Image analysis approach to estimate fecundity of live-bearer rockfishes (Sebastes spp.) along the California Coast Environmental Biology of Fishes

Current gravimetric methods, involving manual counts of oocytes or embryos to estimate fecundity for live-bearing rockfishes (Sebastes spp.) of the California Current Ecosystem, are time-consuming, hindering the rapid assessment of reproductive potential. To improve efficiency in data collection, we adapted the autodiametric method, which uses image analysis to relate mean oocyte diameter to a weight-based oocyte density to rapidly estimate fecundity. We developed autodiametric calibration curves for unfertilized oocytes of five California Current rockfishes: chilipepper (Sebastes goodei), rosy rockfish (S. rosaceus), widow rockfish (S. entomelas), yellowtail rockfish (S. flavidus), and vermilion rockfish (S. miniatus). The autodiametric method was nearly five times faster than the gravimetric method while maintaining precision and unbiased estimates. Autodiametric calibration curves were species-specific due to significant differences in oocyte density for a given mean diameter. The difference in densities was attributed to differences in mean oocyte weight, standardized by diameter, likely due to different stages of pre-fertilized development and/or energy investment among species. The autodiametric method was unable to detect and measure fertilized stages due to poor edge detection and non-circularity of embryos; however, gravimetric counts of embryos were already faster compared to unfertilized samples. Most rockfishes exhibit strong maternal size effects on the number of offspring produced so the ability to rapidly assess maternal effects and to explore spatio-temporal trends in fecundity will provide valuable reproductive information to fisheries managers. This is particularly true given the need to evaluate the impacts of exploitation and climate change on reproductive life history parameters in this diverse genus. Grant no. NA18NMF4270216 Grant no. NA18OAR4170323 Grant no. NA20OAR4320278-13