Intraspecific variability (IV) has been proposed as a new track to explain species coexistence. Previous studies generally assumed that IV results from intrinsic differences between conspecifics that widen species’ fundamental niches and blur differences among species, thus impeding stable coexistence, but also slowing down the rate of competitive exclusion. Based on a body of evidence, we here argue that IV does not necessarily imply differences among conspecifics, nor species niches overlap: conspecifics differ in their measured attributes mainly due to differences in the micro-environment they thrive in. Consequently, they respond more similarly to environmental variation than heterospecifics, thereby concentrating competition within species – a necessary condition for species coexistence. We call for new studies exploring observed IV as an outcome of species-specific responses to high-dimensional environmental variations that can lead to inversions of species hierarchy in space and time promoting stable coexistence.

Ecological theory aims to understand how and why species differences allow competitors to coexist, but explanations remain inconsistent with data. Tightly constrained parameter tradeoffs needed for coexistence in models contrast with evidence that forests can support high diversity and be invaded repeatedly by species that lack specialized tradeoffs. By translating environmental responses to individual covariance, a universal coexistence hypothesis shows i) that species differences lead to a natural tendency to concentrate competition within the species, the common feature needed to promote coexistence in models, and ii) the fingerprint of this effect is available in covariances between individuals that can be observed in nature. The many ways in which species differ make high diversity almost inevitable. This covariance not only provides the evidence for this mechanism; it further provides a new direction for earth surface models that currently cannot sustain diverse communities despite large numbers of evidence-based parameters.

Community composition is a primary determinant of how biodiversity change influences ecosystem functioning and, therefore, the relationship between biodiversity and ecosystem functioning (BEF). We examine the consequences of community composition across six structurally realistic plant community models. We find that a positive correlation between species’ functioning in monoculture vs. their dominance in mixture with regards to a specific function (the “function-dominance correlation”) generates a positive relationship between realized diversity and ecosystem functioning across species richness treatments. However, because realised diversity declines when few species dominate, a positive function-dominance correlation generates a negative relationship between realized diversity and ecosystem functioning within species richness treatments. Removing seed inflow strengthens the link between the function-dominance correlation and BEF relationships across species richness treatments but weakens it within them. These results suggest that changes in species’ identities in a local species pool may more strongly affect ecosystem functioning than changes in species richness.

Understanding the processes that shape forest functioning, structure and diversity remains challenging, although an increasing amount of data documents forest systems across scales. Forest models have a long history in assimilating various data and ecological knowledge and can simulate forest dynamics over spatio-temporal scales unreachable by most empirical investigations. Here we describe the trajectories of development different forest modelling communities have followed to demonstrate the leverage that computer models offer for advancing the understanding of forest ecosystems. Using three widely applied but contrasting forest modelling approaches - species distribution models, individual-based models and dynamic global vegetation models - as examples, we show how scientific and technical advances have led models beyond their initial objectives and limitations. We provide an overview of recent model applications on current important ecological topics and pinpoint ten key questions that could, and should, be tackled with forest models in the next decade. This shows that forest models, due to their long history of assimilating empirical knowledge, their iterative and continuous development, and their complementarity, represent an invaluable toolkit to address a wide range of theoretical and applied ecological questions, hence fostering a deeper understanding of forest dynamics, particularly in the context of global change.