Creating a map of Koppen climate zones for a very Earth-like world is quite easy, with the right instructions. However, what if the climate is several degrees cooler? Do I simply shift all the climates equator-ward, or is it more complex than that? If not, how much should I squish them?

  • 1
    $\begingroup$ It gets drier, too. Someone posted a picture of climate zones during an ice age quite recently... I'll see if I can find it, or if they left a source. Or maybe they'll be along themselves soon enough... $\endgroup$ Sep 20, 2019 at 17:13
  • 4
    $\begingroup$ Wikipedia to the rescue: Last glacial maximum vegetation map (and related article). Not quite what you're after, but an excellent start, I'd say. $\endgroup$ Sep 20, 2019 at 17:37
  • 2
    $\begingroup$ @StarfishPrime Mm, that would make sense, as the continents would expand and the water cycle would be slowed down due to less evaporation. Pity the map doesn't use the Koppen system, but thanks for sharing. $\endgroup$
    – SealBoi
    Sep 20, 2019 at 17:43
  • $\begingroup$ @SealBoi Yes. Sea level on Earth went down some 100's of feet. Also, under all that ice the elevation of ground was significantly different because 2 km of ice will depress even rock after a few 1000 years. There are lovely beaches in northern Canada that are still rebounding. en.wikipedia.org/wiki/Post-glacial_rebound $\endgroup$
    – puppetsock
    Sep 20, 2019 at 18:15
  • $\begingroup$ Features like the subtropical ridges wouldn't move as they're a product of the atmospheric cells, however due to increased aridity, the desert bands where these ridges lie would be wider than they are on present day Earth. Another consequence of cooler temperatures is decreased evaporation however, so those regions aren't necessarily desert, but could rather be dry steppe with quite a lot of biomass. $\endgroup$ Sep 20, 2019 at 21:53

1 Answer 1


Ice Age Climate

Climates and Geography

While there are situations where shifting equator-ward (and colder climates shift downward with altitude too remember) may not be the precisely right answer, that is generally the case so long as you remember a few points

  • colder weather may change currents that would bring different weather fronts which would change temperature in specific regions (Europe is warmed by mid-Atlantic current for its latitude. If Ice Age shifted that current Europe would be much colder even if the average temperatures elsewhere were less shifted). Currents are also shaped by the continents around which they move water.

  • shifting rain bands make different areas deserts that get much hotter especially with the right types of soils and bedrocks (see shifting rain bands and the Sahara moving north and south in Africa)

  • the koppen system fixates on latitude/elevation, precipitation levels, season where majority of precipitation is received, summer temperatures, etc. Other systems for classifying climates exist, focus on other factors in varying emphasis and leave out others. Different models have different strengths and weaknesses and the science as a whole is not fully grasped as the factors that could effect this system seem to still perplex the people predicting the weather.

  • science here will change as we improve our knowledge. Fixating on accuracy in arbitrary circumstances such as providing accurate climate maps with this world will make the work more quickly dated. Quality content within the setting while leaving the setting more ambiguous such as mentioning of the Ice Age without Koppen acronym might be a better approach for the narrative happening within it.

There are a lot more examples and things to remember to consider, but the key is not the specific accuracy of any specific climate maps unless you are specifically making maps of hypothetical worlds guaranteeing climate accuracy in their Koppen zoning.

The goal should probably be to produce quality content in the setting While your setting may not be totally accurate, it would behoove you to make sure it is not so inaccurate as to be glaringly obvious while also not specific enough to be easily dated in 5 years. Depending on what the prose is for of course...


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .