GROWTH MODELS OF Cryptomeria fortunei Hooibrenk ex Otto et Dietr BASED ON SIMULTANEOUS EQUATIONS OF NONLINEAR MEASUREMENT ERRORS

摘要

The stand growth model is the basis for decision-makers of forest management. So far, stand basal area (SBA) and stand volume have been modelled effectively. This paper attempts to reduce the subjective errors in measurement and calculation of SBA and stand volume, and improve the precision of and compatibility between SBA model and stand volume model. Firstly, the data on various parameters were collected from 362 Cryptomeria fortunei Hooibrenk ex Otto et Dietr (Cryptomeria fortunei) trees in southeastern China's Fujian Province. Next, the Richard equation, the Logistic model, and the Mitscherlich model were selected as the basic SBA models, while the Richard equation and the Schumacher model were taken as the basic models of stand volume. Then, optimal basic models were screened by the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) with entropy weights, using evaluation indices like the determination coefficient (R-2), root mean square error (RMSE), and absolute bias (AB). Finally, the optimal basic models were improved with the simultaneous equations of nonlinear measurement errors. The results show that: the Logistic model achieved an R-2 of 0.8938, an RMSE of 4.3602, and an AB of 3.5581, and the shortest total distance to the optimal solution after being processed by TOPSIS with entropy weights, suggesting that this model is the optimal SBA model; the Schumacher model was confirmed as the optimal stand volume model, with the higher R-2 and lower RMSE and AB; the results of paired ttest show that the SBA and stand volume predicted by the two optimal error models deviated very slightly from the measured results, indicating that they are more compatible and consistent than the basic models. The research results provide a scientific basis for improving the management and value evaluation of Cryptomeria fortunei.