Habitat selection of three gull species in response to sudden changes in human mobility (dataset)

Data and code for "Habitat selection of gulls in response to sudden changes in human mobility"

Associated journal article can be found here: http://doi.org/10.1098/rspb.2025.2482

Primary contact for this repository:
Robert Patchett; University of St Andrews, Scotland; [email protected]

1. Project Overview
This document accompanies the manuscript titled:
"Habitat selection of gulls in response to sudden changes in human mobility"
and includes the data and R scripts used in the analyses.

Abstract: Developing robust strategies for human–wildlife coexistence is hampered by our limited understanding of how humans impact animal space use. It is challenging to measure the relative effects of landscape modification and human mobility on wildlife, since these factors are typically confounded. The extreme change in human mobility levels that occurred during COVID-19 lockdowns provided an opportunity to disentangle these impacts. Many gull species are considered urban adapters, capable of roosting, foraging and breeding near humans in highly modified environments. We predicted that lockdown-induced changes in human mobility would affect gulls’ selection for urban and beach habitats, because of altered disturbance levels and food availability. We analysed GPS-tracking data from 113 individual gulls over multiple years (2015–2022), across three species in western Europe (herring gull Larus argentatus, lesser black-backed gull L. fuscus, and yellow-legged gull L. michahellis). We found that, during lockdowns, selection for urban areas increased in two of ten colonies and selection for beaches increased in one colony and decreased in two others. This heterogeneous pattern likely reflects differences in how gull populations respond to opportunities and challenges presented by human-modified landscapes. Understanding this context dependence is emerging as a priority for coordinated efforts to promote sustainable human–wildlife coexistence.

Repository information and dataset IDs for GPS tracking data used in analyses can be found in electronic supplementary material, Table S1.

2. Contents
* README.txt - This file
scripts.zip/
* 1_processing.r - Script to process raw GPS tracking data
* 2_annotation.r - Script to annotate processed GPS tracking data
* 3_issa.r - Script to run integrated step selection analyses
* 4_urban_analyses.R - Script to reproduce the urban habitat selection analyses and generate plots
* 5_beach_analyses.R - Script to reproduce the beach habitat selection analyses and generate plots

data.zip/
* corine_landcover_2018 - Folder containing CORINE landcover data for use with 2_annotation.r
* corine_levels.csv - File used in 2_annotation.r
* urban_data.csv - Final dataset used for analysis of urban habitat selection
* beach_data.csv - Final dataset used for analysis of beach habitat selection