The evolution of seasonal migration and the slow-fast continuum of life history in birds

Abstract

Seasonal migration is a widespread adaptation in environments with predictable periods of resource abundance and scarcity. Migration is frequently associated with high mortality, suggesting that migratory species live on the “fast” end of the slow-fast continuum of life history. However, few interspecific comparative studies have tested this assumption and prior assessments have been complicated by environmental variation among breeding locations. We evaluate how migration distance influences the tradeoff between reproduction and survival in 45 species of mostly passerine birds that breed sympatrically in North American boreal forests but migrate to a diversity of environments and latitudes for the northern winter. We find, after accounting for mass and phylogeny, that longer distance migrations to increasingly amenable winter environments are correlated with reduced annual reproductive output, but also result in increased adult survival compared to shorter-distance migrations. Non-migratory boreal species have life history parameters more similar to long-distance migrants than to shorter-distance migrants. These results suggest that long-distance migration and other highly specialized strategies for survival in seasonal environments impose selection pressures that both confer and demand high adult survival rates. That is, owing to the reproductive cost of long-distance migration, this strategy can only persist if balanced by high adult survival. Our results reveal migratory distance as a fundamental life history parameter that predicts, and is inextricable from, the balance of survival and reproduction. Our study further provides evolutionary context for understanding the annual cycle demography of migratory species and the strategies long-distance migrants use to maximize survival on their journeys.

Natural selection and circular pathways to seasonal migration in birds

“…no point sources, see Fig. 1) and short-distance migrants do not pay the cost of lower annual fecundity (14). As predicted, the short-distant migrant C. guttatus has intermediate W/L morphology in both sexes, relative to congeners with sedentary (lower W/L) and long-distance and triple migrant (higher W/L) phenotypes (Fig.…”

“…A recent wave of research has greatly expanded the discussion, shining light on the polygenic nature of the migratory phenotype (e.g. 6,7), functional links between behavior and morphology (8,9), the sources and targets of selection (10)(11)(12)(13)(14)(15)(16), fundamental tradeoffs between fecundity and survival (14), and the heterobathmic distribution of behavioral and morphological components of the migratory phenotype among populations and species (5,6,17). That the expression of migratory behavior originates (in sedentary populations) as an evolutionary response to high mortality during the resource-depleted non-breeding season (11)(12)(13)(14) is an alternative hypothesis that has recently gained traction over the traditional view of migration as an adaptation by which tropical species increase fecundity by exploiting seasonal (pulsed) food resources in temperate regions (18,19).…”

“…Here, I advance a synthetic theory of the evolutionary sequence(s) that produced the observed spectrum of migratory behavior in birds. The "migratory revolutions" (MR) model conceptually integrates morphology and behavior (20,21) and presumes that extreme optimization for long-distance migration (14,22,23) constrains the directionality of evolutionary change. The model explains (predicts) the complete spectrum of the migratory phenotype, from sedentary residents of non-seasonal habitats, to obligate long-distance migrants that also practice short-distance ("intratropical") migration during the non-breeding season (24,25).…”

“…Migratory behavior and its functional morphology are polygenic traits with extensive phenotypic integration, that presumably evolve in tandem as a single "functional phenotype" (20,21). Embryonic growth rates do not differ between migratory and non-migratory taxa (14), and forelimbs (wings, W) and hindlimbs (legs, L) are ontogenetically independent and achieve adult size within the first month after hatching. Notwithstanding this independence, there is an inverse relationship between W and L investment during development, presumably because of performance tradeoffs, which produces, at macroevolutionary scales, a negative relationship between wing length and tarsometatarsus length (26).…”

“…Long-distance migration evolves from short-distance migration in response to a different, persistent selective force: additive migration-related mortality (e.g. crossing the Caribbean Sea during the Atlantic hurricane season, 16), which optimizes the functional phenotype for annual adult survival at the cost of annual fecundity (14). In the model, longdistance migrants can only "escape" this positive feedback loop by forming colonies.…”