and
and
Time
stem
is to
tural
f the
ience
)olar-
ental
ca at
a proposal for an operational environmental monitoring system
for Asia and the Pacific region. (Hielkema, 1990 & 1992;
Kalensky, 1992; Snijders, 1996).
5.2.2 Monitoring Agriculture with Remote Sensing
(MARS)
Status: ongoing, regional - the European Union countries
and several Central/Eastern European counties.
Organization: Institute for Remote Sensing Applications of
the European Union Joint Research Centre.
MARS started in 1988 as a 10-year pilot project for the
improvement of agricultural statistics in Europe through the
use of remote sensing. This required development of new
methodologies appropriate for the European agricultural land
use conditions. MARS implementation strategy consists of the
following four main components:
(a) regional inventories of acreages of agricultural crops;
(b) assessment and monitoring of vegetation conditions;
(c) forecasting average yield of selected crops;
(d) forecasting of crop production for selected crops.
Over 100 public and private sector institutions from about 20
European countries are participating in the implementation of
the MARS project. The main remote sensing inputs consist of
medium resolution Landsat and SPOT imagery, and low
resolution (1 km) NOAA-AVHRR optical and thermal
imagery. The applicability of radar imagery to agricultural
statistics in Europe is being explored. While the medium
resolution imagery is only obtained for coverage of sampling
units, total covearge is provided by low resolution imagery.
During the crop growing season, the NOAA-AVHRR imagery
is daily processed into two regional mosaics: the normalized
difference vegetation index and the surface temperature
differences. MARS priority is now shifting from developing
new methodologies to providing technical support for their
operational applications. (Bernard & Meyer-Roux , 1994).
$2.3 Forest Resources Assessment - 1990 (FRA-90)
Status: ongoing - global
Organization: Food and Agriculture Organization (FAO)
of the United Nations.
FAO, as part of its mandate, is conducting periodic
assessments of global forest resources. The latest one,
completed in 1995, was the Survey of Tropical Forest Cover
and Study of Change Processes (FAO, 1996). Its main
objectives were to provide reliable and globally consistent
information on the state of the tropical forest cover and the
rates of its change during the 1980s. Sampling design was
based on two-stage stratification with 10% sampling intensity
(117 sampling units out of the total of 1203). The area of
sampling units (SU) for the assessment of the state of forest
cover corresponded to one Landsat scene (34000 sq.km). The
SU for change assessment were smaller because they
corresponded to overlapping areas of two multidate (around
1980 and 1990) Landsat images. In the humid tropics, the
overlap area was often further reduced by clouds. If the cloud-
free overlapping area was smaller than 10000 sq.km, the SU
423
was rejected. It happened in two SU, one in Colombia and one
in Papua New Guinea. Landsat hard copy images at scale
1 : 250 000 were interpreted at the regional and national
institutes familiar with forest resources in the SU areas.
Results were compiled in the form of global, regional and sub-
regional statistics. and three types of maps: current and
historical state of forest cover, and change maps. The “state of
forest” maps have 10 land cover classes, the “forest change”
maps 15 change classes. The area of the smallest mapping unit
is 50 hectars. Future plans include testing the usefulness of
satellite SAR imagery for global forest assessment, expanding
the network of participating countries, transforming of FRA-90
into a program for continuous assessment of global forests,
and strengthening the existing linkages with a complementary
project *Tropical Ecosystem Environment Observation by
Satellite” (TREES) of the European Commission Joint
Research Centre - Institute for Remote Sensing Applications.
TREES is developing a methodology for forest mapping in
humid tropics, based on the NOAA-AVHRR 1 km imagery.
(Drigo, 1996; D'Souza et al., 1995; FAO, 1993; Lanly, 1992;
Malingreau et al., 1994; Singh, 1992).
5.2.4. Global Terrestrial Observing System (GTOS)
Status: ongoing - global.
Organizations: Food and Agriculture Organization (FAO)
of the United Nations;
United Nations Environment Programme
(UNEP),
World Meteorological Organization (WMO);
United Nations Educational, Scientific and
Cultural Organization (UNESCO);
International Council of Scientific Unions
(ICSU).
The Global Terrestrial Observing System (GTOS) is one of
three major global monitoring systems established in mid-
1990s as a follow-up to the Earth Summit. The other two
systems, complementary to GTOS, are the Global Ocean
Observing System (GOOS) and Global Climate Observing
System (GCOS). All three systems are interlinked. Close
coordination of their implementation is an essential
requirement for the achievement of their objectives. GTOS
was established in response to a growing need for a
systematic, long term monitoring of changes in the natural and
managed ecosystems at a global level. Its Scientific Secretariat
is hosted by FAO. GTOS plan of operation is based on a
global network of some 100 field sites and transects, covering
all the major terrestrial ecosystems. Its hierarchy of
observational levels includes EO satellites, aerial platforms
and surfase-based systems. Remote sensing will play a vital
role in the following three areas:
(a) monitoring changes in land cover and land use over large
areas,
(b) extrapolating local observations from GTOS field sites to
larger areas;
(c) ensuring a consistent set of measurements worldwide.
Data management will be implemented at three levels: (i) the
field sites (national level), (ii) the regional and thematic
centres; and (iii) the global coordinating centre, managing
the GTOS meta-database. An important characteristic of
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B4. Vienna 1996
