Explaining rarity: Process
A key challenge in ecology is to identify generalities that transcend the idiosyncrasies of specific focal assemblages or regions. Through the use of traits (i.e., morphological, physiological, phenological, behavioural, biochemical, or other characteristics of species that influence their fitness), functional ecology offers a taxonomically-neutral approach to inferring information about ecological processes from species’ functional traits . Similarly, traits may offer a path towards identifying the key processes that drive different rarity dimensions. However, despite the large body of literature that has sought links between traits and rarity (Table 2), identifying generalities has proven difficult for a variety of reasons. First, rarity is defined differently across studies, e.g., species may be considered rare based on range size , conservation status , or frequency of occurrence within a study region . The traits associated with such different types of rarity can be expected to vary substantially. A second challenge is the large diversity of taxa and traits covered in studies focussing on trait–rarity relationships: the traits related to rarity may vary substantially among taxonomic groups owing to differences in physiology and life history. In addition, species exhibit numerous traits, not all of which will necessarily impact or even correlate with their degree of commonness or rarity. Furthermore, traits may interact within a single individual or species, and the effects of traits that promote rarity may be offset by other traits that promote commonness. Finally, studies on traits naturally focus on those traits for which we have adequate data, and it is unclear to what degree the available data correspond to the key drivers of rarity .
Owing to the difficulties in finding consensus among studies of the traits associated with rarity, we propose a top-down approach linking rarity to traits. Instead of inferring trait–rarity relationships by synthesising the available evidence, we instead propose a suite of trait dimensions that can be mechanistically linked to the three proposed rarity dimensions (i.e., range size, occupancy, and local abundance).