Latitudinal variation in breeding time reaction norms in a passerine bird
Article first published online: 8 APR 2010
© 2010 The Authors. Journal compilation © 2010 British Ecological Society
Journal of Animal Ecology
Volume 79, Issue 4, pages 836–842, July 2010
How to Cite
Gienapp, P., Väisänen, R. A. and Brommer, J. E. (2010), Latitudinal variation in breeding time reaction norms in a passerine bird. Journal of Animal Ecology, 79: 836–842. doi: 10.1111/j.1365-2656.2010.01693.x
- Issue published online: 7 JUN 2010
- Article first published online: 8 APR 2010
- Received 30 April 2009; accepted 11 March 2010Editor: Simon Verhulst
- phenotypic plasticity;
- avian breeding time;
- geographical variation;
- proportional hazards model
1. The timing of phenological events, as flowering time in plants or migration and breeding time in animals, is related to environmental conditions, e.g. local ambient temperature, and varies considerably between years. The relationship between the relevant environmental condition(s) and phenology can be described by a reaction norm.
2. Apart from variation among years phenology also varies geographically and the timing of phenological events is generally later at higher latitudes. However, whether this latitudinal trend is caused by a single reaction norm (and locally varying conditions) or whether reaction norms differ spatially is largely unknown.
3. Avian breeding time is determined by a variety of internal and external factors, e.g. photoperiod and temperature. The corresponding reaction norms are normally described by a regression of annual mean against temperatures averaged over a certain period. We circumvented problems inherent in this approach by using proportional hazards models to describe the relationship between breeding time and local temperature and day length.
4. The proportional hazards model describes a ‘time to event’, e.g. survival or in this case time until laying of the first egg, by modelling the probability that the event will occur per unit time as a function of an unspecified baseline hazard and fixed or time-dependent variables. Any variable that changes its value during the ‘time to event’, e.g. ambient temperature, can be included as time-dependent variable.
5. Applying this approach to a large data set of Finnish great tit (Parus major) egg laying dates from populations spanning about 700 km in latitude, we found that temperature- and day length-reaction norms vary with latitude.
6. Optimal reaction norms likely differ between populations and the observed variation in reaction norms among populations may thus reflect local adaptation. Under environmental change, local reaction norms may become suboptimal, and latitude-specific breeding time reaction norms may represent a source of variation that could benefit a species to adapt to such changes.