Invasive alien predators overturn the spatial-scaling laws of biocomplexity: Data and R codes to run analyses and to reproduce figures.

This archive contains data and R codes to run analyses and to reproduce figures for "Invasive alien predators overturn the spatial-scaling laws of biocomplexity". All R codes are carefully commented. They require that you provide, at start of each R session, the location where you downloaded and uncompressed the archive on your computer using the following command: setwd("path_to_archive_location/Supplementary_Materials_Zenodo/"). Paths to files in R codes are for a Linux machine and separate folders with a slash "/". In case you are running these codes on a MS Windows machine, backslashes "\" should be used instead. Any request should be sent to [e-mail]. Data folder contains: Inventory_data_formatted.txt: contain the raw biodiversity data at each sampling session. ROWS: One row is a trophic species at a given sampling session. COLUMNS: Milieux: habitat type (pond or marsh) Id_Sites: sampling site identity Annee: year of sampling Date: month of sampling Taxon: species identity as recording on the field Technique.d.echantillonnage: sampling technique Surface: habitat surface area in m2 sampid: identity of sampling session Observation_matrix.txt: 61 x 61 matrix for the number of times a trophic interaction was studied in the literature, and reported as either present or absent. Contains 0 when no report was available, or a positive integer. This matrix is not used in the analyses presented in the paper. ROWS: a trophic species as a prey. COLUMNS: the same trophic species as predators. Paper_list.txt: list of papers used in the literature review. ROWS: a paper. COLUMNS: Author: first author Year: year of publication Title: title of publication. Journal. Volume. DOI: when available Taxa_names_matrices.txt: contains the list of trophic-species names in the same order as they appear in the metaweb. Trophic_link_matrix.txt: 61 x 61 matrix for the number of times a trophic interaction was reported to be present in the literature. Includes NA if no observation was available (corresponding to a 0 entry in Observation_matrix.txt). ROWS: a trophic species as a prey. COLUMNS: the same trophic species as predators. Metaweb folder contains: Missing_links subfolder contains: Matching_centrality_model.R: R script to run the matching-centrality model. This requires to also have JAGS installed on your machine (https://mcmc-jags.sourceforge.io/). This script is used to produce Matching_centrality_model.jags, Matching_centrality_model_SAMPLES.txt and Matching_centrality_model_SUMMARY.txt files. Matching_centrality_model.jags: latent-link model in JAGS language. Matching_centrality_model_SAMPLES.txt: posterior-samples data from JAGS model. Matching_centrality_model_SUMMARY.txt: summary of the JAGS model. Used to produce Table S1 in Supporting Information. Matching_centrality_cutoff_probability.R: R script to compute the threshold probability that defines the predicted adjacency matrix. Used to produce Matching_centrality_predicted_trophic_matrix.txt file. Impute_missing_links.R: R script to impute missing trophic links from predicted adjacency matrix. Produces Imputed_link_matrix.txt file. Imputed_link_matrix.txt: 61 x 61 adjacency matrix of the imputed metaweb. Metaweb_figure subfolder contains: Truncate_latin_names_function.R: R function to truncate trophic-species names for clearer metaweb plotting. Metaweb_figure.R: R scripts to produce figures of the metaweb (Fig. 1 in manuscript and Fig. S1 in Supplementary Information). Requires the user to choose between the observed (non-imputed) and imputed metawebs. Trophic_relationships_figure subfolder contains: R script to produce figure of trophic relationships in the metaweb (Fig. 2 in manuscript). Surface_bioinvaders folder: Surface_bioinvader_Models_and_Figure.R: R script to produce prevalence-area relationships (Fig. 3 in manuscript) and summaries for the three models (Table S2 in Supporting Information). Summary_tables.txt: Model summaries as reported in Table S2. Local_food_webs folder: Local_food_web_reconstruction subfolder contains: Reconstruct_local_observed_food_webs.R: R script to produce local food-web data at each sampling session. Includes extra food-web metrics. Used to produce data files Local_reconstructed_food_webs.txt and Local_reconstructed_food_webs_imputed_missing_links.txt. Reconstruct_local_food_web_function.R: R function to compute local food-web descriptors at each sampling session from local biodiversity and and metaweb. Computes more descriptors than used in the paper. Surface_food_webs_Models_and_Figure subfolder contains: Model_function_surface_food_webs.R: R function used to run models. Surface_effects_food_webs_Models_and_Figure.R: R script used to produce Fig. 4 in manuscript, Fig. S2 in the Supplement, as well as model summaries and anova tables testing for significance of effects as reported in Table S3. Surface_community folder contains: Model_function_surface_community.R: R function to run models and produce summary statistics in Anova.table.txt and Summary.table.txt. Surface_community_Models_and_Figure.R: R script used to produce Fig. S3.