The Continent of Vindarten

[buzzgunner]

Thanks very much! I'll be posting a revised version of the map, incorporating many of the suggested changes, for further comment.

[Hai-Etlik]

I didn't post it as a WIP as I thought that it was finished. Clearly, based on the responses, that's not the case. Don't get me wrong, I appreciate the feedback and plan to act on all the good suggestions. Is it possible to move this thread to "Region/World Mapping" section as a WIP? (I assume a moderator or admin will need to do this.)

I would, however, like to comment on the observation that Elterio Delgard made regarding the rivers.

Realistically, at the scale of my map, rivers shouldn't even be visible. For example, here's a shot of the Grand Canyon (clipped from Google Earth), at an altitude much much lower than that of my map:

Attachment 84388

As you can see, while Lake Powell and Lake Mead are visible, as is the canyon itself, the Colorado River is too small to be seen.

Regarding the missing vegetation along the sections of river passing through desert regions, this would be a problem if those regions were being irrigated and crops were being grown. However, if no human activity is present, there's no reason why substantial plant life should be visible at the given altitude. Here, for example, is an image (again from Google Earth) of the present-day Upper Nile:

Attachment 84389

This is at roughly the same altitude as the Grand Canyon image (above). Note that while broad areas of vegetation are visible in some locations, much of the land immediately adjacent the river is barren. Given that areas of my map that show rivers through desert regions are unpopulated, I designed them to appear the same.

It's important not to conflate orthophotos with maps. You can build maps on top of orthophotos, and it's often useful to consider a notional top down view when designing the symbology for a map in order to make it more intuitive, but you should keep it at the level of an abstract appeal to intuition (Forests are often symbolized in green for example). If you are trying to fake an orthophoto from scratch, that's really challenging and an "all or nothing" proposition. If you get caught in between you get something that's horribly fake looking as an orthophoto, and unfocused and hard to read as a map.

So, if its a map, and rivers are important to what the map is trying to do, then draw them clearly with sufficient size and contrasting colours to make them easily visible. If it's a map and rivers aren't important to what you are doing, don't include them at all as they would just cause clutter and detract from what is imponrtant. If you are trying to fake an orthophoto, then you need to spend a lot of time looking at comparable orthophotos and learning about physical geography, remote sensing, and image processing to be able to pull it off. If I were going to try to learn to fake orthophotos, I'd probably start by taking real ones, and then trying to replicate them myself from information about the physical geography behind them.

[buzzgunner]

I agree with you, vis-a-vis maps vs. orthophotos. My reply to Elterio Delgard's comment on my rivers was to simply illustrate that a river traversing a desert region does not require adjacent vegetation to be "realistic". I am however redoing my rivers to make them look more organic and less like blunt lines drawn on the map. Whether or not I've succeeded will have to wait until the next round of critiques when I post the revised map!

[buzzgunner]

You also have seriously mismatched information about the projection. The equally spaced straight parallels say this is normal equidistant cylindrical, which would be a horrible projection for something in this range of latitudes; even if you picked a standard parallel to minimize distortion there would be massive distortion somewhere. Over this extent it also wouldn't be either distance or bearing preserving which means the scale bar and compass rose can't possibly be correct. There's no way to even come close to preserving both scale and compass bearings at this extent.

And yet, isn't that what a Robinson projection does? If you assume that compass rose on my map is at 0° longitude, then isn't my map effectively a Robinson projection with the longitude lines removed? I don't mean to be militant about this, but it seems that there are a number of existing accepted map projections that don't meet your specifications.

If I'm wrong (and I'm certainly not an expert cartographer, as illustrated by my map), could you please elaborate for me? Thanks!

[Hai-Etlik]

And yet, isn't that what a Robinson projection does? If you assume that compass rose on my map is at 0° longitude, then isn't my map effectively a Robinson projection with the longitude lines removed? I don't mean to be militant about this, but it seems that there are a number of existing accepted map projections that don't meet your specifications.

If I'm wrong (and I'm certainly not an expert cartographer, as illustrated by my map), could you please elaborate for me? Thanks!

A compass rose indicates the map is bearing preserving (North is always the same direction anywhere on the map, and northeast is always 45 degrees from north anywhere on the map). Robinson is not bearing preserving. There is only one projection that is bearing preserving at small scales (Big area drawn small) and that's Normal Mercator. A compass rose on a Robinson map is wrong, as is a linear scale bar. Using Robinson for a restricted extent is arguably an abuse of the projection as it's a specialist projection for full planet general reference maps. That's a very specialized role and it breaks down outside of it.

It's hard to explain without a lot of typing or diagrams that I'm not in a position to put together until my thumb heals but no normal aspect cylindrical or pseudocylindrical projection (where parallels of latitude project to straight parallel lines) is going to be remotely appropriate for the extent of that map. The quickest explanation I can give is that in order to minimize the distortion, the projection needs to be tilted so lit lays flat over the area being mapped and normal aspect (pseudo)cynindrical projections are flat to the equator. (that's a gross simplification)

[buzzgunner]

A compass rose indicates the map is bearing preserving (North is always the same direction anywhere on the map, and northeast is always 45 degrees from north anywhere on the map). Robinson is not bearing preserving. There is only one projection that is bearing preserving at small scales (Big area drawn small) and that's Normal Mercator. A compass rose on a Robinson map is wrong, as is a linear scale bar. Using Robinson for a restricted extent is arguably an abuse of the projection as it's a specialist projection for full planet general reference maps. That's a very specialized role and it breaks down outside of it.

It's hard to explain without a lot of typing or diagrams that I'm not in a position to put together until my thumb heals but no normal aspect cylindrical or pseudocylindrical projection (where parallels of latitude project to straight parallel lines) is going to be remotely appropriate for the extent of that map. The quickest explanation I can give is that in order to minimize the distortion, the projection needs to be tilted so lit lays flat over the area being mapped and normal aspect (pseudo)cynindrical projections are flat to the equator. (that's a gross simplification)

The diagrams aren't necessary. Your explanation makes perfect sense to me.