It is required to estimate the distribution of stand volume of Hiba (Thujopsis dolabrata S. et Z) in natural forest of Shimokita peninsula for sustainable wood supply in this region. For this purpose, it is necessary to find the index that corresponds to the basal area of Hiba. We applied the Pattern decomposition method, one of the methods for overcoming mixel problem, which decomposite TM data to Water (W), Soil (S), and Vegetation (V) patterns. This method reduces the topographical effect to remote sensing data. It is found that the relation between W pattern value and basal area is useful to estimate the stand volume. Then we produced the distribution map of Hiba stand volume. Using the distribution map, we analysed the relation between stand volume of Hiba and two geographical conditions, the inclination angle and the distance from roads. We assumed that the location where inclination angle is under 25 degrees and distance from road is 200 m is suitable for felling, and estimated avairable volume of Hiba. Finally, we found that less than 20% of the stand volume is available for wood supply.
