Overview

Image courtesy of the Max Planck Institute for Dynamics and Self-Organization
Particle trajectories in turbulence; color corresponds to the particle speed. Image courtesy of the Max Planck Institute for Dynamics and Self-Organization

Introduction

We conduct theoretical and computational research on fluid dynamics, motivated primarily by environmental problems. Where possible, we always seek to collaborate with experimentalists so that the problems may be attacked from all angles in collaborative efforts.

Although our main focus is on systems where the flow is turbulent, we also have interests in problems such as porous media flows and hydrology. Some of the current projects being worked on in our group are listed below. Currently, the main projects are focussed on understanding and predicticting the motion of particles in turbulent flows.

Project 1

Mixing and dispersion of particles in turbulent flows; Lagrangian irreversibility; scalar transport by turbulent velocity fields. We are studying how these problems are affected by the kinds of complexity found in real applications to understand and predict pollution dispersion, mixing in oceans, droplet dynamics in clouds etc.

Project 2

Particle collision rates in turbulent flows, with applications to understanding and predicting the formation of rain droplets in clouds, and water filtration processes. There is much work to be done in order to construct accurate statistical theories for collisions in turbulence. Furthermore, particle collisions in turbulence have only been studied in detail for simplified versions of the real problems of interest. Many aspects of the real problems remain to be understood and described theoretically.

Project 3

Sub-grid models for Large Eddy Simulations (LES) of complex particle-turbulent flows. The motivation behind this work is that in many real applications, the system complexity is too great to be handled analytically. In these cases, LES presents a promising tool to predict the motion of particles in complex flows. However, for many problems the scales removed by the filtering are precisely the scales that dominate the system behavior, such as particle collisions which are mainly affected by the small scales of the turbulence. Constructing subgrid models with sufficient sophistication to represent the small-scale turbulence dynamics required for predicting these problems is a major challenge that is being worked on in our group.