1. Home
  2. Research
  3. Projects
  4. Developing dynamic species distributions models to inform conservation and management of non-target species and communities in the eastern Pacific Ocean
Project N.1.c - Developing dynamic species distributions models to inform conservation and management of non-target species and communities in the eastern Pacific Ocean

01 Mar 2021 - 31 Aug 2022

Program(s) in charge: Ecosystem & Bycatch Program
Funded
Objectives
Contribute to the development of high-resolution dynamic habitat models for key non-target species and ecological functional groups impacted by tuna fisheries to better understand the dynamics of target-bycatch-environment co-occurrence and assess the vulnerability of the species under existing and projected effort and environmental regimes using EASI-Fish.
Background
  • Managing the diverse range of co-occurring species is a significant challenge owing to the dynamic biophysical environment of the EPO at different scales
  • Understanding the likelihood of species-fishery interactions requires knowledge of each species’ spatio-temporal distribution relative to that of the fishing effort under specific environmental conditions
  • Besides, dynamic models can assist in the assessment of the potential vulnerability of species and ecological functional groups (e.g. hammerhead sharks) to existing or predicted levels of fishing effort using EASI-Fish
  • The IATTC has done significant progress on dynamic models of distribution for the main tropical tuna species (e.g. SAC-10-INF-D) but models for some of the most important key bycatch species are missing
  • The project will produce models for a total of 8 species, selected based on IATTC’s current conservation and management priorities and data availability
Relevance for management
Advancing our understanding of the relationship between environment, biological community structure and vulnerable bycatch species to guide the development of alternative and/or complementary bycatch mitigation measures
Duration
18 months
Workplan and status
  • Mar-Apr 2021: Conduct exploratory data analysis and extraction of environmental covariates
  • Apr-Dec 2021: Develop models and evaluations for 8 key bycatch species
  • Dec 2021-Apr 2022: Run model predictions
  • Dec 2021-Aug 2022: Preparation of written reports and peer-reviewed manuscripts
  • Apr 2022-Aug 2022: Development of a beta online portal for decision makers
  • Aug 2021-Aug 2022: Continuous engagement with IATTC CPCs, fishers, and other key EPO resource stakeholders
External collaborators
Stockholm Resilience Center at the University of Stockholm
Deliverables
  • A compendium of spatially-explicit dynamic species distribution models for key non-target bycatch species
  • A beta-version user-friendly online platform to visualize main results and promote engagement and conversations with decision-makers
  • Dissemination of material, including peer review publications, documents and presentations for the IATTC SAC and working groups on Bycatch and FADs, capacity building workshops with stakeholders, and other national and international scientific forums
Updated date: 01 May 2023
Progress summary for the reporting period
  • Long-term empirical data was analyzed to assess the effectiveness of static vs dynamic management options for two vulnerable shark species.
  • Machine-learning species distribution models were run for key bycatch species, including certain species of sharks and the critically endangered leatherback turtle.
  • A set of predictions for those key sensitive bycatch species are being run to help improve EASI-Fish models.
Challenges and key lessons learnt
  • Closing areas of high fishing inefficiency, and reallocating effort proportionally to reflect historical patterns, yearly tuna catch may have increased while the bycatch of certain sharks could have decreased significantly.
  • Static closures seem less effective than dynamic and adaptive measures, which should be considered to more efficiently fulfill conservation and sustainability objectives in the EPO.
  • Machine-learning algorithms are powerful tools to deal with data-limited species and can produce accurate and reliable species distribution models for sensitive species. 
  • Data confidentiality issues were experienced by participants, which delayed the project significantly. However, a solution was found, and analyses are being run preserving all confidentiality aspects of the data.
  • Predictions for sharks are underway, beginning with silky shark.
  • Presentation at BYC-10
  • Presentations and documents at BYC-11 (BYC-11-01, BYC-11-04)
  • A manuscript is under review in a peer-reviewed journal.
Comments
The COVID-19 pandemic and issues with data sharing and confidentiality delayed the project. The number of SDMs to de delivered will be revised to meet conservation priorities and deadlines.
The postdoctoral position of the main collaborator is over, and this work will be revised and taken over by the new members of the Ecosystem and Bycatch Program.