Reaction to perceived degradation provides for assessing the present and 

 past condition or status of selected environmental components, to determine 

 whether, why, and how much change has occurred. The approach is largely 

 descriptive, although it may improve general understanding of ecosystem responses 

 to stress and may identify problems that should be dealt with in the future. 

 One example of such an assessment was the large-scale investigation to determine 

 the effects of oil drilling and production in Louisiana coastal waters (Ward et 

 al. 1979). 



Reaction to planned activities is somewhat more anticipatory in that the 

 present status of the environment is analyzed to determine whether the planned 

 activities are likely to create or cause problems. This approach may lead to 

 the identification of stipulations intended to alleviate the potential problems. 

 The Outer Continental Environmental Assessment Program for oil and gas develop- 

 ment in Alaska (Engelmann 1979), for example, has contributed to the design of 

 stipulations on the sale and operations of lease tracts in the Arctic (Weller 

 et al. 1979). 



Historically, environmental assessment programs have contributed greatly 

 to our ability to describe present or past states of the environment and 

 considerably less to our understanding of how various ecosystem components 

 interact and respond to man-induced perturbations to the system. Present 

 assessment practice has been criticized as both ineffective and wasteful (Ward 

 1978; Holling 1978), primarily because of the reactive and descriptive character 

 of the studies and the lack of coupling with the decision processes which the 

 assessments purportedly support. 



If one accepts as a basic premise that a primary intent or expectation of 

 the environmental assessment is to guide the formulation of policy or to aid 

 decisionmaking through the identification of the optimal course of action from 

 among a selection of proposed alternatives, then there are several implications 

 for the environmental assessment itself. First, it is essential that much 

 greater emphasis be placed on predicting the possible future states of the 

 environment that may result from adoption of those policies and alternatives. 

 Furthermore, the environmental research required for such an assessment should 

 be strongly focused by the concerns of the decision or policy problem. Early 

 recognition of the elements of a rational decision process can help considerably 

 in focusing the environmental assessment on those ecological processes and 

 consequences that are most important and useful to the waste -management decisions 

 that may ultimately influence or modify the magnitude or extent of marine 

 pollution. 



A further implication of the foregoing premise is that the classes of 

 modeling efforts required for the assessment are not restricted to the environ- 

 mentally oriented models, such as ecosystem and population dynamics, circulation 

 and trajectories, etc. These types of effort must be supplemented by probabilistic 

 modeling or risk assessment (Kates 1978) and value modeling, 



both of which have long been recognized as important components of decision- 

 analysis (Howard 1966, 1968; North 1968). In this paper, I review the implications 

 of orienting marine pollution assessment strongly toward environmental management 

 and decisionmaking and identify some of the most important modeling needs from 

 a decision-oriented perspective. 



