PNNL

The Science

Weather and climate models are now approaching kilometer-scale resolution where storms are better represented than in traditional coarser resolution models. However, the deep convecting motions that drive them are still not fully resolved. Pacific Northwest National Laboratory led researchers studying the effects caused by underresolving these motions. The team found that they become too large in scale, too efficiently transporting air vertically and degrading the evolution of the larger storm complex. Increasing model resolution can improve the simulated storm complex evolution but requires greater computing resources that are currently not scalable to regional and global weather and climate models.

The Impact

Deep convective circulations are expected to be underresolved in kilometer-scale models for most storm complex scenarios. However, this can bias vertical redistribution of atmospheric energy that influences how larger storm complexes evolve with the potential to alter predicted regional climate conditions. More research is required to determine how models behave over a wider range of well-observed cases and whether biases can be reduced without significantly increasing model computing expense.

Summary

To take a closer look at the impact of deep convective circulations on larger storm complex evolution, researchers examined a developing squall line storm complex observed on 20 May 2011 during the US DOE Atmospheric Radiation Measurement (ARM) Midlatitude Continental Convective Clouds Experiment (MC3E) in two model simulations with different horizontal resolutions. They found that radar observations of the squall line precipitation and wind structure are better reproduced by the higher resolution simulation, highlighting a bias in the coarser resolution simulation. This bias is caused by underresolved deep convective updrafts and downdrafts in the coarser resolution simulation through the following pathway:

1. Deep convective updrafts and downdrafts in the coarser resolution simulation are wider than those in the higher resolution simulation.
2. Wider convective drafts have greater mass fluxes and carry more condensate than narrower drafts.
3. Relatively wider downdrafts in the coarser resolution simulation more efficiently transport dry mid-level air downward than narrower downdrafts in the higher resolution simulations.
4. The more efficient vertical transport in the coarser resolution simulation accelerates development of a cold pool and downward transport of horizontal momentum, producing a more sheered, wider and weaker deep convective region that propagates, matures, and decays more quickly than in the higher resolution simulation.

Relationships between draft width and other draft properties are very similar in the two simulations. This indicates that the differences in the simulations are primarily a result of the differing draft size distribution as opposed to differences in draft properties for a given draft size. These results imply that underresolved convective circulations in kilometer-scale simulations may vertically transport air too efficiently and too far vertically, potentially biasing buoyancy and momentum distributions that impact entire storm system evolution.

PNNL Contacts

Jerome Fast, Pacific Northwest National Laboratory, jerome.fast@pnnl.gov

Adam Varble, Pacific Northwest National Laboratory, adam.varble@pnnl.gov

Funding

Funding for this study was provided by the US Department of Energy Atmospheric System Research grants DE-SC0008678 and DE-SC0016476 with additional support by the U.S. Department of Energy Office of Science Biological and Environmental Research as part of the Atmospheric System Research program. Pacific Northwest National Laboratory is operated by Battelle for the U.S. Department of Energy under Contract DE-AC05-76RLO1830. The Center for High-Performance Computing (CHPC) at the University of Utah provided computing resources, NOAA provided NEXRAD radar and radiosonde data, and ARM provided radiosonde observations. Mike Dixon and the University Corporation for Atmospheric Research provided RadX software (https://www.eol.ucar.edu/content/lidar-radar-open-software-environment) and Scott Collis and Jonathan Helmus provided Py-ART software (https://arm-doe.github.io/pyart/).