Careers | Phone Book | A - Z Index

Fluctuating Hydrodynamics


In this figure showing the development of spinodal decomposition in a near-critical van der Waals Argon system, the liquid (red) and vapor (blue) domains spontaneously develop when the system is quenched into the unstable portion of the phase diagram and grow over time. The simulations were done using the fluctuating hydrodynamics methodology developed for multiphase systems.

Atmospheric Modeling


This 3-d cloud was simulated using a low Mach number model that accurately incorporates
water phase transitions in moist air. Iso-contours of liquid water are depicted, intercepted by a vertical plane where concentration of water vapor is indicated.

High-Order Combustion Simulations

resizedimage450413 combustionmodeling2

This image was featured on the cover of Combustion Theory and Modeling and was generated using a new numerical algorithm for integrating the multicomponent, reacting, compressible Navier-Stokes equations, targeted for direct numerical simulation of combustion phenomena.

Fluctuating Hydrodynamics


To study the effects of thermal fluctuations in fluids at the microscale, we have developed a new low Mach number fluctuating hydrodynamics code for multicomponent mixtures. The image shows the development of a diffusive layer convection instability as a layer of less-dense salty water is placed on top of a horizontal layer of denser sweet water. The observed giant fluctuations are caused by long-range correlations between fluctuations.

Computational Cosmology


Researchers in CCSE and the Computational Cosmology Center have collaborated to develop a new, massively parallel N-body hydro cosmology simulation code. The code is being used at LBL to study the Lyman alpha forest.

Compressible Astrophysics


This simulation of the death of a massive star used the compressible astrophysics code, CASTRO, and was featured on Nature magazine's Images of the Month. CASTRO is a massively parallel radiation-hydrodynamics code being used to study explosive astrophysical phenomena.

Low Swirl Burner


NOx emissions in a simulation of the low swirl burner experiment in the LBNL Combustion Laboratory, capturing the complex cellular burning structures in this lean premixed hydrogen-air flame that lead to localized hot spots where NOx emissions are enhanced. The simulation, which is at the full scale of the experiment, is the first of it's kind, in terms of domain size, and chemical fidelity.

The Center for Computational Sciences and Engineering (CCSE) develops and applies advanced computational methodologies to solve large-scale scientific and engineering problems arising in the Department of Energy (DOE) mission areas involving energy, environment, and industrial technology.  The primary focus of CCSE researchers is on designing algorithms for multiscale, multiphysics problems described by nonlinear systems of partial differential equations, and in developing implementations of algorithms that target current and next-generation massively parallel computational architectures.  Sample application areas include combustion, porous media flow, fluctuating hydrodynamics, atmospheric modeling, cosmology and astrophysics.  CCSE researchers work collaboratively with application scientists to develop state-of-the-art solution methodologies in these fields.

CCSE has also been the home of BoxLib, a software framework for massively parallel block-structured adaptive mesh refinement (AMR) codes.   BoxLib has been used for the development of new algorithms and is the basis for many mature AMR applications.    As part of DOE's Exascale Computing Project, CCSE and other researchers are developing AMReX, a next-generation block-strucutured AMR framework.

For more about CCSE, see

Group Lead: Ann Almgren