Understory biomass measurement in a dense plantation forest based on drone-SfM data by a manual low-flying drone under the canopy. (15th June 2022)

• Record Type: Journal Article
• Authors: Zhang, Yupan; Onda, Yuichi; Kato, Hiroaki; Feng, Bin; Gomi, Takashi
• Is Part Of: Journal of environmental management. Volume 312(2022)
• Publication Date: 2022-06-15
• Subjects: Canopy height model -- Drone survey -- Plantation -- Structure from motion -- Understory vegetation biomass; Environmental policy -- Periodicals; Environmental management -- Periodicals; Environment -- Periodicals; Ecology -- Periodicals; 363.705
• DOI: 10.1016/j.jenvman.2022.114862
• Languages: English
• ISSNs: 0301-4797

Abstract

Abstract: Forest is an important part of the environmental system, which has a significant impact on soil hydrological characteristics and forest landscapes, because these processes are influenced by forest management and understory vegetation. Quantitative understory vegetation biomass (UVB) measurement and estimation are vital processes in forest ecology and environmental management. However, these estimations are difficult to make on a large scale especially dense planted forest. Here, we applied catchment-scale Structure from Motion and a manually operated ultralow-flying drone under the canopy of a dense planted (2000 stems/ha) Japanese cedar/cypress plantation forest to reconstruct the understory. An understory drone survey was performed over a 1.1-ha sub-catchment to generate a canopy height model based on dense point cloud data. A biomass survey of three 16-m 2 harvesting plots was conducted to compute understory vegetation volume data based on point clouds. Combined with harvested biomass data in the field, quantitative models were developed between the understory vegetation volume and biomass. Subsequently, the models were used to map spatial understory vegetation biomass distribution in the sub-catchments. Aerial photos taken by the ultralow-flying drones under the canopy yielded a high-resolution catchment-scale understory with point cloud density >10/cm 2 . Strong cubic model coefficients of determination ( R 2 = 0.75) predicted the understory vegetation biomass based on the canopy height model. The mean understory vegetation biomass was 0.82 kg/m 2 and dominated by low ferns. In the present study, we successfully reconstructed the multilayered forest structure and generated understory vegetation biomass distribution models. This results also will be essential to evaluate the erosion and evapotranspiration in dense plantation forests and future environmental management. Graphical abstract: Image 1 Highlights: Achieved first understory biomass survey by drone under high-density forest canopy. Applied high-density cloud point dataset in calculating understory vegetation volume. Developed a model for predicting understory vegetation biomass. Provided a new insight for understory biomass estimation and forest management.