Search for a project

Satellite Observations of Air Quality

Prof Martyn Chipperfield (SEE), Dr Richard Pope (SEE), Dr Brian Kerridge (STFC Rutherford Appleton Laboratory), Dr Antje Inness (European Centre for Medium Range Weather Forecasts)

Project partner(s): Rutherford Appleton Laboratory and National Centre for Earth Observation (CASE)

Contact email: M.Chipperfield@leeds.ac.uk

Project Summary

Poor air quality is a global environmental issue which is receiving a lot of public and media attention. Current and recent satellite missions can provide observations of pollutants such as ozone and nitrogen dioxide on global down to city scales (e.g. Pope et al., 2014). However, for these data to be used in air quality forecast systems, the systems have to be adapted to interpret the vertical sensitivities of the satellite observations. This project will further develop and test the European Centre for Medium-Range Weather Forecasts (ECMWF) Copernicus Atmospheric Monitoring Service (CAMS) to use satellite trace gas vertical profile information. The resulting improvements in, for example, ozone forecasts will be tested and the overall assimilated products used to look at decadal trends in global atmospheric pollution.

This PhD project is supervised in Leeds by Prof Martyn Chipperfield, who leads an atmospheric chemistry modelling group, and by Dr Richard Pope, a researcher with the NERC National Centre for Earth Observation. The project will be co-supervised by Dr Brian Kerridge who leads the Remote Sensing Group at the Rutherford Appleton Laboratory. You will work with state-of-the-art models and satellite data. Collaboration with the ECMWF will enable the research to feed through to an improved understanding tropospheric ozone and surface air quality.

Figure 1. Image of the Metop B satellite (with the GOME-2 instrument) in orbit.

Objectives

The primary aim of this project is to improve our understanding of ozone in the lower atmosphere, and thereby our understanding of its role in air quality globally. A main focus will be to allow the assimilation of RAL satellite observations (e.g. from GOME-2 satellite instrument) into the state-of-the-art system employed by ECMWF for the Copernicus Atmosphere Monitoring Service (CAMS). The research will involve:

  1. Analyse long-term trends in tropospheric ozone (and other related gases such as NO2) since 2003 from CAMS reanalyses. This will allow the first assessments of the global trends in atmospheric pollutants based on combined observations and model using a new CAMS dataset to be produced by late 2018 (without inclusion of the RAL ozone profiles). The dataset will be used to determine the magnitude of causes of trends in different regions. Comparisons will also be made with simulations use the Leeds atmospheric chemical transport model TOMCAT.
  2. Implement the assimilation of RAL GOME-2 ozone in the ECMWF CAMS C-IFS system. The accuracy of global estimates of tropospheric ozone from assimilation systems depend on the amount and accuracy of observations which are fed in. Experiments can be performed to test how the assimilation results change (e.g. for surface ozone) between experiments using the full profile information and an experiment which does not. Following this, certain time periods (Objective 1) can be produced with this new data to compare the impact of assimilating this new data (e.g. from GOME-1 and GOME-2) with the standard CAMS product.
  3. Use the new ECMWF CAMS ozone fields (developed in Objective 2) to explore known biases in model simulations. One likely focus will be on the Mediterranean area where past estimates from the CAMS system have systematically underestimated satellite observations (e.g. from GOME-2).

Depending on time and interest of the student, the project could also investigate the assimilation of RAL IASI CH4 profiles into the CAMS system (using a similar methodology for ozone) and test the impact on chemistry and derived emission fluxes.

Background

Society is becoming increasingly aware of the dangers of poor air quality. Air pollutants, such as ozone (O3), nitrogen dioxide (NO2) and particulate matter (PM2.5 and 10—particles with diameters of less than 2.5 and 10 μm, respectively), can have significant impacts on human health. Exposure to significantly elevated levels of surface ozone can cause reduced respiratory function and cardiovascular problems. It is estimated that poor UK air quality results in approximately 50 000 premature deaths annually and costs society £8.5–20.2 billion per year.

A good understanding of air quality – to inform the public and policy makers - depends on extensive and accurate observations. However, measurements at the surface are sparse (especially on a global scale) and do not provide information on how pollutants are transported at higher altitudes. Observations from satellite can provide this information but an important challenge is how to make use of the data that they provide. With the recent launch of TROPOMI on the Copernicus Sentinel 5P (13/10/2017), the use of satellite observations for air quality studies is set to increase significantly.

The best formal way of combining observations from different observing platforms is through data assimilation. In this procedure, observations are ingested into an atmospheric model so that the model is ‘nudged’ towards the observed value. In the absence of observations, an assimilation system will just produce a model prediction – which is dependent on the quality of the model. In a major initiative the ECMWF is leading the Copernicus Atmospheric Modelling Service (CAMS) project. In CAMS the ECMWF forecast model, with detailed schemes for atmospheric chemistry and aerosols is being used to produce daily forecasts of atmospheric composition (see Figure 2). However, these products do not currently use all of the available information from satellites and so, in some cases, may be no better than a model prediction.

Figure 2. CAMS forecast of surface ozone. These products do not currently make use of tropospheric ozone profile information from satellites.

Methodologies

The Rutherford Appleton Laboratory (STFC-RAL) has developed a state-of-the-art retrieval scheme to produce global ozone profile data from satellite ultraviolet sounders with unique sensitivity to the lower atmosphere (Miles et al, 2015). The ECMWF/CAMS has, in turn, implemented RAL ozone in the recent ERA-5. It has also implemented RAL NRT data in its satellite pre-processing system and undertaken preliminary data monitoring and trial assimilation. RAL has also developed the first scheme to retrieve height-resolved methane profiles from the Interferometric Atmospheric Sounder (Siddans et al, 2016). These schemes both employ Optimal Estimation through which each retrieved profile is accompanied by a set of vertical smoothing functions (averaging kernels) for each height level and an error covariance matrix.

To accurately use the RAL ozone (and methane) data through assimilation into CAMS, and thereby improve air quality forecasts, satellite “averaging kernels” must be included into the model to account for vertical smoothing, and careful attention must be paid to the vertical correlation of errors on retrieved profiles. The challenge will be to exploit optimally the satellite-retrieved information on tropospheric ozone and methane in a manner compatible with vertical correlations in the model background error covariance matrix; particularly for ozone.

Requirements

Undergraduate degree in any physical/chemical science, environmental science, computing, engineering or mathematics. An interest in atmospheric pollution and air quality. An interest in developing in-depth understanding of high performance computing and analysis of large model and observation datasets.

Training

Students will receive training in running and visualizing global model results both at the ECMWF and through the institute’s new Centre of Excellence for Modelling the Atmosphere and Climate (CEMAC: https://www.cemac.leeds.ac.uk/).

Co-supervision will involve regular meetings between all partners. In addition the successful PhD student will have access to a broad spectrum of training workshops put on by the Faculty that include an extensive range of supportive workshops in skills such as managing your degree to preparing for your viva (http://www.emeskillstraining.leeds.ac.uk/). There will also be opportunities to take part in field campaigns, international conferences, and training courses offered by other organisations. You will have the opportunity to make regular visits to STFC’s Rutherford Appleton Laboratory at Harwell and the ECMWF in Reading. The NERC National Centre for Earth Observation (NCEO) will provide training in data assimilation.

Research environment

The student will join the vibrant Atmospheric Composition and Aerosols research group of 10 Academic Staff. There are about 50 PhD students across the Institute for Climate and Atmospheric Science (ICAS) covering climate, dynamics, impacts, with extensive programmes in observations, modelling and lab studies. Wider interdisciplinary experience is guaranteed through our new cross-campus Priestley Centre (http://climate.leeds.ac.uk). Peer exchange and learning occurs through frequent institute and group seminars, discussion meetings and paper review groups.

The STFC-RAL Remote Sensing Group has an international standing in satellite sounding of atmospheric composition, drawing on expertise in spectrometry from UV through IR to millimetre wavelengths and VIS/IR imaging. The Group has developed state-of-the-art schemes to produce multi-year global data sets on ozone, methane, aerosol and cloud in NERC’s National Centre for Earth Observation (www.nceo.ac.uk) (NCEO) and in ESA’s “Climate Change Initiative” (cci.esa.int). Current research centres on increasing sensitivity to near-surface ozone and methane, derivation of particulate matter and photosynthetically active radiation (PAR) for applications including air quality, surface flux estimation and vegetation health. The Group collaborates with partners in the UK and international science community to exploit its datasets to test climate models and their representations of composition-climate interactions and surface-atmosphere interactions. A near-real time processing chain has been established on JASMIN (www.jasmin.ac.uk) to support operational users such as Copernicus Atmosphere Monitoring Service. The Group also provides scientific direction and expert support to develop future satellite missions and precursor airborne instruments.

ECMWF is one of the world’s leading meteorological services. Within this the CAMS project is unrivalled in its ambition to create a system for the comprehensive forecasts of atmospheric composition. By spending time at ECMWF the student will gain experience of working in an operational weather service.

The CASE support from NCEO will provide additional funding for these extended visits to ECMWF and RAL.

Further reading

Launch of Sentinel 5P Satellite. http://www.bbc.co.uk/news/science-environment-41604186

ECMWF CAMS system: http://atmosphere.copernicus.eu/

Miles, G. M., R. Siddans, B. J. Kerridge, B. G. Latter, and N. A. D. Richards. Tropospheric ozone and ozone profiles retrieved from GOME-2 and their validation, Atmos. Meas. Tech., 8, 385–398, 2015 www.atmos-meas-tech.net/8/385/2015/, doi:10.5194/amt-8-385-2015, 2015.

Pope, R.J., N.H. Savage, M.P. Chipperfield, S.R. Arnold and T.J. Osborn,
The influence of synoptic weather regimes on UK air quality: Analysis of satellite column NO2,
Atmos. Sci. Letts., 15, 211-217, doi:10.1002/asl2.492, 2014.

Siddans, R., Knappett, D., Kerridge, B., Waterfall, A., Hurley, J., Latter, B., Boesch, H., and Parker, R.: Global height-resolved methane retrievals from the Infrared Atmospheric Sounding Interferometer (IASI) on MetOp, Atmos. Meas. Tech., 10, 4135-4164, https://doi.org/10.5194/amt-10-4135-2017 , 2017

Related undergraduate subjects:

  • Applied mathematics
  • Astronomy
  • Atmospheric science
  • Chemical engineering
  • Chemistry
  • Computer science
  • Computing
  • Earth system science
  • Engineering
  • Environmental science
  • Geography
  • Geophysical science
  • Geophysics
  • Geoscience
  • Mathematics
  • Mechanical engineering
  • Meteorology
  • Natural sciences
  • Oceanography
  • Physical science
  • Physics
  • Remote sensing
  • Statistics