A new tool for detecting changes in dynamic rules in population time series data
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
casestudydata
figs
simresults
tests
writing
.gitignore
README.md
harmonia_example.R
monarch_example.R
monarch_regime.Rproj
plot_harmonia_figures.R
plot_monarch_figures.R
plot_simulation_results.R
plot_simulation_results_model_sets.R
regime_shift_detector.R
simulations.R
simulations_weight_analysis.R

README.md

The Monarch Regime project

In this repo, we develop a novel break-point analysis tool for population time series data, building on the methods developed by Bahlai et al 2015, Ecological Applications. The tool uses the Ricker model as the basis for a dynamic regime, and iterates through alll break point combinations, using information-theoretic decision tools (ie AICc) to determine best fits. In this repo, we develop the tool, simulate data under a variety of conditions to test the tool, and apply the tool to case studies of monarch butterfly overwintering and multicolored Asian ladybeetle invasions. This tool is scripted entirely in R.

File navigation

regime_shift_detector.R - contains functions to detect regime shifts in population time series. The function RSdetector() takes raw time series data and generates a complete report on fits, best fits, break points, and regression parameters for models with best fits

monarch_example.R - applies the regime shift detector analysis to monarch overwintering data from Mexico

plot_monarch_figures.R - plots output from monarch example, places outputs in figs folder

harmonia_example.R - applies the regime shift detector analysis toharmonia ladybeetle population data from Kellogg Biological Station. Includes data cleaning/manipulation code after Bahlai et al 2015.

plot_harmona_figures.R - plots output from harmonia example, places outputs in figs folder

simulations.R- a set of functions that creates time series data using secified parameters, and then a set of funtions to test if the parameters input match the ones detected by the regime shift detector, and the code that creates simulations under a variety of conditions, runs it through the comparison functions, and tallies the outputs, outputs a CSV file to the 'simresults' folder

plot_simulation_results.R - takes the simulation outputs and creates plots based on varying one input at a time to see the RSdetector's performance under differing conditions- outputs figures as PDF vector graphis to the 'figs' folder

casestudydata folder contains data for Harmonia case study. Monarch study data are proprietary

simresults folder contains simpulation outpu CSVs, in sets of ~50 iterations of each case at a time

figs folder PDF versions of all figs produced be scripts are held in this folder

tests folder contains a version of the RSdetector code and test data used in development

writing folder contains bits of writing, abstracts, etc, relevant to presentations and publications on this project

by regime, I mean the set of dynamic rules, or constants, in an equation describing how populations of an organism change throught time. I am not referring to authoritarian government and/or the rules imposed by them. Nature (and the things people do to nature) is the government of butterflies.