Oral Presentation Australian Society for Fish Biology Conference 2025

CheckEM: an open-source toolkit for standardising, cleaning, and visualising stereo-video fish survey data (123920)

Brooke Gibbons 1 , Claude Spencer 1 , Jordan Goetze 2 , Dianne McLean 3 , Todd Bond 1 , Jacquomo Monk 4 , Matthew Navarro 1 , Daniel Agnello 1 , Charlotte Aston 1 , Charlie Huveneers 5 , Stephen Newman 6 , Andrew Hoey 7 , Ben Radford 3 , Nathan Knott 8 , Conrad Speed 3 , Daniel Ierodiaconou 9 , Shaun Wilson 3 , Tim Langlois 1
  1. The University of Western Australia, Crawley, WA, Australia
  2. Department of Biodiversity, Conservation and Attractions, Kensington, Western Australia, Australia
  3. Australian Institute of Marine Science, Perth, Western Australia, Australia
  4. Institute for Marine and Antarctic Studies, University of Tasmania, Hobart, Tasmania, Australia
  5. College of Science and Engineering, Flinders University, Adelaide, South Australia, Australia
  6. Department of Primary Industries and Regional Development, Hillarys, WA, Australia
  7. College of Science and Engineering, James Cook University, Townsville, QLD, Australia
  8. Fisheries Research, NSW Department of Primary Industries and Regional Development, Huskisson , NSW, Australia
  9. School of Life and Environmental Science, Deakin University, Warrnambool, Victoria, Australia

Background:
Effective research and monitoring requires accurate, interoperable and representative data. Video-based methods are commonly used to survey fish and are rapidly expanding globally due to their cost-effectiveness, ability to provide accurate body size measurements, non-destructive sampling approach and capacity to create permanent data records. However, the effectiveness of video-derived data depends on standardised approaches which produce reliable, reproducible, and error-free data. A national synthesis of video-based fish survey datasets identified numerous errors in data collection and annotation, many of which are also relevant to other survey methods.

Aims:
To develop an open-source toolkit, CheckEM, for quality control checks on fish survey data.

Development:
The CheckEM web application and R package identifies errors in metadata and cross-checks annotations of fish with taxonomic databases, expected spatial distributions, and maximum body sizes. CheckEM flags species observed beyond their known range, outdated scientific names, and body size outliers. CheckEM standardises, cleans, and visualises datasets, offering interactive plots and tables in a user-friendly interface. Downloadable summary data and error reports support iterative checks and improvement of data quality.

Conclusions and Implications:
CheckEM enhances data accuracy, confidence, interoperability, and reusability, improving collaboration and cross-dataset comparisons to support robust analyses and informed marine resource management.