Stand site index scale development using the generalized algebraic difference approach
| Autor: | A. V. Lebedev, V. V. Kuzmichev |
|---|---|
| Jazyk: | English<br />Russian |
| Rok vydání: | 2022 |
| Předmět: | |
| Zdroj: | Сибирский лесной журнал, Vol 9, Iss 3, Pp 48-58 (2022) |
| Druh dokumentu: | article |
| ISSN: | 2311-1410 2312-2099 |
| DOI: | 10.15372/SJFS20220306 |
| Popis: | The basis for forecasting the growth of forest stands is the site index scales. Expansion of knowledge about the processes of functioning of forest ecosystems, the nature of changes in the process of growth of morphological indicators of trees and stands, their interconnections and interdependencies, and the creation of a more suitable mathematical apparatus and appropriate software for describing biological processes form the prerequisites for formulating and solving the problem of forecasting the growth of stands on higher methodological level. The aim of this work is to assess the predictive ability of the growth equations obtained using the generalized algebraic difference approach (GADA) to describe the course of growth of tree stands at the average height and to construct a site index scale. The data for the study were general tables of the course of growth of complete (normal) pine stands North Eurasian. A total of 25 equations obtained using the GADA approach are analyzed. Comparative analysis showed that the best quality of data alignment is provided by an equation based on the Mitscherlich function (also known as DrakinVuevsky, ChapmanRichards) with the replacement of parameters responsible for the limiting values of the height and the shape of the curve. This model is polymorphic, has the shape of a sigmoid curve and variable asymptotes, i.e. takes into account most of the properties imposed on the growth rate models in height. Model errors are distributed depending on the selected forecasting interval and the site index of the stand. With an increase in the forecasting period, an increase in the error occurs. For all intervals of the forecast range, the value of the mean absolute error is not exceeding 2.01 %. The highest average absolute forecast error (1.1–2.2 %) is characteristic of the extreme site index classes (Ib, V, Va and Vb). The methodology considered in the study can be applied to develop models of the growth rate of stand inventory variables of other forestforming species in Russia. |
| Databáze: | Directory of Open Access Journals |
| Externí odkaz: |
|