Accuracy and Time Efficiency of ForestScanner App when Measuring Plot-Level DBH under Diverse Forest Ecosystems

Abstract

Accurate and efficient measurement of tree diameter at breast height (DBH) is essential for forest inventory and management. While traditional methods are time-consuming, new smartphone-based LiDAR applications like ForestScanner promise rapid, cost-effective solutions. However, their performance across diverse forest ecosystems requires thorough evaluation. This study aimed to assess the accuracy and time efficiency of the ForestScanner app for plot-level DBH measurements compared to manual caliper methods under varied growing conditions in Romania. One hundred circular plots (approx. 300 m² each) were established in forests near Brașov City, encompassing diverse forest tree species, ages, topographies, and understory conditions. DBH of 987 trees was measured manually with calipers and digitally using the ForestScanner app on a LiDAR-equipped iPhone. Time consumption for plot establishment, manual DBH, and app-based DBH measurements was recorded. Accuracy was assessed using bias, mean absolute error (MAE), and root mean squared error (RMSE), with heteroskedasticity checked via Breusch-Pagan and White tests. ForestScanner showed a negligible overall bias (-0.003 cm), but MAE reached 3.66 cm when all measurements were included. Occlusion by vegetation or nearby trees significantly impacted the app’s accuracy; for non-obstructed trees (n = 824), bias was +0.26 cm with an MAE of 2.07 cm. Manual DBH measurement averaged 14 seconds/tree, while ForestScanner averaged 16 seconds/tree. Plot establishment time and measurement time were influenced by tree density. ForestScanner offers a user-friendly, free tool for DBH measurement and tree mapping, but its accuracy may be affected by occlusion. On the other hand, the app comes equipped with several useful features, such as documenting the plots by LiDAR point clouds, real-time DBH measurement, and data storage, while returning comparable time efficiencies. Future work should focus on more diverse forest types to refine its practical application in forestry.