hatchery trap efficiency, which can be different for natural and hatchery 

 fish. Of the natural fish that are available, the hatchery removes the 

 smaller of hatrem (the hatchery target) or those available in excess of 

 minrun (the minimum acceptable wild escapement). The remainder of 

 hatchery needs are met using returning hatchery adults, if available. 

 Adults not needed by the hatchery and those escaping the trap are subject 

 to a terminal harvest, with independent rates for hatchery and natural 

 fish. Hatchery adults which escape harvest stray into natural spawning 

 areas in proportion to the parameter, hstray. (This parameter may also in- 

 corporate the differential spawning success between natural and hatchery 

 fish in the wild.) The total number of adults spawning naturally includes 

 both natural and hatchery adults, and an additional number of adults, 

 adtsup, which originate from outside the system (simulating adult supple- 

 mentation). All adults are subject to prespawning mortality, either within 

 the hatchery or natural areas. Jacks (one-ocean fish) can be excluded from 

 the spawning population by using the parameters, jackspn and hjackspn. 



Operational Notes The SLCM and ancillary programs are written in the SAS programming 



language. While the model has been developed on a SUN workstation, it 

 can be transferred easily to any system equipped to run the SAS language. 

 Copies of files needed to run the model and produce summary analyses are 

 available from the authors. The modular code can be modified to suit the 

 particular needs of various user groups. 



Once the SLCM files are installed, users run the model by setting up 

 three parameter files, a file containing initial numbers, and a control file. 

 Examples are shown in appendix C. These files can be modified by using 

 a convenient text editor. Only the control file needs to be submitted as an 

 SAS program. If the control file is set up properly, it will make all neces- 

 sary calls to other files and programs. As output, the model creates an 

 SAS data set containing all variables requested in the control file. Model 

 results can be analyzed using ancillary SAS programs supplied with the 

 model. Users may write additional ancillary programs to meet their spe- 

 cific needs. 



The control file includes macros that allow users to perform multiple 

 runs of the model using stepped parameters (each run would consist of 

 multiple games), and to change parameters for specific years within games. 

 This is useful in performing sensitivity analyses and in simulating environ- 

 mental changes over time. The stochastic properties of the model can be 

 changed through this mechanism by introducing additional uncertainty or 

 changing the distributional properties of the parameters. 



The control file also contains an SAS macro variable, &HATCHERY. 

 Setting & HATCHERY to "off' shuts down all hatchery production. The 

 simulated population then will contain only naturally produced fish. 



ILLUSTRATIVE MODEL RESULTS 



We demonstrate the SLCM using parameter sets listed in appendix C. 

 These parameters were chosen to reflect a stock that is not limited by habi- 

 tat, but by a low smolt-to-adult survival rate. Thus, the density-dependent 

 parameter {beta) is set at a very low level. In this example, subbasin es- 

 capement (natural and hatchery) is the principal output examined. Our 

 base case is a contrived fall chinook example that is not meant to mimic 



14 



