One of ecology's grand challenges is developing general rules to explain and predict highly complex systems. Understanding and predicting ecological processes from species' traits has been considered a 'Holy Grail' in ecology. Plant functional traits are increasingly being used to develop mechanistic models that can predict how ecological communities will respond to abiotic and biotic perturbations and how species will affect ecosystem function and services in a rapidly changing world; however, significant challenges remain. In this review, we highlight recent work and outstanding questions in three areas: (i) selecting relevant traits; (ii) describing intraspecific trait variation and incorporating this variation into models; and (iii) scaling trait data to community- and ecosystem-level processes. Over the past decade, there have been significant advances in the characterization of plant strategies based on traits and trait relationships, and the integration of traits into multivariate indices and models of community and ecosystem function. However, the utility of trait-based approaches in ecology will benefit from efforts that demonstrate how these traits and indices influence organismal, community, and ecosystem processes across vegetation types, which may be achieved through meta-analysis and enhancement of trait databases. Additionally, intraspecific trait variation and species interactions need to be incorporated into predictive models using tools such as Bayesian hierarchical modelling. Finally, existing models linking traits to community and ecosystem processes need to be empirically tested for their applicability to be realized.

Predicting changes in community composition and ecosystem function in a rapidly changing world is a major research challenge in ecology. Traits-based approaches have elicited much recent interest, yet individual studies are not advancing a more general, predictive ecology. Significant progress will be facilitated by adopting a coherent theoretical framework comprised of three elements: an underlying trait distribution, a performance filter defining the fitness of traits in different environments, and a dynamic projection of the performance filter along some environmental gradient. This framework allows changes in the trait distribution and associated modifications to community composition or ecosystem function to be predicted across time or space. The structure and dynamics of the performance filter specify two key criteria by which we judge appropriate quantitative methods for testing traits-based hypotheses. Bayesian multilevel models, dynamical systems models and hybrid approaches meet both these criteria and have the potential to meaningfully advance traits-based ecology.

[1] We conducted measurements of methane (CH 4 ) emission and ecosystem respiration on >200 points across the Arctic coastal tundra near Barrow, Alaska, United States, in July 2007 and August 2008. This site contains broad diversity in tundra microtopography, including polygonal tundra, thaw lakes, and drained lake basins. In 2007, we surveyed CH 4 emissions across this landscape, and found that soil water content was the strongest control of methane emission rate, such that emission rates rose exponentially with water content. However, there was considerable residual variation in CH 4 emission in the wettest soils (>80% volumetric water content) where CH 4 emissions were highest. A statistical analysis of possible soil and plant controls on CH 4 emission rates from these wet soils revealed that vegetation height (especially of Carex aquatilis) was the best predictor, with ecosystem respiration and permafrost depth as significant copredictors. To evaluate whether plant height served as a proxy for aboveground plant biomass, or gross primary production, we conducted a survey of CH 4 emission rates from wet, Carex-dominated sites in 2008, coincidently measuring these candidate predictors. Surprisingly, vegetation height remained the best predictor of CH 4 emission rates, with CH 4 emissions rising exponentially with vegetation height. We hypothesize that taller plants have more extensive root systems that both stimulate more methanogenesis and conduct more pore water CH 4 to the atmosphere. We anticipate that the magnitude of the climate change-CH 4 feedback in the Arctic Coastal Plain will strongly depend on how permafrost thaw alters the ecology of Carex aquatilis.

Recent studies have increasingly turned to graph theory to model more realistic contact structures that characterize disease spread. Because of the computational demands of these methods, many researchers have sought to use measures of network structure to modify analytically tractable differential equation models. Several of these studies have focused on the degree distribution of the contact network as the basis for their modifications. We show that although degree distribution is sufficient to predict disease behaviour on very sparse or very dense human contact networks, for intermediate density networks we must include information on clustering and path length to accurately predict disease behaviour. Using these three metrics, we were able to explain more than 98 per cent of the variation in endemic disease levels in our stochastic simulations.

One contribution of 17 to a theme issue 'Biodiversity and ecosystem functioning in dynamic landscapes'. The importance of intraspecific trait variability for community dynamics and ecosystem functioning has been underappreciated. There are theoretical reasons for predicting that species that differ in intraspecific trait variability will also differ in their effects on ecosystem functioning, particularly in variable environments. We discuss whether species with greater trait variability are likely to exhibit greater temporal stability in their population dynamics, and under which conditions this might lead to stability in ecosystem functioning. Resolving this requires us to consider several questions. First, are species with high levels of variation for one trait equally variable in others? In particular, is variability in response and effects traits typically correlated? Second, what is the relative contribution of local adaptation and phenotypic plasticity to trait variability? If local adaptation dominates, then stability in function requires one of two conditions: (i) individuals of appropriate phenotypes present in the environment at high enough frequencies to allow for populations to respond rapidly to the changing environment, and (ii) high levels of dispersal and gene flow. While we currently lack sufficient information on the causes and distribution of variability in functional traits, filling in these key data gaps should increase our ability to predict how changing biodiversity will alter ecosystem functioning.