---

作 者 ：徐兰英;李凤日*

期 刊 ：植物研究 2009年 29卷 3期 页码：333-338

关键词：分形；适应性群团抽样；白刺；乌兰布和沙漠；

Keywords:fractal, adaptive cluster sampling, Nitraria tangtorum, Ulan Buh Desert,

摘 要 ：以乌兰布和沙漠的白刺沙丘为主要研究对象，采用分形理论将遥感数据和实际调查数据相结合的方法，进行了适应性群团抽样技术的研究。基于分形理论建立描述白刺沙丘周长和面积关系的线性回归方程，利用周长估计了白刺沙丘的面积，估计精度达89.8%。在沙丘面积估计值的基础上，进行了适应性群团抽样分析。研究结果表明：当适应性标准（每个样方内白刺沙丘的面积）等于50 m²时，即接近含有白刺沙丘的样方均值时，适应性群团抽样的估计精度最高达到97.9%，适应性群团抽样效率是简单随机抽样效率的5.56倍。这一研究为沙漠地区具有群团分布特点的植被提供更有效的调查方法。

Abstract:Based on the fractal theory, the adaptive cluster sampling technique for Nitraria tangtorum sand dune was studied by using the method of integrated remote sensing data and the actual survey data in Ulan Buh Desert. The linear regression equation for describing the relationship between perimeter and area of N. tangtorum sand dunes was established based on fractal theory. On the basis of the regression equation, the area of N. tangtorum sand dunes was estimated using perimeter and the prediction precision is 89.8%. Based on the estimated value of the area, effect of the adaptive cluster samples was analysed. The results showed that the precision of the adaptive cluster sampling can reach 96.9% and the relative efficiency of adaptive cluster sampling is 5.65 times of the simple sampling when the adaptability standard value (each plot area of N. tangtorum sand dune) is 50 m² which nearly equals the sample average value of the dune. The results will provide more effective inventory method for vegetation with cluster distributing in desert.

---