leaflet Exercises in R: 20 Real-World Map Problems
Twenty hands-on leaflet exercises grouped by skill: base maps, markers and popups, polygons and choropleths, color palettes and legends, layer controls, and full ops-team dashboards. Every problem is framed around a real ask from analysts, dispatchers, and city planners. Solutions are hidden behind a click so you can attempt each one first.
Section 1. Base maps and tile providers (3 problems)
Exercise 1.1: Render a default OpenStreetMap base layer
Task: A new analyst onboarding to a logistics tool needs a baseline map to confirm the leaflet stack is wired up. Initialize a leaflet widget and add the default OpenStreetMap tiles so the analyst sees a world map at the lowest zoom. Save the result to ex_1_1.
Expected result:
#> A leaflet htmlwidget:
#> - one base tile layer: OpenStreetMap default
#> - view: world extent, default zoom 1, center (0, 0)
#> - no markers, polygons, or overlaysDifficulty: Beginner
An empty map widget shows nothing until you attach a basemap; create the widget first, then give it a tile layer.
Pipe leaflet() into addTiles() called with no arguments to get the default OpenStreetMap tiles.
Click to reveal solution
Explanation: leaflet() returns an empty htmlwidget; tiles are not implied. addTiles() with no arguments calls the OpenStreetMap provider, which is the canonical free baseline. You will compose every other layer (markers, polygons, choropleths) on top of this base. Skipping the tile layer leaves a blank grey canvas, which is a common first-time gotcha.
Exercise 1.2: Center and zoom on a specific city for a courier briefing
Task: Dispatch wants the morning briefing centered on Manhattan at zoom 12 so the courier team sees neighborhood-level streets, not boroughs. Build a leaflet map centered on longitude -74.0060 and latitude 40.7128 at zoom level 12 with default tiles. Save the result to ex_1_2.
Expected result:
#> A leaflet htmlwidget:
#> - tile layer: OpenStreetMap
#> - center: lng = -74.0060, lat = 40.7128 (Manhattan)
#> - zoom: 12 (neighborhood-level streets visible)Difficulty: Beginner
A basemap on its own opens at world extent; you need one more step to pin both the center point and the zoom level.
After addTiles(), call setView() with lng = -74.0060, lat = 40.7128, and zoom = 12.
Click to reveal solution
Explanation: setView() controls both the geographic center and the zoom level simultaneously. Zoom 12 is the sweet spot for urban dispatch: streets are labeled, but a whole borough still fits in view. Compare to fitBounds(), which auto-fits a bounding box and is preferred when you have a set of points whose extent you want shown. Use setView() when the focal point is fixed regardless of data.
Exercise 1.3: Compare three tile providers for a real-estate listing page
Task: A real-estate marketing team is A/B testing the base-map style on listing pages. Render three leaflet maps centered on San Francisco (lng -122.42, lat 37.77, zoom 12) using OpenStreetMap, CartoDB Positron, and Esri WorldImagery. Save the three maps as a named list to ex_1_3 with names "osm", "carto", and "esri".
Expected result:
#> A named list of length 3 holding leaflet htmlwidgets:
#> $osm : OpenStreetMap tiles, SF view
#> $carto : CartoDB.Positron clean light tiles, SF view
#> $esri : Esri.WorldImagery satellite tiles, SF viewDifficulty: Beginner
Each map is built the same way but with a different basemap style; collect the three into a named container.
Use addProviderTiles() with providers$CartoDB.Positron and providers$Esri.WorldImagery, plain addTiles() for OSM, all inside a list() with names osm/carto/esri.
Click to reveal solution
Explanation: addProviderTiles() switches to any of 200+ free tile providers exposed through the providers object. Positron is the de facto choice for analytics dashboards because its muted palette lets data colors pop. Esri WorldImagery is the satellite default. For commercial use, check provider terms: OpenStreetMap requires attribution, and some providers cap free tile requests per day.
Section 2. Markers, popups, and labels (4 problems)
Exercise 2.1: Drop a single marker with a popup for a store opening
Task: A retail expansion lead is announcing the new Brooklyn flagship and wants a one-pin map for the press release. Add one marker at lng -73.9442, lat 40.6782 with a popup that reads "Brooklyn Flagship Opening Sept 2026". Use OpenStreetMap tiles centered on the marker at zoom 13. Save to ex_2_1.
Expected result:
#> A leaflet htmlwidget:
#> - tile layer: OpenStreetMap
#> - center: (-73.9442, 40.6782), zoom 13
#> - one marker with popup "Brooklyn Flagship Opening Sept 2026"Difficulty: Beginner
Set up the basemap and view first, then drop a single point that carries a click message.
Call addMarkers() with lng, lat, and popup = "Brooklyn Flagship Opening Sept 2026".
Click to reveal solution
Explanation: addMarkers() accepts either inline coordinates or a data frame with lng/lat columns. Popups appear on click; for hover behavior you would use label = ... instead. Popups support HTML by default, so something like popup = "<b>Brooklyn</b><br>Sept 2026" will render bold. For untrusted content always sanitize, or pass popup = htmlEscape(text) to avoid XSS through user-submitted strings.
Exercise 2.2: Plot multiple store locations from a data frame
Task: The retail ops team has a tibble of US flagship stores with name, lng, and lat. Plot all five stores as markers on a leaflet map with each popup showing the store name. Use OpenStreetMap tiles and let fitBounds() auto-zoom to show all five. Build the data inline first. Save the map to ex_2_2.
Expected result:
#> A leaflet htmlwidget:
#> - tile layer: OpenStreetMap
#> - 5 markers (NYC, Chicago, LA, Austin, Seattle)
#> - view auto-fit to bounding box of all 5
#> - each popup shows the store name on clickDifficulty: Intermediate
Hand the table to the map up front so columns can be referenced by name, then let the view auto-frame to all points.
Pass stores to leaflet(), use addMarkers(lng = ~lng, lat = ~lat, popup = ~name), then fitBounds() with the min and max of the lng and lat columns.
Click to reveal solution
Explanation: Passing the data frame to leaflet(data = stores) lets you reference columns with formula notation (~lng, ~lat, ~name). This is far cleaner than passing vectors. fitBounds() auto-zooms to the smallest box that contains all points, which is exactly what you want for a multi-store overview. For a single-coast subset you would use setView() with a hardcoded center to keep the framing stable across updates.
Exercise 2.3: Use awesome icons to encode store status
Task: The retail ops dashboard needs to flag store status visually: "open" stores use a green storefront icon, "closing_soon" uses an orange clock, and "closed" uses a red ban icon. Use awesomeIcons() from the FontAwesome library. Build a tibble with three stores, one per status, then render them with status-driven icons. Save to ex_2_3.
Expected result:
#> A leaflet htmlwidget:
#> - 3 awesome-icon markers, one per status
#> - green "store" icon for "open"
#> - orange "clock" icon for "closing_soon"
#> - red "ban" icon for "closed"
#> - popup shows store name on clickDifficulty: Intermediate
Map each status category to a color and a glyph through lookup vectors, then build one vectorized icon definition.
Use awesomeIcons() with library = "fa" and named lookup vectors for icon and markerColor, then render with addAwesomeMarkers(icon = icons).
Click to reveal solution
Explanation: awesomeIcons() builds a vectorized icon definition; you pass parallel vectors for icon, markerColor, and iconColor. Named lookup vectors are the idiomatic way to map a categorical column to icon properties without nested if_else(). library = "fa" selects FontAwesome; alternatives are "ion", "glyphicon", "mdi". Always pick icons that read clearly at the smallest zoom you support, since FA icons rasterize poorly below 16px.
Exercise 2.4: Cluster dense markers for a sales call list
Task: A regional sales rep loaded 80 prospect accounts inside the New York metro and the map is unreadable from the overlap. Use clusterOptions = markerClusterOptions() to collapse overlapping markers into expandable cluster bubbles. Generate the 80 points by sampling around NYC and render with clustering. Save to ex_2_4.
Expected result:
#> A leaflet htmlwidget:
#> - tile layer: OpenStreetMap
#> - 80 prospect markers clustered into ~6 expandable bubbles
#> - clicking a cluster zooms in and splits the bubble
#> - center: NYC, zoom 10Difficulty: Intermediate
When points pile up in a small area, group them so overlapping pins collapse into expandable bubbles.
Pass clusterOptions = markerClusterOptions() to addMarkers(), and setView() on NYC at zoom 10.
Click to reveal solution
Explanation: markerClusterOptions() activates the MarkerCluster plugin, which bundles overlapping markers into a single bubble showing the count. Clicking the bubble zooms in and splits the cluster recursively. This is essential for any dataset over ~50 points in a small area, where individual markers stack and become useless. Tune visual density with maxClusterRadius and disableClusteringAtZoom to control when clusters dissolve into raw markers.
Section 3. Shapes, lines, and polygons (3 problems)
Exercise 3.1: Draw a delivery route as a polyline
Task: A last-mile dispatcher needs to visualize today's planned route between four Manhattan dropoffs in sequence. Build a tibble with the four stops in order, draw a blue polyline connecting them with addPolylines(), and overlay numbered markers at each stop. Use OpenStreetMap tiles, fitBounds() for framing. Save to ex_3_1.
Expected result:
#> A leaflet htmlwidget:
#> - tile layer: OpenStreetMap
#> - 1 blue polyline connecting 4 stops in order
#> - 4 markers labeled "Stop 1", "Stop 2", "Stop 3", "Stop 4"
#> - view auto-fit to route bounding boxDifficulty: Intermediate
Connect the stops in their table order with a line, then layer the points on top and frame the whole route.
Use addPolylines(lng = ~lng, lat = ~lat, color = "blue"), then addMarkers(popup = ~stop), then fitBounds().
Click to reveal solution
Explanation: addPolylines() connects points in the order they appear in the data frame, so route ordering matters. The line is a straight geodesic between consecutive points; for real road-following lines you would call a routing service like OSRM or Mapbox Directions and feed the returned coordinate string back in. Use weight for line thickness and dashArray = "5,5" for dashed lines (useful for planned vs actual routes).
Exercise 3.2: Size circles by sales volume for a territory map
Task: The VP of Sales wants a map where each city's circle radius is proportional to its quarterly revenue, so high-grossing markets are visually obvious. Use addCircles() with radius scaled to revenue (treat radius in meters). Tooltip should show name and revenue. Save to ex_3_2.
Expected result:
#> A leaflet htmlwidget:
#> - 5 circles, radius proportional to revenue
#> - LA circle largest (revenue 5.2M), Austin smallest (1.1M)
#> - hover label shows "<name>: $<revenue>M"
#> - view auto-fit to all 5 citiesDifficulty: Intermediate
Drive each circle's ground size from the revenue column, scaled by a constant so the biggest market reads clearly.
Use addCircles() with radius = ~revenue * 25000 and label = ~paste0(name, ": $", revenue, "M").
Click to reveal solution
Explanation: addCircles() interprets radius in meters on the ground, so the visual size changes with zoom (geographic). For a marker that keeps a constant pixel size regardless of zoom, use addCircleMarkers() where radius is in pixels. The 25000 multiplier is empirical: pick a value so the largest bubble fits the city footprint at your default zoom. For perceptual accuracy, scale by sqrt(revenue) so circle AREA (not radius) is proportional to the metric.
Exercise 3.3: Rectangles for delivery-zone bounding boxes
Task: A logistics planner divides New York into three rectangular delivery zones (Lower Manhattan, Midtown, Uptown) and wants to overlay them as semi-transparent rectangles with hover labels showing the zone name. Use addRectangles() with lng1/lat1/lng2/lat2. Save the map to ex_3_3.
Expected result:
#> A leaflet htmlwidget:
#> - 3 semi-transparent rectangles tiling Manhattan
#> - colors: Lower (red), Midtown (green), Uptown (blue)
#> - hover label shows zone name
#> - view centered on Manhattan, zoom 12Difficulty: Intermediate
Each zone is defined by two opposite corners, so a simple box overlay fits without a coordinate list.
Use addRectangles() with the lng1/lat1/lng2/lat2 columns, fillOpacity = 0.3, and label set to the zone name.
Click to reveal solution
Explanation: addRectangles() is the simplest geometric overlay because you specify two opposite corners, not a coordinate list. For non-rectangular zones use addPolygons() with a list of lng/lat vectors. Note that the data frame's columns are passed as raw vectors here rather than with ~ formula notation: this works when you call leaflet() without data =. Both styles are valid; pick one per project for consistency.
Section 4. Choropleth and color scales (4 problems)
Exercise 4.1: Build a continuous color palette with colorNumeric
Task: A city analyst wants to color neighborhoods on a 0-100 deprivation index, going from light yellow to dark red. Build a continuous palette function with colorNumeric(palette = "YlOrRd", domain = 0:100). Then apply it to a vector of five scores and return both the palette function and the colors as a named list. Save to ex_4_1.
Expected result:
#> List of 2:
#> $ pal : function (x) ... (colorNumeric palette)
#> $ colors: chr [1:5] "#FFFFCC" "#FED976" "#FD8D3C" "#E31A1C" "#800026"
#> (5 colors going light yellow to dark red for scores 10, 30, 55, 80, 95)Difficulty: Intermediate
Build a reusable mapping from a fixed numeric range onto a yellow-to-red color ramp, then apply it to the scores.
Call colorNumeric(palette = "YlOrRd", domain = 0:100) and store both that function and its output on scores in a list().
Click to reveal solution
Explanation: colorNumeric() returns a CLOSURE: a function that, given a numeric vector, returns a vector of hex colors. The domain argument fixes the scale so the same score always maps to the same color across different subsets of your data, which is critical for comparable maps. For discrete bins use colorBin(), for quantile-based binning use colorQuantile(), and for categorical data use colorFactor().
Exercise 4.2: Bin a continuous metric into 5 quantile-based colors
Task: A retail planner wants to color 30 zip codes by foot-traffic into 5 quantile bins so the top quintile pops. Generate 30 random foot-traffic counts, build a 5-quintile palette with colorQuantile(), and return a tibble with the count, the bin index, and the hex color. Save to ex_4_2.
Expected result:
#> # A tibble: 30 x 3
#> foot_traffic bin color
#> <dbl> <int> <chr>
#> 1 2350 3 #FD8D3C
#> 2 8200 5 #BD0026
#> 3 450 1 #FFFFB2
#> ...
#> # 27 more rows hidden; bin values 1-5; colors light yellow to dark redDifficulty: Intermediate
Skewed counts need equal-count bins rather than equal-width ones, so each color holds the same number of zip codes.
Use colorQuantile(palette = "YlOrRd", domain = foot_traffic, n = 5) and derive the bin index with cut() over quantile() breaks.
Click to reveal solution
Explanation: colorQuantile() is the right tool when raw values are highly skewed: equal-width bins (colorBin()) would lump 90% of zip codes into one color and leave the top 10% as outliers. Quantile binning guarantees each bin holds the same number of observations, so the visual contrast is preserved even when distributions are heavy-tailed. The trade-off is that bin boundaries are no longer round numbers; document them in the legend.
Exercise 4.3: Choropleth polygons with a continuous palette and legend
Task: A real-estate marketing team wants a choropleth of 4 neighborhood polygons colored by median home price, with a legend in the bottom-right. Build the neighborhood polygons inline as a tibble of lng/lat vectors per neighborhood. Use colorNumeric for the fill, addPolygons(), and addLegend(). Save to ex_4_3.
Expected result:
#> A leaflet htmlwidget:
#> - 4 polygon overlays, fill color by median_price
#> - palette: YlGnBu (light yellow to dark blue), domain 200k-1.2M
#> - legend in "bottomright" titled "Median Price ($)"
#> - hover label shows neighborhood name and priceDifficulty: Intermediate
Color each neighborhood polygon from a shared price scale, then add a key so the colors are decodable.
Build the palette with colorNumeric("YlGnBu", ...), loop addPolygons(fillColor = pal(price)), and finish with addLegend(position = "bottomright").
Click to reveal solution
Explanation: Iterating with a for loop is the most readable pattern when each polygon has its own vector of coordinates rather than a tidy long table. For a real project you would typically load a shapefile or GeoJSON with sf::st_read() and pass the resulting sf object directly to addPolygons(data = sf_obj, fillColor = ~pal(price)). addLegend() rounds out the choropleth: a colored polygon without a legend is unreadable.
Exercise 4.4: Highlight polygons on hover for an interactive choropleth
Task: The same real-estate map from 4.3 should highlight the polygon under the cursor (thicker border, full opacity) so the user can clearly see which neighborhood they are inspecting. Wrap addPolygons() with highlightOptions(). Reuse the same data. Save to ex_4_4.
Expected result:
#> A leaflet htmlwidget:
#> - same choropleth as ex_4_3
#> - hovering a polygon thickens its border to 3px and lifts opacity to 1.0
#> - hover label still shows name + priceDifficulty: Advanced
Give each polygon a temporary style that activates only while the cursor is hovering over it.
Add highlightOptions = highlightOptions(weight = 3, fillOpacity = 1.0, bringToFront = TRUE) to the addPolygons() call.
Click to reveal solution
Explanation: highlightOptions() swaps in temporary styling on mouseover, then restores the original style on mouseout. bringToFront = TRUE ensures the highlighted polygon overlaps its neighbors visually, which matters when borders touch. This is the single biggest readability win for choropleths shown to non-technical viewers: it converts a static color blob into something users can explore.
Section 5. Layer control and interactive features (3 problems)
Exercise 5.1: Toggle two base maps and two overlay groups
Task: A city dashboard needs the user to toggle between OpenStreetMap and CartoDB.Positron base layers, and to toggle two overlay groups: "Police Stations" (10 markers) and "Fire Stations" (5 markers). Use addLayersControl(). Generate both station tibbles inline by sampling around Chicago. Save the map to ex_5_1.
Expected result:
#> A leaflet htmlwidget:
#> - 2 base layers selectable via radio: "OSM", "Positron"
#> - 2 overlay groups toggleable via checkbox: "Police", "Fire"
#> - 10 blue police markers, 5 red fire markers
#> - layer control collapsed by default in top-rightDifficulty: Advanced
Tag every layer with a group name, then expose those groups as switchable base maps and toggleable overlays.
Pass group = to each addTiles/addProviderTiles/addCircleMarkers call, then addLayersControl(baseGroups = ..., overlayGroups = ...).
Click to reveal solution
Explanation: Layer groups are the leaflet idiom for showing/hiding sets of features as a unit. Pass group = "name" to every add*() call you want grouped, then list those groups in addLayersControl(). baseGroups are mutually exclusive (radio buttons); overlayGroups are independent (checkboxes). For dashboards with many layers, set collapsed = TRUE so the control starts as a small icon and expands on hover.
Exercise 5.2: Add fullscreen and scale-bar controls
Task: The retail leadership reviews territory maps on a projector in the boardroom and needs the map to expand to fullscreen with one click, plus a scale bar so they can eyeball distances. Use addFullscreenControl() from leaflet.extras and addScaleBar() from leaflet. Build a five-store map and add both controls. Save to ex_5_2.
Expected result:
#> A leaflet htmlwidget:
#> - 5 sales markers (NYC, Chicago, LA, Austin, Seattle)
#> - fullscreen button in top-left corner
#> - scale bar in bottom-left, units "metric" (km)
#> - view auto-fit to all 5 citiesDifficulty: Advanced
Build the five-store map first, then bolt on two presentation widgets - one to expand the view, one to show distance.
Add addFullscreenControl(position = "topleft") and addScaleBar(position = "bottomleft", options = scaleBarOptions(imperial = FALSE)).
Click to reveal solution
Explanation: Boardroom and conference-room maps benefit enormously from addFullscreenControl(): a projected map at 1024x768 with a sidebar of controls is unreadable, but fullscreen mode reclaims the entire screen. addScaleBar() is the cheapest credibility win on any geographic figure presented to non-technical stakeholders. Set imperial = TRUE for US audiences; metric for international. The scale bar auto-adjusts to the current zoom level.
Exercise 5.3: Custom HTML popup styling a delivery zone
Task: A logistics ops team wants a polygon overlay for the Newark delivery zone with a click-popup that styles content in HTML: bold zone name, 2-row table of "Orders today" and "Avg delivery time". Use addPolygons() with a custom HTML popup string. Save to ex_5_3.
Expected result:
#> A leaflet htmlwidget:
#> - 1 blue polygon outlining Newark delivery zone
#> - click triggers popup with HTML table:
#> - <b>Newark Zone</b>
#> - Orders today: 142
#> - Avg delivery time: 38 minDifficulty: Advanced
Assemble a small HTML snippet with a bold heading and a two-row table, then attach it as the polygon's click content.
Build the string with paste(...) and pass it to addPolygons() through the popup argument.
Click to reveal solution
Explanation: Popups support arbitrary HTML, so you can embed tables, images (<img>), and links to operational dashboards. For data-driven popups across many polygons, build the HTML strings vectorized with glue::glue() or sprintf() and pass them as a vector. Keep popup HTML lean: every popup is rendered on first click, so embedding large images can slow down dense maps. For richer interactivity (forms, charts), use Shiny instead.
Section 6. Realistic ops-team dashboards (3 problems)
Exercise 6.1: Build a daily delivery-coverage map with depot and route layers
Task: A regional logistics dispatcher wants a daily ops map showing one depot, three delivery zones as rectangles, and the planned routes from depot to each zone center as polylines. Color rectangles by load (light=low, dark=high). Toggle "Zones" and "Routes" via layer control. Use the inline data below. Save the map to ex_6_1.
Expected result:
#> A leaflet htmlwidget:
#> - 1 depot marker (red, "Newark Depot")
#> - 3 zone rectangles fill-colored by load (light yellow to dark red)
#> - 3 dashed polylines from depot to each zone center
#> - layer control: "Zones" and "Routes" toggleable, OSM baseDifficulty: Advanced
Combine a fixed depot, metric-colored zone boxes, and depot-to-zone lines, tagging zones and routes so each can be toggled.
Compute zone centers with mutate(), color with colorNumeric("YlOrRd", ...), loop addRectangles(group = "Zones") and addPolylines(group = "Routes", dashArray = "6,6"), then addLayersControl().
Click to reveal solution
Explanation: This pattern is the backbone of a dispatch dashboard: one fixed asset (depot), several variable assets (zones) styled by a metric (load), and the relationships between them (routes). The loop builds the layers incrementally rather than vectorizing because each zone has unique coordinates. In production you would swap in real polygons (a shapefile) and real road-following routes from a routing API. The toggle control lets dispatchers focus on one perspective at a time.
Exercise 6.2: Crime hotspot heatmap for a city analyst
Task: A city analyst wants to surface crime hotspots from a sample of 200 incident points clustered around three latent epicenters in Chicago. Render a heatmap with addHeatmap() from leaflet.extras, with radius = 20 and blur = 15. Center on Chicago at zoom 11. Save to ex_6_2.
Expected result:
#> A leaflet htmlwidget:
#> - tile layer: CartoDB.DarkMatter
#> - heatmap of 200 incident points, 3 visible hotspots
#> - red core for dense areas, fading through orange/yellow to blue
#> - center: Chicago (-87.63, 41.88), zoom 11Difficulty: Advanced
Turn the scattered incident points into a smooth density surface, and use a dark basemap so the hot cores stand out.
Use addHeatmap(lng = ~lng, lat = ~lat, radius = 20, blur = 15) on top of providers$CartoDB.DarkMatter.
Click to reveal solution
Explanation: addHeatmap() aggregates point density into a smooth gradient using a kernel: radius is the influence radius in pixels and blur softens transitions between bands. The max parameter sets the intensity ceiling and is the single most consequential knob: too high and the map looks flat, too low and a few points saturate. CartoDB.DarkMatter is the conventional base for heatmaps because dark backgrounds make red/orange peaks pop visually. For high-precision hotspot detection, consider addCluster() or formal spatial statistics like Getis-Ord G*.
Exercise 6.3: Multi-layer sales territory dashboard
Task: A VP of Sales wants a single map showing three territories: each territory's polygon colored by Q1 revenue (choropleth), markers for top accounts inside each territory, and the territory headquarters as a starred icon. Toggle "Territories", "Top Accounts", and "HQs" independently. Reuse a single color palette across the choropleth and the legend. Save to ex_6_3.
Expected result:
#> A leaflet htmlwidget:
#> - 3 colored territory polygons (West, Central, East)
#> - 6 top-account circle markers (2 per territory)
#> - 3 starred HQ markers
#> - layer control with 3 overlay groups, all visible by default
#> - legend bottom-right titled "Q1 Revenue ($M)"Difficulty: Advanced
Stack three feature types - colored territory polygons, account points, and HQ icons - each in its own toggleable group sharing one revenue scale.
Build one colorNumeric("YlGnBu", ...) palette, loop addPolygons(group = "Territories") and addAwesomeMarkers(icon = awesomeIcons(icon = "star"), group = "HQs"), add addCircleMarkers(group = "Top Accounts"), then addLegend() and addLayersControl().
Click to reveal solution
Explanation: Production dashboards stack five to ten layer types on one map; this exercise compresses that pattern into three. Two design rules to remember: keep one palette per metric (the legend implicitly trusts a single domain), and always assign every overlay a group = so the user has fine-grained control. For very large datasets (10k+ markers), pair this pattern with marker clustering or feature-group simplification to keep render time under one second.
What to do next
- Brush up on the core API with the leaflet for R tutorial parent post.
- Practice manipulating the data that feeds these maps with dplyr Exercises.
- For shapefile and GeoJSON handling, work through the sf package Exercises.
- Compare with a different mapping toolkit in the ggplot2 Maps Exercises.
r-statistics.co · Verifiable credential · Public URL
This document certifies mastery of
leaflet Mastery
Every certificate has a public verification URL that proves the holder passed the assessment. Anyone with the link can confirm the recipient and date.
315 learners have earned this certificate