This repository contains the data and R code needed to reproduce the analyses published in the Nature Communication paper entitled "Impact of artificial light at night on diurnal plant-pollinator interactions" (authors: Simone Giavi, Colin Fontaine and Eva Knop).

'data.Rdata' : This R object contains two datasets. The first ('DAT') reports the number of diurnal interaction between plants and pollinator. Description of each column:

'treatment' : light treatment with control site (C, dark during night) and treatment site (T, artificially illuminated).
'plant' : selected plant species (21 species).
'insect_group' : pollinator groups (Hymenoptera, Diptera, Coleoptera).
'site' : meadow pair.
'repeate' : number of sampling replicate.
'plab' : plant abundance, specific for each repeate and plant species.
'dVis' : number of diurnal visits (plant-pollinator interactions).
'pair' : factor including site and repeate.

The second R object contained in 'data.Rdata' ('gradient') reports the light intensity measurements on three illuminated sites at each meter from 0 to 50-60 from the street lamp. Measurements were taken at night and in absence of light sources other than the street lamp.
'site' : site name.
'distance' : distance from the lamp in meters.
'Mean' : mean light intensity in that specific point. Several measurements were taken and were then averaged. SEM were very small and therefore not considered meaningful.

'script.R': In this script two models are fitted on the dataset 'DAT' and the effect size of the light treatment for each plant species (first model) and for each plant species and pollinator group (second model) is computed from the posterior distribution of the models. Subsequently, the code for figure 1 is provided.  Then 'DAT' is rearranged to build figure S1 and finally 'gradient' is used to build figure S2.