Getting the most from animated maps

The reach of cartographers’ ideas often far exceeds the grasp of the available technology. Animated maps—maps that move, change, and can be watched like a movie—are a great example. Starting in the 1940s, a few animated maps were hand-drawn, one frame at a time, for news reels to show current events. These maps were prohibitively expensive to produce. And, because they had to be seen in a movie theater and took months to complete, they were also difficult to distribute and out-of-date by the time they were shown.

The animated map at the start of the story shows the  extent of sea ice in the polar regions . Compare this to the hand-crafted,  narrated animated map from a 1939 news reel  showing the German invasion of Poland. The clever use of 3D stylized maps effectively conveys the emotion of the invasion as much as specific details.

While the growth of computing made the production of animated maps much easier, until quite recently most animated maps had to be “pre-rendered,” saved as videos, and passively watched. Bandwidth wasn’t fast enough nor the hardware powerful enough to make animated maps on-demand and interactive. 

In the past decade that has all changed. In fact, readers often now expect a story to contain  dynamic, rich, animated maps that behave like a hybrid between a movie and a video game. Indeed,  some of the best storytelling and journalism  I’ve seen in the past 10 years has been built around animated maps. 

Three of the most common ways time and animation are used on maps are showing:

1. Fly-bys
2. Data values changing over time, and
3. The movement of objects

There are other kinds of animated maps, but we will focus on these here. 

Here on the StoryMaps team we regularly include animated maps in our stories. Why? Put simply, they move. And in their dynamic, dancing nature they can reveal previously unseen subtleties in data. They can show where things are changing, and by how much. And they can push gigabytes of data into our neocortex at a glance. This is possible because we live (and evolved) in an animated world. As a result, without much conscious effort, our eyes and brains are deeply wired to make sense of a moving environment.

When you add time to your maps you are tapping into ancient visual processing skills of humans. 

Detecting movement: Like the shadow of a predator moving in a forest, things that move on maps “pop” against a static background. Even out of the corner of our eye or when the motion is very subtle, we have dedicated visual processes for seeing motion.

Rate of change: We’re not just good at seeing movement, but also at quickly figuring out if something is getting faster or slower.

Anticipating trajectories: If we couldn’t do this, we couldn’t catch a baseball or avoid another pedestrian.

Seeing emergent groupings: One bird flying in the sky is a single object. But when it’s thousands in a flock turning and wheeling we no longer see the individuals but “abstract out” a single coherent object with shared fate.

Synchronization and rhythm: In an animated map we can see if events are happening simultaneously, in an orderly sequence, or randomly. We do this by calling on the same kind of temporal neuro-circuitry that allows us to play music.

Animated maps can really engage your audience in profound ways. They invite questions like “What’s going on here?” or create reactions like “Wow! Look at that pattern!” 

These two maps below show the power of using time to animate non-temporal data. Instead, animation here is used to represent either flow rates or distance and gives our brains time to absorb what we're seeing. They're also beautiful and require no legend to be understood—a rare feat for any thematic map.

Animating raw data is often a bad idea and can lead to really boring maps; our data often need some preparation before they can shine.

Don't be afraid to aggregate your time series data before animating it. In fact, animating raw data is often a bad idea and can lead to really boring maps if not much is happening from moment to moment. Sometimes we need to derive averages, or sum totals, or spatially bin occurrences before meaningful patterns can emerge.

Take this  lovely animated map made by cartographer John Nelson  showing the location and intensity of tornados by month using 60+ years of data. Since tornados are (mercifully) rare events and there can be long stretches without them, "playing back" this raw data wouldn't help us much. Instead, by adding together all tornados for a given month and counting how many occur in a region, we can step back from the raw data.

What we see now is a springtime "bloom" of tornado outbreaks as they move north with the warming weather. But look closely and you'll see there Is more than a north-south axis, there is also a westward migration to the high plains during the summer (Nebraska and the Dakotas). Also note the way the circles "evaporate" as number and intensity of tornados die down in September and October.

Despite all of this, like any narrative tool, animated maps can easily be over-used. That is not just a stylistic note of caution; they’re actually not suited for many kinds of data and are going to make some map-reading tasks harder than they need to be. There are seven questions that I use when deciding how to cast the maps in my stories:

On the whole, animated maps are more work to create than static maps. Sometimes, a lot more work. That’s why I keep them for where they are most needed. For something simple like a locator map or a map with a few pushpins, static is better. 

But it’s more than just labor costs. An animated map grabs our attention and it can overpower the story. If the map should be in a supporting role and not the star, a static map is likely the better choice.  

Animated maps are slower than static maps in two ways: They take longer to load and they take time to watch. If I’m asking my readers to make an investment of their time, I try to reward that with some real payoff. Put another way, if the static map would have done the job just as well in half the time, I’m likely going to use it. But if the animated map adds something that static map can’t (see below), it’s a justified cost.

Though often captivating, animated maps can make some map reading tasks harder than they need to be.

Most animated maps allow readers to stop, rewind, and skip ahead, just like a video. But even with this control, if I want my readers to compare two very specific times—say the unemployment rate in 1950 vs 1970—and confidently say how much they’ve changed and where they’ve changed the most, static maps are going to work better. Most people need time to study those maps, this is hard stuff!! Two static maps side by side or overlain with a simple toggle would be perfect for this. But if I bury 1950 and 1970 in the middle of a 50-year-long animation, I’m making my reader’s lives a lot harder. Those dates will only be on-screen very briefly and—short of a photographic memory or a lot of jumping around in the animation—it’s not going to be easy to figure out what’s going on. 

The previous point was about quantitative reading tasks involving things like calculating numbers and comparing specific places. But if I just want my readers to appreciate how things are trending overall, where they’re moving or changing more or less, faster or slower, animation can be great. That same animation of the last 50 years of unemployment will show when things are stable, when recessions hit, and the overall cyclic nature of the economy. You won’t need to be an expert to see those things, they will jump out. And there is no need to dig for specific numbers to glean this, it’s about detecting overall patterns of change.

These are qualitative reading tasks and it is here that animation really shines. If you have time-series data with large overall trends or cycles, animation will literally bring this to life. 

On an animated map the facts of the story are revealed to the readers over time. This gives them time to digest what they’re seeing. It also gives you the chance, like a film director, to focus their attention on different things. Like cinematic pacing, you can decide how fast to show things, what sequence to show things in, and when to do a big reveal. 

When you think of map animation as cinema, you realize how many established storytelling tools are at your disposal: like moving the virtual camera through a 3D landscape or directing the emotional arc of the narrative. If this weren’t true—if story telling wasn’t a temporal relationship with our audience—we could just share the spreadsheet of raw data and call it a day.

Taking maps-as-cinema even further, animation allows you to add a narrator to the map. I have done this many times when a introducing an unfamiliar subject to a wide audience. I have used a narrator to ensure a critical insight in the map isn’t missed by my readers or to call out specific moments or locations. A narrator can make an intimidating map more approachable. And that is never a bad thing.

The last item on this list is perhaps the simplest and also the most frustrating. So often I find myself in search of historic data that I can’t get for a variety of reasons, or it simply never existed. For example, I might know when the battles occurred, but not the exact path an army took 500 years ago across the landscape. And to animate its march with false precision on top of modern maps would be misleading. 

Sometimes you have some historic data but not enough data to make an animation worthwhile. For example, you might have sales data for only the last four quarters and the resultant animation is just…underwhelming. In fact, that sales data might be interesting and a chart showing breakdowns or a series of small-multiple maps will do it justice in a way a one-second long, four-frame animation won’t. 

Even if our software can interpolate missing time periods, that’s almost never statistically a good idea

More often than not you won’t have high quality, reliable data going back as far as you want. In those cases it is safer and more ethical to not gloss over this. Even if our software can interpolate missing time periods, that’s almost never statistically a good idea. A static map will still reveal what we do know and not oversell what we don’t.

Animated maps can add a lot of visual punch to your story. They can be the star of the show, but they can be equally hard to ignore. And they require “buy in” from your audience to really work. I love them because they work equally well for novices and experts alike by tapping into that deep wiring in our eye brain system. They’re also a natural fit with  StoryMaps . Using the latest generation of mapping tools like  ArcGIS Pro  it’s never been easier to experiment with different kinds of animated maps. By blurring the lines between maps, cinema, and interactive games, cartographic animation is worth the time and expense if used wisely.

Not sure where to get started? Have a look at  Maps We Love  for some winning examples. Happy mapping!