论文部分内容阅读
以高密度土壤养分采样数据为数据源,通过随机抽取生成不同采样尺度的样点数据,分析采样尺度对土壤养分空间变异特征分析的影响。研究结果表明:区域土壤养分预测均值随采样尺度减小呈下降趋势,而变异系数增加;养分空间分布的全局趋势随采样尺度增大而增强,但不影响半方差模型;当采样尺度较大,样点间自相关较弱时,相对较少的样点也能满足区域统计参数估测分析需要,但不能用于空间变异特征和插值分析;当样点数大于最佳采样数时,养分统计参数、空间变异特征和插值分析随着采样尺度减小而精度提高,当采样尺度达到0.2左右时,能够满足中等空间变异的土壤养分空间插值分析需要;样点空间布局对相关距和空间插值分析精度的影响比采样尺度本身更为显著。
Taking high-density soil nutrient sampling data as data source, sampling data of different sampling scales were randomly generated to analyze the influence of sampling scales on the analysis of spatial variations of soil nutrients. The results showed that: the mean value of soil nutrient prediction decreased with the sampling scale and the coefficient of variation increased; the global trend of nutrient spatial distribution increased with the sampling scale, but did not affect the semi-variance model; when the sampling scale was larger, When the autocorrelation between samples is weak, the relatively few samples can meet the needs of regional statistical parameters estimation and analysis, but can not be used for spatial variability and interpolation analysis. When the sampling points are greater than the optimal sampling number, , Spatial Variation and Interpolation Analysis accuracy increases with decreasing sampling scale. When the sampling scale reaches about 0.2, it can meet the need of medium nutrient spatial interpolation analysis of medium spatial variability. The spatial distribution of sampling points has a great influence on the accuracy of correlation space and spatial interpolation analysis The impact is more pronounced than the sampling scale itself.