1. Recent research has highlighted the existence of significant intraspecific trait variation within and among populations of plant species. This inherent variation within species means there is a wealth of trait diversity from which to source germplasm for use in ecological restoration. However, it remains unclear how to source materials from this pool of trait diversity in order to achieve desired outcomes and support ecosystem function in a restoration context.
2. We provide a framework to study the structure of trait variation across populations and genotypes in an effort to bridge functional trait research with the sourcing of native plant materials for restoration. We investigated the structure of intraspecific functional trait variation in three forb species used in restoration on the Colorado Plateau to (i) understand the structure of functional trait variation within and among populations, and (ii) determine if individual and multivariate functional traits differ between populations while accounting for trait variation within and among genotypes.
3. We found considerable functional trait variation at all sampling levels, with variation within populations often surpassing variation among populations. Still, we observed population-level differences in trait values in eight of the twelve species-by-trait combinations and populations largely segregated in multivariate trait space.
4. Synthesis and applications. Using micropropagation techniques, we uncovered population-level differences in functional trait variation, suggesting that mixing populations to create restoration germplasm following a regional admixture provenancing approach could lead to increased functional diversity in restorations. However, the substantial trait variation identified within some populations of our study species also suggests a similar potential when utilizing genotypically diverse material from even a single population. Further research on these and other species is needed to understand the structure of intraspecific functional trait variation and how it impacts ecosystem function. The approach outlined in this study can assist researchers in assessing the underlying trait variation present in various restoration materials and provide managers with more detailed information that can help make germplasm sourcing decisions.

Individual level trait data collected from micropropagated plants grown in a growth chamber. Individual plants are identified by species, population, genotype, and replicate (clone). Biomass measurements were manually collected with an analytic balance for each individual plant.  Measurements were collected for three leaves of each individual using leaf scans and ImageJ software (v.1.51, Schneider, Rasband, & Eliceiri, 2012).

This dataset includes all of the raw data needed to calculate the functional traits used in the study. Root-mass ratio was calculated as the ratio of dry below-ground biomass to total dry biomass. Specific leaf area (SLA) of each leaf was calculated as leaf area divided by leaf dry mass (mm²/mg) and leaf dry matter content (LDMC) was calculated as the ratio of dry to fresh leaf mass (mg/g). Leaf traits were calculated individually for each of three leaves harvested from each individual plant which were then averaged to provide a single value for each individual plant in subsequent analyses.