Trees provide various ecosystem services. Given the impact of climate change and increasing
focus on nature-based solutions, effective management of forests and urban forests is increasingly
crucial. However, field tree inventory data often suffers from human errors, and there is a lack of
standardized quality requirements for both data and collectors in Taiwan. A management system is
needed to ensure quality control and quality assurance, and the development of tools is also needed
to improve measurement precision. This study conceived an experiment to assess the measurement
quality of collectors based on the number of outlier results detected using the boxplot method. We
also developed a tool - breast height pole (BH pole) - to aid in measuring diameter at breast height
(DBH) and evaluated its impact on measurement precision. We also conducted analyses taking
into account different tree shapes and DBH classes, and proposed a suggested tolerance range.
The results showed that the outlier detection mechanism facilitates rapid examination of inventory
quality. When the measurement quality objective (MQO) was set to 90%, approximately 80% of
collectors passed the test. Evaluation of measurement precision using standard deviation and coefficient
of variation of the data revealed a high level of error in DBH inventory. The BH pole proved
helpful in improving measurement precision, especially for S-type trees among the four tree types
analyzed. Finally, we recommend two tolerance standards based on the results: the first utilizes the
inner fence of the boxplot method, setting the tolerance as median DBH ± 0.68 cm; and the second
employs standard deviation, setting the tolerance as mean DBH ± 0.56 cm. The research aimed
to develop results to serve as a reference for establishing inventory quality standards in Taiwan,
thereby complementing existing tree inventory guidance and regulations.