When it comes to fishery management, good quality data is the foundation for making good decisions. Thanks to an innovative research project underway at the Institute for Marine and Antarctic Studies (IMAS), getting data to inform scalefish stock assessments could be about to become a lot cheaper and easier.
Stock assessments for scalefish species usually rely on interpreting patterns in commercial catch data. But for many of our scalefish species, recreational fishing makes up an increasingly large amount of total catch.
Funded by the Fisheries Research Development Corporation (FRDC), IMAS are exploring how new technologies, like electronic measuring boards and AI image recognition, can help fill research gaps by collecting data about fish length. If successful, these new data collection methods could revolutionise stock assessments for these fisheries.
IMAS are training AI to recognise fish species from underwater footage, providing a new way to collect data on our important scalefish species. Photo: IMAS
Adding fish length data to stock assessments
One technology IMAS are trialling are electronic fish measuring boards for commercial fishing and recreational charter fishing boats. Fishers with the boards can enter what species of fish they’re measuring, and the board then keeps a digital record of the length of each fish that’s measured. At the end of a fishing trip, fishers can then send this data directly to IMAS scientists, giving researchers access to more fish length data for more species.
Dr Alyssa Marshell from the IMAS research team demonstrating how the electronic measuring boards works at the Australian Wooden Boat Festival.
Why the length of fish is important
Understanding
the length of fish, as well as how many fish there are of different
lengths, is key to assessing how healthy our fish stocks are. That’s
because in fish, length and weight are related to reproduction, and the range of
sizes and weights in a fish stock helps us understand what’s going on under the water
in terms of breeding and fish mortality.
For example, if a fish stock has a wide range of
different length fish in good numbers, that’s a great indicator of good
stock health. On the other hand, if some sizes are missing (e.g. very
few large, old fish), it can be a warning sign that something is awry
with the stock.
|
Another aim of the research project is taking automated data collection a step further by training AI to collect length and species data of fish from underwater footage. If successful, this approach will allow researchers to quickly and efficiently collect valuable data, without having to catch any fish or spend hours in the lab watching footage.
Baited Remote Underwater Video (BRUV) footage from 50-60m deep off Schouten Island. Footage like this is being used to help train AI models to recognise different scalefish species. Video credit: IMAS
These innovative approaches are set to transform fisheries monitoring over the next decade. By improving the accuracy of length-based stock assessments, this research aims to contribute to sustainable management decisions that support the long-term health of Tasmania’s scalefish fisheries and guide future harvest strategies.