Age at Maturation Predicted from Routine Scale Measurements In Norwegian Spring-Spawning Herring (Clupea herengus) Using Discriminant and Neural Network Analysis

We evaluate two methods allowing the prediction of age at maturation from the width of annual growth layers in scales (or otoliths) in a case study on Norwegian spring-spawning herring. For this stock, scale measurements have been collected routinely for may decades. We compare the performance in classifying age at maturation(at 3,4,...9 years) between conventional discriminant analysis (DA) and the new methodology of artificial neural networks trained by backpropagation (NN) against a 'control' comprising historical estimates of age at maturation obtained by visual examination of scales. Both methods show encouraging, and about equally high classification success. The marginal differences in performance are in favour of DA, if the proportion of correctly classified cases is used as criterion (DA 68.0%, NN 66.6.%), but in favor of NN if other criteria are used including the prediction errors (error >1 year: DA 5.2%, NN 2.9 %) and the average degree of under- or overestimation (DA: underestimation 1.1% of mean; NN: overestimation 0.2% of mean). We provide evidence that both methods approach the a priori limit to maximal classification success, caused by overlapping combinations of predictor variables between maturation groups. These methods will allow studies on age at maturation in this important fish stock over a very long time-span including periods well before, during, and after its collapse to commercial extinction. Similar techniques might well be applicable to any other fish stock with long-term data on scale or otolith growth layers available.
KEYWORDS: Age at maturation; Classification; Discriminant analysis; Growth layers; Neural network; Norwegian spring-spawning herring; Scale measurements

VIEW CONTENT