- 上一篇：江苏省生态空间过程与机理研究
- 下一篇：徐州市旅游业发展的困境与对策
摘要土壤有机质（Soil organic matter, SOM）是评价土壤质量的主要指标，选择 合适的样点布设模式与点面拓展方法对准确揭示区域 SOM 空间变异特征具有重 要意义。本研究以江西省余江县中部地区为案例区，基于 7 种样点布设模式，使 用普通克里格方法和两种图斑连接方法（分别连接土壤类型和土地利用图斑）获 得区域 SOM 空间分布特征，并通过验证样点对比各模式与方法的预测精度。结 果表明：普通克里格方法由于平滑效应强烈，预测精度最低；图斑连接方法考虑 了不同土壤类型和土地利用方式间 SOM 含量差异，预测精度大幅提高，其中连 接土地利用图斑要优于土壤图斑；在样点布设模式中，按照各类型样点 SOM 变 异系数比重布点的模式最优，可见 SOM 变异系数的差异对于合理分配样点数量 以提高研究区 SOM 预测精度有重要影响，而图斑面积则影响不大。综合考虑， 基于土地利用 SOM 变异系数的采样模式和土地利用图斑连接方法的组合（PLCv） 可实现对地形复杂区 SOM 变异特征的精确预测。69069
该论文有图 7 幅，表 1 个，参考文献 25 篇。
毕业论文关键词：土壤有机质 样点布设 空间预测
Spatial Variation Characteristics of Soil Organic Matter Based on Different Soil Sampling Designs
Abstract
Soil organic matter (SOM) is an important index to evaluate the quality of the soil. Selecting an appropriate and efficient soil sampling design and point-plane expanding method has important significance on accurately detecting spatial differentiation of regional soil organic matter (SOM). Taking the central part of Yujiang County in Jiangxi Province as study area, based on seven soil sampling design patterns, ordinary kriging (OK) and two different polygon-based methods, including linking SOM data of samp les to corresponding soil polygon and land use polygon, were used for revealing soil organic matter (SOM) variation, and their prediction uncertainty were compared in this study. Results showed that the prediction precision of ordinary kriging (OK) was the worst due to its strong smoothing effect. However, the prediction precision of the polygon-based methods improved owing to the consideration of great difference of SOM content between different soil types and land use patterns, and the method which linked SOM data of samples to corresponding land use polygon was superior to that which linked SOM data of samples to corresponding soil polygon. Among soil sampling setting methods, the method had the highest prediction which was designed by SOM variance coefficients. It showed that taking SOM variance coefficients into consideration played an important role in allocating sample points and improving the prediction accuracy, but the effect of polygon area was not significant. All things considered, the combination of the soil sampling design by SOM variance coefficients of different land use types and the polygon-based method which linked SOM data of samples to corresponding land use polygon, could precisely predict SOM variation characteristics in the regions with complex terrain.
Key Words: Soil organic matter (SOM) Sampling design patterns Spatial prediction
目 录
摘要 I
Abstract II
目录 III
图清单. Ⅳ
表清单. Ⅳ
1 绪论 1
1.1 研究意义