Pollution yields longer-lasting storm clouds

Nov 26, 2013
Pollution decreases the size of cloud and ice particles and increases their lifespans, making clouds grow bigger.

(Phys.org) —A new study reveals how pollution causes thunderstorms to leave behind larger, deeper, longer lasting clouds. Appearing in the Proceedings of the National Academy of Sciences November 26, the results solve a long-standing debate and reveal how pollution plays into climate warming. The work can also provide a gauge for the accuracy of weather and climate models.

Researchers had thought that pollution causes larger and longer-lasting by making thunderheads draftier through a process known as convection. But atmospheric scientist Jiwen Fan and her colleagues show that pollution instead makes clouds linger by decreasing the size and increasing the lifespan of cloud and ice particles. The difference affects how scientists represent clouds in climate models.

"This study reconciles what we see in real life to what computer models show us," said Fan of the Department of Energy's Pacific Northwest National Laboratory. "Observations consistently show taller and bigger anvil-shaped clouds in storm systems with pollution, but the models don't always show stronger convection. Now we know why."

Also, pollution can decrease the daily temperature range via such clouds: High clouds left after a thunderstorm spread out across the sky and look like anvils. These clouds cool the earth during the day with their shadows but trap heat like a blanket at night. Pollution can cause clouds from late afternoon thunderstorms to last long into the night rather than dissipate, causing warmer nights.

Secret Life of Clouds

Models that predict weather and climate don't reconstruct the lives of clouds well, especially storm clouds. Usually these models replace storm clouds with simple equations that fail to capture the whole picture.

Because of the poor reconstructions, researchers have been faced with a dilemma: Pollution causes the anvil-shaped clouds to linger longer than they would in clean skies—but why?

Possible reasons revolve around tiny natural and manmade particles called aerosols that serve as seeds for cloud droplets to form around. A polluted sky has many more aerosols than a clean sky—think haze and smog—and that means less water for each seed. Pollution makes more cloud droplets, but each droplet is smaller.

More and smaller droplets change things for the clouds. Researchers have long thought that smaller droplets start a chain reaction that leads to bigger, longer-lasting clouds: Instead of raining down, the lighter droplets carry their water higher, where they freeze. The freezing squeezes out the heat the droplets carry with them and causes the thunder cloud to become draftier. The stronger convection lifts more water droplets, building up the cloud.

But researchers don't always see stronger convection every time they see larger and longer-lasting clouds in polluted environments, indicating a piece of the puzzle was missing.

To solve this dilemma, Fan and colleagues decided to compare real-life summer storm clouds to a computer model that zooms deep into simulated clouds. The model included physical properties of the cloud particles as well as the ability to see convection, if it gets stronger or weaker. Most models run in days or weeks, but the simulations in this study took up to six months.

"Modeling the details of cloud microphysical properties is very computationally intensive, so models don't usually include them," said Fan.

Convection Vexation

The researchers started with cloud data from three locations that differ in how polluted, humid and windy they typically are: the tropics in the western Pacific, southeastern China and the Great Plains in Oklahoma. The data had been collected through DOE's ARM Climate Research Facility.

With support from DOE's Regional and Global Climate Model program, the research ran simulations on PNNL's hometown supercomputer Olympus. Their simulations of a month of storms ended up looking very similar to the actual observed clouds, validating that the models re-created the storm clouds well.

The team found that in all cases, pollution increased the size, thickness and duration of the anvil-shaped clouds. However, only two locations—the tropics and China—showed stronger convection. The opposite happened in Oklahoma—pollution made for weaker convection.

This inconsistency suggested that stronger convection isn't the reason. Taking a closer look at the properties of water droplets and ice crystals within clouds, the team found that pollution resulted in smaller droplets and ice crystals, regardless of location.

In addition, the team found that in clean skies, the heavier ice particles fall faster out of the anvil-shaped clouds, causing the clouds to dissipate. However, the ice crystals in polluted skies were smaller and too light to fall out of the clouds, leading to the larger, longer-lasting clouds.

Lastly, the team estimated how much warming or cooling the storm clouds contributed. Overall, the polluted clouds cooled the day and warmed the night, decreasing the daily temperature range.

Most models don't simulate convection well, take into account the microphysical processes of storm clouds, nor address how pollution interacts with those processes. Accounting for effects on storm clouds in this way could affect the ultimate amount of warming predicted for the earth in the next few decades. Accurately representing in is key to improving the accuracy of predicted changes to the climate.

Explore further: Synchronization of North Atlantic, North Pacific preceded abrupt warming, end of ice age

More information: Jiwen Fan, L. Ruby Leung, Daniel Rosenfeld, Qian Chena, Zhanqing Lid, Jinqiang Zhang, and Hongru Yan. Microphysical effects determine macrophysical response for aerosol impacts on deep convective clouds, Proc Natl Acad Sci, Early Edition online the week of November 11-15, 2013, DOI: 10.1073/pnas.1316830110

Related Stories

Pollution teams with thunderclouds to warm atmosphere

May 18, 2012

Pollution is warming the atmosphere through summer thunderstorm clouds, according to a computational study published May 10 in Geophysical Research Letters. How much the warming effect of these clouds offsets the cooling that o ...

Fair-weather clouds hold dirty secret

Apr 26, 2013

(Phys.org) —Their fluffy appearance is deceiving. Fair-weather clouds have a darker side, according to scientists at Pacific Northwest National Laboratory. Fair-weather cumulus clouds contain an increasing ...

A better picture of clouds

Feb 13, 2012

Some of us look at clouds and see animal shapes. Scientists are looking beyond. For the first time, a team of scientists led by Pacific Northwest National Laboratory used actual measurements of clouds and ...

Recommended for you

Fires in Central Africa During July 2014

3 hours ago

Hundreds of fires covered central Africa in mid-July 2014, as the annual fire season continues across the region. Multiple red hotspots, which indicate areas of increased temperatures, are heavily sprinkled ...

NASA's HS3 mission spotlight: The HIRAD instrument

13 hours ago

The Hurricane Imaging Radiometer, known as HIRAD, will fly aboard one of two unmanned Global Hawk aircraft during NASA's Hurricane Severe Storm Sentinel or HS3 mission from Wallops beginning August 26 through ...

Fires in the Northern Territories July 2014

Jul 23, 2014

Environment Canada has issued a high health risk warning for Yellowknife and surrounding area because of heavy smoke in the region due to forest fires. In the image taken by the Aqua satellite, the smoke ...

User comments : 2

Adjust slider to filter visible comments by rank

Display comments: newest first

verkle
1.5 / 5 (8) Nov 26, 2013
Another finding in the climate models arena that refutes what we thought up until now. How many more of these findings are coming? We are taking small baby steps in refining climate models, which are still far from being practical.
animah
not rated yet Nov 29, 2013
I normally never comment on this stuff but this is too much. Verkle, have you read up on these models in any detail that allows you to so disdainfully dismiss them as "impractical"?

For example, would you kindly comment on the GFDL's work at Princeton and where you find the math they crunch for weeks at a time on their 30,000 processor array wanting?

Because I am sure you would be keen on dispelling any misunderstanding as to the casually uncouth and trivially shallow impression you may have inadvertently created with your post among the majority here.