Simulations

Large Eddy Simulation (LES) Large-Eddy Simulation (LES) can be used to simulate convectively-driven flow in the planetary boundary layer (PBL). It explicitly resolves the large turbulent eddy (plus the mean) field of momentum and thermodynamic variables by applying a filter to the Navier-Stokes equations. Motions on a scale smaller than the grid scale (subgrid-scale), typically 50 to a 100 meters, are parameterized. In my work I use a recent version of the UCLA-LES code (Stevens, B. and A. Seifert - J.Meteor.Soc.Japan, 86A, 2008). An example of a simulated flow field of a shallow cumulus case (RICO) is shown in below movie, in which the left panels represent vertical velocity and the right panels cloud (liquid) water. Wind speed experiments The cartoon below summarizes the differences in winds, surface fluxes, boundary layer height and profiles of equivalent potential temperature during RICO, comparing periods that were characterized by little to moderate precipitation (the top and middle panel), and periods with moderate to widespread precipitation (the middle and bottom panel). Based on the analysis of RICO data described in my [paper], it is hypothesized that wind speed regulates humidity (hence convection) and explains the differences in the sparsely-precipitating regime, whereas subtle variations in subsidence regulate humidity and convection in the moderately precipitating regime. These results have motivated me to ask the question: do we really understand the influence of wind speed on shallow cumulus convection? In other words: ~ How do stronger winds and variations in subsidence influence convection and precipitation, do stronger winds continue to drive an increase in surface forcing, convective activity and precipitation? Currently I am analyzing the results from (idealized) wind speed experiments that I performed with the LES and that will help us to get more insight in above questions. Some initial results are included in this [talk] that I gave at the GCSS meeting last Summer. 2D cloud movie One nice and fun way to understand how clouds work and develop in a simulation, is by looking at some simple animated movies of the flow field. I made the movies below to better understand how clouds and rain are impacted by the numerics of the LES. Although the buoyant liquid water "bubbles" in these movies look more like mushrooms than clouds, they are fun to watch! The top two panels represent cloud water and the bottom two panels are rain water. That looks like one heavy rain shower in the bottom right panel! The RICO GCSS LES Intercomparison Case of precipitating cumulus In 2006 I helped set-up a Large Eddy Simulation (LES) intercomparison case for precipitating shallow cumulus, an initiative from the GEWEX Cloud System Studies (GCSS) Boundary Layer Group. This work was performed under the guidance of Pier Siebesma and Margreet van Zanten while doing an internship at the Royal Dutch Meteorological Institute - KNMI, The Netherlands. The case description and results can be found here website.

Research links: Publications Observations LES RICO Case Go to: Home Research Curriculum Vitae