6 May 2021 - GRSF - Additional requirements for fisheries traceability¶
Meeting Notes¶
Participants:
FishSource (Merul Patel, Susana Segurado, Patricia Amorim, Braddock Spears)
FAO (Marc Taconet, Bracken van Niekerk, Anne-Elise Nieblas, Aureliano Gentile)
Main topics¶
i) Development of a concept of traceability unit, linking a fishing unit to a single stock
a. Semantic code handled as decided here https://support.d4science.org/projects/stocksandfisherieskb/wiki/17-09-26_GRSF_validation
b. Single UUID generated per traceability unit.
c. Non-technical: need to refine definition of fishing area.
ii) Modify rules for Marine Resources
a. Permit no more than one Marine Resource record per species/taxon
b. Define semantic code as: + <”nei”>
c. Allow link to traceability unit
Background (by Susana Segurado)¶
As we near the roll-out of the GRSF codes with industry, and following on from our most recent call where these questions re-arose, I wanted to, as agreed, bring up the remaining issues from our perspective that may affect implementation, hoping we can discuss these and reach solutions.
1 The definition of "Fishing unit (for traceability)": this concept was added, and differentiated from "other fishery" for the purposes of traceability.
One important concern here is that we not permit overlapping units. e.g. not allow one fishery in 27.6.a and a second in 27.6.b, and then the addition of a third fishery in 27.6. This is a simple example which could be easily spotted, but if more complex could very easily be missed. For this reason, I recommend we tighten the definition of "fishing area", to not allow for overlaps.
Another important concern with the "fishing area" (further solidified by reading the FAO draft document on traceability you shared) is whether we will be able to meet legal reporting needs. We need to be able to evaluate the legality and the management . If we leave the "fishing area" broadly or flexibly defined then in theory a FAO major fishing area like FAO 27 could be included. For most species though that would not allow us to meet our objectives.
Given these concerns, I still believe that the fishing area must follow the concept of a management area or reporting area (such as a TAC area). Even if you decide you do not want to restrict the definition of a "fishing unit" in this way, could we alternatively add a new concept of a "traceability unit" that is defined in this way?
2 Still on the "fishing unit" or perhaps a new concept of "traceability unit", as we've discussed before, in order to be useful for traceability the definition must include reference to a single stock. For traceability purposes, the main (even only!) product of interest is a unique code for a fishing/traceability unit. We discussed how to handle this in the semantic code for a fishery unit (https://support.d4science.org/projects/stocksandfisherieskb/wiki/17-09-26_GRSF_validation) but this has not yet been implemented. And we do not yet have a solution as to how to handle this in terms of UUIDs. An ad hoc web service was mentioned (5th May 2020) but I'm not sure I understand this. And a modification or new definition is needed to include the stock in the concept, as fisheries cannot be unique for traceability without it, e.g.
There is an EU TAC set for "union waters of 2a and 4" for Nephrops. This covers 4 functional units (FU 7, 9, 10 and 32) and the area outside functional units that is not assessed. If one fishery were created for areas 2.a and 4 it would need to link to all four stocks plus the resource, thereby not being unique. The fishing unit by itself would also not immediately tell us which stock is being caught. The stock areas are the differentiating factor here as they are the smaller unit. However I do not believe we would want to add the complexity into the "fishing area" field of distinguishing the stocks. Instead, that differentiation can be made by the construction of the code itself.
Similarly, there is an EU TAC set for haddock in "union and international waters of 5b and 6a". Area 5b is part of one assessment unit: Faroes; and 6a is part of another one: North Sea, West of Scotland, Skagerrak. A single fishery created for the 5b and 6a management unit (or even "fishing area") would also need to link to both stocks, which is why these would need to be different fisheries. This is a case where we have clear area numbering, but where that is not the case, it becomes much more difficult to detect.
This happens anywhere there is a management (fishing) area that is not a perfect subunit of the assessment area, which is not infrequent.
So creating a fishing unit without a stock is problematic (and these are currently flagged with Traceability flag "true" but are not sufficient for traceability). I see the FAO draft document on traceability you shared addresses this by introducing a "Traceable Resource Unit (TRUs)", which would correspond to this "traceability unit" but the impracticality of not having a single UUID, arguably for the purpose that could make most use of a UUID, remains.
I will add this comment too to the FAO draft traceability document although it is partially a GRSF question, but there will need to be an ability during the harvest to look up both the stock code and the fishery code (or just the traceability code). So either the "fishing zone" mentioned in the draft doc will need to be at a resolution to allow both the stock and the fishery to be identified (looked up), or both the stock and the fishery IDs will need to be known preemptively or in real-time. Even if the reporting area is considered for the "fishing zone", in the examples above that does not allow the stock to be unequivocally identified. To be able to identify a traceability unit, we will need to know the smallest of the area needed for legal reporting or the area that will allow us to define a stock, which is why I argue both these areas must be included.
3 We also still have a decision to take around "resources" - we have not approved any resource records yet while a decision is pending. Sort of following an earlier proposed model, we could create one "resource" record per species for the unassessed portion and add "nei" to the assessment areas field, following FAO reporting standards. The alternative of leaving it blank, which we discussed at some point during the "other fishery" discussions, could be misinterpreted to mean all areas, which would include assessment areas.
E.g. Stock: Species + Assessment Area(s)
and Resource: Species + nei
Would this be possible to implement under the current GRSF structure or would this imply new tech needs?
Again here, when creating a fishery/traceability unit with a resource, the importance of tightly defining the fishing area is critical for uniqueness.
Actions¶
- Susana and Aureliano to draft concept note for "Traceability Unit" and then submit to the group.
- Aureliano to report on main outcomes for the public consultation on the "Advancing end-to-end Seafood traceability" (https://seafoodtraceability.konveio.site/)"
