1. The continued controversy over the type or types of land classification to be 

 used in resource inventories, assessments, and planning has not been resolved. 

 Despite interagency cooperative agreements, few objective evaluations of the 

 process appear to be underway. There is a need to evaluate the effectiveness 

 and efficiency of various classification approaches. This will require an 

 analysis of both the role of land classification and the kind of information it is 

 expected to provide. With such analysis, the ability of various systems to 

 deliver appropriate information can be evaluated. 



2. Ecological and classification is meant to be an integrated approach to land 

 survey. As such, the physical and biotic characteristics of land and their 

 interactions must be studied and integrated. Wildlife is perhaps one of the most 

 difficult components to fit into an ecological land survey. Taylor^'' has 

 summarized the reasons as follows: 



a. Animals are less conspicuous than components such as vegetation and 

 landforms. Although the largest species may be checked by aerial census, 

 very few species are suitable for remote sensing; 



b. Whereas landforms or vegetation are sedentary, the mobility and behavior 

 patterns of animals make them difficult to study within a short time; and 



c. Habitat units perceived by an animal may or may not coincide with 

 identified land ecosystems, or with all parts of any particular land 

 ecosystems. For example, a pika may recognize only one talus slope as 

 important, in contrast, elk may recognize various areas throughout the 

 mountain range as important at different times of the year. 



Despite these difficulties, our approach should incorporate the wildlife 

 component if we are to conduct a fully integrated land survey. To accomplish 

 this, we need a clear definition of the term "wildlife" and a rationale for 

 incorporating wildlife into ecological land classification schemes (i.e., of what 

 value would wildlife information be?). 



3. Existing classification systems usually emphasize the soil/ vegetation complex, 

 mainly because land classifiers do not understand aquatic habitats. In most 

 landscapes, water bodies are so intricately associated that integrated survey is 

 essential. This need is emphasized by the fact that aquatic ecosystems are 

 controlled by the lands around them. This is a key point because a holistic 

 approach should be capable of recognizing integrated terrestrial/ aquatic 

 systems. Attempts to relate to land through separate systems have, in part, had 

 limited success because they were regarded as independent systems. 



We need a nationally accepted method that will compatibly incorporate water with 

 the surrounding terrain. Such a method being developed is part of the work on the 

 water component of the National Site (Land) Classification System (T. Terrell, Fish 

 and Wildlife Service, personal communication). Platts^' has also reported on studies 

 that integrate aquatic ecosystems with terrestrial ecosystems through the land 

 systems inventory concept. The Environmental Protection Agency is working on a 

 rationale for unified and practical land/water classification. ^^ Although this effort is 

 promising, much additional work needs to be done to implement such an approach. 



Clearly, to accomplish these tasks, coordination is necessary. Besides interagency 

 agreements to coordinate programs for classification and inventory of natural 

 resources, there is a need to assure that coordination takes place within a broader 

 context to deal with the questions of philosophy, application, and definitions. A 

 vehicle for such coordination could be modeled after the very successful Canada 

 Committee on Ecological Land Classification." 



SUMMING UP 



Systems for classifying and evaluating land as ecosystems have evolved in different 

 agencies of the federal government over the past decade. Such systems involve the 



23 



