多种空间插值法在土地质量地球化学评价中对比分析——以四川省叙永县麻城镇和摩尼镇为例

A variety of spatial interpolation methods compared in the geochemical evaluation of land quality——a case study of Macheng and Moni town,Xuyong county,Sichuan province

使用R语言的空间统计包,综合采用分层克里金法、多变量协同克里金法和反距离加权平均法,对研究区1:10000土地质量地球化学调查评价项目表层土壤数据进行插值,进而进行土壤养分单指标和土壤质量地球化学综合评价。借鉴土地利用强度分析(LULCC)方法,对评价等级进行划分,分析不同插值方法的评价结果等级间的差异度、土壤质量地球化学综合等级类型转变强度和等级增量来源强度。最终探讨发现:“有样”单元赋反距离加权平均值与“无样”单元赋不同方法的插值结果相结合,能避免空间插值计算范围过大,导致的高值或低值数据被平均后较小的评价单元评价等级降低或提高的问题。

In this study,the methods of stratified kriging d,multivariate collaborative kriging and inverse distance weighted average(IDW)using the spatial statistical package of R language were comprehensively adopted to interpolate the surface soil data of 1:10000 geochemical survey and evaluation project of land quality in studing area,sichuan province,for geochemical comprehensive evaluation of earth quality.Based on land use intensity analysis(LULCC)method,the geochemical grade of land quality was analyzed,and the difference between the geochemical evaluation results of land quality with different interpolation methods and the conversion intensity of land quality grade type and the source intensity of grade increment were analyzed.Finally,it is found that the combination of the weighted average value of inverse distance assigned by"sample"element and interpolation results of different methods assigned by"no sample"element can be used to avoid large spatial interpolation calculation range,which makes the data of high value or low value be averaged,resulting in lower or higher evaluation grade of smaller evaluation units.