基于区域剖分算法的土壤碳通量空间采样策略

发布时间：2018-05-09 05:57

本文选题：土壤呼吸 + 碳通量　；参考：《应用生态学报》2017年08期

【摘要】：由于土壤碳通量在空间分布上具有很强的异质性,传统的采样方法难以对区域土壤碳通量进行精确估算,因此确定适当的采样策略对区域土壤碳通量的估算具有重要意义.本文提出一种逐点递增式采样的区域剖分部署策略(RDPG):设定初始采样点,使用改进的凸包插值算法构造Delaunay三角网,根据邻近已知采样点插值计算三角形各边垂直平分线的交点的离散度,选择离散度最大的点作为新增采样点.采用该方法对变异系数为0.42~0.59的仿真试验区域进行多次试验,结果表明:在相同试验条件下,RDPG布局策略能够获得比随机采样和均匀采样策略更高的区域土壤碳通量估算准确度.RDPG方法考虑了区域土壤碳通量的空间异质性,提高了区域土壤碳通量拟合精度.
[Abstract]:Because of the heterogeneity of soil carbon fluxes in spatial distribution, it is difficult for traditional sampling methods to estimate soil carbon fluxes accurately, so it is very important to determine appropriate sampling strategies for estimating regional soil carbon fluxes. In this paper, an incremental sampling strategy for regional partition and deployment is proposed. The initial sampling points are set up, and the improved convex hull interpolation algorithm is used to construct the Delaunay triangulation network. Based on the interpolation of adjacent sampling points to calculate the dispersion of the intersection points of vertical bisections of each side of a triangle, the points with the largest degree of dispersion are selected as the additional sampling points. The method is used to test the simulation test area with a coefficient of variation of 0.42g 0.59. The results showed that under the same experimental conditions, the RDPG layout strategy could obtain higher accuracy of estimating regional soil carbon flux than random sampling and uniform sampling. RDPG method considered the spatial heterogeneity of regional soil carbon flux. The fitting accuracy of regional soil carbon flux was improved.
【作者单位】：浙江农林大学信息工程学院;浙江省林业生态工程管理中心;
【基金】：国家自然科学基金项目(61190114,61303236) 浙江省科技计划项目重大科技专项(2012C13011-1) 浙江省省院合作林业科技项目(2015SY12)资助~~
【分类号】：S153

【参考文献】