The purpose of this project is to provide a level of mapping and 
description of wetland ecosystems that will be useful to wildlife manage- 
ment, land use planning and the assessment of the environmental impact 
of development. As no single level of ecological classification can 
meet all these needs, it will, in effect, be a level of mapping and des- 
cription of patterns of wetland ecosystems which balances user needs 
against what is a practical program in terms of funding, time, personnel 
and contemporary technology. The minimal objectives of this project are, 
therefore, as follows: 
  
  
      
    
      
      
    
       
      
   
   
  
  
  
  
   
  
  
  
  
  
  
   
  
  
  
  
  
  
   
  
  
   
  
  
     
1. To provide a synoptic overview of the ecosystem patterns of the entire 
region which will be useful in highlighting ecosystem patterns which 
have significantly different wildlife habitat values and would have 
different responses to a particular kind of development manipulation. 
2. To provide an overall framework of wetland ecosystem patterns which 
will be useful in the planning of more intensive levels of ecosystem 
study in small areas and in extrapolating the results of these studies. 
Figure 
Landsat imagery is the ideal tool for use in an undertaking such as 
this. It provides a synoptic overview of distinctive patterns of wetland 
ecosystems in that it records differences in reflectance values from veg- 
etation growth. As the vegetation of a wetland type is the best indic- 
ation of the transactions among the biotic and environmental features of 
the ecosystem, a Landsat image provides "ready-made" delineations of wet- 
land types which will survive, with only minor alterations, to the final 
published maps. The fact that Landsat provides temporal data also makes 
it an invaluable aid as it records the seasonal changes that occur in 
wetland types and which serve to identify them. For example, fens have 
a distinctive appearance on spring imagery obtained just after the ice 
has melted - dark,because of the higher water level, and surrounded by 
drainage patterns - which helps to delineate them and distinguish them 
from sphagnum-covered wetland types such as bogs. The location of a fen 
would be much less distinguishable on summer imagery (Figures 3a and 3b). 
Preliminary identification of the wetland types is, therefore, done 
directly onto the 1:1,000,000-scale Landsat imagery, in colour composite 
format, which enhances recorded variations in reflectance across the scene. 
The basic classification system used is that developed by Jeglum et al 
(1974), but the level within this framework to be employed within one 
Landsat scene depends upon the level of detail which the imagery can 
provide. Since any level of detail which it is possible to obtain would Figure 
not be sacrificed merely to achieve a consistent level of classification 
throughout a map area, the full spectrum of types and patterns used in 
mapping will not be known until the entire area has been mapped. These ! 
virtual 
on Fig: 
  
Aerial photography (1:60,000) is used to specify the locations of 
sampling stations, only accessible by helicopter, which are chosen on the 
basis of the following objectives as related to interprecation of the 
Landsat data: