How Was the Sahara Desert Formed?

The Sahara Desert formed primarily because an ancient sea vanished. Climate model simulations point to the shrinkage of the Tethys Sea during the Late Miocene, roughly seven to eleven million years ago, as the pivotal event that weakened the African summer monsoon and allowed arid conditions to spread across North Africa. But the Sahara’s story is far more complicated than a single drying event. The desert has expanded and contracted dozens of times, sometimes flipping from lush grassland to barren sand within a few centuries, driven by shifts in Earth’s orbit, vegetation feedbacks, and ocean currents that researchers are still working to untangle.

The Tethys Sea and the Birth of Aridity

North Africa was not always dry. For most of the past hundred million years, the region sat at latitudes and under atmospheric conditions that kept it considerably wetter. The transformation into a desert traces back to a period geologists call the Tortonian stage of the Late Miocene, roughly seven to eleven million years ago. During this window, the Tethys Sea, a vast body of water that once separated Africa from Eurasia, was steadily shrinking as tectonic plates pushed the two continents closer together. Climate modeling shows that this shrinkage drastically weakened the African summer monsoon, the seasonal rain engine that had kept North Africa green. With less moisture being pulled inland from the sea, arid conditions expanded across the continent, and the Sahara as a recognizable desert began to take shape.

The Tethys connection did something else, too. It made the monsoon far more sensitive to changes in Earth’s orbit. Once the sea had shrunk enough, even small wobbles in orbital geometry could push the monsoon stronger or weaker, causing the Sahara’s boundaries to pulse in and out. That orbital sensitivity became the dominant driver of the desert’s extent for millions of years afterward.

1PubMed. Aridification of the Sahara desert caused by Tethys Sea shrinkage during the Late Miocene

North Africa Before the Desert

Long before the Sahara existed, the land underneath it experienced radically different climates. During the Late Ordovician period, roughly 445 million years ago, ice sheets expanded northward across what is now North Africa, depositing thick layers of glacial sediment. Outcrop studies in the Al Kufrah Basin in Libya reveal sand-dominated glacial formations and deformed braidplain deposits from that era, evidence that the region once sat under a continental ice cap.

2Sedimentary Geology. Glaciation and deglaciation of the Libyan Desert: The Late Ordovician record

Between that ancient glaciation and the Miocene drying, the region cycled through tropical forests, shallow seas, and savannah landscapes. Dinosaur fossils, petrified wood, and marine sediments found throughout the Sahara all testify to these earlier incarnations. The point is worth dwelling on: the Sahara is not a permanent feature of the planet. It is a relatively recent geological development, and its land has worn many different faces.

Orbital Wobbles and the Sahara’s Green Phases

Earth’s orbit around the Sun is not perfectly stable. It wobbles on cycles of roughly 20,000 to 100,000 years, changing how much solar energy different parts of the planet receive during different seasons. These orbital precession cycles have been the main pacemaker of Saharan climate for at least the past several million years. When the orbital geometry increases summer sunshine over North Africa, the monsoon strengthens, rain pushes northward, and the desert greens. When the geometry shifts back, the monsoon retreats, and the sand returns.

High-resolution geochemical records from the Eastern Mediterranean spanning the past 5.2 million years capture these oscillations in detail. The records show that around 3.2 million years ago, the system intensified: fluvial sediment inputs during green Sahara periods roughly doubled, even though the underlying monsoon rain intensity stayed about the same. Something in the landscape changed so that each green phase left a bigger environmental footprint. Researchers interpret this as a shift in how the land surface responded to monsoon rainfall, likely linked to changes in vegetation cover and soil dynamics.

3Communications Earth & Environment. Organic carbon burial in Mediterranean sapropels intensified during Green Sahara Periods since 3.2 Myr ago

The African Humid Period

The most recent and best-studied green phase is the African Humid Period, which lasted from roughly 15,000 to 5,000 years ago. During this time, the Sahara was barely recognizable as a desert. Grasslands, lakes, and river systems covered much of what is now barren sand and rock. Lake Mega-Chad, the largest pluvial lake in Africa, reached its peak extent by about 11,500 years ago and held that level until around 5,000 years ago. The lake was enormous, dwarfing the modern remnant of Lake Chad by orders of magnitude.

4PubMed Central. West African monsoon dynamics inferred from abrupt fluctuations of Lake Mega-Chad

The mountains of the central Sahara received especially heavy rainfall during this period. Mid-Holocene precipitation records from the Tibesti mountains, which rise above 3,000 meters in northern Chad, show that the massif received at least ten times more rainfall than the surrounding plains. Surprisingly, much of this moisture came from north-easterly Mediterranean winds rather than the West African monsoon pushing up from the south. Strong orographic uplift of these moist air masses as they hit the mountains created conditions far wetter than anything the flat desert around them experienced.

5PubMed Central. Mid-Holocene extreme precipitation in the Tibesti, Central Sahara

Ancient waterways crisscrossed the green Sahara, forming a connected network of lakes, rivers, and inland deltas. Analysis of the region’s zoogeography shows that more animal species dispersed across the Sahara via these waterways than along the Nile corridor, and many of these species were aquatic, confirming that the linked waterway system was extensive enough to support fish and other freshwater life across what is now one of the driest places on Earth. Human populations followed these waterways too, with Nilo-Saharan speakers hunting aquatic fauna spreading across the southern Sahara while other groups hunting savannah animals moved southward.

6Proceedings of the National Academy of Sciences. Ancient watercourses and biogeography of the Sahara explain the peopling of the desert

How Vegetation Amplified the Swings

Orbital forcing alone cannot explain how rapidly the Sahara greened or dried. The monsoon strengthens gradually as orbital geometry shifts, but the landscape transitions happened within decades to centuries, far too fast for a purely astronomical explanation. The missing piece is vegetation feedback. When plants cover the ground, they darken the surface, absorbing more solar energy than bare sand would. This lower albedo heats the land surface, draws in more moist air, and produces more rain. The additional rain supports more plants, which further darken the surface, creating a self-reinforcing loop.

Modeling studies estimate that vegetation feeds back to precipitation by a factor of two to three when plant cover exceeds about 60 percent, amplifying whatever rainfall the orbital shift initially provides.

7Climate of the Past. Comparison of the green-to-desert Sahara transitions between the Holocene and the last interglacial

The feedback works in both directions. As vegetation thickens, it increases evapotranspiration, the process by which plants pull water from the soil and release it as vapor into the atmosphere. This extra moisture recycles as local rainfall, so a greener landscape essentially manufactures some of its own rain. Conversely, once vegetation starts dying off, the surface brightens, less moisture recycles, and the drying accelerates. The same feedback loop that helped green the Sahara also helped destroy it.

8One Earth. The Greening of the Sahara: Past Changes and Future Implications

The Abrupt Return of the Desert

The end of the African Humid Period around 5,000 to 5,500 years ago was strikingly sudden. Marine sediment records off the West African coast show a well-defined period of low dust influx between about 14,800 and 5,500 years ago, when the Sahara was nearly completely vegetated. The onset and termination of this period both happened within decades to centuries, and both transitions occurred when summer insolation crossed nearly the same threshold value, about 4.2 percent higher than today’s levels.

9Quaternary Science Reviews. Abrupt onset and termination of the African Humid Period: rapid climate responses to gradual insolation forcing

The picture is a bit more nuanced than a single switch flipping across the whole continent, though. A synthesis of hydrologic records from across Africa shows that while the termination was locally abrupt at any given site, it occurred progressively later at lower latitudes. The northern margins of the Sahara dried first, and the southern edges held on longer as the tropical rain belt gradually migrated southward. So the desert did not snap into existence all at once; it rolled southward over a few thousand years, even though each individual location experienced its own transition as a sudden collapse.

10Nature Geoscience. The time-transgressive termination of the African Humid Period

Did Humans Help Create the Desert?

A persistent idea holds that early pastoralists, by overgrazing the Sahara’s grasslands, helped tip the ecosystem into desert. The logic sounds intuitive: herders stripped the vegetation, soil dried out, and the feedback loop did the rest. But modeling work that tested this hypothesis found the opposite. The dominant collapse of the green Sahara occurred 500 to 1,000 years after the period when the landscape was most sensitive to human disturbance. Rather than accelerating desertification, pastoralism appears to have actively delayed the region’s environmental deterioration, possibly because managed grazing maintained some vegetation cover longer than an unmanaged landscape would have kept it.

11Nature Communications. Pastoralism may have delayed the end of the green Sahara

This does not mean humans had zero effect on the landscape, but it does push back hard against the narrative that herders caused or even meaningfully sped up the Sahara’s formation. The orbital and vegetation-feedback drivers were simply operating on a much larger scale than anything early pastoral societies could have influenced.

What the Ice Ages Did to the Sahara

The Sahara’s behavior during glacial periods, when massive ice sheets covered the Northern Hemisphere, was different from what you might expect. A 140,000-year climate simulation shows that during the most recent glacial period, roughly 70,000 to 15,000 years ago, the ice sheets and lowered greenhouse gas levels actually pushed the northern limit of summer monsoon rains only slightly southward. The desert did not expand as dramatically as the “cold equals dry” assumption suggests. At the same time, the winter Mediterranean storm track intensified, delivering more rain to the northern Sahara. The result was a complex mosaic of wet and dry zones rather than a uniform expansion of sand.

12PubMed Central. African climate response to orbital and glacial forcing in 140,000-y simulation with implications for early modern human environments

Mountain Refuges and Underground Water

Even at its driest, the Sahara has never been completely lifeless. Isolated mountain ranges like the Aïr in Niger and the Tibesti and Ennedi in Chad harbor pockets of vegetation that survived the worst arid phases. Studies of the Aïr mountains found that boulders and gullies provided physical shelters where relict plant species persisted, protected from extreme aridity and likely from grazing animals. These mountain habitats acted as biological time capsules, preserving species from wetter eras that could potentially recolonize surrounding areas if conditions improved.

13Journal of Arid Environments. Elevation and local refuges ensure persistence of mountain specific vegetation in the Nigerien Sahara

Beneath the surface, ancient water systems also persist. The Nubian Sandstone Aquifer System, one of the largest fossil water reserves in the world, underlies parts of Chad, Libya, Egypt, and Sudan. Modern recharge does still occur in northern Chad, but only in very limited areas and for brief periods. In the southern Ennedi mountains, diffuse and concentrated recharge happens during August; farther north, water reaches the aquifer only through seasonal flooding of dry riverbeds, or wadis, and the extent of these flooded areas varies wildly from year to year.

14Hydrogeology Journal. Infiltration and recharge dynamics in the Nubian Sandstone Aquifer System of northern Chad

The Bodélé Depression and the World’s Biggest Dust Source

The Sahara is the largest source of airborne mineral dust on Earth, and the single most productive spot within it is the Bodélé Depression in northern Chad. This dried-up lakebed, a remnant of the much larger Lake Mega-Chad, produces roughly half of all the mineral aerosols emitted from the Sahara. On average, about a hundred dust storms a year originate there, launching plumes that can be tracked by satellite across thousands of kilometers.

15PubMed Central. Dust as a tipping element: the Bodele Depression, Chad
16Eos, Transactions American Geophysical Union. Active sand dunes are largest dust source in the Sahara Desert

The depression also contains some of the fastest-moving barchan dunes on the planet, crescent-shaped sand formations that migrate steadily to the southwest, continuously exposing fresh diatomite sediment that the wind picks up and carries away.

17Frontiers in Environmental Science. Assessment of Aeolian Activity in the Bodélé Depression, Chad

Saharan Dust Feeds the Amazon

The dust that leaves the Sahara does not just disappear into the atmosphere. A significant portion of it travels across the Atlantic Ocean and lands in the Amazon Basin, where it plays a surprisingly important ecological role. The Amazon rainforest grows on nutrient-poor soils, and phosphorus, which is essential for plant growth, is chronically scarce. Satellite measurements over a seven-year period estimated that roughly 28 million metric tons of Saharan dust are deposited in the Amazon Basin annually, delivering about 22,000 metric tons of phosphorus per year. That imported phosphorus roughly matches the amount the basin loses through river drainage, suggesting that Saharan dust effectively prevents the Amazon from running out of a critical nutrient over timescales of decades to centuries.

18Geophysical Research Letters. The fertilizing role of African dust in the Amazon rainforest: A first multiyear assessment based on data from Cloud‐Aerosol Lidar and Infrared Pathfinder Satellite Observations

Phosphorus is not the only nutrient making the journey. Saharan dust also delivers soluble iron, another micronutrient that limits plant growth in parts of the Amazon. The deposited iron is bioavailable, meaning plants, bacteria, and fungi can actually use it. Seasonal dust deposition enriches both the topsoil and the canopy surface, supporting microbial communities and benefiting species that are especially efficient at absorbing minerals from the air.

19Atmospheric Chemistry and Physics. Soluble iron nutrients in Saharan dust over the central Amazon rainforest

Saharan Dust and Atlantic Hurricanes

Saharan dust plumes also interact with tropical weather systems over the Atlantic in ways that researchers are still sorting out. The Saharan Air Layer, a mass of hot, dry, dust-laden air that streams off the continent at altitudes of roughly one to three miles, tends to suppress deep convection over the tropical ocean. This dry, stable layer can choke developing tropical disturbances before they grow into organized storms.

20Atmosphere. On Saharan Air Layer Stability and Suppression of Convection over the Northern Tropical Atlantic

The relationship with hurricanes that do form, however, is less straightforward. Modeling work suggests that a strengthened dust plume can shift tropical cyclone activity westward, pushing storms from the cooler central Atlantic toward the warmer waters off the North American coast, where they have more energy available to intensify. The implication is counterintuitive: while dust suppresses storm formation in the open ocean, it may increase the number of storms that actually make landfall in North America, and particularly the fraction that reach major hurricane strength.

21Atmospheric and Oceanic Science Letters. Impact of Saharan dust on landfalling North Atlantic tropical cyclones over North America in September

Reading the Desert’s Past in Ocean Mud

Much of what scientists know about the Sahara’s long history comes not from the desert itself but from the ocean floor beside it. As conditions on land shift between wet and dry, the amount of dust blown offshore changes dramatically, and that dust accumulates in layers on the seafloor. Sediment cores drilled from the Atlantic off the coast of West Africa preserve a continuous record of Saharan dust deposition stretching back hundreds of thousands of years. These records serve as a proxy for conditions on land: thick dust layers correspond to dry periods, thin layers to green ones. A core covering the past 240,000 years confirms that the pattern is dominated by monsoon variability tied to orbital precession, with each major green-to-desert transition leaving a clear signature in the geochemistry of the sediment.

22PubMed Central. Monsoon-driven Saharan dust variability over the past 240,000 years

These marine records are what allowed researchers to pin down the abruptness of past transitions and to establish that the Sahara has been cycling between green and desert states for millions of years. They also provide the baseline against which modern changes can be measured, a useful thing to have when the question of whether the Sahara might green again inevitably comes up. The orbital geometry that would push the monsoon northward again is thousands of years away, but the records make clear that the desert is not a permanent fixture. It is one phase of a cycle that will, on geological timescales, eventually swing the other way.