Despite three years of experience with the National Environmental 

 Policy Act, no Federal agency has developed a "feedback" procedure so 

 it can quantify the types of problems being encountered in preparing 

 impact statements. This relatively simple procedure of screening their 

 impact statements for "unknowns" would allow agencies to sub- 

 stantiate their need for research in specific areas and would lend credi- 

 bility to the results. It would be of great value if all agencies periodical- 

 ly review their impact statements and compile a list of ecological 

 difficulties encountered from gathering preparatory data through 

 preparing the final impact statement and implementing the proposed 

 activity. 



Recommendation 6 Prepare environmental impact statements with high priority 



consideration to: 



a. Maximum practicable use of quantitative analysis, 

 prediction, and evaluation of ecological impacts; 



b. Post-impact evaluation of predicted impacts and 

 unpredicted consequences on a selective basis; and 



c. Identification and initiation of research of problems 

 amenable to solution by additional ecological research. 



Agencies should publish follow-ups concerning their predictions of 

 environmental impact. Federal agencies should (a) estimate the kind 

 and amount of environmental impact they believe their activity will 

 cause, (b) provide information to decision makers to permit weighing of 

 environmental costs against activity benefits, and (c) if the activity is 

 terminated, estimate how long it will take the disturbed ecosystem to 

 recover or succeed to another "useful" condition. Agencies should, after 

 project implementation, follow-up and evaluate actual impact. 



Agencies should develop a mechanism to inspect their impact state- 

 ments for ecological problems amenable to research. This action will 

 require the aid of ecologists to translate environmental problems into 

 the proper types of research categories so a systematic approach may 

 be made in solving the problem. 



A number of benefits could accrue from these activities. The reliability 

 of agency predictions could be validated. Accurate predictions would 

 indicate that agencies have "adequate" environmental information; 

 widely missed predictions would indicate that the agency needs to 

 conduct additional research and perhaps their proposed activities need 

 to be delayed until adequate data becomes available. At present, once 

 an activity is initiated no matter what its resulting impact is, the 

 activity often continues. 



Recommendation 7 Complete the existing National System of Natural Areas, with 



full representation of major ecosystems, to provide sites for 

 studies of naturally functioning systems that can serve as 

 ecological reference points for baseline monitoring, and as 

 controls for experimental research. Utilize existing Federal 

 lands to the maximum extent possible. Classify the natural 

 areas into a useful system. 



24 



