Volume 10, Issue 20 (11-2022)
Ecol Iran For 2022, 10(20): 120-128 |
Back to browse issues page
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
Hoseinpour A, fallah A, niknejad M, hejazian M, kalbi S. (2022). Investigating of Kriging Geostatistic Method Capability for Forest Stand Volume Zoning (Case Study: Haftkhal Area). Ecol Iran For. 10(20), 120-128. doi:10.52547/ifej.10.20.120
URL: http://ifej.sanru.ac.ir/article-1-249-en.html
URL: http://ifej.sanru.ac.ir/article-1-249-en.html
1- Sari Agricultural Sciences and Natural Resources University
Abstract: (2658 Views)
Extended Abstract
Introduction and Objective: Due to the difficulty of this process, determining the volume inventory of the forest is one of the most fundamental objectives and duties of forest management. It has always been connected to difficulties for the department in charge of implementation. Mapping of forest stand volume has an important role in sustainable forest management. In Iran’s north forestry plans, the systematic random sampling method and arithmetic average method is used to estimate the stand volume and spatial relationships between inventory sample plots will not be considered. The aim of this study is to compare kriging method and ordinary one used to map stand volume of Haftkhal forestry plan.
Material and Methods: For this purpose, we used 243 circular sample plots with 10R area and 150×200meter grid (Forestry Department standard in Iran) for inventory.
To evaluate the performance of the kriging-method-created map the approach of poisonous variograms, histograms, and their indices was first used as the referential map to determine whether the data were normal. The best parameters of Lag Size, Nugget, and Partial Sill were then tested using the remaining points and the conventional approach to create the necessary kriging map using 30% of the sample pieces (73 sample pieces) randomly separated as ground control points. Finally, the ideal mode was decided upon. The produced map was then verified using ground control sample pieces (73 sample pieces), and the amount of difference was measured by comparing the head volumes of the control sample pieces and those calculated using the Kriging method as corresponding points.
Results: Map produced by the arithmetic mean method (ordinary method of Forestry Department) and maps of Kriging method was compared. Accuracy of kriging map was controlled by using ground samples plots. The results showed that Root Mean Square error (RMSe) of the best model was 30 percent and Bias was 0.11 percent. Using SAS software showed the correlation coefficient between kriging map and ground sample plots and corresponding relationship was at the level of 99 percent, but the correlation between arithmetic mean method was not corresponded even in 95% probability level. T-test, using this software, also showed similar results.
Conclusion: In conclusion, Kriging capability is superior to conventional methods for estimating forest stand volume.
Introduction and Objective: Due to the difficulty of this process, determining the volume inventory of the forest is one of the most fundamental objectives and duties of forest management. It has always been connected to difficulties for the department in charge of implementation. Mapping of forest stand volume has an important role in sustainable forest management. In Iran’s north forestry plans, the systematic random sampling method and arithmetic average method is used to estimate the stand volume and spatial relationships between inventory sample plots will not be considered. The aim of this study is to compare kriging method and ordinary one used to map stand volume of Haftkhal forestry plan.
Material and Methods: For this purpose, we used 243 circular sample plots with 10R area and 150×200meter grid (Forestry Department standard in Iran) for inventory.
To evaluate the performance of the kriging-method-created map the approach of poisonous variograms, histograms, and their indices was first used as the referential map to determine whether the data were normal. The best parameters of Lag Size, Nugget, and Partial Sill were then tested using the remaining points and the conventional approach to create the necessary kriging map using 30% of the sample pieces (73 sample pieces) randomly separated as ground control points. Finally, the ideal mode was decided upon. The produced map was then verified using ground control sample pieces (73 sample pieces), and the amount of difference was measured by comparing the head volumes of the control sample pieces and those calculated using the Kriging method as corresponding points.
Results: Map produced by the arithmetic mean method (ordinary method of Forestry Department) and maps of Kriging method was compared. Accuracy of kriging map was controlled by using ground samples plots. The results showed that Root Mean Square error (RMSe) of the best model was 30 percent and Bias was 0.11 percent. Using SAS software showed the correlation coefficient between kriging map and ground sample plots and corresponding relationship was at the level of 99 percent, but the correlation between arithmetic mean method was not corresponded even in 95% probability level. T-test, using this software, also showed similar results.
Conclusion: In conclusion, Kriging capability is superior to conventional methods for estimating forest stand volume.
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
