Using automatic open-source analysis of backscatter data from fishing vessels to implement feedback management of the Antarctic krill fishery

The implementation of a feedback management system for a sustainable Antarctic krill fishery is challenged by a lack of krill abundance data. Here we present the Krillscan python package to automatically process and share echosounder data from both scientific surveys and fishing operations. This enables an accelerated, transparent, and reproducible analysis of krill backscatter data. We tested our algorithms against manual scrutinization with the LSSS software and found satisfactory agreement with some minor offsets in the estimated nautical area scattering coefficients. In addition to the automatic processing and data transfer, Krillscan also contains a graphical user interface to supervise the automatic krill swarm detection. We found that echogram size can be compressed up to 100 times and raw data are processed faster than they are generated, enabling near-real time analysis and data transfer on manned or unmanned vessels.