Bridging the model-data divide for shallow convection



Gustafson, William I. — Pacific Northwest National Laboratory
Vogelmann, Andrew M. — Brookhaven National Laboratory

Area of research:

Cloud Distributions/Characterizations

Journal Reference:

Gustafson W, A Vogelmann, Z Li, X Cheng, K Dumas, S Endo, K Johnson, B Krishna, T Toto, and H Xiao. 2020. "The Large-Eddy Simulation (LES) Atmospheric Radiation Measurement (ARM) Symbiotic Simulation and Observation (LASSO) Activity for Continental Shallow Convection." Bulletin of the American Meteorological Society, 101(4), 10.1175/BAMS-D-19-0065.1.


Combining large-scale atmospheric models and observations presents a long-standing challenge for scientists because of the inherent mismatch between different space and time scales. For example, shallow convective clouds—low, puffy clouds that reflect sunlight back to space—are so small that typical atmospheric models cannot resolve them. The U.S. Department of Energy’s Large-Eddy Simulation (LES) Atmospheric Radiation Measurement (ARM) Symbiotic Simulation and Observation (LASSO) activity seeks to bridge such scale gaps. To get a more detailed look at the atmosphere around an ARM observatory, LASSO packages LES modeling with real-world observations. In a new foundational paper in the Bulletin of the American Meteorological Society (BAMS), researchers describe the first scenario of focus for LASSO: shallow convection at ARM’s Southern Great Plains atmospheric observatory in Oklahoma.


LES is an approach that uses fine-scale models such that most turbulent mixing of near-surface air is essentially resolved. Compared with other long-term LES activities, LASSO is unique in that it uses a LES ensemble for each simulated day with shallow convection. The ensemble helps account for uncertainty in the model input data representing the atmosphere that surrounds the LES, and the air around the model is used to drive the LES as it advances in time. This LES ensemble is packaged into data bundles that combine model input and output data, observations, and evaluation data that show how the LES results compare with what was observed. These bundles make it easier for researchers to interact with and choose the LES simulations that they need for their work. So far, example uses of LASSO include improving theoretical understanding of shallow clouds, cloud representations in models, and radar observation methodologies. The number of researchers using LASSO continues to increase as more simulations are released and the library of data bundles grows.


Shallow convective clouds are critical to the Earth’s energy balance, and they are important for solar forecasting at the surface for applications such as solar farms. However, these clouds cannot be resolved even in the highest-resolution operational weather forecast model, let alone in climate models that must have accurate handling of the Earth system’s radiative balance. LASSO seeks to contribute to this and other active areas of research.

The BAMS foundational article introduces LASSO and describes its audience, core concepts, LES production for shallow convection, and available data and general LES behavior. As of April 2020, LASSO data bundles are available for 78 case dates from 2015 to 2018 at ARM’s Southern Great Plains atmospheric observatory. The Southern Great Plains is one of the most heavily instrumented long-term observatories in the world. Processing of data bundles is underway for the 2019 shallow-convection season (spring/summer). The BAMS paper also touches on LASSO’s future, including a new deep convection scenario using data from ARM’s Cloud, Aerosol, and Complex Terrain Interactions (CACTI) field campaign, which took place from October 2018 through April 2019 in Argentina.