A “thermatic map” is almost certainly a search for thematic map, a common misspelling. A thematic map shows the spatial distribution of one or more specific data themes across geographic areas. Instead of simply showing where things are (like a road map or atlas), a thematic map visualizes data tied to those places: population density, average income, election results, rainfall, or crop types. It’s one of the most widely used tools in data visualization, public health, urban planning, and journalism.
How Thematic Maps Differ From Reference Maps
The maps most people grew up with are reference maps (also called base maps). These display natural and man-made features of general interest: roads, rivers, city names, borders. Their job is to help you find a location or understand the physical layout of a place.
A thematic map does something fundamentally different. It takes a single topic, or theme, and shows how that topic varies across space. A reference map of the United States shows state boundaries and capital cities. A thematic map of the United States might shade each state by median household income, instantly revealing which regions are wealthier and which are not. The geography is just the canvas; the data is the point.
Thematic maps can be qualitative or quantitative. A qualitative version might color-code regions by dominant land use (farmland, forest, urban). A quantitative version assigns shading or symbols based on numerical data like percentage population change or crime rates. Many thematic maps combine both, layering categorical boundaries with numerical data on top.
Choropleth Maps
Choropleth maps are the type most people picture when they hear “thematic map.” They shade predefined geographic areas (states, counties, countries) to reflect the value of a variable like population density or birth rate. Darker shading typically means a higher value, lighter means lower. Election maps showing which party won each county are a classic example.
The power of a choropleth map is instant pattern recognition. You can spot regional clusters and outliers at a glance. The tricky part is how the data gets grouped into color categories. If you divide data into five color buckets, the method you use to set the boundaries between those buckets can dramatically change the story the map tells.
Several standard methods exist. Quantile classification puts an equal number of areas into each bucket, so if you have 50 states and five color classes, each class contains exactly 10 states. Equal interval classification divides the range of values into evenly spaced groups, which works well when data is spread evenly but can be misleading when it’s clustered. The Jenks natural breaks method looks for gaps in the data and places boundaries there, minimizing variation within each group while maximizing variation between groups. Jenks is often considered the best default for choropleth maps because it reflects real patterns in the data rather than imposing an arbitrary structure.
Proportional Symbol Maps
Instead of shading entire regions, proportional symbol maps place symbols (usually circles) at specific locations and scale them based on a data value. A city with 5 million people gets a much larger circle than one with 500,000. This approach works especially well for data tied to points rather than areas, like city populations, earthquake magnitudes, or factory output.
The scaling isn’t as simple as making one circle twice as wide for twice the value. Because humans perceive area, not diameter, a circle representing twice the data should have twice the area. This requires a square root adjustment to the diameter. Research on visual perception has shown that people still tend to underestimate the size of larger symbols, so some mapmakers apply a perceptual correction called Flannery compensation that slightly enlarges bigger symbols to account for this bias.
A related but distinct approach is the graduated symbol map, which groups data into classes (like small, medium, and large) rather than scaling each symbol individually. Proportional symbols are more precise; graduated symbols are easier to read quickly.
Dot Density Maps
Dot density maps use small, identical dots to represent a set quantity. On a population map, for example, each dot might represent 50,000 people. The dots are distributed within each geographic area based on total count, creating a visual impression of density without assigning specific colors or sizes. Where dots cluster together, density is high. Where they’re sparse, density is low.
These maps have a few practical advantages. They’re intuitive for most readers, they show variation within regions (not just between them), and the original data is recoverable since you can count the dots and multiply by the unit value. You can also layer multiple datasets on the same map using different colored dots, like showing two ethnic groups simultaneously to reveal patterns of segregation or integration.
Cartograms
Cartograms take a more radical approach: they distort the geography itself. Instead of shading or adding symbols, a cartogram resizes each geographic unit based on a data value. On a population cartogram of the United States, Texas and California balloon in size while Montana and Wyoming shrink, because the map encodes population rather than land area.
Two main styles exist. Contiguous cartograms maintain borders between neighbors, so Alabama still touches Georgia even as both areas warp. The result looks like a fun-house mirror version of the real map. Noncontiguous cartograms let go of shared borders entirely, resizing each unit independently and leaving gaps between them. This is easier to read but loses the sense of geographic connection.
Cartograms are particularly effective for countering a common visual bias. On a standard map, geographically large but sparsely populated regions dominate the visual field, while small but densely populated areas are almost invisible. Cartograms correct for this by making visual prominence proportional to the data, not the land area.
Other Common Types
Several other thematic map styles serve specific purposes:
- Heat maps use color gradients to show intensity across a surface, common in weather forecasting (temperature) and web analytics (user clicks).
- Isoline maps connect points of equal value with contour lines. Topographic maps showing elevation are the most familiar example, but the same technique maps barometric pressure, temperature, and rainfall.
- Flow maps use lines of varying thickness to show movement between locations: trade routes, migration patterns, or commuter traffic.
Essential Map Elements
A thematic map needs a few elements to be readable. The legend is the most critical. It explains what the colors, symbols, or shading mean. Good legend design is minimal: it identifies features that aren’t self-evident and skips obvious ones like water or major roads. Contrary to instinct, labeling it “Legend” is unnecessary since readers already know what it is.
A scale bar or scale indicator helps readers estimate real-world distances. A north arrow provides orientation, though it’s often omitted on maps of well-known areas. A title and data source round out the essentials, telling the reader what they’re looking at and where the numbers came from. The date of the data matters too, since a population map from 2010 tells a very different story than one from 2024.
Tools for Creating Thematic Maps
Professional thematic mapping once required expensive desktop software. That’s no longer the case. QGIS is a free, open-source geographic information system capable of producing publication-quality thematic maps. ArcGIS, the industry standard in professional mapping, offers both desktop and online tools, including free web-based versions for simpler projects. For quick, interactive web maps, platforms like Mapbox and CartoDB let you upload a spreadsheet or database and visualize it directly in a browser. Leaflet, a JavaScript library, powers many of the interactive maps you see on news sites. Even Google Sheets can produce basic choropleth maps from tabular data.
The barrier to entry is now almost entirely about understanding your data and choosing the right map type, not about technical skill or software cost. The most important decision is matching your visualization method to your data structure: choropleth for rates and ratios tied to areas, proportional symbols for counts tied to points, dot density for showing distribution within regions, and cartograms for correcting visual bias in area-based data.