Sentinel-2 EVI2 Composites, Phenological Metrics, and Analysis Scripts for Monitoring Native Grassland Restoration in South Australia (2015–2024)

README — Sentinel-2 EVI2 Dataset and Analysis Scripts for Grassland Restoration Study This archive contains processed remote sensing data and R scripts used to analyse vegetation phenology across restored, reference, and control native temperate grassland sites in South Australia. Contents:1. EVI2_Composite_Data/ - Sentinel-2 EVI2 image composites (.tif) and corresponding acquisition date files (.csv), generated using Google Earth Engine (GEE). 2. Extracted_Phenology_Metrics/ - CSV files containing site-level phenological metrics, including start of season (SOS), end of season (EOS), length of season (LOS), and peak of season (POS). These were calculated from the EVI2 time series using the R script `phenology_modeling.R`. 3. Scripts/ - phenology_modeling.R: R script for smoothing EVI2 time series and extracting phenological metrics per site. - hypothesis_testing.R: R script for statistical comparison of phenological metrics across management categories (restored, reference, control) using ANOVA and Tukey’s tests. File Naming Convention:- S2-composited_evi2_[site]_[start_date]_[end_date].tif: Composite EVI2 image- DATES_S2-composited_evi2_[site]_[start_date]_[end_date].csv: Dates included in each composite- Copy of [Site] dataset.csv: Extracted phenological metrics per site used for statistical comparison Tools used:- Google Earth Engine (for image generation)- R 4.2.3 (for time series analysis and hypothesis testing) Prepared by: Maximilian McQuillan [e-mail] April 2025