CmlMagneville / Herbivory_BioInv_Paper

Herbivory_BioInv R Project

This project contains the code and data for the paper “…” (202..).

📦 It is organised as follow:

• the outputs folder contains high quality PDF files of each Figure and Supplementary igure of the paper

• the R folder contains the R scripts used to prepare the basic dataframes, compute each plot and diverse values used in the paper. Most Figure has one script for preparing data for the plot and plotting data and another one for computing interesting values about these plots.

• the raw_data folder contains the raw data used in the paper. The precise content of each file is described in the following Raw Data section.

• the transformed_data folder contains the data build from the raw data files computed to do the plots. The precise content of each file is described in the following Transformed Data section.

✨ To get all the figures and analysis of the paper, just run the make.R file

Goals of the Project

The goal of this project is to study the herbivory activity and its temporal variation within a day and across days in an invasive context using long-duration remote underwater cameras.

To learn more >>> add article

Raw Data

The raw_data folder contains:

• annot_data.rds: a dataframe containing informations about when annotations has been realised. 🔧 It has four columns : frame_nb_start -> the frame number of the beginning of the annotation sequence, frame_nb_stop -> the frame number of the end of the annotation sequence, real_time_start -> the time corresponding to the beginning of the annotation sequence, real_time_stop -> the time corresponding to the end of the annotation sequence

• final_0_1_all_cam_all_days.rds: the presence absence dataframe of the four species of interest. 🔧 It has seven columns : time -> the time for each second of each annotation sequence (thus rows of this dataframe does not represent continuous time but time of annotation sequences i.e. 1min40s each 10min from 8:00:00 to 18:11:40), cam_nm -> the name of the camera on which annotations has been realised (four cameras), recording_day -> the days on which annotations has been realised (three cameras), Then one column for each species of interest filled with 1 or 0 if species annotated or not.

💡 The dataframe has 75 144 row for 101 seconds of annotations * 62 sequences of annotation * 3 days * 4 cameras

• final_df_abund_all_cam_all_days.rds: organised as the final_0_1_all_cam_all_days dataframe but shows the abundance of species for each annotated second.

• final_fd_behav_all_cam_all_days.rds: organised as the final_0_1_all_cam_all_days dataframe but shows the bites (grazing activity) of species for each annotated second.

• fish_mouth_size.csv and metadata_fish_mouth_size.csv are data collected on several sites thanks to the EXOFISHMED campaign (https://exofishmed.cnrs.fr/). Only data collected in Crete - October 2019 was used.

Transformed Data

The tranformed_data folder contains:

• all_info_df.rds: organised as the final_0_1_all_cam_all_days dataframe but Species columns are divided into grazing and swimming behaviour and two columns have been added for sequence number (from 1 to 62 per day and camera) and a “day-camera” string.

• bites_nb_all_cam_top_df.rds: shows the number of bites of each species per sequence and mean maxN across the four spots of each species per sequence.

• bites_nb_all_sp_top_df.rds: shows the number of bites for the four species altogether per sequence and mean maxN across the four spots of each species per sequence.

• bites_nb_cam_day_df.rds : shows the proportion of time where each species is annotated grazing for one day and one camera compared to the amount of time where the species is seen.

• bites_nb_seq_model_df.rds: shows the data used for the generalized linear model.

• bites_sp_prop_df.rds: shows the number of bites given by each species for one day and one camera and the total number of bites done for one day and one camera.

• final_all_info_df.rds: shows for each species and each sequence of each day and camera: the maxN, the total bites number, the maximal number of individuals seen grazing, the mean number of individuals seen grazing, the number of frame where individual are seen grazing and the number of frame where individual are seen.

• final_all_info_df2.rds: is the same as the final_all_info_df but the same of species is changed.

• grazing_sc_sr2.rds: shows the number of bites of Sparisoma cretense in sequences where it is seen grazing and the number of individual of Sparisoma cretense grazing. A column indicate whether Sparisoma cretense individual(s) is/are grazing alone or with Siganus rivulatus and the last column refers to the category of abundance used for the plot.

• grazing_ss_sr2.rds: shows the number of bites of Sarpa salpa in sequences where it is seen grazing and the number of individual of Sarpa salpa grazing. A column indicate whether Sarpa salpa individual(s) is/are grazing alone or with Siganus rivulatus and the last column refers to the category of abundance used for the plot.

Contributions

CM, SV and TC conceived the ideas and designed methodology; MLLB, TD, GS, SV and CM collected the data; MLLB and CM analysed the data; CM and SV led the writing of the manuscript. All authors contributed critically to the drafts and gave final approval for publication.