Flood-driven survival and growth of dominant C4 grasses helps set their distributions along tallgrass prairie moisture gradients
Data files
Jan 27, 2025 version files 35.64 KB
-
Fld2.csv
4.36 KB
-
README.md
3.57 KB
-
TableS1.xlsx
27.70 KB
Abstract
Five C4 grasses (Bouteloua curtipendula, Schizachyrium scoparium, Andropogon gerardii, Sorghastrum nutans, Spartina pectinata) dominate different portions of a moisture gradient from dry to wet prairies in Wisconsin and other parts of the Upper Midwest. We hypothesized that the distributions of these species may, in part, reflect differences among species in flooding tolerance, and in context-specific growth relative to each other. We tested these ideas with greenhouse flooding and drought experiments, mesocosm experiments conducted under heavy rainfall, and a natural experiment involving graminoid response to a month-long flood in two wet-mesic prairies. Bouteloua appears to be excluded from wet and wet-mesic prairies by physiological intolerance of flooding; it promptly succumbs to inundation in greenhouse and mesocosm experiments. Given its low stature, Bouteloua is likely excluded from more productive, well-drained sites by competition. Schizachyrium is excluded from wet prairies because of its low tolerance to flooding, demonstrated by the natural experiment. Sorghastrum ranked low in flooding tolerance in both the greenhouse and natural experiments, suggesting that physiological intolerance excludes it from wet prairies. Spartina had by far the greatest growth under the wettest conditions in the mesocosms, suggesting that competition helps it dominate wet prairies. Indeed, the quadrat presence of Spartina increased by 57% two years after the flooding of two prairies, while that of upland grasses declined by 44%. Andropogon’s surprisingly high flooding tolerance, its lack of significant difference from other species in drought tolerance, and its tall stature suggest that broad physiological tolerance combined with competitive ability allows it to occur across the prairie moisture gradient.
README: Flood-driven survival and growth of dominant C4 grasses helps set their distributions along tallgrass prairie moisture gradients
https://doi.org/10.5061/dryad.m37pvmd8t
Description of the data and file structure
Five grass species were grown in greenhouses and outdoor mesocosms and subjected to flooding. Data on survival as a function of flooding time were gathered for the greenhouse experiment, and on survival and growth were gathered for the mesocosm experiment.
Files and variables
Fld2.csv
contains tissue score data: leaf tissue condition as a function of time in a drought-to-death experiment for individual plants in five different species grown in two greenhouses. Data are used for Table S1 and Table 2.
- Greenhouse: The number of greenhouses in the facility, either 7 or 26
- Days Flood: Number of days of exposure of plants to fully flooded condition
- Species:
- And = Andropogon
- Bout = Bouteloua
- Schiz = Schizachyrium
- Sorg = Sorghastrum
- Rating May 14: visual score of percent of live tissue, 1-10, conducted 6 days after the last pots were removed from treatment. 1 = no functional leaf tissue 10 = all functional leaf tissue
TableS1.xlsx
gives dry biomass as a function of species, water treatment, and occurrence in monoculture vs. mixtures in the mesocosm experiment. Barrels are 55-gallon mesocosms labeled to separate each for purposes of analysis. See methods.
- Barrel Labels consist of a letter indicating species in monoculture, or an M for mixed, i.e., barrels that had all five species. Barrel labels then have a treatment number, then another letter A, B, C, D, and E added to create a unique designation for each barrel.
- Species:
- A = Andropogon
- B = Bouteloua
- C = Schizachyrium
- O = Sorghastrum
- P = Spartina
- M = Mixed treatment in planting (all 5 spp. planted)
- Treatments
- 1 = zero depth to the water table
- 2 = 20 cm to the water table
- 3 = 40 cm to the water table
- 4 = 60 cm to the water table
- 5 = 80 cm to the water table.
- PlantSeedNet is the dry weight in grams of the plant stem, leaves, and seeds.
An Example: Barrel Labeled P5E is Spartina in monoculture, treatment 5, randomly assigned to barrel E to create a unique barrel code. Five plants were planted in each barrel, so each monoculture unique barrel label is repeated in the data 5 times, one for each plant that was dried and weighed separately.
A mixed treatment example. Barrels labeled M5E have five different species in a mixture with Treatment 5, randomly assigned to barrel E to create a unique barrel code. Column 2 then indicates which species is weighed to return the PlantSeedNet weight.
FloodCode_2023_JC.Rmd (hosted on Zenodo) gives the code for conducting the logit analysis. Fld2.csv provides the data input matrix for the logit analysis and should be placed in the R directory to run.
Logit_Description_Brett_Larget.pdf (hosted on Zenodo) is the logit analysis description as a pdf. It does not include data or code that can be used directly for analysis.
TableS1.xlsx provides data on biomass of five different species grown in monoculture and mixture at five different water-level treatments.
Code/Software
FloodCode_2023_JC.Rmd is R code. Fld2.csv is a comma-separated datasheet to be used as input for the R code. The graph will output in the same directory used for the input file. The R Studio is Version 2023.03.1+446
Change log
Table S2 was renamed Table S1 and uploaded as 'Data'. R code added to 'Software'.
Methods
See Wernerehl and Givnish 2025, American Journal of Botany for full details.