combination and the resulting image pattern was compared
with the ground truth map. The results were highly
satisfactory and the factor combination in the G.E 100
produced results which were confirmed by the ground truth
maps.
The effect of "shadowing" because of terrain form was found
to be separable from the absorbtion effects from Araucaria
by stretching the "contrast". Invariably the Araucaria
areas showed a darker image than the non-Araucaria areas.
The subsequent use of thirteen computer compatible tapes
to cover the uhole of the temperate and subtropical forest
areas was put in hand and applying the same procedure the
Araucaria forests were identified. As they were confirmed
on the Television Screen, the image produced was recorded
on slides and indexed precisely. Subsequently, the slide
was projected onto a base map and the results traced off
onto the map, and this permitted the mozaic of slides to
be built up into a controlled map of the whole area.
At all times, the results uere checked against any aerial
photographs and forest survey reports, to ensure that the
basic methodology was being applied consistently.
A similar procedure was subsequently applied to mixed
forests in South Wales using the Plessey IDP 3000. In this
case the mapping procedure was facilitated by pro jecting the
slide onto a sand blasted glass screen which permitted work
to be done without the operator getting into his own light.
CONCLUSION
In this paper the general methodology for successfully
identifying Araucaria and other forests using computer
technology,has been discussed. This Conference has