Morning counts and afternoon counts were 

 summed to determine daily counts. New Jersey 

 personnel counted boats passing through inlets 

 for entire days while New York personnel con- 

 centrated efforts obtaining interviews and did 

 not make inlet counts. 



Inlets in the survey area were grouped into 

 three size classes (small, a; medium, b; large, c) 

 according to the maximum expected numbers of 

 each type of vessel using the inlet. The mean and 

 vai'iance of the number of vessels sailing daily 

 through inlets in each class was determined 

 separately for weekend days and weekdays as 

 boat traffic was frequently much greater on 

 weekends. The mean and variance for weekend 

 days and weekdays were combined using the 

 following formulae (Cochran 1977): 



s, = 



10 we, + 23 wd, 

 10 + 23 



where s, = estimated mean number of vessels 



sailing daily in inlet class /, where 



i = a, b, c 



we^ = estimated mean number of vessels 



sailing daily on weekend days in 



inlet class i 



wd, = estimated mean number of vessels 

 sailing daily on weekdays in inlet 

 class i 

 10 = mean number of weekend days in 



season 

 23 = mean number of weekdays in season 



and the estimated variance of s, is: 



vis) = 



(.o^y-^.-do^T^'^. 



where v (s^) = estimated variance of mean num- 

 ber of vessels sailing daily in in- 

 let class i 



v (we-) = estimated variance of mean num- 

 ber of vessels sailing daily on 

 weekend days in inlet class i 



V iwd^} = estimated variance of mean num- 

 ber of vessels sailing daily on 

 weekdays in inlet class i 

 10 and 23 = constants as above. 



The mean, variance, and confidence interval of 

 the number of vessels of each type sailing daily in 

 all inlet classes was determined by combining the 



means and variances according to the following 

 formulae (Cochran 1977): 



s = 



NgSg + NhSh + NcSc 



where s = mean number of vessels sailing 

 daily through all inlets where the 

 vessel type occurs 

 N^,,Nf^,N^. = number of inlets in inlet classes 

 a, b, and c 

 s^, s^, s. = mean number of vessels sailing 

 daily through inlet classes a, b, 

 and c 



v(s) 



( 



N, ^' 



y 



+ 



+ 



Na+Nb+ Nc . 



V iSa) 



( 

 ( 



Na+Nb+Nc _ 



Na ^' 



A^a + A^6 + Nc 



y 

 y 



V (Sb) 



V iSr) 



where vis) = estimated variance of s 



V (sa),y isb),v isc) = estimated variance of Sq, 



Sb , and Sc 



CI = s±1.96\ v(s) 



where CI = 95*7^ confidence interval about s. 



Estimation of Catch Rates 



Interviews were made at dock sites, ma- 

 rinas, and launching ramps to determine vessel 

 catches. Vessel catches were determined rather 

 than individual angler catches since most pri- 

 vate boat and charter boat anglers share their 

 total catches. In addition, while some party boat 

 anglers may fish as individuals, it is common 

 practice for family or social gi'oups to share a 

 common fish container making it impossible to 

 determine the exact catch per angler. During 

 interviews the type of vessel, fishing location, 

 interview site location, number of Atlantic 

 mackerel caught, and fork lengths (FL) of 

 Atlantic mackerel were recorded. 



Inspection of the distribution of catch 

 per vessel indicated a lognormal distribution. 

 Therefore, the catch numbers were first con- 

 verted to natural logs and then the means and 



801 



