-Dr. Makoto Miyake,
Visiting Researcher at the National Research Institute of Far Seas Fisheries-

In stock assessments, catch per unit of effort (CPUE) is considered to reflect the abundance of a fish stock.For example, if you caught 10 tons of tuna per fishing day, the abundance of that tuna stock was about twice as high as the stock from which only 5 tons were caught per fishing day. Of course, if there is some difference in fishing procedure (e.g. depth of gear, target species, area, time) between these two cases, they can not be directly compared and hence need to be adjusted to account for such differences. This adjustment is called, ‘standardization’ and CPUE modified for various factors is called standardized CPUE (index). In the case of longline, number of fish caught per 1000 hooks is widely used and considered to reflect the abundance of stocks well.
For purse seine, catch (in weight) per purse seine fishing day (including searching day without catch) is generally accepted as an abundance index. However, purse seine fishery is relatively selective. For example, if there are schools of big yellowfin and those of small skipjack, the fishing vessels will only set the net on yellowfin schools, skipping the skipjack schools. In other words, longline is a passive and more random fishing method, whilst purse seine is an active and more selective fishing method. For this reason, the standardized CPUE for purse seine is not as indicative of stocks as longline CPUE.
In the case of purse seine fishery on schools associated with FAD (fish aggregating device), the situation is worse. For example, even whilst stock size keeps declining, FADs still attract fish and the catch level may be maintained. Therefore the amount of fish under FAD may not reflect the abundance of total stocks.
Consequently, in stock assessment work, longline abundance indices are often given high weight compared to other indices, either deliberately by researchers or automatically within assessment models. Therefore, the results of analysis appear to be influenced more directly by the tendencies shown in the longline indices. This close dependency on longline indices had been initially accepted when the longline catch contributed a major part of the tuna catches. However, in more recent years, an overwhelmingly large part of the total tuna catches is made by purse seine. Besides, purse seine catches much smaller (younger) fish than longline. This leads us naturally to doubt whether we can continue assessment methods so heavily dependent on longline indices.
The abundance of large fish is the residuals of population after the accumulative mortality of fish through younger ages. In that sense, it is more indicative of the stock status than the small fish indices. However, when the abundance of large fish of a species gets reduced, the longline has changed its behaviour (e.g. targets on another species of tuna) and the representation of index of the stock status would be also reduced. This has been pointed out for many years in the past. Unfortunately, no good index for purse seine has been developed. It should be the duty of scientists, who work with purse seine data, to give some serious effort to solve this problem.