Aims Recent theoretical and empirical work suggests that species diversity enhances the primary productivity and stability of communities. However, the relationships between the diversity of different species types (i.e., total species, response species, effect species and common species), the special function of ecosystems, and the potential mechanism driving stability remain unclear. Our objective is to address the question by comparing the diversity effect of these different types on aboveground net primary productivity (ANPP) and community stability.
Methods Our experiment was conducted in alpine meadow at the Haibei Research Station of the Chinese Academy of Sciences from 2007 to 2011. We used a split-plot design with clipping treatment in the whole plot using three clipping levels (stubbled 1 cm, 3 cm and unclipped). Subplots were treated with fertilizer (urea 7.5 g·m^-2·a^-1+ ammonium phosphate 1.8 g·m^-2·a^-1 and unfertilized) and watering (20.1 kg·m^-2·a^-1 and unwatered).
Important findings We observed that the diversity of different species types affected ecosystem functioning differently. ANPP was mainly affected by the diversity of response species and effect species, whereas community stability was largely affected by that of common species. The maintenance of stability depended on increasing diversity in common species, and the potential mechanism was the portfolio effect. Both the over-yielding effect and asynchrony effect, however, had no influence on stabilizing the community. Clipping had enormous effects on the diversity of total species, whereas the changes in diversity of response species mainly connected with resource availability. Thus, clipping and fertilization had reverse effects on species diversity, ANPP and stability, i.e., the former increased both species diversity and stability and decreased ANPP, while the latter had opposite effects on them. Our results suggest that ANPP is driven by the diversity of a few effect species because they have a great influence on ANPP, while stability is driven by the diversity of a large number of common species because they can coexist stably in the community. The portfolio effect is the main mechanism of the diversity-stability relationship. The diversity effect of different species differs among each other; therefore, in terms of specific ecosystem functioning, we infer that “functional identity” of species in community is more important than diversity per se and it may be incorrect if we did not discriminate when defining the relationship between species diversity and ecosystem function in any situation.

Aim Seasonal snow cover may play an important role in litter decomposition in subalpine forest, but little information has been available on the effects of different snow depths on litter mass loss. Therefore, our objective was to characterize the rate of mass loss during litter decomposition under different snow cover conditions.
Methods A field litterbag experiment was conducted in an alpine fir forest of western Sichuan, China from October 2010 to October 2012. Samples of air-dried leaf litter of fir (Abies faxoniana), larch (Larix mastersiana), cypress (Sabina saltuaria) and birch (Betula albosinensi) were placed in nylon litterbags (20 cm × 20 cm, 10 g per bag), and the edges were sealed. Mass loss rates were investigated at different critical stages (onset of freezing period, deep freezing period, thawing stage and growing season) during two years of decomposition.
Important findings Two years of decomposition resulted in mass loss of 33.98%-39.55% for fir, 46.49%-48.22% for birch, 42.30%-44.93% for larch and 40.34%-43.84% for cypress. Compared with no snow cover, thick snow cover marginally increased mass loss by 1.73%-5.57%. The k values from the Olson decomposition constant for three coniferous litters (fir, larch and cypress) were highest under thick snow cover and lowest under no snow cover. However, the k value for birch, a broad-leaved species, showed the ranked order of no snow cover﹥thin snow cover﹥thicker snow cover﹥thick snow cover﹥medium snow cover. Although snow cover did not significantly promote decomposition of birch litter during growing season in the second year, snow cover significantly promoted decomposition of fir, larch and cypress litters at all investigated stages in two years. Additionally, mass loss during snow cover period in the first year accounted for 42.5%-65.5% of the entire first year decomposition, indicating that seasonal snow cover dramatically changed the decomposition of leaf litters in winter, especially at the deep frozen stage. In conclusion, litter decomposition in this alpine forest would be delayed by the decrease of winter snow cover predicted with climate change. Compared with broad-leaved litter, coniferous litter could display stronger responses to such changes of snow cover.

Aims Litter decomposition of understory species is one of the essential components in material cycling and other important processes in alpine/subalpine forest ecosystems. Natural snow patches with different snow depths in winter could play an important role in litter decomposition due to significantly different freeze-thaw characters, but little information has been available. Therefore, our objective was to understand the effects of snow patches on litter decomposition of dwarf bamboo (Fargesia nitida) and salix (Salix paraplesia), two representative understory shrubs in alpine forest.
Methods A field experiment using litterbags was conducted in an alpine forest in western Sichuan, China. Samples of air-dried leaf litter were placed in nylon litterbags, and the litterbags were placed on the forest floor along a snow depth gradient from forest gap to canopy cover. Five snow patches with different snow depths were selected: thickest snow cover patch (SP1), thicker snow cover patch (SP2), middle thick snow cover patch (SP3), thinner snow cover patch (SP4) and no snow cover patch (SP5). Mass loss was measured at five critical periods as decomposition proceeded (onset of soil freezing period, soil freezing period, soil thawing period, early growth period and later growth period) of the first year of decomposition.
Important findings Mass loss of dwarf bamboo and salix litters in the freeze-thaw season accounted for (48.78 ± 2.35)% and (46.60 ± 5.02)% of the first year of litter decomposition, respectively. Both litters displayed higher mass loss rate under the patches with snow cover compared with no snow patch in the freeze-thaw season, but showed higher mass loss rate under SP5 in the growth season. Over the first year of decomposition, although mass loss rate of bamboo litter increased with the increase of winter-snow depth, salix litter showed the highest value under SP4 and lowest value under SP5. In addition, correlation analysis indicated that mass loss rate in the freeze-thaw season was positively related to daily mean temperature and negative cumulative temperature, whereas mass loss rate in the growth season was not related to any investigated temperature factors. However, 1-year mass loss rate was significantly related to daily mean temperature and negative/positive cumulative temperature. These results indicated that change of snow pattern would have significant effects on understory litter decomposition in the alpine forest in the scenario of warmer winters, but the effects could be various in different kinds of litter.

Aims Our objective was to explore seasonal variations of leaf C:N:P stoichiometry in plants with the same growth form.
Methods We chose six shrubs in the desert of the Alxa Plateau in north-central China (Zygophyllum xanthoxylum, Nitraria tangutorum, Reaumuria soongorica, Ceratoides lateens, Oxytropis aciphylla and Ammopiptanthus mongolicus) and observed their phenological stages from May to October 2010. Leaf samples were collected during this period, and leaf C, N and P contents and C:N:P stoichiometry were monitored.
Important findings Seasonal dynamics of leaf C, N and P contents and C:N, C:P and N:P mass ratio in the six shrubs were species-specifics, and the variation of leaf C, N and P and C:N, C:P and N:P mass ratios in different species were also dramatically different. Based on variation analysis among different seasons within species, there were less seasonal dynamics in C and N contents and C:N mass ratio than the other three parameters including P contents and C:P and N:P mass ratios. The range of values of coefficient of variation (CV) for C and N contents and N:P mass ratio was 0.60%-10.20%, 6.09%-20.50% and 5.87%-18.78%, respectively. For the other three parameters, the range of CV values for P content was 16.43%-43.43%, and C:P and N:P mass ratios were 8.48%-31.95% and 11.86%-40.73%, respectively. With the comprehensive analysis based on the total variation (resulting from two factors: season and species) for each parameter in these six shrubs, the rank of CV for each parameter was P (28.85%) > C:P (25.02%) > N:P (22.18%) > N (14.22%) > C:N (12.48%) > C (4.62%). Factorial analysis of variation for each parameter, with sampling date (season) and species as independent variables, showed that leaf C and N contents and leaf C:N, C:P and N:P mass ratios were mainly determined by plant species. For leaf P contents, it was the sampling date (season).

Aims Understanding the ecological restoration functions of the Coriaria nepalensis-Erianthus rufipilus community in phosphorus-enriched degraded mountain areas can guide ecological restoration and community assembly in degraded mountain areas. Our aim was to assess the degree of ecological restoration of the Coriaria nepalensis- Erianthus rufipilus community in three aspects: community assembly process and non-point source pollution control of dominant species and the community.
Methods We conducted a point pattern analysis of three dominant species (Coriaria nepalensis, Erianthus rufipilus and Eupatorium adenophorum) in a plot located on a degraded mountain of the phosphorus-enriched area in the Lake Dianchi Watershed in Kunming, Yunnan Province, China. The effects of dominant species and the reconstructive community on surface runoff and soil nutrient loss were monitored using runoff plots in the wet season. We compared surface runoff and nitrogen and phosphorus loss in these plots under natural rainfall conditions with reference plots.
Important findings Coriaria nepalensis, Erianthus rufipilus and the reconstructive community of these two species effectively decreased surface runoff, soil loss and nutrients loss. The individual number of dominant species was low, and no significant associations were found in species pairs, which suggest each species’ distribution patterns may be associated with specific habitats. These results indicate that the reconstructive community of C. nepalensis and Erianthus rufipilus could lead to effective ecological restoration in the phosphorus-enriched degraded mountain area of the Lake Dianchi watershed, but individual numbers of dominant species should be increased in this reconstructive community in order to develop community structure and control non-point source pollution loss.

Aims Spatial heterogeneity in soil nutrients is common in nature, and clonal plants are supposed to be able to better use spatially heterogeneously distributed soil nutrients. Although scale and contrast are two of the most important elements of spatial heterogeneity, few studies have tested effects of scale and contrast of soil nutrient heterogeneity on the growth of clonal plants.
Methods We conducted a greenhouse experiment in which we grew individual ramets of a rhizomatous clonal plant, Bolboschoenus planiculmis (synonym Scirpus planiculumis), in heterogeneous conditions consisting of nutrient-rich and nutrient-poor patches. The experiment had two levels of patch scale (large vs. small patch) crossed with two levels of patch contrast (high vs. low contrast), thereby consisting of four treatments. In each treatment, the total area of the nutrient-rich patches and that of the nutrient-poor patches were the same, and the total amount of soil nutrients was also the same in all the four treatments.
Important findings At both clone (whole plant) level and patch level, patch scale, contrast or their interaction did not significantly affect biomass, number of ramets, total rhizome length or number of tubers of B. planiculmis. At the patch level, however, biomass, number of ramets, total rhizome length and number of tubers of B. planiculmis were significantly greater in the nutrient-rich patches than in the nutrient-poor ones, whereas spacer length (i.e., distance between adjacent ramets) was smaller. Such effects depended on neither patch scale nor patch contrast. Therefore, when growing in environments with heterogeneous soil nutrients, B. planiculmis was able to shorten its spacer length and likely also increased branching intensity of the rhizomes, such that more ramets and tubers (i.e., potential ramets) were placed in nutrient-rich patches. These responses are generally considered one aspect of the foraging behavior of clonal plants and are thought to be adaptive. However, the patch scale or contrast designed in the present experiment could not affect the placement of the ramets in the heterogeneous environments. We presume that the responses of B. planiculmis are likely different if a wider range of scale and contrast is used and thus future studies to test effects of scale and contrast of resource heterogeneity should consider a wider range of scale and/or contrast.

Aims The invasive alien plant Solanum rostratum has different phenotypes in different distribution areas in China. Our objectives were to determine the reason for the phenotype variation and to find typical traits of the phenotype variation.
Methods Seeds of S. rostratum collected from nine areas of distribution in China were sowed in a common garden. We investigated 10 phenotypic traits and used one-way ANOVA, principal component analysis, UPGMA cluster analysis and correlation analysis to analyze results.
Important findings Inter-population variation was significant in all morphological characters (p < 0.01), indicating that the phenotypic variation had a genetic foundation. The coefficient of variation of reproductive organs (CV = 18.2%) was higher than that of vegetative organs (CV = 9%). Analysis of CV and the principal component analysis of phenotypic traits both indicated that the traits of radius, corolla diameter, internode space and plant height were the main factors accounting for the phenotypic variations. According to UPGMA cluster analysis, the nine populations could be divided into three groups. These cluster results were not due to geographic distances. Altitude had greater influence on phenotypic traits than longitude and latitude.

Aims Alhagi sparsifolia, which occurs in arid regions, is commonly considered a type of sun plant. However, the effects of shade on A. sparsifolia are complex, because in addition to reduced light intensity, shade often changes other environmental factors. Our goal is to study the impact of different shading on the physiological form-especially in terms of water-of A. sparsifolia.
Methods We used natural light, moderate shade (70% natural light) and severe shade (30% natural light) and after 90 days observed reactions of water status, leaf morphology and stress resistance indicators in A. sparsifolia.
Important findings With the increase in shading, there are rises in soil moisture content, water potential, stomatal conductance, specific leaf area, chlorophyll (Chl) and carotenoid content in A. sparsifolia, as well as declines in leaf thickness, proline, malondialdehyde, soluble sugar content and Chl a/Chl b. Moderate shade for A. sparsifolia can reduce the temperature and increase the humidity. This can improve the habitats of A. sparsifolia by avoiding damage caused by high temperature, high light and low water potential and then promoting the growth of plants, although long-term shade is detrimental. Therefore, we recommend protecting A. sparsifolia and promoting its growth with short-term shade, especially during the high temperatures and strong light at noon in the summer.

Aims Our objectives were to optimize the positions for root sampling of winter wheat and maize and to simulate the distribution of root length density (RLD) throughout the root zone profile in the North China Plain.
Methods Soil cores were taken at different locations in fields of winter wheat and summer maize at the grain-fill stage, and results were compared to decide the positions for root sampling. The bulk density parameter was used to modify the Gerwitz and Page model to increase the accuracy in simulating the distribution of RLD throughout the soil profile.
Important findings There was a larger spatial difference in RLD in the 0-10 cm soil layer for winter wheat. The difference became smaller below the 10 cm soil layer. For the top 20 cm soil layer, RLD on the row was greater than that between two rows. However, below 20 cm, the situation was reversed, i.e., RLD between two rows was greater than that on the row, indicating the strong proliferation ability of the root system of winter wheat. For maize, RLD sampled at the stem was greater than that sampled 10 cm away from the stem, and much greater than that sampled 20 cm away from the stem for the 0-10 cm soil layer. Below the 10 cm soil layer, RLD was highest sampled 10 cm away from the stem. RLD sampled at the stem around the 30-50 cm soil layer was the lowest among the three positions. The results indicated that in the top soil layer, RLD immediately under the plants was the highest, and RLD decreased with increased distance away from the plant stem. However, with increased soil depth, root proliferation reduced the differences among locations. The high bulk density of soil pan around 10-30 cm significantly influenced the shape of the curve of RLD distribution along the soil profile. Introducing the bulk density parameter into the Gerwitz and Page model significantly reduced the errors of simulated and measured RLD throughout the root zone profile.