The test sites are classified with the highest thematic detail
(crop or crop type). By combining the (classified) medium
resolution images with the (classified) high resolution
images this thematic detail will be extrapolated to the entire
test region.
In order to identify some specific smaller land cover classes
or local (small-scale) phenomena, level 3 may even have
not a sufficiently high spatial resolution. As a result, the
methodology is further extended to high resolution satellite
images, like the TK-350 and KVR-1000 Russian
photographic space information sources with a spatial
resolution of less than 10 meters (level 4).
Panchromatic images of very high resolution (TK-350,
KVR-1000) can be used as a source of vector data for GIS
layers, providing information on
- urban areas and settlements;
- roads, railways;
- hydrographical elements (rivers, lakes, ponds, channels);
- some types of land cover (forests, shrubs, meadows,
marshes, etc.).
In comparison with scanner data, usage of panchromatic
data leads to a more precise extraction of contours and
gives possibility to take texture and geometric peculiarities
of the images into account. Together with ancillary data,
panchromatic images allow to carry out a more detailed
stratification of the territory.
In this study, an integrated technology based on using
multispectral scanner images in combination with
photographic data is developed to improve the accuracy of
land cover classification. An overview of the mentioned
systems is given in Table 1.
3. AGRICULTURAL LAND USE IN RUSSIA
3.1 General Information
The territory of the Russian Federation is about 17 million
square kilometres (mln.sq.km). The territory is divided into
autonomous republics, autonomous regions and okrugs,
krajs, oblasts (further on this hierarchical level will be
referred to as "region", or "oblast", without taking into
account the difference between federal units).
The area of the regions differs significantly, varying from
18 to about 600 thousand square kilometres (ths.sq.km) in
the European part of the country, and up to three million
sq.km in Siberia. The average size of a region in the zone of
intensive agricultural production in the European part of
Russia makes up about 50 ths.sq.km. Every region is
divided on its turn into "districts" ("rayons"), the average
district area in the European part being about 1.0-2.5
ths.sq.km.
Table 1. Specifications of some satellite sensor systems.
Sensor Spectral Spatial Swath
bands resolution width
NOAA- 0.58-0.68 um 1.1 km 2800 km
AVHRR 0.72-1.0 um (nadir)
3.55-3.93 um
10.3-11.3 um
11.5-12.5 um
MSU-SK 0.5-0.6 um 175 m 600 km
0.6-0.7 um
0.7-0.8 um
0.8-1.0 um
10.3-11.8 um 590 m
MSU-E 0.5-0.6 um 45 m 45 km
0.6-0.7 um
0.8-0.9 um
Landsat-TM 0.45-0.52 um 30m 185 km
0.52-0.60 um
0.63-0.69 um
0.76-0.90 um
1.55-1.75 um
2.08-2.35 um
10.4-12.5 um 120m
SPOT 0.50-0.59 um 20m 117 km
0.61-0.68 um
0.79-0.89 uum
0.51-0.73 um 10m
TK-350 BW film 10m 200*
Stereo 300 km
KVR-1000 BW film 2m 40 km
In the economics and statistics a notion "an economic
region or an economic area" is often used, which means a
group of regions with similar natural, climatic and
economic conditions. There are 12 economic regions in the
territory of the Russian Federation.
According to the natural-agricultural zonation of land, there
are eight zones in Russia from the polar-tundra to the semi-
desert ones, excluding mountains.
3.2 Agricultural Production
The agrarian sector in Russia produces (at an average)
about 15% of the gross national product. The agricultural
production is based mainly upon the big collective
enterprises. The total number of enterprises occupied in
agricultural production is about 30,000. A specific feature
of such collective enterprises is the rather large size of
fields (50-100 and more ha in the European part), which
simplifies the identification process of land use mapping
and monitoring.
98 International Archives of Photogrammetry and Remote Sensing. Vol. XXXII, Part 7, Budapest, 1998
lar
av
spi
cei
49 2
Tv
to
