In order to conduct further verification of measuring
broad-leaved forests using a low-cost portable Terrestrial Laser Scanner (TLS),
the present study measured a secondary broad-leaved forest of the Funyu
experimental forest, Utsunomiya University, Japan. Then, DBH, height, top end
diameter, sweep, and stem volume were analyzed and compared with results of
coniferous plantation forests using terrestrial LiDAR. RMSE of DBH was 1.91 cm,
which was higher than that of coniferous plantation forests. However, DBH was
typically rounded to 2 cm; therefore, RMSEs of DBH were within the allowable
range. Furthermore, species did not affect the accuracy of DBH mensuration. The
RMSE of height was 2.26 m, which was similar to 2.29 m of the 32-year-old
coniferous plantation forest with high stand density. RMSE of height outside
the plot was higher than that inside the plot because distances from the
portable TLS to trees were too short to measure the heights of trees. The log
detection rate was 79.22% in the present study, whereas it was 85.23% in the
coniferous plantation forest. RMSE of top end diameter was lower than that of
the coniferous plantation forest because the broad-leaved trees were extracted
manually in the present study, whereas coniferous trees were extracted
automatically. Errors in top end diameters increased with increasing top end
heights because the number of points decreased. The RMSE of sweep was much
higher than that of coniferous plantation trees. The RMSE of stem volume was
14.3%, which was lower than that of the coniferous plantation forest, despite
the low-cost portable TLS, because of lower RMSEs of top end diameters and
shorter distances from the portable TLS.
Subjects | Engineering |
---|---|
Journal Section | Research Articles |
Authors | |
Publication Date | June 29, 2017 |
Published in Issue | Year 2017 Volume: 3 Issue: 1 |
The works published in European Journal of Forest Engineering (EJFE) are licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.