Forest volume is of great interest to forest managers.Plot observations of forest volume are available from planning reports.For managers,however,what is relevant are forest volume surfaces.In this study,three interpo...Forest volume is of great interest to forest managers.Plot observations of forest volume are available from planning reports.For managers,however,what is relevant are forest volume surfaces.In this study,three interpolation methods were employed to construct continuous surfaces for the evaluation of forest volume at positions for which no measurements are available.For the Municipal Forest of Skyros Island,we compared spatial predictions derived from Inverse Distance Weighting(IDW),Block Kriging(BK),and Block Co-Kriging(BCK)interpolation methods,as applied to data from 120 sample plots.The existence of spatial autocorrelation in the data was identified using correlograms of Moran’s I index.Spatial outliers were identified and excluded from the analysis using the local Moran’s I index.Only slope,of the examined environmental factors,showed an acceptable correlation with forest volume and was used as an auxiliary variable for BCK.The performance of the three methods was evaluated,using an independent validation set and comparing these indices:mean error(ME)and root mean square error.Additionally,for BK and BCK,the indices,standardized ME,average standard error,and the standardized root-mean-square error were used.For the three interpolation methods,the BK method gave the more accurate results;BCK is the next more accurate method and IDW the third.展开更多
文摘Forest volume is of great interest to forest managers.Plot observations of forest volume are available from planning reports.For managers,however,what is relevant are forest volume surfaces.In this study,three interpolation methods were employed to construct continuous surfaces for the evaluation of forest volume at positions for which no measurements are available.For the Municipal Forest of Skyros Island,we compared spatial predictions derived from Inverse Distance Weighting(IDW),Block Kriging(BK),and Block Co-Kriging(BCK)interpolation methods,as applied to data from 120 sample plots.The existence of spatial autocorrelation in the data was identified using correlograms of Moran’s I index.Spatial outliers were identified and excluded from the analysis using the local Moran’s I index.Only slope,of the examined environmental factors,showed an acceptable correlation with forest volume and was used as an auxiliary variable for BCK.The performance of the three methods was evaluated,using an independent validation set and comparing these indices:mean error(ME)and root mean square error.Additionally,for BK and BCK,the indices,standardized ME,average standard error,and the standardized root-mean-square error were used.For the three interpolation methods,the BK method gave the more accurate results;BCK is the next more accurate method and IDW the third.
