An Open Science approach to infer fishing activity pressure on stocks and biodiversity from vessel tracking data

Vessel tracking data help study the potential impact of fisheries on biodiversity and produce risk assessments. Existing workflows process vessel tracks to identify fishing activity and integrate information on species vulnerability. However, there are significant data integration challenges across the data sources needed for an integrated impact assessment due to heterogeneous nomenclatures, data accessibility issues, geographical and computational scalability of the processes, and confidentiality and transparency towards decision making authorities. This paper presents an Open Science data integration approach to use vessel tracking data in integrated impact assessments. Our approach combines heterogeneous knowledge sources from fisheries, biodiversity, and environmental observations to infer fishing activity and risks to potentially impacted species. An Open Science e-Infrastructure facilitates access to data sources and maximises the reproducibility of the results and the method's reusability across several application domains. Our method's quality is assessed through three case studies: The first demonstrates cross-dataset consistency by comparing the results obtained from two different vessel data sources. The second performs a temporal pattern analysis of fishing activity and potentially impacted species over time. The third assesses the potential impact of reduced fishing pressure on marine biodiversity and threatened species due to the 2020 COVID-19 lockdown in Italy. The method is meant to be integrated with other systems through its Open Science-oriented features and can rapidly use new sources of findable, accessible, interoperable, and reusable (FAIR) data. Other systems can use it to (i) classify vessel activity in data-limited scenarios, (ii) identify bycatch species (when catchability data are available), and (iii) study the effects of fisheries on habitats and populations' growth.