Taggart et al.: Estimating abundance of Cancer magister 



495 



6- 



04- 



0.2 



Pots Transects 



(n=250) (n=75) 



-O- -•- 05 change/year 



-O  03 change/year 



•V- -T- 02 change/year 



2 4 6 8 10 



Study duration (years) 



12 



Figure 6 



Relationship between power and study duration in comparing catch from 

 crab pots and density on dive transects for male Dungeness crabs (C. 

 magister) at three levels of population change. To hold effort constant, 

 we set the sample size (n ) to the number of pots and dives that could 

 be accomplished in five days. 



larger with pots than with transects and the variance 

 with pots is lower. 



Despite their low power, the independent measures 

 of abundance provided by dives helped us identify bias 

 in our Dungeness crab survey method. Our analysis of 

 these two techniques demonstrates that it is possible 



to avoid most biases with pots if sampling is conducted 

 at optimal times of year. Similar comparisons could be 

 conducted in other areas to identify sampling biases so 

 that they could be minimized and important param- 

 eters, such as abundance, size, and sex ratio, could be 

 monitored effectively. 



