The problem of Changing scales

[Durakken]

How do you guys solve the problem of changing scales? Like say you want to create a map of a kingdom in a world... Obviously you can just places the mountains and forests where they are because they are only really broadly placed on the larger scales and it's arbitrary where you consider them stopped... Rivers and borders on the other hand need more of a concrete placement, but if you scale up they lose definition and visibility. A river might be incredibly important to an area but going to a world map might make it invisible and likewise, using techniques to make rivers (such as using a program) can create all sorts of problems as the area map will make the rivers go in ways that even a broad understanding of the world map it wouldn't go... So How do you guys get past this? Or do you just ignore inconsistencies like this?

[xpian]

Speaking personally, I take my cue from existing cartography experts, like National Geographic. For instance, if you look at a map of the U.S. and you trace the Mississippi river, you'll see it more or less flowing down to the Gulf, north to south, with lots of good little "meanders" or squiggles thrown in. But if you zoom in on sections of the Missouri-Arkansas-Louisiana region by getting another, much more detailed map, you'll see a ridiculous amount of river meandering, with tons of oxbow lakes all over the place, and marshy wetlands where the river once flowed, etc. All that stuff was, of course, not drawn on the big map of the U.S. because you'd never be able to see it at that scale.

As far as I can tell, this is an issue that cartographers have always had to deal with, down through the centuries. You draw the level of detail that is appropriate for the scale of map you're making. You assume, and hope your viewership assumes, that there are lots of smaller details that would be apparent on any zoomed-up map.

Here's the important thing, as far as I'm concerned. When you do the zoomed up map of the smaller region, you can't put in any details that would blatantly contradict the look of the bigger, low-detail map. Sure, you make your zoomed-up river do lots more squiggly meandering, but if your simple river on the zoomed-out map flows south of a particular mountain, you've got to make the zoomed-up river flow in roughly the same course. If the zoomed-out map shows two rivers clearly converging upstream from a particular city, it might be a mistake to then draw the zoomed-up map with those same rivers actually converging within the city walls. That would confuse people looking at both maps. Maybe, when you're working on the zoomed-up map, you utilize a semi-transparent overlay of the zoomed-out map to remind you where everything goes.

The biggest issue for me, personally, is how to represent mountains at the various scales. In fantasy mapping, we tend to draw a lot of isometric-style mountains that are fun and evocative, but which are less realistic than they could be, compared to real-world cartography. If you do a wide map with mountains like that, and then do a zoomed-up region that includes some of those mountains...well, they're going to look silly if you just scale them right up. Do you re-draw the whole mountain range at the new scale? Do you take cues from the zoomed-out map while doing so? It's an artistic problem to solve, for sure.

[J.Edward]

Much of what xpian said is on point.
That's why I work at doing better top down mountains.
You can solve a lot of problems that way. That said, they're not as popular as iso mountains and often for good reason.
Top down mountains are often not as interesting to look at but they allow you a lot more accuracy when laying out land features.

I usually draw up where all my ridge lines are and those often determine water flow.
To do more zoomed in maps, I often do what xpian said and utilize a layer from the larger scale map.
In the end you still have to redraw everything but most of your details are there to work with.

I also try not to let the story get too far ahead of the map. That seems to be where people start getting the inconsistencies.
They have this and that here in the story but that doesn't work with the geography.

[waldronate]

A critical part of mapmaking is abstraction. Elements that are important to the purpose of the map will show, regardless of scale. Less important details will drop out first.

I often see people who want to draw whole-world maps that have detail down to a fraction of a mile and then complain when they can't get their system to handle that. I hear whining when I suggest that they would be much better served by drawing a whole-world map that only shows details relevant at a world scale and drawing several maps that show smaller areas in greater detail. It's a lot of work!
A world map might show just large mountain ranges, major rivers, large country outlines, capital cities, that teeny monastery that is critical to the plot of the story, and so on. A regional map might show smaller mountain ranges, interesting topography, smaller country outlines, villages, and (once again) that teeny monastery that is critical to the plot of the story. A local map of the teeny monastery will show things that are important to the plot and a few red herrings.

A fun part of historic maps is their accuracy. Many elements on a map are inaccurately placed or perhaps don't exist at all! Similarly, important physical elements in the world might not be represented on a map at all if those elements aren't important to the purpose of the map. Somewhere here on the guild is a map from a few years back that shows a map that a group of adventurers bought in a major city and it has redlines showing what they actually found.

[J.Edward]

I agree with waldronate. Most land features really don't show strongly on a world map.
Looking at Earth from space, you're hard pressed to even see a river. They're just not that big.
You can see evidence of rivers but seldom can you actually see the river itself.
Seas and very large lakes show but they have to be quite large.
You can see where ranges of mountains are by their shadows and maybe snow and glacier cover but not individual mountains.

There is a certain mystery in making a larger region map that doesn't show you everything.
Then you have to go in to a region or closer to get any detailed information.
We're so used to satellite imagery where we can see the whole world and zoom in and out.
That's cool and useful for world building but doesn't do quite as well for a map.
There is also an appeal to being able to see everything at once. So I can appreciate both points.

[waldronate]

As with literature, reading works similar to what you want to achieve can help with spacing and elements. David Rumsey's map collection http://www.davidrumsey.com/view/atla...ases-1570-1800 has an excellent selection of classic world atlases (the rest of the site is nice as well). Looking at how the zooms are handled shows one way that details can be handled. It also points up the significant differences in scale shape between mapmakers and over time. It wasn't until the late 1700s that even the coastlines were reasonably accurate. It was another century before continental interiors were well-mapped and not until the late 20th century that that things were accurately mapped worldwide.

[xpian]

I agree with J.Edward -- you can rarely see rivers from a satellite view. One notable exception being, perhaps, my favorite river: the Zambezi. At flood stage in the rainy season.

[J.Edward]

I agree with J.Edward -- you can rarely see rivers from a satellite view. One notable exception being, perhaps, my favorite river: the Zambezi. At flood stage in the rainy season.

I just had to map the Zambezi for a commission a few months back.

[waldronate]

The big crow's-foot thing in the middle is the Okavango delta, unrelated to the Zambezi, isn't it? The "lake" in the upper-right would be the Zambezi and the lake would be flooded marshland that's greened up from the flood. The coloration on the Okavango delta is also due more to the vegetation than to actual water surface. The champions for visibility due to vegetation are, of course, the Nile and the Indus rivers, because they flow through otherwise tan desert (the Volga and Amu are also impressive, but much smaller)

There are a few rivers that you can see the water from satellite images, but usually only right at their mouths (the Ganges and Amazon are the best two examples). It's as much from their coloring as it is from their width, though, because the muddy water contrasts nicely with the dark green and blue of the land and sea.

Anyway... Enough rambling for me for now.

[xpian]

@waldronata - You're right about the Okavango...it just spreads out into that vast delta and evaporates, never reaching the Zambezi or the sea. [ Actually, it turns out that every so often, when the water is really high, it overflows into the Zambezi after all: "In very wet years, a part of the river's flow may extend along the Magweggana River (actually a northeastern distributary of the Okavango delta) and enter the Zambezi river, bypassing the Kalahari." ]

However, the Okavango IS quite flooded during the rainy season, with the surface water spreading out over a 250 km by 150 km region. The people who live there get around on boats in many places. The maps of the region depict it as almost a seasonal lake. I've watched hours of glorious nature documentaries showing animals swimming through the Okavango flood. https://en.wikipedia.org/wiki/Okavango_Delta

The Zambezi, by contrast, does indeed flow to the sea, and in so doing forms the world's largest sheet of falling water: Victoria Falls or Mosi-oa-Tunya, "the Smoke that Thunders." This satellite view shows the Zambezi flowing in from the upper left. It then hits the gorge the that forms the falls at the bright line of white foam horizontally across the upper middle of the picture.