【摘要】 在生境要素及植被发育程度相似的黔中喀斯特石漠灌丛林内,一次性完成12个200 m2样地及其中96个25 m2样方的调查,采用改进的巢式样方法研究植物多样性指数值随样地面积增加的变化规律,并为确定植物多样性研究的最小调查样地面积提供科学依据。结果表明:随着有序样地面积增大,灌木植物Patrick指数、Gleason指数及Margalef指数值增加,Shannon指数、Pielou指数及Alatalo指数值先增后减,样地间6种植物多样性指数值的差异性显著; Simpson指数及McIntosh指数值也先增后减,但样地间2种植物多样性指数值的差异性不显著。8种植物多样性指数按信息保存量≥80%划分,最小调查样地面积被划分为2种类型:100 m2样地包括Patrick指数,反映物种数量多少; 25 m2样地包括Gleason指数、Margalef指数、Simpson指数、Shannon指数、Pielou指数、Alatalo指数及McIntosh指数,反映植物多样性及均匀度。8种植物多样性指数值的Connor&McCoy模型拟合效果较好、de Caprariis模型及Michaelis-Menton模型拟合效果较差,Patrick指数和Margalef指数的Arrhenius模型及Logistic模型拟合效果也较好。8种植物多样性指数与96个样方的灌木植物平均高度和草本植物平均盖度的相关性显著,与草本植物平均高度的相关性不显著。主成分分析表明,立地要素、林分密度及林分空间属性对石漠灌丛林灌木植物多样性的贡献率分别为4.3%～11.7%、35.4%～67.9%和20.4%～60.2%,灌木植物多样性主要受林分结构及发育程度的影响。更多还原

【Abstract】 Stony desert shrub forests with similar habitat factors and vegetation development were investigated using twelve 200 m~2( 20 m×10 m) plots,including ninety-six 25 m~2( 5 m×5 m) subplots in central Guizhou karst desertification area. We assessed the impacts of plot size on plant diversity using the improved nest method,to provide a scientific framework for determining the minimum sampling plot size. Our results showed that an increase in the ordered sampling area in stony desert shrub forest caused the increase of the values of Patrick index,Gleason index and Margalef index,while caused the values of Shannon index,Pielou index and Alatalo index to increase first and then decrease. There was a significant difference in the values of six plant diversity indices among the different plot sizes. Simpson index and McIntosh index increased first and then decreased,but there was no significant difference between the two indices for the plot sizes.We classified the eight diversity indices into two groups according to the information retention rate≥80%: for the 100 m~2 plot size,the Patrick index that reflects the number of species was selected,and for the 25 m~2 plot size,the Gleason index,Margalef index,Simpson index,Shannon index,Pielou index,Alatalo index and McIntosh index that reflect plant diversity and evenness were selected. The Connor & McCoy model,with eight plant diversity index values,had a better fitting effect,the de Caprariis model and Michaels-Menton model had a worse fitting effect,and the Arrhenius model and Logistic model with Patrick index and Margalef index had a better fitting effect. There was a significant correlation between the eight plant diversity indices and the average height of shrubs and the average coverage of herbs,but no significant correlation between these indices and the average height of herbs across the 96 subplots. Results of principal component analysis showed that the contribution rates of site elements,stand density and stand spatial properties to the shrub diversity were 4.3%-11.7%,35.4%-67.9% and 20.4%-60.2%,respectively.The shrub diversity was mainly affected by forest structure and development level.更多还原