Data and code from: Suppression of soil microbiota rather than neighbours facilitates absinthe (Artemisia absinthium) invasion in native grasslands

Published Apr 30, 2026 on Dryad. https://doi.org/10.5061/dryad.8pk0p2p3k

Data files

Apr 30, 2026 version files 42.09 KB

Absinthe_Allelopathy_Oikos_Code__For_Final_Publication_.R

10.03 KB
Absinthe_Allelopathy_Oikos_Dataset.xlsx

26.71 KB
README.md

5.35 KB

Abstract

Many mechanisms can lead to successful plant invasion, but their importance is often context dependent. One such mechanism is allelopathy: chemical inhibition of neighbouring plants. The importance of allelopathy may be mediated by soil microbiota and environmental conditions, and depend upon the species or functional group affected. To better understand how these factors alter allelopathy and competition, we conducted a combined field and greenhouse experiment focusing on the invasive species absinthe (Artemisia absinthium). We experimentally introduced absinthe into native grassland via disturbances and collected soil from beneath these plants after three years. We then tested how invaded versus uninvaded field soil, within burned or unburned grassland, affected the performance of absinthe, two forbs (Achillea millefolium and Medicago sativa), and two grasses (Bromus inermis and Elymus lanceolatus) in a greenhouse competition experiment. We included activated carbon and soil sterilization treatments to test for allelopathic and microbial effects. Microbial inhibition, rather than allelopathy, drove competitive interactions. In uninvaded, unburned soil, all species were inhibited, and absinthe had no competitive effect. However, when absinthe was released from microbial inhibition by soil sterilization or prescribed fire, it suppressed other species. Microbial inhibition was also reduced for all species in invaded soil, suggesting suppression of pathogens by absinthe. Responses also varied between functional groups, with grasses being more vulnerable to microbial inhibition. Our findings highlight the complexity of invasion effects on soils and the potential for feedbacks on invasion outcomes. They also suggest that disturbance may facilitate invasion via effects on both plants and soil microbes.

Dataset DOI: 10.5061/dryad.8pk0p2p3k

Description of the data and file structure

Files and variables

File: Absinthe_Allelopathy_Oikos_Code__For_Final_Publication_.R

Description: Code associated with Wasan et al 2026 Oikos

File: Absinthe_Allelopathy_Oikos_Dataset.xlsx

Description: Data on plant growth responses to absinthe-conditioned soils from Wasan et al 2026 Oikos. Also includes data on soil analyses. The file contains five tabs: one for each analysis performed. These are "Soil Analysis" which contains data on total nitrogen, "Absinthe Alone" which contains biomass data for absinthe (Artemisia absinthium) when grown without competitors, "Absinthe competition" which contains absinthe biomass data when grown in competition with other species, "Alfalfa" which contains Medicago sativa biomass when grown alone and in competition with absinthe under various treatments, and "Brome-NWG-Yarrow" which contains biomass data for Bromus inermis, Elymus lanceolatus, and Achillea millefolium when grown alone or in competition with absinthe under various conditions.

Variables

Invaded - Whether the area where the soil was collected had been experimentally invaded by absinthe or not
Fire - Whether the field plot where the soil was collected had been burned or not
Block - Block number; i.e., the replicate identity
Treatment - Soil treatment following collection from the field prior to seeding: Activated carbon added (AC), Control, or Sterilized with heat and steam
TN - Total nitrogen (%)
Species - Identity of the plant species for which the biomass was collected in the experiment (Smooth brome (Brome; Bromus inermis), northern wheatgrass (NWG; Elymus lanceolatus), yarrow (Achillea millefolium), alfalfa (Medicago sativa))
Density - Whether the plant was grown alone or in competition with absinthe
Shoots - Shoot biomass for plants grown in the greenhouse experiment in g

Code/software

Code for linear models testing for treatment effects on plant growth, along with code for generating the figures within the dataset. Included is code for loading the data into R from the Excel sheet using the readxl package. The treatment variables were then leveled so that all contrasts were from treated to control. Following the revealing code is code for mixed models.

First is the model testing for nutrient differences as a function of absinth invasion and prescribed fire in the field (experiment described in Wasan et al. 2023), with the experimental block as a random effect. The results of these analyses were then graphed using the afex package (Singmann et al. 2023), which graphs the estimated marginal means for each treatment and the associated 95% confidence intervals.

The next model tests treatment effects on absinthe biomass, with field invasion, prescribed fire, and soil treatments (control, sterilized, activated carbon) included as factorial fixed effects, and block again as the random effect. There are two models, one for absinthe biomass when grown alone ("shoots.absinthe") and one for absinthe grown in competition (" absinthe.comp"). This model includes an additional factorial fixed effect called "Species", which signifies the identity of the species that was competing with absinthe. Absinthe grown with alfalfa was not included in this model because not all treatments were applied to pots including alfalfa. Additional code is included for pairwise comparisons among significant treatments using the R package emmeans (Lenth 2023), with contrasts always nested within the treatments according to the scale at which they were applied (prescribed fire > experimental invasion > soil treatments and competition treatments in the greenhouse). Results were again graphed using afex.

The next two models examine the responses of the species against which absinthe competed. There are two separate models because alfalfa was never grown in sterile soil. The alfalfa model and analysis framework was structured similarly to the above analyses, except the variable "Density" was used to denote whether the plant was grown in competition or alone. The final model examined the effects of absinthe competition on the biomass of the three other focal species and included both "Species" and "Density" as factorial fixed effects. Otherwise, the analysis was as above. Pairwise comparisons and graphs were also generated following the same format.

Lenth R (2023). _emmeans: Estimated Marginal Means, aka Least-Squares Means_. R package version 1.8.7, https://CRAN.R-project.org/package=emmeans.

Singmann H, Bolker B, Westfall J, Aust F, Ben-Shachar M (2023). _afex: Analysis of Factorial Experiments_. R package version 1.3-0, https://CRAN.R-project.org/package=afex.

Wasan JPM, Pyle LA, Bennett JA (2023). Disturbance and nutrient availability drive absinthe (Artemisia absinthium) invasion in a native rough fescue grassland. Écoscience, 30(1), 27–38. https://doi.org/10.1080/11956860.2023.2165283