initial commit

man/calculate_ga_aoa.Rd

@@ -0,0 +1,22 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/functions.R
+\name{calculate_ga_aoa}
+\alias{calculate_ga_aoa}
+\title{Calculate Area of Applicability Data}
+\usage{
+calculate_ga_aoa(train_data, test_data, predictors)
+}
+\arguments{
+\item{train_data}{Dataframe. The training data from Washington.}
+
+\item{test_data}{Dataframe. The extrapolation data from Georgia.}
+
+\item{predictors}{Character vector. The list of predictor variable names.}
+}
+\value{
+An \code{sf} object containing the Georgia data with an added 'di' (Dissimilarity Index) column.
+}
+\description{
+Generates the Area of Applicability (AOA) scores (Dissimilarity Index)
+for the Georgia extrapolation dataset based on the Washington training data.
+}

man/combine_forest.Rd

@@ -0,0 +1,28 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/functions.R
+\name{combine_forest}
+\alias{combine_forest}
+\title{Combine Washington and Georgia Forest Data}
+\usage{
+combine_forest(wa_data, ga_data)
+}
+\arguments{
+\item{wa_data}{A data frame containing the Washington forest inventory data.}
+
+\item{ga_data}{A data frame containing the Georgia forest inventory data.}
+}
+\value{
+A single combined data frame with an additional column \code{.id}
+(renamed to "state") indicating the source ("WA" or "GA").
+}
+\description{
+Merges the Washington and Georgia datasets into a single data frame, adding a
+column to identify the source state.
+}
+\examples{
+\dontrun{
+  combined <- combine_forest(wa_raw, ga_raw)
+  table(combined$state)
+}
+
+}

man/create_stats_summary.Rd

@@ -0,0 +1,20 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/functions.R
+\name{create_stats_summary}
+\alias{create_stats_summary}
+\title{Create Statistical Summary of Forest Data}
+\usage{
+create_stats_summary(data)
+}
+\arguments{
+\item{data}{A data frame or sf object containing the forest data.}
+}
+\value{
+A data frame with descriptive statistics (mean, sd, min, max, etc.),
+sorted by descending absolute kurtosis.
+}
+\description{
+Generates descriptive statistics for numeric variables in the dataset,
+excluding spatial coordinates (lat/lon) and year. It sorts the results
+by kurtosis to highlight non-normal distributions.
+}

man/forestedAnalysis-package.Rd

@@ -0,0 +1,15 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/forestedAnalysis-package.R
+\docType{package}
+\name{forestedAnalysis-package}
+\alias{forestedAnalysis}
+\alias{forestedAnalysis-package}
+\title{forestedAnalysis: Spatial Cross-Validation and AOA Analysis of Forest Cover}
+\description{
+A research compendium analyzing forest cover data in Washington and Georgia. It evaluates the Area of Applicability (AOA) and demonstrates model failure during spatial extrapolation.
+}
+\author{
+\strong{Maintainer}: Rob Wiederstein \email{khuon68@gmail.com}
+
+}
+\keyword{internal}

man/get_epa_ecoregions.Rd

@@ -0,0 +1,23 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/functions.R
+\name{get_epa_ecoregions}
+\alias{get_epa_ecoregions}
+\title{Download EPA Level III Ecoregions Data}
+\usage{
+get_epa_ecoregions(url, dest_dir = "data/epa")
+}
+\arguments{
+\item{url}{Character string. The direct URL to the EPA Ecoregions zip file.}
+
+\item{dest_dir}{Character string. The local directory where the data should
+be saved. Defaults to "data/epa".}
+}
+\value{
+A character string containing the full file path to the downloaded zip file.
+This return value is designed to be tracked by \code{targets}.
+}
+\description{
+Downloads the EPA Level III Ecoregions shapefile (zip format)
+to a local directory. Implements a caching check to avoid re-downloading
+if the file already exists.
+}

man/helper_save_fig.Rd

@@ -0,0 +1,35 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/functions.R
+\name{helper_save_fig}
+\alias{helper_save_fig}
+\title{Helper: Save Plot for Quarto Slide}
+\usage{
+helper_save_fig(
+  plot_obj,
+  name,
+  type = c("map", "plot"),
+  width = 10,
+  height = 6.18,
+  dpi = 300
+)
+}
+\arguments{
+\item{plot_obj}{The ggplot object to save.}
+
+\item{name}{A short descriptive name (e.g., "wa_ecoregions").}
+
+\item{type}{Either "map" or "plot". Adds this prefix to the filename.}
+
+\item{width}{Width in inches (default: 10).}
+
+\item{height}{Height in inches (default: 6.18).}
+
+\item{dpi}{Resolution (default: 300).}
+}
+\value{
+The full file path (invisibly).
+}
+\description{
+Saves a ggplot object as a PNG, sized to fit comfortably
+below a standard slide title, with robust font handling.
+}

man/plot_ecoregion_comparison.Rd

@@ -0,0 +1,20 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/functions.R
+\name{plot_ecoregion_comparison}
+\alias{plot_ecoregion_comparison}
+\title{Plot Ecoregion Complexity Comparison (WA vs GA)}
+\usage{
+plot_ecoregion_comparison(eco_data)
+}
+\arguments{
+\item{eco_data}{An \code{sf} object containing ecoregion polygons. Must contain
+columns \code{STATE_NAME} and \code{US_L3NAME}.}
+}
+\value{
+A \code{patchwork} object containing the combined plot.
+}
+\description{
+Generates a side-by-side comparison of Level III ecoregions for Washington
+and Georgia. It uses a "void" theme, qualitative colors, and carefully tuned
+label repulsion settings to avoid overlapping text.
+}

man/plot_failure_mechanism.Rd

@@ -0,0 +1,23 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/functions.R
+\name{plot_failure_mechanism}
+\alias{plot_failure_mechanism}
+\title{Plot Failure Mechanism Comparison}
+\usage{
+plot_failure_mechanism(aoa_data, pred_data, boundaries)
+}
+\arguments{
+\item{aoa_data}{Dataframe containing the AOA results (must have 'di', 'lon', 'lat').}
+
+\item{pred_data}{Dataframe containing prediction results (columns: .pred_class, forested, lon, lat).}
+
+\item{boundaries}{An \code{sf} object containing state boundaries (must include "GA" or "Georgia").}
+}
+\value{
+A \code{patchwork} object containing the side-by-side comparison.
+}
+\description{
+Creates a side-by-side diagnostic plot returning a patchwork object.
+(a) The Area of Applicability (Dissimilarity Index) showing where the model is extrapolating.
+(b) The spatial distribution of actual classification errors.
+}

man/plot_ga_comparison_map.Rd

@@ -0,0 +1,21 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/functions.R
+\name{plot_ga_comparison_map}
+\alias{plot_ga_comparison_map}
+\title{Plot Georgia Forest Comparison}
+\usage{
+plot_ga_comparison_map(pred_data, boundaries)
+}
+\arguments{
+\item{pred_data}{Dataframe containing prediction results (columns: .pred_class, forested, lon, lat).}
+
+\item{boundaries}{An \code{sf} object containing state boundaries (must include "GA" or "Georgia").}
+}
+\value{
+A \code{patchwork} object containing the labeled comparison plot.
+}
+\description{
+Creates a side-by-side comparison of forest cover for Georgia.
+The left plot (a) is the Model Prediction, and the right plot (b) is the Actual Data.
+Features a shared right-side legend and standardized spatial styling.
+}

man/plot_georgia_aoa.Rd

@@ -0,0 +1,17 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/functions.R
+\name{plot_georgia_aoa}
+\alias{plot_georgia_aoa}
+\title{Plot Georgia Area of Applicability (AOA)}
+\usage{
+plot_georgia_aoa(aoa_sf)
+}
+\arguments{
+\item{aoa_sf}{An \code{sf} object containing the 'di' column (output of \code{calculate_ga_aoa}).}
+}
+\value{
+A \code{ggplot} object showing the Dissimilarity Index map.
+}
+\description{
+Plots the pre-calculated Dissimilarity Index (DI) for Georgia.
+}

man/plot_precip_hex_comparison.Rd

@@ -0,0 +1,34 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/functions.R
+\name{plot_precip_hex_comparison}
+\alias{plot_precip_hex_comparison}
+\title{Plot Annual Rainfall Comparison (Clipped Hexes)}
+\usage{
+plot_precip_hex_comparison(
+  wa_data,
+  ga_data,
+  boundaries,
+  bins = 30,
+  max_limit = 2500
+)
+}
+\arguments{
+\item{wa_data}{Dataframe containing Washington data (requires 'precip_annual', 'lat', 'lon').}
+
+\item{ga_data}{Dataframe containing Georgia data (requires 'precip_annual', 'lat', 'lon').}
+
+\item{boundaries}{An \code{sf} object containing state boundaries.}
+
+\item{bins}{Integer. Number of hexes across the state width. Default is 30.}
+
+\item{max_limit}{Numeric. The visual cap for rainfall (mm) to ensure comparable scales. Default is 2500.}
+}
+\value{
+A \code{patchwork} object containing the side-by-side comparison.
+}
+\description{
+Creates a polished side-by-side comparison of annual precipitation.
+Hexagons are spatially generated and clipped to the exact state boundaries
+to eliminate "bleeding" edges. Uses \code{theme_forestry_void} with explicit
+font sizing to match topographic maps.
+}

man/plot_regional_comparison.Rd

@@ -0,0 +1,23 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/functions.R
+\name{plot_regional_comparison}
+\alias{plot_regional_comparison}
+\title{Plot Regional Comparison of Forested Data (WA vs GA)}
+\usage{
+plot_regional_comparison(data, boundaries)
+}
+\arguments{
+\item{data}{A data frame containing the forest point data. Must contain
+columns \code{lon}, \code{lat}, \code{state}, and \code{forested}.}
+
+\item{boundaries}{An \code{sf} object containing state boundaries. Must contain
+a \code{NAME} column.}
+}
+\value{
+A \code{patchwork} object containing the combined side-by-side maps.
+}
+\description{
+Generates a side-by-side comparison of forest cover for Washington
+and Georgia. It handles font registration (Atkinson Hyperlegible), spatial
+transformations, and creates a combined plot with a shared legend.
+}

man/plot_rf_importance.Rd

@@ -0,0 +1,17 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/functions.R
+\name{plot_rf_importance}
+\alias{plot_rf_importance}
+\title{Plot Random Forest Variable Importance}
+\usage{
+plot_rf_importance(data)
+}
+\arguments{
+\item{data}{An sf object or data frame containing the 'forested' target and predictors.}
+}
+\description{
+Fits a ranger Random Forest model to the provided data, calculates
+permutation importance, and generates a lollipop chart. It distinguishes
+between spatial (lat/lon) and biophysical predictors.
+Uses the project's 'Atkinson' font theme via theme_forestry_plot().
+}

man/plot_spatial_exploration.Rd

@@ -0,0 +1,18 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/functions.R
+\name{plot_spatial_exploration}
+\alias{plot_spatial_exploration}
+\title{Plot Spatial Autocorrelation Exploration}
+\usage{
+plot_spatial_exploration(wa_data)
+}
+\arguments{
+\item{wa_data}{A dataframe or tibble containing elevation, lat, and lon columns.}
+}
+\value{
+A ggplot object showing standardized elevation vs. spatially lagged elevation.
+}
+\description{
+Generates a Moran Scatterplot to visualize spatial
+autocorrelation in elevation data using a 5km neighborhood.
+}

man/plot_state_topo.Rd

@@ -0,0 +1,11 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/functions.R
+\name{plot_state_topo}
+\alias{plot_state_topo}
+\title{Create Single State Topo Plot}
+\usage{
+plot_state_topo(data, boundary_sf, raster_path, state_name)
+}
+\description{
+Generates a topo map with manually tuned label placement for Georgia.
+}

man/plot_theme_diagnostic.Rd

@@ -0,0 +1,11 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/functions.R
+\name{plot_theme_diagnostic}
+\alias{plot_theme_diagnostic}
+\title{Simplified Theme Diagnostic}
+\usage{
+plot_theme_diagnostic()
+}
+\description{
+Uses built-in NC data to verify theme_forestry_spatial.
+}

man/plot_us_map.Rd

@@ -0,0 +1,11 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/functions.R
+\name{plot_us_map}
+\alias{plot_us_map}
+\title{Plot US Map with Forestry Theme}
+\usage{
+plot_us_map()
+}
+\description{
+Highlights Washington and Georgia using standardized presentation fonts.
+}

man/process_ecoregions.Rd

@@ -0,0 +1,32 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/functions.R
+\name{process_ecoregions}
+\alias{process_ecoregions}
+\title{Process and Clip EPA Ecoregions}
+\usage{
+process_ecoregions(
+  zip_path,
+  target_states = c("Washington", "Georgia"),
+  simplify_tol = 0.05
+)
+}
+\arguments{
+\item{zip_path}{Character string. The file path to the zipped EPA shapefile.}
+
+\item{target_states}{Character vector. The names of the states to clip the
+ecoregions to. Defaults to \code{c("Washington", "Georgia")}.}
+
+\item{simplify_tol}{Numeric. The simplification tolerance passed to
+\code{rmapshaper::ms_simplify}. Range is 0-1, where higher numbers remove more detail.
+Defaults to 0.05.}
+}
+\value{
+An \code{sf} object containing the processed ecoregions with standardized
+columns \code{US_L3NAME} and \code{STATE_NAME}.
+}
+\description{
+Extracts EPA Level III ecoregion data from a zipped shapefile,
+standardizes column names, and clips the geometry to specified state boundaries.
+It includes robust steps for geometry repair (handling spherical validity),
+small island removal, and simplification for optimized plotting.
+}

man/save_combined_topo.Rd

@@ -0,0 +1,19 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/functions.R
+\name{save_combined_topo}
+\alias{save_combined_topo}
+\title{Save Combined Side-by-Side Topo Plot}
+\usage{
+save_combined_topo(
+  wa_data,
+  ga_data,
+  wa_boundary,
+  ga_boundary,
+  wa_raster_path,
+  ga_raster_path,
+  output_path
+)
+}
+\description{
+Save Combined Side-by-Side Topo Plot
+}

man/save_error_map_png.Rd

@@ -0,0 +1,26 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/functions.R
+\name{save_error_map_png}
+\alias{save_error_map_png}
+\title{Save Model Error Diagnostic Map}
+\usage{
+save_error_map_png(data, boundary_sf, raster_path, output_path)
+}
+\arguments{
+\item{data}{A data frame containing model predictions (must include '.pred_class',
+'forested', '.pred_Yes', 'lon', and 'lat').}
+
+\item{boundary_sf}{An \code{sf} object representing the state boundary.}
+
+\item{raster_path}{Character string. File path to the elevation raster (.tif).}
+
+\item{output_path}{Character string. File path where the PNG will be saved.}
+}
+\value{
+The \code{output_path} (invisible), for integration with \code{targets}.
+}
+\description{
+Generates a diagnostic map highlighting prediction errors. It plots
+misclassified points colored by the magnitude of the error (confidence in the wrong answer)
+over a hillshaded elevation background.
+}

man/save_outlier_map_png.Rd

@@ -0,0 +1,24 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/functions.R
+\name{save_outlier_map_png}
+\alias{save_outlier_map_png}
+\title{Save Outlier Diagnostic Map}
+\usage{
+save_outlier_map_png(data, boundary_sf, raster_path, output_path)
+}
+\arguments{
+\item{data}{A data frame containing the analysis dataset (must include numeric columns and 'forested' factor).}
+
+\item{boundary_sf}{An \code{sf} object representing the state boundary (e.g., Washington).}
+
+\item{raster_path}{Character string. File path to the elevation raster (.tif).}
+
+\item{output_path}{Character string. File path where the PNG will be saved.}
+}
+\value{
+The \code{output_path} (invisible), for integration with \code{targets}.
+}
+\description{
+Generates a diagnostic map highlighting multivariate outliers (Z > 3)
+overlaid on a hillshaded elevation raster. Uses the standardized forestry theme.
+}

man/setup_forestry_fonts.Rd

@@ -0,0 +1,15 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/functions.R
+\name{setup_forestry_fonts}
+\alias{setup_forestry_fonts}
+\title{Register Project Fonts}
+\usage{
+setup_forestry_fonts()
+}
+\value{
+NULL (called for side effects)
+}
+\description{
+Registers 'Atkinson Hyperlegible Next' (Sans) and 'Atkinson Hyperlegible Mono'
+(Monospace) with the sysfonts package for use in R graphics.
+}

man/style_audit_table.Rd

@@ -0,0 +1,21 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/functions.R
+\name{style_audit_table}
+\alias{style_audit_table}
+\title{Style Audit Table (GT)}
+\usage{
+style_audit_table(data, title = NULL, subtitle = NULL)
+}
+\arguments{
+\item{data}{A data frame to be formatted.}
+
+\item{title}{Character string. The title of the table (optional).}
+}
+\value{
+A \code{gt_tbl} object ready for rendering.
+}
+\description{
+Converts a data frame into a formatted \code{gt} table with consistent
+styling for audit reports. Includes row striping, numeric formatting, and
+standardized font sizes.
+}

man/theme_forestry_plot.Rd

@@ -0,0 +1,18 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/functions.R
+\name{theme_forestry_plot}
+\alias{theme_forestry_plot}
+\title{Standard Forestry Plot Theme (Cowplot + Atkinson)}
+\usage{
+theme_forestry_plot(font_size = 14, grid = TRUE)
+}
+\arguments{
+\item{font_size}{Integer. Base font size. Default is 14 (good for slides).}
+
+\item{grid}{Logical. If TRUE, adds a light gray grid (useful for presentations).}
+}
+\description{
+A standardized theme for non-spatial plots (scatter, bar, line).
+Based on cowplot::theme_cowplot(), it includes clean axes and a minimalist look.
+Uses 'Atkinson Hyperlegible Next' for all text.
+}

man/theme_forestry_spatial.Rd

@@ -0,0 +1,11 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/functions.R
+\name{theme_forestry_spatial}
+\alias{theme_forestry_spatial}
+\title{Standardized Spatial Theme (Atkinson)}
+\usage{
+theme_forestry_spatial(base_size = 16)
+}
+\description{
+High-visibility map theme for presentations using Atkinson fonts.
+}

man/theme_forestry_void.Rd

@@ -0,0 +1,11 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/functions.R
+\name{theme_forestry_void}
+\alias{theme_forestry_void}
+\title{Standardized Void Theme (Maximal Data Ink)}
+\usage{
+theme_forestry_void(base_size = 16)
+}
+\description{
+Removes axes/grids for shape-focused maps, but keeps project fonts.
+}