摘要
以三峡库区王家沟小流域为研究区,在对土壤样本的理化性质进行分析和测量的基础上,用均方根误差、平均相对误差、一致性系数三个精度评价指标对克里克插值和案例推理方法用于土壤养分预测的精度进行了评价,并对这两种方法所获得的土壤养分空间分布图的效果进行了比较.研究结果显示,与克里克插值方法相比,案例推理方法不仅获得了较好的土壤养分预测精度,还获得了较好的制图效果.尤其是在没有采样点分布的区域,案例推理方法获得的土壤养分空间分布图也具有更高的详细度.
Taking Wangjiagou small watershed of Three Gorges Reservoir Area as the researching zone and based on the analysis and measurement of soil physicochemical properties, prediction precisions of cased-based reasoning and Kriging interpolation were evaluated by root mean squared error, average relative er-ror and agreement coefficient. Moreover, the soil nutrient-distribution maps produced by the two methods were compared. The results showed that compared with the Kriging interpolation method, case-based rea-soning had higher precision and better mapping effect. In areas without soil sampling, in particular, case-cased reasoning gave more detailed spatial distribution maps of soil nutrients.
出处
《西南大学学报（自然科学版）》
CAS
CSCD
北大核心
2013年第9期105-110,共6页
Journal of Southwest University(Natural Science Edition)
基金
国家水体污染控制与治理科技重大专项课题(2012ZX07104-003)
西南大学博士基金(SWU112065)
关键词
克里克插值
案例推理
全氮
预测精度
制图效果
Kriging interpolation
case-based reasoning
total nitrogen
prediction precision
mappingeffect
