Help for world elevation map

[waldronate]

If you're after accuracy, don't forget that the Earth is about 20km shorter in radius through the poles than it is through the equator (it's roughly 1 part in 300 flattening through the poles). That pretty much swamps the 6km altitude of Kilimanjaro at the equator. SJS also forgot to mention that 90+% of the Earth's surface is less than 1000 meters is elevation (assuming that you want sea level to be roughly flat). That's very approximately 1/400 of an inch; I've always found 400 grit sandpaper (1/400 inch particles) to be delightfully smooth to the touch.

The biggest problem that I've always had with paper is that it changes size with humidity (typically a few percent). I live in a desert that has lots of indoor humidity in summer and almost none in winter, so I've watched quite a few projects crack or wrinkle over the years...

[SJS]

I was coming back to mention the bulge, but waldronate beat me to the punch... in some respects, the equatorial bulge is the biggest "mountain" on the planet, since one could, I suppose, measure heights on the earth from the earth's center, rather than sea level. In that case, the tallest "mountain" would likely be a mountain in the Andes, near the equator.

Even the bulge is barely visible to the eye, so that, from the moon, the Earth still looks like a circle rather than a flattened circle (an "oblate spheroid").

That sandpaper analogy was excellent, waldronate.

[SeaAngel]

If you're after accuracy, don't forget that the Earth is about 20km shorter in radius through the poles than it is through the equator (it's roughly 1 part in 300 flattening through the poles). That pretty much swamps the 6km altitude of Kilimanjaro at the equator. SJS also forgot to mention that 90+% of the Earth's surface is less than 1000 meters is elevation (assuming that you want sea level to be roughly flat). That's very approximately 1/400 of an inch; I've always found 400 grit sandpaper (1/400 inch particles) to be delightfully smooth to the touch.

The biggest problem that I've always had with paper is that it changes size with humidity (typically a few percent). I live in a desert that has lots of indoor humidity in summer and almost none in winter, so I've watched quite a few projects crack or wrinkle over the years...

I was coming back to mention the bulge, but waldronate beat me to the punch... in some respects, the equatorial bulge is the biggest "mountain" on the planet, since one could, I suppose, measure heights on the earth from the earth's center, rather than sea level. In that case, the tallest "mountain" would likely be a mountain in the Andes, near the equator.

Even the bulge is barely visible to the eye, so that, from the moon, the Earth still looks like a circle rather than a flattened circle (an "oblate spheroid").

That sandpaper analogy was excellent, waldronate.

Uh, to be honest... I thought that the sinusoidal projection would take care of the equatorial bulge....? It doesn't? :-/

Also, I'm afraid I didn't catch the sandpaper analogy. Could you explain it, please? :-)

[SJS]

Uh, to be honest... I thought that the sinusoidal projection would take care of the equatorial bulge....? It doesn't? :-/

Also, I'm afraid I didn't catch the sandpaper analogy. Could you explain it, please? :-)

I wish I knew how big the sinusoidal projection is (that is, how big the globe would be) but my guess is that even the bulge, which is 3 times the size of most mountains, is still so small proportionally to the diameter of the earth that it makes no difference. In other words, if someone made a sinusoidal projection map with the correction and a second one without the correction, the two would be indistinguishable (unless, maybe, the globe was 50 miles across).

The analogy I thought of which is functionally equivalent to the sandpaper analogy is the giant titan analogy. Imagine you were some humanoid god-like being who was positively enormous (like bigger than the planet Jupiter). Lets say your finger is the size of Europe. If you were to encounter the Earth, hanging in space, and you were to rub your finger over Asia, you wouldn't even feel the bumps of the Himalayas. The Earth would seem to you, the planetary giant, as being perfectly smooth.

The sandpaper analogy was going in the other direction. Imagine you shrink the Earth to the size of a desktop globe (even a big globe, 4 feet across). As you shrunk it, you'd see Mt. Everest clearly, but it would get smaller and smaller in proportion to the rest of the Earth. How big would that bump of Everest be when you had the Earth all shrunk down? Invisible. Unfeelable. Similarly, sandpapers are rated by how gritty they are (how big the bumps are). The finest sandpaper technically still has bumps, but so slightly that, rubbing your finger across it, it feels perfectly smooth.

[SeaAngel]

Oh, I get it now, thanks! :-)

I also understand why, in terms of accuracy, a globe which can be held with one hand, should show neither mountains nor the equator as bulges... But I haven't clarified why I wanted to make the globe in the first place: I want to get a feel of how the earth's geography allows or inhibits organisms from expanding in various places. Biologically speaking, a population of animals can easily (given enough time) enhabit a large flat landmass. But if the altitude changes abruptly, it becomes a barrier due to temperature and moisture changes.
This anomaly between the effects of distance and altitude (distance being a smaller obstacle for populations compared with altitude) is why I'm willing to overlook the inaccuracies you have all mentioned :-/

[waldronate]

If you're interested in how terrain affects distribution, then you're probably much more likely to get use out of a map that shows altitude (and perhaps temperature and/or rainfall) as color or that has significant distortion of altitude. As I've mentioned in other places, dime-store or thrift-store globes are pretty cheap and most of the basic models from Replogle have terrain molded in already. You can paint them or add an overlay layer with just terrain to get rid of the distracting human things like country boundaries and names. Or you can go with a digital globe like the basic Cesium - WebGL Virtual Globe and Map Engine model.

To reuse one of my favorite pictures:


and http://s3images.coroflot.com/user_fi...mFqD3GTPK6.jpg is also helpful with understanding how things work at a gross level.

http://pubs.usgs.gov/of/1993/0380a/report.pdf is fun, http://volcanoes.usgs.gov/about/edu/.../ballglobe.pdf is cleaner, and http://astrogeology.usgs.gov/maps/pl...aps-and-globes has good examples of a common globe manufacturer layout (none for earth, but I can gen up some of these if you'd like). http://www.vendian.org/mncharity/dir3/planet_globes/ is fun, but I don't think that it adds much beyond the others. And I just came across http://www.brighthubeducation.com/he...for-a-project/ that looks like fun.

[SJS]

I have an old globe at home that has bumps for mountains - clearly not to scale of the size of the earth, and I don't know how accurate it is anyway (the Rockies feel higher than the Appalachians, but I don't know if each individual mountain is represented or if it's just a regional-type effect). Anyway, you could always do what my globe does - just pick a different scale for elevations than you do for overland distances. Alternatively, rather than doing the whole earth you could do a small enough region that the elevations can be seen in scale with distances.

[SeaAngel]

If you're interested in how terrain affects distribution, then you're probably much more likely to get use out of a map that shows altitude (and perhaps temperature and/or rainfall) as color or that has significant distortion of altitude. As I've mentioned in other places, dime-store or thrift-store globes are pretty cheap and most of the basic models from Replogle have terrain molded in already. You can paint them or add an overlay layer with just terrain to get rid of the distracting human things like country boundaries and names. Or you can go with a digital globe like the basic Cesium - WebGL Virtual Globe and Map Engine model.

To reuse one of my favorite pictures:


and http://s3images.coroflot.com/user_fi...mFqD3GTPK6.jpg is also helpful with understanding how things work at a gross level.

http://pubs.usgs.gov/of/1993/0380a/report.pdf is fun, http://volcanoes.usgs.gov/about/edu/.../ballglobe.pdf is cleaner, and Maps and Globes | USGS Astrogeology Science Center has good examples of a common globe manufacturer layout (none for earth, but I can gen up some of these if you'd like). Making globes of the planets is fun, but I don't think that it adds much beyond the others. And I just came across How to Make a Homemade Globe Using Print-and-Assemble Capability that looks like fun.

Thanks for the info I especially liked the vendian site, it has great source maps!

Anyway, you could always do what my globe does - just pick a different scale for elevations than you do for overland distances.

This is exactly what I plan to do :-) According to wikipedia, when it comes to temperature, moving up 100 metres (330 ft) on a mountain is roughly equivalent to moving 80 kilometres (45 miles or 0.75° of latitude) towards the pole. I'll see if I can use that ratio...