CGIAR/aquatic-macroinvertebrate-images

Dataset Card for Aquatic Macroinvertebrate Images

This image dataset contains 1,300 photos (one hundred images of each of the thirteen mini stream assessment scoring system, or miniSASS aquatic macroinvertebrate groups), used in the development of machine learning models to automatically assess the water quality score.

Dataset Description

This dataset accompanies the working paper, Digitally improving the identification of aquatic macroinvertebrates for indices used in biomonitoring.

Koen, R. C. J.; Koen, F. J.; Pattinson, N. B.; Dickens, Chris W. S.; Graham, P. M. 2023. Digitally improving the identification of aquatic macroinvertebrates for indices used in biomonitoring. Colombo, Sri Lanka: International Water Management Institute (IWMI). CGIAR Initiative on Digital Innovation. 10p. https://hdl.handle.net/10568/138246

The report provides an overview of the mini Stream Assessment Scoring System (miniSASS) and South African Scoring System Version 5 (SASS5) as biomonitoring techniques for assessing the ecological condition of streams and rivers based on the identification of aquatic macroinvertebrates. While miniSASS relies on minimally trained citizen scientists to identify macroinvertebrates at the Order-level, SASS5 utilizes expertly accredited practitioners for finer resolution, even up to the family-level. However, the reliance on citizen scientists for miniSASS identification introduces limitations in terms of precision, accuracy, and reliability. To address these limitations, ongoing developments within the CGIAR Initiative on Digital Innovation include the creation of a miniSASS smartphone application, an upgraded website, an interactive online course, and a machine-learning identification algorithm to assist with photo identification. Additionally, a revised dichotomous key has been developed to improve operator identification during miniSASS surveys. The outcomes of these developments and explorations presented in the paper aim to enhance the overall effectiveness and reliability of both the miniSASS and SASS5 techniques. By leveraging digital innovation and incorporating machine-learning technology, we anticipate the efficiency, accuracy, and accessibility of biomonitoring assessments will significantly improve, ultimately contributing to a better understanding and management of our aquatic ecosystems.

• Curated by: Nicholas Pattinson, GroundTruth
• Funded by: CGIAR Initiative on Digital Innovation
• Shared by: Chris Dickens, CGIAR/IWMI
• License: CC-BY-4.0

Dataset Sources

• Repository: https://doi.org/10.7910/DVN/1QQPJ5
• Paper: https://hdl.handle.net/10568/138246
• Demo: https://minisass.org

Source data producers

• GroundTruth https://www.groundtruth.co.za
• Internatoinal Water Management Institute https://iwmi.org

Annotators

• R.C.J. Koen (North-West University, South Africa)
• F.J.K Koen (North-West University, South Africa)

Dataset Card Contact

• Jawoo Koo, CGIAR/IFPRI