摘要:以吉林蛟河阔叶红松林的木本植物为研究对象，将30hm²的样地面积划分为5m×5m，10m×10m，20m×20m，25m×25m的连续取样单元，在4个不同尺度下分别统计各物种在每个取样单元中的有无，得到每个物种在不同尺度下的取样单元数。利用随机分布模型和负二项分布模型分析物种的多度分布，对比预测多度与观测多度讨论两个模型的科学性与实用性。结果表明：对于阔叶红松林而言，负二项分布模型在所有研究尺度上的预测精度都要优于随机分布模型。随机分布和负二项分布的模型预测误差随着研究尺度的增大而增大，因此选取较小的取样单元可以切实提高物种多度的预测精度。利用随机分布和负二项分布模型对多度较小的物种进行预测的效果要优于多度较大的物种。负二项分布模型适合用来模拟阔叶红松林的物种多度分布格局，并且模型的拟合效果受取样单元大小影响。



Abstract:Taking the broad-leaved pine forest in Jiaohe, Jilin as the research object, the sample plot area of 30 hm² was divided into continuous sampling units of 5 m×5 m, 10 m×10 m, 20 m×20 m and 25 m×25 m. The samples of each species in each sampling unit were counted under four different scales to obtain the number of sampling units of each species in different scales. The random distribution model and negative binomial distribution model were used to analyze the species abundance distribution, and the scientific and practical applicability of the two models were discussed by comparing the predicted abundance and observed abundance. The results showed that the prediction accuracy of the negative binomial distribution model was better than that of the random distribution model at all research scales for broad-leaved pine forests. The model prediction error of random distribution and negative binomial distribution increased with the increase of research scale, so choosing smaller sampling unit could improve the prediction accuracy of species abundance. The random distribution and negative binomial distribution models were more effective than the ones with greater abundance in predicting species. The negative binomial distribution model is suitable for simulating the species abundance distribution pattern of broad-leaved Korean pine forest and the fitting effect of the model is affected by the size of the sampling unit.





PDF全文下载地址:

https://www.ecologica.cn/stxb/article/pdf/stxb202010302785