spaces and farmland, (2) block shape and building layout, (3) building density, and
(4) building height, for every 250-m unit.
We applied this method to the Jakarta metropolitan area to clarify the distribution
of four residential types: urban village, rural, regular form, and high-rise types. Since
there were no available data on the four indicators in Jakarta, we performed
classification using visual detection. In addition, the use of a multinomial logit
model with the available data as alternative variables showed that visual detection
had acceptable validity. The future scope of this method is as follows. By
constructing a more suitable model, it may be possible to create a visual detection–
based model in only some parts and apply it to the entire area. On the other hand, the
development of remote sensing technology and image analysis processing will
facilitate the automatic recognition of the features that can be identified by visual
detection. This will enable us to prepare global data for indicators and, thereby, make
the model globally applicable to automatically classify residential areas, instead of
resorting to visual detection.
Furthermore, this chapter utilizes the results of the classification of settlements in
Jakarta to describe the characteristics of Jakarta’s residential landscape. We particularly focused on the urban village and rural types to estimate the current kampung’s
spatial distribution and its relationship with the kampung that originated during the
colonial period. The kampung is an important research object in Indonesian urban
studies, particularly studies on Jakarta, and has been researched by many scholars.
However, due to the ambiguity of its definition and unavailability of data on its
distribution, kampung dynamics has not been adequately analyzed on the city scale.
Accordingly, this chapter analyzed the kampung by approximating the urban village
and rural types to kampung. Further, although a kampung is not determined solely by
physical features, certain valid trends of the kampung can be identified. The finding
that approximately half the number of existing kampungs are historical kampungs
has not been mentioned in earlier studies, and this finding indicates the classification
method’s usefulness.
A residential area is a place where people live their daily lives and, hence, has a
small spatial scale within the built environment. Consequently, various residential
areas are formed in megacities and have different problems such as high CO 2
emissions, air pollution, poverty, social inequality, and other issues. To address
these issues effectively, it is important to understand the conditions and implement
the policies according to the diversity of residential areas. The classification method
presented in this chapter is one way to clarify this diversity within the Jakarta
megacity. However, more research is required on this method before it can be
applied to other cities and for the creation of new data and quantitative classification.
Such research will significantly help us understand the mechanisms by which each
megacity accommodates its large population and the diversity of residential landscapes in megacities and achieve global sustainability.
4 Diversity and Historical Continuity of the Residential Landscape of a. . .
63
(4) building height, for every 250-m unit.
We applied this method to the Jakarta metropolitan area to clarify the distribution
of four residential types: urban village, rural, regular form, and high-rise types. Since
there were no available data on the four indicators in Jakarta, we performed
classification using visual detection. In addition, the use of a multinomial logit
model with the available data as alternative variables showed that visual detection
had acceptable validity. The future scope of this method is as follows. By
constructing a more suitable model, it may be possible to create a visual detection–
based model in only some parts and apply it to the entire area. On the other hand, the
development of remote sensing technology and image analysis processing will
facilitate the automatic recognition of the features that can be identified by visual
detection. This will enable us to prepare global data for indicators and, thereby, make
the model globally applicable to automatically classify residential areas, instead of
resorting to visual detection.
Furthermore, this chapter utilizes the results of the classification of settlements in
Jakarta to describe the characteristics of Jakarta’s residential landscape. We particularly focused on the urban village and rural types to estimate the current kampung’s
spatial distribution and its relationship with the kampung that originated during the
colonial period. The kampung is an important research object in Indonesian urban
studies, particularly studies on Jakarta, and has been researched by many scholars.
However, due to the ambiguity of its definition and unavailability of data on its
distribution, kampung dynamics has not been adequately analyzed on the city scale.
Accordingly, this chapter analyzed the kampung by approximating the urban village
and rural types to kampung. Further, although a kampung is not determined solely by
physical features, certain valid trends of the kampung can be identified. The finding
that approximately half the number of existing kampungs are historical kampungs
has not been mentioned in earlier studies, and this finding indicates the classification
method’s usefulness.
A residential area is a place where people live their daily lives and, hence, has a
small spatial scale within the built environment. Consequently, various residential
areas are formed in megacities and have different problems such as high CO 2
emissions, air pollution, poverty, social inequality, and other issues. To address
these issues effectively, it is important to understand the conditions and implement
the policies according to the diversity of residential areas. The classification method
presented in this chapter is one way to clarify this diversity within the Jakarta
megacity. However, more research is required on this method before it can be
applied to other cities and for the creation of new data and quantitative classification.
Such research will significantly help us understand the mechanisms by which each
megacity accommodates its large population and the diversity of residential landscapes in megacities and achieve global sustainability.
4 Diversity and Historical Continuity of the Residential Landscape of a. . .
63