Estimating age at maturation and energy-based life-history traits from individual growth trajectories with nonlinear mixed-effects models
|Author(s)||Brunel Thomas1, Ernande Bruno2, Mollet Fabian M.1, 3, Rijnsdorp Adriaan D.1, 4|
|Affiliation(s)||1 : Wageningen IMARES, NL-1970 AB Ijmuiden, Netherlands.
2 : IFREMER, Lab Ressources Halieut, F-62321 Boulogne Sur Mer, France.
3 : Blueyou Consulting Ltd, CH-8003 Zurich, Switzerland.
4 : Wageningen Univ, Aquaculture & Fisheries Grp, NL-6700 AH Wageningen, Netherlands.
|Source||Oecologia (0029-8549) (Springer), 2013-07 , Vol. 172 , N. 3 , P. 631-643|
|WOS© Times Cited||16|
|Keyword(s)||Bioenergetics growth model, Individual growth trajectory, Life-history trade-offs, Energy acquisition, Maintenance, Reproductive investment, Sexual maturation|
|Abstract||A new method is presented to estimate individuals’ (1) age at maturation, (2) energy acquisition rate, (3) energy expenditure for body maintenance, and (4) reproductive investment, and the multivariate distribution of these traits in a population. The method relies on adjusting a conceptual energy allocation model to individual growth curves using nonlinear mixed-effects modelling. The method’s performance was tested using simulated growth curves for a range of life-history types. Individual age at maturation, energy acquisition rate and the sum of maintenance and reproductive investment rates, and their multivariate distribution, were accurately estimated. For the estimation of maintenance and reproductive investment rates separately, biases were observed for lifehistories with a large imbalance between these traits. For low reproductive investment rates and high maintenance rates, reproductive investment rate estimates were strongly biased whereas maintenance rate estimates were not, the reverse holding in the opposite situation. The method was applied to individual growth curves back-calculated from otoliths of North Sea plaice (Pleuronectes platessa) and from scales of Norwegian spring spawning herring (Clupea harengus). For plaice, maturity ogives derived from our individual estimates of age at maturation were almost identical to the maturity ogives based on gonad observation in catch samples. For herring, we observed 51.5 % of agreement between our individual estimates and those
directly obtained from scale reading, with a difference lower than 1 year in 97 % of cases. We conclude that the method is a powerful tool to estimate the distribution of correlated life-history traits for any species for which individual growth curves are available.