7.2 Thematic mapping techniques

Thematic mapping techniques are the core methods for turning raw geographic data into visual stories about spatial patterns. Choosing the right technique determines whether your audience can actually interpret the data you're presenting, so understanding the strengths and limitations of each approach matters as much as knowing how to build the map itself.

Types of thematic maps

Thematic maps differ from general reference maps because they focus on communicating a specific subject rather than showing where things are. Each type below handles a different kind of data and spatial question, so picking the right one is your first design decision.

Choropleth maps

Choropleth maps shade predefined areas (countries, states, counties) according to a data value. Darker shading typically means higher values, lighter means lower. They're probably the most common thematic map you'll encounter.

• A population density choropleth might shade counties from pale yellow (low density) to deep red (high density)
• A per capita income map might use light-to-dark blue, with the darkest blues representing the highest income brackets

One thing to watch: choropleth maps imply that the entire area has a uniform value, which is rarely true. A large county might show low average density even though most people live in one small town within it. This is exactly the problem dasymetric maps try to solve (see below).

Proportional symbol maps

Proportional symbol maps place scaled symbols (usually circles) at specific locations, where the symbol's size corresponds to the data value. Bigger symbol = bigger value.

• A map of city populations might place circles over each city, with Tokyo's circle much larger than Oslo's
• A map of earthquake magnitudes might use squares scaled to the Richter value at each epicenter

The main challenge is symbol overlap in dense areas. You also need to decide between mathematical scaling (symbol area is directly proportional to the value) and perceptual scaling (adjusted so readers perceive the size differences accurately, since people tend to underestimate the area of larger circles).

Dot density maps

Dot density maps scatter dots within areas, where each dot represents a fixed quantity. The clustering of dots reveals spatial concentration without implying uniform distribution the way a choropleth does.

• An agricultural map might place one dot per 1,000 acres of farmland, so heavily farmed regions appear dense with dots
• A racial/ethnic distribution map might assign one dot per 100 people of a given group, revealing neighborhood-level patterns within cities

Choosing the right dot value (how much each dot represents) is critical. Too few dots and the pattern disappears; too many and the map becomes an unreadable blob.

Isarithmic maps

Isarithmic maps (also called contour maps) draw isolines connecting points of equal value across a continuous surface. The spacing between lines tells you the rate of change: tightly packed lines mean rapid change, widely spaced lines mean gradual change.

• Topographic maps use contour lines to connect points of equal elevation
• Weather maps use isobars (equal pressure), isotherms (equal temperature), or isohyets (equal precipitation)

These maps only work for truly continuous phenomena. You wouldn't use isolines for categorical data like land use types.

Dasymetric maps

Dasymetric maps refine choropleth maps by incorporating ancillary data (like land cover or building footprints) to redistribute values more realistically within each area.

For example, a standard choropleth might show a county's average population density spread evenly across the whole county. A dasymetric version would use land cover data to concentrate the population in urban zones and show forests or water bodies as uninhabited. The result is a much more accurate picture of where people actually live.

Cartograms

Cartograms distort the geometry of areas so that each region's size is proportional to a chosen variable rather than its actual land area. They maintain topology (which regions are neighbors) while warping shapes.

• A population cartogram makes India and China enormous while Canada and Australia shrink
• A GDP cartogram inflates the United States and Western Europe relative to their geographic size

Cartograms are powerful for making a point, but they sacrifice geographic accuracy. Readers sometimes struggle to identify distorted regions, so clear labeling is essential.

Cartographic design principles

Good data on a poorly designed map is still a bad map. These principles guide how you organize visual elements so readers understand your message quickly and correctly.

Visual hierarchy

Visual hierarchy controls where the reader's eye goes first, second, and third. You establish it through size, color, contrast, and placement.

• The map title and thematic content should dominate
• The legend and supporting elements should be easy to find but not compete with the map face
• Use larger or bolder fonts for the most important labels; smaller, lighter fonts for secondary information

If everything on your map has equal visual weight, nothing stands out and the reader doesn't know where to look.

Figure-ground organization

This is about making the main subject (figure) visually distinct from the background (ground). Your thematic data should pop; the base geography should recede.

• Use contrasting colors or shading between the mapped theme and the background
• Simplify or mute background features (coastlines, neighboring regions) so they provide context without competing
• Whitespace and subtle base tones help the figure stand out

Legibility and readability

Legibility is whether you can physically distinguish individual text and symbols. Readability is whether you can understand the map's message as a whole.

Factors that affect both:

• Font choice and size (too small = illegible; too decorative = hard to read)
• Symbol design and spacing
• Color contrast (low contrast between symbol and background kills legibility)
• Map scale and level of detail (too much detail at a small scale creates clutter)

Contrast and color selection

Color does heavy lifting on thematic maps. Poor color choices can mislead readers or make the map inaccessible.

• Ensure enough contrast between adjacent colors so classes are distinguishable
• Use color-blind friendly palettes (tools like ColorBrewer are designed for this)
• Match the color scheme to your data type (see the Symbolization section below for sequential, diverging, and qualitative schemes)
• Avoid using red and green as the only distinguishing colors, since red-green color blindness is common

Typography in thematic mapping

Font choices affect both aesthetics and clarity.

• Sans-serif fonts (like Helvetica or Arial) tend to work well for small labels and body text
• Serif fonts can work for titles or larger text elements
• Adjust font weight and style to reinforce the visual hierarchy
• Avoid label overlap by adjusting placement, using leader lines, or reducing label density
• Keep spacing consistent so text doesn't crowd other map features

Data classification methods

How you divide continuous data into classes dramatically changes what patterns the reader sees. Two maps of the same dataset can tell very different stories depending on the classification method. Understanding these methods is essential for both creating and critically reading thematic maps.

Equal interval classification

Equal interval divides the data range into classes of the same width.

How it works:

1. Find the minimum and maximum values in your dataset
2. Subtract the minimum from the maximum to get the total range
3. Divide the range by the number of desired classes
4. Set class breaks at equal steps from the minimum

For a dataset of household incomes ranging from $\$0$ to $\$100{,}000$ with 5 classes, each class spans $\$20{,}000$: $0\text{–}20{,}000$, $20{,}000\text{–}40{,}000$, $40{,}000\text{–}60{,}000$, $60{,}000\text{–}80{,}000$, $80{,}000\text{–}100{,}000$.

This method is simple to understand, but it performs poorly with skewed data. If most values cluster at the low end, most of your map may end up in one or two classes while the upper classes contain very few observations.

Quantile classification

Quantile classification puts an equal number of observations into each class, regardless of the value range each class spans.

With 100 observations and 5 classes, each class gets exactly 20 observations. The trade-off is that class boundaries can be arbitrary, and two very different values might end up in the same class while two similar values get split into different classes.

This method guarantees a visually balanced map (each color appears roughly equally), which makes it popular for choropleth maps. But be cautious: it can obscure the actual distribution of values.

Natural breaks (Jenks) classification

Natural breaks (Jenks optimization) finds class boundaries that minimize within-class variance and maximize between-class variance. The algorithm looks for natural groupings or clusters in the data.

This is often the default in GIS software because it tends to produce intuitive classes that reflect real patterns. The downside is that class breaks are data-specific, so you can't easily compare maps of different datasets that used Jenks classification, since their breaks will differ.

Standard deviation classification

Standard deviation classification centers classes on the mean and sets breaks at intervals of the standard deviation (commonly $\pm 0.5$, $\pm 1$, or $\pm 1.5$ standard deviations).

This method highlights how far values deviate from the average, making it useful for showing outliers or anomalies. It works best when the data is approximately normally distributed. With heavily skewed data, most observations pile into one or two classes.

Manual classification

Manual classification lets you define class breaks yourself based on domain knowledge, policy thresholds, or the story you need to tell.

For example, you might set income breaks at poverty line, median income, and high-income thresholds that correspond to established economic categories. This gives you full control but introduces subjectivity. Always document your rationale so readers understand why you chose those breaks.

Symbolization techniques

Symbolization is how you translate data values into visual marks on the map. The right symbolization makes patterns obvious; the wrong choice obscures them.

Color schemes for thematic maps

Color scheme selection should match the type of data you're mapping:

• Sequential schemes (light to dark in a single hue) work for ordered data like population density or income. Higher values get darker colors.
• Diverging schemes (two contrasting hues meeting at a neutral midpoint) work for data with a meaningful center value, like temperature anomalies above and below average, or election margins.
• Qualitative schemes (distinct, unrelated hues) work for categorical data like land use types or soil classifications. No color should imply "more" or "less."

Tools like ColorBrewer (colorbrewer2.org) provide tested palettes for each scheme type, including color-blind safe options.

Proportional scaling of symbols

When scaling symbols to data values, you need to decide on a scaling approach:

• Mathematical (absolute) scaling: symbol area is directly proportional to the data value. Accurate but readers tend to underestimate larger symbols.
• Perceptual (apparent magnitude) scaling: symbol sizes are adjusted using Flannery's compensation (an exponent of approximately $0.57$ applied to the radius) so that readers perceive the size differences more accurately.

Define a reasonable size range so the smallest symbol is still visible and the largest doesn't overwhelm the map. Always include a legend showing reference symbol sizes.

Dot distribution and density

Key decisions for dot density maps:

• Dot value: how much each dot represents (e.g., 1 dot = 500 people). This controls the visual density.
• Dot size: must be small enough to avoid merging in dense areas but large enough to be visible in sparse areas.
• Dot placement: dots can be placed randomly within each area or guided by ancillary data (like placing population dots along road networks rather than in lakes).

The legend must clearly state the dot value and units.

Isolines and isopleth mapping

Isolines connect points of equal value. Isopleth maps go a step further by filling the zones between isolines with colors or patterns to show value ranges.

• Choose an isoline interval that captures meaningful variation without cluttering the map. Too many lines obscure the pattern; too few hide important detail.
• Label isolines with their values, either directly on the line or at the map edges.
• For isopleth zones, use a sequential color scheme that creates a logical visual progression from low to high values.

Areal interpolation techniques

Areal interpolation estimates data values when source boundaries don't match your target mapping units. This comes up frequently when combining datasets from different administrative systems.

• Areal weighting assumes uniform distribution within each source unit and allocates values proportionally based on the area of overlap with target units. Simple but often inaccurate.
• Dasymetric interpolation uses ancillary data (land cover, building density) to redistribute values more realistically before transferring them to target units.
• Pycnophylactic interpolation creates a smooth surface that preserves the total value within each source unit. It avoids the sharp boundaries of areal weighting while maintaining volume (total count) preservation.

Map elements and layout

A well-designed layout ensures the reader can find and interpret all the information they need. Every element should serve a purpose.

Map title and subtitles

The title tells the reader what the map is about. It should be specific enough that someone can understand the map's subject at a glance.

• Place the title prominently, typically at the top of the layout
• Keep it concise: "Median Household Income by County, 2020" is better than "A Map Showing the Distribution of Household Income Across Counties in the United States for the Year 2020"
• Use subtitles for additional context like geographic extent, time period, or data source

Legend design and placement

The legend translates your symbols and colors into meaning. Without a clear legend, the map is just a pretty picture.

• Position the legend where it's easy to find but doesn't cover important map content
• Organize legend items in a logical order (e.g., low to high for sequential data)
• Use labels that match the terminology in the title and annotations
• Include units (e.g., "persons per sq km" or "USD") and any necessary explanatory notes

Scale and north arrow

The scale bar shows the relationship between map distance and real-world distance. The north arrow orients the reader.

• Place both in a visible but unobtrusive location, often near the bottom of the layout
• Label the scale bar with appropriate units (kilometers, miles) for your audience
• Keep the north arrow simple. An elaborate compass rose can distract from the thematic content.
• For small-scale maps (large areas), a scale bar may be misleading due to projection distortion. Consider whether a representative fraction or verbal scale is more appropriate.

Inset maps and locator maps

Inset maps provide additional context or zoom into areas of interest. Locator maps specifically show where the mapped area sits within a larger region.

• Place insets where they don't obscure important map content (corners or margins work well)
• Use a different scale or level of detail to distinguish the inset from the main map
• Label the inset clearly so readers know what it shows
• For locator maps, a simple outline with the study area highlighted is usually sufficient

Marginalia and metadata

Marginalia includes everything outside the main map frame: data sources, author, date, projection, and notes. Metadata documents the data's origin, accuracy, and limitations.

• Place this information along the bottom or side margins in a smaller font
• Always credit your data sources
• Include the map projection, especially if the map covers a large area where distortion matters
• Note any data processing steps, date of data collection, or known limitations
• Add copyright information when required by data providers

Thematic mapping software

A range of tools exists for creating thematic maps, from full GIS platforms to specialized web applications.

GIS software for thematic mapping

Full GIS platforms like ArcGIS Pro, QGIS (free and open-source), and MapInfo provide comprehensive thematic mapping capabilities alongside spatial analysis tools.

Key features relevant to thematic mapping:

• Support for multiple data formats (shapefiles, geodatabases, GeoJSON, web services)
• Built-in data classification methods (equal interval, quantile, Jenks, standard deviation)
• Flexible symbolization and labeling tools
• Layout editors for composing print-ready or digital map products
• Export options for various formats (PDF, PNG, SVG, web tiles)

QGIS is a strong option for students since it's free, cross-platform, and supports most of the same thematic mapping workflows as commercial software.

Specialized thematic mapping tools

Some tools focus on specific map types or streamlined workflows:

• ScapeToad and go-cart.io: cartogram generation tools that distort area geometries based on attribute values
• Surfer (Golden Software): specialized in isarithmic/contour mapping and surface interpolation
• ColorBrewer (colorbrewer2.org): not mapping software per se, but an essential tool for selecting appropriate color schemes
• Mapbox and Kepler.gl: web-based platforms useful for interactive thematic maps and large dataset visualization
• D3.js: a JavaScript library for building custom, interactive thematic maps for the web (requires coding knowledge)