The OP output above is directly from the PNG export.
Attached is an example of zooming into the Corpus area.
Where possible, I'm trying to use pre-existing GIS data (for example, roads, land use, rivers, lakes, water).
For settlement data, I have a settlement database shapefile (that's where the smaller text "Robstown: 11,487" comes from - it's settlement name, and current 2022 population).
In T2K terms, settlements need to be re-input.
Right now, what I am doing is creating 4 different shapefiles.
1. Destroyed urban areas (Houston et al)
2. Intact large urban areas (e.g., San Antonio)
3. Settlement symbols (village, town, small city) - large cities are visualized by #2.
4. Settlement names (also has a size component)
This could technically be loaded into an Excel file also, here's basically the data needed:
1. Coordinates (lat, lon)
2. Name (string)
3. Size (village, town, small city, large city)
4. Intact (yes/no)
5. Hidden (large metro areas have dozens of towns, usually suppress all but the major ones).
From a sizing perspective, the basic heuristic I've settled on is using current 2022 pops:
<4,000 = village
<40,000 = town
<200,000 = small city
>200,000 = large city
City size affects city symbol and font size display rendering.
Comment