What is hypsometric tone

Choose a color ramp for elevation?


This may be more subjective as far as style goes, but I'd like some decent gradients for my DEM when it's over my hill shadow.

I designed the Swiss Hillshade model shown here, but I'm not too convinced that the color scheme is good.

Any examples or screenshots would be great.

Reply:


I usually come to this question from the perspective of "What will improve my data and not obscure it?"

Tufte talks about the use of colors in cards: Label , measure up , Represent and Animate . The choice of DEM colors is usually mostly for the latter (enlivening) - so that they look good. For example, the standard "atlas coloring" of many of the maps I see is really pretty (and is used in the Swiss shading example) - it comes from something that seems "natural": white (snow) at high altitude, green (forests ) on the lower slopes yellow / brown (the plains) and blue (the sea). It looks beautiful combined with a hill shade.

However, if you apply it everywhere, you get card colors that aren't are representative - they do not reflect reality and (worse) can be completely misleading. A map of the Luangwa Valley I created made it look like the Alps, even though it is a deep, hot valley bordered on one side by a cool escarpment (no snow, no large bodies of water).

Also, all of these colors don't photocopy well, and many offices don't have color printers that I work in, so all the colors disappear or just black spots appear.

Colorbrewer lets you explore some of the sequential color schemes that you can use on your cards. You can choose "Photocopiable" and other options - but it's kind of depressing how few colors are left when you choose more restrictions.

The other problem ( Measurement ) with color gradients is that the eye interprets subdivisions between colors if they do not exist - grayscale can show different amounts much better (nice article on this in the visual.ly blog).

So ... I almost always use either:

  • Grayscale hillshading or
  • A two-tone scheme depicting the ecology or habitat of the area - often from dry (yellow or red) to wet (green).



First, let me say how much I appreciate this question. I've seen so many examples of inappropriate color palettes for digital elevation models that it's good to see people think about it. There are some really good answers here too, but here's my take. I doubt there is a generally good palette, but a group of palettes that are good for rendering elevations under different terrains and at different landscape scales. Another factor to consider is whether you plan on capping the DEM as well. These are my favorites:

(1) This is your standard height range, and is a great choice for a variety of terrains and scales:

(2) If you're dealing with slightly more mountainous terrain, you may want the palette to start with deeper greens and end with white. This is best if you want to make a background image transparent, otherwise the white can be interpreted as the background.

(3) The following is a common palette used in atlases, and I find it best for rendering smaller scales (i.e. large geographic regions).

(4) This palette course is useful if you have suppressed the topography or if you have overlaid a lot of information and you want the topography to take a back seat. It is also best if a silhouette is used.

(5) Finally, the following spectrum color palette is useful if you do not want to use the typical green for the valley floor and brown for the relief of the mountains. This could either be because you are dealing with a very large scale (e.g. a finely resolved DEM of a small area where there is no relief on the mountain scale) or because the DEM is not of height depends, but on another attribute.

If you are displaying the DEM transparently over a shadow image (or vice versa), the transparency level is an equally important consideration for the palette. Compare the above image to the following, both using the same palette:

How vivid should the colors be compared to the shade tint? This likely depends on the application (i.e. the purpose of the card) and the other information overlaid.

You will notice that I did not include a grayscale image above. I prefer not to render DEMs in grayscale: 1) you can't make a composite relief model (i.e. a DEM and a Hillshade image like above), and 2) all of the palettes used above will automatically interpret these as height while I'm on a Grayscale Image knows my immediate reaction is to interpret it as a satellite image or an aerial image. It takes me a moment to realize that it is indeed a height. So unless it is an application that is supposed to be used to print the map (i.e. the editor told me that the image must be in grayscale) then I am not using a grayscale palette. And usually, in those cases where it has to be grayscale, I'd rather use a shaded image to convey the topography than DEM (if I have to choose one then I have

Thank you again for asking such a wonderful and important question!







Simbamangu's very good answer deals with one of the topics: The basic problem with shades of elevation that do not use neutral grays is the inevitable tendency to interpret the meaning of the colors. A common rendering technique, for example, is to use deep greens for the valley floor, which, with increasing lightening, move on an incline through brown and cream tones, to finally reach the mountain peaks in white or light blue / purple. (The technique is known as hyspometric or elevation tint.)

This is very effective in conveying the shape of the terrain. The lows are low and the heights are high with no need to interpret contour lines. However, that particular valley floor, drawn in lush shades of green, may actually be a semi-arid desert, and that bare looking mountain slope at high altitude may actually be a cloud rainforest.

The solution to counteract the tendency to misinterpret color values ​​is to use a technique known as " Cross-Blended Hypsometric Tints " and is excellently described by the undoubtedly well-known Tom Patterson. In cross-blended hyspometry, the gray-shaded relief provides the light and dark of the shadow, while another layer, e.g. B. classified vegetation cover is used to provide color values. The result is something where green actually means vegetation, and brown is rocky or barren, and white is covered in snow and ice, and light or darkness indicates relative altitude.

Just use grayscale in projects where there is no time or resources to blend in.



In addition to what has been mentioned, there are a few other things to keep in mind. You can improve the three-dimensional impression of the map by varying not only the hue, but also the saturation, brightness and vibrancy. Saturated colors appear closer and are suitable for mountain peaks, while the plains and valleys can be more unsaturated or dull in color. The famous Swiss cartographer Eduard Imhof used these contrasts in his color gradients:

Hill shading combined with hypsometric colors Plate 6 from the book: Eduard Imhof, Kartographische Reliefpresentation, Berlin and New York 1982. Sheet format 26 x 17 cm; Source: library.ethz.ch:

He also suggests not using ramps of the same spacing and color, but using geometric steps like 0, 200, 500, 1000, 2000, 4000 m, etc. However, this always depends on the area you want to map. I've seen examples of mapping a full hypsometric color ramp onto low topography countries like the Netherlands with bizarre results. It is best to know the geography of your area and the features of the topography that you want to convey on the map. Example: I would like to color a map of Germany hypsometrically. The 200 meter mark is here, as is often the case, a good mark for visually separating the lowlands from the more mountainous areas (one could go from a greenish to a yellowish color here). The mountains in Germany can be divided into two different geographic types. 1) the low mountain range characteristic of many areas of central Germany and 2) the Alps in the extreme south. 1) has all the highest peaks just under 1500 m, while the German Alps rise above 2900 m. By limiting the color ramp to 1500, these two topographical areas are optically separated. See how the Alps stand out from the rest in the first example (probably not that easy to spot on this small scale). In the second example, they don't come into their own in terms of color (the shading helps here), but some ridges in the lowland mountains are more accentuated. There is no clear right or wrong. You just have to know what you want. By limiting the color ramp to 1500, these two topographical areas are optically separated. See how the Alps stand out from the rest in the first example (probably not that easy to spot on this small scale). In the second example, the colors don't come out as well (the shading helps here), but some ridges in the lowland mountains are more accentuated. There is no clear right or wrong. You just have to know what you want. By limiting the color ramp to 1500, these two topographical areas are optically separated. See how the Alps stand out from the rest in the first example (probably not that easy to spot on this small scale). In the second example, the colors don't come out as well (the shading helps here), but some ridges in the lowland mountains are more accentuated. There is no clear right or wrong. You just have to know what you want.

A colored Germany relief created by me:

Example from Wikipedia of the user Botaurus-stellaris:




So for most of the maps I use, I use Swiss shades (here is a screenshot from California's Central Sierra) and tend to use a red / beige-gray theme that is not too insistent on the philosophy described by Simbamangu Face looking loosely follows (i.e. it's not super colorful and in some places it's even a bit monotonous, but the information is conveyed and I can layer a lot more over it. I want the highs - in the middle - pop and the flat land - left - go back.

This version is a flattened GeoTiff, but the color values ​​(all r, g, b) I use are:

  • Aerial perspective (from this Esri toolbox): low values, white; high values, 79.79.79
  • Filtered hillshade: low values, white; high values, 84,84,84 - layer at 35% transparency
  • DEM: low values, 255, 252, 252; high values ​​242.220.208 - layer with 25% transparency

I hope this shows an approach. Good luck finding one for your card




I think choosing a good gradient depends on your intended use and geographic location. You can start with a few examples on ESRI's MappingCenter blog


I'm sure a lot of people already know this, but I came across an excellent source of lots of gradients for QGIS (SVG format). I use QGIS 2.4 for this.

  1. Install the SVG2ColR plugin
  2. Navigate to this site: SVG ramp

  3. Download the gradient you want

  4. Load the SVG gradient and export it as a style
  5. Settings -> Style Manager -> Gradient -> Share -> Import

Now you have an amazing range of gradients to choose from. There are lots of great elevation ramps on the site that I listed above. Many thanks to the site operators for the excellent work!

We use cookies and other tracking technologies to improve your browsing experience on our website, to show you personalized content and targeted ads, to analyze our website traffic, and to understand where our visitors are coming from.

By continuing, you consent to our use of cookies and other tracking technologies and affirm you're at least 16 years old or have consent from a parent or guardian.

You can read details in our Cookie policy and Privacy policy.