The Extent, Drivers, and Consequences of Intraspecific Variation in Plant Functional Traits
MetadataShow full item record
Classifying species and communities by their collection of functional traits (traits that indirectly or directly quantify plant fitness) rather than by species identity and richness alone is increasingly being used to identify global patterns in plant and ecosystem function, to identify abiotic filters which shape community composition, and to redefine plant strategies. Typically, studies using plant functional traits have relied on trait means or community level trait averages to quantify functional change. Recent research, however, suggests that intraspecific variation in plant functional traits is extensive, and may be of importance for a number of ecological processes. My dissertation research focuses on quantifying the extent, drivers, and consequences of intraspecific variation in plant functional traits. I conducted a study in Pacific Northwest grasslands which found that, after applications of carbon in the form of either sugar or activated carbon, perennial forbs, but not grasses, suffered significant decreases in cover and biomass. These results indicated that perennial forbs, as a functional group, were more susceptible to carbon addition; however some species of forb were less affected than others. One possible explanation for this difference was that some species exhibited greater variation in functional traits that were affected by carbon addition, and this variation may have allowed some individuals to pass unaffected through the treatment filter. To test this hypothesis, I developed a study to investigate the extent of intraspecific variation in two commonly measured plant functional traits thought to reflect growth strategy, specific leaf area (SLA) and rosette diameter, within Hypochaeris radicata, a globally distributed perennial herb. I simultaneously examined the extent of intraspecific variation in plant functional traits, and the efficacy of methods used in the literature to quantify this variation. I collected trait measurements on 241 individuals from 10 populations spanning a 480 km transect. I found significant variation in plant traits: there was a 27-fold variation in SLA, and a 34-fold variation in rosette diameter. Furthermore, the method of analysis used to quantify these traits affected the interpretation of these results, as inclusion of intraspecific variation may violate the underlying assumptions of some commonly applied statistical tests (e.g. ANOVA), and thus this variation is frequently reduced through corrective measures. Variation in plant traits can be driven by biotic, as well as abiotic factors. To determine whether and how plants shift their traits in response to biotic conditions, I measured SLA and plant height for a suite of 8 forbs in an experimental grassland. Traits were collected from individuals in one of three treatments: high forb cover, balanced grass/forb cover, or high grass cover. Using a combination of permutational analysis of variance and linear mixed models, I found that plants shifted traits in terms of both average expression and trait variation in response to the biotic community, but that the direction of the shift was highly idiosyncratic. In addition, the suite of plants shifted their overall niche breadth and niche spacing in both SLA and plant height to limit similarity in the face of high competition from dominant grasses. Based on the results of the previous experiments, I then set out to determine whether this large variation was beneficial to Hypochaeris radicata in stressful abiotic conditions, and whether the observed variation was due to plasticity or adaptation. I collected seeds from 3 populations (25 maternal lines from each) that differed in growing conditions and trait expression. Maternal lines also differed in their trait expression; using PCA, maternal lines were ranked by seed weight and rosette diameter, and lines with low, intermediate, and high trait values were selected for each treatment. Plants were grown in the greenhouse under three treatments: ambient, drought, or shade conditions. I collected trait data for SLA, rosette diameter, chlorophyll content, and leaf shape, as well as calculating relative growth rate and devotion to above- and below-ground tissues. Analyses of these data using hierarchical linear mixed models revealed that, for plant diameter, maternal line of origin contributed significantly to variation (30%), but for all other traits, population of origin and maternal line explained very little of the observed variation (<10%). Furthermore, trait variation in the field was not predictive of trait variability in the greenhouse. These results indicate that, although there may be local genetic adaptation, plant trait expression is largely driven by plasticity to prevailing abiotic conditions. In summary, I found that plant functional traits are highly variable and can respond to both abiotic and biotic drivers. In addition, I found that for H. radicata, variation in plant traits was largely driven by plasticity.
- Forestry