Feasibility of Terrestrial Laser Scanning for Plotwise Forest Inventories

Detailed, up-to-date forest information is increasingly important in quantitative forest inventories. The accuracy of the information retrieval is highly dependent on the quality and quantity of the reference data collected on field sample plots. In practice, the plotwise forest data are used as a reference for the calibration of large-area inventory data measured by aerial and space-borne remote sensing techniques. Field reference data are conventionally collected at the sample plot level by manual measurements. Because of the high costs and labor intensity of manual measurements, the number of tree attributes collected is limited. Some of the most important tree attributes are not even measured or sampled. Terrestrial laser scanning (TLS) has been recently shown to be a promising technique for forest-related studies. Many tree attributes have been correlated with measurements from TLS data. Numerous TLS methods have been proposed. 6However, the feasibility of applying TLS in plotwise forest inventories is still unclear. The major missing factor is automation of data processing. Other factors hampering the acceptance of the technology include the relatively high cost of the TLS instrument, the low measurement accuracy achieved using the automated data processing currently available, and the shortage of experimental results related to the retrieval of advanced stem attributes (e.g., stem curve) and to different forest conditions. In this study, a series of methods to map sample plots were developed, and their applicability in plotwise forest inventories was analyzed. The accuracy of stem mapping, the efficiency of data collection, and the limitations of the techniques were discussed. The results indicate that TLS is capable of documenting a forest sample plot in detail and that automated mapping methods yield accurate measurements of the most important tree attributes, such as diameter at breast height and stem curve. The fully-automated TLS data processing that was developed in this study resulted in measurement accuracy similar to that of manual measurements using conventional tools or models and of manual measurements from point cloud data. The results of this study support the feasibility of TLS for practical forest field inventories. Further research is needed to explore new protocols for the application of TLS in field inventories. Three possible new directions are the integration of detailed tree attributes (e.g., stem curve, volume, and biomass) in large-area inventories, the utilization of TLS field plots in national forest inventories, and the mapping of large sample plots, e.g., in operational harvest planning. More studies need to be performed on sample plots under different forest conditions (development class, tree species, and amount of ground vegetation).