Public Climate Data Explorations: Habitable Venus

This is part of my series looking into publicly available climate data I can use to produce climate maps. Unlike my regular climate explorations, none of these are models I've run myself, and the available data I can access is often rather more limited, but they still make for an interesting look. Today, I've found several studies looking at Venus, and investigating scenarios for a more habitable surface climate.

The geological history of Venus is rather more obscure than that of Mars. What limited data we have seems to indicate that most of the modern surface was formed by a massive volcanic event around half a billion years ago, erasing evidence of Venus's previous surface conditions. It has clearly gone through a runaway greenhouse event, where all of its surface water evaporated and was lost to space, allowing atmospheric CO₂ levels to rise with no weathering process to curtail them, but it's not clear when this happened. Several of the studies we'll look at today are intended to show that a more temperature climate may have been possible in the past, thanks to a dimmer sun and Venus's slow rotation, but these climates are only stable if Venus can reach a cool state in the first place. Other research has suggested that Venus must have started with a hot, steam-dominated atmosphere after it formed, and, despite the dimmer sun, may not have been able to cool down from this state, and so may simply never have had a temperate, habitable period.

So unlike our previous look at early Mars, these models are less reliable as specific scenarios for Venus's past and more just broad thought experiments in what sorts of climates might be possible for early Venus or a similar exoplanet. Regardless, even if these results aren't borne out as a model for Venus's past, they can still give us a more general idea of what climates we might expect for habitable worlds with low obliquity, slow but asynchronous rotation, and high insolation.

Let's familiarize ourselves first: Venus's topography was mostly mapped by the Magellan probe but there are still gaps over about 2% of the surface, which are often just filled in with some type of interpolation for world maps or modelling. The terrain is generally fairly gentle, though with some steeper plateaus and trenches probably formed by local stretching and squeezing of the crust in its current "squishy lid" mode of tectonics. Elevation ranges about 14 km between the highest and lowest point but most terrain is within a couple km of the global "datum", corresponding to the average radius (though different sources can be a tad inconsistent on where they set datum, I've tried to give measures consistent with the below map). 

There are some clear highland regions, but no true continents like Earth, so if we imagine the surface inundated by oceans to any particular level, we would expect there to be some large landmasses but also numerous smaller islands scattered over the planet. Most of the models we'll look at today use this modern topography flooded with oceans to some level, even though it all would have formed well after any potential habitable period, because we have nothing better to work with and we can perhaps expect early Venus might have broadly resembled its modern topography in its broad qualities.

Venus flooded to 800 meters above datum, as used in many of the models we'll look at today, with some landmasses labeled based on the names of roughly corresponding highlands on modern Venus.

At its orbit of 0.72 AU, Venus has an orbital period of about 225 days, with an eccentricity of just 0.0068 and axial tilt of 2.6°. Most unusual is that it has retrograde rotation with a sidereal period of 243 days, longer than the orbit, which works out to give it solar days a bit under 117 days. It's not clear when Venus adopted this rotation—it seems unlikely that it initially formed with it—which complicates the question of when Venus might have had a habitable climate, given that a slower rotation tends to allow for the formation of highly reflective cloud formations that can forestall runaway warming at the inner edge of the habitable zone. Many of these models use the modern rotation rate, but a few will test different rotation rates.

But for our purposes, this also creates a bit of a quandary in terms of how to classify climate zones in any model of a habitable Venus. Given the negligible eccentricity and axial tilt, I think we can mostly ignore the orbit and presume any seasonal cycle would be tied to the planet's slow rotation: the long days will be warm and could potentially allow for substantial vegetation growth, and the long nights will be cold and dark, without much variance depending on where they fall in the orbital year, so any hypothetical vegetation or other life that might appear on a habitable Venus would more likely tie any seasonal growth cycles to the day and night cycle rather than the orbit.

However, such a short seasonal cycle falls into a bit of an awkward range for the new bioclimate system I've started using for these explorations: Much of the internal logic of the system is based on the accumulation of growth or growth interruption periods, based in large part on the idea that different balances of such periods influence the evolutionary pressure to maintain growing structures through cold or dark periods or enter a deeper dormancy by e.g. shedding leaves or completing their life cycle and producing seeds. On Earth, for example, plants use these dormancy measures to survive winters, but take no such measures for nights even though no photosynthesis is possible then, essentially just waiting them out. We can infer that this would likely still be true even if days were many times longer, in part because subtropical vegetation similarly take no special measures to survive their short and mild winters. By extension, we can probably expect that on a planet with very short years, perhaps less than a month or two, the brief winters would likely affect life more similarly to Earth's days than Earth's years, prompting perhaps brief dormancy but probably not the conclusion of whole life cycles or loss of major growth structures. In the Pasta bioclimate system, this is handled by only interrupting the accumulation of growth periods when there is a sufficiently large period of growth interruption, similar to that which causes the transition from subtropical to temperate deciduous biomes on Earth; any shorter and it is assumed that any flora would just wait out the period and continue growth afterwards. If there are no substantial growth interruptions through the year, then the growth period is counted as effectively infinite, which allows for a planet with short years to have forest zones even if the growth period within a single year is below what would usually be the threshold for tundra climates on a planet with more substantial seasons.

But Venus's 117-day cycles fall in an awkward middle ground where it's not clear by extrapolation from Earth whether we should expect the cycles to have a more day-like or year-like effect on life. The threshold for a "substantial growth interruption" in the Pasta system corresponds to about 83 days of zero growth (though of course it can also include longer periods of low growth), so it is possible for some regions to reach that threshold, but only where the growing season is quite short, so it's hard to say that the same balance of evolutionary pressures would really apply; and this also has the perhaps curious implication that much of the planet would have day-like responses to the cycle but marginal areas would have more year-like responses.

The upshot of this is just that I will be applying the Pasta climate algorithm largely as I first designed it, but be aware that this is one of the trickier cases where I'm not too confident on how I've extrapolated the system from Earth's example.

Complicating matters further are the limitations of the data itself: All the data we'll be looking at today comes from ROCKE-3D modelling, and as typical for that model, the available outputs contain only averaged annual data. In past explorations I've worked out how to reconstruct enough information from that to apply most of my bioclimate classification system (excluding just Mediterranean zones, so you won't see any of those here), but that was working with cases fairly similar to Earth, with long, obliquity-driven seasons, and also with models cold enough that we didn't have to worry about the potential of H or E type climates with summers hot enough to potentially inhibit vegetation growth. Here, I'll need to take a slightly different approach:

So far as I can tell, these models define their year around the Venerean orbital period rather than the solar day, but that corresponds to just under 2 Venerean days and some of these are averages over multiple years, so any sampling biases probably aren't too bad. However, the growing season length parameter I've used previously to estimate GDD is unlikely to be reliable, and at any rate wouldn't account for potential growth interruptions from hot periods. A workaround which I've alluded to previously is to instead infer the whole seasonal cycle based on the reported average near-surface air temperature, average diurnal highs and lows, and diurnal high of the coldest day. So far as I can tell these "diurnal" ranges are sampled from 24-hour periods rather than the actual Venerean days, but we can still use them to infer the likely temperature of the coldest part of the day, and then if we assume the whole temperature cycle roughly follows a sinusoidal curve symmetric around the average temperature, we can use that to infer the maximum temperature and the pattern of temperatures throughout the day-night cycle. I can then count up growing-degree-days for the warm half of the cycle, presuming that this roughly corresponds to the day (probably not exactly true so this may produce a slight overestimate); no growth will accumulate at night regardless of temperature. From that I can also count up growth interruption periods (though the heating degree-days parameter should still work to indicate growth interruption from cold periods, so I'll check against that as well), which gives me enough information to classify all the thermal climate categories.

In principle we could similarly reconstruct the patterns of light across each day, but I don't think that's necessary; based on some rough calculations, with a short seasonal cycle and higher insolation than Earth in most of these models, it seems unlikely for any part of the planet to accumulate a substantial growth interruption period from low light levels alone, outside of perhaps a thin strip near the poles too small to appear at the model resolution. The very thick daytime cloud cover expected in some of these models may complicate matters, but I'm not sure it would change the outcome much and there's no good way to reconstruct the exact pattern of light would result form it anyway.

It's a lot of somewhat ugly hacks and nested assumptions, but that's about the best I can do with the data available to me, so let's finally take a look.

"Was Venus the first habitable world of our solar system?"

Way et al. 2016

Paper

Dataset

The first in this series of papers and one of the earliest uses of the ROCKE-3D model. As the name suggests, it's intended to demonstrate the potential of a habitable Venus in the past, specifically modelling habitability scenarios 2.9 and 0.715 billion years ago—why those specific dates, I'm not sure, perhaps because good models of the solar spectrum at those points was already available from early modelling of precambrian Earth. Their models use Venus's modern topography filled with oceans up to 800 meters above datum, covering about 60% of the surface (though with some smaller islands filtered out and a strip of land added to the south pole due to limitations of the ocean model), and a broadly Earth-like atmosphere of 1 bar N₂, 400 ppm CO₂, and 1 ppm CH₄.; neither of these are based on any particular indication that these would be likely for early Venus, just an assumption of Earth-like conditions by default.

The first model attempts to simulate Venus 2.9 billion years ago, with a somewhat dimmer and cooler sun but insolation still 146% that of modern Earth.

This lies well outside the typical limits of the conventional habitable zone, but with Venus's slow rotation it's not only temperate, but even a bit chilly: the global average temperature is about 11 °C, but this is a split between the pleasant oceans, much of which are warm even through the night, and the vast inland tundras that likely rarely rise above 10 °C. Some of this may be due to the topography: though Venus lacks many sharp mountain ranges, it also lacks many flat plains, and many of the big landmasses here rise fairly quickly from their coasts to above 1 km; in part this may be caused by the different tectonics, but in a habitable climate like this we would probably expect erosion and deposition processes to start forming broader coastal plains. Some of the climate contrast may also be due to albedo differences: the land is a bit more reflective than the oceans and also seems to be more prone to gather snow overnight which will delay its warming in morning. But a major factor may just be the short seasonal cycle: compared to somewhere on Earth with a similar average temperature, these areas just have far less time to warm up to growing temperatures, and it also doesn't help that a major cloud formation covers much of the day side and so further inhibits warming through much of the day.

Of course, this doesn't guarantee that a habitable Venus at this point would have been this cold—it could have just had more greenhouse gasses—but that such a cold climate is possible at this high insolation is an interesting enough result in itself. It's also notable that, despite over 50 days of dark and then the same period of light, temperature variation on land or sea seems generally moderate (of course much of this is inferred, but even sticking with the directly reported data, the difference between the average "diurnal" high and lowest diurnal high averages around 5-8 °C over most of the landmasses and peaks at just 11 °C in the drier parts of Aphrodite, implying not more than around 20 °C total temperature variation over most of the planet). Easy global air circulation due to the weak Coriolis effect and thick cloud cover (both reflecting light away by day and providing some greenhouse warming at night) seem to help keep a pretty even global climate at all times of day.

The next model moves forward to 715 million years ago (interestingly enough, about the same time Earth was plunging into a snowball period), when insolation on Venus had increased to 170% that of modern Earth.

For all that extra light, average temperature only rises about 4 °C, perhaps because as the surface temperature and evaporation rises, the cloud cover increases; but this only goes so far, cloud cover can only increase so much and more evaporation also drives a stronger greenhouse effect. At any rate, this modest warming seems to have a dramatic impact on the prospects for vegetation, with subtropical and mild boreal zones stretching across much of the landmasses. Another warning that this sharp change may be an artifical product of how I've constructed the climate system: Having a slightly warmer day reduces the length of the low-growth period in many areas below what I classify a "substantial growth interruption", so now growth periods are considered to carry across cycles rather than being limited to a single Venerean day, so this sudden change is very sensitive to what exactly I set that threshold of significant interruption to be.

Still, the transition between boreal forests and open tundra can be fairly sharp in some areas on Earth, demonstrating that a small change in growing season length can cause a substantial difference in vegetation patterns (note as well that the boreal zones in the highlands here are areas where the substantial growth interruption threshold is reached but the daytime growing period is still long enough to reach above the threshold for tundra, and these are also much more widespread than in the last scenario), so in principle a small shift in global climate could cause a dramatic shift in global biomes like this on a planet with fairly uniform temperatures, I'm just not so certain that it would occur exactly where I've predicted it (and of course none of this necessarily occured at these specific points of time in Venus's path, this is more just something that might happen to a Venus-like planet in general). The thick cloud cover at midday would also complicate matters in a way I just can't properly account for here, though from past experience thick cloud cover may be enough to reduce growth but probably not stop it, so with that accounted for we might expect to see more boreal rather than subtropical zones here.

One prediction that I do think might be more robust is the lack of any temperate zones: with such short seasonal cycles, a deciduous habit is unlikely to be worth it, as the extra productivity of fresh leaves by day and reduced cost of maintaining leaves by night cannot make up for the cost of having to regrow leaves every day. Thus something analagous to a temperate deciduous biome is less likely to appear here and we might expect dominance by something like evergreens—though it's not inconceivable that a niche might appear for some sort of briefer, less expensive dormancy that just hasn't been selected for on Earth with its long seasons.

Regardless, this case does also give us a better view of precipitation patterns, which may look familiar from our previous look at climates with long days: a wide band of wet climates across the low latitudes, with some particularly wet spots near the equator, and arid high latitudes. In reality the "band" is more of a wet spot over the planet's day side, which moves across the equator as the planet rotates, so we can expect a fairly regular weather pattern of wet, cloudy noons and drier mornings and evenings (moreso the mornings because the cloud formation will probably lag behind the movement of the substellar point). This also helps explain why we don't see much rainshadowing from Venus's highlands: as the warm wet spot moves, the winds converging on it follow, so any given spot will have winds crossing it at many different directions and so there will likely be some part of the day where there's a clear path for moisture to travel from the ocean.

The last two models in this study are essentially controls to compare against, both using the same parameters for the 2.9 billion years ago scenario with 146% of Earth's moden insolation. First, a model with Earth's current topography (or the simplified version often used for ROCKE-3D models, anyway) but still with Venus's low tilt and slow rotation.

This turns out warmer than either of the Venus models, which the paper attributes to the larger area of equatorial oceans causing more evaporation and so a stronger greenhouse effect from more water vapor, with notably heavier rains as well.

Finally, there's a model with Venus's topography where the rotation period is reduced to only 16 times that of Earth's, but still retrograde such that it works out to solar days about 15 times longer than Earth's.

Without the large substellar cloud formation of a slow-rotating planet, the average temperature shoots up to 56 °C, and this planet is likely well on its way to a moist or runaway greenhouse. Hence one of the main conclusions of the paper, that the early rotation rate of Venus or a similar exoplanet may be a major factor in whether it could have ever retained a habitable climate.

For simplicity I used essentially the same algorithm to assess this world but with shorter days, but unsurprisingly everywhere outside some highlands is just too warm to see much growth with Earthlike vegetation. In case you want to imagine some far more heat-tolerant alien life on such a world or just want to see the precipitation patterns better, I also made a map where I simply ignored any high-temperature limits on growth, essentially presuming full growth throughout the day:

Though precipitation is about double what it was in the second model, it's just not enough to keep up with greater evapotranspiration, outside of some cooler highlands (which may also benefit from greater orographic rains). There's notably a large area of monsoon savanna climates, indicating some periods where rainfall greatly exceeds evaporation and then others where available surface moisture falls well short of potential evaporation. We can perhaps imagine a cycle of torrential noontime rains and then the surface rapidly drying out in evening and remaining dry through night and morning (though I can't be sure from this data when exactly the rain falls in the day; perhaps instead it comes in evening as the surface cools, with dry conditions through most of the day).

Now, if we ignore variation over the day-night cycle and treat all these cases as essentially seasonless, we can produce Koppen-Geiger climate maps from them, so here's that:

I don't think these are terribly informative for this sort of global climate, so I won't bother with it for any of the other cases we'll look at.

"Venusian Habitable Climate Scenarios: Modeling Venus Through Time and Applications to Slowly Rotating Venus-Like Exoplanets"

Way and Del Genio 2020

Paper

Dataset

A followup paper from the two lead authors of the first paper, exploring a wider range of climate scenarios for Venus or Venus-like exoplanets. With generally similar parameters and setup to the first paper, they pick 9 scenarios with various combinations of insolation (and solar light spectrum),  atmosphere, and day length, and then run each with 5 different topography setups:

• Venus with no oceans and just a small amount of initial soil water, equivalent to a global layer of about 20 cm; let's call it "Dry Venus".
• Venus with a somewhat larger stock of surface water equivalent to a global layer 10 meters deep; this is not placed in permanent oceans but allowed to move through the water cycle and collect into small lakes. As I've done with some ROCKE-3D models in the past, for these cases (and Dry Venus) I'll mark cells in as "seas" where they have above 94% lake cover (as the model seems to cap it at 95%), and we can call this case "Lake Venus".
• Venus with global oceans over 60% of the surface, the same as the first paper; "Ocean Venus".
• An "aquaplant" with globally uniform 158-meter deep oceans and no land, which I'll skip over here.
• The simplified Earth topography, with oceans uniformly 310 meters deep for better comparison to the similarly shallow oceans of the Ocean Venus cases.

That gives us a lot more to look at, so I'll move through these a bit quicker:

The first several sets of simulations look at potential scenarios for Venus 4.2 billion years ago, with insolation 140% of modern Earth. The first set uses a 10-bar atmosphere of pure CO₂, and as you might imagine that doesn't work out too well; Dry Venus manages to equilibriate at an average of 262 °C for lack of enough water to drive a runaway, and the other models are all just terminated after they pass a 100 °C average, as this is beyond the reliable temperature range for ROCKE-3D anyway. They do provide averages for the last 5-10 years of each case, but there's nothing much to see; interpreted normally these are all just hot barren land and torrid oceans, and even ignoring high temperature thresholds, they all lack precipitation and so are just deserts.

The next 3 sets all use the same insolation for 4.2 billion years ago but a more modest 1-bar atmosphere of 97% CO₂ and 3% N₂, but have varying rotation rate. First, a set with a rotation period of 16 Earth days (solar day of 15 Earth days):

These manage to equilibriate below boiling but are still quite hot, ranging from 60 °C average for Earth to 96 °C for Dry Venus, so the wetter cases at least are likely bound for a runaway, but for now there are at least hints of habitable refuges in the polar highlands. For interest I did again test what results we get ignoring heat limits on growth, and much as in the last study this seems to indicate a generally dry climate but with bursts of rain at some point in the day.

The next set increases the rotation period to 64 Earth days (giving a solar day of about 50 Earth days), which finally manages to bring the temperature down to something resembling a globally habitable climate but still with questionable stability in the long term.

Dry Venus is dry, but Lake Venus shows us something like what we saw with some of the early Mars models, where the lowlands remain parched but the cooler, rugged highlands collect more moisture. The wetter worlds, meanwhile, just about cling to the edge of habitability; I won't bother showing the alternative interpretation without heat growth limits here, as it essentially just turns all the hot barren into desert or semidesert.

The next set increases rotation back up to Venus's modern value, with 117-day cycles, which improves matters further, but these climates are still very much on the hot end of habitable thanks to their CO₂-rich atmospheres, with the coolest (Ocean Venus) averaging 37 °C—though the daytime cloud cover keeps it from reaching too high by day, so perhaps it could be stable and reasonably hospitable to life.

The remaining sets all use Venus's long modern days but explore different scenarios for later points in Venus's history. The next moves forward to 2.9 billion years ago, with 147% of modern Earth's insolation (I suppose they slightly recalculated) and swaps out the CO₂-rich atmosphere for the same 1-bar N₂-dominated atmosphere with 400 ppm CO₂ and 1 ppm CH₄ used in the last study.

So Ocean Venus and Earth are essentially just a repeats of the 1st and 3rd cases in the last study, happening to balance out a tad cooler due perhaps to minor changes to the solar spectrum used or tweaks to the ROCKE-3D model. More interesting are the drier cases: Lake Venus has an impressive oasis in the Aphrodite highlands, as well as a small sea at the permanently dim north pole, and even Dry Venus is showing some hints of something other than a global desert. Most of the Dry Venus cases here have fairly substantial cloud formations in the upper atmosphere and an equatorial rain belt like we can see more clearly in many of the Lake Venus cases, it just rarely passes the threshold of being classed other than a desert, peaking here at a few mm/day over Aphrodite and a couple other high peaks and less than 1 mm/day over the rest of the planet. The sense we might get is of a world near to the minimum possible water content for an active water cycle, providing perhaps just enough moisture for some life to get by but not enough for any substantial surface life.

Something else you might have picked up on by now is that Ocean Venus is usually the coldest of each of these sets, having enough surface water in the warm latitudes to feed into thick daytime clouds but not so much as to cause excess evaporation and greenhouse warming as in the case with Earthlike topography. This perhaps implies there could be an even better optimization of ocean coverage to maximize the cooling of a slow-rotating planet like this at high insolation (though possibly this could vary depending on the exact rotation rate; the paper points out that previous studies have tended to show that much drier planets are cooler at high insolation with Earthlike rotation rates, where clouds are less efficient at cooling the planet).

The next set uses the same atmospheric proportions and other parameters, but reduces the surface pressure to 0.25 bar, based on the idea that studies suggesting a lower atmospheric pressure for Archean Earth might indicate a similarly thin atmosphere for ancient Venus.

As you might expect, the weaker greenhouse effect makes for quite cold climates, with average temperature as low as 3 °C for Ocean Venus.  But the long, bright days manage to prevent runaway ice formation and even keep some of the oceans fairly warm, giving us a global tundra climate of a sort likely impossible on a more Earth-like planet that would be prone to snowballing. Lake Venus maintains a similar climate partially by virtue of having too little moisture to snowball, so is a cold desert but still with some oases. Dry Venus is hotter but drops below 0.2 mm/day average rainfall over almost its whole surface, as most of its water becomes trapped in polar icecaps.

The last few sets keep a 1-bar, N₂-dominated atmosphere and move forward in time, increasing the insolation and shifting the solar spectrum for a slightly hotter star. The next models Venus as it might have been 715 million years ago, with 171% of Earth's modern insolation.

Ocean Venus is again a repeat of the second model in the last study, but a tad cooler, but of course it's nice to have the others to compare against: the Earthlike case is starting to show signs of extreme temperature variation between night and day, but still remains broadly habitable; Lake Venus has a more impressive strip of wet climates across its equator; and Dry Venus actually forms a lake large enough to appear as an ocean cell in the highlands of Ishtar (the data also indicates a few smaller lakes in Aphrodite, which appear in some other cases as well).

Next, they model Venus with its modern insolation, with 190% of Earth's insolation:

Dry Venus is approaching dangerous territory, but Ocean Venus remains moderate, and Lake Venus and Earthlike have sweltering but potentially stable average temperatures of around 30 °C.

The final set goes even farther, increasing insolation to 240% of modern Earth, which by my reckoning would put it about 2-3 billion years in the future, though they keep the sun's modern light spectrum. Dry Venus, Lake Venus, and the Earthlike case all crash or are terminated at high temperatures within a few years, but Ocean Venus still maintains an average temperature of 26 °C, with what appears to be a persistently wet and cloudy climate across most of the low and middle latitudes, with large parts of the continents averaging over 80% cloud cover.

The conclusion the study takes from these results is that if Venus did have a habitable period with something like its modern rotation and a stable atmospheric composition, increasing insolation alone wouldn't have been enough to push it into a runaway greenhouse; instead some other influence would have had to tip it over the edge, with the authors suggesting a large ill-timed volcanic event raising CO₂ levels.

"Early Habitability and Crustal Decarbonation of a Stagnant-Lid Venus"

Höning et al. 2021

Paper

Dataset

This final paper attempts to move beyond broad guesses at what the early atmosphere of Venus would have been like and instead attempts to construct a geochemical model for how carbon would have moved between the mantle, crust, and atmosphere of early Venus as it evolved, assuming a stagnant lid mode of tectonics. Most of the paper discusses the model and its results: the short version is that while a carbon-silicate cycle could help keep a stagnant-lid Venus habitable for a while, without subduction helping to recycle carbon back into the mantle, it would build up in the crust, which would cause more CO₂ outgassing, which would eventually cause a runaway greenhouse when Venus was a bit under a billion years old.

There are all sorts of caveats about the assumptions in that model and how well it represents a stagnant or squishy lid planet and weathering feedbacks on the surface, but at any rate to check their results they take the insolation and CO₂ values at 4 points in their model (ranging from 139 to 145% Earth's insolation and 0.112 to 0.206 bar CO₂) and run them in ROCKE-3D, with Venus's modern topography and slow rotation, 1 bar N₂, and surface water equivalent to a global layer 22 m deep, so a bit over twice Lake Venus in the last study. This is again allowed to dynamically collect in lakes, and I identified "sea" areas here based on lake cover.

They broadly resemble the hotter Lake Venus cases, again implying a mostly dry world but with large wet, potentially even lush pockets near the equator and north pole. But there's not much distinguishing them aside from a slight warming from a 27 °C average for the first to 33 °C for the last.

The data repository also contains one more model that isn't mentioned in the paper but appears quite similar other than being far colder, averaging around 9 °C, which gives it more of a cold global desert but still what looks like a fairly hospitable wet belt along the equator.

Based on some poking around in the repository, I think this is a control model run with just 100 ppm CO₂; perhaps a better fit if we imagined a similar world with a cooler interior and less outgassing.

To wrap things up a bit, as with our previous look at Mars I think the results here are fascinating not necessarily for what they might imply about Venus (going by the latest research at the moment, the prospect for a habitable period on early Venus seems a bit dubious compared to Mars), but what sort of climate they imply could be possible elsewhere, especially with this sort of slow asynchronous rotation that otherwise receives little attention: desert worlds with lush equators, global tundra climates that hover just above freezing, and habitable wet planets at shockingly high insolation.

But that'll do for today; there are a few other models with available datasets that use Venus's modern topography for modelling exoplanets, but we'll leave those for another day. Probably the next major public data exploration I'll do will look at tidal-locked planets, but there's also a couple individual studies I may want to take a quick look at first. See you then.

Buy me a cup of tea (on Patreon—and join the Discord!)