New idea: define climates in terms of four different variables:
Temperature: glacial, polar, subarctic, temperate, subtropical, tropical.
Temperature Pattern: oceanic, continental.
Precipitation Amount: 80+ inches, 60-80 inches, 40-60 inches, 20-40 inches, 10-20 inches, <10 inches.
Precipitation Distribution: Winter, Summer, Year-Round.
Technically- very technically- this means that you have 216 different climate types to work with. Luckily for us, probably less than half are going to show up in the vanilla raws, because not all these climates actually exist in the real world.
For example, let me create a tag for the city where I live in, that could show up in a raw. Pittsburgh gets a goodly amount of rain- 37 inches- per year, spread fairly evenly throughout it, and is temperate in climate, but it's quite cold in winter and quite hot in summer. So if, in vanilla DF, we accept that Pittsburgh's is the sort of climate that we want to appear, then in the raw file climate.txt, we will put in
[TEMPERATE:CONTINENTAL:20_40:YEAR],
Likewise, if we wanted to put in Ketchikan's climate, it would probably look something like this:
[SUBARCTIC:OCEANIC:80:WINTER]
(I'm just going to say that an oceanic subarctic climate as we've defined it will usually be above freezing, but not by much. Ketchikan will still get summer rain- it'll just get much more winter rain.)
London would look like this:
[TEMPERATE:OCEANIC:20_40:YEAR]
Iquitos, Peru, would look like this:
[TROPICAL:OCEANIC:80:YEAR]
(Really, all tropical climates will be defined as oceanic in the real world, because of their constant temperatures.)
Phoenix, Arizona, would look like this:
[SUBTROPICAL:CONTINENTAL:10:YEAR]
The South Pole:
[GLACIAL:CONTINENTAL:10:YEAR] (although glaciers will get so amazingly little precipitation that [YEAR] is basically meaningless.)
Cambridge Bay, Nunavut:
[POLAR:CONTINENTAL:10:SUMMER]
However, what are we to make of the following tag?
[POLAR:OCEANIC:40_60:YEAR]
That doesn't even really exist in the real world. Tundra climates are always continental, and they always get less than 10 inches of precipitation a year. Therefore, in vanilla DF, climate.txt will not include this tag, because this tag describes a climate that specifically does not exist in the real world.
(Actually, it sort of exists- not with this much precipitation, but some subantarctic islands like Bouvet Island are indeed oceanic polar with year-round precipitation. They're quite rare, though, and I don't think they get this much precipitation.)
Feel free to mod it in; but if you don't like snow, you'd be a fool to build a fort here; it'd get loads of it. If you wished, even, you could redo the climate raws from scratch to create a world that's all temperate rainforests, or none of which is warm enough for trees to grow, or which never dips below freezing, or (shock) all of which is a pretty good climate for doing just about anything. Potentially this could even have an effect on crop growth and civilization growth à la Guns, Germs and Steel. A tundra and glacier world will support only those civs which grow nothing above ground (that's goblins and dwarves). A taiga world would support no elves. It might support a few humans- you can grow, for example, potatoes and carrots and the like in a taiga climate- but they will be geopolitically inconsequential. Could even force some species to be solely hunter-gatherers. Dwarven ingenuity in a world of just desert will allow them to survive and develop civilization on their underground crops, but humans will never develop agriculture- they'll just be hunter-gatherers- and elves will never even evolve. Goblins, which eat raw meat, will be marginal in a desert world, but they'll do perfectly fine in a grassland climate; they'll just probably be nomads. Thinking about it now, that's an interesting possibility. Humans have four lifestyle options: either they can be hunter-gatherers who move around (the Inuit), hunter-gatherers who stay put (the Northwest Coast Indians before whites moved in), people who move around but also raise their own food sources (the Mongols), and people who are agricultural and sedentary (every civilization, ever.) Depending on climate, certain groups would be slated to be this or that. Dwarves will, thanks to the underground, always be agricultural and sedentary, but it's not altogether impossible to imagine that in a rainforest world, elves will never develop agriculture because they'll never have to, despite having settlements- they'll just pick fruit from the abundant trees. Or a world where humans develop agriculture along rivers where there's water for irrigation, but otherwise herd animals around. That'll affect who comes to trade and what they have to trade with you.