505
Symposium on Remote Sensing for Resources Development and Environmental Management / Enschede / August 1986
Global vegetation monitoring using NOAA GAC data
H.Shimoda, K.Fukue, T.Hosomura & T.Sakata
Tokai University Research & Information Center, Tokyo, Japan
ABSTRACT: In the last decade, the necessity of global monitoring of vegetations or bio-mass
has become a more urgent matter. In order to prevent the large disaster which may be caused
by vegetation decrease, accurate conditions or stage of the world vegetation should be
monitored. The only satellite system and observation data which can be used for the purpose
is TIROS/NOAA system and Vegitation Index Data(VID) made of AVHRR, respectively. The
purpose of this study is to establish the method to derive a global vegetation map from a
VID set.
One data set was used in this study. A large shading effects mainly caused by sun angle
deviations were first eliminated. The classification was done using a maximum liklihood
method with four channels of VID. Training data composed of 67 categories were chosen
according to bhe World Vegetation Map made by Preston James et al.
After the classification, these 67 categories were unified to 17 categories. Then the
classified image were transformed to longitude and latitude coordinates. As a result of
this study, NOAA Vegetation Index Data were proved to be a suitable data for world wide
vegetation monitorings.
1. INTRODUCTION
In the last decade, the necessity of
global monitoring of vegetations or
bio-mass has become a more urgent matter.
Large forest areas in Asia and South
America are dissapearing because of cut
and burn agricultures as well as soil
erosions. In Africa and also in Asia,
Sahel areas are penetrated by deserts.
The total amount of vegetation dis
appearing- areas is estimated to be about
3 00,000Kin for a year, which corresponds
to be about the same of the area of Japan.
These vegetation decrease in a world scale
are causing heavy shortage of foods
production, which results in many people
starved especially in developing countries
in Africa. It may also cause a world
scale meteorological change.
In order to prevent the large disaster
which may be caused by these vegetaion
decrease, accurate conditions or stages of
the world vegetaion should be monitored.
It is obvious that this kind of monitoring
could be accompalished only through the
use of earth observation satellites.
However, past earth observation
satellite data were not appropriate for
this global monitoring purpose. Landsat
MSS data has proved that they are very
good tools for vegetation monitorings, but
it is almost impossible to use those data
in a global scale. There are two other
operational satellite systems. They are
TIROS/NOAA series satellites and weather
satellites in geosynchronous orbits.
However, the latter satellites are not
appropriate for vegetation monitoring
because of their wavelength ranges. They
lack the near infra-red channels which are
best for vegetaion discriminations.
Thus, the only satellite system which
can be used for this purpose is TIROS/NOAA
system. The mam sensor of these
satellites is AVHRR (Advanced Very High
Resolution Radiometer) and it has one band
in visible and one band in near infra-red
regeon. Their repetitive rate is twice a
day at least, and the ground resolution is
about lKm at the nadir. This ground
resolution is stil too high for monitoring
purposes, because it means that we need
about 500 million pixels to cover the
whole earth, and also we must mosaic about
18 paths of each ground coverage.
But now, we have more convenient data
for our purpose. It is called as a
vegetation index data made of these NOAA
data. In this data set, each hemisphere
is composed of 1024 x 1024 pixels in a
Polar Stereo projection, which is a
moderate data quantity for data pro
cessings and analyses. The purpose of
this study is to establish the method to
derive a global vegetation map from a NOAA
vegetation index data set.
2. NOAA SATELLITES
The TIROS/NOAA sereies satellites were
first launched in 1960 by NOAA (United
States National Oceanic and Atmospheric
Administration) for weather and ocean
monitoring. Recently these two satellites
were unified and called as NOAA satel
lites .
The orbit of this series satellites is a
polar orbit with about 99 inclination and
the altitudes are about 850Km. The main
sensor is AVHRR and it has 4(5) spectral
bands in visible, near and thermal
infra-red regeon. The swath width of
AVHRR is about 3000Km and one satellite
covers the same area twice a day in
ascending and descending mode.