Increasing the resolution of global change gradients along which traits are measured
Species traits are often measured along major environmental gradients (i.e. elevation or latitude), with measurements focused on capturing intraspecific variation (Chalmandrier et al. 2017), interspecific variation (Balasubramaniam & Rottenberry, 2016), or both (Classen et al. 2017). Such gradients usually encompass a broad range of environmental factors that can be static or dynamic across space and time (i.e. water temperature, acidity, soil quality, wind, etc.). Constructing trait-based predictions requires ecologists to identify at what scale specific traits metrics change in response to particular aspects of the gradient of interest, yet trait variation is often not explicitly linked to measures of important aspect of the gradient at the same resolution (e.g. environmental data collected at the region or site level, while trait data collected at the individual level). Designing field data collection with trait and environmental data sets gathered at equal resolutions allows ecologists to quantify the scale at which variation in both response and explanatory variables matters for the environmental filtering process(s) under investigation (i.e. Figure 1). Advanced multivariate techniques (e.g. fourth corner an RQL analysis) can be used to parse out relationships between traits and different aspects of environmental gradients, as well as the spatial and temporal scale at which these relationships hold, subsequently allowing for the identification of potential ecological mechanisms driving trait and phylogenetic patterns across land and seascapes (Anderson et al. 2019). Without explicit links between directional trait variation and the aspect(s) of the environment driving that change, ecologists may continue to miss likely causal relationships that are useful for prediction.