User menu
A Workshop developing assessing CPUE trends and techniques used by IOTC RFMO was held in AZTI Tecnalia in San Sebastian (Spain), from 21 to 22 October 2013. The following were the key issues and recommendations identified at the meeting:
1. Use of newer standardization techniques (GLMM, GEOSTATISTICAL APPROACHES and CORE AREA APPROACHES) using operational data to assess divergence in CPUE‟s across fleets.
1.1. While standard approaches worked well in most cases, the CPUE WG recommended that newer approaches should be tested. The results of the workshop conducted over the two day period, indicated that the GLMM models tend to capture the trends better. In addition incorporating more vessel specifics and using geostatistical techniques is another approach to pursue over time. Finally, the use of core-area approaches may be informative for by-catch species. Moreover, the majority feeling of the group was that during CPUE standardization the use of operational data when they are available is recommended as it will allow to capture the covariates that are important during the standardization process.
1.2. The strongest recommendation that came out of the workshop was that in areas where CPUE’s diverged the CPC’s were encouraged to meet inter-sessionally to resolve the differences. In addition, the major CPC’s were encouraged to develop a combined CPUE from multiple fleets so it may capture the true abundance better. Approaches to possibly pursue are the following: i) Assess filtering approaches on data and whether they have an effect, ii) examine spatial resolution on fleets operating and whether this is the primary reason for differences, and iii) examine fleet efficiencies by area, iv) use operational data for the standardization, and v) have a meeting amongst all operational level data across all fleets to assess an approach where we may look at catch rates across the broad areas.
1.3. Simulation studies could also be developed to assess which models work best (delta log-Normal, zero inflated versus standard GLM+constant, Tweedie).
1.4. Operational level data is useful if we want to quantify fishing fleet efficiency using fleet dynamic covariates. More applications could be developed using the methods developed by Hoyle and Okamoto (2010), or Hoyle (2009), and preliminarily presented by Dr. Okamoto at the CPUE workshop.
1.5. Assess how core area Standardization works along with out of core or boundary area effects.
1.6. Environmental data would be useful to consider in relation to standardization approaches. However, the way it is usually performed in GLMs, where an environmental covariate is associated to each observation (in regular 1°, 5° or even 10° grids) , may not be the most pertinent as it does not allow to identify the ecological processes which may affect CPUE. Alternatively, GLMs could be performed in sub-areas where the variability pattern of the environmental signature is well identified (using spatial EOFs to delineate those sub-areas). In such sub-areas, GLMs could be designed with and without environmental covariates to understand the potential effect of the environment. Environmental covariates should be in limited numbers (the lesser the better) and selected in order to test hypothesis on the ecological processes at stake.
Develop robust CPUE series for other species and Working Parties.
2.1. The Working Group recommended to also focus the efforts in other species such as Temperate Tuna. In addition the WG recommended, developing better CPUE data for Neritic Tuna, and also improving the data and standardization on marlins and sharks.
2.2. Develop a reference manual for use in performing a CPUE standardization for any fleet in any working party (e.g. neritic tuna WP or temperate tuna WP). Criteria for inclusion of the data in a stock assessment should also be developed (possibly using ICCAT techniques as a baseline).
With regard to Purse Seine data the following were recommended:
3.1. Approaches being pursued by EU scientist have some promise, and more work should be put in the development of an index of abundance for the PS fleet on Skipjack Tuna, Yellowfin Tuna and Bigeye Tuna.
3.2. The availability of Vessel Monitoring System (VMS) data is a major requirement for the purse seine fishing fleet as it enables to spatialize the nominal effort (i.e. fishing or searching time), which is key to appreciate the temporal changes in the spatial extent of the prospected areas. VMS data may also be used to analyze PS trajectories with the aim to discriminate sets on FADs equipped with buoys from free school sets, log sets and foreign FADs sets (see after)
3.3. Purse seiners currently fish during the same trip on a combination of free-swimming and drifting FAD-associated schools. In addition, fishing on FADs results from both the detection of vessel-owned FADs through GPS geolocation systems as well as from the finding of 'foreign' FADs through bird detection for instance. Future analyses should focus on the separation of fishing time between searching and running towards FADs. Classification methods based on indicators describing spatial behaviour of vessels could enable to define typical fishing strategies and categorize trip components into such categories.
3.4. Data available on FAD activities collection have improved recently. Future analyses should focus on the definition of a fishing effort for purse seiners using FADs by (i) looking at the influence of the number of FADs owned by a vessel on individual CPUEs, (ii) by investigating the CPUEs in areas characterized by strong contrasts in FAD density. The influence of supply vessels on catch rates (e.g. the number of sets per day) and on the overall fishing capacity of the PS fleet should also be investigated at the vessel level through the information available from supply logbooks.
3.5. Analyses of temporal changes in individual and overall fleet catchability from CPUEs should be conducted to estimate fishing power creep and investigate how such changes are related with some major technological changes known for the PS fleet (e.g. bird radar). Including vessel effects into GLMs can reveal useful insights on vessel efficiency for such analyses. Attention should be paid also for change over time of fishery indicators which are part of the CPUE (e.g., number of set by day, % of successful set, catch size of the set).
4. The CPUE Workshop participants recommended that a thorough analysis of the history of the fishery would be useful for references for each species. In addition, the Group agreed that a central body (the Secretariat) should undertake additional activities in key areas (Neritic tunas where they can develop/collate the existing data on catch and effort and analyse this for some key species (eg. Longtail and Kawakawa)).
5. The CPUE Standardization Working Group agreed that a reference document that IOTC could use in what criteria should be used in utilizing a dataset for CPUE Standardization for all WP would incorporate the specifics of the temporal and spatial coverage of the data, and useful covariates that could quantify the fishing activity and the environment in which the fish lived.
