松嫩平原盐碱草地土壤种子库小尺度格局研究

用样线法、调查法、幼苗萌发法研究松嫩平原盐碱草地种子库在小尺度空间范围的水平分布格局。结果表明：①土壤种子库的密度为（1179．27±917．78）粒/m2，其中有757株植物幼苗，单子叶42．93％，双子叶57．07％，隶属15个科，23个属，37个种。②土壤种子库Shannon—Wiener指数H、Simpson指数、Margalet丰富度指数R、Pielou均匀度指数的范围分别为：1．42～2．03、0．70～0．78、1．50～3．33、0．77～0．89。选取出现频数较高的8个物种进行空间格局分析，离散系数的分析表明，只有苣荬菜在样方18显示出聚集分布，其他情况均为随机分布。Lloyd平均拥挤指数分析显示，繁缕、红足蒿、全叶马兰、寸草在该地区为聚集分布，而苣荬菜、碱蓬、稗草、羊草为随机分布。此外，在试验的取样间距（1m）前提下，8个物种在18个样方的空间相关性差别较大，8个物种在多数样方中显示空间负相关，表现正相关性的样方较少。

Study on Small-scale Heterogeneity of Soil Seed Bank in Saline-alkali Grasslands of the Songnen Plain

Soil seed banks play an important role in vegetation restoration. The small-scale heterogeneity of soil seed bank density was analyzed using the method of line transect, survey, seedling germination in saline-alkali grasslands of the Songnen Plain, China. Results revealed ： （ 1 ） the density of soil seed bank was （ 1 179.27 ± 917.78） （ seeds/m2 ） , among the soil seed bank there existed 757 seeds including 37 species, 23 genera 15 families, with 57.07% Dicotyledon and 42.93% Monocotyledon. （2） The range of Shannon-Wiener diversity index, Simpson index, Margalet richness index, Pielou＇ s evenness index in the soil seed bank was 1.42 - 2.03, 0.70 - 0.78, 1.50 - 3.33, 0.77 - 0.89, respectively. The species with high frequency in the soil seed bank were selected to carry on the analysis of the spatial pattern u- sing variance/mean ratio, Lloyd＇ s index of mean crowding and Moran＇ s I statistic of spatial autocorrelation. The result of dispersion coefficient indicated that only Sonchus brachyotus was aggregated in quadrats 18 and other species was random in rest of circumstances. The Lloyd＇ s index of mean crowding showed that Stellaria media, Artemisia rubripes, Kalimeris integrifolia and Carex duriuscula were aggregated in this area, while Sonchus brachyotus, Suaeda glauca,Echinochloa crusgalli ,Leymus chinensis were randomly distributed. Moreover, on the premise of the experimental patch size（ 1 m）, the spatial correlation of 8 species in the 18 quadrats differentiated greatly, the number of quadrats in which 8 species were negatively autocorrelated was large, and the number of quadrats in which 8 species were significantly autocorrelated was small.