Effect of bin width on variogram model accuracy: A case study of teak tree volume specific to Solomon clone in Tawau, Sabah, Malaysia
Jamalul Kiram, Johannah | Yunus, Rossita Mohamad | Japarudin, Yani | Lapammu, Mahadir | Monteuuis, Olivier | Goh, Doreen Kim Soh
The volume of teak trees (Tectona grandis Linn. f.) was analyzed using data from a teak plantation managed by the research and development team at Sabah Softwood Berhad in Brumas Camp, Tawau, Sabah, Malaysia. To fit the exponential model to the experimental variogram, various bin widths were used to obtain different variogram models. These models were plotted for comparison, and the root mean square error (RMSE) was calculated. Additionally, cross-validation was performed to assess the predictive accuracy of each model for the unseen data. The analysis indicated that a bin width of 0.003—that was approximately 333 m apart—was ideal for fitting the exponential model. This width demonstrated the lowest RMSE and ideal graphical observations. This study confirms that selecting the optimal bin width significantly affects the accuracy of model predictions, regardless of the sample size.
Показать больше [+] Меньше [-]