image’, is a high spatial resolution monochromatic 
reference image which permits the spectral data to 
be accurately located with respect to targets on 
the ground. This ability to act as a spectrometer 
and imaging device at the same time greatly 
facilitates acquisition of high quality spectral data 
which are accurately geo-located. 
4.0 FORESTRY APPLICATIONS 
Airborne linear array imagers can be used 
for a variety of forestry applications to provide 
forest information in a more efficient and cost- 
effective manner: 1) as an alternative to aerial 
photography for forest inventory, 2) for mapping of 
major forest changes (e.g. clearcuts, burned areas), 
3) for assessing insect and disease damage, and 4) 
for providing sample data for various forestry 
applications. To meet these requirements, a 
airborne linear array system must provide efficient 
cost-effective area coverage and efficient data 
processing. 
4.1 Forest Inventory 
Forest inventories are the basis for the 
management of British Columbia’s forest resource. 
The Province’s forests cover an area of 
approximately 52 million hectares, contain close to 
8 billion cubic metres of merchantable timber and 
represent over 40 percent of the country’s timber 
supply. The Province’s forest inventory database 
is currently reinventoried on a ten year cycle 
using large-scale aerial photography as the 
primary information source. Between reinventories 
the database is updated on a regular basis for 
major disturbances and new harvesting activity 
using Landsat Thematic Mapper (TM) imagery. 
Forest inventory mapping for management 
purposes requires the delineation and description 
of forest stands on the basis of species 
composition, density, age and height. In order to 
accomplish this, it must be possible to discern and 
interpret individual trees (implying resolution of 
less than 1 meter), and high quality stereo 
imagery must be available. Research has 
demonstrated that high resolution airborne 
imagery can provide such detail (Gougeon and 
Moore, 1988). 
The advantages of linear array imagers for 
forest inventory are large when a GIS 
infrastructure is in place. Linear array imagers 
through the use of optimized wavelength bands 
and optimized enhancements will likely provide 
forest typing of equivalent or perhaps better 
accuracy than conventional aerial photography but 
with a greater ease of interpretation due to the 
digital nature of the data. 
The major benefits will result from the 
impact on the total inventory production system, 
particularly the cost savings of integrating the 
data and interpretation directly with existing data 
on a GIS. This eliminates the costly process of 
transferring interpretation onto a map base. 
4.2 Mapping Forest Change 
The capability to efficiently geometrically 
correct linear array imager data and superimpose 
multiple dates of imagery makes such data 
suitable for mapping forest change and for directly 
updating forest inventories on a GIS. Major 
changes such as clearcuts and many burned areas, 
which may require only delineation and no 
quantification of levels of change can be easily 
detected and delineated by single or multi-date 
high level linear array imager data (Robertson and 
Nesby, 1986). 
If operational post-flight processing 
facilities can be developed it is possible that near 
real-time airborne imagery can acquired and be 
used for highly time-sensitive applications such as 
monitoring forest fires or insect migrations. 
4.3 Insect and Disease Damage 
Assessment 
Insect damage surveys are necessary for 
planning control strategies, for assessing the 
effectiveness of control strategies, for planning 
harvest schedules and salvage logging, and for 
predicting timber losses which are likely to be 
incurred. For damage assessment, the main need 
is for accurate assessments and geographic 
placement of damage levels over large areas. 
Airborne linear array data at a spatial resolution 
of approximately 5 m over a wide swath could be 
used for insect damage surveys. The use of 
narrow wavelength bands (5 to 10 nm) is 
important for identifying affected areas at the 
earliest possible stage of infestation. 
Recent experiences of forest insect 
infestations and their resulting impact highlight 
the requirement for efficient and accurate methods 
of assessing insect and disease damage. Large 
areas would have to be covered, generally on an 
annual basis. In western Canada, for example, 
over one half million hectares are effected by 
mountain pine beetle damage involving over 8,000 
infestations. Kneppeck and Ahern (1989) found 
that MEIS data at 1.4 m resolution was able to 
detect 1.36 times as many red attacked trees as 
were detected on 1:10,000 natural colour 
photography. Ahern (1988) concluded that 
airborne linear array data should be able to detect 
bark beetle stress in conifers long before the red 
attack stage by acquiring narrow band imagery in 
the red edge spectral region. 
Sketch mapping, the most common 
technique used for mapping damage is expensive 
and inappropriate for broad area estimates of 
damage. Increasingly, however, there is a demand 
for defoliation assessment on a stand basis so that 
damage information can be used more effectively 
389