Nikolai Lofving Hersfeldt
patreon
Another tilted Earth climate map
🕑 Added 2024-06-19 16:48:07 +0000 UTCRemember that xkcd what if way back with Earth rotated 90 degrees? https://what-if.xkcd.com/10/ Well I gave it a go in exoplasim, and the results aren't a bad match overall, though I haven't compared the prevailing winds yet and this may be a good opportunity to try out the storm climatology module a bit.
In other news, been hitting a bit of a snag both with that ice model and my plans for an alternate climate classification system. The ice model is straightforward enough to run and goes at a good clip at very coarse resolutions but slows way down once you try to refine the resolution, which may make it difficult to get down to the kind of resolutions necessary to get good data input to the terrain model. It may end being I need to just use the ice model to inform some manually drawn glacial valleys rather than being able to apply the data directly.
With the alternate climate system, it seems like some pretty easy improvements over Koppen can be made just switching to more reliable measures (actual minimum temperature rather than coldest month average, growing degree-days rather than various indirect measures of the growing season), but figuring out a reliable measure for precipitation patterns like Mediterranean or monsoon in a way that doesn't rely on too many assumptions about a particular region's seasonality or sampling period of the inputs is trickier. Still, that's a less urgent project so I can just poke at it every now and then.
Regardless, next post will probably be another public data exploration of some models of an ancient habitable Mars. It's fairly patchy data so don't get too excited but it's interesting nonetheless.
Comments
Nikolai Lofving Hersfeldt
I was more worried about things like defining areas by wettest or driest month, which then is a bit sensitive to how long your sampling months are. Something like halves of the years might be a bit more reliable if we can assume you'll probably always have an even number of samples, but can get unreliable when we have stuff like dry summer and winter and wet midseasons, whech we could then try to determine by dividing the year in 4 but that ultimately just gives us some of the same vulnarabilities, where exact distribution of rains between, say, february and march might place a location in different zones without actually implying much about actual conditions in reality. Part of my main goal here is to try to work back to the most important factors rather than indirect measures; like rather than using Koppen's approach of trying to measure growing seasons by hottest month or number of months above 10 C, growing degree-days is a more flexible measure of growing season regardless of the exact pattern of seasons that seems to more reliably relate to actual vegetation patterns. Seasonal precipitation patterns matter mostly in the ways they restrict water availability to plants during growth periods, but I haven't found a direct measure of that so far; I've been experimenting with things like precipitation or evaporation averages weighted by growing rates, but they haven't proved too reliable so far.
Vera Lycaon
Given the nature of different precipitation patterns I feel like sample periods are more or less unavoidable here; even if they just end up being things like "warmest/coolest ½ of the year" (perhaps more strictly defined as those times where daily average temperatures are above or below the median to account for equatorial areas on high obliquity worlds) - I don't see a particularly intuitive way to define these patterns without making at least a few assumptions about how they'd pan out in the data, even with the addition of a set of extra categories for dry periods focused on both summer and winter (with wet equinoxes) and on spring/fall (with wet solstices). Slow rotators (or at least, climates with especially big day/night differences) may need their own category too, come to think of it; an ostensibly temperate climate that swings between freezing nights and days of 20°C+ temperatures, both several Earth days long each is obviously going to impact life very differently from say, what you'll find here on Earth in places like western Europe or southern South America.
Stephen Rider
Storm climatology? This sounds interesting.
