DISTRIBUTIONAL ANALYSIS 
OF GREEN TRACTS AND VEGETATION-COVERED AREAS 
Kiichiro KUMAGAI 
Department of Civil and Environmental System Engineering, Setsunan University 
17-8 Ikedanakamachi, Neyagawa, Osaka 572-8508, JAPAN 
kumagai@civ.setsunan.ac.jp 
KEY WORDS: Remote Sensing, Land Cover, Land Use, Comparison, Decision Support, Distributed, Urban 
ABSTRACT: 
The green tract of land is one of the significant factors for urban planning and urban management to improve and conserve the 
urban environment. Planning the green tracts is often carried out as a part of land use strategy based on relation to 
neighborhood, their scale, functions, and locational condition. On the other hand, a vegetation-covered area is a considerable 
focus of investigation for urban environment. For making the vegetation cover map, it is required to estimate vegetation cover 
ratio widely. Many previous studies have discussed methods to calculate vegetation cover ratio applying satellite remotely 
sensed data. In general, the green tracts contain both vegetation-covered areas and vegetation-uncovered areas because there are 
various kinds of parks/green zones assigned to the green tracts. In addition, the vegetation-covered areas usually include the 
places allocated to other land use categories excluding the green tract. In case of producing a plan for open spaces, it is 
necessary to confirm the current status of not only the green tracts of land use but also the vegetation-covered areas. The 
objective of this study is to develop the comparison rhethod for the difference of distribution between the green tracts and the 
vegetation-covered areas using a land use map and satellite remotely sensed data. The decision tree method including the two 
viewpoints from parcel level and local area level was examined as the comparison method. It was shown that the developed 
method might provide some suggestions for producing a plan for open space. 
1. INTRODUCTION the green tracts and the vegetation-covered areas using a land 
use map and satellite remotely sensed data. We estimated 
The green tract of land is one of the significant factors for vegetation-covered areas applying normalized vegetation 
urban planning and urban management to improve and index (NDVI) derived from QuickBird data and land use 
conserve the urban environment. It plays an important role in data. The decision tree method including the two viewpoints 
the evaluation of urban environment; fire spread prevention, from parcel level and local area level was examined as the 
refuge place in disaster, field of relaxation, etc. Planning the comparison method. It was shown that the developed method 
green tracts is often carried out as a part of land use strategy might provide some suggestions for producing a plan for 
based on relation to neighborhood, their scale, functions, and open space. 
locational condition. On the other hand, a vegetation-covered 
area is a considerable focus of investigation for urban 
environment. It is expected to use a vegetation covered map 2. METHODS AND MATERIALS 
for the urban climatic simulation, the evaluation of fire spread 
prevention, the estimation of the sensible temperature, and so 2.1 Study Area 
on. For making the vegetation cover map, it is required to 
estimate vegetation cover ratio widely. Many previous studies Figure 1 shows a study area in this research. The area is 
have discussed methods to calculate vegetation cover ratio located in the Kansai district, the western part of Japan. It 
applying satellite remotely sensed data. covers 10 km horizontally by 10 km vertically. There have 
been a variety of urbanization types since the end of World 
In general, the green tracts contain both vegetation-covered War II in this area. Therefore, many kinds of land use exist: 
areas and vegetation-uncovered areas because there are industrial areas, residential areas, residential areas ( new 
various kinds of parks/green zones assigned to the green town type), commercial areas, etc. There is a large 
tracts. In addition, the vegetation-covered areas usually difference of greenery status between their locations. 
include the places allocated to other land use categories 
excluding the green tract. In case of producing a plan for 2.2 Data 
open spaces, it is necessary to confirm the current status of 
not only the green tracts of land use but also the We used two kinds of spatial information for applying the 
vegetation-covered areas. Through the comparison of the calculation of a green tract map and a vegetation-covered 
distribution between the green tracts and the map as follows. 
vegetation-covered areas, the regions, where the green tract 
covered with vegetation should be improved, might be 2.2.1 Satellite Data: Multispectral data of QuickBird, 
suggested. The objective of this study is to develop the observed on 21 February 2002 with a spatial resolution of 2.4 
comparison method for the difference of distribution between m/pixel, were, adopted. The off nadir angle in the 
488 
Internc 
Figur 
p 
Fig 
obser 
corre 
NDV 
calcu 
resan 
2.2.2 
Infor 
analy 
data 
road, 
park/ 
the « 
study 
level 
inves 
use | 
categ 
out € 
local 
1S no 
some 
cons 
relati