relationships can be hypothesized to complete 

 the abstract model. For each time period, the 

 mortality rate in each age category depends 

 upon the density. Over time this density will 

 change, as will the inventory of deer in each 

 category. Hence, mortality rates will differ 

 over time, even if, the same functional re- 

 lationships are hypothesized. 



Now, add in the complicating factor of 

 changes in some or all of these mortality func- 

 tions consistent with an improved habitat and 

 a higher plane of nutrition. In the real world, 

 changes in the deer habitat — and its counter- 

 parts in other fish and wildlife species — are 

 occurring continuously. 



In making management decisions, some 

 knowledge is assumed of the structure of the 

 relevant biosystem. However, knowledge is, 

 at best, uncertain, and heroic assumptions 

 are aften made about the effect of a structural 

 change. Thus, decisions may be made which 

 move the biosystem toward the objectives 

 desired in an unpredictable manner. Manage- 

 ment is usually carried out within the 

 boundaries described by legally authorized 

 regulations, which are, hopefully, both con- 

 sistent with a set of objectives and flexible 

 enough to afford the on-the-spot manager some 

 discretionary action. When regulations are 

 for more than one distinct resource unit this 

 flexibility is desirable because each unit is 

 unique. 



For example, regulations for deer hunting 

 in a particular state usually embrace more 

 than one herd. No two herds will be identical 

 at any point in time, and the regulations must 

 be sufficiently flexible to allow for these dif- 

 ferences. Regulations are ideally formulated 

 with regard to the structure of the relevant 

 biosystems, but knowledge of these biosy stems 

 is not complete. The response of the biosystems 

 to particular management actions cannot be 

 predicted with certainty. Therefore, there is 

 a limit to the rigidity of the regulations. Be- 

 yond this limit, management will be ineffec- 

 tive in attempts to satisfy the set of objectives. 



Thus far, we have briefly described three com- 

 ponents of the management system of a public 

 resource. These are the complex biosystem, 

 the set of objectives, and the set of regulations 

 relating to the particular resource. One more 

 component is necessary to complete a work- 



able management system; that is, a means of 

 monitoring the system is required. For any 

 biosystem, the selection of the parameters to 

 be monitored is the result of experience and 

 expertise. However, to be useful to manage- 

 ment, the selected parameters must be indica- 

 tive of the performance of the biosystem so that 

 it can be determined whether, or to what 

 extent, objectives are being accomplished. 



Typically, only relatively few parameters 

 can be monitored accurately and rapidly enough 

 to be useful. Information on the state of the 

 system is of most value when it is current. 

 The role of time in monitoring systems cannot 

 be overemphasized. Information on the state 

 of a biosystem at any time is usually incomplete. 

 For example, the total number of deer in a 

 herd is a useful parameter in developing man- 

 agement strategies. In most herds it is im- 

 possible to take an accurate annual census, 

 and estimates of the total population must 

 be based on samples, which often may be col- 

 lected only at certain times of the year. 



Historically, researchers and managers have 

 been restricted to experimentation on the real 

 biosystem. However, with the advent of com- 

 puters and programming languages, it is now 

 feasible to perform simulated experiments on 

 biosystems that can be described by mathe- 

 matical equations. This paper is concerned 

 with the computer simulation of the deer 

 population in Mendocino County, California. 

 The model shows the population dynamics 

 and some of the economic and recreational 

 consequences associated with various hunting 

 strategies. 



COMPUTER SIMULATION METHODOLOGY 



Simulation involves building and operating 

 a model designed to represent those features 

 of the real system under study and to provide 

 information about the performance of the 

 system under assumed controlled conditions. 



Three classes of simulation models can be 

 distinguished: (1) physical models, such as 

 scale models of river systems and planetar- 

 iums, (2) mathematical models where a set of 

 equations describing the system under study 

 is written and these equations are solved, per- 

 haps analytically, and (3) computer simula- 

 tion where the system is described and the 



122 



