Project

General

Profile

GRSF ยป

4 August 2016 GRSF requirements wrap-up

Meeting Notes
Topics: Objective of the call was to wrap-up the discussions held so far, highlighting the outstanding points. The outcome is stored in the wiki page GRSF_database_overview, which is expected to serve as the basis for the development of the first version of the GRSF.

Participants: FishSource (Susana Segurado); RAM (Daniel Hively); FORTH (Yannis Marketakis); FAO (Marc Taconet, Kiran Viparthi, Anton Ellenbroek, Giulia Gorelli, Aureliano Gentile)

Notes

The wrap up document GRSF_database_overview was presented, introducing the overall logic for the creation of GRSF records, including data importing and processing (dissection, mapping and assignment of identifiers), competency questions, pending issues and how to move forward. A simulation for creation of GRSF records was also presented (see GRSF_UUIDs.xlsx Worksheet "GRSF_record_creation" http://goo.gl/0RLZ8f) and various scenarios were discussed.

It was recognized the importance to adopt shared agreed standards for all fields of GRSF records to satisfy the needs of the business cases. To achieve this, two options were envisaged: 1) the update of source contents; 2) the provision of detailed mapping between local classifications and GRSF standards.

The proposal to use two kind of identifiers, the machine generated alpha-numeric UUID and the Semantic ID resulting from a concatenation of relevant GRSF fields was recognized by the participants as the viable solution to identify unique records and establish proximities.

The issue on the time component of the records was debated, and it was concluded to have one reference year per source record. The wrap-up document has been updated accordingly.

The pending issues on the fields Management entity and Fishing vs Jurisdiction Area were further discussed. In particular:

  • Management entity: in case of multiple management entities, the legitimate authority for traceability is the one issuing binding decisions. At present, the FIRMS data model does not fully capture the distinction of responsibilities among different management entities, while FishSource identifies management units upon review of the management entities responsibilities (see how FishSource identifies Management Units https://support.d4science.org/attachments/download/618/FishSource_AboutManagementEntity-ManagementUnitConcepts_July2016.docx).

  • Fishing vs Jurisdiction Area: The need to have two types of areas for describing a fishery was confirmed. Particular importance was given the jurisdiction area defining the management unit where the management entity exerts its mandate. For example, in the case of a bigeye tuna fishery described in the Atlantic High Seas (intersecting FAO area 34), the jurisdiction area would be "ICCAT Management Unit Area BET_ATL - Atlantic", while the fishing are field would be one or more sub-divisions of the FAO area 34.
    Further examples are in the Excel file (Worksheet GRSF_record_creation).

  • Given the heterogeneity of management regimes, specific guidelines on the use of the agreed standards need to be developed and adopted according to the different areas. In general, we should use the standard preferred by the legitimate authority to describe their fisheries (EEZ, statistical areas, management areas and/or a combination). The following example on the development of guidelines was provided during the meeting: the management decisions in CCAMLR are clearly assigned by statistical are/sub-area or division (https://www.ccamlr.org/en/organisation/explanation-terms). There is also the notion of Small scale research unit. In terms of guidelines, this leaves us with the need to decide among these two options: Option 1) we put these statistical areas in the Fishing area field, or Option 2) we put these statistical areas in the Jurisdiction field because they are used by CCAMLR to enact binding decisions. The decision needs to be related to one of the two following logics: 1) we give prevalence to concepts (Jurisdiction includes all type of areas considered for defining binding management); 2) we give prevalence to users and systematically put jurisdictions (RFBs comp areas, EEZs, specifically defined FMUs such as ICCAT or Australian) in the jurisdiction area field, and statistical or LMEs in the fishing area field.

At the moment, the heterogeneity of standards in the sources and the lack of appropriate mapping rules to the GRSF standards is forcing to accept "local" standards. This simplification will allow the implementation of the first version of the GRSF.
Therefore, the conventional rule for selecting codes to fill GRSF fields and Semantic ID in case of multiple sources (i.e. FIRMS>FishSource>RAM) is intended to be temporarily enforced and functional exclusively to the first implementation. After enforcement of GRSF standards those records not compliant would not be eligible as GRSF records.
After the release of the first version (Autumn 2016), the development of appropriate mapping rules is expected by partners in order to comply with the agreed standards for the creation of GRSF records and Semantic IDs. FishSource and RAM agreed on the need to provide such mapping or to update their data adopting the GRSF standards.

Follow-up actions

  • FishSource and RAM to provide a brief description on their expectations of GRSF functionalities and possible uses.
  • FORTH and CNR to start implementation upon the requirements summarized in the wrap-up document and Excel working files.
  • FORTH to liaise with FIRMS, FishSource and RAM to refine the data importing rules according to requirements.
  • FAO to assist FORTH and CNR colleagues as required

Working files and materials

Add picture from clipboard (Maximum size: 8.91 MB)