385
EVALUATING AIRBORNE LINEAR ARRAY TECHNOLOGY
FOR OPERATIONAL FOREST INVENTORY APPLICATIONS IN BRITISH COLUMBIA
Paul G. Pilon
Remote Sensing Section
Inventory Branch, Ministry of Forests
British Columbia, CANADA
ABSTRACT
Canada, over the last decade, has played a leading role in the design and operation of airborne linear
array sensors and the development of specific applications for forestry. These state-of-the-art imaging systems
have the potential to revolutionise forest management by providing more efficient and cost-effective sources
of digital information which are readily compatible with computer-based information systems. The improved
radiometric, spatial and spectral performance of data from these sensors are able to provide improved feature
discrimination when compared to existing optical-mechanical sensors.
A review of the pertinent published literature on the use of airborne linear array data for forestry
applications indicates that such data can provide improved, forest inventory mapping, regrowth monitoring,
insect and disease damage assessment, inventory update and forest sampling, and can offer multispectral
imaging with stereoscopic and superior cartographic accuracies.
Research performed in forestry using airborne linear array data combined with recent advances in
inertial navigation and global positioning systems, image analysis and GIS technology, optical data storage
and artificial intelligence are at a stage where operational applications are becoming feasible.
Key Words: Airborne Linear Array, Forest Inventory, MEIS, FU, CASI
1.0 INTRODUCTION
Obtaining detailed and up-to-date resource
information is essential for effective forest
management. With the advent of computer-based
information systems it is becoming increasingly
important that resource data be computer
compatible and be capable of being computer
processed in an efficient and cost-effective manner.
Airborne linear array sensors can provide detailed
digital data at resolutions comparable to
conventional aerial photography. The improved
radiometric, spatial and spectral performance of
data from these sensors provide improved feature
discrimination when compared to existing optical-
mechanical sensors. Canadian scientists, over the
last decade, have played a leading role in the
advancement of linear array technology and the
development of specific applications for forestry.
Research using the MEIS (Multispectral
Electro-Optical Imaging Sensor) linear array
imager has demonstrated a large potential of such
technology for forestry applications. MEIS data
has shown excellent potential for softwood species
identification and discrimination (Leckie and
Dombrowski, 1984), regrowth monitoring
(Kneppeck and Ahem, 1987) and insect and
disease damage assessment (Ahem et al., 1986;
Epp and Reed, 1986; Kneppeck and Ahem, 1989).
imaging areas of different forest type (Edel and
Bianchi, 1986) and has been used to monitor forest
stress and forest disease and damage (Miller et al.,
1986; Rock et aL, 1988; Banninger, 1988). The
success of research results from these sensors, to
date, has been significant enough to warrant
further investigation directed at developing
operational forestry inventory systems which would
utilize airborne linear array imager technology.
2.0 BACKGROUND
Linear array imagers are a significant
advancement in electro-optical sensor technology
and have provided unique opportunities for the
development of operational forestry applications.
Electro-optical sensors have undergone three stages
of evolution since the late 1960’s:
First Generation
single detector optical-mechanical line
scanners (e.g., Landsat MSS and TM and
the airborne Daedalus MSS);
Second Generation
multi-element linear array pushbroom line
scanners (e.g., MEIS II and SPOT HRV);
Third Generation
two-dimensional multi-element array
imaging spectrometers (e.g., the Canadian
FLI and CASI systems and the American
AIS).
The FLI (Fluorescence Line Imager)
system, although designed primarily for ocean
applications, has been demonstrated over land for
