| 要旨トップ | 目次 | | 日本生態学会第69回全国大会 (2022年3月、福岡) 講演要旨 ESJ69 Abstract |
一般講演(口頭発表) C02-06 (Oral presentation)
Tropical storms, such as typhoons in Asia, cause significant damage to forests. Thus, efficient and cost-effective methods are needed to evaluate wind-induced damages. In this study, high-resolution satellite images were used to delineate damaged forest areas, and high-resolution airborne LiDAR (ALS) was used to estimate standing biomass and its loss due to a major typhoon from ALS metrics. From analyses of satellite image detections of the area north of Kyoto City, Japan, with the random forest algorithm, we found that 2.47 km2 forest land out of 230 km2 (1.07%) were damaged. Overall, the ALS-derived canopy height agreed well with actual canopy height (R2 = 0.87), and was used to make a generalized linear mixed model that predict aboveground biomass (R2 = 0.62). The total estimated standing biomass of the study area was 6396 Gg (295 MG/ha), of which 1.44% were damaged by the typhoon. Among the vegetation types, plantations were mainly damaged because of their higher canopy height. The rate of biomass loss in plantations (2.4%) was more than four times higher than other vegetation types (0.16 ~ 0.61%). Our estimated damage rates were likely underestimated because of the inability of ALS to detect small-scales damages which may prevail. Therefore, the actual biomass loss should be higher and could not be quickly offset by primary productivity of the forest after the typhoon.