Body mass dynamics of migratory nightjars are explained by individual turnover and fueling

Abstract

Abstract Body mass is a commonly used indicator of the energy stores of migratory animals and there is considerable evidence that it is a critical determinant of migration decisions and outcomes. Mean population mass often increases during the post-breeding period in most migratory species. Usually, this increase is interpreted as the result of fuel accumulation for migration based on the assumption that mean population mass mirrors mean individual mass. However, an empirical test of this assumption is lacking, and it is unknown whether the general increase in mean population mass is entirely the result of within-individual mass gain, or if it rather reflects a change in the nature of individuals in the population (mass-dependent turnover). We investigated changes in body mass during the post-fledging period of a migratory bird, the Red-necked Nightjar (Caprimulgus ruficollis), and combined longitudinal and cross-sectional data collected over 9 years to disentangle the relative contribution of individual-level (mass gain) and population-level (selective appearance and disappearance) processes. We found that the average body mass of fully-developed juveniles increased as the season progressed and that both individual mass gain and the selective disappearance of lighter individuals contributed to this increase. Contrary to the general expectations for migrants, the turnover of individuals contributed 3.5 times more to the seasonal increase in average body mass than individual mass gain. On a practical note, this differential contribution implied a discrepancy of over 40% between the time-average rates of mass gain (fuel deposition rates) estimated from population-level and individual-level data. Our study calls for caution in the use of population-level changes in body mass to make inferences about individual fuel deposition rates and, more generally, indicates that longitudinal and cross-sectional approaches need to be combined to uncover phenotype-time correlations in natural populations.