This repository contains materials used to support the analyses for the paper titled: “Inferences about phenological shifts in an Arctic community vary with time-windows”. This study aims to examine how signals of phenological change are manifested at different temporal scales using data collected through the BioBasis program at the Zackenberg Valley from 1996 to present.
-
assess how trend directionality changes with time-series duration and choice of start years;
-
identify the minimum time-series length needed to obtain a high probability of agreement between shorter- and long-term trends.
-
data: contains raw and cleaned data used in the analysis
-
RScripts: contains analysis scripts for both plants and arthropods
-
diagnostics: contains taxon-specific files of model diagnostics
-
models: contains taxon-specific model objects
-
predictions: contains taxon-specific figures visualizing model parameters and predictions
-
Misc: contains exploratory analysis materials
To reproduce this analysis, the following scripts in the RScripts folder may be useful:
arthropod_data_cleaning.R: clean arthropod data, remove year-plot combinations that do not form a bell curve
covariates_data_cleaning.R: clean air temperature and spring snow cover data
plant_data_cleaning.R: clean plant data, remove year-plot combinations that do not form a bell curve
arth:phen2.stan: final Stan code for arthropod data
plant_phen6b.stan: final Stan code for plant data
arthropod_functions.R: source code for fitting and examining arthropod models
plant_functions.R: source code for fitting and examining plant models
allreps_models.R: code for fitting models across all time-windows for all plants and arthropods
create_reps.R: code for creating bubble plots for all plants and arthropods
all_repc_ci.R: source code for creating CI width-TSL plot for all plants and arthropods
arthropod_stanmod_PD.R: code to run for interfacing with Stan in R, and perform post-processing steps for arthropod models (used if not re-fitting models)
plant_stanmod_PD.R: code to run for interfacing with Stan in R, and perform post-processing steps for plant models (used if not re-fitting models)
phen_slide.R: script to extract slopes at different time-windows for both plants and arthropods (used if not re-fitting models)
signal_strength.R: script to examine trend directionality and strength for plants and arthropods (used if not re-fitting models)