2010 年 96 巻 8 号 p. 517-523
Steel is the most widely used material in the world. Many studies related to material flow analysis of steels have been conducted for Japan, Asian countries and the world. Top-down and bottom-up approaches were used in these material flow analyses and stock accounting. However, applicability of these approaches is quite dependent on data availabilities. To overcome this problem, we proposed a methodology to estimate steel stock based on satellite images. Previous studies told us that night time light images collected by satellites have strong correlation with human activities, e.g. electricity consumption, CO2 emissions, GDP, etc., which also links to the amount of steel stock. In this study, night time light image was used along with land cover data to estimate the building and civil construction steel stock in Japan and Asian countries. Analysis was first performed within Japan for each prefecture, and then the result was applied countrywide for Japan, China, South Korea and Taiwan. It was found that the building steel stock showed high relativity with urban night time light while civil construction steel stock was better described by total night time light.