基于GIS的近海沉积物痕量元素空间分布特征插值方法选择  

Research on the space distribution features of trace metal in the offshore sediments based on GIS

作　　者：段志鹏[1] 林进清[2] 利锋[1] 甘华阳[2] 梁开[2] 吴昌华[1] 曾祥云[1] 刘鑫[2] 霍振海[2] 

机构地区：[1]华南理工大学土木与交通学院,广东广州510641 [2]广州海洋地质调查局,广东广州510760

出　　处：《海洋环境科学》2014年第2期262-268,共7页Marine Environmental Science

基　　金：国家自然科学基金(41001341);国家海洋地质保障工程(1212010914020)

摘　　要：基于GIS空间分析模块,对雷州半岛近海沉积物实测同步提取态金属砷(simultaneously extracted metals As,SEMAs)数据分别使用反距离权重法(inverse distance weighted,IDW)、张力样条法(tension spline,TS)和普通克里格法(ordinary kriging,OK)进行空间插值;通过交叉验证法和插值图像对比分析,对各插值方法的精度进行了评价。所得结果表明:三种方法的平均相对误差(mean relative error,MRE)的排序为OK<IDW<TS,平方根误差(root mean squared error,RMSE)的排序为OK<IDW<TS,其中TS法的MRE及RMSE显著高于OK和IDW法,OK法的MRE及RMSE略低于IDW法。OK法所得的插值等值线图较平滑,在采样点密集处比稀疏区域更好的反映了近海沉积物中SEMAs分布特征,IDW法所得结果图与OK法相类似,但IDW法存在明显的局部误差。TS法所得数据结果溢出原始数据区间且出现了负值,与实际不符。总体来看,OK法更适合于描述近海沉积物痕量元素的空间分布特征。With the space analysis module of GIS being applied and using three kinds of methods,which are respectively inverse distance weighted（ IDW）,tension spline（ TS） and ordinary kriging（ OK）,spatial interpolations are made upon the SEMAsdata obtained from on-the-spot survey of the offshore sediments in Leizhou Peninsular. An comparison and evaluation about the accuracy of each interpolation method is conducted through cross-validation and interpolation image comparison analysis method. The results of the study indicate that Mean Relative Error（ MRE） sort of these three methods is OK IDW TS. Root Mean Squared Error（ RMSE） sort of them is OK IDW TS. The MRE and RMSE though TS is significantly higher than those through OK and IDW. The MRE and RMSE amount by OK is slightly lower than that of IDW. Moreover,the contour figure drawn based on the results obtained through OK is relatively more smooth and can better reflect the distribution of SEMAsin the surface sediment especially at the area with intensive sampling points. The contour figure based on IDM results is similar with that based on OK,but there exists obvious localized error in the figure based on IDM. The interpolated data by TS overflow the raw data range and negative values,which is inconsistent with the reality. Overall,OK is more suitable to be used to describe the spatial distribution of the trace elements in offshore sediments.

关 键 词：GIS SEMAs 空间分布特征 插值方法 

分 类 号：P736[天文地球—海洋地质]