Climate Mini-Exploration: Terraformed Mars

A short but interesting one today: a while back I was commissioned to run some models in ExoPlaSim representing the climate of Mars terraformed to have an Earth-like climate, and I figured it was finally time to have a look at them in more detail. 

Mars today sits at the outer edge of the conventional habitable zone, with only about 43% the light that Earth receives. To warm it to near-Earth temperatures would likely take one of three potential approaches:

• Increase CO₂ pressure to something like multiple bar.
• Use some artificial, much more potent greenhouse gas or aerosols with a similar effect.
• Place reflective structures in space near Mars to direct more sunlight to the surface.

ExoPlaSim isn't terribly well equipped to model any of these approaches. CO₂ is the only greenhouse gas that can be set directly (water vapor provides additional greenhouse heating but is controlled by the internal hydrology model, and greenhouse heating from ozone can be toggled but not set to any arbitrary value), but the simplistic way the model handles light wavelengths causes it to significantly underestimate greenhouse heating from over about 0.1 bar of CO₂. Light from the sun can be set freely, but reflected light from space mirrors probably wouldn't be distributed in the same way as light from a single light source. Nevertheless, that last option is still the best we can do, so we can perhaps imagine the mirrors are intentionally set up to match the heating pattern of direct sunlight.

Thus, the parameters of the planet and orbit have been set to match modern Mars:

• A radius of 53.2% of Earth's.
• Surface gravity 37.9% of Earth's 
• An orbital period of 687 Earth days, rounded to 685 for the purpose of the model
• A day length of 24 hours and 40 minutes
• An obliquity of  25.19°
• An orbital eccentricity of 0.0934, with periapsis coming shortly after the northern winter solstice

But the atmosphere was set to be a close analogue to Earth (0.78 bar N₂, 0.21 bar O₂, 0.01  bar Ar, 300 mbar CO₂), and the light from the sun was set at near-Earth levels and then tweaked for the desired temperature. As it happens, while trying to reach a temperate average temperature, I first overshot to a somewhat hot climate, and then overcorrected to a cool climate, so we have a range of potential results to look at.

Greyscale heightmap used, with seas marked blue and a maximum elevation at this resolution of 21885 meters above the sea level

The topography is that of modern Mars filled to a sea level 800 meters below the generally agreed "datum" of 0 elevation (corresponding to an equipotential surface with the planet's average radius at the equator). This ocean coverage amounts to about 45% of Mars's surface and would likely require around 160 million km³ of water, 12% of Earth's ocean volume. That much water may not be available on Mars (not to mention the nitrogen required to fill out the atmosphere), but I leave that as an exercise for the terraformer.

This happens to place the coastline close to the "Great Escarpment" that divides the northern lowlands—now a single vast ocean—from the more rugged southern highlands, which means there's not much in the way of coastal plains; most of the land even close to the oceans is about 2 kilometers above sea level. The swings of the escarpment give us  an alternating series of landmasses and seas, which I've labelled here based on roughly corresponding planums and other features on modern Mars so I can reference them later. In the west are the Tharsis highlands, with Olympus Mons still reaching to over 21 km above sea level and a complex eastern coastline formed by features like Valles Marineris; and in the east the gentler but more heavily cratered Sabaea and a large island formed by the Elysium volcanoes. In the south there are two isolated seas formed by the Hellas and Argyre basins; in reality we might expect the water level in these and many smaller basins to vary due to different balances of precipitation and evaporation, but that's fairly tricky hydrology to work out and it probably doesn't make a huge difference at the resolution of this model anyway.

Be sure to contrast these results against our previous look at published data on a speculative warm early Mars (and the followup look with my bioclimate system, if you want to compare these climate maps a little more directly), which used a more advanced but lower-resolution climate model and imagined even lower light levels than today but thick atmosphere rich in greenhouse gasses, and sea levels 3.1 km lower than here. If you're unfamiliar with my climate explorations, you may also want to reference my notes on the limitations and quirks of the ExoPlaSim model in the first exploration and my explanation of the bioclimate classification system I use to map out these climates (though the upper right map in these plates corresponds to the more prominent Koppen-Geiger system).

Cool Mars

We'll start with the cold case, with about 85% Earth's sunlight resulting in a global average temperature of -4.9 °C. You can perhaps think of this as representing a relatively early stage in the terraforming process, though that implies a somewhat curious sequence where the atmosphere and oceans are fully in place well before the climate has warmed.

Much as we've seen before, the combination of obliquity and eccentricity produces a somewhat complex and asymmetric seasonal pattern, here enhanced by Mars's geographic asymmetry. Summer in the north is long but cool, with the northernmost landmasses seeing a brief thaw in midsummer only to promptly start cooling again to a global low in September (or whatever you'd want to call the corresponding period about 9/12 of the year after northern winter solstice). Only the warmest coastal lowlands remain warm enough through this period to avoid nightly frosts and occasional snowfall (and even some of those areas then freeze as the northern hemisphere swings into winter).

The south sees a shorter but more intense summer: much of it is still frozen as late as December, then shoots to over 20 °C just two months later, and is frozen again by April. Bear in mind, though, that months here are almost twice as long, so that still works out to a pretty decent growing season. But that also implies a long winter, as long as an entire year on Earth, with bitter midwinter nights as cold as -90 °C near the south pole. For their part, Hellas and Argyre manage to neatly divide the year into ice-covered and ice-free halves, and many of the southern craters may be similar, with their unfrozen depths providing crucial refuges for freshwater life.

And even the equator isn't free of seasons, with an apoapsis winter from about July to November, a rapid December thaw and then periapsis summer with peak temperatures around March. Some of the lower parts of Sabaea and Tharsis manage to avoid the worst chills, and Marineris remains stubbornly ice-free even as snow gathers on the surrounding cliffs, but there are no tropical parts of this world. The high Tharsis peaks, meanwhile, never thaw, with glaciers gathering on their flanks; A more detailed model of the full process of terraforming might find a greater tendency for large masses of ice to become trapped on Tharsis as moisture is added to the surface, as we saw with some of the models of ancient Mars, but the greater direct light might help induce more melt here, especially as we're presuming some sort of mirror arrangement that could perhaps focus even more onto the problem areas.

Atmospheric circulation is roughly organized into 4 circulation cells, though at the seasonal extremes—especially southern summer—one Hadley cell dominates over the other, pushing the convergence zone far from the equator. Perhaps because they have very uniform surfaces—ice-covered ocean in the north and flattish plains in the south—the polar regions have very consistent east-blowing winds, forming a vortex over each pole, though the southern vortex almost disappears for a few months of summer. This overall pattern is quite similar to modern Mars, despite the substantially greater air pressure and heating, but in detail the patterns are complicated by the ocean and landmass distribution: winds blowing north onto Acidalia in northern summer then diverge towards the warm landmasses on either side; and in their early summer, the southern seas form prominent anticyclones as cool air from their still-partially-frozen surfaces blows out onto the surrounding warm plains.

There is something like a tropical rain belt, but unlike Earth, where trade winds predominantly deliver rains to the east side of continents, here the winds cross the equator and curve around to the west for much of the year, and so the equatorial landmasses are generally wetter on their western sides. To be fair, this is similar to the monsoon patterns around the Indian Ocean on Earth, and here there is a prominent monsoon cycle across the north-facing coastlines flanking Utopia and Amazonas, where seasonal winds alternate between directly onshore in summer and offshore in winter. To the north of the equator, the patches of Mediterranean climates reflect some of the oddities of the seasonal cycle: slow cooling of the northern hemisphere in early summer means the winds are slow to shift and so these areas remain dry, and by the time the winds do shift the world is starting to cool as apoapsis approaches, so the heaviest "summer" rains miss the warmest part of the year.

The south is generally dry as you might expect, but does receive two bursts of precipitation: First in early summer as moist air blows from the cool southern seas onto the surrounding warm plains, which ceases as temperatures start to equalize in February; and then a flurry of snow in fall when the strengthening polar vortex picks up water off the still-unfrozen seas and distributes it around the pole. For most of the southern plains this might be just enough for some annual grasses and shrubs, though around Hellas itself are a few potential patches of cool Mediterranean woodland, alternating between balmy summers and long, quiet winters blanketed in snow, overlooking the shores of the frozen sea.

Overall it's a somewhat dry planet, with less than half the average precipitation of Earth, but these wide seasonal swings in rain help to distribute it well over most of the land area, with the notable exception of Tharsis. Here, the monsoon winds tend to divert moisture from Acidalia to the east, while numerous broad peaks and escarpments block what onshore winds do come from either there or Amazonas in the west. Olympus Mons has a particularly impressive contrast between torrential rains and snow on its southwestern face, up to 7 cm/day in summer, and a rainshadow desert to its northeast, which rarely receives more than 1 cm of rain over an entire Martian year.

As harsh as this climate is in some ways, it is still very much habitable. Thick forests could extend over much of Sabaea and cling to the wetter edges of Tharsis. Marineris gives some tantalizing hints of warm channels flanked in Mediterranean vegetation cutting through the cool, dry Tharsis uplands. And the southern plains may not be terribly attractive for settlement at this stage, but they could still support a rich steppe ecology like the drier parts of northern Siberia or Canada, with generous room for herds of migrating animals, like the mammoth steppe of Pleistocene Eurasia.

I'm not saying it would necessarily be the wisest move for us to populate a partially terraformed Mars with herds of genetically reconstructed mammoths, but you can't deny that it makes for a striking aesthetic.

Temperate Mars

We'll look next at the ultimate result of balancing the climate to match Earth, with 93% of Earth's light giving it an average temperature of 14.5 °C. This is perhaps a little closer to what people generally have in mind for a terraformed Mars, but it's not without its tradeoffs.

Due perhaps to its smaller size and slightly higher obliquity, this world has a smaller overall range of average temperatures than Earth, but this belies the significant seasonality in many areas. The overall shape of seasonal variation is much the same as in the last case: on the northernmost coasts of Tharsis and Sabea, a long, mild summer peaks in June, though overall temperature variation through the year isn't much more than 10 °C, and the entire northern ocean remains ice-free through winter.

Even a bit to the south, the cycle is nearly reversed, with an apoapsis chill reaching its nadir in September, with the nighttime frost line in this case crossing the equator from the south but not reaching much past it, and then a hot but generally tolerable periapsis summer peaking around March (though barely higher than in February or April, hence why I chose to show those instead as they better represent the shifts in the south). There's a notable disagreement here between the Koppen-Geiger and Pasta climate classifications on the extent of tropical climates, which we've seen before in cases with cool equators: The Koppen tropics are marked based on winter average temperatures of 18 °C, which is a good fit to the transition to subtropical biomes on Earth, but in those areas winters also have longer nights and less direct sunlight by day. But that won't be the case for somewhere closer to the equator (especially here where the coldest month happens to fall near an equinox), so they won't necessarily have the same nighttime lows and will be less prone to frosts, as represented by defining tropical climates by minimum temperatures in the Pasta system. Tropical climates here are a bit more seasonal in temperature than we might expect on Earth, but never particularly cold.

The southern seasons actually follow almost the same timing as the equator, coldest in September and hottest in February, but with rather more impressive extremes. Winters are shorter, with the spring thaw generally coming about a month earlier and fall freeze a month or two later compared to the previous case, but still sinks to nighttime chills as low as -70 °C near the south pole. The more serious issue, however, may be summers, with daytime highs leaping to over 50 °C for two months, leaving only a month or two before and after of more moderate conditions (though again, these are long months).

There are no permanent glaciers or sea ice predicted anywhere on Mars in this model, though that may be due in part to the model's limited resolution around the highest mountain peaks.

Wind patterns are almost the same as in the last case, alternating equator-crossing winds and polar vortices. The vortices are are perhaps slightly weaker and the Hellas anticyclone a bit more persistent, but the differences are all quite subtle. Now that the northern ocean is unfrozen, the norther polar vortex appears to form some fairly consistent cloud bands, though there's no land beyond 40° north to witness them.

This brings much the same pattern of monsoon rains, though a bit more concentrated in the hotter climate. Southern Sabaea benefits from a fortuitous arrangement of winds carrying water alternately from Acidalia and Hellas, supplying a broad tropical to subtropical rainforest. Elysium is another winner in this new climate, with the shifting winds alternately pelting its slopes with rain from every side, steadier in the east and more torrential in the west.

But the heaviest rains come again in Tharsis; Olympus Mons again catches heavy downpours on its southwest flank, up to 12 cm/day, but a larger wet basin has formed on the slopes of Arsia Mons to the south, where onshore winds from Amazonas are only briefly interrupted during northern summer. Funneled up the broad volcanic slopes, this creates a steady gradient from tropical lowlands to snowy montane forests and tundra. Even the Tharsis interior sees a bit more rain, either carried over the mountains from the west or working its way in from Acidalia and Marineris to the east in the brief period around periapsis where the equator is substantially warmer than both poles and something closer to Earth's tropical rain belt forms along it. Northern Tharsis, however, is drier, with a particularly vicious drought across much of periapsis summer.

The south, on the other hand, appears to be somewhat the loser here. Precipitation is a bit higher on average, and again comes in two periods related to the shifting wind and temperature patterns: in late spring, moist air off the cool Hellas and Argyre seas circulates into the polar vortex and produces light snow and rain over the polar plains, though the mid latitudes are left a bit more dry. Come summer, the polar vortex collapses and moisture off the seas is diverted north, leaving the plains to bake under the midsummer heat. In fall, the vortex returns and the shifting Hadley cell winds also carry some moisture across the mid-latitudes, producing a longer rainy period over most of the southern hemisphere, that slackens and switches to snow as the south cools, cutting off in winter as the seas partially freeze over. Things may then not be as dire as they appear, as there's a substantial potential growing period in fall for grasses and shrubs. But this is no mammoth steppe, and prospects for wooded shores on Hellas seem dimmer.

On the whole, this is a lusher world , particularly in the areas around Tharsis not trapped behind rainshadows, and perhaps a closer match to the general range of Earth's climate, though the smaller planet lacks the warmer and colder ends of our spectrum, and seems fundamentally unsuited to forming the broad subarctic forest of Earth by virtue of its topography and orbit: the north lacks any land at high latitudes and is fairly aseasonal at the mid latitudes, while the south is too dry and viciously seasonal.

Compared to cooler Mars, the marginal areas here are more marginal, with much of the semiarid regions there now overtaken by desert and semidesert in the hotter climate. One can perhaps imagine an awkward transitional period, where areas seeded with life and perhaps settled early in the terraforming process must then be abandoned, and much debate can be had between different regions of settlers on exactly what the final goal of the process should be.

Hot Mars

This was the result of my initial attempt to run the model with the same light level as Earth, resulting in an average temperature of 34.8 °C, like because the smaller, drier planet has less ice and cloud cover and so a lower albedo (around 0.27 here). I'm not sure what reasonable scenario would give this result, perhaps some severe mismanagement or orbital mirrors left unattended after some societal collapse, but now that I have this result it's worth at least a brief look.

Though the northernmost tips of Tharsis and Sabaea still follow the northern obliquity seasons, most of the lower latitudes are again dominated by the eccentricity cycle, with sweltering summers averaging over 40 °C for a third of the year in many areas. In the south, there is still a winter freeze, though it doesn't quite reach the southern seas, but this alternates with torrid summers with daytime highs over 80 °C in some areas (more closely following the obliquity cycle in this case and peaking in January). Even the seas there don't offer much respite. The best refuges for human life in these climates are likely to be the northern tips of the landmasses, including perhaps northeast Elysium, and the Tharsis highlands. 

Winds are much the same, though the summer heat seems to help pull a little more air off the oceans into northern Tharsis, making it a bit wetter.

(The line across the northern oceans is likely a modelling artifact, though there may be some consistent cloud bands associated with the polar vortex.)

The hot climate is wetter, but average precipitation is still only moderately greater than Earth. Heavy monsoon rains sweep across the low latitudes, with more total droughts in the winter hemisphere but fewer areas that never receive rain. Olympus Mons manages to catch up to 17 cm/day of rain on its western slopes in August. In the south, the precipitation is more consistent through winter as the seas never freeze over, but still interrupted for several months in the hottest part of summer.

This wetter climate could be quite lush with vegetation that could tolerate the heat, but plant life brought from Earth is likely to struggle, retreating as well into the cooler refuges. Don't regard this one as an intentional goal for any terraforming effort we might undertake, but perhaps a template for a Mars-like hot world elsewhere in the cosmos.

That'll do for this small exploration. There are of course other potential approaches to terraforming Mars that might imply different climates, but this is about what I can manage with my available tools. If anyone feels like illustrating Martian mammoths, let me see the results.

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