Estimating fecundity using image analysis in Cervimunida johni and Pleuroncodes monodon

Fecundity is directly related to the reproductive potential in all exploited marine species. The gravimetric method has been widely used for the estimation of fecundity in the commercially valuable squat lobsters Cervimunida johni and Pleuroncodes monodon due to its simplicity and precision. However, the procedure is time-consuming because it requires counting of all eggs from the subsample. The fecundity of some fish species has been successfully estimated using the autodiametric method, which consists of the conversion of oocyte size into oocyte density per weight of the ovary. The purpose of this study was to apply and assess an automated image analysis and the autodiametric method for the estimation of fecundity in C. johni and P. monodon. A total of 204 ovigerous females of C. johni and 176 of P. monodon were collected from a fishing survey in August 2015. The results showed that the “counting error” was lower than 12% in the particles analyzed in both squat lobster species. In addition, the “measurement error” generated an average particle exclusion of 16% and 42% for the measurement of egg size in C. johni and P. monodon females, respectively. During the processing time to analyze the mass of eggs from a female in laboratory, the autodiametric method proved to be faster and more efficient because it lasted on average 61% and 73% less than the gravimetric method in C. johni and P. monodon, respectively. The potential fit between the volume-based mean egg diameter (D_V) and egg density (g⁻¹ egg mass) was significant for both species. The relationship between female size and fecundity was significant for both squat lobster species. Furthermore, the analysis of covariance indicated that the coefficients were not significantly different between both methods. We discussed the autodiametric procedure and the precision of fecundity estimates for both squat lobster species.