Assessing primary vegetation recovery from earthquake-induced landslide scars: A real-time kinematic unmanned aerial vehicle approach

Globally, post-disturbance site formation is caused by large-scale natural disturbances. In such areas, it is necessary to restore vegetation rapidly as it regulates the outflow of sediment and water. However, the relationship between post-disturbance site vegetation recovery and environmental factors has not yet been elucidated. This study aimed to estimate the environmental factors affecting early vegetation recovery after large-scale disturbances (i.e., earthquakes). We used a novel analysis of multidimensional high-definition earth surface data derived from real-time kinematic unmanned aerial vehicles (RTK-UAVs) with high positioning accuracy to reveal primary vegetation recovery conditions, accounting for ground surface changes and environmental information of earthquake-induced landslide scars in Japan's boreal forests. A vegetation survey was conducted to understand the species composition and coverage that had recovered from earthquake-induced landslide scar in Japan's boreal forests. The herbaceous species Petasites japonicus subsp. giganteus and the woody introduced species Larix kaempferi were identified as the dominant species that help restore vegetation, with high frequency and plant coverage in the primary succession of the landslide scar. The reproductive characteristics of these pioneer species suggest that they have an advantage for establishment and growth, particularly on landslide scars with substantial topsoil dynamics. An optimal generalized linear model was used, which considered the distance from the vegetation inside and outside the landslide scars, ground surface change, inclination angle, annual amount of solar radiation, and the local region as explanatory variables. The ideal environmental conditions for vegetation recovery are summarized as follows: 1) Maximum biological legacy preserved inside and outside landslide scars. 2) Vegetation recovery expected in stable areas with little erosion and accumulation. 3) Vegetation recovery expected on slopes other than steep slopes or those with extremely high amounts of sunlight. The vegetation recovery conditions and RTK-UAV remote-sensing techniques used in this study can be applied to vegetation restoration zoning plans immediately after a disaster in similar environments.