106 



3.1.2 Discussion: Value of a Process Oriented Ecosystem Study? 



Since the development of process oriented ecosystem studies are a new undertaking for MMS, it is appropriate 

 that we consider the overall value of this approach. A process study is markedly different from previous MMS 

 ecosystem studies which stressed comprehensive species inventory rather than system function. Ecosystem 

 inventories are of limited overall value since it is impossible to determine the importance of component species 

 and there are no objective criteria for setting the scope of the study. By contrast, process studies dispense with 

 efforts to be comprehensive and devote all resources to determination of the most important functional 

 components of a system. Properly done, a process oriented study will have greater management value than a 

 description. 



The primary intellectual tool of process studies is systems analysis or general systems theory. Systems analysis 

 can be characterized as an approach to research in which the objective is to gain a predictive capability about 

 the behavior of complex systems. Rather than structuring research around an often artificial dichotomy in which 

 a null hypothesis is either proven or not proven false, system analysis seeks to partition complex systems into 

 more tractable natural subdivisions. While simpler than the whole system under analysis, the subsystems usually 

 retain too much complexity for dichotomous hypothesis testing. Therefore, research is designed to estimate 

 parameters associated with state variables and state transitions. Typically, the goal of developing predictive 

 capability is achieved in the form of an environmental systems model. 



Systems analysis contributes to ecosystem management in several ways: (1) differentiating between natural 

 variability and anthropogenic environmental stress (and thus minimizing incorrect assumptions about impacts 

 where none exist, or uncovering impacts where none are obvious), (2) in the efficient use of study resources, (3) 

 answering specific questions about target species (e.g., important fisheries or important food resources), (4) 

 predicting the severity of proposed oil and gas impacts, and, (5) offering substantial predictions (hindcast and 

 forecast) about the behavior of the system under varying natural and man-imposed regimes. For example: 



• Variability - What is the explanation for the observed variability? Will it be assigned to the wrong 

 cause? All previous monitoring studies on this shelf have led to the conclusion that the natural long- 

 term variability and influence of the river make definitive causal relationships between ecosystem 

 behavior and OCS impacts impossible to detect. In the present absence of understanding how the 

 ecosystem works, there is a sure and significant risk that actual impacts arc either left undiscovered, or 

 that inconsequential relationships are assumed to be severe impacts. The relationship between the 

 observed variability and true impacts may be of three types: coincidental (either causal or non-causal), 

 random, or partially related (see Figure 14). One may incorrectly assume that causal relationships are 

 absent if the natural variability is disguised by the natural noise in the system. 



• Experimental design - Building the best sampling protocol depends on how and what questions are 

 asked. For example, general impacts on community primary production may be important concerns of 

 interest. However, knowing how production changes due to variability of physical, toxicological, or river- 

 nutrient related processes involves an understanding of how biological communities integrate all three 

 driving forces, not just one of them. Simple toxicological studies (single species) are not usually sufficient 

 to answer questions about the community behavior, because of the integrated nature of ecosystem parts. 



• Prediction - Monitoring alone is not prediction; prediction, needed for management, requires knowledge 

 of how the whole ecosystem works, not of just the parts. 



• Interdependence - A system is a set of interacting interdependent units, of which an ecosystem is an 

 example. Only by investigating the key components, e.g., nutrients and phytoplankton, can we expect 

 to understand what causes variations of the benthos in time and space, and ultimately the fisheries 

 resources that might be affected by offshore oil and gas recovery activities. The result is understanding 

 how the TEXLA shelf functions, rather than stochastic relationships that are empirical and behind 

 which there is no insight into how one part of the system affects another. 



