considerable redundancy among the 

 predictors; and (4) the fraction of 

 the total variance accounted for by 

 the forced St. Lucie variable model 

 and the equally presented St. Lucie 

 variable model should be sizable, 

 0.30 or more. Otherwise, one would 

 argue that the appropriate predictor 

 variables simply were not made avail- 

 able for selection. If they were, it 

 is likely that they would dominate or 

 even replace the effects of the St. 

 Lucie variables. After a tentative 

 model had been selected by the fore- 

 going proceedings, ridge regression 

 analysis was performed to examine the 

 stability of the regression coeffi- 

 cients . 



the consequence of heavy rainfall. 

 Our results, therefore, can be inter- 

 preted as being representative of 

 fishing conditions on the estuary as 

 affected by the present schedule for 

 normal allowable discharges. 



Overall, during the one-year 

 period from June 1978 through June 

 1979, 440,399 acre-feet were released 

 through the St. Lucie Canal, 91,260 

 acre-feet through C-23, and 88,187 

 acre-feet through C-23A. In addi- 

 tion, a total of 49.29 inches of rain 

 fell on the estuary. Three distinct 

 rainy periods correspond to high 

 weekly cumulative discharge rates 

 from each of the three discharge 

 structures . 



COMPARISON WITH EARLIER STUDIES 



Two resource reports were writ- 

 ten about the St. Lucie River by the 

 Fish and Wildlife Service, Vero 

 Beach, Florida office in the late 

 1950's. (U.S. Department of the 

 Interior, Fish and Wildlife Service 

 1959, 1960). Comparisons of the raw 

 data from those early studies with 

 current estimates of the sportfishing 

 catch were made. Using the analysis 

 techniques from the old report, the 

 current raw data were converted to a 

 similar comparable form. 



RESULTS 



SINGLE-SOURCE, CROSS 

 CORRELATION ANALYSES 



Single-source, cross correlation 

 analyses (Table 1) indicate that 

 neither rain nor the St. Lucie Canal 

 discharges had significant correla- 

 tions with fishing pressure. The 

 agricultural canals, on the other 

 hand, affected pressure with some 

 indications that discharges in the 

 North Fork may increase South Fork 

 fishing pressure while decreasing 

 pressure in the North Fork and inlet 

 sections . 



In response to outcries that 

 declared the St. Lucie River a "bio- 

 logical desert," the Corps of Engi- 

 neers agreed to limit eastward dis- 

 charges from Lake Okeechobee to 2,500 

 cfs whenever possible. The study 

 team was fortunate to have had three 

 releases to that level within the 

 sampling period. One was the sched- 

 uled release in the beginning of the 

 summer of 1978 and the other two were 



MULTIPLE-SOURCE, STEPWISE 

 REGRESSION ANALYSES 



Fifteen species of fish, chosen 

 from the 56 species sampled, either 

 because of their importance as sport 

 fishes or their significant harvests, 

 were tested for correlations with St. 

 Lucie Canal discharges. These fish 



273 



