When Longlei Li and a team of Cornell scientists pointed a NASA satellite at Earth's deserts from the International Space Station, they solved a puzzle that has clouded climate science for decades: what exactly is blowing through our atmosphere, and how much does it warm or cool the planet?
Mineral dust—those fine particles lifted from the Sahara, Middle East, and East Asia—is everywhere in our skies, yet researchers have struggled to understand its true climate impact. These particles scatter and absorb sunlight, seed clouds, and fertilize distant ecosystems, but their effect on planetary energy balance has remained frustratingly uncertain. The problem was iron oxides, the light-absorbing minerals embedded in dust grains. A single uncertainty about these compounds was throwing off virtually every climate model on Earth.
The breakthrough came from EMIT, the Earth Surface Mineral Dust Source Investigation instrument mounted aboard the International Space Station. Using imaging spectroscopy, EMIT maps mineral composition across Earth's arid regions at an unprecedented resolution of 60 meters—fine enough to understand the ground sources that eventually become airborne dust. For the first time, scientists could reliably identify key dust minerals like hematite and goethite across vast remote deserts where travel is difficult or impossible. "It was amazing to see how much the quality of the instrument improved our understanding of the mineral composition of these areas," said Natalie Mahowald, deputy principal investigator on EMIT and Irving Porter Church Professor in Engineering.
When Li's team fed this new EMIT data into four independent Earth system models, the results were stunning. The uncertainty surrounding iron oxides in dust models plummeted from 0.62 watts per square meter to just 0.1 watts per square meter—a reduction by more than a factor of six. Iron oxides, which had been the single largest obstacle to accurate climate estimates, were no longer dominating the uncertainty. The field could finally move forward.
The gains were most dramatic over the Sahara Desert, Earth's largest source of atmospheric dust. EMIT-enabled models reduced errors in simulated radiative effects by as much as 80%, bringing simulations into line with actual satellite observations. Across all major global dust source regions—North Africa, the Middle East, and Asia—uncertainty dropped by more than half. Globally, dust's overall effect on solar radiation remained within previously estimated ranges, but with vastly greater confidence behind those numbers.
What emerged were clearer regional patterns. Dust from parts of North Africa tends to be iron-rich, enhancing solar absorption and potentially warming the atmosphere under certain conditions. Dust from some Asian regions is more reflective and cooling. These distinctions matter enormously for understanding regional climate responses. "This makes our understanding more physically grounded and that's essential for improving climate projections," Li explained.
The research, published in Nature Geoscience, has fundamentally shifted what scientists will focus on next. Rather than asking only what dust is made of—a question that now has a much clearer answer—researchers can increasingly concentrate on how dust moves through the atmosphere, how it breaks apart, and how it distributes across particle sizes. The mystery of mineral composition, a mystery that seemed almost impossible to solve from Earth, has been cracked from 400 kilometers above it. And with that solved, the next generation of climate models can be more honest about one of the planet's most pervasive and consequential particles.