Intraspecific functional trait variation and coordination in Schizachyrium scoparium

Plant functional traits are increasingly recognized as vital tools in ecological restoration and biodiversity conservation. While functional traits and functional diversity are increasingly being used to inform restoration efforts, challenges remain in the characterization of trait variation in many systems, including within-species. This dataset supports the investigation of intraspecific trait variation and coordination among five populations of Schizachyrium scoparium (little bluestem), a species commonly used for restoration, from different habitat types across a gradient from southern Wisconsin to Northern Illinois. The dataset contains leaf, root, and chemical functional trait measurements recorded at the individual plant level as well as identifying information associating trait measurements with individual plant genotype and population membership. In addition to the data, this repository contains an R script demonstrating the analytical workflow used to model trait variation and coordination in the system.

Methods
Trait values contained in this repository reflect measurments from plants that were germinated and propagated in vitro using tissue culture techniques (see manuscript for details). All plants were harvested for trait measurement after growing for 13 weeks ex vitro. Plants were removed from their pots, and the growing media was gently washed from the root systems. The above-ground and below-ground tissues were separated at the crown, and the root systems were temporarily wrapped in damp paper towels, stored in plastic bags, and refrigerated for subsequent root scanning. Five fully expanded leaves were randomly chosen from each plant for leaf trait measurements and removed at the leaf collar, taking only the lamina and leaving the sheath tissue behind. The selected leaves were scanned at 600 dpi, and the surface area (mm2) of each leaf was calculated using ImageJ software (Schneider et al. 2012). The selected leaves were then weighed to retrieve the fresh leaf mass (g), dried at 60°C for 72 hours, and re-weighed to retrieve the dry leaf mass (mg). With these measurements, we calculated specific leaf area (SLA) as the ratio of leaf area to leaf dry mass (mm2/mg) and leaf dry matter content (LDMC) as the ratio of leaf dry mass to leaf fresh mass (mg/g). The resulting five leaf trait values for SLA and LDMC were then averaged within each individual. After weighing, the dried leaf samples were pooled per plant and sent to the Danforth Plant Science Center in St. Louis, MO for chemical analysis where % N content was measured via combustion by an elemental analyzer (Elementar vario ISOTOPE cube).

The cleaned root system of each plant was individually scanned at 600 dpi using an Epson Expression 10000XL large-format flatbed scanner with a transparency attachment, following the protocol from (York 2023). The root systems were floated in a 300mm x 420 mm x 20 mm acrylic box filled with ~400 ml of water for scanning (York 2020). The entire root system of each plant was scanned, though in some cases the root systems needed to be sectioned to fit in the scanning area and ensure that the roots remained submerged. Minor edits were made to the root images to remove the borders of the acrylic box and any shadows from partially submerged roots using the open source GIMP software (v. 2.10; (The GIMP Development Team 2019). Following the scanning procedure, roots were patted dry to remove surface moisture and weighed to retrieve fresh root weight (g). The root samples were then dried at 60°C for 72 hours and weighed to obtain dry root mass (mg). These measurements were used to calculate root dry matter content (RDMC) as the ratio of dry root biomass to fresh root biomass (mg/g).

The root scan images were analyzed using the “broken roots” analysis method in RhizoVision explorer (v. 2.0.2; (Seethepalli and York 2020). Various settings were tested to analyze root images, and segmented images were previewed to assess accuracy. Following testing, all root images were analyzed using the maximum recommended pruning threshold of 20, a non-root object filter of 1, edge-smoothing disabled, and an image threshold of 180 to produce the clearest root skeletonization. We extracted total root length (m) and average root diameter (mm) from the RhizoVision analyses and calculated specific root length (SRL, m/g) for each plant by dividing the total root length (m) by the dry root mass (g).

References: 

Schneider, C. A., W. S. Rasband, and K. W. Eliceiri. 2012. NIH Image to ImageJ: 25 years of image analysis. Nature Methods 9:671–675.

Seethepalli, A., and L. M. York. 2020, October. RhizoVision Explorer - Interactive software for generalized root image analysis designed for everyone. Zenodo.

The GIMP Development Team. 2019, June 12. GIMP.

York, L. 2020, October. Plans for root scanning trays to use on flatbed scanners. Zenodo.

York, L. 2023. Root scanning using a flatbed scanner with transparency unit.