Aah, thanks for the clarification! I see the problem. (Purely technologically, I find it much easier to give references for snippets of information on Wikidata.)

I’m not so pessimistic about the discrepancies between this 100 year old source and current usage. Sure, there are errors and systematic shortcomings in the gazetteer, and there are bound to be changes over time. But I wouldn’t go so far as to predict that thousands of names have become obsolete – after all, there hasn’t been any massive shift in population or language.

Ok, maybe I’ll try your free Green Marble version 1 as base map next time :-)

As a linguist, I don’t quite think that names and their referents are completely non-verifiable – it seems ‘just’ a matter of elicitation and cognitive theory (fuzzy boundaries!) to me. However, I absolutely agree that the marginality or peripherality of this region certainly has a big impact on quality of data curated – broadly generalized – in the West. The number of duplicate GeoNames entries in the area – automatically propagated to Wikidata and by Lsjbot to svwiki and cebwiki – is rather disheartening.

Thanks for your suggestion regarding the image(s)! I only created the last one, and I roughly described the processing steps here. I’m completely self-taught in remote sensing, so I’m totally prepared to hear that I made rookie mistakes! If you have any constructive feedback you’d be willing to share, I’d be curious! (Btw, you’re gonna … uh … love what I did to the surface reflectance data.)

Inspired, among others, by this.