Japanese flounder Paralichthys olivaceus is one of the most important coastal fish species in Japan. However, its catch in the southwestern Sea of Japan has been decreasing. Therefore, there is a need to improve the accuracy of stock assessment. Catch-at-age, which is the basis for stock assessment, is calculated based on the length–frequency distribution using age-length keys. It is recommended that this basic information is revised annually. However, the information for this species has not been updated for a relatively long period. In this study, growth curves were estimated and compared with those recorded from other areas. A multinomial logistic age-length key was generated and compared with the classical catch-at-age calculation method to examine potential errors. On the basis of the results obtained, a sex-specific von Bertalanffy growth curve was selected using the Bayesian information criterion. A model of sex and season was chosen for the multinomial logistic age-length key. The relative error was smaller and the estimated age composition was more accurate with the multinomial logistic age-length key, even when data with missing values were used. The improved accuracy of catch-at-age achieved due to the superior characteristics of the multinomial logistic age-length key is expected to be useful for the advancement of stock assessment.

日本牙鲆Paralichthys olivaceus是日本最重要的沿海鱼类之一。然而，日本西南海的捕获量却在不断减少。因此，有必要提高库存评估的准确性。年龄捕获量是种群评估的基础，是根据使用年龄长度键的长度频率分布来计算的。建议每年修订此基本信息。然而，该物种的信息已经相当长的一段时间没有更新了。在这项研究中，估计了生长曲线并与其他地区记录的生长曲线进行了比较。生成多项逻辑年龄长度密钥，并将其与经典的捕获年龄计算方法进行比较，以检查潜在的错误。根据所获得的结果，使用贝叶斯信息标准选择了性别特异性冯贝塔朗菲生长曲线。选择性别和季节模型作为多项逻辑年龄长度密钥。即使使用具有缺失值的数据，使用多项 Logistic 年龄长度密钥的相对误差更小，估计的年龄构成也更准确。由于多项逻辑年龄长度密钥的优越特性，捕获年龄的准确性得到提高，预计将有助于种群评估的进步。