Research location description:
To measure temporal trends in leaf traits, we sampled 14 specimens each of D. oligosanthes and P. virgatum at the Kansas State University Herbarium (KSC) and 17 specimens each at the Ronald L. McGregor Herbarium at the University of Kansas (KANU). For the years 2021 and 2022, plants were collected in the field at Konza Prairie Biological Station, Kansas and pressed and dried before sampling.

To compare how D. oligosanthes and P. virgatum leaf traits vary across grasslands of the Great Plains of the United States, we sampled individuals from eight sites over the summers of 2021 and 2022: (1) Woodworth Station Waterfowl Production Area, North Dakota, (2) Cedar Creek Ecosystem Science Reserve, Minnesota, (3) Valentine National Wildlife Refuge, Nebraska, (4) T. L. Davis Preserve, Nebraska (5) Kish-Ke-Kosh Prairie, Iowa, (6) Konza Prairie Biological Station, Kansas, (7) Wah’Kon-Tah Prairie, Missouri, and (8) Joseph H. Williams Tallgrass Prairie Preserve, Oklahoma.

At each grassland site, one to nine replicates of each species (with most sites having 5 replicates) were measured for their specific leaf area (SLA), leaf dry matter content (LDMC), leaf thickness, C:N, δ13C, stomatal density, and stomatal length using standardized sampling methods. For leaf measurements (SLA, LDMC, and leaf thickness), the most recently produced, but mature leaf was sampled from each replicate. Leaf area and leaf thickness were measured in the field. Leaf area was measured using Leafscan, a mobile app for measuring the surface area of leaves and leaf thickness using calipers. To calculate LDMC, leaves were rehydrated by being submerged in water for 24-72 hours for wet mass measurements and dried in a drying oven at 60 °C for at least 48 hours for dry mass.

Stable isotope measurements for leaf δ13C, δ15N, total C, and total N were performed at the Stable Isotope Mass Spectrometry Laboratory at Kansas State University. Multiple leaves from each replicate were dried for at least 48 hours at 60 °C and homogenized with an amalgamator. Total C and N of homogenized leaf samples were measured using an Elementar vario Pyro cube coupled to an Elementar Vision mass spectrometer for isotope analysis.

For temporal trends, all δ13C values were corrected for changes in atmospheric δ13C by converting to carbon isotope discrimination values Δ13C. Atmospheric [CO2] and δ13C air measurements were retrieved from McCarroll & Loader (2004) for the years preceding 2004 and measurements from the Mauna Observatory Data were used for years 2004 – 2022 (Keeling et al., 2005).

We measured stomatal density and length using stomatal peels on herbarium samples and pressed and dried field samples collected from each study site. Stomatal peels were created by applying clear nail varnish to leaves of the specimens and peeling the varnish once dry with clear tape. Both D. oligosanthes and P. virgatum are amphistomatous, so peels were made on both the abaxial and adaxial surfaces of the leaves. For herbarium specimens, the leaves of P. virgatum were long and folded to fit on the mounting sheet, exposing both sides of the same leaf. Thus, abaxial and adaxial peels were taken from the same leaf where the leaf was folded. For D. oligosanthes, the leaves were short and not folded to fit on the herbarium sheets, so only one side of each leaf was readily available to perform peels. To circumvent this issue, peels of the abaxial and adaxial surfaces were made on different (but similarly-developed) leaves of the same individual. For field-collected material, abaxial and adaxial peels were taken from the same leaf.

Two counts of stomatal density were taken for each peel, and five replicates of stomatal lengths were measured for each count of stomatal density (10 total per specimen). Stomata were counted under 20x magnification on the objective lens and 10x magnification on the ocular lens using an Olympus BH-2 Microscope (Shinjuku City, Tokyo, Japan). An image was taken of each leaf section using a Lumenera Infinity 2 microscopy camera (Ottawa, Canada). The area of the image field of view was determined by using a stage micrometer and was 0.120 mm2 for each image. Stomatal densities were then converted to stomata per 1 mm2. Total stomatal density was measured as the sum of the abaxial and adaxial stomatal densities. Stomatal length (horizontal length of the guard cell from end to end) was measuring using ImageJ; pixel length was converted to mm using a reference length determined from the stage micrometer. Five herbarium specimens of P. virgatum that were measured for stable isotopes were unable to be sampled for stomatal densities or lengths, as either the specimens had leaves that were too curled or wrinkled to obtain peels, or stomata were too sunken and not visible on the peels. Additionally, we note that because leaves shrink during dehydration, these measurements are likely overestimations of stomatal densities and underestimations of stomatal lengths compared to fresh leaf tissue. However, because all tissue in this study was dry, the values are all comparable.

The four corresponding R scripts can be found here:

R Code for the Herbarium Leaf Trait Analysis (Stomatal Traits): http://lter.konza.ksu.edu/sites/default/files/data/LTM011_Rscript.R

R Code for the Herbarium Leaf Trait Measurements (Non-Stomatal): http://lter.konza.ksu.edu/sites/default/files/data/LTM012_Rscript.R

R Code for the Grassland Sites Leaf Trait Measurements (Stomatal): http://lter.konza.ksu.edu/sites/default/files/data/LTM013_Rscript.R

R Code for the Grassland Sites Leaf Trait Measurements (Non-stomatal): http://lter.konza.ksu.edu/sites/default/files/data/LTM014_Rscript.R