To 3D or not to 3D?
Both 2D and 3D maps are great, but knowing when to use them—and why—is key.
It has never been easier to create amazing 2D and 3D maps.
Mapping technology has come so far in the last 15 years that things that seemed like science fiction, or required a million dollars of hardware and thousands of hours of labor, can now be produced on your laptop in minutes. As a storyteller, it’s fun to take advantage of that power to engage and impress your audience. But when should you use 2D maps versus 3D maps, and how do you get the most out of each? An even better question to ask is, For a given dataset or communication task, how do we know when to use which? Let’s explore these questions.
What follows is not a step-by-step tutorial on how to make these maps. I want share some of the editorial thought processes that we use on the StoryMaps team, and how we work through pros and cons of 2D and 3D maps. Spoiler alert: This isn’t a simple black-and-white checklist or yes/no decision tree with a clear winner. Rather, cartographic decisions like this inevitably involve a series of trade-offs.
Ever wonder why your transatlantic fight goes so far north? On this 3D interactive globe the answer is obvious; it's actually the shortest path and a straight shot across the pond.
The exact same airline routes on a flat Web Mercator map don't make nearly so much sense in 2D.
The latest mapping tech can let us see the world simultaneously from both a 2D and 3D perspective giving us the best of both worlds. Here the 3rd dimension shows additional information over time.
Some background
2D and 3D are broad categories that include many types of maps. This topic easily fills multiple textbooks , so what follows is very condensed. Generally speaking, a 3D map extrudes the map into the third dimension, what is often called the z-dimension, either because it is a) directly representing 3D objects like buildings, mountains, or the globe itself, or b) symbolically representing data which can be conceived of as having a third dimension.
A recent winner from the prestigious Information is Beautiful Awards shows light pollution as a series of mountains. Author: Jacob Wasilkowski
We often speak of “data surfaces” in cartography which don’t have actual hills and valleys but do have conceptual ones. For example, we can think of geographic phenomena like light pollution, or unemployment, or housing prices, as a traditional 2D map—or we can visualize them as a 3D surface. In the map above, cities with their strong lights are mountains and uninhabited areas are plains and valleys. The same concept of mountains and valleys can apply to housing prices or unemployment rates. In all of these cases 2D and 3D are valid depictions of the same data. But they make your brain work in different ways and—here’s the important part—one isn’t necessarily better. They’re just better suited to different tasks. No map can do everything for everyone, so be deliberate in your choices.
Ask yourself, “What do I need my readers to do here? What do I want them to learn from this map?” Once you answer that, the choice of how to depict the data may become obvious.
3D versus 2D
I often hear “3D has got to be better than 2D, I mean just look at it!” Yes, 3D maps can have a big wow factor, can look cool, and are great at grabbing attention. These are all valid reasons to use them sparingly, but too much of a good thing can spoil the effect. You might be surprised to learn that 2D maps do some things much better than 3D maps.
A word of caution: With flexible mapping tech like ArcGIS Scene Viewer or ArcGIS Pro , it’s simplistic to frame this as an either-or argument. These tools let us explore the world simultaneously from both 2D and 3D perspectives. It’s truly the best of both worlds!
However, in stories where static maps are needed, limited interactivity is possible, or when your audience is quickly skimming through a story, I highly recommend using fast-loading static maps. And that requires us to choose one approach, whether it's 2D or 3D.
Eight Things to Consider
To help simplify this discussion, let’s break the world of maps in two: maps that deal with things at human scales or just a small slice of the planet (buildings, cities, mountains) and those that depict the entire globe. We need to do this because the answer to the questions below can change based on how much of the earth is being depicted. It’s complicated, and that’s OK.
The bottom line really is just “Can my audience understand and learn what they need to from this map?” If the answer is YES, you’ve got a winner.
1. Do your readers need to accurately estimate relative sizes and distances?
When a cityscape is viewed obliquely in a 3D map, scale varies throughout the scene. Judging relative distances and sizes requires a lot of cognitive horsepower and flying around. By comparison, a quick glance at a simple 2D map of the city makes answering those questions a breeze.
Because most 2D maps are carefully projected and viewed looking straight down, they do a much better job of faithfully depicting distances, relative angles, and sizes at local scales. If I need my readers to understand one of those things at a glance (and not spend time picking and prodding the map for the answer), 2D is simpler, faster, and just works better.
Not convinced? How confident are you that one red line is longer than the other?
In 2D...
...it takes no more than a glance to know which line is longer (not guess).
2. Is the 3rd dimension critical to your story?
For some tasks, like climbing mountains in a national park, the 3rd dimension is very relevant, and seeing the terrain ahead before leaving can be helpful. Yes, we depict terrain on 2D maps, but for a visceral sense of what lies ahead, 3D rocks (pun intended).
It’s not just rock climbing. The same is true for hydrologists, geologists, meteorologists, engineers, architects, or any context in which the 3rd dimension is relevant to visualizing the problem and solving it.
Here the 3rd dimension is used brilliantly to show us the depth of earthquakes.
Author: Reluca Nicola
...Or is the 3rd dimension really that critical to your story?
If, on the other hand, you find yourself saying “let’s make it 3D because it looks cool,” but 3D doesn’t communicate anything more to your readers than the 2D map did, I might reconsider.
How different are the populations of Kenya, Uganda, and Tanzania? There's obviously some serious population in Nigeria, but is it twice as much as Ethiopia? Here I worry the 3rd dimension makes reading the map harder than it ought to be.
Data: World population estimate 2016 (Esri)
3. Will your readers be navigating?
There is a lot of research on this topic, but, for in-the-moment, turn-by-turn navigational tasks, many people prefer to view the world obliquely, like we do with our own eyes every day. The theory is that there's a shorter conceptual leap between a 3D real world and a 3D depiction of that world, especially when decisions need to be made quickly—for example, when piloting a car.
However, change the context from driving to pre-planning the optimal route to get all of your errands done on the way home, and many people prefer the abstraction and simplification of 2D maps. Judging relative distances and synoptic decision-making require seeing things from a less situational and subjective perspective. The added details of the immersive 3D scene aren’t helpful.
In short, 2D and 3D are both great at navigation, but it all depends on the task and the context.
Right: Downtown Zurich in 3D with Esri's world traffic service vs. 2D proportional traffic flows by Jim Herries
4. Need to see the entire globe at once?
Stepping back to the scale of the entire planet, 3D digital globes are awesome, and one the best things to have happened to mapping in a long time. They suffer, however, from a big drawback: You can only see half the planet at once. If it is critical that your audience sees all locations at once, project the earth onto something flat.
Here we see how much of the Earth is suitable as pasture land in a 3D map ...
Need to see the entire globe at once?
...and here is the same data as a 2D map. Some tasks, like comparing the latitudes at which pasture land is most plentiful, are going to be easier here.
5. Is occlusion going to bother your readers?
This one is a toss-up and is the local-scale version of seeing all of the data at once. 3D maps at more human scales often result in some things being hidden behind other things.
Sometimes, however, that is precisely the point of the map: to simulate what can and can’t be seen from a location. This is known as viewshed analysis .
...Is occlusion going to bother your readers?
On the other hand, for many maps the hiding of things behind other things might be frustrating, requiring readers to move or interact with the maps.
If you just want your readers to know where the bus route goes, perhaps you shouldn't hide half of the route behind 3D buildings.
6. In the real world where your readers live, how much interactivity is going to be possible?
3D maps really come to life when your readers can interact with them by rotating, touching, pinching, tapping-on, and spinning them to their hearts' content. When you start adding things like digital 3D rulers, fly-bys, and scene slicing, these maps can be magical. In other words, they’re incredibly engrossing, and keep us glued to our screens like a video game powered by geographic data.
3D interactive tools allow us to see inside our data
However, lots of interactivity might not be possible in some contexts. On slow or small devices, in busy-use environments in which hands are not free (e.g., pilots), or for folks with compromised dexterity, the immersive experience of these maps is not appropriate and a simpler, static 2D map will likely make your readers’ lives easier.
7. Do you have the data?
Sometimes our efforts to employ 3D maps are stymied simply because we don’t have the data. For example, I may know building footprints but not building heights. Or I have a few spot readings of something like air temperature or voter reaction but not enough to safely interpolate a fancy, smooth 3D “data surface” even if our software will let us (see: Mark Monmonier's How to Lie with Maps ).
At other times, you may have the data but it's just not needed. Does it really matter to your readers how deep the subway tunnel is? Or is it just the sequence of stops that matters, and where to make transfers?
Engineers are interested in knowing how deep the stations are, riders don't need to. Author: Ken Field
8. Does it convey the right emotion?
This is a big one, and probably the hardest for me to give advice on. But we’ve all been readers of maps that make us stop in our tracks and go “wow!” Those maps can become time-sinks in the best possible way. On the other hand, I also admire elegant, spartan maps that get the job done and get to the point with a minimal amount of ink. Even better, I love maps that work so well that I’m not even aware there is a representation of data before me: when the map becomes transparent, and I feel as if I’m in the data itself, learning and problem solving.
Experiment with the emotion of 2D abstractions and rich 3D immersives, and try to match their emotional impact with the seriousness or rigor of your story and content.
Final Thoughts
There are many wonderful examples of 2D and 3D maps out there to inspire you. Maps We Love is a great place to start, as are many of the stories we feature on our Resources Page . And lastly, it doesn’t have to be an all-or-nothing approach; One of my favorite stories from the past few years is The Human Reach , which carefully mixes 2D and 3D maps to move the narrative along and create a sense of immersion.
Happy mapping!
Acknowledgments
Big thanks to Greyson Harris for his skills with ArcGIS Pro and for making many of the maps in this story.
