mathematical population model whether or not those who headstart 

 turtles are aware of this or are willing to admit it. 



As I said, my purpose is not to criticize headstarting 

 projects or those who run them. My purpose is to show that any 

 management effort we are using is based, even if unknowingly, 

 upon a mathematical population model. The same argument could 

 easily be shown to apply to hatcheries, the implementation of 

 TEDs, the setting of size limits on harvests, or any other 

 management decision that has been made. 



In the past, when some of us have called for the development 

 and implementation of explicit mathematical population computer 

 models to help determine management options, opponents have said 

 things like: We don't have enough information to develop 

 population models or it's too early to base management on 

 population models. 



Well, as I have shown, current management practices are 

 already based on (tacit) population models. To those who say "We 

 don't have enough information to develop population models," I 

 find myself asking "Why do current (tacit) models ignore 

 information that is available?" To those who say, "It's too 

 early to base management on population models," I ask, is it too 

 late to base management decisions on more explicit population 

 models? 



Some of the current, informal models guiding our present 

 management decisions may not be based on the best information 

 available, or may be based on only a portion of the information 

 available. We must base our management decisions on explicit 

 population models that incorporate all we know about a particular 

 species. Where specific information is lacking, well constructed 

 models that incorporate general knowledge about sea turtle 

 population attributes can be used to determine just how critical 

 the missing information is and how sensitive the predictions of 

 the model are to inaccurate information. 



Inexplicit, vaguely constructed models, put into use by 

 those who do not recognize or admit that they are using models, 

 are potentially counterproductive and will continue to limit our 

 ability to assess which management practices should receive 

 credit for any observed increase in sea turtle populations and 

 which should receive blame for any decline. 



It is possible to incorporate all we know about a species or 

 a given population into a model. This is not to say that we now 

 know everything we need to know. But by being explicit in 

 spelling out our models and the values we incorporate into them, 

 we can continue to build better models as more information 

 becomes available. 



202 



