Interconnected partnerships: Mapping collaborations in Australian fisheries stock assessment

Collaboration in science is important because it can lead to efficient knowledge exchange and grow scientific understanding. Examining the nature of co-authorship on reports and peer- reviewed papers can give insight into aspects of science collaboration. Here we examine the bibliographic networks of scientists who provide the scientific basis for Australian fisheries decision making using the methods of stock assessment. We focus on Australian stock assessments and determine co-author clustering around organisation, geolocation, and fished species. For the peer- reviewed literature there is evidence of clustering around a few individual authors characterised by their large number of publications and their relatively frequent first authorship. These stock assessment networks are vulnerable because knowledge exchange may be constrained by repeat collaboration with the same central co-authors who are mostly from the same organisation, leading to siloing. However, this does not mean the stock assessments are necessarily wrong. We also find that network connectedness is highly influenced by a few papers that have a higher-than-average number of co-authors, many of which involve “one-off” authors. This study highlights the need for increased and diversified collaboration, facilitation of interactions, and greater information sharing, among stock assessment scientists who provide the scientific basis for fisheries decision making. This needs to be balanced against the capacity of the available pool of stock assessment scientists.