Estimates of demographic relationships are needed to evaluate allocation tradeoffs and how species differences contribute to biodiversity. To quantify variation within individuals, among individuals, and among species and its contribution to life history tradeoffs, demographic heterogeneity, and biodiversity, we constructed models of interacting life history schedules and fitted them to long-term experimental and monitoring data on tree population dynamics. Spatio-temporal data include increment cores, seed traps, canopy and reproductive status, remote sensing of canopy architecture, light, soil moisture, and tree, seedling, and sapling censuses. Data derive from mapped plots with superimposed treatments that include herbivore exclosures, gap creation, and CO2 fumigation, all applied as intervention designs over multiple years. A hierarchical model is used to estimate the interactions among life history traits and to predict life history schedules for all dominant canopy species in nine stands from the Piedmont Plateau and southern Appalachians of North Carolina that include a range of soils, hydrologic conditions, and elevations. Covariates are included with random effects taking up additional variation at individual, plot, or stand level. We find large differences among species in many traits that only become apparent from estimates of latent variables, which can be represented as predictive distributions. Although many species appear to have similar demographic rates when considered along only a few resource or environmental axes, underlying interactions differ substantially. Species differ further in terms of the extent to which life history schedules interact over time. While some species show clear year-to-year correlations in growth and fecundity, others do not. Given the high dimensionality of species differences, data showing no differences among species is best taken as ‘no evidence', rather than ‘evidence against' niche differences.