Skip to Main Content
Skip Nav Destination

Global observations and improved models track atmospheric dust

9 December 2022

Mineral dust affects health and climate. New tracking methods are finally able to monitor and predict its behavior.

Several times every spring and autumn, in early mornings, red rain falls on Belgrade, Serbia. The rain gains its unusual color from dust that originates a thousand kilometers to the south in the Sahara, where hundreds of tons of aerosol microparticles periodically go airborne over the African drylands, most often during dramatic dust events known as haboobs. That dust travels at a high altitude across the Mediterranean and falls on many other southern European cities as well, from Belgrade to Madrid (and can make it as far west as the Amazon), affecting precipitation, people’s health, air transportation, and the climate.

Cars covered in a layer of reddish dust
Cars in Madrid, Spain covered by red dust that originated in the Sahara. Credit: Ladanifer/Alamy Stock Photo

Some episodes of strong dust storms, such as one in Spain in March 2022, receive global media attention. During such episodes, morning TV shows and social media are swamped with photographs of copper-red dust on cars and balconies. The phones of meteorologists and atmospheric physicists ring nonstop. Among these scientists is Slobodan Nickovic of the Institute of Physics Belgrade (IPB), who has dedicated his career to the modeling of dust atmospheric transport. He is one of the architects of the Sand and Dust Storm Warning Advisory and Assessment System (SDS-WAS) and a member of the SDS-WAS steering committee at the World Meteorological Organization (WMO).

“Mineral dust is one of the most abundant aerosols in the atmosphere,” says Nickovic. The IPB tracks dust with lidar, a pulsed laser that locates particles in the atmosphere based on the time it takes for reflected light to return to a receiver. Lidar is just one of many instruments used to observe mineral dust transport from the ground in recent years.

Lately, a significant surge in the general interest in dust, beyond communities of physicists and climate scientists, has invigorated dust research and strengthened institutional support. The WMO established SDS-WAS in 2007; now, because of the international cooperation of scientists from several countries, there is sufficient real-time data from the atmosphere to both observe and model dust transport in new ways.

Millions of tons in the atmosphere

Mineral dust is made of tiny particles of various shapes and sizes, from submicron to several tens of microns in diameter. These fragments, composed mainly of the oxides and carbonates that constitute Earth’s crust, are born out of complex geomorphological processes. Recent research has demonstrated that most major dust sources are located in arid regions in topographic depressions, where deep alluvial deposits have resulted from intermittent flooding.

Dust particles are brought to high atmospheric levels by turbulent updrafts. After the particles travel long distances, they settle again on the ground because of rain or gravitational sedimentation.

Dust contributes to the richness of the living world. In the ocean, it’s a crucial source of minerals for phytoplankton, which produces up to 45% of the oxygen in the atmosphere. Both iron from Saharan dust and macronutrients like nitrate and phosphate are essential for phytoplankton’s growth. Phytoplankton, in turn, forms part of the base of the marine food chain, providing nutrition for shrimp, snails, jellyfish, and many other sea creatures. Likewise, Saharan dust contributes essential nutrients such as phosphorus to the flora of the Amazon rainforest.

But dust also has potentially devastating effects. “The risks associated with dust are often underappreciated,” says Daniel Tong of George Mason University in Virginia. “We need a systematic view. It begins with a full list of potential damage.”

For example, dust represents an enormous everyday challenge to air travel: It reduces visibility, damages aircraft engines and instrument sensors, and causes dangerous icing during flight. Meanwhile, on the ground, airborne dust in the Pan-American region can carry a soil fungus called Coccidioides, which can cause an infectious and sometimes deadly disease called Valley fever. The incidence rate of Valley fever increased by 700% from 1998 to 2011 in US areas frequented by dust storms. Tong says that a range of other health issues can be brought about by dust, including increased risk of asthma, allergies, and heart attacks.

In the atmosphere, dust absorbs and scatters incoming solar radiation. Dust particles also nucleate ice, which makes heterogeneous cloud glaciation more efficient, and thus contribute to the formation of precipitation. As a result, dust has various effects on climate change. The Intergovernmental Panel on Climate Change says dust can increase or decrease global temperatures, depending on particle size, altitude, and nature.

A recent paper reported on Arctic sources of dust, which accelerates warming in the Arctic and Antarctica. As glaciers retreat, they produce new bare land areas, which become new dust sources at high latitudes. The dust then settles on and darkens polar glaciers, which subsequently absorb more radiation and melt faster.

Human activity is also increasing the amount of atmospheric dust through a process called desertification. Low topsoil moisture, unfrozen soil, and near-surface wind velocity above a certain threshold all increase airborne dust. New studies have shown that humans have considerably contributed to the creation of new dust sources because of land disturbance, desiccation of lakes, and invasive agricultural practices. As much as one-fifth of total dust emissions have origins in those types of anthropogenic sources.

Online and offline models

A recent assessment of dust emissions based on global models indicates that the Sahara and Sahel regions of northern Africa emit the most dust (790–840 million tons/year), followed by the Gobi and Taklamakan deserts in China and Mongolia (140–220 million tons/year). Other dust sources in the Middle East, Central Asia, Eastern Australia, South America, the southern US, and northern Mexico produce less than 100 million tons/year.

Map of the world color coded to show where dust originates
A global map of the sources of sand and dust storms. Credit: UNCCD

Meteorologists and climatologists have an arsenal of models that effectively monitor airflow, temperature, and other atmospheric parameters locally and globally. But modeling the transport of atmospheric dust has proved difficult. Trouble begins on the ground: Dust emissions depend nonlinearly on surface wind speed and soil mosture and erodibility. The complex land–atmosphere interactions are a critical issue in dust transport models. Another major problem is that models inevitably rely on a coarse recording of meteorological conditions. For example, atmospheric models are often unable to reproduce the small-scale wind events that are responsible for a large part of dust behavior.

For those reasons, models of dust transport developed before 1990 had limitations. They employed the so-called offline dust modeling approach, which relied on a model of the global atmosphere. In that technique, the atmospheric model was run first, and the wind, temperature, and moisture were recorded every three or six hours. Then, using those parameters, a standalone dust model was executed. But dust particles don’t quietly wait for three hours to make their next move, so offline models were not sufficiently accurate.

In the mid 1990s, Nickovic and his associates developed the online dust model, which is now widely used for dust predictions and climate models. “The dust concentration equation is embedded in a regional numerical weather prediction model,” says Nickovic. The dust concentration is updated on the order of tens of seconds when the atmospheric parameters are calculated.

Physicist’s DREAM

Based on Nickovic’s 1996 model, the Dust Regional Atmospheric Model (DREAM) was developed in 2001 as an add-on to an atmospheric model that solves the equation for the transport of a conserved mass of dust. “Numerical modeling through DREAM and other initiatives allowed us to translate complex physical processes, interactions, and mechanisms into algorithms,” says Vassilis Amiridis, research director at the National Observatory of Athens Institute for Astronomy, Astrophysics, Space Applications, and Remote Sensing.

Amiridis says that DREAM is one of the first initiatives that made it possible to routinely predict dust production and transport patterns. It and other new models can finally answer lingering questions about dust physics. One such question is how large desert dust particles, with diameters greater than 20 μm, frequently end up far from their sources. Another open question is what other physical mechanisms may affect dust transport. For example, recent measurements suggest that particles may acquire charge and orient along electric fields in the atmosphere, explains Amiridis.

Dust researchers can now make more reliable predictions of dust behavior than in the past. SDS-WAS, which started as an unassuming scientific initiative within the WMO, now includes 9 global and 15 regional models, and as many as 25 organizations around the world create daily dust prognoses. Among other things, SDS-WAS is able to forecast dust storms before they happen.

“Despite the remarkable progress in atmospheric dust modeling over the last two decades and the growing availability of sophisticated ground-based and spaceborne aerosol observations, several critical points regarding dust monitoring and forecasting are not yet well addressed,” says Amiridis. With the support of the European Research Council, Amiridis has developed the Panhellenic Geophysical Observatory on the Greek island of Antikythera. The research there focuses on air masses and the different aerosol types they transport, including mineral dust from Africa, smoke from regional forest fires, anthropogenic pollution from megacities, and sea-salt particles.


Although a dust storm is an unforgettable sight, high-altitude dust is not easy to see. Dust transport differs with altitude, so researchers need to measure aerosol layers with high vertical resolution. The best data so far come from surface-based lidar systems. Most such systems perform vertical measurements at heights of up to 10 km and record atmosphere profiles with a spatial resolution from 3.75 m to 7.5 m and a time resolution of 60 seconds. The systems that are not surface-based, such as those carried by satellite, usually have a much lower spatial resolution, from 1 m to 100 m, but they have global coverage.

“The high variability of aerosol fields requires characterization with high resolution in time and space, and hence a higher density of observations,” says Zoran Mijić of the Environmental Physics Laboratory at the IPB, where one of the European watchtowers for the Saharan dust is located.

Photo of a lidar facility emitting a laser into the sky
The lidar facility at the Institute of Physics Belgrade. Credit: Institute of Physics Belgrade

The Aerosol, Clouds and Trace Gases Research Infrastructure (ACTRIS) has made it possible to deliver lidar data sets to users in near real time. New lidar advancements within ACTRIS have also made it possible to deliver particle depolarization-ratio measurements. When the lidar-emitted light, which is always linearly polarized, scatters off spherical particles, its polarization does not change. But when scattered by nonspherical particles, the light is depolarized. By measuring the ratio between polarized and depolarized backscattered light, lidar can distinguish spherical from nonspherical particles.

The ratio between the number of spherical and nonspherical particles speaks volumes about the nature of aerosols. For example, raindrops in a cloud are spherical, whereas ice crystals are not, so the ratio reveals the inner structure of a cloud. Similarly, with a depolarization ratio, lidar can distinguish between dust and marine aerosols, because dust particles are irregularly shaped and marine aerosols are fairly spherical.

Similar lidar techniques can provide other useful information about dust particles, such as dust mass concentrations, dust surface area, cloud condensation nuclei, and ice-nucleating particle concentrations. “Numerous such lidar methods have been developed in the last few years,” says Albert Ansmann, an investigator at the Leibniz Institute for Tropospheric Research who has contributed considerably to those efforts.

Global monitoring

“Desert dust can be transported over long distances, so it is never a regional event,” says Ansmann. He emphasizes the importance of strengthening international cooperation, which has already improved data quality and quantity. For example, the European Aerosol Research Lidar Network, founded in 2000, has expanded from its original 17 stations in 10 countries to 31 stations in 16 countries.

Satellites offer truly global dust observation. After using a series of lidar instruments—which for decades recorded the atmosphere from various spacecraft, specialized airplanes, and even the space shuttle Discovery—NASA’s Langley Research Center and the French space agency launched the Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observation satellite in 2006. The satellite is equipped with an instrument that combines active lidar with passive IR and visible imagers to probe the vertical structure and properties of aerosols over the globe.

In addition, Ansmann says, lidar on a satellite called EarthCARE will be able to directly determine quantities, such as backscatter and extinction height profiles, that aren’t currently measurable. In cooperation with the Japan Aerospace Exploration Agency, the European Space Agency plans to launch EarthCARE in September 2023. One of the EarthCARE instruments will operate in the UV range at a wavelength of 355 nm, with a high spectral resolution receiver and depolarization channel. The instrument will provide atmospheric profiles of aerosols with a vertical resolution of about 100 m when measuring from the ground up to an altitude of 20 km and a resolution of 500 m when measuring from an altitude of 20 km to 40 km.

Predictive power

Along with a dramatic surge in the number and types of lidars and the new physical properties they can measure, the amount of dust transport data has also increased. That increase gives dust models more predictive power.

Similarly, dust models are continually being adapted to the observed vertical profiles recorded by lidar instruments. Ever more varied lidar data provide better inputs and daily verification and validation of models that help improve them.

Although dust is a phenomenon whose impact on climate is not easy to determine, advances in the understanding of desertification and dust’s impact on Arctic and Antarctic warming represent fertile ground for more accurate predictions of climate change based on new knowledge about mineral dust.

Subscribe to Physics Today
Get our newsletters
Close Modal

or Create an Account

Close Modal
Close Modal