Public Climate Data Explorations: Habitable Mars

This is a second in a series of posts looking at data from previously published climate modelling that I can work into climate maps. To reiterate from the first post, I did not run these models myself, I’m just making maps based on them, and because much of this data contains only average data, I’m mostly making maps of Holdridge life zones using estimates of average biotemperature that are based on Earth’s seasons so may not exactly apply here (aside from a couple cases with no seasons where I can assume average biotemperature is about the same as average temperature).

Today, we’ll look at a couple studies modelling the climate of a potentially habitable early Mars. For those out of the loop, the general consensus among researchers based on pretty clear geological evidence is that Mars probably had a period with a substantial atmosphere and bodies of water on the surface at some point in the first billion or so years after it formed, but exactly how much water and air it had and how long this period lasted remain open discussions.

The papers we’re discussing today attempt to reconstruct that early Martian climate at a couple points in its history based on reasonable guesses at the potential atmosphere and water content. But you might also use these results as a rough template for how a terraformed Mars might appear as it warmed up, though the lower insolation than modern Mars and somewhat human-unfriendly atmospheric composition in these models mean they’re not a perfect fit (I looked and managed to track down one attempt to simulate the climate of a modern terraformed Mars, but it’s not especially detailed).

Don’t take these models as totally authoritative on the climate of Early Mars, though; other models have proposed cooler or more transiently warm climates that still match up decently well with at least some of the geological evidence, but the models we’ll look at today are still interesting as one possibility—and even if they don’t pan out for Mars’s case, they still hold some lessons for the potential climates of other small, dry, cool planets.

For reference, here’s the topography of modern Mars:

 

 

Notable highlights include the ~8-km deep Hellas basin in the southeast, an enormous impact crater, the ~12-km high Tharsis volcanic plateau in the equatorial west, and a ~4 km escarpment ringing the planet between the younger, flatter northern lowlands and older, more cratered southern highlands. The planet orbits at 1.52 AU, giving it a 687-day year and about 43% the insolation of Earth, and it currently has 0.093 eccentricity, a 24.7-hour day, and 25.19° eccentricity, though the obliquity may have varied significantly over time. It has a 0.006-bar CO₂ atmosphere and a bracing average temperature of -60 °C, though equatorial regions can occasionally rise above freezing; but the thin atmosphere means any liquid water on the surface would promptly sublimate anyway.

"3D Simulations of the Early Martian Hydrological Cycle Mediated by a H2-CO2 Greenhouse"

Guzewich et al. 2021

Paper

Dataset

This study simulates a set of scenarios for a potentially habitable Mars about 3.8 billion years ago, when the sun was 3/4 of its modern luminosity so the planet had just 32% of the insolation of modern Earth. Despite this, the study shows how various combinations of CO₂, H₂, and CH₄ could have allowed for habitable surface temperatures.

The study includes a lot of model outputs, so I’ll pick out some highlights: most of the models are just testing various atmospheric mixes without any surface water, so you can see the paper for temperature maps, but the short version is that as you warm the planet, Hellas reaches above-freezing temperatures first, then a few other basins before the whole north thaws, but much of Tharsis remains frozen over even in the warmest models.

For our purposes the interesting models are all run with a 2-bar atmosphere of 94% CO₂, 5% H₂, and 1% CH₄. The study first experiments with a few different levels of initial water in the soil or in shallow lakes, but in all cases most of the water promptly becomes trapped in ice on Tharsis; more initial water actually leaves less leftover because it allows larger glaciers to form and cool the plateau further. Some water always remains in circulation and brings light rain over parts of the desert, with a few particularly wet patches on the west flank of Tharsis that catch moisture from the tropical winds (due in part to its smaller size, Mars has a single equator-crossing tropical circulation cell with winds curving around to blow predominantly east at low latitudes, rather than Earth’s paired Hadley cells with west-trending winds), but never enough to show up on the climate maps. Meltwater out of these glaciers could potentially be another major source of water to the area around Tharsis in these scenarios, but isn’t extensively modelled here.

 

 

Things get a bit more exciting when they introduce a small sea in the Hellas basin, bringing more substantial rains to the southern hemisphere:

 

They then also add a larger ocean in the north as well (the paper implies this model excluded the Hellas sea, but the data here seems to include it):

 

 

The heavily cratered terrain gives us a pretty noisy precipitation distribution, but we can see some overall patterns: the surface is still mostly arid, but with wetter regions around the oceans that could likely support some vegetation, and again Tharsis catches some moisture on its western flank but casts a long rainshadow to its east.

The study also includes a few models with a speculative prehistoric Martian topography assuming that Tharsis formed after this point in Mars’s history and that its rise caused the rotational axis to shift. This reconstruction remains controversial, but without Tharsis trapping water (the pre-Tharsis highlands in the southwest of this map are much lower and trap far less water), adding small amounts of initial water more substantially increases global precipitation:

 

With 10-meter-deep initial lakes:

 

 

With 100-meter-deep initial lakes:

 

 

With 500-meter-deep initial lakes:

 

The last model seems to imply a lush global climate despite the lack of seas, but that’s quite a bit of initial water and the output data is a little unclear about exactly how much area this is covering (ROCKE-3D allows for modelling lakes with dynamically adjusting water levels rather than set seas, but I’m not totally sure how to interpret the data as actual lake area); I suspect that much of this terrain is covered in small seas caught in the irregular terrain that’s providing rain to the surrounding area, whereas in reality this climate would probably be accompanied by widespread erosion, forming channels which would allow water to drain into a few larger seas, resulting in a somewhat drier overall climate.

As well as being an interesting look at a habitable Mars, this also gives us a potential template for cool, dry worlds generally: mostly arid, but with potential for wet regions even with low amounts of global water, but heavily dependent on topography and particularly the presence of potential cold traps.

One other trend I want to highlight is that—at least outside the frozen southern highlands—the higher, cooler areas tend to be the wettest here, somewhat the reverse of the tendency on Earth; those two wet spots in the north are highlands and the hot, dry patches in the southeast are deep basins (sometimes higher areas are marked as wetter climates on climate maps just because they’re cooler and so have less evaporation, but this isn’t the case here: precipitation is substantially higher in these cool highlands and very low in the basins). The study doesn’t discuss this much, but presumably this is due to orographic rains in the highlands and adiabatic heating in the lowlands; as air descends several kilometers into the basins, it warms and dries. On Earth these tend to be secondary effects relative to the main circulation cells that control the distribution of precipitation, but due either to Mars’s different circulation or the lower total water availability, topographic effects are just more prominent here.

“Circumpolar Ocean Stability on Mars 3 Gy ago”

Schmidt et al. 2022

Paper

Dataset

This study brings us to a point a bit later, 3 billion years ago. Geological evidence indicates that Mars had significantly cooled or dried by this point, but exactly how much remains a point of contention): Erosion by water over the southern highlands seems to have stopped, which is generally taken to indicate that the planet must have frozen over or dried out by this point and so lost any earlier oceans, but erosion still continued in the lower northern regions and there is even evidence for unfrozen seas. Transiently warm climates or even brief warming after major impact events have been proposed as explanations, but this study attempts to model a stable cool but wet climate.

Insolation here is set to a slightly higher 33% of modern Earth, and modern Mars topography is used (save that the northern ice cap is removed) with deep oceans filling in Hellas and the northern lowlands. A 1-bar atmosphere of mostly CO₂ is used with either 10% or 20% H₂, and for each H₂ level 4 models are run with 0°, 20°, 40°, and 60° obliquity (another set of models for all these cases is also run with a non-dynamic ocean for comparison’s sake, but I’ll skip those).

I’ll start with the 20% H₂ models, because these are about the same as the last study and probably not actually a good match for Mars at this point in time.

0° tilt (the eccentric orbit would still give a bit of seasonality but I’ve assumed constant average temperature for simplicity’s sake, which also allows me to make a seasonless Koppen map):

 

 

20°, which is the closest to modern Mars:

 

 

40°, which the authors consider most likely for ancient Mars:

60°, which is high enough that my assumed average temperature-biotemperature relation probably doesn’t work too well:

 

The 10% H₂ models seem to be closer to the authors’ expectations for Mars at this time and also have monthly climate data included, so I can provide proper Koppen-Geiger maps, though I’ll still do Holdridge zones for comparison (using the monthly data for biotemperature rather than my estimation scheme from average temperature).

 

0°:

 

 

20°:

 

 

40°:

 

 

60°: 

 

There's maybe a bit of disagreement here between the Koppen and Holdridge schemes about how dry these areas are, but that's always a bit hard to nail down for very cool regions where evaporation rates are low and a lot of moisture may be present as ice and snow throughout the year, which is inaccessible to plants while frozen but can provide significant meltwater in spring.

 

The main argument of this study is that ocean circulation (as is properly modelled in ROCKE-3D) could allow for unfrozen Martian oceans to persist even in a globally cold average climate. Again, aside from the specific interest in Mars, this gives us a template for another type of marginal climate, a near-snowball world with large areas of cold desert but also a surviving warm sea and some reasonably hospitable patches along the coastlines (and this might also be taken as a template for a partially terraformed Mars if seas are added early).

That’s it for today. Other modelling has been done for various scenarios of Mars at various points in its history, but either didn’t fully model processes like rainfall or don’t have publicly available data. Regardless, the general consensus seems to be that episodes of at least locally wet conditions might have occurred for a while, but by about 2 billion years ago the planet had completely dried up; any water not trapped in ice or subsurface aquifers was lost to space, and probably most of the atmosphere as well.

Next of these posts will probably be climate models of Venus; see you then.

Buy me a cup of tea (on Patreon)