This repository contains all code for "Climate change could fuel urinary schistosomiasis transmission in Africa and Europe" (2024) by van der Deure, T., Maes, T., Huyse, T., and Stensgaard,. A-S and published in Global Change Biology.
In this publication, we use correlative and mechanistic modelling to predict the current and future distribution of schistosomiasis intermediate host Bulinus truncatus.
The code used for correlative and mechansitic modelling are found in separate folders.
To reproduce the correlative modelling, first install the necessary packages using the script provided. Note that we used biomod2 version 4.2.4. Then download environmental layers by running 03_layers.R, fit the models using 04_models.R and run the projections using 06_projections.R. All model settings used are specified in 04_models.R. The other scripts in this folder apply spatial thinning to the occurrence points, build background layers, and investigate correlations between environmental variables.
Most mechanistic modelling was performed in Julia. The packages used including their versions are provided in Manifest.toml. Use Pkg.instantiate() to install all packages. Run climate_data.jl to download the necessary climate layers. run_model.jl generates model runs, calling the other scripts as required. Under trait_fits, the R script fit_traits.R fits Bayesian hierarchical models to the snail life history data, which is imported by the same script.
Figures are generated by two Julia scripts plot_maps.jl and plot_thermal_curves.jl, both in the figures folder.
To use another folder to store big files (e.g. when using a hard drive), simply change the paths specified in globals.jl. These paths are used to write to from mechanistic models, and are then used to read from to plot their outputs.
Most of the life history trait data used in the mechanistic models stem from an experiment that has been separately published as a preprint.
All primary, intermediate, and output data used in and generated by this code is avaiable for download on Dryad at the following link: https://doi.org/10.5061/dryad.98sf7m0s9.
Anyone is welcome to use, adapt, or redistribute (parts of) the code provided here. When using this code as part of a scientific publication, please cite the forthcoming publication.
The corresponding author for the publication is Tiem van der Deure, email tvd@sund.ku.dk.