Data from: Minority-group incubators and majority-group reservoirs for the diffusion of climate change adaptations
Data files
Sep 17, 2023 version files 84.10 MB
-
main_parts.zip
8.60 MB
-
README_Dryad.md
4.92 KB
-
supp_parts.zip
75.49 MB
-
Variables_Table.xlsx
9.66 KB
Abstract
These data are part of a data portal that accompanies the special issue 'Climate change adaptation needs a science of culture,' published in Philosophical Transactions of the Royal Society B in 2023. To access the data portal, please visit 10.5061/dryad.bnzs7h4h4.
This repository contains code and supporting documentation for the agent-based model analyzed in our paper, "Minority-group incubators and majority-group reservoirs for climate change adaptation". We performed and analyzed a suite of agent-based models that simulated the spread of an adaptation, i.e., a beneficial behavior, in a population with a minority and majority group, defined by group size and tendency to interact with others from one's own group versus another group (homophily). We ran 1000 trials per parameter setting, where parameters were systematically varied to test different homophily levels in each group, and the effect of whether the minority group, majority group, or both groups start with one member knowing the adaptation. The adaptation either spread from one agent in one or both groups to the rest of the members of both groups in the case of adaptation success, or the adaptation disappeared from the entire population (adaptation failure). We then calculated the success rate across all 1000 trials. We also measured the mean time to either adaptive success or failure.
The data was generated through agent-based modeling of adaptation diffusion in a simulated population. The data is broken out into several CSV files from simulations that were each run on one cluster node. There are two main archives of CSV files: (1) `main_parts.zip` contains 30 CSV output files used in the main text analysis; and (2) `supp.zip` contains 270 CSV output files used in the supplemental analyses. The R analysis code (`scripts/plot.R`) contains utilities for combining and processing this raw output. See the `Analysis` subsection in the README_Dryad.md file for more information on the data combination and processing steps.
All required software is free and open-source. The simulations were run using the Agents.jl library in the Julia programming language. Model output analysis was performed with the ggplot2 library in the R programming language.
