Above- and below-ground functional trait coordination in the Neotropical understory genus Costus

Abstract

The study of plant functional traits and variation among and within species can help illuminate functional coordination and trade-offs in key processes that allow plants to grow, reproduce and survive. We studied 20 leaf, above-ground stem, below-ground stem and fine-root traits of 17 Costus species from forests in Costa Rica and Panama to answer the following questions: (i) Do congeneric species show above-ground and below-ground trait coordination and trade-offs consistent with theory of resource acquisition and conservation? (ii) Is there correlated evolution among traits? (iii) Given the diversity of habitats over which Costus occurs, what is the relative contribution of site and species to trait variation? We performed a principal components analysis (PCA) to assess for the existence of a spectrum of trait variation and found that the first two PCs accounted for 21.4 % and 17.8 % of the total trait variation, respectively, with the first axis of variation being consistent with a continuum of resource-acquisitive and resource-conservative traits in water acquisition and use, and the second axis of variation being related to the leaf economics spectrum. Stomatal conductance was negatively related to both above-ground stem and rhizome specific density, and these relationships became stronger after accounting for evolutionary relatedness, indicating correlated evolution. Despite elevation and climatic differences among sites, high trait variation was ascribed to individuals rather than to sites. We conclude that Costus species present trait coordination and trade-offs that allow species to be categorized as having a resource-acquisitive or resource-conservative functional strategy, consistent with a whole-plant functional strategy with evident coordination and trade-offs between above-ground and below-ground function. Our results also show that herbaceous species and species with rhizomes tend to agree with trade-offs found in more species-rich comparisons.

Keywords: Ecophysiology; functional strategies; rhizome traits; specific root length; stem specific density; tropics; variance component analysis.

Figure 1.

Map of Costa Rica and Panama showing the geographic location of the eight field sites. Shading corresponds to elevation (m asl).

Figure 2.

Principal components analysis (PCA) biplot of the studied functional traits. Groupings denote species with resource-acquisition or resource-conservation strategies. Species are abbreviated as shown in Table 1. Chl: chlorophyll concentration; g_s: stomatal conductance; LT: leaf thickness; LA: leaf area; LDMC: leaf dry matter content; SLA: specific leaf area; LM:PM: lamina dry mass to petiole dry mass ratio; LA:PM: leaf area to petiole dry mass ratio; P: leaf phosphorus concentration; K: leaf potassium concentration; δ ¹³C: leaf carbon isotopic composition; C: leaf carbon concentration; δ ¹⁵N: leaf nitrogen isotopic composition; N: leaf nitrogen concentration; SSD: stem specific density; RhWC: rhizome water content; RhSD: rhizome specific density; SRL: specific root length; FRD: fine-root diameter; RTD: root tissue density.

Figure 3.

Correlation plots showing significant correlations only (P < 0.05). (A) Cross-species correlations. (B) Correlations using phylogenetic contrasts. Chl: chlorophyll concentration; g_s: stomatal conductance; LT: leaf thickness; LA: leaf area; LDMC: leaf dry matter content; SLA: specific leaf area; LM:PM: lamina dry mass to petiole dry mass ratio; LA:PM: leaf area to petiole dry mass ratio; P: leaf phosphorus concentration; K: leaf potassium concentration; δ ¹³C: leaf carbon isotopic composition; C: leaf carbon concentration; δ ¹⁵N: leaf nitrogen isotopic composition; N: leaf nitrogen concentration; SSD: stem specific density; RhWC: rhizome water content; RhSD: rhizome specific density; SRL: specific root length; FRD: fine-root diameter; RTD: root tissue density.

Figure 4.

(A) Cross-species correlation between stomatal conductance and stem specific density and (B) between stomatal conductance and rhizome specific density. (C) Correlation between stomatal conductance contrasts and stem specific density contrasts, and (D) stomatal conductance contrasts and rhizome specific density contrasts. Contrasts were calculated as the difference between trait values of sister species divided by branch length. Trend line is included when correlations were significant. Species are abbreviated as shown in Table 1.

Figure 5.

Variance component analysis of the traits studied. We partitioned the total trait variation into three levels: sites, species and individuals. Chl: chlorophyll concentration; g_s: stomatal conductance; LT: leaf thickness; LA: leaf area; LDMC: leaf dry matter content; SLA: specific leaf area; LM:PM: lamina dry mass to petiole dry mass ratio; LA:PM: leaf area to petiole dry mass ratio; P: leaf phosphorus concentration; K: leaf potassium concentration; δ ¹³C: leaf carbon isotopic composition; C: leaf carbon concentration; δ ¹⁵N: leaf nitrogen isotopic composition; N: leaf nitrogen concentration; SSD: stem specific density; RhWC: rhizome water content; RhSD: rhizome specific density; SRL: specific root length; FRD: fine-root diameter;, RTD: root tissue density.