<b>Code and data for: “Hydro-geomorphological drivers across scales shape the trajectory of coastal wetland restoration”</b>

README: Code and data for "Hydro-geomorphological drivers across scales shape the trajectory of coastal wetland restoration" Dataset OverviewThis repository contains the remote sensing classification maps, source data, and analysis code supporting the findings of the manuscript Hydro-geomorphological drivers across scales shape the trajectory of coastal wetland restoration submitted to Nature Communications. The files are organized into raw geospatial data, underlying figure data, and the associated scripts required for spatial and statistical analyses. Repository StructureGEE_Classification_Code.js: Google Earth Engine (GEE) JavaScript code used to generate the 3-year window tidal wetland classification maps.Classification_Maps_TIF/: Folder containing 69 GeoTIFF files representing the latest time-period remote sensing classification results for all study sites.Source_Data_Files/: A set of .xlsx files (Fig1_source_data.xlsx, Fig2b_source_data.xlsx, Fig2d_source_data.xlsx, Fig2f_source_data.xlsx, Fig3_source_data.xlsx) containing the raw tabular data used to generate the respective figures.R_Analysis_Scripts/: A set of .R scripts (Fig2b_analysis.R, Fig2d_analysis.R, Fig2f_analysis.R) used to perform statistical models and generate the plots.1. Remote Sensing Data (GeoTIFFs)Coordinate Reference System (CRS): WGS 84 (EPSG:4326)Resolution: 30 meters (derived from Landsat 5/7/8)Pixel Values: * 0 = Background / No Data1 = Vegetation2 = Mudflat3 = WaterUsage: These maps represent the final observational window for each site. Users should exclude pixels with a value of 0 when calculating areal statistics.2. Source Data DescriptionThe underlying data used to generate the figures and perform statistical analyses are provided as separate .xlsx files: Fig1_source_data.xlsx: Contains geographical coordinates (longitude, latitude) and site-level compositional attributes (e.g., recent_tidal_flat_composition, recent_tidal_marsh_composition, tidal_wetland_extent_change_rate) used for mapping the global distribution and status of the study sites.Fig2b_source_data.xlsx: Contains site-level variables including tidal_wetland_extent_change_rate, total_suspended_matter, tidal_range, and the categorization of wetland change status. This dataset is used to analyze the relationship between wetland persistence/extent change and sediment availability.Fig2d_source_data.xlsx: Contains the extracted variables (recent_tidal_marsh_composition and relative_elevation) required for the robust GAM (beta regression) analysis, assessing the nonlinear effect of relative elevation on tidal marsh composition.Fig2f_source_data.xlsx: Contains site-level variables including initial_tidal_marsh_change_rate and tidal_creek_density used to model the linear relationship between hydro-geomorphological features and early-stage marsh colonization.Fig3_source_data.xlsx: Contains the global gridded data and threshold variables (e.g., tsm_threshold and relative_elevation) used for the sensitivity scenario analysis and global potentially restorable wetland area calculations.3. Code and Software RequirementsGoogle Earth Engine Code: The GEE_Classification_Code.js file requires a registered Google Earth Engine account. It can be executed directly within the GEE Code Editor.R Analysis Code: The provided R scripts (Fig2b_analysis.R, Fig2d_analysis.R, and Fig2f_analysis.R) require R (version 4.0 or higher) and RStudio.Required R Packages: ggplot2, dplyr, readxl, grid, mgcv (for beta regression GAMs), MASS (for ordinal logistic regression), FSA (for Kruskal-Wallis post-hoc Dunn tests), ggsignif (for plot annotations), and nnet (for multinomial logistic regression robustness checks). The brant package is also optionally used to test the proportional odds assumption. Users must ensure these packages are installed prior to execution.Execution: Ensure that each R script and its corresponding source data file (e.g., Fig2b_source_data.xlsx) are placed in the same working directory before running the code.4. LicenseThis dataset is made available under the Creative Commons Attribution 4.0 International License (CC-BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.