抄録
The objective of the present study is to develop a transparent and verifiable model of volume estimation for conifer plantations using remote sensing. The model is to be independent from observation, that is, ground truths are not necessary for parameter fitting and model construction. To achieve it, the estimation process is divided into two steps: direct measurements of physical parameters of stand, and a stand volume estimation by an external model. As an external model, we adopt the stand density chart, which is a robust and general model of stand volume growth and estimation based on a semi-empirical growth model of even-aged stands. It can estimate a stand volume from a dominant height and a density, which can be directly measured by remote sensing. In the present study, firstly, the iso-height curves, a sub-model of the stand density chart, are created from the external inventory data. Then, the dominant heights and densities are directly observed by means of the airborne remote sensing. Finally stand volumes are estimated by the observed dominant heights and densities using the iso-height curves. We found that the new iso-height curves predicted the stand volume very well. The dominant height estimation was reasonably accurate, too. However, there was a room for improvement for the stand density estimation. This method will contribute the implication of the remote sensing technology to forest management by sharing the concept and values with foresters.
