Sick without signs. Subclinical infections reduce local movements, alter habitat selection, and cause demographic shifts
https://doi.org/10.5281/zenodo.13934755
In wildlife populations, parasites often go unnoticed, as infected animals appear asymptomatic. However, these infections can subtly alter behaviour. Field evidence of how these subclinical infections induce changes in movement behaviour is scarce in free-ranging animals, yet it may be crucial for zoonotic disease surveillance. We used an ultra-high-resolution tracking system (ATLAS) to monitor the movements of 60 free-ranging swallows every 8 seconds across four breeding seasons, resulting in over 1 million localizations. About 40% of these swallows were naturally infected with haemosporidian parasites. Here, we show that infected individuals had reduced foraging ranges, foraged in lower quality habitats, and faced a lowered survival probability, with an average reduction of 7.4%, albeit with some variation between species and years. This study highlights the impact of subclinical infections on movement behaviour and survival, emphasizing the importance of considering infection status in movement ecology. Our findings provide insights into individual variations in behaviour and previously unobservable local parasite transmission dynamics.
This R-Project contains the following code files (in folder R):
01_cmr_data_cleaning: Code to clean the raw capture-mark-recapture data
02_pathogen_data_cleaning Code to clean the raw pathogen data
03_movement_data_cleaning: Code to clean and filter the raw ATLAS movement data
04_ctmm_models: Code to prepare data, applying continuous time movement models (incl. model selection) on movement data. Moreover, track reconstruction is performed
05_HMM_behaviour_models: Code to perform Hidden-Markov models (HMMs) to assess behavioural states from movement data
06_ISSF_habitat_selection_models: Code to perform Integrated Step Selection Functions (iSSF) to assess habitat selection during foraging behaviour
07_multivariate_analyses: Code to perform multivariate analyses on morphological traits in relationship to infection status
08_population_models_nimble: (nimble)-Code to perform candidate Multievent models, including model selection to obtain the best fitting models.
09_simulation_study: (nimble)-Code to perform simulation study on Multievent models, assessing if we could retrieve unbiased estimates given our sampling regime
10_resampling_30min: Code to perform analyses on resampled data (resampled to 30 minutes), to verify that we could not detect subtle parasite-induced effects on movement behaviour by using coarser resolution
This repository contains the following data files (in folder data-raw) :
birds_cmr_2019: capture-mark-recapture data collected in 2019
birds_cmr_2020: capture-mark-recapture data collected in 2020
birds_cmr_2021: capture-mark-recapture data collected in 2021
birds_cmr_2022: capture-mark-recapture data collected in 2022
birds_cmr_2023: capture-mark-recapture data collected in 2023
swallows_filtered: Filtered movement data of swallow species (full data set published on movemebank: Movebank-ID: 3053965481)
pathogen_data: Blood parasite infection data of swallow species
stacked.tiff: Stacked environmental covariates used in anylses (in folder geo-proc)
NA
Multievent models (08_XXX) and simulations (09_XXX) require package nimble (de Valpine et al., 2017): https://doi.org/10.1080/10618600.2016.1172487
Please see R scripts for additional packages used in processing data.