Types of Maps: 12 Map Types, What Each Shows and When to Use It

There are dozens of names for maps, but almost every map belongs to one of two families. Reference maps show many features at once so you can find places and your way between them. Thematic maps pick one subject, like population density, election results or rainfall, and turn the rest of the geography into a quiet backdrop.

This guide covers the 12 types of maps you will meet most often: five reference types and seven thematic types. For each one it explains what the map shows, what data it needs and when to use it, and it links to our detailed guides where we have them. The last sections cover how to choose between thematic types and how to build one with the MapAtlas Maps API.

Reference maps and thematic maps

A reference map, such as an atlas page or a hiking map, shows many features at once and treats them as roughly equal. It answers "what is here?". A thematic map demotes the geography to a base layer and promotes a single dataset to the foreground. It answers "where is this concentrated, and what is the pattern?".

The two families work together. A thematic map almost always sits on top of a muted reference map, often greyed-out borders or a soft road network, so the eye goes to the colours, dots or symbols that carry the data. Every flat map in both families also uses a map projection, which decides how the round Earth is stretched onto the page.

The 12 types of maps at a glance

Reference maps

1. Political maps

A political map shows borders, capitals and major cities, with each country or administrative unit in its own flat colour. It is the classic wall map in a classroom. The colours only separate neighbours; they carry no data, which is the main difference from a choropleth map that looks similar at first sight.

2. Physical maps

A physical map shows the natural landscape: mountain ranges, plains, rivers, lakes and coastlines. Relief is usually drawn with hypsometric tints, a colour ramp from green lowlands through brown uplands to white peaks, often combined with hillshading. Physical maps work at small scales, a country or a continent, so you see the shape of the land but cannot measure a single hill.

3. Topographic maps

A topographic map shows the same landscape at large scale, usually 1:25,000 or 1:50,000, and measures it with contour lines that join points of equal height. It adds water, vegetation, roads, buildings and names, which is why hikers, engineers and planners all use it. How to read the contours is covered in our guide to contour lines, and the elevation tool returns the height of any point.

4. Road maps

A road map puts the road network first: motorways, main and minor roads, place names and points of interest such as fuel stations. The default style of most web maps is a road map, because navigation is what most people open a map for.

5. Cadastral maps

A cadastral map shows land parcels: the boundary and the identifier of every plot, as recorded by a national cadastre or land registry. It is the base for property transactions, taxation and planning permission. Cadastral maps are drawn at very large scales, often 1:500 to 1:5,000, because a boundary error of a few metres matters to the owners.

Other reference types follow the same idea for a specific job. Nautical charts show water depths, hazards and buoys. Aeronautical charts show airspace classes and navigation aids.

Thematic maps

6. Choropleth maps

A choropleth map fills predefined regions, such as countries, provinces, postcodes or hex grids, with a colour that encodes a value. Voter turnout per region is a classic example. Always normalise the data first: raw counts in a choropleth inflate large regions and hide patterns in small ones. The choropleth map guide covers classification and colour schemes in detail.

7. Dot density maps

A dot density map places one dot for every N occurrences of a phenomenon, for example one dot per 1,000 people. The eye reads the cloud of dots as density without any numbers on the map. It suits raw counts over an area and avoids the area bias of choropleths. See the dot density map guide for how to choose the dot value and place the dots.

8. Proportional symbol maps

A proportional symbol map places a circle, or another shape, at each location and sizes it by a value: customers per store, population per city, cameras per country. The reader can compare two places directly by comparing two circles, regardless of how much land each covers. The proportional symbol map guide explains why the circle area, and not the radius, should carry the value.

9. Cartograms

A cartogram resizes regions so that their area matches the data, so a country with a large population becomes large on the map regardless of its physical size. It is the strongest correction for the area bias of choropleths, at the cost of distorted shapes. Our cartogram guide builds one from our own enforcement camera counts.

10. Isoline maps

An isoline map draws lines of equal value across a continuous surface. Contour lines on a topographic map, isobars on a weather chart and isotherms on a climate map are all isolines. When the value only exists per area, such as population density, the result is called an isopleth map. The contour lines guide explains how to read them and how they are computed.

11. Flow maps

A flow map draws movement between places as lines or arrows whose width is proportional to the volume: commuters between towns, freight between ports, migration between countries. Charles Joseph Minard's map of Napoleon's 1812 campaign in Russia is the best-known historical example. Flow maps need origin and destination pairs with a value for each pair, and they become unreadable quickly, so aggregate small flows or show only the largest ones.

12. Heat maps

A web heat map smooths individual points into a continuous coloured surface, so dense clusters glow and empty areas fade out. It is a quick way to show where activity concentrates, such as taxi pickups in a city or incidents along a road network. A heat map shows relative density only, so it suits exploration better than exact comparison, and the result depends heavily on the smoothing radius you choose.

Other map types worth knowing

A few types bend the categories. Schematic maps, like the London Underground map, space stations for legibility instead of true position, because connections matter more than distance. Dasymetric maps refine a choropleth with ancillary data, such as land use, so the value sits where it actually occurs within a region. Routing-derived maps, such as isochrones, show the area reachable from a point within a time budget; the Isochrone API returns those polygons ready to render. Real-time maps overlay live vehicle positions, weather radar or transit arrivals on a web map.

Choosing the right thematic map type

Start with the data:

• A rate or percentage per region: choropleth map.
• A raw count at point locations: proportional symbol map.
• A raw count spread over an area: dot density map.
• A continuous variable, such as temperature or elevation: isoline map.
• Movement between places: flow map.
• Many individual points, where only clustering matters: heat map.
• Physical area would distort the message: cartogram.

Then pick a colour scheme. Sequential schemes (light to dark) work for ordered data. Diverging schemes, two colours meeting at a midpoint, work for data with a meaningful zero, like change against a baseline. Categorical palettes are for unordered categories. The ColorBrewer palettes remain the standard reference for all three. Finish with a legend that explains the classes or symbol sizes; the map legend guide covers the layouts that work.

Building a thematic map with MapAtlas

The MapAtlas Maps API serves vector tiles and styling controls suitable for any of the thematic types above. You load a base style with administrative boundaries, fetch your data as GeoJSON, and add a fill layer for choropleths, a circle layer for proportional symbols, or a heatmap layer for density.

For administrative regions, host your boundaries as GeoJSON, attach the value you want to map as a property on each feature, and add a fill layer with a data-driven colour expression that maps the property to a colour ramp. For point data, attach the value as a property and use a circle-radius expression sized by the value, with a circle-stroke for legibility on busy maps.

For interactive analysis, such as clicking a region to see exact numbers, filtering by category or animating over time, pair the Maps API with a small client-side state layer. For one-off static maps for blog posts or reports, the same API can render a static image you can save as PNG. To place a symbol for an address, turn it into latitude and longitude with the coordinates tool.

Common mistakes to avoid

The most common failure is using raw counts in a choropleth: a region twice as big will look twice as important. The second is over-classifying with too many colour bins, which hides the pattern; five to seven classes are usually enough. The third is ignoring colour blindness: the default red-green palette is hard to read for around 1 in 12 men and 1 in 200 women.

Whatever the type, a map is a piece of communication. Pick the type that matches your data, keep the encoding honest, and the reader should see the pattern in seconds.