752 



Fishery Bulletin 94(4), 1996 



as a proxy for water temperature, was negatively 

 correlated with Tortugas pink shrimp landings dur- 

 ing the following July-September. The present analy- 

 ses for the Tortugas fishery rarely included air tem- 

 perature as a predictor variable, perhaps because 

 predictors were limited to the warmest months of 

 May-October. Costello and Allen (1970) listed no 

 known negative impacts of high summer tempera- 

 tures on pink shrimp. Air and water temperature 

 have been found to be potential predictors in a vari- 

 ety of other fisheries (Fogarty, 1989). 



One final type of data consistently entered my 

 models, i.e. September wind speed at Key West. Wind 

 speed and direction were postulated by Vance et al. 

 (1985) to affect recruitment and settlement of post- 

 larval Penaeus merguiensis and thus subsequent 

 harvest. Off Florida, high winds in September, asso- 



ciated with passage of tropical storms, could break 

 up advective processes that deliver planktonic pink 

 shrimp to the nursery areas during the summer. 

 However, concurrent studies of oceanographic fea- 

 tures and larval pink shrimp abundance remain to 

 be conducted during this period (Criales and Lee, 

 1995). It is also possible that high winds in Septem- 

 ber reduce fishing effort and thus subsequent har- 

 vest, but high winds occur during the low point of 

 the fishing year. 



My forecasting method involves generating new 

 prediction equations each year, rather than employ- 

 ing a single fixed equation, because 1 ) the models do 

 not describe cause and effect relationships, 2) causal 

 relationships between most variables and subsequent 

 recruitment to the fishery are not well known (lead- 

 ing to the "hazards of correlative studies" noted by 



