Species diversity and functional diversity relationship varies with disturbance intensity

The relationship between species diversity and functional diversity is a fundamental topic in conservation ecology but its direct evaluation in naturally heterogeneous systems and its relationship with disturbance intensity is limited. We examined the relationship between taxonomy-based species diversity (in terms of richness and evenness) and trait-based functional diversity of riparian and upland vegetation and determined if the slopes of this relationship vary with disturbance intensity. We used data from riparian and upland plant communities subjected to three categorical disturbance intensities (low = uncut mature forests, intermediate = riparian buffers around clearcuts and riparian buffers around clearcuts + soil scarification, and high = clearcuts and clearcuts + soil scarification) along 24 small streams of boreal forests in northwestern Ontario, Canada. For all disturbance intensities and habitat types (riparian vs. upland), the relationships between species diversity and functional diversity were positive and non-zero, but the slopes of these relationships varied significantly with disturbance intensities. In species richness vs. functional richness relationship, the slopes were steep at low and high disturbances and relatively less steep at moderate disturbance; however, patterns were less clear in species evenness vs. functional evenness relationship. Differential distributions of species richness per trait were, in part, associated with this variation in slope. In richness (opposite for evenness), less steep slopes were associated with relatively uniform distribution of species richness per trait compared to non-uniform distribution at steep slopes, where competitive (low disturbance) and disturbance-tolerant (high disturbance) traits dominated. Our results confirm that species diversity is a good indicator of ecosystem stability but it depends on ecosystem disturbance history. More importantly, this study suggests that different components of species diversity (e. g., richness and evenness) and the distribution of species richness per trait together can summarize the stability of an ecosystem more accurately than prediction made from species diversity alone.