Data Storytelling with Maps

Aishwarya Ramakrishnan
Nightingale
Published in
9 min readOct 20, 2020

As a kid, I was fascinated by the concept of maps during my Geography classes. Eventually, I became more and more curious so I started to explore the design of maps. There did exist visual data storytelling in my school books with the help of maps and legends and some story around it. For example, I can recollect one map that speaks about the soil type in India.

Map showing major soil types in India
Source: Maps of India — Major Soil Types

We are living in an era where stories are more attractive than straight facts. The ability to present the raw facts and numbers with interactive visualizations has led to modern data storytelling. When we use maps to tell a story, we give an exact idea of what is happening in and around and it attracts the audiences even more compared to the traditional videos and images. Maps show the spatial relationship that exists for a longer period of time whereas images and videos are used to capture a scene at a particular instant involving temporal information.

In this article, we will look at some useful strategies and alternatives to build maps that are meant to tell effective and compelling stories.

But first, here are a few things to keep in mind while constructing a map:

  • Make the geographic area of the map within its subject area of the map. For example, if we were speaking about the state of Massachusetts, the right choice would be to focus on the state rather than a country map.
  • Make design choices that are based on target audiences. If children are the target audience, it is good to be creative with color mix and attractive symbols. For the general audience, opt for colors, labeling, and icons based on the subject matter. Keep accessibility in mind.
  • Choose the relevant data that cover the main subject area of the map visualization and avoid cluttering with too much background data and information. Use labels sparingly by focusing only on information that needs attention.
  • Follow good color consistency when marking legends on a map. For example, colors should vary from light to dark if it represents low to high values and when representing changes, it is better to opt for colors that show the variations in a distinguishable way.
  • When designing maps for the color blind people, it is advisable to neither rely on just a single color nor use too many colors. Carefully select contrasting colors and shades, and don’t hesitate to check out tools that help build color blind safe palettes.

Fun fact: Facebook’s logo and blue color scheme were specifically chosen because Mark Zuckerberg is red-green color blind and sees blue the best.

  • All map elements should reward the readers with an enriching experience. These include a proper title to the map, helpful legends, information about the datasets, instructions for how to read and interact with the map, and finally information about the author for further feedback.

With that being said, keep in mind to use maps only if location is a part of your data-story. As a data artist, I believe that one could create the most delightful user experience through stories around maps that flow well together. Now, let’s look into how to choose the right map type for your stories.

Selecting the right map type for effective storytelling

All maps tell a story, but not all types of maps are suited for every type of story. Let’s dive into choosing the right type of map to effectively narrate a story. We will look into map types used to narrate about natural calamities and atmosphere, maps for depicting density and quality differences, maps used to narrate about the data or information flow and distance, and finally, maps for everybody.

1. Map types used to narrate about natural calamities

Below are the different types of maps that can be leveraged to show the potential impacts of natural calamities. These are firefly maps, vector wind maps, time travel maps, isopleth maps.

Firefly maps. They use glowing symbols with a dark backdrop as a base map. The background is comparatively dark and the glowing effects make it the best when it comes to narrations about earthquakes, major fires, hurricanes, tornado effects, and night-lights.

Firefly map that shows Earthquakes since 1898 by magnitude
Source: Adventures in mapping — Firefly cartography

Vector wind maps. They use vector arrows that show how the wind blows. The length of the arrow indicates the strength of the wind. They are also used to depict the speed of the winds.

A composite vector wind map that shows the strength and direction of the winds
Source: Visualizing the Wind on Weather Maps

Time travel maps. They divide the map canvas into different chunks of time. Different colors are used to represent the different times of arrival of calamities like Tsunami. Map contours:

  • Red: 1–4-hour arrival times
  • Yellow: 5–6-hour arrival times
  • Green: 7–14-hour arrival times
  • Blue: 15–21-hour arrival times
Map showing the estimated travel time for Tsunamis based on earthquake epicenters
Source: Tsunami Time Travel Maps

Isopleth maps. They show data as a third dimension and simplify data with continuous distribution. Isopleth maps are used to map surface elevations, atmospheric pressure, and measurements that can be viewed as a third dimension.

Isopleth map showing the mean water level fluctuation
Source: Isopleth map of the mean WLF of the study area

2. Map types for depicting density and quality differences

The following types of maps can be leveraged if you are planning to show the variations in your datasets from a density perspective. These are dot maps, graduated symbol maps, proportional symbol maps, Dorling maps, density equalizing maps, choropleth maps, hexbin maps, and heatmaps.

Dot maps. They are used to represent geographic densities and distribution by placing equally sized points over a geographic region. More dots mean a higher value and fewer dots mean a lower value.

Map showing the total number of deaths state-wise due to lightning
Source: The ambiguity of dot density maps

Graduated symbol maps. The variation in the size of the symbols is based on the quantity or value associated with a location. Symbols could be circles, any shapes, or pictograms.

The graduated symbol map shows the highest Total Insurance Value on the southern tip with the largest symbol
Source: ArcGIS Insights — create a graduated symbol map

Proportional symbol maps. They use the same symbols with variations in size to represent the severity of impacts. For example: In the below map, the bigger the circle, the larger is the amount of oil consumed in barrels per year by state.

The map contains legends based on barrels per year by state associated to the US Oil Consumption
Source: The complexity of drawing good proportional symbol maps

Dorling maps. They are used to depict areas with different shapes. The emphasis is more towards the shape that represents the area than the geographic area itself. The larger the size, the higher is the severity of impact.

Map representing the support by state focusing on the total percentage of population irrespective of the geographic area
Source: Carbon Design System

Density Equalizing Maps. They are used to depict geographic areas with extreme values by bulging the areas on the map. By doing so they do not lose the connection in the map.

First map shows the divisions of Mainland China and Taiwan and second map shows the areas that are proportional to GDP
Source: Fast flow-based algorithm for creating density-equalizing map projections

Choropleth maps. They are used to represent values by shading the respective geographic area from a lighter to a darker shade of the same color.

Map showing ratio of number of males compared to the females from the data for 50 US States and Puerto Rico
Source: Choropleth Maps and Census Data

Hexbin maps. They plot densities rather than points. In a hexbin map, all the areas will be represented with the same size without any bias. We can also use squares to split the map area into several parts instead of hexagons.

A Hexbin map that shows wedding per 1000 people in 2015 state by state
Source: Hexbin map in R: an example with US states

Heat maps. They are used to color code based on the density of points. Heat maps use a warm to cool color spectrum to indicate the parts that receive the most attention.

Heat Map that shows US Covid 19 cases by County
Source: Tracking daily new COVID-19 cases in United States, mapping hot spots

3. Maps used to narrate about data or information flow and distance

The following types of maps can be leveraged if your use case revolves around data flow and distances between a set of variables like cities and towns. we will see distributive flow maps, network flow maps, radial flow maps, schematic flow maps, voronoi diagrams, isochrone maps.

Distributive flow maps. Flow maps are used to depict the flow of information and commodity from origin to one or multiple destinations. Distributive flow maps are used to show the flow of information from origin to multiple destinations.

Distributive flow map showing the cash flows associated to Budget in USD
Source: Flow maps and diagrams

Network flow maps. They use link lines to show the interconnections between a group of entities and helps us determine the relationship between them. For example, the below image shows that Europe is the central hub around which the world’s major airports are connected.

Network map showing the world’s air traffic connections
Source: A Network Map of the World’s Air Traffic Connections

Radial flow maps. In these types of maps, the lines radiate from an origin to one or many destination nodes. These types of maps can be used for example to depict the exports made by one country to multiple countries.

Radial flow map showing from where the Cardinals arrived to the Vatican based on their office
Source: Map of Cardinals

Schematic maps. They omit irrelevant details and make essential information easier to grasp. They are used to simplify complex navigations and systems. For example subways, transits, electrical networks…

Schematic map showing the subway stations of Boston
Source: CRS route map — Schematic

Voronoi diagrams. They are used to find the minimum distance needed to reach a point or a landmark and understand proximity as well as distance features. In the example below, the voronoi map could help choose a new location to build a school that is meant to be far from the existing ones by helping find the largest empty circle amid a collection of points.

Voronoi map showing which directions to take to reach the destination faster in a particular region ‘Richmond’
Source: Visualising Supermarkets with a Voronoi Diagram

Isochrone maps. They show how much time is required to travel from one place to another. Isochrone maps are very useful in the areas of transport and urban planning. The below map shows the travel time to Johannesburg from different places in South Africa.

Isochrone map that shows the time taken to travel to Johannesburg
Source: Isochrone Maps 2017 in Review

4. Maps for everybody

Word cloud maps. In this type of map, the size of the word shows the importance of the word. Emphasis is given to large words that suggest an influential topic and small words represent topics with less importance. It is fast, engaging, and a very good starting point for interesting questions and conversations during presentations.

Word cloud map that represents the text about US Constitution
Source: US Constitution Word Cloud Map

Topographic maps. They are all-purpose maps and are used to represent terrains, natural, and man-made features. They still play a key role in infrastructure, military, and resource planning.

Topographic map of the state of Massachusetts
Source: Massachusetts Physical Map

Maps are a powerful way to express and visualize data. They can be easy to understand and do not always require any expertise in any specific field. It is however important to choose the right map type based on the data at hand and the story that you are trying to tell. I hope that my article would help to share your data story with fellow workers, volunteers, organizations, and the public, and to get people excited about your story and the critical information that you want to portray.

As a passionate Data Scientist, I always like to share information in the form of stories for better understanding. This is my first article and I’ll be writing more such articles at https://medium.com/@aishwarya.ramky. Thanks!

--

--

Aishwarya Ramakrishnan
Nightingale

Passionate about Data and Analytics Story Telling Cultures!