Why Maps Paint Lowlands Green, and Why That Misleads

Find a classroom wall map and look at the Persian Gulf. The coast is dark green. Green is the colour of forest and farmland, and the arid Gulf coast is neither.

The green means "low ground." It does not mean "lush." Cartographers call this system hypsometric tints, which is jargon for elevation colours. Atlases, airline in-flight maps and television weather maps all use it. In their 2011 paper on the subject, Tom Patterson and Bernhard Jenny describe a general concern among mapmakers that readers mistake elevation colours for vegetation or climate. They call the evidence "mostly anecdotal," but they took the concern seriously enough to redesign the scale.

The story behind those greens and browns has a rivalry, a school atlas, and a lesson for anyone who builds colour scales for charts and interfaces.

Before the green: brown washes and blue mountains

The first elevation colours were not green at all. According to Patterson and Jenny, Leonardo da Vinci's map of central Italy, drawn around 1503 to 1504, uses indistinct brown tints to show relief. Darker brown means higher ground. Leonardo almost certainly had no surveyed heights to work from, so he was painting an impression.

Real elevation colour had to wait for real elevation data. Barometers arrived around 1800 and made it possible to measure heights, and scientists began compiling tables of spot heights. Coloured layers followed. The French engineer Jean Louis Dupain-Triel filled the space between contour lines with tints on a map from 1791. In the late 1820s, the Austrian Franz von Hauslab drew manuscript maps with elevation as colour layers.

His technique spread a decade later, once chromolithography made colour printing practical. Layer-tint maps then became a staple of school maps and atlas plates.

Early schemes were wild. August Papen's 1857 to 1859 map of central Europe used a series of colours picked for maximum contrast, with no logic to the order. Patterson and Jenny note that Papen uses blue for a middle elevation, and that the effect resembles the rainbow scales scientists still apply to computer-generated data. Imhof, the Swiss cartographer we will meet shortly, complained that such schemes destroy the continuity of the terrain.

The long argument over "higher is darker"

Printing many colours was expensive, so mapmakers moved to something cheaper: one ink, in steps of intensity. Cartographers in Leipzig preferred brown for school and wall maps, and followed a rule of "the higher, the darker."

It had a practical logic. Lowlands cover more area on a small-scale map, and they carry more labels, so light tones there keep the text readable.

Then shaded relief arrived. Printing dark shadows over dark mountain tints produced mountains so heavy they swallowed everything else on the page, with unreadable labels and no sense of depth. Patterson and Jenny describe this as the death knell for "higher is darker." Most hypsometric tints today follow the opposite rule, "the higher, the lighter."

Light summits and dark valleys come with a catch. You get a clean relief effect and you give up quiet lowlands, which is why mapmakers keep reworking the ramp.

Peucker versus Imhof

By the early twentieth century, academic cartographers were debating how to make flat paper look three-dimensional. The German word for the effect was Farbenplastik, "colour plasticity." Two camps formed.

Karl Peucker, a Viennese cartographer (1859 to 1940), was among the first to analyse elevation tints systematically, separating colours by brightness, saturation and hue. His principle was "the higher, the richer the color." The ramp started with dull gray-green in the lowlands and finished with bright red on the peaks. He believed some colours appear closer to the eye than others and that a good ramp would exploit that. If this sounds familiar, our piece on how color temperature tricks the eye into seeing depth covers the same idea, and the same doubts about it.

The doubts came from Eduard Imhof (1895 to 1986). Imhof questioned the mathematical and physical basis of Peucker's claims. But he agreed on one point: a good elevation ramp should imitate aerial perspective, the way distance weakens and cools colour.

Imhof's own scheme ran from green or blue-green, through light yellow-green and light yellow-red, to white. He developed it in 1962 for the Swiss Middle School Atlas. Peucker's ramp jumps to a bright yellow at mid elevations. Imhof's is subtler and moves evenly from dark lowlands to light highlands, with a faint blue haze over the valleys. He went further and argued for smooth blends in place of stepped layers.

Neither man won outright. Peucker's original tints are now rarely if ever used. Imhof's are still common in Switzerland, including the current Schweizer Weltatlas, and less so elsewhere. What spread instead was a mixture. The 1962 International Map of the World starts with Imhof-like blue-green at the bottom, turns to yellow, adds a medium red-brown in a nod to Peucker, then climbs through light gray-violet to white. Patterson and Jenny read those last two colours as a sign of John Bartholomew Jr.'s earlier influence.

Bartholomew, a Scottish cartographer, had published his own tints in 1878, with greens and tans low down, orange-browns and purples higher up, and white at the top. Cartographers of the period did not love it. Patterson and Jenny quote the historian Gardiner: they "seemed to consider it a catchpenny trick, liable to take the mystique out of geography and make it intelligible to laymen." Within twenty years, atlas publishers everywhere had adopted it anyway.

These swatches are my approximation of the familiar sequence, from green lowlands through yellow and red-brown to gray-violet and white. They are not the published values of any atlas.

Imhof himself saw the weak spot in all of this. Patterson and Jenny quote his verdict on how mapmakers choose colours: "Tradition, partiality and whim, preconceived opinions, aesthetic sensitivity or barbarity of taste often play leading roles in the selection of colors."

A scale designed by Alpine eyes

Why green at the bottom? Part of the answer is geography. Patterson and Jenny point out that Imhof's tints generally mimic the Alps: lush valleys, glacier peaks, warm light at sunset. Green lowlands, yellow middle slopes and white summits are logical choices there.

They also ask what would have happened if the pioneers had lived in Phoenix, Reykjavik or Nairobi. Brown or gray might now be the default lowland colour.

Imhof and Peucker had a second advantage. They worked in countries that teach map reading in school, and every Swiss middle school student gets a world atlas. Readers learned what the colours meant. The authors add that this cannot be assumed today, and that many maps, television weather maps among them, show the tints with no legend at all.

The fix: let the lowlands change colour

In 2009, Patterson and Jenny introduced cross-blended hypsometric tints, a scale that gives up the idea of one lowland colour for the whole planet. Their reasoning was blunt. Teaching 6.9 billion people to read hypsometric tints was unlikely to happen, so they adapted the tints to what people already assume.

The lowland colour now changes with the regional environment, and the regions blend into one another. They kept the number of environments at four, to avoid needless complexity. Warm humid regions are a rich yellow-green. Cool humid regions are blue-green. Arid regions are khaki brown. Polar regions are cold gray.

Again, these are my approximations of the four descriptions. The published CMYK and RGB values sit in the paper's appendix.

The details show how much judgment goes into one colour. The khaki brown was the hardest to get right. Giving it the same darkness as the greens made huge areas of the map look gloomy, and adding red made lowlands read as highlands, because many readers link red with height. The compromise was a neutral brown, two percent lighter than the greens when converted to grayscale.

The environment colours apply only below 1,000 metres. Above that line, the tints are conventional and identical everywhere, which the authors chose to keep the reds, grays and whites readers seem to like. The polar scale is a special case. It runs from gray-violet lowlands to nearly white highlands, since warm yellow and red do not suit an icecap.

Why bother with the lowlands only? The authors cite a 1998 study in the journal PNAS by Cohen and Small, which found that 74 percent of the world's people live below 500 metres. The bottom of the scale is where the readers are.

Did it catch on? At the time of writing, the cross-blended tints had been downloaded 25,172 times from Natural Earth, ahead of the 13,554 downloads for the site's land cover map over the same period. The authors admit they did not yet know whether ordinary readers understand the colours as both elevation and environment.

What a map scale teaches a chart designer

Hypsometric tints are a sequential colour scale with a story attached. The lessons carry over.

Lightness carries the order. The long fight over "higher is darker" and "higher is lighter" was about which direction readers take as "more." Whatever you choose, make the lightness steps move one way only. Our colour scale generator builds ramps you can check against that rule, and the glossary entry on luminance explains why it works.

Familiar colours carry baggage. Green already means vegetation to your readers, and red already means "high" or "danger." A scale that borrows those meanings gets read through them, whether you wanted that or not. The same lesson runs through our guide to building a colour system for data visualisation that does not lie.

Spectral scales are contested. Papen's blue middle elevation resembles the rainbow colour maps used in scientific visualisation. A 2007 article in IEEE Computer Graphics and Applications by David Borland and Russell Taylor, titled "Rainbow Color Map (Still) Considered Harmful," argued that the rainbow is a poor choice because it confuses viewers and can mislead. Colin Ware, Maureen Stone and Danielle Albers Szafir answered in 2023 that the advice had hardened into dogma and that rainbows can be useful for selected applications. The disagreement is live, so pick the scale for the task and test it.

Ready-made schemes exist for a reason. ColorBrewer, by Cynthia Brewer, Mark Harrower and Penn State, sorts map colour advice into sequential, diverging and qualitative schemes. For a first pass at a chart, start there, then check the result with a colour blindness simulator. Our accessible palettes guide for data visualisation covers the testing step.

Say what the colours mean. Patterson and Jenny conclude that a physical map needs a legend and a backdrop map may not. The same goes for any chart with a colour scale. If a reader has to guess whether green means "low" or "lush," you have already lost them.

Next time you see a green Persian Gulf, remember that nobody is claiming the desert is a forest. A century of cartographers just agreed to use green as shorthand for "low." The agreement is clear to the people who made it and a puzzle to everyone else.

Sources

• Patterson and Jenny, "The Development and Rationale of Cross-blended Hypsometric Tints," Cartographic Perspectives 69 (2011)
• Borland and Taylor, "Rainbow Color Map (Still) Considered Harmful," IEEE Computer Graphics and Applications (2007), DOI record
• Ware, Stone and Szafir, "Rainbow Colormaps Are Not All Bad," IEEE Computer Graphics and Applications 43(3) (2023), NOAA Institutional Repository
• ColorBrewer 2.0, Cynthia Brewer, Mark Harrower and The Pennsylvania State University