Fisheries scientists are analysing images of giant crabs for shell colour patterns, size and sex data, which will contribute to better managing harvests in this important fishery.
Giant crab is a high-value species that supports small fisheries in Tasmania, South Australia and Victoria, but all three states face ongoing difficulties in collecting detailed data on the catch.
Project leader, Dr Scott Hadley from IMAS at the University of Tasmania, said crab size is a critical step in stock assessment modelling.
“Collecting data at sea is especially difficult in this fishery, because conditions are often rough and only a small number of individual crabs are captured on most days. This means it’s difficult to track the status of the stock,” Dr Hadley said.
“Currently, the assessment of giant crab relies on catch rate data, but this data can be unclear due to the small number of operators in each fishery.”
To address these data collection difficulties in the fishery, giant crab fishers are working with researchers from IMAS, the South Australia Research and Development Institute (SARDI) and the Victorian Fisheries Authority (VFA), in a project supported by the Fisheries Research and Development Corporation (FRDC).
“Our aim is to develop an effective, low-cost data collection method using an image recognition system, which will take photos of the giant crabs’ carapace, or top shell, and analyse the images to determine their length and sex,” Dr Hadley said.
“Automated computer formulas will also use the photos to identify individual crabs through colour patterns on their carapace. This speckled colouration remains unaltered after shell moulting, so is potentially equivalent to a unique fingerprint.”
In the future, researchers may be able to use this ‘fingerprint’ shell pattern to track changes in the size of recaptured crabs and determine their growth.
“To make sure our measurements are accurate, we’ll also test the computer formulas to see if they can account for different factors that may affect size estimates during image analysis, such as vibration from fishing vessels, the angle of the crab and varying light conditions,” Dr Hadley said.
The final results will rely on the evaluation of size measurement accuracy and a risk assessment to investigate the impact of inaccurate estimates on the stock biomass assessment.
The researchers expect that the project will significantly contribute to improving the giant crab stock assessment, and give greater certainty in fishery management decisions.
The project is funded by the Fisheries Research and Development Corporation on behalf of the Australian Government.
Images: Researchers using cameras and sensors to capture images of giant crabs. The data will enable researchers to estimate crab size, classify them by sex and identify individual crabs. (Photos: Weijun An)
Published 28 September 2021