Do community-weighted mean functional traits reflect optimal strategies?

Abstract

The notion that relationships between community-weighted mean (CWM) traits (i.e. plot-level trait values weighted by species abundances) and environmental conditions reflect selection towards locally optimal phenotypes is challenged by the large amount of interspecific trait variation typically found within ecological communities. Reconciling these contrasting patterns is a key to advancing predictive theories of functional community ecology. We combined data on geographical distributions and three traits (wood density, leaf mass per area and maximum height) of 173 tree species in Puerto Rico. We tested the hypothesis that species are more likely to occur where their trait values are more similar to the local CWM trait values (the'CWM-optimality' hypothesis) by comparing species occurrence patterns (as a proxy for fitness) with the functional composition of forest plots across a precipitation gradient. While 70% of the species supported CWM-optimality for at least one trait, nearly 25% significantly opposed it for at least one trait, thereby contributing to local functional diversity. The majority (85%) of species that opposed CWM-optimality did so only for one trait and few species opposed CWM-optimality in multivariate trait space. Our study suggests that constraints to local functional variation act more strongly on multivariate phenotypes than on univariate traits.

Keywords: ecological niche models; functional diversity; leaf mass per area; maximum height; tropical forests; wood density.

Figure 1.

(a) Variation of CWM traits suggests selection towards a locally optimal trait value (inward-pointing arrows). However, local functional diversity implies opposing mechanisms that promote local functional diversity (outward-pointing arrows). We used ENMs to test the hypothesis that species would be less likely to occur in areas where their trait values are more distant from the local CWM values (the ‘CWM-optimality hypothesis’). See online version for colour legend. for each species and trait, we calculated ΔCWM as the absolute difference between the species-mean trait value and the CWM value at each study plot. (b) Summarized predictions of the CWM-optimality hypothesis. (Online version in colour.)

Figure 2.

Variation in (a) WD, (b) LMA, (c) and H_max along a gradient of mean annual precipitation for 12 forest plots. Large circles represent CWM trait values, smaller points represent trait values of species that occurred in each plot. Trend lines and R²-values correspond to regressions of CWM values and mean annual precipitation. (Online version in colour.)

Figure 3.

Panels (a–d) show histograms of species-specific slopes for the regression between ΔCWM values and ROR. Light grey bars show slopes for all species; statistically significant slopes (i.e. p < 0.05) are darkened (negative and positive slopes are orange and blue in the online version of the figure, respectively). Panels (e–h) show species-specific slope values (as in a–d) plotted against their value of precipitation at ROR_max. In (e–g), point colours represent species trait values (see legends) and species with non-significant slopes are grey without black circles. In (h), species with significant slopes are shown with black circles. Units of slope values are ROR divided by 1 s.d. of the relevant trait. (Online version in colour.)

Figure 4.

Barplot with the proportion (and number) of study species (N = 173) with significantly negative and positive regression slopes between ΔCWM and ROR for different numbers of univariate trait axes, and in multivariate trait space. Because some species exhibited both positive and negative slopes for different trait axes, the sum of the numbers of species shown is greater than the total number of study species. (Online version in colour.)