| 英文摘要 |
We compared nominal and standardized bigeye and yellowfin CPUE using Taiwanese longline catch and effort data, both operational and aggregated by 5x5 degree square and month, in the tropical Indian Ocean between 10N-15S (core area) from 1979 to 2012. Nominal and standardized bigeye CPUE for both data types showed quite similar trends. The standardized bigeye CPUE trend from operational data kept at nearly the same level from 1979 to 2012. In the case of yellowfin, CPUE trends of both data types showed similar trends but with small differences in amplitude; overall they were relatively stable until 2005, then decreased to less than half in 2009 from the 2003-2005 level. Historical change in the fishing efficiency of the Taiwanese longline fishery was estimated for bigeye and yellowfin by including Vessel ID in the standardization using operational data. The estimated fishing efficiency for bigeye across all core areas increased from 0.9 to 1.1 during 1979 ~ 2012. When areas were viewed separately, in the west core area efficiency increased continuously from 0.8 in 1979 to 1.1 in 1989 and then remained stable. In the east core area efficiency was estimated to have remained stable throughout the 33 years. In the southern area there was a continuous increasing trend, from 0.7 to 1.4 during the 33 years analyzed. On the other hand, fishing efficiency for yellowfin tuna estimated in core areas showed no clear trends, varying between 0.7 and 1.0 from 1979 to 1993 and then remaining at a similar level of about 1.1 (in the case of west core) or remaining stable for the 33 years analyzed (east core). In the south area there were relatively high values of 1.2 in 1979 and 1989 and low value of 0.6 for 1992 and 1993, stability of around 1.0 until 1996, and then a steady increase to 1.3 in 2012. Two types of Taiwanese operational longline data, the first with resolution of 5 degree square from 1970 to 2012 (noted as the ‘long series’) and the second with resolution of 1 degree square from 1995 to 2012 (noted as the short series’), were standardized and their trends compared. As well as differing in resolution, data on the number of hooks between floats (NHBF) was available in the short series. Cluster analysis was used to classify longline sets in relation to species composition of the catches. Five effort clusters were identified. Four clusters comprised ~90% of the total sets, and the catch compositions suggested targeting for either tropical tuna (bigeye and yellowfin) or albacore. The clusters were used as a target proxy in GLMs to adjust the effectiveness of fishing effort units. NHBF was used an alternative target proxy in GLMs for operational data from 1995 to 2012. In all core areas and the south area, bigeye CPUE trends were similar between different types of data. For yellowfin tuna, quite similar trends were observed between standardized CPUEs derived from the two types of data except in the east core area. In this case the long CPUE series showed a decreasing trend from 1992 to 2012, but the short series did not. |