MeadoWatch Analysis Project
This github repository contains the raw data, cleaned data, scripts (R markdown), figures and output files from analysis characterizing the wildflower season and visitor sentiment. Specifically:
The main folder contains the markdown files (with analyses), including:
- MW_DataWrangling.RMD: this markdown reads in raw data, and filters out outliers and plot-species combinations with too few observations. This script takes a long time to run, and could probably use some optimizing.
- MW_SDDModeling.RMD: this markdown reads in SDD data (processed by MW_Microclimate.RMD) from collected from various locations with Mt Rainier NP, and gap fills missing MeadoWatch observations.
- MW_Modelfitting(byyearspeciesplot).RMD: this markdown reads in clean phenology data and fits phenological curves to data from each species / year / plot.
- MW_Modelfitting(byspecies).RMD: this markdown reads in clean phenology data and clean snow disappearance data and fits a phenological model to all data collected from each species (assuming a linear relationship between SDD and peak flowering).
- MW_PredictedFloweriness.RMD: this markdown reads in species specific parameters (description the relationship between peak flowering and snow for each species), and estimates flowering richness along each trail.
- MW_WTAanalysis: This markdown reads in WTA data and explores it and predicted flowering (from MeadoWatch data)
The data folder contains
- Phenology data
- WTA data
- Snow disappearance data
The clean data folder contains
- Phenology data (minus outliers, plot-year-species data with less than a certain number of observations)
- MeadoWatch data (with predictions for missing data)
The output folder contains
- parameters from maximum likelihood models (fit per species / year / plot, fit per species)
- AIC from models
- outliers
The figs folder contains various figures (many also plotted within the markdown file)