GREENBLATT ASSOCIATIONS OF TUNA WITH FLOTSAM 



Table 2. — Spatial distribution of fork length measurement 

 ( centimeters) of yellowfin and skipjack tuna m the Commission's 

 Yellowfin Regulator>- Area. 1973-75. C YRA subareas are shown 

 in Figure 1. (Source of data, SWFC.) 



Monthly rainfall in Central America was calcu- 

 lated by averaging the stations reporting to the 

 Environmental Data Service (U.S. Department of 

 Commerce 1963-1975). 



In order to achieve the objectives of this paper, 

 the data obtained from lATTC and SWFC were 

 examined and analyzed in several ways. The main 

 sources of flotsam were inferred by examining the 

 average distribution of flotsam-associated sets 

 and consideration of the average surface circula- 

 tion. 



Two methods may be used to determine if differ- 

 ent set types are related: correlation of set types 

 occurring in an area and comparison of fork length 

 distributions (length-frequency gi-aphs) stratified 

 by species and set type. Spearman's rank correla- 

 tion coefficient (Siegel 1956) was calculated to ex- 

 pose possible correlations of numbers of sets. Fork 

 length distributions were weighted by the catch in 

 each set. A high positive correlation between set 

 types occuring in an area would indicate a rela- 

 tionship between set types. Similar looking 

 length-frequency distributions would serve as 

 further evidence that set types are related. 



An increase in the percentage of flotsam- 

 associated sets would be evidence that flotsam- 

 associated sets have become more important to the 

 fishery. The CYRA was subdivided into three 

 nearshore areas and one offshore area (Figure 1). 

 Stratifying the number of flotsam-associated sets 

 by area allows the determination of area effects. 

 Hence, the importance of flotsam to the fishing 

 industry may be determined by the percentage of 

 flotsam-associated sets occurring each year and 

 stratifying the number of flotsam-associated sets 

 by area. 



Average rainfall was tabulated to determine if 

 any connection existed between river runoff and 

 the number of flotsam-associated sets. 



Catch rate is an indicator of the importance of 

 flotsam-associated sets to the tuna fishery. Calcu- 

 lation of the average yearly catch per set (includ- 

 ing zero catch sets) for different set types should 



demonstrate any trends as well as the relative 

 value of making one set type over another. 



Calkins (1965) examined tuna length distribu- 

 tions from single sets in the eastern tropical 

 Pacific, finding that single unassociated schoolfish 

 sets caught tuna of a relatively uniform size (i.e., 

 small variance in length). If the fish caught in 

 flotsam-associated sets represent aggregations of 

 solitary tuna, portions of schools, or several 

 schools, then one would expect the variance of 

 tuna length to be greater than for unassociated 

 schools. In order to examine if floating objects act 

 as aggregators, length-frequency data were strat- 

 ified by species and set type. Mean length and 

 standard deviation were calculated on a single set 

 basis for each category and compared using 

 Kruskal-Wallis one-way analysis of variance by 

 ranks (Siegel 1956). 



If flotsam does aggregate tuna, one would expect 

 more of the larger flotsam-associated sets than 

 unassociated sets. Differences in catch distribu- 

 tion were observed by plotting histograms of tons 

 of tuna caught per set stratified by year and set 

 type and by calculating catch per successful set for 

 each year and set type. The single set catch data, 

 collected by SWFC technicians, existed for the 

 1974-76 period. 



RESULTS 



The availability of logs or other flotsam in an 

 area are determined by the source of the flotsam 

 and the currents in the area. Large rivers flow into 

 the Pacific from southern Mexico (lat. 20°N) and 

 continue down the coast of South America (to lat. 

 20°S). These rivers are capable of releasing many 

 logs into the Pacific during the rainy season. 

 SWFC and lATTC observers reported large densi- 

 ties of logs near the Gulf of Tehuantepec (lat. 

 16°N, long. 100°W), the Gulf of Nicoya (lat. lO'^N, 

 long. 85°W), and the Gulf of Fonseca (lat. 13°N, 

 long. 87°W). 



The average yearly number of flotsam- 

 associated sets in 1972-75 were plotted by 5° 

 squares (Figure 1). In general, most flotsam- 

 associated sets occurred in Areas 1 and 2. Most of 

 the offshore flotsam-associated sets (i.e., Area 4) 

 occurred quite close to Areas 1 and 2. Area 3 did 

 not have large numbers of flotsam-associated sets. 

 If the main source of logs and other flotsam is the 

 rivers of Central America, then it is important to 

 examine the major current patterns in the eastern 

 tropical Pacific to determine if the currents can 



149 



