Abstract: Accurate remote sensing measurement and batch extraction of stand factors are the development trends of forest resources survey. In this study, UAV and ground-based handheld LiDAR were used for remote sensing of Cinnamomum camphora woodland of Pujiang Country Park in Shanghai. Combined with field survey data, we explored forest stand factor extraction method and evaluated the accuracy of 3D LiDAR space-ground integration technology in urban artificial pure forest. The results showed that: (1) 3D LiDAR space-ground integration technology had high measurement accuracy and application value in remote sensing measurement of forest stand factors, and the bias of the extracted tree height was -0.13 m (-1.17%), with R² being 0.95 (P<0.001). The bias of the diameter at breast height (DBH) was -2 cm (-5.59%), with R²=0.98 (P<0.001). Effective correction of systematic errors in DBH extraction was accomplished using the formula DBH=0.9670×DBH_p-0.0057. (2) UAV and ground-based handheld LiDAR had advantages in remote sensing measurement of forest stand factors. UAV LiDAR had advantages in measuring tree height and crown width, while ground-based LiDAR was more accurate in measuring DBH and canopy volume. Therefore, LiDAR air-ground fusion technology has broad application space in forest carbon sink assessment and forest resources survey, which would provides technical support for the automation and intelligence of largescale forest resources survey.

Key words: forest resources survey, carbon sink, stand factor, remote sensing survey, LiDAR