Quantitative assessment of the impact of typhoon disturbance on a Japanese forest using satellite laser altimetry

• Published in: Remote sensing of environment Vol. 156; pp. 216 - 225
• Main Authors: Hayashi, Masato, Saigusa, Nobuko, Oguma, Hiroyuki, Yamagata, Yoshiki, Takao, Gen
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 01.01.2015
• Subjects: Canopies; Canopy height; Damage; Forest disturbance; Forests; ICESat/GLAS; Larix; Larix kaempferi; Lidar; Mathematical models; Risk assessment; Slopes; Typhoon; Typhoons; Wind damage; INW, Japan, Hokkaido; Japan, Hokkaido; Typhoon; ICESat/GLAS; Canopy height; Risk assessment; Forest disturbance
• ISSN: 0034-4257; 1879-0704
• DOI: 10.1016/j.rse.2014.09.028

Spaceborne light detection and ranging (LiDAR) sensors can accurately observe forest parameters, such as canopy height and forest biomass. In this study, we investigated the potential of spaceborne LiDAR to quantify the effects of forest disturbances caused by typhoons on Hokkaido Island, Japan, after typhoon Songda in 2004. We developed a model to estimate canopy height from the Geoscience Laser Altimeter System (GLAS) of the Ice Cloud and land Elevation Satellite. GLAS waveforms are broadened by slopes (“pulse broadening”), so we used the lead10 and trail10 values (edge extents referring to 10% and 90% of the cumulative energy returned) to correct for this effect, because they were the parameters most strongly correlated with pulse broadening. We developed the estimation model by dividing GLAS data between gentle and steep slopes, using an empirically determined threshold, and obtained a root-mean-square error of 3.5m for canopy height. We then applied the model to GLAS data in the area damaged by typhoon Songda to estimate changes in canopy height caused by the typhoon. We used a wind damage map produced in a previous study to identify the damage severity. The model showed that canopy height decreased by an average of 2.7±1.8m (95% confidence interval) in heavily damaged areas. The canopy height decreased most in coniferous forests, and especially in Japanese larch forests, that they have very shallow roots. We also assessed the geographic factors that most strongly influenced wind damage risk by means of multivariate analysis. We found serious damage in areas with a gentle slope that sustained strong winds. Understory plants and topography type also affected the risk. Our results show that spaceborne LiDAR can be used to quantify the severity of wind damage caused by forest disturbances, thereby providing guidance for forest management. •Spaceborne LiDAR can observe forest disturbances quantitatively.•We developed a model for canopy height estimates with an accuracy of 3.5m.•Canopy height decreases caused by typhoon Songda averaged about 2.7m.•Topography and forest conditions strongly affected the risk of wind damage.