Plant co-occurrence and functional trait data from drylands, along with soil, grazing pressure, and climate data

Smit, Imke Chrissie 1 ; le Roux, Peter C.1; Soliveres, Santiago2; Gross, Nicolas3; Le Bagousse-Pinguet, Yoann4; Saiz, Hugo5; Abedi, Mehdi6; Ahumada, Rodrigo7; Arroyo, Antonio I.8; Asensio, Sergio2; Benoit, Richard9; Blaum, Niels10; Bu, Chongfeng11 12; Eldridge, David13; Fajardo, Alex14; Farzam, Mohammad15 16; Fateh, Amghar17; García-Gómez, Miguel18; Gaitán, Juan J.19; Gonzalez, Sofia20; Gozalo, Beatriz2; Greenville, Aaron C.21; Guirado, Emilio2; Hering, Robert10; Leder, Cintia V.22; Li, Xinhao11; Liu, Jushan23; Louw, Michelle A.1; Manzaneda, Antonio J.24; Marais, Eugene25; Martínez-Valderrama, Jaime2; Mendoza, Betty J.26; Mora, Juan P.14; Moreno, Gerardo27; Ochoa, Victoria28; Oñatibia, Gastón R.29; Peter, Guadalupe22; Pueyo, Yolanda8; Quiroga, Emiliano7 30; Rahmanian, Soroor31; Rey, Pedro J.22 24; Rolo, Victor27; Plaza, César28; Sarig, Shlomo32; Selyari, Javad33; Travers, Samantha13; Valencia, Enrique34; Wang, Deli23; Wang, Lixin35; Wardle, Glenda21; Yahdjian, Laura29; Zaady, Eli32; Zhang, Yuanming36; Zhou, Xiaobing36; Maestre, Fernando T.37

Published Dec 11, 2025 on Dryad. https://doi.org/10.5061/dryad.9cnp5hqx5

Data files

Dec 11, 2025 version files 882.57 KB

1._PCA.R

2.74 KB
2._model_selection_nint_nurse_traits.R

13.29 KB
3._Trait_difference_analysis.R

6.64 KB
4._Graphs_and_figures.R

143.28 KB
alltraits.xlsx

181.78 KB
Chisq_results_6Feb2024_for_submission.csv

73.30 KB
nint_nurse_trait_env_data.xlsx

198.10 KB
README.md

13.58 KB
trait_difference_data.xlsx

249.85 KB

Abstract

Plant functional traits can influence interaction outcomes between nurse and target plants through a “functional trait match”, which occurs when the traits of nurse plants ameliorate their environment, and target plants possess traits that allow them to benefit from this ameliorated environment. We investigated how the traits of putative nurse species affect interaction outcomes across global drylands and determined the functional match that promotes facilitation. We also investigated how grazing pressure and global climatic and edaphic gradients affected this trait match. We used a collaborative survey conducted across 29 sites from five continents, where we gathered in situ co-occurrences of dominant species (‘nurses’) and other vascular plant species, as well as their functional traits [plant height and leaf dry matter content (LDMC)]. Climate, edaphic variables, and grazing pressure were measured in situ or extracted from databases. We used a model-building approach to determine the effect of dominant plant traits on interaction outcomes, and how the functional trait match between nurse and target species is affected by environmental variables. Tall dominant plants with conservative leaves generally had a greater positive effect on species richness and cover beneath their canopies, but these effects were strongly modulated by grazing pressure and soil pH. Target plants that were significantly associated with dominant plants tended to be shorter and have more acquisitive leaves than dominant plants, regardless of environmental conditions. However, the difference in height and LDMC between dominant plants and negatively associated target plants was strongly affected by environmental conditions. Functional traits play a significant role in determining interaction outcomes between dryland plants. Facilitation in drylands is driven by a conservative-acquisitive trait match, a pattern observed regardless of grazing pressure, climate, and soil conditions.

Dataset DOI: 10.5061/dryad.9cnp5hqx5

Description of the data and file structure

This data is linked to the research article titled "Plant trait matching occurs in facilitative interactions across global drylands". This paper investigates the role of functional traits and environmental gradients in determining plant-plant interactions in global drylands. Specifically, the paper investigates whether the traits of dominant plants (putative nurse plants), along with soil, climate, and grazing factors, affect interaction outcomes. Furthermore, the paper investigates the trait difference between facilitative pairs of dominant and target species (the trait match) and whether the trait match changes along environmental gradients.

This data was collected as part of the BIODESERT project. Refer to Maestre et al. (2022) for details of the sampling design used to collect this data.

Maestre, F. T., Eldridge, D. J., Gross, N., Le Bagousse-Pinguet, Y., Saiz, H., Gozalo, B., Ochoa, V., & Gaitan, J. J. (2022). The BIODESERT survey: assessing the impacts of grazing on the structure and functioning of global drylands. Web Ecology, 22(2), 75-96. doi:10.5194/we-22-75-2022

Files and variables

File: alltraits.xlsx

Description: Trait data of species collected in plots. Trait values represent the mean trait value for each specific species in each plot, except for height and lateral spread which represents the maximum value for each species in each plot. Data has been gapfilled with trait data collected from other plots with the same grazing pressure.

Attribute	Meaning
ID	Unique ID of the plot.
site_ID	Unique ID of the site.
taxon	Species name
trait	Trait that was measured. MeanLL = mean leaf length (cm), MeanSLA = mean specific leaf area (cm² /g). MeanLDMC = mean leaf dry matter content (%), MeanLA = mean leaf area (cm²), MaxH = maximum plant height (cm), MaxLS = maximum lateral spread (cm²). PercentN = elemental concentration of Nitrogen in the leaf (%), PercentC = elemental concentration of Carbon in the leaf (%).
value	The value of the trait.

File: Chisq_results_6Feb2024_for_submission.csv

Description: Results from χ2 analyses to determine whether species are associated with bare or dominant microsites.

Attribute	Meaning
site_ID	Unique ID of the site.
ID	Unique ID of the plot.
species	Name of target species.
association	Association of the target species. nurse = associated with dominant microsite; bare = associated with bare microsite; neutral = no significant association to either microsite; too_rare = occurrence is too low to perform χ2 test.

File: nint_nurse_trait_env_data.xlsx

Description: Used in the first model building procedure to determine the effect of dominant plant traits and environmental variables on interaction outcomes. Contains the Nint value and dominant species associated with each replicate, as well as the functional traits of the dominant species. Additionally, contains the climate and soil variables associated with each plot.

Attribute	Meaning
ID	Unique ID of the plot.
site_ID	Unique ID of the site.
replicate_no	Number of the replicate.
nurse_sp	Name of the dominant species.
NIntc_richness	Value of NIntC for the replicate calculated with species richness.
NIntc_cover	Value of NIntC for the replicate calculated with vascular cover.
NIntc_richness_binom	NIntC richness transformed to range between 0 and 1
NIntc_cover_binom	NIntC cover transformed to range between 0 and 1
Aridity	Aridity of the plot
graz	Grazing pressure of the plot. 0 = ungrazed; 1 = low grazing; 2 = medium grazing; 3 = high grazing.
RASE	Coefficient of variation of precipitation.
pH	Soil pH measured in bare patches without vegetation.
AMT	Annual mean temperature, ℃.
SAC	% mass of soil that consists of sand, collected in bare patches without vegetation.
nurse_mean_H	Mean Height of the dominant species, cm.
nurse_meanLDMC	Mean Leaf dry matter content of the dominant species, as a percentage.
log_nurse_meanH	nurse_mean_H, log transformed
log_nurse_meanLDMC	nurse_mean_LDMC, log transformed
sin_lat	The sine of the latitude of the plot
sin_long	The sine of the longitude of the plot

File: trait_difference_data.xlsx

Description: Used in the second model building exercise to investigate whether trait matching shapes interaction outcomes and whether this changes along environmental gradients.

Attribute	Meaning
site_ID	Unique ID of the site.
ID	Unique ID of the plot.
replicate_no	Number of the replicate.
trait	Trait considered.
trait_difference	Difference in the trait value between the dominant and target species. (trait value of dominant – trait value of target)
nurse_sp	Name of dominant species.
target	Name of target species.
association	Association of the target species. nurse = associated with dominant microsite; bare = associated with bare microsite
Aridity	Aridity of the plot
graz	Grazing pressure of the plot. 0 = ungrazed; 1 = low grazing; 2 = medium grazing; 3 = high grazing.
RASE	Coefficient of variation of precipitation.
pH	Soil pH measured in bare patches without vegetation.
AMT	Annual mean temperature, ℃.
SAC	% mass of soil that consists of sand, collected in bare patches without vegetation.
sin_lat	The sine of the latitude of the plot
sin_long	The sine of the longitude of the plot

Code/software

File: 1._PCA.R

Description: Script to perform principal component analysis of the trait data in order to reveal the two main axes of trait variation. The dataset associated with this script is alltraits.xlsx

File: 2._model_selection_nint_nurse_traits.R

Description: Script to perform model selection to determine the drivers of interaction outcomes. The data associated with this script is nint_nurse_trait_env_data.xlsx This model selection procedure uses multithreading and takes 96 hours to run.

File: 3._Trait_difference_analysis.R

Description: Script to perform model selection to determine the drivers of the difference in traits between dominant and target species. The data associated with this script is trait_difference_data.xlsx This model selection procedure takes 15 minutes to run.

File: 4._Graphs_and_figures.R

Description: Script wherein the graphs and figures shown in the main text were created.

All scripts were created in R4.4.2. R and R studio are required to run these scripts. All required libraries are loaded at the start of each script.

Access information

Other publicly accessible locations of the data:

none

Data was derived from the following sources:

This data represents a processed subset of the BIODESERT dataset