Nick Schutgens

profile image

Nick Schutgens

postdoctoral researcher

I am member of the Climate Processes Group in the subdepartment of Atmospheric, Oceanic and Planetary Physics.

I have a PhD in astrophysics and have since worked in atmospheric remote sensing (GOME, SCIAMACHY, EarthCARE Doppler radar, MODIS), aerosol data assimilation (using an ensemble Kalman filter and smoother) and aerosol modelling (MIROC-SPRINTARS, ECHAM-HAM, HadGEM-UKCA, WRF-Chem). My main interest are the climate, biosphere and health impacts of aerosol. In my work I use observations to try to improve global aerosol models. This involves different lines of approach: trying to understand the consequences of different spatio-temporal scales in observations and models on comparisons; evaluating models with observations; process studies of models; and parameter estimation and identification of structural errors of these models.

I am a Dutch national and have worked before in the Netherlands (KNMI) and Japan (NICT and University of Tokyo).

I am and have been involved in the supervision of master and PhD students, both here at the University of Oxford and before at the University of Tokyo. For the past three years, I have given the "Flows, Fluctuations and Complexity" tutorials at Worcester college (I'm currently on a break).

My most recent work concerns an assessment of the effect of spatio-temporal sampling on model-observation comparisons. Global models employ grid-boxes (~ 200 by 200km) far in excess of the typical field-of-view of observations (1-10km). Also, often time-averaged (daily, monthly, yearly) datasets are used in these comparisons even though observational sampling (depending on the measurement) shows a lot of intermittency. I have shown that these sampling issues can easily introduce errors of 50-100% and suggested strategies for mitigating them. I have also started evaluating several global aerosol models with remote sensing data (MODIS, AERONET, MAN).

Also, I have recently conducted a pathway analysis for processes in the global aerosol model ECHAM-HAM. This shows in unprecedented detail when and where what processes are important. This analysis helps to understand model simulation; prioritise model improvements; interpret evaluation with observations; simplify aerosol modules (and reduce CPU requirements); and to identify potential structural errors. Currently, I am extending this analysis to HadGEM-UKCA.

A bit further back in my scientific past, I developed an ensemble Kalman filter and smoother for global aerosol (at the University of Tokyo). This system integrates information on aerosols from observations and model predictions into a single, comprehensive picture and is still in use, by myself and other researchers. Two institutes (NIES in Japan and BSC in Spain) are implementing it for forecasts. My own interest in EnKF is its potential for parameter estimation and identification of structural errors.

I am involved with the development of the Community Intercomparison Suite, as scientific project manager. I advise both the National Institute for Environmental Science (NIES in Tsukuba, Japan) and the Barcelona Super Computing Center (BSC in Barcelona, Spain) on the implementation of an ensemble Kalman filter (developed by myself) for improving forecasts of aerosol. I am also providing feedback to the UK Met Office on the comparison of HadGEM-UKCA with aerosol remote sensing observations.