Every cloud has a silver lining, the saying goes, but as it turns out, it's more of a giant halo.
It was already known that clear sky up to several kilometres away from clouds appears brighter than cloud-free sky elsewhere. People had assumed that this was due to varying levels of tiny particles in the air â€“ now research shows that the effect is actually mostly due to light reflected off the cloud and bouncing off the particles.
This seemingly innocuous finding could have a surprisingly big knock-on effect because it means there may be fewer cooling particles in the sky than previously thought. And that could change the way we model climate change.
To discover why the air near clouds appears so aglow, Tamas Varnai and Alexander Marshak at the NASA Goddard Space Flight Center in Greenbelt, Maryland, used MODIS satellite observations from a piece of sky above the Atlantic just southeast of the UK.
Previous studies had speculated that whilst some of the glare may be due merely to measurement artefacts, it might also be the case that more of the small airborne particles aggregate around clouds or swell in size as they absorb some of the cloud's water.
Because they reflect some sunlight back into space, such aerosol particles are thought to make an important contribution to keeping the planet cool. Spraying particles into the atmosphere has even been considered to build a "sun shield" to combat global warming, although that plan was scrapped due to concerns it could damage the ozone layer.
"Aerosols measure only about a tenth of a micron, which is really, really tiny compared to cloud droplets, which are often roughly 10 microns across," says Marshak.
Because of their small size, the sky appears clear even where aerosol particles abound. To work out the amount of particles suspended in the air, Marshak's team looked at the number of clear sky pixels picked up by MODIS and subtracted the reflection that was estimated to come from the planet's surface and air molecules.
"This leaves us with the remaining reflection bouncing off aerosol particles, and so we can estimate their density," explains Marshak.
Using this idea, it makes sense to assume that where the sky appears brighter, light must be being reflected off more or bigger aerosol particles â€“ unless there is more light in the first place, of course.
In their current analysis, Marshak and Varnai found that the bright sky effect was stronger on the sunlit sides of clouds or when the clouds were denser.
Because more light reflects off a denser or sunlit cloud, this suggests that the clear sky brightness near clouds is caused by extra light reflecting off the clouds sideways and then scattering again between the particles in the clear sky area before reaching the satellite. "It's essentially extra energy bouncing off the clouds that enhances the glow of the clear sky," he says.
This effect â€“ called 3D radiative interaction â€“ had been previously identified as a factor cranking up the sky's brightness, but the new data elevates it to the most important factor. This, in turn, means that many estimates of aerosol density may be plain wrong, because most clear sky analyses are close enough to clouds to be affected by the effect, says Marshak.