LIFE HISTORY DATA SOURCES
Quality data for quantifying life histories takes much effort to collect
as life histories cannot be described simply by observing static
phenotypes. Measuring life history traits can be challenging: some
infrequent, brief life history events can be at best rarely observed.
Examples include reproduction in semelparous perennial species, or
mortality in long-lived species. Much effort has been devoted to the
inference of unobserved or imperfectly detected life history events in
population studies [35]. The demographic context of traits like
generation time and life expectancy means that they must be measured
using samples of individuals within populations or species. Hence, life
history trait data vary greatly in their abundance, accessibility, and
quality (Table 1) [12,36], even among charismatic species [12].
The gold standard in demographic data is detailed schedules of age- or
stage-specific vital rates . These rates of survival,
development, and reproduction are becoming increasingly available
through open databases of demographic data (e.g.,
[10,11,16–18,31,37,38]). Despite the diversity of model dimensions
and complexity, life history traits can be derived from these schedules,
yielding a set of common traits across species. This derivation
overcomes some of the issues with simple trait data outlined in Table 1.
Demographic quantities such as stationary [39] and transient
[40] population dynamics can also be calculated from life history
schedules, as can selection pressures [41]. However, although these
databases are accessible, models vary in quality and data remain
taxonomically biased [36]. Moreover, the traits derived from these
models are mathematical functions of the same underlying schedules of
survival, development, and reproduction, hence correlations among them
are a combination of the real and the artefactual [42].