Improving spine-based age estimation in centrarchid fishes using otolith-derived training sets

Area of Science:

• Ichthyology
• Fisheries Science
• Quantitative Biology

Background:

• Nonlethal age determination in fish is crucial but often lacks precision and accuracy.
• Otoliths provide accurate estimates but require fish sacrifice, limiting their use.
• Dorsal spines offer a nonlethal alternative but require improved estimation methods.

Purpose of the Study:

• To assess the effectiveness of lake-specific training sets for calibrating nonlethal age estimates from dorsal spines.
• To improve the precision and accuracy of fish age estimation using dorsal spines.
• To reduce underestimation biases and reader variability in fish aging.

Main Methods:

• Paired otolith-spine images from incidental mortalities were used to create training sets.
• Five readers estimated ages from spine images before and after calibration.
• Calibration involved familiarization with lake-specific otolith-spine image pairs.

Main Results:

• Calibration consistently increased estimated ages, especially for older fish.
• Variability among readers in age estimates and length-at-age trajectories decreased post-calibration.
• Training sets effectively reduced underestimation biases and reader-based variance.

Conclusions:

• Lake-specific training sets derived from incidental mortalities offer a practical method to improve nonlethal spine-based age estimation.
• This approach enhances the reliability of age data for fisheries management and biological studies.
• Utilizing incidental mortalities for calibration is recommended for improving nonlethal aging programs.