bul 2004 
orest. In: 
: Legacy. 
shington, 
' Natural 
chens for 
ins. — In: 
onitoring 
rth and 
, Kluwer 
J. 2001, 
ies Using 
65: 470- 
Manus, J. 
sensing - 
te Sens. 
Sterling, 
odiversity 
Evolution 
ness on a 
rsity and 
>neralized 
11-156. 
odiversity 
) in the 
rogram of 
zler, Ariel 
able help. 
| Michael 
cial thank 
itor. 
EXAMINATION FOR INFLUENCE ASSESSMENT METHOD ON FOREST VOLUME 
ESTIMATE USING REMOTE SENSING AND GIS 
Hisao. Ito ^*, Susumu. Ogawa " 
* Dept of geography, Graduate school of science, Tokyo Metropolitan university 1-1 Minami-Ohsawa, Hachiohji-city, Tokyo 192-0397, Japan 
ito-hisao(@c.metro-u.ac.jp 
? Dept of environmental system, faculty of geo-environmental science, Rissho university 1700 Magechi, Kumagaya, Saitama, 360-0194, Japan 
ogawa(@ris.ac.jp 
Key words: Remote Sensing, Forestry, Resources, Statistics, Calibration, Estimation, Measurement 
Abstract: 
Since waming of the global warming by carbon dioxide, the biomass distribution of forest has been concemed and its study has been requested. Much 
expectation has focused on the remote sensing corresponding to wide area. We noticed small area and estimated the spatial distribution of the forest 
volume using the actual GIS and remote sensing data. We comprehended the spatial distribution of the vegetation types, the geological formation, and 
the water cycle. Moreover, considering the seasonal change, we assessed influence and relationship between the estimated forest volume and other 
natural factors. The forest volume was largely influenced by the forest types. 
1. INTRODUCTION 
Since warning of the global warming by carbon dioxide, the 
biomass distribution of forest has been concerned and its study 
has been requested. Much expectation has focused on the 
remote sensing corresponding to wide area. However, many 
existing studies have discussed global forest distribution. We 
noticed small area and estimated the forest volume using the 
actual GIS and remote sensing data. Next, the estimated forest 
volume was compared with the vegetation types, the geological 
formation, and the water environments which may affect forest 
volume. 
2. METHODS 
The spatial distribution of forest volume (timber volume, whole 
volume, and biomass) was estimated by comparing the digital 
numbers of satellite images and actual measurement data. At 
first, the effects of haze and shade were reduced by Tasseled 
Cap correction. Next, the forest volume each small catchment 
was compared with the spatial distribution of the vegetation 
types, the geological formation, and the water cycles estimated 
by the satellite images, the GIS data, and the atmospheric data. 
1.1 Study Area 
Okutama forest area was selected as a study area and is located 
in the headwaters of the Tamagawa River, the western part of 
Tokyo. It functions for a municipal drinking-water source, river 
recharge, sediment run-off prevention, and water purification 
for Tokyo. Most Okutama forest areas exist in intensely uneven 
mountainous area. The altitude ranges 200m to 2000m high. 
First, the small catchments were divided by DEM. The Figure 1 
shows study area divided to small catchments . 
  
* Corresponding author. 
851 
  
  
  
  
  
Yamanashi up 
Figure 1. Study area 
1.2 Data 
We chose each seasonal satellite data from Landsat/TM images 
(Path = 107, Row = 35; Path = 108, Row = 35) and 
TERRA/ASTER images covering Okutama forest area. We also 
used a Digital Map 25000, a digital elevation model 50-m Grid 
(Nippon-II) produced by Japan Geographical Survey Institute, 
and monthly meteorological grid data made by National Digital 
Land Information. Moreover, we used a vegetation map and 
timber volume statistics (1999), breast-height diameter, height, 
ages, species, and the number of trees in Okutama forest area, 
offered by the Bureau of Waterworks of Tokyo Metropolitan 
Government.