Page 10 - DiNatale2010

1987, 1988), small boats, mostly used for the leisure fishery, are able to catch important numbers of juveniles per
day and the total number of these boats and vessels in the Mediterranean Sea is quite huge.

Monitoring these catches is very difficult, because these fish are used for personal consumption or for the local
markets, but it is essential to understand the recruitment and follow the yearly variations. If these data are not
collected any more of if they are not reported to the SCRS for the risk of compliance problems, then the SCRS
has no tools to properly assess recruitment, in the absence of other indicators.

Figure 11 shows a true example of two data sets, collected in exactly the same area in 2008, one by scientific
observers monitoring the effective landing by all vessels and the other as landings of vessels having a bluefin
tuna quota reporting their catches in landing declarations and then in Task II statistics. This has been occurring
for many years and in this way the SCRS lost the opportunity to follow the high recruitment noticed in an
anecdotic way by many scientists since 2003 in almost all the Mediterranean areas, thanks to prolonged high
temperatures at seas.

This lack of basic data on recruitment clearly negatively affected the result of the assessment, independently
from the attempt to apply correction factors under the hypothesis of various recruitment scenarios.

Another problem affecting the size distribution in the model is caused by the rare presence of large fish. Usually,
giant tunas are not abundant because of natural factors, but their presence among the catches in the
Mediterranean fisheries was always reported in the past. In recent times, the number of large fish in the national
reports decreased and this was a further cause of concern in the SCRS bluefin tuna assessment. It was decided to
pay particular attention to this issue and it was possible to identify several factors affecting the data. It is very
clear that the different composition of the bluefin tuna schools reported in the previous chapter was one of the
causes, together with the different fishing strategy of the purse seine fleet, but was not the only one.

Giant bluefin tunas in cages were present anyway and even very large specimens were identified (a few over 600
kg, several over 400 kg and many over 200 kg). In the case of large fish, the effects of caging on the fork length,
independently from the number of months, is usually minimal, particularly if compared to the same effect on
juvenile tunas. These catches moved to cages were not properly monitored for many years and these specimens
often escaped from the statistics.

For gears other than the purse seine a different problem was detected: due to the mandatory logbooks and
landing declarations, the only possibility for some fishermen to escape from a quota overshot when landing in
the authorised harbours was to lie about the weight of the individual fish, because the number is always easy to
check. Figure 12 clearly shows the effects of these actions on the Task II data. This fact causes an important bias
in the statistics, because of the decreasing number of large fish, along with the disappearance of juveniles in the
reports and together with the higher presence of small and medium size fish. Logbook data from vessels having
quota are used by many CPs for their official catch declarations to ICCAT.

Length data used in the recent bluefin tuna assessment are affected by several other factors. One of them is the
poor correspondence between the real purse seine catches and the corresponding data input in the VPA. As a
matter of fact, after the adoption of the caging system, it became almost impossible to get a proper number of
length samples at the moment of catching the bluefin tuna, before they are moved to the fattening cages. Only a
few specimens die during the catch operations or the transport and these are not considered as representative of
the structure of the catch. To partly overcome the problem, the ICCAT adopted Rec.06-07, making mandatory a
sampling procedure at the harvesting. This sampling, even considering various factors affecting the growth
during the different fattening periods, was considered able to provide more reliable information about the size
structure of the catches, particularly for the length frequencies of medium and large specimens, where the growth
in length is minimal when they are kept in cages for a few months.

During the last (2008) assessment, it was decided not to use these data that were collected spending a huge effort
and an important amount of money by some CPs. At the same time, it was decided to use the very few data
concerning the fish that died during the purse seine catch operations (a little more than 100 specimens) and raise
them to the total catch by using the Montecarlo suite of the VPA. It is a common point of view of many
scientists that this was able to create really a “virtual population” very far from the real structure of the catches.
As a matter of fact, fish which die during the purse seine operations are usually not the larger ones and the
casuality makes these catches not representatives of the length frequencies in the haul. It is very possible that this
questionable procedure was the causes of the lost of large specimens in the assessments, while very large
specimens were included in the length frequencies collected at harvesting and were not used. The effect of this

1013