Skip to main content
Dryad

Shortgrass steppe and northern mixedgrass prairie plant species traits

Cite this dataset

Blumenthal, Dana; Kray, Julie; Mueller, Kevin; Ocheltree, Troy (2020). Shortgrass steppe and northern mixedgrass prairie plant species traits [Dataset]. Dryad. https://doi.org/10.5061/dryad.8sf7m0cjr

Abstract

Despite progress in trait-based ecology, there is limited understanding of the plant traits that structure semiarid grasslands. In particular, it remains unclear how traits that enable plants to cope with water limitation are related to traits that influence other key functions such as herbivore defense and growth. The hypothesis that drought and herbivory exert convergent selection pressures is supported for morphological traits, but largely untested for struct­ural, physiological, and phenological traits. Drought and economic traits can also covary, but where and to what degree remains uncertain.

Here we address these uncertainties in semiarid shortgrass steppe and mixedgrass prairie, the largest remaining grasslands in North America. Using a broad selection of traits for 37 of the most common plant species in each ecosystem, we ask whether traits that confer drought tolerance, avoidance and escape covary with herbivore resistance traits and economic traits.

Results reveal that both drought tolerance and escape are coordinated with other functions, but in opposite fashion. Drought tolerant species (low leaf osmotic potential and high leaf dry matter content, LDMC) were also herbivore resistant (high leaf toughness and cellulose) and at the ‘slow’ end of the economic spectrum (low leaf nitrogen, leaf phosphorus, and high stem density). Conversely, drought escape via early senescence was associated with lower drought tolerance, lower herbivore resistance, and ‘fast’ economic traits. Drought avoidance, as indicated by thick leaves, may also be associated with lower drought tolerance (LDMC). Senescence date and LDMC appear to be key traits in these semiarid grasslands, differentiating species along multiple axes of function.

Synthesis – Covariation between drought, herbivory and economic traits means that, of the many potential trait combinations, few actually exist within these grasslands. Consequently, changes in land management and climate should have predictable effects on drought resistance, forage quality and productivity in the western Great Plains. 

Methods

Sampling methods-general: We sampled 5-10 replicate individual plants per species, depending upon the trait. For most traits, we sampled during flowering, thereby standardizing each measurement by plant developmental stage. For leaf osmotic potential, which can vary as water availability changes within a growing season, we constrained our sampling campaigns to 3-4 week periods of favorable soil moisture conditions when species diversity was at its seasonal maximum.

The majority of traits were measured in mixedgrass prairie in 2013 and shortgrass steppe in 2014. Exceptions were leaf senescence date, leaf pubescence, plant height (measured in 2015 at both sites), and leaf osmotic potential (measured in mixedgrass in 2015 and shortgrass in 2017).

Usage notes

Trait description and measurement details:   

Trait

Description

Measurement

Specific leaf area (SLA, cm2 g-1)

Area of an individual leaf divided by the leaf’s dry mass

Area of fully hydrated leaves measured on multiple leaves from each replicate plant and divided by oven-dry mass of the same leaves; mean SLA determined for each replicate (n=10).

Leaf nitrogen

(N, %)

Percentage of leaf dry mass that is N

N content determined from ground oven-dry leaf tissue run through an elemental analyzer (n=10).

Leaf phosphorus (P, %)

Percentage of leaf dry mass that is P

P content determined via wet chemistry analysis of ground oven-dry leaf tissue (n=5).

Stem specific density (SSD, mg mm-3)

Stem mass per unit volume of stem tissue

Volume calculated from dimensions measured via calipers near base of hydrated stem, divided by oven-dry mass of the same stem section (n=10).

Leaf osmotic potential

o, MPa)

Leaf cell solute potential at full hydration

Solute concentration of leaf cell water vapor (osmolality) measured via an osmometer, converted to osmotic potential (n=5).

Leaf thickness (mm)

Thickness of leaf lamina

Thickness measured perpendicular to primary axis of extension from the stem, at a central location along the leaf lamina, using calipers (n=10).

Leaf pubescence (%)

Percent of leaf surface covered by hairs

Mean percent cover of hairs determined using a grid of sample points placed over 40X magnified digital images of upper & lower surfaces of each leaf (n=5).

Leaf dry matter content

(LDMC, g g-1)

Ratio of dry leaf mass to leaf mass at full hydration

Mass of fully hydrated leaves measured on multiple leaves from each replicate plant paired with oven-dry mass of the same leaves; mean LDMC determined for each replicate (n=10).

Senescence date (DOY)

Day of year on which leaf senescence began

Mean day of year on which leaf canopy greenness of observed plants declined from 75-100% to 50-75% green (n=5).

Individual leaf area (cm2)

Projected area of an individual leaf

See leaf area notes for SLA (n=10).

Plant height (mm)

Height of uppermost leaf above ground surface

Height of uppermost leaves (i.e. top of canopy), excluding reproductive structures, above ground surface (n=25).

Leaf toughness (N)

Force required to punch a hole through the leaf lamina

Force-to-punch measured on fully hydrated leaves by a LF-Plus materials testing machine (n=10).

Leaf cellulose (%)

Percentage of leaf dry mass that is cellulose

Cellulose content, calculated from lignin and acid-detergent fiber content determined via wet chemistry analysis on ground oven-dry leaf tissue (n=5).

Leaf lignin (%)

Percentage of leaf dry mass that is lignin

Lignin content determined via wet chemistry analysis on ground oven-dry leaf tissue (n=5).

                       

Data flag: "P" indicates three instances where senescence date was not recorded for a species during the observation year. In these cases, we predicted senescence date from regression models constructed using other phenology information across species, sites and years.