Funded
Click to see projects by theme, goal or target:
- Objectives
- To use the EASI-Fish ERA approach to assess vulnerability status and the efficacy of conservation and management measures prescribed under IATTC Resolution C-19-04 for reducing fishing impacts on the East Pacific stock of leatherback turtle (Dermochelys coriacea).
- Background
- IATTC is committed, through the Antigua Convention, to ensure the long-term sustainability of all non-target species impacted by EPO tuna fisheries.
- On 1 January 2021 a revised resolution on sea turtles (C-19-04) entered into force that requires EPO tuna fisheries to implement various measures designed to reduce the bycatch of sea turtles, in particular the use of circle hooks and finfish baits in shallow longline sets.
- EASI-Fish has been used by the IATTC as an alternative approach to traditional population models to assess the efficacy of management measures on data-limited bycatch species, including the critically endangered spinetail devil ray.
- In collaboration with the Inter-American Convention on the Protection and Conservation of Sea Turtles (IAC) and EPO stakeholders, the staff developed a preliminary EASI-Fish assessment for 2018. The project was extended to improve on this model through the development of a dedicated species distribution model and an update of the fishing effort by coastal artisanal fisheries.
- Relevance for management
- EASI-Fish can rapidly and cost-effectively quantify the cumulative impacts of multiple data-limited fisheries on species under proposed management measures—either individually or in combinations—under IATTC Resolution C-19-04 to determine their potential efficacy of reducing the vulnerability of the EP leatherback turtle stock to becoming unsustainable in the long-term. This will ultimately simplify the choice of management measures required to meet conservation and fisheries objectives.
- Duration
- 12 months
- Workplan and status
- Jun-Sept 21: Collaborate with stakeholders to collate available fishing effort and leatherback presence data in the EPO.
- Sept 21-Jan 22: Develop a new approach to use presence and absence records to produce a dedicated species distribution model (SDM) for the East Pacific leatherback turtle stock.
- Jan 22-Apr 22: Populate EASI-Fish model with biological and fisheries data and run 70 hypothetical scenarios
- May 22: Present final EASI-Fish assessment results and the special distribution model to the Bycatch Working Group (BYC-11).
- External collaborators
- IAC, CPCs
- Deliverables
- Papers and oral presentations for BYC-11
- Scientific journal publications
- Updated date: 01 May 2023
- Progress summary for the reporting period
- Jun-Sept 21: Collaborated with IAC, CPCs and stakeholders to collate available fishing effort and leatherback occurrence data in the EPO.
- Sept 21-Jan 22: Developed a new machine-learning approach to use presence and absence records and a series of environmental variables to produce a dedicated species distribution model (SDM) for the East Pacific leatherback turtle stock.
- Jan 22-Apr 22: Populated EASI-Fish model with biological and fisheries data and ran 70 hypothetical scenarios
- Apr 22: Prepared EASI-Fish assessment and SDM results to present at BYC-11.
- May 22: Presented EASI-Fish assessment and SDM results at BYC-11.
June 22: Presented EASI-Fish assessment to the IAC COP10 via video.
Both technical documents presented at BYC-11 were processed and submitted, as a joint submission, to a peer-reviewed journal.
- Challenges and key lessons learnt
- The machine learning algorithm used to generate the SDM and predictions for the EP leatherback turtle is capable of depicting hotspots of species habitat suitability and describe the species environmental preferences.
- The estimated fishing mortality, and hence vulnerability status, is strongly influenced by predictions from an SDM and also the threshold value used to define cells where the species is predicted to be present. Although the new SDM was greatly improved, further exploration of how to best determine threshold values is desirable.
- The complex life history of leatherback turtles presented new technical challenges for the EASI-Fish model that is constructed using a single annual timestep. Further model development is required to better represent spatial heterogeneity in fishing impacts and the potential impacts of spatial closures. For example, different size classes of animals are present in different regions during the breeding season, so a 2-stage model is desirable to characterize this aspect.
- International highly collaborative projects can be successful to develop studies on data-limited species that require a significant amount of data and explore and assess the potential effect of different conservation and management measures, both individually or collectively.
- BYC-11-01 – A machine learning SDM for the EP leatherback turtle
- BYC-11-02 – An EASI-Fish assessment of the EP leatherback turtle and effectiveness of measures in Res. C-19-04
- Lopez J, Griffiths SP, Wallace B, Caceres V, Bustos LC, Cocas L, Vega R, Zárate P, Clavijo L, Cari I, Rodriguez-Baron JM, Carvajal JM, Piedra R, Andraka S, Rendón L, Herrera M, Suárez J, Santana H, Abrego M, Veelenturf C, Quiñones J, Perez M, Alfaro J, Mangel J, de Paz N (In Review) A machine learning species distribution model for the critically endangered east Pacific leatherback turtle (Dermochelys coriacea). Endangered Species Research.
- Griffiths SP, Wallace B, Cáceres V, Rodríguez LH, Lopez J, Abrego M, Alfaro-Shigueto J, Andraka S, Brito MJ, Bustos LC, Cari I, Carvajal JM, Clavijo L, Cocas L, Paz Nd, Herrera M, Lauritsen AM, Mangel JC, Perez M, Piedra R, Dávila JAQ, Rendón L, Rguez-Baron JM, Santana H, Stacy B, Suárez J, Swimmer Y, Veelenturf C, Vega R, Zárate P (In Review) Vulnerability status and efficacy of potential conservation measures to reduce bycatch of the critically endangered East Pacific leatherback turtle (Dermochelys coriacea). Endangered Species Research.