Skip to Main Content U.S. Department of Energy
Science Directorate
Page 183 of 740

Atmospheric Sciences & Global Change
Research Highlights

June 2016

Little Big Parameter

Factoring in small-scale winds has a big impact on global climate dust simulations

dust devil in Arizona
A large dust devil, a small-scale wind event, is sending dust aerosol particles into the atmosphere in Arizona. Such a small-scale event is typically not accounted for in global climate models. PNNL’s research has shown that using their computationally efficient technique to account for small-scale dust events like these improves models to more closely resemble real-world results, especially for dust emissions world-wide. Photo courtesy of NASA. Photo courtesy of NASA. zoomEnlarge Image.

They can cloud your vision or sting your face, depending on how they are tossed into the atmosphere. Scientists know that these tiny particles of dust and sea salt can enter the air naturally to affect the climate as well, but when and how varies based in part on local surface wind speeds. This small-scale wind speed variance is often a missing component of large-scale climate models.

Results: Now, Pacific Northwest National Laboratory scientists have developed a new method to represent how local wind speeds can lift natural aerosol particles into the atmosphere. When researchers considered these effects in a global climate model, they found two surprising results. In some cases, the amount of dust produced a year (yearly mean emission) increased by more than 50 percent. In others, the mean remained the same, but the amount of dust raised by weaker winds was higher.

"We knew that small-scale wind variances might greatly affect natural aerosol emissions in models," said PNNL atmospheric scientist Dr. Kai Zhang, who led the study. "With the new method we can quantitatively count that small-scale effect in global climate models, with a very small increase in the computational cost."

Why It Matters: In global climate models, typically simulated at scales of 100 kilometers (62 miles) per grid box, many of the small-scale details are averaged out. For instance, important features like clouds are considered small-scale, and so are wind events for the most part. Yet, scientists have known for decades that things happening at scales smaller than the grid boxes of global and regional climate models can have a large effect on a model's results. That's one of the reasons researchers continue to evaluate which small-scale processes are important to consider.

Surface wind speed, for example, affects how sea salt and dust are sent into the atmosphere, both in computational simulations and in the real world. Most global climate models factor in only a single speed for wind, despite the fact that wind speed can vary considerably across a typical model grid box 100 to 200 kilometers across. Unfortunately, including small-scale calculations such as those for local wind speed in global climate models often comes with a large cost in computer time. The PNNL method, however, proved to be computationally efficient, making the approach more feasible to adopt.

Methods: Scientists analyzed results of local surface wind speeds from finely detailed global and regional models and determined how much variability was currently included. They then developed a way to predict the small-scale wind speed variability and tested their approach in the widely accepted global climate model, Community Atmosphere Model version 5. They used a range of data to represent wind speed variations, taking into account the impact from turbulence (irregular air movement), convection (how heat is transferred in the atmosphere), and topography (the surface features of land, i.e. mountains). Finally, they calculated how emissions might change in each model grid box using different wind speeds.

Percent of dust predicted in the model CAM5
The new method PNNL scientists developed to factor small-scale wind speed variability into global climate models had a surprising impact on the amount of dust the model predicted, increasing levels by more than 50 percent in many cases.

Their results indicated that, while the changes in small-scale wind speeds have a rather minor impact on sea salt emission calculations, such changes strongly affect the modeling of when and how much dust is released from the ground.

What's next? The new method of representing wind speed variability provides the basis for future work to make simulated, wind-driven aerosol emissions more realistic. PNNL scientists are currently studying additional aerosol-related processes that are affected by differences in local meteorological conditions to improve state-of-the-art climate models.


Sponsors: The research was supported by the U.S. Department of Energy's Office of Science, Office of Biological and Environmental Research as part of the Scientific Discovery Through Advanced Computing project in the Earth Systems Modeling Program.

Research Team: Kai Zhang, Chun Zhao, Hui Wan, Yun Qian, Richard Easter, Steven Ghan, and Koichi Sakaguchi, PNNL; and Xiaohong Liu, University of Wyoming

Research Area: Climate & Earth Systems Science

Facilities: The National Science Foundation and the State of Wyoming provided computational resources at the National Center for Atmospheric Research's (NCAR) Wyoming Supercomputing Center, with the support of NCAR's Computational and Information Systems Laboratory. PNNL Institutional Computing also provided computational resources.

Reference: Zhang K, C Zhao, H Wan, Y Qian, R Easter, S Ghan, K Sakaguchi, and X Liu. 2016. "Quantifying the impact of subgrid surface wind variability on sea salt and dust emissions in CAM5." Geoscientific Model Development 9:607-632. DOI:10.5194/gmd-9-607-2016.

Page 183 of 740

Science at PNNL

Core Research Areas

User Facilities

Centers & Institutes

Additional Information

Research Highlights Home


Print this page (?)

YouTube Facebook Flickr TwitThis LinkedIn

In short...

In one sentence: PNNL scientists developed a new method to represent the effects of small-scale wind variability on natural aerosol particle emissions such as dust and sea salt at less computational cost, improving global climate models.

In 100 characters: PNNL developed a better way to model small-scale winds at less computational cost in climate models