17. REMOTE SENSING REQUIREMENTS TO SUPPORT FOREST
INVENTORIES
157
it is possible to get some information at low cost that would be very
expensive with field measurements, e.g., ecological information,
landscape diversity, ecological corridors
output data is a model of forests, it can be used for simulation studies,
e.g., for simulation of sampling errors of different field sampling designs
and for simulation alternative future scenarios of forests.
6.
THE FINNISH MULTI-SOURCE FOREST
INVENTORY
The Finnish Multi-Source National Forest Inventory (MS-NFI) is
described shortly below because it is one of the few operative inventories
utilizing space-borne remote sensing data. Finland’s forest resources have
been investigated by means of eight National Forest Inventories since the
year 1921 (Ilvessalo 1927). The information has been utilized in large area
forest management planning, such as determining the level of cuttings and
other treatments needed, and has formed the information basis for official
forest policy making and for the strategic planning of the forest industries.
During the eighth inventory (1986–1994), a multi-source inventory
system was developed. It utilizes satellite images and digital map data in
addition to ground measurements. For a real forest inventory, it is not
sufficient to utilize remote sensing data in such a way that only some classes
are derived, e.g., on the basis of tree species dominance. The image analysis
and parameter estimation method has been designed in the Finnish MS-NFI
so that it is possible to estimate all inventory parameters for areal units of
about 40 hectares or larger (Tomppo, 1991, 1993 and 1998). The ninth
inventory, which started in 1996, also contains the measurement of some
additional characteristics describing forest biodiversity.
Examples of parameters measured in MS-NFI in the field are given
below to provide an idea of the kind of estimates that should also be
produced in a remote sensing based inventory. Stand level variables
measured in Finnish MS-NFI are:
general data of field plot cluster (inventory type, record type, crew
leader, coordinates of cluster, date, inventory area),
plot identification data (plot number, coordinates, administrative
information, multiple use, plot size, etc.),
site data (e.g., land class and its changes, direction and distance to the
closest stand boundary, main site fertility type, mixture of site fertility
types, specification of mire type, type of soil, quality and thickness of
INVENTORIES
157
it is possible to get some information at low cost that would be very
expensive with field measurements, e.g., ecological information,
landscape diversity, ecological corridors
output data is a model of forests, it can be used for simulation studies,
e.g., for simulation of sampling errors of different field sampling designs
and for simulation alternative future scenarios of forests.
6.
THE FINNISH MULTI-SOURCE FOREST
INVENTORY
The Finnish Multi-Source National Forest Inventory (MS-NFI) is
described shortly below because it is one of the few operative inventories
utilizing space-borne remote sensing data. Finland’s forest resources have
been investigated by means of eight National Forest Inventories since the
year 1921 (Ilvessalo 1927). The information has been utilized in large area
forest management planning, such as determining the level of cuttings and
other treatments needed, and has formed the information basis for official
forest policy making and for the strategic planning of the forest industries.
During the eighth inventory (1986–1994), a multi-source inventory
system was developed. It utilizes satellite images and digital map data in
addition to ground measurements. For a real forest inventory, it is not
sufficient to utilize remote sensing data in such a way that only some classes
are derived, e.g., on the basis of tree species dominance. The image analysis
and parameter estimation method has been designed in the Finnish MS-NFI
so that it is possible to estimate all inventory parameters for areal units of
about 40 hectares or larger (Tomppo, 1991, 1993 and 1998). The ninth
inventory, which started in 1996, also contains the measurement of some
additional characteristics describing forest biodiversity.
Examples of parameters measured in MS-NFI in the field are given
below to provide an idea of the kind of estimates that should also be
produced in a remote sensing based inventory. Stand level variables
measured in Finnish MS-NFI are:
general data of field plot cluster (inventory type, record type, crew
leader, coordinates of cluster, date, inventory area),
plot identification data (plot number, coordinates, administrative
information, multiple use, plot size, etc.),
site data (e.g., land class and its changes, direction and distance to the
closest stand boundary, main site fertility type, mixture of site fertility
types, specification of mire type, type of soil, quality and thickness of