In Simple Terms
Winds carry dust from North Africa across the ocean, and some of it falls on South America. The dust brings phosphorus, a nutrient that plants need, but scientists still disagree about how much reaches the Amazon. Newer estimates suggest less dust than one earlier satellite estimate, and researchers still need measurements deep inside the forest to know what arrives there.
A desert-to-rainforest connection
Every year, winds sweep clouds of mineral particles out of North Africa and send them on a long journey across the Atlantic. Some of the dust reaches South America and settles over parts of the Amazon Basin, carrying phosphorus with it. The amount may seem small compared with the vast rainforest, but phosphorus matters in soils that lack many nutrients. This airborne journey links deserts and tropical forests in a striking way.
How do you measure dust you cannot collect?
There is no practical way to catch every dust particle that lands across the Amazon Basin. Instead, researchers combine satellite observations, air-monitoring stations, and computer models. Each approach captures a different part of the story: satellites track dust layers as they travel, ground-based filters show what actually reaches particular sites, and models use those observations to estimate how much dust settles across a much wider area.
In a study led by Hongbin Yu and colleagues, researchers used data collected by the lidar instrument aboard the CALIPSO satellite from 2007 to 2013. The instrument sends pulses of light into the atmosphere and analyzes the light reflected back by particles, revealing whether dust layers are present and how high they are. The team combined the satellite observations with wind data to estimate how much dust entered the region, then compared that amount with how much left. They treated the difference as dust deposited along the way.
The team estimated that about 182 million tons of dust crossed the western coast of Africa each year, and that dust settling over the Amazon carried roughly 22,000 tons of phosphorus. Their estimate for annual dust deposition across the basin was 27.7 million tons, but the uncertainty range was wide: 8 to 48 million tons. In other words, the headline figure was not a precise direct measurement. It was a central estimate surrounded by a range that included much lower totals.
Yu and colleagues published their findings in Geophysical Research Letters in 2015. In a separate study published in Remote Sensing of Environment, the team used CALIPSO’s seven-year record to examine dust transport across the Atlantic. The amounts varied from year to year during the observation period, and the record was too short to reveal a long-term trend, such as whether deposition had increased or declined over several decades.
Filters in Cayenne reshape the estimate
A different kind of test came from Cayenne, the capital of French Guiana, on the northeastern coast of South America. The local air-quality agency, ATMO Guyane, monitored airborne particles from 2002 to 2017. In 2014, the agency began using filters at an elevated, forested site to collect particles from the air. A filter can capture sea salt from the Atlantic as well as mineral dust that has crossed the ocean from Africa, so researchers first had to separate the different materials before estimating how much Saharan dust was present.
Joseph Prospero and colleagues at the University of Miami washed the sea salts off used filters, then heated the remaining material to isolate the mineral dust. The samples offered a way to check whether model predictions matched what was actually in the air. The team used NASA’s MERRA-2 model and found that its account of seasonal changes was close to the pattern measured in Cayenne.
The researchers then used the model to estimate deposition across a large area of the Amazon Basin. When they included the southernmost strip covered by their analysis, the total reached 10.4 million tons per year. But they were less certain that all the particles in that area had come from Africa. Excluding the strip brought the estimate down to 8 million tons. For that reason, the result is usually given as a range of 8–10 million tons rather than as one definitive figure.
Prospero and colleagues published their study in Global Biogeochemical Cycles in 2020. Their estimate was far below the central value derived from the satellite data, but it fell within the uncertainty range reported in the 2015 study. That distinction matters: the difference between the estimates does not necessarily mean that dust levels changed over time. The studies covered different years and areas, and they did not define dust transported into the basin in exactly the same way.
Why light can make dust seem heavier
In 2021, Jasper Kok and colleagues proposed another explanation for the gap between some estimates. Larger particles tend to fall out of the air as they cross the Atlantic, while finer particles stay aloft longer and continue the journey. For each ton of material, small particles can block more light than some satellite calculations assumed. As a result, the optical signal could make the dust mass appear larger than it really is.
Kok’s team estimated that about 8.5 million tons of dust settled over the rainforest each year during 2004–2008. However, their estimate included dust from all sources, not just Africa. The study appeared in Atmospheric Chemistry and Physics. Its figure therefore cannot be compared directly with every other estimate without accounting for differences in the area studied, the dust’s origin, and the years covered.
A 2023 study by Xirong Wang and colleagues brought the investigation farther inland. The team tested its model against air measurements in Cayenne and at the Amazon Tall Tower Observatory near Manaus, Brazil. For 2013–2017, the researchers estimated that about 10 million tons of African dust settled each year, with an uncertainty of 1.7 million tons either way. They calculated that this dust carried roughly 8,500 tons of phosphorus annually. The findings were published in Atmospheric Chemistry and Physics.
Season and source shape the dust’s journey
Dust does not reach South America at the same rate throughout the year. In Prospero’s model, 89% of the dust deposited near Cayenne fell between February and May. During this period, rain helps remove particles from the air and bring them down to the ground. In the Northern Hemisphere summer, the belt of tropical rain shifts north, and some dust plumes follow routes closer to the Caribbean and the southern United States.
The model also suggests that the heaviest deposition occurs in northern and northeastern South America, while much smaller amounts reach parts of the central Amazon. That pattern means that a basin-wide total cannot tell us how much dust reaches any particular forest or patch of soil. Where rain falls, how high a dust plume travels, and which way the winds blow all affect where the particles eventually land.
North Africa is not a single, uniform dust source, either. A 2020 study comparing desert regions found that dust from the El Aoukb region, which extends across Mauritania and Mali, has a more favorable route to the Amazon than dust from the Bodélé Depression in Chad. The geographic fingerprint of the dust also shifted between winter seasons. Chemical analysis of particles collected in Cayenne attributed the largest share in winter 2014 to western North Africa, while samples from winter 2016 pointed to central North Africa. A study tracking these year-to-year changes appeared in Geophysical Research Letters in 2022.
Phosphorus: a small load with a possible impact
Dust carries more than minerals: it also adds phosphorus, an element plants need, to soils that are often short of it. In the forest, decomposing leaves return some phosphorus to the soil, where roots can take it up again. But rain washes some of it into streams and rivers, carrying it out of the ecosystem. Yu’s team estimated that the phosphorus brought in by African dust is about equal to the amount the Amazon Basin loses through this process.
That comparison does not mean dust alone determines the forest’s phosphorus budget, or that all the phosphorus it carries reaches tree roots. Other sources contribute, too. A study by Aaron Barclay and colleagues found that vegetation fires in southern Africa also transport phosphorus to Cayenne, including some in a form that dissolves readily in water. The findings were published in the Proceedings of the National Academy of Sciences in 2019.
A field experiment in old-growth forest north of Manaus offers evidence that adding phosphorus can affect plant growth in these soils. Researchers added the nutrient to experimental plots. After two years, fine-root production in the treated plots was 29% higher than in plots that received no phosphorus. But the experiment tested the effect of adding phosphorus; it did not establish how much Saharan dust reached the site or how much of its phosphorus the trees absorbed.
For that reason, Prospero and colleagues called for more measurements within the forest itself, including particles in the air and material carried by rainwater. Coastal samples are useful, but they cannot by themselves settle what is happening far from the coast. Direct measurements deep in the basin could help researchers distinguish African dust from other phosphorus sources and show how deposition changes from place to place and from season to season.
The Bottom Line
Satellite and ground-based evidence agrees that dust from North Africa crosses the Atlantic and adds phosphorus to parts of South America. But the estimated amount depends on how it is measured and which area is included. The 2015 CALIPSO estimate gave a central value of 27.7 million tons per year, with a broad range of 8 to 48 million tons, while models tested against measurements in Cayenne put African dust deposition at 8–10 million tons. Wang’s 2023 study supports a similar figure, estimating 10 ± 1.7 million tons for 2013–2017. Rainfall timing, dust source, and storm paths determine where the particles land, and comparisons across different studies and years cannot establish a long-term decline. Measuring air and rain inside the Amazon remains essential to finding out how much of this airborne fertilizer reaches the forest far from the coast.