cymbopogon liangshanensis, Digitaria sanguinalis, Debregeasia longifolia, Achnatherum pappiforme, Artemisia argyi, Buddleja asiatica, and Euptelea pleiospermum; Subtropical evergreen broad-leaved plants grow on the slope with the dominant species being Cinnamomum camphora, Cyclobalanopsis gambleana, Morus australis, Pyracantha crenulata; broad-leaved mixed forests distributes from 2200 m to 2800 m with Cyclobalanopsis gambleana, Betula spp., Buddleja asiatica, Acer laxiflorum, Tetracentron sinense, Cinnamomum camphora, Cercidiphyllum japonicum, and Euptelea pleiospermum growing at lower altitudes, with Tsuga chinensis, Acer laxiflorum, and Betula spp. at higher altitudes; frigid dark coniferous forests occurs between 2800 m and 3600 m, and the dominant species are Tsuga chinensis, Picea spp., Abies fabric; alpine sub-frigid shrub and meadow vegetation is from 3600 m to 4200 m with a variety of Rhododendron spp., Spiraea spp., Salix spp., Fastuca spp. and Carex finitima; alpine frigid meadow vegetation is between 4200 m and 4600 m with dominant species being Polygonum viviparum, Rhodiola spp., Potentilla fruticosa, Kobresia spp., Saxifraga densifoliata and Anaphalis lactea; an alpine frigid sparse grass and desert zone is from 4600 m to 4800 m; an alpine ice and snow zone is above 4900 m.

Results

Variations in leaf N_mass across all species and between functional groups

Leaf N_mass of Mount Gongga's plants varied greatly, ranging from 2.8 to 59.1 mg/g, across all species. The arithmetic mean N_mass ± SD of overall species, seed plants, ferns, and mosses were 22.2±11.9 mg/g (n = 291), 22.5±11.8 mg/g (n = 269), 25.2±10.2 mg/g (n = 9), and 10.0±4.8 mg/g (n = 13), respectively (Fig. 1). One-way ANOVA showed significant differences in leaf N_mass between mosses and seed plants and between mosses and ferns (p = 0.000).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 1. Arithmetic mean values ± SD of leaf N_mass across different plant functional groups (different letters indicate significant difference at 0.05 level).

https://doi.org/10.1371/journal.pone.0044628.g001

In seed plants, leaf N_mass was significantly higher in C₃ species (24.2±11.5 mg/g, n = 252) than in C₄ species (12.8±5.1 mg/g, n = 21) (p = 0.000). Leaf N_mass was lower in perennial herbs (21.3±12.4 mg/g, n = 119) than in annual herbs (24.0±11.0 mg/g, n = 20), but the difference was not significant (p = 0.35). Leaf N_mass was slightly higher in woody plants (23.5±10.3 mg/g, n = 136) than in herbs (21.8±12.1 mg/g, n = 139), but the difference between them was not significant (p = 0.34). Leaf N_mass was remarkably lower in evergreen woody plants (15.6±8.6 mg/g, n = 19) than in deciduous woody plants (24.3±10.1 mg/g, n = 117) (p = 0.001).

Altitudinal pattern of leaf N_mass

For all species pooled together, leaf N_mass showed a significant and nonlinear altitudinal trend, with N_mass increasing below about 2200 m and then decreasing with increasing elevation above the altitude (R² = 0.243, p = 0.000) (Fig. 2a). Moreover, most plant groups, including seed plants, fern, C₃ plants, herbs, woody plants, annual herbs, perennial herbs and deciduous woody plants, displayed similar altitudinal trends in leaf N (Fig. 2b–f). In their respective sampling range, leaf N_mass in mosses decreased above 2200 m, and leaf N_mass increased in C₄ plants below 2200 m (Fig. 2b, 2c). For evergreen woody plants, leaf N showed a monotonic decline with increasing altitude (Fig. 2f). Fig. 2g, Fig. 2i and Fig. 2j respectively shows that after excluding evergreen woody plants, mosses and C₄ plants, the remaining plants still had trends similar to the overall altitudianl trend for all species pooled.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 2. Variation in leaf N_mass with altitude.

(a) the red line indicates all samples (R² = 0.243, n = 460); (b) those cross dots and the red regression line are seed plants (R² = 0.230, n = 432), those full diamonds and the black solid line represent ferns (R² = 0.397, n = 12), and those empty diamonds and the dash line indicate mosses (R² = 0.651, n = 16); (c) those cross dots and the red line are C₃ plants (R² = 0.302, n = 408), and those empty diamonds and the dash line indicate C₄ plants (R² = 0.160, n = 31); (d) the red line and the cross dots are all herbs (R² = 0.213, n = 230), the black line and those diamond are all woody plants (R² = 0.264, n = 214); (e) the red line and those cross dots are perennial herbs (R² = 0.242, n = 196), the black line and those diamonds are annual herbs (R² = 0.045, n = 34); (f) those cross dots and red line indicate deciduous woody plants (R² = 0.242, n = 185), the blue line and those diamond dots with a cross are evergreen woody plants (R² = 0.218, n = 29); (g) the cross dots and the red line are all plants excluding evergreen woody plants (R² = 0.237, n = 431); (h) the cross dots and the red line are all plants excluding mosses (R² = 0.228, n = 444); (i) the cross dots and the red line are all plants excluding C₄ plants (R² = 0.315, n = 429).

https://doi.org/10.1371/journal.pone.0044628.g002

Leaf N_mass versus MAT

Leaf N_mass showed a significant nonlinear relationship with MAT for all plants species pooled together (p = 0.000; Fig. 3a). Leaf N_mass increased until MAT decreased to about 8.5°C and then decreased with decreasing temperature. Leaf N_mass of seed plants, woody plants, herbs, deciduous woody plants, and C₃ plants all showed an N_mass–MAT trend similar to that of all plants pooled together (p = 0.000). Leaf N_mass of C₄ plants correlated negatively with MAT (p = 0.032). For mosses and evergreen woody plants, leaf N_mass decreased with a decrease in MAT (p = 0.000 and p = 0.001, respectively; Fig. 3b, c).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 3. Correlations between leaf N_mass and mean annual temperature (MAT).

(a) all plants (R² = 0.229, n = 460), (b)moss (R² = 0.651, n = 16) and (c) evergreen woody plant (R² = 0.218, n = 29).

https://doi.org/10.1371/journal.pone.0044628.g003

Discussion

Körner (1989) reported that herbaceous plants displayed an increasing trend in leaf N_mass with increasing altitude, but N_mass in evergreen woody plants remained stable. Our observations on Mount Gongga are different from those of Körner (1989). For most herbs, we found that leaf N_mass first increased and then decreased (Fig. 2e). For evergreen woody plants, we observed a reduction in leaf N_mass with altitude. Limited species and samples in the study by Körner (1989) might be responsible for the inconsistency between his findings and ours.

This study shows that about 2200 m elevation was a turning point, which determined the change direction of leaf N with altitude, and the corresponding MAT of the elevation was about 8.5°C (Fig. 3). There are certain comparability between the findings of this study and the observation by Reich and Oleksyn [3]. They found that leaf N decreased with increasing temperature when MAT was greater than 5–10°C but increased with increasing temperature when MAT was lower than 5–10°C.

Leaf N may show opposite responses to temperature: on one hand, leaf N increases as temperature decreases because high leaf N can offset reduced biochemical reaction rates caused by the diminished efficiency of N-rich enzymes at low temperatures (the temperature–plant physiological hypothesis, TPPH); on the other hand, low leaf N may be favored by cold climates because low temperature reduces the rates of decomposition and mineralization of organic matter, resulting in low soil-N availability [3]. Moreover, low temperature may suppress root nutrient uptake (the biogeochemical hypothesis, BH). The altitudinal variation in leaf N may be the result of the competition between the two mechanisms (Fig. 2): except for C₄ plants, mosses and evergreen woody plants, when MAT was greater than 8.5°C, leaf N of all plants increased with decreasing temperature, suggesting that the influence of TPPH exceeded that of BH; yet, when MAT was lower than 8.5°C, leaf N of all plants declined with decreasing temperature, suggesting that the effect of TPPH was less than that of BH. For the evergreen woody plants, leaf N linearly decreased with increasing altitude (decreasing temperature) (Fig. 2d), and it suggests that in most cases, the influence of BH mechanism on their leaf N may exceed that of TPPH, irrespectively of whether MAT is above or not 8.5°C. For the mosses, because all samples were from the sites above 2200 m (or below 8.5°C), influence of BH mechanism may exceed that of TPPH when MAT was below 8.5°C. How the leaf N of mosses responses to temperature when MAT is above 8.5°C remains unclear. Similarly, for C₄ plants, we can only suggest that the role of TPPH mechanism may be greater than that of BH when MAT was above 8.5°C, and we do not know the response of C₄ plants to temperature below 8.5°C.

Precipitation is also an important factor that influences the biogeographic pattern of leaf N_mass. Han et al. [5] analyzed the leaf N of 1900 plant species across China and suggested that precipitation explained leaf N variation to a greater extent than temperature. On the eastern slope of Mount Gongga, precipitation is expected to increase with elevation [17]. We believe that precipitation played a role in the altitudinal trend in leaf N_mass because precipitation increased remarkably with elevation on Mount Gongga. MAP is expected to be greater than 2000 mm when elevation is greater than 3000 m. High soil-water content contributes to the formation of an anaerobic environment, resulting in reduced rates of decomposition and mineralization of organic matter. Consequently, low leaf N_mass occurs at higher altitudes. However, we were not able to determine the effect of precipitation because reliable precipitation data were not available.

Species compositional shifts can also affect the overall biogeographic pattern of N_mass for all species pooled, because different plants or plant functional types may show significant differences in leaf N (the species composition hypothesis, SCH) [3]. Several comprehensive investigations in China [5]–[8], [18] favored SCH. He et al. [7] suggested that at the biome scale, temperature affects leaf N mainly through a shift in plant species composition rather than by temperature itself. Although large differences in leaf N occurred across functional groups of Mount Gongga's flora, leaf N increased up to an elevation of about 2200 m, and then decreased with increasing elevation, for overall species and for different plant groups except evergreen woody plants, C₄ plants and mosses (Fig. 2). In fact, only evergreen woody plants displayed an explicit decline trend across the altitude gradient, the altitudinal trend in mosses and C₄ plants cannot be ascertained. Mosses show similar declining trend of leaf N as the overall species and other plant groups, at least in their distribution range of above 2200 m (Fig. 2b). No mosses were sampled below 2200 m perhaps because the dry and hot climate in the valley does not benefit the growth of mosses. Thus we could not ascertain whether the trend of leaf N in mosses is different from those in other plant groups below 2200 m. For C₄ plants, leaf N also showed the same trend as other plant groups below 2200 m (Fig. 2c). No C₄ plants were collected above 2200 m mainly because too low temperature at high altitudes does not favor C₄ growth. The altitudinal patterns of leaf N in C₄ plants and mosses were not clear because their samples distributed a limited elevational range. On the eastern slope of Mount Gongga, evergreen woody plants are distributed from 1500 m to 4200 m [19]. Especially at elevations above 2800 m, evergreen woody plants have a high biomass ratio in the community. Still, the number of evergreen species is small, and generally, only 1–3 evergreen woody species exist at each elevation. Thus, evergreen woody plants did not have great contribution to the overall altitudinal pattern of leaf N for all species pooled. Compared with the altitudinal trend in leaf N for all species pooled, after excluding evergreen woody plants, the trend for remaining plants was almost unchanged (Fig. 2g). Similarly, the number of C₄ plants and mosses also are very small at each altitude; they did not show great influences on the overall altitudinal trend in leaf N because the remaining plants still had the trends similar to all species pooled after excluding C₄ plants and mosses (Fig. 2i and 2j). The results may suggest that species compositional shifts did not play a major role in the overall altitudinal pattern of leaf N for all species pooled.

In conclusion, the general pattern of leaf N in Mount Gongga's plants was that leaf N first increased up to an altitude of 2200 m, with a corresponding mean annual temperature (MAT) of 8.5°C, and then decreased with increasing elevation. Temperature may be one of influential factors on the altitudinal pattern of leaf N given that the linear relationship between temperature and altitude. Other environmental factors, such as precipitation, air pressure, and herbivores, may also have effects on the altitudinal pattern, but these effects was not explored here for lack of related data. The future studies should address these environmental controls. In addition, more investigations in other mountains are needed to test the altitudinal pattern of leaf N_mass revealed here.

Acknowledgments

We would like to thank Ma Yan for analyzing leaf N at the Stable Isotope Laboratory of the College of Resources and Environment, China Agricultural University.