长白山苔原草本植物入侵与土壤环境的关系

在对长白山高山苔原西坡样带内132个样方进行植被调查和土壤取样分析的基础上,应用冗余分析(RDA)和典型相关分析方法,探讨了草本植物入侵苔原带程度与苔原土理化性质及环境之间的关系。研究表明,长白山苔原带西坡草本入侵程度区域差异明显,可分为5个不同的入侵等级;海拔、坡度、全钾含量、粘粒含量、有机质含量等10种土壤环境因子与草本植物入侵程度明显相关。RDA分析表明土壤环境因子能解释93%的植物物种多度信息,影响草本植物入侵的主要土壤因子是有机质含量,粘粒含量和坡度;第一对典型变量说明有机质含量与牛皮杜鹃多度正相关,与大白花地榆多度负相关,粘粒含量则相反;第二对典型变量说明海拔、速效氮含量与笃斯越桔、长白老鹳草多度呈负相关。长白山苔原带西坡草本植物在空间分布上是离散的,呈斑块状。可见,草本植物入侵,对生境是有所选择的。土壤有机质含量与灌木多度呈正相关,说明在草本入侵过程中,土壤有机质含量会减少;或者有机质含量减少的地方,灌木生长退化,草本由此开始侵入定植。

Relationship between herbaceous plant invasion and soil properties in the Changbai Mountain Tundra

The Changbai Mountain tundra, which is located in the upper part of the Changbai Mountain volcanic cone (2000-2500m), is the only typical alpine tundra in eastern China, and because its environmental characteristics are similar to those of polar tundra, the Changbai Mountain tundra is highly valuable for scientific research. Herbaceous species invasion has been found in both polar tundra and alpine tundra, this phenomenon may caused by the climate change. In general, there is a close correlation between tundra vegetation and soil, the change of vegetation community structure directly or indirectly led to changes in soil properties. The progression of invasion resulted in a decrease in the organic matter content.The growth of tundra vegetation is affected by both temperature and soil composition (i.e., nutrients), whereas the composition of the plant community affects local soil processes. The vegetation of the Changbai Mountain tundra is undergoing significant changes in response to global climate change. On the western slopes of the mountain, herbaceous plants have become invasive and have expanded their distribution, whereas the importance of shrubs as the dominant plant type has been significantly reduced. The present study investigated the relationship between herbaceous plant invasion and soil properties in 100m×1600m transects and 132 plots in August 2014. The latitude, longitude, and elevation of the sample plots were measured using GPS, whereas slope measure, direction, and position were measured using a compass, and the plant community characteristics (species, height, number, coverage) and soil thickness were recorded in the field, whereas the physico-chemical properties (n=12) of sieved soil samples (0-10cm) measured in the laboratory. Redundancy and canonical correlation analyses indicated that the invasion degree of the Changbai Mountain tundra varied significantly and could be categorized into five invasion levels. In addition, ten of the soil factors (e.g., elevation, slope, Total K, clay content, organic matter content) were significantly correlated with the degree of herb invasion, and the samples with higher herbaceous invasion were mainly located in areas with lower elevations, greater slopes, relatively low mineral, organic nutrient, sand, and silt contents, and relatively high clay content. However, there was no obvious relationship between herbacious invasion and available nutrient contents (available N, available P, available K). Redundancy analysis indicated that soil factors explained 93% of the variation in plant species abundance and that the main soil factors affecting herb invasion were organic matter content, clay content, and slope, whereas shrub growth was mainly affected by organic nutrient content. Furthermore, the distributions of the herbaceous plants Deyeuxia angustifolia and Sanguisorba stipulata were related to soil thickness and clay content, respectively. Meanwhile, canonical correlation analysis indicated that soil factors and plant species abundance were significantly correlated and the habitats of different herbaceous species were significantly different. More specifically, first canonical variables indicated that organic matter content was positively correlated with Rhododendron chrysanthum abundance and negatively correlated with S. stipulate abundance, whereas the second canonical variables indicated that elevation and available nitrogen content were negatively correlated with the abundance of Vaccinium uliginosum and Geranium baishanense.