RTCA Prof Roy Harrison OBE - What’s next for monitoring and modelling?

Transcript

1. What’s next for monitoring and modelling? Roy M. Harrison University of Birmingham and National Centre for Atmospheric Science
2. 2024/2881 DIRECTIVE (EU) 2024/2881 OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 23 October 2024 on ambient air quality and cleaner air for Europe (recast) The Directive requires: (a) Air quality supersites (b) Measurement of Ultrafine Particles (c) Measurement of Oxidative Potential
3. Birmingham Air Quality Supersite (BAQS) • Instrumentation for highly advanced near real-time characterisation of atmospheric composition • Defra support for supersites in Birmingham, Manchester and London • Elucidate air quality trends and key processes
4. Pros of measuring Oxidative Potential • Oxidative Potential is a measure of the capacity of airborne particles (or gases) to oxidise organic molecules or trace metal ions • The most plausible explanation for the adverse effects of airborne particles is that they cause oxidative stress to human organs, which in turn causes harmful inflammation • Measurements of Oxidative Potential may reflect the capacity of particle exposures to cause inflammatory effects, leading to exacerbation of disease
5. Cons of measuring Oxidative Potential • There are several OP tests; they do not correlate strongly with one another • Inter-laboratory comparison has revealed poor reproducibility of OP measurements • Real-time measurement of OP shows it to be highly unstable and hence subject to rapid decay, so lab measurements do not reflect the lung dose • Few studies have shown OP to be a better predictor of adverse health effects than PM2.5 mass; most do not
6. Air quality and health implications of exposure to ultrafine particle pollution From Lelieveld et al.,Cardiovasc Res, cvag136, https://doi.org/10.1093/cvr/cvag136
7. Annual country-average mortality density per 100 000 population due to UFP exposure. From Lelieveld et al. (2026) Cardiovasc Res, cvag136, https://doi.org/10.1093/cvr/cvag136 The content of this slide may be subject to copyright: please see the slide notes for details.
8. In the 2021 Air Quality Guidelines Good Practice Statement for Ultrafine Particles, the World Health Organisation (WHO) recommend: Quantification of UFP as particle number concentration (PNC) >10nm Extended monitoring and characterisation of UFP Low concentration of PNC considered as <1000 cm-3 (24-hour mean) High PNC considered as >10,000 cm-3 (24-hour mean), or >20,000 cm-3 (1-hour mean)
9. Section 5 – Measurement of Ultrafine Particles D. Minimum number of sampling points for fixed measurements of ultrafine particles where high concentrations are likely to occur. Ultrafine particles shall be measured at selected locations in addition to other air pollutants. Sampling points to measure ultrafine particles shall coincide, where appropriate, with sampling points for particulate matter or nitrogen dioxide referred to in Point A of this Annex, and be sited in accordance with Section 4 of Annex VII. For that purpose, at least one sampling point per 5 million inhabitants shall be established at a location where high UFP concentrations are likely to occur. Member States that have fewer than 5 million inhabitants shall establish at least one sampling point for fixed measurements at a location where high UFP concentrations are likely to occur. For member States with more than 2 million inhabitants, monitoring supersites at urban background or rural background locations established in accordance with Article 10 shall not be included for the purpose of meeting the requirements on the minimum number of sampling points for UFP set here.
10. Trends in Particle Number at London, Marylebone Road Trends in particle number: 2010 -2021 Nucleation : <30nm Aitken: 30-100nm Accumulation: >100nm Annual mean concentration of each particle size mode split by year (2010-2021) and wind direction The 180° and 225° data show the greatest contribution of Marylebone Road traffic The 0° and 315° data are more reflective of background north London air Damayanti et al., Limited impact of diesel particle filters on road traffic emissions of ultrafine particles. Environ. Int, 107888(2023)
11. Temporal trends in Black Carbon and Accumulation Mode particle concentrations in relation to heavy duty truck traffic which did not meet the Euro VI emissions standard, 2013-2021.
12. Trends in particle number • We interpret the rapid reduction in BC and PNC as being largely attributable to the progressive uptake of diesel particle filters as Euro 5 and 6 standard vehicles have entered the fleet since 2011. • The greatest impact has been upon the BC and Aitken and Accumulation mode particles, with little change seen in the Nucleation mode particles, which are comprised largely of condensed lubricating oil, and form in the cooling exhaust after passage through the particle filter (Harrison et al., 2018)
13. Conclusions from Marylebone Road Measurements • Diesel particle filters (DPF) have been hugely successful in reducing emissions of solid Black Carbon Particles • Semi-volatile vapours arising from unburnt fuel and lubricating oil pass through the DPF • Subsequent condensation of the vapours in cooler ambient air leads to formation of nucleation mode particles whose concentrations have changed little since the introduction of DPFs.
14. Daily cycle of: Black Carbon (BC) and number concentration (N) levels for monitoring sites across Europe (black line: BC; red line: Particle number) 3 N (>5nm) 45000 N (>7nm) BC 4.0 BC N (>7nm) BC 40000 30000 2.5 25000 2.0 20000 1.5 15000 1.0 10000 BCN 5000 0.5 NK LUG 0.0 0 2 4 6 8 10 12 14 16 18 20 22 0 2 4 6 8 BC 6 35000 4 30000 25000 3 20000 2 15000 10000 1 BERN 0 0 4 6 8 6 8 10 12 14 16 18 20 22 Hour (UTC) 12 N (>7nm) 45000 BC 10 40000 35000 8 30000 25000 6 20000 4 15000 10000 2 MR 5000 0 0 0 10 12 14 16 18 20 22 2 4 6 8 10 12 14 16 18 20 22 Hour (UTC) Hour (UTC) 4.0 50000 3 45000 N (>2.5nm) HU 40000 BC N (>2.5nm) SCO BC 3.5 3.0 35000 2.5 30000 25000 2.0 20000 1.5 15000 1.0 10000 0.5 5000 0.0 0 0 2 4 6 8 10 12 14 16 18 20 22 Hour (UTC) 0 2 4 6 8 10 12 14 16 18 20 22 Hour (UTC) Black Carbon (µg/m 3) Number of particles (n/cm ) Black Carbon (µg/m 3) 5 40000 2 4 50000 3 3 N (>7nm) 45000 Black Carbon (µg/m 3) Number of particles (n/cm ) 50000 0 2 Hour (UTC) Hour (UTC) 5000 0 10 12 14 16 18 20 22 Number of particles (n/cm ) 0 From Reche et al. (2011) 3.5 3.0 35000 Black Carbon (µg/m 3) Number of particles (n/cm ) 50000
15. Particle Formation: Why Less Pollution Can Mean More Particles PM2. 5
16. Variations in source contributions of particle number concentration under long-term emission control in winter of urban Beijing Shang et al., https://doi.org/10.1016/j.envpol.2022.119072
17. Ongoing uncoordinated anthropogenic emission abatement promotes atmospheric new particle growth in a Chinese megacity Tang et al., https://doi.org/10.1038/s41467-025-62011-6
18. Carbon Capture and Storage: a Potential Enhancer of New Particle Formation Zhu et al., Real-Time Monitoring of Emissions from Monoethanolamine-Based Industrial Scale Carbon Capture Facilities. Environmental Science & Technology 2013 47 (24), 14306-14314. DOI: 10.1021/es4035045 Xie et al., Atmospheric Fate of Monoethanolamine: Enhancing New Particle Formation of Sulfuric Acid as an Important Removal Process. Environmental Science & Technology 2017 51 (15), 8422-8431. DOI: 10.1021/acs.est.7b02294
19. The Direction of Modelling (1) • Receptor models of particulate matter and VOC for source apportionment • Models capable of prediction of secondary organic aerosol from known precursors • Models of ultrafine particles, including secondary particles and formation in plumes from CCS plants
20. Future concentrations of UFP • There are two reasons why air pollution controls may lead to increased concentrations of ultrafine particles from secondary formation • Reducing PM2.5 reduces the condensation sink which favours the formation and survival of new particles • Reducing NOx favours the formation of larger Highly Oxygenated Molecules of lower volatility which enhance particle growth rates and hence reduce particle losses by coagulation. • Amine releases from CCS may enhance UFP formation.
21. Annual map for PNC between 10 and 100 nm (PNC10_100) at 10 m × 10 m resolution for the 2019 BAU case in UK West Midlands
22. The Direction of Modelling (2) Machine Learning Algorithms: ➢Now heavily used in weather forecasting ➢Able to de-weather air quality datasets revealing trends more clearly ➢Offer enhanced capability for air quality prediction ➢Caveat: these models are valid only within the parameter space used to train them. They cannot reliably predict a very different future!
23. THANK YOU