Guo Hua

Effects of ecological factors on plant communities of Ziwuling Mountain, Shaanxi Province, China

Abstract This paper explores the relationships between environmental factors and community patterns. In 60 plots in the Malan Forest Region of the Loess Plateau, 17 variables were used as indicators of spatial position, soil and topography characteristics. Plant community types were analyzed by TWINSPAN (Two-way Indicator Species Analysis). Significant environmental factors were selected by forward selection and Monte Carlo tests. The relationship between communities and environmental factors was identified by DCCA (Detrended Canonical Correspondence Analysis). Partial CCA (Canonical Correspondence Analysis) partitioned observed compositional variation into components explained by environment, locality and covariation between them. The results showed that: (1) the plant communities can be divided into 13 types including 4 different successional stages. (2) Salt and alkali-hydrolyzable nitrogen are the significant factors related to plant community distribution at early stage. Elevation and humus thickness are the significant factors in the transitional stage. Elevation, slope, aspect, thickness of forest floor, humus thickness and pH are the significant factors for the sub-climax stage. Elevation, slope, aspect, pH and available phosphorus are the significant factors for the climax stage. (3) The ecological characters and the distribution patterns of plant communities in different successional stages are consistent with environmental gradients. (4) As succession proceeds, the impact of environmental conditions on plant communities increases, while the total influence of spatial location, and the interaction between the spatial location and environmental factors decrease.

Predicting plant traits and functional types response to grazing in an alpine shrub meadow on the Qinghai-Tibet Plateau

The identification of easily measured plant functional types (PFTs) that consistently predict grazing response would be a major advance. The responses to grazing of individual traits and PFTs were analyzed along a grazing gradient in an alpine shrub meadow on the Qinghai-Tibet Plateau, China. Three response types were identified; grazing increaser (GI), grazing decreaser (GD), and neutral (NE) for both traits and PFTs. Seven traits were measured: plant height, economic group, cotyledon type, plant inclination, growth form, life cycle, and vegetative structure. The first five were significantly affected by grazing. Ordinal regressions for grazing response of the seven traits showed that the best single predictors of response were growth form (including the attributes “Scattered”, “Bunched” or “Closely Bunched”), and plant inclination (“Rosette”, “Prostrate”, or “Erect”), followed by economic group (“Shrub”, “Grass”, “Sedge”, “Legume”, “Forb”, or “Harmful”) and plant height (“Tall”, “Medium”, or “Small”). Within the four optimal traits, the summed dominance ratio (SDR) of small plants, forbs, rosette and bunched plants, invariably increased, while that of tall plants, shrubs, grasses, and erect plants decreased, when grazing pressure was enhanced. Canonical correspondence analysis (CCA) identified eleven explanatory PFTs based on 195 defined PFTs, by combining the different attributes of the four optimal traits. Among explanatory PFTs, the most valuable in predicting the community response to grazing were Tall×Shrub×Erect×Scattered and Small×Forb×Rosette, as these have the closest connections with grazing disturbance and include fewer species. Species richness, diversity, and community evenness, did not differ among grazing treatments because turnover occurred in component species and their relative abundances along the grazing gradient. We have demonstrated that a minimum set of PFTs resulting from optimal individual traits can provide consistent prediction of community responses to grazing in this region. This approach provides a more accurate indicator of change within a changing environment than do univariate measures of species diversity. We hope to provide a link between management practices and vegetation structure, forming a basis for future, large scale, plant trait comparisons.