ANALYSIS OF MIGRATIONS AND MORTALITY OF 



BLUEFIN TUNA, THUNNUS THYNNUS, 



TAGGED IN THE NORTHWESTERN ATLANTIC OCEAN^'^ 



F. J. Mather iii,^ B. J. Rothschild,'* G. J. Paulik,^ and W. H. Lenarz* 



ABSTRACT 



An analysis is presented on the release and return data from bluefin tuna, Thiinnus thynnus, tagged 

 in the northwest Atlantic Ocean from 1954 to 1970. There was an apparent northward movement of 

 fish from the New Jersey area as the fishing seasons progressed. Tag returns from bluefin released in 

 the Long Island and southern New England areas tended to be to the north at first and then to the 

 south. Mean distances between release and return tended to be greater for fish released in the New 

 Jersey area than for the other two areas. Estimates of mortality rates for tagged bluefin were made 

 using the Chapman-Robson method and then adjusted for Type-I and Type-II tag shedding and Type-I 

 tagging mortality. The average estimate of instantaneous fishing mortality is 0.57 and other losses 

 (natural, tagging, and emigration) is 0.68 on an annual basis. The estimate of other losses is consid- 

 erably higher than the natural mortality that would be expected for bluefin. Evidence is presented 

 suggesting that the rate of emigration may be quite high. The average single season exploitation rate 

 of tagged bluefin was estimated to be 0.33. It was noted that since bluefin may be both immigrating to 

 and emigrating from the fishery the estimate of exploitation may not be representative of the entire 

 population. Even though validity of available effort data is questionable, regression estimates of 

 mortality and survival rates were made using catch per effort data. These estimates of survival are 

 lower than those obtained using the Chapman-Robson method. 



The data which form the basis for this report 

 were assembled by the first author. This study is 

 based upon releases of tagged bluefin tuna, 

 Thunnus thynnus, that were made by a variety of 

 organizations and individuals under the coordi- 

 nation of the Woods Hole Oceanographic Institu- 

 tion Cooperative Game Fish Tagging Program at 

 various locations along the middle Atlantic bight 

 of North America from July 1954 to August 1970 

 and returns of these tags to the end of 1970. Addi- 

 tional returns are expected in the future from 

 more recent releases. 



MIGRATIONS 



Because of the variety of methods, locations, 

 and dates of release, we needed to assemble the 

 data by relatively homogeneous release groups. 

 The criterion for constructing a release group for 



'This paper is dedicated to the memory of Gerald J. Paulik. 

 He was a good friend and colleague and made important con- 

 tributions to the theory of tagging. 



^Contribution No. 3180, Woods Hole Oceanographic Institu- 

 tion, Woods Hole, MA 02543. 



'Woods Hole Oceanographic Institution, Woods Hole, MA 

 02543. 



^Southwest Fisheries Center, National Marine Fisheries 

 Service, NOAA, P.O. Box 271, La Jolla, CA 92037. 



^College of Fisheries, University of Wtishington, Seattle, WA 

 98195. Deceased. 



analysis of migrations was to develop homoge- 

 neous time-location strata of releases from which 

 a minimum of 20 tags were recovered. This proce- 

 dure allowed us to work with homogeneous 

 groups, but eliminated roughly 10'7r of the recov- 

 ered tags from our analysis. Table 1 summarizes 

 these release groups and Figure 1 shows their 

 localities. We can see from Table 1 that during the 

 study period the tagging operations tended to shift 

 from the New Jersey coast, to the New York coast, 

 to the southern New England coast, and that re- 

 leases in July tended to be south of those made in 

 August or September. 



Tagging data have been used to show some of 

 the longer migrations of the bluefin tuna (Food 

 and Agriculture Organization, 1972). We ex- 

 amined the shorter term recoveries from an ana- 

 lytic point of view. In order to do this, we made use 

 of a method developed by Rothschild (Bayliffand 

 Rothschild, in press). Using this method each release 

 group was stratified into intervals of time at lib- 

 erty. Release vectors (latitude and longitude) for 

 each release group were used to compute an aver- 

 age or common release vector. Each recapture vec- 

 tor for the group was then standardized to the 

 common release vector. The standardized vectors 

 were then used to find 1) the average recapture 

 vector and 2) the determinant of the recapture 



Manuscript accepted January 1974. 

 FI.SHERY BULLETIN: VOL. 72, NO. 4, 1974. 



900 



