Skip to contents

explodemap

R-CMD-check License: MIT Hugging Face Space Pipeline Studio

explodemap makes exploded-view maps from polygon data. It separates crowded regions for display while keeping every polygon’s shape intact.

Use it when you want to:

  • separate dense counties, municipalities, districts, or service areas
  • build larger grouped layouts such as national region maps
  • inspect one area interactively with focus_map()
  • hand a computed layout to dragmapr for manual composition

Install 

install.packages("explodemap")

# Development version
# install.packages("pak")
# pak::pak("PrigasG/explodemap")

Try It 

shiny::runApp(system.file("shiny/pipeline-studio", package = "explodemap"))

Pipeline Studio is the bridge app for explodemap and dragmapr: compute a layout, refine it by dragging, apply the edits, and export the final map.

Quick Start 

library(sf)
library(explodemap)

sq <- function(x, y, size = 1000) {
  st_polygon(list(rbind(
    c(x, y), c(x + size, y), c(x + size, y + size),
    c(x, y + size), c(x, y)
  )))
}

x <- st_sf(
  id = c("a1", "a2", "b1", "b2"),
  region = c("A", "A", "B", "B"),
  geometry = st_sfc(
    sq(0, 0), sq(3000, 0),
    sq(12000, 0), sq(15000, 0),
    crs = 3857
  )
)

layout <- explode_sf(x, region_col = "region", plot = FALSE)

plot(layout, "both")

Main Workflows

Explode Any Projected sf 

layout <- explode_sf(my_sf, region_col = "district", plot = FALSE)

Input should be polygon or multipolygon sf data in a projected CRS.

Explode A US State 

nj <- explode_state(
  state_fips = "34",
  crs = 32111,
  level = "cousub",
  region_map = list(
    North = c("Bergen", "Essex", "Hudson", "Morris"),
    Central = c("Hunterdon", "Mercer", "Middlesex"),
    South = c("Atlantic", "Camden", "Cape May")
  ),
  plot = FALSE
)

Use level = "county" for county maps, or level = "cousub" for municipal and county-subdivision maps.

Build A Grouped Layout 

grouped <- explode_grouped(
  my_sf,
  region_col = "region",
  mode = "auto_collision",
  plot = FALSE
)

Grouped layouts are useful for national or multi-region maps where entire region blocks need space between them.

Inspect The Result Interactively 

focus_map(
  grouped,
  label_col = "name",
  group_col = "region",
  group_palette = group_palette(grouped$sf_grouped$region),
  show_group_labels = TRUE
)

In Shiny:

ui <- fluidPage(focusmapOutput("map", height = "650px"))

server <- function(input, output, session) {
  output$map <- renderFocusmap({
    focus_map(grouped, label_col = "name", group_col = "region")
  })
}

Prepare And Check Data

Release A adds small helpers that are useful in scripts, apps, and tests:

check <- validate_explodemap_input(
  my_sf,
  group_col = "region",
  id_col = "geoid",
  label_col = "name"
)

prepared <- prepare_explodemap_input(
  my_sf,
  id_col = "geoid",
  label_col = "name",
  group_col = "region",
  target_crs = 5070,
  target_vertices = 350000
)

layout <- explode_grouped(prepared$data, region_col = "region")

Other app-friendly helpers:

count_geometry_vertices(my_sf)
simplify_to_vertex_budget(my_sf, target_vertices = 350000)
assign_spatial_groups(my_sf, method = "clusters", groups = 6)
group_palette(my_sf$region)
explodemap_fingerprint(my_sf, id_col = "geoid", group_col = "region")

Improve A Layout

Automatic layout is collision-aware, but cartographic composition often still needs judgment. Use diagnostics and bounded search before manual editing:

report <- diagnose_layout(grouped, label_col = "name")
plot(report)

better <- optimize_grouped_layout(
  my_sf,
  region_col = "region",
  label_col = "name"
)

Then hand the layout to dragmapr:

library(dragmapr)

state <- as_dragmapr_state(better)
dragmapr_edit(better, state = state)

Render the same edited state in either package:

focus_map(better, state = state)

render_dragged_map(
  better$sf_grouped,
  region_col = "region",
  state = state
)

Common Helpers 

print(layout)
summary(layout)
plot(layout)
plot(layout, "both")
calibration_row(layout)

Export TopoJSON when mapshaper is installed:

export_topojson(layout, "exploded.topojson")

Learn More 

vignette("getting-started", package = "explodemap")
vignette("workflow-guide", package = "explodemap")
vignette("grouped-layouts", package = "explodemap")
vignette("state-first-composition", package = "explodemap")

The method is described in:

Arthur, G. (2026). A Hierarchical Vector-Based Framework for Multi-Scale Exploded-View Cartography: Centroid-Driven Spatial Displacement for Dense Administrative Maps.