Articles • scari

Quantify risk to specific viticultural regions

Example usage of create_risk_report(), applied to create reports for French Viticultural areas

Example usage of create_risk_report(), applied to create reports for French Viticultural areas
Create risk reports for global countries and provinces
Create risk reports specifically for the USA and US States

Initialize scari

Initialization of renv package for dependencies

Initialization of renv package for dependencies

Retrieve and tidy input data for MaxEnt

Retrieve occurrence data and climate rasters

Bioclim variable rasters: CHELSA
SLF occurrence records: GBIF, lydemapR, literature

SDM modeling pipeline

Setup and create a global-scale model and 3 regional-scale models

Setup for global-scale MaxEnt model
Run global-scale MaxEnt model
Setup for regional-scale MaxEnt models
Run 'Ri.NAmerica' model trained on SLF invaded region in North America
Run 'Ri.Asia' model trained on SLF invaded range in Japan and South Korea
Run 'Rn' model trained on SLF native range in east Asia

Ensemble Regional-scale SDMs

Analyze extrapolation and use as weighting factor to ensemble regional-scale models

Run an Extrapolation Detection (NT2) analysis and calculate MIC
Ensemble regional-scale models using a weighted mean

Quantify SLF risk

Create suitability maps, scatter plots and figures of results

Create risk maps of each model scale and intersect predictions
Plot shift in SLF risk to important viticultural regions under climate change
Infer stability of known SLF populations due climate change effects
Generate IVR/SLF risk tables and other figures

Quantify model fit

Assess model fit using ROC plots, variable contribution, confusion matrices, and response curves

Analyze biological realism of model response curves
Assess model goodness of fit via AUC, TSS and confusion matrices
Plot variable importance for all models

Workflows

Outlines specific workflows for functions

Workflow for the functions 'compute_MaxEnt_summary_statistics()'