Climate Explorations: Pressure
 |
| "True color" approximation for Earth with 0.25 bar sea level pressure. |
Hello, this is the sixth part of a series where I explore different possible climates with the ExoPlaSim model. As per usual, you can check the first post in the series for details on my approach and some of the limitations of the model, and if you’re unfamiliar with the new climate classification system I’ve started using to map these climates, you may want to look at that.
Most of the previous explorations have been playing with the planet’s orbital and rotational motion, but for the next few posts we’ll keep to our baseline days, years, orbit, and tilt (though I’ll continue adjusting stellar flux to control temperature, which you can think of as either changing the luminosity of the star or changing the planet’s orbital distance, even though I’m not altering the year length) and look more at the planet’s physical properties. Today, we’ll alter the total sea-level atmospheric pressure.
Earth’s average sea-level atmospheric pressure is 101.3 kilopascals (kPa); for mathematical convenience, scientists will often measure pressure in the slightly lower unit bar, equal to 100 kPa. Within the solar system—setting aside the gas giants and very thin atmospheres of some small icy bodies—Mars has around 0.006 bar at the surface, the moon Titan has 1.47 bar, and Venus has 92 bar, but these don’t point to any particular overall trend, and they all have very different climates and geological histories from Earth such that none necessarily give us a good sense of what range of atmospheric pressures an Earthlike exoplanet should have. There seems to be a common belief that atmospheric pressure should be directly linked to planetary mass or gravity; that a planet larger than Earth must have a thicker atmosphere, or equivalently that a planet can’t have a thicker atmosphere without being larger. There is some logic to this: a larger planet with higher escape velocity may accrete more gasses as it forms, lose less to atmospheric escape, and perhaps even outgas more from its greater interior volume. But a lot of other factors may contribute to how much gas or volatiles a planet gain as it forms and how its atmosphere develops, so in reality I’d expect that while there may be some overall correlation between planet mass and atmospheric pressure (and at the extremes, very small planets will simply fail to retain any significant atmosphere while very large planets will tend to form gas giants), there could still be a lot of individual variation, and so it may not be implausible or necessarily even that unusual for a planet smaller than Earth to have a thicker atmosphere, or vice-versa for a larger planet to have a thinner atmosphere. For this exploration, I’ve just retained the same mass and size as Earth in all cases.
ExoPlaSim can handle a range of pressures from about 0.1 to 10 bar—and even at those extremes isn’t necessarily totally reliable—so even though that’s not necessarily the full range of potentially habitable pressures, that’s what we’ll look at today. For the most part we’ll leave the oxygen partial pressure at 0.2 bar, argon at 0.01 bar, and CO2 at 0.3 mbar and adjust only the nitrogen pressure, though for the 0.1 bar case we’ll run out of nitrogen (and argon) to remove and have to reduce oxygen as well.
Based on simple atmospheric physics, there’s a couple effects we should expect to see here:
First, higher-pressure, higher-density air can hold more heat. This means it should take longer to warm up and cool down, reducing temperature variation across days and years, and air should carry more heat as it circulates across the planet, transporting more heat from the equator to the poles and so reducing their relative temperature difference. Thus, a higher-pressure atmosphere should reduce temperature variation in all respects. For reference, here's a quick look at temperature variation and extremes in the baseline model with about 1 bar pressure (note that because the max and min temperatures are sampled from the surface temperature but the average is taken from the 2-meter air temperature, the max can sometimes be a bit lower than the average over ice where the surface is frozen but the air is slightly warmer):
Second, even though we’re keeping a constant partial pressure of CO2, higher pressure should still tend to increase the greenhouse effect through pressure broadening: higher pressure increases the infrared light that CO2 and other greenhouse gasses absorb, increasing the greenhouse effect. We’ll be compensating for this with reduced light from the sun to keep a global average temperature of around 15 °C, but the shifted balance of direct solar heating to greenhouse heating may still have some subtle effects, even further tending to reduce temperature variation through time and across the surface, as greenhouse heating is more-or-less globally even rather than concentrated like sunlight.
Air pressure also influences evaporation of water in several potentially conflicting ways, but experimental studies tend to find that, all else being equal, greater atmospheric pressure tends to reduce evaporation rates on balance, probably because it makes it harder for individual water molecules to escape the water surface and then diffuse away.
And of course these trends should tend to be reversed at lower pressure.
Previous Exploration: Combining Obliquity and Eccentricity
Higher Pressure
2 bar
Prior Model: Baseline
Stellar Flux Relative to Earth: 0.943
Average Temperature: 15.3 °C
This is about as high of an atmospheric pressure as humans can handle before starting to suffer the effects of nitrogen narcosis, but in terms of climate the effect is fairly subtle.
Compared to the baseline, the most obvious effect is more moderate seasons: at low latitudes, the extreme summers erroneously predicted by ExoPlaSim in the Americas are gone, and at higher latitudes the winters are less extreme but the summers are also more modest, so overall there’s a greater spread of subarctic climates. The overall spread of tropical and temperate climates still hasn’t shifted much, but worth noting that, as expected, the stellar flux has to be reduced here to offset the greater greenhouse heating from pressure broadening.
More modest seasons means a bit less swing in the circulation cells with the seasons, and winds are also slower, averaging about 19 km/h to the 22 km/h in the baseline. But that doesn’t mean they’re weaker; going by the drag equation, the force exerted by the wind on any static should scale with the density of the air (which will scale directly with pressure for constant composition, which we are close enough to with a nitrogen-dominated atmosphere) and the square of the windspeed; given double air density, this means that the average wind force is actually 54% higher than on Earth.
The steadier rains and reduced summer heat has made some areas a bit wetter, but overall this is a drier world, with only around 81% the average precipitation of the baseline.
4 bar
Prior Model: 2 bar
Stellar Flux Relative to Earth: 0.912
Average Temperature: 14.5 °C
With another doubling of pressure, this planet is likely beyond the safe limits for long-term human habitation, except perhaps in the highest mountain plateaus. Nevertheless, the distribution of climates that still looks pretty similar to real Earth in the broad strokes, but with a variety of odd details.
First off, temperature variation has continued to decline, including seasonally, latitudinally, and daily; Peak daytime highs even in the subtropical deserts rarely exceed 35 °C, and more generally few areas have more than 10 °C of daily temperature variation. Modest winters could be of some benefit to life at high latitudes, but modest summers as well could hamper growth, with a wider spread of tundra climates.
This has in turn shifted circulation patterns again, with a notable concentration of rains in central Africa now that moisture from the Indian Ocean isn’t being diverted north into Asia by the summer monsoon. Still, global average precipitation has continued to decline, to only 61% of the baseline. This may partially be a direct effect of the higher pressure suppressing evaporation, but previous studies on pressure variation have suggested it may have more to do with shifts in the planet’s energy balance: the stellar flux has been reduced further to offset greater greenhouse heating, and on top of that the thicker atmosphere reflects and absorbs more light; the overall Bond albedo has increased from 0.3 in the baseline to 0.39, but more of that reflection is happening in the atmosphere such that on average the surface only receives 67% of the direct sunlight as in the baseline. The increased greenhouse warming compensates for this, but less direct light also reduces evaporation, even at the same temperature (though reduced summer temperatures probably also contribute); this may help some semiarid regions stay a bit moister between rains, but it also means there’s less moisture in the atmosphere to rain elsewhere. Ice also melts slower, hence the appearance of new glaciers in Greenland and Tibet.
This also means less light for photosynthesis, especially in areas that also have more consistent cloud cover, hence the appearance of quasitropical zones in the Pasta bioclimate map. Thus, despite the stable climate, vegetation is generally predicted to be a bit thinner here, even in the equatorial rainforests.
10 bar
Prior Model: 4 bar
Stellar Flux Relative to Earth: 0.911
Average Temperature: 15.6 °C
Here we see a more dramatically distinct climate pattern, though looking closely it is largely just continuing the trends we’ve already seen.
First off, temperature variation has of course continued to decline. The maximum reported temperature for the whole planet at any time of year is a scant 26 °C. Winters over the glaciers can be more extreme in the other direction, and winter nighttime frosts may still be common at high latitudes, but still averages over much of the planet remain in the comfortable 5-20 °C range, with barely any shift at night.
Winds are seemingly fairly sedate, with little global shift in circulation patterns over the year and an average speed of only 13 km/h; but accounting for the higher air density, the actual force of the wind is over 3 times greater on average.
Such a moderate climate comes with disadvantages, though; the stellar flux hasn’t been reduced much, probably because the greater Bond albedo (now 0.47) is helping to offset the increased greenhouse warming, but that pans out to the surface only receiving 48% the direct sunlight as in baseline. This of course has suppressed the water cycle in turn, such that average precipitation is only 42% of baseline. This is somewhat mitigated by lower evaporation, such that fewer areas remain completely dry through the year but there are broader stretches of semidesert and steppe. ExoPlaSim’s simple vegetation model actually predicts no forests, though still widespread lower vegetation cover; I’m not sure I’d be so pessimistic myself, but it’s fair to say this might be a generally less photosynthetically productive world overall, with more open ground and fewer dense rainforests.
Lower Pressure
0.5 bar
Prior Model: Baseline
Stellar Flux Relative to Earth: 1.069
Average Temperature: 14.1 °C
This is roughly equivalent to the pressure at about 5-6 km on Earth, but because we’re keeping a constant partial pressure of oxygen, you wouldn’t expect to have equivalent effects of altitude sickness; rapidly shifting from 1 to 0.5 bar could cause nitrogen bubbling and decompression sickness, but this can be avoided with a more gradual transition.
Unsurprisingly, the trends we’ve seen generally continue here in the reverse direction, with a wider swing in seasonal temperatures. Much of the mid-latitudes must contend with searing summer days, and in the subtropics even winter doesn’t bring much respite as the weaker global circulation traps more heat in the tropics, and greater daily temperature variation causes hotter daytime peaks. Correspondingly, higher latitudes have large ice caps and brutal winter nights, though at least a few areas may benefit from the hotter summers as well.
Wind patterns show a bit more swing with the seasons and are slightly quicker here, averaging 23 km/h, but in terms of force are substantially weaker, only 57% of the baseline.
And the trend in energy balance continues as well; in addition to the increased stellar flux, the atmosphere is less reflective and opaque, and so the surface receives 15% more direct light, which in turn drives 17% more global precipitation. Combined with the greater shifts in circulation, this means few areas are completely dry, with even the large deserts receiving bursts of summer rain (though of course this has to contend with greater summer evaporation, so moisture availability may still be fairly transient).
0.25 bar
Prior Model: 0.5 bar
Stellar Flux Relative to Earth: 1.185
Average Temperature: 14.1 °C
With nitrogen almost totally removed, this atmosphere is mostly oxygen, which may make it highly prone to large fires (which in reality might consume enough oxygen to prevent it becoming this prevalent in the first place), but that’s not represented in this model. It’s also worth noting that 10-20% greater flux is around where many models predict that an Earthlike planet should enter a moist greenhouse; ExoPlaSim isn’t designed to detect or model the details of such a transition, so I can’t be certain this is a stable climate in the long term (but I’m not sure it isn’t either, details on the exact effect of lower pressure on the habitable zone limits are rather scant).
It's certainly a much wetter world, with 45% greater precipitation on average (despite only 32% greater direct sunlight, so the linear relation we’ve seen between the two so far doesn’t seem to hold—perhaps the direct effect of lower pressure is getting more significant), such that growing tropical rainforests are squeezing out the last of the deserts.
All this extra moisture and cloud cover is actually having a moderating effect on temperatures, with many of the tropical and subtropical forests seeing less extreme seasons and days than in the last case. But drier areas aren’t so lucky, with deserts suffering summer days as hot as 70 °C, while Antarctica in winter dips to as low as -120 °C.
Oddly enough, these temperature contrasts aren’t driving particularly strong winds; the average wind speed is still about the same as baseline, which means 1/4 the wind force.
But regardless, outside of these few trouble spots, this world is blanketed in thick forests (though perhaps also wracked with monstrous summer firestorms).
0.1 bar
Prior Model: 0.25 bar
Stellar Flux Relative to Earth: 1.356
Average Temperature: 15.9 °C
This world’s atmosphere now totally lacks nitrogen and argon, and oxygen has been reduced to half its partial pressure on Earth. At low altitude this might be just about tolerable for human habitation if given plenty of time to acclimate—though I’m not certain, other gasses still play some role in our respiration that may be lacking here. This is probably about the oxygen level Earth had in the Cambrian and Ordovician, so it’s probably sufficient for complex life more generally, though the lack of any atmospheric nitrogen for soil bacteria to fix into nutrients may be a serious issue for photosynthetic life.
I want to emphasize off the bat that we’re operating at limits of ExoPlaSim’s design range, and again the model isn’t fully equipped to assess if this climate would really be stable in the long term or experience runaway warming or loss of water to space. There has been previous study of varying atmospheric pressure using ExoPlaSim to the same range which seems to say that the atmospheric water content likely doesn’t get high enough even at low pressure to be worried about a moist greenhouse, but they only tested stellar fluxes up to about 14% greater than Earth. If we are willing to accept these results, though, they have some pretty striking implications.
We’ll start with the obvious: global average precipitation is about double the baseline, but this isn’t just an even doubling but a substantially broader spread of precipitation into previously dry areas, with no part of the planet dropping below 9 mm/month of precipitation at any time of year. Notably, though, this world doesn’t have particularly greater humidity or cloud cover than baseline (the air does have a higher proportion of water vapor, but in absolute terms the total mass of atmospheric water vapor isn’t much higher), and there are still large areas where the soil dries out for part of the year. This perhaps implies an odd hydrological cycle where water rapidly evaporates but then almost as quickly reaches saturation and rains back down, and this perhaps helps retain water within regions rather than allowing it to be carried far away by the prevailing winds.
This is not without its downsides, however. At 0.1 bar, the boiling point of water is around 45 °C, which is easily surpassed across much of the low latitudes in summer days, implying that all open bodies of water will completely boil away (though much of these areas dip well below the boiling point at night, so the water may promptly rain back down—which perhaps accounts for some of that rapid water cycling, but not all of it as that pattern extends beyond these boiling regions). I emphasize again that this isn’t really a situation ExoPlaSim is designed to handle and perhaps not healthy for the long-term stability of this climate, but do note that none of the ocean surface ever approaches the boiling point. If we can take this at face value, that seems to imply a fairly challenging biome where life may need to take extreme measures to retain water and avoid thermal stress in summer (shown by the hot and extraseasonal zones in the Pasta bioclimate map, as I’ve partially tied their definitions to the boiling point).
The broader distribution of moisture again does have some moderating effect, and in particular rapid evaporation near the boiling point will consume a good deal of thermal energy and so help slow warming. But even beyond that, heat is a bit more evenly distributed by latitude compared to the last model, with a cooler equator and no ice caps at the poles. At a guess, this may come down to some combination of greater summer melt at the poles from more direct light, this model happening to balance out at 1.8 °C hotter than the previous one, and a thinner atmosphere and weaker greenhouse effect allowing for faster radiative cooling at night in the tropics.
Different circulation patterns may also help; the ITCZ swings far with the seasons, such that at the seasonal extremes one equator-crossing Hadley dominates and the other almost disappears. The polar cells also seem to be fairly insubstantial here in any season, so overall prevailing winds are dominated by fewer, wider circulation cells. Though these winds are weak (averaging 26 km/h but only 15% of the baseline wind force) and carry less heat, this more direct circulation may still help distribute heat a bit more from tropics to poles. Notably, this is also a bit more similar to the predominantly equator-crossing winds patterns of Mars, so there may be some fundamental relationship between pressure and circulation patterns that I don’t know the exact causes of.
ExoPlaSim’s simple vegetation model predicts global lush forests, but wasn’t really built with the potential for literally boiling summers in mind, so take that as you will, and of course this also doesn’t account for fire rates or issues with nitrogen supply either. In general this whole scenario must be taken with a generous helping of salt.
That’ll do for today, though I’ll leave off with a couple quick charts of the trends in wind I've been discussing:
And the different energy balance across models:
There’s probably some fascinating ways in which different atmospheric pressures could be combined with parameters we’ve previously explored like obliquity and day length, but for now the next planned exploration will look at planet size.
Very cool!
ReplyDelete