02 October 2014
BLOOM’s position on deep-sea quotas for 2015/2016
02 October 2014
In November 2012, the Council of Fisheries Ministers had decided of a drastic increase of quotas for French fleets involved in deep-sea fishing. This was inconsistent with the precautionary approach that has been repeatedly called for by ICES with regards to the management of deep-sea fisheries and led to increased captures of bycatch species, some of which are threatened with extinction. Below we state what we hope will be taken into consideration for setting deep-sea TACs and quotas for 2015 and 2016:
1. French scientists speak with one voice to remind decision makers that ONLY ONE deep-sea stock meets criteria of sound management
First, as French scientists from various research institutes, gathered in the “Association française d’halieutique” (AFH) have stated in their joint analysis (September 2014) about the upcoming TAC & Quota decisions: “the situation of marine resources in the European Atlantic remains fragile”. Firstly because the number of overexploited or degraded stocks is still important and because “numerous stocks are still poorly known”. They emphasize deep-sea fish stocks as being in the latter case highlighting that there is “sufficient data to provide a scientific advice” for “only 3 stocks out of 26.” They insist on the fact that only 1 stock out the 3 known deep-sea stocks meets criteria of sound management currently in place.
The Association française d’halieutique calls on decision makers to respect scientific advice. BLOOM reinforces this recommendation.
2. There is only one “properly managed” deep-sea stock out of 26 stocks that fall under a TAC regime and 22 more that are unregulated.
Only 26 stocks are currently under a TAC regime while 22 other deep-sea species are unregulated and a total of 100 species are reported to be caught by IFREMER observer programs in French bottom trawl fisheries.
The NGO OCEANA reviewed the 2012 ICES advice regarding the 54 deep-sea species that the European Commission had included in its proposal (COM(2012) 371 final) and found that:
- 21 species were considered to be depleted or at risk of depletion
- 5 species had one or more stocks that were in stable or slightly negative condition
- 4 species had one or more stocks that were of unknown status, due to a lack of data
- 3 species had one or more stocks that were considered to be in good condition
- The status of a further 26 species was completely unknown, because they had never been assessed.
We call on decision makers to respect the ecosystem approach and to decrease total allowable catches of deep-sea species in order to reduce fishing pressure on sensitive and sometimes even endangered species.
3. There is insufficient scientific knowledge and data about most deep-sea fish stocks and bycatch species
The latest STECF advice (REVIEW OF SCIENTIFIC ADVICE FOR 2015 – PART 2 (STECF-14-11) reminds that (p. 357):
- « Productivity is very low in the deep-water. The diversity of deep-water life history strategies is considerable, but some species of fish targeted by fisheries are particularly vulnerable to disturbance because they grow slowly, mature late in life, and form aggregations easily accessible to fisheries. Recovery rates are much slower than in shallower waters. The knowledge of central biological characteristics such as stock identity, migration, recruitment, growth, feeding, maturation, and fecundity of most deep-water species still lags considerably behind that of commercially exploited shelf-based species. »
- « Fisheries data including length and age compositions, discards, and CPUE, are slowly increasing for deep-water stocks but time-series data are often short and are not available in sufficient spatial resolution for some stocks e.g. orange roughy and alfonsinos. »
- « VMS data are not readily available for most fleets. »
- « In many cases, information on stock structure of deep-water species is lacking. (…) Assessment data are improving for several stocks/species (…) But for the majority of deep water species there is still no conclusive information on stock structure. »
4. For a few targeted fish, 100 species are caught and discarded in bottom-trawl fisheries
Observer data from the IFREMER shows that 100 species are caught in French deep-sea bottom trawl fisheries, which only target three main commercial species. Deep-sea bottom trawl fisheries should be managed respecting a global ecosystem approach.
The STECF advice (REVIEW OF SCIENTIFIC ADVICE FOR 2015 – PART 2 (STECF-14-11) repeatedly insists, as here for the advice for roundnose grenadiers (p. 387), that the ecosystem approach should be applied:
“Given that roundnose grenadier is taken in a deepwater mixed fishery, there is a need to harmonise management measures to account for the management requirements for other species taken.”
5. The MSY is an ill-fitted tool for the management of long-lived species and STECF calls for lower quotas than those advised by ICES
The STECF advice also notes about the MSY approach that (p. 11):
« While recognising that the standard approach to advise catches corresponding to fishing at FMSY should, lead to stock biomasses that on average are capable of delivering MSY, STECF considers that using such a standardised approach may not be wholly appropriate, especially for long-lived, low productivity species. The following comments which relate to the advice for roundnose grenadier in Vb, VI, and VII (Section 9.9.2 of this report) provide an example of where this may be the case but similar arguments may also be appropriate for other stocks.
STECF notes that the ICES advice for roundnose grenadier (in Vb, VI, VII and XIIb) is based on an assessment and catch forecast for the component of the stock in Division Vb, and Subareas VI and VII only. Furthermore, while the mean estimate of biomass for the stock in Vb, VI, VII has increased slightly over the last two years, it remains at only about 30% of the estimated mean level for the beginning of the time-series (1988) and is close to the MSY Btrigger reference point.
Based on ICES’ projections, fishing at FMSY implies landings of 3952 t in 2015 and 4019 t in 2016 and will give rise to a small (2%) increase in stock biomass. Such a slow response in biomass to fishing at FMSY implies that the recovery of the stock to BMSY (68,935 t) will take many years (on the order of 20 years), assuming no other changes in the environment or fishery). STECF notes that even with no fishing, the biomass is only predicted to increase by 10% by 2017.
Furthermore, the low landings from Vb, VI and VII observed over the last three years (2011 = 1577 t, 2012 = 2501 t, 2013 = 1498 t; average = 1862 t) have not resulted in any significant increase in stock biomass. STECF notes that fishing at FMSY potentially implies more than a 2-fold increase in landings in 2015 compared to the average landings over the most recent 3 years. Given the uncertainty in the assessment results, such an increase in landings (and catch), increases the risk that recovery of the stock biomass to levels that will deliver MSY will be impeded. STECF considers that restricting landings in 2015 and 2016 to less than the recent average level of 1862 t would be a more appropriate risk-averse approach and is likely to lead to a more rapid recovery of the stock biomass. »
6. The data-poor approach applied to deep-sea species is fundamentally flawed
In order to deal with these data-poor species, special methods using catch data have been invented. One method, referred to as “Catch-MSY”, uses catch or landings per unit effort information obtained from fishing captain logbooks or tallybooks. A second method, the “gislasim() function”, uses length data and the von Bertalanffy growth model.
Problems and critiques
Both methods came under criticism during WKLIFE II. For the Catch-MSY method, the problem came from treating catch data as equivalent to fisheries independent survey data. Unfortunately, that can be a serious mistake, since it is possible for catch data to remain stable or even increase while the stock abundance is actually decreasing.
Flawed categories attributed to French-caught deep-sea species
There are 6 categories in the data-poor approach, ranking 1 to 6. Generally there is no doubt about the best-known stocks (top categories) and the worst-known ones (last categories). But middle categories 3 and 4 are highly problematic. Both of these are considered the “best” of the data-poor stocks. If a species is classified under category 4 to 6, any estimate of catch (using whatever model) will be subject to both an “uncertainty cap” and a “precautionary buffer” which both trigger reductions of 20%. BUT if a species is classified as category 3, then the application of the 20% reduction for uncertainty and 20% for precautionary buffer is optional.
Blue ling, black scabbardfish and roundnose grenadiers were all classified in category 3 which means they do not have any reduction mechanisms applied to them due to uncertainty or precaution caps.
This should not have happened because the CPUE data which is used to estimate the status of the stock CANNOT BE regarded as equivalent to fishery independent survey data.
- Deep-sea species caught by French trawlers, namely blue ling, black scabbardfish and roundnose grenadier should fall in a lower category than cat. 3 in the data-poor ranking system of WKLIFE, because fisheries data is unquestionably not as reliable as independent scientific data.
- Given the absence of sufficient scientific information – as repeatedly reminded by scientists, the uncertainty cap and precautionary buffer should systematically apply to deep-sea stocks. If they had not been erroneously placed in category 3, this would already be the case for deep-sea species.
We call on Fisheries Ministers of the EU to bear in mind the particularly vulnerable nature of deepwater ecosystems and organisms when making decisions about deep-sea TACs and quotas.
From the WKLIFE 2 REPORT 2012:
On page 37: Section 4.4 “Commentary on DLS categories 3 and 4”
« Category 3 stocks are those for which survey indices (or other indicators of stock size such as reliable fishery-dependent indices [cpue and mean length in the catch) are available that provide reliable indications of trends in total mortality, recruitment and biomass.
In this category, survey indices, such as CPUE, are treated as equivalent whether they come from fisheries-independent data or from fisheries-dependent data. The scientific justification for this is weak. Survey data are obtained from sampling stations located in a random or stratified random design. It should go without saying that fisheries stations are located using various biases, but generally are repeats of locations that produced reasonable numbers of fish on previous occasions. Consequently, there is no way to make an unbiased estimate of any of the parameters called for under Category 3.
This problem is discussed in Walters and Martel’s (2004) book Fisheries Ecology and Management. They note: ‘Two main problems have caused dangerously misleading overestimates of abundance, recruitment, and net production during population declines and the onset of overfishing: (1) the use of commercial catch per unit of effort (CPUE) or other relative-abundance indices that are, in fact, not proportional to abundance (Harley et al., 2001) and (2) changes over time in the size/age selectivities that confuse the interpretation of population composition data’ (p. 94).
The first of these is especially problematical. Harley et al. (2001) showed for several ICES fish stocks, the commercial CPUE remained high while the survey data showed that populations were declining (so-called hyperstability). That this can happen is due at least in part to the fact that experienced fishing boat captains can find areas where fish are concentrated, and some species may show severe contractions of their range just before the final collapse (for example, North American cod; see Myers and Cadigan, 1995).
The importance of these distinctions has to do with the fact that category 3 species can be exempt from applying the uncertainty cap or the precautionary reduction if it is felt that the abundances of the species are increasing or are at least stable. However, if the commercial CPUE data are likely to be underestimating the degree to which the fish population may be maintaining itself, then such an exemption would be ill-advised.
Category 3 species should be those for which there is survey data to substantiate population abundance claims based on commercial CPUE. However, if all that is available for estimating abundance is fishery-dependent data, those species should be assigned to category 4 and the application of the uncertainty cap and precautionary buffers be mandatory. As an example, one might look at the deep-sea fish, Black scabbardfish (Aphanopus carbo) in Subareas VI, VII, and Divisions Vb and XIIb. For that species, the claim has been made that abundances are stable at least, and perhaps increasing (WGDEEP, 2012). However, the only fishery-independent data for this species comes from surveys taken off Scotland but at depths much greater than where the commercial fishery operates, and those CPUE values have very wide variances indicating extreme patchiness in the abundance values. With regard to fisheries-dependent CPUE data, most come from French vessels that land most of the Black scabbardfish in these northern areas. But they may have contracted their fishing effort into a smaller number of ICES rectangles (compare ICES CM 2012/ACOM:17 with ICES CM 2009/ACOM:14) thus perhaps unwittingly producinghyperstable estimates.”
 The Association française d’halieutique gathers more than 200 scientists from the Ifremer, IRD, Irstea, INRA, various universities & French « grandes écoles » http://sirs.agrocampus-ouest.fr/AFH/index.php/presentation.
« Dans le même temps, la situation de l’ensemble des ressources exploitées dans les eaux européennes de l’Atlantique reste fragile. D’abord parce que le nombre de stocks surexploités ou dégradés reste important (de l’ordre d’un tiers, en nombre de stocks comme en volume potentiel de capture). Et ensuite, parce que de nombreux stocks restent mal connus. C’est le cas des stocks profonds pour lesquels seuls 3 stocks sur 26 disposent de suffisamment d’information pour faire l’objet d’une évaluation scientifique. Parmi eux, un seul remplit les critères de bonne gestion actuellement en vigueur. »
Starting on page 357 for deepwater resources:
Dubé et al., 2012. Observations à bord des navires de pêche professionnelle. Bilan de l’échantillonnage 2011 http://archimer.ifremer.fr/doc/00109/21976/19586.pdf
Starting on page 357 for deepwater resources:
 Ibid, WKLIFE 2012, p. 9ff.
 Ibid, WKLIFE 2012, p. 14-15.
 Ibid, WKLIFE 2012, p. 15 ff.
 WKLIFEII, 2012, Report of The Workshop to Finalize the ICES Data-limited Stock (DLS) Methodologies Documentation in an Operational Form for the 2013 Advice Season and to make Recommendations on Target Categories for Data-limited Stocks (WKLIFE II). ICES CM 2012/ ACOM: 79. See Sections 4.3 and especially 4.4, on page 37 and following. http://www.ices.dk/sites/pub/Publication%20Reports/Expert%20Group%20Report/acom/2012/WKLIFE/wklife2_2012.pdf
 WKLIFEII, p. 37.
 See for ICES species: Harley, S. J., Myers, R. A., and Dunn, A. 2001. Is catch-per-unit-effort proportional to abundance? Canadian Journal of Fisheries and Aquatic Sciences, 58: 1760–1772 and for a history of the collapse of Canadian cod: Walters, C. J. and Martell, S. J. D. 2004. Fisheries Ecology and Management. Princeton University Press. In the case of Canadian cod, fisheries managers were predicting on the basis of catch numbers future increases in cod abundance even as the stock was in its final stages of collapse.