Test Earth climate maps (Patreon)
Content
So the koppenpasta script rework has gone well, and I've been testing out with some Earth climate data. I've also added a feature to attempt to automactically construct a map key of all climate zones present in the map (though this does depend on the default PIL font which isn't always super legible). In the order I think these should be appearing, you've got standard Koppen-Geiger; Holdridge, now properly indexed by biotemperature, precipitaiton, and potential evapotranspiration ratio; A somewhat simplified Koppen-Trewartha; Thornthwaite-Feddema, which is based on precipitation and potential evapotranspiration; and the Prentice et al. 1992 biome model, which uses minimum nighttime winter temperatures, total growing degree-days, and ratio of actual evapotranspiration to potential evapotranspiration.For my own scheme, I'm working largely from that last approach, but trying to expand it and tweak some of the approach to better account for other potential types of year lengths and seasonal cycles. Replicating the major climate regions is easy enough, but the tricky bits are 1, trying to get a bit more detail in semiarid grasslands, savannas, and dry forests, which aren't well represented in the Koppen scheme, and 2, trying to find parameters that most directly represent the effect of climate on vegetation in a way that'll work across different year lengths and types of seasonal cycle, which is not really considered in a lot of classification schemes; e.g., there's a complex back-and-forth pattern of trees switching from evergreen subtropics to deciduous temperate forests and then back to evergreen boreal conifers, and this seems to relate to a complex balance of relative gains and costs (evergreens gain more energy in early spring and late fall and don't have to use resources to regrow leaves, but must maintain leaves through unproductive winter months), and it's not totally clear which is more important in these transitions, and what factors best represent that balance. I may end up having to make some broad guesses, but in the process I have found some neat tricks, such as a good approach for identifying temperate rainforests.