From the Sahara to the Pacific Rim: How Atlantic-Built Deserts Illuminate Today's Accelerating Aridification

 

The Sahara was built over millions of years by the same physical machinery — a descending Hadley-cell branch anchored by the Azores/North Atlantic Subtropical High, cold Atlantic SSTs and the Canary Current, and a weakened monsoon — that is now drying the Pacific Rim's subtropical margins (US Southwest, Chile/Peru, southern Australia); the difference is timescale, not mechanism.

• Where orbital precession swung the Sahara between desert and "Green Sahara" savanna over ~20,000-year cycles and tectonic/oceanic cooling drove the secular trend over millions of years, anthropogenic greenhouse forcing is now expanding the subtropical dry zones at ~0.1–0.5° latitude per decade and converting episodic "drought" into permanent "aridification" within mere decades.
• The paleo record's clearest warning is nonlinearity: the Green Sahara collapsed within centuries once vegetation and dust feedbacks crossed a threshold, and scientists now explicitly invoke this analogy (Seager's "perpetual drought," Overpeck & Udall's "aridification") for the modern Southwest — abrupt, hard-to-reverse transitions are the central risk.

Key Findings

1. The Sahara is a two-timescale phenomenon. A multi-million-year secular aridification (Tethys closure, mountain uplift, ocean cooling) sets the arid background state; orbital precession (~21–23 kyr, modulated by 41 kyr obliquity and 100 kyr eccentricity) oscillates the desert in and out of "green" phases on top of it.
2. Atlantic systems are central to both timescales. The North Atlantic Subtropical High and Hadley descent create the dry baseline; the cold Canary Current upwelling suppresses coastal convection; AMOC and cross-equatorial SST gradients set the ITCZ/West African Monsoon position that determines how far north the rains reach.
3. The Pacific Rim has near-exact analogues. North Pacific High and South Pacific High (subtropical ridges), the California and Humboldt/Peru currents (cold upwelling coastal deserts), and the Walker circulation/ENSO are the Pacific mirror of the Atlantic mechanisms.
4. Modern aridification is the same mechanism, compressed. Greenhouse warming expands the Hadley cell and shifts subtropical highs poleward, drying the subtropical margins. The post-2000 Southwest megadrought is the driest 22-year period in at least 1,200 years, with human-caused warming responsible for ~42% of the soil-moisture deficit.
5. Nonlinearity is the key danger. The Green Sahara's termination was abrupt because of vegetation-albedo and dust feedbacks; modern drylands show that ~50% of productivity changes are abrupt, tipping-point-like shifts — raising the risk of fast, hard-to-reverse transitions around the Pacific.

Details

PART 1 — How Atlantic systems built the Sahara

Deep-time origins. The conventional view held the Sahara to be ~2–3 million years old, coinciding with Northern Hemisphere glaciation at the onset of the Quaternary. Zhang et al. (2014, Nature 513:401–404) revised this dramatically. Using the Norwegian Earth System Model (NorESM) and the Community Atmosphere Model, they identified the Tortonian stage (~7–11 Ma) of the Late Miocene as the pivotal period: the shrinkage of the Tethys Sea (progenitor of the Mediterranean, Black and Caspian seas) "drastically weakened" the African summer monsoon, allowing arid, desert conditions to expand across North Africa. Crucially, they found the Tethys shrinkage "also enhanced the sensitivity of the African monsoon to orbital forcing, which subsequently became the major driver of Sahara extent fluctuations" — establishing the two-timescale system. Independent evidence supports an old, orbitally-paced Sahara: a Nature Geoscience study (2022) found Saharan dust pulses entering the North Atlantic since at least 11 Ma, "a result of astronomically paced cycles between arid and humid conditions," while Canary Islands paleosols record African dust delivery for at least 4.8 million years. Tectonic uplift (Tibetan Plateau, Atlas Mountains) and the secular cooling of the Atlantic and global ocean provided additional forcing.

The Atlantic Ocean's specific roles. Three Atlantic mechanisms govern North African aridity:

• The North Atlantic Subtropical (Azores) High and Hadley descent. The descending branch of the Hadley cell at ~15–30°N produces persistent subsidence, adiabatic warming and drying — the fundamental reason subtropical deserts sit where they do. This is the same blocking-anticyclone dynamic at the core of the prior discussion of the North Pacific and Azores Highs.
• The cold Canary Current and coastal upwelling. The Canary Current, the eastern limb of the North Atlantic subtropical gyre, brings cool water south along NW Africa; wind-driven upwelling (year-round north of Cap Blanc/Ras Nouadhibou) further chills coastal SSTs. Cold water suppresses evaporation and stabilizes the lower atmosphere, suppressing moist convection and limiting precipitation to under ~100 mm/yr along much of the Saharan west coast — the textbook cold-current coastal-desert mechanism. (Note: trends in Canary upwelling intensity under modern warming are contested, with some records showing ~1.2 °C of coastal cooling at Cape Ghir over the last century but no coherent region-wide intensification.)
• AMOC and ITCZ control. The Atlantic Meridional Overturning Circulation transports ~0.5 PW of heat into the Northern Hemisphere; this cross-equatorial energy transport keeps the annual-mean ITCZ north of the equator. A vigorous AMOC pulls the ITCZ northward; a weakened AMOC (or a cold North Atlantic) shifts it south, weakening the West African Monsoon. Zhang & Delworth (2005) established that AMOC strength controls the annual-mean tropical Atlantic ITCZ position, with a weakening producing a southward ITCZ shift. North Atlantic SST gradients thus directly meter moisture delivery to the Sahel and Sahara.

The orbital "Green Sahara" story. Superimposed on the arid baseline, precession of the equinoxes shifts the timing of perihelion and hence Northern Hemisphere summer insolation. During the early-to-mid Holocene, NH summer insolation was substantially higher than today — by roughly 7% in relative terms, equivalent to anomalies of more than 25 W/m² at low-to-mid northern latitudes (the maximum occurred ~10–7 ka). This extra summer heating intensified the land–sea thermal contrast, deepened the Saharan heat low, and pulled the West African Monsoon and ITCZ far northward. As deMenocal & Tierney (2012, Nature Education) describe, precession "controls the strength and northward penetration of the monsoonal rains."

The magnitude of the transformation was extraordinary. During the African Humid Period (~11,000–5,000 years ago; broader wet phase ~14,800–5,500 BP in the marine dust records), the Sahara was savanna, woodland and lakes. Lake Mega-Chad attained an area of ~361,000 km² and a depth of up to ~160 m (Armitage, Bristow & Drake 2015, PNAS; Drake & Bristow 2006, The Holocene) — larger than all the North American Great Lakes combined, and the largest freshwater lake in Africa, draining south via the Mayo Kebbi outlet into the Benue/Niger system. Today's Lake Chad is a tiny remnant (~1,500 km² or less), and its northern sub-basin, the Bodélé Depression, is the planet's single largest dust source. Proxy data indicate the Green Sahara extended to ~31°N (Tierney et al. 2017, Science Advances), and modeling with the classic Kutzbach & Liu (1997) result shows that a ~7% increase in summer radiation yields at least a 17% increase in African monsoonal rainfall — and up to ~50% when ocean feedbacks are included.

Feedbacks and abrupt collapse. Orbital forcing alone cannot reproduce the observed greening; climate models systematically fail to push rainfall north of ~24°N without feedbacks. The Charney (1975, QJRMS; Charney, Stone & Quirk 1975, Science) vegetation-albedo mechanism is central: vegetation lowers surface albedo (e.g., ~0.20 vs. ~0.33 for desert), increasing absorbed solar energy, surface heating, convection and rainfall, which sustains more vegetation — a positive feedback supporting multiple stable states. Dust feedbacks are equally important. Pausata, Messori & Zhang (2016, Earth and Planetary Science Letters) showed that an ~80% reduction in atmospheric Saharan dust (consistent with the 60–80% dust-flux reductions inferred from proxies) under a vegetated Sahara pushed the monsoon's northern boundary to ~30°N — roughly 1,500 km north of the control — with the dust reduction alone contributing about one-third (~500 km) of the total northward shift, a WAM displacement "twice as large as the shift due to orbital forcing only." Matching the proxy record required both vegetation and dust feedbacks.

When summer insolation declined after the mid-Holocene, these feedbacks ran in reverse. Tierney & deMenocal (2013, Science 342:843–846), using a high-resolution Horn of Africa leaf-wax record, found the AHP termination (~5,500 years ago) occurred "within centuries rather than over millennia," underscoring "the nonlinearity of the region's hydroclimate"; deMenocal et al. (2000, ODP Site 658C off Mauritania) similarly documented onset and termination "within decades to centuries" (often cited as ~100–200 years) in response to gradual orbital forcing — the classic nonlinear-threshold evidence. There is genuine debate: Shanahan et al. (2015, Nature Geoscience) argue the termination was time-transgressive — spatially variable, migrating to progressively lower latitudes — rather than synchronous everywhere.

How the two timescales weave together. The secular trend (Tethys retreat, uplift, ocean cooling) progressively lowered the background moisture state and increased the system's sensitivity to orbital forcing. Once the arid baseline existed, precession became "the major driver of Sahara extent fluctuations" (Zhang et al. 2014). The Sahara is therefore best understood as a slow tectonic/oceanographic drying trend with fast orbital oscillations riding on top, both amplified by land-surface feedbacks that can flip the system between stable states.

PART 2 — The Pacific Rim today: same machinery, different ocean

The deserts and their mechanisms. The Pacific basin reproduces the Atlantic pattern with striking fidelity:

• Western North America. The North Pacific High is the Pacific analogue of the Azores High; in summer it migrates north and steers storm tracks away from California, producing the Mediterranean dry-summer regime. The cold California Current (eastern limb of the North Pacific gyre) creates cool, foggy, arid coastal conditions — the direct analogue of the Canary Current. The Great Basin, Sonoran, Mojave and Chihuahuan deserts sit under this regime, reinforced by Sierra Nevada/Coast Range rain shadows.
• Western South America. The Atacama is arguably Earth's oldest and driest non-polar desert, driven by the South Pacific High, the cold Humboldt/Peru Current and upwelling, and the Andean rain shadow (the "double rain shadow"). Its hyperarid core receives <1–2 mm/yr. Dating is being revised deeper in time: Hartley et al. placed the switch to hyperaridity ~3 Ma (global cooling plus Central American Seaway closure enhancing Humboldt upwelling), while a 2026 Nature Communications cosmogenic-nuclide study (Dunai and colleagues) argues extreme aridity in the hyperarid core began ~45 Ma, triggered by global cooling after the Early Eocene Climate Optimum, with the Humboldt Current and Andean uplift later intensifying and expanding the dry zone — a deep-time analogue to the Saharan secular trend.
• Australia and East Asia. Southern Australia sits under the subtropical ridge, which blocks rain-bearing fronts when positioned poleward in the warm season; ~three-quarters of the continent is arid or semi-arid. The Gobi is reported to be expanding on its southern (Chinese) edge — ~3,600 km² of grassland overtaken per year in some accounts — but attribution is genuinely mixed between land use (overgrazing, water depletion) and climate, and one satellite study (2000–2012) found net contraction driven by precipitation variability, echoing the 1980s–90s controversy over the Sahara's southern boundary.

The Pacific overlays — Walker circulation and ENSO. Unlike the Atlantic, the Pacific is dominated by the Walker circulation and ENSO. La Niña-like states tend to dry the Southwest; El Niño tends to wet it. Richard Seager has noted that recent decades' La Niña-leaning pattern contributed to Southwestern drought, with a "background steady drying of the region ... occurring due to rising greenhouse gases, and variability moving around that" — directly parallel to the Sahara's "secular trend plus oscillation" structure, with ENSO playing the role precession plays for the Sahara, but on an interannual-to-decadal rather than orbital timescale.

Megadrought vs. aridification. This conceptual distinction is crucial. Williams et al. (2020, Science 368:314–318) showed the 2000–2018 Southwest drought was "the second driest 19-year period since 800 CE," with anthropogenic trends in temperature, humidity and precipitation accounting for "47% (model interquartiles of 35 to 105%) of the 2000–2018 drought severity." The 2022 update (Williams, Cook & Smerdon, Nature Climate Change 12:232–234) found 2000–2021 to be the driest 22-year period since at least 800 CE, with "human-caused climate change ... responsible for about 42% of the soil moisture deficit since 2000." But a megadrought is an episodic event; Overpeck & Udall (2020, PNAS) argue the correct frame is "aridification" — a more permanent shift to a drier climate-hydrology state, driven substantially by temperature (warmer "thirsty" air increases evapotranspiration and dries soils even when precipitation is steady). They document declining Colorado and Rio Grande flows — rivers supplying >40 million people — and explicitly call for replacing "drought" with "aridification" in scientific and policy language.

Hadley expansion reaches the Pacific Rim. The same Hadley-cell expansion that fixes the Sahara's latitude is now widening, pushing the subtropical dry zones poleward. Per IPCC AR6 (citing Grise & Davis 2020 and Studholme & Gulev 2018), "the annual mean Hadley cell extent has shifted poleward at an approximate rate of 0.1°–0.5° latitude per decade over the last about 40 years." IPCC AR6 WGII concluded greenhouse forcing has contributed to drying in dry-summer climates including "the Mediterranean, southwestern Australia, southwestern South America, South Africa and western North America (medium to high confidence)" — essentially every Pacific- and Atlantic-rim Mediterranean-climate region. Under the most extreme scenario, subtropical highs are projected to migrate poleward by ~1.5° by century's end, consistent with Hadley-cell expansion.

PART 3 — The acceleration thesis

The same physics, compressed from millions of years to decades. The mechanism that built the Sahara — Hadley descent under a subtropical high, reinforced by cold eastern-boundary currents and a suppressed monsoon — is now being driven directly by greenhouse warming on the Pacific Rim. As the planet warms and the equator-to-pole temperature gradient changes, the Hadley cell expands poleward and the subtropical dry zones expand with it. Seager et al. (2007, Science 316:1181–1184) wrote that the Southwest "will dry in the 21st century" as "part of a general drying of the subtropics and poleward expansion of the subtropical dry zones," and that "the transition to a more arid climate should already be under way" — with "the levels of aridity of the recent multiyear drought or the Dust Bowl ... becom[ing] the new climatology of the American Southwest within a time frame of years to decades." The Lamont-Doherty press framing was blunt: "a perpetual drought." Seager & Vecchi (2010, PNAS) sharpened the timing, anticipating that "anthropogenic aridity will be as large in magnitude as the droughts caused by natural decadal variability ... by around 2050," with the Southwest unlikely to see a return of the prolonged moist conditions of the 20th century.

Comparing rates — the heart of the matter. The Saharan secular aridification unfolded over ~7–11 million years (Tethys retreat, uplift, ocean cooling). The orbital green-desert oscillations cycle over ~20,000 years, with the most recent collapse compressed by feedbacks into perhaps 100–200 years at the most responsive sites. Modern anthropogenic subtropical expansion is running at ~0.1–0.5° latitude per decade (Grise & Davis 2020 narrow the robust observed rate to ≤0.5°/decade, within the model range). In other words, greenhouse forcing is delivering a poleward dry-zone migration in decades that orbital cycling would take millennia to accomplish and that the tectonic/oceanographic trend took millions of years to build. The forcing is being applied orders of magnitude faster than any natural analogue in the relevant records.

Feedbacks and tipping points. The paleo lesson is that vegetation-albedo and soil-moisture feedbacks can turn gradual forcing into abrupt regime shifts. Berdugo et al. (2022, PNAS 119:e2123393119), assessing 41,830 dryland sites over 2000–2019, found that "50% of all dryland ecosystems exhibiting gains or losses of NDVI are characterized by abrupt positive/negative temporal dynamics" — the signature of tipping-point behavior — and that human pressure reduces dryland resilience and impedes recovery. Zeng & Yoon (2009, GRL) projected expansion of the world's warm deserts (Sahara, Kalahari, Gobi, Great Sandy) under global warming specifically via the vegetation-albedo (Charney) feedback. The Green Sahara's abrupt collapse is the cautionary template: once feedbacks engage, a slowly-forced system can flip fast and prove difficult to reverse — exactly the risk for Pacific-rim drylands now being forced far faster than the orbital pacing under which those feedbacks historically operated.

Do scientists explicitly draw the Sahara analogy? Yes, though carefully. The "perpetual/permanent drought" and "aridification" framings (Seager; Overpeck & Udall) are functionally Sahara-analogues — they describe the Southwest acquiring a permanently desert-like mean state via the same subtropical-expansion physics that fixes the Sahara. Educational and review sources explicitly group the Sahara, the Sonoran/Chihuahuan deserts, and the southern Andes/Patagonia as the regions facing climate-driven precipitation decline. Scientists generally avoid claiming the Southwest will literally become hyperarid like the Atacama or Saharan core, but the mechanistic identity (Hadley descent + subtropical high + cold current + poleward expansion) is widely accepted.

Uncertainties and debates. (1) Some dynamical metrics put the observed Hadley expansion near or above 1°/decade — substantially faster than greenhouse-gas-forced models (~0.1–0.3°/decade) — an unresolved model–observation gap; Grise & Davis (2020) attribute much of the recent Northern Hemisphere expansion to natural variability rather than forcing. (2) California's precipitation future is genuinely uncertain: it sits between an expanding subtropical high (drying) and a possibly southeastward-extending Aleutian low plus intensifying atmospheric rivers (wetting), with large intermodel spread. (3) AHP-termination synchrony (abrupt-everywhere vs. time-transgressive) remains debated. (4) Gobi trends are contested (expansion vs. contraction; land use vs. climate). (5) IPCC AR6 WGI assigns only medium confidence that greenhouse forcing caused the observed Hadley expansion and low confidence in how that expansion translates into subtropical land drying — an essential honesty check on the strongest versions of the acceleration thesis.

Recommendations

• For framing the synthesis: Treat the Sahara as the "completed experiment" and the Pacific Rim as the "experiment in progress." The rigorous throughline is the subtropical-high/Hadley-descent/cold-current triad plus monsoon suppression, with vegetation-albedo and dust feedbacks as the amplifier, and orbital vs. anthropogenic forcing as the speed contrast (millions of years → millennia → decades).
• Lead with the rate comparison, because it is the most defensible and striking claim: the mechanism is identical and uncontroversial; the novelty is the speed of forcing, which has no natural analogue.
• Watch these benchmarks that would change the assessment: (1) reconciliation of Hadley-edge trends between reanalyses and models — convergence would raise attribution confidence; (2) Colorado River flow and Southwest soil-moisture trajectories against the Williams et al. tree-ring baseline; (3) AMOC strength — a major weakening would shift the Atlantic ITCZ south with global rainfall consequences, the modern echo of the AHP-termination mechanism; (4) detection of abrupt (vs. gradual) vegetation shifts in Pacific-rim drylands as an early-warning tipping signal.
• For honest communication: Always distinguish "drought" (episodic, recoverable) from "aridification" (a shifted mean state). The policy implication — that the new dry baseline will not be reversed by a few wet years — follows directly from the paleo evidence of stable desert states held in place by feedbacks.

Caveats

• Mechanistic analogy is not quantitative equivalence: the Southwest becoming "more arid" is not the same as becoming the hyperarid Saharan or Atacama core. The claim is about direction and mechanism, not a literal future Sahara in North America.
• Paleo rates (especially AHP-termination abruptness) carry dating and proxy uncertainties; "within centuries" is well-supported at specific sites (Horn of Africa, Cap Blanc) but spatial synchrony is debated (Shanahan et al. 2015).
• Attribution of the modern Hadley expansion is partly to natural variability, and the model–observation rate gap is real and unresolved; IPCC confidence on the land-drying link is only low-to-medium.
• ENSO and decadal Pacific variability can mask or temporarily reverse the underlying greenhouse trend for years to decades, complicating detection and making single wet or dry years poor evidence either way.
• Desert-age estimates (Sahara, Atacama) are actively being revised deeper in time by new dust and cosmogenic-nuclide records; specific onset dates should be read as best-current-estimates, not settled values.