ecosystems. The first level involves the radical change or complete removal 

 of a given parcel of the system. Individual and cumulative perturbations of 

 the substrate and floral components of wetlands will lead to changes in the 

 fauna. These changes are predictable because major interdependent relation- 

 ships between these elements of the system are known. 



If the isolated alterations are of small magnitude, the functioning of 

 the ecosystem is unimpaired. The preceding discussion of incremental losses 

 of habitat and species displacement in coastal uplands is also applicable to 

 marsh ecosystems. The pristine wetlands inherently experience natural fluctua- 

 tions in various components, and they contain compensating mechanisms to 

 accommodate such fluxes. Only when threshold levels of critical links and/or 

 components are reached does the ecosystem change noticeably. This, of course, 

 represents the second level of ecological impacts. Unfortunately, these 

 ecosystem threshold levels are unknown. 



Water is the major factor in maintaining marsh systems. Salt marshes and 

 fresh marshes depend primarily on inputs of a single type of water - saline 

 and fresh, respectively. Brackish marshes and delta marshes characteristi- 

 cally receive inputs of both water types, although not in equal proportions. 

 The movement of water through the system is the important force driving and 

 controlling the wetlands. Because roads, levees, and canals alter water 

 flows, they are considered to be the source of the most important impacts at 

 the ecosystem level. Indeed, they often become the boundaries separating two 

 ecosystems. 



The effects of man-induced landforms on the ecosystem are not nearly as 

 predictable as their impacts on the small specific sites they occupy. Two 

 reasons for this lack of predictability are apparent. First, the ecosystems 

 operate over a larger area and a longer time scale than do the alterations. 

 Second, the orientation of canals, levees, and roadways is such that the water 

 flows of the ecosystems do not experience complete major changes: "all or 

 none" situations are few. This means a relatively long period of time must 

 pass before natural fluxes/cycles encounter these partial alterations. By the 

 end of such a time period, additional and/or different alterations have fre- 

 quently occurred. 



These facts indicate that the way to maintain an ecosystem during petro- 

 leum exploration and extraction is to manage the ecosystem. But the dilemma 

 is unavoidable - management decisions are typically concerned with relatively 

 small areas and must occur over a short time scale; moreover, there is insuf- 

 ficient time to observe the wisdom of the choices, reflected by long-term 

 changes. It would be highly desirable for the land manager to have, before 

 major petroleum operations occur, a development plan for an entire oil or gas 

 field. Presently, this does not and cannot occur. Drilling and extraction 

 decisions are intermittent and based on cumulative information. Each new well 

 provides data which are utilized in the decision of whether to drill another; 

 in addition, the degree of speculation varies with the decision maker, the 

 price of petroleum products, and other factors. There is no reason to expect 

 this dilemma to change in the near future. 



140 



