Abstract:

Objective: The aim of this study is to test whether there is any bias in the DBH measurement data of large survivor trees in continuous forest inventory, and to provide scientific basis for improving the measurement method. Method: Based on the DBH measurement data of 2501 large survivor trees in forest sample plots from the continuous forest inventory in Beijing from 2020 to 2024, the tree growth rate models and difference significance test method were used to test whether the difference between measured and estimated DBH growth was significant and there was bias in DBH measurement data, and the relationship between the bias size and interval length (1–8 years) was analyzed, and the data in Hebei Province was employed for verification. Result: The results showed that there was a significant bias in the DBH measurement data of large survivor trees from the inventories in Beijing from 2020 to 2024. The total bias of measured DBH growth was 38.0%, among which the bias of 8-year interval data was 21.9%, and that of 1-year interval data was 88.7%. The DBH measurement data of large survivor trees in Hebei Province from 2020 to 2024 also confirmed that the total bias of measured DBH growth was 50.0%, and the bias of five sets of plots ranged from 32.8% to 86.6%. Conclusion: There are positive biases in the DBH measurement data of large-diameter survivor trees in some extent, and the size of bias is highly correlated with the interval length. The shorter the interval length, the greater the bias in DBH measurement data. It is suggested to develop tree DBH growth rate models based on previous inventory data, and reasonably set the upper limit for warning according to the predicted DBH growth. For very large survivor trees over a certain size (e.g. 50 cm), the predicted estimates from models can be directly used as the measured DBH.

Key words: continuous forest inventory, large survivor tree, diameter at breast height (DBH), bias