- 3 
PHILOSOPHIC APPROACHES TO DAMAGE DETECTION 
A. Initial Detection and Analysis of Damage 
From the above discussion, it should be clear that it is pos- 
sible to classify damage to vegetation. However, the suggested approach 
differs radically from the classical entomology and pathology textbook 
approach--including the approach suggested in the Manual of Remote 
Sensing published by the American Society of Photogrammetry in 1975 
(Thorley, et al. 1975), e.g., leafeating, sap sucking and meristematic 
insects, root and stem diseases including diebacks and wilts, foliage 
diseases, etc. The difference is based on the realization that one 
particular agent can cause a wide variety of damage syndromes and con- 
versely a given damage syndrome can be caused by any one of a large 
number of agents. Nautiyal and Waters (1975) have indicated that "The 
simultaneous occurrence of different symptoms, and the association of 
more than one organism with a single effect should be recorded, other- 
wise analysis and interpretation of data are limited and possibly 
erroneous." Because of this fact: "Surveys of particular areas should 
cover the entire life span of a crop." Surveys should be periodic, 
designed primarily to assess the crop at critical periods, either during 
any possible attack, or at given intervals for surveillance or check-up 
purposes. With periodic surveillance and a damage classification system, 
damage symptoms may be described on an evolutionary basis. It may be 
noted that general progression of damage symptoms ranges from Type IV 
damage to Type I damage. The rapidity with which the symptoms of damage 
approach Type I damage (total defoliation), indicates to a large degree 
the variety of the cause, and the subsequent urgency or control action. 
What if damage occurs, or becomes obvious to remote sensing 
detection the day after the flight occurs? A present problem is that 
many remote sensing programs are of a "one-shot nature", but the satel- 
lites have made us aware of the benefits of repeated coverage. A 
second temporal aspect is the need to respond to emergency situations, 
such as those that arise from industrial accidents, e.g., SO, leakages. 
In many areas this capability is already being used on a crisis-basis. 
We should have the facility on a permanent basis. 
It is suggested here that the approach to remote sensing 
damage detection should: a) involve an holistic approach, with attention 
given to detection and description on the basis of damage types for 
entire vegetation communities, and b) include a temporal surveillance 
aspect. However, the sheer size of the extensively managed areas in- 
volved, presently precludes the implementation of such a system because 
of "economic constraints",