How to Color Your Map Using SCIENCE!

Sketching out a map for a setting can be a lot of fun. Drawing a map gives you a bird’s-eye view of the world, a way to spatially organize plot arcs, and can be a great piece of artwork in its own right. But like most works of fiction, the creator should remember to keep it as believable as possible. This might be less important or less possible for unrecognizably alien worlds.  Maps of Earth-like settings, however, can benefit from following some basic rules. Forests, tundras, deserts and plains don’t appear arbitrarily. These biomes are located where they are on Earth due to the way air and water circulate in the atmosphere – and any Earth-like world should follow the same basic rules for its atmosphere that Earth does.

But who wants to spend time researching atmospheric science just to know which parts of their map to color green, brown, or beige?  Well, I do, so let me save you some trouble by relaying what I’ve learned.

Before we dive into the biome-painting guidelines, we’ll need a brief background in planetary-scale meteorology, the reasons behind the rules:

  • Heat rises.
  • The equator gets hotter than higher latitudes.
  • Warmer air holds more water.
  • The planet rotates from east to west.

These simple facts dictate the basic weather and climate patterns of our globe; most notably, the atmosphere’s banded circulatory system.*

Circulation of the Earth's Atmosphere: the Six Cells. Circulation of the Earth’s Atmosphere: the Six Cells. Image by NASA; public domain.

At the sun-soaked equator, hot air rises high into the atmosphere and begins to travel away from the center. At 30° latitude to the north and south, a third of the way to the poles, the air has finished cooling again, and falls back to Earth. Its journey has wrung out much of its original water content; as it descends on latitude 30, it brings almost no rain. But this air isn’t about to stay put. It travels now along the surface, either poleward atop the mid-latitude cell, or back to the equator. When moving poleward, it picks up evaporating water along the way, and when it crashes into the polar air to the north or south, it drops a deluge before ascending again.

This has three important implications:

  1. The latitudes where air descends as it converges (about 30°N and 30°S, and arguably the poles) tend to be clear-skied and dry, and latitudes where air masses collide and ascend are cloudy, rainy, and lush.
  2. Earth is spinning east to west at 1,042 miles per hour. Even though we aren’t aware of it on the surface, the weather is. The Coriolis effect causes any object or air mass moving toward the poles to curve west, while traveling toward the equator causes trajectories to tack east. Thus, winds between 35° and 50° latitude will be predominantly westerly, and winds between latitudes 0° and 15° will be predominantly easterly.*
  3. Since these winds travel along the surface, mountains can divert them or push them upward, causing them to spill their water on the windward side while leaving the leeward side dry. The result: rain shadows.

So, knowing all this, what mapmaking rules can we deduce?  Let’s try an example.  If we start from the map below (mountains in gray), which biomes can we expect in which regions?


If this landmass is the size of, say, Hawaii, then our job is easy; we just need to know its latitude.  If it’s in the dry belt near 30° latitude, it’s, well, dry.  Otherwise it’s fairly wet, and therefore can be colored green, except for the leeward side of the mountains (west near the equator, east in the colder latitudes) which will be in a rain shadow.

Of course, most world building is done closer to actual world scale. What if this hunk of land is a full-blown continent?


In this case, things get a bit more complicated.  We’re going to want to overlay an Earth-like chart of the climate zones and wind patterns to get a full picture of what’s going on, and which regions need to be wet, dry, or in-between.


I’ve filled in the map with the basic climate zones: tan is desert, dark green is wet (forest, jungle, bog, and bayou), purple is polar (tundra, icy bogs peppered with stunted trees, bare ice, and rock) and pale green is anything in between those extremes that isn’t too wet or too dry; terrain like steppe, plain, savannah, and scrubland.*

The arrows on the left are key to understanding the climate patterns I’ve filled in.  They represent the normal directions of surface winds, and how wet or dry they are.  Blue is wet wind, orange is dry, green is in-between, and purple* is frigidly cold and polar.

The hurricanes (blue swirls) are a bit special; at semi-random intervals, they dump lots of water (and destruction) on coastlines between between about 5° and 30° latitude.*

In most cases, water from the blue lines follows the arrows until it hits a mountain range, beyond which there’s a dry rain shadow (desert, scrubland, arid plain, et cetera). However, there are a few other quirks and complexities to consider:

Poles are cold, and more cold equals more wet.

This is pretty basic, but it’s worth a reminder: get close enough to the top or bottom end of a planet and everything is just ice, regardless of how wet or dry everything is.  Antarctica is the Earth’s largest “desert,” in terms of precipitation, yet it’s paradoxically the Earth’s largest reservoir of fresh (and solid) water.  The reason is simple: it’s too cold for water to evaporate.* Stepping back from the poles a bit, we find that stereotypical deserts (sand dunes and scrubland) are rare; cold water evaporates so slowly that wet forests and bogs are common even where rainfall (or snowfall) is low. Conversely, regions that receive a lot of rain can still be relatively barren, if they’re hot enough. Nigerian grassland receives more rainfall than a Minnesota forest.

So where should your deserts be, exactly?

Even though the planet is driest  in the band around 30° latitude, deserts tend to skew equator-ward a bit, due to more direct sun exposure – and thus increased evaporation rates – nearer to the equator. Similarly, rain-shadows in the far north don’t tend to cause deserts, just drier plains or tundra. Coastlines of any kind will help moderate and moisten the climate, since incoming winds will be able to easily gain evaporating water from the ocean; they’re not always enough, though. Western coasts in near-equatorial regions can be quite dry, especially if a mountain range rises directly to their east. An Earth example would be coastal Peru.

What do your mountains look like?

Mountains don’t just stop rain, they CATCH it* – their windward sides tend to be very lush.  They’re also colder than their surroundings due to altitude, which makes them even more likely to support forests. Tall mountains located deep in otherwise arid regions are often islands of green.

What impact will lakes have?

Large lakes act just like miniature versions of oceans; the winds that pass over them become moistened, watering the lands beyond. They’re great at breaking up deserts – so long as they’re deep enough that they can survive existing between deserts.* Note that lakes stop moistening the air when they freeze over – but a lake that freezes over is probably located somewhere so cold that it really doesn’t need much precipitation to stay forested.

Most regions are only as wet as whatever’s upwind.

Places downwind from oceans and lakes (with no mountains in the way) are obvious choices for lush regions, but bodies of water aren’t the only wet things out there. Forests, wetlands, and other soggy regions behave in nearly the same way: water that evaporates from them moistens land downwind.* This gives rise to a significant feedback effect: chop down a large forest, and in climate terms it’s as if you’ve replaced an ocean with dry land. Regions downwind will receive less rain.

East coasts in the temperate zone…

Are frequently wetter than one would expect. The exact reasons are complicated. Partly it’s because they tend to be colder* than west coasts at the same latitude, thus requiring less rain to keep from drying out.  It’s also partly because of the forest feedback effect: if the winds blow in from a forest to the west, they’re more likely to bring rain.*

Examples: Rotate!

Examples are probably the best way to get a feel for how these climate rules work. One example is Earth, of course, and another appears in the image above.  But what would happen to our example continent if we rotated it?


Here we have a 90° rotation from the original.  Deserts still appear in the middle, forests are still on top and bottom.  The polar region is, as always, on the very top* – but it also stretches southward in regions of high elevation.  The temperate forest band has shrunk somewhat, but the equatorial jungle has grown; the rich southern forests would be thoroughly (but transiently) watered by the seasonal monsoon.


A 180° rotation makes for a complicated mix of climate patterns.  This map has a good deal of wiggle room; much depends on just how tall these mountains are, and how often rain-bearing hurricanes slam into the mid-latitude coastlines. But the general patterns hold.


And now, the final rotation. This is, in a way, the most believable map – the fjords are all in the north, where they would have naturally been carved by glaciers during a recent ice age. In the northeast, a long inlet of water functions like Canada’s Hudson Bay: it cools the climate during the summer, but since it’s frozen in the winter, it can’t keep the surrounding land warm. The net effect is the colder, polar climate penetrating farther south than normal.*

Throughout these examples, we see some regions (like the Alaska-esque peninsula with the fjords, and the large island-continent) stay mostly wet, while other regions (e.g., the basin in the final map’s southeast, surrounded by mountains on almost all sides) are always drier than their surroundings.

Boring Details: Disclaimer

It’s worth noting that, despite having fairly distinct air circulation patterns, Earth occasionally deviates from these rules in dramatic ways. These deviations are often a result of the specific arrangements of continents, which heat up or cool down differently than surrounding oceans, and can shift the positions of the cells.  Most notably, the large continents of the Eastern Hemisphere push and pull the “equatorial” rain belt, or ITCZ,* dramatically north and south over the course of the year, depending on the season.  This phenomenon is responsible for the monsoon rains common throughout the tropical regions of Africa, Asia, and Australia.  In the Western hemisphere, the rain belt mostly stays put in the Amazon Rainforest.*

For more examples of biome prediction and an exploration of Earths with different axes of rotation, check out Randall Munroe’s What If? #10 and Chris Wayan’s Planetocopia.*

More Guidelines Than Rules

The rules of speculative fiction and fantasy cartography are really only guidelines. But that doesn’t mean we should take for granted how things work on Earth, especially since fantasy worlds tend to be Earth-like by default.  Even when the world deviates from Earth in a major way, we can still expect it to follow a set of climate rules; we can look to the solar system’s other atmospheres* for guidance, or extrapolate from known Earth phenomena.* And if parts of your world don’t jive with the predictions of climatology, it’s still worth identifying and justifying them. Maybe there’s an out-of-place desert on that equatorial island because…a wizard did it? A terraforming experiment gone wrong? The Sand Spirits finally vanquished the Moisture Lords? Creative deviations are up to you – but hopefully you’re now able to identify them for what they are, or avoid them where believability is crucial.

P.S. Our bills are paid by our wonderful patrons. Could you chip in?

Read more about



  1. Alyssa

    Wow. There is so much information here! I will admit, I took more than an hour to go through it all. But it was worth it! My main problem in terms of worldbuilding is INFORMATION OVERLOAD, and yet I absolutely love thorough articles on interesting “what ifs” like this one. I will now retreat from the internet for the next few days and let all this data sink in.

  2. Amit Patel

    This is really great. Thank you for compiling these guidelines!

  3. Brandon Kier

    Thanks Alyssa and Amit! Glad this could be of use. Now might be a good time to plug the sequel I have coming this Sunday.

    By the way Amit, I love your hex grid resources. Good stuff.

Leave a Comment

By submitting a comment, you confirm that you have read and agree to our comments policy.