Abstract:

The indices of rapid assessment of forest health (RAFH) were given and their rationalities were tested based on BP neural networks in terms of convergence effects of BP networks established according to pattern recognition of training data and the consistency tests between simulation outputs and expected outputs with three methods including percent error, linear regression and Nash-Sutcliffe efficiency. The results of convergence effects showed that the networks could converge properly with 16 or more neurons in hidden layer, which indicated that there was a significant, nonlinear correlation between the inputs derived from training data values of 3 indices consisting of stand structure, severity of pest and disease and soil thickness and the target outputs resulting from precision assessment of forest health (PAFH). The reults of consistency tests demonstrated with mean relative error (-6.140 9%), the Nash-Sutcliffe efficiency (0.905 4) as well as the slope (α=0.968 3) and the intercept (b=0.049 0) of the regression equation indicated high consistency between simulation outputs and excepted outputs. Therefore, stand structure, severity of pest and disease and soil thickness could be considered as indices of rapid assessment of forest health (RAFH).

Key words: rapid assessment of forest health, indices, BP neural networks, rationality test