Urban Aggregation Index

Advances in Space Research

Prepare the following files for the online submission of your manuscript:
File 1 – manuscript text: Title, authors and affiliations (+ name and e-mail of the corresponding
author), followed by abstract, key words, text, list of references, appendices, Figure captions
File 2-N Tables in order (can be in one file, may be a compressed file)
File N-M Figures in order (can be in one file, may be a compressed file)

Highlights (3sentences of 100characteristics)

This paper presented the quantitative relationships between the SE and the SAL of the UBA using Landsat land cover analysis and Aggregation Index. From this study, the following results were obtained: 1) the extensive urban development had occurred in the eastern Nagoya city during the period 1985 – 1997, and also the SAL of UBA had increased in the same region during the study period. 2) the SE and the SAL of OFA had considerably decreased due to the urban development. In other words, OFA decreased and became fragmented due to the encroachment of UBA. 3) a fairly strong positive correlation was found between the SE and the SAL of UBA, but the similar correlation for OFA was not found.

Our most important finding was a strong correlation between the SE and the SAL of the UBA, but it could be only because of the unified urban development policies in the study area. If this methodology is applied to other cities, it will be difficult to expect a similar result from other cities where various urban development policies are taken. Therefore, the future research is necessary to confirm whether a good positive correlation between the SE and the SAL does exist in other cities.

We propose the use of the diagrams of AP and AI calculated from Landsat data and aggregation index as an evaluation methodology for urban development in the view of spatial compactness. Figure 7 (a)(b)(c) show the typical examples of three scatter diagrams of relationships between the AP and the AI. These could be evaluated as good, bad and moderately in spatial compactness. Figure 7 (a), which is a good example in spatial compactness, shows the high AI values for any the SE values, and good spatially aggregately distributed. It can be considered that the city which has the relationship between the AP and AI like this, is spatially effectively composed for maintaining public facilities. Figure 7(b), which is a bad example in spatial compactness, shows the low AI values for the low SE values, and spatially dispersedly distributed. Figure 7(c), which is a moderately example in spatial compactness, shows the intermediately relationship between the AI and the SE comparing to Figure 7 (a) and Figure 7(b). By using this evaluation methodology, the relationship between the AI and the SE in 1985 is evaluated as moderately and that in 1997 is good.

In order to get more accurate, detailed, and derivative results, the spatial resolution of satellite data and the availability of various landscape metrics might be examined. The optimal spatial resolution of satellite data might be selected according to the size of study area and the spatial distribution pattern of land cover, but it is always necessary to concern about what the appropriate spatial resolution of satellite data is in the satellite remote sensing research. Although we chose Landsat TM data for our study this time, higher resolution satellite data (like Spot with 15 m spatial resolution) could be more useful to acquire better results. Furthermore, there are various landscape metrics as well as AI. Using the various landscape metrics, we could try to investigate the effective urban spatial analysis method from several different aspects.