# Title of Dataset: Combining environmental niche models, multi-grain analyses, and species traits identifies pervasive effects of land use on butterfly biodiversity across Italy --- This dataset is associated with the manuscript "Combining environmental niche models, multi-grain analyses, and species traits identifies pervasive effects of land use on butterfly biodiversity across Italy", by F Riva, Barbero F, Balletto E, and Bonelli S., accepted for publication in the journal Global Change Biology. It contains a script (Script_Riva_et_al_GCB_2023.txt) and the data (Data_Riva_et_al_GCB_2023.RData) needed to reproduce the results of the paper. Within the .RData files, we provide Maxent Environmental Niche Models for 288 butterflies in Italy, which were used in the manuscript to test how land use affected biodiversity across spatial scales. The analysis demonstrated that both biodiversity of butterflies and the distribution of each species are affected by land use across large spatial scales across Italy. ## Description of the Data and file structure Beyond the README file, the repository contains only two files - a script written with software R, and a .RData file associated with the script. The script contains comments that detail all the steps of the analysis and connect the code to three sections of results in the manuscript. To run the script (Script_Riva_et_al_GCB_2023.txt), one must have software R installed (at least version 4.1) and edit the section "LOAD DATA AND DATA PREPARATION" to specify where the data included in the repository (Data_Riva_et_al_GCB_2023.RData) is located in their computer. Inside the .RData file, we hace included all the models run for the analysis, as well as the raster files representing environmental gradients and bioregions used to calibrate the models (see below). Given that we run more than 13.000 Maxent models, and this requires several hours of computation, we have provided the models already fitted and retained after model selection (see Methods in the manuscript). Still, we have shared the function we created to prepare occurrence data and to fit Maxent model in the section "SUPPLEMENTARY FUNCTIONS" of the script. This section also includes supplementary analyses, e.g., a test for spatial autocorrelation in our models, and a fucntion to calculate variable importance estimates. The version of software R and of the packages used in the analysis is provided at the end of the script file. ## Sharing/access Information Links to other publicly accessible locations of the data: NA We used the following four main datasets as covariates in our models. - Climatic data were downloaded from (i) WorldClim (http://worldclim.org/version2) and (ii) ENVIREM (https://envirem.github.io/). - Land use (iii) was downloaded from the Corine Land Cover project, accessing the release v18_5, reference year = 2012 (https://land.copernicus.eu/pan-european/corine-land-cover), and reclassified according to table S1 in the Supplementary information of the manuscript. - Biogeographic regions (iv) from (https://ecoregions2017.appspot.com/) were aggregated to represent four regions representative of the biogeography of Italian butterflies; see Fig. S2 in the Supplementary information of the manuscript. The response in our model was the occurrence of 288 butterfly species known to occurr in Italy. Biodiversity data was available in the CkMap dataset (https://faunaitalia.it/documents/CKmap_ENG.pdf), and combined with all data available on December 2 2012 in the Global Biodiversity Information Facility (GBIF; https://www.gbif.org/) GBIF Occurrence Download https://doi.org/10.15468/dl.7hhtym, accessed from R via rgbif (https://github.com/ropensci/rgbif) on 2022-12-02