The general task of image classification seems to be solved due to the development of modern convolutional neural networks (CNNs). However, the high intraclass variability and interclass similarity of plankton images still prevents the practical identification of morphologically similar organisms. This prevails especially for rare organisms. Every CNN requires a vast amount of manually validated training images which renders it inefficient to train study‐specific classifiers. In most follow‐up studies, the plankton community is different from before and this data set shift (DSS) reduces the correct classification rates. A common solution is to discard all uncertain images and hope that the remains still resemble the true field situation. The intention of this North Sea Video Plankton Recorder (VPR) study is to assess if a combination of a Capsule Neural Network (CapsNet) with probability filters can improve the classification success in applications with DSS. Second, to provide a guideline how to customize automated CNN and CapsNet deep learning image analysis methods according to specific research objectives. In community analyses, our approach achieved a discard of uncertain predictions of only 5%. CapsNet and CNN reach similar precision scores, but the CapsNet has lower recall scores despite similar discard ratios. This is due to a higher discard ratio in rare classes. The recall advantage of the CNN decreases with increasing DSS. We present an alternative method to handle rare classes with a CNN achieving a mean recall of 96% by manually validating an average of 6.5% of the original images.

Daily formation of fish otolith micro-increments is frequently assumed, however applying inferences about timing of life history events and formation of otolith macro-structures requires further validation of the periodicity of micro-increment formation. We analysed micro-increments from Western Baltic cod (WBC, Gadus morhua) otoliths marked with tetracycline-hydrochloride as part of an age validation study to test the assumption of daily formation of micro-increments. We found that the number of counted micro-increments consistently underestimated the age of cod aged 1 and older. Time at liberty was also underestimated, especially for fish at liberty during winter. In contrast, micro-increment counts of otoliths from wild-caught young-of-the-year (YOY) cod could be used to realistically estimate timing of hatch and translucent zone formation. Under ambient conditions, settlement did not correspond to any visible pattern within the otoliths, but could be inferred from the prey switch observed from stomach content analyses. We therefore conclude that micro-increments can be assumed to form on a daily basis until the first winter, and can therefore be used to investigate early life history of YOY WBC. However, the periodicity of micro-increment formation appears to vary seasonally in older individuals, with the number of micro-increments formed during the winter period being particularly low.