Skip to main content

​
Overview

This guide demonstrates advanced plotting techniques for OpenLand visualizations, including customizing plot appearance, applying ggplot2 themes, exporting high-resolution figures, and combining multiple visualizations.

​
Understanding OpenLand Plot Architecture

OpenLand uses S4 classes and methods for plotting:
  • Interval plots: Display change rates across time intervals
  • Category plots: Show gains/losses for all categories
  • Transition plots: Detail specific category transitions
All plots are built with ggplot2 and gridExtra, allowing extensive customization.

​
Customizing Plot Appearance

​
Basic Plot Parameters

The plot() method accepts several key arguments: 
library(OpenLand)

# Load example data (use your own data here)
data(SL_2002_2014)

testSL <- intensityAnalysis(
  dataset = SL_2002_2014,
  category_n = "Ap",
  category_m = "SG"
)

# Basic plot with custom labels
plot(testSL$category_lvlGain,
     labels = c(
       leftlabel = bquote("Gain Area (" ~ km^2 ~ ")"),
       rightlabel = "Intensity Gain (%)"
     ),
     labs = c("Categories", "Uniform Intensity"),
     marginplot = c(.3, .3),
     leg_curv = c(x = 1, y = .5),
     fontsize_ui = 8)

​
Key Parameters Explained

​
labels

Axis titles for left and right panels: 
# Using plain text
labels = c(
  leftlabel = "Area (square kilometers)",
  rightlabel = "Change Intensity (percent)"
)

# Using mathematical expressions with bquote()
labels = c(
  leftlabel = bquote("Area (" ~ km^2 ~ ")"),
  rightlabel = bquote("Intensity (" ~ "%·yr"^-1 ~ ")")
)

# Using plotmath for complex equations
labels = c(
  leftlabel = bquote("Area (" ~ 10^3 ~ "ha)"),
  rightlabel = expression(paste("Rate (", km^2, "/", yr, ")"))
)

​
title

Add a main title to the plot: 
plot(testSL$interval_lvl,
     title = "Interval Level Analysis: São Lourenço Basin",
     labels = c(
       leftlabel = "Interval Change Area (%)",
       rightlabel = "Annual Change Area (%)"
     ))

​
labs

Legend labels: 
# For interval plots
labs = c(type = "Change Type", ur = "Uniform Rate")

# For category/transition plots
labs = c(type = "Land Cover Categories", ur = "Expected Rate")

​
marginplot

Adjusts spacing between left and right panels: 
# Negative values pull panels closer
marginplot = c(lh = -10, rh = 0)  # For interval plots

# Positive values create more space
marginplot = c(lh = 0.5, rh = 0.5)  # For category/transition plots

# Fine-tune based on your data
marginplot = c(.3, .3)  # Common balanced setting

​
leg_curv

Controls the arrow pointing to the uniform intensity line: 
# x: controls horizontal arrow length
# y: controls vertical arrow position

# Short arrow, lower position
leg_curv = c(x = 0.1, y = 0.1)

# Long arrow, centered
leg_curv = c(x = 5/10, y = 5/10)

# Position relative to plot dimensions
leg_curv = c(x = 1, y = 0.5)

​
fontsize_ui

Size of uniform intensity percentage text: 
# Small text for crowded plots
fontsize_ui = 7

# Default size
fontsize_ui = 10

# Larger text for presentations
fontsize_ui = 14

​
color_bar (Interval plots only)

Colors for fast/slow change bars: 
plot(testSL$interval_lvl,
     color_bar = c(
       fast = "#B22222",   # Red for fast change
       slow = "#006400",   # Dark green for slow change  
       area = "gray40"     # Gray for area bars
     ))

# Using R color names
plot(testSL$interval_lvl,
     color_bar = c(
       fast = "firebrick",
       slow = "forestgreen",
       area = "steelblue"
     ))

​
Using ggplot2 Themes

OpenLand plots accept ggplot2 theme arguments via ...:

​
Built-in Themes

# Classic theme (white background, no grid)
plot(testSL$category_lvlGain,
     labels = c(
       leftlabel = bquote("Gain (" ~ km^2 ~ ")"),
       rightlabel = "Intensity (%)"
     ),
     theme_classic())

# Minimal theme
plot(testSL$category_lvlGain,
     labels = c(
       leftlabel = bquote("Gain (" ~ km^2 ~ ")"),
       rightlabel = "Intensity (%)"
     ),
     theme_minimal())

# Black and white theme
plot(testSL$category_lvlGain,
     labels = c(
       leftlabel = bquote("Gain (" ~ km^2 ~ ")"),
       rightlabel = "Intensity (%)"
     ),
     theme_bw())

​
Custom Theme Elements

library(ggplot2)

# Customize text elements
plot(testSL$category_lvlGain,
     labels = c(
       leftlabel = bquote("Gain (" ~ km^2 ~ ")"),
       rightlabel = "Intensity (%)"
     ),
     # ggplot2 theme customizations
     axis.text = element_text(size = 12, color = "black"),
     axis.title = element_text(size = 14, face = "bold"),
     legend.text = element_text(size = 11),
     legend.title = element_text(size = 12, face = "bold"),
     panel.background = element_rect(fill = "white"),
     panel.grid.major = element_line(color = "gray90"),
     panel.grid.minor = element_blank())

​
Publication-Ready Theme

# Create a reusable theme function
pub_theme <- function() {
  theme_bw() +
  theme(
    axis.text = element_text(size = 11, color = "black"),
    axis.title = element_text(size = 12, face = "bold"),
    legend.text = element_text(size = 10),
    legend.title = element_text(size = 11, face = "bold"),
    legend.position = "right",
    panel.grid.minor = element_blank(),
    strip.background = element_rect(fill = "gray95", color = "black"),
    strip.text = element_text(size = 11, face = "bold")
  )
}

# Apply to plots
plot(testSL$category_lvlGain,
     labels = c(
       leftlabel = bquote("Gain Area (" ~ km^2 ~ ")"),
       rightlabel = "Intensity Gain (%)"
     ),
     pub_theme())

​
Exporting High-Resolution Figures

​
Saving Individual Plots

OpenLand plots use grid.arrange(), so use appropriate save functions: 
library(ggplot2)
library(gridExtra)

# Method 1: Using png()
png("category_gain.png", 
    width = 3000, 
    height = 2000, 
    res = 300)  # 300 DPI for publication
plot(testSL$category_lvlGain,
     labels = c(
       leftlabel = bquote("Gain Area (" ~ km^2 ~ ")"),
       rightlabel = "Intensity Gain (%)"
     ))
dev.off()

# Method 2: Using pdf() for vector graphics
pdf("category_gain.pdf", 
    width = 10,   # Width in inches
    height = 6.5)  # Height in inches
plot(testSL$category_lvlGain,
     labels = c(
       leftlabel = bquote("Gain Area (" ~ km^2 ~ ")"),
       rightlabel = "Intensity Gain (%)"
     ))
dev.off()

# Method 3: Using tiff() for maximum quality
tiff("category_gain.tiff",
     width = 3000,
     height = 2000,
     res = 300,
     compression = "lzw")
plot(testSL$category_lvlGain,
     labels = c(
       leftlabel = bquote("Gain Area (" ~ km^2 ~ ")"),
       rightlabel = "Intensity Gain (%)"
     ))
dev.off()
For journals: 
# High resolution TIFF
tiff("figure1.tiff", width = 3000, height = 2000, res = 300, compression = "lzw")
plot(testSL$interval_lvl)
dev.off()
For presentations: 
# PNG with moderate resolution
png("slide_figure.png", width = 1920, height = 1080, res = 150)
plot(testSL$interval_lvl, fontsize_ui = 14)
dev.off()
For web: 
# Smaller PNG
png("web_figure.png", width = 1200, height = 800, res = 96)
plot(testSL$interval_lvl)
dev.off()
For vector graphics: 
# PDF (scalable, no quality loss)
pdf("figure1.pdf", width = 10, height = 6.5)
plot(testSL$interval_lvl)
dev.off()

​
Combining Multiple Visualizations

​
Creating Multi-Panel Figures

Since OpenLand plots use grid.arrange() internally, save plots first: 
library(gridExtra)
library(grid)

# You cannot directly save OpenLand plot() outputs
# Instead, create combined figures with other plot types

# Example: Combine intensity plot with net-gross plot
png("combined_analysis.png", width = 4000, height = 2000, res = 300)

par(mfrow = c(1, 2))

# First panel: interval level
plot(testSL$interval_lvl,
     labels = c(
       leftlabel = "Interval Change (%)",
       rightlabel = "Annual Change (%)"
     ))

# Second panel created separately
# (OpenLand plots cannot be easily combined in grid.arrange)
# Best approach: save separately and combine in image editor

dev.off()

​
Sequential Plot Export

# Save all analysis levels systematically
output_dir <- "./figures/"
dir.create(output_dir, showWarnings = FALSE)

# Interval level
pdf(paste0(output_dir, "01_interval_level.pdf"), width = 10, height = 6)
plot(testSL$interval_lvl,
     title = "A) Interval Level Analysis",
     labels = c(
       leftlabel = "Interval Change Area (%)",
       rightlabel = "Annual Change Area (%)"
     ))
dev.off()

# Category gain
pdf(paste0(output_dir, "02_category_gain.pdf"), width = 10, height = 6)
plot(testSL$category_lvlGain,
     title = "B) Category Level - Gains",
     labels = c(
       leftlabel = bquote("Gain Area (" ~ km^2 ~ ")"),
       rightlabel = "Intensity Gain (%)"
     ))
dev.off()

# Category loss
pdf(paste0(output_dir, "03_category_loss.pdf"), width = 10, height = 6)
plot(testSL$category_lvlLoss,
     title = "C) Category Level - Losses",
     labels = c(
       leftlabel = bquote("Loss Area (" ~ km^2 ~ ")"),
       rightlabel = "Loss Intensity (%)"
     ))
dev.off()

# Transition gain
pdf(paste0(output_dir, "04_transition_gain.pdf"), width = 10, height = 6)
plot(testSL$transition_lvlGain_n,
     title = "D) Transition Level - Gain of Pasture",
     labels = c(
       leftlabel = bquote("Gain of Ap (" ~ km^2 ~ ")"),
       rightlabel = "Intensity Gain of Ap (%)"
     ))
dev.off()

# Transition loss
pdf(paste0(output_dir, "05_transition_loss.pdf"), width = 10, height = 6)
plot(testSL$transition_lvlLoss_m,
     title = "E) Transition Level - Loss of Savannah",
     labels = c(
       leftlabel = bquote("Loss of SG (" ~ km^2 ~ ")"),
       rightlabel = "Intensity Loss of SG (%)"
     ))
dev.off()

print("All figures saved successfully!")

​
Customizing Other Visualization Types

​
Net-Gross Plot Customization

# Basic customization
netgrossplot(
  dataset = SL_2002_2014$lulc_Multistep,
  legendtable = SL_2002_2014$tb_legend,
  xlab = "Land Cover Category",
  ylab = bquote("Area (" ~ km^2 ~ ")"),
  changesLabel = c(
    GC = "Gross Changes", 
    NG = "Net Gain", 
    NL = "Net Loss"
  ),
  color = c(
    GC = "gray70", 
    NG = "#006400", 
    NL = "#EE2C2C"
  )
)

# Save with high resolution
png("netgross_plot.png", width = 3000, height = 2000, res = 300)
netgrossplot(
  dataset = SL_2002_2014$lulc_Multistep,
  legendtable = SL_2002_2014$tb_legend,
  xlab = "Land Cover Category",
  ylab = bquote("Area (" ~ km^2 ~ ")"),
  changesLabel = c(GC = "Gross", NG = "Net Gain", NL = "Net Loss"),
  color = c(GC = "gray70", NG = "darkgreen", NL = "darkred")
)
dev.off()

​
Chord Diagram Customization

# Chord diagrams use the circlize package
# Save as high-resolution PNG
png("chord_diagram.png", width = 3000, height = 3000, res = 300)
chordDiagramLand(
  dataset = SL_2002_2014$lulc_Onestep,
  legendtable = SL_2002_2014$tb_legend
)
dev.off()

# For PDF (vector graphics)
pdf("chord_diagram.pdf", width = 10, height = 10)
chordDiagramLand(
  dataset = SL_2002_2014$lulc_Onestep,
  legendtable = SL_2002_2014$tb_legend
)
dev.off()

​
Sankey Diagram Customization

# Sankey diagrams are interactive HTML widgets
# Save using htmlwidgets package
library(htmlwidgets)

# Create the Sankey
sankey_plot <- sankeyLand(
  dataset = SL_2002_2014$lulc_Multistep,
  legendtable = SL_2002_2014$tb_legend
)

# Save as HTML
saveWidget(sankey_plot, "sankey_multistep.html", selfcontained = TRUE)

# For static image, use webshot
library(webshot)
webshot("sankey_multistep.html", "sankey_multistep.png", 
        vwidth = 1200, vheight = 800)

​
Bar Plot Customization

# Evolution bar plot
png("evolution_barplot.png", width = 3000, height = 2000, res = 300)
barplotLand(
  dataset = SL_2002_2014$lulc_Multistep, 
  legendtable = SL_2002_2014$tb_legend,
  xlab = "Year",
  ylab = bquote("Area (" ~ km^2 ~ ")"),
  area_km2 = TRUE
)
dev.off()

​
Advanced Styling Examples

​
Example 1: Grayscale for Print

# Convert colors to grayscale
SL_2002_2014$tb_legend$color <- gray.colors(
  n = nrow(SL_2002_2014$tb_legend),
  start = 0.9,
  end = 0.1
)

# Plot with grayscale theme
plot(testSL$category_lvlGain,
     labels = c(
       leftlabel = bquote("Gain Area (" ~ km^2 ~ ")"),
       rightlabel = "Intensity Gain (%)"
     ),
     theme_bw(),
     panel.grid = element_blank())

​
Example 2: Large Font for Presentations

plot(testSL$interval_lvl,
     labels = c(
       leftlabel = "Interval Change (%)",
       rightlabel = "Annual Change (%)"
     ),
     fontsize_ui = 18,
     # Increase all text sizes
     axis.text = element_text(size = 16),
     axis.title = element_text(size = 18, face = "bold"),
     legend.text = element_text(size = 14),
     legend.title = element_text(size = 16, face = "bold"),
     strip.text = element_text(size = 16))

​
Example 3: Colorblind-Friendly Palette

# Use colorblind-friendly colors
library(RColorBrewer)

# Get colorblind-safe palette
cb_colors <- brewer.pal(n = nrow(SL_2002_2014$tb_legend), name = "Set2")
SL_2002_2014$tb_legend$color <- cb_colors

plot(testSL$category_lvlGain,
     labels = c(
       leftlabel = bquote("Gain Area (" ~ km^2 ~ ")"),
       rightlabel = "Intensity Gain (%)"
     ))

​
Complete Publication Workflow

library(OpenLand)
library(ggplot2)

# Load and prepare data
data(SL_2002_2014)
testSL <- intensityAnalysis(
  dataset = SL_2002_2014,
  category_n = "Ap",
  category_m = "SG"
)

# Create output directory
dir.create("publication_figures", showWarnings = FALSE)

# Define publication theme
pub_theme <- theme_bw() +
  theme(
    axis.text = element_text(size = 11, color = "black"),
    axis.title = element_text(size = 12, face = "bold"),
    legend.text = element_text(size = 10),
    legend.title = element_text(size = 11, face = "bold"),
    legend.position = "right",
    panel.grid.minor = element_blank(),
    strip.background = element_rect(fill = "white", color = "black"),
    strip.text = element_text(size = 11, face = "bold")
  )

# Export all figures as TIFF (300 DPI)
figures <- list(
  list(name = "Fig1_interval", obj = testSL$interval_lvl,
       labels = c(leftlabel = "Interval Change Area (%)",
                  rightlabel = "Annual Change Area (%)")),
  list(name = "Fig2_category_gain", obj = testSL$category_lvlGain,
       labels = c(leftlabel = bquote("Gain Area (" ~ km^2 ~ ")"),
                  rightlabel = "Intensity Gain (%)")),
  list(name = "Fig3_category_loss", obj = testSL$category_lvlLoss,
       labels = c(leftlabel = bquote("Loss Area (" ~ km^2 ~ ")"),
                  rightlabel = "Loss Intensity (%)")),
  list(name = "Fig4_transition_gain", obj = testSL$transition_lvlGain_n,
       labels = c(leftlabel = bquote("Gain of Ap (" ~ km^2 ~ ")"),
                  rightlabel = "Intensity Gain of Ap (%)")),
  list(name = "Fig5_transition_loss", obj = testSL$transition_lvlLoss_m,
       labels = c(leftlabel = bquote("Loss of SG (" ~ km^2 ~ ")"),
                  rightlabel = "Intensity Loss of SG (%)"))
)

for (fig in figures) {
  tiff(
    filename = paste0("publication_figures/", fig$name, ".tiff"),
    width = 3000,
    height = 2000,
    res = 300,
    compression = "lzw"
  )
  
  plot(
    fig$obj,
    labels = fig$labels,
    marginplot = c(.3, .3),
    leg_curv = c(x = 0.5, y = 0.5),
    fontsize_ui = 10,
    pub_theme
  )
  
  dev.off()
}

print("All publication figures saved!")

​
Tips for Best Results

  1. Always use vector formats (PDF, SVG) when possible - They scale without quality loss
  2. Set appropriate DPI - 300 DPI for print, 150 DPI for presentations, 96 DPI for web
  3. Test your figures - View exported files before submission
  4. Use consistent styling - Create theme functions to ensure uniform appearance
  5. Label clearly - Use mathematical expressions (bquote, expression) for professional labels
  6. Consider colorblind readers - Use colorblind-friendly palettes or add patterns
  7. Size appropriately - Check journal requirements for figure dimensions
  8. Save source code - Keep R scripts to regenerate figures if needed

​
Troubleshooting

Issue: Text cut off in exported figure Solution: Increase plot dimensions or adjust margins 
png("figure.png", width = 3500, height = 2000, res = 300)
plot(testSL$category_lvlGain, marginplot = c(.5, .5))
dev.off()
Issue: Legend overlaps with plot Solution: Adjust leg_curv parameter 
plot(testSL$category_lvlGain, leg_curv = c(x = 0.7, y = 0.3))
Issue: Uniform intensity label not visible Solution: Increase fontsize_ui 
plot(testSL$category_lvlGain, fontsize_ui = 12)
Issue: Colors look different when exported Solution: Use explicit color profiles and check in target medium 
png("figure.png", width = 3000, height = 2000, res = 300, type = "cairo")
plot(testSL$category_lvlGain)
dev.off()

Build docs developers (and LLMs) love

Get started for freeTalk to us