News & Media

Latest Stories

258 results found
Filtered by Atmospheric Science, Cybersecurity, Emergency Response, Energy Efficiency, Grid Architecture, Reactor Licensing, Stakeholder Engagement, and Subsurface Energy Systems
APRIL 6, 2020
Web Feature

Blue Light? Orange Light?

PNNL study evaluated "tunable" lighting and its effects on sleep at study in a California nursing home. Tunable refers to the ability to adjust LED light output and the warmth or coolness of the light color.
DECEMBER 11, 2019
Web Feature

PNNL to Lead New Grid Modernization Projects

PNNL will lead three new grid modernization projects funded by the Department of Energy. The projects focus on scalability and usability, networked microgrids, and machine learning for a more resilient, flexible and secure power grid.
DECEMBER 6, 2019
Web Feature

Converging on Coastal Science

Advancing a more collective understanding of coastal systems dynamics and evolution is a formidable scientific challenge. PNNL is meeting the challenge head on to inform decisions for the future.
NOVEMBER 26, 2019
Web Feature

Conquering Peak Power

PNNL’s Intelligent Load Control technology manages and adjusts electricity use in buildings when there’s peak demand on the power grid.
NOVEMBER 13, 2019
Web Feature

Let There Be (Acceptable) Light

Advancements such as LEDs have changed consumers’ experience with lighting. Whereas there was once a simple choice of how much light a consumer desired, there’s now a variety of choices to be made about the appearance of light.

Moments Matter When It Comes to Modeling Rain


Getting rain properties correct in atmospheric models is critical for accurately representing the structure and evolution of cloud systems.

Improved representation of rain microphysics led to more accurate simulations of surface precipitation

March 12, 2019
October 22, 2019

The Science
In atmospheric models, raindrop properties such as rainfall rate are usually described based on the raindrop size distribution. For example, heavy rain rates may have a wider raindrop size distribution than light rain. To improve predictions of rain that reaches the surface, the primary question has been, how can models better represent the evolution of raindrop size distribution in space and time in the atmosphere? A team led by researchers at the U.S. Department of Energy’s (DOE) Pacific Northwest National Laboratory improved the representation of rain microphysics by predicting the shape parameter of raindrop size distributions in a recently developed cloud microphysics scheme. They found that under a wide range of atmospheric conditions, their advanced representation delivered surface rain properties similar to those produced by a benchmark scheme, but with less computational resources.

The Impact
Because raindrops play a major role in the vertical redistribution of heat and moisture in the atmosphere, they are a critical component for modeling the structure and evolution of cloud systems such as mesoscale convective systems. These systems are major sources of heavy rain in the central United States. A proper representation of rain in numerical models is not only vital to predict surface precipitation, but also to accurately simulate environmental conditions and circulation patterns. The advanced rain microphysics representation from this study improves simulations of rain properties under various atmospheric conditions, and it can be used to increase accuracy of weather and climate models. The scheme will be implemented in DOE’s Energy Exascale Earth System Model (E3SM).

Cloud microphysics schemes in weather and climate models usually predict two moments—the total number and mass—of the raindrop size distribution. Researchers upgraded the Predicted Particle Properties (P3) cloud scheme by adding another predicted variable—shape parameter—for raindrop size distribution, turning the two-moment scheme into a three-moment scheme for raindrop representation. They also developed and incorporated a new parameterization for drop-drop collisions—when two drops collide—and the breakup of large drops into smaller ones. 
To evaluate those new developments, the research team tested them with an idealized rain model. The model simulated 450 rain scenarios that were initialized by different raindrop size distributions and environmental conditions. Researchers compared the simulated surface rain properties against those from a detailed and computationally costly reference scheme. The team found that, depending on initial rain intensity, up to 95 percent of simulations with the new developments produced raindrop sizes and surface rain rates within ±20 percent biases from the reference results. This was a considerable improvement from the original two-moment scheme, which only reached 4 percent using the same criteria for comparisons with the reference results. Sensitivity tests showed that both the added degree of freedom—the additional variable for raindrop size distribution—and the new process parameterization contributed to the improvements.

PI Contact
Jiwen Fan, Pacific Northwest National Laboratory,  

This research was supported by the Climate Model Development and Validation program funded by the Office of Biological and Environmental Research in the U.S. Department of Energy Office of Science. Model simulations were performed using PNNL Institutional Computing.

Paukert M, J Fan, PJ Rasch, H Morrison, JA Milbrandt, J Shpund, and A. Khain. 2019. “Three-Moment Representation of Rain in a Bulk Microphysics Model.” Journal of Advances in Modeling Earth Systems 11(1):257−277,