The periphery of this map is composed of the watersheds of Asia, Europe and Africa. Each basin being attached at the ocean, sea or bay into which it drains. Inland basins, such as the Caspian, are attached at their lowest overflow points, as it were, their spouts. Deserts, such as the Sahara, follow the same logic. The threadlike white line seen in the Americas is the continental divide; it forms the edge of the Style “N” map (following this post).
The large red arrowhead at the map’s bottom, just a bit left of center, is where — so they say, if there are such things — great waves occur; the Indian Ocean surface current trying its best to squeeze its way around the southern tip of Africa and getting pinched now and then by the three currents to the south of it going the other direction.
world maps with constant-scale natural boundaries 
August 27, 2009 at 4:23 pm
awesome! where’d you get the ocean currents layer from?
August 27, 2009 at 4:59 pm
Ocean currents are from several sources, mostly National Geographic, with a few added tweaks — I emphasized the Antarctic Bay’s down-spirals to emphasize the association between the currents and the bays. National Geographic was vague, although to cut them some slack, they used a standard projection that distorted polar regions, making the Bay-Current relationship appear uncertain. I may have overemphasized it. Maybe an ocean current expert will weigh in and let us know the real scoop.
Also, in order to focus on the global patterns, I omitted many local (sub-basin) surface currents. In my opinion, most sources overdo these local surface currents at the expense of global patterns.
February 4, 2010 at 8:58 pm
This is phenomenal.
I read somewhere once that the current through the Bering Strait is mostly fresh water from the Yukon river.
You would have to do some research about where to make the cut through the Bering Sea, but then you could unzip the Americas and have more of a Spilhaus type map. It would be a great teaching tool for the thermohaline ciruclation.
February 4, 2010 at 9:16 pm
Michael, Thank you for the comment. I’m not sure what you mean by “you could unzip the Americas and have more of a Spilhaus type map” If you mean to imagine taking map “M” and ALSO unzipping the Americas, you’d wind up, if I understand you, with this map: https://rightbasicbuilding.com/2008/05/06/csnb-map-2/ Or perhaps you mean something else? Chuck
February 5, 2010 at 3:02 am
Yes, that’s the map I meant.
If I understand correctly, the difference between “M” and what I’m calling “Spilhaus” is that “M” keeps the oceans contiguous.
There are places in the middle of the ocean that aren’t “contiguous.” They are kind of like continental divides, where a drop of rain might end up going north with one current or south with another.
February 5, 2010 at 3:42 am
I think we both are accurately interpreting what each is saying. Pronouns always throw me into a bit of uncertainty. To be clear, none of the maps here are “Spilhaus” maps. On CSNB Map #2, the interruptions in the oceans are what I call “current divides.” I find myself quite taken by how the narrowness of the Humboldt Current (up the west edge of South America) pairs up with the narrowness of the western continental divide of South America. But you may be focusing on something else. Not quite sure where you’re going with this. cc
February 6, 2010 at 1:12 am
It gets more and more impressive.
I had thought that CSNB#2 had been drawn by Mr. Spilhaus, and the interruptions in the oceans were arbitrary.
Where would I find a copy of Spilhaus & Snyder’s paper?
February 6, 2010 at 1:45 am
Spilhaus, A.F., and Snyder, J.P., Oct 1991. World Maps With Natural Boundaries. CGIS-ACSM, vol. 18, 4. pp. 246-254.
That’ll teach me to keep track of my acronyms.
ACSM is, I believe, the American Congress on Surveying and Mapping.
The CGIS is the The Cartography and Geographic Information Society Journal.
Slips my mind at the moment which library I tracked it down in. It may not be online.
February 7, 2010 at 4:07 am
I wasn’t expecting much online either, but there appears to be an atlas by Spilhaus on Google Books – full preview.
http://books.google.com/books?id=GEHVbeZwLGQC&printsec=frontcover&dq=spilhaus&cd=1#v=onepage&q=&f=false
February 7, 2010 at 9:29 am
Yeah, the Spilhaus Atlas, published by American Philosophical Society, is good. Astounding, really. When you compare it to what else was available.
Though I was working on CSNB Map #2 by 1994, & CSNB #1 in 1990, I didn’t stumble on Spilhaus’ work until 1996. As far as my own precedents, Spilhaus gets credit for hoisting the phrase “natural boundaries” (my terms were nowhere near as articulate), and Snyder gets credit for teaching me the importance of where a projection system locates “constant scale.”
I put together the two concepts.
In 2005, Mike Purucker added the necessary hyphen in “constant-scale” (it’s an adjective modifying “natural boundaries”).
I never met Snyder, though I’m told he was very particular about starting a project; you had to be very specific about what you wanted him to do. Spilhaus I talked to a couple of times in his last year — he’d just turned eighty-six when I tracked him down (I think that conversation is recounted somewhere on this blog, if not, then in my 2003 ISPRS paper Visual Calculus or Perceptual Fribble? — World Maps with Constant Scale Edges . . .). He was mentally strong, and continued working on a new ocean-centered map right up until his passing.
February 8, 2010 at 1:31 am
As you point out in your 2003 paper, this would work for any surface with hills and dales. Have you tried other data sets besides elevation?
The elevation maps of other planetary bodies are neat, but I’m thinking specifically of Waldo Tobler’s maps of demographic and economic flows. http://csiss.ncgia.ucsb.edu/clearinghouse/FlowMapper/
February 8, 2010 at 3:02 am
Sure. Constant-scale natural boundary mapping works with any natural “tree” you can mark on a globe.
Look at my main menu, click on any of the Earth tectonic maps; they use as map boundary various lines of trenches and mid-ocean spreading centers.
The crania maps also use boundary systems that are not strictly topographic ridges and valleys. And in CSNB map #2, the water interruptions are not elevationally derived either — they are edges of slipstreams.
I haven’t done it, but systems drawn from political boundaries (or any other geographic situation) would work; the main thing is to have a global focus, what Marston Morse called “in the large.” (Morse invented Morse Theory (so named by his admirers) a mathematical method of keeping track of all the critical points on an essentially globular object; Morse himself called it A Calculus of Variations in the Large. Morse Theory, with later improvements, continues to be basic for study of dynamics on 2-D surfaces embedded in three-space, things like earth’s climate, tectonics and global hydrology.)
When I look at Tobler’s material, I don’t see a global focus — many important phenomena don’t rise to the in-the-large category. Nothing in that to be ashamed of, and a world map showing, for example, this Tobler material wouldn’t, I think, add anything to the discussion. Maybe I’m not looking at the same maps you are?
Not that there aren’t global issues within demography and economics; it’s just that Tobler’s horizon, at least in the maps you cite, is too low to benefit from putting it on a world map of any projection, conventional or novel. Think about it this way: those USA maps he uses to show the data, they also have a “boundary” at the edge of the map, which limits rather than enlarges the scope. This is fine if your focus is regional. CSNB maps only come into play when a subject is “in the large.”
Not sure if that really helps. You tell me.
Another thing worth clarifying: much of this fuzzy geographical data like demography and economics may be organized topographically. The data has density and scarcity, which generate contour lines, which generate ridges and valleys, which are meat and potatoes for constant-scale natural boundary mapping.
Not to change scale over-much in a post on Earth ocean currents, but pardon me a moment while I Google-image-search electron density and imagine what those objects (and maps) would look like in a CSNB projection . . .
And then there are all sorts of possibilities with edges of slipstreams . . . Jupiter & Saturn; what else? You tell me . . .
July 19, 2010 at 4:41 pm
I would like to talk to you about Spilhaus maps and the phone conversations you had with Dr. Spilhaus. I am writing a biography about him for the National Sea Grant Program and would very much appreciate getting a “user’s perspective” on what his maps meant/mean to the world of cartography.
I think Dr. Spilhaus would be very pleased with the map you created for Watersheds “M” with Currents. Keep up the great works!