Improving national borders - LSIB import?

I just noticed that parts of the Thai-Malaysian border had been replaced with data from an LSIB import. I checked, and it appeared to be much more detailed and accurate than the previous lines from the CIA World Database. I’ve also been looking at some portions of the currently mapped Thai-Myanmar border and found accuracy to be rather lacking. An LSIB import to replace these borders (and those with Cambodia) sounds like a good idea to me. Sections that follow rivers which have been manually traced should of course be kept. Any comments?

If you’re willing to do the significant amount of work this will require, of course I think you should proceed. Too bad we can’t find something like this for the national parks.


I’ve gone ahead and done most of the Thai-Myanmar border.

Between Mae Sot and Myawaddy there’s a section of the border from the import (dark line below) that doesn’t agree with the currently drawn one.

The border is supposed to follow the river, but apparently the river seems to have changed course since the border was demarcated. In this case, which is the “correct” boundary?

If the import is deemed a more accurate representation of the border, should we just overwrite the manually drawn boundaries with the import?

A belated update. (I looked this up a while ago but forgot I’d asked here.) It appears to be pretty much standard practice to adhere to the text of the treaty, so when the river changes course the boundary follows it.

OTOH, the Lao-Thai boundary is incorrect in several places. Except for a handful of named locations, all islands are deemed to be on the Laotian side. I might get around to fixing it when I have the time.

I’ve redone the Lao-Thai border.