The heightmap is based on SRTM3 data with a bit of manual scaling of the heights. When I load it into the scenario editor I usually specify a maximum map height of 21, but you can obviously use any value you wish. The southern part of the map is mountainous while the northern part is virtually flat.
The scenario has towns and larger lakes positioned as in the real world, subject to the constraints of a 1024x1024 grid and not wanting to have towns too close together. Latitudes/longitudes for deriving positions were taken from either Wikipedia or OpenStreetMap+Maperitive. The town sizes are loosely based on the current populations, but scaled back to allow for growth as the scenario starts in 1926. Non-Guatemalan towns have their two letter country code after their name (BZ for Belize, SV for El Salvador, HN for Honduras, MX for Mexico); Guatemala is privileged because it's the centre of the map and has more towns on the map than any other country.
The rivers are not based on the real world; there were just too many of them and none of them seemed particularly iconic. I am not saying I won't revist this in the future, but I don't have any appetite for it at the moment. As a compromise I used ic111's "Rainfall River Generator" (viewtopic.php?f=33&t=71970&start=260) to generate the rivers, and while I'm no expert they look fairly good to me - some of them even seem to agree with rivers I can see on the real map.
This is a sub-tropical scenario, although with very little desert; the industries and resources from the sub-tropical climate seemed a better fit for the region than the temperate ones, and as I'd never really played in sub-tropical before it seemed an interesting experiment. I'm not sure how accurately placed the desert or rainforest is compared to the real world; I eyeballed http://www.ecologicalregions.info/htm/sa_eco.htm and winged it a little bit here.
Industries were placed at random using the scenario editor.
The scenario uses the Tropic Refurbishment Set and Tropic Refurbishment Set Extender NewGRFs to overcome the limited default selection of sub-tropical trains. I've also included the Earlier Road Vehicles NewGRF to allow the scenario to start in 1926 for the early TRS trains without leaving road vehicles unavailable. Finally, the redFISH NewGRF is included to add some interesting ships; I quite like using the log tug on the rivers in the rainforest-heavy northern part of the map, for example.
I've never created a scenario before, this was started on a sudden inexplicable whim and then turned into a minor obsession. Any constructive feedback is welcome.
Feel free to make your own derivatives if you wish. This work is licensed under the Creative Commons Attribution-ShareAlike 3.0 Unported License. To view a copy of this license, visit http://creativecommons.org/licenses/by-sa/3.0/ or send a letter to Creative Commons, PO Box 1866, Mountain View, CA 94042, USA.
- NASA for the SRTM3 heightmap data
- Python "elevation" package for the eio tool used to download the SRTM3 data
- gdaltransform from the Geospatial Data Abstraction Library was used to convert between latitude/longitude and heightmap X/Y coordinates
- ic111 for the Rainfall River Generator
- wallyweb for providing a binary of the Rainfall River Generator
- Maperitive (used to browse the OpenStreetMap data and extract latitudes/longitudes from it)
- http://www.ecologicalregions.info/ecoregions.htm (approximate source for the rainforest/desert regions)