My overarching research interests involve using numerical models and data to understand our impact on the world in which we live.  This broad classification masks a hierarchy of models and techniques with greatly varying complexity and range of applicability.  In addition, the research area is highly multidisciplinary, covering: Physics, Chemistry, Numerical methods, Data and Computational Science.

Like most, I do have a specific focus which one might call my ‘professional’ remit. This involves researching the evolution and impact of atmospheric aerosol particles.

It is worth Aerosol particles are ubiquitous components in the atmosphere. Ranging from a few nanometres to microns in size, they are comprised of potentially thousands of chemical components, their size and composition key determinants in their environmental impacts. They remain one of the most uncertain contributors to climate change:directly influencing the radiative flux by scattering/absorbing radiation and indirectly through their ability to act as cloud condensation nuclei (without aerosol particles there would be no clouds!). I use a wide range of holistic approaches to attempt reduction of these uncertainties.

• At the single particle level, highly detailed theoretical tools have been developed to gain further understanding on the link between the chemical nature of organic aerosol, its interaction with inorganic components, and the associated impact on aerosol and cloud properties.
• The partitioning of material between the gaseous and condensed phase is key to determining the evolving chemical composition and size, thus environmental impacts. In addition to detailed modelling studies, I use laboratory studies to validate and improve methods for predicting fundamental properties such as pure component vapour pressures.  These studies offer insights into the applicability of methods, largely designed for chemical engineering purposes, for use in atmospheric studies.
• In addition to the detailed studies mentioned above, the highly complex modelling tools are also used to drive process parametrisation in large scale models. These large scale models demand numerical and chemical complexity be kept to a minimum.  This presents a new challenge to those listed above as the trade-off between accuracy and numerical becomes paramount. Without a retention of significant accuracy there is a danger of erroneously prescribing confidence to the effect of certain detailed phenomena on large scale effects such as climate change and air quality.

The research I carry out as part of my job can spill out into areas that might be considered ‘tangential’. On the contrary, I find it hugely exciting to identify those disciplines that might benefit my current area of research at the university. Presently, these include the following:

• Informatics. In my ‘professional’ research, atmospheric aerosol particles could potentially be comprised of millions of different chemical species. Manually predicting all of the relevant properties of such species, or their behaviour in mixtures, would be impossible. For this I have built software based on open-source informatics developments that allow us to automate such predictions. For more information, please visit a website I have put together with help from a local SME called UManSysProp: http://umansysprop.seaes.manchester.ac.uk/.  The core ethos behind this is that researchers can now represent molecules using textual notation. In our case, we use SMILES strings (http://www.daylight.com/dayhtml/doc/theory/theory.smiles.html) Still in working this area, I also look into ways in which we can use such technology to simplify composition dependent features of atmospheric aerosol. This might include hygroscopicity (potential for absorbing water) or volatility (potential for components to evaporate or condense). For tutorials on these properties, please keep an eye for updates via the UManSysProp website listed above.

• Emerging computational technology. The landscape of high-performance computing is rapidly changing. Alongside traditional multi-core CPU approaches, graphics cards (GPUs) with thousands of cores have become available as ‘accelerators’ for compute-intensive work. Other accelerators include Intel’s XeonPhi or FPGA (Field programmable gate arrays). I became very interested in this technology a few years ago, with an aim to bring such potential into my research. As a result, I am now on the committee for an exciting annual conference organised with other academics at the University of Manchester called EMiT (Emerging Technology): http://emit.manchester.ac.uk/  Why don’t you come along to our next event this year. The programme promises to be hugely interesting and can be found here: http://emit.tech/programme/