PART FOUR
(4.1–3.7 billion years before the present)
SAILING ON AN ENDLESS SEA
This is much more like the adventure you were promised. You are about to sail on the ocean of Mars – or you would if you had a sail, but this is the next best thing: a little inflatable boat equipped with a small, robust and necessarily enclosed electric motor to move you about.
At the shore, the waves behave strangely; the one-third gravity makes them tall and thin. It’s difficult to launch the boat in these conditions, as the white horses try to drive the hull back onto the beach. The wash pours in and drags out across the rocky beach, booming and roaring. You wrestle the boat through the surf with help, and a spacesuit that’s totally sealed against vacuum is good enough for the shallows. The motor works perfectly and you dip the fast-turning propeller into the water – you’re off.
Every time you skip over a wave, you feel like you’re airborne for a moment: the propeller spins wildly as it claws at the air, then gives you a surge of acceleration as it submerges again. You finally escape from the line of breaking waves and reach deeper water, where the swell is still severe but manageable. No one wants to throw up in a spacesuit.
You have been given the task by virtue of your summers by the sea. You know how to handle small craft, but this is the strangest shore you’ve ever departed from. The land is a dense black line. The rock is dark. The sand is black. There’s little variation in colour: there’s no oxygen in the air to react with the minerals and turn them red, or any other colour. The land slopes down inexorably from the south to the north; any visible change in height is either due to an eroding crater wall or the scar of a river.
You call them rivers, but they’re more like wadis – dry riverbeds that burst into short-lived exuberance after a storm has passed inland. There’s nothing to slow the run-off, no soil, no vegetation to soak up or impede the flow of water from the land back to the ocean. Every time it rains, the land is scraped clean and the rivers rise in their gorges so fast that a wall of water charges downhill, bearing everything from fine silt to boulders, battering and cracking the banks and the beds, and then subsiding almost as quickly as it appeared. A pour, a trickle and stop.
And now you’re afloat on that reservoir of water. There are lakes in the highlands, where huge craters have warped the surface and produced deep local lows that will never be overtopped. They have their own weather, but you’re on the great northern ocean and there’s nothing but waves between you and the pole. In fact, as there’s no ice, you could – in theory – cross that point and head south again, to land a full 180 degrees of longitude from where you started, a journey of almost 10,000 kilometres. Wider than the Atlantic. Almost a Pacific crossing.
It’s why the swell is so high. The winds rattle around the hemisphere of water with nothing to deflect them. They’re always dragging on the surface, always raising waves. Sometimes there’s a storm – there’s always a storm, it just depends where – which gathers itself up and bulls its way inland, releasing torrents of water onto the north-facing Dichotomy rise: as the rain clouds are pushed by the slope into higher, colder air, every last drop is wrung out. The southern high latitudes are almost a desert.
And sometimes, there’s a bigger wave still. With half a planet of ocean, a meteorite strike can produce a tsunami of gargantuan proportions. This colossal wave can smash against the shore, tearing it up, flinging debris up the slope and boring deep along the same river valleys that carry the water down. Some of the waves you’re skipping along the top of are echoes of those events, passing and repassing around the circular sea.
While you still have an important scientific mission to carry out, you can’t help but feel the joy of zipping off into the distance with your load of instruments and sample containers. You reach the first waymark and cut the motor. Your sampling device is no more complicated than a weighted length of line with a bottle strapped to it. When you press the trigger at your end, the bottle will open and suck in the water, and it will close again when you let go. There are also temperature gauges and a device to log your position for that sampling event. The most difficult thing you have to remember is to use the numbered bottles in the right order.
As you go further out, the shoreline sinks rapidly. If it was perfectly calm, you’d lose sight of it after a mere five kilometres. But it’s never calm, so the distant dark line pops up as you crest a wave. As it passes, it blocks your view until the next time. You’re surrounded by a bowl of water, with only the cloud-studded sky above. You can imagine being out here with no idea of where land was, and instead of turning south, going in any other direction and becoming hopelessly, wildly lost. Except you’re a better navigator than that. The Sun might be confusingly high now, but it still sets in the west and you know south is to its left.
You restart the motor and plough on through the waves to the next sampling point. You could almost forget where you are, and be tempted to unseal your suit and let the salt water slap against your bare skin. If you held your breath, you might even survive the experience. There’s no one out here with you, either to stop you or to help if you get into difficulties. It’s just too risky; the moment passes. You’ll always remember it, though, and wonder if you shouldn’t have been a little more reckless.
THE HELLAS IMPACT
Hellas is the single largest extant crater in the entire solar system, at 2,300 kilometres across and 9 kilometres deep (but it’s only the second largest impact feature on Mars, after the enormously eroded Utopia Planitia basin, or the third if we credit the Great Dichotomy impact theory). Its creation marks the end of the pre-Noachian and the beginning of the Noachian. In a history which already contained repeated huge impacts of global significance, Hellas was still an absolute monster. The ejecta was thrown pole to pole, enough to bury the entire surface of Mars half a kilometre deep; the temperature soared and the air pressure rose dramatically as liquid water turned back into steam. The immediate effects lasted for decades and centuries, but others persisted for far longer.
The amount of energy released by such an impact is almost impossible to comprehend. As well as excavating some seventy million cubic kilometres of Martian rock and launching it up to halfway around the planet, and even into space, much of the violence of the explosion went down into the crust. Cold rock is brittle. It’s strong when compressed, under a heavy load bearing down from above, but it’s extraordinarily weak in tension. Breaking rock is easy when it’s simultaneously being pulled apart.
In the moments after impact, when the pressure wave was driving through the crust and into the mantle, everything was in compression. After the explosion, all that pressure abruptly disappeared: the energy stored in the rock was released and everything was thrown into tension. Things broke. A lot of things broke.
The rock beneath the crater, and for a thousand kilometres around, shattered, deeply and irrevocably, into huge blocks and arcs, metres and kilometres across. Those fault lines stayed in the basement of the rock forever – no matter that the land appeared to be healed by subsequent overlying deposits of material. The crust would remain vulnerable to movement along those lines. As well as later impacts reopening the old scars, the fractures were exploitable by other processes, such as volcanism and water.
Further below, in the plastic mantle where rock was hot and flowed on a timescale of millions of years, the cracks sealed up, but given the sudden removal of so much overburden, the mantle also pushed up into the absence left behind. The crater was initially even deeper than it appears now, but as the relaxing mantle pressed against the rigid crust above, all those broken lines and points of flexure shifted to accommodate the movement.
This had one further effect that would only be seen later. With the mantle now able to penetrate closer to the surface, retaining its heat but relieved of its pressure, it spontaneously melted. Then, by exploiting the shattered pathways to the surface, the less dense magma worked its way towards outlets around the crater rim. There are six separate sites in total where this happened, forming the earliest known volcanoes on Mars and the only major volcanic province in the southern highlands: Tyrrhena Patera and Hadriaca Patera in the north-east, and to the south and southwest Amphitrites Patera, Malea Patera, Peneus Patera and Pityusa Patera. Whenever it was that these volcanoes began erupting, they continued to be active for hundreds of millions of years, so deep were the scars in the rock beneath them.
As the shockwave spread out from the impact, the energy was distributed across the wavefront. Surface waves – a marsquake – rattled through the existing landscape, bringing down crater walls, collapsing peaks and throwing them into valleys, losing energy all the while. The deep waves, travelling through the bedrock of the crust, shifted faults and moved blocks that measured kilometres across.
But Mars is a sphere. The shockwave radiating from Hellas reconvened on the exact opposite side of the planet. It’s pure conjecture that this refocusing of impact energies had any effect, but Alba Mons – Mars’s largest volcano by volume – is precisely there, 180 degrees from the Hellas ground zero. Hellas and Alba Mons are understood to be separated by at least a billion years of history, and their relative positions may be entirely coincidental. But there’s a strong suspicion that Hellas produced a weakness that volcanism later exploited to create Alba Mons, having fractured the crust for tens of kilometres down.
The Hellas impact falls within a time period in the solar system known as the Late Heavy Bombardment. The evidence suggests that after the formation of the planets from coalescing planetary embryos, planetesimals and the other detritus of the debris disc, collision rates reduced from almost constant to sparse and sporadic. This happened not just for Mars but for all four planets in the inner solar system, as well as the Moon. Then, around four billion years ago, the rate of major impacts may have spiked again. This phenomenon was originally suggested to explain the formation of the dark lunar basins, which are craters that subsequently flooded with lava – none of the returned samples from those areas date older than 3.9 billion years.
In a simple model of planetary formation, cratering rates start high and quickly decrease, thereafter slowly dwindling away to currently observed rates. The model doesn’t explain how a second peak in impacts could have occurred. We have, however, already suggested a mechanism that could explain it: the Grand Tack theory. The frozen outer reaches of the solar system still contain a great many objects, some of them very large indeed: Pluto, Eris, Haumea, Makemake, Sedna and Quaoar are all over 1,000 kilometres across. Add a myriad of other early solar system objects of all sizes, almost all rock and ice, and there was a ready store of projectiles to aim at targets lying sunward of them. Giant planets wandering outwards beyond the orbit of Mars would have undoubtedly perturbed the movements of any planetesimals that they encountered.
As Jupiter slid back out to its current orbit over the course of a few hundred thousand years, it would have rudely shoved Saturn towards the other gas giants, Uranus and Neptune. These planets in turn would have also pushed outwards, directly into the area occupied by the frozen accumulations of ice and dust that had previously escaped entrapment in the gravity wells of larger objects.
At this point, the orbits of these minor bodies would have changed, and many of them would have ended up heading on wildly elliptical paths towards the Sun. A barrage of unclaimed material, rich in volatiles, would have spun towards the centre of the solar system. Undoubtedly, Jupiter soaked up a great number of them, but many would have flashed past, accelerating all the way, until they turned around the Sun trailing comas of dust and gas, before heading back out to the cold depths of space beyond Neptune in their new, altered orbits.
The chances that any individual rock – we should probably call them comets, because that is what they were – would strike a planet were small. But given a couple of hundred million years of cycling from the very edge of the solar system to its heart and back again, those odds started to become more of a certainty. Impacts from these comets could quite possibly have added millions of tonnes of water and other volatile chemicals to the now-depleted Martian atmosphere.
There remains doubt as to whether the Late Heavy Bombardment ever took place: some of the lunar evidence has been questioned, although a sudden, discrete influx of large, water-rich meteorites does explain some of the physical and chemical anomalies we find on the rocky planets. It also makes a tidy conclusion to the Grand Tack theory. But it could also be that the impact rate did just slowly ebb away, and the creation of Hellas was merely an act of cosmic tidying up – one of the last great impacts of the early solar system.
THE START OF THE NOACHIAN
The picture we have of Mars after the first half billion years of its life is inevitably provisional. Reaching back so far in time, peeling back younger layers to reveal the ancient landscape, painstakingly reconstructing it from behind the fault lines, craters and volcanoes that came later is far from a straightforward task.
But the image we’ve built up so far looks something like this. In the south, there was a heavily cratered surface, sitting kilometres higher than the land to the north. There were the scars of relatively new craters – Argyre, Chryse and the great fist-mark of Hellas – plain and fresh, and of such depth that lava eventually broke through the cracks on the crater floor and puddled as sheets of infill. In addition, Hellas developed discrete volcanic sites on the crater rim, where explosive eruptions – highly likely given the reaction of hot lava meeting groundwater – created hot, thick ash clouds that rushed down from the summits, leaving in their wake dense, metres-thick, heat-welded jumbles of water-rich glass fragments and pulverised lava. Geysers and gases spewed from the ground – water, carbon dioxide and sulphides – and mixed with the atmosphere. The air pressure, outside of spikes caused by huge impacts, was somewhere in the region of one or two bars, and the temperature fluctuated wildly around a slowly cooling average.
There was an ocean, a mighty ocean that not only covered the northern half of the globe, but also turned the deep craters in the south into inland seas. And as difficult as it is to imagine that ocean and those seas as temporary, they were there at the whim of the local environmental conditions. They had disappeared before, repeatedly: one big meteorite strike and the soaring temperatures would evaporate every last trace of surface water and turn it into atmosphere. When it eventually came down again as rain, there would always be less of it than before, because some of it would have bled away into space.
Those times of great disruption were largely over, though smaller meteorites still fell, creating craters tens or hundreds of kilometres in diameter. Those that struck the land threw ejecta out in great blankets. Those that hit water caused tsunamis hundreds of metres high that washed across the shorelines and travelled far inland, up valleys and down again, carrying off all the sediment they could hold, which they dumped back into the oceans as they quietened.
Those ocean sediments formed stratified layers of rock: gently laid sandstones and mudstones made of eroded material from the highlands deposited in more placid times, interleaved with layers of churned flood deposits from the ebb and flow of impact-induced tsunamis. Meanwhile, on land, flood basalts – immense, region-wide eruptions of very fluid lava from fissures fed directly from the mantle–crust boundary – accumulated huge thicknesses of black rock. These were cut with lobe-fronted sheets of locally erupted lava from volcanoes, impact debris and wind-blown ash. The land was scoured from above, eroded by corrosive rain tainted with sulphur and carbonic acids.
In a sky that was primarily water vapour and carbon dioxide, there were bands and stacks of clouds, and possibly even a permanent fog that reached from ground level to the highest peaks. If the clouds did part, the revealed sky would be blue, because an atmosphere scatters shorter (blue) wavelengths of light more than longer (red) ones, effectively spreading that colour across the sky.
Rain-fed rivers cut through the surface deposits of broken ejecta and volcanic ash and used the debris they carried to grind away at the valley sides and riverbeds, carving channels and gorges in even the strongest rocks. They flowed downhill: away from the region dominated by the slope of the Great Dichotomy, water accumulated in craters or found its way into the faults that split the surface, where it was added to the deep reservoir of groundwater that was connected underground to the northern ocean.
From equator to north pole the ocean was a kilometre deep, with currents that carried the warm equatorial water to the pole, and transported the colder water back to the shore at the Dichotomy, making the equator more temperate than it might otherwise have been. And the weather: storms, immense and powerful, swept in off the northern ocean, bearing rain which turned to snow as the fronts moved inland and were pushed higher over the rising ground.
What about below the seabed? The basement rocks were brutally cracked, and water, seeking its lowest level, migrated into these cracks, filling up every available space, descending and heating up until it became too hot to stay liquid. Dissolved salts and other minerals filtered both up and down through the open matrix of loose sediments and volcanic debris, gluing them together to form solid rock. Hydrothermal vents formed along fault lines: boiling, mineral-laden water squirted into the cold depths of the sea. As we’ll see shortly, if we ever want to find evidence of life on Mars, the remnants of the Noachian ocean is where we ought to look.
It is important to remember that catastrophic events – floods of lava and water, meteorites falling from the sky – were just punctuation to the main sentence. Hundreds or thousands of years might pass without incident. But it’s just as important to realise that, given the long reach of the Noachian, those events were certain to take place. Within a span of a million years, a thousand once-in-a-thousand-years occurrences happened – a statistical certainty of instability and chaos contained within a semblance of what we might have considered normal.
The forces the planet had been subjected to already were monumental. And there was more still to come.
OBLIQUITY AND ECCENTRICITY
Before we feel our way further into the Noachian, we need to talk about Mars’s cavalier attitude towards both maintaining a stable orbit around the Sun and where its axis of rotation is pointing, because they have severe implications for the Martian climate at every point in its history.
Mars spins on its axis like a top; one revolution is a Martian day. It also circles the Sun once in a Martian year. So far, so normal. Because Mars’s axis is tilted over, it has seasons, just like Earth does. When the north pole is angled towards the Sun, it’s summer in the northern hemisphere; 334 sols later, or half a Martian year, the north pole is pointing away from the Sun and it’s winter in the northern hemisphere.
Mars has two added complications to this entirely average planetary scenario. The first is this: no planet’s orbit about the Sun is exactly circular. It is almost circular, but not quite, and this not-quiteness is called eccentricity. Comets, which have high eccentricity – that is, highly elliptical orbits – can be very distant from the Sun most of the time, well outside of Neptune’s orbit, and then swing in fast and close to the Sun, inside the orbit of Mercury. Planetary orbits don’t vary anywhere near as much, but Mars currently has the largest eccentricity of any planet barring Mercury, and its furthest and closest distances to the Sun vary by more than 40 million kilometres.
If summer coincides with the closest approach, it’ll be unusually hot. Likewise, if winter comes around when Mars is at its furthest distance from the Sun, then it’ll be colder than average. Moreover, the eccentricity itself changes gradually over time, driven by the gravity of other planets, chiefly by massive Jupiter. These periods of greater-to-lesser eccentricity occur in two cycles: a smaller one of 96,000 years and a much larger one of two million years.
The second complication is that the Great Dichotomy, however it was caused, altered the distribution of mass on Mars and left it unstable. As a result, the angle at which Mars tilts from straight up changes over time – and not by just a little, but by an enormous amount. The angle that a planet’s rotational axis makes compared to the plane of its orbit is called its obliquity; this would be zero degrees if the axis was pointing straight up, and ninety degrees if the axis was lying flat. A planet is able to maintain a steady obliquity if it has a comparatively massive moon to stabilise it, but Mars doesn’t have that: its two tiny moons do it no favours at all, so the axis wobbles back and forth.
Mars’s current obliquity is twenty-five degrees, giving it moderately strong summer and winter seasons; it’s been this way for a hundred thousand years, maybe more. But there have been times when the obliquity was much closer to zero, meaning that summer and winter were abolished for the duration. At other times the axial tilt has reached sixty degrees: under these conditions, the Sun would be higher in the sky at the poles than at the equator, and the polar regions would be subjected to 334 sols of freezing cold and dark, followed by 334 sols of burning heat and light. Year on year. Again, for hundreds of thousands of years.
The obliquity constantly changes, within a range, by a fraction of a degree a year. Over time, the changes accumulate, back and forth. A low-obliquity period, like Mars is in now, can last millions of years. But the swings between different obliquity ranges happen quickly, in a matter of a few thousand years. It’s a chaotic system, feeding off of tiny changes in gravity due to Mars’s eccentric orbit around the Sun. Going back even four million years – let alone four billion – to find out what the axial tilt would have been is a futile exercise.
While the effect of changing obliquity on the climate might not be predictable in terms of time, it is in terms of scale. Depending on the reinforcing factors of high obliquity and high eccentricity, the amount of energy from the Sun reaching Mars’s poles can increase fourfold in summer. The opposite is also true. Mars has sustained million-year periods where low obliquities meant very little sunlight ever reached the poles: ice caps formed, locking up water that might have otherwise fallen as rain.
Young Mars was energetic enough, and had a thick enough atmosphere, to buffer its environment against the more extreme effects of its astronomical wanderings. But for older Mars – Mars of later, sparer periods – its entire climate came to depend on this chaotically changing obliquity.
INTRODUCING THARSIS
Tharsis is Mars’s largest volcanic province, covering a quarter of the planet’s entire surface. It has several descriptive names: Tharsis Rise, the Tharsis Bulge or the Tharsis Plateau. It’s roughly 5,000 kilometres across, with a surface area of – depending on how we measure it – 30 million square kilometres. It straddles the Great Dichotomy and rises even higher than the southern highlands, up to 7 kilometres above datum: that’s without considering the vast heights of the individual volcanoes that perch on it. The weight of Tharsis is estimated at a billion billion tonnes, and it’s another feature, alongside the Dichotomy, that’s large enough to interfere with Mars’s axis of rotation. The weight of Tharsis has grown so great that the crust it rests on has failed in several places, opening up huge swarms of visible parallel cracks in the surface layers that presumably also extend into the depths.
Tharsis is a feature that dominates Mars. Our eye is inevitably drawn to this great bolus of rock that has erupted from the face of the planet, and the question of how it could possibly have got there and sustained itself over such an incredible length of time is one that will take some careful unpicking. There are several stories that need to be told here, but rather than two opposing tales, the formation of Tharsis is more of a rambling narrative that lasts far longer than it has any right to, complete with anecdotes and digressions.
Because as massive as it is, it has necessarily grown and changed over the four billion years of its existence. What we see of Tharsis today looks both permanent and ancient, but while some of it is very old, other parts are startlingly young. Centres of volcanic activity and tectonic uplift have shifted across the whole Tharsis region over time, either as a response to the vagaries in the precise direction of the mantle plume that is its primary cause, or as a result of the opening and closing of channels that allowed magma to rise to the surface as the crust flexed and filled. Tharsis is complicated. Different things have happened to different areas of it at the same time, and there’s no straightforward presentation of the facts that fits.
We know that the only explanation for Tharsis is that it was, and continues to be, caused by a mantle plume: a single rising column of hot mantle material pushing against the base of the crust underneath it. We’ve already encountered mantle plumes in connection with the Great Dichotomy, but the plume associated with the Dichotomy was in the wrong place to power Tharsis. Either that plume moved from the south pole, or the mantle moved, or the crust moved, or all three – or another plume started. Is there any evidence of migration, either of a southern plume moving towards the equator, or of the crust having slid over it? Furthermore, why would it have stopped just the other side of the Great Dichotomy, rather than continuing on its journey?
To answer the first question: potentially, yes. There’s a path from the south pole heading towards Tharsis, roughly along the longitude line eighty degrees west, that is suspiciously smooth given that this region is part of the heavily cratered southern highlands. There are large craters, visibly shallow and flat-bottomed – arguably filled – but the smaller ones we would have expected to see have mostly disappeared, giving a much younger age for the surface. Clearly, something buried the previous landscape, and close inspection reveals the telltale lobate edges of lava flows.
The first stage of Tharsis, then, was the migration of the plume northwards. Volcanism was limited to small, temporary structures expelling thick layers of water-rich ash and expansive fissure lavas, turned on and off as the plume passed by. Where the plume’s progress slowed, the land rose upwards, both from the heat of expansion and from the deep-seated emplacements of magma within the crust itself adding to the bulk of the rock. Those intrusions pushed up through the fractured crust, forcing it apart and solidifying in place, time and time again, forcing the already-high highland to an even greater altitude. To either side of the plume’s track, faults formed or were reactivated, and the movement cut across existing features.
But when the plume approached the Great Dichotomy boundary, it slowed its northward advance and then stalled. With some drift, both northward and southward, the plume has remained under Tharsis up to the present. It created the massive pile of rocks we see, and it is so large that it obscures almost all the evidence of its own earlier movement, much of its early history and also the boundary between northern and southern crusts. Tharsis’s long rise might have started in the Noachian, but the process ran and ran, long after common sense would have expected it to stop.
The answer to our second question is that the plume stalled for a simple dynamic reason: a mantle convection current relies on the sinking of cool mantle material as much as it does on the rising of hot material. Where the crust was especially thick, in the southern hemisphere, it acted as insulation – the heat was sealed underneath and built up. Where it was thin, in the north, the heat radiated away more quickly through the rock, and cooled faster. As the plume crossed the Dichotomy, the mantle sank more quickly, and it trapped the rising plume just beyond the margin of the southern crust.
We’ll inevitably return to this area, as it was – and remains – the most significant indicator of Mars’s internal processes. It is the source of much of the planet’s second atmosphere, the site of the largest rift valley and, due to its height, a place where equatorial glaciers form. Features unambiguously associated with water and ice occur side by side with those associated with fire and impact, and some of the landscapes found here are unique on Mars.
WE NEED TO TALK ABOUT WATER
However and wherever we look on Mars, we find evidence of water. With our satellites, we look for minerals that have been altered by water and turned into clays, and we find that they’re abundant. With our rovers, we look for thin layers of exposed rock that could only have been transported by flowing water and deposited in lakes, and there they are. We see features on the surface that look so uncannily like dried-up riverbeds, feeding downhill into successively larger channels before washing out onto broad plains, that the only possible explanation for them is that they were indeed rivers. We see the results of catastrophic floods that have scoured the land into a chaotic jumble of broken boulders and carved shapes. We see shorelines where water once lapped.
But the water has gone. Where there were rivers, lakes, an ocean, all there is now is dust and ice. These are the bald facts: Mars was once a place where it was warm enough and pressurised enough to allow liquid water to flow at the surface, and now it’s not. The Martian atmosphere of today is too thin and too cold to permit it. Under current conditions, water as ice sublimates directly into water as gas – it’s not hot so we can hardly grace it with the label steam – and it precipitates out again from the vapour phase directly to the solid. We know this happens because we can see seasonal winter growth and summer retreat of the ice caps, north and south. Water and frozen carbon dioxide are transferred from one pole to the other on the ephemeral Martian winds.
And every year, in every summer–winter cycle, Mars loses a little more water as it moves from solid to gas. Vulnerable molecules, let loose in the atmosphere, are broken by the unfiltered sunlight, and the parts drift ever higher, eventually to be driven off into interplanetary space, scrubbed away from the tenuous top of the atmosphere by the constant scour of particles emanating from the Sun. For every year that passes, Mars gets drier and colder.
But in the Noachian, Mars had free water – enough that it would have been possible to sail around the world. The questions are, then, where did that water come from and where did it go?
As ever, we have only partial answers. We know some of the constraints. We know that Mars had to have had significant quantities of water: there are ancient shorelines that suggest times when large bodies of water existed, and the complex systems of rivers and lakes had to mean clouds and rain – a full water cycle of evaporation and precipitation. We know that when this cycle was working, the atmosphere had to be thick enough and warm enough to permit it. We can use our satellites to search the surface not just for minerals that still contain water bound up inside them, but also for ice, both on the surface and buried below it, shrouded by the rock and dust. We can, by crater counting, try to work out when the water flowed and when it stopped. Nothing is absolutely certain, but the story started something like this.
A planet, even one as short-changed on mass as Mars, only needs a tiny percentage of water within its original composition to be effectively rich with the stuff. These things are measured in metres of global equivalent layer: if you averaged the surface to remove all its high and low points, and spread out all the water across it, how deep would it be?
The exact figure is unknowable, but it is guessable. Taking an average from volatile-rich meteorites, perhaps half a per cent of the weight of Mars when it formed was water. If we gathered all of that together at the surface, we’d be kilometres deep in it. It’s not the only version of the story of course; other interpretations present a much drier Mars, where the water layer was less than 100 metres deep, and Mars’s atmosphere was always thin and cold. But a water-rich early Mars is the more likely scenario.
So where did the water go? When Mars melted, most of the volatiles – the water and the gases – boiled out to form the first atmosphere. Most, but not all. The mantle held on to some of its water, inside the minerals that make it up. We have supporting evidence for this: a wet mantle flows more easily and at lower temperatures than one which is devoid of water, allowing for mantle convection and the potential for plumes, such as those that may have formed Tharsis and the Great Dichotomy.
A lot of water was simply lost to space. We know that when a planet has no magnetic field the Sun’s solar wind knocks out molecules from the top of the atmosphere. Over billions of years, an atmosphere can be stripped down to the soil.
Then there was the locking-in of water into minerals in the crust: clays are the obvious sink, but if conditions were right, carbonates are too. Clays are definitely present on much of the surface of Mars – although how thick these layers might be is yet to be determined – but there are also subsurface layers of clay minerals that would have formed through the action of underground water, and have been subsequently exposed at the floors and sides of large craters.
We can add to that all the other hydrated minerals that formed in cracks in the crust, deposited from hot water that percolated throughout the broken subsurface. Volcanic activity increases this, so wherever there was rising magma, there would inevitably have been mineral-rich water sluicing around nearby. Satellites have detected opal – hydrated silica – on the surface. There will be more to find beneath it.
On Mars, there was nothing – is nothing – to bind a loose surface together. Rain would immediately run off, dragging dust and sand downhill, the drops joining to form puddles, which then overflowed to form riverlets and then rivers, always seeking the lowest point. Craters would turn into lakes. Fissures in the ground would become sinkholes. The water might run to the sea, or it might disappear underground again, to be forced back to the surface in a different place, seeping through layers of rock as it went, altering minerals on the way.
So it travelled, round and round. And with each successive epoch there was a little less free water; a slow drying of Mars.
THE NORTHERN OCEAN
At a rough estimate, a third of Mars was covered by open water in the Noachian. Almost all of that water was in the northern hemisphere, in the lowlands beyond the Great Dichotomy. The ocean was vast and deep, and it sustained itself over hundreds of millions of years by a natural cycle of evaporation, condensation, precipitation and collection. In short, weather.
The atmosphere at this time had to be thick enough to support liquid water, otherwise there wouldn’t have been so much of it for so long. Likewise, the temperature had to be locally high enough to prevent freezing, but not so high as to allow boiling. But before we dream of balmy days floating on the seas of Mars, we need to remember that the atmosphere was composed primarily of carbon dioxide and water vapour, with no free oxygen, and was potentially at a pressure of several bars.
As we’re discovering, a warmer world is a wetter world. The storms on Mars would have been tremendous. Great torrents of rain would have descended on the bare ground, eroding the surface, seeping through cracks, running into rivers and ending up back in the sea. Most rain would have fallen within a narrow band to the south of the Dichotomy. Beyond that, in the shadow of the huge rise in elevation, there was very little rain – a desert. And further still, the south pole, and possibly ice.
When we look for evidence of this ocean and the rivers that fed it, of the ancient shorelines where the waves rolled up black beaches of basalt sand, of deltas of deposits laid down as the inflowing water dropped its sediment load, we find it. And more exotic tellings: the marks of impact-generated tsunamis, of broken boulders dragged up slopes by a wall of water and the wash as it receded, and later of stones that might have been carved from solid rock by glaciers, entrained in ice and dropped on the seabed as the icebergs melted. The story is surprising, but everything points to it being true.
When the ancient shoreline was first tentatively identified, it wasn’t long before it was pointed out that portions of the sea level weren’t at the same height. By the laws of nature and gravity, any large body of water finds its own level: add more water and it raises the whole ocean; take it away and sea level drops everywhere. A contemporaneous shoreline cannot be at two different heights.
But across four billion years? Mars has utterly changed during that time: the massive, ever-growing bulk of Tharsis flexed both the crust and the mantle beneath it, even affecting the position of Mars’s axis of rotation. All this could account for the measured deviations in sea level. We’re allowed our ocean.
The water that washed up on the beaches would have been of a different quality to our seas on Earth – no seaweed, no shells, no driftwood, no strand-line of debris. Twice a day there were weak solar tides, not significant lunar ones: Mars’s two moons are mere peanuts and would have been unable to drag water from its basins and pile it up on its coasts.
There would have been storms, though. Waves are driven by the wind, and the wind is driven by differences in atmospheric pressure, which are ultimately caused by sunlight heating the surface more strongly when it’s overhead than when it’s at an oblique angle. The longer the wind blows on the same wavefront, the larger that wave will grow, and there were no shoals, islands or continents for the waves to break on in the northern ocean. The distance over which the wind can blow unobstructed, called the fetch, is essentially limitless in a circular sea. Our own southern ocean is feared by sailors for this reason. The Martian northern ocean would likewise have been a ferocious place.
And that’s without the sporadic but inevitable meteorite impacts. Water, when hit hard enough, becomes as incompressible as rock. It forms the head of the hammer, even as it shatters, not cushioning the blow but transferring it into the seabed.
Imagining one of these events is difficult. All the energy of the meteorite would turn both the water above and rock below into superheated vapour that would want to do nothing else but expand. The shockwave would drive everything before it: the air, the ground and the sea. A circular wall of water would rear up, hundreds of metres high at the start, chasing and being chased by a blast of heat containing molten rock buoyed up on a raft of steam. The energy of the water wave would drop as it expanded out from the impact point. At twice its diameter, it would have less than a third of its original force, but moving so quickly that it would still strike the coast hard enough to break it. Pieces of rock would be torn out by their roots and carried inland. The coastline itself would change, with high ground laid low and riverbeds scoured deep by the surging water. The wave would bore inland until finally it could no longer support its uphill passage. The energy of it would dissipate, its momentum gone.
Then would come the great draining of the land as the water washed back into the sea, down those same rivers, pulling on those same rocks once again. Rip currents would tear at the seabed and gouge at river deposits laid down in less apocalyptic times. As the transient crater collapsed, the water would rush in, quenching the rock and violently turning into steam. There would be secondary waves, and the churning and roaring would continue. But for the most part, it would be over in a day. Until the next time.
The land would have been marked, though: the scars of those impact-created tsunami events were wild and chaotic, and only the long passage of time has softened them but not quite managed to erase them completely. The wave-carried rocks are still there, up on the mountainsides of Mars.
LIFE
We don’t think we’ve found any Martian fossils. Back in 1996, there was a flurry of activity and some bold claims about the nature of the iron oxide particles discovered in one Martian meteorite, but these were – and remain – contentious.
ALH84001 – yes, it’s that rock again, the one that haunts me – dates to around four billion years old, but it was only recently ejected from the surface of Mars, some seventeen million years ago, and even more recently entrained in a block of Antarctic ice, 13,000 years ago. This rock, when recovered from the ice and put under an electron microscope, revealed chains of tiny iron oxide crystals that would, in any terrestrial rock, shout loudly of a biological origin. Our oceans are full of magnetotactic bacteria: simple, single-celled organisms that navigate their position in the water column by sensing the direction and angle of the Earth’s magnetic field using almost identical chains of crystals made of the iron oxide magnetite. Therefore, the investigating scientists thought the iron oxide chains in ALH84001 could be evidence for single-celled Martian life.
Sometimes the simplest answer is the right answer, and sometimes it isn’t. If you try to emulate conditions on Mars four billion years ago, you can make iron crystals spontaneously line up in chains using purely chemical processes. Biology – life – is not required. So we have precisely one Martian rock with tiny lines of iron in it and several explanations as to how they got there. The position is far from satisfactory and I can only apologise for that.
But the absence of evidence is not the evidence of absence. Just because we have not discovered, with our very limited explorations, anything that’s obviously a fossil – a bone, a shell, a tooth – that in no way confirms that they’re not there to be found. Fossil formation relies on several factors all occurring successfully and simultaneously, including burial rates, mineral content and mechanical and chemical events. Something that has lived and died is more likely to be preserved for millions, potentially billions of years if it’s covered over quickly, pressed into a layer of rock and its soft parts subsequently mineralised. We don’t know if the conditions for that ever existed on Mars.
Fossils can be hard to spot, too. Even the large ones. And especially if we don’t exactly know what shape these novel early Martians might have taken. If they were individual single-celled bacteria, then there’s not going to be anything to see until we get more samples back and put them under an electron microscope. But if, one day, a rover or an astronaut digs out a piece of rock – most likely from an old lake or seabed, or from the jumble of river deposits – and reveals something that is unmistakeably a fossil, even though we might not be able to say what it’s a fossil of, ought we to be surprised?
Life is tough and tenuous, persistent and rare. In our world, it’s common and yet precious. It grew from a series of poorly understood chemical reactions into a living, breathing skin that stretches almost pole to pole, from the highest peaks to the most Hadean depths. Where it’s most abundant, it’s riotous. So where did it come from?
Back in 1952, Stanley Miller and Harold Urey conducted a simple but profound experiment. Using basic laboratory equipment, they made an assumption of early terrestrial conditions in terms of gases, added some water, heated it up and put a continuous electric spark across the mixture. They made all twenty amino acids necessary for life. They didn’t make life itself, but life would need all those chemicals. It was a stunning result; it was also the first of many investigations into how complex organic chemicals might be formed – not just on Earth, but everywhere – and how that chemical soup might tip over into self-replicating organisms.
But we struggle with the whole notion of how to draw a line between self-sustaining chemical reactions and very simple life. We don’t even know if there is a line, or whether putting down the markers on one side or other of it will mean we miss something important or end up mislabelling a purely physical process. The very simplest cells are nothing but a series of chemical reactions, with inputs and outputs. What could be the line is the way even the most basic life – stuff we might dismiss as slime – has the ability to self-replicate. Put simply, life creates copies of itself.
We have precisely one example to go on here: carbon-based amino acids, the basis of life on Earth. There’s no guarantee this form of life is mirrored anywhere else, let alone on Mars. What appears to be important is the ability to chemically encode information in a way that’s transferable, and the harnessing of some energy-producing reaction to power that transfer. We do that with DNA and aerobic respiration. But viruses don’t have DNA and yet they manage to replicate and evolve: are they alive? Our arbitrary line says they are, but without living cells to infect and reproduce in, they’re inert, so perhaps not.
We still don’t know exactly what to look for, so what do we mean when we ask, ‘Is there life on Mars?’ In its baldest sense, we’re asking if there’s life that we’d recognise as life. Does it share the characteristics of life as we might understand it? Is it self-organising, self-replicating; does it conform to our idea of what being alive would be, even if it is just slime?
We’ve mapped Mars thoroughly from orbit and sent cameras to the surface. If aliens had done the same to us, their conclusion would be clear – yes, there’s abundant life on Earth. But wherever we look on Mars, we don’t see that abundance. We see instead a profound absence. There’s nothing visible that we’d recognise as living. The conditions that exist on Mars now appear to be inimical to life as we understand it. But perhaps it exists within the ground, below the level where lethal radiation can penetrate. Perhaps it exists where there’s free water – beneath the ice caps or deep under the permafrost. Until we look, we can’t say, but we can test the hypothesis.
Extremophiles are organisms that can withstand conditions that would annihilate other forms of life. Some are adapted for high temperatures. Some for low temperatures, as they have internal fluids rich in natural antifreeze. Some for highly acidic or alkaline conditions. Some for crushing pressures. Some for radiation levels that ought to destroy DNA. Some live in rocks and breathe sulphur. Some live without oxygen. Some live without water.
Given that we can replicate the conditions on the surface of Mars in a laboratory, we can run experiments on our own extremophile organisms to see if they might survive there. The answer is yes: some of them can. Some of them can even survive being inside a simulated meteorite striking a surface at tens of kilometres per second.
These results are, however, disappointing. Given that we know that life might be able to exist on Mars, the fact that we cannot immediately detect it is suspicious. Where are the fields of lichen on the slopes of Olympus Mons? Where are the seasonal blooms of bacteria across the vast northern plains? Where are the odd, calcified structures of algae colonies jostling around the edges of the ice caps? Given that only the very simplest, hardiest life could exist, there would be little to no competition for the (admittedly scant) resources. If an organism can exploit an ecological niche, it’ll usually exploit it to its maximum capacity, and we just don’t see that on Mars. There’s no obvious evidence of life at all.
We might have read it all wrong, though. Martian extremophiles could just be hanging on, millennium after millennium, in an environment too toxic even for them. Or they might never have evolved in the first place, or had evolved for different extremes and then vanished as Mars changed. Perhaps they were too slow to adapt to the ever-stricter climatic conditions. We know about mass extinctions: if Mars once had a tree of life, the changing environment might simply have taken an axe to it.
But what about conditions in the past? Might they have been more conducive to life? We find the very oldest examples of fossilised life on Earth in the preserved relics of a process that carries on today, deep down on the ocean bed. Hydrothermal vents are places where water percolates through cracks in the sea-floor and, having been heated by a subsurface source, re-emerges as a geyser of hot, mineral-laden water. These geysers are often so rich in dissolved minerals that the water is stained dark, which gives them their popular name, ‘black smokers’.
The boiling water gushing out of the seabed into the cold, abyssal depths prompts the mineral load to precipitate out and form craggy columns of rock. Despite the darkness and the heat, isolated and complex ecosystems live and breed and fight and die on those steep slopes.
The exact same processes would have taken place on Noachian Mars. There was a deep ocean. There was water sinking down through a broken crust, heating up, dissolving minerals from the rock it travelled through and rising up again to be ejected forcibly into a cold, dark sea. If life is inevitable, then this is where we would have found it.
The question is: if raw chemicals made the leap to self-replication on Mars, how far along did the journey of life progress before the seas dried up? Mere freezing over of the surface would have mattered little to anything living hundreds or thousands of metres down. But it’s as big a stretch from chemicals to self-replicating life as it is from single-celled organisms to creatures that might swim or crawl or float, let alone hunt, feel or think. Perhaps there was never anything more complicated than slime: when the seas finally sublimated away and the black smokers of Mars were deprived of the water that drove them, that was it. Global mass extinction, life irrevocably snuffed out.
Or perhaps life followed the heat and the water ever further underground, below the thick permafrost layer. Until we dig down, we’ll struggle to know for sure. If we ever do find Martians, that’s most likely where they will be.
There are reasons for our anticipation: we’ve found methane on Mars, when there’s apparently no volcanic reason for it to be there. Methane can be a by-product of organic processes too, and the gas has a very short survival time in an atmosphere – ten to a hundred years – before it breaks down. Something on Mars is creating it. We don’t know, yet, what it might be.
THARSIS RISES
As the Noachian progressed, more material was added to Tharsis, inside and out. The ground it started from was, on average, four kilometres above the northern plains, but the plume’s presence, and the magma it created, pushed it higher still. The line of the Great Dichotomy became buried under Tharsis: not just the surface slope, but the structural depths too. The crust, fractured at a deep and fundamental level, shifted in response. If there was one aspect of Tharsis at this time that was as important as the volcanic eruptions, it was the creation – or reactivation – of crustal fractures.
Fault-bounded valleys, known as graben, are one of Mars’s enduring features. They formed – just like they do on Earth – when kilometre-sized blocks rose or sank in response to tension within the crust. To make a graben, a central block sinks relative to the blocks either side of it, creating a wide, flat-bottomed valley with sharply sloping sides. Most grabens are suspiciously straight, as if designed by some intelligence: imagine bricks resting on a slowly inflating air bed, shifting and jostling together. Noachian Tharsis was both covered and surrounded by these valleys: sometimes complex, sometimes buried, the ground stretched as it gradually pushed upwards. Some parts rose much higher than others and created the ring of ridges around the high plateau of Syria Planum. These ridges remain today: the most significant is the Claritas ridge in the west, and only the north-east is unguarded, as the land slopes down towards the Chryse impact basin on the margin of the Great Dichotomy.
To the south-east of Tharsis lies the huge and deep Argyre crater – a lake at that time, just a little short of four billion years ago. The mid-to-late Noachian Argyre was fed by meltwater from the south pole, which ran through rivers that carved their indelible way through the highlands and ended up collecting in Argyre’s large inland sea. The water appeared periodically to escape the almost 2,000 kilometre-wide basin, making its way north through the Uzboi Vallis and its nested chain of rivers – the Uzboi, the Ladon and the Morava. The Uzboi Vallis potentially started as a graben, but was modified later by the water flowing down through it. The river system itself was augmented by run-off from the burgeoning Tharsis uplands on its way out to the ocean: water always seeks the lowest point.
And it wasn’t just water that poured off the Tharsis region. Lava emerging from fissures in the ground flooded out, covering the floors of the plateaus and filling craters until they were all but buried. Eruptions also belched steam and other gases – carbon dioxide and sulphur – into the air, changing the quality as well as the quantity of the atmosphere. The primordial mix that blanketed Mars straight after it melted had gone. What was there now was on its way to becoming a more nuanced, sparser secondary atmosphere in which the volcanic component was increasingly important for boosting the pressure. Since the core-generated magnetic field had failed, the solar wind had been stripping the atmosphere faster than it was being created below.
It’s difficult to say whether there were any true volcanoes present on Tharsis in the beginning. Eruptions from a single, sustained vent can form mountains from a flat plain in a remarkably short time frame, but these seemed to have occurred later in Tharsis’s long life. If there were examples back then, they’ve been either buried by subsequent lavas or levelled by ground movements. Instead, cracks in the ground spewed out copious floods of lava that flowed almost as freely as the water. The land bucked and rose, cracked and heaved, as far below it a planetary-scale fist of rising mantle pushed up against the underside of the crust.
LAKE ERIDANIA
One of the extraordinary characteristics of Mars is the sheer persistence of its features. When we point to a geological feature on Earth and call it old, it will almost certainly be less than a billion years old. Almost all of the rocks we see are far, far younger than that, even though they’re still tens or hundreds of millions of years old. On Mars, the landscape is a mosaic of the genuinely ancient – billions of years old – with younger deposits on top.
So identifying a series of basins in the southern highlands as the site of a late Noachian inland lake, larger than any lake elsewhere in the solar system, is remarkable, but by Martian standards it is entirely reasonable.
Evidence for the existence of a lake in Eridania is threefold: firstly, there’s its geography. Eridania is a series of ancient circular impact craters nested together to form a linked chain of low points, surrounded by the higher ground of two plateaus, or terrae: Cimmeria and Sirenum. The outflow from the lakes is where a huge river – the Ma’adim Vallis – once met the old crater rim. This low spillover point is where water would have crested the lake’s edge and poured downhill in a torrent, towards the sea. The Ma’adim Vallis traced its way a thousand kilometres to the ocean, where the river spilled out onto the northern plains through the Gusev crater – the site of the Spirit rover landing in 2004.
The second piece of evidence is Eridania’s relationship with the still-visible river channels that extend across the highland surface, in a band that starts at the Great Dichotomy and stretches some thirty degrees further south. Rivers were abundant in this region, but we don’t see them in the area covered by Eridania: the river valleys finish at a much higher level, almost as if they flowed into a large body of existing water.
The third sign is the presence of hundreds of metres of clay deposits within the craters, which can be best interpreted as having been laid down over a significant period of time – hundreds of millions of years – in a deep, replenishing body of water. These deposits vary in thickness across the lake bed, and where they’ve been excavated by more recent craters, the clay appears to go as deep as two kilometres.
The clays came from erosion of the highlands. Water mechanically and chemically broke down the rock of the volcanic hinterland and carried it to the cold, still depths of the lake. Rivers reduced the particles’ size as they flowed downstream – boulders, cobbles and pebbles clashed in the energetic headwaters to form the sands, silts and muds found in mature lowland streams. Once the sediment found its way into the lake, it settled. If there had been filter-feeders present, they would have collectively siphoned their way through vast quantities of water in the hopes of a meal. The regurgitated clay-sized particles would have ended up as pellets, stuck together with mucus. This would have made the settling process much quicker and more effective, but in the absence of sponges, shellfish, worms and other crustacean-like creatures, the clays had to settle naturally, through the water column, onto the lake bed.
Standing on the slowly curving shoreline of Eridania, we might have watched the rain clouds roll in from the north. The far side of the lake could easily be one, two, three hundred kilometres away, but we wouldn’t be able to see it, as anything further than a few kilometres would be below the foreshortened Martian horizon. Waves would climb to improbable heights in the low gravity before looping over and breaking on the headlands and in the bays. The land around us would be necessarily barren and scarred – nothing but layer after layer of dark volcanic rock and impact ejecta, piled up into mountains and eroded back down into battered, worn highlands and filled-in hollows. All around the lake, silver rivers would run down across shallow beaches of black grit and into the water.
The rivers were not the sum total of water entering Eridania, though. Water was percolating through the basement cracks at the bottom of the craters and feeding the lake through underground springs that might have been superior in volume to the surface water. Water also left the lake in the same way, heading underground and into the crust.
However, what we know of Mars is that where there are deep impact basins, there’s often volcanic activity. The twin effects of decreased pressure on the mantle below and increased infiltration of water often meant that rock spontaneously melted at depth and rose up to resurface the floor of the crater, seeping out along the same cracks made by the impact.
Now that the water is long gone and the lake bed is exposed to view, this is what we find: that hydrothermal activity – those black smoker vents pouring out mineral-rich hot water – was present. Not only that, but even though Eridania was flooded to depths of a kilometre, lava was forcing its way out into the lake-bed sediments. That both water and lava can coexist is perhaps counter-intuitive, but once the lava forms a supercooled crust on its outside, it can remain molten inside and still flow. Likewise, a skin of steam on the water side stops the whole lake from quenching the lava. In several places on Eridania’s floor, we’re treated to the sight of once-wet lake-bed clays cut into a chaotic terrain of islands by dense submarine lava flows.
The longevity of the Martian landscape means that it can be difficult to unravel the sequence of events that took place both above and below the surface of the lake. If the water retreated during dry spells, then the lake as a whole would have grown much saltier, and evaporites – water-soluble minerals such as salts, carbonates and gypsum – would have formed at the shoreline. But the craters would have been flooded again when the climate turned. Eruptions increased the temperature of the lake as a whole, but given the latitude and height of the area, ice would also have covered some or all of the surface at various times.
Extreme weather events meant that water periodically overtopped the basin spillway, sending a flood down the Ma’adim Vallis and turning what was a steady if unimpressive river into a torrent that scoured the channel sides and floor and pushed debris down to the vallis’s end, the Gusev crater, and across the northern plains.
That this huge lake ended up a dried husk, slowly and relentlessly drained by a changing climate and a thinning atmosphere, was inevitable. As the shoreline contracted, the clays shrank and subsequent lavas laid down ash and solid flows across the saturated ground. Meteorites smacked into the lake bed and splashed out sloppy ejecta as mudflows, exposing the layers of deposited and sometimes altered sediment in their craters. Dust eventually covered everything and frost gnawed at the exposed rock.
But that there was once such a place is enough. It represents a snapshot of what conditions were like across large areas of Mars as the Noachian slid towards the Hesperian: wet, broken, active, yet always with the spectre of a cold, still, airless future ahead of it.