The spatial division of a metropolis from the viewpoint of the commuting pattern of employed persons classified by industry and occupation in the Tokyo metropolitan area

The main purpose of this study is to identify the spatial patterns of commuting flows in the Tokyo metropolitan area. Using census data classified by industry and occupation, the author examined these patterns from 1975 to 1995. Recent studies have demonstrated that the spatial structure of the Tokyo metropolitan area is divided into several sectoral zones. For example, Chiyoda ward has been regarded as the center of Tokyo in name and in reality. However, the volume of commuting inflow to the newly-developed city centers such as Shinjuku ward has rapidly increased over the past 20 years. As a result, Chiyoda ward has come to be just' one of the centers' of Tokyo. In this way, the mono-centric structure of the centers of Tokyo has been transformed into a multi-nucleation one.First, the author examined commuting patterns of each of these 11 wards, calculating "an index of daily migration for work (DMW)". In this paper, the author calculated this index for each industry and occupation. Comparing the distributional patterns of the largest DMW classified by industry with those by occupation, concentric patterns were shown in some wards from the former result, and sectoral patterns in almost all of the 11 wards from the latter. Professional and technical workers who lived in the northeast zone of the Tokyo metropolitan area commuted more to Minato, Shinjuku and Shibuya wards, and managers and officials who lived in the southwest zone commuted more to Chuo, Taito and Sumida wards. In this way, professional and technical workers, and managers and officials who live in outer suburbs commute less to an adjacent core. On the other hand, sales workers and service workers who live in the inner suburbs commute more to an adjacent core. For example, sales workers who lived in the northeast zone commuted more to Chuo and Taito wards, and service workers who lived in the southwest zone commuted more to Shibuya and Toshima wards.Second, using cluster analysis, the author classified the 11 wards into several groups in 1975 and 1995. For that purpose, the author calculated a location quotient for six major industrialand occupational groups in each ward. As a result, the wards were classified into five clusters in 1995. Comparing the results of this analysis in 1995 to that for 1975, the former result corresponded to the group of commuting patterns classified by occupation.In conclusion, the author subdivided the Tokyo metropolitan area into six commuting zones. Each zone includes one of multiple urban cores. A commuting zone of workers living in the same area varies depending on the occupation. For example, the commuting zone of professional and technical workers commuting from Saitama Prefecture to Tokyo was different from that of sales workers. These results showed that the traditional spatial subdivision of Tokyo termed "Shitamachi and Yamanote" has become less clear-cut and more complex.